a refined method for the detection of baculovirus occlusion bodies in forest terrestrial and aquatic...

A refined method for the detection ofbaculovirus occlusion bodies in forestterrestrial and aquatic habitatsPeter M Ebling* and Stephen B HolmesCanadian Forest Service, Natural Resources Canada, 1219 Queen St E, Sault Ste Marie, Ontario, Canada P6A 2E5

Abstract: A sensitive and efficient method was developed for the detection of genetically modified and

wild-type baculovirus occlusion bodies (OB) in forest terrestrial and aquatic habitats. The protocol

facilitates the analysis of a large number of samples collected and frozen to maintain viral integrity.

Lyophilization was used to standardize the size of both field-collected soil samples and test substrates

inoculated with OBs for the determination of minimum detection threshold. To simulate natural

conditions, terrestrial test substrates were inoculated at a standardized moisture content determined

using a soil pressure plate apparatus. OBs, extracted from lyophilized test substrates by washing,

sieving and centrifugation, were subjected to alkaline lysis and viral DNA isolated using a purchased

DNA purification kit. PCR amplified DNAwas visualized using agarose gel electrophoresis. Minimum

detection thresholds in terrestrial substrates were 103, 102, 102 and 101 OBs from 0.5g of lyophilized L,

F-H and mineral soil horizons, and 1.0ml of leachate, respectively. Detection thresholds in aquatic

substrates were 100 and 103 OBs from 1.0ml of pond water and 1.0g of bottom sediment, respectively.

# 2002 Society of Chemical Industry

Keywords: baculovirus; Choristoneura fumiferana; detection; spruce budworm

1 INTRODUCTIONResearchers worldwide are devoting considerable

effort to the development of biological alternatives to

the use of broad-spectrum chemical pesticides for both

forestry and agricultural applications. Baculoviruses,

although relatively slow-acting, are attractive candi-

date biopesticides due to their host specificity. In

recent years, genetic engineering is increasingly being

used to accelerate the action of baculoviruses. With the

introduction into the environment of these genetically

modified organisms, it is essential to elucidate their

environmental fate and to assess the risks associated

with their release. This has promulgated the need to

develop rapid and reliable methods for the low-level

detection of baculoviruses from samples derived from

various environmental substrates.

While approaches employed by other researchers1,2

have been effective for the direct extraction of viral

DNA from test substrates, such methods are incapable

of distinguishing the source of the DNA as being

occlusion bodies (OB), free nucleocapsids or naked

DNA. Methods involving the extraction and detection

of OBs alone have not been very sensitive. Reported

here is a refined method with improved detection

capabilities for OBs of both genetically modified and

wild-type baculovirus.

The objective of this study was to develop a sensitive

method for detecting OBs of genetically modified

baculovirus from a field-release into outdoor forest

terrestrial and aquatic microcosms. Quantification of

the success of the protocol was determined by

obtaining minimum detection thresholds of OBs in

various environmental substrates inoculated with the

baculovirus. The protocol presented has facilitated a

multi-year study in which a large number of samples

were collected and stored frozen to maintain viral

integrity until analysis. Results of the field-release will

be reported separately.

Terrestrial samples of equivalent mass collected

from the field over a period of time may vary

substantially in their moisture content, and thus the

relative quantity of organic and mineral material they

contain. Therefore, the quantity of OBs extracted

from these samples may vary even when expressed on a

per weight basis. Many researchers have not taken this

into consideration. Some researchers standardize

sample size by oven-drying representative samples,

and then calculating either the desired mass of moist

field sample to be taken for analysis or the actual dry

mass that was analyzed when using a fixed mass of

moist field-collected material. This approach is not

practical when sampling over numerous time periods,

since moisture content and sample size need to be

calculated for each sample time. Its other disadvantage

(Received 18 January 2002; revised version received 13 May 2002; accepted 8 July 2002)

* Correspondence to: Peter M Ebling, Canadian Forest Service, Great Lakes Forestry Centre, 1219 Queen Street East, Sault Ste Marie,Ontario, Canada, P6A 2E5E-mail: [email protected]/grant sponsor: Enhanced Timber Product Network of the Canadian Forest Service

# 2002 Society of Chemical Industry. Pest Manag Sci 1526–498X/2002/$30.00 1216

Pest Management Science Pest Manag Sci 58:1216–1222 (online: 2002)DOI: 10.1002/ps.591

is that it requires the storage of large numbers of

samples prior to analysis. The protocol presented here

utilizes lyophilization to facilitate the standardization

of frozen field-collected samples containing OBs.

