Methods for Detection of Anticarsia gemmatalis Nucleopolyhedrovirus DNA in Soil

Abstract

Two methods, phenol-ether and magnetic capture-hybridization (MCH), were developed and compared with regard to their sensitivities and abilities to extract the DNA of the insect baculovirus Anticarsia gemmatalis nucleopolyhedrovirus (AgMNPV) from soil and to produce DNA amplifiable by PCR. Laboratory experiments were performed with 0.25 g of autoclaved soil inoculated with different viral concentrations to optimize both methods of baculovirus DNA extraction and to determine their sensitivities. Both procedures produced amplifiable DNA; however, the MCH method was 100-fold more sensitive than the phenol-ether procedure. The removal of PCR inhibitors from the soil appeared to be complete when MCH was used as the viral DNA isolation method, because undiluted aliquots of the DNA preparations could be amplified by PCR. The phenol-ether procedure probably did not completely remove PCR inhibitors from the soil, since PCR products were observed only when the AgMNPV DNA preparations were diluted 10- or 100-fold. AgMNPV DNA was detected in field-collected soil samples from 15 to 180 days after virus application when the MCH procedure to isolate DNA was coupled with PCR amplification of the polyhedrin region.

Baculoviruses are large, double-stranded, circular DNA viruses, which are infective to invertebrates. These viruses are mostly found in insects and have been reported to infect over 600 insect species (21). Baculoviruses have been employed to control important agricultural (10, 26) and forest (2, 20, 30) insect pests. However, this use of baculoviruses has been limited by the long length of time required for these viruses to kill their insect hosts. Genetic engineering of baculoviruses has enhanced their speed of killing, and it could potentially expand the use of these viruses as agents of insect control.

Baculoviruses are classified in the family Baculoviridae, presenting two genera, Nucleopolyhedrovirus and Granulovirus (27). Virions of nucleopolyhedroviruses (NPVs) and granuloviruses (GVs) are occluded in large protein crystals, which are mostly formed by a single protein, called polyhedrin in NPVs and granulin in GVs. The occlusion bodies protect the virus particles, conferring resistance to solubilization, except under strong alkaline conditions (31). These properties enable baculoviruses to remain viable for many years outside the insect host, with the soil being the major long-term reservoir for these viruses in the environment (5, 6). Many studies have reported the persistence of baculoviruses in soil for periods of up to 5 years (4, 16, 23). Therefore, the detection of baculoviruses in the environment, especially in the soil, is extremely important to the study of the ecology and environmental fate of both wild-type and genetically engineered baculoviruses.

The detection of baculovirus polyhedra varies in sensitivity according to soil type (7, 9) and pH (11). Polyhedra are adsorbed by soil particles mainly by Coulomb forces; that is, negatively charged polyhedra are retained on the positively charged sites of the soil particles (11). Different reagents have been used to extract baculovirus polyhedra from soil, with efficiencies of polyhedra recovery ranging from 7% for cytoplasmic polyhedrosis virus (12) to 24% for Autographa californica NPV (AcMNPV) (39). There are no reports of the direct detection of baculovirus DNA from soil. Nonetheless, DNA from bacteria (1, 13, 14, 29, 37) and viruses such as enteroviruses (34) has been directly extracted from soil and sediments before amplification by PCR. The main challenge regarding the extraction of DNA from soil is the presence of humic acids and phenolic compounds that are coextracted with the DNA and are inhibitory to enzymes used in DNA manipulation, such as restriction endonucleases and DNA polymerases (35, 36).

The objectives of this study were to develop a method for extracting baculovirus DNA from soil for further PCR amplification and to use the developed method to detect baculovirus DNA from field samples collected over a 1-year period after virus application. The Anticarsia gemmatalis NPV (AgMNPV) was chosen as a model because of its widespread use in Brazil and potential for use in southeastern United States to control the velvet bean caterpillar in soybeans. The abilities to detect baculovirus polyhedra and DNA in the soil will be useful tools in studies seeking to better understand baculovirus epizootics in general, to elucidate their environmental fate, and to assess risks associated with the release of genetically improved baculoviruses.

