Vesicular stomatitis New Jersey virus (VSNJV) is an insect-transmitted Rhabdovirus causing vesicular disease in domestic livestock including cattle, horses, and pigs. Natural transmission during epidemics remains poorly understood, particularly in cattle, one of the most affected species during outbreaks. This study reports the first successful transmission of VSNJV to cattle by insect bite resulting in clinical disease. When infected black flies (Simulium vittatum Zetterstedt) fed at sites where VS lesions are usually observed (mouth, nostrils, and foot coronary band), infection occurred, characterized by local viral replication, vesicular lesions, and high neutralizing antibody titers (>1:256). Viral RNA was detected up to 9 d postinfection in tissues collected during necropsy from lesion sites and lymph nodes draining those sites. Interestingly, when flies were allowed to feed on flank or neck skin, viral replication was poor, lesions were not observed, and low levels of neutralizing antibodies (range, 1:8–1:32) developed. Viremia was never observed in any of the animals and infectious virus was not recovered from tissues on necropsies performed between 8 and 27 d postinfection. Demonstration that VSNJV transmission to cattle by infected black flies can result in clinical disease contributes to a better understanding of the epidemiology and potential prevention and control methods for this important disease.
Vesicular stomatitis (VS) is a disease characterized by the development of vesicular lesions on the mouth (lips, gums, tongue), nostrils, and coronary bands of the hooves of cattle, swine, and horses. Lesions are also observed in the prepuce in male horses and pigs or teats of lactating cattle (Letchworth et al. 1999). The causative agents, vesicular stomatitis virus serotypes New Jersey (VSNJV) and Indiana (VSIV), are members of the genus Vesiculovirus in the family Rhabdoviridae. VSNJV is the main cause of vesicular disease throughout the Americas and occurs sporadically in the western United States (Rodriguez 2002). The most recent outbreaks occurred in 2004–2006 in nine western states and resulted in >700 premises and >19,000 cattle under quarantine (USDA–APHIS, Animal Health Monitoring and Surveillance; http://www.aphis.usda.gov/vs/nahss/equine/vsv/usaha_2006_VSV_presentation.ppt) (Rainwater-Lovett et al. 2007). Despite significant economic losses, knowledge surrounding VSNJV transmission remains unclear.
Efforts to answer fundamental questions regarding VSNJV biology and epidemiology in cattle, including transmission, have been limited to sporadic entomological studies during previous epidemics(Walton et al. 1987). Little is known about the natural routes of transmission among animals and the factors determining the occurrence of clinical disease. Clinical disease by insect transmission to domestic swine and horses has been previously documented (Mead et al. 2004a, b). Subclinical infection after insect bite has also been reported in cattle (Perez de Leon and Tabachnick 2006). However, clinical disease resulting from insect transmission has not been previously shown in cattle, one of the most commonly affected species during VS outbreaks throughout the Americas.
Until recently, scientific data to support the long-standing hypothesis that biting insects transmitted VSNJV to livestock during epidemics were lacking (Heiny 1945). In a previous study, Scherer et al. (2007) showed that cattle could be consistently infected with VSNJV by scarification; however, the development of infection and clinical signs was dependent on the site of inoculation. Only animals inoculated in the mouth or coronary band developed clinical disease, whereas animals inoculated on the skin of the neck or flank did not. In this study, our objective was to determine the ability of black flies (Simulium vittatum Zetterstedt) to transmit VSNJV to cattle and cause clinical disease after feeding at sites where VS lesions occur versus those sites where they do not. The development of clinical VS, characterized by virus shedding, the development of vesicular lesions, and a four-fold rise in VSNJV neutralizing antibody titer was considered evidence of virus transmission. Black flies were chosen for this study because they are a natural host for VSNJV and a confirmed VSNJV vector (Mead et al. 1997, 2004a, b). The development of an experimental insect bite inoculation model that consistently results in clinical disease will be useful for future transmission and pathogenesis studies aimed at understanding the viral, vector, and host factors mediating disease in cattle.
Materials and Methods
One- to 3-d-old Simulium vittatum females (IS-7 cytotype) from a continuous laboratory colony (Bernardo et al. 1986) were infected through intrathoracic inoculation with a 1-μl suspension containing 103.5 plaque-forming units (pfu) of VSNJV. The virus used in this study was isolated from the tongue epithelium of a bovine naturally infected during the 1995 Colorado epidemic (NJ95COB) as previously described (Rodriguez et al. 2000). Baseline infection rates and viral counts were determined in three black flies immediately after injection as previously described (Mead et al. 1997). Remaining infected black flies were maintained by feeding on 15% dextrose at 26°C for a 3- to 4-d extrinsic incubation.
