Vaccine Vectors Ankara-Human Immunodeficiency Virus Virus- and ...

Published Ahead of Print 4 April 2007. 2007, 81(13):7022. DOI: 10.1128/JVI.02654-06. J. Virol.

Mary Marovich and Paul SpearmanCurrier, Bonnie M. Slike, Xuemin Chen, James Tartaglia, Xiugen Zhang, Farah Cassis-Ghavami, Mike Eller, Jeff Vaccine VectorsAnkara-Human Immunodeficiency Virus Virus- and Modified Vaccinia Virusand Induction of Apoptosis by Canarypox Direct Comparison of Antigen Production

http://jvi.asm.org/content/81/13/7022Updated information and services can be found at:

These include:

REFERENCEShttp://jvi.asm.org/content/81/13/7022#ref-list-1at:

This article cites 49 articles, 18 of which can be accessed free

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on February 23, 2013 by P

EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/cgi/alerts

http://jvi.asm.org/

JOURNAL OF VIROLOGY, July 2007, p. 7022–7033 Vol. 81, No. 130022-538X/07/$08.00�0 doi:10.1128/JVI.02654-06Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Direct Comparison of Antigen Production and Induction of Apoptosisby Canarypox Virus- and Modified Vaccinia Virus Ankara-Human

Immunodeficiency Virus Vaccine Vectors�

Xiugen Zhang,1 Farah Cassis-Ghavami,2 Mike Eller,3 Jeff Currier,3 Bonnie M. Slike,3 Xuemin Chen,1James Tartaglia,4 Mary Marovich,3 and Paul Spearman1*

Departments of Pediatrics and Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia1;Department of Pediatrics, Vanderbilt University, Nashville, Tennessee2; U.S. Military HIV Research Program,

Rockville, Maryland3; and Sanofi Pasteur, Toronto, Ontario, Canada4

Received 30 November 2006/Accepted 29 March 2007

Recombinant poxvirus vectors are undergoing intensive evaluation as vaccine candidates for a variety ofinfectious pathogens. Avipoxviruses, such as canarypox virus, are replication deficient in mammalian cells byvirtue of a poorly understood species-specific restriction. Highly attenuated vaccinia virus strains such asmodified vaccinia virus Ankara (MVA) are similarly unable to complete replication in most mammalian cellsbut have an abortive-late phenotype, in that the block to replication occurs post-virus-specific DNA replication.In this study, an identical expression cassette for human immunodeficiency virus gag, pro, and env codingsequences was placed in canarypox virus and MVA vector backbones in order to directly compare vector-borneexpression and to analyze differences in vector-host cell interactions. Antigen production by recombinant MVAwas shown to be greater than that from recombinant canarypox virus in the mammalian cell lines and in theprimary human cells tested. This observation was primarily due to a longer duration of antigen production inrecombinant MVA-infected cells. Apoptosis induction was found to be more profound with the empty canarypoxvirus vector than with MVA. Remarkably, however, the inclusion of a gag/pro/env expression cassette alteredthe kinetics of apoptosis induction in recombinant MVA-infected cells to levels equal to those found incanarypox virus-infected cells. Antigen production by MVA was noted to be greater in human dendriticcells and resulted in enhanced T-cell stimulation in an in vitro antigen presentation assay. These resultsreveal differences in poxvirus vector-host cell interactions that should be relevant to their use as immu-nization vehicles.

The search for a safe and effective human immunodeficiencyvirus (HIV) vaccine has stimulated the development of recom-binant live vectors as vehicles for the induction of specificcellular immune responses. Attenuated poxvirus vectors have anumber of desirable features as HIV vaccine candidates, in-cluding promising safety profiles and the ability to incorporatesubstantial genetic material for the expression of foreign geneproducts (35, 38, 46, 48). The leading poxvirus vectors arederived from two genera within the Poxviridae family. Canary-pox virus and Fowlpox virus vectors belong to the Avipoxvirusgenus. Avipoxviruses are naturally restricted to replication inavian species. In human cell lines, avipoxviruses fail to repli-cate their DNA genomes (47). The precise mechanisms under-lying this host cell restriction have not been elucidated. Despitea complete block to replication, avipoxvirus recombinants ex-press appropriately engineered foreign gene products in hu-man cells and can elicit specific cellular and humoral immuneresponses to inserted genes in animals and in human volun-teers (22, 49).

Vectors derived from Vaccinia virus, a member of the Ortho-poxvirus genus, constitute another leading group of poxvirus-

based vectors under evaluation as HIV candidate vaccines.Because of potential safety concerns with wild-type vacciniavirus strains, highly attenuated vaccinia virus strains are ac-tively being pursued as HIV vaccine candidates. NYVAC is ahighly attenuated vaccinia virus strain that was derived by theprecise deletion of 18 open reading frames from the Copen-hagen vaccine strain of vaccinia virus (47). The attenuation andimmunogenicity of NYVAC have been demonstrated in sev-eral animal models (26, 47), and the block to replication ap-pears to be less complete than that for canarypox virus inmammalian cells (47). Modified vaccinia virus Ankara (MVA)was generated by passaging the Ankara strain of vaccinia virusmore than 500 times on chicken embryo fibroblasts (CEFs)(36). MVA replicates permissively in CEF and Syrian hamsterBHK cells but does not complete its replication cycle in othermammalian cell lines or in human primary cells (9). In contrastto the avipoxvirus vectors, both NYVAC and MVA demon-strate an abortive-late phenotype in most human cell lines (45,47). Thus, although both avipoxvirus-based vectors and highlyattenuated vaccinia virus vectors are replication deficient inhuman cells, they are blocked at distinct stages of replication.

