risk assessment and risk management plan for dir … · risk assessment and risk management plan...

Risk Assessment and Risk Management Plan for

DIR 059/2005

Commercial Release of herbicide tolerant (Roundup Ready Flex® MON 88913) and

herbicide tolerant/insect resistant (Roundup Ready Flex® MON 88913/Bollgard II®) cotton

south of latitude 22° South in Australia

Applicant: Monsanto Australia Ltd

February 2006

Executive Summary Introduction The Gene Technology Regulator (the Regulator) has made a decision to issue a licence for dealings involving the intentional release (DIR) of herbicide tolerant and herbicide tolerant/insect resistant genetically modified (GM) cotton into the Australian environment, in respect of application DIR 059/2005 from Monsanto Australia Ltd (Monsanto).

The DIR 059/2005 licence permits the commercial release of the two GM cotton lines on an unrestricted basis south of latitude 22º South. The licence also permits the use of seed from the GM cotton plants as stockfeed, including in northern Australia where measures to limit spread and persistence of the GMOs have been imposed.

The Gene Technology Act 2000 (the Act) and the Gene Technology Regulations 2001 (the Regulations) govern the process undertaken by the Regulator before a decision is made on whether or not to issue a licence. The decision is based upon a risk assessment and risk management plan (RARMP) prepared by the Regulator in consultation with a wide range of experts, agencies and authorities and the public.

More information on the process required for the comprehensive assessment of licence applications to release a genetically modified organism (GMO) into the environment is available from the Office of the Gene Technology Regulator (OGTR) (Free call 1800 181 030) or at <http://www.ogtr.gov.au/ir/process.htm>.

The application Monsanto applied for a licence to release herbicide tolerant and herbicide tolerant/insect resistant GM cotton, Roundup Ready Flex® MON 88913 (referred to as Roundup Ready Flex®) and Roundup Ready Flex®/Bollgard II®, into the environment. Monsanto is seeking approval for unrestricted, commercial scale planting of the two GM cotton lines in all areas south of latitude 22º South in Australia.

The GM cotton lines have either one or five introduced genes. Roundup Ready Flex® cotton contains two copies of a herbicide tolerance gene (cp4 epsps) isolated from a common soil bacterium, Agrobacterium sp. strain CP4. The gene expresses a protein that provides tolerance to glyphosate, the active ingredient in Roundup Ready® Herbicide, and enables the herbicide to be applied for weed control in the GM cotton crop.

Roundup Ready® cotton, which was approved for commercial release south of latitude 22º South in 2000 (refer DIR 023/2002) and is now widely grown, contains only one copy of the cp4 epsps gene. Glyphosate can only be applied to Roundup Ready® cotton plants up to the four-leaf stage of growth (ie prior to flower formation, approximately 3 to 5 weeks after planting) because later application can lead to yield loss. If weather conditions prevent spraying in this short window, then farmers must use more intensive methods of weed management (eg inter-row cultivation, hand weeding, shielded/spot spraying). As Roundup Ready Flex® cotton has increased and prolonged expression of the protein from the two herbicide tolerance genes, it is tolerant to glyphosate application throughout the growing season (approximately 24 to 28 weeks) and will provide greater flexibility in the timing of spraying.

Roundup Ready Flex®/Bollgard II® cotton was produced by conventional crossing of Roundup Ready Flex® cotton with Bollgard II® cotton (approved for commercial release south of latitude 22º South in 2002 under DIR 012/2002) and contains all the genes introduced into each of the parent GMOs. This means that (in addition to the two copies of the herbicide tolerance gene) the plants contain two insect resistance genes (cry1Ac and cry2Ab)

Executive summary (February 2006) I

from a common soil bacterium, Bacillus thuringiensis. The genes express two insect resistance proteins that are specifically toxic to caterpillars of some lepidopterans (butterflies and moths), including Helicoverpa armigera and H. punctigera. These are the two major pests of cotton in Australia.

Roundup Ready Flex®/Bollgard II® cotton also contains three bacterial genes from the latter parent (nptII, aad, and uidA) that were used to select modified plants in the laboratory.

Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton have been approved previously for limited and controlled releases under DIR licences 035/2003 and 055/2004.

The application requests approval for commercial scale cultivation without containment measures south of latitude 22º South, and the use of the GM cotton plants and their by-products in the same manner as non-GM or other commercially released GM cotton. This would include sale of seed for commercial planting, use in human food and stockfeed, sale of lint, export of seed and unrestricted transport south of latitude 22º South. Monsanto also proposes restricted transport of ginned cotton seed from the release to areas north of latitude 22º South for use only as stockfeed.

Under Australia’s integrated framework for the regulation of genetically modified organisms Monsanto will also require approval from Food Standards Australia New Zealand for the use of oil and linters from the Roundup Ready Flex® cotton in food. In addition, Monsanto will require approval from Australian Pesticides and Veterinary Medicines Authority to extend the current registration of Roundup Ready® Herbicide to allow its application to Roundup Ready Flex® cotton (including application beyond the seedling stage).

Risk assessment Background The risk assessment first considered what harm to the health and safety of people or the environment could arise as a result of gene technology, and how it could happen, during the proposed release of the GM cotton lines into the environment (hazard identification refer to Chapter 2 for more information).

The risks to people and the environment from the proposed commercial release were assessed in comparison to non-GM cotton and GM cotton lines previously approved for commercial release by the Regulator, in the context of the intended agronomic management practices, and the environmental conditions in the regions proposed for the release.

Hazards are particular sets of circumstances (events) that might give rise to adverse outcomes (i.e. cause harm). When an event was considered to have some chance of causing harm, it was identified as posing a risk that required further assessment.

Each event associated with an identified risk was then assessed to determine the seriousness of harm (consequence—ranging from marginal to major) and the chance of harm (likelihood—ranging from highly unlikely to highly likely). The level of risk (ranging from negligible to high) was then estimated using a Risk Estimate Matrix (refer to Chapter 2 for more information).

Hazard identification Of the 41 events compiled during the hazard identification process, eight were selected for further assessment. The potential adverse outcomes to the environment associated with these events were: toxicity for non-target invertebrates and weediness. The remaining 33 events were not assessed further as they were considered not to give rise to an identified risk to human health and safety or the environment (refer to Chapter 3 for more information).

Executive summary (February 2006) II

Risk of toxicity to non-target invertebrates One event was considered that might cause toxicity in non-target invertebrates via direct or indirect ingestion of the insect resistance proteins in the GM cotton lines as a result of this release (event 1).

The risk assessment considered the consequence and likelihood of harm that might result from the above event. The estimate of risk for this event is negligible.

Risk of weediness Seven events were considered that might result in the GM cotton lines exhibiting greater weediness than non-GM cotton or other GM cotton lines previously approved for commercial release:

• Expression of the herbicide tolerance gene increasing spread and persistence of the GM cotton plants through tolerance to glyphosate (event 2)

• Expression of the herbicide tolerance and insect resistance genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced insect attack on the plants (event 3)

• Dispersal of GM seed during transport or storage north of latitude 22º South (event 4)

• Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South (event 5)

• Dispersal of GM seed via flooding north of latitude 22º South (event 6)

• Expression of the herbicide tolerance gene in other cultivated (non-GM or commercially released GM) or naturalised cotton plants increasing spread and persistence through providing glyphosate tolerance (event 7)

• Expression of the herbicide tolerance and insect resistance genes in combination in other cultivated (non-GM or commercially released GM) or naturalised cotton plants increasing spread and persistence through providing glyphosate tolerance and reduced insect attack on the plants (event 8).

The risk assessment considered the consequence and likelihood of harm that might result from each of the above events. The estimates of risk for events 2, 3, 6, 7 and 8 are negligible and for events 4 and 5 are low.

Risk management The Risk Analysis Framework defines negligible risks as insubstantial, with no present need to invoke actions for their mitigation. Low risks are defined as minimal but may invoke actions for mitigation beyond normal practices.

The level of risk to the health and safety of people or the environment for six of the eight events that were assessed was estimated as negligible. Therefore, no risk management measures have been imposed. The risk estimate for the two remaining events was low. Hence, treatment measures have been imposed to manage these risks.

The licence conditions, detailed in Chapter 6 of the RARMP, require the applicant to minimise dissemination of GM cotton seed, and spread and persistence of GM cotton plants, in areas north of latitude 22º South.

Executive summary (February 2006) III

Conclusions of the RARMP The risk assessment concludes that this commercial release of herbicide tolerant and herbicide tolerant/insect resistant GM cotton lines poses low to negligible risks to the health and safety of people and the environment as a result of gene technology.

The risk management plan concludes that the negligible risks do not require specific risk treatment measures. However, licence conditions have been imposed to treat the low risks to the environment which relate to the use of GM cotton seed as stockfeed in northern Australia. Therefore, licence conditions have been imposed to minimise spread and persistence of the GMOs north of latitude 22º South.

Executive summary (February 2006) IV

Table of Contents EXECUTIVE SUMMARY.................................................................................................................................... I

INTRODUCTION .................................................................................................................................................... I THE APPLICATION ................................................................................................................................................ I RISK ASSESSMENT...............................................................................................................................................II

Background ...................................................................................................................................................II Hazard identification .....................................................................................................................................II Risk of toxicity to non-target invertebrates ................................................................................................. III Risk of weediness ........................................................................................................................................ III

RISK MANAGEMENT .......................................................................................................................................... III CONCLUSIONS OF THE RARMP ........................................................................................................................ IV

ABBREVIATIONS VII TECHNICAL SUMMARY .................................................................................................................................. 1

INTRODUCTION ................................................................................................................................................... 1 SECTION 1 APPLICATION .............................................................................................................................. 1 SECTION 2 RISK ASSESSMENT ....................................................................................................................... 3 SECTION 3 RISK MANAGEMENT .................................................................................................................... 8

3.1 Risk treatment measures for identified risks .............................................................................. 8 3.2 Other regulatory considerations.................................................................................................. 9

SECTION 4 CONCLUSIONS OF THE RARMP................................................................................................. 10 CHAPTER 1 RISK ASSESSMENT CONTEXT ........................................................................................ 11

SECTION 1 BACKGROUND........................................................................................................................... 11 SECTION 2 THE GMOS AND PROPOSED DEALINGS ...................................................................................... 12

2.1 The proposed dealings.............................................................................................................. 12 2.2 The parent organism................................................................................................................. 13 2.3 The GMOs................................................................................................................................ 13 2.4 The introduced genes and their products .................................................................................. 14 2.5 Method of genetic modification ............................................................................................... 21

SECTION 3 THE RECEIVING ENVIRONMENT ................................................................................................. 22 3.1 Size and duration of the proposed release ................................................................................ 22 3.2 Major cotton growing regions of Australia .............................................................................. 22 3.3 Environmental conditions suitable for growing cotton............................................................. 23 3.4 Presence of the introduced proteins in the receiving environment ........................................... 23 3.5 Presence of similar GM cotton in the receiving environment .................................................. 24 3.6 Agronomic practices for these GM cotton lines ....................................................................... 25

SECTION 4 PREVIOUS AUSTRALIAN AND INTERNATIONAL APPROVALS....................................................... 25 4.1 Previous Australian approvals of the same or similar GMOs................................................... 25 4.2 International approvals ............................................................................................................. 27

CHAPTER 2 RISK ASSESSMENT............................................................................................................. 28 SECTION 1 INTRODUCTION.......................................................................................................................... 28 SECTION 2 HAZARD CHARACTERISATION ................................................................................................... 29

2.1 Production of a substance toxic to people ................................................................................ 37 2.2 Production of a substance allergenic to people......................................................................... 41 2.3 Production of a substance toxic to organisms other than people .............................................. 44 2.4 Spread and persistence of the GM cotton in the environment .................................................. 50 2.5 Gene flow by vertical gene transfer.......................................................................................... 52 2.6 Gene flow by horizontal gene transfer...................................................................................... 54 2.7 Unintended changes in toxicity ................................................................................................ 56 2.8 Unintended changes in biochemistry or physiology................................................................. 57 2.9 Unintended effects on existing pests or weeds ......................................................................... 59 2.10 Secondary impacts.................................................................................................................... 62 2.11 Unauthorised activities ............................................................................................................. 65

SECTION 3 RISK ESTIMATE PROCESS FOR IDENTIFIED RISKS........................................................................ 66

Table of contents (February 2006) V

CHAPTER 3 ......................RISK ESTIMATES FOR TOXICITY FOR NON-TARGET INVERTEBRATES................................................................................................................ 67

SECTION 1 BACKGROUND........................................................................................................................... 67 SECTION 2 CONSEQUENCE AND LIKELIHOOD ASSESSMENTS ....................................................................... 68

2.1 Toxicity of non-GM cotton ...................................................................................................... 68 2.2 Event 1: Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by

non-target invertebrates as a result of this release .................................................................... 68 SECTION 3 RISK ESTIMATES........................................................................................................................ 72

CHAPTER 4 RISK ESTIMATES FOR WEEDINESS.............................................................................. 73 SECTION 1 BACKGROUND........................................................................................................................... 73 SECTION 2 CONSEQUENCE AND LIKELIHOOD ASSESSMENTS ....................................................................... 74

2.1 Weediness of non-GM cotton................................................................................................... 74 2.2 Event 2: Expression of the cp4 epsps gene construct increasing spread and persistence of the

GM cotton plants through tolerance to glyphosate. .................................................................. 75 2.3 Event 3: Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing

spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory.............................................................................................................. 79

2.4 Event 4: Dispersal of GM seed during transport or storage north of latitude 22º South........... 81 2.5 Event 5: Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude

22º South. ................................................................................................................................. 83 2.6 Event 6: Dispersal of GM seed via flooding north of latitude 22º South. ................................ 85 2.7 Event 7: Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense

cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance ................................................................................................................................... 86

2.8 Event 8: Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory...................................... 88

2.9 Uncertainty ............................................................................................................................... 90 SECTION 3 RISK ESTIMATES........................................................................................................................ 90

CHAPTER 5 RISK MANAGEMENT ......................................................................................................... 95 SECTION 1 BACKGROUND........................................................................................................................... 95 SECTION 2 OTHER AUSTRALIAN REGULATORS ........................................................................................... 95 SECTION 3 RISK TREATMENT MEASURES FOR IDENTIFIED RISKS ................................................................. 96 SECTION 4 GENERAL RISK MANAGEMENT................................................................................................... 96

4.1 Other risk management considerations..................................................................................... 96 SECTION 5 MONITORING AND COMPLIANCE ............................................................................................... 98 SECTION 6 CONCLUSIONS OF THE RARMP................................................................................................. 98

CHAPTER 6 LICENCE CONDITIONS ..................................................................................................... 99 SECTION 1 INTERPRETATIONS AND DEFINITIONS ........................................................................................ 99 SECTION 2 CONDITIONS............................................................................................................................ 101

REFERENCES ........................................................................................................................................... 106 APPENDIX A DEFINITIONS OF RISK ANALYSIS TERMS ............................................................... 123 APPENDIX B SUMMARY OF SUBMISSIONS RECEIVED FROM PRESCRIBED EXPERTS,

AGENCIES AND AUTHORITIES ON THE APPLICATION ...................................... 125 APPENDIX C SUMMARY OF PUBLIC SUBMISSIONS RECEIVED ON THE APPLICATION .... 126 APPENDIX D SUMMARY OF SUBMISSIONS RECEIVED FROM PRESCRIBED EXPERTS,

AGENCIES AND AUTHORITIES ON THE CONSULTATION RARMP ................... 127 APPENDIX E SUMMARY OF PUBLIC SUBMISSIONS RECEIVED ON THE CONSULTATION

RARMP ................................................................................................................................. 129

Table of contents (February 2006) VI

DIR 59/2005—Risk Assessment and Risk Management Plan Office of the Gene Technology Regulator

Abbreviations aad Gene encoding the aminoglycoside adenyltransferase from

Escherichia coli AAD Aminoglycoside adenyltransferase from E. coli ANZFA Australia New Zealand Food Authority (now FSANZ) APVMA Australian Pesticides and Veterinary Medicines Authority AQIS Australian Quarantine Inspection Service bp Basepair of nucleic acid Bt Bacillus thuringiensis Btk Bacillus thuringiensis variety kurstaki CaMV Cauliflower mosaic virus CCI Confidential Commercial Information as declared under section 185 of

the Act cp4 epsps epsps gene from Agrobacterium sp. strain CP4 CP4 EPSPS EPSPS protein from Agrobacterium sp. strain CP4 cry Gene encoding a Cry protein Cry protein Crystal insecticidal protein isolated from Bacillus thuringiensis CSIRO Commonwealth Scientific and Industrial Research Organisation DIR Dealing involving Intentional Release DNA Deoxyribonucleic acid EFSA European Food Safety Authority EPSPS 5-enolpyruvylshikimate-3-phosphate synthase FMV Figwort mosaic virus FSANZ Food Standards Australia New Zealand (formerly ANZFA) g Gram GM Genetically Modified GMAC Genetic Manipulation Advisory Committee GMO Genetically Modified Organism GTTAC Gene Technology Technical Advisory Committee GUS The β-glucuronidase protein from E. coli ha Hectare IgE Immunoglobulin E kg Kilogram km Kilometre LD50 Amount of a substance given in a single dose that causes death in 50% of a

test population of an organism mg Milligram MRL Maximum residue limit mRNA Messenger ribonucleic acid µg Microgram ng Nanogram NHMRC National Health and Medical Research Council NICNAS National Industrial Chemicals Notification and Assessment Scheme nos Gene encoding nopaline synthase nptII Gene encoding the neomycin phosphotransferase type II protein from

E. coli NPTII Neomycin phosphotransferase type II from E. coli OGTR Office of the Gene Technology Regulator PCR Polymerase Chain Reaction

Abbreviations (February2006) VII


T-DNA Transfer deoxyribonucleic acid TIMS committee Transgenic and Insect Management Strategy committee TGA Therapeutic Goods Administration uidA Gene encoding the β-glucuronidase protein from E. coli USDA United States Department of Agriculture US EPA United States Environmental Protection Agency US FDA United States Food and Drug Administration

Abbreviations (February2006) VIII


Technical summary Introduction The Gene Technology Regulator (the Regulator) has decided to issue a licence (DIR 059/2005) to Monsanto Australia Ltd (Monsanto) for dealings involving the intentional release of herbicide tolerant and herbicide tolerant/insect resistant genetically modified (GM) cotton into the environment.

The DIR 059/2005 licence permits commercial release of the two GM cotton lines on an unrestricted basis south of latitude 22º South. The licence also permits the use of seed from the GM cotton plants as stockfeed, including in northern Australia where measures to limit the spread and persistence of the GMOs have been imposed.

The Gene Technology Act 2000 (the Act), the Gene Technology Regulations 2001 (the Regulations) and corresponding state and territory law govern the comprehensive and highly consultative process undertaken by the Regulator before making a decision whether or not to issue a licence to deal with a GMO.

The Regulator’s Risk Analysis Framework explains the approach used to evaluate licence applications and to develop the Risk Assessment and Risk Management Plans (RARMPs) that form the basis of her decisions1.

The RARMP for DIR 059/2005 has been finalised in accordance with the gene technology legislation. Matters raised in the consultation process regarding risks to the health and safety of people or the environment from the dealings proposed by the applicant were taken into account by the Regulator in deciding to issue a licence and the conditions that have been imposed.

Consistent with Australia’s integrated regulatory framework for gene technology, the Regulator has also liaised closely with other regulatory agencies that have been considering parallel applications relating to this release, namely Food Standards Australia New Zealand (FSANZ) and the Australian Pesticides and Veterinary Medicines Authority (APVMA), to avoid duplication and enable coordinated decision making.

Section 1 Application Title: Commercial release of herbicide tolerant (Roundup Ready Flex® MON 88913) and herbicide

tolerant/insect resistant (Roundup Ready Flex® MON 88913/Bollgard II®) cotton south of latitude 22° South in Australia*

Applicant: Monsanto Australia Ltd Common name of the parent organism: Cotton Scientific name of the parent organism: Gossypium hirsutum L. Modified trait(s): Prolonged herbicide tolerance and/or insect resistance, antibiotic resistance, reporter gene

expression Identity of the gene(s) responsible for the modified trait(s):

• Two copies of the cp4 epsps gene from the bacterium Agrobacterium sp. strain CP4 (herbicide tolerance)

• cry1Ac and cry2Ab genes from the bacterium Bacillus thuringiensis (insect resistance) • nptII gene from the bacterium Escherichia coli (antibiotic resistance) • uidA gene from the bacterium Escherichia coli (reporter gene)

Proposed location(s): South of latitude 22° South and use of seed as stockfeed, including in northern Australia

1 More information on the assessment of licence applications and copies of the Risk Analysis Framework are available from the Office of the Gene Technology Regulator (OGTR). Free call 1800 181 030 or at <http://www.ogtr.gov.au/ir/process.htm> and <http://www.ogtr.gov.au/pdf/public/ raffinal2.2.pdf> respectively.

Technical summary (February 2006) 1


Proposed release size: Phased introduction commencing with 20,000 hectares in current cotton growing areas of NSW and QLD in 2006 to commercial scale planting in subsequent years, potentially including other areas suitable for cotton growing south of latitude 22° South

Proposed time of release: Ongoing from 2006 *The title of the licence application submitted by Monsanto is Licence Application to the OGTR for agricultural use of Roundup Ready Flex® technology (MON 88913) in cotton

Monsanto applied for a licence to release herbicide tolerant and herbicide tolerant/insect resistant GM cotton lines, Roundup Ready Flex® MON 88913 (referred to as Roundup Ready Flex®) and Roundup Ready Flex®/Bollgard II® respectively, into the environment. Monsanto is seeking approval for unrestricted, commercial scale planting of the two GM cotton lines south of latitude 22º South.

The company anticipates a phased introduction in the current cotton growing areas of New South Wales and southern Queensland. Future plantings may occur in other areas south of latitude 22° South that are suitable for growing cotton. The GM cotton lines may also be planted on a small scale for evaluation, demonstration, education and research purposes.

Some details of the DNA sequence of the Roundup Ready® Flex gene construct have been declared Confidential Commercial Information (CCI) under section 185 of the Act. However, this information was made available to the prescribed expert groups and agencies that were consulted in the preparation of the RARMP.

The GM cotton lines have either one or five introduced genes. Roundup Ready Flex® cotton contains two copies of the cp4 epsps gene (encoding the CP4 EPSPS protein) from Agrobacterium sp. strain CP4. The bacterial CP4 EPSPS protein provides tolerance to glyphosate, the active constituent in Roundup Ready® Herbicide.

Roundup Ready® cotton, which was approved for commercial release south of latitude 22º South in 2000 and is now widely grown, contains only one copy of the cp4 epsps gene. Glyphosate herbicide can only be applied over the top of Roundup Ready® cotton plants up to the four-leaf stage of growth (ie prior to flower formation, approximately 3 to 5 weeks after planting) because later applications can lead to yield loss. As Roundup Ready Flex® cotton has increased and prolonged expression of the cp4 epsps gene, it is tolerant to glyphosate throughout the growing season (approximately 24 to 28 weeks).

Roundup Ready Flex®/Bollgard II® cotton was produced by conventional crossing of Roundup Ready Flex® cotton with Bollgard II® cotton (approved for commercial release south of latitude 22º South in 2002 under DIR 012/2002) and contains all the genes introduced into each of the parent GMOs. This means that the plants contain (in addition to the two copies of the herbicide tolerance gene) the cry1Ac and cry2Ab genes (encoding the Cry1Ac and Cry2Ab Crystal insecticidal proteins). The insect resistance proteins, Cry1Ac and Cry2Ab, are specifically toxic to caterpillars of some lepidopterans (butterflies and moths), including Helicoverpa armigera and H. punctigera. These are the two major pests of cotton in Australia.

Roundup Ready Flex®/Bollgard II® cotton also contains the nptII and aad antibiotic resistance marker genes (encoding the kanamycin and neomycin resistance protein, and the streptomycin and spectinomycin resistance protein, respectively) and the uidA reporter gene (encoding the β-glucuronidase protein). The aad gene is not expressed in the GM cotton plants because it is under the control of a bacterial regulatory sequence that is not active in plants. The uidA gene enables visual identification of plant material that has incorporated the intended genes.

More detailed information on the GMOs, the introduced genes and their products is provided in Chapter 1.



No specific containment measures have been proposed for the plantings in areas south of latitude 22° South. Monsanto proposes to use the GM cotton plants and their by-products in the same manner as non-GM or other commercially released GM cotton. This would include the sale of seed for commercial planting, use of oil and linters in human food, use of cotton seed in stockfeed, sale of lint, export of seed and unrestricted transport south of latitude 22º South. Monsanto also proposes restricted transport of ginned cotton seed from the release to areas north of latitude 22º South for use as stockfeed.

Section 2 Risk assessment The risk assessment considered information contained in the application, previous GM cotton assessments, current scientific knowledge, and issues relating to risks to human health and safety and the environment raised in submissions received during consultation with a wide range of prescribed experts, agencies and authorities, including all local councils in southern Australia, on the application (summarised in Appendix B) and on the RARMP (see Appendix D).

Advice received from the public on the application and from consultation on the RARMP is summarised in Appendices C and E, respectively.

The risk assessment first considered what harm to the health and safety of people or the environment could arise due to gene technology, and how it could happen during this release of GMOs into the environment (hazard identification), in comparison to non-GM and commercially released GM cotton and in the context of the proposed release.

A hazard (source of potential harm) may be an event, substance or organism (OGTR 2005). The hazard identification process resulted in the compilation of a list of 41 events that describe sets of circumstances (events) by which the proposed release could potentially give rise to adverse outcomes.

A risk is identified when a hazard is considered to have some chance of causing harm to people and/or the environment. Those events that do not lead to an adverse outcome, or could not reasonably occur, do not advance in the risk assessment process. The events that are considered to have the potential to lead to adverse outcomes are assessed further to determine the seriousness of harm (consequence) that could result and how likely it is that the harm would occur. The level of risk is then estimated using the Risk Estimate Matrix (see below and Chapter 2).

RISK ESTIMATE

Highly Likely Low Moderate High High

Likely Negligible Low High High

Unlikely Negligible Low Moderate High

LIK

ELIH

OO

D

Highly Unlikely Negligible Negligible Low Moderate

Marginal Minor Intermediate Major CONSEQUENCES

Risk Estimate Matrix: A negligible risk is considered to be insubstantial with no present need to invoke actions for mitigation. A low risk is considered to be minimal but may invoke actions for mitigation beyond normal practices. A moderate risk is considered to be of marked concern that will necessitate actions for mitigation that need to be demonstrated as effective. A high risk is considered to be unacceptable unless actions for mitigation are highly feasible and effective.



Eight of the 41 events characterised in the hazard identification process for the proposed release were identified as requiring further assessment. The potential adverse outcomes associated with these events were: toxicity for non-target invertebrates and weediness. These identified risks were assessed in comparison to the parent organism and other GM cotton lines previously approved for commercial release, in the context of the intended agronomic management practices, and the environmental conditions in the regions where the proposed release might occur.

The consequence and likelihood assessments used to derive risk estimates for these eight events are summarised in Table 1 (the detailed risk assessments are in Chapters 3 and 4). More information on the remaining 33 events that were considered not to give rise to an identified risk is provided in Chapter 2.

If a risk is estimated to be higher than negligible, risk treatment measures may be required to protect the health and safety of people or the environment.

Table 1 Summary table for the risk assessment Potential adverse outcome

Event that may give rise to the

adverse outcome

Consequence assessment

Likelihood assessment Risk estimate

Risk evaluation

Toxicity for non-target invertebrates (see Ch 3)

Event 1 Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates as a result of this release.

Minor • The Cry1Ac and

Cry2Ab proteins may be toxic to some non-target lepidopteran insects.

• The Cry1Ac and Cry2Ab proteins may be toxic to some dipteran insects at high concentrations.

• Field studies have indicated that growing Bollgard II® cotton plants does not have any significant effect on the sizes or variety of non-target invertebrate populations in GM cotton fields, as compared to unsprayed non-GM cotton.

Highly unlikely • Roundup Ready® Flex/

Bollgard II® plants are expected to progressively replace the commercially released Roundup Ready®/ Bollgard II® plants already grown in the release areas.

• Non-target invertebrates do not appear to be sensitive to the levels of Cry1Ac and Cry2Ab expressed in commercially released Bollgard II® plants.

• The insect resistance of the Roundup Ready Flex®/Bollgard II® plants has been shown to be similar to that of Bollgard II® plants, suggesting that expression levels of the insecticidal proteins are equivalent.

Negligible No specific treatment options are required.

Weediness (see Ch 4)

Event 2 Expression of the cp4 epsps gene construct increasing spread and persistence of the GM cotton plants through tolerance to glyphosate

Minor • Cotton is not a

serious weed in southern Australia because of the limited availability of water, nutrients and suitable temperature conditions.

• Although glyphosate is the most widely used herbicide in Australia today, it is not generally used to

Highly unlikely • The proposed release would

result in the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton.

• Similar commercially approved glyphosate tolerant cotton lines are already extensively cultivated and transported and have not become




Potential adverse outcome


adverse outcome



Risk evaluation

control established cotton plants as the herbicide is not effective on cotton beyond the seedling stage (plants are damaged but not killed).

• Glyphosate tolerant cotton volunteers are effectively controlled by mechanical means or, if still at the seedling stage, by the use of alternative herbicides.

problematic weeds. • Expression of the cp4 epsps

gene construct is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

• The chance of volunteer GM plants establishing as weeds by finding suitable ecological niches would be no greater than for the non-GM parent.

Event 3 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory

Minor • Cotton is not a

serious weed in southern Australia because of the limited availability of water, nutrients and suitable temperature conditions.

• Lepidopteran herbivory is not an important limiting factor on the spread and persistence of cotton in southern Australia.

• The herbicide tolerance and insecticidal genes operate through independent, unrelated biochemical mechanisms and there is no evidence of any interaction.

Highly unlikely • The proposed release would

result in the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton.

• Similar commercially approved GM cotton plants containing the cp4 epsps, cry1Ac and cry2Ab genes in combination are already extensively cultivated and transported and have not become problematic weeds.

• Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).







adverse outcome



Risk evaluation

Event 4 Dispersal of GM seed during transport or storage north of latitude 22º South

Minor • Although protection

from lepidopteran insect herbivory may provide some selective advantage in northern Australia, current weed management practices would be able to control cotton volunteers around storage areas and on roadsides.

• The use of glyphosate to control weeds on roadsides would be ineffective on established cotton plants (whether GM or non-GM) and other means are required to control these plants.

• Glyphosate tolerant cotton volunteers are susceptible to other forms of chemical and mechanical control.

Unlikely • The applicant intends to

apply the industry standard of transporting ginned cotton seed in covered containers/vehicles.

• Seed from similar GM cotton plants containing the same introduced genes have been similarly transported to areas north of latitude 22º South for stockfeed since 2000 and have not become problematic weeds.

• Any incident involving spillage of GM cotton seed in northern Australia can be readily controlled through cleaning of the site of the spill.

• Expression of the introduced genes is not expected to alter susceptibility to the environmental conditions that limit the establishment and persistence of cotton in northern Australia (eg plant competition, fire, herbivory by non-lepidopteran insects and variable availability of water and nutrients).

• Surveys of transport routes indicate that survival of cotton volunteers is limited to disturbed environments.

• Cotton volunteers around storage areas are easily controlled and those on roadsides are generally controlled by plant competition and roadside management practices (slashing).

Low Treatment options proposed to minimise dissemination of seed from the GM cotton lines, and spread and persistence of GM cotton plants in areas north of latitude 22º South.

Event 5 Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South

Minor • Protection from

lepidopteran insect herbivory may provide some selective advantage in northern Australia.

• Glyphosate tolerant cotton volunteers are effectively controlled by mechanical means or, if still at the seedling stage, by the

Unlikely • Seed from similar GM cotton

lines containing the same introduced genes has been used as stockfeed in northern Australia since 2000 and these GM cotton lines have not become problematic weeds.

• Surveys of the incidence of cotton volunteers in areas where stock is fed cotton seed indicate that cotton

Low Treatment options proposed to minimise dissemination of seed from the GM cotton lines, and spread and persistence of GM cotton plants in





adverse outcome



Risk evaluation

use of alternative herbicides.

volunteers are not problematic.

• Animal trampling and grazing are known to limit spread and persistence of cotton plants in areas where stock is fed cotton seed or where stock grazes after being fed cotton seed.

areas north of latitude 22º South.

Event 6 Dispersal of GM seed via flooding north of latitude 22º South

Marginal • Protection from

lepidopteran insect herbivory may provide some selective advantage in northern Australia.


Unlikely • Similar GM cotton lines

containing the same introduced genes have been commercially grown and transported in southern Australia, and their seed used as stockfeed in northern Australia, since 2000 and have not become problematic weeds.

• Although habitats close to waterways may be favourable for cotton establishment, expression of the introduced genes is not expected to alter susceptibility to the environmental factors (eg plant competition, fire or herbivory by non-lepidopteran insects) that will limit the establishment and persistence of cotton plants in these or other areas where seed may be dispersed to during flooding.


Event 7 Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance

Minor • Although transfer of

the cp4 epsps gene construct to similar commercially released GM cotton plants could result in plants containing three copies of the cp4 epsps gene, this would not alter the limited effectiveness of glyphosate in controlling cotton plants beyond the seedling stage.

• The expression of the cp4 epsps gene construct is not expected to alter

Highly unlikely • Cotton is primarily

in-breeding and gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies.

• If transfer of the cp4 epsps gene construct to other cotton plants occurred, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants.






adverse outcome



Risk evaluation

susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

Event 8 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory

Minor • Crossing of

Roundup Ready Flex®/Bollgard II® plants with similar commercially released GM cotton plants could result in plants containing three copies of the cp4 epsps gene (assessed in event 7).

• The expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

Highly unlikely • Cotton is primarily

in-breeding and gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies.

• If transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to other cotton plants occurred, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants.


Section 3 Risk management A risk management plan builds upon the risk assessment to consider whether any action is required to mitigate the identified risks, and what can be done to protect the health and safety of people and the environment.

The risk assessment considered eight events that might lead to a risk to the environment. The risk estimates for the adverse outcomes associated with six events are negligible (ie insubstantial with no present need to invoke actions for their mitigation). The risk estimates for the remaining two events are low (ie minimal but may invoke actions for mitigation beyond normal practices).

3.1 Risk treatment measures for identified risks The risks of the following two events that may lead to the GM cotton lines exhibiting greater weediness than non-GM cotton or other GM cotton lines previously approved for commercial release were estimated as low:

• dispersal of GM seed during transport or storage north of latitude 22º South



• dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South.

Therefore, risk treatment measures have been imposed to minimise dissemination of GM cotton seed, and spread and persistence of GM cotton plants, in areas north of latitude 22º South.

The following licence conditions have been imposed:

• transport of GM cotton seed to areas north of latitude 22º South must only be conducted in covered vehicles

• requirement to label the consignments of GM seed with contact phone numbers to call in case of spillage or misdirection

• the licence holder is required to inform people who transport the GM seed to areas north of latitude 22º South of the conditions associated with shipments

• north of latitude 22º South, GM cotton seed is only allowed to be fed to stock inside stockyards, feedlots or dairies

• the licence holder is required to provide information about cotton volunteers and their control to end-users of cotton seed north of latitude 22º South.

3.2 Other regulatory considerations In July 2003, the Australian Pesticides and Veterinary Medicines Authority (APVMA) registered the use of the insecticidal proteins produced by the cry1Ac and cry2Ab genes in GM Bollgard II® cotton as insecticidal products. This registration also covers the Roundup Ready Flex®/Bollgard II® GM cotton line, including a requirement to implement the insect resistance management plan developed for Bollgard II® cotton by the Transgenic and Insect Management Strategy Committee to maintain the efficacy of the product.

Accordingly, when the possibility of insects developing resistance to the Cry1Ac and Cry2Ab proteins as a result of this proposed release was considered in the risk assessment (see Section 2.10.1), no chance of harm to the environment was identified. Hence no conditions have been imposed in relation to management of insecticide resistance.

The use of Roundup Ready® Herbicide on the GM cotton lines proposed for release is also subject to regulation by the APVMA. Roundup Ready® Herbicide is currently registered for the use on Roundup Ready® cotton up to the four-leaf stage of growth. Monsanto requires APVMA approval of its application to vary the registration to allow the herbicide to be applied to Roundup Ready Flex® lines (including application after the four-leaf stage) to undertake the commercial release of these GM cotton lines. As the APVMA generally imposes conditions on the use pattern of herbicides, there will be restrictions on the number of applications that can be made and on the spraying window (ie up to what stage of crop growth herbicide applications can be made) eg in order to limit resistance development and comply with residue limits.

Accordingly, when the possibility of the use of glyphosate on the GM cotton lines resulting in changes in the weed spectrum (Section 2.9.2) or the development of herbicide resistant weeds (Section 2.10.3) as a result of this release was considered, no chance of harm was identified and no additional conditions have been imposed.

Food Standards Australia and New Zealand (FSANZ) is responsible for food safety standards. FSANZ previously approved food (oil and linters) derived from Roundup Ready® cotton and Bollgard II® cotton. Roundup Ready Flex® cotton has recently received similar approval.




These assessments have been conducted in parallel and involved mandated consultation between the agencies. This ensures the exchange of information, sharing of expertise and coordinated decision making.

Section 4 Conclusions of the RARMP The risk assessment concludes that this commercial release of herbicide tolerant and herbicide tolerant/insect resistant GM cotton lines poses low to negligible risks to the health and safety of people and the environment as a result of gene technology.

The risk management plan concludes that the negligible risks do not require specific risk treatment measures. However, licence conditions have been imposed to treat the low risks to the environment which relate to the use of GM cotton seed as stockfeed in northern Australia. Therefore, conditions have been imposed to minimise spread and persistence of the GMOs in areas north of latitude 22º South.


Chapter 1 Risk assessment context Section 1 Background 1. This Chapter describes the parameters within which risks that may be posed to the health and safety of people and the environment by the proposed release are assessed, based on scientific evidence. These include the scope and boundaries for the evaluation process required by the gene technology legislation2, details of the intended dealings, the GMO(s) and parent organism(s), previous approvals and releases of the same or similar GMOs in Australia or overseas, environmental considerations and relevant agricultural practices. The parameters for the risk assessment context are summarised in Figure 1.1.

Figure 1.1 Components of the risk context considered during the preparation of the Risk Assessment

RISK ASSESSMENT CONTEXT

GMO The genetic modification (genotype)

The modified traits (phenotype)

DEALINGS PARENT ORGANISM Activities involving the GMO

RECEIVING ENVIRONMENT Agronomic practices

Environmental conditions Previous releases

2. Sections 49 to 51 of the Gene Technology Act 2000 (the Act) outline the matters which the Regulator must take into account, and who she must consult with, in preparing the RARMPs that form the basis of her decision on licence applications.

3. For this application, establishing the risk assessment context includes consideration of:

• comparisons with the parent organism and other commercially released and widely grown GM cotton lines

• the nature and effect of the genetic modification

• relevant agricultural practices

• conditions proposed for transport of cotton seed for stockfeed

• the proposed size, duration and regions requested by the applicant

• the environmental conditions in the regions where the release would occur

• previous approvals for release of these GMOs in Australia and overseas

• any previous releases of these or other GMOs relevant to this application.

4. Initial consideration of the application under section 49 of the Act determined that public consultation was not required for the preparation of the consultation version of the

Chapter 1 Risk context (February 2006) 11

2 The legislative requirements and the approach taken in assessing licence applications are outlined in more detail at <http://www.ogtr.gov.au/ir/process.htm> and in the Risk Analysis Framework (OGTR 2005) <http://www.ogtr.gov.au/pdf/raffinal2.2pdf>.



RARMP.Even though public comment was not sought on the preparation of the consultation version of the RARMP, three submissions from the public were received (summarised in Appendix C).

5. In accordance with section 50 of the Act, the Gene Technology Technical Advisory Committee (GTTAC), State and Territory governments, Australian Government agencies, the Minister for Environment and Heritage and local councils where the release may take place3 were consulted on matters relevant to the preparation of the RARMP. This advice, and where it was taken into account in the RARMP, is summarised in Appendix B.

6. In accordance with section 52 of the Act, the Regulator notified the public that a RARMP had been prepared and invited written submissions. Advice on the RARMP was also sought from the same expert groups and agencies as mentioned above. The issues raised and how they were addressed in the RARMP are summarised in Appendices D and E, respectively.

Section 2 The GMOs and proposed dealings 2.1 The proposed dealings 7. Monsanto proposes to release herbicide tolerant (Roundup Ready Flex® MON 88913, abbreviated here as Roundup Ready Flex®) and herbicide tolerant/insect resistant (Roundup Ready Flex®/Bollgard II®) GM cotton into the environment.

8. Monsanto is seeking approval for unrestricted commercial scale planting of the two GM cotton lines south of latitude 22° South. The company anticipates a phased introduction commencing with up to 20,000 hectares in spring 2006 in the current cotton growing areas of New South Wales (NSW) and southern Queensland (QLD). The area is expected to increase in subsequent years and may include plantings in other areas south of latitude 22° South that are suitable for growing cotton. The GM cotton lines may also be planted on a small scale for evaluation, demonstration, education and research purposes, including in areas where cotton is not currently grown.

9. No specific containment measures have been proposed for the plantings in areas south of latitude 22° South and Monsanto intends that the GM cotton plants and their products would be used in the same manner as non-GM or other commercially released GM cotton. Hence the dealings would include:

• sale of seed for commercial planting in areas south of latitude 22° South

• conventional crossing with elite non-GM cotton varieties suitable for use under Australian conditions

• use of oil and linters from the GM cotton in human food (subject to approval by Food Standards Australia and New Zealand)

• use of cotton seed from the release as stockfeed anywhere in Australia

• sale of lint

• export of seed

• unrestricted transportation south of latitude 22° South (including transport of seed to release areas for planting and transport of seed cotton after harvest to cotton gins for ginning).

10. The waste produced from growing and harvesting of the GM cotton plants will be treated in the same way as waste from non-GM cotton without segregation.

3 All local councils south of latitude 22º South were consulted.



11. Although cotton seed is intended to be used for stockfeed anywhere in Australia, Monsanto also proposes restricted transportation of ginned cotton seed from the release to areas north of latitude 22° South in accordance with OGTR guidelines for transportation of GMOs.

12. The cotton lint (long cellulose fibres) removed from seed cotton during ginning are used to produce cotton fabrics for clothing, upholstery, towels and other household products. Processed lint does not contain protein or genetic material.

13. De-linted cotton seed is processed into four major products: oil, meal, hulls and linters (a type of short cellulose fibre) (Cherry & Leffler 1984). Whole cotton seed, meal and hulls are used in stockfeed. The oil is used in a variety of food products (including frying oil, salad dressing and margarine) and the linters are used as a cellulose base for several consumer food and hygiene products. Protein or genetic material is not detectable in processed cotton seed oil and linters (Sims et al. 1996; USDA 2004).

14. Food Standards Australia New Zealand (FSANZ) has approved food (oil and linters) derived from Roundup Ready Flex® cotton. Roundup Ready® cotton and Bollgard II® cotton have previously received such approvals.

15. Monsanto has applied to the Australian Pesticides and Veterinary Medicines Authority (APVMA) to extend the registration of Roundup Ready® Herbicide to allow the herbicide to be applied to Roundup Ready Flex® lines (including application after the four-leaf stage).

16. Monsanto anticipates that Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton will gradually replace its current GM cotton lines, Roundup Ready® and Bollgard II®/Roundup Ready® cotton. The length of the phase out period will depend on a number of factors such as availability of seed and grower acceptance.

2.2 The parent organism 17. The parent organism is cultivated cotton (Gossypium hirsutum), which is exotic to Australia and is grown as an agricultural crop in NSW, southern and central QLD and on a trial basis under limited and controlled conditions in WA, the NT and northern QLD. More detailed information on cotton can be found in the document, The Biology and Ecology of Cotton (Gossypium hirsutum) in Australia (OGTR 2002b), which was produced to inform the risk assessment process for licence applications involving GM cotton plants. This document is available at <http://www.ogtr.gov.au> under ‘Publications & Forms’.

2.3 The GMOs 18. Two GM cotton lines are proposed for release: Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II®. Both GMOs contain two copies of the cp4 epsps gene derived from Agrobacterium sp. strain CP4, a common soil bacterium. This gene encodes the enzyme CP4 EPSPS, which confers tolerance to glyphosate, the active ingredient of Roundup Ready® Herbicide.

19. EPSPS enzymes are critical for the synthesis of aromatic amino acids (essential building blocks for proteins). The EPSPS enzyme naturally produced in plants is inhibited by glyphosate, whereas the CP4 EPSPS enzyme produced by the bacterial cp4 epsps gene is insensitive to the effect of glyphosate (for more details see Section 2.4.1 of this Chapter).

20. Roundup Ready® cotton, which was approved for commercial release south of latitude 22º South in 2000 and is now widely grown, contains only one copy of the cp4 epsps gene and has little tolerance to glyphosate in reproductive tissues. This means that glyphosate can currently only be applied up to the four-leaf stage of growth (ie prior to flower formation, approximately 3 to 5 weeks after planting) to control weeds. After this stage, application of the herbicide can lead to yield loss. As Roundup Ready Flex® cotton has increased and



prolonged expression of the cp4 epsps gene, it is tolerant to glyphosate throughout the growing season (approximately 24 to 28 weeks). Hence, the window in which glyphosate can be applied for weed control is longer, giving growers increased flexibility in timing herbicide applications for integrated weed management.