MATERIALS AND METHODS

Polyhedral extraction and viral DNA purification.

A series of experiments were performed to optimize the extraction of baculovirus DNA from soil samples for subsequent PCR amplification. The treatments consisted of 0.25 g of an autoclaved Dothan loamy fine sand soil collected from soybean fields being placed in 1.5-ml microcentrifuge tubes and inoculated with 50 μl of different dilutions of AgMNPV polyhedra. The soil samples were autoclaved to ensure that the only baculovirus present in the samples was AgMNPV at the viral concentration inoculated per treatment, in order to compare the sensitivities of two DNA extraction methods. The viral concentrations tested were 10⁷, 10⁵, 10⁴, 10³, 10², 10¹, and 1 polyhedron per 0.25 g of soil. The last concentration of virus was tested only in the magnetic capture-hybridization (MCH) experiments. Two different methods were tested, and their efficiencies in extracting baculovirus DNA were compared.

Phenol-ether extraction.

The phenol-ether extraction method consisted of the extraction of humic acids by incubating soil samples in 1.0 ml of 1 M Na₄P₂O₇ at pH 7.0 for 4 h in a rotary shaker at room temperature. The solution was centrifuged at 12,000 × g for 10 min in a microcentrifuge, and the pellet was resuspended in 1.0 ml of TE buffer (10 mM Tris, 1 mM EDTA [pH 8.0]). The solution was centrifuged again at 12,000 × g for 10 min. The pellet was subsequently resuspended and incubated for 2 h in a solution containing 500 μl of TE buffer, 200 μl of a 3× dilute alkaline solution (0.3 M Na₂CO₃, 0.03 M EDTA, 0.51 M NaCl) plus 500 μl of 0.2 M NaOH to disrupt AgMNPV polyhedra. Sodium hydroxide is also known to extract humic acids from soil (33). The alkali-released virus was centrifuged at 1,800 × g for 5 min, to remove unsolubilized polyhedra and soil particles, and the supernatant was centrifuged again at 12,000 × g for 20 min at 4°C. The pellet was resuspended in 500 μl of TE buffer and incubated with 20 μl of proteinase K (5 mg/ml) overnight at 37°C or for 2 h at 65°C in order to degrade the viral envelopes and capsids. The virus DNA was then purified with successive phenol-ether extractions (three times each) and was subsequently used for PCR amplification of the polyhedrin gene (polh).

MCH.

The MCH method was modified from that described by Jacobsen (14, 15). The method uses streptavidin-coated magnetic beads (M-280; Dynal, Inc., Lake Success, N.Y.) conjugated to a biotinylated probe specific for the AgMNPV polh region to capture AgMNPV DNA by hybridization from soil samples. The biotinylated probe contained a biotin molecule on a five-carbon atom spacer arm incorporated at the 5′ end of the oligonucleotide. The oligonucleotide was 103 bp long (positions +240 to +343 of polh) (40), and it was purified by high-pressure liquid chromatography (DuPont, Inc., Wilmington, Del.). Before conjugation, 200 μl of a 10-mg/ml suspension of streptavidin magnetic beads was washed three times with a solution containing 400 μl of phosphate-buffered saline (PBS) (2.65 mM KCl, 1.46 mM KH₂HPO₄, 136.9 mM NaCl, 8.0 mM Na₂HPO₄) and 0.1% sodium dodecyl sulfate at pH 7.1 to remove sodium azide (part of the storage buffer). This amount of beads is enough for 10 hybridization reactions. The magnetic beads were resuspended in a solution containing 200 μl of TE buffer and 1 M NaCl and conjugated to the biotinylated probe (approximately 100 ng) by 1 h of rotatory incubation at room temperature in a hybridization oven (OV3; Biometra, Inc., Tampa, Fla.). Following conjugation, the beads were washed three times with 400 μl of TE and 1 M NaCl and incubated for 15 min in the hybridization oven at room temperature in a solution containing 400 μl of 0.125 M NaOH and 0.1 M NaCl. After this incubation, the conjugated beads were washed three times with the solution containing 400 μl of TE and 1 M NaCl to remove any NaOH and residual cDNA. The beads were then resuspended in 200 μl of sterile distilled (SD) H₂O and used for hybridization.