In the determination of minimum detection thresh-

old of OBs in terrestrial substrates, researchers either

fail to standardize the moisture content of the test

material prior to inoculation, or standardize using dry

test material, thereby possibly affecting adsorption of

OBs to soil particles. The protocol presented here

utilizes a pressure plate extractor method3 to standar-

dize the moisture content of test material to field

capacity, the moisture of the soil when water has been

drained by gravity, prior to inoculation.

2 EXPERIMENTAL METHODS2.1 Virus inoculumA genetically modified baculovirus, Choristoneurafumiferana MNPVegt�/lacZþ,4 was used as the inocu-

lum for all substrates tested. Cf MNPVegt�/lacZþ was

constructed from the Ireland strain of Cf MNPV by

inserting into the egt locus a 3.55-kb fragment

containing lacZ under the control of the p10 promoter

of Autographa californica NPV.

A diet surface contamination method was used to

conduct two successive in vivo amplifications of OBs

of the genetic recombinant to provide a sufficient

quantity for testing. For each amplification, 50ml ofaqueous suspension containing 1�106 OBs, enumer-

ated using a stained-film method,5 were applied to the

surface of formaldehyde-free artificial diet6 in half-

filled creamer cups (Portion Packaging, Etobicoke,

Ontario) and allowed to dry for 30min at room

temperature. Eight laboratory-reared fifth-instar

Choristoneura occidentalis Freeman larvae per container

were allowed to feed continuously (25°C, 60% RH,

and 16:8h light:dark photoperiod) on the contami-

nated diet until death. Cadavers were collected daily

for 10 days, pooled, and stored at �20°C.

OBs were separated from larval cadavers by filtra-

tion and centrifugation. Infected insects were macer-

ated in distilled water for 1min and the solution

filtered through three layers of cheesecloth. OBs were

pelleted out of suspension by centrifugation at 2053 gfor 30min at 4°C. The pellets were washed by

suspension in distilled water and centrifugation at

2053 g for 30min at 4°C. The pellets were then re-

suspended in distilled water, OBs enumerated using a

stained-film method5 and standardized to a concen-

tration of 2�109 OBml�1 before storage at 4°C.

Presence of the gene insert was confirmed by restric-

tion enzyme digestion and agarose gel electrophoresis,

although methods and results are not presented here.

2.2 Viral DNA for positive controlsViral DNA for use as positive controls in subsequent

PCR reactions was extracted from the OBs by alkaline

lysis and purification using the QIAGEN1 Blood

& Cell Culture DNA Maxi Kit (Cat No 13362;

QIAGEN Inc, Mississauga, Ontario). Virions were

released from the OB suspension (2�109 OBml�1;

40ml) by treating with alkaline solution (sodium

carbonate 0.5M, sodium thioglycolate 0.2M; pH

11.3; 15ml) and incubating at room temperature for

10min on a magnetic stirrer. Virions were collected

from the suspension by centrifugation in two tubes for

10min (1369 g; 4°C), followed by centrifugation of

the supernatants for 1h (30000 g; 4°C). Each pellet

containing the virions was re-suspended in TE

(Tris-HCl 10mM, EDTA 1mM; pH 8.0; 5ml). DNA

was released from each virion suspension by the

addition of G2 lysis buffer (5ml; supplied with the

QIAGEN DNA Maxi Kit) and protease K (20mg

ml�1; 250ml) followed by incubation in a 50°C water-

bath for 18h. DNA was isolated from each resulting

mixture using protocols supplied with the QIAGEN

DNA Maxi Kit, with the exception that the DNA was

never vortexed, as described below. Vortexing was

omitted to minimize shearing of the DNA. Briefly,

each sample was loaded onto an equilibrated Geno-

mic-tip and the DNA entered the resin by gravity flow.

Each Genomic-tip was washed twice with the buffer

supplied and the DNA was extracted from the resin

using pre-heated elution buffer (50°C), also supplied

with the kit. DNA was precipitated from the elution

buffer by adding 0.7 volumes of room-temperature

isopropanol, mixing by inversion 15 times, incubating

for 20min at room temperature and centrifuging for

20min (16000 g; 4°C). The resulting DNA pellets

were washed by the addition of cold 70% ethanol

followed by centrifugation for 10min (16000 g; 4°C).

