of January 29, 2018.This information is current as

-Herpesvirus-68γLatency-Deficient Recombinant Murine

and Vaccine-Mediated Protection by a Importance of Antibody in Virus Infection

Ren Sun, Ting-Ting Wu and Marcia A. BlackmanMichael L. Freeman, Claire E. Burkum, David L. Woodland,

http://www.jimmunol.org/content/188/3/1049doi: 10.4049/jimmunol.1102621December 2011;

2012; 188:1049-1056; Prepublished online 23J Immunol 

        average*  

4 weeks from acceptance to publicationSpeedy Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

?The JIWhy

Referenceshttp://www.jimmunol.org/content/188/3/1049.full#ref-list-1

, 28 of which you can access for free at: cites 47 articlesThis article

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2012 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on January 29, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Importance of Antibody in Virus Infection andVaccine-Mediated Protection by a Latency-DeficientRecombinant Murine g-Herpesvirus-68

Michael L. Freeman,* Claire E. Burkum,* David L. Woodland,* Ren Sun,† Ting-Ting Wu,†

and Marcia A. Blackman*

The human g-herpesviruses EBV and Kaposi’s sarcoma-associated herpesvirus establish lifelong latent infections, can reactivate

in immunocompromised individuals, and are associated with the development of malignancies. Murine g-herpesvirus-68 (gHV68),

a rodent pathogen related to EBV and Kaposi’s sarcoma-associated herpesvirus, provides an important model to dissect mech-

anisms of immune control and investigate vaccine strategies. Infection of mice with gHV68 elicits robust antiviral immunity, and

long-term protection from gHV68 reactivation requires both cellular and humoral immune responses. Vaccination of mice with

AC-replication and transcription activator (RTA), a highly lytic latency-null recombinant gHV68, results in complete protection

from wild-type gHV68 infection that lasts for at least 10 mo. In this report, we examine the immune correlates of AC-RTA–

mediated protection and show that sterilizing immunity requires both T cells and Ab. Importantly, Ab was also critical for

mitigating viral infection in the brain, and in the absence of Ab-mediated control, amplification of the AC-RTA virus in the brain

resulted in fatality. Our results highlight important considerations in the development of vaccination strategies based on live-

attenuated viruses. The Journal of Immunology, 2012, 188: 1049–1056.

The human g-herpesviruses EBV and Kaposi’s sarcoma-associated herpesvirus (KSHV) are widely disseminatedpathogens that establish lifelong latent infections requiring

constant immune surveillance. Latent infection is largely benign inthe majority of infected individuals, and it has been demonstratedthat g-herpesvirus latency can be beneficial to the host by en-hancing resistance to bacterial infections (1). However, the viruseshave oncogenic potential, and infection is associated with the de-velopment of malignancies in immunosuppressed individuals andspecific high-risk groups. For example, posttransplant lymphopro-liferative diseases are a well-known consequence of EBV in immu-nosuppressed transplant recipients, and both KSHVand EBV causea high rate of malignancy in AIDS patients. In addition, regions ofSoutheast Asia have a 20-fold higher than normal incidence ofEBV-associated nasopharyngeal carcinoma, and EBV is associatedwith a high frequency of Burkitt’s lymphoma in areas of holoen-demic malaria in equatorial Africa (2, 3). Thus, the development ofvaccines capable of preventing the establishment of g-herpesviruslatency will lower the risk of associated malignancies. Because of

the species specificity of the g-herpesviruses, EBV and KSHVcannot be directly used in a mouse model. However, the mouseg-herpesvirus-68 (gHV68) is highly homologous to EBV andKSHV, and mechanisms of immune control are similar (4–6).Importantly, gHV68 provides a natural in vivo infection model inwhich host immune evasion mechanisms are preserved.Many important principles have been established by vaccination

studies using the mouse gHV68model. Early vaccination strategiesthat focused on subunit and inactivated viral vaccines met withlimited success. For example, similar to results in human trials (7),strategies targeting gHV68 gp150 (the homolog of EBV gp350)were shown to lessen infectious mononucleosis-like symptoms,but failed to impact the establishment of long-term latency (8). Inaddition, dominant lytic and latent epitopes that have been iden-tified in the mouse model were used in targeted vaccination strat-egies. Although vaccination with well-characterized lytic T cellepitopes reduced the extent of lytic infection upon challenge, thisstrategy failed to prevent the establishment of latency and did notsubstantially lower the long-term latent viral load (9, 10). Thisis consistent with accumulating data showing that latency canbe established independent of the lytic infection (11–14), whichexplains why targeting the lytic phase of infection does not sig-nificantly impact the establishment of latency. Subunit vaccinesthat directly target the establishment of latency were also unsuc-cessful, as vaccination with a defined latency epitope yielded onlytransient protection (15, 16). Recent vaccination studies haveshown only partial protection with live attenuated mutant virusesincapable of replication or reactivation (14, 17–19). Because of theoncogenic nature of the g-herpesviruses, engineered viruses thatare incapable of establishing latency hold the greatest promise forsuccessful and safe vaccine candidates (20–22).We have recently generated a live-attenuated virus, termed AC-

replication and transcription activator (RTA), that fails to establishlatency and elicits strong and long-lasting protection against in-fection with wild-type (WT) virus (23). A dual strategy was used togenerate AC-RTA. First, the gHV68 genome was modified to allow

*Trudeau Institute, Saranac Lake, NY 12983; and †Department of Molecular andMedical Pharmacology, University of California at Los Angeles, Los Angeles, CA90095

Received for publication September 13, 2011. Accepted for publication November28, 2011.

This work was supported by National Institutes of Health Grants AI084327 (toM.L.F.), AI049823 (to D.L.W.), AI042927 and AI082919 (to M.A.B.), CA148250(to M.A.B. and R.S.), and DE019085 (to R.S.), and by funds from the TrudeauInstitute.

