Adenovirus hemorrhagic disease affects primarily mule deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus), Rocky Mountain elk (Cervus canadensis nelsoni), and moose (Alces alces) in their first year of life. The method by which the causative virus, Deer atadenovirus A, is maintained in the environment and transmitted to neonates is unknown. In this study, we investigated the potential transmission of the virus from dam to offspring in Rocky Mountain mule deer (Odocoileus hemionus hemionus) and elk in western Wyoming, US. We sampled dams before parturition during placement of vaginal implant transmitters and at parturition and sampled neonates during capture in their first days of life. We also tested for the virus in mortalities submitted for pathologic examination and laboratory analysis. We detected viral DNA in samples from all time points tested but did not find a connection between positive dams and offspring mortalities associated with adenovirus hemorrhagic disease. Although we did not find direct evidence of transmission events between dams and offspring, asymptomatic animals shedding of Deer atadenovirus A, are a likely source of infection in neonates.

Adenovirus hemorrhagic disease (AHD) is caused by Deer atadenovirus A (OdAdV-1), a nonenveloped, double-stranded DNA virus (Woods et al. 1997; King et al. 2018). The known hosts are mule deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus), Rocky Mountain elk (Cervus canadensis nelsoni), and moose (Alces alces; Woods et al. 2018). The most common manifestation of AHD is a hemorrhagic disease grossly indistinguishable from blue-tongue and epizootic hemorrhagic disease (Woods et al. 1996). Experimental studies have shown the virus is transmitted via direct contact, but how the virus is maintained in the ecosystem outside of outbreaks is unknown (Woods et al. 2018).

Understanding the mechanisms of OdAdV-1 maintenance in the ecosystem and transmission to susceptible host species is vital to predicting the impacts of AHD on host populations. Viruses are maintained in a host population through a range of possible transmission cycles that include short-term infections that spread continuously to new susceptible members of the population, persistent infections in individuals that shed on an intermittent or continuous basis, persistence in reservoir species with periodic spillover into susceptible animals, or by persisting outside of a host as resistant particles in the environment (Knipe and Mahan 2013). A health surveillance study in British Columbia, Canada, found that mule deer from herds without any recognized AHD cases had OdAdV-1 antibodies by using an unpublished enzyme-linked immunosorbent assay (Mathieu et al. 2018). High antibody prevalence suggested that OdAdV-1 exposure was high, even though AHD cases had not been observed (Mathieu et al. 2018). Determining the mechanism of maintenance of OdAdV-1 in the ecosystem and evaluating transmission from dams to offspring would expand our understanding of the epidemiology of AHD.

Stressed and young animals are most susceptible to adenovirus-related diseases, including AHD, based on observational and experimental data (Woods et al. 1997; Boyce et al. 2000; Knipe and Mahan 2013). During AHD outbreaks in California, US, free-ranging juveniles younger than 6 mo were more likely to develop AHD than were yearlings or adult mule deer (Boyce et al. 2000; Woods et al. 1996, 2018). Experimental OdAdV-1 infections demonstrated that juveniles were much more susceptible than were yearling black-tailed deer (Odocoileus hemionus columbianus), with eight of 10 developing disease compared with two of seven yearlings (Woods et al. 1997, 1999).

In considering the epidemiology of AHD, it is important to understand the life histories of the vulnerable host species. Neonatal mule deer typically spend the first weeks of their life in hiding, which isolates them from potential conspecific shedders besides their dam (Riley and Dood 1984). In contrast, neonatal elk are isolated for only a few days postpartum and, therefore, have potential exposure from their dam and other conspecifics (Altmann 1963; Geist 1982). If dams are persistently infected and intermittently shedding OdAdV-1, the perinatal period could be an opportunity for them to transmit the virus to their offspring (Woods et al. 1999, 2001; Knipe and Mahan 2013). Maternal stress and immune suppression during the periparturient period could increase the likelihood of virus reactivation and shedding (von der Ohe and Servheen 2002).

