Abstract
Corvids can be a sensitive indicator for West Nile virus (WNV) prevalence and are a component of many WNV surveillance programs. An improved sampling procedure using a bilateral intraocular cocktail (BIC) was developed for testing corvid carcasses for WNV. This new procedure was substantially faster than harvesting internal organs, required less specialized equipment and training, and yielded excellent diagnostic sensitivity.
Corvids, especially American crows (Corvus brachy-rhynchos), are extremely susceptible to West Nile virus (WNV; family Flaviviridae, genus Flavivirus) and serve as sentinels for WNV activity (Centers for Disease Control and Prevention [CDC]: 2003, Epidemic/epizootic West Nile virus in the United States: guidelines for surveillance, prevention, and control, 3rd revision. Available at: http://www.cdc.gov/ncidod/dvbid/westnile/resources/wnv-guide-lines-aug-2003.pdf. Accessed November 11, 2008). 3,5,8,14 West Nile virus has been detected in many organs in crows, including brain, heart, kidney, liver, lung, spleen, skin, and feather pulp. 1,2,5,12,15,18 Although sensitive, necropsy and harvesting of internal organs for real-time reverse transcription polymerase chain reaction (RT-PCR) is labor intensive, time-consuming, and requires specialized tools, training, and biological containment facilities (CDC: 2003, Epidemic/epizootic West Nile virus in the United States; CDC: 2007, Biosafety in microbiological and biomedical laboratories, 5th ed. Available at: http://www.cdc.gov/OD/ohs/biosfty/bmbl5/BMBL_5th_Edition.pdf. Accessed November 13, 2008). In an attempt to increase efficiency, reduce risk to personnel, and maintain or improve diagnostic sensitivity, an improved sampling method using a bilateral intraocular cocktail (BIC) for detecting WNV in corvids was developed.
Dead corvids were collected by the San Diego County Vector Control program from June through December 2008. Corvids in good-to-fair postmortem condition (not dismembered or ridden with insects) and assessed to be dead for ≤48 hr were analyzed for WNV. If the birds could not be processed on the same day, they were frozen to −80°C and processed the following business day. Each corvid was placed in a separate bag to avoid cross-contamination. All tissue and BIC collection procedures were performed within a biological safety cabinet a using appropriate personal protective equipment. Fifty to 100 mg each of heart, spleen, kidney, and brain (cerebellum) were collected and pooled in a 1.5-ml microfuge tube (without diluent) using a separate set of cleaned and autoclaved instruments for each bird. Bilateral intraocular cocktail sampling was performed by fitting a 16-gauge sterile needle to a sterile 3-ml syringe. Without removing the eyes from the carcass, the needle was inserted through the center of the cornea and used to scrape, vigorously mix, and aspirate the interior of the eye, yielding fragments of retina, uvea, pecten, and vitreous and aqueous humors. Some eyes were transected post-BIC extraction to verify the collection of these structures. This procedure yielded 1–3 ml of BIC aspirated from the globes of each bird. Moderately dehydrated and collapsed eyes yielded adequate amounts of BIC as well. All collected samples were frozen to −70°C, unless processed the same day for nucleic acid extraction.
Bilateral intraocular cocktail (BIC) and tissue pool real-time reverse transcription polymerase chain reaction West Nile virus results for corvids tested from June–December 2008.
For nucleic acid extraction, approximately 25 mg total of heart, spleen, kidney, and brain were suspended in 600 μl tissue lysis buffer b with 1% β-mercaptoethanol (BME). Tissues were homogenized with ceramic beads c by mechanical lysis d for 30 sec. Approximately 550 μl of tissue homogenate suspension was recovered per sample and passed through a commercially available spin column e to filter out insoluble debris from the lysates. Purified RNA was extracted from the lysates with a commercially available kit. b The RNA concentration for each sample was determined by using a spectrophotometer f and then standardized to 100 ng/μl prior to being tested by real-time RT-PCR.
Bilateral intraocular cocktail samples were processed using the same procedure as described above for the tissues except for the following modifications: 140 μl of BIC was added to 600 μl tissue lysis buffer with 1% BME. This mixture was homogenized via either mechanical lysis as mentioned above or vortexed on a thermomixer g at 2,000 × g for 10 min at room temperature. All other steps were identical to those used with the tissue homogenates.
Hybridization real-time RT-PCR was performed on the RNA extracted from pooled tissue homogenates, as well as BIC. A commercially available kit specific for the NS4 gene of WNV was used per the manufacturer's specifications. h All RT-PCR reactions were performed on a Roche LightCycler 480. i Amplification cycling times were as follows: 1 cycle for reverse transcription at 61°C for 30 min; 1 cycle for denaturation at 95°C for 30 sec at 4.4 ramp rate; 45 cycles for amplification at 95°C/4 sec/4.4 ramp, 50°C/6 sec/2.2 ramp, 72°C/15 sec/4.4 ramp. A cycle threshold (Ct) of 38 or less was considered positive for target sequence amplification.