Address correspondence and reprint requests to Dr. Marcia A. Blackman, TrudeauInstitute, 154 Algonquin Avenue, Saranac Lake, NY 12983. E-mail address:mblackman@trudeauinstitute.org

Abbreviations used in this article: AID, activation-induced cytidine deaminase;AID2/2ms2/2, activation-induced cytidine deaminase/secretory m-chain double-knockout mice; gHV68, g-herpesvirus 68; i.n., intranasally; KSHV, Kaposi’ssarcoma-associated herpesvirus; mMT, B cell-deficient; p.i., postinfection; RTA, rep-lication and transcription activator; ms, secretory m-chain; WT, wild-type.

Copyright� 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102621

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

constitutive overexpression of the viral RTA, which drives the viruspreferentially into the lytic cycle. Second, viral sequences neces-sary for the establishment of latency and oncogenicity were de-leted, including ORF73 (a latency-associated nuclear Ag, LANA,homolog), ORF72 (v-cyclin), and M11 (a viral bcl-2 homolog).AC-RTA exhibits enhanced lytic replication demonstrated by largerplaque size in vitro, higher peak viral titers in vivo, and faster clear-ance from immunocompetent mice (23). We previously showed thatinfection of BALB/c mice with AC-RTA generated a strong lyticinfection in the lung that was efficiently cleared by the immunesystem. Although the virus failed to establish latency, strong pro-tective immunity was generated that prevented subsequent infectionwithWT virus (23); both virus-specific T cells and Abs were induced(23, 24). In the current studies, we examined the vaccination efficacyof AC-RTA in C57BL/6 mice and determined the immune correlatesof protection. Consistent with the finding that both humoral andcellular immunity are required to control latent gHV68 infection, wefound that both Ab and T cells contribute to protective immunityinduced by AC-RTA. Unexpectedly however, we found that Ab-deficient mice failed to control AC-RTA infection and died withhigh viral titers in the brain.

Materials and MethodsMice and virus infections

Female 6–12-wk-old C57BL/6, activation-induced cytidine deaminase(AID)2/2, secretory m-chain (ms)2/2, AID/ms double-knockout mice(AID2/2ms2/2), B6.129S2-Igh-6tm1Cgn (B cell-deficient [mMT]), andB6.129S-H2dlAb1-Ea (MHC class II2/2) mice were obtained from theTrudeau Institute animal facility and maintained in specific-pathogen-freeconditions. Mice were infected intranasally (i.n.) with 400 PFU or i.p.with 104 PFU gHV68 (strain WUMS) or AC-RTA (23). All animal studieswere approved by the Institutional Animal Care and Use Committee of theTrudeau Institute.

Plaque assays

Plaque assays were performed as described previously (25). In short,spleens, lungs, and brains were homogenized, serially diluted, and platedon NIH3T3 cell monolayers for 1.5 h. Homogenates were then removedand overlaid with 0.75% carboxymethylcellulose in media. Plates wereincubated for 6 d at 37˚C, 10% CO2. Cells were then fixed with methanol,stained with Giemsa violet stain, and plaques counted.

Flow cytometry

Splenocytes were treated with Fc Block, then stained with Ab against CD8a(BD Biosciences) and allophycocyanin-conjugated MHC class I tetramersspecific for gHV68 epitopes H-2Db/ORF6487–495 (AGPHNDMEI) or H-2Kb/ORF61524–531 (TSINFVKI) (obtained from the Trudeau Institute MolecularBiology Core Facility). Samples were collected on a BD FACSCanto IIcytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star).

ELISA

ELISAs for virus-specific IgG were performed as described previously (26).In short, plates were coated with gHV68 virions (0.5 mg/well), incubatedwith 4-fold serially diluted sera from experimental or control mice, andthen incubated with appropriate alkaline phosphatase-conjugated goat anti-mouse Ab (Southern Biotechnology Associates). Plates were then washedand incubated with p-nitrophenyl phosphate substrate. OD was then mea-sured at 405 nm; OD values higher than twice the naive serum sampleswere considered a positive reaction.

Serum transfers

Serum was harvested from whole blood collected from mice at 1 or 4 moafter gHV68 infection or AC-RTA vaccination. To test the protective ef-ficacy of immune serum, naive mice were treated i.p. with 250 ml serum1 d prior to infection with gHV68. Lungs were harvested at 3 and 7 dpostinfection (p.i.) and analyzed by plaque assays.

In vitro neutralizing Ab assay

The neutralizing activity of serum Ab was assessed in vitro as describedpreviously, with slight modifications (27). Briefly, heat-inactivated serum

samples were serially diluted 2-fold in media and mixed 1:1 with 103 PFU/ml gHV68 for 1.5 h at 37˚C, then overlaid on NIH3T3 cell monolayers forplaque assays.

In vivo T cell depletion with Ab

T cell subsets were depleted from naive or AC-RTA–vaccinated mice bytreatment with mAbs to CD4 (GK1.5), CD8 (2.43), or anti-Thy1.2 (30H12)(BioXCell). For each group, 250 mg Ab was administered i.p. every 2 to3 d beginning 2 d prior to gHV68 challenge. Lungs and spleens wereharvested at 7 d p.i. and analyzed by plaque assays.

Quantitative real-time PCR

DNAwas isolated from homogenized brain tissue using the DNeasy Minikit (Qiagen) and quantified with a Nanodrop ND-1000 spectrophotometer(Thermo Fisher Scientific). The copy number of the gHV68 ORF50 genein 200 ng DNA was determined by quantitative real-time PCR essen-tially as described (16), using a standard curve quantitation methodon an Applied Biosystems 7500 Real-Time PCR System (Applied Bio-systems), using Taqman Gene Expression master mix (Applied Bio-systems). Primers, probes, and reaction cycles used are as described inUsherwood et al. (16).

Statistical analysis

Data were analyzed using the Student t test, one-way ANOVA, or log-rank(Mantel–Cox) test for survival curves. All analyses were performed usingPrism 5 software (GraphPad).