In this study, we collaborated with ongoing recruitment studies in Rocky Mountain mule deer and elk herds in western Wyoming, US, both of which experienced AHD related die-offs in their neonates in 2015 and 2016 (French 2016). Given that perinatal neonates are in relative isolation with their dam, we hypothesized that dams were shedding the OdAdV-1 virus and transmitting it to their offspring in the perinatal period, resulting in AHD-related neonatal mortalities. Discovering how OdAdV-1 is transmitted to neonates and recognizing how AHD affects that age category, would further the understanding of AHD and its population-level effects.

Study area

Animals sampled in this study were part of three long-term recruitment studies in western Wyoming, US. The Wyoming Range Mule Deer herd study, located in western Wyoming (42°25′N, 110°42′W), began in 2015 and has the goal of maintaining GPS collars on 70 female mule deer for multiple years. The South Rock Springs Mule Deer and Elk studies, in the Greater Little Mountain ecosystem located in south-central Wyoming (42°1′N, 108°18′W), began in November 2015 and have the goals of maintaining GPS collars on 50 female mule deer and 35 female elk. The Wyoming State Veterinary Laboratory in Laramie, Wyoming received all mortality submissions since the onset of those studies. All studies captured collared animals in later winter or early spring and fitted animals determined to be pregnant by ultrasound, with a vaginal implant transmitter (VIT; Advance Telemetry Systems, Isanti, Minnesota, USA; Vectronics Aerospace, Berlin, Germany). At the time of parturition, signified by the expulsion of the VIT, researchers located the VIT and the neonate to collect morphometric measurements and fit the neonates with GPS collars. They tracked the survival of those animals and located all mortalities to ascertain the proximal cause of death and to retrieve intact carcasses or tissues to submit to the Wyoming State Veterinary Laboratory for necropsy and diagnostic testing. The University of Wyoming Institutional Animal Care and Use Committee approved both studies (Wyoming Range Mule Deer, no. 20170215KM00260; South Rock Springs Mule Deer and Elk, no. 20170322KM00267).

Sampling

Sampling for our study lasted from March 2018, when VIT were fit to pregnant females, through April 2019 to include recovery and sampling of mortalities. We collected samples during VIT placement, shortly after parturition during neonate collaring and VIT recovery, and from necropsied dead animals. At the time of VIT placement, we obtained a vaginal sample by swabbing the pipe used to insert the VIT with a sterile Dacron Fiber–tipped, plastic applicator swab (ThermoFisher Scientific, Waltham, Massachusetts, USA) or with a Pur-Wraps sterile cotton-tipped applicator (Puritan Medical Products, Guilford, Maine, USA). Swab storage until DNA extraction was at –20 C in 1 mL of Eagle's minimal essential media with 1× glutamine (Corning, Corning, New York, USA) with 4% Seradigm premium-grade fetal bovine serum (VWR, Radnor, Pennsylvania, USA). The pipe used to place VIT was cleaned and sanitized in 2% clorohexidine solution for a minimum of 2 min between animals (Durvet Inc., Blue Springs, Missouri, USA). Placement of VIT occurred 17 March 2018 to 19 March 2018 for the Wyoming Range study, which was at about 110 d of gestation. For the South Rock Springs study, VIT placement occurred between 20 April 2018 to 23 April 2018, which was at about 150 d of gestation for mule deer and about 220 d for elk.

After parturition, we recovered VIT in 5 to 48 h after expulsion from the animal. Upon retrieval, we placed VIT in individual Whirl-Pak bags and froze them within 12 h. The VIT remained frozen until they were returned to the laboratory, where we used sterile cotton-tipped applicators moistened with sterile phosphate-buffered saline (PBS; National Diagnostic, Atlanta, Georgia, USA) to sample the expelled VIT and then stored the swabs in sterile PBS and froze them at –20 C until DNA extraction. Some VIT models had silicon covering around the wings, which we cut using a single-use razor blade to allow for swabbing underneath.

