Dennis J. Kohler

Large-Scale Avian Influenza Surveillance in Wild Birds throughout the United States

Avian influenza is a viral disease that primarily infects wild and domestic birds, but it also can be transmitted to a variety of mammals. In 2006, the United States of America Departments of Agriculture and Interior designed a large-scale, interagency surveillance effort that sought to determine if highly pathogenic avian influenza viruses were present in wild bird populations within the United States of America. This program, combined with the Canadian and Mexican surveillance programs, represented the largest, coordinated wildlife disease surveillance program ever implemented. Here we analyze data from 197,885 samples that were collected from over 200 wild bird species. While the initial motivation for surveillance focused on highly pathogenic avian influenza, the scale of the data provided unprecedented information on the ecology of avian influenza viruses in the United States, avian influenza virus host associations, and avian influenza prevalence in wild birds over time. Ultimately, significant advances in our knowledge of avian influenza will depend on both large-scale surveillance efforts and on focused research studies.

Diffusion of influenza viruses among migratory birds with a focus on the Southwest United States

The Southwest United States, including Arizona and New Mexico, has a diverse climate and is home to many different avian species. We sequenced the hemagglutinin (HA) gene of twenty influenza specimens for the years 2007-2009. This included four from Arizona, and sixteen from New Mexico. We analyzed the sequences and determined the following HA subtypes: H3, H4, H6, H8, and H11. For each subtype, we combined our virus sequences with those from a public database, and inferred phylogeographic models of influenza diffusion. Statistical phylogeography indicated that overall evolutionary diffusion of avian influenza viruses is geographically structured (p<0.05). In addition, we found that diffusion to the Southwest was often from nearby states including California. For H3, H4 and H6, the intra-flyway gene flow rates were significantly (p<0.001) higher than those of inter-flyway. Such rate difference was also observed in H8 and H11, yet, without statistical significance (p=0.132, p=0.190, respectively). Excluding any one flyway from the calculation generated similar results, suggesting that such barrier effect on gene flow rates is not exclusively produced by any single flyway. We also calculated the Bayes factor test for the significant non-zero rates between states and identified significant routes both within and across flyways. Such inter-flyway spread of influenza was probably the result of birds from four flyways co-mingling on breeding grounds in northern regions or marshaling on staging areas post breeding in Canada or Alaska, before moving south each fall. This study provides an initial analysis of evolutionary diffusion of avian influenza virus to and from the Southwest United States. However, more sequences from this region need to be generated to determine the role of host migration and other factors on influenza diffusion.

Viremia in North American Mammals and Birds After Experimental Infection with Chikungunya Viruses

Chikungunya virus (CHIKV) is an arthropod-borne virus, which is known to cause severe disease only in humans. To investigate its potential zoonotic host range and evaluate reservoir competence among these hosts, experimental infections were performed on individuals from nine avian and 12 mammalian species representing both domestic and wild animals common to North America. Hamsters and inbred mice have previously been shown to develop viremia after inoculation with CHIKV and were used as positive controls for infection. Aside from big brown bats (Eptesicus fuscus), none of the mammals or birds developed detectable viremia or overt clinical disease. However, most mammals and a smaller proportion of birds developed neutralizing antibody responses to CHIKV. On the basis of these results, it seems unlikely that CHIKV poses a significant health threat to most domestic animals or wildlife and that the species examined do not likely contribute to natural transmission cycles. Additional studies should further evaluate bats and wild rodents as potential reservoir hosts for CHIKV transmission during human epidemics.

