Journal of General Virology (2001), 82, 2191–2197.
Printed in Great Britain
Immunogenicity of an E1-deleted recombinant human
adenovirus against rabies by different routes of administration
Ad Vos,1Andreas Neubert,1Elke Pommerening,1Thomas Mu ? ller,2Leopold Do ? hner,3
Larissa Neubert1and Kenneth Hughes4
1Impfstoffwerk Dessau-Tornau GmbH, PO Box 214, 06855 Rosslau, Germany
2Institute for Epidemiological Diagnostics, Federal Research Centre for Virus Diseases of Animals, Seestrasse 55, 16868 Wusterhausen,
3Micromun GmbH, Walther-Rathenau Strasse 49a, 17489 Greifswald, Germany
4Microbix Biosystems Inc., 341 Bering Avenue, Toronto, Ontario, Canada M8Z 3A8
The immunogenic properties of an E1-deleted, human adenovirus type 5 (Ad5) vaccine virus with
activity against rabies were examined in mice, foxes and dogs using different routes of
administration. NMRI mice received 105?8, 105?3, 104?3, 103?3and 102?3TCID50by peroral or
intramuscular (i.m.) administration. Furthermore, six mice received 105?8TCID50intracerebrally
(i.c.). The construct elicited marked seroconversion in mice after oral administration. Immuno-
reactivity in mice was even more pronounced i.m. and i.c. After direct oral administration (108?0
immunized by direct injection (107?7TCID50) in the membrane of the jejunum were shown to
seroconvert. Pre-existing immunity against canine adenovirus did not hinder the development of
rabies VNA after oral application of the construct (108?0TCID50). Fox cubs (24–29 days old) born
from rabies-immune vixens were shown to develop very high levels of rabies VNA after i.m.
administration (108?0TCID50), indicating that the immunogenicity of the construct could surpass
maternally transferred immunity. In dogs, the construct (108?0TCID50) induced a very strong
immune response after i.m. administration. However, no immune response was detectable in dogs
(108?0TCID50). Hence, it must be concluded that the construct is not suitable for oral vaccination of
dogs against rabies.
Oral vaccination of foxes (Vulpes vulpes) against rabies has
been developed as a preferred method to control, and
eventually to eradicate, vulpine rabies (Sto ? hr & Meslin,
1996; Mu ? ller & Schlu ? ter, 1998). This approach has also been
those that cannot be vaccinated by the parenteral route.
However, due to the close relationship between dogs and
humans, more stringent safety conditions are required for oral
vaccination of dogsagainst rabies (Wandeler, 2000). Presently,
no commercially available oral rabies vaccine is completely
without risk. In order to address this issue, new vaccine
Author for correspondence: Ad Vos.
Fax ?49 34901 885797. e-mail ad.vos?idt-direct.de
candidates have been under investigation. One such family of
candidates are recombinant human adenovirus type 5 (Ad5)
vectors that express the highly immunogenic rabies gly-
coproteinG (RG) (Yaroshet al., 1996; Xiang et al., 1996; Wang
et al., 1997). One of these prototype vaccines has already been
shown to induce protective immunity in animals upon
challenge with rabies virus (Prevec et al., 1990; Charlton et al.,
1992). However, the potential for in vivo replication of this E3-
inserted recombinant adenovirus is an undesirable attribute
from a safety perspective. In response to this shortcoming, a
replication-deficient, E1-deleted, human Ad5 vector has been
designated human Ad5 HCMV–intron–ERA, was evaluated
for its ability to induce an immune response by different routes
of administration in mice (Mus musculus), foxes and dogs. It
was also determined whether pre-existing canine adenovirus
0001-7747 ? 2001 SGM
A. Vos and others A. Vos and others
(CAV)-specific immunity inhibited the development of rabies
Finally, the ability of the construct to surpass maternally
transferred immunity against rabies in young animals was
examined. The variability of construct immunogenicity in
terms of target species and route of administration was
? Vaccine virus. The vaccine virus was produced by a homologous
Biosystems Inc., Toronto, Canada). Briefly, cDNA for the ERA (Evelyn–
Rochitniki–Abelseth) strain RG was inserted into the XbaI site of plasmid
pCA13 just downstream of the forward-orientated HCMV immediate-
early promoter and ahead of the SV40 polyadenylation signal. This
generated plasmid pCA13-RG.
by Berkner & Sharp (1985), is known to increase protein expression in
multiple systems. It was cloned into pCA13-RG, downstream of the
promoter region and before the RG gene. The intron sequence in
pMLPsp1a contains the 5? splice-donor site from the Ad2 major late
promoter (MLP) fused to the 3? acceptor site from an immunoglobulin
light-chain gene. The primary sequence of the intron is as follows: 5?