ResultsAC-RTA confers long-lasting protection following vaccination

Our previous study examined protection from WT gHV68 chal-lenge in BALB/c mice up to 3 mo after vaccination with AC-RTA(23). To determine if AC-RTA vaccination induced long-lastingprotection in C57BL/6 mice, we i.n. vaccinated mice with AC-RTA, challenged them i.n. with WT gHV68 either 1 or 6 mopostvaccination, and examined viral titers in the lungs 7 d afterchallenge. As shown in Fig. 1A, AC-RTA vaccination of C57BL/6mice was highly protective against WT virus challenge, with pro-tection lasting for at least 6 mo.We next examined the cellular and humoral immunity elicited by

AC-RTA vaccination compared with WT virus infection (Fig. 1B).We have previously examined virus-specific CD8 T cell responsesin the spleen p.i. with AC-RTA and WT virus. As we previouslyreported, numbers of tetramer-positive cells for two differentepitopes, ORF6487/D

b (ORF6) and ORF61524/Kb (ORF61), were

comparable 12 d p.i. with either virus, and 1 mo p.i., ORF61-specific cells, but not ORF6-specific cells, were reduced in theabsence of latency amplification (24). In this study, we examinedtetramer levels 7 mo after WT infection or AC-RTA vaccination(Fig. 1B). Although ORF61-specific cells are normally maintainedlong-term at a higher level than ORF6-specific cells after WTvirus infection, in the absence of latency, both pools of epitope-specific T cells were maintained comparably. This shift in immu-nodominance is similar to a previous report using a differentlatency-deficient recombinant virus (28) and implicates latencyor reactivation from latency in the preferential maintenance ofORF61-specific T cells.We have previously shown that the levels of virus-specific IgG

elicited by WT virus infection of C57BL/6 mice are maintainedover the long-term (22 mo), but neutralizing titers progressivelydecline (29). Therefore, we examined virus-specific Ab responsesand found that both WT virus and AC-RTA vaccination elicitedrobust class-switched Ab responses that were sustained as long as6 mo p.i. as measured by ELISA, although AC-RTA vaccinationelicited lower Ab titers than WT gHV68 infection (Fig. 1C).Together, the data show that vaccination with AC-RTA provideslong-lasting protection from WT viral challenge in C57BL/6 miceand induces sustained cellular and humoral immunity.

1050 Ab IN IMMUNITY TO AND PROTECTION BY AVACCINE VIRUS

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Both virus-specific T cells and Ab contribute to protection

To determine the relative roles of antiviral CD4 and CD8 T cells inAC-RTA vaccine-mediated protection, mice were depleted of CD4T cells, CD8 T cells, or both by mAb administration 30 d aftervaccinationwithAC-RTA, beginning2 d prior toWTvirus challenge.Protectionwas assessed bymonitoring the presence ofWTgHV68 inthe lung 7 d after challenge (Fig. 2A). Depletion of T cells resulted in

reduced vaccine-mediated protection, with indications of a greaterrole for CD4 T cells than CD8 T cells (two out of five fully protectedafter anti-CD4 treatment versus four out of five fully protected afteranti-CD8 treatment), although there was variation between indi-vidual mice. However, even total T cell depletion did not completelyabrogate protection, suggesting a role for Ab. Challenge of micedepleted of T cells 4 or 8 mo after AC-RTA vaccination againresulted in partial loss of protection (Fig. 2B), consistent with a rolefor both T cells and Ab in long-term protection.Because Ab was playing a role in protection, we addressed the

contribution of antiviral Ab in two ways. First, we passivelytransferred WT- or AC-RTA–immune serum into naive mice, thenchallenged with WT virus (Fig. 3A). Sera from both WT- or AC-RTA–infected mice were protective compared with naive serum. Aprevious report has shown that the class-switched and neutralizingAb response to gHV68 rises rapidly the first 3 wk p.i. and continuesto increase over several months (27). Consistent with that report,we found that serum from WT-immune mice 4 mo p.i. providedgreater protection than serum from WT-immune mice 30 d p.i.,especially when protection was measured 7 d after challenge. AC-RTA–immune serum did not exhibit the same increase in protectiveefficacy over time after vaccination: AC-RTA–immune serum 4 mop.i. elicited significantly less protection 7 d after challenge thanWT-immune serum frommice 4 mo p.i. (p = 0.0415, Student t test),consistent with the lower overall antiviral IgG after AC-RTA vac-cination compared with WT gHV68 infection (Fig. 1C). Second,we compared the ability of serum obtained from WT- and AC-RTA–infected mice 4 mo p.i. to neutralize virus in an in vitro assay(Fig. 3B). At each dilution, serum fromWT-immune mice 4 mo p.i.had more neutralizing activity than serum from AC-RTA–vacci-nated mice, although both sera neutralized gHV68 more effectivelythan naive serum. The enhanced protective efficacy of serum frommice infected with WT virus raises the possibility that the presenceof latent virus drives maturation of the neutralizing response. Thepossibility that the Ab-specific response generated after vaccina-tion by vectors that do not induce latency is deficient may haveimportant implications for vaccination efficacy in cases of declin-ing T cell function.Taken together, our data show that T cells and Ab are both influ-

enced by the presence of latent virus. Importantly, each contributesto the long-lasting protection afforded by AC-RTA vaccination.

Ab-deficient mice succumb to AC-RTA vaccination

To further investigate the role for Ab in AC-RTAvaccine-mediatedprotection in vivo, we infected Ab-deficient AID2/2ms2/2 with

FIGURE 1. Long-lasting protection following AC-RTA vaccination. A,

The number of PFU per lung of AC-RTA– or mock-immunized mice 7 d

after challenge with WT gHV68. B, The number of tetramer-positive CD8

T cells in the spleen 7 mo after WT gHV68 infection or AC-RTA vacci-

nation. C, Reciprocal titers of virus-specific IgG Ab 6 mo after WT gHV68

infection or AC-RTA vaccination. *p # 0.05, **p # 0.01, ***p # 0.001,

Student t test.

FIGURE 2. Contribution of T cells to AC-RTA–mediated protection. A, The number of PFU per lung 7 d after WT gHV68 challenge of mock-vaccinated

mice, mice vaccinated 30 d prior with AC-RTA, or AC-RTA–vaccinated mice treated with anti-CD4, anti-CD8, or anti-Thy1.2 mAb prior to challenge. B,

The number of PFU per lung 7 d after WT gHV68 challenge of mock-vaccinated mice, mice vaccinated 4 or 8 mo prior with AC-RTA, or AC-RTA–

vaccinated mice treated anti-Thy1.2 mAb prior to challenge. **p # 0.01, ***p # 0.001, one-way ANOVA.