We collected perinatal samples from neonates during capture, 5 to 48 h after birth, using sterile-tipped applicators to swab the buccal and anal cavities. Swabs were stored frozen at –20 C in 1 mL of sterile PBS. Parturition occurred between 24 May 2018 and 27 June 2018 for the Wyoming Range mule deer herd, whereas parturition occurred between 21 May 2018 and 7 July 2018 for mule deer in the South Rock Springs herd, and for elk, between 19 May 2018 and 27 June 2018.

Adult and offspring mortalities brought back to the Wyoming State Veterinary Laboratory for necropsy had varying states of autolysis and consumption. We tested all necropsied carcasses from the onset of the study to the end of April 2019 for OdAdV-1, using the best-available tissue samples but did not include estimated proximal cause of death in our analyses.

DNA extraction and PCR

Following the manufacturer's instructions, we isolated DNA using a MagMax™ total nucleic acid isolation kit (ThermoFisher Scientific). Swabs in their respective transport tubes were vortexed and clarified by centrifugation, and the supernatant was removed for DNA extraction. We tested all swabs individually. For tissue samples, we prepared a pool of lung, liver, spleen, kidney, and brain, listed in order of preference, as available by homogenizing more than 25 mg tissue in 0.5 mL Bovarnick's medium (0.218 mM sucrose, 3.8 mM KH2PO4, 7.2 mM K2HPO4, 4.9 mM glutamic acid, 1% phenol red; Sigma-Aldrich, St. Louis, Missouri, USA) in a bead beater tub at 20 Hz for 2 min on the Tissue Lyzer (Qiagen, Germantown, Maryland, USA). The supernatant was clarified by 3,000 × G centrifugation before DNA extraction. We included a negative-extraction control consisting of the corresponding swab and medium for each sample type.

To detect OdAdV-1 DNA, we used the Wyoming State Veterinary Laboratory real-time PCR assay. Briefly, 25 µL reactions were prepared using 5 µL of the extracted DNA, 1 µL of primer-probe mix containing 15 µM forward (5′-CCAAA-TAAACCACATCCCGTA-3′) and reverse (5′-TTGTGTGGCGTGCTTAACTA-3′) primers and a 5-µM probe (5′-FAM-TCCGCATTTGCTCCT GGAAA-TAMRA-3′; Eurofins Scientific, Luxembourg City, Luxembourg), 0.5 µL Xeno Internal Control Positive Control LIZ Assay (Applied Biosystems, Foster City, California, USA), 12.5 µL SsoAdvanced Universal Probes Supermix (Bio-Rad Laboratories, Hercules, California, USA), and 6 µL nuclease-free water. We performed PCR amplification and detection with a CFX96 real-time PCR detection system (Bio-Rad). Thermal cycling conditions were a 10-min denaturation step at 95 C followed by 40 cycles of 95 C for 15 s and 60 C for 30 s. We used negative extraction and negative template controls in all reactions. The test was previously validated using multiple standard curves (10–1 to 10–8) and could detect 13.51±21.91 DNA copies. We found no cross reactions with Bovine adenovirus 5, an atadenovirus, Bovine adenovirus 3, a Mastadenovirus, and bovine respiratory viruses (Infectious bovine rhinotracheitis, Bovine herpesvirus type 1, Parainfluenza type 3, Bovine respiratory syncytial virus, and bovine vial diarrhea viruses). To classify samples with a cycle threshold (Ct) value >37 as positive, we verified the results with repeated testing, those that were not positive in multiple tests were considered as suspect.

To assess the potential for extraction and PCR inhibition, we tested 10% of samples during both DNA extraction and PCR. To test for inhibition of DNA extraction, we combined equal parts positive cell culture material and sample material before extraction. We tested for PCR inhibition by spiking sample DNA with an equal volume of the positive extract control with a known Ct value. We considered inhibited any sample with a three or more point increase in Ct value from the positive extract control. The quality of the sample was also assessed by determining the presence of cellular DNA using a cytochrome c oxidase subunit 1 mitochondrial gene target, end-point PCR (Folmer et al. 1994).

