Protection of non-human primates against rabies with an adenovirusrecombinant vaccine

Z.Q. Xiang a, L. Greenberg b, H.C. Ertl a,n, C.E. Rupprecht c,d

a The Wistar Institute of Anatomy & Biology, Philadelphia, PA, United Statesb Centers for Disease Control & Prevention, Atlanta, GA, United Statesc The Global Alliance for Rabies Control, Manhattan, KS, United Statesd Ross University School of Veterinary Medicine, Basseterre, St. Kitts, West Indies

a r t i c l e i n f o

Article history:Received 24 September 2013Returned to author for revisions17 October 2013Accepted 20 December 2013Available online 9 January 2014

Keywords:Adenoviral vaccineNHPRabies virusVNA

a b s t r a c t

Rabies remains a major neglected global zoonosis. New vaccine strategies are needed for human rabiesprophylaxis. A single intramuscular immunization with a moderate dose of an experimental chimpanzeeadenovirus (Ad) vector serotype SAd-V24, also termed AdC68, expressing the rabies virus glycoprotein,resulted in sustained titers of rabies virus neutralizing antibodies and protection against a lethal rabiesvirus challenge infection in a non-human primate model. Taken together, these data demonstrate thesafety, immunogenicity, and efficacy of the recombinant Ad-rabies vector for further consideration inhuman clinical trials.

& 2013 Elsevier Inc. All rights reserved.

Introduction

Rabies virus causes an acute progressive viral encephalitis, withthe highest case fatality of any infectious disease (Jackson, 2009).Although rabies is one of the oldest recognized zoonoses, itremains a neglected disease, resulting in tens of thousands ofhuman fatalities, and tens of millions of annual exposures (http://www.who.int/mediacentre/factsheets/fs099/en/). Rabies virus ismost commonly transmitted by dogs in the developing world.Children are at high risk of exposure, and nearly half of rabies-related fatalities affect children below the age of 16. In parts ofAfrica, Asia, and Latin America, outdated nerve tissue-basedvaccines are still used, some unchanged little from the time ofPasteur (Wu et al., 2011). Modern human vaccines, while highlyefficacious, are expensive, and require multiple doses for theelicitation of protective immunity (Rupprecht et al., 2009; health.costhelper.com/rabies-vaccine.html). In general, rabies vaccinesare given after a bite, as post-exposure prophylaxis (PEP). In casesof severe exposure, active immunization has to be combined withpassive immunization with a rabies virus immune globulin pre-paration of human (HRIG) or equine origin (ERIG), both of whichare costly and habitually in short supply (Jentes et al., 2013; http://www.who.int/ith/vaccines/rabies/en/). Development of novel

inexpensive, safe and effective rabies biologics for human popula-tions at risk would be a major advantage for rabies prevention inthe developing world. It is well established that protection againstrabies virus is mediated by neutralizing antibodies and titers at orabove 0.5 international units (IU) in general protect. Induction ofrabies virus neutralizing antibodies (VNA) have been shown to becritical in protection of experimental animals exposed to rabiesvirus, as well as humans bitten by rabid animals (Plotkin, 2010;WHO Position Paper, 2007). Previous studies showed that inclu-sion of a rabies vaccine into childhood immunizations resulted inthe development of memory B cells that rapidly mounted potentrabies VNA recall responses after booster immunization(Malerczyk et al., 2007; Suwansrinon et al., 2006). Nevertheless,one study conducted in Thailand concluded that inclusion of rabiesvaccines into vaccination programs for children would not be costeffective when compared to standard PEP (Chulasugandha et al.,2006). A less expensive vaccine able to induce sustained rabiesVNA titers and durable memory B cell responses in a shortenedschedule may render preventative childhood vaccination morecost-effective and thereby reduce the impact of rabies in devel-oping countries, as well as of those deemed at risk of rabies virusexposure in developed countries, such as veterinarians, diagnosticworkers, animal handlers, wildlife researchers, cavers, certaintravelers, selected members of the military, etc. (Manning et al.,2008; http://www.who.int/ith/vaccines/rabies/en/).

The objective of this research was to test the comparativesafety, immunogenicity, and efficacy of an adenovirus (Ad) vector

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/yviro

Virology

0042-6822/$ - see front matter & 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.virol.2013.12.029

n Corresponding author. Tel.: þ1 215 898 3863; fax: þ1 215 898 3868.E-mail address: ertl@wistar.upenn.edu (H.C. Ertl).

Virology 450-451 (2014) 243–249

expressing the rabies virus glycoprotein in a non-human primatemodel. Most of these studies were conducted with chimpanzee-derived Ad vectors. Humans only rarely demonstrate neutralizingantibodies to such viruses (Chen et al., 2010; Xiang et al., 2006),which may impair vaccine efficacy. Vectors expressed the full-length viral glycoprotein of the Evelyn Rokitnicki Abelseth (ERA)strain of rabies virus. Immunization of nonhuman primates withthe Ad vectors induced high and sustained VNA titers to rabiesvirus, confirming previous studies performed in mice (Chen et al.,2010; Xiang et al., 2002), which showed that Ad vectors, althoughmost commonly used for induction of cellular immune responses,also elicited potent and durable transgene product-specifichumoral responses.

