Plant Cell Reports (1994) 14:69-74 Plant Cell Reports ~ Springer-Verlag 1994

A biolistic approach for the transfer and expression of a gusA reporter gene in embryogenic cultures of Pinus radiata

Christian Walter 1, Dale R. Smith t, Marie B. Connett*, Lynette Grace 1, and Derek W.R. White 2

1 New Zealand Forest Research Institute, Private Gag 3020, Rotorua, New Zealand 2 AgResearch Grasslands, Private Bag 11008, Palmerston North, New Zealand * Present address: Tasman Forestry Ltd. Te Teko, New Zealand

Received 4 February 1994/Revised version received 6 July 1994 - Communicated by J. Widholm

Summary. The biolistic| particle delivery system was used for the delivery of DNA into embryogenic tissue culture cells of Pinus radiata D. Don. Several experiments with varying parameters were performed to increase the delivery efficiency. Six different controlling elements were cloned upstream of the B-glucuronidase coding sequence (gusA reporter gene) and transient expression of the gusA reporter gene was compared three days after bombardment. The results clearly indicate a decrease in transient expression as follows: pEmu-derivatives with the ocs-enhancer-element > 2x CaMV 35S (with Kozak consensus-sequence) > 2x CaMV 35S (without Kozak consensus sequence) > CaMV 35S (with Kozak consensus-sequence) > CaMV 35S (without Kozak consensus sequence). Time course experiments monitoring gusA expression showed a significant decrease in the number of blue spots 10-14 days after bombardment. A few blue clumps however, were still detected 35 days after shooting. Embryo initials expressing the gusA gene in all cells were also detected. The results suggest that it will be possible to develop a reliable biolistic protocol for stable integration of genes into Pinus radiata embryogenic cultures which are capable of plant regeneration.

Keywords: Pinus radiata - Biolistic| - Transformation.

Abbreviations: ccc, covalently closed circular DNA; lin, linearised DNA; E, restriction enzyme Eco RI; Sph, restriction enzyme SpH I; 2,4-D, 2,4- dichlorophenoxyacetic acid; BAP, 6-benzylaminopurine.

Introduction

Pinus species are the most widely distributed trees in commercial forestry use around the world, supplying, with other conifers, the world's main sources of plantation-

grown wood. Pinus radiata, one of the most important species, is extensively grown in Chile, New Zealand, Australia, and other places (Burdon, 1994). Conventional selection and breeding techniques have been applied to Pinus radiata in intensive research carried out for over 40 years, resulting in trees with improved growth rate and stem form, increased resistance to pathogens, and better adaptability to certain areas and climates. Traditional breeding results in the recombination of a wide

spectrum of genes influencing these and other quantitative traits rather than specific alterations in only the trait of interest. Consequently, genetic improvement through molecular techniques has been considered for a wide range of conifers. This has led to the establishment of transformation protocols in conifer species using a naturally occurring gene transfer system involving Agrobacter ium tumefaciens

(Sederoff et al. 1986, Ellis et al. 1989, Morris et al. 1989, Stomp et al. 1990, Bergmann and Stomp 1992) as well as direct gene transfer methods (Wilson et al. 1989, Bekkaoui et al. 1988, Tautorus et al. 1989, Goldfarb et al. 1991, Ellis etal. 1991, Robertson etal. 1992, Ellis etal. 1993). Various promoters to drive gusA expression have been tested in spruce (Bekkaoui etal. 1990, Ellis etal. 1991, Duchesne et al. 1991, Charestet al, 1993).

Pinus radiata has been transformed by wild type Agrobacterium (Stomp et al. 1990; Bergmann and Stomp 1992), but plants were not regenerated. The transformation of plant cells via particle bombardment (Klein et al. 1987) has proven to be a valuable tool for gene expression studies and stable transformation of monocotyledonous ( Gordon- Kamm etal. 1990; From etal. 1990; Chowdhury and Vasil 1992, Cao et al. 1992) as well as dicotyledonous plants (McCabe et al. 1988, Finer et al. 1990).