OdAdV-1 detection

At each sampling window and in every study-group tested we found samples positive for OdAdV-1 (Table 1). Vaginal swab samples taken during VIT placement in elk (4%; 1/24, gestational age 210 d), South Rock Springs mule deer (2%; 1/48, gestational age 180 d), and Wyoming Range mule deer (5%; 3/64, gestational age 102 to 120 d) were positive. All positive expelled VIT samples were from Wyoming Range mule deer (15%; 8/52). Positive perinatal swabs came from both South Rock Springs mule deer (4%; 2/55) and Wyoming Range mule deer (6%; 5/80), including from an opportunistically caught fawn whose dam was not collared. All positive perinatal samples were from oral swabs. Five of the perinatal mule deer samples were from neonates born as a singleton; the remaining samples, one positive and one suspect, came from two sets of twins. In both cases, OdAdV-1 was not detected in the other twin. We detected OdAdV-1 in three of eight necropsied adult females and two of 24 necropsied juveniles from South Rock Springs and in two of 26 necropsied juveniles from the Wyoming Range (Table 1). It was not possible to obtain samples from every animal at every time point for a variety of reasons, including failure to recover the VIT or to locate the neonate (see Supplementary Materials). All four of the OdAdV-1–positive fawns were twins; one of the twins died 2 mo after its sibling and did not test OdAdV-1 positive, the remaining pairs had a twin that died in the month after the positive twin's demise (none of these siblings were necropsied). None of the necropsied OdAdV-1–positive fawns (n=4) at the time of perinatal sampling were positive at necropsy. The remaining positive perinatal fawns were either not necropsied (n=2) or were alive at the end of the study (n=1).

Table 1

Detection of Deer atadenovirus A (OdAdV-1) by real-time PCR. Dams and their offspring from two Rocky Mountain mule deer (MD; Odocoileus hemionus hemionus) and one Rocky Mountain elk (Cervus canadensis nelsoni) herd units were tested for the presence of OdAdV-1 DNA. We collected vaginal swab samples during vaginal implant transmitter (VIT) placement before parturition (approximately days into gestation: South Rock Springs elk=220, South Rock Springs [SRS] MD=150, Wyoming Range [WR] MD=110). At the time of parturition, indicated by VIT expulsion, we retrieved the VITs and collected buccal and anal swabs from neonates 5–48 h after birth. We tested anal and buccal swabs separately; all anal swabs were negative. A subset of all the animals that died during the study (March 2018 to April 2019) were necropsied. One dam whose fawn was OdAdV-1 positive at necropsy had a positive vaginal sample; no other dam-offspring pairs showed any pattern of transmission.

Detection of Deer atadenovirus A (OdAdV-1) by real-time PCR. Dams and their offspring from two Rocky Mountain mule deer (MD; Odocoileus hemionus hemionus) and one Rocky Mountain elk (Cervus canadensis nelsoni) herd units were tested for the presence of OdAdV-1 DNA. We collected vaginal swab samples during vaginal implant transmitter (VIT) placement before parturition (approximately days into gestation: South Rock Springs elk=220, South Rock Springs [SRS] MD=150, Wyoming Range [WR] MD=110). At the time of parturition, indicated by VIT expulsion, we retrieved the VITs and collected buccal and anal swabs from neonates 5–48 h after birth. We tested anal and buccal swabs separately; all anal swabs were negative. A subset of all the animals that died during the study (March 2018 to April 2019) were necropsied. One dam whose fawn was OdAdV-1 positive at necropsy had a positive vaginal sample; no other dam-offspring pairs showed any pattern of transmission.
Detection of Deer atadenovirus A (OdAdV-1) by real-time PCR. Dams and their offspring from two Rocky Mountain mule deer (MD; Odocoileus hemionus hemionus) and one Rocky Mountain elk (Cervus canadensis nelsoni) herd units were tested for the presence of OdAdV-1 DNA. We collected vaginal swab samples during vaginal implant transmitter (VIT) placement before parturition (approximately days into gestation: South Rock Springs elk=220, South Rock Springs [SRS] MD=150, Wyoming Range [WR] MD=110). At the time of parturition, indicated by VIT expulsion, we retrieved the VITs and collected buccal and anal swabs from neonates 5–48 h after birth. We tested anal and buccal swabs separately; all anal swabs were negative. A subset of all the animals that died during the study (March 2018 to April 2019) were necropsied. One dam whose fawn was OdAdV-1 positive at necropsy had a positive vaginal sample; no other dam-offspring pairs showed any pattern of transmission.