Results

Immunogenicity of Ad vectors used in a high dose prime-boostregimen

A pilot experiment was conducted in two Chinese rhesusmacaques to test if Ad vectors expressing the rabies virus glyco-protein induced rabies VNAs and if such responses could beenhanced by booster immunizations. To this end, two monkeysthat had no detectable antibody titers to rabies virus or the Advectors were immunized on day 0 with 1012 vp of the AdC7rab.gpvector given intramuscularly (i.m.). Eight months later, they wereboosted with the same dose of the AdC6rab.gp vector given i.m.Five months later, they were boosted i.m. with 1012 vp of theAdHu5rab.gp vector. Blood was collected at several time pointsafter vaccination. Rabies VNA titers were determined from heat-inactivated plasma. Plasma from naïve monkeys, and monkeysimmunized with vectors expressing an unrelated transgene wereincluded, but none of the latter developed detectable rabies VNAs(o0.2 IU, data not shown). The experimental animals developedVNA titers of approximately 10 IU after the first immunization(Fig. 1). Although values fluctuated, titers were sustained for atleast 8 months. After the boost with AdC6rab.gp, rabies VNA titersvirus increased in both animals �10 fold. In one animal, rabiesVNA titers then contracted to levels obtained after priming, whilethe other animal showed more sustained increases after the firstboost. A second boost with an AdHu5rab.gp vector again increasedrabies VNA titers by more than 10 fold. Overall, these preliminary

results showed that Ad vector immunization induced a potentand sustained rabies VNA response. In addition, vectors inducedmemory B cells that readily differentiate into antibody-secretingcells upon booster immunizations.

Vaccine efficacy in a dose escalation study

The vector dose used in the above pilot study exceeded thedose that would be expected to cause significant toxicity inhumans. Although the AdC vectors we tested have not yet under-gone testing in humans, other E1-deleted Ad vectors of human orsimian serotypes typically cause dose-limiting adverse events at1011 vp (Harro et al., 2009). Therefore, we next tested theAdC68rab.gp vector in a dose escalation experiment at 109 to1011 vp in small groups of two cynomologous monkeys per group.An AdC vector control was used at 1011 vp. A commerciallyavailable human diploid cell rabies vaccine (HDCV), used intra-muscularly at the recommended dose for pre-exposure vaccina-tion, was included. The AdC vector vaccine was only given once,and the HDCV was given three times on days �41, �27 and �20in relation to the AdC vaccine. Animals were challenged at 18weeks after the last vaccine dose with �106.4 (MICLD)50/ml ascalculated via the Spearman–Kaerber method, of a virulent streetstrain of canine rabies virus. The two animals that received thecontrol vector developed signs of rabies. All other vaccinatedanimals had detectable rabies VNA by day 7 post-immunization,and remained disease-free (Fig. 2A).

Effect of pre-existing immunity to a human serotype Ad vector on theefficacy of AdC68rab.gp

Humans are exposed to Ad viruses and in response developneutralizing antibodies, which are serotype-specific, as well asbinding antibodies and T cells, which cross-react between differenthuman and simian serotypes (Hutnick et al., 2010; Xiang et al.,2002). Numerous studies have shown that pre-existing neutraliz-ing antibodies to Ad vectors impact their immunogenicity (McCoyet al., 2007; McElrath et al., 2008; Roberts et al., 2006). Neutraliz-ing antibodies to AdC68 virus are only found rarely in humans(Xiang et al., 2002). Nevertheless, pre-existing binding antibodiesor T cells may also affect vector-induced immune responses.Therefore, in the next experiment, we assessed the effect of pre-exposure to AdHu5 virus on the efficacy of the AdC68rab.gpvaccine. To this end, groups of four animals were either pre-exposed with 1012 vp of an AdHu5 vector expressing a reporterprotein, or they were left untreated. Four weeks later, the pre-treated as well as the untreated animals were vaccinated with 1010

vp of AdC68rab.gp. A third positive control group of two animalsreceived HDCV, given three times on days �41, �27 and �20 inrelation to the AdC vaccination (given on day 0). Four controlanimals received PBS. All vaccinated animals developed detectablerabies VNA after immunization. Although due to the small num-bers of animals there were no significant differences betweenthe vaccine groups, there was a clear trend towards more rapidresponses in the HDCV vaccine group and slower responses in theAdC68rab.gp group, which were even further delayed if animalshad been pre-exposed to the AdHu5 vector (Fig. 2D). Animals werechallenged with �105.2 MICLD50/ml of street rabies virus ofcanine origin 18 weeks after AdC vaccination. As shown inFig. 2B, all of the AdC68rab.gp and HDCV vaccinated animalssurvived the challenge, while two of four control animals devel-oped signs of rabies, indicating that pre-existing immunity tohuman serotype Ad vectors did not reduce the efficacy of theAdC68rab.gp vector. This is in agreement with previous studiesconducted by ourselves and others in laboratory rodents andnonhuman primates, which showed that pre-existing immunity