The Journal of Immunology 1051

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

AC-RTA (30, 31). These mice are deficient in class-switching,somatic hypermutation, and IgM secretion, thereby lacking cir-culating Abs while still retaining expression of Ig on the B cellsurface. These mice maintain a complete B cell compartment,with the exception of plasma cells (30). Because B cells are themajor reservoir of latent virus (32, 33), these mice are excellentmodels for assessing the importance of humoral immunity in AC-RTA–mediated protection. To this end, C57BL/6 and AID2/2ms2/2

mice were infected with WT and AC-RTA. Unexpectedly, at ∼3wk, the AID2/2ms2/2 mice succumbed to AC-RTA infection (Fig.4A). To determine if lethality was a characteristic unique to thisparticular strain of mouse, we infected two additional strains ofAb-deficient mice, MHC class II2/2 mice (Fig. 4B) and mMTmice (Fig. 4C). All three strains succumbed to AC-RTA infectionwith similar kinetics, confirming a requirement for Ab in survivalfollowing AC-RTA infection.Because AID2/2ms2/2 mice cannot undergo class switching or

secrete IgM, we were unsure whether circulating IgM could pro-long survival of Ab-deficient mice after AC-RTA infection. Weinfected AID2/2 mice, which lack class-switched Ab but can stillsecrete IgM (34), and ms2/2 mice, which cannot secrete IgM, butcan still generate and secrete other Ab classes (35), with AC-RTA.AID2/2 mice succumbed to AC-RTA infection with similar ki-netics as above (Fig. 5A), whereas ms2/2 mice survived infection(Fig. 5B). These data show that circulating IgM is not sufficient toallow survival of AC-RTA–infected mice lacking class-switchedAbs.AC-RTA is biased toward lytic rather than latent infection be-

cause it has an ectopic copy of RTA driven by a strong hybridpromoter (23). Consistent with enhanced lytic replication, AC-RTA generates larger plaques and enhanced viral titers in

multiple-step growth experiments in vitro (23). In addition, wehave previously shown that AC-RTA infection of C57BL/6 miceresults in higher and more rapid titers in the lung compared withWT infection (24). Therefore, we asked whether elevated AC-RTAtiters could account for the death of AID2/2ms2/2 mice. WhereasC57BL/6 mice clear both WT and AC-RTA virus by day 10 afteri.n. infection (24), analysis of viral clearance in the lungs ofAID2/2ms2/2 mice showed impaired clearance of both viruses inthe absence of Ab. Interestingly, AC-RTA infection was clearedwith faster kinetics than WT virus in AID2/2ms2/2 mice (Fig.6A), ruling out uncontrolled virus replication in the lungs as acause of death in the mice.Mice that succumbed to AC-RTA infection were moribund,

exhibiting severe weight loss, ataxia, and hind-limb paralysis,consistent with the mice undergoing encephalitis. Therefore, wemeasured the lytic viral load in the brains of C57BL/6 and AID2/2

ms2/2 mice following infection with WT gHV68 or AC-RTA (Fig.6B). Remarkably, high titers of lytic virus were found in all thebrains of AID2/2ms2/2 mice on days 15 and 21 post–AC-RTAinfection, whereas only one out of nine WT gHV68-infectedAID2/2ms2/2 mice had a detectable, but still very low, amountof virus. At no time did we detect lytic WT gHV68 in the brains ofC57BL/6 mice. In Ab-deficient mice, WT gHV68 infection pro-duced high levels of viral DNA in the brain, indicating that Abplays a major role in preventing viral spread to the brain, but thevirus in the brain is mostly maintained in a latent state (Fig. 6C).Regardless of Ab, AC-RTA infection led to significant levels ofviral DNA in the brain, consistent with the highly lytic nature of

FIGURE 3. Contribution of Ab to AC-RTA–mediated protection. A, The

number of PFU per lung 3 or 7 d after WT gHV68 challenge in mice

treated with sera from naive mice, mice 30 d or 4 mo after WT infection, or

mice 30 d or 4 m after AC-RTA vaccination. B, The percent of wells

positive for cytopathic effect (CPE) after in vitro infection with WT

gHV68 pretreated with naive serum or sera from mice 4 mo after WT

infection or AC-RTA vaccination (n = 2–5/group). Dashed line is at 60.3%,

the average percent of CPE when no serum was added. *p # 0.05, **p #

0.01, ***p # 0.001, one-way ANOVA.

FIGURE 4. Requirement for Ab during AC-RTA infection. A, Survival

of AID2/2ms2/2 mice after WT gHV68 or AC-RTA i.n. infection (n = 5–

14/group, p = 0.0013). B, Survival of MHC class II2/2 mice after WT

gHV68 or AC-RTA i.n. infection (n = 6–10/group, p = 0.0002). C, Survival

of mMT mice after WT gHV68 or AC-RTA infection (n = 15/group, p #

0.0001).

1052 Ab IN IMMUNITY TO AND PROTECTION BY AVACCINE VIRUS

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

AC-RTA; once AC-RTA reaches the brain, it undergoes lytic rep-lication that is mitigated by Ab. Taken together, in bothWT gHV68and AC-RTA infections, Ab is essential for controlling viral inva-sion of the brain, and the importance of this Ab-mediated control

is heightened by highly lytic AC-RTA infection, which causesAb-deficient mice to die.We next examined whether the infiltration and expansion of AC-