21. Roundup Ready Flex®/Bollgard II® cotton was produced by conventional crossing of Roundup Ready Flex® cotton with Bollgard II® cotton (approved for commercial release south of latitude 22º South under DIR 012/2002). Roundup Ready Flex®/Bollgard II® cotton plants contain all the genes introduced into each of the parental GMOs. This means that the plants contain, in addition to the herbicide tolerance genes, two insecticidal genes (cry1Ac and cry2Ab) derived from the common soil bacterium Bacillus thuringiensis (Bt). The insecticidal genes encode proteins that are specifically toxic to lepidopteran caterpillars, including the two key Helicoverpa pests of cotton.

22. Roundup Ready Flex®/Bollgard II® cotton plants also contain genes derived from the common gut bacterium Escherichia coli, as a result of the crossing with Bollgard II®, conferring resistance to the antibiotics kanamycin and neomycin (nptII gene), and streptomycin and spectinomycin (aad gene). The aad gene is not expressed in the plants because the bacterial regulatory sequence that controls its expression is not active in plants. This gene was used in the laboratory prior to the production of the genetically modified plants to select for bacteria containing the modified DNA. No antibiotic resistance marker genes are present in Roundup Ready Flex® cotton.

23. A reporter gene (uidA) from the bacterium E. coli is also present from the Bollgard II® parent cotton plants. This gene encodes the enzyme β-glucuronidase (GUS) that enables visual identification of plant tissues in which it is being expressed.

24. Short regulatory sequences (promoters and terminators) that control expression of the introduced genes are also present in the GM cotton lines. These sequences are derived from the cauliflower mosaic virus, figwort mosaic virus, Agrobacterium tumefaciens and from soybean and two other plant species. Although the first three of these organisms are plant pathogens, the regulatory sequences comprise only a small part of their total genome and are not in themselves capable of causing disease.

25. Some details of the DNA sequence of the Roundup Ready Flex® gene construct have been declared Confidential Commercial Information (CCI) under section 185 of the Act. However, substantial information is in the public domain (refer Table 1.1), and the remainder was made available to the prescribed expert groups and agencies that were consulted in the preparation of the Risk Assessment and Risk Management Plan (RARMP). The information was also considered during the preparation of the RARMP.

2.4 The introduced genes and their products 2.4.1 The herbicide tolerance gene (cp4 epsps)

26. The cp4 epsps gene, which confers tolerance to glyphosate (N-phosphonomethyl glycine), the active ingredient of Roundup Ready® Herbicide, was isolated from the Agrobacterium species strain CP4. This cp4 epsps gene encodes a 47.6 kDa EPSPS protein consisting of a single polypeptide of 455 amino acids (Padgette et al. 1996).

27. In plants, the native epsps (5-enolpyruvylshikimate-3-phosphate synthase) gene encodes an enzyme (EPSPS) critical for the biosynthesis of aromatic amino acids (tryptophan, tyrosine and phenylalanine), which are essential building blocks for cellular proteins. During this biosynthetic process the EPSPS enzyme catalyses the addition of the enolpyruvyl moiety of phosphoenolpyruvate to shikimate-3-phosphate. EPSPS performs this function in plants, bacteria, algae and fungi but is absent from mammals, which are not able to synthesise these aromatic amino acids (Bentley 1990; Padgette et al. 1993).



28. Glyphosate herbicide functions by inhibiting the activity of the naturally occurring EPSPS enzyme in plants, thus blocking the biosynthesis of aromatic amino acids and eventually leading to cell death (Steinrucken & Amrhein 1980). The cp4 epsps gene from Agrobacterium is naturally insensitive to the effects of glyphosate (Padgette et al. 1993), as are a number of other microbial EPSPS enzymes (Schulz et al. 1985; Eschenburg et al. 2002), being still able to function normally in the biosynthesis of aromatic amino acids. Consequently, in GM plant cells expressing the Agrobacterium cp4 epsps gene, biosynthesis of aromatic amino acids is not blocked in the presence of glyphosate and the plants are not harmed by application of glyphosate.

29. The coding sequence of the cp4 epsps gene introduced into the GM cotton plants has been modified to achieve optimal expression in plants. Although the gene sequence has been altered, there is only one amino acid difference between the enzyme produced in Roundup Ready Flex® cotton lines and the native Agrobacterium enzyme (information provided by the applicant; (Padgette et al. 1993)), and the protein activity is not altered.

30. In plants, the EPSPS enzyme and the site of aromatic amino acid synthesis are located in the chloroplast. Thus, the cp4 epsps gene in the GM cotton lines was linked to a chloroplast transit peptide (CTP) coding region, ctp2 from the epsps gene of the plant Arabidopsis thaliana (Klee et al. 1987), to provide transport to the cotton chloroplast. The ctp2 sequence present in Roundup Ready Flex® cotton is the same as that used in the development of Roundup Ready® cotton. The CTP targets the CP4 EPSPS enzyme to the chloroplast. In plants, EPSPS is synthesised as a preprotein (ie containing the CTP) by free cytoplasmic ribosomes. The precursor is transported into the chloroplast stroma and proteolytically processed to yield the mature enzyme (della-Cioppa et al. 1986). Once cleaved from the mature protein, chloroplast transit peptides are rapidly degraded (Bartlett et al. 1982; della-Cioppa et al. 1986).

31. Two copies of the cp4 epsps gene are present in the GM cotton lines and each copy is under the control of a different promoter. A promoter is a region of DNA linked to a gene that determines whether a gene is expressed, to what extent and in which plant tissues. Details of the genetic elements of the introduced two-gene construct are presented in Table 1.1. Table 1.1 Arrangement of the genetic elements in the cp4 epsps gene construct introduced into

Roundup Ready Flex® cotton (starting from the right border region)

Genetic element Function Derived from Reference P-FMV/TSF1 chimeric promoter consisting of: enhancer sequence from the FMV

35S promoter and Figwort mosaic virus (Richins et al. 1987)

promoter of the elongation factor EF-1 alpha gene

Arabidopsis thaliana (Axelos et al. 1989)

L-TSF1 and I-TSF1 leader sequence (exon 1) and intron sequence from the elongation factor EF-1 alpha gene

A. thaliana (Axelos et al. 1989)

ctp2 targeting sequence

chloroplast transit peptide sequence of the plant epsps gene, directing the CP4 EPSPS protein into the chloroplast

A. thaliana (Klee et al. 1987)

cp4 epsps coding sequence for the CP4 EPSPS protein

Agrobacterium sp. strain CP4

(Padgette et al. 1996; Barry et al. 1997)

E9 termination region

3’ nontranslated region of the pea ribulose-1,5-bisphosphate carboxylase small subunit E9 gene

Pisum sativum (pea) (Coruzzi et al. 1984)

P-35S/ACT8 chimeric promoter consisting of: enhancer sequence from the

CaMV 35S promoter and Cauliflower mosaic virus (Kay et al. 1987)

promoter of the act8 actin gene A. thaliana (An et al. 1996)



L-ACT8 and I-ACT8

leader sequence and intron and flanking exon sequence from the act8 actin gene

A. thaliana (An et al. 1996)

ctp2 targeting sequence

chloroplast transit peptide sequence of the plant epsps gene, directing the CP4 EPSPS protein into the chloroplast

A. thaliana (Klee et al. 1987)

cp4 epsps coding sequence for the CP4 EPSPS protein

Agrobacterium sp. strain CP4

(Padgette et al. 1996; Barry et al. 1997)

E9 termination region

3’ nontranslated region of the pea ribulose-1,5-bisphosphate carboxylase small subunit E9 gene

Pisum sativum (pea) (Coruzzi et al. 1984)

32. One copy of the cp4 epsps gene is controlled by the chimeric P-FMV/TSF1 promoter consisting of enhancer sequences from the figwort mosaic virus (FMV) 35S promoter (Richins et al. 1987) and the elongation factor EF-1 alpha promoter from the plant Arabidopsis thaliana (EF-1 alpha A1 gene; (Axelos et al. 1989)). The other copy of the cp4 epsps gene is controlled by the chimeric P-35S/ACT8 promoter consisting of enhancer sequences of the cauliflower mosaic virus (CaMV) 35S promoter (Kay et al. 1987) and the act8 actin promoter from A. thaliana (An et al. 1996). These promoters direct the cp4 epsps genes to be expressed in all plant tissues throughout plant growth.

33. The GM cotton lines also contain additional non-coding sequences from other plant species for improved gene expression. These include the non-translated leader (exon 1) and intron sequences (L-TSF1 and I-TSF1, respectively) from the A. thaliana EF-1 alpha A1 gene (Axelos et al. 1989), and the non-translated leader and intron/exon sequences (L-ACT8 and I-ACT8, respectively) from the act8 actin gene of A. thaliana (An et al. 1996).

34. Also required for gene expression in plants is an mRNA termination region, including a polyadenylation signal. The mRNA termination region for both the cp4 epsps genes in the GM cotton lines is the 3’non-translated region derived from the pea ribulose-1,5-bisphosphate carboxylase small subunit E9 gene (Coruzzi et al. 1984).

35. Each cp4 epsps gene in the Roundup Ready Flex® cotton encodes the same protein as is expressed in Roundup Ready® cotton, which was approved for commercial release in Australia (DIR 023/2002). Expression of the CP4 EPSPS protein in the GM cotton lines

36. The two different chimeric promoters, containing the viral enhancer sequences (from the CaMV or FMV 35S promoters), are expected to cause expression of the CP4 EPSPS protein from the two copies of the cp4 epsps gene throughout most or all parts of the Roundup Ready Flex® cotton plant. Roundup Ready Flex® cotton has increased and prolonged expression of the cp4 epsps gene, compared to commercially released Roundup Ready® cotton, including expression in the reproductive parts of the plant.

37. Expression of the CP4 EPSPS protein in Roundup Ready® cotton is significantly lower in reproductive tissues such as stigmas, anthers and floral buds than in vegetative tissue such as leaves (Pline et al. 2002a). The reproductive structures of cotton plants are very sensitive to the effects of glyphosate (Pline et al. 2002a; Pline et al. 2002b) and Roundup Ready® cotton crops sprayed with glyphosate beyond the four-leaf stage of growth exhibit reduced pollination and increased boll abortion (Monsanto Australia Limited 2001). Therefore, application of glyphosate over the top of these GM cotton plants after the four-leaf stage leads to yield loss (Charles 2002b). Tolerance to glyphosate in the reproductive parts of Roundup Ready Flex® cotton plants is intended to enable glyphosate application over the top of the cotton crop throughout the growing season.

38. Field trials conducted in the USA have shown that reproductive parts of Roundup Ready Flex® cotton plants have greater tolerance to glyphosate as compared to the same plant parts



in Roundup Ready® cotton (determined by measurement of pollen viability and the number of pollen grains attached to the stigmatic lobe after treatment with glyphosate).

39. Data on the level of expression of the CP4 EPSPS protein in different Roundup Ready Flex® cotton tissues have been collected from field trials conducted in the USA across four locations during 2002 (Table 1.2) (Bookout et al. 2003). Samples of young leaves collected over the growing season, roots, seeds and pollen were analysed using an enzyme-linked immunosorbent assay (ELISA). All of the values presented in Table 1.2 are well above the level of quantitation, which means that they can be reliably and reproducibly quantitated. Because of the limited quantities of pollen available, protein levels by dry weight could not be determined for pollen. The CP4 EPSPS protein levels detected in all tissues from negative segregants (progeny of backcrosses between Roundup Ready Flex® cotton and non-GM cotton that did not inherit the introduced genes) were less than the assay limits of quantitation. Table 1.2 CP4 EPSPS protein levels in Roundup Ready Flex® tissues from US field trials in 2002

(± standard deviation)

CP4 EPSPS (µg per g fresh weight)

CP4 EPSPS (µg per g dry weight)

Tissue type

Mean Range Mean Range

LOQ/LOD (µg per g fresh

weight) Leafa A 170 ± 64 64 - 260 970 ± 460 270 - 1700 0.23/0.069 Leaf B 270 ± 99 77 - 410 1400 ± 540 480 - 2600 0.23/0.069 Leaf C 170 ± 44 63 - 260 960 ± 210 290 - 1000 0.23/0.069 Leaf D 160 ± 61 66 - 260 630 ± 230 290 - 1100 0.23/0.069 Root 31 ± 11 19 - 64 99 ± 40 57 - 200 0.23/0.073 Seed 310 ± 110 67 - 550 340 ± 120 72 - 580 2.7/1.7 Pollen 4 ± 0.22 3.8 – 4.3 ND ND 0.23/0.11

a represent the newest fully expanded leaves (leaf A to D) collected at different time points throughout the growing season between the seedling stage and crop maturity. LOQ: limit of quantitation LOD: limit of detection ND: not determined

40. Levels of CP4 EPSPS protein in Roundup Ready Flex® cotton tissues collected from field trials conducted under DIR 035/2003 in Australia during 2003–04 were also determined (Table 1.3) (Pineda et al. 2005). All of the values are well above the levels of quantitation.



Table 1.3 CP4 EPSPS protein levels in Roundup Ready Flex® tissues from Australian field trials in 2003–04 (± standard deviation)

CP4 EPSPS (µg per g fresh weight)

CP4 EPSPS (µg per g dry weight)

Tissue type

Mean Range Mean Range

LOQ/LOD (µg per g fresh

weight) Leaf Aa 210 ± 47 120 - 310 1400 ± 370 730 - 2200 0.23/0.069 Leaf Bb 250 ± 44 140 - 330 1500 ± 360 760 - 2000 0.23/0.069 Leaf Cc 290 ± 60 200 - 410 1600 ± 400 1100 - 2400 0.23/0.069 Leaf Dd 270 ± 100 100 - 360 1100 ± 430 430 - 1500 0.23/0.069 Seed 280 ± 30 230 - 340 310 ± 30 260 - 380 2.7/1.7

a Leaf A: newest fully expanded leaf collected at the 2-3 node stage b Leaf B: newest fully expanded leaf collected at the 4-6 node stage c Leaf C: newest fully expanded leaf collected at the 8-10 node stage d Leaf D: newest fully expanded leaf collected at approximately cutout (final stage of plant

growth)

41. The CP4 EPSPS protein levels detected in Roundup Ready Flex® cotton are higher than those detected in Roundup Ready® cotton tissues (42 - 53 and 60 - 299 µg per g fresh weight in leaves and seeds, respectively; detailed information is provided in the RARMP for DIR 023/2002, available from <http://www.ogtr.gov.au>).

42. Expression of the CP4 EPSPS protein in Roundup Ready Flex®/Bollgard II® was determined indirectly in greenhouse and laboratory tests by conducting a glyphosate tolerance bioassay (Burns et al. 2004). In this bioassay, glyphosate was applied at 1.5- and 21-times the rate used for weed control in cotton. No significant difference between Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton in the tolerance to glyphosate was observed. This indicates that CP4 EPSPS protein expression levels in Roundup Ready Flex®/ Bollgard II® cotton are similar to those in Roundup Ready Flex® cotton. 2.4.1 The insect resistance genes (cry1Ac and cry2Ab)

43. Roundup Ready Flex®/Bollgard II® cotton was produced by conventional crossing of Roundup Ready Flex® cotton with Bollgard II® cotton (approved for commercial release south of latitude 22º South under DIR 012/2002) and as a result contains all the genes introduced into each of the parental GMOs.

44. The cry1Ac and cry2Ab genes of Bollgard II® cotton are derived from Bacillus thuringiensis (Bt), a common soil bacterium (Martin & Travers 1989). B. thuringiensis is readily isolated in association with insects (particularly in dry indoor environments such as warehouses), from leaf surfaces (particularly of deciduous and coniferous trees) or from soil samples (Meadows 1993). One survey found B. thuringiensis in 70% of soil samples collected from 30 countries (Martin & Travers 1989). B. thuringiensis produces a range of insecticidal proteins, each with specific toxicity to certain groups of insects.

45. The crystalline (Cry) proteins (also called Bt proteins or Bt toxins) are a diverse family of insecticidal proteins produced by various subspecies of B. thuringiensis. The cry1Ac and cry2Ab genes are derived from B. thuringiensis variety kurstaki (Btk) and encode Bt toxins that are highly specific to a subset of lepidopteran insects (moths and butterflies), including Helicoverpa armigera and H. punctigera, major herbivorous pests of cotton in Australia (Widner & Whiteley 1990; Dankocsik et al. 1990; Macintosh et al. 1990).

46. The Btk HD1 strain is used in commercial Bt sprays in Australia (see APVMA database <http://www.apvma.gov.au/pubcris/subpage_pubcris.shtml>) and elsewhere.

47. The toxic effect of Bt proteins requires alkaline conditions (as provided in the larval insect gut) to dissolve the crystals, partial digestion by specific proteases to release the active core

http://www.ogtr.gov.au/



toxin and binding to specific receptors found on the insect midgut epithelium surface. Binding leads to formation of pores in the cell membrane which leads to leakage of intracellular contents into the gut lumen and water into the cell and eventually to cell death, gut paralysis and starvation. It is these steps that provide the high degree of target specificity of each Bt toxin (Hofmann et al. 1988; Van Rie et al. 1989; English & Slatin 1992; Knowles & Dow 1993).

48. More information about the toxicity of the Cry1Ac and Cry2Ab proteins for a variety of invertebrate species is provided in Chapter 3.

49. Expression of the cry1Ac and cry2Ab genes in Bollgard II® cotton is controlled by the enhanced CaMV 35S promoter (Kay et al. 1987; Odell et al. 1985). The mRNA termination region for the cry1Ac gene in Bollgard II® cotton is derived from the alpha subunit of the beta-conglycinin gene of soybean and for the cry2Ab gene from the nopaline synthase (nos) gene of A. tumefaciens.

50. More detailed information about the cry1Ac and cry2Ab genes present in Bollgard II® cotton, and the Bt toxins, can be found in the Risk Assessment and Risk Management Plans for DIR 012/2002 and DIR 022/2002, available at <http://www.ogtr.gov.au>. Expression of the Cry1Ac and Cry2Ab proteins in the GM cotton

51. The promoters controlling expression of the cry1Ac and cry2Ab genes in Bollgard II® cotton result in expression of the Cry proteins throughout the growing season and in most of the tissues. Conventional crossing with Roundup Ready Flex®/Bollgard II® cotton does not alter this gene expression.

52. Expression levels of the Cry1Ac and Cry2Ab proteins in Bollgard II® cotton have been extensively studied. Detailed information on the expression levels of these proteins in leaves, seeds and pollen of Bollgard II® cotton is presented and discussed in the RARMP for DIR 012/2002, available from <http://www.ogtr.gov.au>. Some of the data are summarised in Table 1.4 below. Table 1.4 Cry1Ac and Cry2Ab protein levels (µg per g fresh weight of plant tissue) in Bollgard II®

cotton tissues (± standard deviation)

Tissue type Year tested Cry1Ac Cry2Ab Leaf 1998

1999 2.75 ± 1.32 2.07 ± 0.61

23.8 ± 6.3 6.4 ± 1.4

Seed 1998 1999

3.35 ± 0.63 2.60 ± 0.66

43.2 ± 5.7 57.4 ± 13.1

Whole plant 1998 1999

0.17 ± 0.08 0.08 ± 0.01

8.8 ± 1.2 20.8 ± 1.8

Pollen 1998 1999

0.02 ± 0.01 0.05 ± 0.07

<0.25 0.32

53. Expression of the Cry proteins in Roundup Ready Flex®/Bollgard II® was determined indirectly in greenhouse and laboratory tests by conducting an insect resistance bioassay (Burns et al. 2004). In this bioassay, second and third instar larvae of tobacco budworm were fed tissues from Bollgard II® and Roundup Ready Flex®/Bollgard II® cotton. Based on the measured larvae mortality and average weight of surviving larvae, there is no significant difference between the efficacy of the two GM cotton lines. This indicates that expression levels of the Cry1Ac and Cry2Ab proteins in Roundup Ready Flex®/Bollgard II® cotton are similar to those in Bollgard II® cotton.



2.4.2 The reporter gene (uidA)

54. Roundup Ready Flex®/Bollgard II® cotton plants contain the uidA reporter gene. The uidA gene is derived from the common gut bacterium Escherichia coli (E. coli) and encodes the enzyme β-glucuronidase (GUS) (Jefferson et al. 1986).

55. The uidA gene is the most widely used reporter gene in GM plants (Miki & McHugh 2004). Reporter genes encode enzymes that are easily assayed and are therefore used to ‘report’ on the activity of a promoter to which the reporter gene is linked. They can also be linked to a gene to report on the cellular location of the encoded protein, or used as a simple biochemical tag to identify GM tissues.

56. The GUS protein is a monomer with a molecular mass of 68 kDa, and the GUS enzyme is active in the form of a tetramer. GUS catalyses the hydrolysis of β-glucuronides and, less efficiently, some β-galacturonides. A large variety of β-glucuronides exist in nature and they have been described as the detoxified excretion forms of xenobiotics (foreign substances, eg drugs) and endogenous compounds (eg steroids) in vertebrates (Jefferson & Wilson 1991). E. coli lives in the digestive tract of vertebrates, including humans (Jefferson et al. 1986), and the GUS enzyme enables it to metabolise β-glucuronides as a main source of carbon and energy.

57. GUS cleaves the chromogenic substrate X-gluc (5-bromo-4-chloro-3-indolyl β-D-glucuronide) to produce an insoluble blue colour (Jefferson et al. 1987). Endogenous GUS enzyme activity is found in many other bacterial species, and also in vertebrates and invertebrates, but there is very little background activity in non-GM plants (Jefferson et al. 1987; Gilissen et al. 1998). Therefore, the production of a blue colour in particular plant cells after staining with X-gluc indicates that these cells express GUS from the introduced uidA gene, and have been successfully genetically modified.

58. The uidA gene was used as a marker in the laboratory for selecting successfully modified Bollgard II® cotton cells after the genetic modification. Expression of this gene in Roundup Ready Flex®/Bollgard II® cotton plants is controlled by the CaMV 35S promoter (Kay et al. 1987; Odell et al. 1985). The mRNA termination region of the gene is the 3’non-translated region of the nos gene from A. tumefaciens (Rogers et al. 1985).

59. Further information on the uidA gene in Bollgard II® cotton is provided in the RARMP for DIR 012/2002 available at <http://www.ogtr.gov.au>. 2.4.2 The antibiotic resistance marker genes (nptII and aad)

60. Roundup Ready Flex® cotton does not contain any antibiotic resistance genes. However, Roundup Ready Flex®/Bollgard II® cotton does contain the nptII and aad antibiotic resistance marker genes from E. coli. These genes were used in the initial laboratory stages of development of Bollgard II® cotton plants, to enable selection of cells containing the desired genetic modification. Both genes occur in the natural environment and are in common use as selectable markers in the production of GM plants.

61. The nptII gene was isolated from the bacterial Tn5 transposon (from E. coli) (Beck et al. 1982). It encodes an enzyme, neomycin phosphotransferase type II (NPTII), which confers resistance to the aminoglycoside antibiotics kanamycin and neomycin. NPTII uses ATP to phosphorylate kanamycin and neomycin, thereby inactivating the antibiotic and preventing it from killing the NPTII-producing cell. The nptII gene functioned as a selectable marker during the laboratory stages of cotton plant cell selection following genetic modification, allowing modified cells to grow in the presence of the antibiotic while inhibiting the growth of non-modified cells.



62. The NPTII enzyme is widespread in the environment and in food chains, in naturally occurring kanamycin-resistant microorganisms found in soil and in mammalian digestive systems (Flavell et al. 1992).

63. Expression of the nptII gene in Roundup Ready Flex®/Bollgard II® cotton plants is controlled by the CaMV 35S promoter (Kay et al. 1987; Odell et al. 1985). The mRNA termination region of the gene is the 3’non-translated region of the nos gene from A. tumefaciens (Rogers et al. 1985).

64. Further information on the antibiotic resistance genes in Bollgard II® cotton is provided in the RARMP for DIR 012/2002 available at <http://www.ogtr.gov.au>.

2.5 Method of genetic modification 65. The Roundup Ready Flex® cotton was produced by Agrobacterium-mediated DNA transformation (Zambryski 1992) of a Coker variety of cotton (transformation event 88913). Coker cotton varieties are US cultivars that are widely used in producing GM cotton plants because they can be readily cultured and regenerated in the laboratory.

66. Agrobacterium tumefaciens is a common gram-negative soil bacterium that causes crown gall disease in a wide variety of plants. Plants can be genetically modified by the transfer of DNA (transfer-DNA or T-DNA, located between specific border sequences on a resident plasmid) from A. tumefaciens through the mediation of genes from the virulence region of Ti plasmids.

67. Disarmed Agrobacterium strains have been constructed specifically for plant transformation. The disarmed strains do not contain the genes responsible for the overproduction of auxin and cytokinin (iaaM, iaaH and ipt), which are required for tumour induction and rapid callus growth (Klee & Rogers 1989). Agrobacterium plasmid vectors used to transfer T-DNAs contain well characterised DNA segments required for their replication and selection in bacteria, and for transfer of T-DNA from Agrobacterium and its integration into the plant cell genome (Bevan 1984; Wang et al. 1984). Agrobacterium-mediated transformation has been widely used in Australia and overseas for introducing new genes into plants without causing any biosafety concerns or any adverse reactions.

68. In this instance, a typical, disarmed binary plasmid vector (PV-GHGT35) was used to introduce the gene construct containing two copies of the cp4 epsps gene (see Table 1.5) into cotton variety Coker using standard Agrobacterium transformation protocols. Following co-cultivation with A. tumefaciens containing the gene construct, cotton cells were cultured in the presence of glyphosate to select for those cells containing the introduced DNA (since the cp4 epsps gene confers tolerance to glyphosate). Subsequently, cotton plants containing the introduced gene constructs were regenerated from these cells. The resulting genetically modified plants were further screened for tolerance to glyphosate over several generations in the laboratory and under field conditions in the USA. Roundup Ready Flex® cotton, as well as crosses with commercial cotton varieties, were also assessed for their agronomic performance. Roundup Ready Flex® cotton is derived from a single genetic modification, or ‘transformation’, event (event 88913).

69. Bollgard II® cotton was produced by inserting the cry2Ab and uidA genes into the genomic DNA of INGARD® cotton variety DP50B (containing the cry1Ac and nptII genes). Genes were delivered into the cotton meristematic cells by microprojectile bombardment (transformation event 15985) (McCabe & Martinell 1993). This technique is a well-established method of plant transformation that uses compressed gas to 'shoot' tiny tungsten or gold particles coated with the genes to be inserted into plant cells. The introduced genes become incorporated into the genome of the bombarded plant cells. The uidA gene is used as a marker to identify plant tissue that contains the introduced gene of interest. INGARD®



cotton was itself produced by Agrobacterium-mediated transformation of a non-GM Coker cotton variety, introducing the cry1Ac, nptII and aad genes. Further information about these constructs can also be found in previous RARMPs (DIR 012/2002 for Bollgard II®, and DIRs 022/2002 and 023/2002 for INGARD®).

70. Roundup Ready Flex®/Bollgard II® cotton was produced by conventional crossing of GM Roundup Ready Flex® cotton with GM Bollgard II® cotton.

71. As part of the proposed commercial release, the Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® traits would be introduced into a range of cotton varieties suitable for various growing conditions in Australia by conventional crossing.

Section 3 The receiving environment 72. The receiving environment forms part of the context in which the risks associated with dealings involving the GMOs are assessed. This includes the size, duration and regions of the dealings, any relevant biotic/abiotic properties of the regions where the release would occur; intended agronomic practices, including those that may be altered in relation to normal practices; other relevant GMOs already released; and any particularly vulnerable or susceptible entities that may be specifically affected by the proposed release (OGTR 2005).

3.1 Size and duration of the proposed release 73. The size and duration of the proposed release are outlined in Section 2.1. The proposed release is to occur in the current cotton growing areas south of latitude 22º South. Future plantings may also occur in other areas south of latitude 22º South that are suitable for growing cotton. The GM cotton lines may also be planted on a small scale in different regions for evaluation, demonstration, education and research purposes.

3.2 Major cotton growing regions of Australia 74. Table 1.5 lists the major cotton growing regions in Australia. Detailed information about cotton production and the individual valleys where cotton was grown commercially in the 2004–05 season is available from Cotton Australia (Cotton Australia 2005b). A map showing the local government areas in which cotton was grown in 2001 is available at <http://www.ogtr.gov.au/pdf/public/cotmaplga.pdf>. Table 1.5 Major cotton growing regions as of 2003a

State Cotton growing region LGAs Towns

QLD Central Highlands Emerald, Peak Downs Emerald

QLD Dawson - Callide Banana Theodore, Biloela, Moura

QLD St George - Dirranbandi Balonne St George, Dirranbandi

QLD Darling Downs Wambo, Dalby, Jondaryan, Chinchilla, Pittsworth, Milmerran

Dalby, Chinchilla, Oakey, Pittsworth, Milmerran, Toowoomba

QLD/NSW Macintyre Valley Waggamba (QLD), Moree Plains (NSW)

Goondiwindi, Mungindi, Bogabilla

NSW Gwydir Valley Moree Plains, Walgett Moree, Collarenebri

NSW Upper Namoi Gunnedah Gunnedah, Boggabri, Curlewis

NSW Lower Namoi Narrabri, Warren Narrabri, Wee Waa, Walgett

NSW Macquarie Valley Narromine, Warren Narromine, Warren, Trangie, Dubbo

NSW Bourke Bourke Bourke

NSW Lachlan - Murrumbidgee Carrathool, Lachlan Hillston, Lake Cargellico, Griffith



a Source: modified from (Reeve et al. 2003)

3.3 Environmental conditions suitable for growing cotton 75. Climates with long, warm summers are typical for summer cotton growing regions in Australia. The areas where cotton can be grown in southern Australia are mainly limited by the amount of irrigation water available (for irrigated cotton) and the length of the summer season. Climatic data for some of these ares are given in Table 1.6.

76. Temperature is the dominant environmental factor affecting cotton development and yield (Constable & Shaw 1988; Australian Cotton Cooperative Research Centre 2002). Cotton is planted when the minimum soil temperature at 10 cm depth is 14°C for at least three successive days. Cotton seedlings may be killed by frost. A minimum of 180–200 frost-free days of uniformly high temperatures (averaging 21–22°C) is required (Duke 1983). Growth and development of cotton plants below 12°C is minimal and a long, hot growing season is crucial for achieving good yields (Constable & Shaw 1988).

77. Crop yields may be lower in southern growing regions as a result of the shorter summer season. The minimum day degrees (heat accumulation, calculated progressively during the season) required from planting of cotton to 60% boll opening is 2050 (information from the Australian Cotton CRC; available at <http://cotton.pi.csiro.au>). For example, cotton planted on 1 October near Warren (Macquarie Valley, NSW) could be expected to reach 60% boll opening by 31 March the following year.

78. Cotton can also be grown as dryland crop (Australian Cotton Cooperative Research Centre 2002). Dryland cotton production strongly depends on rainfall (at the right time during the growing season) and the water holding capacity of the soil. In most areas south of latitude 22º South, the variability of rainfall during the critical months (January to March) is high (see <http://www.bom.gov.au/climate/averages/variability.shtml>) and therefore, dryland cotton production may not always be viable. Table 1.6 Climatic data for some of the current cotton growing regions in Australia.

Emerald Post Office (central QLD)

Narrabri West Post Office (northern NSW)

Bourke Post Office (northern NSW)

Hillston Airport (southern NSW)

Average daily max/min temperature (summera)

34.2°C/20.3°C 32.3°C/17.3°C 35.6°C/20.3°C 32.4°C/17.6°C

Average daily max/min temperature (winterb)

23.3°C/7.8°C 18.9°C/4.5°C 18.9°C/5.5°C 15.8°C/4.6°C

Average monthly rainfall (summer)

84.4 mm 72.5 mm 38.8 mm 28.7 mm

Average monthly rainfall (winter)

27.8 mm 45.7 mm 23.6 mm 32.1 mm

a December, January, February b June, July, August

Source: <http://www.bom.gov.au>.

3.4 Presence of the introduced proteins in the receiving environment 79. CP4 EPSPS, Cry1Ac, GUS and NPTII proteins are widespread in the environment, through the presence of the bacteria from which they are derived. This forms part of the baseline data for assessing any risks from exposure to these proteins that may result from the proposed release of the GM cotton lines.

80. The CP4 EPSPS protein is produced naturally by the CP4 strain of the common soil bacterium Agrobacterium sp (Padgette et al. 1996). This bacterium can also be found on



plants and fresh plant produce. Similar EPSPS proteins are present in all plants, bacteria and fungi.

81. The Cry1Ac protein is naturally produced by the kurstaki variety of the common soil bacterium B. thuringiensis (Bt). The protein is also present in Btk (B. thuringiensis var kurstaki) microbial sprays which are used to protect crops from insect herbivory.

82. The Cry2Ab protein is not naturally expressed in soil bacteria or present in Btk sprays due to an ineffective promoter of the cry2Ab gene. The Cry2Ab protein expressed in the Roundup Ready Flex®/Bollgard II® cotton is 88% identical at the sequence level to the Cry2Aa protein (Widner & Whiteley 1989; Dankocsik et al. 1990). The Cry2Aa protein is naturally expressed in B. thuringiensis var kurstaki and present in Btk sprays (information supplied by the applicant for DIR 012/2002).

83. Related Cry proteins are also produced by other varieties of B. thuringiensis. Spores from a range of Bt varieties and their crystal (Cry) toxins are found widely in both the agricultural and natural environment, including in soil, on plant leaves, in grain stores and in dead insects (Meadows 1993).

84. The presence of Bt toxins in agricultural situations has increased over the past 30 years due to the commercial use of Btk microbial sprays to protect food crops from insect attack (ANZFA 1999). Residues of Btk proteins, including Cry1Ac and Cry2Aa (Widner & Whiteley 1989; Dankocsik et al. 1990), are present on a wide variety of foods, including fresh foods such as lettuce and tomato, with no reported toxic or allergic responses (ANZFA 1999).

85. The GUS and NPTII proteins are widespread in the environment since they are naturally produced by the common gut bacterium Escherichia coli. E. coli is widespread in human and animal digestive systems (Jefferson et al. 1986) as well as in the environment.

86. The GUS protein used in GM plants is 99.8 % identical to the E. coli GUS protein. GUS activity is also found in a wide range of other bacteria, including other microorganisms of the digestive tract and many soil bacteria (Gilissen et al. 1998).

87. GUS enzyme activity has been detected in numerous plant and animal species, including species used as raw food (Gilissen et al. 1998). It is very common in almost all tissues of vertebrates and is also present in invertebrates such as molluscs, nematodes and insects. GUS activity has been detected in over 50 different plant species (Hu et al. 1990). However, when endogenous GUS activity is found in plants, the activity is very low and its function is unknown (Gilissen et al. 1998).

88. Humans (and, by implication, other animals) continually ingest kanamycin-resistant microorganisms, some containing the NPTII enzyme. The diet, especially raw salad, is the major source: estimated conservatively, each human ingests 1.2 x 106 kanamycin-resistant microorganisms daily (Flavell et al. 1992). Large numbers of kanamycin- or neomycin-resistant bacteria already inhabit the human digestive system (Levy et al. 1998), with Flavell et al. (1992) estimating about 1012 per person. Kanamycin-resistant bacteria have been isolated from soil, river water and sewage (Smalla et al. 1993).

3.5 Presence of similar GM cotton in the receiving environment 89. A number of releases of GM cotton plants expressing the same proteins have previously been approved. These releases are summarised in Section 4. The introduced proteins present in Roundup Ready Flex® (CP4 EPSPS) and Roundup Ready Flex®/Bollgard II® cotton (CP4 EPSPS, Cry1Ac, Cry2Ab, GUS and NPTII) are the same as those present in the commercially released Roundup Ready®, Bollgard II® and Bollgard II®/Roundup Ready® cotton lines (approved for commercial release south of latitude 22º South under DIR 023/2002 and DIR 012/2002). The only difference is that the level of CP4 EPSPS protein expressed in



Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton is higher than in Roundup Ready® and Bollgard II®/Roundup Ready®.

90. The insect resistant Bollgard II® GM cotton and the glyphosate tolerant Roundup Ready®

GM cotton are already widely planted in the agricultural environment as a result of their commercial release in southern Australia. In the 2004–05 season, Bollgard II® (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 70% of commercially grown cotton. Roundup Ready® cotton (also including the stacked variety Bollgard II®/ Roundup Ready®) comprised about 50%. This forms part of the baseline data for estimating the risks that may result from the proposed release.

3.6 Agronomic practices for these GM cotton lines 91. Agronomic management of the GM cotton lines would differ from the management of non-GM cotton in that glyphosate herbicide could be applied over the top of the cotton crop to control weeds. In addition, fewer applications of pesticide sprays are expected to be used on Roundup Ready Flex®/Bollgard II® cotton since it is resistant to the major lepidopteran pests of cotton. These management practices are the same as those used for the commercially grown Roundup Ready® and Roundup Ready®/Bollgard II® cotton lines with the exception that glyphosate herbicide can be applied throughout the growing season (approximately 24 to 28 weeks) instead of being restricted to before the four-leaf stage of growth (ie prior to flower formation, approximately 3 to 5 weeks after planting). However, as the APVMA generally imposes conditions on the use pattern of herbicides, there will be restrictions on the number of applications that can be made and on the spraying window (ie up to what stage of crop growth herbicide applications can be made eg in order to limit residue levels). All other crop management practices, including application of water and fertilizer, are expected to be similar to those for non-GM cotton.

92. High levels of farm hygiene are commonly maintained on cotton farms (eg all equipment is cleaned on entry and exit to a field/farm to prevent the transfer of disease or the spread of weeds). Irrigation practices (Good Management Practice of cotton industry) used by cotton growers in Australia retain irrigation water run-off, as well as the first 15 mm of storm water run-off, on-farm to minimise the entry of pesticide residues into natural waterways. Transport of ginned cotton seed is conducted in covered vehicle to minimise loss of seed.

Section 4 Previous Australian and international approvals 4.1 Previous Australian approvals of the same or similar GMOs 4.1.1 Previous releases approved by GMAC or the Regulator

93. Two limited and controlled releases of Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton have been approved previously under DIR 035/2003 and DIR 055/2004. These releases were approved for NSW, QLD, NT and northern WA on 950 and 1815 hectares, respectively. The main part of the approved release areas is in the current cotton growing regions south of latitude 22º South. The field trials under DIR 035/2003 and DIR 055/2004 are being conducted during 2003–05 and 2005–06, respectively. Data relevant to the assessment of the current application were provided from these trials.

94. Roundup Ready® cotton, containing only one copy of the same cp4 epsps gene present in Roundup Ready Flex® cotton, has been approved for commercial release since 2000 (refer licence DIR 023/2002). Bollgard II® and Bollgard II®/Roundup Ready® cotton were approved for commercial release by the Regulator in 2002 (licence DIR 012/2002).

95. Both commercial releases were restricted to areas south of latitude 22º South because of concerns about the potential weediness of the GM cotton lines in the northern tropical areas. North of latitude 22° South, field trials with Bollgard II® cotton and Bollgard II®/Roundup



Ready® cotton are being conducted under limited and controlled conditions (refer licence DIR 012/2002).

96. Prior to obtaining approval for commercial release, numerous field trials with Roundup Ready®, Bollgard II® and Bollgard II®/Roundup Ready® cotton lines were conducted under the voluntary system overseen by the Genetic Manipulation Advisory Committee (GMAC), and four licences for limited and controlled releases of Bollgard II® and/or Bollgard II®/Roundup Ready® cotton have been issued by the Regulator (DIR licences) as listed below.

Roundup Ready® cotton - 23 planned releases (PR) authorised by GMAC:

• CSIRO Division of Plant Industry (PR-55, PR-55X, PR-55X2, PR-55X3, PR-55X5 and PR-55X6)

• Deltapine Australia Pty Ltd (PR-32, PR-52, PR-52X, PR-52X2, PR-52X3, PR-71, PR-71X, PR-71X2, PR-83, PR-83X, PR-83X3, PR-140, PR-140X and PR-143)

• Monsanto (PR-83X2 and PR-83X4)

• Cotton Seed Distributors Pty Ltd (PR 55-X4).

Bollgard II® and Bollgard II®/Roundup Ready® cotton – 14 planned releases authorised by GMAC and 4 limited and controlled releases approved by the Regulator:

• CSIRO Division of Plant Industry (PR-123, PR-123X, PR-123X2, PR-131, PR-131X, PR-131X2 and PR-131X3)

• Deltapine Australia Pty Ltd (PR-51X4, PR-112, PR-112X, PR-112X2, PR-118, PR-118X and PR-118X2)

• Cotton Seed Distributors (Bollgard II® and Bollgard II®/Roundup Ready® in QLD; DIR 005/2001)

• CSIRO (INGARD®, Bollgard II® and Bollgard II®/Roundup Ready® cotton in WA and NT; DIR 006/2001)

• Department of Agriculture (WA) (Bollgard II® cotton in WA; DIR 009/2002)

• Monsanto (Bollgard II® and Bollgard II®/Roundup Ready® cotton in northern WA, NT and northern QLD; DIR 012/2002).

4.1.2 Approvals by other Australian government agencies

97. The Regulator is responsible for assessing risks to the health and safety of people and the environment associated with the use of gene technology. Other government regulatory requirements would also have to be met in respect of release of Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton, including requirements of the Australian Pesticides and Veterinary Medicines Authority (APVMA) and Food Standard Australia New Zealand (FSANZ). This is discussed further in the Technical Summary and in Chapter 5.

98. FSANZ is responsible for human food safety assessment and food labelling, including GM food. FSANZ has approved food (oil and linters) derived from Roundup Ready Flex® cotton. FSANZ has previously issued such approvals for oil and linters from Roundup Ready® cotton and Bollgard II® cotton.

99. Monsanto will also require the APVMA to authorise an extension of the current registration of Roundup Ready® Herbicide to enable its application to Roundup Ready Flex® cotton (including application after the four-leaf stage of growth).


4.2 International approvals 100. In the United States of America, Roundup Ready Flex® cotton was approved for commercial release in December 2004 (USDA-APHIS 2004) and for use in food in March 2005 (US Food and Drug Administration). Bollgard II® was approved for commercial release in the USA in 2002.

101. Roundup Ready Flex®/Bollgard II® cotton is also approved, as under the current US regulatory system a stacked GMO is automatically approved if it was produced by conventional crossing of two GMOs, containing unrelated traits, that have already been approved in the USA.

102. In Japan, Roundup Ready Flex® cotton was approved for use in food in April 2005 and for use in animal feed in February 2006.

103. Field trials of Roundup Ready Flex® cotton are currently in progress in Mexico. Applications have been made for field trials in South Africa and Costa Rica (information supplied by the applicant).

104. Roundup Ready® and Bollgard II® cotton have been approved for commercial release in other countries:

• The USA - the US Department of Agriculture and the Food and Drug Administration approved the commercial release and use in food (respectively) of Roundup Ready® cotton in 1995 and Bollgard II® cotton in 2002

• Canada - the Canadian Food Inspection Agency and Health Canada approved the commercial release and use in food of Roundup Ready® cotton in 1996, and in June 2003 they authorised the use of Bollgard II® cotton event 15985 for livestock feed and human food

• Japan - the Japanese Ministries of Agriculture, Forestry and Fisheries, and Health, Labour and Welfare approved the commercial release of Roundup Ready® cotton in 1997 and its use in animal feed in 1998. Bollgard II® cotton was approved for use in food and animal feed in 2002 and 2003, respectively

• Argentina - approved the commercial release of Roundup Ready® cotton in 1999 and its use in human food and stockfeed in 2000 and 2001, respectively

• The Philippines - approved the use of Roundup Ready® cotton in animal feed and human food in 2003.

105. Other countries where Roundup Ready® cotton has been approved, or is pending approval, include India, Israel, Mexico and the European Union.

106. Limited and controlled releases of Bollgard II® cotton have been approved and carried out in Argentina, Costa Rica, India, Mexico and South Africa.



Chapter 2 Risk assessment Section 1 Introduction 107. Risk assessment is the overall process of identifying the sources of potential harm (hazards) and determining both the seriousness and the likelihood of any adverse outcome that may arise. The risk assessment (summarised in Figure 2.1) considers risks from the proposed dealings with the GMO that could result in harm to the health and safety of people or the environment posed by or as a result of gene technology.

Figure 2.1 The risk assessment process.

RISK CONTEXT GMO Dealings Parent organism Receiving environment

POTENTIAL CONSEQUENCES ADVERSE Severity of harm OUTCOME HAZARD RISK IDENTIFIEDSource of

ESTIMATE potential harm

RISK LIKELIHOOD Chance of harm

Note: words in bold are defined in Appendix A.

108. Once the risk assessment context has been established (see Chapter 1) the next step is hazard identification to examine what harm could arise and how it could happen during this release of GMOs into the environment.