Canarypox virus vectors have been tested extensively as HIVvaccine candidates in human trials (5, 7, 13, 16, 17, 23, 34). Thesafety profiles of these vectors have in general been excellent(15), and HIV-specific immune responses have been elicited(7). The magnitudes and persistence of these responses have

* Corresponding author. Mailing address: Departments of Pediat-rics and Microbiology and Immunology, Emory University School ofMedicine, Atlanta, GA 30322. Phone: (404) 727-5642. Fax: (404) 727-8249. E-mail: [email protected].

� Published ahead of print on 4 April 2007.

7022


EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

been modest, however, leading the NIH National Institute ofAllergy and Infectious Diseases-sponsored HIV Vaccine Clin-ical Trials Network to forego plans for phase III testing of acanarypox virus-based HIV vaccine regimen (44). Neverthe-less, a phase III trial is presently under way in Thailand to testthe efficacy of a canarypox virus-based HIV vaccine candidateemployed as a priming agent in combination with a bivalentrecombinant subunit gp120 protein booster (39). Vectorsbased on the avipoxvirus fowlpoxvirus are also under evalua-tion as HIV vaccines for human use (42). MVA vectors havegenerated very promising results in simian models of HIVinfection (1, 4, 25), and on this basis, they are also currentlyundergoing evaluation in human trials. The promise of thisapproach is highlighted by the ability of a DNA prime/MVAboost regimen to protect macaques from simian-human immu-nodeficiency virus 89.6P challenge (1), and this regimen is nowthe model for a human trial sponsored by the National Insti-tutes of Health-sponsored HIV Vaccine Trials Network.MVA-based vaccination regimens are also safe, as reported ina phase I human clinical trial (10).

Despite intense interest in poxvirus vectors as candidatevaccines for HIV and other infectious diseases, few directcomparisons of poxvirus-based vaccine candidates have beenreported to date. A recent report compared NYVAC andMVA vectors in cell culture systems. Significant differenceswere discovered in vector-generated cytopathic effect, growthcharacteristics, phosphorylation of the translational initiationfactor eIF2-�, stage of replication block, and the degree ofapoptosis induced by these vaccinia virus-derived vectors (37).These findings can now form the basis for a detailed under-standing of the vector-specific factors that may affect the im-munogenicity of vectors employed in vaccine trials. The findingof a number of significant differences between the closely re-lated MVA and NYVAC vectors makes it even more likelythat there will be distinct differences in antigen production andhost interactions seen between the more distantly related or-thopoxvirus vectors and avipoxvirus vectors.

Direct comparisons of vector-elicited antigen expression andhost cell effects generated by avipoxvirus vectors and attenu-ated vaccinia virus vectors are warranted in order to choose theoptimal vector to elicit a desired response and to design im-proved future generations of vaccines. In this study, we per-formed a head-to-head comparison of a canarypox virus vectorand an MVA vector bearing an identical HIV gene expressioncassette. We found significant differences in the timing andmagnitudes of antigen expression exhibited by canarypox virusand MVA vectors. While apoptosis was lower, as expected,with the naked MVA vector, the inclusion of a gag/pro/envexpression cassette had a major influence on apoptosis medi-ated by either recombinant vector. Recombinant MVA gener-ated more antigen in most human cell lines and in primarycells, including human dendritic cells (DC), than did thecanarypox virus recombinant and resulted in enhanced cellularresponses in an in vitro antigen presentation system.

MATERIALS AND METHODS

Construction of recombinant viruses. The construction of the recombinantcanarypox virus vector vCP205 has been described previously (19). This vectorexpresses the HIV type 1 (HIV-1) LAI gag and pro open reading frames underthe control of the vaccinia virus I3L promoter. The expression cassette includes

a modified HIV-1 MN env gene corresponding to the gp120 coding region linkedto the gp41 transmembrane anchor sequence from HIV-1 LAI (28 amino acids)and under the control of the vaccinia virus H6 promoter. The gag/pro and envexpression cassettes were placed between canarypox virus flanking regions inshuttle plasmid pHIV32, and homologous recombination was used to generatethe recombinant canarypox virus vCP205. The recombinant MVA virus MVA205was generated by inserting the identical expression cassette used for creatingvCP205 into the deletion 3 locus of the MVA genome. To do this, the entireexpression cassette was removed from the vCP205 transfer vector pHIV32 byrestriction digestion with BamHI and XbaI and was then inserted by blunt-endcloning into the MVA transfer vector pMC03 (generously provided by LindaWyatt and Bernard Moss, NIH), between SbfI and AscI sites. The resultingtransfer construct lacked the synthetic vaccinia virus early/late promoter ofpMC03 but retained the selection marker beta-glucuronidase (8). After sixrounds of plaque purification and selection using staining with X-Gluc (5-bromo-4-chloro-3-indolyl-�-D-glucuronic acid), a master stock of MVA205 was gener-ated. One experiment in this paper describes the use of vCP1452, a vector thatincludes the vaccinia virus E3L and K3L genes, whose construction was describedin detail previously (20). MVA-GFP was generously provided by Mark Feinberg(11). vCP-GFP is a recombinant canarypox virus vector, previously termedvCP1540 and developed by Sanofi Pasteur, expressing the green fluorescentprotein (GFP) gene under the control of the H6 promoter. MVA/T7 is a recom-binant vector expressing T7 polymerase (50) and was provided by Linda Wyattand Bernard Moss.