Transient expression of the gusA reporter gene in non- embryogenic suspension culture cells ofPinus radiata after

Correspondence to. D.R. Smith

70

part ic le acce lera t ion has been repor ted (Campbel l et al.

1992), but the e f f i c iency was low and stably t ransformed

tissue cou ld not be regenerated,

In addi t ion to ef f ic ient D N A del ivery systems, tissue

cul ture p ro toco ls to regenera te who le plants f rom single

t ransformed cei ls and appropriate se lect ion reg imes are

n e c e s s a r y to p r o d u c e f e r t i l e t r a n s g e n i c p l an t s . A n

embryogen i c t issue cul ture pro tocol for Pinus radiata has

been deve loped at N Z F R I (Smi th 1994) which a l lows the

regenera t ion o f plants f rom single embryogen i c cells.

In this paper the opt imisa t ion o f the bombardmen t protocol

for the t ransient express ion o f genes in Pinus radiata

e m b r y o g e n i c ce l l s w i th a h e l i u m gas d r iven par t ic le

a c c e l e r a t o r ( D u P o n t P D S 1000He) is r epor ted . T h e

ident i f icat ion o f p romo to r -gene constructs that y ie ld high

numbers o f express ing cel ls is also described. Embryo -

initials express ing the g u s A reporter gene could be detected

I0 days after bombardmen t .

Material and Methods

Tissue. Embryogenic tissue cultures ofPinus radiata used in this report were originally initiated from immature zygotic embryos. Tissue was maintained in an undifferentiated state on Embryogenesis Medium (EM, Smith 1994), ideally with embryos never developing past the eight-celled stage before dissociating into simple embryogenic initials (see Fig. 3a). These embryos have large single suspensor cells attached to the embryo initial. Although further development of suspensors to form multiple stranded structures was not encouraged on this medium, the suspensor cells are highly vacuolated and constitute the bulk of the tissue mass. Scattered throughout the tissue mass are 8330-2_583 competent embryogenic initials per gram fresh weight of tissue (Smith, D.R., unpublished). Over 700 plants have been produced from the cell line used in these experiments.

For every transformation experiment, 300 mg tissue (fresh weight) was suspended in 1.5 ml liquid EM and plated onto solid Embryo Development Medium (EDM, Smith 1994), gelled with 3 gL 1 Gelrite and incorporating 1 mg L-I 2,4-D and 0.6 mgL -1BAP, as well as 100 mgL "1 glutamine in a Falcon petri-dish. Aliquots of a single uniform suspension ofembryogenic cells were used across all treatments. Mannitol or sorbitol were added to the medium at a final concentration of 0.25M; sodium-thiosulfate (STS) was used to give a final concentration of 0.2%. The tissue in iidded dishes was left in the laminar flow over-night to evaporate excess water and was bombarded the next day.

Vectors and molecular cloning. The plasmids p6ARE4OCSAAdhIGN (pEmuGN) and p4OCSD35SIGN (Last etal. 1991) were obtained from R. Brettell, CSIRO Division of Plant Industry, Canberra, Australia. The pJIT vectors were obtained from the John lanes Institute of Plant Science Research, Norwich, UK. The vector pANDY 3 was cloned by A.G. Griffiths (unpublished) and contains the npt H gene under the control of a nos promoter. Molecular cloning was performed using standard procedures according to

Maniatis et al. (1982). DNA for bombardment of plant tissue was isolated using a Magic-Maxi-Prep DNA isolation kit (Promega Corporation, Madison, Wisconsin, USA ). Purification of DNA from electrophoresis gels following restriction endonuclease digestion was performed with a Geneclean (BIO I01) kit. The plasmid vectors used in this study were constructed as follows:

pCW103 and pCWI22: The nptll gene with CaMV 35S promoter and CaMV polyadenylation signal was excised from pANDY3 using Kpnl and

cloned into pJIT65 (pCW 103) or pJIT 166 (pCW 122 ), respectively. pCW5 and pCW6: The gusA gene without upstream and downstream controlling elements was excised with EcoRI I HindlII from pJIT 166 or pJIT65 and cloned into pJIT82 to obtain pCW5 and pCW6, respectively.