GAATTCAAAG GCGTCTAACC AGTCACAGTC GCAAGGTA-
GG CTGAGCACCG TGGCGGGCGG CAGCGGGTG GCGGTC-
GGGGT TGTTTCTGGC GGAGGTGCTG CTGATGATGTAATT-
AAAGTA GGCGGTCTTG AGACGGCGGA TGGTCGAGGT GA-
GGTGTGGC AGGCTTGAGA TCTGGCCATA CACTTGAGTG
ACAATGACAT CCACTTTGCC TTTCTCTCCA CAGGTGTCCA
CTCCCAGGTC CAACCGGATC C 3?.
The intron was extracted from pMLPsp1a by PCR under standard
conditions. The PCR product was restricted with EcoRI and ligated into
similarly cut pCA13-RG to generate pCA13(int)-RG. Nucleotide orien-
tation was confirmed by sequencing.
In order to produce recombinant human Ad5 bearing the HCMV–
intron–ERA cassette, pCA13(int)-RG was co-transfected with plasmid
pBHGE3, which contains the right-hand end of Ad5 (rightward of E1),
into 293 cells. 293 cells are a human kidney cell line that has been
transformed with left-end regions of Ad5, including E1. Homologous
recombination between pBHGE3 and pCA13(int)-RG results in the
production of a viral genome that will replicate in 293 cells and produce
progeny virus. 293 cells provide helper E1 function. Individual virus
isolates are selected by plaque purification on immobilized 293 cell
HCMV–intron–ERA virus was propagated routinely in suspension
culture using 293N3S as the host cell line. For cultivation, the suspension
cells were placed in spinner flasks containing growth medium (MEM
Eagle, Joklik modification, Sigma) with 10% foetal calf serum. Virus was
harvested 3 days after infection of cells. HCMV–intron–ERA virus was
enumerated by TCID??on 293N3S cells.
? Animals. All animals were without detectable levels of rabies virus-
neutralizing antibodies (VNA) prior to vaccination, with the exception of
the fox cubs born from rabies-immune vixens. Foxes were marked
individually with electronic identifiers and juvenile specific-pathogen-
free dogs (beagles) were identified individually by a tattoo (ear). All
animal tests were conducted at the Experimental Animal Facility of
Impfstoffwerk Dessau-Tornau GmbH (IDT). Animal experimentation
was performed according to the German Animal Welfare Act of 25 May
1998. The experimental design of the tests, as required, was approved by
the appropriate German authorities.
? Rapid fluorescence focus inhibition test (RFFIT) assays
Rabies virus. Prior to testing, sera were heat-inactivated at 56 ?C
for 30 min and then centrifuged at 926 g for 5 min. Serum samples
were evaluated for rabies VNA by the RFFIT as described by Smith et
al. (1973), with the modifications described by Cox & Schneider (1976).
The rabies VNA titres were converted to international units (IU). The
international standard immunoglobulin (2nd human rabies immuno-
globulin preparation, National Institute for Standards and Control,
Potters Bar, UK), adjusted to 0?5 IU?ml, served as a positive control
for these experiments. A level of 0?5 IU?ml is as an arbitrarily defined
threshold indicative of protection against rabies infection (Anonymous,
1978) and is used here as the threshold for positivity.
CAV2. A simple inhibition test (Mayr et al., 1977) was used to test
sera for VNA against CAV2. Firstly, the serum samples were inacti-
vated at 56 ?C for 30 min. A virus suspension with a known titre was
added to each serum dilution and the mixture was incubated at 37 ?C
for 60 min. This mixture was then inoculated on monolayers of recep-
tive Madin–Darby canine kidney (MDCK) cells for virus replication.
Four days later, CAV2-neutralizing antibody titres were evaluated ac-
cording to Spearman–Ka? rber, based on the CPE. A titre of 1:4 was
? ELISA. The objective of these tests was to detect both general
for human Ad5. For detection of the former, highly purified human Ad5
hexon protein was used. This protein was isolated by DEAE chromato-
graphy and crystallization (Do ? hner & Hudemann, 1972). The hexon
proteinis the most abundant adenovirusprotein (Zabner et al., 1997), and
antibodies directed against this protein could also be the result of earlier
CAV infection. The Ad5-specific ELISA was conducted with highly
purified human Ad5 fibre antigen. Fibre antigen was isolated by anion
exchange and immunoaffinity chromatography.