Virus propagation and plaque titration. CEFs were prepared from 10- to11-day-old chicken embryos and were maintained in minimum essential medium(MEM; Gibco, Gaithersburg, MD) with 10% fetal calf serum (FCS), 100 U/mlpenicillin, and 100 g/ml streptomycin. To generate MVA205 and vCP205 stocks,150-cm2 flasks were seeded from the master stocks at a multiplicity of infection(MOI) of 0.1. The infected cells were incubated at 37°C for 3 to 5 days untilmaximum cytopathic effect was evident. Working virus stocks were generated byfreeze-thawing and sonication of the infected CEF cell pellets three times,followed by removal of cell debris by centrifugation at 1,000 � g for 10 min at4°C. Stocks were plaque titrated by preparing serial 10-fold dilutions of stockvirus in MEM with 2% FCS. Diluted virus was placed on CEFs in 60-mm2 dishesfor 1 h and rocked three or four times. The inoculum was then removed and thedishes overlaid with 2% MEM containing 1% agarose. vCP205 plaques werecounted after being stained with 0.1% neutral red, and the plaque titers of stockswere recorded; MVA205 titers were tested by X-Gluc staining. The resultingvirus stocks were stored at �80°C until use. Gag and Env expression by MVA205and vCP205 stocks was confirmed by Western blotting using pooled sera fromHIV-positive individuals for detection, and the presence of intact gene insertswas confirmed by sequencing the viral DNA extracted from cells infected withthe master stock.

Quantitation of antigen expression. Gag-Env expression was measured aspreviously described (19, 20). Cells were plated on the night prior to infection,and infections were performed when the cells were noted to be 80% confluent onthe dish. An accurate count of the number of cells present per dish was achievedby counting the cells from one 10-cm2 dish at the time of infection (about 1.2 �107 CEF cells or 0.8 � 107 HeLa cells at 80% confluence). Cells were infectedwith MVA205 and vCP205 at an MOI of 10 in 2 ml of MEM with 2% FCS. After1 hour of infection with intermittent gentle agitation, the medium was removed,and 10 ml of fresh 2% MEM was added to each dish. Medium and/or cells wereharvested at the time points outlined in Results for detection of p24 antigen orfor other assays. Samples for protein analysis were prepared as follows. Super-natants were filtered through a 0.45-�m filter and pelleted by centrifugationthrough a 5-ml 20% sucrose cushion in a Beckman SW28 rotor at 100,000 � g for3 h. The pellets were then resuspended in phosphate-buffered saline (PBS).Analysis of cellular protein expression was performed by lysing cells in PBS with1% NP-40; nuclei were separated via centrifugation at 10,000 � g for 10 min. Gagand Env protein contents were analyzed by Western blotting using sera pooledfrom HIV-positive donors.

Cell lines and primary cells. In addition to the CEFs described above, anumber of cell lines were employed in this study. These included murine C2C12myoblasts (ATCC CRL-1772), murine 3T3 fibroblasts (ATCC CRL-1658),African green monkey kidney BSC-40 and Vero cells (ATCC CCL-81), humanrhabdomyosarcoma TE671 cells (ATCC CRL 8805), human melanoma MelJuSocells, human embryonic kidney 293 cells (ATCC CRL 1573), human cervicalcarcinoma HeLa cells (ATCC CCL-2), human U937 macrophages (ATCC CRL-1593.2), human Jurkat T cells, and human kidney epithelial MRC-5 cells (ATCCCCL-171). All cell lines were obtained from the American Type Culture Col-lection, with the exception of BSC-40 (provided by Bernard Moss, NIH), Jurkat(obtained from Barney Graham, NIH), and MelJuSo (obtained from Markus

VOL. 81, 2007 COMPARISON OF POXVIRUS-BASED HIV VACCINE VECTORS 7023


EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

Thali, University of Vermont). Adherent cell lines were maintained in Dulbecco’smodified Eagle medium with 10% FCS; suspension cells were maintained inRPMI 1640 with 10% FCS. Primary human peripheral blood mononuclear cells(PBMCs) were isolated by Ficoll-Hypaque separation and were maintained inRPMI 1640 medium with 10% FCS, 2 mM L-glutamine, penicillin G (100 U/ml),and streptomycin (100 �g/ml). After poxvirus infection, PBMCs were cultured in2% RPMI 1640 medium. Human monocytes were isolated from the PBMCfraction by negative depletion with a monocyte isolation kit (Miltenyi Biotec,Auburn, CA). Isolated monocytes (CD45� CD14�) were �95% pure, as indi-cated by flow cytometry. Monocytes were cultured in RPMI with 10% FCS and2% granulocyte-macrophage colony-stimulating factor. After 4 days of matura-tion in culture, monocytes were infected with MVA205 or vCP205, using thesame methods as those used for the adherent cell infection process.