The DNA pellets were each re-suspended in TE (pH

8.0; 100ml) and pooled. Two hours (þ18h of incu-

bation) were required to complete the lysis portion of

the procedure and an additional 5h for purification of

the DNA using the QIAGEN DNA Maxi Kit. Yield

and purity of DNA from seven extractions was deter-

mined by spectrophotometric measurement.7 Extrac-

tion efficiency of the DNA kit was calculated using an

estimate (PM Ebling andW Fick, unpublished) of 108

nucleocapsids OB�1 (0.1516fg DNA nucleo-

capsid�1).

2.3 Test substratesMinimum detection thresholds of OBs were deter-

mined in test substrates obtained from forest terrestrial

and aquatic habitats.

2.3.1 Terrestrial substratesTerrestrial substrates were obtained from the forest

floor beneath a 50-year-old mixed-wood stand (50%

Pinus strobus L, 40% Populus tremuloides Michx and

10% Pinus resinosa Ait) within the Great Lakes–St

Lawrence Forest Region, near Thessalon, Ontario,

Canada (46° 21.66747’N, 083° 35.46039’W). A soil

core sampler was used to collect the top 20cm of forest

floor, consisting of 4cm of humus overlaying medium

to coarse sand. The humus horizon was separated into

two substrates for testing: a 3-cm deep L layer (a layer

Pest Manag Sci 58:1216–1222 (online: 2002) 1217

Detection of OBs in forest terrestrial and aquatic habitats

of accumulated organic matter in which the original

structures of the plants were easily discernible), and a

1-cm deep F-H organic horizon (a layer of decom-

posed or partly decomposed organic matter). Transi-

tional layers did not exist within the mineral soil

horizon and, therefore, this substrate was tested

without further division. The fourth terrestrial sub-

strate tested was leachate, obtained by collecting

rainwater that had percolated through artificially

constructed soil microcosms, acquiring soluble com-

ponents. Microcosms consisted of two stackable

polypropylene buckets of unequal depth. The shal-

lower inner bucket contained intact soil profiles

(0.16m2) as described above. Holes were previously

drilled into the bottom of the inner buckets and the

bottoms lined with landscape fabric, enabling rain-

water, but not soil, to drain from them and collect in

the larger outer bucket. Microcosms were buried at

ground level in the mixed-wood stand described above

and leachate collected after rainfall had occurred. The

L, F-H, mineral soil and leachate test substrates had

pHs of 4.7, 4.5, 4.9, and 6.8, respectively.

2.3.2 Aquatic substratesOB detection thresholds were determined using

aquatic substrates obtained from a nearby natural

pond. Two substrates were examined, pond water and

bottom sediment. The pond water (pH 6.9) contained

few suspended organic particulates but was dis-

coloured by solubilized materials. The bottom sedi-

ment (pH 4.6) consisted primarily of silt and partly

decomposed or non-decomposed organic matter,

intermixed with a small quantity of medium to coarse

sand.

2.4 Inoculation of test substratesThe four terrestrial and two aquatic test substrates

were inoculated using 10-fold serial dilutions of OBs

obtained from the virus inoculum above for the

determination of minimum detection thresholds. All

substrates and experimental controls were replicated

three times to ensure repeatability of the protocol.

2.4.1 Inoculation of terrestrial substratesTo ensure equal representation of organic components

in all replicates of each test substrate, large particles

were cut into pieces less than 1cm in length and mixed

throughout the sample. Although grinding could have

been used to reduce sub-sampling error, it was not

done for two reasons: (1) to minimize changes in

surface area, which might affect adsorption of OBs,

and (2) to minimize the release of humic substances

that could interfere with subsequent analysis.

To simulate natural conditions, ie adsorption of

OBs to organic or mineral material under natural

moisture conditions, test substrates were inoculated at

field capacity. Field capacity is the moisture content of

soil when water has been drained by gravity. This

condition usually develops in a well-drained soil 2 or 3

days after rain. Field capacity was determined for five

replicates of each soil substrate (L, F-H and mineral

horizons) using a pressure plate (Soil Moisture Equip-

ment Co, Santa Barbara, CA) extractor method.3

Briefly, test substrates were equilibrated with water,

allowed to drain in a pressure plate apparatus (24h;

33kPa) and then oven-dried (24h; 105°C). Moisture

content was determined by comparing the wet and dry

weight of each test substrate.

For this protocol to be consistent with the metho-

dology used (ie lyophilization) to standardize the size

of field-collected samples (refer to the discussion

below), field capacity was also determined for an

additional five replicates of each test substrate using

lyophilization (24h; Unitrap 10–100 freeze-drier, The

VirTis Co, Inc, Gardiner, NY) instead of oven-drying.

