to Heterologous and Homologous Rabies Viruses
April D. Davis,aJodie A. Jarvis,aCraig E. Pouliott,aShannon, M. D. Morgan,bRobert J. Rudda
Rabies Laboratory, Wadsworth Center, New York State Department of Health, Slingerlands, New York, USAa; Department of Biological Sciences, University of Albany, State University of New York, Albany, New York, USAb
Rabies virus (RABV) maintenance in bats is not well understood. Big brown bats (Eptesicus fuscus), little brown bats (Myotis lucifugus), and Mexican free-tailed bats (Tadarida brasiliensis) are the most common bats species in the United States. These colonial bat species also have the most frequent contact with humans and domestic animals. However, the silver-haired bat ( La-sionycteris noctivagans) RABV is associated with the majority of human rabies virus infections in the United States and Canada. This is of interest because silver-haired bats are more solitary bats with infrequent human interaction. Our goal was to deter-mine the likelihood of a colonial bat species becoming infected with and transmitting a heterologous RABV. To ascertain the potential of heterologous RABV infection in colonial bat species, little brown bats were inoculated with a homologous RABV or one of two heterologous RABVs. Additionally, to determine if the route of exposure influenced the disease process, bats were inoculated either intramuscularly (i.m.) or subcutaneously (s.c.) with a homologous or heterologous RABV. Our results demon-strate that intramuscular inoculation results in a more rapid progression of disease onset, whereas the incubation time in bats inoculated s.c. is significantly longer. Additionally, cross protection was not consistently achieved in bats previously inoculated with a heterologous RABV following a challenge with a homologous RABV 6 months later. Finally, bats that developed rabies following s.c. inoculation were significantly more likely to shed virus in their saliva and demonstrated increased viral dissemina-tion. In summary, bats inoculated via the s.c. route are more likely to shed virus, thus increasing the likelihood of transmission.
L
yssavirus infections have been reported in numerous species ofterrestrial and flying mammals (1,2). Several regions of
enzo-otic rabies virus (RABV) activity occur in raccoon, skunk, and fox populations within the continental United States. These enzootic foci are generally limited to homologous infections; e.g., raccoons
are infected with a raccoon rabies virus variant (2). Despite less
geographical isolation, chiropteran RABVs are also typically
lim-ited to their host species (3). However, RABV spillover into
het-erologous hosts does occur and has been implicated in the origin
of raccoon- and skunk-adapted RABVs (4).
Two of the most common species of bats in the continental
Unites States are the big brown bat (Eptesicus fuscus) and the little
brown bat (Myotis lucifugus) (5). These highly colonial species are
adapted to living in both urban and rural areas. On the basis of data from public health rabies laboratories, big brown bats and little brown bats are the bat species most commonly submitted for rabies testing. The large number of submissions stems from
hu-man or domestic animal exposure (6). Despite frequent
interac-tion, only three human rabies cases acquired in the United States have been associated with the little brown bat or big brown bat RABV since 1990 (Fox News). The RABV most often associated with human
rabies cases acquired in the United States is the silver-haired bat (
La-sionycteris noctivagans) RABV (LnRV) (7;http://www.cdc.gov/rabies /location/usa/surveillance/humanrabies.html). Unlike big brown or little brown bats, silver-haired bats are tree-dwelling bats that form small colonies, and thus, human interaction with this species
is infrequent (7). Previous studies have postulated that the
silver-haired bat RABV possesses an increased pathogenicity, including the ability to replicate at lower temperatures and infect
nonneu-ronal cell lines (8,9). These unique characteristics increase the
transmissibility of LnRV, posing a greater exposure risk to more gregarious bat or domestic animal species, thereby increasing the possibility of human-LnRV interaction.
In nature, the amount of RABV and the depth at which it is inoculated in a bite wound vary considerably between bats and terrestrial RABV vector species, such as raccoons, skunks, and foxes. The puncture wounds following a bat bite may easily go unnoticed, yet the bite of a raccoon would likely result in trauma requiring medical attention and thus rabies treatment. Addition-ally, the depth of inoculation would be greater following the bite of a raccoon than the bite of a bat. Bat bites are less likely to penetrate into muscular tissue, generally resulting in a superficial laceration or subcutaneous (s.c.) exposure. Studies in which bats were inoc-ulated with European bat lyssavirus-1 (EBLV-1) demonstrated the ability of s.c. and intramuscular (i.m.) inoculation of RABV to
result in RABV infection and possible transmission (10). As a
result, bat RABV may have an increased ability to infect and rep-licate in epithelial or subcutaneous tissue.
Is human RABV infection the result of direct contact with a silver-haired bat or another more common bat species or domes-tic animal that may be infected with LnRV? Although spillover of LnRV into heterologous bat species and domestic animals has been reported, adaptation has not. However, a study comparing heterologous RABV infections among bat species demonstrated that little brown bats were most likely to be infected with a
heter-ologous RABV, including LnRV (3). Thus, heterologous RABV
transmission and dissemination among bat species may be more common than previously acknowledged.
Received31 December 2012Accepted31 May 2013
Published ahead of print5 June 2013
Address correspondence to April D. Davis, [email protected]. Copyright © 2013, American Society for Microbiology. All Rights Reserved. doi:10.1128/JVI.03554-12
on November 7, 2019 by guest
http://jvi.asm.org/
To determine the potential of heterologous RABV infection in
a colonial bat species, little brown bats (Myotis lucifugus) were
inoculated with a homologous or heterologous RABV. Addition-ally, to determine the influence of the route of exposure on infec-tion and transmissibility, bats were inoculated either i.m. or s.c.
MATERIALS AND METHODS
Ethics statement.The experiments were designed and animal care was done in compliance with the guidelines of the USDA Animal Care and Welfare Act (AWA) and the Association for Assessment and Accredita-tion of Laboratory Animal Care (AAALAC). The use of bats in this exper-iment was approved by and conducted in accordance with the Wadsworth Center IACUC.