Virus infection efficiency assay. Determination of the efficiency of infectionwas performed using viruses expressing GFP, namely, MVA-GFP (11) and vCP-GFP (expressing the GFP gene under control of the vaccinia virus H6 promoter;originally called vCP1405). For this analysis, cells were infected at an MOI of 10PFU/cell, as measured by CEF plaque titration. Cells were harvested at 24 hpostinfection, washed three times with cold PBS containing 0.1% bovine serumalbumin, fixed with 2% formaldehyde, and analyzed using a BD LSR II flowcytometer. Gates were determined using an identical population of uninfectedcells.

Measurement of virus-induced apoptosis. HeLa cells were infected withMVA205, vCP205, or vCP1452 at an MOI of 10 PFU/cell. Cells were harvestedat 12 and 24 h postinfection, and DNA fragmentation was analyzed with anApoptotic DNA Ladder kit (Roche Applied Science, Indianapolis, IN), using themanufacturer’s instructions. A second method used a cell death detection en-zyme-linked immunosorbent assay (ELISA; Roche) according to the manufac-turer’s instructions. Flow cytometry was used to measure apoptosis, using an-nexin V and propidium iodide (PI) staining. Cells were detached from plates withAccutase (eBioscience, San Diego, CA) at 12 or 24 h postinfection with vCP205or MVA205 or at the same time following treatment with camptothecin. Cellswere washed in cold PBS, stained with annexin V-fluorescein isothiocyanate orannexin V-allophycocyanin and PI, and analyzed on a BD LSR II flow cytometeror a BD FACSCanto flow cytometer.

Real-time PCR. To examine the quantity of viral DNA present in cells infectedwith vCP205 or MVA205, HeLa cells were infected at an MOI of 10 PFU/cell insix-well plates. The cells were harvested at 0, 2, 4, 8, 12, 24, 48, and 72 hpostinfection. Each experiment was performed in triplicate. Cells were washedonce with cold PBS, and DNAs were extracted using a QIAamp DNA Mini kit(QIAGEN, Valencia, CA). The following real-time PCR primers located withinthe HIV gag region were used: 2463-2486, AGAGAAGGCTTTCAGCCCAGAAGT; and 2639-2616, TGCACTGAATGCACTCTATCCCAT. These primersamplified a 176-bp fragment. Quantitative real-time PCR was performed in aDNA Engine Opticon 2 system (Bio-Rad Laboratories, Hercules, CA). Serialdilutions of 20 ng/ml of pHIV32 plasmid at 1:10 dilution increments were usedto generate standard curves at the same time in the same plate.

Antigen presentation in human myeloid DC. Myeloid DC were derived fromleukapheresis blood products from healthy HIV-seropositive donors, obtainedunder an Internal Review Board-approved protocol (RV149) and prepared aspreviously described (33). The isolated peripheral blood monocytes were treatedwith recombinant human interleukin-4 (R&D Systems, Minneapolis, MN) andrecombinant human granulocyte-macrophage colony-stimulating factor (FisherClinical Services, Rockville, MD) to drive DC differentiation. The DC wereexposed to the vectors at various MOI within a working range of vector/DC ratiosof 1 to 5 PFU/cell. At 2 h postinfection, the DC were washed and then resus-pended in medium to a concentration of 106/ml. Samples from each experimentwere set aside for a functional DC enzyme-linked immunospot (DC-ELISPOT)assay, and the remaining DC were incubated at 37°C for 2 to 6 h for laterexpression analysis. Infection rates were determined using labeling with antivec-tor antibodies (rabbit anti-canarypox virus [Sanofi Pasteur]) and an anti-vacciniavirus monoclonal antibody (MAb; clone TW2.3) (29). HIV transgene expressionwas measured by intracellular labeling with rabbit anti-Env (LS086-4-D51; agenerous gift from Florence Boudet at Sanofi Pasteur) and an anti-p24 MAb(Dako, Glostrup, Denmark) and then was analyzed by flow cytometry (33).Functional effects of canarypox virus and MVA infection and HIV gene expres-sion were assessed after vector loading of the DC in a DC antigen presentationassay. The DC-ELISPOT assay uses autologous DC and responder PBMCs fromhealthy HIV-seropositive donors to detect antigen-specific gamma interferon(IFN-�), as previously described (15). Briefly, 96-well ELISPOT plates wereprecoated with anti-IFN-� MAb (Mabtech AB, Sweden) and then loaded withcells. The cells included infected (or mock-infected) DC distributed in triplicatewells, with or without PBMCs at a 1:10 ratio of DC to PBMCs. In some exper-

iments, the DC were titrated and used across a 3-log dilution to verify doseeffects (1:10, 1:100, and 1:1,000 ratios of DC to responders). The plates wereincubated for 20 to 24 h at 37°C, washed, incubated with a biotinylated anti-IFN-� MAb (Mabtech), washed again, and then developed with a horseradishperoxidase system (Vectastain ABC kit; Vector Labs, CA). The data were col-lected and analyzed using simple statistics (linear regression and paired Student’st test).