109. It is important to note that the word ‘hazard’ is used in a technical rather than a colloquial sense in this document. The hazard is a source of potential harm. There is no implication that the hazard will necessarily lead to harm. A hazard can be an event, a substance or an organism (OGTR 2005).

110. Hazard identification involves consideration of events (including causal pathways) that may lead to harm. These events are particular sets of circumstances that might occur through interactions between the GMO and the receiving environment as a result of the proposed dealings.

111. A number of hazard identification techniques are used by the Regulator and staff of the OGTR, including the use of checklists, brainstorming, commonsense, reported international experience and consultation (OGTR 2005). In conjunction with these techniques, hazards identified from previous RARMPs prepared for licence applications of the same or similar GMOs are also considered.

112. The hazard identification process results in the compilation of a list of events. Some of these events lead to more than one adverse outcome and each adverse outcome can result from more than one event.

Chapter 2 Risk assessment: hazard identification (February 2006) 28


Section 2 Hazard characterisation 113. The list of events compiled during hazard identification are characterised to determine which events represent a risk to the health and safety of people or the environment posed by or as a result of gene technology.

114. A risk is identified only when there is some chance that harm will occur. Those events that do not lead to an adverse outcome or could not reasonably occur do not represent an identified risk and will not advance in the risk assessment process. Risks associated with the remaining events are assessed further to determine the seriousness of harm (consequence) and chance of harm (likelihood). The identified risks must be posed by or result from gene technology.

115. The criteria used by the Regulator to determine harm are described in Chapter 3 of the Risk Analysis Framework (OGTR 2005). Harm is assessed in comparison to the parent organism and other GMOs previously approved for commercial release, in the context of the proposed dealing and the receiving environment. The risk assessment process focuses on measurable properties for determining harm.

116. The following factors are taken into account during the analysis of events that may give rise to harm:

• the proposed dealings, which may include experimentation, development, production, breeding, propagation, use, growth, importation, possession, supply, transport or disposal of the GMOs

• the characteristics of the non-GM parent (OGTR 2002b)

• routes of exposure to the GMOs, the introduced gene(s) and its product(s)

• potential effects of the introduced gene(s) and its product(s) expressed in the GMOs

• potential exposure to the introduced gene(s) and its product(s) from other sources in the environment

• properties of the biotic and abiotic environment at the site(s) of release

• agronomic management practices for the GMOs

• the size, duration and regions of the release.

117. Limited and controlled releases of the same GMOs have occurred in field trials approved under licence DIR 035/2003 and DIR 055/2004. GM cotton lines expressing the same introduced proteins have been commercially released south of latitude 22º South under DIR 023/2002 (Roundup Ready® cotton) and DIR 012/2002 (Bollgard II® and Bollgard II®/Roundup Ready® cotton) since 2000 and 2002, respectively. There have been no reports of adverse effects on the health and safety of people or the environment resulting from any of these releases.

118. Forty one events that are discussed in detail later in this Section are summarised below in Table 2.1. Events that share a number of common features are grouped together in broader hazard categories as indicated in the table. Eight of the events that were characterised are considered to lead to an identified risk that required further assessment.



Table 2.1 Summary of events that may give rise to adverse outcomes Hazard

category Event that may give rise to an

adverse outcome Potential adverse outcome

Identified risk?

Reason

1. Ingestion of GM plant materials and food products.

Toxicity for people

No People usually only consume processed products of cotton plants (oil and linters) which do not contain detectable protein or genetic material. Evidence from feeding studies indicates that the introduced proteins have very low acute oral toxicity for mammals. Other GM crops (eg soybean and corn) containing the same or similar CP4 EPSPS, Cry1Ac, GUS or NPTII introduced proteins are approved for use in food.

2. Contact with, or inhalation of, GM plant materials containing the introduced CP4 EPSPS, Cry1Ac, Cry2Ab, GUS or NPTII proteins.

Toxicity for people

No On the basis of the very low acute oral toxicity of the introduced proteins, it is also expected that they will have very low acute dermal and inhalation toxicity. People are already exposed to commercially released GM cotton lines containing the same proteins without evidence of toxicity.

3. Consumption of animals that were fed GM plant material.

Toxicity for people

No The introduced proteins and DNA of the GM material are rapidly broken down into smaller components in the digestive tract of animals that are fed cotton seed. As a result, meat produced from these animals would be no different to meat from animals that were fed seed from non-GM cotton.

4. Consumption of honey produced by bees that pollinated GM plants.

Toxicity for people

No The pollen content of honey is less than 0.1%. The introduced proteins exhibit very low acute oral toxicity.

SECTION 2.1

Production of a substance toxic to people

5. Consumption of mushrooms cultivated on cotton trash/compost.

Toxicity for people

No The introduced proteins and DNA would be degraded during composting. Mushrooms can only take up break-down products of proteins (eg amino acids), not intact protein.

1. Use of GM plant material in food. Allergic reactions in people

No People usually only consume processed products of cotton plants (oil and linters) which do not contain detectable protein or genetic material. Evidence indicates that the introduced proteins are not allergenic.

SECTION 2.2

Production of a substance allergenic to people

2. Contact with items containing GM cotton fibre.

Allergic reactions in people

No Processed lint does not contain detectable amounts of protein.



Hazard category

Event that may give rise to an adverse outcome


Identified risk?

Reason

3. Contact with GM plant materials containing the introduced proteins.

Allergic reactions in people

No Evidence indicates that the introduced proteins are not allergenic. The same proteins are also expressed in other commercially released GM cotton lines. There have been no reports of allergenicity from this route of exposure.

1. Direct or indirect ingestion of the introduced proteins by vertebrates as a result of this release.

Toxicity for vertebrates

No Evidence from feeding studies indicates that the introduced proteins have very low acute oral toxicity for mammals. The same or similar proteins are widespread in the environment and vertebrates are therefore already exposed to them. These same proteins are also expressed in other commercially release GM cotton lines.

2. Direct ingestion of GM plant material by stock

Toxicity for vertebrates

No Compositional analysis indicates that cotton seed and raw cotton seed meal from Roundup Ready Flex® cotton are compositionally equivalent to that derived from non-GM cotton. Use of other commercially approved GM cotton lines (containing the same introduced proteins) for stockfeed has not shown increased toxicity compared to non-GM cotton.

3. Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates as a result of this release.

Toxicity for non-target invertebrates

Yes See Chapter 3, Event 1.

4. Direct or indirect ingestion of the CP4 EPSPS, GUS or NPTII proteins by invertebrates as a result of this release.

Toxicity for invertebrates

No There is no evidence suggesting that the CP4 EPSPS, GUS or NPTII proteins are toxic to invertebrates. They are exposed to these proteins through natural sources and through commercially released GM cotton lines.

SECTION 2.3

Production of a substance toxic to organisms other than people

5. Contact with the introduced proteins by microorganisms as a result of this release.

Toxicity for microorganisms

No There is no evidence suggesting that the introduced proteins are toxic to microorganisms. They are exposed to the same, or similar, proteins through natural sources and through commercially released GM cotton lines.

SECTION 2.4

Spread and persistence of the GM

1. Expression of the cp4 epsps gene construct increasing spread and persistence of the GM cotton plants through tolerance to glyphosate.

Weediness Yes See Chapter 4, Event 2.



Hazard category



Identified risk?

Reason

2. Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory.


3. Expression of the uidA or nptII gene or the presence of the regulatory sequences in the GM cotton plants.

Weediness No The uidA or nptII genes are not expected to have any influence on the spread and persistence of the GM cotton plants.

4. Dispersal of GM seed during transport or storage north of latitude 22º South.


5. Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South.


6. Dispersal of GM seed via flooding north of latitude 22º South.


7. Increased exposure of people to the introduced proteins as a result of spread and persistence of the GM cotton plants in the environment.

Toxicity/Allergic reactions in people

No These proteins are not toxic or allergenic and people are already exposed to them via bacteria expressing the proteins naturally or via other commercially released GM cotton lines and other GM crops that are approved for use in food containing the same or similar introduced proteins. The amount of exposure expected as a result of spread and persistence of the GM cotton lines would be small in comparison to the exposure from the extensive cultivation of the GM cotton plants, if they were approved for commercial release, and from similar GM cotton lines that are already commercially released (Roundup Ready® and Bollgard II®/Roundup Ready®).

cotton in the environment

8. Increased exposure of non-target invertebrates to the Cry1Ac and Cry2Ab proteins as a result of spread and persistence of the GM cotton plants in the environment.


No Non-target invertebrates are already widely exposed to the Cry1Ac and Cry2Ab proteins through the commercially released Bollgard II® GM cotton. The amount of exposure expected as a result of spread and persistence of the GM cotton lines would be small in comparison to the exposure from the extensive cultivation of the GM cotton plants, if they were approved for commercial release, and from a similar GM cotton line that is already commercially released (Bollgard II®).



Hazard category



Identified risk?

Reason

1. Gene transfer to native Gossypium species.

Weediness No Well established genetic incompatibility prevents vertical gene transfer to native Gossypium species.

2. Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance.


3. Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory.


4. Increased exposure of people to the introduced proteins as a result of gene transfer to G. hirsutum or G. barbadense plants.

Toxicity/Allergic reactions in people

No These proteins are not toxic or allergenic and people are already exposed to them via bacteria expressing the proteins naturally or via other commercially released GM cotton lines and other GM crops that are approved for use in food containing the same or similar introduced proteins. Compared to this amount of exposure, the amount of exposure to the introduced proteins expected as a result of vertical gene transfer would be small.

5. Increased exposure of non-target invertebrates to the Cry1Ac and Cry2Ab proteins as a result of gene transfer to G. hirsutum or G. barbadense plants.


No Non-target invertebrates are already widely exposed to the Cry1Ac and Cry2Ab proteins through the commercially released Bollgard II® GM cotton. Compared to this amount of exposure, the amount of exposure to the introduced proteins expected as a result of vertical gene transfer would be small.

SECTION 2.5

Gene flow by vertical gene transfer

6. Presence of the introduced regulatory sequences in G. hirsutum or G. barbadense plants.

Unpredictable effects

No The introduced regulatory sequences do not behave any differently than endogenous regulatory sequences in plants.

SECTION 2.6

Gene flow by horizontal gene transfer

1. Presence of the cp4 epsps, cry1Ac, cry2Ab, uidA, nptII or aad genes, or the introduced regulatory sequences, in other organisms.

Toxicity, weediness, increased pathogenicity or decreased effectiveness of antibiotic medicines

No Horizontal gene transfer of the introduced sequences to other organisms is not expected to result in any adverse outcomes during this release.



Hazard category



Identified risk?

Reason

1. Altered levels of innate toxic or anti-nutritional compounds as a result of random insertion of the gene construct into the cotton genome during development of the GM cotton lines.

Toxicity for people and other organisms

No Seed from the GM cotton plants has no significant changes in any of the toxic or anti-nutritional compounds in comparison to seed from the non-GM parent cotton line.

2. Altered metabolism of glyphosate in the GM plants expressing the CP4 EPSPS protein resulting in the production of toxic compounds.


No The introduced CP4 EPSPS protein functions in the same way (in the biosynthesis of aromatic amino acids) as the native enzyme in plants and is not involved in glyphosate metabolism.

SECTION 2.7

Unintended changes in toxicity

3. Synergistic effects of the introduced proteins when ingested in combination resulting in altered toxicity.


No The introduced proteins are not expected to have synergistic effects when ingested in combination (except, potentially, for the two Cry proteins for invertebrates. This is discussed in Chapter 3, Event 1).

SECTION 2.8

Unintended changes in biochemistry or physiology

1. Altered biochemistry or physiology of the GM cotton plants resulting from insertion or expression of the introduced genes.

Toxicity for people and/or other organisms or weediness

No The introduced CP4 EPSPS protein functions in the same way as the native enzyme in plants. Phenotypic and compositional analysis demonstrate that Roundup Ready Flex® cotton is equivalent to non-GM cotton indicating that biochemical pathways and plant physiology are not altered in the GM plants.

1. Decreased use of insecticide sprays on the GM cotton resulting in the increased prevalence of other insects.

Increased non-lepidopteran insect herbivory

No Bollgard II® cotton is already widely grown south of latitude 22º South and the proposed Roundup Ready Flex® /Bollgard II® would not pose an increased risk for higher prevalence of non-lepidopteran insect pests compared to Bollgard II®. Application of integrated pest management strategies developed by the cotton industry would be able to control any non-lepidopteran insect pests.

SECTION 2.9

Unintended effects on existing pests or weeds

2. Changes in the weed spectrum due to altered agricultural management practices.

Emergence of weeds that are more difficult to control

No The proposed Roundup Ready Flex® cotton crop management plan submitted as part of an application to the APVMA to vary the registration of Roundup Ready® Herbicide to enable its application to Roundup Ready Flex® specifies an integrated weed management strategy that would be implemented through a Technology User Agreement. This includes the requirement for a weed management audit which would allow identification and management of shifts in the weed spectrum.



Hazard category



Identified risk?

Reason

3. Expression of the introduced genes resulting in increased disease burden

Increased disease burden

No Previous releases of the same or similar GM cotton lines in Australia did not show increased disease burden. No differences were observed in the pest or disease status between Roundup Ready Flex® cotton and non-GM cotton during agronomic performance testing in the USA and Australia.

1. Development of Cry1Ac and Cry2Ab insecticide resistance.

Loss of insecticidal trait efficacy

No This is an issue that has been assessed by the APVMA previously (Bollgard II® registration). The resistance management conditions imposed by the APVMA would apply to Roundup Ready Flex®/Bollgard II® cotton.

2. Secondary effects on populations of organisms that interact with lepidopteran insects.

Potential to impact biodiversity

No In non-GM cotton, lepidopteran insect larvae are killed by commonly used broad-spectrum insecticides and this affects organisms that interact with lepidopteran insects. This hazard also relates to toxicity to non-target invertebrates and is addressed in Event 1 in Chapter 3.

3. Use of glyphosate on the GM cotton lines resulting in development of herbicide resistant weeds (in the agricultural environment)

Development of herbicide resistant weeds

No This issue is under consideration by the APVMA. The proposed Roundup Ready Flex® cotton crop management plan submitted as part of an application to the APVMA to vary the registration of Roundup Ready® Herbicide to enable its application to Roundup Ready Flex® is designed to minimise the potential for development of glyphosate resistant weeds. This plan would be implemented through a Technology User Agreement. The APVMA would also impose conditions on the use of the herbicide if it considered this necessary.

SECTION 2.10

Secondary impacts

4. Use of glyphosate resulting in reduced plant diversity.

Potential to impact biodiversity

No Cotton fields are generally kept free of other plants/weeds. The use of glyphosate on glyphosate tolerant GM cotton would mainly replace other methods of keeping cotton weed free and plant diversity would be no lower than it already is in non-GM cotton.



Hazard category



Identified risk?

Reason

5. Expansion of cotton cultivation into new areas.

Potential to impact biodiversity and ecosystems in these new areas

No The GM cotton lines have the same water and climatic requirements as non-GM cotton and the area where they may be grown south of latitude 22º South would therefore be restricted in the same way as it is for non-GM cotton.

SECTION 2.11

Unauthorised activities

1. Use of GMOs outside the proposed licence conditions (non-compliance).

Weediness No The Act provides for substantial penalties for non-compliance and unauthorised dealings with GMOs and also requires a test of the suitability of the applicant to hold a licence prior to the issuing of a licence by the Regulator.



2.1 Production of a substance toxic to people 119. Toxicity is the cascade of reactions resulting from exposure to a dose of a chemical that is sufficient to cause direct cellular or tissue injury, or otherwise inhibit normal physiological processes (Felsot 2000). Toxic proteins are known to act via acute mechanisms rather than through chronic exposure (Sjoblad et al. 1992). Toxicity may occur through ingestion, contact or inhalation. The level of toxicity is often expressed as the LD50. This is the amount of a substance given in a single dose that causes death in 50% of a test population of an organism.

120. Toxicity assays generally use the purified toxin of interest rather than the product that expresses the protein (eg GM plant material). This is necessary because the aim of the assays is to determine the concentration of toxin at which an adverse effect is seen. The level of expression in the product is used to determine the level of exposure to the toxin and comparison to the results of the toxicity assay indicate whether or not this is a safe level of exposure (OECD 1998; Konig et al. 2004). The use of purified toxin also increases the reproducibility of the assays. 2.1.1 Ingestion of GM plant materials and food products containing the introduced

proteins

121. Exposure to the introduced proteins could occur as a result of ingestion of material from the GM cotton plants. However, people do not normally eat cotton plants or any unprocessed material from cotton plants. The applicant intends to use cotton seed oil and linters from the proposed release in human food (subject to approval by FSANZ). However, protein or genetic material is not detectable in processed cotton seed oil and linters (Sims et al. 1996; USDA 2004). Any ingestion of material from the GM cotton plants containing the introduced proteins is expected to be very limited and result in extremely low exposure to the introduced proteins.

122. The introduced proteins present in the GM cotton lines are the same as those present in commercially released Roundup Ready® and Bollgard II® cotton lines. Studies using the purified forms of the introduced proteins present in Roundup Ready® and Bollgard II® cotton have been conducted to determine the level of acute oral toxicity of the proteins for mammals. These studies did not find any evidence of acute toxicity. Detailed descriptions of the results of these studies are available in the Risk Assessment and Risk Management Plans for DIR 012/2002 (Bollgard II® and Bollgard II®/Roundup Ready® cotton), DIR 022/2002 (INGARD® cotton) and DIR 023/2002 (Roundup Ready® and Roundup Ready®/INGARD® cotton). The key results of the toxicity studies are summarised in the following Sections.

The CP4 EPSPS protein

123. The CP4 EPSPS protein expressed in the GM cotton plants proposed for release is the same as that present in commercially released Roundup Ready® cotton. Toxicity studies using the purified form of the introduced CP4 EPSPS protein present in Roundup Ready® cotton have been conducted. Detailed descriptions of the results of these studies are available in the RARMPs for DIR 022/2002 and DIR 023/2002. Key results are summarised below.

124. Purified CP4 EPSPS protein, at acute doses of up to 572 mg/kg body weight, produced no adverse effects in mice (Harrison et al. 1996a). This is more than a thousand times the anticipated potential consumption of CP4 EPSPS in commercial food derived from all GM food crops expressing this enzyme under development by Monsanto at that time (soybean, potato, tomato, corn)(Harrison et al. 1996a). Sequence alignment with proteins in the toxin database did not show structurally relevant similarity between the CP4 EPSPS protein and any



known toxic or pharmacologically active protein relevant to human health (information supplied by the applicant).

125. The amino acid sequence of the CP4 EPSPS protein expressed in Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton is identical to, or shares greater than 99% sequence identity with, the amino acid of the CP4 EPSPS protein produced in a number of other Roundup Ready crops that are produced commercially, including soybean, corn and canola, in a number of countries. People have consumed these crops and/or their processed products since 1996 (James 2004) without any adverse effects reported.

126. In some countries, including the USA (but not Australia), cotton seed meal is approved for use in human diets (OGTR 2002b). Levels of CP4 EPSPS protein in seed from Roundup Ready Flex® cotton have been measured at 340 µg/g dry weight. A 70 kg person would need to consume approximately 117.7 kg of dried cotton seed at one sitting in order to be acutely exposed to 572 mg/kg body weight. As indicated above, no toxic effects were observed in mice at this level.

The Cry1Ac or Cry2Ab proteins

127. Toxicity studies using the purified forms of the introduced Cry1Ac protein present in INGARD cotton, or the Cry1Ac and Cry2Ab proteins present in Bollgard II® cotton, have been conducted. Detailed descriptions of the results of these studies are available in the RARMPs for DIR 012/2002 and DIR 022/2002. Key results are summarised below.

128. Purified Cry1Ac protein, at acute doses of up to 4300 mg/kg body weight, produced no adverse effects in mice (Naylor 1993a; Naylor 1993b). Likewise, acute oral toxicity studies in mice with purified Cry2Ab protein at doses of up to 1450 mg/kg have not shown any adverse effects (Bechtel 1999).

129. Multiple studies on the acute oral toxicity of Bt microbial preparations, containing Cry1Ac and Cry2Aa (to which Cry2Ab is 88% identical), in mammals such as rats and rabbits have revealed no adverse effects at very high doses (Carter & Ligget 1994; McClintock et al. 1995; Barbera 1995; Spencer et al. 1996). Two separate studies on people found no observable health effect of an oral dose of 1000 mg of Bt microbial spores per day for 3 or 5 days (McClintock et al. 1995; Betz et al. 2000).

130. A study that investigated the effects of the Cry1Ab protein (86% identical to the Cry1Ac protein) on a bovine hepatocyte culture concluded that there were no significant changes to cell morphology or the secretion of albumin or the enzyme lactate dehydrogenase. These results indicate that Cry1Ab has little acute toxicity on mammalian cells even when applied directly (Shimada et al. 2003).

131. In some countries, including the USA (but not Australia), cotton seed meal is approved for use in human diets (OGTR 2002b). Levels of Cry1Ac or Cry2Ab protein in cotton seed from Bollgard II® cotton have been measured at 3 and 50 µg/g fresh weight, respectively. A 70 kg person would need to consume approximately 100,333 kg of cotton seed at one sitting in order to be acutely exposed to 4300 mg Cry1Ac protein per kg body weight and 2,030 kg of cotton seed to be exposed to 1450 mg Cry2Ab protein per kg body weight. As indicated above, no toxic effects were observed in mice at these levels.

The GUS protein

132. Acute oral toxicity studies in mice with purified GUS protein at doses of up to 100 mg/kg did not show any adverse effects (Naylor 1992). Studies feeding humans and



animals with 1010 GUS-containing E. coli bacteria per ingestion also did not show any toxic or pathogenic reactions (Gilissen et al. 1998).

133. When the sequence of the GUS protein expressed in the GM cotton lines was compared to all protein sequences in publicly available databases, it only shared sequence similarities to homologous E. coli and other glucuronidase proteins, as expected (information provided by the applicant). These proteins have not been described as toxic to people. Metabolites of the E. coli GUS enzyme are non-toxic (Gilissen et al. 1998).

The NPTII protein

134. NPTII was the most commonly used marker gene recorded in the US field trial database for 2001 and 2002 (Miki & McHugh 2004). The insertion of the nptII gene into a wide range of GMOs has not resulted in any adverse effects (Flavell et al. 1992). The nptII gene was introduced into mammalian cell lines with no effects on viability or growth. During gene therapy experiments, mammalian cells expressing the NPTII protein have been infused into cancer patients. Again, no adverse effects have been observed (Flavell et al. 1992).

135. The NPTII protein produced in GM tomatoes has been fed to rodents and reported to be rapidly inactivated and degraded (Calgene 1990). An acute oral toxicity study in mice, in which the purified NPTII protein was fed at doses of up to 5000 mg/kg of body weight (2500 mg/kg administered twice, four hours apart), did not show any adverse effects (Berberich et al. 1993). Another similar study in mice also reported no adverse effects of NPTII at 5000 mg/kg of body weight (Fuchs et al. 1993c).

136. Protein and DNA sequence comparisons using sequences from four separate databases (Genbank, EMBL, PIR29, Swiss-Prot) indicated that NPTII does not have significant homology to any proteins listed as food toxins in these databases (FDA 1994).

Assessments by other agencies

137. FSANZ has previously assessed food products (oil and linters) derived from Roundup Ready® and Bollgard II® cotton plants (ANZFA 2000a; ANZFA 2002c), and GM food crops (eg soybean and corn) expressing the same or similar CP4 EPSPS, Cry1Ac, GUS or NPTII introduced proteins (ANZFA 2000a; ANZFA 2000b; ANZFA 2000c; ANZFA 2001e; ANZFA 2001f; ANZFA 2002a; ANZFA 2002c; FSANZ 2003), as safe for use in human food. FSANZ has also assessed food products derived from the Roundup Ready® Flex cotton plants proposed for release in this application. The assessment concludes that food products (oil and linters) derived from these GM cotton lines are as safe as those derived from non-GM cotton plants (FSANZ 2005).

138. The US Environment Protection Agency (EPA) considers the introduced proteins (CP4 EPSPS, Cry1Ac, Cry2Ab, GUS, NPTII) to be non-toxic to mammals and has established exemptions from residue tolerance requirements in all plants for these proteins (EPA 1994; EPA 1996; EPA 2000; EPA 2001b; US EPA 2002).

139. In Australia, the APVMA has also determined that a maximum residue limit (MRL) for Bt toxins in human food and animal feed is not necessary, indicating that they are of no toxicological significance (see The MRL Standard, Table 5 at: <http://www.apvma.gov.au/residues/mrl_standard.shtml>).

Conclusion

140. Any ingestion of material from the GM cotton plants containing the introduced proteins is expected to be very limited and evidence indicates that the introduced proteins have very



low acute oral toxicity for mammals. Therefore, no risk is identified and the potential for toxicity for people as a result of ingestion of the introduced proteins will not be assessed further. 2.1.2 Contact with, or inhalation of, GM plant materials containing the introduced proteins

141. Dermal and inhalation toxicity studies have not been conducted with the CP4 EPSPS, Cry1Ac, Cry2Ab, GUS or NPTII proteins. However, on the basis of their very low acute oral toxicity, they are also expected to be of very low acute dermal and inhalation toxicity. Since the introduced proteins are retained within the GM plant cells, dermal and inhalation contact may occur only when the plant cells have been damaged or broken. Therefore, only people who work with damaged GM cotton plants may come into contact with the introduced proteins during handling and/or processing the GM cotton or its products that contain the protein. The Roundup Ready® and Roundup Ready®/Bollgard II® cotton lines containing the same proteins have been commercially released since 2000 and 2002 respectively without reports of any adverse impacts on people working with the cotton lines. The concentration of protein available for contact via dermal and inhalation routes is expected to be low since the concentration of protein in GM cotton cells is low. The applicant intends to use cotton seed oil and linters from the proposed release in human food (subject to approval by FSANZ). However, protein or genetic material is not detectable in processed cotton seed oil and linters (Sims et al. 1996; USDA 2004). Therefore, no risk is identified and the potential for toxicity for people as a result of contact with, or inhalation of, GM plant materials containing the introduced proteins will not be further assessed. 2.1.3 Consumption of animals that were fed GM plant material

142. Mammals generally avoid feeding on cotton plants. The presence of gossypol and cyclopropenoid fatty acids in cotton seed limits the use of whole cotton seed as a protein supplement in animal feed, except for cattle which are less affected by these components. Inactivation or removal of these components during processing enables the use of some cotton seed meal for farmed fish, poultry and swine. The meal and hulls of cotton seed can also be used for cattle feed. Its use as stockfeed is limited, nonetheless, to a relatively small proportion of the diet and it must be introduced gradually, to avoid potential toxic effects.

143. The introduced proteins are rapidly degraded in mammalian digestive systems (see Section 2.2 of this Chapter for details).

144. The introduced proteins and DNA of the GM material are rapidly broken down into smaller components (eg protein and DNA fragments, amino acids, sugars etc.) in the digestive tract of animals that are fed cotton seed. As a result, meat produced from these animals would be no different to meat from animals that were fed seed from non-GM cotton.

145. The introduced proteins present in the Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton lines are the same as those present in commercially released Roundup Ready® and Bollgard II® cotton lines. Meat from cattle that were fed seed from these commercially released cotton lines has been consumed for several years (since 2000 and 2002, respectively) with no adverse effects reported.

146. Therefore, no risk is identified and the potential for toxicity for people as a result of consumption of animals that were fed GM plant material will not be assessed further. 2.1.4 Consumption of honey produced by bees that pollinated GM plants

147. Honey usually contains some pollen grains. The average pollen content of sieved honey is normally less than 0.1% (Agrifood Awareness Australia 2001). Pollen grains contain



protein and, therefore, may also contain the introduced CP4 EPSPS, Cry1Ac, Cry2Ab, GUS or NPTII proteins. However, levels of the introduced proteins in pollen from the GM cotton lines are expected to be very low. Levels of CP4 EPSPS protein in pollen from Roundup Ready Flex® cotton have been measured at 4 µg/g fresh weight. Amounts of the Cry proteins detected in pollen from Bollgard II® cotton are even smaller (≤ 0.32 µg/g fresh weight; detailed information is provided in the RARMP for DIR 012/2002, available from <http://www.ogtr.gov.au>).

148. All of the introduced proteins are of very low acute oral toxicity and other GM crops containing the same or similar introduced proteins are approved for use in food (see Section 2.1.1 of this Chapter). Therefore, no risk is identified and the potential for toxicity for people as a result of consumption of honey produced by bees that pollinated GM plants will not be assessed further. 2.1.5 Consumption of fungi cultivated on cotton trash/compost

149. Cotton trash can be used as a bulking agent to improve the efficacy of animal manure composting. The compost may be used for cultivation of fungi. However, the introduced proteins and DNA would be degraded during composting.

150. Fungi are heterotrophic (ie require carbon in organic form) and release enzymes into their environment that help degrade organic matter. Only break-down products of proteins (eg amino acids), not intact protein, can be taken up by the growing fungi.

151. Furthermore, the introduced proteins are of very low acute oral toxicity and other GM crops containing the same introduced proteins are approved for use in food (see Section 2.1.1 of this Chapter).

152. Therefore, no risk is identified and the potential for toxicity for people as a result of consumption of fungi cultivated on cotton trash/compost will not be assessed further.

2.2 Production of a substance allergenic to people 153. The possibility that exposure of people to proteins expressed from the introduced genes or their enzymatic products in the GM cotton plants may result in an allergic reaction is considered. Routes of exposure to the CP4 EPSPS, Cry1Ac, Cry2Ab, GUS or NPTII proteins could include consumption of food containing cotton products, accidental ingestion of material from the cotton plants, contact with clothing or household items containing cotton, or contact with material from cotton plants, either as a result of occupational exposure or from living near the proposed release. 2.2.1 Use of GM plant material in food

154. The applicant intends to use cotton seed oil and linters from the proposed release in human food and has received FSANZ approval for this use. Protein or genetic material is not detectable in processed cotton seed oil and linters (Sims et al. 1996; USDA 2004). Oil and linters from similar GM cotton lines (Roundup Ready® and Bollgard II®) expressing the same introduced genes have previously been approved by FSANZ for use in human food.

155. The only difference between the GMOs proposed for release and those already approved is that the CP4 EPSPS protein levels detected in Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton are higher than those detected in Roundup Ready® and Bollgard II®/Roundup Ready® cotton tissues. Leaves and seeds from Roundup Ready® cotton contain 42 – 53 and 60 – 299 µg CP4 EPSPS protein per g fresh weight, respectively (detailed information is provided in the RARMP for DIR 023/2002, available from




<http://www.ogtr.gov.au>) whereas Roundup Ready Flex® leaves and seeds contain 160 - 290 and 280 – 310 µg per g fresh weight, respectively (see Chapter 1).

CP4 EPSPS

156. The CP4 EPSPS protein is 47.6 kDa, which is in the typical range for allergenic proteins. However, it is rapidly denatured by heat, enzymatic digestion and acid in simulated mammalian digestive or gastric fluid ((Harrison et al. 1996a; Canadian Food Inspection Agency 1997; ANZFA 2001g), and information provided by the applicant). The protein shows no significant sequence homology to allergens assembled in the Genpept, PIR and SwissProt protein databases (Mitsky 1993). No immunologically relevant sequences (eight contiguous amino acid identities) were detected when the amino acid sequence of the CP4 EPSPS protein was compared to the ALLERGEN3 sequence database and the protein shows no sequence similarity to any proteins known to pose human health risks (information provided by the applicant). Roundup Ready soybean expressing the identical introduced CP4 EPSPS protein has been shown not to be allergenic (Batista et al. 2005).

Cry1Ac and Cry2Ab

157. The Cry1Ac protein is approximately 133 kDa in size, which is significantly larger than typical allergenic proteins. It is heat labile and rapidly degraded (under 30 seconds) under simulated mammalian gastrointestinal conditions (Fuchs et al. 1993a). The Cry2Ab protein is approximately 71 kDa in size, which is at the upper end of the typical size range for allergenic proteins, and is also easily digested (Leach et al. 2000). Neither the Cry1Ac nor Cry2Ab protein displays characteristics common to known food allergen proteins. Searches of allergen sequence databases have shown no significant matches of the Cry1Ac or Cry2Ab proteins to known allergens (Metcalfe et al. 1996).

158. While there have been reports in the USA claiming allergic reactions to Bt microbial products in topical insecticidal sprays, these are not due to the Cry1Ac or Cry2Aa proteins present in the Bt sprays. A survey conducted among farm workers who picked vegetables treated with Bt microbial products indicated that exposure to Bt products may lead to allergic skin sensitisation, however there was no clinical allergic disease in any of the workers. Most reactions in these workers were shown to be due to other constituents of the Bt sprays, and there was no evidence of antibodies specific to the Cry proteins of the Bt sprays (Bernstein et al. 1999). The US EPA have also determined that reports of reactions to Bt microbial products have been due to non-Cry proteins produced during fermentation or to other ingredients added to the insecticidal formulations (EPA 2001a).

GUS

159. The GUS protein is approximately 68 kDa in size, which is within the typical size range of allergenic proteins. However, the widespread occurrence of GUS and the constant exposure of humans to the protein without ill effect indicates that the likelihood that GUS being an allergen is extremely low (Gilissen et al. 1998). In addition, the GUS protein does not possess glycosylation sites and is rapidly denatured in the simulated mammalian digestive system (Fuchs & Astwood 1996; ANZFA 2002b). The GUS protein from E. coli is rapidly (<15 seconds) degraded in simulated gastric fluid and loses its activity by heating/cooking (Fuchs & Astwood 1996). Protein sequence comparisons using sequences from four separate protein databases (EMBL, Genbank, PIR29 and Swiss-Prot) indicated that GUS does not have significant sequence identity to any known protein food allergens.




NPTII

160. The NPTII protein is approximately 29 kDa in size, which is within the typical size range of allergenic proteins. However, it does not possess glycosylation sites, is not stable in the mammalian digestive system and is heat labile, decreasing the probability that it is allergenic (US FDA 1998; Fuchs et al. 1993b; FDA 1994; ANZFA 2001g)). Fuchs et al. (1993) reported that no NPTII was detected 10 seconds after addition of simulated gastric fluid as measured by both Western blot and enzymatic activity. Protein sequence comparisons using sequences from four separate protein databases (EMBL, Genbank, PIR29 and Swiss-Prot) indicated that NPTII does not have significant sequence identity to any known protein food allergens (Fuchs & Astwood 1996).

161. Other regulatory agencies, in Australia and in other countries, have previously assessed the use of the nptII gene in food crops (eg (ANZFA 2001a; ANZFA 2001b; ANZFA 2001c; ANZFA 2001d; FSANZ 2003). The FDA has evaluated data submitted for deliberate releases of GMOs expressing the NPTII protein and concluded that NPTII does not have any of the characteristics associated with allergenic proteins (US FDA 1998).

162. A number of genetically modified food crops containing the nptII gene have been approved for commercial release both in Australia (DIRs 012/2002, 021/2002 and 022/2002) and overseas. No adverse effects on humans, animals or the environment have been reported from these releases (US FDA 1998; Flavell et al. 1992; EFB 2001).

Conclusion

163. None of the introduced proteins in Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton are derived from a known allergenic source. The UK Royal Society have concluded that there is, at present, no evidence that available GM foods cause allergic reactions, and that the risks posed by GM plants are in principle no greater than those posed by conventional breeding or by plants introduced from other areas of the world (The Royal Society 2002).

164. Therefore, no risk is identified and the potential for allergic reactions in people resulting from exposure through food will not be assessed further. 2.2.2 Contact with items containing GM cotton fibre

165. Fibres are removed from the surface of cotton seed in two separate processes (Gregory et al. 1999). During ginning, the long fibres are removed from the seed. These fibres are called cotton lint. Cotton fabrics, used in clothing, upholstery, towels and other household products, are made from cotton lint. Following the ginning process, the seeds can be delinted, which involves the removal of the shorter fibres (linters) still attached to the seed. These cotton linters are used in a variety of products including medical dressings, felt, fine quality paper (including banknotes in many countries), twine and mops. Cellulose derivatives produced from the linters may be used in pharmaceuticals, cosmetics, toothpaste, lacquers, paints, ice cream, salad dressing and a variety of plastics (Gregory et al. 1999; FSANZ 2004). Cotton fibre is widely used in pharmaceutical and medical applications because of its very low allergenicity.

166. A study of the accumulation of mineral nutrients in cotton fruit found no detectable nitrogen in fibre fractions (Leffler & Tubertini 1976). Using more sensitive methods, specific proteins were detected at very low levels in raw, but not processed, linters and lint (Sims et al. 1996). Therefore the safety of wearing cotton clothing or using other products made from cotton is not expected to be affected by the genetic make-up of the cotton plants from which these components have been derived, that is, whether or not it is derived from GM or non-GM



cotton plants. Therefore, no risk is identified and the potential for allergic reactions in people resulting from exposure through cotton fibre will not be assessed further. 2.2.3 Contact with GM plant materials containing the introduced proteins

167. People working with cotton plants would be exposed primarily to the outer waxy cuticle layer at the plant surface, to the seed coat or to the cotton fibres, all of which are essentially free of protein. However, dermal exposure to proteins (including the CP4 EPSPS, Cry1Ac, Cry2Ab, GUS and NPTII proteins) or to other cellular components of the cotton plants may occur if damage to the plants during handling results in rupture of plant cells.

168. Inhalation of pollen by workers or people living near cotton crops could potentially result in allergic reactions if either of the introduced proteins were allergenic and were expressed in the pollen. The introduced proteins have been shown (CP4 EPSPS and Cry proteins), or are expected, to be present in most parts of the plants including pollen, leaves and seed. However, evidence indicates that none of the introduced proteins is allergenic (see Section 2.2.1 of this Chapter for details).

169. The primary processing of cotton at cotton gins, and the bulk handling of cotton seed and cotton fibre, can create and stir up fine dust and lint particles. Use of personal protective equipment by exposed workers is commonplace in such facilities to prevent respiratory irritations.

170. Therefore, no risk is identified and the potential for allergic reactions in people resulting from contact with GM plant material containing the introduced proteins will not be assessed further.

2.3 Production of a substance toxic to organisms other than people 171. A range of organisms may be exposed directly, through feeding on the GM cotton plants, or indirectly through eating organisms that feed on GM cotton plants. These organisms include vertebrates, invertebrates and microorganisms.

172. Cotton tissue (from either GM or non-GM plants), particularly the seeds, can be toxic to mammals if ingested in large quantities because of the presence of toxic and anti-nutritional factors including gossypol and cyclopropenoid fatty acids (eg dihydrosterculic, sterculic and malvalic acids).

173. Mammals generally avoid feeding on cotton plants. The presence of gossypol and cyclopropenoid fatty acids in cotton seed limits the use of whole cotton seed as a protein supplement in animal feed, except for cattle which are less affected by these components. Inactivation or removal of these components during processing enables the use of some cotton seed meal for farmed fish, poultry and swine. The meal and hulls of cotton seed can also be used for cattle feed. Its use as stockfeed is limited, nonetheless, to a relatively small proportion of the diet and it must be introduced gradually, to avoid potential toxic effects.

174. Neither cotton trash nor stubble are used as animal feed, due to the possible presence of pesticide residues. 2.3.1 Direct or indirect ingestion of the introduced proteins by vertebrates as a

result of this release

175. Mammals generally avoid feeding on cotton plants. In the field, seed cotton is present as large lint-covered bolls that are unattractive to avian species (OGTR 2002b), so birds are not likely to be exposed to the introduced proteins in the seeds of the GM cotton lines.

176. Cotton seed and pollen from the release are not expected to enter aquatic habitats in any significant quantities (OGTR 2002); therefore the level of exposure of aquatic vertebrates to



the GM cotton lines will be low. Irrigation practices (Good Management Practice of cotton industry) used by cotton growers in Australia retain irrigation water run-off, as well as the first 15 mm of storm water run-off, on-farm to minimise the entry of pesticide residues into natural waterways.

177. As discussed under Section 2.1.1, a number of studies aimed at determining the level of acute oral toxicity of the introduced proteins (CP4 EPSPS, Cry1Ac, Cry2Ab, GUS and NPTII) for mammals, have been performed. These studies did not find any evidence of acute toxicity.

178. The Cry1Ac and Cry2Ab proteins are known to bind to a specific receptor located in the gut of the larvae of some insects. Current evidence indicates that only lepidopteran insects are affected (see Chapter 3), although it is possible that insect species from some other invertebrate orders may be sensitive. Due to this high degree of specificity, it is unlikely that this protein will interact with receptors in vertebrates.

179. CP4 EPSPS, Cry1Ac, GUS and NPTII proteins are widespread in the environment, through the presence of the bacteria from which they are derived (see Chapter 1). The Cry1Ac protein is also expressed in Btk (B. thuringiensis var kurstaki) microbial sprays which are used to protect crops from insect herbivory.

180. The Cry2Ab protein is not naturally expressed in soil bacteria or Btk sprays but it is 88% identical at the sequence level to the Cry2Aa protein (Widner & Whiteley 1989; Dankocsik et al. 1990) that is naturally expressed in B. thuringiensis var kurstaki and present in Btk sprays (information supplied by the applicant for DIR 012/2002).

181. Related Cry proteins are also produced by other varieties of B. thuringiensis. Spores from a range of Bt varieties and their crystal (Cry) toxins are found widely in both the agricultural and natural environment, including in soil, on plant leaves, in grain stores and in dead insects (Meadows 1993).

182. The GUS and NPTII proteins are widespread in the environment since they are naturally produced by the common gut bacterium Escherichia coli. This bacterium is widespread in human and animal digestive systems as well as in the environment. The GUS protein used in GM plants is 99.8 % identical to the E. coli GUS protein. GUS enzyme activity has been detected in numerous microbial, plant and animal species, including species used as raw food (Gilissen et al. 1998). Humans (and, by implication, other animals) continually ingest kanamycin-resistant microorganisms, some containing the NPTII enzyme. Kanamycin-resistant bacteria have been isolated from soil, river water and sewage (Smalla et al. 1993).

183. It is therefore likely that many vertebrates will have been exposed to the introduced or similar proteins through natural sources, without evidence of toxic effects. Vertebrates will also have been exposed to the introduced proteins through their expression in the commercially released Roundup Ready®, Bollgard II® and Bollgard II®/ Roundup Ready® cotton lines (approved for commercial release south of latitude 22º South under DIR 023/2002 and DIR 012/2002 in 2000 and 2002, respectively). No adverse effects have been reported from these releases.

184. Since the same or similar proteins are produced by bacteria that are widespread in the environment, the introduced proteins are not expected to be a novel source of harm for vertebrates. The above evidence (Section 2.1.1 of this Chapter) also indicates that the introduced proteins are of very low acute oral toxicity for vertebrates. Therefore, no risk is identified and the potential for toxicity for vertebrates resulting from the expression of the introduced genes in the GM cotton lines will not be assessed further.



2.3.2 Direct ingestion of GM plant material by stock

185. The introduced proteins present in the GM cotton lines are the same as those present in commercially released Roundup Ready® and Bollgard II® cotton lines, which have been approved for use in stockfeed since 2000 and 2002, respectively. The use of cotton seed products derived from these GM cotton lines has not shown any increased toxicity for livestock as compared to non-GM cotton.

186. Acute oral toxicity studies in mice with each of the introduced proteins have not demonstrated any adverse effects ((Naylor 1992; Naylor 1993b; Fuchs et al. 1993c; Harrison et al. 1996b; Bechtel 1999); see Section 2.1.1 of this Chapter for more detail). In studies with rats, quail or catfish fed cotton seed meal at 5 to 20% of their diets, no significant differences were found in weight gain and feed conversion, nor during gross autopsy, for animals fed Roundup Ready® cotton seed meal (containing the CP4 EPSPS and NPTII proteins) compared to those fed non-GM cotton seed meal (Canadian Food Inspection Agency 1997). Similarly, there was no significant difference between catfish, quail or broiler chicken fed Bollgard II® cotton seed meal and animals fed non-GM cotton seed meal (Li & Robinson 2000; Gallagher et al. 2000; Mandal et al. 2004).

187. Compositional analyses of delinted cotton seed collected from Roundup Ready Flex® cotton grown in the USA or in Australia were conducted (Obert et al. 2003a; Obert et al. 2003b; George et al. 2005). The composition of raw cotton seed meal was also analysed (George et al. 2005). Cotton seed was analysed for its proximates, acid detergent fibre, neutral detergent fibre, crude fibre, total dietary fibre, amino acids, fatty acids, cyclopropenoid fatty acids, vitamin E, minerals and gossypol contents. Raw cotton seed meal was analysed for its proximates, different fibre fractions, amino acids, cyclopropenoid fatty acids, minerals and gossypol contents. Carbohydrate and caloric contents were determined by calculation.

188. Only minor differences were observed between Roundup Ready Flex® cotton, the negative segregant and commercially grown non-GM cotton lines. All differences detected were within the ranges expected for non-GM cotton, and cotton seed and raw cotton seed meal from Roundup Ready Flex® cotton were therefore considered to be compositionally equivalent to that derived from non-GM cotton.