Fig. 1. Induction of rabies VNA upon sequential immunizations with Advectors. Two rhesus macaques, shown individually in circles or cross-filled squares,were primed with AdC7rab.gp, and then boosted with AdC6rab.gp, with AdHu5rab.gp administered at the time points indicated by the arrows. The rabies VNA titerswere measured from sera at various time points. Titers are shown as internationalunits (IU) according to a reference serum tested in parallel. Control animalsimmunized with the same vectors, but expressing an antigen of HIV-1, were testedas well, and their sera showed titers consistently below 0.05 IU (not shown).

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249244

Fig. 2. Vaccine efficacy in pre-exposure and post-exposure regimens. (A) Dose escalation: Cynomologous monkeys were vaccinated once with various doses (109–1011 vp)of the AdC68rab.gp vector, a control AdC68 vector given at 1011 vp, or three doses of commercial human rabies vaccine, HDCV. Additional control animals received PBS only.Animals were challenged with canine rabies virus at 18 weeks after AdC68 vaccination. Due to the small number of animals differences between the vaccine groups andcontrol animal were not significant (p¼0.082 by Mantel Cox test). Combining all AdC68rab.gp-vaccinated animals resulted in a significant difference to controls (p¼0.0082).(B) Effect of pre-existing HAd-V5 neutralizing antibodies: Groups of monkeys were injected with 1012 vp of an AdHu5 vector expressing a reporter protein, or they were leftuntreated. Four weeks later, pre-treated, as well as untreated animals, were vaccinated with 1010 vp of AdC68rab.gp. A positive control group received commercial humanrabies vaccine, HDCV given 3 times, and a negative control group received PBS. Animals were challenged with canine rabies virus 18 weeks after AdC vaccination and survivalwas recorded. There was no significant difference between vaccinated animals that had or had not been pre-exposed. (C) Post-exposure prophylaxis: Nonhuman primateswere infected with canine rabies virus. Six hours later, one group was vaccinated with 1010 vp of AdC68rab.gp and received commercial HRIG infiltrated at the rabies virusinoculation site. An additional group was vaccinated with AdC68rab.gp, but without HRIG. A positive control group was treated with the traditional vaccine regimen, i.e.,HRIG combined with 4 doses of commercial PCEC human rabies vaccine, the latter given on days 1, 8, 15 and 29 relative to challenge. Additional control groups only receivedHRIG, a control AdC68 vector or PBS. Survival was recorded. There was no significant difference between controls (PBS and Ad control combined) and animals that onlyreceived AdC68rab.gp or AdC68rab.gp with RIG. Graphs (A)–(C) show percent of animals that survived without developing signs of rabies. Numbers of animals per group areshown within each graph above the bars. (D) shows VNA titers of the animals that are also shown in Fig. B. Data are shown as median titers at different time points7 IQR.Differences between VNA titers of the different cohorts were analyzed by 2-way Anova with Dunnett correction.

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249 245

to AdHu5 virus does not reduce the overall immunogenicity orefficacy of vaccines based on alternative Ad serotypes (McCoyet al., 2007; Roberts et al., 2006; Xiang et al., 2002).

Efficacy in post-exposure prophylaxis

In general, prevention of rabies occurs after exposure hasoccurred. Prophylaxis after severe exposures, which should com-mence as soon as possible, consists of careful wound cleaning,active immunization given repeatedly over a course of �2–48weeks, and a RIG preparation injected once into the site ofexposure (Manning et al., 2008; http://www.cdc.gov/rabies/medical_care/). To assess if AdC68rab.gp would induce protectiveimmunity in a post-exposure model, a total of 18 nonhumanprimates were infected with �106.1 MICLD50/ml of a virulentstreet rabies virus of canine origin. Six hours later, five animalswere vaccinated with 1010 vp of AdC68rab.gp. In addition, theseanimals received commercial HRIG injected close to the rabiesvirus challenge site, at a dose of 20 IU/kg. An additional fiveanimals were vaccinated with AdC68rab.gp, without HRIG. Twoanimals were administered a traditional PEP regimen, i.e., HRIGcombined with four doses of a commercial purified chick embryocell (PCEC) human rabies vaccine, the latter given on days 1, 8, 15and 29 relative to challenge. Two animals received HRIG only, twoanimals received a control AdC68 vector, and two animals weretreated with PBS only. Thereafter, both animals given PBS only, andone of two animals administered the AdC68 vector, displayedclinical signs of rabies, were sedated, euthanized, and diagnosedrabid. The combination of PCEC vaccine and HRIG providedcomplete protection, as did HRIG infiltration, without addition ofvaccine. The AdC68rab.gp given with, or without HRIG, onlyprovided partial protection in four or two of five animals, respec-tively. All surviving animals developed rabies VNA40.5 IU/mlwithin 7 days of PEP initiation, and all had titers 42.8 IU/ml byconclusion of the study at 4 months. These data, while based onsmall groups of animals, suggest that AdC68rab.gp, althoughhighly efficacious in a preventative setting, may not reliablyprotect if given after exposure to rabies virus (Fig. 2C).