RESULTS

Expression of HIV proteins by MVA and canarypox virusrecombinant vectors bearing identical HIV gene expressioncassettes. In order to define vector-specific effects on antigenproduction and on host cells, we constructed an MVA recom-binant virus that incorporates an HIV expression cassette iden-tical to that in vCP205. vCP205 is a gag/pro/env HIV vaccinecandidate vector that has been evaluated extensively in thelaboratory and in human trials (6, 12, 24). This vector incor-porates the HIV-1LAI gag and protease coding sequences un-der the control of the vaccinia virus I3L promoter and a gp120gene from HIV-1 MN fused to the coding sequence for thetransmembrane domain of gp41 from HIV-1LAI under thecontrol of the vaccinia virus H6 promoter. The detailed con-struction of this vector has been described elsewhere (19, 20).The gag/pro and env expression cassette from vCP205 wasplaced within the deletion 3 region of MVA, using homologousrecombination and marker selection. The resulting virus,MVA205, contains the identical HIV gene segments under thesame poxvirus promoters as those in vCP205. MVA205 was

FIG. 1. HIV protein expression by vCP205 and MVA205. (A) West-ern blot of HIV antigen expression in infected CEFs. Cells wereharvested 24 h following infection with test viruses. Cells were har-vested (C) and supernatants pelleted (S) for detection of virus-likeparticle production. The blot was probed using HIV-positive patientsera. The positions of molecular mass markers are shown on the left.(B) HeLa cells were infected and treated as described for panel A.

7024 ZHANG ET AL. J. VIROL.


EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

plaque purified and grown to a high titer (1 � 108 PFU/ml).HIV protein expression was verified in CEFs by Western blot-ting (Fig. 1A). Both constructs produced roughly equivalentamounts of Gag protein in CEFs, although there was moreintracellular cleavage of Pr55Gag in cells infected with vCP205.Gag-Env pseudovirion particles were released from CEFs in-fected with MVA205 and vCP205 (Fig. 1A, supernatant lanes).Unexpectedly, the amount of gp120 produced by vCP205 inCEFs was lower than that produced by MVA205. Protein ex-pression and particle release by HeLa cells appeared to bebasically equivalent for the two vectors by Western blotting(Fig. 1B). The magnitudes and patterns of Gag and Env pro-tein expression in these cells appeared to be identical, validat-ing the construction of heterologous vectors bearing the samegene expression cassette.

In order to compare more precisely the temporal patternsand magnitudes of antigen expression by MVA205 andvCP205, we measured p24 antigen release in cellular superna-tants over time. The full-length gag gene expressed by bothvectors allows efficient pseudovirion particle formation and

release, and the levels and timing of supernatant antigen re-flect cellular production (19). In CEFs, the patterns and mag-nitudes of p24 production were nearly identical (Fig. 2, CEFpanel). However, substantial differences were observed in aseries of mammalian cell lines. Production and release of an-tigen by MVA205 were more efficient than those by vCP205 inmammalian cells, with the exception of 293 cells (Fig. 2). Thepeak of antigen production by the MVA vector was delayedcompared with that for canarypox virus in each mammaliancell line tested. This is best illustrated with the murine myo-blast cell line C2C12, where peak particle production was seenat 24 h with vCP205 and at 48 h with MVA205 (Fig. 2, C2C12panel). These data indicate that both the magnitude and tem-poral pattern of protein expression differ for MVA and canary-pox virus, despite the use of identical promoter-gene expres-sion cassettes.

Expression of antigen by a vaccine vector in immortalizedepithelial or muscle cell lines may not reflect that in morerelevant cells. To address this issue, we extended this anal-ysis to human T-cell and monocytic cell lines and to primary

FIG. 2. Gag-Env pseudovirion production and release by recombinant poxviruses in mammalian adherent cell lines. Each cell line shown wasinfected at an MOI of 10 with vCP205 or MVA205. Cellular supernatants were harvested at multiple time points following infection, and p24antigen was measured by an antigen capture ELISA. p24 antigen production by MVA205 is indicated by black squares, and antigen productionby vCP205 is shown by open circles. Error bars represent standard deviations for triplicate wells.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

human cells. The production of antigen was significantlylower in these cells than in adherent mammalian cell lines.However, the magnitudes of antigen expression and release byMVA205 were higher than those by vCP205 in Jurkat T cells,the monocyte-derived U937 cell line, and human lung MRC-5cells (Fig. 3, top panels). Antigen production and releasefrom PBMCs were low (2 to 3 ng/1 � 106 cells), but these cellsalso demonstrated greater antigen production by MVA205than by vCP205. Human monocyte-derived macrophages dem-onstrated a marked difference in antigen production, and thelater peak of antigen production in these cells by MVA205 wasclearly demonstrated (Fig. 3, macrophage panel). Taken to-gether, these data demonstrate that the higher level of antigenproduction and later peak of antigen production/release seenwith MVA205 in cell lines are also seen in primary humancells.