Animals.Little brown bats were removed from a hibernaculum lo-cated in upstate New York. Bats of both genders were quarantined within a biosafety level 3 facility in groups of 5 to 8 for 6 months. All bats entering the captive colony were weighed, and an oral swab specimen was col-lected. Bats were identified with uniquely colored wing bands. Bats were provided fresh water andad libmealworms daily. Bats were given 1 drop of flax seed oil twice a week to prevent dry skin. The room temperature was maintained at 24 to 27°C, and the humidity was approximately 60 to 80%. Twice a week bats were given a brief physical exam and weighed, and an oral swab specimen was obtained. A bat that had lost 0.5 g between examinations was reweighed daily. If weight loss continued, the bat was placed in a smaller isolation cage, monitored more closely, hand-fed mealworms and beef baby food, and, if necessary, administered 0.5 ml lactated Ringer’s saline s.c. every 24 h. If a bat was demonstrating clinical signs of rabies and did not improve within 24 to 48 h, it was euthanized, necropsied, and tested for rabies via a direct fluorescent-antibody test (dFAT). Sera were collected to assay for anti-rabies virus neutralizing antibodies (VNAs) as previously described (11).
One week prior to inoculation, bats were divided into six mixed gen-der groups of five individuals. Each group was inoculated with 10450% tissue culture infective doses (TCID50s) of either little brown bat (Myotis
lucifugus) RABV (MlV1), big brown bat (Eptesicus fuscus) RABV (EfV2), or silver-haired bat (Lasionycteris noctivagans) RABV (isolate LnV1) in a volume of 10l in the right deltoid muscle or the subcutaneous tissues superficial to the right deltoid muscle using a Hamilton syringe and a 30-gauge needle (Table 1). Inoculations were performed under a magni-fying lens/lamp to ensure proper administration of virus. To confirm that each bat was inoculated correctly, multiple checks were established: two individuals verified bat identification, virus variant, and route of inocula-tion and visualized virus administrainocula-tion. One individual handled the bat, while the other applied the inoculum. Serum was not collected prior to inoculation. Testing of serum prior to RABV inoculation is ideal when working with wild animals, including larger and heartier species of bats, such as big brown bats. However, little brown bats are much smaller, and the amount of serum that can be safely obtained from an⬃10-g bat may not be adequate for testing. Furthermore, the fragility of little brown bats compelled the researchers to be cautious. Thus, the decision was made that maintaining a healthy colony of little brown bats outweighed the benefits of serum collection.
Serum was not collected from all bats that survived the study. Because of the population decline due to white nose syndrome (WNS), the ability to collect little brown bats for use in laboratory studies is limited. Some of the surviving bats were transferred to a WNS study.
Regardless of the primary inoculum, all surviving wild-caught bats were administered a secondary challenge with 104TCID
50s MlV1 in 10l in the right deltoid muscle 6 months after the primary inoculation.
Virus.Virus was isolated from the salivary glands ofMyotis lucifugus
(MlV1),Eptesicus fuscus(EfV2), andLasionycteris noctivagans(LnV1) bats (12,13). To obtain adequate amounts of virus for inoculation, isolates were passaged in either neuroblastoma (NA) (LnV1, EfV2) or BHK (MlV1) cell lines. To confirm the genotype of RABV, the N gene of the virus isolate was sequenced as previously described (14).
Oral swabs.Oral swabs were placed in 500l Eagle’s minimum essen-tial medium (EMEM) supplemented with 100 IU penicillin G, 50g streptomycin, and 2.5 mg amphotericin B per ml. Oral swabs were tested for RABV via virus isolation and PCR. For PCR, RNA was extracted from the oral swab using 200l of sample added to TRIzol LS reagent and processed per the manufacturer’s recommendations (Invitrogen, Carls-bad, CA). The cDNA was generated from extracted RNA as described in the Quanta qScript cDNA synthesis kit (Quanta BioSciences, Gaithers-burg, MD) using random primers.
Two hundred microliters of the oral swab suspension was vortexed in a class II biological safety cabinet. Suspensions were centrifuged for 30 min at 4°C and 10,000 rpm. Supernatant (100l) was placed into a 1-ml microtube, and 200l of NA cells at a concentration of 5⫻105per ml was added to the suspension. The tube was held at 4°C for 15 to 20 min, and the contents were mixed by inversion every 5 min. One milliliter of oral swab growth medium (OSGM; composed of Eagle’s minimum essential medium supplemented with 10% fetal bovine serum, 2.0 mM glutamate, 100 IU penicillin G, 50g streptomycin, and 2.5 mg amphotericin B per ml) was added to duplicate wells in a 96-well plate. Incubation was in a moist chamber at 34°C with 5% CO2for 4 days.
After incubation, one well from each sample was trypsinized, and the contents were seeded into five wells of a new 96-well plate and allowed to grow for 4 days. The OSGM was aspirated from the remaining well; the cell sheet was flooded with 0.01 M phosphate-buffered saline (PBS; pH 7.6) for 1 min to remove the OSGM, the wash was aspirated, and the cell sheet was fixed overnight with methanol-formalin fixative (1:1 100% methanol and 10% formalin solution). Each well received one quick wash followed by two 30-min PBS washes to remove the fixative. Cells were stained with light diagnostics rabies dFAT reagent (catalog no. 5100; Chemicon International, Temecula, CA) for 30 min, followed by two 2-min PBS washes. Before examining the wells on a Zeiss Axiovert 200 fluorescence microscope at magnifications of⫻200 and⫻400, each well was flooded with 0.20 ml of 0.85% saline buffered with 0.05 M Trizma, pH 9.0. The blind-passaged plates were fixed, stained, and examined by fluo-rescence microscopy as described above.
Frozen tissue histology.A full necropsy was performed on any animal that died or was euthanized during the course of the study. A brain tissue sample was collected for routine rabies testing. Multiple tissue samples were collected to assay for the presence of viral antigen and viral RNA. Each tissue sample was divided between two 0.5-ml Sarstedt microtubes (catalog no. 72.730.006) containing 200l of sterile 0.01 M PBS and stored at⫺80°C.
Tissues were cut in 7-m-thick sections on a microtome cryostat (HM 505 N; Microm International GmbH). Two to four sections were flash-frozen and then later thawed and affixed to positively charged slides (cat-alog no. 12-550-19; Fisher brand Colorfrost Plus). The slides containing sections were air dried for 20 min in a fume hood before acetone fixation or put in short-term storage in a⫺20°C freezer.