To standardize sample sizes of test substrates and to

maintain consistency with the methodology to be used

to standardize sample sizes for field-collected samples,

L, F-H and mineral horizon substrates were lyophi-

lized for 24h prior to inoculation. Six 0.5-g lyophilized

samples from each test substrate were saturated to

field capacity (less 10ml) with distilled water and OBs

added (10ml volume) yielding doses of 105, 104, 103,

102, 101 and 100 OB per sample. One additional

sample of each substrate was inoculated with distilled

water in a similar manner and used as an experimental

control to ensure that the substrate did not contain

substances yielding positive signals in subsequent

analysis. Inoculated test substrates were stirred briefly

to ensure uniform distribution of OBs and allowed to

stand for 2h at room temperature, a length of time

consistent with that required to collect the first

samples in the field trial. Samples were subsequently

frozen for 3 days and lyophilized for 24h before

analysis. Three days was selected for the storage time

because freezing had been shown to increase adsorp-

tion of OBs to soil particles, although the effect had

diminished by the third day.8

Leachate test substrates were standardized by

volume (1.0ml). Six samples of leachate were inocu-

lated by adding 10ml of diluted virus inoculum to

990ml of test substrate yielding doses of 105, 104, 103,

102, 101 and 100 OB per sample. One additional

sample was inoculated with distilled water in a similar

manner and used as an experimental control. Inocu-

lated substrates were mixed briefly by inversion to

ensure uniform distribution of OBs, adsorption

allowed for 2h at room temperature, then frozen for

3 days before analysis.

2.4.2 Inoculation of aquatic substratesPond water test substrates were prepared and inocu-

lated in a manner identical to that for leachates.

Bottom sediments are saturated with water and

consequently vary little in moisture content. Therefore

sample size (1.0g) was standardized by wet weight

only. No additional standardization for moisture

content was required to simulate natural conditions

of the sediment prior to inoculation with OBs. Six

1.0-g samples of bottom sediment were inoculated by

1218 Pest Manag Sci 58:1216–1222 (online: 2002)

PM Ebling, SB Holmes

the addition of 10ml of diluted virus inoculum yielding

doses of 105, 104, 103, 102, 101 and 100 OB per

sample. An experimental control was inoculated with

distilled water. Inoculated substrates were mixed

briefly by stirring to ensure uniform distribution of

OBs, adsorption allowed for 2h at room temperature

and then frozen for 3 days before analysis. For

comparison of the minimum detection threshold of

OBs in sediments with other substrates tested, five

1.0-g samples of bottom substrate were lyophilized for

24h and freeze-dried weights determined.

2.5 Extraction of OBs from test substratesInoculated test substrates consisted of six 0.5-g

samples of lyophilized L, F-H, and mineral soil

horizons, 1.0-g samples of saturated aquatic sediment,

and 1.0-ml samples of pond water and leachate.

OBs were washed from L, F-H, mineral soil and

aquatic sediment test substrates using the vacuum

apparatus represented in Fig 1. The apparatus con-

sisted of a 76-mm diameter sieve (USA. Standard

Testing Sieve, No 40, 425mm openings) nested into a

filtercup (Whatman Cat No 1600 113, 30-mm pore

size) fitted with a vacuum support unit (Whatman Cat

No 1600 900). The filtration unit was fitted with a

rubber stopper (size 612) and inserted into a 50-ml

screw-top collection tube. Vacuum was applied to the

collection tube and an in-line vacuum flask installed in

the event of overfilling. A clamp on the collection tube

supported the vacuum apparatus. A sterile apparatus

was used for each sample analyzed.

Test substrates were applied to the sieve and stirred

for 10–15s prior to the addition of distilled water, to

allow small particles to pass. Containers holding the

test substrates were rinsed with distilled water and

added to the sieve. Test substrates were washed on the

sieve with approximately 10ml of distilled water from a

squeeze bottle while gently stirring for 20–30s using a

Teflon spatula. The intention was to wash OBs from

the surface of the test material, not to break them into

smaller pieces thereby increasing surface area and

further releasing humic substances. Vacuum was then

applied for 3–5s. Washing was repeated twice more,

resulting in approximately 30ml of rinsate in the

collection tube. The sieve was removed and the

substrate within the filter cup rinsed with an additional

5–7ml of distilled water while stirring 20–30s to re-

suspend the sediment. Vacuum was then applied for

3–5s. Washing of the filter cup was repeated twice

more, resulting in a final rinsate volume of approxi-

mately 50ml. Vacuum was switched off and the seal

broken (ie vacuum flask disconnected) between each

wash to allow sieved sediment to remain suspended in

the filter cup during stirring, thereby increasing

desorption of OBs. Washing of test substrates through

the filtration apparatus required about 45min for one

replicate of seven samples. The rinsate was centrifuged

for 30min (2053 g; 4°C) and the supernatant poured

off and discarded. The pellet containing OBs was re-

suspended in 900ml distilled water and transferred to a

1.5-ml microtube.

