© 2009 Plant Management Network.
Evaluation of a High-Throughput Protocol for Detecting Blueberry shock virus in Vaccinium Using ELISA
Mohamed (Sid) Sedegui and Nancy K. Osterbauer, Oregon Department of Agriculture, 635 Capitol Street NE, Salem, OR 97301-2532
Sedegui, M., and Osterbauer, N. K. 2009. Evaluation of a high-throughput protocol for detecting Blueberry shock virus in Vaccinium using ELISA. Online. Plant Health Progress doi:10.1094/PHP-2009-0710-01-RS.
The Oregon Department of Agriculture’s official virus testing program for Vaccinium has grown 915% since its inception. To address this increased workload, we evaluated a high-throughput ELISA protocol that uses a 96-well format for sap extraction. To verify the efficacy of the protocol, it was used to test leaves from known infected and healthy plants (positive and negative controls) and from 940 Vaccinium plants grown in a nursery with Blueberry shock virus (BlShV) problems. Each leaf collected was cut in half; one-half was macerated in an extraction buffer using mesh sample bags and a roller press (standard protocol) and the other half was placed into 10 × 96 collection microtubes and macerated in the extraction buffer using a Mixer Mill MM300 (high-throughput protocol). All sample sap extractions were then tested with a commercial ELISA kit. In experiments with the BlShV positive and negative control plants, the high-throughput protocol had high sensitivity (100%) and specificity (100%), matching the standard protocol. In experiments with the blueberry plants of unknown BlShV infection status, the high-throughput protocol had a sensitivity of 94.85% and a specificity of 96.70%, compared to the standard protocol. It is concluded the high-throughput protocol can be used successfully with ELISA to detect BlShV in Vaccinium leaves.
In 2002, at the request of the nursery and blueberry industries, Oregon adopted an Administrative Rule to protect blueberries (Vaccinium L.) from the inadvertent importation of the Sheep Pen Hill variant of blueberry scorch virus (BlScV) (6,9). In 2004, the Oregon Department of Agriculture (ODA) began offering an official testing program for viruses of regulatory concern including BlScV and blueberry shock (BlShV) to nurseries shipping Vaccinium nursery stock out-of-state (5,8). That first year, five nurseries participated in the program, requiring the testing of 2,480 plants (8). In 2007, the 10 nurseries participating in the program requested testing for 22,695 samples, a 915% increase since 2004 (11).
When possible, ELISA is used by ODA for virus testing because it is relatively inexpensive and reliable (2,7,15). However, the plant sap extraction step in the standard ELISA testing protocol is applied to one sample at a time, requiring large amounts of time and labor for large numbers of samples (4). Due to its rapid growth, ODA’s official blueberry testing program became costly and time consuming. By Administrative Rule, these costs had to be passed onto the participating nurseries (9,10). Delays in such a testing program become problematic for nurseries attempting to ship plants to meet customers’ needs.
To address these problems, a high-throughput 96-well format sap extraction protocol (14) was adapted for use with a commercial ELISA kit and tested for efficacy. Here, the diagnostic sensitivity and specificity of ELISA or BlShV in blueberry is compared for the high-throughput sap extraction protocol and a standard sap extraction protocol. Also, the potential time and cost savings associated with the high-throughput protocol is explored. Previously, we tested the efficacy of the high-throughput sap extraction protocol for detecting Phytophthora in leaf tissue (14). To our knowledge, this protocol has not been applied to testing leaf tissue for viruses using ELISA.
Testing Known Infected and Healthy Plants
Leaf samples were collected from known BlShV-infected and healthy blueberry plants. Dr. Joseph Postman (USDA National Clonal Germplasm Repository, Corvallis, OR) provided the known plants. Initial symptom expression in BlShV-infected plants consists of sudden death of the flowers and young vegetative leaf shoots in the spring just as the flowers are about to open (1). Typically, a second flush of leaves is produced later in the season, such that infected plants look normal but have little fruit. Then, the plants become asymptomatic carriers of BlShV; our known infected plants were at this latter stage. Thus, we collected leaves from anywhere within the foliage of our infected and healthy plants.
One leaf was collected per plant to be tested, with samples from five plants combined into one composite sample for ELISA testing. For this controlled study, each leaf collected was cut in half, with one-half of the leaf extracted with the standard protocol and the other with the high-throughput protocol. For each known infected composite sample, tissue from one infected half leaf was mixed with tissue from four healthy half leaves. This was done to mimic the lowest amount of infected leaf material that may be present in a composite sample. For each known healthy composite sample, five healthy half leaves were used.