The ELISA was performed under standard conditions using hexon or
fibre antigens, at 1 µg?ml, for coating microtitre plates. Peroxidase-
labelled protein A was used as the indicator reagent. Human antibodies
were used as the positive control. Dog, fox and human antibodies react
similarly with protein A (Yamamoto et al., 1985). The absorbances of
‘cut-off values’ at 450?620 nm for specific antibodies were in the range
of 0?5. Adenoviruses are ubiquitous; hence, it is to be expected that
animals have circulating antibodies against adenoviruses. Serum samples
from mice were diluted 1:10 to allow the detection of low reactivity
between IgG and protein A.
? Statistics. Student’s paired t-test was used to detect possible
intraspecific significance in absorbance differences between different
blood samples (Zo ? fel, 1988).
Study 1. Mice [peroral (p.o.), intramuscular (i.m.) and
Three groups of six mice were given 0?02 ml human Ad5
HCMV–intron–ERA (10??? TCID??) p.o., i.m. or i.c. Animals
that received the construct i.c. were anaesthetized by intra-
peritoneal (i.p.) injection with a 1:10 dilution of ketamine
hydrochloride in saline. The animals vaccinated p.o. drank the
construct voluntarily from a syringe. During a subsequent trial
in mice, four groups of 10 animals each received 0?02 ml of the
construct using different concentrations (10???, 10???, 10??? and
10??? TCID??). Five animals of each group received the
construct orally, the other five by the parenteral route (i.m.).
Recombinant adenovirus against rabiesRecombinant adenovirus against rabies
Table 1. Rabies VNA titres of mice inoculated with human Ad5 HCMV–intron–ERA by different routes of administration
Five or six animals were inoculated with the doses shown. Titres for individual animals are given in IU?ml, as determined by RFFIT. (–), No rabies VNA detectable (?0?5 IU?ml); ,
269?2, 226?3, 226?3, 269?3, 320?1, 269?2
67?3, 56?6, 56?6, 80?0, 190?3, 80?0
20?0, 20?0, 40?0, (–), 80?0
33?6, 5?0, 10?0, 8?4, Died*
(–), 56?5, 4?2, (–), 0?9
(–), (–), (–), (–), (–)
10?0, 7?1, 10?0, (–), 67?3, 5?9
47?6, 7?1, (–), 67?3, (–)
(–), 10?0, (–), (–), (–)
(–), (–), (–), (–), (–)
(–), (–), (–), (–), (–)
* Animal died 2 days post-vaccination; cause of death unknown, but unrelated to rabies.
Table 2. Rabies VNA titres of foxes inoculated orally with
108?33TCID50human Ad5 HCMV–intron–ERA
Titres are given in IU?ml, as determined by RFFIT. (–), No rabies
VNA detectable (?0?5 IU?ml); , not done.
Fox B0 B1B2B3 B4
This time, the construct was applied orally with a perfusion
needle in the posterior part of the oral cavity of the animals.
For this purpose, the mice were anaesthetized in an ether jar
prior to administration of the construct. All mice were bled by
retro-orbital puncture and subsequently euthanized 55 or 60
days post-immunization. Blood samples were examined for
serum antibody titres against rabies by using RFFIT and
antibodies against the human Ad5 fibre and hexon proteins
The immune responses against rabies of mice inoculated
with human Ad5 HCMV–intron–ERA vaccine are shown in
Table 1. The highest levels of rabies VNA were observed in
themice inoculatedi.c.Theimmune responsein mice receiving
the human Ad5 HCMV–intron–ERA construct orally was
inferior to mice immunized i.m. with the same dose. It was also
observed that the seroconversion rate was dose-dependent.
With the exception of three (of six) mice vaccinated i.c., none
of the animals showed high levels of antibodies against the
fibre and hexon protein (data not shown).
Study 2. Foxes (p.o.)
Eight juvenile foxes received 10??? TCID??human Ad5
HCMV–intron–ERA by direct oral instillation. Blood samples
werecollected 0 (B0), 21 (B1), 35 (B2), 70 (B3) and 95 (B4) days
post-vaccination. All blood samples were examined for the
presence of rabies virus- and CAV-neutralizing antibodies by
RFFIT. Detection of antibodies against the human Ad5 fibre
and hexon proteins was also performed by ELISA.
Six of the eight foxes seroconverted (threshold 0?5 IU?ml).