Particle bombardment. Pinus radiata tissue was bombarded with DNA- coated gold particles prepared according to a standard protocol (Sanford et aL, 1993) or with the following modifications as mentioned in the Results section: 1M Ca(NO3) 2 was used instead of 2.5 M CaCi2; or spermidine was omitted from the coating procedure. A DuPont PDS 1000He biolistic| delivery system was used in all experiments. Physical parameters were optimised to increase numbers of transiently expressing cells per plate. The following conditions were found to be superior and are currently used as a standard bombardment protocol: Rupture disc-pressure: 1100-1550 psi; gap-distance from rupture disc to

macrocarrier: 6 mm; macrocarrier travel distance: 16 mm; microcarrier travel distance: 6 cm. Gold particles supplied by Aldrich Chemicals (size 1.6-3 I.tm) were found

to yield higher transient expression efficiencies than the particles supplied by Bio-Rad (l[.tm or 1.6 p.m).

fl-glucuronidase assay. Histochemical Gas Assays were performed either with a staining solution according to Kozugi et al. (1990), modified with 1% DMSO but no mannitoi, or according to Charest etal. (1993), modified with a double strength buffer. Tissue was flooded with 1-2 ml of the staining solution 3 days after transformation and subsequently incubated at 37~ for 24-48. Blue spots (cells or clumps of cells) were counted under a stereo microscope. Clumps of blue cells were scored as a single spot. Average numbers of blue spots per plate were calculated from 6-18 plates. Transformation efficiency was calculated as the mean number of blue spots per plate. Treatments were repeated at least twice, bat usually three times (blocks)

with three, or six dishes (replicates) per block. Data were analysed as an unbalanced block design by ANOVA, using the GLM procedure of SAS (SAS Institute 1985). It was necessary to use the square root transformation of data to satisfy the assumptions of analysis of variance (Box and Hunter

1978).

Results

a: Optimisat ion o f bombardmen t parameters

Bombardment condit ions to opt imise the transient expression

in embryogen ic Pinus radiata t issue were eva lua ted and are

s u m m a r i s e d in the s t anda rd b o m b a r d m e n t p r o t o c o l

m e n t i o n e d in M a t e r i a l and M e t h o d s . C h a n g e s to

bombardmen t protocols which have been publ i shed by

others (Perl et aL 1992, Kle in et al. 1988b, Tautorus et aL

1989) have been tested as well. Dur ing bombardmen t o f cel ls on nutr ient agar, the impact

o f the he l ium shock-wave and the microcarr iers penetra t ing

the cell wails appeared to displace the cel ls and consequent ly

the penetrat ing force was reduced. To anchor the cel ls on the

medium, they were suspended on a 7 0 m m diameter Whatman

filter paper on the medium. Results compar ing transformation

eff ic iencies with and without fi l ter paper are g iven in Table

1. There is clearly a benef ic ia l effect o f using a fi l ter paper

support upon bombardmen t o f Pinus radiata embryogen i c

tissue. Linearisat ion o f D N A prior to t ransformat ion was pe r fo rmed

in some experiments. Tautorus et al. (1989) found a 2.5 times higher level of cat expression in electroporated jack pine (Pinus banksiana) protoplasts after electroporation compared to using circular DNA. In back spruce (Picea mariana) however, they reported a decrease of 28% in cat expression when using linear compared to ccc-DNA. Iida et al. (1990) found no apparent influence of the DNA- conformation on transient expression after bombardment of tobacco suspension cultures with a gusA gene. In our experiments, the plasmid vector pEmuGN was used to ascertain any differences using covalently closed circular (ccc) DNA versus linearised DNA. Equimolar amounts of either the linear or ccc -DNA were used in bombardment experiments. The results presented in Table 1 indicate that linear DNA gives a significantly higher transient expression frequency than ccc-DNA. No significant differences could be observed using the complete linearised vector (EcoRI digest) or a mixture of two fragments, one containing the reporter gene with its controlling elements and the other containing the vector (EcoRI / SphI digest).