For the standard sap extraction protocol, a 1:10 (w:v) ratio of leaf tissue to extraction buffer was used in the testing. The leaf tissue for each composite sample was placed into a mesh sample bag (catalog no. ACC00930, Agdia Inc., Elkhart, IN) and the appropriate amount of extraction buffer was added. The tissue was then macerated with a roller press (catalog no. ACC00700, Agdia Inc.) attached to a heavy-duty drill press for approximately 30-sec. The resulting sap extracts were then loaded into an ELISA test plate. One hundred microliters per composite sample was loaded per test well and two test wells were loaded per composite sample.
For the high-throughput sap extraction protocol, a sterile punch was used to remove a 6-mm diameter disk of leaf tissue from each half leaf in the composite sample. The leaf disks were combined and then placed in a 96-well extraction plate (catalog no. 19560, Qiagen Inc., Valencia, CA). In an effort to automate the system as much as possible by determining a standard buffer volume for all composite samples, three extraction buffer volumes were tested: 600 µL, 750 µL, and 900 µL. A single tungsten carbide bead (catalog no. 69989, Qiagen Inc., Valencia, CA) was added to each test well after the extraction buffer was added. The test wells were sealed to ensure sample integrity and then agitated for 1.5-minutes at 30 MHz in a Mixer Mill MM300 (Retsch Inc., Newtown, PA). The extraction plates were rotated and the agitation repeated. The resulting sap extracts were then loaded into an ELISA test plate. As with the standard protocol, 100 µL per composite sample was loaded per test well and two test wells were loaded per composite sample.
The extracted blueberry leaf samples were tested with a commercially available ELISA test kit for BlShV (catalog no. SRA19200, Agdia Inc., Elkhart, IN). ELISA test plates were coated with the BlShV capture antibody in our laboratory according to the manufacturer’s protocol prior to testing. Commercially available positive and negative controls were included on each ELISA plate with two test wells loaded per control (catalog nos. LNC19200 and LPC19200, Agdia Inc., Elkhart, IN). The compound direct (triple antibody sandwich) ELISA test was performed according to the manufacturer’s directions. After a 0.5-h incubation in the dark, the optical density (OD) of the alkaline phosphatase label was read at 405 nm in a microplate reader (BioRad Inc., Hercules, CA). Per the ELISA kit manufacturer’s directions, the positive/negative threshold was set at 2.0× the OD of the negative control.
A total of 12 composite samples were tested with ELISA; six samples included infected leaf tissue and six did not. For the standard sap extraction protocol, a total of 0.5 g of leaf tissue was macerated in 5-mL of extraction buffer for each sample. In samples containing infected leaf tissue, 0.1 g of infected tissue was mixed with 0.4 g of healthy tissue. For the high-throughput sap extraction protocol, samples containing infected tissue included one 6-mm disk of infected tissue to four 6-mm disks of healthy tissue. However, the actual ratio of infected to healthy tissue by weight was 1:2 (Table 1). Why the infected leaf disks weighed more than the healthy disks is unknown. We replicated this experiment and got virtually identical results both times. The results from the first replicate are presented here.
Table 1. Blueberry shock virus-infected and healthy composite samples subjected to the high-throughput sap extraction protocol in the first replicate of the controlled study.
The diagnostic sensitivity and specificity of the high-throughput protocol for sap extraction were calculated using a 2 × 2 contingency table used to assess diagnostic tests (3,12). The high-throughput sap extraction ELISA protocol results were compared to the standard sap extraction ELISA protocol results. Where both protocols resulted in a positive test result, samples were allocated the letter A (true positives) and where both protocols resulted in a negative test result, the letter D (true negatives). Where the samples processed with the high-throughput sap extraction protocol were positive while the standard protocol was negative, the letter B was assigned (false positives). Where the samples processed with the high-throughput sap extraction protocol were negative while the standard protocol was positive, the letter C was assigned (false negatives). The diagnostic sensitivity [A / (A + C)] as a measure of a diseased sample testing positive and the diagnostic specificity [D / (D + B)] as a measure of a disease-free sample testing negative were calculated using these formulae.
The ELISA test results for the 12 samples using the high-throughput sap extraction protocol matched those using the standard sap extraction protocol (Table 2). Thus, the high-throughput protocol’s diagnostic sensitivity and specificity each were 100%. We did exceed the ELISA kit manufacturer’s recommendation of a 1:10 (w:v) ratio of leaf tissue to extraction buffer with the high-throughput protocol (Table 1). Results from this controlled study showed the ELISA kit successfully detected BlShV at tissue to buffer ratios as low as 1:30 (w:v) (Tables 1 and 2). In our study, we chose to use a 1:20 tissue to buffer ratio as the maximum extraction buffer volume in further experiments with the high-throughput protocol. This volume provided ample sap extract for replicated testing. Also, the controlled study indicated the commercial ELISA kit successfully and consistently detected BlShV when this tissue to buffer ratio was used.