The geometric mean titres (GMT) of the sera of immunized
foxes were 2?75 (B1), 2?13 (B2), 1?33 (B3) and 1?48 (B4) IU?ml
(Table 2). None of the animals had detectable levels of CAV-
neutralizingantibodies at any timeduring the study. The mean
absorbances against the fibre protein of serum samples B0 and
A. Vos and others A. Vos and others
Table 3. CAV2 VNA titres of foxes pre-immunized with
Titres were determined by RFFIT. (–), No CAV2 VNA detectable
B2 were 0?296?0?071 and 0?415?0?112, respectively. For
the hexon protein, the following mean absorbances for B0
and B2 were obtained: 0?779?0?361 and 1?439?0?724. The
levels of antibodies against the hexon and fibre proteins of
?2?43, d.f.?7, P?0?05).
Study 3. Foxes (intrajejunal)
Five cubs, 8 weeks old and born from rabies-naı?ve vixens,
received by injection 0?5 ml human Ad5 HCMV–intron–ERA
(10??? TCID??) in the membrane of the jejunum following
surgical exposure (laparotomy). For this purpose, the animals
were sedated with a ketamine–xylazine (10%?2%) solution.
Blood samples were collected 0 (B0), 21 (B1) and 49 (B2) days
post-vaccinationand examinedfor the presenceof rabies VNA
All five foxes had detectable levels of rabies VNA. Sera
GMT for the vaccinated foxes were 32?2 (B1) and 11?9 (B2)
Study 4. Foxes (p.o.) vaccinated against CAV2
The effect of pre-existing CAV-specific immunity on the
immune response against rabies, induced by the construct, was
investigated by administering 10??? TCID??CAV2 antigen
(produced at IDT) i.m. to four adult foxes after a blood sample
(B0) had been taken. Twenty-nine days later, the foxes were
bled (B1) and subsequently received 10??? TCID??human Ad5
HCMV–intron–ERA vaccine by direct oral instillation. Blood
samples were taken 28 (B2) and 55 (B3) days after adminis-
tration of the construct. The blood samples were examined for
the presence of rabies and CAV VNA by RFFIT.
All four foxes developed detectable levels of CAV2 VNA
after vaccination with CAV2 antigen and prior to adminis-
tration of the construct (B0 and B1) (Table 3). All animals pre-
immunized with CAV2 developed rabies VNA upon oral
immunization with the construct. The GMT were 3?61 and
3?27 IU?ml at 28 (B2) and 55 (B3) days post-administration.
These results indicate that antibodies to CAV2 did not inhibit
the immune response to rabies after vaccination with the
Study 5. Fox cubs (i.m.)
In order to investigate whether maternally transferred
immunitytorabies viruswouldimpairthe immuneresponse to
RG upon vaccination with the human Ad5 HCMV–intron–
ERA construct, 13 cubs born from four rabies-immune vixens
were given the construct i.m. The vixens had been vaccinated
orally with SAD (Street Alabama Dufferin) B19 just before
mating. SAD B19 is a live-modified rabies virus vaccine, used
extensively for oral vaccination of wildlife (Schneider & Cox,
(mean 24?3 days), were bled (B0) and subsequently injected
i.m. with 10??? TCID??human Ad5 HCMV–intron–ERA.
Blood samples were collected 14 (B1), 21 (B2), 28 (B3) and 47
(B4) days post-vaccination and examined for the presence of
rabies VNA (RFFIT).
The rabies VNA titres of the four vixens at the time of
vaccination of the cubs were 1?81, 11?02, 11?88 and 5?12 IU?
ml. All cubs born from these rabies-immune vixens developed
rabies VNA upon immunization with the construct. The GMT
of the sera of the cubs were 0?38 (B0), 31?08 (B1), 108?73 (B2)
and 155?26 (B3) IU?ml.
Study 6. Dogs [p.o., i.m. and gastrointestinal (g.i.)]
Six dogs received 10??? TCID??human Ad5 HCMV–
intron–ERA by direct oral instillation. Blood samples were
collected 0 (B0), 21 (B1), 35 (B2), and 118 (B3) days post-
vaccination. After the first oral instillation, none of the animals
seroconverted. Hence, vaccination was repeated using a
different lot of vaccine virus. Vaccination was repeated with
10??? TCID??, four animals by the oral route and two by
parenteral (i.m.) administration. Dogs were bled 28 (B4) and 55
(B5) days after this second vaccination attempt.
The dogs that received the secondary vaccination by the
oral route did not develop detectable levels of rabies VNA.