Two hundred forty-two corvids were tested from June 2008 through December 2008. The corvids consisted of 228 American crows (Corvus brachyrhynchos), 5 common ravens (C. corax), 8 Western scrub jays (Aphelocoma californica), and 1 Steller's jay (Cyanocitta stelleri). Of these birds, 77.3% (187/242) tested positive for WNV in tissues by real-time RT-PCR, and 81.0% (196/242) tested positive in BIC by real-time RT-PCR (Table 1). Ten crows were positive in BIC but negative in tissues. One crow was positive for WNV in tissue and negative in BIC. The diagnostic sensitivity of BIC testing was 99.5% (196/197) versus 95.0% (187/197) for tissue pool testing. The negative predictive value for BIC was 97.4% compared with 80.9% for tissues.
The mean Ct for BIC was greater than the mean Ct of tissues (95% confidence interval: 3.64, 5.42, P value <0.01 using a 1-tailed t-test; Table 2). The cause of the almost 4 Ct difference is not known, but the amount of virus recovered from BIC may have been diluted by aqueous and vitreous humors, thus increasing the BIC mean Ct. 7 Despite this, the diagnostic sensitivity and negative predictive value of BIC were superior to tissues. In birds that were positive for WNV in BIC but negative in tissues, Ct values ranged from 16.95 to 34.95.
Cycle threshold (Ct) values of bilateral intraocular cocktails (BIC) and tissue pools from corvids tested for West Nile virus.
Previous studies have demonstrated endophthalmitis and the frequent presence of WNV antigen and live virus in ocular tissues of several species of raptors. 9,10,13,16,17 In experimentally infected raptors, WNV antigen was most frequently recovered from the eyes, suggesting that the eye could be a sensitive indicator of WNV infection. 9 In contrast to a previous study 13 that showed 56% sensitivity of aqueous humor for detecting naturally occurring WNV infection in raptors, the BIC in the present study yielded close to 100% sensitivity for WNV detection in corvids. This may be due to the inclusion of many ocular structures in BIC, which have been previously shown to harbor WNV antigen, and the relatively high viremia titers that corvids (e.g., American crows) sustain relative to raptors. 5 The presence of WNV in the eye but not tissues in 10 of the corvids may be due, in part, to the immune-privileged property of eyes. 4,7,11
The overall positive rate and sensitivity of BIC was comparable to a previous study that demonstrated WNV in feather pulp from American crows and blue jays (Cyano-citta cristata). 1 In the present study, BIC was compared with feather pulp in 6 birds (data not shown). Two birds were positive for WNV in BIC and feather pulp; the remaining 4 birds were negative in BIC, feather pulp, and internal organs. In the 2 positive birds, the Ct values were higher in feather pulp, suggesting less virus present 6 in the feather pulp than in the BIC samples. In addition, harvesting feather pulp was more difficult and time-consuming than collecting BIC. Pulling flight feathers required significant tension and carcass handling and was ergonomically difficult to perform within the confines of a biological safety cabinet so as not to aerosolize potentially contaminated feces or blood on the surface of the feathers. Collecting BIC required only that the head be exposed from a bag holding the bird within the biological safety cabinet. The risk of aerosolizing contaminated dust, feces, and blood was less with the BIC procedure than either necropsy or pulling out feathers. Sampling BIC may be a viable alternative to feather pulp sampling because of the ease of sample collection and not being limited by molt cycles. 1
Sampling internal organs for diagnosing WNV in corvids requires harvesting internal organs via necropsy (CDC: 2003, Epidemic/epizootic West Nile virus in the United States). 2,5,12 This is labor intensive, requires moderate training and necropsy instruments, and must be performed safely following BSL-3 practices when dissecting field-collected dead birds due to the potentially high levels of WNV found in such samples (CDC: 2007, Biosafety in microbiological and biomedical laboratories). In an attempt to improve the efficiency and safety of WNV testing, the use of a simple bilateral ocular extraction procedure to accurately test corvids for WNV via real-time RT-PCR was explored. By using the BIC protocol, which samples retina, uvea, pecten, and vitreous and aqueous humors, collection time decreased from 10–20 min per bird to less than 1–2 min and yielded a more diagnostically sensitive assay than harvesting other organs. Extreme care must still be exercised when sampling BIC; following BSL-3 standards is recommended. West Nile virus surveillance programs that test corvids may benefit from adopting the BIC sampling method.
Acknowledgements. The authors would like to thank the San Diego County Vector Control program for their corvid submissions and support of this work, as well as the staff at the County of San Diego Animal Disease Diagnostic Laboratory for their assistance. The authors would also like to thank Sonia Hazarika, a Charles Louis Davis Foundation student intern, for her assistance with necropsies; Nicole Nemeth of Colorado State University for her technical assistance; and Arno Wünschmann of the University of Minnesota for manuscript review.
Footnotes
a.
Purifier Class II BSC (Delta Series), Labconco, Kansas City, MO.
b.
RNeasy Mini Kit, Qiagen Inc., Valencia, CA.
c.
MagNA Lyser Green Beads, Roche Diagnostics Corp., Indianapolis, IN.
d.
MagNA Lyser Instrument, Roche Diagnostics Corp., Indianapolis, IN.
e.
Qiashredder, Qiagen Inc., Valencia, CA.
f.
Nanodrop 1000, Thermo Fisher Scientific Inc., Wilmington, DE.
g.
Thermomixer R, Eppendorf, Westbury, NY.
h.
LightCycler WNV Detection kit, Roche Diagnostics Corp., Indianapolis, IN.
i.
LightCycler 480 Instrument, Roche Diagnostics Corp., Indianapolis, IN.