189. A recent study investigated the effects of feeding GM cotton seed to dairy cows (Castillo et al. 2004). Measurements were made of dry matter intake, milk yield, milk composition, body weight and body condition score for each cow during two experiments which lasted for 28 days each. In one experiment, three types of GM cotton seed were tested: Bollgard® (INGARD®), Bollgard II®, Roundup Ready® and compared to a non-GM cotton line of similar genetic background. In the second experiment, one type of GM cotton (Bollgard II®/Roundup Ready®) was compared to three different varieties of non-GM cotton. No significant differences were measured between the treatments in each experiment, although cows in all treatments lost some weight during the experiments. This was ascribed to the very hot temperatures which is known to cause stress in cows.

190. Therefore, no risk is identified and the potential for toxicity for vertebrates resulting from the use of GM plant material as stockfeed will not be assessed further. 2.3.3 Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates as a result of this release

191. Potential non-target effects of the insecticidal proteins expressed in Roundup Ready Flex®/Bollgard II® cotton were considered in detail in the RARMP for Bollgard II® cotton (DIR 012/2002) and INGARD® cotton (DIR 022/2002), available at <http://www.ogtr.gov.au>.



192. Non-target invertebrates are already widely exposed to the Cry1Ac and Cry2Ab proteins through the commercially released Bollgard II® GM cotton. In the 2004–05 season, Bollgard II® (including Bollgard II®/Roundup Ready®) comprised 70% of commercially grown cotton crops. If a licence were issued for the current application and growers accept the new GM cotton lines, invertebrates would be exposed to the Cry proteins via Roundup Ready Flex®/Bollgard II® cotton in place of Bollgard II®/Roundup Ready®.

193. Although the primary targets of the Cry1Ac and Cry2Ab proteins are the two major pests of cotton plants, H. armigera and H. punctigera, other lepidopteran species are also sensitive to these toxins. More recent evidence also suggests that expression of different Cry proteins in combination can have synergistic or antagonistic effects on the toxicity for invertebrates. Therefore, a risk is identified for toxicity to non-target invertebrates resulting from ingestion of the GM cotton. The level of risk of toxicity to non-target invertebrates from this event is estimated in Chapter 3 as Event 1. 2.3.4 Direct or indirect ingestion of the CP4 EPSPS, GUS or NPTII proteins by invertebrates as a result of this release

194. There is no direct evidence to suggest that the CP4 EPSPS, GUS or NPTII proteins and other similar proteins are toxic to invertebrates. However, the genes that produced these proteins were derived from common soil or gut bacteria. Invertebrates are therefore expected to be exposed to these proteins in the agricultural environment already without evidence of toxic effects. EPSPS enzymes are naturally present in all plants, bacteria, algae and fungi; GUS enzymes are present in bacteria, invertebrates and animals; and NPTII is a phosphorylating enzyme, which does not possess any properties that distinguish it toxicologically from other phosphorylating enzymes present in microorganisms, plants and animals (FDA 1994).

195. Any additional exposure resulting from the proposed release of the new GM cotton lines is not expected to result in significant toxic effects. Therefore, no risk is identified and the potential of toxicity for invertebrates as a result of the presence of the CP4 EPSPS, GUS or NPTII proteins in the GM cotton plants will not be assessed further. 2.3.5 Contact with the introduced proteins by microorganisms as a result of this release

196. Soil microorganisms may be exposed to the introduced proteins through cotton plant material tilled into the soil after harvest or as a result of root exudation. However, a recent study shows that Cry1Ac protein is not released in root exudates of GM cotton (Saxena et al. 2004). Research investigating the toxicity of the introduced proteins for microorganisms has been discussed in previous RARMPs, most recently DIR 055/2004, available at <http://www.ogtr.gov.au>. The following is a summary of the key points discussed in the DIR 055/2004 RARMP.

The CP4 EPSPS protein

197. The effect of the CP4 EPSPS protein on carbon and nitrogen metabolism by soil microorganisms was tested based on OECD guidelines (Carson et al. 2004). Results support the conclusion that the CP4 EPSPS protein has no adverse impact on carbon and nitrogen metabolism by soil microorganisms and poses no significant risk to microorganisms and microbially-mediated processes in soil. In addition, another study shows that 90% of CP4 EPSPS protein is degraded in the soil within 9 days (Dubelman et al. 2005).

198. The CP4 EPSPS enzyme expressed by the GM cotton plants is derived from a common soil bacterium. EPSPS enzymes are present in all plants, bacteria, algae and fungi and perform



the same biochemical function in all organisms in which they are expressed. Any additional exposure resulting from the expression of CP4 EPSPS in the GM cotton plants is not expected to result in significant toxic effects.

The Cry1Ac and Cry2Ab proteins

199. Potential effects of the Cry1Ac protein on microorganisms have been considered in detail in the risk assessment for the continued commercial release of INGARD® cotton (DIR 022/2002) available at <http://www.ogtr.gov.au>. The key research is summarised here.

200. The effects on soil microorganisms of the Cry1Ac toxin, either purified or in GM cotton leaves, were examined in greater detail by Donegan et al (Donegan et al. 1995; Donegan & Seidler 1998). Numbers and types of protozoa, bacteria and fungi, as well as substrate utilisation tests and DNA fingerprinting of eubacterial ribosomal sequences, were used to analyse the composition of microbial soil communities. The addition of purified Cry1Ac toxin to soil did not cause any detectable changes to numbers of culturable microorganisms (bacteria or fungi).

201. In contrast, the addition of either non-GM leaf material or GM leaf material from one of two GM cotton lines expressing the Cry1Ac toxin caused increases in the number of bacteria and fungi present in the soil samples. This was expected, due to the increased organic matter. In comparison with the non-GM leaves, GM leaves generally caused a more rapid increase in the number of microorganisms, and when individual bacterial species were identified, it was found that GM and non-GM leaf tissue supported different species during decomposition. The authors speculated that these results may reflect faster decomposition and nutrient release from the GM cotton leaves compared to the non-GM leaves. The addition of purified Cry1Ac protein to non-GM leaves did not reproduce the effect, so this could possibly be due to some unknown and unintended effect, either of the genetic modification or the tissue culture process, on the plant characteristics, that was separate from expression of the Cry1Ac protein. No significant differences were found in the numbers of protozoa between any of the soil samples tested (Donegan et al. 1995; Donegan & Seidler 1998).

202. Purified Btk toxins had no effect on in vitro growth of pure or mixed cultures of gram positive bacteria (Bacillus subtilis, B. cereus, B. thuringiensis (subspecies kurstaki and israelensis) and Arthrobacter globiformis), gram negative bacteria (Agrobacterium radiobacter, Pseudomonas aeruginosa, Proteus vulgaris, P. mirabilis, E. coli, Enterobacter aerogenes, E. cloacae, Oscillatoria sp.), yeast, (Saccharomyces cerevisiae, Candida albicans), filamentous fungi (Rhizopus nigricans, Cunninghamella elegans, Aspergillus niger, Fusarium solani, Penicillium sp.), algae (Chlamydomonas sp., Oedogonium sp, Euglena sp.) and diatoms (Stotzky 2000).

203. In addition, a recent study examined the decomposition rates of INGARD®, Roundup Ready® and Roundup Ready®/INGARD® cotton plants and non-GM cotton plant residues (Gupta & Watson 2004). Results indicated that there were no significant differences between the decomposition rates of any of the samples. Fungal colonisation and total microbial respiratory activity appeared greater on INGARD® cotton residues. However, the authors did not investigate whether these changes were beneficial, detrimental or neutral.

204. Potential effects of the Cry2Ab protein on microorganisms have already been considered in the risk assessment for Bollgard II® cotton (DIR 012/2002) available at <http://www.ogtr.gov.au>. No studies have been performed to specifically examine the effect of the Cry2Ab protein on soil microorganisms. A study investigating soil degradation of Cry2Ab in tissues of Bollgard II® cotton shows that total insecticidal activity (Cry1Ac and Cry2Ab) of the samples declined at a rate similar to, and within the range of, the soil



degradation of Cry2Ab2 derived from corn tissue (1.1–19.9 days; (Martin et al. 2000)) and Cry1Ac alone in cotton tissue (2.2–46 days; (Ream 1994; Palm et al. 1996)). As with Cry1Ac alone in INGARD® cotton, the majority of the insecticidal activity of Bollgard II® cotton tissue dissipated rapidly within the first week, more slowly over the next week, and was undetectable after six weeks (Dubleman et al. 2001).

205. No detectable Cry1Ac protein or biological insecticidal activity was present in soil samples from fields grown to Bt cotton for three to six years despite incorporation of the Bt cotton plants into the soil after harvest (Head et al. 2002).

206. Therefore, both the Cry1Ac and Cry2Ab proteins present in GM plant material are expected to be degraded in soil.

207. A recent study demonstrates that biomass of Bt plants (corn, rice, tobacco, canola and cotton) decomposes less (measured as CO2 production) in soil than biomass of non-GM plants (Flores et al. 2005). However, the number of culturable bacteria and fungi, and the activity of enzymes involved in degradation of plant biomass, were not significantly different between soil amended with biomass of Bt or non-Bt corn. Therefore, the reduced decomposition could not be caused by differences in soil biota. A similar study did not detect any difference in the decomposition of leaves of Bt corn and non-Bt corn (Hopkins & Gregorich 2003).

208. Microorganisms (including bacteria, nematodes, and fungi) are already exposed to the Cry1Ac and Cry2Ab proteins, or similar proteins, in the agricultural environment through expression of these proteins naturally by soil bacteria, commercial use of Btk microbial sprays and commercial cultivation of the insecticidal GM cotton lines INGARD® (containing the cry1Ac gene) and Bollgard II® (containing the cry1Ac and cry2Ab genes) since 1996 and 2002, respectively. Any additional exposure resulting from the low level expression of Cry1Ac and Cry2Ab in the GM cotton plants is not expected to result in significant toxic effects.

The GUS protein

209. The direct effects of the GUS protein have not been tested on microorganisms. The GUS enzyme is derived from a common bacterium and is already present in soil and water ecosystems. Similar enzymes are present in many bacteria, invertebrates, vertebrates and some plants and perform similar biochemical functions in these organisms (Gilissen et al. 1998). Therefore, microorganisms are already widely exposed to the GUS protein or similar proteins. The GUS protein is also expressed in the commercially released Bollgard II® cotton plants. Any additional exposure resulting from the expression of GUS in the GM cotton plants is not expected to result in significant toxic effects.

The NPTII protein

210. NPTII is a phosphorylating enzyme, which does not possess any properties to distinguish it toxicologically from other phosphorylating enzymes present in microorganisms, plants and animals (FDA 1994). The function of this enzyme is the phosphorylation (inactivation) of the antibiotic neomycin (and the related kanamycin).

211. There is no direct evidence to indicate whether or not the NPTII proteins or other similar proteins may be toxic to some microorganisms. However, these proteins are produced by common bacteria to protect them against exposure to the antibiotics neomycin and kanamycin (EFB 2001; EFSA 2004; Goldstein et al. 2005). Microorganisms are therefore expected to be exposed to these proteins in the natural and agricultural environment already. INGARD® and Bollgard II® cotton plants expressingthis protein have been present in the



agricultural environment since their commercial release in 1996 and 2002 respectively. Any additional exposure resulting from the expression of the NPTII protein in the GM cotton plants is not expected to result in significant toxic effects.

Conclusion

212. Microorganisms are already widely exposed to the introduced, or similar, proteins and there is no evidence suggesting that the introduced proteins are toxic to microorganisms. Therefore, no risk is identified and the potential for toxicity for microorganisms as a result of the expression of the introduced genes will not be assessed further.

2.4 Spread and persistence of the GM cotton in the environment 2.4.1 Expression of the cp4 epsps gene construct increasing spread and persistence of the GM cotton plants through tolerance to glyphosate

213. The GM cotton plants produce sufficient CP4 EPSPS protein to provide tolerance to glyphosate, the active ingredient in Roundup Ready® Herbicide throughout the growing season. In environments where glyphosate is used to control weeds, the GM cotton plants would have some selective advantage. This could lead to spread and persistence of the GM cotton lines. Therefore, a risk is identified for weediness of the GM cotton lines as a result of the expression of the cp4 epsps gene. The level of risk of weediness from this event is estimated in Chapter 4 as Event 2. 2.4.2 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory

214. Roundup Ready Flex®/Bollgard II® cotton produces the Cry1Ac and Cry2Ab insecticidal proteins, in addition to the CP4 EPSPS protein, which provides tolerance to glyphosate. These Cry proteins are toxic to lepidopteran insects, including major lepidopteran pests of cotton crops. In environments where glyphosate is used to control weeds and lepidopteran herbivory is a significant limit on the spread and persistence of cotton plants, the GM cotton lines could survive and may become persistent in the environment. Therefore, a risk is identified for weediness of the GM cotton lines as a result of the expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination. The level of risk of weediness from this event is estimated in Chapter 4 as Event 3. 2.4.3 Expression of the uidA or nptII gene or the presence of the regulatory sequences in the GM cotton plants

215. In contrast, neither the uidA or nptII gene, nor the introduced regulatory sequences in the GM cotton lines are expected to have any impact on the spread or persistence of the GM cotton plants. The GUS protein, encoded by the uidA gene, is an enzyme involved in bacterial carbohydrate and energy metabolism and it is not expected to confer any selective advantage to cotton. Kanamycin and neomycin are not applied to plants outside of the laboratory, nor does expression by microorganisms result in high levels in the environment. The introduced regulatory sequences behave no differently to endogenous regulatory sequences in cotton. Therefore, no risk is identified and the potential for weediness of the GM cotton lines as a result of the presence of the uidA and nptII genes and/or introduced regulatory sequences will not be assessed further. 2.4.4 Dispersal of GM seed during transport or storage north of latitude 22º South

216. GM cotton lines are not currently permitted unrestricted release in northern Australia (ie north of latitude 22º South), pending the collection and assessment of data on whether insect resistance confers a survival advantage. The applicant is seeking approval for restricted



transportation of ginned cotton seed from the release to areas north of latitude 22º South for use as stockfeed.

217. Dispersal or accidental spillage of seed during transport or storage could allow the GM cotton plants to spread and persist in the environment where the GM cotton lines are not proposed for commercial release (ie north of latitude 22º South). Therefore, a risk is identified for weediness of the GM cotton lines as a result of dispersal of GM seed during transport or storage north of latitude 22º South. The level of risk of weediness from this event is estimated in Chapter 4 as Event 4. 2.4.5 Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South

218. The applicant proposes to use seed from the GM cotton lines in stockfeed anywhere in Australia. In addition to seed dispersal during feeding, a small percentage of cotton seed consumed by stock can pass through the digestive system intact and is able to germinate. As a result, cotton volunteers could establish in areas where livestock is fed (eg dairies and feed lots), or grazes after being fed, cotton seed north of latitude 22º South. This could allow the GM cotton plants to spread and persist in the environment where the GM cotton lines are not proposed for commercial release (ie north of latitude 22º South).

219. Therefore, a risk is identified for weediness of the GM cotton lines as a result of the dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South. The level of risk of weediness from this event is estimated in Chapter 4 as Event 5. 2.4.6 Dispersal of GM seed via flooding north of latitude 22º South

220. Some seed from the GM cotton plants may be dispersed from areas used for stockfeed and storage of GM cotton seed during flooding north of latitude 22º South. Seed may also be washed into drains, creeks and rivers close by. As a result, cotton seed may be transported from areas used for stockfeed and storage, and cotton volunteers may establish along waterways in areas north of latitude 22º South. This could allow the GM cotton to spread and persist in the environment where the GM cotton lines are not proposed for commercial release (ie north of latitude 22º South).

221. Therefore, a risk is identified for weediness of the GM cotton lines as a result of the dispersal of GM seed via flooding north of latitude 22º South. The level of risk of weediness from this event is estimated in Chapter 4 as Event 6. 2.4.7 Increased exposure of people to the introduced proteins as a result of spread and persistence of the GM cotton plants in the environment.

222. Spread and persistence of the GM cotton plants in the environment could lead to increased exposure of people to the expressed proteins. This could result in toxicity for people or allergic reactions.

223. However, all of the introduced proteins are of very low oral toxicity and other GM crops containing the same or similar introduced proteins are approved for use in food (see Section 2.1.1 of this Chapter for details). Evidence also indicates that none of the introduced proteins is allergenic and people are already widely exposed to the proteins via bacteria expressing the proteins naturally or via commercially released GM cotton lines. Commercially released Roundup Ready® and Bollgard II® GM cotton containing the same introduced proteins is already widely grown (see Chapter 1 for details).

224. The amount of exposure expected as a result of spread and persistence of the GM cotton lines would be small in comparison to the exposure from the extensive cultivation of the GM cotton plants, if they were approved for commercial release, and from similar GM cotton lines



containing the same introduced proteins that are already commercially released (Roundup Ready® and Bollgard II®/Roundup Ready®).

225. Therefore, no risk is identified and the potential for toxicity or allergic reactions in people as a result of spread and persistence of the GM cotton plants in the environment will not be assessed further. 2.4.8 Increased exposure of non-target invertebrates to the Cry1Ac and Cry2Ab proteins as a result of spread and persistence of the GM cotton plants in the environment.

226. Spread and persistence of the GM cotton plants in the environment could lead to increased exposure of non-target invertebrates to the expressed proteins. This could result in toxicity for non-target invertebrates.

227. Non-target invertebrates are already widely exposed to the Cry1Ac and Cry2Ab proteins through the commercially released Bollgard II® GM cotton. In the 2004–05 season, Bollgard II® (including Bollgard II®/Roundup Ready®) comprised 70% of commercially grown cotton crops. If a licence were issued for the current application and the new GM cotton lines are commercially grown, invertebrates would be exposed to the Cry proteins via Roundup Ready Flex®/Bollgard II® cotton in place of Bollgard II®/Roundup Ready®. Compared to this level of exposure, the level of exposure expected as a result of spread and persistence of the GM cotton plants would be minimal, since the numbers of GM cotton plants resulting from spread and persistence will be small in comparison to the numbers of plants that would be cultivated if Roundup Ready Flex®/Bollgard II® cotton were to be approved for commercial release.

228. Therefore, no risk is identified and the potential for toxicity for non-target invertebrates as a result of spread and persistence of the GM cotton plants in the environment will not be assessed further.

2.5 Gene flow by vertical gene transfer 229. Transfer of genetic material to offspring by reproduction (vertical gene transfer) could result in the transfer of the introduced genes or their associated regulatory elements to other plants. The only sexually compatible species present in Australia that could receive genes from the GM cotton lines are G. hirsutum and G. barbadense (including both cultivated (GM and non-GM) and naturalised cotton populations). 2.5.1 Gene transfer to native Gossypium species

230. As discussed in the Biology and Ecology of Cotton (Gossypium hirsutum) in Australia (OGTR 2002b), Australian flora contains 17 native Gossypium species, all of which are diploid (C, G or K genomes), while the cultivated cotton species are tetraploid (AD genomes). The native cotton species with highest potential for hybridising with G. hirsutum is G. sturtianum. Hybrids between these two species have been produced without application of plant hormones, when plants were in close proximity to each other. However, these hybrids were sterile, effectively eliminating any potential for introgression of G. hirsutum genes into G. sturtianum populations.

231. The centre of native Gossypium diversity in Australia is in northern Western Australia and the Northern Territory. Most of the Australian Gossypium species have limited distributions and occur at considerable geographic distance from cultivated cotton fields. Thus, gene transfer from the GM cotton plants to native cotton plants is prevented not only by genetic incompatibility but also by geographic constraints on cross-pollination (OGTR



2002b). Therefore, no risk is identified and the potential for weediness in these sexually incompatible species as a result of gene transfer will not be assessed further. 2.5.2 Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants providing glyphosate tolerance

232. Sexually compatible plants (ie naturalised, volunteer or commercially grown G. hirsutum or G. barbadense cotton plants, including other GM cotton plants) expressing the cp4 epsps gene as a result of vertical gene transfer could become tolerant to glyphosate. This could confer a fitness advantage on the plants in environments where glyphosate is used to control weeds. Therefore, a risk is identified for weediness as a result of vertical gene transfer of the cp4 epsps gene construct into other G. hirsutum or G. barbadense plants. The level of risk of weediness from this event is estimated in Chapter 4 as Event 7. 2.5.3 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants providing glyphosate tolerance and reducing lepidopteran herbivory

233. Sexually compatible plants (ie naturalised, volunteer or commercially grown G. hirsutum or G. barbadense cotton plants, including other GM cotton plants) expressing the cp4 epsps, cry1Ac and cry2Ab genes as a result of vertical gene transfer could become tolerant to glyphosate and resistant to lepidopteran insects. In environments where glyphosate is used to control weeds and where cotton plants are controlled by lepidopteran insects, these plants could survive and may become persistent in the environment. Therefore, a risk is identified for weediness as a result of vertical gene transfer of the cp4 epsps, cry1Ac and cry2Ab genes into other G. hirsutum or G. barbadense plants. The level of risk of weediness from this event is estimated in Chapter 4 as Event 8. 2.5.4 Increased exposure of people to the introduced proteins as a result of gene transfer to G. hirsutum or G. barbadense plants

234. Expression of either of the introduced genes in sexually compatible plants (ie naturalised, volunteer or commercially grown G. hirsutum or G. barbadense cotton plants, including other GM cotton plants) could lead to increased exposure of people to the expressed proteins. This could result in toxicity for people or allergic reactions.

235. However, all of the introduced proteins are of very low oral toxicity and both GM cotton and other GM crops containing the same or similar introduced proteins are approved for use in food (see Section 2.1.1 of this Chapter for details). Evidence also indicates that none of the introduced proteins is allergenic and people are already widely exposed to the proteins via bacteria expressing the proteins naturally or via commercially released GM cotton lines. Commercially released Roundup Ready® and Bollgard II® GM cotton containing the same introduced proteins is already widely grown (see Chapter 1 for details).

236. Compared to this level of exposure, the level of exposure expected as a result of vertical gene transfer to sexually compatible cotton plants would be minimal, since outcrossing of cotton is localised around the pollen source and decreases significantly with distance (OGTR 2002 and references therein).

237. Therefore, no risk is identified and the potential for toxicity or allergic reactions in people as a result of vertical gene transfer of the introduced genes into other G. hirsutum or G. barbadense plants will not be assessed further.



2.5.5 Increased exposure of non-target invertebrates to the Cry1Ac and Cry2Ab proteins as a result of gene transfer to G. hirsutum or G. barbadense plants

238. Expression of the cry1Ac and cry2Ab genes in sexually compatible plants (ie naturalised, volunteer or commercially grown G. hirsutum or G. barbadense cotton plants, including other GM cotton plants) could lead to increased exposure of non-target invertebrates to the Cry1Ac and Cry2Ab proteins. This could result in toxicity for non-target invertebrates.

239. Non-target invertebrates are already widely exposed to the Cry1Ac and Cry2Ab proteins through the commercially released Bollgard II® GM cotton. In the 2004–05 season, Bollgard II® (including Bollgard II®/Roundup Ready®) comprised 70% of commercially grown cotton crops. If a licence were issued for the current application and growers accept the new GM cotton lines, invertebrates would be exposed to the Cry proteins via Roundup Ready Flex®/Bollgard II® cotton in place of Bollgard II®/Roundup Ready®. Compared to this level of exposure, the level of exposure expected as a result of vertical gene transfer to sexually compatible cotton plants would be minimal, since outcrossing of cotton is localised around the pollen source and decreases significantly with distance (OGTR 2002 and references therein).

240. Therefore, no risk is identified and the potential for toxicity for non-target invertebrates as a result of vertical gene transfer of the introduced cry1Ac and cry2Ab genes into other G. hirsutum or G. barbadense plants will not be assessed further. 2.5.6 Presence of the introduced regulatory sequences in G. hirsutum or G. barbadense plants as a result of gene transfer

241. All of the introduced regulatory sequences operate in the same manner as regulatory elements endogenous to cotton plants. The transfer of either endogenous or introduced regulatory sequences could result in unpredictable effects. The impacts from the introduced regulatory elements are equivalent to and no greater than the endogenous regulatory elements. Therefore, no risk is identified and the potential for an adverse outcome as a result of vertical gene transfer of introduced regulatory sequences will not be assessed further.

2.6 Gene flow by horizontal gene transfer 2.6.1 Presence of the cp4 epsps, cry1Ac, cry2Ab, uidA, nptII or aad genes, or the introduced regulatory sequences, in other organisms

242. Transfer of the cp4 epsps, cry1Ac, cry2Ab, uidA, nptII or aad genes, or the introduced regulatory sequences, from the GM plants to sexually incompatible plants, animals or microorganisms (horizontal gene transfer) could occur only rarely without human intervention.

243. Transfer of the cp4 epsps gene to other organisms could cause the spread of this gene in the environment, leading to tolerance to glyphosate in other organisms. The CP4 EPSPS enzyme from the common soil bacterium Agrobacterium sp. is naturally insensitive to glyphosate (Padgette et al. 1993) as are a number of other microbial EPSPS enzymes (Schulz et al. 1985; Eschenburg et al. 2002). Thus, insensitivity to glyphosate is already widespread in microbial populations.

244. Transfer of the cry1Ac or cry2Ab gene to other organisms could cause the spread of these genes in the environment, leading to increased exposure of lepidopteran and other insects to the toxic proteins outside of the agricultural environment, and could also have adverse effects on populations of organisms that interact with lepidopteran insects in the food web. The cry1Ac and cry2Ab genes occur naturally in the common soil bacterium



Bacillus thuringiensis which has been isolated from a wide range of sources such as forest, soil, grain dust, bat dung, sea water and dead insects (Martin & Travers 1989).

245. Transfer of the uidA gene to other organisms could cause the spread of this gene in the environment, leading to increased presence of the GUS enzyme in the environment. The uidA gene is already widely present in bacterial species. The GUS protein is already widespread in the environment since it is naturally produced by the common gut bacterium Escherichia coli. E. coli is widespread in human and animal digestive systems as well as in the environment. GUS enzyme activity has been detected in numerous microbial, plant and animal species, including species used as raw food (Gilissen et al. 1998).

246. Transfer of the nptII or aad antibiotic resistance genes to other organisms could result in antibiotic medicines containing kanamycin/neomycin or streptomycin/spectinomycin, respectively, becoming less effective due to spread of resistance to these antibiotics in the environment. Kanamycin is not currently in human clinical use in Australia, while neomycin has limited and very specific clinical uses (TGA, MIMS Online database). Neomycin is registered with the APVMA for limited veterinary uses (APVMA, Pubcris database), however resistance to neomycin and kanamycin are widespread and other antibiotics are preferred (Flavell et al. 1992). Streptomycin is not currently in human clinical use in Australia, while spectinomycin has limited and very specific clinical uses (TGA, MIMS Online database). Streptomycin and spectinomycin are registered with the APVMA for limited veterinary uses (APVMA, Pubcris database). However, many bacteria are already resistant to these antibiotics (Shaw et al. 1993; Heym et al. 1994).

247. The nptII and aad genes were originally isolated from mobile genetic elements (transposons) found in the plasmids and chromosomes of E. coli. E. coli is widespread in human and animal digestive systems as well as in the environment. Transposons are readily transferable between bacterial species in nature. The nptII gene is associated with transposon Tn5 (Beck et al. 1982) and is observed in gram negative bacteria and Pseudomonas sp. While it is widely dispersed in the environment, other genes that also confer resistance to neomycin and kanamycin are more common, and also readily transferable among bacterial species (Smalla et al. 1994; Belgian Biosafety Server 1999). The aad gene is found in several transposons (eg Tn7 and Tn21) and occurs at high frequency among gram-negative bacteria (Belgian Biosafety Server 1999).

248. Since all of the introduced genes are already widely present in bacterial species, any transfer of these genes is much more likely to occur between bacteria, or bacteria and other organisms, than between the GM plants and other organisms.

249. Transfer of the regulatory sequences to other organisms could alter the expression of endogenous genes in unpredictable ways. However, all of the introduced regulatory sequences operate in the same manner as regulatory elements endogenous to cotton plants. The transfer of either endogenous or introduced regulatory sequences could result in unpredictable effects. As there is no difference between the two events, this does not represent a novel adverse outcome as a result of the genetic modification.

250. Horizontal gene transfer has been examined in detail in a number of other RARMPs (most recently DIR 057/2005), which are available from the OGTR website <http://www.ogtr.gov.au> or by contacting the Office. These assessments have concluded that horizontal gene transfer from plants to other sexually incompatible organisms occurs rarely and usually only on evolutionary timescales. Reports of horizontal gene transfer from plants to bacteria occurring during laboratory experiments have relied on the use of highly similar sequences to allow homologous recombination to occur, and conditions designed to enhance the selective advantage of gene transfer events (Nielsen et al. 1998; Mercer et al. 1999;




Nielsen et al. 2000; Gebhard & Smalla 1998; De Vries et al. 2001). Horizontal gene transfer is not expected to produce any adverse outcomes during this proposed release. Therefore, no risk is identified and the potential for an adverse outcome as a result of horizontal gene transfer will not be assessed further.

2.7 Unintended changes in toxicity 2.7.1 Altered levels of innate toxic or anti-nutritional compounds as a result of random insertion of the gene construct into the cotton genome during development of the GM cotton lines.

251. Cotton tissue (from either GM or non-GM plants), particularly the seeds, can be toxic if ingested in large quantities because of the presence of toxic and anti-nutritional compounds including gossypol and cyclopropenoid fatty acids (eg dihydrosterculic, sterculic and malvalic acids). There is potential for the GM cotton plants proposed for release to have increased levels of toxic or allergenic compounds as a result of the genetic manipulation.

252. Compositional analysis of Bollgard II® was conducted prior to its commercial release in 2002 (refer DIR 012/2002) and no difference was detected compared to other commercially grown cotton varieties. FSANZ approved the use of oil and linters from Bollgard II® cotton for use in food (ANZFA 2002c).

253. A detailed compositional analysis of Roundup Ready Flex® cotton in comparison to the parental line was assessed by FSANZ in deciding to approve Monsanto’s application to use food (oil and linters) derived from this GM cotton in human food. FSANZ concluded in its assessment report that food derived from Roundup Ready Flex® cotton is as safe as food derived from other cotton varieties (FSANZ 2005).

254. Compositional analysis of processed cotton seed oil (refined, bleached and deodorised) derived from Roundup Ready Flex® cotton shows that the levels of toxic and anti-nutritional compounds (cyclopropenoid fatty acids and gossypol) are within the range expected for commercial non-GM cotton (Obert et al. 2003b). Cotton seed oil derived from Roundup Ready Flex® cotton is considered to be compositionally equivalent to oil derived from conventional cotton.

255. Compositional analyses of raw cotton seed meal and delinted cotton seed collected from Roundup Ready Flex® cotton grown in the USA or in Australia were also conducted (Obert et al. 2003a; Obert et al. 2003b; George et al. 2005). There were no significant differences in any of the toxic or anti-nutritional compounds listed above and cotton seed and raw cotton seed meal from Roundup Ready Flex® cotton were considered to be compositionally equivalent to that derived from non-GM cotton.

256. Therefore, no risk is identified and the potential for toxicity as a result of unintended changes in toxicity will not be assessed further. 2.7.2 Altered metabolism of glyphosate in the GM plants expressing the CP4 EPSPS protein resulting in the production of toxic compounds

257. In plants, the native epsps (5-enolpyruvylshikimate-3-phosphate synthase) gene encodes an important enzyme (EPSPS) for the biosynthesis of aromatic amino acids (tryptophan, tyrosine and phenylalanine), which are essential building blocks for cellular proteins. During this biosynthetic process the EPSPS enzyme catalyses the addition of the enolpyruvyl moiety of phosphoenolpyruvate to shikimate-3-phosphate. EPSPS performs this function in plants, bacteria, algae and fungi but is absent from mammals, which are not able to synthesise these aromatic amino acids (Bentley 1990; Padgette et al. 1993).



258. The CP4 EPSPS protein expressed in the GM plants functions in the same way, in the biosynthesis of aromatic amino acids, as the native enzyme in plants. The only difference is that it is insensitive to the effects of glyphosate (Padgette et al. 1993) and therefore is still able to perform its function in the presence of glyphosate. The CP4 EPSPS protein is not involved in glyphosate metabolism and as a result no new metabolic products are expected to occur in the GM plants (OECD 1999). Therefore, no risk is identified and the potential for toxicity as a result of altered metabolism of glyphosate in the GM plants will not be assessed further. 2.7.3 Synergistic effects of the introduced proteins when ingested in combination resulting in altered toxicity

259. Bollgard II®/Roundup Ready® cotton expressing the same proteins as Roundup Ready Flex®/Bollgard II® cotton was approved for commercial release in 2002 (DIR 012/2002) and has been used in food and stockfeed since then. There have been no reports of any unexpected or unintentional adverse effects from this commercial release.

260. The gene conferring tolerance to glyphosate and the insecticidal genes of Bollgard II® cotton operate through independent, unrelated biochemical mechanisms. There is no evidence of any interaction between the genes, their products or their metabolic pathways, and no reason to expect that this is likely to occur. There is no evidence or reasonable expectation that synergistic effects are likely to occur from the combination of the two traits, or that they would result in new or increased risks relating to human health and safety or the environment.

261. Of the five introduced proteins, only Cry1Ac and Cry2Ab could have synergistic effects when ingested in combination, and if so only on toxicity for invertebrates. This is considered in Chapter 3, Event 1.

262. Therefore, no risk is identified and the potential for toxicity as a result of synergistic effects of the introduced proteins will not be assessed further.

2.8 Unintended changes in biochemistry or physiology 2.8.1 Altered biochemistry or physiology of the GM cotton plants resulting from insertion or expression of the introduced genes

263. Gene technology has the potential to cause unintended effects due to the process used to insert new genetic material or by producing a gene product that affects multiple traits. Such effects may include:

• altered expression of an unrelated gene at the site of insertion

• altered expression of an unrelated gene distant to the site of insertion for example, due to changes in chromatin structure, methylation patterns, transcriptional read-through

• increased metabolic burden associated with high level expression of the introduced gene

• novel traits arising from interactions of an introduced gene product with endogenous non-target molecules

• secondary effects arising from altered substrate or product levels in the biochemical pathway of the introduced gene product.

264. Such unintended effects might result in adverse outcomes such as toxicity or allergenicity; weediness, pest or disease burden; or reduced nutritional or agronomic value as compared to the parent organism. However, accumulated experience with genetic modification of plants indicates that the process has little potential for unexpected outcomes



that are not detected and eliminated during the early stage of selecting plants with new properties (Bradford et al. 2005).

265. Unintended changes in gene expression could alter either the biochemistry or the physiology of the GM cotton plants. Biochemical or physiological changes to the GM cotton lines proposed for release could occur either as a result of the expression of the introduced genes or of the transformation process itself. However, unintended changes that occur as a result of gene insertions are rarely advantageous to the plant (Kurland et al. 2003).

266. The GM cotton lines proposed for release were selected from a large number of initial transformants and were selected on the basis that they did not show any altered agronomic properties beyond expression of the introduced proteins (information supplied by the applicant).

267. Previous releases of these GM cotton lines in Australia (performed under DIR 035/2003 and DIR 055/2004) did not show any unintended secondary effects. Phenotypic evaluation, including growth and development, plant architecture at harvest, crop yield and fibre quality, of Roundup Ready Flex® cotton from four sites in the major cotton growing regions of Australia was conducted under DIR 035/2003. There was no statistically significant difference in any of the phenotypic characteristics, seed cotton yield and pest or disease status between Roundup Ready Flex® cotton and the negative segregant control (Dunn 2005). The only significant difference observed was in fibre length of unsprayed cotton plants. However, there was no difference between Roundup Ready Flex® cotton and the non-GM control when Roundup Ready Flex® was sprayed with glyphosate. Therefore, the difference is likely to be a statistical variation. Furthermore, all measurements were within normal ranges for cotton grown commercially under Australian conditions.

268. Field trials of Roundup Ready Flex® cotton conducted in the USA in 2002 also indicate that its agronomic characteristics are within the normal range for commercially grown cotton varieties ((Burns 2004); see Table 2.2). Table 2.2 Summary of results from phenotypic analysis of Roundup Ready Flex® cotton – USA field

trials in 2002 (Burns 2004)

Group of characteristics tested

Number of characteristics tested in this group

Number of characteristics for which a significant difference was detected

Details of difference (p ≤ 0.05) detected

Is difference within the range commercially grown cotton varieties?

Plant growth and development

11 1 Date until 50% flowering was 64 days in Roundup Ready Flex® as compared to 63 days in the negative segregant.

yes

Plant mapping 20 0 NA NA Boll/seed measurements

4 1 Seed index of Roundup Ready Flex® was lower as compared to the negative segregant (9.56g versus 9.83g per 100 fuzzy seed, respectively).

yes

Boll and fibre quality

6 2 Roundup Ready Flex® had reduced boll size and reduced micronaire.

yes

NA: not applicable



269. The Cry1Ac, Cry2Ab, GUS, and NPTII proteins act independently of endogenous metabolic pathways in cotton and are therefore not expected to alter other biochemical or physiological characteristics of the GM cotton plants.

270. The CP4 EPSPS protein expressed in the GM plants functions in the same way, in the biosynthesis of aromatic amino acids, as the native enzyme in plants and the only difference from the native enzyme is its insensitivity to glyphosate (Padgette et al. 1993; OECD 1999).

271. Both phenotypic (see above) and compositional (see Section 2.3 of this Chapter) analysis demonstrate that Roundup Ready Flex® cotton is equivalent to non-GM cotton indicating that biochemical pathways and plant physiology are not altered in the GM plants.

272. Therefore, no risk is identified and the potential for toxicity or weediness as a result of unintended changes in biochemistry or physiology will not be assessed further.

2.9 Unintended effects on existing pests or weeds 2.9.1 Decreased use of insecticide sprays on the GM cotton resulting in the increased prevalence of other insects

273. Roundup Ready Flex®/Bollgard II® cotton is resistant to the major lepidopteran pests of cotton crops, thus it is expected to require fewer applications of pesticide sprays than non-GM cotton crops. This has the potential to allow populations of other insects, normally controlled by pesticide sprays, to increase to problematic levels. This could not only result in an increased pest burden in the environment but may also increase the incidence of gene flow if pollinator species became more abundant.

274. Bollgard II® cotton is already approved for commercial release south of latitude 22º South (licence for DIR 012/2002 issued in 2002) and has been widely grown in this area. In the 2004–05 season, Bollgard II® (including the stacked variety Bollgard II®/Roundup Ready®) comprised 70% of all cotton grown. Bollgard II® cotton requires around 86% less insecticide spray than conventional non-GM cotton (Cotton Australia 2005a). Roundup Ready Flex®/Bollgard II® cotton expresses the same Cry proteins as Bollgard II® cotton and the insecticidal trait of the two cotton lines is identical (Burns et al. 2004). Therefore, the reduced requirement for pesticide sprays is expected to be the same for Roundup Ready Flex®/Bollgard II® and Bollgard II® cotton. If Roundup Ready Flex®/Bollgard II® were to replace the currently grown Bollgard II® cotton varieties, there would be no increased risk for a higher prevalence of insect pests.

275. Insect abundance and diversity in Bollgard II® cotton fields has been assessed in the RARMP prepared for DIR 012/2002 (available at <http://www.ogtr.gov.au>). Integrated pest management guidelines developed by the cotton industry recommend that Bollgard II® cotton must be monitored regularly for pests, similar to conventional non-GM cotton, to determine whether and what spraying is required (Johnson & Farrell 2004; Australian Cotton Cooperative Research Centre 2004). The reduction of insecticide use in Bollgard II® cotton has led to increased incidence of sucking pests, mainly mirids. These pests would be killed incidentally by broad-spectrum insecticides commonly used on non-GM cotton. Non-lepidopteran pest populations may need to be controlled in Bollgard II® if they reach a level where yield loss could occur. A range of insecticides is available to achieve this.

276. Therefore, no risk is identified and the potential for increased non-lepidopteran insect herbivory as a result of decreased use of insecticide sprays on Roundup Ready Flex®/Bollgard II® cotton will not be assessed further.



2.9.2 Changes in the weed spectrum due to altered agricultural management practices

277. The Australian Pesticides and Veterinary Medicines Authority (APVMA) has a complementary regulatory role in respect to this application due to its responsibility for agricultural chemical use in Australia, including herbicides, under the Agricultural and Veterinary Chemicals Code Act 1994.

278. For commercial products, the normal form of approval is through registration, but the APVMA may also issue permits allowing restricted use of a herbicide, for example, for a limited period of time or for a limited area. In considering applications for registration or permits, the APVMA also considers a number of issues that are outside the scope of the Gene Technology Regulator’s assessment, such as the efficacy of a herbicide and herbicide resistance management and, if necessary, imposes conditions in relation to these. The APVMA can impose conditions on both registrations and permits.

279. Changes in agricultural practices (eg adoption of minimal tillage) may cause weed population shifts. Any change in weed management practices (eg changes in herbicide use) will also cause a shift in the weed spectrum. For example, weed species that are inherently more resistant to the herbicide used than other weed species may become more problematic (Owen & Zelaya 2005; Nandula et al. 2005). A change in the weed spectrum is inevitable where Roundup Ready® Herbicide is used to replace other weed management practices. This could result in the emergence of weeds that are more difficult to control.

280. Weed control in non-GM cotton fields is mainly achieved by application of a range of residual and non-residual herbicides (pre- and post-emergent), inter-row cultivation, shielded- and spot-spraying, and hand weeding (chipping) (Charles 2002a).

281. Roundup Ready Flex® cotton plants can be sprayed with glyphosate herbicide to control weeds throughout the growing season. As a result, the following changes in weed management practices are anticipated for Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton, as compared to non-GM cotton:

• use of glyphosate, a non-residual herbicide, for weed control throughout the growing season over the top of the cotton plants

• reduction in the use of residual herbicides

• reduction in shielded- and spot-spraying

• reduced soil disturbance from mechanical cultivation methods

• reduced hand weeding (chipping).

282. Roundup Ready® GM cotton only enables the application of glyphosate herbicide for the control of weeds up to the four-leaf stage of the cotton plant (ie prior to flower formation, approximately 3 to 5 weeks after planting). It is grown widely in the agricultural environment following its commercial release in southern Australia (DIR 023/2002). In the 2004–05 season, Roundup Ready® cotton (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 50% of commercially grown cotton.

283. In contrast to Roundup Ready® cotton, glyphosate can be sprayed over the top of Roundup Ready Flex® cotton plants throughout the growing season (approximately 24 to 28 weeks). Australian field trials show that glyphosate can be applied over the top of Roundup Ready Flex® plants up to 60% boll opening without any yield loss (Dunn 2005). As a result, the main changes in weed management practices for Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II® cotton, as compared to Roundup Ready® and Bollgard II®/Roundup



Ready® cotton, would be that the use of glyphosate over the top of the former after the four-leaf stage of growth would reduce the requirement for the use of other herbicides, inter-row cultivation, shielded- and spot-spraying, and hand weeding.

284. These changes in weed management practices may induce shifts in the weed spectrum.

285. The potential for a shift in the weed spectrum as a result of changes to the use pattern of a herbicide, and how it might be managed, is assessed by the APVMA in considering applications for registration of agricultural chemical products, or changes to the use of a product that is already registered (refer to Section 2.10 for details). The application of Roundup Ready® Herbicide to Roundup Ready Flex® cotton and its use after the four-leaf stage would require a variation to the Roundup Ready® Herbicide registration. The APVMA is currently assessing an application from Monsanto relating to this variation.

286. Similar to the commercial release of Roundup Ready® cotton (see under 2.10.3 of this Chapter), Monsanto has submitted a Roundup Ready Flex® cotton crop management plan, which has been endorsed by the Transgenic and Insect Management Strategy (TIMS) committee of the Australian Cotton Growers Research Association, as part of its application. If the GM cotton lines were to be approved for commercial release, this crop management plan would be enforced through the Technology User Agreement for Roundup Ready® Flex cotton. The crop management plan specifies an integrated weed management strategy and the requirement for a weed management audit which would allow identification and management of shifts in the weed spectrum. If required, additional management strategies to counter changes in the weed spectrum may need to be applied.

287. As the potential emergence of weeds that are more difficult to control as a result of altered agricultural management practices is assessed and managed by the APVMA, this event will not be assessed further. 2.9.3 Expression of the introduced proteins resulting in increased disease burden

288. The function of the NPTII enzyme is the phosphorylation (inactivation) of the antibiotic neomycin (and the related kanamycin). In the environment, this enzyme is not likely to be active outside of living cells, as it requires specific chemical conditions for activity, including the availability of specific co-factors. Although antibiotic production by non-pathogenic bacteria has been implicated in suppression of some plant diseases (Brimecombe et al. 2001), no evidence for the involvement of neomycin or kanamycin has been found in a search of the scientific literature. Nor are these antibiotics used in agriculture for controlling soil-borne disease. Thus, the presence of NPTII in soil is not expected to impact on plant disease susceptibility. Furthermore, expression of NPTII in a variety of crop plants (eg canola, corn, cotton and tomato), over several years of agronomic performance testing and commercial cultivation, has not been linked to any increased occurrence of disease. Similarly, expression of the other introduced proteins (Cry1Ac, Cry2Ab, CP4 EPSPS and GUS) is not expected to affect the disease status of the GM plants.