Soil samples containing different concentrations of AgMNPV polyhedra were incubated in a shaker for 1 h at room temperature with a combination of TE buffer, 3× dilute alkaline solution, and 0.2 M NaOH to disrupt the polyhedra. These samples were centrifuged at 3,000 × g for 5 min. The supernatant was centrifuged again at 10,330 × g for 20 min at 4°C. The pellet was resuspended in 200 μl of PBS buffer and boiled for 10 min to disrupt the viral envelopes and capsids. After boiling, the samples were centrifuged at 10,330 × g for 10 min at 4°C. The supernatants, containing the viral DNA, were saved and used for hybridization to the biotinylated probe. Each hybridization reaction mixture consisted of 50 μl of DNA sample, 330 μl of hybridization solution (5× SSC [1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate] and 0.02% sodium dodecyl sulfate), and 20 μl of conjugated magnetic beads. Hybridization took place in a hybridization oven at 62°C for 1.5 to 2 h, according to the calculated number of hours required to reach 2 × C_ot_1/2, where C_o is the total concentration of DNA and t_1/2 is the time of half-renaturation for any probe (32). After hybridization, the beads were washed with 400 μl of SD H₂O and resuspended in a final volume of 50 μl of SD H₂O. For PCR amplification, 25 μl of resuspended beads was added to 25 μl of PCR master mix.

PCR conditions.

PCR amplification of the AgMNPV polh was performed as described elsewhere (24). The temperature program was run for 40 cycles. The DNA template was diluted 10- or 100-fold after the target DNA was obtained by the phenol-ether procedure. When the template DNA was obtained through the MCH procedure, PCR was performed in a total volume of 50 μl, containing 25 μl of resuspended beads and 25 μl of PCR master mix. The PCRs were carried out at a 200 μM final concentration of each deoxynucleoside triphosphate, 10 pmol of each primer, 2.5 mM MgCl₂, and 0.5 U of Primezyme and corresponding buffer (Biometra, Inc.). Each reaction tube was covered with 50 μl of mineral oil to prevent reagent evaporation. The PCR primers utilized in this study were 100% homologous to AgMNPV. They were derived from conserved sequences within the coding region of polh (40) and their DNA sequences are 5′ TA(CT)GTGTA(CT)GA(CT)AACAA(GA)T 3′ (forward) and 5′ TTGTA(GA)AAGTT(CT)TCCCA(AG)AT 3′ (reverse) (24). The amplification products were analyzed by 0.7% Seakem LE agarose (FMC, Inc.) gel electrophoresis in Tris-acetate-EDTA buffer stained with ethidium bromide.

AgMNPV detection from field soil samples.

Viral inocula of the AgMNPV-2D genomic variant were obtained by injecting 5 μl of extracellular virus (50% tissue culture infective dose = 10⁶ viruses/ml) into the hemocoel of fourth-instar A. gemmatalis. Moribund larvae were frozen, and their polyhedra were purified by standard procedures (22). The amount of polyhedra was estimated by counting in a hemacytometer chamber. AgMNPV-2D was applied to 300-m² soybean plots at the North Florida Research and Education Center, University of Florida, Quincy. The virus was applied on 30 August 1995 with a dose of 1.0 × 10¹¹ polyhedra/ha, as recommended for AgMNPV (26). The experiment consisted of two replicates of AgMNPV-2D-treated soybean plots and two replicates of control plots (i.e., plots in which no virus was applied). The soybean plots were separated by a buffer zone of 30 m, and the four outer soybean rows of each plot served as borders. The soil was classified as a Dothan loamy fine sand and contained 70 to 85% sand, 10 to 20% clay, and 0 to 30% silt. Soil samples from the top 15 cm of five sampling sites in each plot were collected with a core sampler. Soil samples weighed 69 g on average. Control and AgMNPV-2D treatments were sampled at 1, 15, 45, 75, 180, and 330 days after virus application. The soil samples were collected in individual plastic bags, put on ice, and taken to the laboratory to be frozen at −20°C. In the laboratory, each soil sample was homogenized manually, and a 0.5-g subsample was used for baculovirus DNA extraction by the MCH procedure. A total of 10 0.5-g soil subsamples were analyzed by PCR for each treatment and each sampling date. Aliquots of the extracted DNA were used in PCR amplification experiments for the detection of AgMNPV, targeting the coding region of the polh (24).