The protocol for postmortem diagnosis of rabies in animals by dFAT
(www.cdc.gov/rabies/pdf/RabiesDFASPv2.pdf) was followed for fixation,
staining, and results interpretation. Each tissue specimen was tested with two different rabies virus-specific fluorescein isothiocyanate conjugates (Light Diagnostics DFA1 [catalog no. 5100] and Light Diagnostics DFA3 [catalog no. 5500]; EMD Millipore, Billerica, MA). All conjugates were titrated to determine the optimal working dilution in the Rabies Labora-tory at the Wadsworth Center and stored frozen at⫺40°C at the working dilution. Conjugates were dispensed using a syringe fitted with a
0.45-m-pore-size syringe filter that contained low-protein-binding mem-brane material.
Statistical analysis was accomplished using a one-way analysis of vari-ance (ANOVA).
Molecular assay.Using a Precellys 24 homogenizer (Bertin Technol-ogy, Rockville, MD), tissues were processed in 800l TRIzol LS reagent and 200l growth media. RNA was extracted per the manufacturer’s recommendations (Invitrogen, Carlsbad, CA). For conventional PCRs, cDNA was generated from extracted RNA and random primers as
on November 7, 2019 by guest
http://jvi.asm.org/
scribed in the ABI high-capacity cDNA synthesis kit (Applied Biosystems, Carlsbad, CA). A 400-bp region of the RABV N gene was amplified using a Qiagen HotStarTaq DNA polymerase PCR per the manufacturer’s pro-tocol (Qiagen, Germantown, MD) and primers 21G (5=-ATGTAACACC CCTACAATG-3=) and 390 (5=-CTTGTCAACTCCATACC-3=) (15). Am-plicons were separated by agarose gel electrophoresis, and the appropriate band from each reaction mixture was excised with a sterile razor blade and purified using spin prep columns (Qiagen, Germantown, MD). The nu-cleotide sequence of purified PCR products was determined and com-pared to other RABV nucleotide sequences via BLAST analysis of the sequences in GenBank.
Quantification of viral RNA was accomplished by TaqMan-based quantitative reverse transcriptase PCR (qRT-PCR) as previously
[image:3.585.43.543.77.498.2]de-scribed (16) using the RABVD1 probe/primer set. A green fluorescent protein (GFP) mRNA assay was used as an internal extraction and positive control. The real-time assay was performed using a qScrip Fast one-step qRT-PCR kit with Low ROX (Quanta, Gaithersburg, MD) per the man-ufacturer’s recommendations. To identify and avoid inhibition, samples were run undiluted and at 1:10 and 1:100 dilutions. Cycling conditions were as follows: 50°C for 5 min, 95°C for 30 s, and 45 cycles of 95°C for 15 s and 50°C for 1 min. The assay was run on an ABI 7500 real-time PCR system. Our limit of detection (LOD) was 170 gene copies with a threshold cycle of 37. An endpoint standard curve was generated from 10-fold serial dilutions of cDNA of known copy number ranging from 108to 102 tran-scripts. Additionally, a primer/probe set designed to target the spiked internal extraction control GFP was included.
TABLE 1Serological and survival results for little brown bats inoculated with a homologous or heterologous rabies virus variant
Bat group and no.
Primary inoculation Secondary inoculation
VNA titer (IU) at completion of study Virus Route
Incubation time (dpia)
VNA titer (IU)
qPCR results for oral swabsc(dpi)
Status of
batb Virus Route
Incubation time (dpi)
VNA titer (IU)
qPCR results for oral swabs
Status of bat
Group 1 MlV1 i.m. MlV1 i.m.
1 N A N A
2 13 ⱕLODd
N De
3 17 0.5 N De
4 N A N A ⱕLOD
5 N A N A
Group 2 MlV1 s.c. MlV1 i.m.
6 N A N A 1.0
7 N A N A
8 N A N A ⱕLOD
9 N A N A ⱕLOD
10 N A N A ⱕLOD
Group 3 EfV2 i.m. MlV1 i.m.
11 N A 23f ⱕLOD N De
12 N A 12 ⱕLOD N De
13 N A 16 ⱕLOD N De
14 N A N A
15 N A 12 ⱕLOD N De
Group 4 EfV2 s.c. MlV1 i.m.
16 N A N A
17 N A N A
18 N A N A
19 N A N A
20 43 32 P (30, 31) Dg
Group 5 LnV1 i.m. MlV1 i.m.
21 IDh ⱕLOD N D
22 17 ⱕLOD N D
23 N A N A
24 N A N A
25 N A N A
Group 6 LnV1 s.c. MlV1 i.m.
26 1.0 N A N A
27 N A N A
28 N A N A 16.0
29 N A 14 1.0 N D
30 51 ⱕLOD P (33, 43, 50, 51) Dg adpi, days postinoculation.
b
The survival status of the bat is denoted A for alive and D for deceased.
cN, negative; P, positive. dⱕ
LOD, results are below the limit of detection for our test.
eThe bats developed rabies following an i.m. exposure. f
Incubation times in bold are numbers of days after the 6-month challenge.
gThe bats developed rabies following an s.c. exposure. h
ID, indeterminate cause of death.
on November 7, 2019 by guest
http://jvi.asm.org/
Analysis of results was performed using ABI 7500 software. Statistical analysis was accomplished using a one-way ANOVA.
RESULTS
RABV was not detected in the saliva of bats that developed ra-bies following an i.m. inoculation.To determine if a homologous RABV (MlV1) was more virulent than a heterologous RABV
(EfV2, LnV1), five little brown bats were inoculated i.m. with 104
TCID50s of either MlV1, EfV2, or LnV1. Oral swab specimens
were collected twice a week, and serum samples were collected from all euthanized bats to test for VNAs.
At 13 days postinoculation (dpi), bat 2 (inoculated i.m. with MlV1) developed rabies. Within 17 dpi, bat 3 (inoculated i.m. with MlV1) and bat 22 (inoculated i.m. with LnV1) developed rabies. Bat 21 (inoculated i.m. with LnV1) developed clinical signs compatible with RABV infection at 17 dpi. Despite multiple at-tempts to obtain an unequivocal diagnosis for bat 21, both the dFAT and quantitative PCR (qPCR) provided inconsistent re-sults. Thus, we did not include bat 21 in our analysis. None of the
bats inoculated i.m. with EfV2 developed rabies (Table 1). No
significant difference (Pⱖ0.5) was found when comparing the
incubation times between RABV variants following the primary i.m. inoculation.