289. Previous releases of the same GM cotton lines in Australia (conducted under DIR 035/2003 and DIR 055/2004) did not show increased occurrence of disease. GM cotton lines expressing the same introduced proteins have been commercially released south of latitude 22º South under DIR 023/2002 (Roundup Ready® cotton) and DIR 012/2002 (Bollgard II® and Bollgard II®/Roundup Ready® cotton) since 2000 and 2002, respectively. There have been no reports of any increased disease burden of these commercially released GM cotton lines.

290. No differences were observed in the pest or disease status of GM and non-GM cotton during phenotypic evaluation of Roundup Ready Flex® cotton during Australian field trials



conducted in the 2003–04 season (Dunn 2005). Similarly, no increase in pest potential was detected during agronomic performance testing conducted in the USA (Burns 2004).

291. Therefore, no risk is identified and the potential for increased disease burden as a result of the expression of the introduced proteins will not be assessed further.

2.10 Secondary impacts 2.10.1 Development of Cry1Ac and Cry2Ab insecticide resistance

292. The APVMA has a complementary regulatory role in respect to this application due to its responsibility for agricultural chemical use in Australia, including insecticides, under the Agricultural and Veterinary Chemicals Code Act 1994 (Ag Vet Code Act 1994).

293. For commercial products, the normal form of approval is through registration, but the APVMA may also issue permits allowing restricted use of an insecticide, for example, for a limited period of time or for a limited area. In considering applications for registration or permits, the APVMA also considers a number of issues that are outside the scope of the Gene Technology Regulator’s assessment, such as the efficacy of an insecticide, and insecticide resistance management and, if necessary, imposes conditions in relation to these. The APVMA can impose conditions on both registrations and permits.

294. Widespread and long-term use of Bollgard II® cotton varieties, including Roundup Ready Flex®/Bollgard II® cotton, could result in the emergence of resistance to the Cry1Ac and Cry2Ab proteins in the target species (Helicoverpa armigera and H. puntigera), and other susceptible lepidopteran species feeding on cotton. This would result in a reduction in the efficacy of these GM cotton lines for the control of insect pests, and could also have impacts on the efficacy of Bt microbial sprays to control insects in other agricultural systems. Potential adverse effects include attenuation of the potential benefits of growing Bollgard II® cotton to the environment and human health.

295. It should be noted that Bollgard II® cotton was developed with the specific intention of reducing the risk of insects developing resistance to the Cry1Ac or Cry2Ab proteins. The expression of two insecticidal proteins (which differ sufficiently in their mechanisms of action) in Bollgard II® cotton, and the fact that the overall insecticidal activity is increased relative to INGARD® cotton, is expected to delay the emergence of resistant insects. In 2004, INGARD® cotton was withdrawn from the market in Australia in favour of Bollgard II® cotton.

296. Bollgard II® (and Roundup Ready Flex®/Bollgard II®) cotton falls under the Ag Vet Code Act 1994 definition of an agricultural chemical product, due to its production of two insecticidal substances, and is thus subject to regulation by the APVMA. In July 2003, the APVMA registered the use of the insecticidal proteins as produced by the cry1Ac and cry2Ab genes in GM Bollgard II® cotton as insecticidal products.

297. A resistance management plan for Bollgard II® cotton has been developed by the TIMS Committee of the Australian Cotton Growers' Research Association in consultation with the APVMA (Monsanto Australia Limited 2004). The APVMA requires implementation of this plan as a condition of registration. The resistance management plan is designed to minimise the development of resistant insects and requires growers to employ a number of measures designed to achieve this objective. As part of the resistance management strategy, refuge crops must be grown, to allow Bollgard II®-sensitive insects to survive. These refuge crops may be unsprayed non Bt-cotton or other specific plant species (eg pigeon pea, sorghum or corn).



298. Cultivation of Roundup Ready Flex®/Bollgard II® cotton would also need to comply with this resistance management plan and all other relevant conditions imposed by the APVMA.

299. Therefore, no risk to the health and safety of people or the environment is identified as the potential for decreased efficacy of the insect resistance trait is being actively managed by the APVMA. 2.10.2 Secondary effects on populations of organisms that interact with lepidopteran insects

300. Widespread and long-term use of Bollgard II® cotton varieties, including Roundup Ready Flex®/Bollgard II® cotton, expressing the cry1Ac and cry2Ab genes could result in other organisms that interact with lepidopteran insects in the food web being adversely affected, which could result in reduced biodiversity.

301. Bollgard II® cotton is already approved for commercial release south of latitude 22º South (licence for DIR 012/2002 issued in 2002) and has been widely grown in this area. In the 2004–05 season, Bollgard II® (including the stacked variety Bollgard II®/Roundup Ready®) comprised 70% of all cotton grown. Roundup Ready Flex®/Bollgard II® cotton expresses the same Cry proteins as Bollgard II® cotton and the insecticidal action/trait of the two cotton lines is identical (Burns et al. 2004). The effect of Bollgard II® cotton, expressing the cry1Ac and cry2Ab genes, on other organisms that interact with lepidopteran insects has been assessed in the RARMP prepared for DIR 012/2002 (available at <http://www.ogtr.gov.au>). There would be no increased risk for organisms that interact with lepidopteran insects, if Roundup Ready Flex®/Bollgard II® were to replace the currently grown Bollgard II® cotton.

302. This is because lepidopteran insect larvae are killed by the insecticidal proteins in GM cotton expressing the Cry1Ac and Cry2Ab proteins (ie Bollgard II®, including stacked varieties). Whereas in non-GM cotton, they are killed by commonly used broad-spectrum insecticides. Therefore, the indirect adverse effects on other organisms that interact with lepidopteran insects in the food web will be no greater in GM than in non-GM cotton. In addition, invertebrates interacting with lepidopteran insects in the food web are likely to be adversely affected or killed by the use of broad-spectrum insecticides in non-GM cotton.

303. Therefore, no risk is identified and the potential for an adverse outcome for organisms interacting with lepidopteran insects, as a result of the expression of the cry1Ac and cry2Ab genes in Roundup Ready Flex®/Bollgard II® cotton, will not be assessed further. 2.10.3 Use of glyphosate on the GM cotton lines resulting in development of herbicide resistant weeds

304. Herbicide use on weed communities can exert selective pressure that leads to the development of herbicide resistant weeds. The repetitive use of a single herbicide, or herbicide group, increases the chance that development of herbicide resistant weeds will occur. Integrated weed management practices help to avoid selection of resistant weed biotypes (Avcare 2003). Integrated weed management has prevented the development of herbicide resistant weeds in Australian cotton fields up to this point (Roberts & Charles 2002).

305. The first confirmed cases of glyphosate resistance in Australia were in populations of Lolium rigidum (rigid ryegrass) (Powles et al. 1998; Pratley et al. 1999). Subsequently, other resistant populations of L. rigidum have been verified, with 34 confirmed in Australia in 2003 (Preston 2003). The majority of these populations have developed in fallows or horticultural situations with intensive use of glyphosate.



306. Glyphosate resistance has also been reported in a number of other weed species around the world (Heap 2003; Nandula et al. 2005). To date, a total of eight weed species have developed resistance to glyphosate (Nandula et al. 2005).

307. The APVMA operates the national system that evaluates, registers and regulates agricultural and veterinary chemical products. Any changes to the use of a product that is already on the market must also be referred to the APVMA. For commercial products, the normal form of approval is through registration, but the APVMA may also issue permits for experimental work that allow restricted use of an agricultural chemical, for example for a limited period of time or for a limited area.

308. In considering applications for registration or permits, as well as considering potential health and environmental impacts, the APVMA also considers a number of issues that are outside the scope of the Gene Technology Regulator’s assessment, such as efficacy and the trade implications of residues. The hazard of development of resistance to agricultural chemicals is also part of the APVMA’s assessment of agricultural chemical use. The APVMA can impose conditions on the use of chemical products in both registrations and permits. These conditions can include restrictions on use, implementation of a resistance management plan, and ongoing reporting on compliance.

309. As part of the Deed of Agreement between the Australian Government and Monsanto for the commercial release of Roundup Ready® and Roundup Ready®/INGARD® cotton in 2000, Monsanto was required to develop a crop management plan designed to minimise the potential for development of glyphosate-resistant weeds (Monsanto Australia Limited 2001) (refer DIR 023/2002). This plan was endorsed by the herbicide resistance subcommittee of the Transgenic and Insect Management Strategy (TIMS) committee of the Australian Cotton Growers Research Association and is enforced by Monsanto under its Technology User Agreement. The plan includes a requirement to prevent seed set of weeds that have survived exposure to Roundup Ready® Herbicide. Compliance with the current crop management plan and undertaking of a weed management audit endorsed by the TIMS subcommittee is required by the APVMA in connection with the use of Roundup Ready® Herbicide and the continued commercial release of Roundup Ready® cotton, as noted in the licence for DIR 023/2002 issued by the Regulator in June 2003.

310. Data collected by Monsanto from growers of Roundup Ready® cotton as part of the weed management audit (provided in support of licence application for DIR 023/2002) indicates general compliance with this plan. Over three seasons of cultivation of these GM cotton lines, there has been no indication of development of glyphosate resistant weeds and no change in the level of grower satisfaction with this technology.

311. Roundup Ready® Herbicide is not currently registered for the use on cotton (Roundup Ready® cotton) beyond the four-leaf stage of growth. The APVMA is assessing an application from Monsanto for a variation of the registration of Roundup Ready® Herbicide to allow its use on Roundup Ready Flex® cotton (including application after the four-leaf stage). As part of its application to the APVMA, Monsanto has submitted a Roundup Ready Flex® cotton crop management plan (see also 2.9.2 of this Chapter) designed to minimise the potential for development of glyphosate-resistant weeds. If the GM cotton lines were to be approved for commercial release, this plan would be implemented through a Technology User Agreement. The APVMA would also impose conditions on the use of the herbicide (eg restrictions on the number of applications that can be made and at what stage of crop growth these can be made eg in order to limit residue levels), similar to those imposed for the use of Roundup Ready® Herbicide on Roundup Ready® cotton, if it considered this necessary to manage any identified risks.



312. Therefore, no risk is identified as the potential for the use of glyphosate on the GM cotton lines resulting in development of herbicide tolerant weeds will be assessed and managed by the APVMA. 2.10.4 Use of glyphosate resulting in reduced plant diversity

313. Weeds or other plants in cotton fields adversely affect cotton by competing for resources such as nutrients, water and light. Therefore, cotton fields, whether GM or non-GM, are generally kept free of other plants/weeds. In non-GM cotton, this is mainly achieved by use of pre-emergent herbicides, shielded/directed in-crop herbicide application, cultivation and hand chipping (Charles 2002a; Radcliffe 2002b). The use of glyphosate on glyphosate tolerant GM cotton would mainly replace these other means. Plant diversity would be no lower than it already is in non-GM cotton.

314. Therefore, no risk is identified and the potential for the use of glyphosate resulting in reduced plant diversity will not be assessed further. 2.10.5 Expansion of cotton cultivation into new areas

315. The GM cotton lines have similar productivity and the same water, nutrient and climatic requirements as non-GM cotton (see Chapter 1). As a result, the area where they may be grown south of latitude 22º South would be restricted in the same way as it is for non-GM cotton. Expansion of cotton cultivation into new areas south of latitude 22º South would be mainly limited by the length of the season (especially number of days of appropriate temperatures), water availability (irrigation and/or rainfall) and suitability of the soil (good water retention qualities are required).

316. The main difference in cultivating the Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II cotton as opposed to non-GM or other GM lines is essentially an economic one as it would enable a reduction in the intensity of on-farm weed management practices. No changes to off-farm practices are required to control the spread and persistence of Roundup Ready Flex® plants.

317. Whether and where (within the regions suitable for growing cotton) the GM cotton lines would be grown would be determined by other factors such as market acceptance and seed and variety availability.

318. Therefore, no risk is identified and the potential to impact biodiversity and important ecosystems as a result of expansion of cotton cultivation into new areas will not be assessed further.

2.11 Unauthorised activities 2.11.1 Use of GMOs outside the proposed licence conditions (non-compliance)

319. If a licence were to be issued, non-compliance with the proposed conditions of the licence could lead to spread and persistence of the GM cotton plants outside of the proposed release area (south of latitude 22º South). The adverse outcomes that this event could cause are discussed in Section 2.4. The Act provides for substantial penalties for non-compliance and unauthorised dealings with GMOs. The Act also requires that the Regulator has regard for the suitability of the applicant to hold a licence prior to the issuing of a licence. These legislative provisions are considered sufficient to minimise risks from unauthorised activities. Therefore, no risk is identified and the potential for an adverse outcome as a result of unauthorised activities will not be assessed further.



Section 3 Risk estimate process for identified risks 320. Eight events from the hazard identification process (Events 1–8 in Table 2.1) are considered to lead to identified risks for the adverse outcomes of toxicity for non-target invertebrates and weediness.

321. Chapters 3 and 4 provide detailed assessment of the consequences and likelihood of these eight events in order to obtain estimates of the level of risk. The risks are assessed against the baselines established by reference to characteristics of the parent organism and aspects of the receiving environment (including agronomic management practices and other GM cotton lines previously approved for commercial release).

322. Information contained in the application (including information required by the Act and the Regulations on the GMOs, the parent organism, the proposed dealings and potential impacts on the health and safety of people and the environment), current scientific knowledge, and submissions received during consultation with experts, agencies and authorities and the public (Appendices B to E) were also considered.

323. The consequence assessment considers the seriousness of the harm that could potentially result from each event, while the likelihood assessment considers the chance of the event resulting in harm. Consequence and likelihood assessments are then combined to give an overall risk estimate using the Risk Estimate Matrix (Figure 2.2). During the consequence and likelihood assessments, consideration is also given to areas of uncertainty that arise from a lack of data.

Figure 2.2 The OGTR Risk Estimate Matrix (OGTR 2005)

RISK ESTIMATE

Highly Likely Low Moderate High High

Likely Negligible Low High High

Unlikely Negligible Low Moderate High

LIK

ELIH

OO

D

Highly Unlikely Negligible Negligible Low Moderate

Marginal Minor Intermediate Major CONSEQUENCES

Risk Estimate Matrix: A negligible risk is considered to be insubstantial with no present need to invoke actions for mitigation. A low risk is considered to be minimal but may invoke actions for mitigation beyond normal practices. A moderate risk is considered to be of marked concern that will necessitate actions for mitigation that need to be demonstrated as effective. A high risk is considered to be unacceptable unless actions for mitigation are highly feasible and effective.

324. Definitions of risk analysis terms used by the Regulator can be found in Appendix A.

325. After an estimate is obtained for each identified risk, risks higher than negligible are evaluated to determine if risk treatment measures are required to mitigate potential harm (see Chapter 5—Risk Management).



Chapter 3 Risk estimates for toxicity for non-target invertebrates

326. This Chapter estimates the risk associated with one event that could lead to the adverse outcome of toxicity for non-target invertebrates arising from this proposed release. The risk estimate is based on consequence and likelihood assessments for this event.

Section 1 Background 327. Toxicity is the cascade of reactions resulting from exposure to a dose of a chemical that is sufficient to cause direct cellular or tissue injury or otherwise inhibit normal physiological processes (Felsot 2000). Toxic proteins are known to act via acute mechanisms rather than through chronic exposure (Sjoblad et al. 1992).

328. Criteria for measuring toxicity of a substance may include decreased feeding, weight loss, decreased weight gain, altered development, reduced reproductive capacity, and death.

329. The two GM cotton lines proposed for release express either one or five proteins as a result of the genetic modification. Events that may give rise to toxicity for non-target organisms as a result of the proposed release were considered in Chapter 2. Expression of the CP4 EPSPS, GUS and/or NPTII proteins is not expected to provide a novel source of harm to non-target organisms, as these and similar proteins are present in the environment naturally and are expressed by common bacterial species without any indication of toxicity for any organism. Evidence also indicates that the Cry1Ac and Cry2Ab proteins are not toxic for vertebrates or microorganisms and are expected to pose a risk only to some non-target invertebrates. Therefore, this Chapter will be limited to assessing the risk of toxicity for non-target invertebrates as a result of ingestion of the Cry1Ac and Cry2Ab proteins in combination in the Roundup Ready Flex®/Bollgard II® cotton lines.

330. The toxicity of the Cry1Ac and Cry2Ab proteins for insect pests of cotton, in particular the targets H. armigera and H. punctigera, is not considered to be an adverse outcome. Other lepidopteran pest species of Australian cotton (Forrester & Wilson 1988; Johnson & Farrell 2004) for which the toxicity of the Cry1Ac and Cry2Ab proteins has been tested are:

• Pink bollworm (Pectinophora gossypiella) is reported to be sensitive to the Cry1Ac and Cry2Ab proteins (van Frankenhuyzen & Nystrom 2002).

• Armyworms (Spodoptera species) are reported to be unaffected by the Cry1Ac and Cry2Ab proteins at high doses (van Frankenhuyzen & Nystrom 2002)

• Cutworms (Agrotis spp) are reported not to be sensitive to the Cry1Ac protein and have not been tested for sensitivity to the Cry2Ab protein (van Frankenhuyzen & Nystrom 2002)

331. Other lepidopteran pest species such as the Cotton Tipworm (Crocidosema plebejana), the Cotton Webspinner (Loxostege affinitalis), the Cotton Looper (Anomis flava), the Cotton Leaf Perforator (Bucculatrix gossypii) have not been tested for sensitivity to the Cry1Ac and Cry2Ab proteins (van Frankenhuyzen & Nystrom 2002). The Rough bollworm (Earias huegeliana) has also not been tested although a relative, Earias vitella, is reported to be sensitive to the Cry1Ac protein (van Frankenhuyzen & Nystrom 2002). The Cry2Ab protein has not been tested on this genus (van Frankenhuyzen & Nystrom 2002).

Chapter 3 Risk assessment: toxicity for non-target invertebrates (February 2006) 67


Section 2 Consequence and likelihood assessments 332. Consideration is given to event 1 identified in Chapter 2 (Hazard identification) that may give rise to toxicity for non-target invertebrates. For this event the level of risk is estimated from assessment of the seriousness of harm (consequence —ranging from marginal to major) and the chance of harm (likelihood—ranging from highly unlikely to highly likely).

333. The Regulator can only consider risks to the health and safety of people or the environment posed by, or resulting from, gene technology. For this reason, the level of risk from the proposed dealings with the GMOs is considered relative to the baselines of toxicity for invertebrates of the non-GM parent and the environment in which the GM cotton plants are proposed for release. Therefore, other sources of the introduced genes or similar genes in the environment (such as from naturally occurring bacteria) and the widespread commercial plantings of Bollgard II® GM cotton in Australia are relevant to the risk estimate.

2.1 Toxicity of non-GM cotton 334. Information on non-GM cotton is included here to establish a baseline for comparison with the GM cotton lines being considered in this risk assessment. Cotton is a well-established field crop with a long history of use. A comprehensive review of the biology of non-GM cotton is provided in the document The Biology and Ecology of Cotton (Gossypium hirsutum) in Australia (OGTR 2002b).

335. Gossypol, a phenolic compound produced by cotton, is known to be toxic to insects (Percival et al. 1999). However, a study by Wilson et al. (1981) (cited in Percival et al. 1999) showed that high levels of gossypol, even in combination with morphological characteristics that discourage insect infestations, such as okra or laciniate leaf forms, were not sufficient to provide protection against the pink bollworm (Pectinophora gossypiella), a major pest of cotton in the USA.

336. As indicated by surveys carried out under previous cotton DIR licences (eg DIRs 017/2002 and 025/2002), large numbers of invertebrates can be found in unsprayed cotton crops. These include herbivores, predators and beneficial invertebrates.

2.2 Event 1: Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates as a result of this release

337. The Cry1Ac and Cry2Ab proteins are known to be toxic to the two major lepidopteran pests of cotton in Australia, namely H. armigera and H. punctigera and some other more minor pests (see Section 1). Although a number of Australian field studies have been performed to determine the ecological effects of GM cotton plants expressing the Cry1Ac and/or Cry2Ab proteins (ie INGARD® and Bollgard II®), the toxicity range of the Cry1Ac and Cry2Ab proteins has not been fully investigated in regard to Australian invertebrates.

338. In addition, toxicity for non-target lepidopteran insects may also result in indirect impacts that may give rise to harm through:

• effects on populations of specialist parasitoids and predators that feed on lepidopteran insects affected by the Bt toxin

• effects on populations of other organisms that interact with lepidopterans affected by the Bt toxins.

339. The risk of toxicity for non-target invertebrates from direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination would depend upon the toxicity of the these proteins both individually and in combination (consequence assessment) and the level of exposure to the proteins during this release (likelihood assessment). The risk is assessed



against the baseline of the toxicity of the non-GM parent organism for insects, toxicity due to the presence of the introduced genes in the environment in strains of common bacteria and commercial plantings of Bollgard II® GM cotton in the Australian environment that express the same proteins. 2.2.1 Consequence assessment

340. Potential non-target toxicity of the Cry1Ac and Cry2Ab proteins has been considered in detail in the risk assessment for the commercial release of Bollgard II® cotton (DIR 012/2002) and in the risk assessment for DIR 022/2002 for the continued commercial release of INGARD® cotton (expressing only the Cry1Ac protein). More recently, this issue was addressed again in the risk assessment prepared for DIR 055/2004. These risk assessments are available at <http://www.ogtr.gov.au>. A summary of the discussions in these RARMPs is presented below along with new information from the literature and tests commissioned by the OGTR.

Studies conducted under controlled conditions

341. A large number of insect species from a range of orders have been tested for sensitivity to the Cry1A class of proteins (including the Cry1Ac protein) (Macintosh et al. 1990; Sims 1994; Sims 1995; van Frankenhuyzen & Nystrom 2002). This class of Bt toxins is predominantly active against lepidopteran species, although there is some evidence that some dipteran species (eg mosquitoes) may also be affected when very high levels of protein are ingested. Where concentrations have been reported, these range from 50 000 to 100 000 ng of toxin/mL (Widner & Whiteley 1989; Liang & Dean 1994). This is approximately 200 fold lower level of toxicity than for the target insects. Levels of Cry1Ac protein in Bollgard II® leaves have been measured at 2 400 ng per g fresh weight (refer RARMP prepared for DIR 012/2002).

342. Studies have also been performed on key beneficial, non-target species using the same Cry1Ac protein that is produced by INGARD® and Bollgard II® GM cotton plants (Palmer & Beavers 1993a; Maggi 1993a; Palmer & Beavers 1993b; Maggi 1993b; Palmer & Beavers 1993c; Sims 1994; Sims & Martin 1996). There were no adverse effects of Cry1Ac observed, even at concentrations well above the expression levels found in INGARD® and Bollgard II® cotton, for any of the beneficial, non-target species tested in these studies.

343. The toxicity of the Cry2A class of Bt toxins, to which Cry2Ab belongs, has been tested against a large number of insect species from a range of orders (van Frankenhuyzen & Nystrom 2002). Bt toxins of this class are mainly toxic to lepidopteran insects but in some cases, dipteran insects can also be affected. The Cry2Ab toxin present in Bollgard II® cotton is generally considered to have specific toxicity for lepidopterans (Widner & Whiteley 1990; Widner & Whiteley 1989; Dankocsik et al. 1990). However, there is some evidence that Cry2Ab is active against at least one dipteran insect, Anopheles gambiae (African malarial mosquito) (Ahmad et al. 1989). The concentration at which mortality was seen was not reported.

344. Data have also been supplied by the applicant that indicate that Cry2Ab is non-toxic for a number of key beneficial, non-target species (Maggi 2000a; Palmer & Krueger 2000a; Maggi 2000b; Palmer & Krueger 2000b; Palmer & Krueger 2000c; Palmer & Krueger 2000d).

345. The OGTR commissioned additional work by CSIRO Entomology to investigate the toxicity of Bt cotton plants on a range of dipteran insects. Three dipteran species were tested (Culex quinquefasciatus, mosquito; Musca domestica, house-fly; and Chironomus tepperi,



bloodworm) using a freeze-dried powder of cotton leaves containing the Cry2Ab protein. The study found no evidence of toxicity at exposure levels that were equivalent to, or higher than, those that would occur in fresh GM cotton leaves (Akhurst 2005).

346. Recent research has reported the effects of exposure to the Cry1Ac protein on the parasitic wasp, Microplitis mediator, which parasitises the caterpillars of H. armigera (Liu et al. 2005). The Cry1Ac toxin did not have any effect on adult wasps when it was present in the 10% honey solution that the wasps were fed. When exposed female wasps laid eggs, there was a 3% increase in the length of time spent in the egg/larval stages but no significant effects on pupal weight, pupal period, adult weight or adult longevity. When caterpillars fed on Cry1Ac were parasitised, there was a dose-dependent effect on the length of the egg/larval stages of the parasitoid larvae with a 24% increase in this period at the highest dose of Cry1Ac (8 µg/g). In addition, pupal and adult weight were decreased, and adult longevity was also decreased. It is difficult to determine whether some or all of these effects are the result of sensitivity to the Cry1Ac toxin or whether they may be due to the low quality of the parasitised hosts due to their sensitivity to the Cry1Ac toxin.

347. The above studies all address the toxicity of the Cry1Ac or Cry2Ab proteins when they are ingested alone. Synergistic, additive or antagonistic effects can occur when different Cry proteins, or other insecticidal proteins, are ingested by an insect at the same time (Schnepf et al. 1998; del Rincon-Castro et al. 1999). If such effects occurred in the GM cotton lines, they could increase the potency of the GM cotton lines and could potentially lead to an altered toxicity range to that which might be extrapolated from the toxicity of the Cry proteins individually.

348. Studies have shown both synergistic and antagonistic effects between the Cry1Aa, Cry1Ab and Cry1Ac proteins for different lepidopteran species (van Frankenhuyzen et al. 1991; Lee et al. 1996) and have also suggested that the individual toxins may interact synergistically with the spores in spray formulations (van Frankenhuyzen et al. 1991; Liu et al. 1998). The spores in the spray formulations could be expected to contain the Cry2Aa protein, which is naturally expressed in the HD1 strain of B. thuringiensis from which the spray is formulated (see Chapter 1). The spores may also contain the VIP3A protein (Donovan et al. 2001), another insecticidal protein produced by some B. thuringiensis strains, including the HD1 strain. These studies have not directly investigated whether or not synergistic or antagonistic effects occur when the Cry1Ac and Cry2Ab proteins are ingested together.

349. Monsanto conducted studies in the field to investigate the ecological effects of the INGARD® and Bollgard II® cotton plants which have been grown commercially in Australia since 1996 and 2002 respectively. These field studies have been discussed in detail in the RARMP prepared for DIR 012/2002.

350. The overall conclusion from the field studies involving INGARD® cotton plants, expressing the Cry1Ac protein, showed no significant adverse impact on the sizes or variety of non-target invertebrate populations in GM cotton fields, as compared to unsprayed non-GM cotton. Similar studies using Bollgard II® cotton plants have also not shown any significant impacts on non-target invertebrate populations, suggesting that even if there is a synergistic effect occurring between the Cry1Ac and Cry2Ab proteins, this does not result in significant toxicity for non-target invertebrates.

351. The most recent published research compared the species diversity between sprayed and unsprayed non-GM cotton and two unsprayed Bt cotton lines (INGARD® and a stacked Bt cotton line expressing the Cry1Ac and Cry2Aa proteins) (Whitehouse et al. 2005). The greatest reduction on population diversity was caused by spraying of the non-GM cotton.



Approximately 5% of the differences seen between unsprayed non-GM cotton and Bt cotton were due to the type of crop. The most consistent differences were: greatly decreased numbers of Helicoverpa species and slightly decreased numbers of dipteran and hemipteran species on Bt crops in comparison to the unsprayed non-GM crops.

352. In summary, laboratory experiments have demonstrated that the Cry1Ac and Cry2Ab proteins are toxic to some lepidopteran insects and may be toxic to some dipteran species at very high concentrations. Field studies have indicated that growing Bollgard II® cotton plants does not have any significant effect on the sizes or variety of non-target invertebrate populations in GM cotton fields, as compared to unsprayed non-GM cotton. Therefore, expression of the Cry1Ac and Cry2Ab proteins in the Roundup Ready® Flex/Bollgard II® GM cotton plants is not expected to cause significant toxicity for non-target invertebrates and the consequences of direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates are assessed as minor. 2.2.2 Likelihood assessment

353. The commercial release of Roundup Ready® Flex/Bollgard II® cotton in regions south of latitude 22° South that are suitable for growing cotton could result in a large number of non-target invertebrates being exposed to the Cry1Ac and Cry2Ab proteins. However, non-target invertebrates are already exposed to the same proteins in these regions through the commercial release of Bollgard II® and Roundup Ready®/Bollgard II® cotton lines (DIR 012/2002). Roundup Ready® Flex/Bollgard II® cotton is expected to progressively replace Roundup Ready®/Bollgard II® in the cotton growing areas of Australia south of latitude 22° South. If the area where cotton is grown expands beyond the current cotton growing regions, the number of non-target invertebrates being exposed to the Cry proteins might increase. However, any expansion would be limited to areas that are suitable for growing cotton.

354. Non-target invertebrates may be directly exposed to Roundup Ready® Flex/Bollgard II® cotton plants and to the Cry1Ac and Cry2Ab proteins, through feeding on the plants. Exposure may also occur in the soil either when cotton tissues break down following incorporation into the soil or as a result of exudation of the introduced proteins through the roots. Indirect exposure may occur through eating other organisms, including the lepidopteran target pests of Roundup Ready® Flex/Bollgard II® cotton, which have previously fed on the plants.

355. Relative levels of exposure through these routes are discussed in the RARMP prepared for DIR 055/2004. The conclusions are that herbivorous species would have the highest level of exposure, with sap suckers, pollinators, pollen feeders, soil invertebrates and insectivores having reduced exposure in comparison to the herbivores.

356. A recent study has provided evidence that some non-target herbivores and arthropod predators collected from a Bt corn field contained detectable levels of the Cry1Ab protein present in the GM corn plants (Harwood et al. 2005). However, these insects were collected and killed for analysis and the study did not investigate whether there were any adverse effects resulting from this exposure to the Cry1Ab protein.

357. As discussed in the previous Section, populations of non-target invertebrates do not appear to be very sensitive to the levels of Cry1Ac and Cry2Ab expressed in commercially released Bollgard II® plants. The insect resistance of the Roundup Ready Flex®/Bollgard II® plants has been shown to be similar to that of Bollgard II® plants (see Chapter 1), suggesting that expression levels are equivalent. Bollgard II® cotton plants have been widely grown in



Australia since 2002 without evidence of significant adverse effects on non-target invertebrates.

358. In summary, exposure to the Cry1Ac and Cry2Ab proteins in Roundup Ready® Flex/Bollgard II® plants is not expected to be higher than in Bollgard II® plants and is not expected to cause greater impacts on non-target invertebrates than Bollgard II®. The likelihood of toxicity for non-target invertebrates as a result of direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination is estimated to be highly unlikely.

Section 3 Risk estimates 359. Risk estimates (which can range from negligible to high) are based on a combination of the consequences and likelihood assessments, using the Risk Estimate Matrix (see Chapter 2).

360. The risk estimates for the adverse outcome of toxicity for non-target invertebrates as a result of the proposed release of these GM cotton plants have been made relative to the baseline of the toxicity of non-GM cotton for invertebrates and in the context of the widespread use of commercially released Bollgard II® cotton plants in Australia without evidence of significant adverse effects on non-target invertebrates.

361. The consequences of direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in the GM cotton plants by non-target invertebrates has been assessed as minor, and the likelihood of this resulting in toxicity to them as highly unlikely. Therefore the risk estimate for event 1 is negligible.

362. As the risk of Event 1 leading to toxicity for non-target invertebrates is estimated to be negligible, there is no present need to invoke actions for mitigation (OGTR 2005). Therefore, no risk treatment measures for toxicity for non-target invertebrate organisms are proposed.

Table 3.1 Summary of risk assessment Event that may

give rise to toxicity for non-target invertebrates



Does risk require

treatment?

Event 1 Direct or indirect ingestion of the Cry1Ac and Cry2Ab proteins in combination by non-target invertebrates.

Minor • The Cry1Ac and

Cry2Ab proteins may be toxic to some non-target lepidopteran insects.

• The Cry1Ac and Cry2Ab proteins may be toxic to some dipteran insects at high concentrations.

• Field studies have indicated that growing Bollgard II® cotton plants does not have any significant effect on the sizes or variety of non-target invertebrate populations in GM cotton fields, as compared to unsprayed non-GM cotton.

Highly unlikely • Roundup Ready® Flex/

Bollgard II® plants are expected to progressively replace the commercially released Roundup Ready®/ Bollgard II® plants already grown in the release areas.

• Non-target invertebrates do not appear to be sensitive to the levels of Cry1Ac and Cry2Ab expressed in commercially released Bollgard II® plants.

• The insect resistance of the Roundup Ready Flex®/Bollgard II® plants has been shown to be similar to that of Bollgard II® plants, suggesting that expression levels of the insecticidal proteins are equivalent.

Negligible No



Chapter 4 Risk estimates for weediness 363. This Chapter estimates the risks associated with seven events that could lead to the adverse outcome of weediness arising from this proposed release. The risk estimates are based on the consequence and likelihood assessments of each event.

Section 1 Background 364. Weeds are plants that spread and persist outside their natural geographic range or intended growing areas such as farms or gardens. In addition, plants may also be considered weeds if they are simply growing where they are not wanted.

365. Weediness in Australia is often correlated with weediness of the species, or a close relative, elsewhere in the world (Panetta 1993; Pheloung et al. 1999; Pheloung 2001). The chance of weediness is increased by repeated intentional introductions of plants outside their natural geographic range that increase the opportunity for the plants to establish and spread into new environments (eg escapes of commonly used garden plants) (Mulvaney 2001; Groves et al. 2005).

366. Negative characteristics of weeds may include spread and persistence, competitiveness, rambling or climbing growth, toxicity, production of spines, thorns or burrs, or parasitism. In addition, the spread and persistence of weeds is a measure of their potential invasiveness, which may give rise to negative environmental impacts such as:

• reduced biodiversity (including genetic, species and ecosystem diversity) that results from lower abundance of desirable species, reduced species richness, or undesirable changes in species composition

• interference with the intended use of the land they occupy

• degradation of landscape/ecosystems, such as altered water or nutrient availability.

367. Complex interactions between a plant and its environment (including availability of water, nutrients and light) determine the degree to which that plant can spread and persist in the environment. A number of measurable properties of plants that may influence spread and persistence or competitiveness are listed below:

• germination, survival and reproduction under a wider range of environmental conditions

• rates of seedling growth

• rates of growth to reproductive stage

• degree of self-pollination

• use of non-specialist pollinators or wind when out-crossing

• period of seed production

• seed output

• degree of seed dispersal

• longevity of seed and degree of dormancy

• allelopathy (effect on the germination and/or growth of neighbouring plants through chemical exudates)

• resistance to pests or pathogens.

Chapter 4 Risk assessment: weediness (February 2006) 73


368. In the risk assessment, consideration is given to characteristics that may be expected to be altered as a result of the genetic modification and that may increase the spread and persistence of the GMOs, or of sexually compatible relatives that may receive the introduced gene(s). Alterations in these characteristics may indicate potential for weediness.

369. The GM cotton lines proposed for release express one or five proteins as a result of the genetic modification. Events that may give rise to weediness were considered in Chapter 2. Expression of the GUS and NPTII proteins in one of the GM cotton lines (Roundup Ready Flex®/Bollgard II®) is not expected to have any impact on the weediness of the GM cotton. The toxicity of the Cry1Ac and Cry2Ab proteins for lepidopteran pests of cotton and/or the tolerance to glyphosate herbicide conferred by the CP4 EPSPS protein may lead to weediness of the GM cotton plants. Therefore, this Chapter will be limited to assessing the risk of weediness as a result of expression of the cry1Ac, cry2Ab and cp4 epsps genes in the GM cotton lines.

Section 2 Consequence and likelihood assessments 370. Consideration is given to events 2–8 identified in Chapter 2 (Hazard identification) that may give rise to weediness. For each event the level of risk is estimated from assessments of the seriousness of harm (consequence—ranging from marginal to major) and the chance of harm (likelihood—ranging from highly unlikely to highly likely).

371. The Regulator can only consider risks posed by, or resulting from, gene technology. For this reason, the level of risk from the proposed dealings with the GMOs is considered relative to the baselines of weediness of the non-GM parent and the environment in which the GM cotton plants are proposed for release. Therefore, the widespread planting of the herbicide tolerant Roundup Ready®, the insect resistant Bollgard II® and the insect resistant/herbicide tolerant Bollgard II®/Roundup Ready® GM cotton in the current cotton growing regions in NSW and QLD is also considered when assessing the risks posed by the commercial release of Roundup Ready Flex® and Roundup Ready Flex®/Bollgard II®GM cotton lines.

2.1 Weediness of non-GM cotton 372. Information on non-GM cotton is included here to establish a baseline for comparison with the GM cotton lines being considered in this risk assessment. Attributes of non-GM cotton associated with potential weediness are discussed in the document The Biology and Ecology of Cotton (Gossypium hirsutum) in Australia (OGTR 2002b). This document concludes that non-GM cotton is not a serious weed in Australia, because environmental factors including temperature, soil moisture, nutrient limitation and roadside management practices limit the establishment and/or persistence of cotton outside of agricultural and other disturbed environments.

373. Small, persistent cotton populations have been observed, mainly in northern Australia. It has been noted by scientists over many years that the morphology of many of these naturalised cotton populations is distinct from that of the cultivated cotton varieties. When grown in a glasshouse, they tend to have poor architecture and produce small bolls and seed with sparse, grey lint. They also produce mainly tufted rather than fuzzy seeds, which is a strong indication that they are not derived from modern cultivars which are all fuzzy seeded cotton plants (Curt Brubaker and Lyn Craven, CSIRO, pers. comm.).

374. Tufted seeded cotton plants were originally used when hand delinting was required, before the advent of mechanical saw gins in the late 18th century. Tufted seeded cotton plants were subsequently replaced by fuzzy seeded varieties with better lint characteristics and disease resistance. It seems likely, therefore, that many naturalised cotton populations result



from attempts in the early 19th century to establish cotton industries in northern Queensland and the Northern Territory (Curt Brubaker and Lyn Craven, CSIRO, pers. comm.).

375. Other cotton plants appear to be of more recent origin (eg Eastick 2002) but these are confined to areas of disturbed land with at least a seasonal water supply: typical locations are above the high tide mark on beaches and near river banks.

376. An important indicator of potential weediness of a particular plant is its history of weediness in any part of the world and its taxonomic relationship to declared weeds (Bergelson et al. 1998; Panetta 1993; Pheloung 1995). Cotton has been grown for centuries throughout the world without any reports that it is a serious weed. Likewise, cotton is not considered to be a serious weed in Australia (Groves et al. 2000; Groves et al. 2002; Groves et al. 2003). Worldwide, there are about 50 species of Gossypium (Fryxell 1992; Craven et al. 1994) none of which is listed as a serious weed anywhere in the world (Holm et al. 1979; Holm et al. 1997; Randall 2002; Groves et al. 2003).

377. The weed status of cotton has also been considered previously in many of the RARMPs produced during the assessment of a variety of GM cotton lines (eg DIRs 012/2002, 022/2002, 023/2002, 036/2003, 055/2004). In addition to the information in the Biology and Ecology document (OGTR 2002b), these RARMPs have considered new data that has been collected during previous releases of GM cotton lines in Australia.

378. A survey for cotton volunteers was performed along transport routes around Narrabri and from Emerald to the Atherton Tablelands (Farrell & Roberts 2002). This report concluded that volunteer cotton plants had established in roadside environments only rarely during 12 years of transporting ginned cotton seed for stockfeed. Cotton plants were not observed in undisturbed habitats.

379. Small quantities of G. barbadense (Pima cotton) are also commercially grown in Australia. Herbarium records for G. barbadense suggest that naturalised populations may occur, or may have occurred in the past, in northern, central and south eastern Queensland and in the northern regions of the Northern Territory and Western Australia (OGTR 2002b). The presence of remnants of some of these populations has not been confirmed.

2.2 Event 2: Expression of the cp4 epsps gene construct increasing spread and persistence of the GM cotton plants through tolerance to glyphosate.

380. The risk of weediness of the GM cotton plants as a result of the expression of the cp4 epsps gene construct would depend on the weediness of non-GM cotton plants, the importance of the use of glyphosate in limiting the spread and persistence of cotton (consequence assessment), the scale of the release and the chance of progeny establishing as weeds (likelihood assessment). The risk is assessed against the baseline of the low weediness potential in the non-GM parent organism, in the context of the large scale of the proposed release and the receiving environment for the proposed release, which includes the commercial release of GM Roundup Ready® cotton containing the same introduced CP4 EPSPS herbicide tolerance protein.

381. Detailed discussions of the risk from expression of a single cp4 epsps herbicide tolerance gene increasing the weediness of cotton plants in Australia is provided in the risk assessment documents for DIRs 023/2003 (Roundup Ready® and Roundup Ready®/INGARD®) and 012/2002 (Bollgard II®/Roundup Ready®), available at <http://www.ogtr.gov.au>. These risk assessments concluded that expression of a single cp4 epsps gene, which provides tolerance to glyphosate herbicide only up to the four-leaf stage of growth, does not enhance the weediness potential of these GM cotton plants (in



comparison to non-GM cotton plants) in the cotton growing regions of Australia south of latitude 22° South.

382. While the expression of one copy of the cp4 epsps gene resulted in Roundup Ready® cotton being tolerant to glyphosate only up to the four-leaf stage of growth, this assessment takes into account that Roundup Ready Flex® cotton with two copies of the cp4 epsps gene is tolerant to glyphosate throughout the growing season. 2.2.1 Consequence assessment

383. The cp4 epsps gene construct (comprising two copies of the cp4 epsps gene) could confer a selective advantage in areas where glyphosate is used to control weeds. This could result in spread and persistence of the GM cotton lines in the environment.

384. As described in Section 1, the spread and persistence of a plant may have a number of negative environmental impacts including reduced biodiversity, interference with the intended use of the occupied land or degradation of the landscape/ecosystem.

385. As mentioned earlier, cotton is not a serious weed in Australia because of the limited availability of water, nutrients and suitable temperature conditions. Frost and soil moisture are particularly significant in relation to seedling germination and plant growth in southern Australia.

386. Glyphosate is the most widely used herbicide in Australia today, in both the agricultural and non-agricultural environment (Radcliffe 2002a). It is approved for the control of a wide range of annual, perennial, tree, brush and woody weeds. There are currently 253 products registered that contain glyphosate as active constituent (APVMA Pubcris database available at <http://www.apvma.gov.au/pubcris/subpage_pubcris.shtml>).

387. However, glyphosate is not generally used to control established cotton plants because while the application of glyphosate beyond the seedling stage results in yield loss, including reduced boll formation, it does not kill the plants and they can recover. Established or ratoon cotton plants, whether GM or non-GM, are difficult to control by herbicides alone. However, they are effectively controlled by mechanical methods involving mulching, root cutting and cultivation (Roberts et al. 2002).

388. Some GM cotton seed is likely to be dispersed during transport (eg during transport of cotton modules to gins) and therefore GM cotton volunteers may establish on roadsides where seed from the GM plants was transported.

389. Glyphosate may be used to control weeds on roadsides. In this situation, spraying is often limited to fixtures such as signs and guide-posts, while slashing is used in accessible areas.

390. If glyphosate were to be applied for weed control on roadsides, it would be ineffective on established cotton plants, whether GM or non-GM, and other means (ie mechanical methods) would need to be used to control it. However, compared to non-GM cotton, there might be an increased number of Roundup Ready Flex® plants surviving past the seedling stage only in areas where glyphosate is used. The number of seedlings surviving would be similar to that for GM Roundup Ready® cotton.

391. Some GM cotton seed is likely to be dispersed during flooding. As a result, cotton volunteers may establish along waterways (eg drains, creeks and rivers) or in flood prone areas. A number of glyphosate products are registered for use in aquatic areas such as drains, channels or the margins of dams (NRA 1996; APVMA 2004). However, as mentioned above, the use of glyphosate is only considered to be a limiting factor on the growth of cotton



seedlings, but not of established cotton plants. If glyphosate tolerant established cotton plants occurred near waterways, they can be effectively destroyed by mechanical means.

392. In the on-farm environment, a range of herbicides, including glyphosate, may be used to control cotton volunteers (at the seedling stage) that emerge after harvest. Herbicides containing carfentrazone-ethyl or paraquat and diquat as active constituents are currently registered by the APVMA for control of volunteer cotton, including Roundup Ready® cotton volunteers (APVMA Pubcris database available at <http://www.apvma.gov.au /pubcris/subpage_pubcris.shtml>). In addition, the mechanical methods that are used to destroy established cotton plants or ratoon regrowth are effective in controlling GM seedlings that are tolerant to glyphosate (Roundup Ready® or Roundup Ready Flex® cotton) (Roberts et al. 2002).

393. Integrated weed management strategies stress the need to avoid relying on one control method (Roberts & Charles 2002). To avoid the development of glyphosate resistant weeds for example, application of glyphosate alone should not be used as the sole management strategy. Alternation of various strategies would result in the destruction of glyphosate tolerant GM cotton volunteers. Consistent with this, the applicant has developed an integrated weed management strategy in the proposed Roundup Ready Flex® cotton crop management plan that would be implemented through a Technology User Agreement.