In order to be certain that the release of antigen in theseassays was an accurate reflection of the magnitude and tem-poral sequence of cellular protein production and that thedelayed peak of protein production by MVA represented on-going protein production in infected cells, we quantified cellu-lar and supernatant p24 production over time for two of themammalian cell lines. In HeLa cells, cell-associated p24 pro-duction by vCP205 peaked by 18 to 24 h postinfection, whileproduction by MVA205 was still rising at these times (Fig. 4,top panels). Intracellular levels of p24 generated by MVA205peaked at 36 h postinfection, and antigen release continued torise until 48 h postinfection. TE671 cells demonstrated thesame earlier and lower peak of p24 expression for vCP205 (Fig.4, bottom panels) and the same extended production forMVA205. Thus, there was a longer duration of protein expres-sion, and the total amount of antigen produced was two- tothreefold higher, in these cells when the same genes wereexpressed by MVA205 than the case with vCP205.

We considered the possibility that differences in the efficien-cies of MVA versus canarypox virus in entering and produc-tively infecting the cell lines and primary cells tested may havebeen responsible for some of the observed differences in themagnitudes of antigen expression. To test this, we utilizedMVA and canarypox virus vectors expressing GFP and moni-tored the efficiency of infection by flow cytometry. Cells wereinfected at an MOI of 10 PFU/cell of either virus, and thepercentage of cells expressing GFP was assessed at 24 h postin-fection. Both canarypox virus-GFP and MVA-GFP were ableto efficiently infect each adherent cell line (Fig. 5A to J).Although there were modest differences noted in the expres-sion of GFP in CEF and 293 cells, where these differences wereseen the higher levels were with canarypox virus and thus couldnot account for enhanced antigen production by MVA. Mostcell lines and primary cell preparations were infected equallyby either vector. Not surprisingly, the efficiencies of infectionof suspension cells (Jurkat and U937 cells) and of primary cells(PBMCs and macrophages) were markedly lower (10 to 20%of cells infected) than those seen with adherent cell lines.Overall, the ability of both vectors to infect cells at equivalentor near-equivalent levels ruled out differential entry/infectionas the reason for enhanced antigen expression by MVA.

Differential apoptosis induction by canarypox virus andMVA vectors. Apoptosis induction by live vector vaccines mayhave important effects on the resulting immune response.While rapid apoptosis may limit antigen expression and thusadversely impact specific immunity to vector-encoded antigens,apoptotic responses may also be desirable for antigen uptakeand presentation through the alternative class I presentationpathway (antigen cross-priming). We examined apoptosis byMVA205 and vCP205 by first using a traditional DNA frag-mentation assay with vector-infected HeLa cells (Fig. 6A). At12 h postinfection with either virus, little DNA fragmentation

FIG. 3. Gag-Env pseudovirion production and release from suspension cells and primary cells. Suspension cells were infected at an MOI of 10with vCP205 or MVA205, and supernatants were collected at the indicated times for p24 antigen measurement. Note that the y-axis scale isdifferent from that in Fig. 2. p24 production by MVA205 is shown by black squares, and vCP205 production is indicated by open circles. Error barsrepresent standard deviations for triplicate wells.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

was observed. However, at 24 h postinfection, there wasmarked apoptosis induction, appearing approximately equalfor both viruses (Fig. 6A). We included a second recombinantcanarypox virus, vCP1452, in this analysis as a comparator.This recombinant expresses the same gag, pro, and env genes asvCP205 and MVA205 but was also engineered to express thevaccinia virus E3L and K3L genes, resulting in diminishedactivation of the RNA-dependent protein kinase, enhancedantigen expression, and diminished apoptosis in mammaliancells (20). As noted previously, apoptosis induction was mark-edly less prominent with vCP1452. We next performed a morequantitative assessment of apoptosis induction by MVA205versus vCP205 with an ELISA-based assay for detection ofhistone-complexed DNA fragments in the cytoplasm. At 12 hpostinfection, MVA205 and vCP205 demonstrated compara-ble levels of apoptosis in this assay. By 24 h, both MVA205 andvCP205 induced significant apoptosis above the negative con-trol level and above the levels produced by vCP1452. The totalamount of apoptosis induced by vCP205 was greater at 24 hthan that of MVA205 by this assay. We then sought to definethe role of the vector itself versus the role of gag and env geneexpression in apoptosis induction by these poxvirus vectors inmammalian cells. To do this, we used annexin V and PI stain-ing of either empty vector or vector plus insert 12 and 24 hfollowing infection. Annexin V staining detects phosphatidyl-serine translocation to the outer membrane, an early marker ofapoptosis (31), while PI staining reveals later stages of apop-