Longevity of AdC68rab.gp induced immune responses

Lastly, a longer term pre-exposure vaccination experiment wasconducted based upon the above preliminary results. On day 0, six

animals were vaccinated IM with 1 ml containing 109 vps of theAdC68rab.gp vector. All vaccinated animals developed rabies VNAwithin 30 days of immunization (Fig. 3A). Titers were remarkablystable for at least 18 months. By 21 months, after vaccination, fouranimals still had titers above 10 IU, and the other two animals hadtiters of �1 IU. The AdC68rab.gp-immunized animals, together with agroup of five control animals, were challenged with �106.3 MICLD50/ml of a virulent street rabies virus of canine origin, at 22 months aftervaccination. All control animals displayed signs of rabies, but allvaccinated animals survived. No rabies virus antigens were detectedin the brains of the vaccinated animals at the conclusion of the study,in contrast to control animals (Fig. 3B). Vaccinated animals demon-strated a very robust anamnestic response within 7 days post-challenge that at its peak showed on average 240-fold increases inVNA titers compared to those measured pre-challenge. Althoughrabies VNA responses contracted rapidly after challenge, theyremained elevated (�8 fold) for the duration of the study (Fig. 3C).

Discussion

Globally, more than three quarters of the human population areassumed to be at risk for rabies virus exposure (World HealthOrganization, 2013). The five billion people living in Asia andAfrica are at high risk of exposure to rabid dogs, which accountsfor the majority of all human rabies-related fatalities (Knobel et al.,2005). More than 15 million humans receive PEP each year, withan estimated prevention of more than 300,000 deaths (WorldHealth Organization, 2013). In South East Asia, approximately20 million people are bitten each year by dogs. Of those bitten,more than 1.5 million receive PEP. Significantly, PEP is given to�40% of children by the time they reach 16 years of age in Asiaand Africa (Meslin and Briggs, 2013). Based upon such alarmingstatistics, rabies vaccine may be considered preventatively as partof childhood immunization programs in countries with a highincidence of rabies, given as a pre-exposure series (Ravish et al.,2013). In case of subsequent exposure, follow-up with woundcleaning and a booster immunization would be recommended(World Health Organization, 2013). Immunological memoryprovided by modern rabies vaccines is long lasting, on theorder of decades or more (Malerczyk et al., 2007; Suwansrinonet al., 2006). Recall responses develop rapidly, so that RIG isunnecessary.

Fig. 3. Vaccine immunogenicity and efficacy in a long-term study. (A) VNA titers after immunization: Groups of 6 animals were vaccinated with 109 vp of AdC68rab.gp.The rabies VNA titers were tested periodically from sera for a period of 21 months. The graph shows rabies VNA titers of individual animals. Differences between pre- andpost-vaccination samples assigning a titer of 0.06 IU, the limit of detection, to pre-vaccination samples showed significant differences for post-vaccination samples collectedfrom months 2 to month 17 by the multiple t-test with Sidak Holms correction (mo 2 p¼0.0001, all other time points po0.0001). (B) Vaccine efficacy: The vaccinatedanimals shown in (A) were challenged with canine rabies virus at 22 months after immunization. An additional 5 unvaccinated animals were challenged as well. Survival wasrecorded. Difference in survival was significant by the Mantel–Cox test (p¼0.0005). (C) VNA titers after challenge: Sera from the vaccinated animals in (A) were tested theday of challenge and then at the indicated time points after challenge for VNA titers to rabies virus. The graph shows rabies VNA titers of individual animals. The followinganimals showed a significant increase in titers after challenge, as tested for by two-way ANOVA with Sidak correction for multiple comparisons: A38428 (p¼0.032), A28100(p¼0.0011), A28112 (p¼0.0012), A28111 (p¼0.0012), A28114 (po0.0001).

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249246

As an example, a cost analysis was conducted for differentregimens on preventative vaccination vs. traditional PEP in Thaichildren (Chulasugandha et al., 2006). Using the least expensivevaccine regimens (intradermal vaccinationþERIG) it was estimatedthat the cost for preventative vaccination followed by boosterimmunizations would equal the cost of PEP with a dog biteprevalence of 7%. Although the actual dog bite rate is not knownfor Thailand, about 30% of children have been bitten by a dog by age15, suggesting an annual incidence rate of �2.3 bites per 100,000.The authors concluded that preventative childhood vaccination wasnot cost-effective for Thai children (Chulasugandha et al., 2006). Onemay assume that this type of cost analyses based on availablevaccines would be similar for other countries with a high incidenceof rabies.