Table 2. ELISA results from testing known infected and healthy blueberry plants for blueberry shock virus using the high-throughput and standard protocols for sap extraction.
x A true positive sample tested positive using both sap extraction protocols, a false positive tested positive using only the high-throughput protocol, a false negative tested positive using only the standard protocol, and a true negative tested negative with both sap extraction protocols. Results shown are from the first replicate of this experiment.
y Optical density readings are an average across two test wells for each composite sample.
z Not applicable. There were no false positives or false negatives detected.
Testing Nursery Samples of Unknown Infection Status
To verify the efficacy of the high-throughput sap extraction protocol, individual leaves were collected from 940 blueberry plants grown at a nursery with a history of BlShV. Leaves were collected at a rate of one per plant to be tested, with leaf samples from five plants from the same block combined into one composite sample for ELISA testing. As with the controlled study, each leaf collected was cut in half, with one-half of the leaf extracted with the standard sap extraction protocol and the other with the high-throughput sap extraction protocol. A total of 188 composite samples were tested with ELISA as described above.
In this study, the diagnostic sensitivity of the high-throughput protocol as compared to the standard protocol was 94.85% while the specificity was 96.70% (Table 3). There was disagreement between the ELISA test results for the two sap extraction protocols for eight composite samples; five samples were designated false negatives (negative with high throughput and positive with standard) while three were designated false positives (positive with high throughput and negative with standard). Sufficient leaf tissue remained from five of the eight inconclusive samples for re-testing, specifically the three false positive samples and two of the false negative samples. The remaining leaf tissue from these five samples were stored at -20°C for 1-week and then placed in a cooler for same-day delivery to the laboratory of Dr. Robert Martin (USDA-Agricultural Research Service, Corvallis, OR) for confirmatory testing. Based on his results, all five of these inconclusive samples were negative.
Table 3. ELISA results for composite environmental samples of blueberry leaf tissue tested for blueberry shock virus using the high-throughput and standard protocols for sap extraction.
x A true positive sample tested positive using both sap extraction protocols, a false positive tested positive using only the high-throughput protocol, a false negative tested positive using only the standard protocol, and a true negative tested negative with both sap extraction protocols.
y Optical density readings are an average across two test wells for each composite sample.
Efficiency of the Sap Extraction Protocols
One goal of this study was to develop a more efficient and cost effective testing protocol for blueberry samples. We determined the amount of time it took to process one complete ELISA plate (46 samples plus positive and negative controls) using the two sap extraction protocols. For the standard protocol, it required an average of 30 min to record sample origin information and weigh out the sample tissue. Macerating the tissue required an additional 23 min. Finally, loading the sample extracts into the ELISA plate required another 25 min. On average, the overall sap extraction time for one ELISA plate using the standard protocol was 78 min. For the high-throughput protocol, it required an average of 20 min to record the sample origin information and load the 96-well extraction plate because the tissue did not need to be weighed. Macerating the tissue in the Mixer Mill MM300 required an additional 10 min and loading the sample extracts into the ELISA plate another 3 min. On average, the overall sap extraction time for one ELISA plate using the high-throughput protocol was 33 min.
Performing sap extractions using the high-throughput protocol was successful for the detection of BlShV in blueberry leaf tissue by ELISA. The diagnostic sensitivity and specificity of the test results obtained using the high-throughput protocol compared favorably to test results obtained using a standard protocol for sap extraction. In addition, less than half of the time was needed to extract sap with the high-throughput protocol than with the standard protocol.
There was an acceptable level of agreement between ELISA test results for samples extracted with the high-throughput protocol and with the standard protocol. The diagnostic sensitivity and specificity of the high-throughput protocol were both 100% during the controlled study with known infected and healthy plants. When testing nursery plants whose infection status was unknown, the diagnostic sensitivity (94.85%) and specificity (96.70%) of the high-throughput protocol remained favorable when compared to the standard protocol. However, false negatives and false positives were detected.
When testing for pathogens of regulatory interest, false positives are less of a concern than false negatives. This is because a plant that tests positive initially must be re-tested to confirm the initial test results (15). Re-testing by an independent laboratory of the three false positives in this study indicated all three were negative, verifying the need for confirmatory testing. False negatives are of greater concern to plant health regulators; five were detected. Sufficient leaf tissue remained to perform confirmatory testing on two of the five samples. Both samples tested negative for BlShV, indicating the initial ELISA test results with the high-throughput protocol were correct. In a previous study, researchers cited possible reasons for such discrepancies in ELISA testing as a very low level of infection within the leaves tested or poor extraction of the sample (4). Also, the remaining leaf tissue from the samples was frozen at -20°C prior to delivery to the independent laboratory for confirmatory testing. Depending on how samples are stored, freezing can affect the detection of different viruses in leaf tissue with ELISA (13,16). No studies have been done that specifically address the impact of freezing on the detection of BlShV in Vaccinium leaf tissue. However, it is possible that storing the samples at -20°C affected the independent laboratory’s test results.