Fourteen, 9- to 12-mo-old Holstein steers weighing 300-400 kg were obtained from an experimental livestock provider (Thomas-Morris, Reisterstown, MD). Animals were housed in a BSL-3 animal facility and monitored for at least 7 d before the initiation of the experiments. Before virus exposure, steers were sedated intramuscularly with xylazine (0.22 mg/kg), which was adequate to maintain lateral recumbence for the duration of the procedures. To infect the animals, 15–30 VSNJV-infected black flies were placed into feeding cages. The cages were constructed with 5-cm-diameter PVC or polycarbonate tubing cut into 1.3-cm sections and enclosed on the two sides with polyester mesh (12 squares/cm) as previously described (Cupp et al. 1981). The cages were manually held on the designated areas of sedated steers randomly assigned to one of four treatment groups: flank (n = 2; steers 73 and 89), coronary band (n = 4; steers 90, 92, 93, and 98), neck (n = 4; steers 735, 736, 737, and 738), or muzzle (n = 3; steers 78, 79, and 87). On completion of inoculation, sedation was reversed with tolazine (2–4 mg/kg, slow IV). The number of black flies that fed on each animal was determined by black fly dissection and observation of blood in the gut. The virus positive control steer (ID 96) was infected with NJ95COB through direct inoculation of virus around the commissure of the muzzle and by coronary band scarification as previously described (Scherer et al. 2007).
Clinical Observation and Sampling.
Cattle were clinically evaluated daily for 7–27 d after infection. Whole blood and sera were collected through jugular puncture on postinfection days (PID) 1–7, and whole blood was immediately processed for virus isolation and real-time reverse transcriptase-polymerase chain reaction (rRT-PCR). Sera were stored at –70°C until processing for virus isolation, serology, and rRT-PCR, as previously described (Scherer et al. 2007). Additional samples taken for virus isolation and rRT-PCR included nasal cavity swabs, 6-mm punch biopsies (Miltex, Bethpage, NY) of tissues where S. vittatum had fed, and oral swabs or oropharyngeal fluid (OPF) samples. Lesions, if observed, were also swabbed. Swabs and OPFs were placed in minimal essential medium (MEM) containing 400 U/ml penicillin, 400 U/ml streptomycin, and 10 μg/ml amphotericin B (Iwahara et al. 1993) and stored at –70°C. Animals were euthanized according to the Institutional Animal Care and Use Committee protocol and necropsied on PIDs 7 (steers 78, 79, and 98), 8 (steers 87, 93, and 96), 14 (steers 735 and 737), 21 (steers 736 and 738) and 27 (steer 73). During necropsy, samples were collected from lesion and nonlesion sites (coronary bands, tongue, nostril, and lips or skin of neck and flank, respectively), draining lymph nodes (mandibular, retropharyngeal, parotid, prescapular, axilar, popliteal, and inguinal), nasopharynx, and tonsil. Necropsies were not performed on steers 89, 90, and 92.
The development of clinical VS was defined by the development of vesicular lesions, virus shedding, and the detection of a four-fold rise in VSNJV neutralizing antibody titer. Subclinical infection was defined as an animal without clinical signs and a VSNJV neutralizing antibody titer of ≥1:32, which represents a four-fold increase in neutralizing antibodies from baseline sera collection. Either clinical disease or seroconversion was considered evidence of successful virus transmission.
Virus isolation was attempted from blood, sera, swabs, punch biopsies and necropsy tissues as previously described (Scherer et al. 2007) with the following changes. Sample supernatant was inoculated into 24-well tissue culture plates containing confluent African green monkey kidney (Vero) cell monolayers and incubated for 1 h at 37°C, and media were replaced with fresh MEM. Cultures were observed for cytopathic effect (CPE) at 24, 48, and 72 h postinoculation. Wells with CPE were confirmed as VSNJV using rRT-PCR. Total RNA was extracted using the Qiagen RNeasy kit (Qiagen, Valencia, CA) following the manufacturer’s protocol and stored at –20°C. Viral RNA was detected by semiquantitative rRT-PCR specific for the nucleocapsid gene of VSNJV as described previously (Scherer et al. 2007). Neutralizing antibodies were detected by microtiter serum neutralization as previously described (Martinez et al. 2004).