By comparison, the AdC68rab.gp described in this manuscriptappears more immunogenic compared to licensed rabies vaccines byreliably inducing high titers of VNA after a single immunizationwithout the addition of adjuvant. The vaccine also induces sustainedresponses, which as shown here in nonhuman primates, remain atadequate levels for over a year. Equally important, AdC68rab.gpinduces memory B cells, which upon rabies virus challenge mounta rapid and potent recall response within a few days. This wasachieved with a very modest dose of 109 vp of vector, which is wellbelow the dose that is typically used for Ad vectors undergoingclinical testing as vaccines for other infectious diseases, such ashuman immunodeficiency virus-1 (HIV-1). A single dose regimen,compared to a three-dose regimen, which is administered over a 3–4week period, would be more convenient, should produce betterpatient compliance, and by inference, most likely would also be morecost-effective, especially if combined with other childhood vaccines.Oral immunization may be advantageous, not only for free-ranginganimals but also for mass vaccination of children. We had shownpreviously that mice immunized orally with AdC68rab.gp mountedrabies virus-specific VNA and were protected against challenge (Zhouand Ertl, 2006). Nevertheless, oral AdC68rab.gp immunization of non-human primates failed to elicit adequate levels of rabies VNA (datanot shown), suggesting that, as shown by others (Mercier et al., 2007),Ad vectors may need to be encapsidated to retain immunogenicityupon targeting to the gastrointestinal tract of larger mammals.

Currently, Ad vectors are being explored as vaccine vectors forother infectious diseases to which protection is linked to induction ofcellular responses, such as those caused by HIV-1 (Baden et al., 2012;McElrath et al., 2008), hepatitis C virus (Barnes et al., 2012),Plasmodium falciparum (Sheehy et al., 2012) or Mycobacterium tuber-culosis (Hoft et al., 2012). A surrogate of adequate immunizationagainst rabies virus has been well defined (Johnson et al., 2010).Rabies VNAs at titers at or above 0.5 IU are deemed adequate asevidence of a proper immunological response to prevent a productiveinfection. Presumably, CD4þ T cells are needed to promote B cellresponses, but otherwise cell-mediated immunity does not contributeprominently to the prevention of virus acquisition. Results shownhere demonstrate that Ad vectors induce very potent and sustainedantibody responses and are thus suited as vaccine carriers forpathogens that are best prevented by humoral responses. These dataconfirm previous studies in laboratory rodents, where we comparedthe AdHu5rab.gp vector to a DNA vaccine and a vaccinia-rabiesglycoprotein (V-RG) recombinant vaccine, both expressing the sametransgene product (Xiang et al., 1999, 1996). The VRG vaccine islicensed for wildlife immunization, and as such has shown efficacy ina number of species such as raccoons, coyotes, and foxes (Brochieret al., 1996; Weyer et al., 2009). By comparison, Ad-based vectors haveshowed superior induction of rabies VNAs, as compared to the othertwo vaccine platforms.

When used during PEP, the AdC68rab.gp vaccine did not inducereliable protection. All non-lyssavirus recombinant vaccines,including Ad vectors, V-RG, etc., first have to synthesize the

transgene product before an immune response is elicited. In caseof PEP, rapid induction of an immune response is essential and thedelayed induction of responses may be detrimental. In our mind,such recombinant vaccines, including AdC68rb.gp, therefore maynot be ideally suitable in exploration for PEP applications.

Taken together, these data demonstrate the safety, immuno-genicity, and efficacy of the AdC68rab.gp vaccine in a non-humanprimate animal model. Such encouraging results should be sup-portive towards the near-term consideration of human clinicaltrials, particularly considering the low, single-dose delivery, with-out the need for adjuvants. Together with improved local publiceducation, enhanced de-centralized laboratory-based surveillance,mass canine rabies vaccination, and improved human prophylaxis,such modern biologics will have a substantive role to play in globalrabies prevention and control strategies (Lembo, 2012).

Materials and methods

Viral vector

Briefly, all Ad vectors were generated from recombinant viralmolecular clones, as described (Zhou et al., 2010). A number ofdifferent Ad vectors derived from human and simian serotypeswere tested, specifically vectors based on human serotype 5 (HAd-V5 vectors, referred to as AdHu5), and simian serotypes SAd-V23(vectors called AdC6), SAd-V24 (vectors called AdC7) and SAd-V25(vectors called AdC68). Of note, SAd-V24 and SAd-V25 are closelyrelated, and induce cross-neutralizing antibodies, while neutraliz-ing antibodies to SAd-V23 fail to neutralize SAd-V24 or SAd-V25.The viruses were deleted in the E1 region, and hence impaired intheir ability to replicate. An expression cassette containingthe rabies virus glycoprotein gene, under the control of a CMVpromoter, was placed into the E1 domain. Upon rescue andexpansion on HEK 293 cells, vectors were purified by CsCl gradientcentrifugations. Vectors were titrated for content of virus particles[vps] by spectrophotometry at 260 nm using the formula:OD260�dilution�1.1�1012. Content of infectious virus particleswas measured by nested RT-PCR with transgene or Ad (hexon)-specific primers on RNA isolated from HEK 293 cells infected for5–7 days with serial dilutions of the viral vector. Batches weretested for endotoxin using the Limulus Amebocyte Lysate (LAL)gel-clot method and a commercial kit. Genetic integrity andidentity were assessed by isolation of viral DNA. The recombinantDNA, in parallel with the original molecular clones and shuttleplasmids used for generating molecular clones, were digested witha set of restriction enzymes and analyzed by gel electrophoresis.Expression of the transgene product was confirmed by immuno-precipitation of lysates from transfected HeLa cells, and primaryvaccine immunogenicity was demonstrated in mice, as described(Xiang et al., 2002). The Ad vectors expressing green fluorescentprotein were used to measured antibodies to Ad vectors. Thesevectors have been described previously (Xiang et al., 2002).