The high-throughput protocol requires significantly less time to perform leaf sap extractions than the standard protocol. In 2007, our laboratory tested 22,695 blueberry plants for BlShV (11). Based on the average time needed to extract samples with the standard protocol, it took 641.4 personnel hours to extract the blueberry samples. In contrast, the high-throughput protocol would have taken 271.4 personnel hours to extract the same number of samples, more than halving the time needed to provide test results to participating nurseries. Also, in 2007, the wage of an entry-level technician within our laboratory was $10.56 per hour excluding benefits. Using these figures, it would cost $6,773 and $2,865 to perform leaf sap extractions with the standard protocol and the high-throughput protocol, respectively. Thus, the high-throughput protocol would have resulted in a savings of more than $3,907 per year in personnel salary costs. It was estimated that this savings would offset costs for the Mixer Mill MM300 in about 2-years.
As demonstrated here, the high-throughput protocol can be used successfully to extract sap from blueberry leaves for subsequent ELISA testing for BlShV. In our laboratory, this protocol provides for accurate testing that is more efficient and economical because of the large number of samples tested. However, when only a few samples are tested, the standard sap extraction protocol can be as efficient and economical because we can test a large number of samples significantly faster than with a standard sap extraction protocol. We generally use the standard protocol for confirmatory testing on samples that initially tested positive. The potential for inaccurate test results is always a concern for plant health regulators and diagnosticians. Several issues can contribute to inaccurate test results including low virus titer, poor extraction, and cross contamination of test wells. When performing the high-throughput sap extraction protocol, ensuring the collection tubes in the extraction plate are tightly sealed prior to maceration is critical to maintaining sample integrity. Also, using an eight channel pipettor minimizes the opportunity for cross contamination while loading sample sap extracts.
The authors wish to thank Shannon Lane, Christina Bellert, Sylvia Miller, and ODA’s Nursery Inspectors for their excellent assistance collecting and processing the blueberry samples. Mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product by the ODA and does not imply its approval to the exclusion of other products or vendors that may also be suitable.
1. Caruso, F. L., and Ramsdell, D. C., eds. 1995. Compendium of Blueberry and Cranberry Diseases. American Phytopathological Society, St. Paul, MN.
2. Katoch, M., Abdin, M. Z., Ram, R., and Zaidi, A. A. 2003. An overview of diagnostics for viruses infecting gladiolus. Crop Prot. 22:153-156.
3. Lane, C. R., Hobden, E., Walker, L., Barton, V. C., Inman, A. J., Hughes, K. D. J., Swan, H., Colyer, A. and Barker, I. 2007. Evaluation of a rapid diagnostic field test kit for identification of Phytophthora species, including P. ramorum and P. kernoviae at the point of inspection. Plant Pathol. 56:828-835.
4. MacDonald, J. D., Stites, J., and Kabashima, J. 1990. Comparison of serological and culture plate methods for detecting species of Phytophthora, Pythium, and Rhizoctonia in ornamental plants. Plant Dis. 74:655-659.
5. MacDonald, S. G., Martin, R. R., and Bristow, P. R. 1991. Characterization of an Ilarvirus associated with a necrotic shock reaction in blueberry. Phytopathology 81:210-214.
6. Martin, R. R., MacDonald, S. G., and Podleckis, E. V. 1992. Relationships between blueberry scorch and sheep pen hill viruses of highbush blueberry. Acta Hortic. 308:131-139.
11. Oregon Department of Agriculture. 2008. 2007 Plant Health Section Annual Report -Commodity Inspection Division. N. Osterbauer, ed. Commodity Inspection Div., Oregon Dept. of Agric., Salem, OR.
12. Shapiro, D. E. 1999. The interpretation of diagnostic tests. Stat. Meth. Medic. Res. 8:113-134.
13. Torrance, L., and Dolby, C.A. 1984. Sampling conditions for reliable routine detection by enzyme-linked immunosorbent assay of three ilarviruses in fruit trees. Online Ann. Appl. Biol. 104:267-276. doi: 10.1111/j.1744-7348.1984.tb05611.x.
14. Trippe, A., Berghauer, E., and Osterbauer, N. 2008. A high-throughput system for the detection of Phytophthora ramorum in susceptible plant species: A preliminary report. Pages 427-434 in: Proc. of the Sudden Oak Death 3rd Sci. Symp. PSW-GTR-214, USDA Forest Service, Washington, DC.
16. Ward, C. M., Walkey, D. G. A., and Phelps, K. 1987. Storage of samples infected with lettuce or cucumber mosaic viruses prior to testing by ELISA. Online. Ann. Appl. Biol. 110:89-95. doi: 10.1111/j.1744-7348.1987.tb03235.x.