No discernible patterns of transmission were detectable because of the low infection rate in offspring. Of all dam-offspring samples for which we had at least one sample (n=119) from both individuals, only one pair tested positive. Interestingly, the same dam had twins, and we did not detect OdAdV-1 in her other fawn. The rest of the samples had either a positive dam and a negative offspring (n=9), a positive offspring and a negative dam (n=9), or both individuals were negative (n=90). Eight positive samples could not be included in this analysis because of the lack of a corresponding dam or offspring sample.

Sample quality and inhibition testing

Inhibition of DNA extraction and PCR inhibition were a concern because of the variable condition of the VIT samples, which ranged from being covered in birthing material licked clean or even covered in dirt. Extraction of DNA and PCR of expelled VIT samples showed no signs of inhibition. Twenty of 114 (18%) VIT samples had no detectable cellular DNA; however, three of those samples (2%) were OdAdV-1 positive and an additional sample was suspect (Table 2). Cellular DNA was present on 87% (40/46) of VIT with silicon wing covers and 79% (54/ 68) VIT with no wing covers.

Table 2

Cellular DNA detection from vaginal implant transmitters (VITs) with and without silicon covers that were obtained from Rocky mule deer (Odocoileus hemionus hemionus) and Rocky Mountain elk (Cervus canadensis nelsoni) in Wyoming, USA, during the 2018 birthing season. We compared sample quality and Deer atadenovirus A (OdAdV-1) detection rates of VIT with and without a silicon covering over the wings. The covering potentially protected excretions from the environment. We determined the presence of cellular DNA using a cytochrome c oxidase subunit 1 mitochondrial gene target, endpoint PCR to assess sample quality (Folmer et al. 1994). We detected OdAdV-1 using real-time PCR; samples without repeated positive OdAdV-1 tests were considered suspect. We were able to detect OdAdV-1 DNA on low-quality VIT samples on which we were unable to detect cellular DNA.

Cellular DNA detection from vaginal implant transmitters (VITs) with and without silicon covers that were obtained from Rocky mule deer (Odocoileus hemionus hemionus) and Rocky Mountain elk (Cervus canadensis nelsoni) in Wyoming, USA, during the 2018 birthing season. We compared sample quality and Deer atadenovirus A (OdAdV-1) detection rates of VIT with and without a silicon covering over the wings. The covering potentially protected excretions from the environment. We determined the presence of cellular DNA using a cytochrome c oxidase subunit 1 mitochondrial gene target, endpoint PCR to assess sample quality (Folmer et al. 1994). We detected OdAdV-1 using real-time PCR; samples without repeated positive OdAdV-1 tests were considered suspect. We were able to detect OdAdV-1 DNA on low-quality VIT samples on which we were unable to detect cellular DNA.
Cellular DNA detection from vaginal implant transmitters (VITs) with and without silicon covers that were obtained from Rocky mule deer (Odocoileus hemionus hemionus) and Rocky Mountain elk (Cervus canadensis nelsoni) in Wyoming, USA, during the 2018 birthing season. We compared sample quality and Deer atadenovirus A (OdAdV-1) detection rates of VIT with and without a silicon covering over the wings. The covering potentially protected excretions from the environment. We determined the presence of cellular DNA using a cytochrome c oxidase subunit 1 mitochondrial gene target, endpoint PCR to assess sample quality (Folmer et al. 1994). We detected OdAdV-1 using real-time PCR; samples without repeated positive OdAdV-1 tests were considered suspect. We were able to detect OdAdV-1 DNA on low-quality VIT samples on which we were unable to detect cellular DNA.