Mannitol decreases the osmotic pressure in cells and has been reported to give a higher survival rate of transformed maize cells, perhaps by preventing possible leakage after

bombardment (Per et al. 1992). Data from previous experiments with embryogenic Pinus radiata tissue indicated sorbitol to be a better osmoticum for those cells than mannitol (Smith, D.R. and Hargreaves, C., unpublished). The efficiency of transformation with ccc-DNA (pEmuGN) using media with and without mannitol or sorbitol was determined (Table 1). Both had a beneficial effect and sorbitol is therefore currently used in our transformation protocol. The use of sodium-thiosulfate (STS), which is an inhibitor of ethylene action in the culture medium (Perl et al., 1992) seemed to decrease the efficiency (Table 1), but the difference is not significant. Modified microcarrier coating protocols where spermidine is omitted, or Ca(NO3) 2 is used rather than CaCI 2, decreased the transformation efficiency in our system as well (Table 1).

Using the optimal parameters described, we currently obtain 100-150 gusA expressing spots per plate using constructs with 2x CaMV 35S - Kozak-positive controlling elements (pCW 122) or pEmu derivatives. Transient expression was primarily observed in the embryo initials. Where detected in suspensor cells, the staining appeared to have diffused from the adjacent embryogenic initials.

71

Table 1. Optimisation of bombardment parameters for the transient expression of gus in embryogenic Pinus radiata tissue. Tissue was bombarded using the standard protocol with the alterations indicated.

Plasmid DNA Bombardment conditions Av. no. of blue spots !

DNA Sorbitol Mannitol Filter STS used Spermidine Ca-source

,EmuGN ~ ~ ~ ~i~i~::;.iii i i ~ ~i~ ~i~ ~.ii~ ~:.~i~ + CaCl 2 8.8 a ?i if::: ::? :i:;::i:i i:i:i i:i i i:i<:i:? :i:i

)EmuGN :: ii:.!i !i iiii?iii~ii ~::::ii!iii::i iii iiiiil + CaCl 2 21.0 b ii: ilii i:i :!i:: i.i.i i.iii.ii ; . ,ii:i!:

,emu N + Ca% 22.2 h 5:::5 "::>':5:::::: : : . : . ::x::

:ii..iii .'ii:i iiiii.iii:i iii:i i:ii iii:: :iii:.:~!i .:ii :iii i:iii:i ii!:i iili:i :iiii.i i i iii

,EmuGN ccc i.iiiiii:i:iii:i!i::!ii::ii::i:!iiii :ii!i.:ii!i:!i iiiiii.:.iiiiiiiii.:iiiii: i:{iiii !iili: !iiiii::::.iiiii::i:.:i~ :i.!ii!~i::):iiiii)!iiiii !i :i::ii: + + CaCl 2 49.2 a . . . . . . . �9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

~EmuGN ccc ii::i::i!ii !i!ii.i:.ii!:.:.!~!ii::i:!:!i:. !:!ii) ::ii::i::iii.:.iiii !iil !iii! !ii~:::.ii{i !ii}:.!!!ii!:!:::i!i + + CaCl 2 122.5 b

,EmuGN ccc ii~i~ ~ ~i~ ~ ~ ~ ~iili ii! :i~!:.:i~i:::~i~:::~.i::~::~ ~ ~!~ ~ ~i):~i:?.iii::i::~ iiiii:i~i~::~i~ii i~[ + + CaCI 2 70.8 c

)EmuGN ccc + i ii!i:i:!::i ::iiii: ii:ii!:ii(i:i ::i:iiii.i:ii i i iiiii il + CaCI 2 33.0 a