Nonhuman primates

Female or male cynomologous monkeys (Macaca fasicularis) orChinese rhesus macaques (Macaca mulatta) were used in thestudies. The M. facicularis were 1–2 years of age with weightsranging from 1 to 6 kg. They were housed in the animal facility ofthe Centers for Disease Control and Prevention (CDC), Atlanta, GA.The M. mulatta were 2–3 years of age, housed in the NonhumanPrimate Facility of the Division of Medical Genetics of theUniversity of Pennsylvania, Philadelphia, PA. The animals werehoused in pairs, until the time of rabies virus infection. Prior tohandling, animals were sedated with ketamine hydrochloride

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249 247

(10 mg/kg) or Telazol (3–5 mg/kg) intramuscularly (i.m.). All vac-cines were given i.m. Blood samples were collected via venipunc-ture from a peripheral vein and placed in serum separator tubes.Typically, bleeding occurred once a week to once a month, with�6–7 ml collected per kg per month. Upon challenge with rabiesvirus, animals were observed multiple times per day. Any altera-tions in body mass, food consumption, or water intake weremonitored closely. Upon the demonstration of clinical signscompatible with rabies (e.g., paresis, cranial nerve deficits, etc.),animals were sedated, and euthanized by intravenous barbiturateoverdose. Research was approved by the Institutional Animal Careand Use Committees.

Virus neutralization assays

Serum was separated from clotted blot after low speed cen-trifugation. Rabies virus neutralizing antibodies (VNAs) wereassayed using the rapid fluorescent focus inhibition test (RFFIT),with CVS-11 virus propagated upon MNA cells, as described (Louieet al., 1975). As no ‘protective’ titers have been defined, compara-tive levels of rabies VNAs were defined arbitrarily via the RFFIT,using the World Health Organization recommendations of evi-dence, with a level of 0.5 IU/ml considered as a minimumadequate level of acceptable comparable induction, as compatiblewith standard human clinical trial criteria (World HealthOrganization, 2013). Neutralizing antibodies to Ad viruses weremeasured as described (Xiang et al., 2006). Animals with circulat-ing neutralizing antibody titers Z1:20 to the vaccine vectors werenot enrolled into the study.

Rabies virus challenge

Challenge viruses consisted of street rabies viruses of canineorigin, chosen based upon global public health relevance from theNew and Old Worlds, and preliminary non-human primate sus-ceptibility data, prepared as previously described (Franka et al.,2009; Rupprecht et al., 2005). Challenge virus stocks were main-tained at �80 C, and were diluted using sterile PBS/2% heat-inactivated equine serum or FBS. Typical rabies virus challengeconcentrations ranged from �105.2 to 106.4 mouse intracerebrallethal dose (MICLD)50/ml. Sedated animals were inoculated in themasseter muscles with 0.5 ml of canine rabies virus, with anapproximate lethal dose, estimated based upon prior titrations innaïve animals from other studies. Brain tissue was removed fromeuthanized animals. Rabies virus antigens were detected using thedirect fluorescent antibody test, as described (Reid et al., 1983).

Statistical analyses

Comparisons between two sets of samples were conducted byuse of the Student0s t-test, and comparisons between multiplesamples were conducted by use of a 2-way ANOVA. The Sidak-Holms correction was used to adjust for multiple comparisons.Differences in survival were determined by Mantel–Cox tests.Statistical analyses were conducted with GraphPad Prism,version 6.0b.

Acknowledgments

The authors thank staff in the CDC Rabies Program and AnimalResources Branch for their assistance. The views expressed in thiscommunication are those of the authors and do not necessarilyreflect the opinion of their host institutions. Use of commercialproduct names is for comparison purposes only and does not

constitute official endorsement. This work was supported in partby grants from NIH/NIAID 1R01AI055018 and 5R01AI037166.

References

Baden, L.R., Walsh, S.R., Seaman, M.S., Tucker, R.P., Krause, K.H., Patel, A., et al., 2012.First-in-human evaluation of the safety and immunogenicity of a recombinantadenovirus serotype 26 HIV-1 Env Vaccine (IPCAVD 001). J. Infectious Dis. 207,240–247.