We worked alongside recruitment studies to collect prospective samples from dams and offspring under free-ranging conditions. Generally, acquisition of samples from diseased individuals requires a chain of low-probability events to occur: detection of a carcass or disease by an interested individual (often a member of the public), submission of a usable sample to a laboratory capable of diagnosing it, an available diagnostic test with species-specific baseline information, and a life-history knowledge of the animal in hand (Stallknecht 2007). These conditions are especially true for rarely observed newborns. Collaborating with recruitment studies circumvents this set of challenges.

We found OdAdV-1 DNA-positive samples in multiple apparently healthy mule deer and elk dams and their offspring. Samples were OdAdV-1 positive at all sampling points: before parturition at VIT placement, after parturition on VIT, oral swabs of neonates collected between 5 and 48 h after birth, and in necropsied animals. The highest percentage of positive samples came from necropsied animals and from VIT after parturition from one of the three herds. Animals that died from disease were more likely to have an intact carcass for necropsy examination, which likely biased the results from necropsied animals. None of the stillborn (n=3) or fetal (n=1) carcasses necropsied tested positive for the virus. The postpartum-positive VITs were only found in a single herd (Wyoming Range Mule Deer), suggesting that this herd had a higher shedding rate at that time point. Higher detection in the Wyoming Range study area could be a result of a 10 F (5.6 C) cooler average June temperature than the South Rock Springs study area, which would likely prolong the persistence of viral DNA in that region.

Testing dams and their offspring at multiple time points did not reveal clear patterns of transmission. The positive perinatal samples indicated that some dams expose their neonates to OdAdV-1, either in utero or during parturition. To determine definitively whether transmission was vaginal or in utero, as positive perinatal samples suggested, an experimental, controlled challenge study would be necessary. The preferred sample type for OdAdV-1 DNA detection in neonates was oral swabs because all anal swabs were negative. From positive oral swabs alone, we are unable to conclude whether the source of the virus was due to infection resulting in viral shedding or to recent exposure to positive maternal fluids. Transmission that resulted in some of the offspring mortalities may have occurred at a later time because they occurred months after the animals were born (see Supplementary Materials). Conversely, neither the one positive OdAdV-1 nor the one suspect perinatal fawn that died within a month of parturition tested positive at necropsy. There was no apparent pattern between dams that tested positive and their neonates dying close to parturition. Our evidence supports the hypothesis that dams shedding OdAdV-1 expose their neonates to the virus and that many neonates exposed to the virus either do not die or do not test positive for OdAdV-1 at necropsy.

Detection of OdAdV-1 shedding in this study is notable because we demonstrated viral DNA presence in apparently healthy animals in a year with relatively low AHD-related mortalities. Finding shedding of OdAdV-1 that does not result in AHD gives evidence for the hypothesis that virus maintenance in the ecosystem is by a conspecific that is intermittently shedding the virus or by a continual transmission cycle. Potentially, adenovirus shedding occurs in animals that are persistently or latently infected and shedding is stress induced (Knipe and Mahan 2013). Intermittent shedding of human and animal adenoviruses of the genus Mastadenovirus has been documented, even in hosts with neutralizing antibodies, but has not been demonstrated for members of the genus Atadenovirus (Knipe and Mahan 2013). Viral shedding at parturition may be stress induced because dams were likely stressed in the days leading up to giving birth (von der Ohe and Servheen 2002). However, the acute stress of capture was not likely to have resulted in viral shedding at the time we collected vaginal swab samples, given that the virus would not have had adequate time to replicate, although it was possible that animals were under chronic stress because of other factors. Overall, our findings on shedding suggested an endemic cycle of intermittent, low-level shedding with the potential to cause disease in susceptible individuals.