)EmuGN ccc + :ii: ii iiiiiiii i:ii!iii: iiiii~;:iiii!ii :i:.ii!;i; :!i !.iii + CaCI2 75.5 b

~EmuGN ccc + + :: :: ii!::i:iiii::iiii iii~ :iiii:i:ii :::iii ii::i :: ::i + CaCl 2 76.0 a . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . i iii: i.i!.i :i~i::iii i ii~:i ~:ii:~?. !i.i i~

)EmuON ccc + + :i::ii}i i::!ii::i}ili ii iii~i:ii~:i:::.iiii:iiii i i:iii i i + CaCl 2 53.0 b ::::. ::: .:::. : . : : : z :5:.:.::::.::: :::

:iii!:. iill .iili iii i i i i iiii iiil i !i...i:

,CW 120 ccc + + .i.ii:iii{ii:iiiii ii iiiii!:::ii iii!i-:iiiiiiiil ii iiii:ii ::iiiiiiii i i CaC12 14.5 a

pCW 120 ccc + + i %iii.:il ;iiil ~!ii::i::.:iiiiii:ii;:iil CaC12 37.3 b

pCW 120 ccc + + + !ili::J:i~i~:~.~i~i :~ :: i::~ i i :~ :~ 67.3 a . . . . . . . . , . . . . . . . . . . . . . . .;..;...;.

pCW 120 ccc + + + iii:!: ii 60.3 a

Results are mean values of 6-12 replicates. Highlighted areas indicate the experiments. Treatments within an experiment not bearing the same letter are different at the 1% level of significance.

Abbreviations: ccc: covalently closed circular DNA; lin: linearised DNA; E: restriction enzyme Eco RI; Sph: restrictin enzyme Sph I; Av.n.: Average number.

72

b: Comparison of promoters for gusA expression in Pinus radiata embryogenic tissue

The gusA reporter gene used in our experiments was controlled by one of six different promoter constructs: CaMV 35S with or without Kozak-consensus sequence for translation initiation (Kozak, 1989), double CaMV 35S with or without Kozak-consensus sequence, p6ARE4OCSAADHIGN (pEmuGN); andp4OCSA35SIGN. The latter two (pEmu-derivatives) are artificial promoters both containing the ocs enhancer element (Ellis et al. 1987) and are reported to yield high expression of a reporter gene either in monocots or in dicots (Last et al. 1991). Equimolar amounts of each of the six plasmids (ccc- DNA)

were coated onto gold particles and bombarded into embryogenic cells and the number of blue spots determined three days after bombardment. The results (Fig. 1) indicate a significant difference in expression of the reporter gene using different upstream controlling elements. The 2x CaMV 35S promoter yielded approximately six times as many blue cells as the single CaMV 35S promoter. Using the Kozak consensus sequence situated around the ATG start codon of the gusA gene, we obtained an approximately 2.5-fold increase in numbers of blue spots when either the 35S or the double 35S promoter was used to control gusA expression. Furthermore, cells bombarded with the 2x 35S-Kozak- positive construct showed considerably stronger gusA expression (cells showed a dark blue stain compared to a faint blue with single 35S or double 35S-Kozak-negative; data not shown). Using p4OCSA35SIGN, a further increase in delivery efficiency compared to the Kozak-positive construct was observed, but there was no significant difference between the two pEmu constructs. The results were obtained using three different DNA preparations of

each construct and were consistent when repeated.

c: Time course of expression of the gusA reporter gene

Thirty plates with embryogenic tissue of Pinus radiata on EM medium were bombarded with ccc-DNA (pEmuGN) and samples of three plates were examined every 3-4 days to obtain a time course for the expression of the gusA gene in cells. The number of cells expressing gusA decreased after 6-15 days, but reached a plateau thereafter, and expression could be detected after 35 days (Fig. 2).