tosis or necrosis. Therefore, in our flow cytometry analysis, thelower right quadrant (annexin staining) represents early apop-tosis while the upper right quadrant (annexin plus PI) repre-sents late apoptotic events, and total apoptosis is indicated bythe total of these two quadrants. In mock-infected HeLa cells,low levels of early and late apoptosis were observed (Fig. 6C,negative control). Camptothecin treatment resulted in a sig-nificant increase in total apoptosis observed compared to thatfor negative control cells (23.2% versus 5.6%). We then exam-ined MVA and canarypox virus vectors lacking any HIV in-serts. MVA itself generated modestly increased apoptosis 12and 24 h following infection compared with that of control cells(Fig. 6C). In sharp contrast, MVA205 generated markedlyhigher levels of apoptosis at both 12 and 24 h. At the 24-h timepoint, MVA205 generated 41.9% apoptotic cells, in contrast to14.6% apoptotic cells for the MVA empty vector. Apoptosis bycanarypox virus empty vector was higher than that by MVA at24 h (23.4%) but demonstrated the same pattern of greaterapoptosis when expressing the gag-env insert (39% apoptoticcells). A potentially important difference was noted betweenvCP205 and MVA205 at the earlier time point. By 12 h, a largepercentage of vCP205-infected cells were already double pos-itive, indicating that this vector induced very rapid apoptosisthat had progressed to late stages by 12 h. We includedvCP1452, designed to decrease apoptosis, as a further control.This vector induced less apoptosis at both 12 and 24 h than theempty canarypox virus vector, despite expressing HIV gag and

FIG. 4. Cellular versus supernatant antigen expression. Cells were infected at an MOI of 10 with either virus and harvested at the indicatedtimes for p24 antigen quantitation for cells (left) and the corresponding pelleted supernatants (right). Black squares indicate MVA205, and whitecircles indicate vCP205. Error bars represent standard deviations for triplicate wells. (Top) HeLa cells. (Bottom) TE671 (human rhabdomyosar-coma) cells.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

FIG. 5. Comparison of efficiencies of infection by canarypox virus and MVA. Each of the cell types used in this study was infected withvCP-GFP and MVA-GFP, and the percentages of cells expressing GFP were quantified by flow cytometry. Gray histograms represent uninfectedcells, and white histograms represent green fluorescence 24 h following infection. Bars and corresponding numbers indicate percentages of theinfected cell populations expressing GFP. Suspension cells with lower infection efficiencies are shown in panels K and L. Results are representativeof experiments performed three times.

7028


EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

env genes. Taking all of these apoptosis data together, weconcluded that vCP205 and MVA205 both induced substantiallevels of cellular apoptosis in mammalian cells and that theHIV gene insert had a surprising and robust effect on the levelsof apoptosis observed. Furthermore, apoptosis induction byvCP205 was more profound at early time points and thus mayhave contributed to the limitation of late antigen production bythis vector.

To verify that the HIV gene insert was responsible for apop-tosis induction by these recombinant poxvirus HIV vaccinevectors, we compared apoptosis induction by MVA205 or

vCP205 to that by recombinant vectors expressing other genes.MVA205 infection of HeLa cells led to substantially moreapoptosis at 24 h than did infection with MVA expressing T7polymerase or MVA expressing GFP (Fig. 7E versus C and F).The difference was most pronounced in the double-positive (PIplus annexin V) population. Similarly, vCP205 induced a muchhigher level of apoptosis that had progressed to late stages by24 h than did empty vector or a canarypox virus expressingGFP (Fig. 7G to I). These data confirm that the HIV geneinsert exerts substantial effects on apoptosis induction byvCP205 and MVA205.

FIG. 6. Measurement of virus-mediated apoptosis. Three methods were used to assess apoptosis in virus-infected HeLa cells. (A) DNAfragmentation was assessed in cells infected at the indicated times. The DNA marker was lambda DNA digested with HindIII and EcoRI. Thepositive control was a kit positive control from the manufacturer (Roche), and the negative control sample was from uninfected cells. (B) ELISAfor detection of histone-bound DNA in cytoplasm. MVA205, vCP205, and vCP1452 were evaluated in HeLa cells for apoptosis at 12 and 24 h.Blank, no added DNA; Neg Ctr, DNA from uninfected cells; Camp, positive control from camptothecin-treated HeLa cells. (C) HeLa cells wereinfected with the indicated viruses and harvested at 12 or 24 h for annexin V and PI staining by flow cytometry. Gates were set on single-staineduninfected cell populations. Numbers represent percentages of cell populations in each quadrant.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

Quantitation of the differential block to viral DNA replica-tion in HeLa cells. MVA is able to replicate its viral DNA inmammalian cells but is arrested at a late (assembly) stage,while avipoxviruses have been reported to be arrested at astage prior to DNA replication (47, 49). However, detailedquantitation of avipoxviral DNA levels over time followinginfection of human cells has not been reported. We consideredthe possibility that the extended duration and magnitude ofheterologous gene expression by MVA relative to those forcanarypox virus may be linked with the ability to replicate theviral genome and proceed with late-stage transcription. To testthis possibility, we developed a real-time PCR assay for mea-suring the number of copies of gag present in cells infected withMVA205 or vCP205. Serial dilutions of the transfer plasmidused to insert the gag gene into the poxviruses were used togenerate a standard curve for absolute quantitation of DNAcopies. Cells (1 � 106) were infected with 1 � 107 PFU ofeither virus, and the inoculum DNA was measured at timezero. After 1 h, the cells were washed, and subsequent PCRdata were derived from DNAs extracted from the infectedcells. For MVA205, the number of DNA copies of gag repre-senting the poxvirus genome increased from 12 to 24 h, to alevel approximately 50-fold higher than the input, and subse-quently declined (Fig. 8, filled squares). The copy number forgag in the vCP205-infected cells did not increase after infectionbut declined rapidly between 12 and 24 h after infection. Thesedata confirm that avipoxvirus vectors fail to replicate theirDNAs in HeLa cells, while the DNA copy number in MVA-infected cells increases at least until 24 h postinfection. Al-though ongoing foreign antigen expression by MVA may notbe linked directly to the kinetics of DNA replication, this

finding further highlights the difference in the stages of repli-cation arrest of MVA and canarypox virus in mammalian cells.