We used samples of convenience that were collected as part of other field studies, leading to limitations. We collected vaginal swabs, a nontraditional type of sample for adenovirus, during VIT placement, which may have limited out abilities to detect the virus. Furthermore, our study design limited conclusions as to when animals shed during gestation because our testing was done at a single time point. Detection of OdAdV-1 could be compromised because of poor sample quality, particularly from the VIT. The natural behavior of cleaning up the birth site by the dam often included consuming products of the conceptus off the VIT, exposing them to the environment and to the dam's saliva. These behaviors may have removed or introduced viral particles, and PCR inhibitors in the soil are a potential cause of false-negative results, although we did not detect inhibition. The number of samples with no cellular DNA (20/114; Table 2) demonstrated the poor quality of the samples; however, adenoviruses have a strong outer capsid that makes them relatively stable in the environment (Knipe and Mahan 2013). Swab storage medium may also have affected detection. Perinatal swabs were stored in PBS, instead of transport media, and the protein-free buffer is less protective of viral integrity (Johnson 1990). Carcass quality, tissue availability, and the number of carcasses recovered for necropsy limited detection of the virus in infected animals. Virus detection at necropsy also may be limited if the virus is not evenly distributed throughout the tissues. Although PCR is highly sensitive, low viral concentration in samples may have resulted in false-negatives.

Demonstrating that dams expose their offspring to OdAdV-1 does not rule out another reservoir or environmental contamination of the virus, which requires further research. To more fully understand the population-level effects of maternal transmission of OdAdV-1 on mule deer and elk, sampling at different time points or during a high AHD-related mortality year would likely be helpful. Testing for a difference in OdAdV-1 transmission between elk and mule deer was not possible because few elk tested positive. The apparently lower detection rates in elk warrants further research to understand variations in susceptibility between different host species. In cattle, maternal antibodies to another Atadenovirus, Bovine atadenovirus 7, passed to the calf in colostrum, neutralized OdAdV-1 in experimentally infected calves (Woods et al. 2008). Therefore, investigating the transmission of maternal antibodies to OdAdV-1 alongside the transmission of the virus will be an important aid in further understanding the epidemiology of AHD in wild ungulate neonates.

Collaborating with offspring recruitment studies proved to be a simple and convenient way to obtain samples from wild animals at multiple time points. Intermittent shedding of OdAdV-1 in our three study groups with a known history of AHD indicated that dams likely exposed their neonates, the most afflicted age group, to the virus, and that many exposed and OdAdV-1–positive neonates did not die of AHD. Our detection of many OdAdV-1–positive individuals in a year with relatively low AHD-related mortalities further supported a continual transmission cycle. Therefore, we suspect OdAdV-1 affects neonatal survival in herds known carry OdAdV-1 on a continual basis.

Supplementary material for this article is online at http://dx.doi.org/10.7589/JWD-D-20-00034.

We thank Matthew Hayes and the technicians that collected the samples in summer 2018 at the Wyoming Range Mule Deer and South Rock Springs Mule Deer and Elk study areas. Jennifer Makenna and Marce Vasquez at the Wyoming State Veterinary Laboratory were essential in designing and troubleshooting the PCR assays. Jessica Jennings-Gaines in the Wyoming Department of Game and Fish helped to develop the method for testing vaginal implant transmitters. Madison Vance and Isabel Noyes helped to organize hundreds of samples, to plate samples, and to set up PCR reactions.