Barnes, E., Folgori, A., Capone, S., Swadling, L., Aston, S., Kurioka, A., et al., 2012.Novel adenovirus-based vaccines induce broad and sustained T cell responsesto HCV in man. Sci. Transl. Med. 4. (115ra1–115ra1).

Brochier, B., Aubert, M.F., Pastoret, P.P., Masson, E., Schon, J., Lombard, M., et al.,1996. Field use of a vaccinia-rabies recombinant vaccine for the control ofsylvaticrabies in Europe and North America. Rev. Sci. Technol. 15, 647–670.

Chen, H., Xiang, Z.Q., Li, Y., Kurupati, R.K., Jia, B., Bian, A., et al., 2010. Adenovirus-based vaccines: comparison of vectors from three species of Adenoviridae.J. Virol. 84, 10522–10532.

Chulasugandha, P., Khawplod, P., Havanond, P., Wilde, H., 2006. Cost comparison ofrabies pre-exposure vaccination with post-exposure treatment in Thai children.Vaccine 24, 1478–1482.

Franka, R., Wu, X., Jackson, F.R., Velasco-Villa, A., Palmer, D.P., Henderson, H., et al.,2009. Rabies virus pathogenesis in relationship to intervention with inactivatedand attenuated rabies vaccines. Vaccine 27, 7149–7155.

Harro, C.D., Robertson, M.N., Lally, M.A., O’Neill, L.D., Edupuganti, S., Goepfert, P.A.,et al., 2009. Safety and immunogenicity of adenovirus-vectored near-consensusHIV type 1 clade B gag vaccines in healthy adults. AIDS Res. Hum. Retroviruses25, 103–114.

Hoft, D.F., Blazevic, A., Stanley, J., Landry, B., Sizemore, D., Kpamegan, E., et al., 2012.A recombinant adenovirus expressing immunodominant TB antigens cansignificantly enhance BCG-induced human immunity. Vaccine 30, 2098–2108.

Hutnick, N.A., Carnathan, D., Demers, K., Makedonas, G., Ertl, H.C.J., Betts, M.R.,2010. Adenovirus-specific human T cells are pervasive, polyfunctional, andcross-reactive. Vaccine 28, 1932–1941.

Jackson, A.C., 2009. Therapy of rabies encephalitis. Biomédica: revista del InstitutoNacional de Salud 29, 169–176.

Jentes, E.S., Blanton, J.D., Johnson, K.J., Petersen, B.W., Lamias, M.J., Robertson, K.,et al., 2013. The global availability of rabies immune globulin and rabies vaccinein clinics providing direct care to travelers. J. Travel Med. 20, 148–158.

Johnson, N., Cunningham, A.F., Fooks, A.R., 2010. The immune response to rabiesvirus infection and vaccination. Vaccine 28, 3896–3901.

Knobel, D.L., Cleaveland, S., Coleman, P.G., Fèvre, E.M., Meltzer, M.I., Miranda, M.E.G.,et al., 2005. Re-evaluating the burden of rabies in Africa and Asia. Bull. WorldHealth Organiz. 83, 360–368.

Lembo, T., 2012. On Behalf of the Partners for Rabies Prevention, 2012. Theblueprint for rabies prevention and control: a novel operational toolkit forrabies elimination. PLoS Negl. Trop. Dis. 6, e1388.

Louie, R.E., Dobkin, M.B., Meyer, P., Chin, B., Roby, R.E., Hammar, A.H., VJ, C., 1975.Measurement of rabies antibody: comparison of the mouse neutralization test(MNT) with the rapid fluorescent focus inhibition test (RFFIT). J. Biol. Stand. 3,365–373.

Malerczyk, C., Briggs, D.J., Dreesen, D.W., Banzhoff, A., 2007. Duration of immunity:an anamnestic response 14 years after rabies vaccination with purified chickembryo cell rabies vaccine. J. Travel Med., 14, http://dx.doi.org/10.1111/j.1708-8305.2006.00097.x.

Manning, S.E., Rupprecht, C.E., Fishbein, D., Hanlon, C.A., Lumlertdacha, B., Guerra,M., et al., 2008. Human Rabies Prevention—United States, 2008. MMWR.Morbidity and Mortality Weekly Report, 57, 1–28.

McCoy, K., Tatsis, N., Korioth-Schmitz, B., Lásaro, M.O., Hensley, S.E., Lin, S.W., et al.,2007. Effect of preexisting immunity to adenovirus human serotype 5 antigenson the immune responses of nonhuman primates to vaccine regimens based onhuman- or chimpanzee-derived adenovirus vectors. J. Virol. 81, 6594–6604.

McElrath, M.J., De Rosa, S.C., Moodie, Z., Dubey, S., Kierstead, L., Janes, H., et al.,2008. HIV-1 vaccine-induced immunity in the test-of-concept Step Study: acase–cohort analysis. Lancet 372, 1894–1905.