RTA in the brain of AID2/2ms2/2 mice was a consequence of theroute of inoculation. It has been shown that following i.n. admin-istered AC-RTA that a high concentration of virus can be detectedat the site of inoculation (23), raising the possibility that this fa-cilitates localization to and infection of the brain. To address this,we i.p. infected AID2/2ms2/2 mice (Fig. 7A) or MHC class II2/2

mice (Fig. 7B) with AC-RTA andmeasured survival. AID2/2ms2/2

mice infected with AC-RTA i.p. succumbed to infection with fasterkinetics than i.n. infected mice (compare Fig. 7A with Fig. 4A;median survival: i.p. 15 d, i.n. 21 d; p = 0.0139), whereas MHCclass II2/2mice succumbed with similar kinetics (compare Fig. 7Bwith Fig. 4B; median survival: i.p. 25 d, i.n. 23.5 d; p = 0.6177). Ineither case, AC-RTA resulted in death in both strains followingi.p. infection, conclusively demonstrating that AC-RTA is lethalto Ab-deficient mice regardless of the route of infection.We next measured lytic viral titers in the lungs, spleens, and

brains of AID2/2ms2/2 mice following i.n. or i.p. infection withAC-RTA to determine if i.p. infection resulted in increased viralloads in the brain (Fig. 7C). Interestingly, i.p. infection led toaccelerated dissemination to the spleen compared with i.n. in-fection at 7 d p.i., but lytic virus could not be detected in the brainuntil 14 d p.i. The reasons for the delayed kinetics of virus in thebrain compared with the spleen remain unclear.Taken together, our data emphasize the importance of Ab

responses in a recombinant viral vaccine strategy. In immuno-competent animals, AC-RTA vaccination leads to long-lasting

FIGURE 6. Virus in the brain after AC-RTA infection of Ab-deficient mice. A, The number of PFU per lung of AID2/2ms2/2 mice at the indicated times

after WT gHV68 or AC-RTA infection. B, The number of PFU per brain of C57BL/6 or AID2/2ms2/2 mice at the indicated times after WT gHV68 or

AC-RTA infection. C, The number of viral ORF50 gene copies per 200 ng DNA from brains of C57BL/6 or AID2/2ms2/2 mice 21 d after WT gHV68 or

AC-RTA infection. *p # 0.05, **p # 0.01, Student t test and ***p # 0.001, one-way ANOVA.

FIGURE 5. Class-switched Ab is required during AC-RTA infection. A,

Survival of AID2/2 mice after WT gHV68 or AC-RTA i.n. infection (n = 4

to 5/group, p = 0.0249). B, Survival of ms2/2 mice after WT gHV68 or

AC-RTA i.n. infection (n = 5 to 6/group).

The Journal of Immunology 1053

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

protection from viral challenge that is dependent on both cellularand humoral immunity. In Ab-deficient animals, however, thehighly lytic AC-RTA infection results in the infiltration and rep-lication of the virus in the brain, correlating with encephalitis-likesymptoms, morbidity, and death.

DiscussionThis study has examined the immune mechanisms by which AC-RTA, a latency-deficient gHV68, induces long-lasting protectiveimmunity against infection with WT virus. The data show thatboth T cells and Ab contribute to protection. Unexpectedly, wefound that not only did Ab make an important contribution toimmune protection after vaccination, but Ab was also essentialfor controlling the viral spread to the brain. This importance ishighlighted by the lethality of highly lytic AC-RTA infection inAb-deficient mice, which correlated with the presence of highviral titers in the brain. This finding was unexpected because i.n.infection of mice with WT gHV68 leads to a robust acute infec-tion in the lungs followed by the establishment of viral latency indendritic cells, macrophages, and B cells in the spleen and else-where (32), but infection with gHV68 is not normally neuro-invasive. gHV68 can infect the CNS if inoculated intracerebrally,however (36, 37). Additionally, i.n. inoculation with WT virus can

lead to infection in the brains of immunocompromised mice, in-cluding neonatal BALB/c mice (38) and adult mice deficient intype I IFN receptor (36). These observations are also consistentwith a study using an HSV vaccination model in which it wasshown that Ab and T cells elicited by a mutant replication-deficient vaccinating virus contributed to protection of the hostat distinct stages of challenge infection, with Ab preventing en-cephalitic infection of the CNS (39). Therefore, our studies vali-date the importance of the mouse model for examining vaccinestrategies and understanding correlates of immune protection andraise a caution that live vaccines that are attenuated and protectivein an immunocompetent host may have unanticipated virulencein immunocompromised individuals.It has long been known that Ab participates in immunity to

gHV68 (17, 26, 40). Two separate questions are raised by thesestudies: what is the role of Ab in protection elicited by vaccinationwith AC-RTA, a latency-deficient virus, and what is the role of Abin control of the initial virus infection?The observation that T cell depletion lowered the protective

efficacy following vaccination is similar to previous observationsusing a different live attenuated virus that was capable of estab-lishing latency but was incapable of reactivation. In agreementwith our findings, it was shown that protection could be partiallyreconstituted by passive transfer of Ab (17). These earlier studiesalso suggested that Ab-mediated protection was not mediated bythe simple blocking of the initial infection because vaccinationwas effective against high challenge doses (17). This conclusionis consistent with our data showing that passively transferred serahad only modest protection in terms of viral titers at day 3 (andthus did not prevent initial infection), but showed greater than onelog reduced viral titers at day 7 (Fig. 3A).Our data show that passive transfer of serum from WT mice

obtained 4 mo p.i. is more protective than Ab obtained from mice30 d p.i. (Fig. 3A). In contrast, serum from AC-RTA–vaccinatedmice is equally protective over time, consistent with the possibilitythat Ab elicited by AC-RTA infection failed to undergo some as-pects of Ab maturation. It is possible that differences in protectiveefficacy reflect differences in neutralizing activity, as assessed in anin vitro assay (Fig. 3B). However, the contribution of virus neu-tralizing Ab in protection is poorly understood. The observationthat the protective efficacy of serum matures over time in WTgHV68-infected but not vaccinated mice raises the interestingpossibility that latency or reactivation is necessary to drive thisunknown aspect of Ab maturation, as AC-RTA vaccination fails toestablish latency. The differences in protective Ab are not simplya consequence of class switching, as our analysis failed to revealany statistically significant differences in isotype profile as late as6 mo p.i. (data not shown). Differences in protection may reflecta change in specificity of virus-specific Abs associated with dif-ferent phases of infection, as complex serological patterns of Absspecific for different viral proteins have been described during EBVlatency, following reactivation, and associated with tumors (41).It has been reported that human g-herpesviruses can infect the