394. Therefore, the consequences of the expression of the cp4 epsps gene increasing the spread and persistence of the GM cotton plants proposed for release through tolerance to glyphosate are assessed as minor. 2.2.2 Likelihood assessment

395. The proposed release would result in the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton. Monsanto anticipates a phased introduction of the GM cotton lines commencing with up to 20,000 hectares in spring 2006 in the current cotton growing areas of NSW and southern QLD. The area is expected to increase in subsequent years and may include plantings in other areas south of latitude 22° South that are suitable for growing cotton.

396. The large scale of this proposed release would also result in the transport and storage of large quantities of GM cotton seed, which would increase the occurrence of seed dispersal events associated with these dealings. Seed is stored on farms in various ways (eg in sheds) that maintain its quality and protect it from animals and weathering.

397. The opportunity for any adverse outcome from dispersal of seed from the proposed release is further diminished by the limited availability of suitable environmental conditions for germination, survival and persistence of spilt seed.

398. As part of the commercial release of Roundup Ready® and Roundup Ready®/INGARD® cotton in 2000, Monsanto was required to conduct an environmental monitoring program. This three year program included monitoring for the incidence of volunteer GM cotton in non-agricultural situations (eg roadsides and non-crop areas). Details of the results were reviewed in the risk assessment prepared for DIR 023/2002 (available at <http://www.ogtr.gov.au>) and are only summarised here, along with more recent results from these studies (provided to the OGTR in the 2005 Annual Report required under the licence for the commercial release of Roundup Ready® cotton, DIR 023/2002).

399. Roadside surveys were conducted to determine the incidence of volunteer cotton, including testing for the Roundup Ready® trait, along roadsides in the Lower Namoi Valley (NSW; in 2002, 2003 and 2004) and the Darling Downs (QLD; in 2003 and 2004) – two



traditional cotton growing regions south of latitude 22º South (see Table 1.5). Table 4.1 summarises the incidence and frequency of cotton volunteers observed in these surveys.

400. Cotton was not a significant roadside weed in any of the regions surveyed. The number of volunteer cotton seedlings seems to be highly variable between seasons, indicating that it is most probably dependent on ideal germination conditions. The majority of cotton volunteers resulted from new germination rather than survival of plants from the previous season. Survival of cotton volunteers seemed to be limited by plant competition, roadside slashing and predation. Table 4.1 Summary of cotton volunteers observed in the 2002, 2003 and 2004 roadside surveys

across the Lower Namoi Valley and the Darling Downs

Lower Namoi Darling Downs Summary data 2002 2003 2004 2003 2004

Number of sites surveyed 70 77 78 40 24 Number of sites with volunteers

6 (8.6%) 33 (43%) 3 (3%) 5 (13%) 0

Number of sites with mature volunteers

NA 2 (2.6%) 3 (3%) 1 (2.5%) 0

Number of mature volunteers 6 12 19 26 0 Number of seedling volunteers NA ~ 1000 0 108 0 Total number of volunteers NA ~ 1012 19 134 0 Mature volunteers per km (each site = 0.2 km)

NA 0.8/km 1.22/km 3.3/km 0/km

Total volunteers per km 4.2/km ~ 65/km 1.22/km 14/km 0/km NA not available

401. Slashing appeared to be the common method of roadside weed control, and herbicide use tended to be limited to around fixtures (eg signs and guide posts) and drainage points where slashing is difficult. This suggests that glyphosate tolerance is not likely to provide a significant selective advantage.

402. Analysis of the correlation between percentage of Roundup Ready® volunteers detected and the proportion of Roundup Ready® cotton grown per valley in the previous season also did not indicate a selective advantage of Roundup Ready® cotton compared to other cotton varieties.

403. Flooding does occasionally occur in southern Australia and GM cotton seed may be dispersed by flooding. Some of this dispersed seed is not expected to survive as the viability of cotton seed is affected by moisture (Halloin 1975). Irrigation practices (Good Management Practice of cotton industry) used by cotton growers in Australia retain irrigation water run-off, as well as the first 15 mm of storm water run-off, on-farm to minimise the entry of pesticide residues into natural waterways. This practice would also reduce the dispersal of seed.

404. Although habitats close to waterways may be favourable for cotton establishment, tolerance to glyphosate is not expected to provide a significant selective advantage, compared to non-GM cotton, in these environments. Other environmental factors such as plant competition are expected to limit the establishment and persistence of cotton plants in these areas.

405. GM herbicide tolerant cotton plants are already being grown in the current cotton growing areas of NSW and southern QLD. The glyphosate tolerant Roundup Ready® GM cotton is widespread in the agricultural environment as a result of its commercial release (since 2000) in southern Australia and has not become a problematic weed. In the 2004–05



season, Roundup Ready® cotton (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 50% of commercially grown cotton.

406. Roundup Ready Flex® cotton differs from Roundup Ready® cotton only in that glyphosate herbicide can be applied over the top of the cotton crop to control weeds at later stages of growth (after the four-leaf stage) without yield loss. This is not expected to alter the on farm weediness potential of Roundup Ready Flex® cotton as glyphosate is not effective in controlling cotton plants beyond the seedling stage.

407. Some GM cotton seeds may spread from the release sites, germinate and persist in the environment following the release. However, the expression of the cp4 epsps gene construct is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (eg limited availability of water, nutrients and suitable temperature conditions), where frost and water availability in particular are known to be major limitations (see Chapter 1). The chance of volunteer GM plants establishing as weeds by finding suitable ecological niches would be no greater than for the non-GM parent organism. Therefore, the likelihood of weediness as a result of event 2 is assessed as highly unlikely.

2.3 Event 3: Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory.

408. The risk of weediness of the GM cotton plants as a result of the expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination would depend on the weediness of non-GM cotton plants, the importance of the use of glyphosate and lepidopteran herbivory in limiting the spread and persistence of cotton (consequence assessment), the scale of the release and the chance of progeny establishing as weeds (likelihood assessment). The risk is assessed against the baseline of the low weediness potential of the non-GM parent organism and in the context of the large scale of the proposed release and the receiving environment for this proposed release, which already includes the commercial release of a GM cotton line (Bollgard II®/Roundup Ready®) containing all five introduced genes.

409. Detailed discussions of the risk from expression of the CP4 EPSPS herbicide tolerance protein in combination with the two insect resistance proteins (Cry1Ac and Cry2Ab) increasing the weediness of cotton plants in Australia is provided in the risk assessment document for DIR 012/2002 (Bollgard II®/Roundup Ready®) available at <http://www.ogtr.gov.au>. This risk assessment concluded that expression of the proteins in combination does not enhance the weediness potential of these GM cotton plants (in comparison to non-GM cotton plants) in the cotton growing regions of Australia south of latitude 22° South.

410. While the expression of one copy of the cp4 epsps gene resulted in Roundup Ready® cotton being tolerant to glyphosate only up to the four-leaf stage of growth, this assessment takes into account that Roundup Ready Flex® cotton (including Roundup Ready Flex®/ Bollgard II®) with two copies of the cp4 epsps gene is tolerant to glyphosate throughout the growing season. 2.3.1 Consequence assessment

411. The cp4 epsps, cry1Ac and cry2Ab genes in combination could confer a selective advantage in areas where glyphosate is used to control weeds and lepidopteran insect predation limits one or more of the key life stages of cotton. This could result in spread and persistence of the GM cotton lines in the environment.



412. As described in Section 1, the spread and persistence of a plant may have a number of negative environmental impacts including reduced biodiversity, interference with the intended use of the occupied land or degradation of the landscape/ecosystem.

413. As mentioned earlier, cotton is not a serious weed in Australia because of the limited availability of water, nutrients and suitable temperature conditions. Frost and soil moisture are particularly significant in relation to seedling germination and plant growth in southern Australia.

414. The limited effectiveness of glyphosate in controlling cotton beyond the seedling stage is discussed in event 2. The risk assessment prepared for DIR 012/2002 concluded that lepidopteran herbivory is not an important limiting factor in determining cotton distribution in southern Australia compared to other environmental factors. The major impact of herbivory is on cotton production as a result of insect damage to the plants and cotton bolls.

415. The herbicide tolerance and insecticidal genes operate through independent, unrelated biochemical mechanisms. There is no evidence of any interaction between the cry genes and the cp4 epsps gene, their proteins or their metabolic pathways, and no reason to expect that this is likely to occur. There is no significant difference in the herbicide tolerance and insect resistance traits between GM cotton plants containing only one trait (Roundup Ready Flex® and Bollgard II®) and GM plants containing both traits (Roundup Ready Flex®/Bollgard II®) (Burns et al. 2004).

416. Therefore, the consequences of the expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing the spread and persistence of the GM cotton plants proposed for release through tolerance to glyphosate and reduced lepidopteran herbivory are assessed as minor. 2.3.2 Likelihood assessment

417. The proposed release would result in the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton. Monsanto anticipates a phased introduction of the GM cotton lines commencing with up to 20,000 hectares in spring 2006 in the current cotton growing areas of NSW and southern QLD. The area is expected to increase in subsequent years and may include plantings in other areas south of latitude 22° South that are suitable for growing cotton.

418. GM cotton plants containing a combination of a herbicide tolerance and an insect resistance trait are already being grown in the current cotton growing areas of NSW and southern QLD. The insect resistant/glyphosate tolerant Bollgard II®/Roundup Ready® GM cotton is widespread in the agricultural environment as a result of its commercial release in southern Australia (since 2002) and has not become a problematic weed.

419. Roundup Ready Flex®/Bollgard II® cotton plants differ from Bollgard II®/Roundup Ready® plants only in that glyphosate herbicide can be applied over the top of the plants to control weeds at later stages of growth (after the four-leaf stage) without yield loss. This is not expected to alter the on farm weediness potential of Roundup Ready Flex® cotton as glyphosate is not effective in controlling cotton plants beyond the seedling stage.

420. Some GM cotton seeds may spread from the release sites, germinate and persist in the environment following the release. However, the expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (eg limited availability of water, nutrients and suitable temperature conditions), where frost and water



availability in particular are known to be major limitations (see Chapter 1). The chance of volunteer GM plants establishing as weeds by finding suitable ecological niches would be no greater than for the non-GM parent organism. Therefore, the likelihood of weediness as a result of event 3 is assessed as highly unlikely.

2.4 Event 4: Dispersal of GM seed during transport or storage north of latitude 22º South.

421. The applicant is seeking approval for restricted transportation of ginned GM cotton seed from the release to areas north of latitude 22º South for use as stockfeed.

422. The risk of weediness of the GM cotton plants as a result of dispersal of seed during transport and storage in northern Australia would depend on the weediness of non-GM cotton plants, the importance of the use of glyphosate and lepidopteran herbivory in limiting the spread and persistence of cotton in this area (consequence assessment), the scale of the release/transport and the chance of progeny establishing as weeds (likelihood assessment). The risk is assessed against the baseline of the low weediness potential in the non-GM parent organism and in the context of the scale of the proposed release and the receiving environment (such as roadsides, feedlots, cattle yards or dairy farms). No other GM cotton lines are approved for unrestricted, commercial release north of latitude 22º South. 2.4.1 Consequence assessment

423. Dispersal or accidental spillage of GM seed during transport or storage in areas north of latitude 22º South may result in the establishment of GM cotton volunteers. This could allow the GM cotton plants to spread and persist in the environment where the GM cotton lines are not proposed for commercial release.

424. Although protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia (OGTR 2002a), current weed management practices would be able to control cotton volunteers.

425. The limited effectiveness of glyphosate in controlling cotton beyond the seedling stage (in southern Australia) is discussed in event 2 and is not expected to be different for areas north of latitude 22º South.

426. GM cotton volunteers may establish on roadsides where seed from the GM plants was transported. As cotton does not compete well with other plants and has high water and nutrient requirements (see Chapter 1), volunteer establishment is mainly expected in disturbed, favourable habitats such as ditches and roadside drains.

427. The options for controlling weeds along roadsides include: chemical control with herbicides, mechanical control (using cultivators, graders, excavators or chippers) and burning (Charles et al. 2002). Glyphosate may be used to control weeds on roadsides, but spraying is often limited to fixtures such as signs and guide-posts, while slashing is used in accessible areas (see event 2).

428. As mentioned in event 2, established or ratoon cotton plants, whether GM or non-GM, are difficult to control by herbicides alone. Therefore, if glyphosate were to be used to control weeds on roadsides, it would be ineffective on established cotton plants, whether GM or non-GM, and other means (ie mechanical methods) would need to be used to control it. Effective mechanical methods involve mulching, root cutting and cultivation (Roberts et al. 2002).

429. In addition, integrated weed management strategies stress the need to avoid relying on one control method (Roberts & Charles 2002). To avoid the development of glyphosate resistant weeds for example, application of glyphosate alone should not be used as the sole



management strategy. Alternation of various strategies used along roadsides would result in the destruction of glyphosate tolerant GM cotton volunteers.

430. GM cotton volunteers may also establish close to areas where GM cotton seed is stored. Areas where this could occur are limited and volunteers would be easily controlled (see above and event 2).

431. Therefore, the consequences of dispersal of GM seed during transport or storage north of latitude 22º South are assessed as minor. 2.4.2 Likelihood assessment

432. Some GM cotton seed may be dispersed in areas north of latitude 22º South during transport of GM cotton seed harvested in the south to areas north of latitude 22º South for stockfeed. The amount of cotton seed being transported would depend on the amount of cotton seed being used in stockfeed in the north each year. This can be highly variable. For example, the use of cotton seed as stockfeed increases significantly during drought.

433. The applicant proposes to apply the industry standard of transporting ginned cotton seed in covered containers/vehicles. Therefore, any spillage of seed during transport to areas north of latitude 22º South would be rare and any incident involving spillage of GM seed is expected to be readily controlled through cleaning of the site of the spill.

434. Some dispersal of GM cotton seed north of latitude 22º South may also occur in areas where cotton seed is stored. Seed is stored on farms in various ways (eg in sheds) that maintain its quality and protect it from animals and weathering. Dispersal of GM cotton seed during storage is expected to be restricted to areas immediately surrounding these storage areas.

435. The insect resistant Bollgard II® GM cotton and the glyphosate tolerant Roundup Ready® GM cotton are already widespread in the agricultural environment as a result of their commercial release (since 2002 and 2000, respectively) in southern Australia. In the 2004–05 season, Bollgard II® (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 70% of commercially grown cotton. Roundup Ready® cotton (also including the stacked variety Bollgard II®/Roundup Ready®) comprised about 50%. Ginned seed from these GM cotton lines has also been transported to areas north of latitude 22º South for stockfeed since their commercial release. Despite this, these GM cotton lines have not become problematic weeds.

436. The opportunity for any adverse outcome from any dispersal of seed is further diminished by the need for appropriate environmental conditions for germination, survival and persistence of any seed.

437. A survey of the transport routes between Emerald (in the cotton growing region in central Queensland) and the Atherton Tablelands (north of latitude 22º South in Queensland), conducted in 2002, indicated that cotton plants had established in the roadside environment only infrequently, despite 12 years of use of these routes for transporting ginned seed (including GM cotton varieties since their respective commercial releases) for stockfeed (Farrell & Roberts 2002). Only four plants were observed in 1200 km of road surveyed north of latitude 22º South. The study concluded that cotton volunteers tend to establish in highly and regularly disturbed environments and appear to have negligible ability to invade non-disturbed habitats (eg native bush). The following factors that limit survival of cotton volunteers in the roadside environment were identified: competition from already established vegetation, low quantity of seed escapes, high disturbance in areas requiring frequent maintenance and high rate of seed desiccation.



438. Although some GM cotton plants may establish along transport routes or around storage areas, the expression of the cp4 epsps gene construct, or the cp4 epsps, cry1Ac and cry2Ab genes in combination, is not expected to alter susceptibility to the environmental conditions that limit the establishment and persistence of cotton in areas north of latitude 22° South (eg plant competition, fire, herbivory by non-lepidopteran insects and variable availability of water and nutrients). Protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia (OGTR 2002a), but cotton volunteers around storage areas are easily controlled and those on roadsides are generally controlled by roadside management practices.

439. Therefore, the likelihood of weediness as a result of event 4 is assessed as unlikely.

2.5 Event 5: Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South.

440. The applicant proposes to use seed from the GM cotton lines in stockfeed anywhere in Australia. In addition to seed dispersal during feeding, a small percentage of cotton seed consumed by stock can pass through the digestive system intact and is able to germinate (Eastick 2002). As a result, cotton volunteers could establish in areas where livestock is fed cotton seed, or grazes after being fed, in areas north of latitude 22º South.

441. The risk of weediness of the GM cotton plants as a result of dispersal of seed via use of seed as stockfeed in northern Australia would depend on the weediness of non-GM cotton plants, the importance of the use of glyphosate and lepidopteran herbivory in limiting the spread and persistence of cotton in this area (consequence assessment), the scale of the release/use of seed as stockfeed and the chance of progeny establishing as weeds (likelihood assessment). The risk is assessed against the baseline of the low weediness potential in the non-GM parent organism and in the context of the scale of the proposed release and the receiving environment (such as feedlots, cattle yards or dairy farms). 2.5.1 Consequence assessment

442. Volunteers occurring as a result of the feeding of whole cotton seed to stock could allow GM cotton plants to spread and persist in the environment in areas north of latitude 22º South where the GM cotton lines are not proposed for commercial release.

443. GM cotton volunteers may establish in areas where stock is fed cotton seed (eg feedlots, cattle yards or dairy farms) or where stock grazes after being fed cotton seed. Areas where stock is fed are nutrient rich, disturbed habitats and cotton volunteers are expected to establish. However, animal trampling and grazing, as well as plant competition, are expected to limit spread and persistence of cotton plants in these areas.

444. The limited effectiveness of glyphosate in controlling cotton beyond the seedling stage (in southern Australia) is discussed in event 2 and is not expected to be different for areas north of latitude 22º South. Glyphosate tolerant GM cotton volunteers can be easily destroyed by mechanical means or, if still at the seedling stage, with alternative herbicides (see Section 2.2.1).

445. Although protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia (OGTR 2002a), current weed management practices would be able to control volunteers.

446. Therefore, the consequences of dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South are assessed as minor.



2.5.2 Likelihood assessment

447. The amount of cotton seed being used in stockfeed each year, including in areas north of latitude 22º South, can be highly variable. For example, its use increases significantly during drought. However, the quantity of cotton seed used is generally limited to a relatively small proportion of the diet, and must be introduced gradually, to avoid potential toxic effects due to the presence of anti-nutrients (ie gossypol and cyclopropenoid fatty acids) that are normally present in cotton.

448. Roundup Ready® and Roundup Ready®/INGARD® cotton have been in commercial cultivation since 2000 (DIR 023/2002), and Bollgard II® and Bollgard II®/Roundup Ready® cotton since 2002 (DIR 012/2002). Since their commercial release, cotton seed from these GM cotton lines has been used as stockfeed in Australia, including in northern Australia. Over this period there has been no evidence that these GM cotton lines have become problematic weeds.

449. Farrell and Roberts (2002) found cotton volunteers at seven of nine dairy farms surveyed (Atherton Tablelands, March 2002), with GM cotton (Roundup Ready®, Roundup Ready®/INGARD® or INGARD® cotton) identified on four of these. Volunteers were all close to dairy infrastructure, suggesting that their ability to invade is negligible. Such volunteers generally do not complete an entire reproductive cycle to produce new seedlings, being limited by physical damage (eg trampling and grazing), disease and competition, and do not spread into other areas of the farms or natural environment or lead to the development of self-sustaining populations. On farms where both GM and non-GM volunteers were found, there was no indication that the GM plants had enhanced survivorship or reproductive potential in any situation.

450. Eastick (2002) found that although cotton growing in cattle yards may reach reproductive maturity, persistence and seed dispersal from these areas is limited by trampling and grazing, and no cotton volunteers were found in undisturbed bush habitats.

451. Results from a survey conducted over the 2002–03 cotton growing season (as part of research required under licences for DIR 023/2002 and DIR 022/2002) on the incidence of cotton volunteers in areas in northern Queensland where stock are fed cotton seed, or graze after being fed cotton seed, indicate that very little cotton seed is used as stockfeed. Where it has been used, the incidence of cotton volunteers was never observed to be problematic, and volunteer plants never reached flowering or maturity. Cotton seed had not been used for stockfeed in Northern Territory and northern Western Australia and these areas were therefore excluded from this survey.

452. Although some GM cotton plants may establish where stock is fed cotton seed or where stock grazes after being fed cotton seed, the expression of the cp4 epsps gene construct is not expected to alter susceptibility to the environmental conditions that limit the establishment and persistence of cotton in areas north of latitude 22° South (eg plant competition, fire, herbivory by non-lepidopteran insects and variable availability of water and nutrients). The chance of volunteer GM plants establishing as weeds by finding suitable ecological niches would be no greater than for the non-GM parent organism.

453. Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination may alter the environmental limitations on the establishment and persistence of cotton in areas north of latitude 22° South as protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia (OGTR 2002a). However, animal trampling and grazing, as well as plant competition, are expected to limit spread and persistence of cotton plants in these situations.




2.6 Event 6: Dispersal of GM seed via flooding north of latitude 22º South. 455. The risk of weediness of the GM cotton plants as a result of dispersal of seed via flooding north of latitude 22º South would depend on the weediness of non-GM cotton plants, the importance of the use of glyphosate and lepidopteran herbivory in limiting the spread and persistence of cotton in these areas (consequence assessment), the scale of the release, the chance of flooding occurring and the chance of progeny establishing as weeds (likelihood assessment). The risk is assessed against the baseline of the low weediness potential in the non-GM parent organism and in the context of the scale of the proposed release and the receiving environment. 2.6.1 Consequence assessment

456. Some seed from the GM cotton plants may be dispersed from areas used for stockfeed and storage of GM cotton seed during flooding north of latitude 22º South. Seed may also be washed into drains, creeks and rivers close by. As a result, cotton volunteers may establish near areas used for stockfeed and storage, and along waterways close by.

457. The limited effectiveness of glyphosate in controlling cotton beyond the seedling stage (in southern Australia) is discussed in event 2 and is not expected to be different for areas north of latitude 22º South.

458. Although protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia (OGTR 2002a), weed management practices (eg mechanical means) would be able to control cotton volunteers near areas used for stockfeed and storage, and along waterways.

459. Therefore, the consequences of dispersal of GM seed via flooding north of latitude 22º South are assessed as marginal. 2.6.2 Likelihood assessment

460. The amount of GM cotton seed that may be dispersed during flooding north of latitude 22º South would be very limited as the only proposed dealing with the GM cotton lines north of latitude 22º South is the use of seed as stockfeed.

461. Submissions received during consultations on the Risk Assessment and Risk Management Plan prepared for DIR 012/2002 (commercial release of Bollgard II® and Roundup Ready/Bollgard II® cotton in the south) suggested that as some significant river systems in Queensland are located between 20º and 22º South and flow north-west from that area, there was potential for viable GM material to be dispersed into an area where only limited and controlled release will be permitted. Based on these concerns, the Risk Assessment and Risk Management Plan was revised to limit the commercial release of the GM cotton lines to south of latitude 22º South, rather than to south of latitude 20º South as proposed in the original DIR 012/2002 application and its Risk Assessment and Risk Management Plan. As the current application proposes commercial plantings only in areas south of latitude 22º South, no significant dispersal of GM cotton seed from the proposed commercial plantings south of latitude 22º South to areas north of latitude 22º South is expected during flooding.

462. Roundup Ready® and Roundup Ready®/INGARD® cotton have been in commercial cultivation since 2000 (DIR 023/2002), and Bollgard II® and Bollgard II®/Roundup Ready® cotton lines since 2002 (DIR 012/2002). Since their commercial release, cotton seed from these GM cotton lines has been used as stockfeed in northern Australia and may have been



dispersed via flooding. Over this period there has been no evidence that these GM cotton lines have become problematic weeds.

463. Although habitats close to waterways may be favourable for cotton establishment, tolerance to glyphosate is not expected to provide a significant selective advantage, compared to non-GM cotton, in these environments as glyphosate is not expected to be used. Protection from lepidopteran herbivory (in Roundup Ready Flex®/Bollgard II® cotton) may provide some selective advantage in northern Australia (OGTR 2002a). However, other environmental factors (eg plant competition, fire or herbivory by non-lepidopteran insects) are expected to limit the establishment and persistence of cotton plants in areas where seed may be dispersed to during flooding.


2.7 Event 7: Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance

465. The risk of weediness as a result of transfer of the cp4 epsps gene construct to G. hirsutum or G. barbadense plants would depend on the importance of the use of glyphosate in limiting the spread and persistence of cotton (consequence assessment), the chance of gene transfer occurring and the chance of progeny establishing as weeds following gene transfer (likelihood assessment). The risk is assessed in the context of the large scale of the proposed release south of latitude 22º South, the distribution of other cultivated or naturalised cotton plants growing in the vicinity of the crops and the conditions necessary for cross-pollination. 2.7.1 Consequence assessment

466. Transfer of the introduced cp4 epsps gene construct to other sexually compatible plants such as G. hirsutum and G. barbadense could result in the expression of the CP4 EPSPS herbicide tolerance protein in these plants. These plants could be GM cotton plants previously approved for commercial release or non-GM cotton plants. The cp4 epsps gene construct could confer some selective advantage as discussed in Section 2.2.1. This could result in spread and persistence of these cotton plants in an environment where the use of glyphosate is the major constraint on cotton survival.

467. The following GM cotton lines are currently approved for commercial release south of latitude 22° South in Australia:

• INGARD® (DIR 022/2002)

• Roundup Ready®/INGARD® (DIR 023/2002)

• Roundup Ready® (DIR 023/2002)

• Bollgard II® (DIR 012/2002)

• Bollgard II®/Roundup Ready® (DIR 012/2002).

468. In 2004, INGARD® cotton was withdrawn from the market in favour of Bollgard II® cotton and therefore only three GM cotton lines (Roundup Ready®, Bollgard II® and Bollgard II®/Roundup Ready®) are currently grown commercially.

469. Transfer of the cp4 epsps gene construct to Bollgard II® plants would result in the same combination of introduced genes as in Roundup Ready Flex®/Bollgard II® cotton. The risk of weediness of this combination is assessed in event 3.



470. Transfer of the cp4 epsps gene construct to Roundup Ready® (containing one copy of the cp4 epsps gene) or Bollgard II®/Roundup Ready® (containing one copy each of the cp4 epsps, cry1Ac and cry2Ab genes) plants would result in plants containing three copies of the cp4 epsps gene. This may result either in increased or decreased (due to gene silencing) expression of the CP4 EPSPS protein. Decreased expression would reduce the risk of weediness as compared to the GM cotton lines proposed for release. Plants with increased expression may be tolerant to higher concentrations of glyphosate. The limited effectiveness of glyphosate in controlling cotton beyond the seedling stage is discussed in event 2 and would be no different for plants with increased herbicide tolerance.

471. In the long-term, Roundup Ready Flex® cotton varieties are expected to completely replace Roundup Ready® varieties in the commercial production and therefore the occurrence of transfer of the cp4 epsps gene construct from Roundup Ready Flex® cotton to Roundup Ready® cotton varieties would be minimal.

472. Transfer of the cp4 epsps gene construct to other cotton plants is not expected to alter the fact that cotton is not a serious weed in southern Australia due to the limited availability of water, nutrients and suitable temperature conditions. The expression of the cp4 epsps gene construct is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia, where frost and water availability in particular are known to be major limitations (see Chapter 1). Therefore, the consequences of expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance are assessed as minor. 2.7.2 Likelihood assessment

473. The adverse outcome of weediness resulting from an increase in the spread and persistence of other cotton plants is contingent on both of the following steps:

• transfer of the introduced cp4 epsps gene construct to other cotton plants

• weediness of the recipient plants as a result of expression of the introduced gene.

474. Cotton is primarily self-pollinating with pollen that is large, sticky and heavy and not easily dispersed by wind (Jenkins 1992; OGTR 2002b). Overseas studies have shown that insect pollinators can transfer pollen to other nearby cotton plants at rates up to 80% (eg Oosterhuis & Jernstedt 1999). However, cotton pollen dispersal studies conducted in Australia consistently show that outcrossing is localised around the pollen source and decreases significantly with distance (OGTR 2002b and references therein ). For vertical gene transfer to occur, the GM cotton lines would need to be planted within pollination distance of other G. hirsutum or G. barbadense plants.

475. Due to the greatly reduced requirement for insecticidal sprays in Bollgard II® cotton, insect pollinators may become more abundant, which may increase the incidence of gene flow.

476. G. barbadense is the closest relative of G. hirsutum occurring in Australia (OGTR 2002b). It is commercially grown on a small scale in Australia. Hybridisation can occur naturally between these two species (Brubaker et al. 1999). Hybrid progeny exhibit characteristics intermediate to the parents but typically with a lower capacity to produce cotton bolls. G. barbadense and hybrids are not weedier or more difficult to control than G. hirsutum (Warwick Stiller & Greg Constable, CSIRO, pers. comm.).

477. Transfer of the cp4 epsps gene construct to naturalised cotton populations could also occur. It is not known if the herbarium records for G. hirsutum and G. barbadense indicate



current naturalised populations of these plants. A comparison of shires where cotton is cultivated and shires where feral cotton populations occur (in Queensland) indicates that feral populations occur in only three cotton production shires, and one shire adjacent to a cotton production shire. Where significant geographic distances between naturalised populations and the cotton growing regions of NSW and Queensland exist, this will decrease the frequency of gene transfers.

478. The cp4 epsps gene is already available for outcrossing from commercially released Roundup Ready® cotton. In the 2004–05 season, Roundup Ready® cotton (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 50% of commercially grown cotton in Australia. The only difference is that two copies of the cp4 epsps gene would be available for gene transfer from the GM cotton lines proposed for release.

479. The proposed release would result in the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton, which would increase the occurrence of gene transfer events. However, cotton is primarily in-breeding and gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies. Following transfer of the cp4 epsps gene construct to any of these cotton plants, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants (see event 2 and previous Section). Therefore, the likelihood of weediness as a result of event 7 is assessed as highly unlikely.

2.8 Event 8: Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory.

480. The risk of weediness as a result of transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to G. hirsutum or G. barbadense plants would depend on the importance of the use of glyphosate and lepidopteran herbivory in limiting the spread and persistence of cotton (consequence assessment), the chance of gene transfer occurring and the chance of progeny establishing as weeds following gene transfer (likelihood assessment). The risk is assessed in the context of the large scale of the proposed release south of latitude 22º South, the distribution of other cultivated or naturalised cotton plants growing in the vicinity of the crops and the conditions necessary for cross-pollination.

481. Detailed discussions of the risk from expression of the CP4 EPSPS herbicide tolerance protein in combination with the two insect resistance proteins (Cry1Ac and Cry2Ab) increasing the weediness of cotton plants in Australia is provided in the risk assessment document for DIR 012/2002 (Bollgard II®/Roundup Ready®) available at <http://www.ogtr.gov.au>. This risk assessment concluded that expression of the proteins in combination does not enhance the weediness potential of these GM cotton plants (in comparison to non-GM cotton plants) in the cotton growing regions of Australia south of latitude 22° South.

482. While the expression of one copy of the cp4 epsps gene resulted in Roundup Ready® cotton being tolerant to glyphosate only up to the four-leaf stage of growth, this assessment takes into account that Roundup Ready Flex® cotton (including Roundup Ready Flex®/ Bollgard II®) with two copies of the cp4 epsps gene is tolerant to glyphosate throughout the growing season. 2.8.1 Consequence assessment

483. Transfer of the introduced cp4 epsps, cry1Ac and cry2Ab genes in combination (present in Roundup Ready Flex®/Bollgard II® cotton) to other sexually compatible plants such as



G. hirsutum and G. barbadense could result in the expression of the CP4 EPSPS, Cry1Ac and Cry2Ab proteins in these plants. These plants could be GM cotton plants previously approved for commercial release or non-GM cotton plants. The cp4 epsps, cry1Ac and cry2Ab genes in combination could confer some selective advantage as discussed in Section 2.3.1. This could result in spread and persistence of these cotton plants in environments where the use of glyphosate and lepidopteran herbivory are the major constraints on cotton survival.

484. Currently Roundup Ready®, Bollgard II® and Bollgard II®/Roundup Ready® cotton lines are approved for commercial release south of latitude 22º South and for limited and controlled release north of latitude 22º South (for details see event 7).

485. Roundup Ready Flex®/Bollgard II® cotton was produced by crossing Roundup Ready Flex® with Bollgard II® cotton. Therefore, crossing of Roundup Ready Flex®/Bollgard II® plants with Bollgard II® plants would not result in a new combination of introduced genes (as Bollgard II® cotton already contains the cry1Ac and cry2Ab genes at the same location of the genome).

486. Crossing of Roundup Ready Flex®/Bollgard II® plants with Roundup Ready® (containing one copy of the cp4 epsps gene) or Bollgard II®/Roundup Ready® (containing one copy each of the cp4 epsps, cry1Ac and cry2Ab genes) plants would result in plants containing three copies of the cp4 epsps gene. The consequence of this is assessed in event 7.

487. Transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to other cotton plants is not expected to alter the fact that cotton is not a serious weed in southern Australia due to the limited availability of water, nutrients and suitable temperature conditions. As discussed in event 3, insect herbivory is not an important limiting factor. The expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia, where water availability in particular is known to be a major limitation (see Chapter 1). As discussed in event 3, insect herbivory is not an important limiting factor in southern Australia. Therefore, the consequences of expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory are assessed as minor. 2.8.2 Likelihood assessment

488. The adverse outcome of weediness resulting from an increase in the spread and persistence of other cotton plants is contingent on both of the following steps:

• transfer of the introduced cp4 epsps, cry1Ac and cry2Ab genes to other cotton plants

• weediness of the recipient plants as a result of expression of the introduced genes.

489. As discussed in event 7, gene transfer to other G. hirsutum or G. barbadense plants is expected to occur at low levels.

490. Transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to naturalised cotton populations could also occur. The likelihood of this occurring is discussed in event 7.

491. The cp4 epsps, cry1Ac and cry2Ab genes are already available for outcrossing from commercially released Roundup Ready® (containing the cp4 epsps gene) and Bollgard II® (containing the cry1Ac and cry2Ab genes) cotton. In the 2004–05 season, Bollgard II® (including the stacked variety Bollgard II®/Roundup Ready®) comprised about 70% of commercially grown cotton. Roundup Ready® cotton (also including the stacked variety Bollgard II®/Roundup Ready®) comprised about 50%.The only difference is that two copies



of the cp4 epsps gene would be available for gene transfer from Roundup Ready Flex®/Bollgard II® cotton proposed for release.

492. The proposed release would result in the extensive cultivation of GM cotton plants in areas south of latitude 22º South that are suitable for growing cotton, which would increase the occurrence of gene transfer events. However, cotton is primarily in-breeding and gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies. Following transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to any of these cotton plants, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants. Therefore, the likelihood of weediness as a result of event 8 is assessed as highly unlikely.

2.9 Uncertainty 493. There is limited data on the existence and proximity of naturalised cotton (Gossypium hirsutum or G. barbadense) populations to commercial cotton growing regions. This information would be useful for determining the opportunity for, and possible frequency of, gene transfer between GM cotton crops and these populations. However, this information is not required for assessing the risk of this proposed release of herbicide tolerant and herbicide tolerant/insect resistant GM cotton lines as the consequence and likelihood of such gene transfer causing weediness in these plants are expected to be the same as for the GM cotton plants proposed for release.

494. There is also some uncertainty about the weediness potential of insect resistant GM cotton plants in areas north of latitude 22º South. The risk assessment document for DIR 012/2002 (commercial release of Bollgard II® and Bollgard II®/Roundup Ready® cotton) available at <http://www.ogtr.gov.au> concluded that protection from lepidopteran insect herbivory may provide some selective advantage in northern Australia. This uncertainty has been taken into account in the consequence and likelihood assessments relating to dealings with the GM cotton lines in northern Australia and some risk treatment measures are proposed (see Chapter 5).

Section 3 Risk estimates 495. Risk estimates (which can range from negligible to high) are based on a combination of the consequences and likelihood assessments, using the Risk Estimate Matrix (see Chapter 2).

496. The risk estimates for the adverse outcome of weediness of the GM cotton lines as a result of the expression of the cp4 epsps gene construct, or the cp4 epsps, cry1Ac and cry2Ab genes in combination, have been made relative to the baselines of the weediness of non-GM cotton growing in Australia and the current widespread cultivation and use of Roundup Ready® (containing the one copy of the cp4 epsps gene), Bollgard II® (containing the cry1Ac and cry2Ab genes), and a conventional cross of these GM cotton lines in commercial cotton crops in southern Australia.

497. The consequences of increased spread and persistence of cotton resulting from the presence of the cp4 epsps gene in the GM cotton lines (event 2) have been assessed as minor, and the likelihood of this resulting in weediness as highly unlikely. Therefore the risk estimate is negligible.

498. The consequences of increased spread and persistence of cotton resulting from the presence of the cp4 epsps, cry1Ac and cry2Ab genes in combination in one of the GM cotton lines (event 3) have been assessed as minor, and the likelihood of this resulting in weediness as highly unlikely. Therefore the risk estimate is negligible.



499. The consequences of increased spread and persistence resulting from dispersal of GM seed from both lines during restricted transport or storage north of latitude 22º South (event 4) has been assessed as minor, and the likelihood of this resulting in weediness as unlikely. Therefore the risk estimate is low.

500. The consequences of increased spread and persistence resulting from dispersal of seed from both lines via use of GM cotton seed as stockfeed in areas north of latitude 22º South (event 5) have been assessed as minor, and the likelihood of this resulting in weediness as unlikely. Therefore the risk estimate is low.

501. The consequences of increased spread and persistence resulting from the dispersal of GM seed from both lines via flooding north of latitude 22º South (event 6) have been assessed as marginal, and the likelihood of this resulting in weediness as unlikely. Therefore the risk estimate is negligible.

502. The consequences of increased spread and persistence resulting from the presence of the cp4 epsps gene construct in other G. hirsutum and G. barbadense cotton plants, as a result of gene transfer (event 7), have been assessed as minor, and the likelihood of this resulting in weediness as highly unlikely. Therefore the risk estimate is negligible.

503. The consequences of increased spread and persistence resulting from the presence of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum and G. barbadense cotton plants, as a result of gene transfer (event 8), have been assessed as minor, and the likelihood of this resulting in weediness as highly unlikely. Therefore the risk estimate is negligible.

504. The risks of five events that may lead to weediness are estimated to be negligible. As the risks of two events that may lead to weediness are estimated to be low, there may be some need to invoke actions for mitigation beyond normal practices (OGTR 2005). Therefore, some risk treatment measures for weediness were proposed (see Chapter 5).

Table 4.2 Summary of risk assessment Event that may give rise to weediness

Consequence assessment Likelihood assessment Risk estimate

Does risk require

treatment? Event 2 Expression of the cp4 epsps gene construct increasing spread and persistence of the GM cotton plants through tolerance to glyphosate

Minor • Cotton is not a serious weed

in southern Australia because of the limited availability of water, nutrients and suitable temperature conditions.

• Although glyphosate is the most widely used herbicide in Australia today, it is not generally used to control established cotton plants as the herbicide is not effective on cotton beyond the seedling stage (plants are damaged but not killed).

• Glyphosate tolerant cotton volunteers are effectively controlled by mechanical means or, if still at the seedling stage, by the use of

Highly unlikely • The proposed release would result in

the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton.

• Similar commercially approved glyphosate tolerant cotton lines are already extensively cultivated and transported and have not become problematic weeds.

• Expression of the cp4 epsps gene construct is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

• The chance of volunteer GM plants establishing as weeds by finding suitable ecological niches would be no

Negligible No



Event that may give rise to weediness


Does risk require

treatment? alternative herbicides. greater than for the non-GM parent.

Event 3 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination increasing spread and persistence of the GM cotton plants through tolerance to glyphosate and reduced lepidopteran herbivory

Minor • Cotton is not a serious weed

in southern Australia because of the limited availability of water, nutrients and suitable temperature conditions.

• Lepidopteran herbivory is not an important limiting factor on the spread and persistence of cotton in southern Australia.

• The herbicide tolerance and insecticidal genes operate through independent, unrelated biochemical mechanisms and there is no evidence of any interaction.

Highly unlikely • The proposed release would result in

the extensive cultivation of GM cotton plants in current and potential areas south of latitude 22º South that are suitable for growing cotton.

• Similar commercially approved GM cotton plants containing the cp4 epsps, cry1Ac and cry2Ab genes in combination are already extensively cultivated and transported and have not become problematic weeds.

• Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).


Negligible No

Event 4 Dispersal of GM seed during transport or storage north of latitude 22º South

Minor • Although protection from

lepidopteran insect herbivory may provide some selective advantage in northern Australia, current weed management practices would be able to control cotton volunteers around storage areas and on roadsides.

• The use of glyphosate to control weeds on roadsides would be ineffective on established cotton plants (whether GM or non-GM) and other means are required to control these plants.

• Glyphosate tolerant cotton volunteers are susceptible to other forms of chemical and mechanical control.

Unlikely • The applicant intends to apply the

industry standard of transporting ginned cotton seed in covered containers/vehicles.

• Seed from similar GM cotton plants containing the same introduced genes have been similarly transported to areas north of latitude 22º South for stockfeed since 2000 and have not become problematic weeds.

• Any incident involving spillage of GM cotton seed in northern Australia can be readily controlled through cleaning of the site of the spill.

• Expression of the introduced genes is not expected to alter susceptibility to the environmental conditions that limit the establishment and persistence of cotton in northern Australia (eg plant competition, fire, herbivory by non-lepidopteran insects and variable availability of water and nutrients).

• Surveys of transport routes indicate that survival of cotton volunteers is limited to disturbed environments.

Low Yes





Does risk require

treatment? • Cotton volunteers around storage

areas are easily controlled and those on roadsides are generally controlled by plant competition and roadside management practices (slashing).

Event 5 Dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South

Minor • Protection from lepidopteran

insect herbivory may provide some selective advantage in northern Australia.


Unlikely • Seed from similar GM cotton lines

containing the same introduced genes has been used as stockfeed in northern Australia since 2000 and these GM cotton lines have not become problematic weeds.

• Surveys of the incidence of cotton volunteers in areas where stock is fed cotton seed indicate that cotton volunteers are not problematic.

• Animal trampling and grazing are known to limit spread and persistence of cotton plants in areas where stock is fed cotton seed or where stock grazes after being fed cotton seed.

Low Yes

Event 6 Dispersal of GM seed via flooding north of latitude 22º South

Marginal • Protection from lepidopteran

insect herbivory may provide some selective advantage in northern Australia.


Unlikely • Similar GM cotton lines containing the

same introduced genes have been commercially grown and transported in southern Australia, and their seed used as stockfeed in northern Australia, since 2000 and have not become problematic weeds.

• Although habitats close to waterways may be favourable for cotton establishment, expression of the introduced genes is not expected to alter susceptibility to the environmental factors (eg plant competition, fire or herbivory by non-lepidopteran insects) that will limit the establishment and persistence of cotton plants in these or other areas where seed may be dispersed to during flooding.

Negligible No

Event 7 Expression of the cp4 epsps gene construct in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance

Minor • Although transfer of the

cp4 epsps gene construct to similar commercially released GM cotton plants could result in plants containing three copies of the cp4 epsps gene, this would not alter the limited effectiveness of glyphosate in controlling cotton plants beyond the seedling stage.

• The expression of the cp4 epsps gene construct is

Highly unlikely • Cotton is primarily in-breeding and

gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies.

• If transfer of the cp4 epsps gene construct to other cotton plants occurred, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants.

Negligible No





Does risk require

treatment? not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

Event 8 Expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination in other G. hirsutum or G. barbadense cotton plants (including commercially released GM cotton lines) providing glyphosate tolerance and reducing lepidopteran herbivory

Minor • Crossing of Roundup Ready

Flex®/Bollgard II® plants with similar commercially released GM cotton plants could result in plants containing three copies of the cp4 epsps gene (assessed in event 7).

• The expression of the cp4 epsps, cry1Ac and cry2Ab genes in combination is not expected to alter susceptibility to the environmental conditions that limit the spread and persistence of cotton in southern Australia (particularly limited water availability and frost).

Highly unlikely • Cotton is primarily in-breeding and

gene transfer to other cotton plants is expected to occur in close proximity and at low frequencies.

• If transfer of the cp4 epsps, cry1Ac and cry2Ab genes in combination to other cotton plants occurred, the likelihood of it causing weediness in these plants is expected to be the same as for the GM cotton plants.

Negligible No



Chapter 5 Risk management 505. This Chapter evaluates the risks assessed in Chapters 3 and 4 to determine whether or not specific treatments are required to mitigate harm that may arise during the proposed release. Other risk management considerations required under the Act are also addressed in this Chapter.

Section 1 Background 506. Under section 56 of the Act, the Regulator must not issue a licence unless satisfied that any risks posed by the dealings proposed to be authorised by the licence are able to be managed in a way that protects the health and safety of people and the environment.