Pond water and leachate test substrates were not

passed through the filtration apparatus. OBs were

extracted by centrifuging 1.0ml inoculated samples for

30min (2053 g; 4°C) and re-suspending the resulting

pellet in 900ml distilled water in 1.5-ml microtubes.

2.6 Lysis of OBsVirions were released from the 900ml of OB suspen-

sion by adding 100ml of alkaline solution (sodium

carbonate 0.5M, sodium glycolate 0.2M; pH 11.3) and

incubating at room temperature for 10min. Twice

during incubation the suspension was mixed by

inverting the tube four times. Virions were collected

from the suspension by centrifugation for 5min (325 g;4°C), followed by centrifugation of the supernatant for

30min (16000 g; 4°C) and re-suspension of the

resulting pellet in TE (pH 8.0; 500ml). DNA was

released from the virion suspension by the addition of

G2 lysis buffer (1ml; supplied with the QIAGEN

DNA Maxi Kit) and proteaseK (20mg ml�1; 50ml)followed by incubation in a water-bath at 50°C for

18h. The lysis procedure required about 1h (þ18h

incubation) for one replicate of seven samples.

2.7 Isolation of viral DNAViral DNA was isolated from the virion lysate using

protocols and materials supplied with the QIAprep

Spin Miniprep Kit (Cat No 27104; QIAGEN Inc,

Mississauga, Ontario). Briefly, the lysate was neutra-

lized using the buffer supplied and loaded onto a spin

column where the DNA adsorbed onto a silica-gel

membrane during centrifugation (3000 g; 1min; room

temperature). Because the volume of neutralized

lysate was larger than the capacity of the spin column,

the suspension was divided into two equal parts and

centrifuged successively through the same column.

The column was washed twice with the supplied

buffers by centrifugation for 1min (3000 g) at room

temperature. DNA was extracted from the silica-gel

membrane by the addition of a pre-heated elution

buffer (70°C) and by centrifugation for 1min

(16000 g) at room temperature. Isolation of viral

DNA from one replicate of seven samples required

approximately 45min.Figure 1. Vacuum apparatus used for the extraction of OBs from terrestrialand aquatic substrates.



2.8 PCR amplificationViral DNA isolated from OBs extracted from the test

substrates was amplified using PCR prior to visualiza-

tion by agarose gel electrophoresis. PCR amplification

was conducted using a 50-ml PCR reaction mixture

containing sterile ultra-pure water (39.6ml), 10�reaction buffer (Tris-HCl 100mM, magnesium chlor-

ide 15mM, potassium chloride 500mM; pH 8.3; 5ml),nucleotide mix (10mM; 1ml), formamide2 (20g litre�1;

1ml), upstream primer (0.1mg; 1ml), downstream

primer (0.1mg; 1ml), Taq DNA polymerase (2 units;

0.4ml) and template DNA (undiluted; 1ml). Forty

amplification cycles were carried out in a Genius

Thermal Cycler (Techne, Inc, Princeton, NJ). The

first cycle consisted of denaturation of template DNA

at 94°C for 4min, followed by annealing of primers at

65°C for 3min and extension of DNA at 72°C for

3min. Thirty-nine additional cycles were performed

consisting of denaturation at 94°C for 30s, annealing

at 65°C for 1min and extension at 72°C for 1min. A

final extension step at 72°C for 10min was performed

to ensure complete extension. All PCR reactions

included a positive control (12.2ng ml�1 CfMNPVegt�/lacZþ DNA) to ensure amplification of

the desired DNA fragment and a negative control

(PCR reaction mixture excluding viral DNA) to

ensure that PCR reagents were not contaminated.

PCR primers (5’-CTAGAGGATCCGCTAGCA

CG-3’ and 5’-TCA GGC TGC GCA ACT GTT

GG-3’) were designed2 to amplify a 530-bp fragment

from the Autographa californica P10 polyhedrin pro-

moter and the b-galactosidase (LacZ) gene. This

fragment is specific to Cf MNPVegt�/lacZþ and is

not found in nature.2 Primers were synthesized by the

Central Facility of the Institute for Molecular Biology

and Biotechnology, McMaster University, Ontario,

Canada.