RESULTS AND DISCUSSION

Optimization of DNA extraction methods.

The phenol-ether procedure extracted humic acids from soil by sequential incubation in sodium pyrophosphate and sodium hydroxide and then purified the viral DNA by successive phenol-ether extractions. Baculovirus DNA extractions by this method removed large amounts of humic acids from the soil samples, but there was still some inhibition during PCR amplification. Therefore, DNA samples had to be diluted 10- or 100-fold to be used in PCR experiments. A detection limit experiment repeated on three separate occasions showed that the limits of detection for this procedure ranged from 1.1 × 10⁴ (in two experiments) to 1.1 × 10⁵ (in one experiment) AgMNPV genome copies per g of soil, which corresponds to 4 × 10² to 4 × 10³ polyhedra per g of soil (data not shown).

The MCH procedure used to isolate AgMNPV DNA from soil was efficient in removing humic acids and other inhibitors. Therefore, DNA aliquots were added directly in PCRs without dilution. Four replications of the same experiment determined that the detection limit for this procedure was 1.1 × 10² to 1.1 × 10³ genome copies per g of soil, which is equivalent to 4 to 40 polyhedra per g of soil. The lowest limit of detection was obtained in three of the four experiments performed, and the results for one of these experiments are shown in Fig. 1 (lane 9). The size of the polyhedrin PCR product, which was 575 bp, and the decrease in intensity of the PCR products, resulting from smaller amounts of AgMNPV polyhedra being inoculated in the soil samples, are also shown in Fig. 1.

FIG. 1.

Detection limit of the MCH procedure to isolate AgMNPV DNA from soil. Autoclaved soil (0.25 g) was inoculated with AgMNPV polyhedra, and the viral DNA was extracted by the MCH procedure and PCR amplified for the polh region. Lane 1, molecular marker (1-kb ladder); lanes 2 to 9, 575-bp polh PCR products corresponding to 10⁷, 10⁷ (no soil), 10⁵, 10⁴, 10³, 10², 10¹, and 1 polyhedron per 0.25 g of soil, respectively; lane 10, negative control, corresponding to soil with no virus added; lane 11, positive control, corresponding to AgMNPV DNA purified from virions; lane 12, reagents control, corresponding to all PCR reagents with no DNA added. Molecular lengths in kilobases are indicated on the left.

In this study, two different methods were developed and optimized to extract baculovirus DNA directly from soil, and both were suitable for PCR amplification. However, a comparison of the two DNA extraction methods showed that the MCH procedure, detecting as few as 4 polyhedra/g of soil, was 100-fold more sensitive than the phenol-ether extraction method, which required 400 to 4,000 polyhedra/g of soil for detection. In addition, the MCH procedure completely removed the apparent inhibitory effect of humic acids, since direct aliquots of the DNA preparations could be amplified without dilution. In contrast, the phenol-ether method apparently did not completely remove humic acids, because 10- to 100-fold dilutions of DNA were necessary to obtain PCR amplification. The processing time did not differ substantially between the two methods. It took approximately 16 h to process 30 soil samples with the MCH procedure, while it took 19 h to process the same number of samples with the phenol-ether method. The differences in sensitivity and removal of humic acids between the two methods evaluated in this study could be attributed in part to the different conditions to which the samples were subjected. For example, in some experiments, the phenol-ether samples were incubated with proteinase K at 65°C, while the samples subjected to the MCH procedure were boiled in PBS buffer for 10 min. However, the main difference between the MCH procedure and a standard DNA extraction method, such as the phenol-ether procedure, is that the MCH procedure uses streptavidin-coated magnetic beads conjugated to a biotinylated DNA probe, which is specific to the target DNA. During the hybridization step of the MCH procedure, the conjugated beads “fish out” the target DNA, removing it from nontarget DNA and interfering compounds, resulting in sensitive and specific extraction of DNA.