CNS. Acute or reactivating infection with EBV in both immuno-compromised as well as immunocompetent individuals has beenlinked to complications such as meningitis and encephalitis (42,43). In addition, it has been shown that KSHV can be neuro-invasive (44). Neurological complications of KSHV infection areless well defined, but detection of virus in dorsal root ganglia,cerebrospinal fluid, and brain in some patients with CNS com-plications associated with Kaposi’s sarcoma or AIDS-associateddementia have been described, although the relative contributionsof HIV and KSHV to the pathology is controversial (43, 44). Itis unclear whether pathogenesis is a consequence of direct virus

FIGURE 7. AC-RTA is lethal in Ab-deficient mice by i.p. infection

route. A, Survival of AID2/2ms2/2 mice after WT gHV68 or AC-RTA i.p.

infection (n = 3–5/group, p = 0.0453). B, Survival of MHC class II2/2

mice after WT gHV68 or AC-RTA i.p. infection (n = 3–5/group, p =

0.0136). C, The number of PFU per lung, spleen, or brain in AID2/2ms2/2

mice after i.n. or i.p. infection with AC-RTA.

1054 Ab IN IMMUNITY TO AND PROTECTION BY AVACCINE VIRUS

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

invasion of the CNS or whether damage is due to immunologicaleffector mechanisms mediated either by infiltrating CD8 T cells,or deposition of Ag–Ab complexes (43).Our data show that both the absence of Ab and the enhanced lytic

activity of AC-RTA contribute to high viral titers in the brainfollowing AC-RTA infection. Analysis of infectious virus and viralgenome copy number in the brain has allowed us to distinguish therelative roles of enhanced lytic activity of AC-RTA and the absenceof Ab in the inability of the Ab-deficient mice to control AC-RTAinfection within the brain. Whereas AC-RTA infection of both Ab-deficient and WT mice resulted in lytic virus in the brain, the titersrose dramatically with time in the Ab-deficient mice and the micesuccumbed to infection, but titers stabilized in theWTmicewith nolethality. In contrast, no lytic virus was detected in the brains ofeither WT or Ab-deficient mice p.i. with WT gHV68, but analysisof genome copy number was consistent with the presence of latentvirus in the brain of WT gHV68-infected Ab-deficient mice.Taken together, these data show that both WT gHV68 and AC-RTA can infect the brain in the absence of Ab and that the Ab isessential to control viral infection of the brain. However, due tothe highly lytic replication of AC-RTA, which contains an ectopiccopy of RTA under the control of a strong and constitutive hybridpromoter to drive the viral lytic cycle, an unchecked AC-RTAinfection in the brain of Ab-deficient mice leads to death. Howmight Ab prevent the robust replication of AC-RTA in the brainsof WT mice? One intriguing possibility for Ab control of AC-RTAreplication in the brain is through the activity of complementfactor C3, as it has been shown that C3 can mediate gHV68replication during direct CNS infections (45). Alternatively, Ab-neutralizing activity could prevent the spread of lytic virus withinthe brain, possibly by FcR-mediated engulfment by phagocytes.How does the virus get to the brain? In the absence of direct

infection of the CNS, virus could be transported to the brain viacell-free virus in plasma or in a cell-associated manner. Our datashow that both i.n. and i.p. infections lead to dissemination of virusto the brain by 14 d p.i., whereas the lytic titers in the brain are 3logs higher following i.p. infection. Intriguingly, in vitro AC-RTAinfection leads to the overexpression of the gene that encodes theviral regulator of complement activation protein (23, 46), whichhas recently been shown to enhance virus replication withinmacrophages (47). Macrophages can acquire virus, especially af-ter i.p. infection (48), suggesting infected macrophages may carryvirus to the brain. Enhanced macrophage infection and traffickingof AC-RTA could also explain the higher titers of lytic virus in thebrain after i.p. infection compared with i.n. infection. Ab couldplay a role in preventing this trafficking, but this Ab-dependentprevention might be overcome by the overexpression of regulatorof complement activation after AC-RTA infection.Together, these studies have provided important insight into

mechanisms of g-herpesvirus immune control and immune cor-relates of protection elicited by an effective vaccine strategy. BothT cells and Ab contribute to effective protection elicited bya highly lytic latency-deficient virus, AC-RTA. Importantly, ourstudy reveals a previously unrecognized role of Abs in controllingviral infection of the brain. However, whereas Abs prevent WTvirus from accessing the brain, AC-RTA can readily access thebrain regardless of the presence of Abs. AC-RTA was initiallyconstructed to be a nonpersistent virus by replacing the latencylocus with an ectopic RTA expression cassette under the control ofa strong promoter as a vaccine strategy for g-herpesviruses. Theunanticipated complications associated with AC-RTA infection inAb-deficient mice underscore the importance of the mouse in vivoinfection model for studying vaccination with live-attenuated mu-tant viruses.

AcknowledgmentsWe thank Meghan K. Jensen for excellent technical assistance and Drs.

Jacob E. Kohlmeier and William W. Reiley for critical reading of the

manuscript.

DisclosuresThe authors have no financial conflicts of interest.

References1. Barton, E. S., D. W. White, J. S. Cathelyn, K. A. Brett-McClellan, M. Engle,

M. S. Diamond, V. L. Miller, and H. W. Virgin, IV. 2007. Herpesvirus latencyconfers symbiotic protection from bacterial infection. Nature 447: 326–329.

2. Wu, T. T., M. A. Blackman, and R. Sun. 2010. Prospects of a novel vaccinationstrategy for human gamma-herpesviruses. Immunol. Res. 48: 122–146.

3. Rickinson, A. B., and E. Kieff. 2001. Epstein-Barr Virus. In Fields Virology.P. M. Howley, and D. M. Knipe, eds. Lippincott Williams and Wilkins, Phila-delphia, p. 2575–2627.

4. Virgin, H.W., IV, P. Latreille, P.Wamsley, K. Hallsworth, K. E.Weck, A. J. Dal Canto,and S. H. Speck. 1997. Complete sequence and genomic analysis of murinegammaherpesvirus 68. J. Virol. 71: 5894–5904.