PCR-amplified DNA was detected using horizontal

1.2% agarose gel electrophoresis run in TAE buffer

(Tris-acetate 0.04M, EDTA 0.001M) at 5.4Vcm�1 for

30min. BlueJuice2 (Life Technologies Inc) gel-

loading buffer (2.8ml) was added to 25ml of amplified

DNA before loading onto 9-mm thick gels (7.5�

7.5cm). Because positive controls yielded exception-

ally strong signals, only 5ml of DNA (plus 1.1mlloading buffer) was loaded. Gels were stained (45min)

in TAE buffer containing 1mg ml�1 ethidium bro-

mide, de-stained (2�20min) in distilled water and

photographed with Polaroid Type 55 Positive/Nega-

tive film and a UV transilluminator.

3 RESULTSSpectrophotometric measurement7 of viral DNA

derived using the QIAGEN Blood & Cell Culture

DNA Maxi Kit determined that 40ml of OBs (2�109

OBml�1) yielded a mean (SD) of 526(�60)mg DNA

(OD260/OD280=1.63(�0.08)) per extraction. Extrac-

tion efficiency was estimated to be 40.1(�4.6)%.

No significant difference (t test; P =0.05) in

moisture content was observed when comparing

oven-dried and lyophilized samples for the determina-

tion of field capacity of test substrates, thus lyophiliza-

tion was selected to standardize sample size. Mean

moisture content for L, F-H and mineral soil sub-

strates at field capacity was 175.0(�6.2), 91.2(�6.2)

and 20.6(�0.9)%, respectively. Therefore, 0.5g of

lyophilized L, F-H and mineral soil samples weighed

1.375, 0.956 and 0.603g, respectively, when inocu-

lated at field capacity. Moisture content of aquatic

sediment was calculated as 413.3(�21.2)% using five

replicates of lyophilized samples. However, the detec-

tion threshold for bottom sediment was determined

using saturated test substrate.

Minimum detection thresholds of OBs extracted

from three replicates of terrestrial and aquatic sub-

strates are shown in Table 1. Detection thresholds in

terrestrial substrates were 102–103, 101–102, 102–103

and 101 from L, F-H, mineral soil and leachate,

respectively. Detection thresholds in aquatic sub-

strates were 100 and 102–103 from pond water and

bottom sediment, respectively. PCR amplification

products of DNA isolated from OBs that were

extracted from each substrate are shown in Fig 2.

The decrease in intensity of PCR products across the

gels is due to the smaller quantities of OBs used for

Table 1. Minimum detection threshold ofgenetically modified Choristoneura fumiferananucleopolyhedrovirus OBs in inoculated forestterrestrial and aquatic substrates

Sample size

Detection threshold a (OB)

Rep 1 Rep 2 Rep 3

Terrestrial substrates L horizon 0.5gb 102 103 103

F-H horizon 0.5gb 101 102 102

Mineral soil 0.5gb 102 102 103

Leachate 1.0mlc 101 101 101

Aquatic substrates Pond water 1.0mlc 100 100 100

Bottom sediment 1.0gd 102 103 103

a Minimum detection threshold was determined by inoculating test substrates using 10-fold serial

dilutions of occlusion bodies.b 0.5g lyophilized test substrates were saturated to field capacity prior to inoculation, adsorption

allowed for 2h and then frozen for 3 days before analysis.c Inoculated samples were allowed to adsorb for 2h and frozen for 3 days before analysis.d Samples were inoculated when saturated, adsorption allowed for 2h and then frozen for 3 days

before analysis.



inoculation of the test substrates. Detection thresholds

reported indicate that at least two of the three

replicates were identical.

4 DISCUSSIONAcidity of test substrates affects adsorption of OBs to

soil particles,9 thus minimum detection thresholds in

any environmental substrate in question must be

determined prior to using this improved method to

elucidate environmental fate or risk assessment.

Nearly 90% of OBs become adsorbed to soil at pH

4.9–5.0, whereas only 13.2% adsorb to neutral soil

substrates.9 Soil substrates used to establish the

protocol presented here were acidic (pH 4.5–4.9),

thus detection threshold in neutral or alkaline soils

could be further improved. Various buffers and

methods (0.1% sodium pyrophosphate, pH 7.0;

0.1M Tris, pH 7.0; 0.1% SDS; 5-min sonication)

used by other researchers2,8–10 to desorb OBs from test

substrates were tested here, but not reported, since

water was found to be equally effective when applied in

the manner presented. It is evident by comparison of

the detection threshold in pond water (100 OBs) to

those in the soil substrates (101–103 OBs) that either:

(1) desorption from organic or mineral material was

not fully attained, or (2) OBs in soil substrates were

degraded beyond detection. Since incubation and

storage times during the inoculation procedure were

quite short, it is more probable that OBs adsorbed to

soil particles were lost in the filtration process.