Regardless of the infecting variant, RABV was not found in the saliva of any bat inoculated i.m. Although bats 2 and 22 were negative for VNAs, bat 3 had a low titer (0.5 IU) for anti-rabies virus VNAs.
RABV was present in the saliva of bats that developed rabies following an s.c. inoculation.To better understand how the route of inoculation may affect the disease process, we designed an ex-periment to assess an s.c. route of inoculation using three different bat RABVs. Fifteen little brown bats were separated into three
groups of five bats each and inoculated s.c. with 104TCID
50s of
either MlV1, EfV2, or LnV1. Two of the 15 bats inoculated s.c. developed rabies. Bat 20, which received EfV2, and bat 30, which received LnV1, developed clinical signs compatible with rabies
virus infection at 43 and 51 dpi, respectively (Table 1). A
signifi-cant difference (2⫽6.6, degrees of freedom⫽1,Pⱕ0.010) was
present when comparing the number of bats that developed rabies
following an i.m. inoculation (n⫽7) and the number of bats that
developed rabies following an s.c. inoculation (n⫽2).
Virus was found in the saliva of bats 20 and 30 on multiple occasions. Virus was isolated from the oral swab of bat 20 12 and 13 days prior to the development of clinical signs. Virus was first isolated from an oral swab of bat 30 18 days prior to the develop-ment of clinical signs.
Blood was collected during euthanasia, at the terminal stage of disease. VNA was not detected in bat 30 (inoculated s.c. with LnV1), but VNA (32 IU) was present in the serum of bat 20.
Prior RABV inoculation was not always protective against i.m. challenge.To evaluate the impact that a previous exposure to RABV may have on subsequent RABV exposure(s), all animals
that survived the primary inoculation were challenged with 104
TCID50s of homologous RABV (MlV1). Bats were inoculated i.m.
6 months after the primary inoculation. Twelve of the 17 (71%) bats that survived the primary inoculation remained healthy fol-lowing the 6-month challenge. Four of the bats previously inocu-lated i.m. with EfV2 developed clinical rabies (12, 12, 16, and 23 days after the challenge inoculation), and one bat initially
inocu-lated s.c. with LnV1 developed rabies 14 days after the challenge inoculation.
Despite oral swab specimen collection biweekly and immedi-ately prior to euthanasia, RABV was not detected in any of the samples. Production of VNAs at the terminal bleed was negligible. The incubation time in bats that developed rabies following the challenge inoculation was similar to that in bats that developed
rabies after the primary inoculation (Table 1).
Although bats inoculated i.m. with EfV2 survived the primary exposure, 80% developed rabies following the challenge with MlV1. This is twice the number that developed rabies following a primary i.m. exposure to MlV1. Although no significant
differ-ence was noted (P⫽0.19), this trend suggests that exposure to
certain RABVs may increase the susceptibility to infection when challenged with a homologous RABV. However, additional re-search is required to support this hypothesis.
RABV dissemination may depend on the route of exposure, incubation time, and infecting RABV.Although i.m. inoculation is believed to be the most common route of RABV transmission between terrestrial mammals, it may be different among
chirop-terans (17,18) On the basis of our data, i.m. inoculation is more
likely to result in clinical infection yet is less likely to result in dissemination of the virus to the salivary glands. Centrifugal spread to the salivary glands is integral for the maintenance of RABV in nature, and thus, a lack of dissemination through saliva would result in a dead-end host.
To compare the virulence of homologous and heterologous RABVs, three RABV variants were inoculated either i.m. or s.c. To ascertain the tropism of the RABV as a result of the variant and route of inoculation, bats were necropsied and specimens of sev-eral tissues were collected for cryosectioning and tested for the
presence of RABV via dFAT and qPCR (Tables 2and3).
The greatest numbers of RABV antigen-positive tissues were collected from bats inoculated s.c. (bats 20 and 30) or inoculated i.m. with LnV1 (bat 22). Extensive centrifugal spread was evident in bats with the longest incubation periods or following inocula-tion with the putatively more virulent RABV, LnV1. There was a
significant difference (P⫽0.008) in dissemination between
lon-ger (⬎23 dpi) and shorter (⬍17 dpi) incubation times.
Addition-ally, a significant difference (Pⱕ0.05) was noted when comparing
homologous and heterologous virus dissemination.
With the exception of the salivary glands (SGs) and tissues of neurological origin, the other tissues did not contain RABV anti-gen; only the nerves coursing through the tissues possessed RABV antigen. The anterior spinal cord (ASC), left sciatic nerve (LSN), and right sciatic nerve (RSN) were positive for rabies virus antigen in all cryostat sections examined. The ASC was the only tissue that was positive for both viral RNA and RABV antigen (N protein) in all samples. Regardless of the route or variant, the sample that contained the largest amount of viral RNA was the ASC, with the
cycle threshold values ranging from 15 to 16, equal to 108copies.
DISCUSSION
Human rabies cases as a result of vaccine failure have not been
reported following the use of cell culture rabies vaccines (15). The
current vaccine is protective against all rabies virus variants when
administered i.m. or s.c. (19). However, the protection afforded a
bat following a previous exposure to RABV is unknown. Previous studies have demonstrated that the route of inoculation can be
important in predicating survival (20, 21). Ndaluka (2011)
on November 7, 2019 by guest
http://jvi.asm.org/
ported that intranasal inoculation of RABV in mice was highly
pathogenic (21). Conversely, exposing bats to aerosolized RABV
did not result in rabies virus infection. Despite the presence of VNAs, previous aerosol exposure was not protective following an
i.m. challenge (20). Clinical RABV infection is more likely
follow-ing inoculation with higher viral titers (17,21). Indeed, the route
to clinical RABV infection is multifactorial.
Due to their unique ecology, bats may be exposed to rabies several times during their lives via multiple exposure routes, such as the aerosol route (during echolocation and sound production), as well as the subcutaneous (during grooming) and intramuscular (during fighting) routes. Although aerosol exposure is not protec-tive against an i.m. RABV challenge, previous i.m. and s.c. expo-sure may provide some level of protection against future RABV
exposures (17,21).