Both methods were compared only in laboratory experiments which used autoclaved soil samples collected from the field plots before AgMNPV application. MCH was the only method used to detect AgMNPV from the field experiments, because its sensitivity is greater than that of the phenol-ether method in laboratory studies.

The MCH procedure coupled with PCR amplification of the polh region was more sensitive than enzyme-linked immunosorbent assay and radioimmunoassay techniques. Monoclonal antibodies detected 3 × 10⁵ AcMNPV polyhedra (28), while direct and indirect radioimmunoassay detected 5 × 10³ and 1.2 × 10³ polyhedra of Wiseana NPV, respectively (3). DNA techniques such as dot-blot hybridization were shown to be less sensitive than our method, since detection levels between 10⁵ and 10⁶ genome copies were reported (17–19, 38). Many of the detection techniques discussed above were limited by the use of radioisotopes, which increases safety concerns. In addition, they detected baculovirus polyhedra or DNA from larval homogenates instead of soil. Polyhedron or viral DNA extraction from soil is challenging due to the strong adsorption of polyhedra to positively charged particles in the soil (11) and due to the presence of compounds that inhibit PCR amplification (35, 36).

AgMNPV detection from field soil samples.

MCH followed by PCR amplification of the polh region did not detect AgMNPV DNA in soil samples collected from control plots and AgMNPV-2D treatments 1 day after virus application. In contrast, AgMNPV DNA was detected in 70, 100, 80, and 80% of soil samples collected from the AgMNPV-2D treatments at 15, 45, 75, and 180 days postapplication, respectively. AgMNPV DNA was not detected in soil from control plots 15 days postapplication, but it was detected in 40, 20, and 10% of the soil samples from control plots collected at 45, 75, and 180 days postapplication, respectively. No viral DNA was detected from soil samples collected 330 days postapplication (Fig. 2).

FIG. 2.

PCR detection of the AgMNPV in soil samples. Percentages correspond to the number of soil samples that were PCR positive for AgMNPV in relation to the total number of soil samples collected per treatment on each sampling date. PCR amplification targeted the polh region. Error bars indicate standard errors.

The fact that AgMNPV DNA was not detected in soil samples 1 day after virus application was expected, since viral application was aimed at the soybean foliage, not at the soil. In addition, there was no larval mortality due to virus disease, and therefore, no polyhedral release from dead larvae 1 day after virus application. However, this virus was detected in the larval host and in the predator population from 1 to 45 days postapplication with PCR of the polh region as a detection method (25). The same study detected AgMNPV in the predators collected from untreated soybean samples, at least 5 days earlier than it was detected in the host population, indicating that predators may be implicated in viral dispersal. The detection of AgMNPV DNA in soil samples collected from control plots may be due to the movement of predators that had fed upon virus-diseased larvae, coupled with the actions of rain and wind, which may have carried the virus to the control plots.

Some of the PCR signal in the control and experimental plots could be due to DNA from baculoviruses other than AgMNPV, especially because the biotinylated DNA probe was designed from the polyhedrin region, which is a highly conserved region among baculoviruses. To rule out this possibility, PCR products corresponding to a total of 15 soil samples from control and treated plots at 15, 45, 75, and 180 days postapplication were digested with the restriction enzymes HincII and HhaI. These enzymes differentiate AgMNPV from at least seven other baculovirus species (24). The digestion results showed the expected restriction enzyme profiles for AgMNPV (data not shown). In addition, no PCR signal was observed when samples of larval hosts and predators from the same plots were collected before AgMNPV application and analyzed by PCR (25). Finally, the PCR products corresponding to a subsample (20%) of host larvae collected from these plots from 1 to 45 days postapplication were also digested with the enzyme HincII and found to present the expected profile for AgMNPV (25). Collectively, these results strongly indicate that there were no other baculovirus species present, at detectable levels, in the control and treated plots.