Altmann
M.
1963
.
Naturalistic studies of maternal care in moose and elk.
In:
Maternal behavior in mammals
,
Rheingold
HL,
editor.
John Wiley and Sons
,
New York, New York
, pp.
233
253
.
Boyce
WM,
Woods
LW,
Keel
MK,
MacLachlan
NJ,
Porter
CO,
Lehmkuhl
HD.
2000
.
An Epizootic of adenovirus-induced hemorrhagic disease in captive black-tailed deer (Odocoileus hemionus).
J Zoo Wildl Med
31
:
370
373
.
Folmer
O,
Black
M,
Hoeh
W,
Lutz
R,
Vrijenhoek
R.
1994
.
DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates.
Mol Mar Biol Biotechnol
3
:
294
299
.
French
B.
2016
.
Disease identified as killer of Wyoming mule deer fawns.
Billings Gazette, Lifestyles section on February 4, 2016.
Geist
V.
1982
.
Adaptive behavioral strategies.
In:
Elk of North America: Ecology and management
,
Thomas
JW,
Toweill
DE,
editors.
Stackpole Books
,
Harrisburg, Pennsylvania
, pp.
237
242
.
Johnson
FB.
1990
.
Transport of viral specimens.
Clin Microbiol Rev
3
:
120
131
.
King
AMQ,
Lefkowitz
EJ,
Mushegian
AR,
Adams
MJ,
Dutilh
BE,
Gorbalenya
AE,
Harrach
B,
Harrison
RL,
Junglen
S,
Knowles
NJ,
et al.
2018
.
Changes to taxonomy and the International Code of Virus Classification and Nomenclature ratified by the International Committee on Taxonomy of Viruses (2018).
Arch Virol
163
:
2601
2631
.
Knipe
BN,
Mahan
D.
2013
.
Adenoviruses.
In:
Fields virology.
5th Ed.
Lippincott Williams and Wilkins
,
Philadelphia, Pennsylvania
, pp.
2355
2366
.
Mathieu
A,
Flint
M,
Stent
PM,
Schwantje
HM,
Wittum
TE.
2018
.
Comparative health assessment of urban and non-urban free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
PeerJ
6
:
e4968
.
Riley
SJ,
Dood
AR.
1984
.
Summer movements, home range, habitat use, and behavior of mule deer fawns.
J Wildl Manage
48
:
1302
1310
.
Stallknecht
DE.
2007
.
Impediments to wildlife disease surveillance, research, and diagnostics.
Curr Top Microbiol Immunol
315
:
445
461
.
von der Ohe
CG,
Servheen
C.
2002
.
Measuring stress in mammals using fecal glucocorticoids: Opportunities and challenges.
Wildl Soc Bull
30
:
1215
1225
.
Woods
LW,
Hanley
RS,
Chiu
PH,
Lehmkuhl
HD,
Nordhausen
RW,
Stillian
MH,
Swift
PK.
1999
.
Lesions and transmission of experimental adenovirus hemorrhagic disease in black-tailed deer fawns.
Vet Pathol
36
:
100
110
.
Woods
LW,
Hanley
RS,
Chiu
PHW,
Burd
M,
Nordhausen
RW,
Stillian
MH,
Swift
PK.
1997
.
Experimental adenovirus hemorrhagic disease in yearling black-tailed deer.
J Wildl Dis
33
:
801
811
.
Woods
LW,
Lehmkuhl
HD,
Hobbs
LA,
Parker
JC,
Manzer
M.
2008
.
Evaluation of the pathogenic potential of cervid adenovirus in calves.
J Vet Diagn Invest
20
:
33
37
.
Woods
LW,
Lehmkuhl
HD,
Swift
PK,
Chiu
PH,
Hanley
RS,
Nordhausen
RW,
Stillian
MH,
Drew
ML.
2001
.
Experimental adenovirus hemorrhagic disease in white-tailed deer fawns.
J Wildl Dis
37
:
153
158
.
Woods
LW,
Schumaker
BA,
Pesavento
PA,
Crossley
BM,
Swift
PK.
2018
.
Adenoviral hemorrhagic disease in California mule deer, 1990–2014.
J Vet Diagn Invest
30
:
530
537
.
Woods
LW,
Swift
PK,
Barr
BC,
Horzinek
MC,
Nordhausen
RW,
Stillian
MH,
Patton
JF,
Oliver
MN,
Jones
KR,
MacLachlan
NJ.
1996
.
Systemic adenovirus infection associated with high mortality in mule deer (Odocoileus hemionus) in California.
Vet Pathol
33
:
125
132
.

Supplementary data