This study was designed to better understand the maintenance of bat RABV in nature and its impact on public health. Big brown
bats (Eptesicus fuscus), little brown bats (Myotis lucifugus), and
Mexican free-tailed bats (Tadarida brasiliensis) are the most
com-mon bat species in the United States. Although these species have the most frequent interaction with humans and domestic animals, theL. noctivagansand closely relatedPipistrellus subflavusRABV is associated with the majority of human rabies virus infections in the
United States and Canada (5,6). However, unlike Mexican
free-tailed bats, big brown bats, and little brown bats, silver-haired bats are not in frequent contact with humans. Prior to the emergence of white nose syndrome (WNS), 26% and 69% of the bats
submit-ted to the New York State Department of Health (NYSDOH) Ra-bies Laboratory were little brown bats and big brown bats,
respec-tively (22). These data are similar to those from most public health
laboratories in the United States (6). Submissions are generally the
result of contact between a bat and a human or domestic animal
(6, 22). Following the appearance of WNS,Myotisspecies
ac-counted for less than 10% of the submissions and greater than 87% were big brown bats. The proportion of silver-haired bats submitted to the NYSDOH Rabies Laboratory has remained sta-ble, comprising approximately 1% of the bats submitted since 1993. Regardless of the presence or absence of WNS, the rate of positivity for rabies virus among bat species has remained consis-tent between 1993 and 2011. Rabies virus infection was confirmed
in 0.8% and 3.5% of theM. lucifugusandE. fuscusbats,
respec-tively. Rabies virus infection was confirmed in 4% of theL.
noc-tivagansandP. subflavusbats submitted to the NYSDOH Rabies Laboratory over the past 18 years. However, due to the low num-ber of submissions to our laboratory, the percentage is dispropor-tionately high compared to the percentages for big brown bats and little brown bats, which are submitted in much greater numbers
(6,22). Because silver-haired bats are less likely to have contact
with other bat species, domestic pets, or humans, the number of human rabies cases associated with LnV1 remains a conundrum.
To determine the susceptibility ofM. lucifugusto RABV
[image:5.585.41.544.78.362.2]infec-tions, we inoculated bats with a homologous RABV or one of two heterologous RABVs. Although the lack of significance between the heterologous and homologous infection groups may be the
TABLE 2Distribution of RABV in tissues evaluated by cryostat sections using the dFAT
Tissuea
Detection of the following antigenb
:
No. of tissue specimens positive/total no. tested (%) Myo 2
(MlV1, i.m.)
Myo 3 (MlV1, i.m.)
Myo 11 (MlV1, i.m.)
Myo 12 (MlV1, i.m.)
Myo 13 (MlV1, i.m.)
Myo 15 (MlV1, i.m.)
Myo 20 (EfV2, s.c.)
Myo 22 (LnV1, i.m.)
Myo 29 (MlV1, i.m.)
Myo 30 (LnV, s.c.)
SG ⫺ ⫹ ⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ 6/10 (60)
Nose ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ 6/10 (60)
Tongue ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ 4/10 (40)
Lung ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫹ ⫺ ⫺ 4/10 (40)
Stomach ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ 4/10 (40)
Heart NAc NA ⫺ NA ⫺ NA ⫹ NA NA ⫹ 2/4 (50)
Kidney ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ ⫹ ⫺ ⫺ 3/10 (40)
Diaphragm ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ 4/10 (40)
Bladder ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ 3/10 (30)
B. plexus, right ⫹d ⫺ ⫹ ⫹ ⫹ NA NA ⫹ ⫹ ⫹ 7/8 (88)
B. plexus, left ⫹d ⫹ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹
6/8 (75)
RSN ⫹d ⫹ ⫹ ⫹ NA ⫹ ⫹ ⫹ ⫹ ⫹ 9/9 (100)
LSN ⫹d ⫹ ⫹ ⫹ ⫹
NA ⫹ ⫹ ⫹ NA 8/8 (100)
GI ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ 5/10 (50)
ASC ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ NA ⫹ ⫹ ⫹ 9/9 (100)
PSC ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ NA 8/9 (89)
Buccal tissue ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ 4/10 (40)
Trachea NA ⫹ ⫹ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ 6/9 (67)
Nuchal skin ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ 2/10 (20)
Liver ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ 1/10 (10)
Inoculation site, muscle ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫹ ⫺ 6/10 (60) Inoculation site, skin ⫺ ⫺ ⫺ NA ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ 3/9 (33)
Spleen NA ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫺ 3/9 (33)
No. of RV-positive tissues/total no. of tissues positive (%)
5/18 (17) 8/22 (36) 9/22 (41) 12/21 (57) 6/22 (27) 7/20 (35) 17/21 (81) 20/22 (91) 12/22 (55) 16/21 (76)
aSG, salivary gland; B. plexus, brachial plexus; RSN, right sciatic nerve; LSN, left sciatic nerve; GI, gastrointestinal tissue; ASC, anterior spinal cord; PSC, posterior spinal cord. b
The information in parentheses represents the infecting variant, route of inoculation.
cNA, some tissues were not available for cryostat sections. d
Samples from the right and left were combined.