In two cases, gusA stain was found in all cells of the embryo indicating totally transformed embryos, presumably arising from a single transformed cell. Fig. 3b shows an example of such an embryo initial, photographed ten days after bombardment.

Discussion

In this paper an optimised protocol for the biolistic delivery of DNA into embryogenic cultures of the important forestry species Pinus radiata is presented. An embryogenic tissue culture protocol has been used to generate tissue which can be transformed with a gusA reporter gene under the control of several promoter constructs. Several protocols published elsewhere to increase the number of gusA expressing cells have been tested in this study. Some of them have a beneficial effect in our hands (use of filter paper, use of linear instead of circular DNA), while others were shown to cause a decrease in efficiency (omission of spermidine in the coating protocol, use of STS in the bombardment medium, Ca(NO3) 2 as a Ca-source for coating).

Plasmid Promoter Construct Av. no of blue spots

pCW 5 35S K-

pCW 6 35S K+ ~ ACCATGG[

pCW 103 2x 35S K-

pCW 122 2x 35S K+ ~ ACCATGG I

pEmuGN 6ARE4OCSAADH -~6ARE4OCSAADH ]

p4OCSA35SIGN 4 O C S A 3 5 S --~4OCSA35S ]

~ C a M V po lyA~ 8.7 a

~ ~ CaMV polyA }--- 25 .6b

~ CaMV polyA ~ 50.8 c

~ - { NOS polyA ~ 171.0 d,e

~ . ~ u s ~ - - ~ NOS polyA ~ 187.1 e

Fig. 1. Different promoter constructs driving gus expression in Pinus radiata cells. Embryogenic cells supported on Whatman filter paper on EM medium containing sorbitol were bombarded with equimolar amounts of circular DNA. Transformation efficiency is calculated as the mean number of blue spots per plate. Three DNA preparations were made of each vector, and used in bombardment experiments with six replicate plates per preparation (12-18 replicates per experiment). Transformed data were analysed by ANOVA, using the GLM procedure in SAS. Treatments not bearing the same letter are different at the 1% level of significance.

Abbreviations: 35S: CaMV 35S promoter; 2x35S: 2xCaMV 35S promoter; Gus: fl-glucuronidase coding sequence; K-: Kozak consensus sequence not present; K+: Kozak consensus sequence present; ACCATGG: Kozak consensus sequence; CaMV polyA: CaMV polyadenylation signal; NOS poly A: Nopaline - Synthase polyadenylation signal.

15 o

1 0 - - , - ~ o

0

\ I I

0 5 10 I I I I

15 20 25 30 35 Days after shooting

Fig. 2. Expression of the gus reporter gene in embryogenic cells of Pinus radiata, monitored over a time period of 35 days after bombardment with circular pEmuGN DNA on a Whatman filter paper support on media with sorbitol. Number of gus expressing cells decrease, but a few cells expressing gus could be found after 35 days.

The double CaMV 35S promoter used in this study has been reported to be a stronger promoter for white spruce than the single CaMV 35S promoter (Bekkaoui et al. 1990). Their finding matched our results, with a six- fold increase in numbers of transiently expressing cells after bombardment with a double CaMV 35S promoter compared to single CaMV 35S. Furthermore, the Kozak-consensus sequence, which is reported to provide a better ribosome binding site around the ATG codon, potentially enhancing translation (Kozak 1989), has a beneficial effect in our system too. Interestingly, both pEmu derivatives used in this study give an even higher number of cells expressing gusA in Pinus

radiata embryogenic tissue. Both contain the ocs enhancer element, which has been shown to enhance gene expression

73

in tobacco and maize (Ellis et al. 1987), but has not been used in g y m n o s p e r m s before . The cons t ruc t p6ARE4OCSAADHIGN is reported to be a strong promoter for monocotyledonous plants, whereas p4OCSA35SIGN is found to be a strong promoter in dicotyledonous plants. Our findings that both of them lead to strong expression in Pinus

radiata (even stronger than the double CaMV 35S Kozak- positive constructs) suggests that the gymnosperm transcription system recognises both monocotyledonous and dicotyledonous promoters. We expect that other strong monocotyledonous or dicotyledonous promoters (such as the rice actin promoter, McElroy et al. 1990), could be used to control the expression of genes in Pinus radiata as well. Duchesne and Charest (1991) have demonstrated that the En promoter directs a 4.5 fold higher gusA activity than the35S promoter in Picea mariana callus.