Only the i.m.-inoculated animals seroconverted, with high
levels of VNA [dog 3160, 99?00 (B4) and 160?89 (B5)
IU?ml; dog 3161, 18?80 (B4) and 41?67 (B5) IU?ml]. Dogs
3160 and 3161 also showed a clear immune response against
route. The other four orally vaccinated dogs showed no
detectable immune response against these proteins (Fig. 1).
The four dogs that remained seronegative after secondary
oral administration received 10??? TCID??
HCMV–intron–ERA by direct intestinal instillation (endo-
scopic deposition) in the duodenum. The animals were
anaesthetized with a ketamine–xylazine (10%?2%) solution;
pre-medication was with atropine–diazepam. Blood samples
Recombinant adenovirus against rabiesRecombinant adenovirus against rabies
B0B2 B3B4 B5
Fig. 1. Absorbances of serum samples (B0, B2–B5) from dogs inoculated
with 108?33and 108?0TCID50human Ad5 HCMV–intron–ERA during
primary and secondary vaccination against the human Ad5 hexon (a) and
fibre (b) proteins, as determined by ELISA. Samples were included from
dogs that received the construct p.o. during both primary and secondary
vaccinations (?) and from dogs that received the construct respectively
p.o. and i.m. during primary and secondary vaccinations (?).
Blood samples were examined by RFFIT for the presence of
rabies (B0–B8) and CAV (B0–B5) VNA. Antibody levels
against the human Ad5 fibre and hexon proteins were also
determined by ELISA (B0–B5).
VNA after endoscopic deposition of the HCMV–intron–ERA
construct in the duodenum.
An effective yet safe rabies vaccine is needed for the oral
immunization of a variety of wild animals, as well as the oral
vaccination of (stray) dogs. Human adenoviruses have been
studied intensively as virus vectors for vaccine delivery and
gene therapy applications (for reviews, see Klonjkowski et al.,
1999; Russell, 2000), and have been shown to induce
protective responses after oral administration (Fooks et al.,
1998). The adenovirus genome has been well characterized,
vaccine vectors. This application is supported further by the
widehostrangeof human adenoviruses and the stabilityof the
by routine culture techniques.
The strong rabies immune response induced by replication-
defective adenovirus recombinants that express RG and the
absence of many of the safety risks inherent to multiplying
agents led us to test the E1-deleted human Ad5 HCMV–
intron–ERA as a candidate for oral vaccination of carnivores
against rabies. This replication-defective human Ad5 can only
be propagated in cells that express the E1-region proteins
exogenously and thereby complement the deletion in the viral
genome. This virus is unable to replicate in cells that do not
complement the deleted E1 region, but it is still able to infect
these cells and to induce the synthesis of very high levels of
foreign protein (RG). Hence, the potential spread of recom-
binant, replication-competent adenovirus (RCA) by horizontal
transmission from vaccinated to unvaccinated animals, as
mentioned by Hammond et al. (2000), is not a concern with
human Ad5 HCMV–intron–ERA. However, the E1 deletion
may not block the expression of virus genes completely
following in vivo administration (Von Seggern & Nemerow,
1999). The absence of RCA after serial passaging in 293 and
293N3S cells attests further to the safety of this virus vector.
Thereasonfor theabsence ofRCAinthe HCMV–intron–ERA
virus preparation is not understood. All primary sequence
elements required for homologous recombination have been
confirmed as present by sequencing analysis. However, the
absence of RCA is clearly linked to the presence of the intron
moiety, as other recombinant rabies vaccines containing the
frequencies in the order of 10−?–10−? after five serial passages
(M. Moore and K. Hughes, unpublished).
In contrast to studies with similar replication-defective
humanAd5 constructs (Prevec et al.,1990; Xiang &Ertl,1999),
mice developed serum antibody to rabies upon oral adminis-
tration of human Ad5 HCMV–intron–ERA. However, the
immune response of mice vaccinated orally was inferior to that
of animals vaccinated i.m. or i.c., and was clearly dose-
dependent. In rabies-naı?ve foxes, six of eight animals did
seroconvert after direct oral administration. However, the
immune response was lower than that in foxes vaccinated
P5?88, showing a sub-optimal immunization process (Neubert
et al., 2001; Schuster et al., 2001).