on November 7, 2019 by guest
http://jvi.asm.org/
TABLE 3 Distribution of RABV in tissues evaluated by qPCR Tissue a Day of detection of the following antigen b : No. of tissue specimens positive/total no. tested (%) Myo 2 (MlV1, i.m., 13) Myo 3 (MlV1, i.m., 17) Myo 11 (MlV1, i.m., 23) Myo 12 (MlV1, i.m., 12) Myo 13 (MlV1, i.m., 16) Myo 15 (MlV1, i.m., 12) Myo 20 (EfV2, s.c., 43) Myo 22 (LnV1, i.m., 17) Myo 29 (MlV1, i.m., 14) Myo 30 (LnV1, s.c., 51) SG 34 26 23 ⱕ LOD c ⱕ LOD 31 27 23 29 23 8/10 (80) Nose 31 32 30 29 36 ⱕ LOD 27 24 30 21 9/10 (90) Tongue 33 ⱕ LOD 27 26 30 35 ⱕ LOD 24 29 25 8/10 (80) Lung ⱕ LOD 2 30 ⱕ LOD ⱕ LOD ⱕ LOD 31 22 23 37 23 6/10 (60) Stomach 38 28 38 27 ⱕ LOD 33 22 23 25 19 9/10 (90) Heart NA d NA 27 NA ⱕ LOD NA 23 NA NA 23 3/4 (75) Kidney ⱕ LOD ⱕ LOD 37 30 ⱕ LOD ⱕ LOD ⱕ LOD 18 27 23 5/10 (50) Diaphragm 34 32 26 ⱕ LOD 36 27 27 24 34 23 9/10 (90) Bladder ⱕ LOD 34 ⱕ LOD 28 ⱕ LOD ⱕ LOD 25 32 ⱕ LOD ⱕ LOD 4/10 (40) B. plexus, right 25 e 23 22 24 24 24 23 21 22 21 10/10 (100) B. plexus, left 25 e 23 27 27 28 29 23 21 25 21 10/10 (100) RSN 29 e 31 33 25 27 32 23 23 29 27 10/10 (100) LSN 29 e 29 28 27 29 ⱕ LOD 26 21 30 23 9/10 (90) GI 33 ⱕ LOD 31 ⱕ LOD ⱕ LOD ⱕ LOD 23 24 ⱕ LOD ⱕ LOD 4/10 (40) ASC 15 16 15 15 18 15 15 16 16 11 10/10 (100) PSC 27 24 31 24 26 27 23 14 29 ⱕ LOD 9/10 (90) Buccal tissue 32 33 35 28 ⱕ LOD ⱕ LOD 27 25 30 28 8/10 (80) Trachea ⱕ LOD 24 28 24 28 27 25 23 24 25 9/10 (90) Nuchal skin 34 32 29 27 ⱕ LOD 26 ⱕ LOD ⱕ LOD 26 23 7/10 (70) Liver 30 36 29 ⱕ LOD ⱕ LOD ⱕ LOD 26 24 ⱕ LOD 23 6/10 (60) Inoculation site, muscle 23 24 25 24 28 26 24 24 24 22 10/10 (100) Inoculation site, skin 31 26 35 37 ⱕ LOD 30 24 26 33 23 9/10 (90) Spleen ⱕ LOD ⱕ LOD 34 ⱕ LOD ⱕ LOD ⱕ LOD 30 22 ⱕ LOD 27 4/10 (40) No. of RABV-positive tissues/total no. of positive tissues (%) 16/21 (76) 17/22 (77) 21/23 (91) 16/22 (73) 11/23 (48) 14/22 (66) 22/23 (96) 21/22 (95) 18/22 (82) 20/23 (87) a SG, salivary gland; B. plexus, brachial plexus; RSN, right sciatic nerve; LSN, left sciatic nerve; GI, gastrointestinal tissue; ASC, anterior spina l cord; PSC, posterior spinal cord. b The information in parentheses represents the infecting variant, route of inoculation, time of incubation (in days). c ⱕ LOD, results are below the limit of detection for our test. d NA, some tissues were not available for cryostat sections. e Samples from the right and left were combined.
on November 7, 2019 by guest
http://jvi.asm.org/
[image:6.585.145.466.76.717.2]result of the small sample size, our study demonstrated thatM. lucifugusbats are susceptible to infections with heterologous and homologous RABVs. Our results indicate that primary exposure using a homologous RABV has virulence similar to that of expo-sure using a heterologous RABV. These results were unexpected;
in a previous study withE. fuscus, infection with the homologous
RABV was more likely to result in clinical illness than infection
with a heterologous RABV (23). Streicker et al. (2010) reported
that although cross-species transmission events were uncommon,
they occurred more frequently among closely related species (3).
Consequently, less heterogeneity allowed an easier adaptation to a new but closely related host. On the basis of an analysis of data for
the partial sequence of cytochromecoxidase subunit I,M.
lucifu-gusis similarly related toL. noctivagans(83%) andE. fuscus(81%)
(3). Analysis of a 1,414-nucleotide region of the rabies virus N
gene demonstrated the close relatedness of theM. lucifugusRABV
to the L. noctivagans RABV (89.5%) and the E. fuscus RABV
(90%). To further pursue the relationship between species and viral relatedness, future studies should include a more distant bat
species, such asT. brasiliensis.
While it is doubtful that a bite from an insectivorous bat could penetrate into the muscle of a human, bites between microchirop-tera may penetrate muscle during aggressive encounters. Subcu-taneous RABV exposure between bats may occur more frequently, e.g., during commensal grooming or routine territorial behaviors that involve biting. Inoculation of RABV into the musculature or subcutaneous tissues via aggression or commensal grooming may be the most likely routes of RABV transmission. How the route of inoculation affects the outcome of disease is unknown. Further-more, the ability of bat rabies viruses to infect and/or replicate in epithelium or subcutaneous tissues is not well understood.
Mortality rates following experimental i.m. or s.c. RABV inoc-ulation have been variable. Baer and Bales reported longer
incu-bation times inT. brasiliensisbats inoculated s.c. (24). Johnson et
al. (2008) reported no illness following i.m. inoculation of the
homologous host (Myotis daubentonii) with EBLV-2 but illness
among 14% of bats following s.c. inoculation (18). An experiment
completed by Freuling et al. (2009) with EBLV-1 demonstrated that 42% of bats developed rabies following s.c. exposure and 14% developed rabies following i.m. inoculation in the homologous
host (Eptesicus serotinus) (10). Conversely, Franka et al. (2008)
reported an approximately 38% fatality rate following i.m.
inocu-lation with EBLV-1 in a heterologous host (E. fuscus), yet all bats
inoculated s.c. survived (25). These studies reveal that i.m.
inocu-lation is not required to produce clinical rabies virus infection and suggest that s.c. exposure may be the preferred route of inocula-tion for the maintenance of rabies in bat populainocula-tions.
In this study,M. lucifugusbats did not develop rabies following
s.c. inoculation with a homologous RABV, yet they were suscep-tible to s.c. exposure with a heterologous RABV. The variation between this and the previous experiments may be multifactorial, including the lyssavirus to which the animal was exposed, the species of bat, previous exposure in nature, and the health of the bat.