The expression of a gusA reporter gene was monitored in embryogenic cells over a time period of 35 days by scoring the number of blue spots on each plate. A considerable decrease in expression after an initial phase with high numbers o f g u s A expressing cells was found. Such transient expression is commonly found after gene transfer. However, the number of blue spots showing gusA expression levelled off to a plateau by 15 days after the actual bombardment event. Cells that were still expressing at 35 days represented up to 3% of these expressing at day 1 after bombardment. Previous authors have reported that the number of transiently expressing cells that give rise to stable integration events was 0.4% for Picea abies tissue (Robertson et al. 1992) or 4.8% for Nicotiana tabacum (Klein et al., 1988a). This

F i g . 3 . a: Embryogenie Pinus radiata initial after suspending in liquid EM and plating onto solid EDM. Bar represents 50~tm. b: Embryo initial of Pinus radiata with an embryo head consisting of 8-9 ceils attached to a suspensor cell. All cells in the embryo head express the gus reporter gene. Embryogenic cells had been shot with circular pEmuGN DNA on a filter paper on medium without sorbitol. The standard coating and bombardment protocols were used.

74

existence of this long-term expression, plus the observation of embryo initials in which all the cells expressed gusA (Fig. 3b), provide indicative evidence for stable integration. Ellis et al. (1993) reported that transformation of Picea glauca by particle acceleration was dependent on the developmental stage of somatic embryos. In their work, transient and stable expression were reported for somatic embryos where an embryogenic axis was well established. We report here transient expression in simple embryogenic initials, at a frequency in excess of 500 blue spots per gram of embryogenic tissue. Experiments are underway to develop new vectors for the

stable transformation of embryogenic Pinus radiata tissue involving several resistance genes under the control of the pEmu promoter or the CaMV 2x35S (with Kozak-consensus sequence) promoter.

Acknowledgments. The authors are thankful to R. Gardner and R. D. Appleby for their critical review of the manuscript and helpful discussions. We wish to thank M. Houg for data analysis and A.G. Griffiths for the supply of pANDY3 prior to publication.

References

Bekkaoui F, Pilon M, Raju DSS, Crosby WL, Dunstan DI (1988) Plant Cell Reports 7:481-484

Bekkaoui F, Datla RSS, Pilon M, Tautoms TE, Crosby W, Duustan DI (1990) Theor. Appl. Genet. 79:353-359

Bergmann BA, Stomp A-M (1992) Phytopathology 82(12): 1457-1462 Birch RG, Franks T (1991) Aust. J. Plant Physiol. 19:453-469 Box GEP, Hunter WG, Hunter JS (1978) Statistics for experimenters, an introduction to design, data analysis, and model building. John Wiley and Sons.

Burdon RD (1994) chapter no 2 in: Ecosystems of the world. Tree crops. Pinus radiata. Elsevier (in press).

Campbell MA, Kinlaw CS, Neale DB (~992) Can. J. For. Res. 22: 2014- 2018

Cao J, Duan X, McElroy D, Wu R (19~2) Plant Cell Reports 11:586-591 Charest P J, Calero N, Lachance D, Datla, RSS, Duchesne LC, Tsang EWT

(1993) Plant Cell Reports 12:189-193 Chowdhury MKU, Vasil IK (1992) Plant Cell Reports 11:494-437 Duchesne LC, Charest PJ (1991) Plant Cell Reports 10:4 191-194. Ellis JG, Llewellyn D J, Walker JC, Dennis ES, Peacock WS (1987) EMBO