Insect Feeding and Infection.
Baseline inoculation titers in black flies determined immediately after injection ranged from 102 to 102.8 pfu/fly (average titer of three inoculated flies). Viral titers of the inoculated flies after incubation were determined in six randomly selected black flies 4 d after intrathoracic injection and averaged 104.3 pfu/fly. For animal infections, fly cages containing 15–30 infected or noninfected flies were manually held on the designated area of each sedated steer for 20–30 min. The number of black flies feeding ranged from 0 to 14, with an average of 6 (Table 1). In cases where flies were not observed feeding, a second set of 15–30 infected flies was allowed to feed at the same site 24 h later and, in most cases, flies fed well the second time. In one case (steer 92), flies did not feed, and viral transmission was not achieved (Table 1). In all cases where infection occurred, there were between 5 and 12 flies with evidence of feeding as determined by the presence of blood in the gut on fly dissection.
Clinical Outcome and Route of Exposure.
A total of 14 steers were used for black fly transmission experiments. Flies were allowed to feed on the neck (n = 4 steers), flank (n = 2), muzzle (n = 3), or coronary bands of the rear and front feet (n = 4). Fever, defined as temperature of ≥40°C, was never observed in any of the animals during the experiment. Typical vesicular lesions were observed in nostrils, lips, or tongue in all three muzzle-infected steers (steers 78, 79, and 87) after feeding by four to eight infected black flies (Table 1). Lesions appeared on PID 3 in steers 78 and 79 and on PID 7 in steer 87 (Fig. 1a-c). Lesions started as blanched areas 24 h before vesicles appeared. Steer 78 had a vesicle on the right lower lip, and steer 79 showed lesions on the right nostril and on the right lower lip. No foot lesions were observed in steers 79 and 87 up to PID 8 when they were euthanized. Interestingly, steer 78 had a vesicular lesion on the right rear coronary band at PID 9 in addition to the mouth lesions described above (Fig. 1d). This lesion yielded VSNJV by virus isolation and was also positive by rRT-PCR (see below). This was the only animal in this study that developed lesions at a site not contiguous to an inoculated site.
Of the four steers exposed to infected black fly bites on the coronary bands, two developed vesicular lesions on the right coronary bands where infected flies fed. No lesions were observed on the left coronary bands where noninfected flies fed (Table 1). Steers 90 and 93 showed remarkable similarity in development of vesicular lesions, beginning with blanched, hyperemic coronary bands on PID 3 (Fig. 1e). By PID 7, the vesicles had spread to interdigital spaces and ruptured (Fig. 1f). Despite the feeding of five infected black flies on the right coronary band of steer 98, no vesicular lesions were observed up to PID 8 when the animal was euthanized. Fly-feeding on steer 92 was unsuccessful, and no lesions developed on this animal.
A total of six animals were exposed to infected fly bites on shaved neck or flank skin where between 2 and 14 infected black flies successfully fed (Table 1). Pinpoint-sized blood droplets were observed where fly feeding occurred, and feeding was confirmed by the presence of blood in the dissected insect gut. However, none of the six steers developed vesicular lesions or any other signs of disease (Table 1). Steer 96, used as a positive control, was inoculated in the right coronary band and right side of the muzzle by scarification and needle inoculation, respectively. The right coronary band became blanched, and a small vesicle was detected on the right lower lip by PID 3. On PID 9, this animal showed a lesion on the right side of the tongue as well as sloughing of the right coronary band. No lesions were observed on the left coronary bands (Table 1).
Virus was not detected in the blood of any animal at any time regardless of exposure route or clinical status. All lesions were confirmed positive for VSNJV by virus isolation and rRT-PCR. In muzzle-infected animals, swabs taken from lesions, oral, and nasal cavities yielded VSNJV at various time points, with the highest titers obtained from lesions (Table 1). Virus was intermittently isolated from oral swabs as early as PID 1 and as late as PID 7 (Table 1). Among the coronary band-inoculated steers that showed clinical signs, virus was only isolated from lesion swabs and never from nasal and oral swabs. Animals exposed to infected black fly bites on the flank or neck did not yield VSNJV from any of the samples collected. Virus isolated from a neck skin punch biopsy collected on PID 1 from steer 735 confirmed the transfer of VSNJV to this site, but subsequent punch biopsies did not result in the recovery of infectious virus or viral nucleic acid. Postmortem tissue collection focused on local tissues and regional lymph nodes draining the inoculation site. One muzzle-inoculated steer (steer 78) yielded viral RNA from mouth tissues and draining lymph nodes as well as infectious virus from a rear-right coronary band lesion and the corresponding draining popliteal lymph node when necropsied at PID 9 (Table 2). Two other muzzle-inoculated steers tested positive for viral RNA in oral draining lymph nodes, tonsil, and tissues in the oral cavity. In two coronary band-inoculated steers, viral RNA was detected in lesion sites and corresponding draining lymph nodes. No samples tested positive in steer 98, which had a subclinical infection (Table 2). All tissues from necropsies of animals exposed to VSNJV by fly bites on the neck or flank tested negative for infectious virus or viral RNA (Table 2).