Rabies virus was intermittently isolated from the oral swab specimens obtained from the two bats that developed rabies fol-lowing s.c. inoculation. Virus was detected up to 18 days before clinical signs compatible with RABV infection were first noted. This is considerably longer than the length of time in previous studies, in which RABV was detected in saliva within 10 days of the
appearance of clinical signs (14,18,26).
In the present study, primary inoculation with the homolo-gous RABV may have provided the immune priming necessary for surviving a subsequent exposure. This was not apparent following primary inoculation with the two heterologous RABVs. In fact, primary i.m. inoculation with EfV2 (group 3) appeared to have increased the susceptibility to the secondary homologous RABV inoculation. The number of bats in group 3 that developed rabies was twice the number of bats that developed rabies following pri-mary inoculation with the homologous RABV (group 1). To the best of our knowledge, increased susceptibility to RABV following a primary heterologous RABV inoculation has not been demon-strated in previous bat RABV studies.
The increased susceptibility may be the result of immune sup-pression and evasion, lack of available antibody, or antibody en-hancement. RABV is capable of immune subterfuge via decreasing or delaying the immune response or evading the immune system
completely (27,28). Previous studies have demonstrated that the
ability to avert the immune system is not equivalent among RABV
variants (4,6,8,9). A primary i.m. exposure to EfV2 may be more
immune modulating than exposure to another bat RABV. Thus, an impaired ability to mount an innate or humoral response fol-lowing the secondary challenge could provide the virus with an increased opportunity for infection. Alternatively, VNAs pro-duced after the primary inoculation may have been bound to RABV antigen at the time of the second inoculation and thus unavailable to prevent infection.
Antibody enhancement has previously been reported
follow-ing RABV exposure (29–31). Animals inadequately immunized or
lacking passive immunization developed rabies more rapidly than their properly immunized or immunocompetent counterparts. Antibody enhancement may have played a role in the develop-ment of clinical infection in bats first inoculated i.m. with EfV2. However, since serum was not collected following the primary inoculation, the presence of VNAs following the primary inocu-lation was undetermined. To determine if antibody enhancement occurs during rabies virus infection in bats, considerable addi-tional research is required.
Baer and Bales (1967) suggested that lengthened incubation times allowed the virus to replicate to higher titers in the brain and
peripheral tissues and increased centrifugal dissemination (24).
Our results investigating RABV antigen dissemination support greater dissemination following longer incubation times and het-erologous RABV infections. However, using the more sensitive qPCR technique to assay the viral load in tissues, no significant difference was found between the groups.
The presence or absence of VNAs at the terminal bleed in this study was not significantly impacted by the route of exposure. The high titer present in bat 20 may have been due to a previous expo-sure in the wild. One drawback of this study design was the inabil-ity to collect preinoculation and routine samples for serology. However, it is clear from this study that the lack of VNAs cannot corroborate rabies virus naiveté.
Viral antigen and RNA were present in multiple tissues, includ-ing the salivary glands. However, virus could be isolated only from the saliva of bats that developed clinical disease following s.c. in-oculation and, thus, longer incubation times. Following s.c. inoc-ulation, less virus may be available to infect the nerves coursing through the subcutaneous or muscle tissues. Alternatively, the decreased number of motor nerves in the s.c. tissue may affect the ability of RABV to infect the central nervous system (CNS). This
on November 7, 2019 by guest
http://jvi.asm.org/
could diminish centripetal spread, resulting in less virus entering the CNS and brain and providing the opportunity for increased viral replication and dissemination prior to clinical illness. The additional time could allow the virus time to replicate to levels that result in shedding in saliva. Although viral RNA and antigen were present in the salivary glands of most bats that developed rabies, there was no infectious virus found in the saliva of i.m. inoculated bats. The decreased incubation time may have allowed the virus to have time to disseminate to the salivary glands but not the time to replicate and be shed in the saliva. Alternatively, infectious virus may have been immunologically cleared prior to virus isolation. However, given the short incubation times, this is unlikely.
The current study demonstrates the variability in pathogenesis and clinical disease following experimental homologous and
het-erologous RABV inoculation inMyotis lucifugus bats. The
ex-tended incubation time following subcutaneous inoculation al-lowed increased RABV dissemination, thereby resulting in infectious saliva and the ability to transmit disease. The saliva of bats inoculated via the i.m. route was negative for infectious RABV, thus resulting in a dead-end host. These results suggest that although the subcutaneous route of inoculation may decrease the incidence of disease, it is more likely to result in disease
trans-mission. Additionally, the susceptibility ofM. lucifugusto
infec-tion with LnV1 suggests an addiinfec-tional way in which humans may be exposed to the silver-haired bat virus. Increasing our under-standing of RABV maintenance in bats will expand our knowledge of public health risks and transmission of zoonotic diseases.
ACKNOWLEDGMENTS
We are grateful to Carl Herzog, Jeremy Coleman, Carmen Fagnani, and Kelly Broussard for their assistance in acquiring and maintaining the cap-tive bat colonies. We thank Barbara French and BCI for their help and guidance in understanding bat health and husbandry. We are grateful to Michelle Dupuis for her assistance with our quantitative PCR develop-ment. We appreciate our colleagues at the Wadsworth Center for helpful discussions and the Applied Genomics Technologies Core.
Financial support was received from the National Institutes of Health (grant 5K08AI085031).
REFERENCES
1.Childs JE.2002. Epidemiology, p 114 –149.InJackson AC, Wunner WH (ed), Rabies. Academic Press, San Diego, CA.
2.Hughes GJ, Orciari LA, Rupprecht CE.2005. Evolutionary timescale of rabies virus adaptation to North American bats inferred from the substi-tution rate of the nucleoprotein gene. J. Gen. Virol.86(Pt 5):1467–1474. 3.Streicker DG, Turmelle AS, Vonhof MJ, Kuzmin IV, McCracken GF, Rupprecht CE.2010. Host phylogeny constrains cross-species emergence and establishment of rabies virus in bats. Science329:676 – 679. 4.Kuzmin IV, Shi M, Orciari LA, Yager PA, Velasco-Villa A, Kuzmina
NA, Streicker DG, Bergman DL, Rupprecht CE.2012. Molecular infer-ences suggest multiple host shifts of rabies viruses from bats to mesocar-nivores in Arizona during 2001-2009. PLoS Pathog.8:e1002786. doi:10
.1371/journal.ppat.1002786.