IMMUNOLOGICAL AND CLINICAL RESPONSE OF COYOTES (CANIS LATRANS) TO EXPERIMENTAL INOCULATION WITH YERSINIA PESTIS

Abstract Multiple publications have reported the use of coyotes (Canis latrans) in animal-based surveillance efforts for the detection of Yersinia pestis. Coyotes are likely exposed via flea bite or oral routes and are presumed to be resistant to the development of clinical disease. These historic data have only been useful for the evaluation of the geographic distribution of Y. pestis in the landscape. Because the canid immunologic response to Y. pestis has not been thoroughly characterized, we conducted experimental inoculation of captive-reared, juvenile coyotes (n = 8) with Y. pestis CO92 via oral or intradermal routes. We measured the humoral response to Y. pestis fraction 1 capsular protein (anti-F1) and found a significant difference between inoculation groups in magnitude and duration of antibody production. The anti-F1 titers in animals exposed intradermally peaked at day 10 postinoculation (PI; range = 1∶32 to 1∶128) with titers remaining stable at 1∶32 through week 12. In contrast, orally inoculated animals developed higher titers (range = 1∶256 to 1∶1,024) that remained stable at 1∶256 to 1∶512 through week 6. No clinical signs of disease were observed, and minimal changes were noted in body temperature, white blood cell counts, and acute phase proteins during the 7 days PI. Gross pathology was unremarkable, and minimal changes were noted in histopathology at days 3 and 7 PI. Rechallenge at 14 wk PI via similar dosage and routes resulted in marked differences in antibody response between groups. Animals in the orally inoculated group produced a striking increase in anti-F1 titers (up to 1∶4,096) within 3 days, whereas there was minimal to no increase in antibody response in the intradermal group. Information gathered from this experimental trial may provide additional insight into the spatial and temporal evaluation of coyote plague serology.

Determination of Acetaminophen Residues in Whole Body Brown Treesnakes

Abstract Acetaminophen was extracted from brown treesnakes (Boiga irregularis) and analyzed by reversed‐phase high‐performance liquid chromatography (HPLC). Acetaminophen was quantified by UV absorbance at 250 nm. Recoveries were determined by analyzing acetaminophen‐fortified blank homogenized tissue. The mean recovery of acetaminophen in whole body brown treesnakes was 87.9% ± 5.9% and 92.2% ± 5.8% for the fortification levels of 20 and 2400 µg/g, respectively. The methods limit of detection (MLOD) with UV detection was 0.70 µg/g.

EFFECT OF STORAGE TIME AND STORAGE CONDITIONS ON ANTIBODY DETECTION IN BLOOD SAMPLES COLLECTED ON FILTER PAPER

Abstract Using filter paper to collect blood from wildlife for antibody analysis can be a powerful technique to simplify the collection, transport, and storage of blood samples. Despite these advantages, there are limited data that detail how long these samples can be stored and how storage conditions affect antibody longevity. We used blood samples collected on filter paper from coyotes experimentally infected with Yersinia pestis to determine optimum sample storage conditions over time. Blood samples collected on filter paper were stored for 454 d or more in four groups: 1) at ambient temperature and at ambient relative humidity, 2) at ambient temperature with desiccant, 3) at 4 C with desiccant, and 4) at −20 C with desiccant. Samples stored at 4 C or −20 C with desiccant had detectable antibody for a longer period of time than the samples stored at room temperature.

Response of Captive Skunks to Microencapsulated Tetracycline

A captive striped skunk (Mephitis mephitis) study was conducted between February and June 2004 at the United States Department of Agriculture, Animal and Plant Health Inspection Service, Wildlife Services National Wildlife Research Center, Fort Collins, Colorado, USA. The main objective was to determine the percentage of adult striped skunks that were marked after consuming placebo oral rabies vaccine (ORV) baits containing 100 mg of an experimental microencapsulated (coated microparticle) tetracycline hydrochloride biomarker. Biomarkers were identified in the canine teeth and mandibles of five of five skunks that consumed an ORV bait. A second objective was to determine if the microencapsulated tetracycline was resistant to photochemical conversion from tetracycline to epitetracycline. After 15 days of exposure, conversion from tetracycline to epitetracycline concentration in the microencapsulated product (mean 1.9% conversion, SD=1.24) was significantly less (P=0.006) than the pure-grade tetracycline powder (mean 7.5% conversion, SD=1.37). Results support the use of microencapsulated tetracycline hydrochloride as a biomarker in circumstances where the use of conventional powdered tetracycline hydrochloride is not feasible due to ORV bait design constraints.