Antibody responses, measured by virus neutralization against the same VSNJV strain used for inoculation, were considered positive at titers 1:32 or higher according to the Office International des Epizooties (OIE) standards (OIE 2004). Neutralizing antibody titers higher than 1:256 were detected in the sera of all clinically affected steers whether inoculated on the muzzle or coronary bands (Table 1). Additionally, steer 98, which was inoculated on the coronary band but did not show lesions, had a titer >1:256, confirming the occurrence of a subclinical infection. All but one steer exposed to infected fly bites on the flank or neck and steer 92, where infective fly feeding on the coronary band was unsuccessful, were negative by antibody neutralization, with titers ranging from 1:8–1:16. Only one animal exposed in the neck, where 14 infected flies fed, had a minimal positive titer of 1:32 (Table 1).
Evidence for the role of insect vectors in VSNJV transmission during epidemics comes from field data indicating transmission during the months when insects are present, virus isolation from various biting insects during epidemics, and dramatic declines in the number of cases after the first frosts occur (Hanson and Brandly 1957, Walton et al. 1987). Experimental data for insect transmission include the identification of several competent vectors including sand flies, biting midges, and black flies (Mead et al. 1999, Perez de Leon and Tabachnick 2006, Perez de Leon et al. 2006). However, transmission from an experimentally infected insect to livestock resulting in clinical disease has only been previously shown with black flies (Simmulium vittatum) in swine (Mead et al. 2004a, b). In the case of biting midges (Culicoides sonorensis), transmission resulting in subclinical infection, as shown by seroconversion, was observed in cattle exposed to VSNJV-infected midge bites (Perez de Leon and Tabachnick 2006).
This study is the first report of clinical vesicular stomatitis in cattle after experimental VSNJV transmission from biting insects. Clinical outcome was strongly determined by the site where black flies fed. Steers developed lesions after infected S. vittatum fed on the muzzle or coronary bands but not when they fed on the neck or flank. In one of six neck-inoculated animals, virus was isolated from a neck skin punch biopsy at PID 1, providing evidence of transfer of VSNJV to that site. Neither virus nor viral nucleic acid was detected at any other time point, suggesting that VSNJV is incapable of sustained replication and productive infection after insect bite inoculation in the neck or flank skin. Lesions at these sites have never been reported in natural or experimental infections with VSV. Our results enhance findings of previous studies where site of insect feeding or direct skin inoculation determined the clinical outcome of infection in pigs, horses, and cattle, respectively (Mead et al. 2004a, Howerth et al. 2006, Scherer et al. 2007).
In five of the six animals exposed to infected fly bites on the neck or flank, the neutralizing antibody titer was 1:8–1:16. These antibody levels do not constitute seroconversion as defined above but are consistent with antibody levels found in animals living in endemic areas that might be exposed to infected vectors (Rodriguez et al. 1990). Perez de Leon and Tabachnick (2006) showed that steers exposed to VSNJV-infected C. sonorensis bites on the flank or top butt developed a vigorous neutralizing antibody response (>1:512 at PID 8) without clinical signs. This difference may be related to unique characteristics of salivary components of C. sonorensis, biological differences among VSNJV strains used in these studies, or larger numbers of feeding insects (42–239 Culicoides per animal as opposed to 2–14 black flies per animal). Mead et al. (2004a) also showed seroconversion with relatively high titers (1:256 at PID 7) in pigs after black flies infected with a 1997 equine VSNJV fed on the abdomen. This may further illustrate the pathogenic differences between viruses or may be a reflection of differences among host species in response to VSNJV exposure.