As exposure to CAV is relatively common in wild and
domestic canids (Garcelon et al., 1992; Laurenson et al.,
1997; Gese et al., 1997; Truyen et al., 1998; Cypher et al.,
1998; Spencer et al., 1999), it is to be expected that many
animals have developed VNA against CAV. However, pre-
immunity to CAV did not impair the ability of the human Ad5
HCMV–intron–ERAconstruct toelicitan immune response to
rabies. We have shown previously that cubs whelped by
upon oral immunization with the highly efficacious vaccine
virus SAD B19. This inhibition of the immune response
outlasted the presence of detectable levels of maternal
antibodies (Mu ? ller et al., 2001). The rabies-specific VNA
response to the human Ad5 HCMV–intron–ERA construct,
after i.m. administration in cubs whelped by rabies-immune
vixens, was significantly higher than the response elicited to
the oral rabies virus vaccine SAD B19 in cubs born from naı?ve
vixens (Mu ? ller et al., 2001). This demonstrates that maternal
A. Vos and others A. Vos and others
as presented by the human Ad5 recombinant vaccine.
Therefore, the construct is highly suitable for neonatal
immunization of dogs. Many human rabies cases are caused by
bites from rabid puppies that are too young to be eligible for
Dogs that received the human Ad5 HCMV–intron–ERA
construct by direct oral administration did not develop
detectable levels of VNA against rabies. Furthermore, the
construct did not elicit an immune response after g.i.
immunization by endoscopic deposition. This supports the
results of Papp et al. (1997), who could not observe an immune
response after g.i. immunization of cotton rats (Sigmodon
hispidus) with another replication-defective human Ad5 ex-
pressing glycoprotein D of bovine herpesvirus-1. The relative
inefficiency of mucosal delivery for adenovirus infection is
striking, since adenoviruses naturally infect mucosal tissues
and human epidemiological data suggest frequent adenovirus
infection. Indeed, human Ad5 is a common pathogen that
infects most humans at an early age (Xiang et al., 1996; Wang
et al., 1997; Walters et al., 1999). A likely explanation is that
wild-type Ad5 infection requires a large initial inoculum
(Walters et al., 1999). This explains the relative inefficiency of
replication-defective adenovirus as virus vectors for vaccine
delivery when compared with similar recombinant RCA.
Entry of human adenoviruses into cells is a stepwise
process. After binding to the fibre receptor (CAR;
coxsackievirus?adenovirus receptor), interaction of penton
base with integrins facilitates internalization via receptor-
mediated endocytosis (Dmitriev et al., 2000). It has been
suggested that the adenovirus fibre receptors are polarized to
the basolateral plasma membrane of the epithelium, and the
lack of fibre binding at the apical surface appears to be the rate-
limiting step that explains insufficient adenovirus-mediated
gene transfer (Walters et al., 1999). According to Zabner et al.
(1997), intentional injury of the epithelium increased the
efficiency of gene transfer. However, in our study, the lumen
epithelia in the intestines of three dogs were damaged
(gastritis) and the animals still did not respond to the vaccine
delivery. This suggests that the cells of the intestinal smooth
muscle lack the appropriate cell-surface receptors for adeno-
virus attachment (Dmitriev et al., 2000; Schmidt et al., 2000).
Surprisingly, direct injection of the construct in the membrane
of the jejunum in foxes elicited a strong immune response. It
adenovirus vaccines in eliciting an immune response after oral
delivery is partially species-dependent (Monteil et al., 2000).
Although human Ad5 HCMV–intron–ERAwasable toinduce
an immune response in mice after oral delivery, dogs could not
be vaccinated against rabies by this route and foxes developed
that, despite high levels of RG expression in vitro and
significant VNA production by i.m. administration, the human
Ad5 HCMV–intron–ERA construct is not a suitable candidate
for oral vaccination of the tested carnivores against rabies.
We are grateful to Dr Ludvik Prevec of McMaster University,
Anonymous (1978). WHO informal consultation on reference prepara-
tions and potency tests for rabies vaccines. Developments in Biological
Standardization 40, 265–271.
Berkner, K. L. & Sharp, P. A. (1985). Effect of the tripartite leader on
synthesis of a non-viral protein in an adenovirus 5 recombinant. Nucleic
Acids Research 13, 841–857.
Charlton, K. M., Artois, M., Prevec, L., Campbell, J. B., Casey, G. A.,
Wandeler, A. I. & Armstrong, J. (1992). Oral rabies vaccination of
skunks and foxes witha recombinanthuman adenovirusvaccine. Archives
of Virology 123, 169–179.
Cox, J. H. & Schneider, L. G. (1976). Prophylactic immunization of
humans against rabies by intradermal inoculation of human diploid cell
culture vaccine. Journal of Clinical Microbiology 3, 96–101.