5. Nash, A. A., B. M. Dutia, J. P. Stewart, and A. J. Davison. 2001. Natural historyof murine gamma-herpesvirus infection. Philos. Trans. R. Soc. Lond. B Biol. Sci.356: 569–579.

6. Doherty, P. C., J. P. Christensen, G. T. Belz, P. G. Stevenson, and M. Y. Sangster.2001. Dissecting the host response to a gamma-herpesvirus. Philos. Trans. R.Soc. Lond. B Biol. Sci. 356: 581–593.

7. Sokal, E. M., K. Hoppenbrouwers, C. Vandermeulen, M. Moutschen, P. Leonard,A. Moreels, M. Haumont, A. Bollen, F. Smets, and M. Denis. 2007. Recombi-nant gp350 vaccine for infectious mononucleosis: a phase 2, randomized,double-blind, placebo-controlled trial to evaluate the safety, immunogenicity,and efficacy of an Epstein-Barr virus vaccine in healthy young adults. J. Infect.Dis. 196: 1749–1753.

8. Stewart, J. P., N. Micali, E. J. Usherwood, L. Bonina, and A. A. Nash. 1999.Murine gamma-herpesvirus 68 glycoprotein 150 protects against virus-inducedmononucleosis: a model system for gamma-herpesvirus vaccination. Vaccine 17:152–157.

9. Liu, L., E. Flano, E. J. Usherwood, S. Surman, M. A. Blackman, andD. L. Woodland. 1999. Lytic cycle T cell epitopes are expressed in two distinctphases during MHV-68 infection. J. Immunol. 163: 868–874.

10. Stevenson, P. G., G. T. Belz, M. R. Castrucci, J. D. Altman, and P. C. Doherty.1999. A gamma-herpesvirus sneaks through a CD8(+) T cell response primed toa lytic-phase epitope. Proc. Natl. Acad. Sci. USA 96: 9281–9286.

11. Tibbetts, S. A., J. Loh, V. Van Berkel, J. S. McClellan, M. A. Jacoby,S. B. Kapadia, S. H. Speck, and H. W. Virgin, IV. 2003. Establishment andmaintenance of gammaherpesvirus latency are independent of infective dose androute of infection. J. Virol. 77: 7696–7701.

12. Coleman, H. M., B. de Lima, V. Morton, and P. G. Stevenson. 2003. Murinegammaherpesvirus 68 lacking thymidine kinase shows severe attenuation of lyticcycle replication in vivo but still establishes latency. J. Virol. 77: 2410–2417.

13. Flano, E., Q. Jia, J. Moore, D. L. Woodland, R. Sun, and M. A. Blackman. 2005.Early establishment of gamma-herpesvirus latency: implications for immunecontrol. J. Immunol. 174: 4972–4978.

14. Moser, J. M., M. L. Farrell, L. T. Krug, J. W. Upton, and S. H. Speck. 2006. Agammaherpesvirus 68 gene 50 null mutant establishes long-term latency in thelung but fails to vaccinate against a wild-type virus challenge. J. Virol. 80: 1592–1598.

15. Usherwood, E. J., D. J. Roy, K. Ward, S. L. Surman, B. M. Dutia,M. A. Blackman, J. P. Stewart, and D. L. Woodland. 2000. Control of gamma-herpesvirus latency by latent antigen-specific CD8(+) T cells. J. Exp. Med. 192:943–952.

16. Usherwood, E. J., K. A. Ward, M. A. Blackman, J. P. Stewart, andD. L. Woodland. 2001. Latent antigen vaccination in a model gammaherpesvirusinfection. J. Virol. 75: 8283–8288.

17. Tibbetts, S. A., J. S. McClellan, S. Gangappa, S. H. Speck, and H. W. Virgin, IV.2003. Effective vaccination against long-term gammaherpesvirus latency.J. Virol. 77: 2522–2529.

18. Kayhan, B., E. J. Yager, K. Lanzer, T. Cookenham, Q. Jia, T.-T. Wu,D. L. Woodland, R. Sun, and M. A. Blackman. 2007. A replication-deficientmurine gamma-herpesvirus blocked in late viral gene expression can establishlatency and elicit protective cellular immunity. J. Immunol. 179: 8392–8402.

19. Tibbetts, S. A., F. Suarez, A. L. Steed, J. A. Simmons, and H. W. Virgin, IV. 2006.A gamma-herpesvirus deficient in replication establishes chronic infection in vivoand is impervious to restriction by adaptive immune cells. Virology 353: 210–219.

20. Boname, J. M., H. M. Coleman, J. S. May, and P. G. Stevenson. 2004. Protectionagainst wild-type murine gammaherpesvirus-68 latency by a latency-deficientmutant. J. Gen. Virol. 85: 131–135.

21. Fowler, P., and S. Efstathiou. 2004. Vaccine potential of a murinegammaherpesvirus-68 mutant deficient for ORF73. J. Gen. Virol. 85: 609–613.

22. Rickabaugh, T. M., H. J. Brown, D. Martinez-Guzman, T. T. Wu, L. Tong, F. Yu,S. Cole, and R. Sun. 2004. Generation of a latency-deficient gammaherpesvirusthat is protective against secondary infection. J. Virol. 78: 9215–9223.

23. Jia, Q., M. L. Freeman, E. J. Yager, I. McHardy, L. Tong, D. Martinez-Guzman,T. Rickabaugh, S. Hwang, M. A. Blackman, R. Sun, and T. T. Wu. 2010. In-

The Journal of Immunology 1055

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

duction of protective immunity against murine gammaherpesvirus 68 infectionin the absence of viral latency. J. Virol. 84: 2453–2465.

24. Freeman, M. L., K. G. Lanzer, T. Cookenham, B. Peters, J. Sidney, T. T. Wu,R. Sun, D. L. Woodland, A. Sette, and M. A. Blackman. 2010. Two kineticpatterns of epitope-specific CD8 T-cell responses following murine gamma-herpesvirus 68 infection. J. Virol. 84: 2881–2892.

25. Cardin, R. D., J. W. Brooks, S. R. Sarawar, and P. C. Doherty. 1996. Progressiveloss of CD8+ T cell-mediated control of a gamma-herpesvirus in the absence ofCD4+ T cells. J. Exp. Med. 184: 863–871.