In the present study, the MCH procedure followed by PCR amplification detected AgMNPV DNA in soil samples collected in a period ranging from 15 to 180 days postapplication. This method did not detect any virus DNA in control plots or AgMNPV-2D treatments 330 days after virus application. The same virus species has been detected over a period of 7 months in soil collected from treated soybean plots in Louisiana (8). AgMNPV polyhedra were detected with neonate bioassays from soil over a period of approximately 1.5 years after virus application (39). Although the detection levels reported in that study are similar to ours, the virus was detected over a longer period. This difference could be attributed to factors such as soil type, which is not mentioned in that study, sample number, subsample number, or sample size, among others. In this study, we detected virus until 6 months postapplication. It is possible that AgMNPV detection from soil could have been achieved for a longer period if we had increased the number of samples or the subsample size in our study.

Neonate bioassays have detected 4 × 10⁴ Spodoptera frugiperda NPV polyhedra/g in sandy soils (45% sand) and 10 polyhedra/g in soils with high contents of silt or clay (9). In a subsequent study, 16 AgMNPV polyhedra/g were detected in a soil containing 2.4% sand, 76.0% silt, and 21.6% clay, and 318 polyhedra/g were detected in a soil composed of 25.0% sand, 54.0% silt, and 20.9% clay (7). In another study, 7 AcMNPV polyhedra per g of soil were detected by using bioassays with neonates; however, soil type was not mentioned (39). The studies cited above (7, 9) may indicate a trend: sensitivity of baculovirus detection in the soil may be inversely correlated with sand content. In those studies, more sensitive polyhedron detection was associated with low sand contents, while less sensitive detection was associated with higher sand contents in the soil. In our study, we detected as few as 4 AgMNPV polyhedra/g in loamy sand soils containing 70 to 85% sand. While this study evaluated the MCH procedure’s effectiveness on only one soil type, this method should effectively extract DNA from other soil types. However, the sensitivity of the method would be expected to vary according to soil type, as is reported in the literature addressing other methods of baculovirus detection from soil. Preliminary studies with the MCH procedure followed by PCR amplification have successfully extracted and amplified AcMNPV inoculated into a heavy clay-type soil (22a). Evaluation of the MCH method with different soil types is certainly warranted but was beyond the scope of this study.

A comparison of our MCH results with those found in the literature regarding neonate bioassays shows that the detection levels obtained in soils with intermediate sand content (7, 9) are higher than those observed in this field study, in which the sand levels were high. Therefore, we speculate that the MCH procedure followed by DNA amplification of the polh region is potentially more sensitive in detecting baculoviruses in soil than are neonate bioassays. However, it is important to note that no direct comparisons were made between the MCH assay and neonate bioassays. Furthermore, the MCH assays may detect released baculovirus DNA in addition to intact baculoviruses.

The MCH procedure can be modified to provide quantitative data on the number of AgMNPV polyhedra in soil. Competitive PCR (cPCR) is one of the quantitative strategies available. In preliminary studies, we attempted to quantify the AgMNPV DNA present in field-collected samples by using a cPCR approach. The basic requirements for quantitation were achieved; however, the quantitation of AgMNPV DNA by cPCR was not validated due to high inter- and intra-assay variabilities and a narrow range of quantitation (25a). Future quantitative studies using cPCR should focus on decreasing the variability between and within experiments, as well as expanding the range of quantitation.

In summary, we developed two DNA isolation procedures for the direct extraction of baculovirus DNA from soil and subsequent PCR amplification. The MCH procedure was more sensitive in extracting AgMNPV DNA from soil than were the phenol-ether extraction and other methods such as enzyme-linked immunosorbent assay, radioimmunoassay, and DNA dot-blot hybridization. The MCH procedure was also more effective in removing the humic acids that are usually coextracted with DNA from soil and are inhibitory to PCR amplification and other enzymatic procedures. This technique enabled the detection of AgMNPV DNA from field soil samples over a 6-month period. The procedures developed in this research provided a sensitive way to detect baculoviruses in soil, and they should facilitate ecological studies of baculoviruses, since soil is a major reservoir of these viruses. The ability to quantify baculovirus DNA in the soil would certainly add to the usefulness of the techniques presented in this paper, especially in risk assessment studies associated with the release of genetically modified baculoviruses.