Although minimum detection thresholds in soil sub-

strates are higher than in water, they appear to be a

significant improvement over previously reported

methods.2,10

Direct comparison of results cannot be made with

other methods developed to determine minimum

detection thresholds for various soil organisms since

researchers have not adequately standardized sample

size. Although sample sizes of 0.1–5.0g1,2,11,12 were

reported, soil moisture content was not determined or

was calculated using oven-drying procedures. Results

for the lyophilized soil substrates tested here indicate

that 0.5-g samples weigh 1.375, 0.956 and 0.603g for

L, F-H and mineral soil, respectively, when saturated

to field capacity. Thus, sample size for these substrates

varies by as much as 2.75 times when collected at field

capacity and even more when collected shortly after

rain. Bottom sediment varies 5.1 times between

lyophilized and ambient weight. Although oven-drying

has been used by others2 to standardize sample size, it

was not appropriate to meet the objectives of the

protocol presented here for two reasons. First, the

heating of field-collected samples would compromise

any OBs they might contain. Detection thresholds for

test substrates inoculated with OBs prior to oven-

drying were determined but are not reported here,

since the thresholds were drastically diminished.

Second, test substrates oven-dried prior to the addi-

tion of OBs would be inoculated under conditions not

representative of natural moisture content, thereby

possibly affecting adsorption to organic or mineral

material. The use of oven-dried test substrates

saturated to field capacity prior to inoculation would

be appropriate but would not be consistent with

Figure 2. PCR amplification products of Cf MNPVegt�/lacZþ DNA isolatedfrom OBs extracted from forest terrestrial and aquatic substrates:(A) L horizon; (B) F-H horizon; (C) mineral soil; (D) leachate; (E) pondwater; (F) bottom sediment. Lanes 1 through 7 were extracted from testsubstrates inoculated with 105, 104, 103, 102, 101, 100 and 0 OB. Anamplified PCR reaction mixture (excluding DNA) is included in lane 8 as anegative control. A positive control (530-bp template DNA) is included inlane 9 and a 1-kb ladder in lane 10. Minimum detection thresholds interrestrial substrates were 103, 102, 102 and 101 OB from 0.5g lyophilized L(A3), F-H (B4) and mineral soil (C4) horizons, and 1.0ml leachate (D5),respectively. Detection thresholds in aquatic substrates were 100 and 103

OB from 1.0ml pond water (E6) and 1.0g bottom sediment (F3),respectively.



procedures required for the analysis of field-collected

samples. Lyophilization is the only procedure that can

be used for both the analysis of field-collected samples

and the determination of minimum detection thresh-

old in test substrates.

Extraction efficiency of OBs from inoculated soil

samples is reduced by storage at �20°C due to

increased adsorption.8 After an initial phase of rapid

decline, the rate of OB recovery remains relatively

constant. In addition, there is a highly significant

difference in extraction efficiency between unfrozen

samples and those stored frozen for 3 days or more

(P<0.0001).8 The protocol presented here facilitates

the investigation of a large number of frozen samples.

However, detection thresholds could be further

improved by the analysis of samples immediately after

collection.

Most recently established methods1,2 used to detect

baculovirus from soil substrates involve direct DNA

extractions and extrapolation to determine theoretical

concentrations of OBs detected. This is appropriate

for the determination of detection threshold when

inoculating sterile test material, but not for the analysis

of field-collected samples that may contain free

nucleocapsids or naked DNA along with OBs. The

procedure presented utilizes centrifugation to ensure

that OBs are separated from extraneous viral material

and many dissolved humic substances. Humic sub-

stances, shown to be PCR inhibitors,13,14 appear to

have been minimized in the extraction and purification

process. Dilution of the viral DNA product prior to

PCR amplification, as required in other

methods,1,2,11,12 did not improve detection thresholds

(data not shown). Although substrates having the

greatest organic content (L horizon and bottom

substrate) yield a higher detection threshold than

pond water or leachate, thresholds appear to be

increased by adsorption of OBs to organic material

rather than by inhibitory effects in PCR reactions.