Mercier, G.T., Nehete, P.N., Passeri, M.F., Nehete, B.N., Weaver, E.A., Templeton, N.S.,et al., 2007. Oral immunization of rhesus macaques with adenoviral HIVvaccines using enteric-coated capsules. Vaccine 25, 8687–8701.

Meslin, F.X., Briggs, D.J., 2013. Antiviral research. Antiviral Res. 98, 291–296.Plotkin, S.A., 2010. Correlates of protection induced by vaccination. Clin. Vaccine

Immunol. 17, 1055–1065.Ravish, H.S., Srikanth, J., Narayana, D.H.A., Annadani, R., Vijayashankar, V., Undi, M.,

2013. Pre-exposure prophylaxis against rabies in children: Safety of purifiedchick embryo cell rabies vaccine (Vaxirab N) when administered by intradermalroute. Hum. Vaccines Immunother. 9, 1910–1913.

Reid, F.L., Hall, N.H., Smith, J.S., Baer, G.M., 1983. Increased immunofluorescentstaining of rabies-infected, formalin-fixed brain tissue after pepsin and trypsindigestion. J. Clin. Microbiol. 18, 968–971.

Roberts, D.M., Nanda, A., Havenga, M.J.E., Abbink, P., Lynch, D.M., Ewald, B.A., et al.,2006. Hexon-chimaeric adenovirus serotype 5 vectors circumvent pre-existinganti-vector immunity. Nature 441, 239–243.

Rupprecht, C.E., Briggs, D., Brown, C.M., Franka, R., Katz, S.L., Kerr, H.D., et al., 2009.Evidence for a 4-dose vaccine schedule for human rabies post-exposureprophylaxis in previously non-vaccinated individuals. Vaccine 27, 7141–7148.

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249248

Rupprecht, C.E., Hanlon, C.A., Blanton, J., Manangan, J., Morrill, P., Murphy, S., et al.,2005. Oral vaccination of dogs with recombinant rabies virus vaccines. VirusRes. 111, 101–105.

Sheehy, S.H., Duncan, C.J., Elias, S.C., Choudhary, P., Biswas, S., Halstead, F.D., et al.,2012. ChAd63-MVA–vectored blood-stage malaria vaccines targeting MSP1 andAMA1: assessment of efficacy against mosquito bite challenge in humans. Mol.Ther. 20, 2355–2368.

Suwansrinon, K., Wilde, H., Benjavongkulchai, M., Banjongkasaena, U., Lertjarutorn,S., Boonchang, S., et al., 2006. Survival of neutralizing antibody in previouslyrabies vaccinated subjects: a prospective study showing long lasting immunity.Vaccine 24, 3878–3880.

WHO Position Paper. Rabies Vaccines. (2007). Rabies Vaccines. WHO Position Paper.Weekly Epidemiological Record/Health Section of the Secretariat of the Leagueof Nations, 82, 425–435.

Weyer, J., Rupprecht, C.E., Nel, L.H., 2009. Poxvirus-vectored vaccines for rabies—areview. Vaccine 27, 7198–7201.

World Health Organization, 2013. WHO Expert Consultation on Rabies. SecondReport. World Health Organization Technical Report Series, 1–139.

Wu, X., Smith, T.G., Rupprecht, C.E., 2011. From brain passage to cell adaptation: theroad of human rabies vaccine development. Expert Rev. Vaccines 10, 1597–1608.

Xiang, Z.Q., Pasquini, S., Ertl, H.C., 1999. Induction of genital immunity by DNApriming and intranasal booster immunization with a replication-defectiveadenoviral recombinant. J. Immunol. (Baltimore, MD: 1950) 162, 6716–6723.

Xiang, Z.Q., Yang, Y., Wilson, J.M., Ertl, H.C., 1996. A replication-defective humanadenovirus recombinant serves as a highly efficacious vaccine carrier. Virology219, 220–227.

Xiang, Z., Gao, G., Reyes-Sandoval, A., Cohen, C.J., Li, Y., Bergelson, J.M., et al., 2002.Novel, chimpanzee serotype 68-based adenoviral vaccine carrier for inductionof antibodies to a transgene product. J. Virol. 76, 2667–2675.

Xiang, Z., Li, Y., Cun, A., Yang, W., Ellenberg, S., Switzer, W.M., et al., 2006.Chimpanzee adenovirus antibodies in humans, sub-Saharan Africa. Emerg.Infectious Dis. 12, 1596–1599.

Zhou, D., Ertl, H.C., 2006. Therapeutic potential of adenovirus as a vaccine vector forchronic virus infections. Expert Opin. Biol. Ther. 6, 63–72.

Zhou, D., Zhou, X., Bian, A., Li, H., Chen, H., Small, J.C., et al., 2010. An efficientmethod of directly cloning chimpanzee adenovirus as a vaccine vector. Nat.Protocols 5, 1775–1785.

Z.Q. Xiang et al. / Virology 450-451 (2014) 243–249 249