507. Under section 62 of the Act, the Regulator can direct a licence holder to take any steps the Regulator deems necessary to protect the health and safety of people or the environment. Licence conditions can be imposed to limit and control the scope of the dealings and the possession, supply, use, transport or disposal of the GMO for the purposes of, or in the course of, a dealing. In addition, the Regulator has extensive powers to monitor compliance with licence conditions under section 152 of the Act.

Section 2 Other Australian regulators 508. Australia’s gene technology regulatory system operates as part of an integrated legislative framework (OGTR 2005). Other agencies that also regulate GMOs or GM products include FSANZ, APVMA, TGA, NICNAS, NHMRC and AQIS. Dealings conducted under any licence issued by the Regulator may also be subject to regulation by one or more of these agencies.

509. The Gene Technology Act 2000 requires the Regulator to consult these agencies during the assessment of DIR applications. The Gene Technology (Consequential Amendments) Act 2000 requires the agencies to consult the Regulator for the purposes of making certain decisions regarding their assessments of products that are, or contain a product from, a GMO.

510. FSANZ is responsible for human food safety assessment and has recently approved an application from Monsanto to approve food (oil and linters) derived from Roundup Ready Flex® cotton in Australia (FSANZ 2005). Oil and linters from Roundup Ready® cotton and Bollgard II® cotton, have previously been approved for this purpose (ANZFA 2000a; ANZFA 2002c).

511. One of the GM cotton lines proposed for release (Roundup Ready Flex®/Bollgard II®) meets the definition of an agricultural chemical product under the Agricultural and Veterinary Chemicals Code Act 1994, due to its production of two insecticidal substances, and therefore it is subject to regulation by the APVMA. In July 2003, the APVMA registered the use of the insecticidal proteins as produced by the cry1Ac and cry2Ab genes in GM Bollgard II® cotton (covering also stacked varieties) as insecticidal products. An insect resistance management plan for Bollgard II® cotton has been developed by the Transgenic and Insect Management Strategy (TIMS) committee of the Australian Cotton Growers' Research Association in consultation with the APVMA (Monsanto Australia Limited 2004). The APVMA requires implementation of this plan as a condition of registration. Implementation of this insect resistance management plan would also be required for Roundup Ready Flex®/Bollgard II® cotton (refer Section 2.10.1 for detailed discussion).

512. The use of Roundup Ready® Herbicide on the GM cotton lines proposed for release is also subject to regulation by the APVMA. Roundup Ready® Herbicide is currently registered

Chapter 5 Risk management (February 2006) 95


for the use on Roundup Ready® cotton up to the four-leaf stage of growth. Monsanto requires APVMA approval of its application to vary the registration to allow the herbicide to be applied to Roundup Ready Flex® lines (including application after the four-leaf stage) to undertake the proposed commercial release of these GM cotton lines. As the APVMA generally imposes conditions on the use pattern of herbicides, there will be restrictions on the number of applications that can be made and on the spraying window (ie up to what stage of crop growth herbicide applications can be made) eg in order to limit resistance development and comply with residue limits (refer to 2.10.3 for detailed discussion).

513. The Regulator has liaised closely with FSANZ and the APVMA during the parallel assessment of applications pertaining to this commercial release of GM cotton lines.

Section 3 Risk treatment measures for identified risks 514. The detailed risk assessment of events 1–8 contained in Chapters 3 and 4 concluded that the risk estimates are negligible for six of the events and low for two of the events. These events were considered in the context of the large scale of the proposed release and the receiving environment for this proposed release, including other commercially approved GM cotton lines.

515. The Risk Analysis Framework (OGTR 2005), which guides the risk assessment and risk management process, defines negligible risks as insubstantial with no present need to invoke actions for their mitigation. Low risks are defined as minimal but may invoke actions for mitigation beyond normal practices.

516. The risks of the following two events that may lead to weediness were estimated to be low:

• dispersal of GM seed during transport or storage north of latitude 22º South

• dispersal of seed via use of GM cotton seed as stockfeed in areas north of latitude 22º South.

517. Risk treatment measures have been imposed to minimise dissemination of GM cotton seed, and spread and persistence of GM cotton plants in areas north of latitude 22º South.

518. In addition to requiring that transport of seed from the GM cotton lines to/within areas north of latitude 22º South for use as stockfeed occur in covered vehicles as proposed by Monsanto, the company is required to inform people who transport the GM seed of their responsibilities. These include the requirement to label the load of GM seed with contact phone numbers to call in case of spillage or misdirection. Similar licence conditions have been shown to be effective in relation to the commercial release of GM Roundup Ready® (DIR 023/2002) and Bollgard II® (DIR 012/2002) cotton.

519. Other measures to minimise the spread and persistence of the cotton lines in northern Australia have also been imposed. North of latitude 22º South, GM cotton seed is only allowed to be fed to stock inside stockyards, feedlots or dairies. The licence holder is required to provide information about cotton volunteers and their control to end-users of cotton seed north of latitude 22º South. A communication strategy is also required similar to that imposed under the licences for DIR 023/2002 and DIR 012/2002.

Section 4 General risk management 4.1 Other risk management considerations 520. All DIR licences issued by the Regulator contain a number of general conditions that relate to risk management. These include, for example:



• identification of the persons or classes of persons covered by the licence

• applicant suitability

• contingency and compliance plans

• reporting structures, including a requirement to inform the Regulator if the applicant becomes aware of any additional information about risks to the health and safety of people or the environment.

4.1.1 Applicant suitability

521. In making a decision whether or not to issue a licence, the Regulator must have regard to the suitability of the applicant to hold a licence. Under section 58 of the Act matters that the Regulator must take into account include:

• any relevant convictions of the applicant (both individuals and the body corporate)

• any revocation or suspension of a relevant licence or permit held by the applicant under a law of the Commonwealth, a State or a foreign country

• the applicant’s history of compliance with previous approved dealings

• the capacity of the applicant to meet the conditions of the licence.

522. Before making the decision to issue a licence for this application (DIR 059/2005), the Regulator determined that Monsanto Australia Ltd is suitable to hold a licence.

523. Conditions in the licence include a requirement for the licence holder to inform the Regulator of any circumstances that would affect their suitability or their capacity to meet the conditions of the licence.

524. In addition, any applicant organisation must have access to a properly constituted Institutional Biosafety Committee and be an accredited organisation under the Act. 4.1.2 Compliance plan

525. The licence requires Monsanto to submit a plan detailing how it intends to ensure compliance with the licence conditions and document that compliance. This plan is required before the planting of any of the GM cotton lines occurs.

526. Monsanto is also required to provide a method to the Regulator for the reliable detection of the presence of the GMOs and the introduced genetic material in a recipient organism. This instrument is required within 30 days of the issue date of the licence. 4.1.3 Reporting structures

527. The licence obliges the licence holder to immediately report any of the following to the Regulator:

• any additional information regarding risks to the health and safety of people or the environment associated with the release

• any contraventions of the licence by persons covered by the licence

• any unintended effects of the release.

528. The licence holder is also obliged to submit an Annual Report within 90 days of the anniversary of the licence containing any information required by the licence.



Section 5 Monitoring and Compliance 529. A range of monitoring and compliance activities are undertaken on behalf of the Regulator (OGTR 2005) to check compliance with licence conditions.

530. If monitoring activities identify changes in the risks associated with the authorised dealings, the Regulator may also vary licence conditions, or if necessary, suspend or cancel the licence.

531. In cases of non-compliance with licence conditions, the Regulator may also instigate an investigation to determine the nature and extent of non-compliance. The Act provides the Regulator with extensive powers of enforcement to ensure compliance. These include the provision for criminal sanctions of large fines and/or imprisonment for failing to abide by the legislation, conditions of the licence or directions from the Regulator, especially where significant damage to the health and safety of people or the environment could result.

Section 6 Conclusions of the RARMP 532. The risk assessment concludes that this commercial release of herbicide tolerant and herbicide tolerant/insect resistant GM cotton lines poses low to negligible risks to the health and safety of people and the environment as a result of gene technology.

533. The risk management plan concludes that the negligible risks do not require specific risk treatment measures. However, licence conditions have been imposed to treat the low risks relating to the proposed use of GM cotton seed as stockfeed in northern Australia. Therefore, licence conditions have been imposed to minimise spread and persistence of the GMOs in areas north of latitude 22º South.



Chapter 6 Licence conditions Section 1 Interpretations and Definitions This licence does not authorise dealings with GMOs that are otherwise prohibited as a result of the operation of State legislation declaring areas to be GM, GM free, or both, for marketing purposes. In this licence:

(a) unless defined otherwise in this licence, words and phrases used in this licence have the same meanings as they do in the Gene Technology Act 2000 (the Act) and the Gene Technology Regulations 2001;

(b) words importing a gender include any other gender; (c) words in the singular include the plural and words in the plural include the

singular; (d) words importing persons include a partnership and a body whether corporate or

otherwise; (e) references to any statute or other legislation (whether primary or subordinate) are

a reference to a statute or other legislation of the Commonwealth of Australia as amended or replaced from time to time and equivalent provisions, if any, in corresponding State law, unless the contrary intention appears;

(f) where any word or phrase is given a defined meaning, any other part of speech or other grammatical form in respect of that word or phrase has a corresponding meaning.

In this licence: ‘Annual Report’ means a written report provided to the Regulator within 90 days of each anniversary of the date of issue of this licence containing all the information required by this licence to be provided in the Annual Report. ‘Cotton’ means plants of the species Gossypium hirsutum L. ‘Covered Vehicles’ means vehicles that use tight fitting covers to prevent spillage or escape of the load during transporting (for example a trailer with sides moulded or sealed to the base fitted with a roll-over tarp). ‘Deal with’ in relation to a GMO means any one or more of the following as permitted by this licence,

(a) conduct experiments with the GMOs; (b) make, develop, produce or manufacture the GMOs; (c) breed the GMOs; (d) propagate the GMOs; (e) use the GMOs in the course of manufacture of a thing that is not the GMOs; (f) grow, raise or culture the GMOs; (g) import the GMOs;

and includes the possession, supply, use, transport or disposal of the GMO for the purposes of, or in the course of, a dealing mentioned in any of paragraphs (a) to (g).

‘GM’ means genetically modified. ‘GMOs’ means the genetically modified organisms listed in Attachment B and authorised for release by this licence.

Chapter 6 Licence conditions (February 2006) 99


‘OGTR’ means the Office of the Gene Technology Regulator. ‘Location’ means an area of land where the GMOs are planted and grown for the purposes of a licence. ‘Regulator’ means the Gene Technology Regulator. ‘Restricted Zone’ means north of latitude 22º South anywhere in Australia. ‘Technology Users Agreement’ means an agreement by which the licence holder licenses, authorises or otherwise permits the use of the GMO. ‘Volunteer plant’ means progeny of the GMO.



Section 2 Conditions Duration of licence

1. This licence remains in force until it is suspended, cancelled or surrendered. No dealings with the GMOs are authorised during any period of suspension.

Holder of licence

2. The holder of this licence (‘the licence holder’) is Monsanto Australia Limited. Project Supervisor

3. The Project Supervisor in respect of this licence is identified at Attachment A.

4. The licence holder must immediately notify the Regulator in writing if any of the contact details of the Project Supervisor change.

GMOs covered by this licence

5. The GMOs covered by this licence (‘the GMOs’) are identified and described at Attachment B.

Dealings authorised by this licence SOUTH of the Restricted Zone

6. Subject to condition 7, any person, including the licence holder, may conduct any Dealing south of the Restricted Zone.

7. Where the GMOs authorised by this licence are planted or in any other way Dealt with as part of a subsequent licence, or subsequent licence variation, authorising a dealing under the Gene Technology Act 2000, then, for purposes of the subsequent licence, or subsequent licence variation,

(a) only the persons covered by the subsequent licence or licence variation are permitted to grow or otherwise deal with the GMOs, and,

(b) the conditions of the subsequent licence or the licence containing the licence variation, and not the conditions of this licence, will apply to the dealing with the GMOs.

Example: If a subsequent licence contemplates the planting of these GMOs in a Location containing another GMO authorised by that later licence, the conditions of the subsequent licence and not the conditions of this licence will apply to the GMOs for purposes of the dealings conducted under that licence. Dealings authorised by this licence IN the Restricted Zone

8. Any person, including the licence holder, may, subject to the conditions of this licence,

(a) transport GM whole cotton seed into the Restricted Zone and/or

(b) use GM whole cotton seed for stock feed.

Note: This licence does not authorise any other dealing in the Restricted Zone. Dealing with the GMO in the Restricted Zone may however be permitted by another licence issued by the Regulator. Transport and use of GM whole cotton seed in the Restricted Zone

9. If cotton seed from the GMOs is used as stock feed in the Restricted Zone, feeding must take place inside stock yards, feedlots or dairies.



10. GM whole cotton seed derived from the GMOs must not be transported to the Restricted Zone other than in accordance with the conditions of this licence.

Note: This condition does not apply to the transport of GM whole cotton seed derived from the GMOs into the Restricted Zone if it is permitted by another licence.

11. Transporters of GM whole cotton seed to destinations within the Restricted Zone must:

(a) only transport the GM whole cotton seed in Covered Vehicles or in unbreakable sealed containers within a vehicle;

(b) sign Covered Vehicles, or unbreakable containers and transporting vehicle, to indicate that they contain GM whole cotton seed, and with instructions to contact the licence holder in the event that the GM whole cotton seed is spilt or misdirected, including telephone contact numbers.

12. The licence holder must prepare and distribute to cotton gins from which GM whole cotton seed will be transported into the Restricted Zone sufficient copies of the sign required by Condition 11. This sign must accompany each shipment of cotton seed into the Restricted Zone.

13. Cotton gins from which GM whole cotton seed is transported into the Restricted Zone must provide transporters with a sign to accompany every shipment of whole cotton seed into the Restricted Zone, and must maintain a record of this action.

14. The licence holder must, in consultation with the OGTR, develop a communication strategy, including documentation for distribution, to convey the importance of appropriate control of cotton volunteers to all recipients of GM whole cotton seed. A copy of the documentation must be provided in the Compliance Management Plan.

15. The licence holder must take all reasonable steps to distribute the documentation specified in condition 14 to:

(a) the cotton gins from which GM whole cotton seed is sourced for transport into the Restricted Zone;

(b) the transporters of GM whole cotton seed into the Restricted Zone; and

(c) all recipients of GM whole cotton seed within the Restricted Zone, including retailers and the end users of the GM whole cotton seed.

16. A copy of the documents required by conditions 14 must be provided in the Compliance Management Plan.

Terms of Technology Users Agreement

17. The licence holder must not enter into a Technology Users Agreement or any other agreement which would permit the GMOs to be grown or in any other way dealt with in the Restricted Zone other than in a manner permitted by the licence. A copy of the standard Technology Users Agreement must be included in the Compliance Management Report.

Note: This condition does not prevent the licence holder from entering into a Technology Users Agreement permitting the GMOs to be dealt with in the Restricted Zone if the dealing is permitted pursuant to another licence.



Informing people of their obligations

18. The licence holder must inform any person covered by this licence, to whom a particular condition of this licence applies, of the following:

(a) the particular condition (including any variations of it); (b) the cancellation or suspension of the licence; (c) the surrender of the licence.

19. Where the licence holder or an agent of the licence holder enters into a Technology Users Agreement or any other agreement to licence, authorise, or in any other way permit a person to use the GMOs, the licence holder, after having regard to the intended use of the GMO by the person, must,

notify the user of the GMO of the conditions which will apply to the user, and

explain those conditions, and

notify the user that breach of the conditions is an offence under the Gene Technology Act 2000

20. The licence holder must provide written notification to cotton gins from which GM whole cotton seed will be transported into the Restricted Zone which contains the following,

notice of the conditions which will apply to transporters of whole cotton seed into the Restricted Zone, and

an explanation of those conditions, and

a notice that breach of those conditions is an offence under the Gene Technology Act 2000 (‘the Act’), and

a request that the cotton gins convey the information contained in (a), (b) and (c) to transporters of GM whole cotton seed into the Restricted Zone.

21. Cotton gins from which GM whole cotton seed is transported into the Restricted Zone must comply with a request of the licence holder made under condition 20(d).

22. The licence holder must provide the Regulator, on the Regulator’s written request, evidence that the licence holder has informed people of the conditions of this licence that apply to them.

Applicant to notify of circumstances that might affect suitability

23. The licence holder must immediately, by notice in writing, inform the Regulator of:

(a) any relevant conviction of the licence holder occurring after the commencement of this licence;

(b) any revocation or suspension of a licence or permit held by the licence holder under a law of the Australian Government, a State or a foreign country, being a law relating to the health and safety of people or the environment;

(c) any event or circumstances occurring after the commencement of this licence that would affect the capacity of the holder of his licence to meet the conditions in it.



Licence holder must provide information on matters related to suitability

24. The licence holder must provide information related to the licence holder’s ongoing suitability to hold a licence when requested to do so in writing by the Regulator and must provide the information within a time period stipulated by the Regulator.

Additional information to be given to the Regulator

25. The licence holder must inform the Regulator if the licence holder becomes aware of:

(a) additional information as to any risks to the health and safety of people, or to the environment, associated with the dealings authorised by the licence; or

(b) any contraventions of the licence by a person covered by the licence; or (c) any unintended effects of the dealings authorised by the licence.

26. The licence holder must provide the information required by paragraphs (a) (b) and (c) of the immediately preceding condition to the Regulator as soon as practically and reasonably possible and must also include the information in the Annual Report.

Remaining an accredited organisation

27. The licence holder must, at all times, remain an accredited organisation in accordance with the Gene Technology Act 2000 and comply with any conditions of accreditation set out in the licence holder’s instrument of accreditation.

Compliance management plan

28. Prior to planting the GMOs, a written Compliance Management Plan must be provided to the Regulator. The Compliance Management Plan must describe in detail how the licence holder intends to ensure compliance with these conditions and to document that compliance.

Note: The Compliance Management Plan must include copies of documents required to be provided under conditions 11, 14 and 20. Annual Report

29. The licence holder must provide an Annual Report to the Regulator. Testing Methodology

30. The licence holder must provide a written instrument to the Regulator describing an experimental method that is capable of reliably detecting the presence of the GMOs and any transferred genetically modified material that might be present in a recipient organism. The instrument must be provided within 30 days of the issue date of the licence.



References Agrifood Awareness Australia (2001). GM canola, pollen, bees and honey. http://www.afaa.com.au/.

Ahmad, W., Nicholls, C., Ellar, D.J. (1989). Cloning and expression of an entomocidal protein gene from Bacillus thuringiensis galleriae toxic to both lepidoptera and diptera. FEMS Microbiology letters - Federation of European Microbiological Societies, 59: 198-202.

Akhurst, R.J. (2005). A report from CSIRO Entomology to the Office of the Gene Technology Regulator on the testing of the toxicity of the Cry2Ab Bacillus thuringiensis insecticidal protein for diptera. Unpublished.

An, W.Q., McDowell, J.M., Huang, S., McKinney, E.C., Chambliss, S., Meagher, R.B. (1996). Strong, constitutive expression of the Arabidopsis ACT2/ACT8 actin subclass in vegetative tissues. Plant Journal 10: 107-121.

ANZFA (1999). Full assessment report and regulatory impact statement, A341: Oil and linters derived from insect resistant cotton. Report No. A341, Australia New Zealand Food Authority, Canberra, Australia.

ANZFA (2000a). Draft risk analysis report, application A355: Food produced from glyphosate-tolerant cotton line 1445. Australia New Zealand Food Authority, Canberra, Australia.

ANZFA (2000b). Final risk analysis report application A363: Food produced from glyphosate-tolerant canola line GT73. http://www.foodstandards.gov.au/_srcfiles/A363%20draft%20IR.pdf.

ANZFA (2000c). Full assessment report and regulatory impact assessment. Subject: A338, Food derived from Glyphosate tolerant soybeans.

ANZFA (2001a). Draft risk analysis report Application A384 Food derived from insect and potato virus Y-protected potato lines RBMT15-101, SEMT15-02 and SEMT15-15.

ANZFA (2001b). Final assessment report - Application A379: Oil and linters from Bromoxynil tolerant cotton transformation events 10211 and 10222. Report No. A379, Food Standards Australia New Zealand, Canberra, Australia.

ANZFA (2001c). Final assessment report - Application A382: Food derived from insect-protected potato lines BT-06, ATBT04-06, ATBT04-31, ATBT04-36, and SPBT02-05. Report No. A382, Food Standards Australia New Zealand, Canberra, Australia.

ANZFA (2001d). Final assessment report - Application A383: Food derived from insect and potato leafroll virus-protected potato lines RBMT21-129, RBMT21-350, and RBMT22-82. Report No. A383, Food Standards Australia New Zealand, Canberra, Australia.

ANZFA (2001e). Final assessment report application A378: Food derived from glyphosate-tolerant sugarbeet line 77 (GTSB77). Report No. A378, Australia New Zealand Food Authority, Canberra, Australia.

References (February 2006) 106

http://www.afaa.com.au/

http://www.foodstandards.gov.au/_srcfiles/A363 draft IR.pdf


ANZFA (2001f). Final assessment report. Application A380: Food from insect-protected and glufosinate ammonium-tolerant DBT418 corn. Australia New Zealand Food Authority,

ANZFA (2001g). Food derived from glyphosate-tolerant cotton line 1445 - A safety assessment. Australia New Zealand Food Authority, Canberra, Australia.

ANZFA (2002a). Draft Assessment (Full Assessment - s.15) Application A416: Food derived from Glyphosate-tolerant Corn Line NK603.

ANZFA (2002b). Draft final risk analysis report application A386: Food derived from insect-protected, herbicide-tolerant Bt-11 corn. Report No. Application A386, Australia New Zealand Food Authority, Canberra, Australia.

ANZFA (2002c). Final assessment report (Full assessment - S.15) Application A436: Oil and linters derived from insect-protected cotton containing event 15985. Report No. Full assessment - S.15 Application A436, Australia New Zealand Food Authority, Canberra, Australia.

APVMA (2004). PUBCRIS - Registered products database. http://www.apvma.gov.au/pubcris/subpage_pubcris.shtml. Accessed on 22 July 2004.

Australian Cotton Cooperative Research Centre (2002). Australian dryland cotton: production guide. Cotton Research & Development Corporation, Narrabri, Australia. pp 1-108.

Australian Cotton Cooperative Research Centre (2004). An Australian guide to Bollgard II® management. http://cotton.pi.csiro.au/publicat/agro/BG204.htm.

Avcare (2003). Herbicide resistance management strategies. http://www.avcare.org.au/files/resistancestrategie/Herbicide%20resistance%20management%20strategies.pdf.

Axelos, M., Bardet, C., Liboz, T., Le Van, T.A., Curie, C., Lescure, B. (1989). The gene family encoding the Arabidopsis thaliana translation elongation factor EF-1 alpha: molecular cloning, characterization and expression. Molecular and General Genetics 219: 106-112.

Barbera, P.W. (1995). Toxicity/pathogenecity testing of Bacillus thuringiensis strain EG 7826 following acute oral challenge in rats. Report No. IITRI Project Number L08574, IIT Research Institute, Chicago IL.

Barry, G., Kishore, G., Padgette, S., Stallings, W.C. (27-5-1997). Monsanto Company (St.Louis, MO, USA. United States Patent: Glyphosate-tolerant 5-enolpyruvylshikimate-3-phosphate synthases. World patent no: 5633435.

Bartlett, S.G., Grossman, A.R., Chua, N.H. (1982). Methods in chloroplast molecular biology. Edelman, M., Hallick, R.B., Chua, N.H. (eds). Elsevier, Amsterdam. pp 1081-1091.

Batista, R., Nunes, B., Carmo, M., Cardoso, C., Sao Jose, H., Bugalho de Almeida, A., Manique, A., Bento, L., Pinto Ricardo, C., Oliveira, M.M. (2005). Lack of detectable allergenicity of transgenic maize and soya samples. Journal of Allergy and Clinical Immunology 116: 403-410.


http://www.apvma.gov.au/pubcris/subpage_pubcris.shtml

http://cotton.pi.csiro.au/publicat/agro/BG204.htm

http://www.avcare.org.au/files/resistancestrategie/Herbicide resistance management strategies.pdf

http://www.avcare.org.au/files/resistancestrategie/Herbicide resistance management strategies.pdf


Bechtel, C.L. (1999). St Louis, MO,Acute oral toxicity study of insect protection protein 2 (IPP2) in mice. Unpublished. Monsanto Report No. MSL:16649. Monsanto company.

Beck, E., Ludwig, G., Auerswald, E.A., Reiss, B., Schaller, H. (1982). Nucleotide sequence and exact localization of the neomycin phosphotransferase gene from transposon Tn5. Gene 19: 327-336.

Belgian Biosafety Server (1999). Types of antibiotics and related resistance genes. http://www.antibioresistance.be/ARmenu.html.

Bentley, R. (1990). The shikimate pathway - a metabolic tree with many branches. Critical Reviews in Biochemistry and Molecular Biology 25: 307-384.

Berberich, S.A., Leimgruber, R.M., Regan, G.J. (1993). Monsanto Company,Preparation and verification of dose for a mouse acute oral toxicity study with neomycin phosphotransferase II protein (NPTII), study ML-91-409. Unpublished. Monsanto Report No. MSL:13277. Monsanto Technical Report.

Bergelson, J., Purrington, C.B., Wichmann, G. (1998). Promiscuity in transgenic plants. Nature 395: 25.

Bernstein, I.L., Bernstein, J.A., Miller, M., Tierzieva, S., Bernstein, D.I., Lummus, Z., Selgrade, M.K., Doerfler, D.L., Seligy, V.L. (1999). Immune responses in Farm workers after exposure to Bacillus thuringiensis pesticides. Environmental Health Perspectives 107: 575-582.

Betz, F.S., Hammond, B.G., Fuchs, R.L. (2000). Safety and advantages of Bacillus thuringiensis-protected plants to control insect pests. Regulatory Toxicology and Pharmacology 32: 156-173.

Bevan, M. (1984). Binary Agrobacterium vectors for plant transformation. Nucleic Acids Research 12: 8711-8721.

Bookout, J.T., Pineda, N.G., Nicholas, N.R., and Jennings, J.C. (2003). Assessment of CP4 EPSPS protein levels in leaf, root, seed and pollen tissues from Roundup Ready® Flex Cotton MON 88913 produced in 2002 field trials. Unpublished. Report No. Technical Report MSL-18849, Monsanto. St Louis, Missouri.,

Bradford, K.J., van Deynze, A., Gutterson, N., Parrot, W., Strauss, S.H. (2005). Regulating transgenic crops sensibly: lessons from plant breeding, biotechnology and genomics. Nature Biotechnology 23[4], 439-444.

Brimecombe, M.J., De Leij, F.A., Lynch, J.M. (2001). The effect of root exudates on rhizosphere microbial populations. Chapter 4. In: R Pinton, Z Varanini, P Nannipieri, eds. The rhizosphere: biochemistry and organic substances at the soil-plant interface. Marcel Dekker, Inc., New York, USA. pp 95-140.

Brubaker, C.L., Brown, A.H.D., Stewart, J.M., Kilby, M.J., Grace, J.P. (1999). Production of fertile hybrid germplasm with diploid Australian Gossypium species for cotton improvement. Euphytica 108: 199-213.

Burns, J.A. (2004). Petition for the determination of non-regulated status for Roundup


http://www.antibioresistance.be/ARmenu.html


Ready® Flex Cotton MON 88913. Report No. Petition # 04-CT-112U., Monsanto Company, St Louis, Missouri.,

Burns, J.A., Sivasupramaniam, S., Hurst, J., and Vaughn, T. (2004). Bioefficacy of the combined trait product produced through traditional breeding: Roundup Ready Flex® x Bollgard II® cotton. Unpublished. Monsanto Company, St Louis, Missouri.,

Calgene, I. (1990). Request for advisory opinion - kan, gene: safety and use in the production of genetically engineered plants. Report No. FDA Docket Number 90A-0416,

Canadian Food Inspection Agency (1997). Decision Document 97-21: Detemination of the safety of cotton lines with Roundup Ready genes (Gossypium hirsutum L.). Report No. 97-21 , Plant health and Produciton Division, Plant Biotechnology Office,

Carson, D.B., Beach, S.B., and Nickson, T.E. (2004). Effect of the CP4 EPSPS protein on carbon and nitrogen transformation by soil microorganisms. Unpublished. Report No. Monsanto Technical Report MSL-18910., Monsanto, St Louis, Missouri.,

Carter, J.N. and Ligget, M.P. (1994). Acute oral toxicity and infectivity/pathogenicity to rats of EG 7841. Report No. HRC Study Report number ECO 6/942538, Huntingdon Research Centre Ltd., Huntington Cambridgeshire England.

Castillo, A.R., Gallardo, M.R., Maciel, M., Giordano, J.M., Conti, G.A., Gaggiotti, M.C., Quaino, O., Gianni, C., Hartnell, G.F. (2004). Effects of feeding rations with genetically modiried whole cottonseed to lactating Holstein cows. Journal of Dairy Science 87: 1778-1785.

Charles, G. (2002a). Managing weeds in cotton. In: WEEDPak. Australian Cotton CRC,

Charles, G. (2002b). Research results with Roundup Ready® cotton. In: WEEDpak. Australian Cotton CRC,

Charles, G., Sullivan, A., Christiansen, I., Roberts, G. (2002). Managing weeds on roads, channels and water storages. In: WEEDpak. Australian Cotton CRC, Canberra.

Cherry, J.P., Leffler, H.R. (1984). Cotton;Seed. Kohel, R.J., Lewis, C.F. (eds). pp 511-558.

Constable, G.A. and Shaw, A.J. (1988). Temperature requirements for cotton. Report No. Agfact P5.3.5, NSW Agriculture & Fisheries,

Coruzzi, G., Brogue, C., Edwards, C., Chua, N.H. (1984). Tissue-specific and light-regulated expression of a pea nuclear gene encoding the small subunit of ribulose-1,5-bisphosphate carboxylase. EMBO Journal 3: 1671-1679.

Cotton Australia (2005a). Biotechnology. http://www.cottonaustralia.com.au/factSheets/factSheets.aspx.

Cotton Australia (2005b). Cotton Australia - Annual Report 2004-2005.

Craven, L. A., Stewart, J. M., Brown, A. H. D., and Grace, J. P. (1994). Challenging the future; the Australian wild species of Gossypium. In "Proceedings of the 1st World Cotton Research Conference", pp. 278-281.


http://www.cottonaustralia.com.au/factSheets/factSheets.aspx


Crossan, A. and Kennedy, I. (2004). A snapshot of Roundup Ready® cotton in Australia: Are there environmental benefits from the rapid adoption of Roundup Ready® cotton in Australia? The University of Sydney,

Dankocsik, C., Donovan, W.P., Jany, C.S. (1990). Activation of a cryptic crystal protein gene of Bacillus thuringiensis subspecies kurstaki by gene fusion and determination of the crystal protein insecticidal specificity. Molecular Microbiology 4: 2087-2094.

De Vries, J., Meier, P., Wackernagel, W. (2001). The natural transformation of the soil bacteria Pseudomonas stutzeri and Acinetobacter sp. by transgenic plant DNA strictly depends on homologous sequences in the recipient cells. FEMS Microbiology Letters 195: 211-215.

del Rincon-Castro, M.C., Barajas-Huerta, J., Ibarra, J.E. (1999). Antagonism between Cry1Ac1 and Cyt1A1 toxins of Bacillus thuringiensis. Applied and Environmental Microbiology 65: 2049-2053.

della-Cioppa, G., Bauer, S.C., Klein, B.K., Shah, D.M., Fraley, R.T., Kishore, G.M. (1986). Translocation of the precursor of 5-enolpyruvylshikimate-3-phiosphate synthase into chloroplasts of higher plants in vitro. Proceedings of the National Academy of Sciences of the United States of America 83: 6873-6977.

Donegan, K.K., Palm, C.J., Fieland, V.J., Porteous, L.A., Ganio, L.M., Schaller, D.L., Bucao, L.Q., Seidler, R.J. (1995). Changes in levels, species and DNA fingerprints of soil microorganisms associated with cotton expressing the Bacillus thuringiensis var. kurstaki endotoxin. Applied Soil Ecology 2: 111-124.

Donegan, K.K., Seidler, R.J. (1998). Effect of transgenic cotton expressing the Bacillus thuringiensis var. kurstaki endotoxin on soil microorganisms- Risk assessment studies. Chapter V.4. In: YPS Bajaj, ed. Biotechnology in Australia,Volume 42:Cotton pp 300-312.

Donovan, W.P., Donovan, J.C., Engleman, J.T. (2001). Gene knockout demonstrates that vip3A contributes to the pathogenesis of Bacillus thuringiensis toward Agrotis ipsilon and Spodoptera exigua. Journal of Invertebrate Pathology 78: 45-51.

Dubelman, S., Ayden, B.R., Dudin, Y.A., Bookout, J.T., and Jiang, C. (2005). Aerobic soil degradation of the CP4 EPSPS protein. Unpublished. Report No. Monsanto Technical Report, MSL-19332., Monsanto, St Louis, Missouri.,

Dubleman, S., Martin, J.W., and Bhalgat, M.K. (2001). Aerobic soil degradation of the Bacillus thuringiensis insect protection protein 2 in cotton leaf tissue. Report No. MSL-16185, Monsanto Company, St Louis.

Duke, J.A. (1983). Gossypium hirsutum L. In: Handbook of Energy Crops (unpublished)

Dunn, C. (2005). Field phenotypic evaluation of Roundup Ready Flex Cotton MON 88913 from Australian field trials conducted in the 2003-2004 season. Unpublished. Report No. Study Report # MAL2005-5-RRF. Melbourne, Australia., Monsanto Australia Limited. Melbourne, Australia.,

Eastick, R. (2002). Evaluation of the potential weediness of transgenic cotton in northern Australia. Report No. Technical Bulletin no. 305, Northern Territory Government and



Australian Cotton Cooperative Research Centre,

EFB (2001). Antibiotic resistance markers in genetically modified (GM) crops. Report No. Briefing Paper no. 10, European Federation of Biotechnology,

EFSA (2004). Opinion of the scientific panel on genetically modified organisms on the use of antibiotic resistance genes as marker genes in genetically modified plants. The EFSA Journal 48: 1-18.

English, L., Slatin, S.L. (1992). Mode of action of delta-endotoxins from Bacillus thuringiensis: a comparison with other bacterial toxins. Insect Biochemistry and Molecular Biology 22: 1-7.

EPA (2001a). Biopesticides registration action document: Bacillus thuringiensis plant-incorporated protectants. US EPA,

EPA (1994). Neomycin phosphotransferase II; tolerance exemption. Federal Register 59: 49351-49353.

EPA (1996). Plant pesticide Inert ingredient CP4 Enolpyruvylshikimate-3-D and the genetic material necessary for its production in all plants. Federal Register 61: 40338-40340.

EPA (2000). Biopesticide fact sheet: Bacillus thuringiensis subsp. kurstaki Cry1Ac delta-endotoxin and its controlling sequences as expressed in cotton (006445) Issued: 4/00.

EPA (2001b). β-D Glucuronidase from E. coli and the genetic material necessary for its production as a plant pesticide inert ingredient: exemption from the requirement of a tolerance. Federal Register 66: 42957-42962.

Eschenburg, S., Healy, M.L., Priestman, M.A., Lushington, G.H., Schonbrunn, E. (2002). How the mutation glycine96 to alanine confers glyphosate insensitivity to 5-enolpyruvyl shikimate-3-phosphate synthase from Escherichia coli. Planta 216: 129-135.

Farrell, T. and Roberts, G. (2002). Survey of cotton volunteers north of latitude 22º south. Australian Cotton CRC and CSIRO Plant Industry, Narrabri.

FDA (1994). Secondary food additives permitted in food for human consumption; food additives permitted in feed and drinking water of animals; aminoglycoside 3'-phosphotransferase II; final rule. Report No. 59, United States Food and Drug Administration, Washington, USA.

Felsot, A.S. (2000). Insecticidal genes. Part 2: Human health hoopla. Agrichemical & Environmental News 168: 1-7.

Flavell, R.B., Dart, E., Fuchs, R.L., Fraley, R.T. (1992). Selectable marker genes: safe for plants? Bio/Technology 10: 141-144.

Flores, S., Saxena, D., Stotzky, G. (2005). Transgenic Bt plants decompose less in soil than non-Bt plants. Soil Biology and Biochemistry 37: 1073-1082.

Forrester, N.W. and Wilson, A.G.L. (1988). Insect pests of cotton. NSW Agriculture,



Fryxell, P.A. (1992). A revised taxonomic interpretation of Gossypium L. (Malvaceae). Rheedea 2: 108-165.

FSANZ (2003). Final Assessment Report - Application A484: Food from Insect Protected MON863 Corn.

FSANZ (2004). Final assessment report - Application A509: Food derived from insect protected cotton line COT102.

FSANZ (2005). Final assessment report - Application A553: Food derived from glyphosate-tolerant cotton line MON 88913.

Fuchs, R.L., Astwood, J.D. (1996). Allergenicity assessment of foods derived from genetically modified plants. Food Technology 50: 83-88.

Fuchs, R.L., Berberich, S.A., Serdy, F.S. (1993a). Safety evaluation of genetically engineered plants and plant products: insect resistant cotton. In: JA Thomas, LA Myers, eds. Biotechnology and Safety Assessment. Raven Press Ltd, New York, pp 199-212.

Fuchs, R.L., Heeren, R.A., Gustafson, M.E., Rogan, G.J., Bartnicki, D.E., Leimgruber, R.M., Finn, R.F., Hershman, A., Berberich, S.A. (1993b). Purification and characterization of microbially expressed neomycin phosphotransferase II (NPTII) protein and its equivalence to the plant expressed protein. Biotechnology (NY) 11: 1537-1542.

Fuchs, R.L., Ream, J.E., Hammond, B.G., Naylor, M.W., Leimgruber, R.M., Berberich, S.A. (1993c). Safety assessment of the neomycin phosphotransferase II (NPTII) protein. Bio/Technology 11: 1543-1547.

Gallagher, S., Grimes, J., Beavers, J.B. (2000). St. Louis,Insect protection protein 2 in cottonseed meal: a dietary toxicity study with the northern Bobwhite. Unpublished. Monsanto Report No. MSL:1678. Monsanto Company.

Gebhard, F., Smalla, K. (1998). Transformation of Acinetobacter sp. strain BD413 by transgenic sugar beet DNA. Applied and Environmental Microbiology 64: 1550-1554.

George, C., Trujillo, W., Sorbet, R., and Riordan, S. (2005). Compositional analyses of cottonseed collected from Roundup Ready® Flex Cotton, MON 88913, control and conventional cotton from Australian field trials in 2003-2004. Report No. Monsanto Technical Report MSL-19673., Monsanto, St Louis, Missouri.,

Gilissen, L.J.W., Metz, P.L.J., Stiekema, W.J., Nap, J.P. (1998). Biosafety of E.coli β-glucuronidase (GUS) in plants. Transgenic Research 7: 157-163.

Goldstein, D.A., Tinland, B., Gilbertson, L.A., Staub, J.M., Bannon, G.A., Goodman, R.E., McCoy, R.L., Silvanovich, A. (2005). Human safety and genetically modified plants: a review of antibiotic resistance markers and future transformation selection technologies. Journal of Applied Microbiology 99: 7-23.

Gregory, S.R., Hernandez, E., Savoy, B.R. (1999). Cottonseed processing. Chapter 4.5. In: CW Smith, JT Cothren, eds. Cotton: Origin, History, Technology and Production. John Wiley & Sons, New York. pp 793-819.



Groves, R.H., Boden, R., and Lonsdale, W.M. (2005). Jumping the garden fence: Invasive garden plants in Australia and their environmental and agricultural impacts. Report No. CSIRO Report prepared for WWF Australia, WWF-Australia,

Groves, R.H., Hosking, J.R., Batianoff, D.A., Cooke, D.A., Cowie, I.D., Keighery, B.J., Rozefelds, A.C., and Walsh, N.G. (2000). The naturalised non-native flora of Australia: its categorisation and threat to native plant biodiversity. Unpublished report to Environment Australia by the CRC for Weed Management Systems.

Groves, R.H., Hosking, J.R., Batianoff, G.N., Cooke, D.A., Cowie, I.D., Johnson, R.W., Keighery, G.J., Lepschi, B.J., Mitchell, A.A., Moerkerk, M., Randall, R.P., Rozefelds, A.C., Walsh, N.G., Waterhouse, B.M. (2003). Weed categories for natural and agricultural ecosystem management. Bureau of Rural Sciences, Canberra.

Groves, R.H., Hosking, J.R., Cooke, D.A., Johnson, R.W., Lepschi, B.J., Mitchell, A.A., Moerkerk, M., Randall, R.P., Rozefelds, A.C., and Waterhouse, B.M. (2002). The naturalised non-native flora of Australia: its categorisation and threat to agricultural ecosystems. Unpublished report to Agriculture, Fisheries and Forestry Australia by the CRC for Weed Management Systems.

Gupta, V. and Watson, S. (2004). Ecological impacts of GM cotton on soil biodiversity - below ground production of Bt by GM cotton and Bt cotton impacts on soil biological processes. Australian Government Department of Environment and Heritage,

Halloin, J. M. (1975). Solute loss from deteriorated cotton seed: relationships between deterioration, seed moisture and solute loss. Crop Science 15: 11-15.

Harrison, L.A., Bailey, M.R., Naylor, M.W., Ream.J.E., Hammond, B.G., Nida, D.L., Burnette, B.L., Nickson, T.E., Mitsky, T.A., Taylor, M.L., Fuchs, R.L., Padgette, S.R. (1996a). The expressed protein in glyphosate-tolerant soybean, 5-enolpyruvylshikimate-3-phospate synthase from Agrobacterium sp. strain CP4, is rapidly digested in vitro and is not toxic to actutely gavaged mice. Journal of Nutrition 126: 728-740.

Harrison, L.A., Biest, N.A., Leimgruber, R., and Padgette, S. (1996b). Preparation, characterisation, and confirmation of doses for an acute mouse feeding study with B-D-glucuronidase. Unpublished. Report No. Monsanto Report No. MSL:12979, Monsanto Company, St. Louis.

Harwood, J.D., Wallin, W.G., Obrycki, J.J. (2005). Uptake of Bt endotoxins by nontarget herbivores and higher order arthropod predators: molecular evidence from a transgenic corn agroecosystem. Molecular Ecology 14: 2815-2823.

Head, G., Surber, J.B., Watson, J.A., Martin, J.W., Duan, J.J. (2002). No detection of Cry1Ac protein in soil after multiple years of transgenic cotton (Bollgard®) use. Environmental Entomology 31: 30-36.

Heap, I.M. (2003). International survey of herbicide resistant weeds. http://www.weedscience.org.

Heym, B., Honore, N., Truffot-Pernot, C., Banerjee, A., Schurra, C., Jacobs, W.R., Jr., van Embden, J.D., Grosset, J.H., Cole, S.T. (1994). Implications of multidrug resistance for the future of short-course chemotherapy of tuberculosis: a molecular study. Lancet 344: 293-298.


http://www.weedscience.org/


Hofmann, C., Vanderbruggen, H., Hofte, H., Van Rie, J., Jansens, S., Van Mellaert, H. (1988). Specificity of Bacillus thuringiensis δ-endotoxins is correlated with the presence of high-affinity binding sites in the brush border membrane of target insect midguts. Proceedings of the National Academy of Sciences of the United States of America 85: 7844-7848.

Holm, L., Doll, J., Holm, E., Pancho, J., Herberger, J. (1997). World weeds. Natural histories and distribution. John Wiley and Sons, Inc, USA.

Holm, L., Pancho, J., V, Herberger, J.P., Plucknett, D.L. (1979). A geographical atlas of world weeds. John Wiley and Sons, Brisbane, Australia. pp 471-04393.

Hopkins, D.W., Gregorich, E.G. (2003). Detection and decay of the Bt endotoxin in soil from a field trial with genetically modified maize. European Journal of Soil Science 54: 793-800.

Hu, C.Y., Chee, P.P., Chesney, R.H., Zhou, J.H., Miller, P.D., O'Brien, W.T. (1990). Intrinsic GUS-like activities in seed plants. Plant Cell Reports 9: 1-5.

James, C. (2004). Preview: Global status of commercialized biotech/GM crops: 2004. Report No. 32-2004, International Service for the Acquisition of Agri-Biotech Applications, Ithaca, NY.

Jefferson, R.A., Burgess, S.M., Hirsh, D. (1986). β-Glucuronidase from Escherichia coli as a gene-fusion marker. Proceedings of the National Academy of Sciences of the United States of America 83: 8447-8451.

Jefferson, R.A., Kavanagh, T.A., Bevan, M.W. (1987). GUS fusions: β-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. The EMBO Journal 6: 3901-3907.

Jefferson, R.A., Wilson, K.J. (1991). The GUS gene fusion system. Plant Molecular Biology Manual B-14: 1-33.

Jenkins, J.N. (1992). Cotton. In: OECD Historical Review of Traditional Crop Breeding

Johnson, A., Farrell, T. (2004). Cotton pest management guide 2004-2005. NSW Department of Primary Industries,

Kay, R., Chan, A., Daly, M., McPherson, J. (1987). Duplication of CaMV 35S promoter sequences creates a strong enhancer for plant genes. Science 236: 1299-1302.