Preliminary tests conducted using tenfold increases

in sample size with the procedures presented suggested

that minimum detection thresholds may be reduced by

similar amounts. Depending on the type of soil

analyzed, modifications to the procedure may be

required to prevent plugging of the filter cup during

extraction of OBs from larger samples of test substrate.

Because of the limited sample size available from the

field study for which this protocol was developed,

further testing of larger samples was not pursued.

5 CONCLUSIONThe protocol presented should be relatively simple to

adapt to detect a range of viruses from various

environmental substrates. Although this method has

been developed using a genetically modified baculo-

virus, wild-type viruses can also be detected if the

appropriate PCR primers, specific to the target, are

selected. Novel methods of standardizing sample size

and moisture content of test substrates for the deter-

mination of detection threshold of OBs have been

presented and should form the basis upon which new

methods are compared.

Thus, this protocol is a reproducible, efficient and

sensitive method for the detection of Cf MNPV in

various forest terrestrial and aquatic habitats and may

facilitate ecological and risk assessment studies.

ACKNOWLEDGEMENTSThanks are extended to the Environmental Assess-

ment unit of the Canadian Forest Service (CFS) for

supplying test substrates and for the first amplification

of OBs, as well as to Dr Basil Arif for provision of the

recombinant virus inoculum. Funding was provided

by the Enhanced Timber Production Network of the

CFS.

REFERENCES1 De Moraes RR, Maruniak JE and Funderburk JE, Methods for

detection of Anticarsia gemmatalis nucleopolyhedrovirus DNA

in soil. Appl Environ Microbiol 65:2307–2311 (1999).

2 England LS, Trevors JT and Holmes SB, Extraction and detec-

tion of baculoviral DNA from lake water, detritus and forest

litter. J Appl Microbiol 90:630–636 (2001).

3 Kalra YP and Maynard DG, Method manual for forest soil and

plant analysis, Forestry Canada Information Report NOR-X-319

(1991).

4 Barrett J, Xie W, Krell PJ, Sohi SS and Arif BM, Characteriza-

tion and identification of ecdysteroid UDP-glycosltransferase

gene from Choristoneura fumiferana nuclear polyhedrosis virus,

Proc Sixth Internat Colloq Invertebr Pathol and Microbial Contr

Montpellier, France, pp 425–426 (1994).

5 Wigley PJ, Counting micro-organisms, in Microbial control of

insect pests, ed by Kalmakoff J and Longworth JF, NZ Dep Sci

Ind Res Bull 228, Wellington, NZ, pp 29–35, (1980).

6 McMorran A, A synthetic diet for the spruce budworm, Choristo-

neura fumiferana (Clem) (Lepidoptera: Tortricidae). Canad

Entomol 97:58–62 (1965).

7 Sambrook J, Fritsch EF and Maniatis T, Appendix E:

Commonly used techniques in molecular cloning, in Molecular

cloning: a laboratory manual, 2nd edn, Cold Spring Harbor

Laboratory Press, New York, NY (1989).

8 Evans HF, Bishop JM and Page EA, Methods for the quanti-

tative assessment of nuclear-polyhedrosis virus in soil. J

Invertebr Pathol 35:1–8 (1980).

9 Hukuhara T and Wada H, Adsorption of polyhedra of a cyto-

plasmic-polyhedrosis virus by soil particles. J Invertebr Pathol

20:309–316 (1972).

10 Hukuhara T, Purification of polyhedra of a cytoplasmic-poly-

hedrosis virus from soil. J Invertebr Pathol 25:337–342 (1975).

11 Picard C, Ponsonnet C, Paget E, Nesme X and Simonet P,

Detection and enumeration of bacteria in soil by direct DNA

extraction and polymerase chain reaction. Appl Environ

Microbiol 58:2717–2722 (1992).

12 Volossiouk T, Robb EJ and Nazar RN, Direct DNA extraction

for PCR-mediated assays of soil organisms. Appl Environ

Microbiol 61:3972–3976 (1995).

13 Tsai Y-L and Olson BH, Rapid method for separation of

bacterial DNA from humic substances in sediments for

polymerase chain reaction. Appl Environ Microbiol 58:2292–

2295 (1992).

14 Tebbe CC and Vahjen W, Interference of humic acids and DNA

extracted directly from soil in detection and tranformation of

recombinant DNA from bacteria and yeast. Appl Environ

Microbiol 59:2657–2665 (1993).



a refined method for the detection of baculovirus occlusion bodies in forest terrestrial and aquatic...

Documents