26. Kim, I. J., E. Flano, D. L. Woodland, and M. A. Blackman. 2002. Antibody-mediated control of persistent gamma-herpesvirus infection. J. Immunol. 168:3958–3964.

27. Stevenson, P. G., and P. C. Doherty. 1998. Kinetic analysis of the specific hostresponse to a murine gammaherpesvirus. J. Virol. 72: 943–949.

28. Obar, J. J., S. Fuse, E. K. Leung, S. C. Bellfy, and E. J. Usherwood. 2006.Gammaherpesvirus persistence alters key CD8 T-cell memory characteristicsand enhances antiviral protection. J. Virol. 80: 8303–8315.

29. Yager, E. J., I. J. Kim, M. L. Freeman, K. G. Lanzer, C. E. Burkum,T. Cookenham, D. L. Woodland, and M. A. Blackman. 2010. Differential impactof ageing on cellular and humoral immunity to a persistent murine gamma-herpesvirus. Immun. Ageing 7: 3.

30. Kumazaki, K., B. Tirosh, R. Maehr, M. Boes, T. Honjo, and H. L. Ploegh. 2007.AID-/-mus-/- mice are agammaglobulinemic and fail to maintain B220-CD138+plasma cells. J. Immunol. 178: 2192–2203.

31. Carragher, D. M., D. A. Kaminski, A. Moquin, L. Hartson, and T. D. Randall.2008. A novel role for non-neutralizing antibodies against nucleoprotein in fa-cilitating resistance to influenza virus. J. Immunol. 181: 4168–4176.

32. Flano, E., S. M. Husain, J. T. Sample, D. L. Woodland, and M. A. Blackman.2000. Latent murine gamma-herpesvirus infection is established in activatedB cells, dendritic cells, and macrophages. J. Immunol. 165: 1074–1081.

33. Flano, E., I. J. Kim, D. L. Woodland, and M. A. Blackman. 2002. Gamma-herpesvirus latency is preferentially maintained in splenic germinal center andmemory B cells. J. Exp. Med. 196: 1363–1372.

34. Muramatsu, M., K. Kinoshita, S. Fagarasan, S. Yamada, Y. Shinkai, andT. Honjo. 2000. Class switch recombination and hypermutation requireactivation-induced cytidine deaminase (AID), a potential RNA editing enzyme.Cell 102: 553–563.

35. Boes, M., C. Esau, M. B. Fischer, T. Schmidt, M. Carroll, and J. Chen. 1998.Enhanced B-1 cell development, but impaired IgG antibody responses in micedeficient in secreted IgM. J. Immunol. 160: 4776–4787.

36. Terry, L. A., J. P. Stewart, A. A. Nash, and J. K. Fazakerley. 2000. Murinegammaherpesvirus-68 infection of and persistence in the central nervous system.J. Gen. Virol. 81: 2635–2643.

37. Cho, H. J., S. Kim, S. E. Kwak, T. C. Kang, H. S. Kim, H. J. Kwon, Y. W. Kim,Y. S. Kim, E. K. Choi, and M. J. Song. 2009. Age-dependent pathogenesis ofmurine gammaherpesvirus 68 infection of the central nervous system. Mol. Cells27: 105–111.

38. Hausler, M., B. Sellhaus, S. Scheithauer, M. Engler, E. Alberg, A. Teubner,K. Ritter, and M. Kleines. 2005. Murine gammaherpesvirus-68 infection of mice:A new model for human cerebral Epstein-Barr virus infection. Ann. Neurol. 57:600–603.

39. Morrison, L. A., and D. M. Knipe. 1997. Contributions of antibody and T cellsubsets to protection elicited by immunization with a replication-defective mu-tant of herpes simplex virus type 1. Virology 239: 315–326.

40. Stevenson, P. G., R. D. Cardin, J. P. Christensen, and P. C. Doherty. 1999. Im-munological control of a murine gammaherpesvirus independent of CD8+T cells. J. Gen. Virol. 80: 477–483.

41. Gulley, M. L., and W. Tang. 2008. Laboratory assays for Epstein-Barr virus-related disease. J. Mol. Diagn. 10: 279–292.

42. Di Carlo, P., M. Trizzino, L. Titone, G. Capra, P. Colletti, G. Mazzola, D. Pistoia,and C. Sarno. 2011. Unusual MRI findings in an immunocompetent patient withEBV encephalitis: a case report. BMC Med. Imaging 11: 6.

43. Volpi, A. 2004. Epstein-Barr virus and human herpesvirus type 8 infections ofthe central nervous system. Herpes 11(Suppl 2): 120A–127A.

44. Chan, P. K., H. K. Ng, J. L. Cheung, and A. F. Cheng. 2000. Survey for thepresence and distribution of human herpesvirus 8 in healthy brain. J. Clin.Microbiol. 38: 2772–2773.

45. Kapadia, S. B., B. Levine, S. H. Speck, and H. W. Virgin, IV. 2002. Critical roleof complement and viral evasion of complement in acute persistent, and latentgamma-herpesvirus infection. Immunity 17: 143–155.

46. Kapadia, S. B., H. Molina, V. van Berkel, S. H. Speck, and H. W. Virgin, IV.1999. Murine gammaherpesvirus 68 encodes a functional regulator of comple-ment activation. J. Virol. 73: 7658–7670.

47. Tarakanova, V. L., J. M. Molleston, M. Goodwin, and H. W. Virgin, IV. 2010.MHV68 complement regulatory protein facilitates MHV68 replication in pri-mary macrophages in a complement independent manner. Virology 396: 323–328.

48. Weck, K. E., S. S. Kim, H.W. Virgin, IV, and S. H. Speck. 1999. Macrophagesare the major reservoir of latent murine gammaherpesvirus 68 in peritoneal cells.J. Virol. 73: 3273–3283.

1056 Ab IN IMMUNITY TO AND PROTECTION BY AVACCINE VIRUS

by guest on January 29, 2018http://w

ww

.jimm

unol.org/D

ownloaded from