In this study, one of the coronary band inoculated animals did not develop clinical signs but had a high titer (1:256) of neutralizing antibodies (ID 98). The fact that this was the only subclinically infected animal developing high titers of neutralizing antibodies suggests that some level of viral replication might have occurred after fly bite infection at or near the coronary band. This experimental result considered alongside with the results of others (Howerth et al. 1997; Stallknecht et al. 1999; Mead et al. 2004a, b; Howerth et al. 2006; Perez de Leon and Tabachnick 2006; Scherer et al. 2007) supports the occurrence of subclinical infections after insect transmission and is consistent with serological surveys conducted after VSNJV outbreaks identifying seropositive animals that did not show signs of disease (Walton et al. 1987, Francy et al. 1988, Rodriguez et al. 1990, Hayek et al. 1998, Mumford et al. 1998).
Virus was not detected in the blood of any bovine in this study despite seroconversion and lesion development. This is consistent with previously reported results and poses interesting questions about vector transmission of VSV (Thurmond et al. 1987; Howerth et al. 1997; Stallknecht et al. 1999; Stallknecht et al. 2001; Mead et al. 2004a, b; Perez de Leon and Tabachnick 2006; Scherer et al. 2007). Unlike VSV, most insect-borne viruses use a viremic amplifying host or reservoir species for viral maintenance in the insect population and sustained epidemic transmission (Kuno and Chang 2005). Interestingly, VSV seems to use alternative means to achieve its transmission among insects. Mead et al. (2000) discovered that VSV can use horizontal transmission between VSNJV-infected and uninfected black flies while co-feeding on noninfected hosts. In this way, the mammalian hosts can serve as amplifying hosts without necessarily becoming viremic. Additionally, Mead et al. (2004b) showed that black flies can become infected with VSNJV when feeding on or near vesicular lesions of clinically affected hosts. The presence of mouth and coronary band lesions yielding high titers of infectious virus provide a potential source for insects to become infected and maintain the role of infected livestock as amplifying hosts during VSV epidemics.
In all but one animal, virus distribution was restricted to inoculated or contiguous sites or the lymph nodes draining these regions. This is consistent with previous studies of direct inoculation of VSNJV in cattle (Scherer et al. 2007). The development of vesicular lesions in the coronary band and recovery of virus from these lesions and draining lymph nodes of one of the muzzle-infected animals was an interesting finding because generalization of lesions has not been previously observed in experimentally inoculated cattle (Scherer et al. 2007). The mechanism of spread to the coronary band is puzzling in the absence of detectable viremia. The fact that the foot lesion was not observed until PID 9 suggests an external source of contact spread from the mouth lesions.
Although this study focused on insect transmission of VSNJV, the high viral titers in lesion swabs highlight the possibility of VSNJV animal-to-animal contact transmission among livestock that are kept in close contact such as in feed-lots or dairy farms. Circumstantial evidence during a 1982 VSNJV outbreak in the western United States, which continued well into the winter months when insect vectors were not present, suggests that direct transmission might also play a role, particularly in confined cattle populations (Walton et al. 1987). In swine, contact transmission of VSNJV occurs readily, and indirect transmission may also occur through contaminated food and water sources caused by viral shedding from the oral cavity (Howerth et al. 1997; Stallknecht et al. 1999, 2001; Martinez et al. 2004).
In conclusion, transmission of VSNJV from experimentally infected black flies to cattle resulting in clinical disease was shown for the first time. Our findings are consistent with earlier studies of black fly transmission to domestic swine and direct inoculation of cattle and horses in that the site of virus inoculation determined the clinical outcome (Mead et al. 2004a, b; Scherer et al. 2007). The mechanisms mediating the differences in clinical outcome observed by inoculation site and the role of cattle as amplifying hosts for VSNJV after clinical or subclinical infections requires further research. The infection methodology presented herein will be useful in experiments designed to address these important questions. Our findings provide valuable information for further understanding of the natural history of VSNJV.
The authors thank J. M. Pacheco, E. Bishop, and E. J. Hartwig for advice and technical assistance and the animal care takers at PIADC for assistance and help collecting the data. This research was supported by the U.S. Department of Agriculture (CRIS-1940-32000-04000D) and by National Research Initiative of the USDA Cooperative State Research, Education and Extension Service Grant 2005-35204-16102. K.R.L. and M.M. were recipients of a Plum Island Animal Disease Center (PIADC) Research Participation Program fellowship, administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement between the U.S. Department of Energy (DOE) and the U.S. Department of Agriculture (USDA).