Cypher, B. L., Scrivner, J. H., Hammer, K. L. & O’Farrell, T. P. (1998).
Viral antibodiesincoyotesfrom California.Journal ofWildlife Diseases 34,
Dmitriev, I., Kashentseva, E., Rogers, B. E., Krasnykh, V. & Curiel,
D. T. (2000). Ectodomain of coxsackievirus and adenovirus receptor
genetically fused to epidermal growth factor mediates adenovirus
targeting to epidermal growth factor receptor-positive cells. Journal of
Virology 74, 6875–6884.
Do?hner, L. & Hudemann, H. (1972). Untersuchungen zur Kristallisation
von Hexonen der Adenoviren der Gruppe III. Archiv fu ? r die gesamte
Virusforschung 38, 279–289.
Fooks, A. R., Jeevarajah, D., Lee, J., Warnes, A., Niewiesk, S., ter
Meulen, V., Stephenson, J. R. & Clegg, J. C. S. (1998). Oral or
responses in rodents. Journal of General Virology 79, 1027–1031.
Garcelon, D. K., Wayne, R. K. & Gonzales, B. J. (1992). A serologic
survey of the island fox (Urocyon littoralis) on the Channel Islands,
California. Journal of Wildlife Diseases 28, 223–229.
Gese, E. M., Schultz, R. D., Johnson, M. R., Williams, E. S., Crabtree,
R. L. & Ruff, R. L. (1997). Serological survey for diseases in free-ranging
coyotes (Canis latrans) in Yellowstone National Park, Wyoming. Journal
of Wildlife Diseases 33, 47–56.
Hammond, J. M., McCoy, R. J., Jansen, E. S., Morrissy, C. J., Hodgson,
A. L. M. & Johnson, M. A. (2000). Vaccination with a single dose of a
gp55 (E2) gene protects pigs against classical swine fever. Vaccine 18,
Klonjkowski, B., Denesvre, C. & Eliot, M. (1999). Adenoviral vectors
for vaccines. In Adenoviruses: Basic Biology to Gene Therapy, pp. 163–173.
Edited by P. Seth. Austin, TX: Landes.
Laurenson, K., Van Heerden, J., Stander, P. & Van Vuuren, M. J.
(1997). Seroepidemiological survey of sympatric domestic and wild
dogs (Lycaon pictus) in Tsumkwe District, north-eastern Namibia.
Onderstepoort Journal of Veterinary Research 64, 313–316.
Mayr, A., Bachmann, P. A., Bibrack, B. & Wittmann, G. (1977).
Virologische Arbeitsmethoden, Band II. Jena: Gustav Fischer.
Monteil, M., Le Pottier, M. F., Ristov, A. A., Cariolet, R., L’Hospitalier,
R., Klonjkowski, B. & Eloit, M. (2000). Single inoculation of replication-
defective adenovirus-vectored vaccines at birth in piglets with maternal
antibodies induces high level of antibodies and protection against
pseudorabies. Vaccine 18, 1738–1742.
Recombinant adenovirus against rabiesRecombinant adenovirus against rabies
Mu?ller, T. & Schlu?ter, H. (1998). Oral immunization of red foxes (Vulpes
vulpes L.) in Europe – a review. Journal of Etlik Veterinary Microbiology 9,
Mu?ller, T., Schuster, P., Vos, A., Selhorst, T., Wenzel, U. & Neubert, A.
(2001). Effect of maternal immunity on the immune response of young
foxes to oral vaccination against rabies with SAD B19. American Journal
of Veterinary Research (in press).
Neubert, A., Schuster, P., Mu?ller, T., Vos, A. & Pommerening, E.
(2001). Immunogenicity and efficacy of the oral rabies vaccine SAD B19
in foxes. Journal of Veterinary Medicine B 48, 179–183.
Papp, Z., Middleton, D. M., Mittal, S. K., Babiuk, L. A. & Baca-Estrada,
M. E. (1997). Mucosal immunization with recombinant adenoviruses:
induction of immunity and protection of cotton rats against respiratory
bovine herpesvirus type 1 infection. Journal of General Virology 78,
Prevec, L., Campbell, J. B., Christie, B. S., Belbeck, L. & Graham, F. L.
(1990). A recombinant human adenovirus vaccine against rabies. Journal
of Infectious Diseases 161, 27–30.
Russell, W. C. (2000). Update on adenovirus and its vectors. Journal of
General Virology 81, 2573–2604.