5.Fenton B.2001. And now some bats, p 119 –148.InBats, revised edition. Checkmark Books, New York, NY.
6.Patyk K, Turmelle A, Blanton JD, Rupprecht CE. 2012. Trends in national surveillance data for bat rabies in the United States: 2001-2009. Vector Borne Zoonotic Dis.12:666 – 673.
7.Messenger SL, Smith JS, Rupprecht CE.2002. Emerging epidemiology of bat-associated cryptic cases of rabies in humans in the United States. Clin. Infect. Dis.35:738 –747.
8.Morimoto K, Patel M, Corisdeo S, Hooper DC, Fu ZF, Rupprecht CE, Koprowski H, Dietzschold B.1996. Characterization of a unique variant of bat rabies virus responsible for newly emerging human cases in North America. Proc. Natl. Acad. Sci. U. S. A.93:5653–5658.
9.Roy A, Phares TW, Koprowski H, Hooper DC.2007. Failure to open the blood-brain barrier and deliver immune effectors to central nervous sys-tem tissues leads to the lethal outcome of silver-haired bat rabies virus infection. J. Virol.81:1110 –1118.
10. Freuling C, Vos A, Johnson N, Kaipf I, Denzinger A, Neubert L, Mansfield K, Hicks D, Nuñez A, Tordo N, Rupprecht CE, Fooks AR, Müller T.2009. Experimental infection of serotine bats (Eptesicus seroti-nus) with European bat lyssavirus type 1a. J. Gen. Virol.90(Pt 10):2493– 2502.
11. Davis A, Gordy P, Rudd R, Jarvis JA, Bowen RA. 2012. Naturally acquired rabies virus infections in wild-caught bats. Vector Borne Zoo-notic Dis.12:55– 60.
12. Rudd RJ, Trimarchi CV.1987. Comparison of sensitivity of BHK-21 and murine neuroblastoma cells in the isolation of a street strain rabies virus. J. Clin. Microbiol.25:1456 –1458.
13. Rudd RJ, Trimarchi CV.1989. Development and evaluation of an in-vitro virus isolation procedure as a replacement for the mouse inoculation test in rabies diagnosis. J. Clin. Microbiol.27:2522–2528.
14. Davis AD, Dupuis M, Rudd RJ.2012. Extended incubation period of rabies virus in a captive big brown bat (Eptesicus fuscus). J. Wildl. Dis. 48:508 –511.
15. Centers for Disease Control and Prevention.1999. Human rabies pre-vention—United States. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recommend. Rep.48 (RR-1):1–21.
16. Nadin-Davis SA, Sheen M, Wandeler AI.2009. Development of real-time reverse transcriptase polymerase chain reaction methods for human rabies diagnosis. J. Med. Virol.81:1484 –1497.
17. Turmelle AS, Jackson FR, Green D, McCracken GF, Rupprecht CE. 2010. Host immunity to repeated rabies virus infection in big brown bats. J. Gen. Virol.91(Pt 9):2360 –2366.
18. Johnson N, Vos A, Neubert L, Freuling C, Mansfield KL, Kaipf I, Denzinger A, Hicks D, Núñez A, Franka R, Rupprecht CE, Müller T, Fooks AR.2008. Experimental study of European bat lyssavirus type-2 infection in Daubenton’s bats (Myotis daubentonii). J. Gen. Virol.89(Pt 11):2662–2672.
19. Dietzschold B, Hooper DC.1998. Human diploid cell culture rabies vaccine (HDCV) and purified chick embryo cell culture rabies vaccine (PCECV) both confer protective immunity against infection with the sil-ver-haired bat rabies virus strain (SHBRV). Vaccine16:1656 –1659. 20. Davis AD, Rudd RJ, Bowen RA.2007. Effects of aerosolized rabies virus
exposure on bats and mice. J. Infect. Dis.195:1144 –1150.
21. Ndaluka C.2011. Characterization of big brown bat (Eptesicus fuscus) rabies virus in a mouse model. Ph.D. dissertation. Colorado State Univer-sity, Fort Collins, CO.
22. Rudd RJ.8 August 2012, posting date. Rabies annual report. New York State Department of Health Rabies Laboratory, Slingerlands, NY.http:
//www.wadsworth.org/rabies/annualsum.htm.
23. Davis A, Jarvis JA, Pouliott C, Rudd RJ.2013. Rabies virus infection in
Eptesicus fuscusbats born in captivity (naïve bats). PLoS One8:e64808.
doi:10.1371/journal.pone.0064808.
24. Baer GM, Bales GL.1967. Experimental rabies infection in the Mexican freetail bat. J. Infect. Dis.117:82–90.
25. Franka R, Johnson N, Müller T, Vos A, Neubert L, Freuling C, Rup-precht CE, Fooks AR.2008. Susceptibility of North American big brown bats (Eptesicus fuscus) to infection with European bat lyssavirus type 1. J. Gen. Virol.89(Pt 8):1998 –2010.
26. Hughes GJ, Kuzmin IV, Schmitz A, Blanton J, Manangan J, Murphy S, Rupprecht CE.2006. Experimental infection of big brown bats (Eptesicus fuscus) with Eurasian bat lyssaviruses Aravan, Khujand, and Irkut virus. Arch. Virol.151:2021–2035.
27. Wilson PJ, Oertli EH, Hunt PR, Sidwa TJ.2010. Evaluation of a post-exposure rabies prophylaxis protocol for domestic animals in Texas: 2000-2009. J. Am. Vet. Med. Assoc.237:1395–1401.
28. Lafon M.2011. Evasive strategies in rabies virus infection. Adv. Virus Res. 79:33–53.
29. Porterfield JS.1981. Antibody-mediated enhancement of rabies virus. Nature290:542.
30. Prabhakar BS, Nathanson N.1981. Acute rabies death mediated by an-tibody. Nature290:590 –591.
31. King AA, Sands JJ, Porterfield JS.1984. Antibody-mediated enhance-ment of rabies virus infection in a mouse macrophage cell line (P388D1). J. Gen. Virol.65(Pt 6):1091–1093.