Journal 6(11 ) 3203-3208 Ellis DD, Roberts D, Sutton 6, Lazaroff W, Webb D, Flinn B (1989) Plant

Cell Rep. 8:16-20 Ellis DD, McCabe D, Russell D, Martinell B, McCown BH (1991) Plant

Molecular Biology 17:19-27 Ellis DD, McCabe DE, Mclnnis S, Ramachandran R, Russell DR, Wallace

KM, Martinell B J, Roberts DR, Raffa KF, McCown BH (1993) Bio/ Technology 11:84-89

Finer J J, McMullen MD (1990) Plant Cell Reports 8:586-589 Fromm ME, Morris F, Armstrong C, Williams R, Thomas J, Klein TM

(1990) Bio/Technology 8:833-839 Goldfarb B, Strauss SH, Howe GT, Zaerr JB (1991) Plant Cell Reports I0:

517-521 Gordon-Kamm W J, Spencer TM, Mangano M-L, Adams TR, Dairies RJ, Start WG, O'Brien JV, Chambers SA, Adams WR, Willets NG, Rice TB, Mackey C J, Krueger RW, Kausch AP, Lemanx PG (1990) The Plant Cell 2:603-618 lida A, Seki M, Kamada M, Yamada Y, Morikawa H (1990) Theor. Appl.

Genet. 80:813-816 Imaseki H (1986) In: Takahashi N (ed) Chemistry of plant hormones,

CRC Press, Boca Raton, 249-264 Klein TM, WolfED, Wu R, Sanford JC (1987) Nature 327, 70-73 Klein TM, Harper EC, Svab Z, Sanford JC, Fromm ME, Maliga P (1988a) Proc. Natl. Acad. Sci, 85, 8502-8505

Klein TM, Gradziel T Fromm ME, Sanford JC (1988b) Bio/Technology 6, 559-563

Kozugi S, Ohashi Y, Nakajima K, Arai Y (1990) Plant Science 70, 133-140 Kozak M (1989) J Cell Biology, Vol. 108, 229-241 Last DI, Brettel RIS, Chamberlain DA, Chaudhury AM, Larkin P J, Marsh

EL, Peacock WJ, Dennis ES (199l) Theor. Appl. Genet. 81:581-588 ManiatisT, Fritsch EF, Sambrock J (1982): Molecular Cloning, alaboratory

manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY McCabe DE, Swain WF, Martinell B J, Christou P (1988) Bio/Technology 6:923-926

McElroy D, Zhaug W, Cao J, Wu R (1990) The Plant Cell 2, 163-171 Morris JW, Castle LA, Morris RO (1989) Physiol. Mol. Plant Pathol. 34:

451-461 Perl A, Kless H, Blumenthal A, Galili G, Galun E (1992) Mol. Gen. Genet.

235:279-284 Robertson D, Weissinger AK, Ackley R, Glover S, Sederoff RR (1992)

Plant Molecular Biology 19:925-935 Sanford JC, Smith FD, Russell JA (1993) Methods in Enzymology 217, 483-509

SAS Institute (1985) SAS users guide: statistics. SAS Institute, Inc., Cary, NC

Sederoff R, Stomp A-M, Scott Chilton W, Moore LW (1986) Eio/ Technology 4:647-649

Smith DR (1994) U.S. Patent application 08-219879 (under review - publication in preparation)

Stomp A-M, Loopstra C, Chilton WS, Sederoff RR, Moore LW (1990) Plant Physiol. 92:1226-1232

Tautorus TE, Bekkaoui F, Pilon M, Datla RSS, Crosby WL, Fowke LC, Dunstan DI (1989) Theor. Appl. Genet. 78:531-536

Vain P, McMullen MD, Finer JJ (1993) Flant Cell Reports 12: 84-88. Weber H, Graessmann A (1989) FEBS Letters 253:163-166 Wilson SM, Thorpe TA, Moloney MM (1989) Plant Cell Reports 7: 704- 707