Schmidt, M. R., Piekos, B., Cabatingan, M. S. & Woodland, R. T.
permits adenovirus infection of primary lymphocytes. Journal of Im-
munology 165, 4112–4119.
Schneider, L. G. & Cox, J. H. (1983). Ein Feldversuch zur oralen
Immunisierung von Fu ? chsen gegen die Tollwut in der Bundesrepublik
Deutschland. I. Unscha ? dlichkeit, Wirksamkeit und Stabilita ? t der Vakzine
SAD B19. Tiera?rztliche Umschau 38, 315–324.
Schuster, P., Mu?ller, T., Vos, A., Selhorst, T., Neubert, L. &
Pommerening, E. (2001). Comparative immunogenicity and efficacy
studies with oral rabies vaccine SAD P5?88 in raccoon dogs and foxes.
Acta Veterinaria Hungarica (in press).
Smith, J. S., Yager, P. A. & Baer, G. M. (1973).Arapidreproducibletest
for determining rabies neutralizing antibody. Bulletin of the World Health
Organization 48, 535–541.
Spencer, J. A., Bingham, J., Heath, R. & Richards, B. (1999). Presence
of antibodies to canine distemper virus, canine parvovirus and canine
adenovirus type 1 in free-ranging jackals (Canis adustus and Canis
mesomelas) in Zimbabwe. Onderstepoort Journal of Veterinary Research 66,
Sto?hr, K. & Meslin, F.-M. (1996). Progress and setbacks in the oral
immunisation of foxes against rabies in Europe. Veterinary Record 139,
Truyen, U., Mu?ller, T., Heidrich, R., Tackmann, K. & Carmichael, L. E.
(1998). Survey on viral pathogens in wild red foxes (Vulpes vulpes) in
Germany with emphasis on parvoviruses and analysis of a DNA
sequence from a red fox parvovirus. Epidemiology and Infection 121,
Von Seggern, D. J. & Nemerow, G. R. (1999). Adenoviral vectors for
protein expression. In Gene Expression Systems: Using Nature for the Art of
Expression, pp. 111–156. Edited by J. M. Fernandez & J. P. Hoeffler. San
Diego: Academic Press.
Vos, A., Mu?ller, T., Schuster, P., Schlu?ter, H. & Neubert, A. (2000).
Oral vaccination of foxes against rabies with SAD B19 in Europe,
1983–1998: a review. Veterinary Bulletin 70, 1–6.
Walters, R. W., Grunst, T., Bergelson, J. M., Finberg, R. W., Welsh,
M. J.&Zabner, J.(1999).Basolaterallocalizationoffiberreceptorslimits
adenovirus infection from the apical surface of airway epithelia. Journal of
Biological Chemistry 274, 10219–10226.
Wandeler, A. I. (2000). Oral immunization against rabies: afterthoughts
and foresight. Schweizer Archiv fu ? r Tierheilkunde 142, 455–462.
Wang, Y., Xiang, Z., Pasquini, S. & Ertl, H. C. J. (1997). The use of an
E1-deleted, replication-defective adenovirus recombinant expressing the
rabies virus glycoprotein for early vaccination of mice against rabies
virus. Journal of Virology 71, 3677–3683.
Xiang, Z. & Ertl, H. C. J. (1999). Induction of mucosal immunity with a
replication-defective adenoviral recombinant. Vaccine 17, 2003–2008.
Xiang, Z. Q., Yang, Y., Wilson, J. M. & Ertl, H. C. J. (1996). A
replication-defective human adenovirus recombinant serves as a highly
efficacious vaccine carrier. Virology 219, 220–227.
Yamamoto, S., Omura, M. & Hirata, H. (1985). Isolation of porcine,
canine and feline IgG by affinity chromatography using protein A.
Veterinary Immunology and Immunopathology 5, 195–200.
Yarosh, O. K., Wandeler, A. I., Graham, F. L., Campbell, J. B. & Prevec,
L. (1996). Human adenovirus type 5 vectors expressing rabies
glycoprotein. Vaccine 14, 1257–1264.
Zabner, J., Freimuth, P., Puga, A., Fabrega, A. & Welsh, M. J. (1997).
Lack of high affinity fiber receptor activity explains the resistance of
ciliated airway epithelia to adenovirus infection. Journal of Clinical
Investigation 100, 1144–1149.
Zo?fel, P. (1988). Statistik in der Praxis. 2. Auflage. Stuttgart: Gustav
Received 22 March 2001; Accepted 15 May 2001