Klee, H.J., Muskopf, Y.M., Gasser, C.S. (1987). Cloning of an Arabidopsis thaliana gene encoding 5-enolpyruvylshikimate-3-phosphate synthase: sequence analysis and manipulation to obtain glyphosate-tolerant plants. Molecular and General Genetics 210: 437-442.

Klee, H.J., Rogers, S.G. (1989). Plant gene vectors and genetic transformation: plant transformation systems based on the use of Agrobacterium tumefaciens. Cell Culture and Somatic Cell Genetics of Plants 6: 1-23.

Knowles, B.H., Dow, J.A.T. (1993). The crystal delta-endotoxins of Bacillus thuringiensis: models for their mechanism of action on the insect gut. Bioassays 15: 469-476.

Konig, A., Cockburn, A., Crevel, R.W., Debruyne, E., Grafstroem, R., Hammerling, U.,



Kimber, I., Knudsen, I., Kuiper, H.A., Peijnenburg, A.A., Penninks, A.H., Poulsen, M., Schauzu, M., Wal, J.M. (2004). Assessment of the safety of foods derived from genetically modified (GM) crops. Food and Chemical Toxicology 42: 1047-1088.

Kurland, C.G., Canback, B., Berg, O.G. (2003). Horizontal gene transfer: a critical view. Proceedings of the National Academy of Science of the United States of America 100: 9658-9662.

Leach, J.N., Hileman, R.E., Martin, J.W., Nemeth, M.A., and Astwood, J.D. (2000). Assessment of the in vitro digestibility of insect protection proteins 2 (IPP2) utilising mammalian digestive fate models. Report No. Study No. 98-01-39-04; MSL No. 16640 (Amendment), Monsanto Company St Louis, MO.

Lee, M.K., Curtiss, A., Alcantara, E., Dean, D.H. (1996). Synergistic effect of the Bacillus thuringiensis toxins Cry1Aa and Cry1Ac on the Gypsy moth, Lymantria dispar. Applied and Environmental Microbiology 62: 583-586.

Leffler, H.R., Tubertini, B.S. (1976). Development of cotton fruit: accumulation and distribution of mineral nutrients. Agronomy Journal 68: 858-861.

Levy, S.B., Marshall, B., Schluederberg, S., Rowse, D., Davis, J. (1998). High frequency of antimicrobial resistance in human fecal flora. Antimicrobial Agents and Chemotherapy 32: 1801-1806.

Li, M.H. and Robinson, E.H. (2000). Evaluation of cottonseed meal derived from insect protected cotton lines 15813 and 15985 as a feed ingredient for catfish. Report No. MSL-16179, Trad Cochran National Warmwater Aquaculture Centre (Testing facility) Mississippi State University Stoneville MS 38776-0197.

Liang, Y., Dean, D.H. (1994). Location of a lepidopteran specificity region in insecticidal crystal protein CryIIA from Bacillus thuringiensis. Molecular Microbiology 13: 569-575.

Liu, X., Zhang, Q., Zhao, J.-Z.C.Q., Xu, H., Li, J. (2005). Effects of the Cry1Ac toxin of Bacillus thuringiensis on Microplitis mediator, a parasitoid of the cotton bollworm, Helicoverpa armigera. Entomologia Experimentalis et Applicata 114: 205-213.

Liu, Y.-B., Tabashnik, B.E., Moar, W., Smith, R.A. (1998). Synergism between Bacillus thuringiensis spores and toxins against resistant and susceptible diamondback moths (Plutella xylostella). Applied and Environmental Microbiology 64: 1385-1389.

Macintosh, S.C., Stone, T.B., Sims, S.R., Hunst, P.L., Greenplate, J.T., Marrone, P.G., Perlak, F.J., Fischhoff, D.A., Fuchs, R.L. (1990). Specificity and efficacy of purified Bacillus thuringiensis proteins against agronomically important insects. Journal of Invertebrate Pathology 56: 258-266.

Maggi, V.L. (1993a). Evaluation of dietary effects of purified B.t.k. endotoxin proteins on honey bee adults. Report No. CAR 181-92, Monsanto Australia Limited.

Maggi, V.L. (1993b). Evaluation of dietary effects of purified B.t.k. endotoxin proteins on honey bee larvae. Report No. CAR 180-92, Monsanto Australia Limited.

Maggi, V.L. (2000a). Evaluation of dietary effects of purified Bacillus thurigiensis Cry2Ab



protein on honey bee larvae. Report No. MSL-16961, Monsanto Company, Biotechnology Regulatory Science Final Report.

Maggi, V.L. (2000b). Evaluation of the dietary efects of insect protection protein 2 on adult honey bees (Apis mellifera L.). Report No. MSL 16176, California Agricultural Research Inc.

Mandal, A.B., Elangovan, A.V., Shrivastav, A.K., Johri, A.K., Kaur, S., Johri, T.S. (2004). Comparison of broiler chicken performance when fed diets containing meals of Bollgard II hybrid cotton containing Cry-X gene(Cry1Ac and Cry2Ab gene), parental line or commercial cotton. British Poultry Science 45: 657-663.

Martin, J.W., McKee, M.J., Dubleman, S., Dudin, Y.A. (2000). St. Louis,Aerobic soil degradation of the B.t. protein 11098 as a component of insect protected corn. Unpublished. Monsanto Report No. MSL:16440. Monsanto.

Martin, P.A.W., Travers, R.S. (1989). Worldwide abundance and distribution of Bacillus thuringiensis isolates. Applied and Environmental Microbiology 55: 2437-2442.

McCabe, D.E., Martinell, B.J. (1993). Transformation of elite cotton cultivars via particle bombardment of meristems. Bio/Technology 11: 596-598.

McClintock, J.T., Schaffer, C.R., Sjoblad, R.D. (1995). A comparative review of the mammalian toxicity of Bacillus thuringiensis-based pesticides. Pesticide Science 45: 95-105.

Meadows, M.P. (1993). Bacillus thuringiensis in the environment: ecology and risk assessment. Chapter 9. In: PF Entwistle, JS Cory, MJ Bailey, S Higgs, eds. Bacillus thuringiensis, an environmental biopesticide: theory and practice. John Wiley and sons, Chichester, UK. pp 193-220.

Mercer, D.K., Scott, K.P., Bruce-Johnson, W.A., Glover, L.A., Flint, H.J. (1999). Fate of free DNA and transformation of the oral bacterium Streptococcus gordonii DL1 by plasmid DNA in human saliva. Applied and Environmental Microbiology 65: 6-10.

Metcalfe, D.D., Astwood, J.D., Townsend, R., Sampson, H.A., Taylor, S.L., Fuchs, R.L. (1996). Assessment of the allergenic potential of foods derived from genetically engineered crop plants. Critical Reviews in Food Science and Nutrition 36(S): S165-S186.

Miki, B., McHugh, S. (2004). Selectable marker genes in trangenic plants: applications, alternatives and biosafety. Journal of Biotechnology 107: 193-232.

Mitsky, T.A. (1993). 700 Chesterfield Parkway North, St Louis, MO, USA 63198,Comparative alignment of CP4 EPSPS to known allergenic and toxic proteins using Fasta algorithm. Unpublished. Monsanto Report No. MSL:12820. Monsanto Company.

Monsanto Australia Limited (2001). Roundup Ready® Cotton Technical Manual. 2. Monsanto Australia Ltd.

Monsanto Australia Limited (2004). A guide to the 2004/05 Bollgard II resistance management plan.

Mulvaney, M. (2001). The effect of introduction pressure on the naturalization of ornamental woody plants in south-eastern Australia. Chapter 15. In: RH Groves, FD Panetta, JG Virtue,



eds. Weed Risk Assessment. CSIRO Publishing, Melbourne. pp 186-193.

Nandula, V.K., Reddy, K.N., Duke, S.O., Poston, D.H. (2005). Glyphosate-resistant weeds: current status and future outlook. Outlooks on Pest Management August 2005: 183-187.

Naylor, M.W. (1993a). One month feeding study with insect-resistant cottonseed meal in Sprague-Dawley rats. Monsanto Company The Agricultural Group, St Louise Missouri USA.

Naylor, M.W. (1992). Acute oral toxicity study of beta-glucuronidase (GUS) protein in albino mice. Unpublished. Report No. Monsanto Report No. MSL:12485, Monsanto Company, St. Louis.

Naylor, M.W. (1993b). Acute oral toxicity of Bacillus thuringiensis var. kurstaki [Cry1Ac] HD-73 protein in albino mice. The Agricultural group Monsanto Company, St. Louise Missouri USA.

Nielsen, K.M., Bones, A.M., Smalla, K., van Elsas, J.D. (1998) Horizontal gene transfer from transgenic plants to terrestrial bacteria - a rare event? FEMS Microbiol Rev 22 (2): 79-103.

Nielsen, K.M., van Elsas, J.D., Smalla, K. (2000). Transformation of Acinetobacter sp strain BD413(pFG4 Delta nptII) with transgenic plant DNA in soil microcosms and effects of kanamycin on selection of transformants. Applied and Environmental Microbiology 66: 1237-1242.

NRA (1996). Special Review of Glyphosate. Report No. NRA Special Review Series 96.1 , National Registration Authority for Agricultural and Veterniary Chemicals,

Obert, J.C., Trujillo, W., Sorbet, R., and Riordan, S. (2003a). Compositional analyses of cottonseed collected from Roundup Ready® Flex Cotton MON 88913, negative segregant control, MON 88913(-), and commercial non-transgenic cotton varieties from U.S. field trials in 2002. Unpublished. Report No. Monsanto Technical Report MSL-18776., Monsanto, St Louis, Missouri.,

Obert, J.C., Trujillo, W., Sorbet, R., and Riordan, S. (2003b). Compositional analyses of cottonseed, raw cottonseed meal and cottonseed oil derived from Roundup Ready® Flex Cotton MON 88913, negative segregant control MON88913(-), and commercial conventional varieties produced in U.S. field trial in 2002 and processed in 2003. Unpublished. Report No. Monsanto Technical Report MSL-18809., Monsanto, St Louis, Missouri.,

Odell, J.T., Nagy, F., Chua, N.H. (1985). Identification of DNA sequences required for activity of the cauliflower mosaic virus 35S promoter. Nature 313: 810-812.

OECD (1998). OECD Guidelines for the Testing of Chemicals.

OECD (1999). Consensus document on general information concerning the genes and their enzymes that confer tolerance to glyphosate herbicide. Report No. ENV/JM/MONO(99)9, OECD - Organisation for Economic Co-operation and Development, Paris.

OGTR (2002a). Risk Assessment and Risk Management Plan. DIR 012. Monsanto Australia Ltd. Commercial release of Bollgard II® cotton. Office of the Gene Technology Regulator, Canberra, Australia.



OGTR (2002b). The Biology and Ecology of Cotton (Gossypium hirsutum) in Australia.

OGTR (2005). Risk Analysis Framework. Australian Government, Canberra, ACT.

Oosterhuis, D.M., Jernstedt, J. (1999). Morphology and anatomy of the cotton plant. Chapter 2.1. In: CW Smith, JT Cothren, eds. Cotton: Origin, History, Technology and Production. John Wiley & Sons, New York. pp 175-206.

Owen, M.D., Zelaya, I.A. (2005). Herbicide-resistant crops and weed resistance to herbicides. Pest Management Science 61: 301-311.

Padgette, S.R., Barry, G.F., Re, D.B., Eichholtz, D.A., Weldon, M., Kolacz, K., Kishore, G.M. (1993). USA,Purification, cloning and characterisation of a highly glyphosate-tolerant 5-enolpyruvylshikimate-3-phosphate synthase from Agrobacterium sp.strain CP4. Unpublished. Monsanto Report No. MSL:12738, 1-66. Monsanto Company.

Padgette, S.R., Re, D.B., Barry, G.F., Eichholtz, D.E., Delannay, X., Fuchs, R.L., Kishore, G.M., Fraley, R.T. (1996). New weed control opportunities: development of soybeans with a Roundup Ready gene. Chapter 4. In: SO Duke, ed. Herbicide-resistant crops: agricultural, environmental, economic, regulatory and technical aspects. CRC Press, Boca Raton. pp 53-84.

Palm, C.J., Schaller, D.L., Donegan, K.K., Seidler, R.J. (1996). Persistence in soil of transgenic plant produced Bacillus thuringiensis var. kurstaki δ-endotoxin. Canadian Journal of Microbiology 42: 1258-1262.

Palmer, S.J. and Beavers, J.B. (1993a). B.t.k. HD-73 protein: Dietary toxicity study with green lace wing larvae (Chrysopa carnea). Report No. WL 93-233, Monsanto Company, St. Louis.

Palmer, S.J. and Beavers, J.B. (1993b). B.t.k. HD-73 protein: Dietary toxicity study with ladybird beetles (Hippodamia convergens). Report No. WL 93-232, Monsanto Company, St. Louis.

Palmer, S.J. and Beavers, J.B. (1993c). B.t.k. HD-73 protein: dietary toxicity study with parasitic Hymenoptera (Nasonia vitripennis). Report No. WL 93-234, Monsanto Company, St. Louis.

Palmer, S.J. and Krueger, H.O. (2000a). Insect protection protein 2: a dietary toxicity study with green lacewing larvae (Chrysomera carnea). Report No. MSL-16171, Monsanto Company St Louis Misouri.

Palmer, S.J. and Krueger, H.O. (2000b). Insect protection protein 2: a dietary toxicity study with parasitic hymenoptera (Nasonia vitripennis). Report No. MSL-16173, Wildlife International, Ltd. Easton Maryland, USA.

Palmer, S.J. and Krueger, H.O. (2000c). Insect protection protein 2: a dietary toxicity study with the ladybird beetle (Hippodamia convergens). Report No. MSL- 16172, Wildlife International Ltd.

Palmer, S.J. and Krueger, H.O. (2000d). Insect protection protein 2: an acute toxicity study with the earthworm in an artificial soil substrate. Report No. MSL - 16177; Study number WL-99--067, Wildlife International Ltd.



Panetta, F.D. (1993). A system of assessing proposed plant introductions for weed potential. Plant Protection Quarterly 8: 10-14.

Percival, A.E., Wendel, J.F., Stewart, J.M. (1999). Taxonomy and Germplasm Resources. Chapter 1.2. In: CW Smith, JT Cothren, eds. Cotton: origin, history, technology, and production. John Wiley & Sons, New York, USA. pp 33-63.

Pheloung, P.C. (1995). Determining the weed potential of new plant introductions in Australia. Department of Agriculture, Perth, Australia.

Pheloung, P.C. (2001). Weed risk assessment for plant introductions to Australia. Chapter 7. In: RH Groves, FD Panetta, JG Virtue, eds. Weed Risk Assessment. CSIRO Publishing, Melbourne. pp 83-92.

Pheloung, P.C., Williams, P.A., Halloy, S.R. (1999). A weed risk assessment model for use as a biosecurity tool evaluating plant introductions. Journal of Environmental Management 57: 239-251.

Pineda, N.G., Bookout, J.T., and Jennings, J.C. (2005). CP4 EPSPS protein levels in seed and leaf tissues from Roundup Ready® Flex Cotton MON 88913 produced in Australia during 2003-2004. Unpublished. Report No. Technical Report MSL-19576, Monsanto. St Louis, Missouri.,

Pline, W.A., Viator, R., Wilcut, J.W., Edmisten, K.L., Thomas, J., Wells, R. (2002a). Reproductive abnormalities in glyphosate-resistant cotton caused by lower CP4-EPSPS levels in the male reproductive tissue. Weed Science 50: 438-447.

Pline, W.A., Wilcut, J.W., Duke, S.O., Edmisten, K.L., Wells, R. (2002b). Tolerance and accumulation of shikimic acid in response to glyphosate applications in glyphosate-resistant and nonglyphosate-resistant cotton (Gossypium hirsutum L.). Journal of Agricultural and Food Chemistry 50: 506-512.

Powles, S.B., Lorraine-Colwill, D.F., Dellow, J.F., Preston, C. (1998). Evolved resistance to glyphosate in rigid ryegrass (Lolium rigidum) in Australia. Weed Science 604-607.

Pratley, J., Urwin, N., Stanton, R., Baines, P., Broster, J., Cullis, K., Schafer, D., Bohn, J., Krueger, R. (1999). Resistance to glyphosate in Lolium rigidum. I. Bioevaluation. Weed Science 405-411.

Preston, C. (2003). Latest developments in herbicide resistance. Grains Research and Development Corporation Research Update - Southern Region - February 2003, http://www.grdc.com.au/growers/res_upd/south/03/herbicide.htm.

Radcliffe, J.C. (2002a). Pesticide use in Australia. Australian Academy of Technological Sciences and Engineering, Ian McLennan House, 197 Royal Parade, Parkville, Victoria 3052.

Radcliffe, J.C. (2002b). Pesticide Use in Australia - A review undertaken by the Australian Academy of Technolgical Sciences and Engineering. Australian Academy of Technolgical Sciences and Engineering, Parkville, OGTR Library.

Randall, R.P. (2002). A global compendium of weeds. R.G. & F.J. Richardson, Meredith, Victoria. pp 1-905.


http://www.grdc.com.au/growers/res_upd/south/03/herbicide.htm


Ream, J.E. (1994). Aerobic soil degradation of Bacilus thuringiensis var. kurstaki B.t.k. HD-73 protein bioactivity. Report No. MSL 13299, Monsanto Company, St. Louis USA.

Reeve, I., Stayner, R., Doyle, B., and McNeil, J. (2003). A scoping study on socio-economic indicators for the cotton industry - Report to the Cotton Research and Development Corporation. Institute for Rural Futures, University of New England,

Richins, R.D., Scholthof, H.B., Shepherd, R.J. (1987). Sequence of figwort mosaic virus DNA (caulimovirus group). Nucleic Acids Research 15: 8451-8466.

Roberts, G., Charles, G. (2002). Integrated weed management (IWM) guidlelines for Australian cotton production. In: WEEDPak. Australian Cotton CRC, Canberra.

Roberts, G., Kerlin, S., Hickman, M. (2002). Controlling volunteer cotton. In: WEEDpak. Australian Cotton CRC, Canberra.

Rogers, S.G., O'Connell, K., Horsch, R.B., Fraley, R.T. (1985). Biotechnology in plant science. Zaitlin, M., Day, P., Hollaender, A., Wilson, C.A. (eds). Academic Press Inc, New York. pp 219-226.

Saxena, D., Stewart, C.N., Altosaar, I., Shu, Q., Stotzky, G. (2004). Larvicidal Cry proteins from Bacillus thuringiensis are released in root exudates of transgenic B. thuringiensis corn, potato, and rice but not of B. thuringiensis canola, cotton, and tobacco. Plant Physiology and Biochemistry 42: 383-387.

Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R., Dean, D.H. (1998). Bacillus thuringiensis and its pesticidal crystal proteins. Microbiology and Molecular Biology Reviews 62: 775-806.

Schulz, A., Kruper, A., Amrhein, N. (1985). Differential sensitivity of bacterial 5-enolpyruvyl-shikimate-3-phosphate synthases to the herbicide glyphosate. FEMS Microbiology Letters 28: 297-301.

Shaw, K.J., Rather, P.N., Hare, R.S., Miller, G.H. (1993). Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglyciside-modifying enzymes. Microbiological Reviews 57: 138-163.

Shimada, N., Kim, Y.S., Miyamoto, K., Yoshioka, M., Murata, H. (2003). Effects of Bacillus thuringiensis Cry1Ab toxin on mammalian cells. Journal of Veterinary Medical Science 65: 187-191.

Sims, S. and Martin, J. (1996). Effect of the Bacillus thuringiensis insecticidal proteins Cry1Ab, Cry1Ac,Cry2A and Cry23A on Folsomia candida and Xenylla grisea (Insecta: Collembola). Report No. 93-081E1, Monsanto Company, St. Louis.

Sims, S.R. (1994). Sensitivity of insect species to the purified Cry 1Ac insecticidal protein for Bacillus thuringiensis subsp. kurstaki (B.t.k. HD-73). Report No. MSL 13273, Monsanto Company, St. Louis, USA.

Sims, S.R. (1995). Bacillus thuringiensis subsp. kurstaki (Cry1Ac) protein expressed in transgenic cotton: effects on beneficial and other non-target insects. Southwestern Entomologist 20: 493-500.



Sims, S.R., Berberich, S.A., Nida, D.L., Segalini, L.L., Leach, J.N., Ebert, C.C., Fuchs, R.L. (1996). Analysis of expressed proteins in fibre fractions from insect-protected and glyphosate-tolerant cotton varieties. Crop Science 36: 1212-1216.

Sjoblad, R.D., McClintock, J.T., Engler, R. (1992). Toxicological considerations for protein components of biological pesticide products. Regulatory Toxicology and Pharmacology 15: 3-9.

Smalla, K., Gebhard, F., van Elsas, J. D., Matzk, A., and Schiemann, J. (1994). Bacterial communities influenced by transgenic plants.Jones, D. D. eds, Division of Agriculture and Natural Resources, University of California, Oakland, USA. pp. 157-167.

Smalla, K., Van Overbeek, L.S., Pukall, R., van Elsas, J.D. (1993). Prevalence of npt II and Tn5 in kanamycin-resistant bacteria from different environments. FEMS Microbiology Ecology, 13: 47-58.

Spencer, T.M., Orozco, E.M., and Doyle, R.M. (1996). Petition for determination of non-regulated status: insect protected corn (Zea mays L.) with Cry1Ac gene from Bacillus thuringiensis subsp. kurstaki. DEKALB Genetics Corporation,

Steinrucken, H.C., Amrhein, N. (1980). The herbicide glyphosate is a potent inhibitor of 5-enolpyruvyl-shikimic acid-3-phosphate synthase. Biochemical and Biophysical Research Communications 94: 1207-1212.

Stotzky, G. (2000). Monitoring for indirect effects on non-target species, soil microbes, earthworms, and nematodes. In "Workshop on Ecological Monitoring of Genetically Modified Crops", National Research Council, Standing Committee on Biotechnology, Food and Fiber Production, and the Environment, Washington DC USA. pp. 13-14.

The Royal Society (2002). Genetically modified plants for food use and human health—an update. Report No. Policy document 4/02, The Royal Society, UK.

US EPA (2002). Bacillus thuringiensis Cry2Ab2 protein and its genetic material necessary for its production in cotton.

US FDA (1998). Guidance for industry: use of antibiotic resistance marker genes in transgenic plants.1-26. http://vm.cfsan.fda.gov/~dms/opa-armg.html.

USDA (2004). USDA National Nutrient Database for Standard Reference . http://www.nal.usda.gov/fnic/foodcomp/Data/SR17/sr17.html.

USDA-APHIS (2004). Monsanto Co.; Availability of Determination of Nonregulated Status for Cotton Genetically Engineered for Tolerance to the Herbicide Glyphosate. United States Department of Agriculture, Animal and Plant Health Inspection Service, web site.

van Frankenhuyzen, K., Gringorten, J.L., Milne, R.E., Gauthier, D., Pusztai, M., Brousseau, R., Masson, L. (1991). Specificity of activated Cry1A proteins from Bacillus thuringiensis subsp. kurstaki HD-1 for defoliating forest Lepidoptera. Applied Environmental Microbiology 57: 1650-1655.

van Frankenhuyzen, K. and Nystrom, C. (2002). The Bacillus thuringiensis toxin specificity database. http://www.glfc.cfs.nrcan.gc.ca/bacillus .


http://vm.cfsan.fda.gov/~dms/opa-armg.html

http://www.nal.usda.gov/fnic/foodcomp/Data/SR17/sr17.html

http://www.glfc.cfs.nrcan.gc.ca/bacillus


Van Rie, J., Jansens, S., Hofte, H., Degheele, D., Van Mellaert, H. (1989). Specificity of Bacillus thuringiensis delta -endotoxins: importance of specific receptors on the brush border membrane of the mid-gut of target insects. European Journal of Biochemistry 186: 239-247.

Wang, K., Herrera-Estrella, L., Van Montagu, M., Zambryski, P. (1984). Right 25 bp terminus sequence of the nopaline T-DNA is essential for and determines direction of DNA transfer from Agrobacterium to the plant genome. Cell 38: 455-462.

Whitehouse, M.E.A., Wilson, L.J., Fitt, G.P. (2005). A comparison of arthropod communities in transgenic Bt and conventional cotton in Australia. Environmental Entomology 34: 1224-1241.

Widner, W.R., Whiteley, H.R. (1990). Location of the dipteran specificity region in a lepidopteran-dipteran crystal protein from Bacillus thuringiensis. Journal of Bacteriology 172: 2826-2832.

Widner, W.R., Whiteley, H.R. (1989). Two highly related insecticidal crystal proteins of Bacillus thuringiensis subsp. kurstaki possess different host range specificities. Journal of Bacteriology 171: 965-974.

Zambryski, P. (1992). Chronicles from the Agrobacterium-plant cell DNA transfer story. Annual Review Plant Physiology and Plant Molecular Biology 43: 465-490.



Appendix A Definitions of risk analysis terms (* terms defined as in Australia New Zealand Risk Management Standard AS/NZS 4360:2004) Consequence

outcome or impact of an adverse event

Marginal: there is minimal negative impact Minor: there is some negative impact Major: the negative impact is severe

Event*

occurrence of a particular set of circumstances Hazard*

source of potential harm Hazard identification

the process of analysing hazards and the events that may give rise to harm Intermediate

the negative impact is substantial Likelihood

chance of something happening

Highly unlikely: may occur only in very rare circumstances Unlikely: could occur in some circumstances Likely: could occur in many circumstances Highly likely: is expected to occur in most circumstances Quality control

to check, audit, review and evaluate the progress of an activity, process or system on an ongoing basis to identify change from the performance level required or expected and opportunities for improvement Risk

the chance of something happening that will have an undesired impact

Negligible: risk is insubstantial and there is no present need to invoke actions for mitigation

Low: risk is minimal but may invoke actions for mitigation beyond normal practices Moderate: risk is of marked concern requiring mitigation actions demonstrated to be

effective High: risk is unacceptable unless actions for mitigation are highly feasible and

effective Risk analysis

the overall process of risk assessment, risk management and risk communication Risk analysis framework

systematic application of legislation, policies, procedures and practices to analyse risks

Appendix A Definitions of risk analysis terms (February 2006) 123


Risk assessment

the overall process of hazard identification and risk estimation Risk communication

the culture, processes and structures to communicate and consult with stakeholders about risks Risk Context

parameters within which risk must be managed, including the scope and boundaries for the risk assessment and risk management process Risk estimate

a measure of risk in terms of a combination of consequence and likelihood assessments Risk evaluation

the process of determining risks that require treatment Risk management

the overall process of risk evaluation, risk treatment and decision making to manage potential adverse impacts Risk management plan

integrates risk evaluation and risk treatment with the decision making process Risk treatment*

the process of selection and implementation of measures to reduce risk Stakeholders*

those people and organisations who may affect, be affected by, or perceive themselves to be affected by a decision, activity or risk States

includes all State governments, the Australian Capital Territory and the Northern Territory governments Uncertainty

imperfect ability to assign a character state to a thing or process; a form or source of doubt

Appendix A Definitions of risk analysis terms (February 2006) 124


Appendix B Summary of submissions received from prescribed experts, agencies and authorities4 on the application

All issues raised relating to risks to human health and safety and the environment were considered in the context of currently available scientific evidence that was used in the preparation of the consultation RARMP.

Issues relating to the Risk Assessment and where they have been considered:

• Molecular characterisation requirements (see Chapter 1)

• Human health effects (see Chapter 2)

• Hazards arising from occupational exposure (see Chapter 2)

• Development of herbicide resistant weeds (see Chapter 2)

• Development of resistance in target insects (see Chapter 2)

• Potential for altered pesticide metabolism due to expression of the cp4 epsps gene (see Chapter 2)

• Risks from expansion into new areas (see Chapter 2)

• Non-target toxicity (see Chapters 2 and 3)

• Risk of weediness (see Chapters 2 and 4)

• Risks arising from gene flow to other cotton plants including related species (see Chapters 2 and 4)

• Potential for gene stacking with other GM cotton crops (see Chapters 2 and 4)

• Environmental effects (see Chapters 2, 3 and 4)

Issues relating to the Risk Management Plan:

• Changed herbicide use patterns as a result of the release (APVMA assesses this issue—refer Chapter 5)

• Conditions required to manage risks from transport of GM cotton seed north of latitude 22° South (see Chapters 5 and 6)

Issues for future consideration:

• Consideration of gene stacking during possible future GM cotton releases of other herbicide tolerant cotton plants (would be considered in applications for these releases)

Issues that are outside the scope of assessments under the Gene Technology Act 2000:

• Marketing concerns (Outside the scope of the assessment)

• General ethical concerns (Outside the scope of the assessment).

4 Gene Technology Technical Advisory Committee, State and Territory governments, Australian Government agencies, the Minister for Environment and Heritage and local councils where the release may occur. For this application all councils south of latitude 22º South were consulted.

Appendix B Summary of agency submissions on the application (February 2006) 125


Appendix C Summary of public submissions received on the application

All issues raised relating to risks to human health and safety and the environment were considered in the context of currently available scientific evidence that was used in the preparation of the consultation RARMP.

Three submissions were received that raised the following issues:

• Use of GM products in human food and animal feed (see Chapter 2; FSANZ approval also required for food use—refer Chapter 5)

• Concerns about liability following contamination (Outside the scope of the assessment)

• Marketing concerns following contamination (Outside the scope of the assessment).

Appendix C Summary of public submissions on the application (February 2006) 126


Appendix D Summary of submissions received from prescribed experts, agencies and authorities on the consultation RARMP

None of the experts, agencies and authorities prescribed for consultation under the Gene Technology Act 2000, other than local councils, raised any issues on the RARMP relating to human health and saftey and the environment that required further consideration.

Twenty of the 590 local councils consulted in southern Australia provided submissions that raised a number of such concerns, as well as some matters that are outside the scope of assessments required by the Act but are the responsibility of other regulatory agencies. A summary of the submissions and how they were considered is provided below.

All issues relating to risks to human health and safety and the environment were considered in the context of currently available scientific evidence in finalising the RARMP that formed the basis of the Regulator’s decision to issue the licence.

Summary of consideration of advice received from local councils on consultation RARMP

Issues Consideration in RARMP

Summary of Consideration in RARMP

Unintended environmental effects

Chapters 2, 3 & 4 A range of potential environmental effects were examined in Chapter 2. The hazard identification process identified two risks: impacts on non-target invertebrates and weediness. Eight events that could lead to these adverse outcomes were assessed in more detail in Chapter 3 and 4. The risks were estimated as negligible for six events (1, 2, 3, 6, 7, 8) and low for two events (4 & 5). Management conditions have been imposed for the latter. The licence holder is also required to report any unintended effects of the release to the Regulator.

Toxicity to non-target organisms

Chapters 2 & 3 The toxicity of the introduced proteins for non-target organisms was discussed in Chapter 2. A risk was identified only for non-target invertebrates. The level of risk was estimated as negligible (Chapter 3, Event 1).

Risk of weediness Chapters 2 & 4 The potential for expression of the introduced herbicide tolerance gene or a combination of the herbicide tolerance and insecticidal genes to increase the spread and persistence of the GM cotton lines in the environment was identified as a risk in Chapter 2 and assessed in more detail in Chapter 4. The risks were estimated as negligible for five events (2, 3, 6, 7, 8) and low for two events (4 & 5). Management conditions have been imposed for the latter.

Transfer of introduced genes to organisms other than cotton

Chapter 2 Genetic incompatibility would prevent the transfer of the introduced genes to non-cotton plants, animals and microorganisms (Section 2.6).

Dissemination of GM seed

Chapters 2 & 4 The potential for spread and persistence of the GM cotton lines in areas north of latitude 22° South as a result of dispersal of seed from the GM plants was identified as a risk in Chapter 2 and assessed in more detail in Chapter 4. The risks were estimated as negligible for one event (6) and low for two events (4 & 5). Management conditions have been imposed for the latter that have been shown to be effective in relation to similar commercial releases of other GM cottons.

Appendix D Summary of agency submissions on the consultation RARMP (February 2006) 127


Monitoring of long-term effects that could be caused by use of GM cotton seed as stockfeed

— Results from feeding and toxicity studies considered in Chapter 2 do not justify long-term monitoring. The licence requires that any adverse effects must be reported to the Regulator. Similar GM cotton lines containing the same proteins have been commercially released since 2000 and 2002. GM cotton seed has been used in stockfeed since then with no reports of adverse impacts on animals/stock.

In general more research required on the effects of GMOs and GM food

RARMP The information available was considered adequate to assess the GMOs proposed for commercial release. Food Standards Australia New Zealand has concluded that food (oil and linters) derived from the GM cotton lines is as safe as food derived from non-GM cotton.

Changed herbicide use patterns as a result of the release

OSA The Australian Pesticide and Veterinary Medicines Authority (APVMA) is responsible for addressing this issue during its parallel assessment of Monsanto’s application to vary the registration of Roundup Ready® Herbicide for use on these GM cotton lines. If necessary, the APVMA will impose conditions on the use of the herbicide.

Labeling of food containing GM material

OSA This issue is outside the scope of the assessment. Food labelling is the responsibility of Food Standards Australia New Zealand.

Opposition to the import of GM plants and animals until proven safe

OSA The Australian Quarantine Inspection Service (AQIS) regulates the import of plants and animals. AQIS and the OGTR liaise in ensuring imported GMOs have the appropriate authorisations. While the GM cotton lines containing the original transformation events were imported from the USA, they have subsequently been crossed with elite locally-adapted cotton lines here in Australia.

Appendix D Summary of agency submissions on the consultation RARMP (February 2006) 128


Appendix E Summary of public submissions received on the consultation RARMP

The Regulator received 10 submissions from the public on the consultation RARMP. All were analysed in detail. Eight were from individuals and organisations with direct experience of cotton growing that supported the application. Two were from individuals that raised a range of concerns about the use of the GMOs.

All issues relating to risks to human health and safety and the environment were considered in the context of currently available scientific evidence in finalising the RARMP that formed the basis of the Regulator’s decision to issue the licence.

Abbreviations used: APVMA: Australian Pesticides and Veterinary Medicines Authority; Ch: Chapter; FSANZ: Food Standards Australia New Zealand; LC: licence; OSA: Outside the scope of the assessment Issues raised: EN: environmental risks; H: human health and safety; HR: herbicide resistance; IR: insect resistance; S: socio-political concerns; RC: regulatory coordination; Res: research; W: weediness. a Submission from: A: agricultural/industry organisation; I: Individual

Sub. No. Typea Summary of issues raised Issue Consideration of issue

I • Has been involved with and exposed to the cotton industry for more than 10 years and has seen the benefits of Roundup Ready® cotton. Believes that the enhanced characteristics of Roundup Ready Flex® will be of further value to farmers and the Australian Cotton Industry.

None Noted 1

• Believes that Roundup Ready Flex® will have clear benefits for growers including: • increased window of application without crop damage• economic benefit from effective weed control and

enhanced crop growth and potential yield • environmental benefit from reduced use of residual

herbicides • greater budgeting certainty • enhanced planting flexibility and potential water use

efficiency.

None Noted

• Knows a number of farmers who currently rely on the Roundup Ready® system and would value the introduction of Roundup Ready Flex®, especially in central QLD where crop growth is rapid making the current application window somewhat restrictive.

None Noted

• Is in support of the Roundup Ready Flex® application and the process that OGTR follows with these applications, and respects the need for scientific rigour when reviewing such applications to ensure that new products are safe.

None Noted

2 I • Writes in support of Monsanto’s application and of the OGTR process

None Noted

Appendix E Summary of public submissions on the consultation RARMP (February 2006) 129


• Understands the rigorous scientific assessment that needs to take place in order for the product to be approved.

None Noted

• Has been growing cotton for 30 years in the central QLD district and says that: “The introduction of Roundup Ready to our farming system has provided us with easy, flexible weed control. Along with other cotton producers we have experienced increased benefits from using the technology allowing greater flexibility controlling weeds, and less reliance on residual herbicides providing a healthier farming system”.

None Noted

3 I • Has been a user of Roundup Ready® cotton technology for 4 years and now considers it to be an integral part of the weed management system.

None Noted

• Targets different species with different control options and Roundup Ready® cotton has given him the ability to be flexible and at times decrease the use of residual herbicides.

None Noted

• The use of Roundup Ready Flex® cotton will further enhance this flexibility of weed management and at times further decrease the use of residual herbicides.

None Noted

• Some other benefit of Roundup Ready Flex® technology are reduced levels of use of more toxic chemistry, and the convenience of using familiar chemistry.

None Noted

• Fully supports the process the OGTR goes through, and understands and supports the scientific approach that must be taken to keep the assessments rigorous and unemotional so that products like this can be approved.

None Noted

4 I • Expresses wholehearted disapproval of genetically engineered (GE) produce, in this particular instance, GE cotton and protests against Monsanto gaining approval for this application.

None Noted

Reasons for the protest are: • GE foods have not been proven to be safe H Ch 2; Assessment of the

safety of GM food is the responsibility of FSANZ.

• Scientists cannot predict the long-term impact (on our health or the environment) of laboratory constructed foodswhich could never have occurred in nature

H, EN Ch 2, 3 and 4

• Genes from bacteria, viruses, plants and animals have been inserted into soybeans, canola, corn and cotton, and the problem with GE produce is that once it’s released into the environment it cannot be recalled.

EN Ch 2, 3 and 4

• Quotes Dr Judy Carmen: “Concerns have been raised internationally over the safety of consuming GE foods. Some of these concerns include the use of antibiotic resistant genes in GE plants, which may lead to antibiotic resistance in us. Another concern is our potential exposure to unfamiliar or unexpected proteins, toxins and allergens through eating GE foods….An independent reviews of reports published by FSANZ has concluded that tests are inadequate, and that GE foods have never been tested on humans….”

H Ch 2; FSANZ



• Companies like Monsanto often tell us we need their technology to feed the world, but this is just propaganda. The world is producing more food then ever in history and the problem is equitable distribution.

S OSA

• Urges the Regulator to stop Monsanto from experimenting with our environment and our health.

H, EN Ch 2, 3 and 4

5 I Supports the OGTR approval process, including the extensive scientific assessment, for Roundup Ready Flex® cotton.

None Noted

Comments on this GM cotton’s value to him as a cotton grower:

None Noted

• For the past 5 years Roundup Ready® cotton has been an essential part of the weed control programme on his farm, providing easy cost effective control on a broad spectrum of weeds.

• Roundup Ready Flex® cotton would decrease late season reliance on environmentally damaging herbicides and give increased flexibility and convenience in weed management programme.

• To remain competitive in a global market place Australian cotton growers need to have access to technologies which reduce production costs and environmental impacts.

6 A • The industry has successfully managed the introduction of Roundup Ready® cotton over the last 6 years and further health and environmental benefits can be gained from the introduction of Roundup Ready Flex® cotton.

None Noted

• Roundup Ready® cotton has dramatically reduced the need for ‘hand weeding’ and the introduction of Roundup Ready Flex® cotton will further reduce the need for hand weeding in cotton.

None Noted

• In the past it has been difficult for growers to achieve fully effective late season weed control. This has resulted in the need for hand chipping (hoeing) of weeds as the cotton crop approached full canopy closure. Potential exposure of ‘chippers’ to insecticides used to control pests and ultra-violet radiation (sun exposure) was an on-going concern for growers and the industry.

None Noted

• Roundup Ready® cotton has reduced the use of residual herbicides some 31% (Cotton Research and Development Corporation, 2005). This has contributed to favourable environmental outcomes in reduced levels of residual cotton herbicides in river systems (CRDC, various studies) and likely positive benefits to soil biology. It has also lead to an overall reduction in tillage and improved tillage ‘timing’ enabling improved soil structural and conservation practices.

None Noted

• Believes the key issues surrounding the health and environmental issues of the introduction of Roundup Ready Flex® cotton have been addressed by the RARMP and appreciates the rigorous scientific assessment that has taken place.

None Noted

7 I • From time to time warmer and wetter conditions may occur south of latitude 22ºSouth which offer enhanced conditions for weed development in areas chosen for planting of the GMOs.

W Ch 2 and 4



• Short term field trials and subjective observations do not seem adequate tests for potential weediness.

Res, W The information available was sufficient to conduct the risk analysis for this release under the Gene Technology Act 2000 (the Act). Potential weediness of the GM cotton proposed for release has been comprehensively assessed. Similar GM cotton lines have been commercially released since 2000 and 2002 with no adverse impact.

• The effectiveness claimed for subsequent treatment of emergent weeds by whatever process would in practice be dependent on a great variety of factors (e.g. where and when it occurred, how soon it was observed, the approach of the farmer or licensee, notification to the OGTR etc.) and may therefore be less than anticipated.

W Ch 4

• Two copies of the cp4 epsps gene construct imply doubling herbicide resistance to allow multiple sprayings of the cotton. The RARMP notes that introgression with native cotton species, non-GM, naturalised or other GM cotton varieties is unlikely but possible. (In the latter case there seems to be potential for tripling of the herbicide resistance gene). If this enhanced degree of resistance to glyphosate did introgress into related species, this would compound the weediness issue mentioned above (and potentially cost us a most useful and fairly benign weedicide that may not be replaceable by something equally benign).

HR, W

Ch 4, Event 7, multiple copies of herbicide tolerance genes would not alter limited effectiveness of glyphosate in controlling established cotton plants. The APVMA will apply registration conditions regarding the herbicide use and resistance management (refer Technical Summary and Ch 2 Section 2.10.3 ).

• Experience suggests that there is a high probability that the development of insect resistance to the Bt toxins is just a matter of time, however well managed?

IR Responsibility of the APVMA. To date, the resistance management plan implemented by the APVMA has shown to be effective. Review of its effectiveness will be ongoing.

• As the RARMP notes, if managed appropriately, risk of dispersal of seed during transport, or by events such as extreme floods or winds, is low, but over time it will certainly occur as other examples have shown and appropriate management will only delay the event. Subsequent events are not predictable and this is the great problem with such major environmental experiments.

EN The RARMP specifically considered the effect of the dispersal of seed via these routes in Ch 4 and concluded the risk of weediness would be low for transport (Event 4 - conditions imposed) and negligible for floods (Event 6).



• Do the risks outweigh the benefits? Has this been realistically assessed?

OSA The Act requires the Regulator to assess risks to human health and safety and the environment; benefits are excluded from consideration.

8 I • Supports OGTR and its scientific assessment process that has to be carried out for such a product to be approved.

None Noted

• Has been growing Roundup Ready Cotton for the last four years and has found it an invaluable tool in his weed management strategy. For cotton growers, Roundup Ready Flex Cotton would add flexibility and would form part of future weed management programs.

None Noted

9 A • Agrees with the OGTR’s assessment that GM cotton lines carrying the Roundup Ready Flex genes pose negligible to low risk to people and the environment.

None Noted

• Believes that the commercial production of GM cotton in Australia (insect tolerant and herbicide tolerant GM cotton for ten and five years, respectively) has contributed significantly to enabling Australian cotton growers to achieve reduced chemical inputs (ie insecticide and residual herbicides).

None Noted

• Herbicide tolerant GM cotton has contributed to changes in weed control practices including reduced residual herbicide use and reduced cultivation, both of which can have short-term negative impacts on soil microbial bio-diversity.

None Noted

• The cotton industry is aware of and has demonstrated the discharge of its responsibilities in relation to the use of GM technology. It has been actively involved in the development of the resistance management plans for both insects and weeds that are associated with commercial production of these technologies. The cotton industry also seeks to minimise off target impacts of all the pesticides used by cotton growers by encouraging them to participate in the industry’s own environmental management system – the Best Management Practices program.

None Noted

• Commends the rigour with which the OGTR conducts its risk assessments.

None Noted

10 A • Supports the OGTR’s extensive and thorough assessment of the application and the general conclusion of the RARMP.

None Noted

• Highlights the benefits that have already been accrued due to the release of GM cotton in Australia such as: o significant reduction in pesticide use and a shift in the

types of pesticides used o reduction in residual herbicide applications o reduced tillage, maintaining soil structure and

decreasing erosion.

None Noted



• Cotton has been grown in NSW and QLD for over 40 years and has not established itself as a weed despite the seed cotton and fuzzy seed sometimes spilling on farms and roadsides during processing and transport. This has not changed with the initial release of herbicide tolerant cotton.

None Noted

• Notes that the Regulator has a coordinated decision making policy. In the past this has meant the Regulator has not approved the use of a herbicide tolerant crop until the herbicide has been registered for use in that crop. However, the seed can still be planted and managed conventionally by growers without the use of a herbicide in the crop. In fact, some growers may choose to do this just to take advantage of varietal characteristics not related to the GMO trait.

RC Noted

• The harvest of a crop, monitoring of locations and use of seed after harvest under current trial restrictions for Roundup Ready Flex cotton add substantial costs compared to non regulated activities. Yet these activities are completely independent of herbicide registration status, therefore if the Regulator determines the technology is safe, a licence should be issued at that time, and leave herbicide use issues to the APVMA to regulate. This would avoid the unnecessary cost which would be passed back to the primary producers in the price of seed.

RC Noted


risk assessment and risk management plan for dir … · risk assessment and risk management plan...

Documents