Differential antigen production in human DC by MVA andcanarypox virus vectors. The most relevant cell type for study-ing vector-specific antigen expression is not known with cer-tainty, as antigen cross-presentation may occur from uptake ofdying cells or apoptotic bodies by professional antigen-present-ing cells (APCs). However, the ability of vectors to infect andproduce antigens directly in DC may be desirable. We there-fore compared antigen production by vCP205 and MVA205 inmyeloid DC derived from multiple human donors. In a seriesof direct comparison experiments (MOI 5), the vCP205 andMVA205 constructs consistently infected DC at the same rates(59.0% versus 60.2%) (Fig. 9A, white bars). The vector-loadedDC viability for each construct was comparable and stable(range, 80 to 90%). However, expression of the HIV antigensEnv and Gag was consistently and significantly higher forMVA205 than for vCP205 (Fig. 9A, filled bars) (Env expres-sion, 49.9% 7.7% versus 5.0% 2.5% [P � 0.0001]; Gagexpression, 62.8% 13.2% versus 10.5% 1.8% [P � 0.001]).The different expression levels translated into differences inIFN-� production, as assessed by ELISPOT assay. Figure 9B toD show IFN-� ELISPOT data for three representative donors.MVA205, whether loaded onto autologous DC or droppedinto the well directly, resulted in more IFN-� production ineach experiment. There was a strong positive correlation be-tween IFN-� production in the antigen presentation assay andHIV Env expression (r 0.78; P � 0.0001) or HIV Gagexpression (r 0.80; P � 0.0001), as determined by the Spear-man correlation (data not shown). These results suggest thatthe ability of a vector to sustain antigen production and achievehigher levels of antigens in DC could have functional effectsthat may impact vector-induced immune responses.

DISCUSSION

Live viral vectors represent central components of manycurrent HIV vaccine regimens. The poxvirus family includes anumber of vectors that are under evaluation as HIV vaccinecandidates. In order to understand how these vectors shouldbest be used in vaccination regimens, it is essential to analyze

FIG. 7. Effect of insert on virus-induced apoptosis. HeLa cells wereinfected with the indicated viruses and harvested at 12 or 24 h forannexin V and PI staining by flow cytometry. Gates were set on single-stained uninfected cell populations. (A) Negative control, (B) camp-tothecin, (C) MVA/T7, (D) MVA empty vector, (E) MVA205,(F) MVA-GFP, (G) canarypox virus empty vector, (H) vCP205,(I) vCP-GFP. Numbers represent percentages of the cell populationsin each quadrant.

FIG. 8. Quantitative real-time PCR for viral DNA. Quantitation ofviral DNA was performed using primers within the HIV gag genecompared with a purified DNA standard. Time zero represents inputviral DNA. Gray squares represent DNA copies in MVA205-infectedcells, and white circles represent DNA copies in vCP205-infected cells.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

both vector-specific influences on antigen expression and theeffect of the vector on host cell responses, such as the inductionof apoptosis. In this study, we analyzed the pattern and mag-nitude of antigen expression by an avipoxvirus vector in directcomparison with those of an attenuated orthopoxvirus vector.Our study was unique in that it utilized a common gag/pro/envgene expression cassette placed within these disparate vectorbackbones. This allowed us to make conclusions regarding thetiming and magnitude of antigen production characteristic ofeach vector backbone and to compare the effects of each viruson cellular apoptosis when utilized as an HIV antigen expres-sion vehicle.

Antigen production by MVA in a variety of cell lines wasfound to be greater than that of canarypox virus. This resultwas consistently seen in the majority of human cell lines tested,including PBMCs, macrophages, and DC. Interestingly, alonger period of protein production seemed to account formost of the observed differences. Cellular production of anti-gen by canarypox virus diminished after 18 to 24 h, whileproduction of antigen by MVA205 continued until 36 to 48 hpostinfection. The difference in the magnitudes of protein pro-duction was unrelated to the ability to infect target cells, asboth vectors infected a wide variety of cells at similar efficien-cies. The differences in antigen production in our cell cultureexperiments were in the range of two- to fivefold. It will beinteresting to determine if these magnitudes of difference inthe expression of a common antigen result in important dif-ferences in immune responses in different animal species, es-pecially humans. The results from infected DC in the in vitromodel presented here suggest that this difference may indeed

result in significant benefits in antigen presentation. However,a number of factors in addition to antigen expression by thevector have the potential to significantly influence the resultingimmune response in immunized animals or humans. Amongthese, apoptosis induction has been studied intensively andmay be critical.

Poxviruses have developed a number of means to preventcellular apoptosis (18, 40). These include mechanisms to avoidcytotoxic T-lymphocyte (CTL)-mediated killing by down-regu-lation of major histocompatibility complex class I molecules,expression of viral proteins that interfere with IFN and tumornecrosis factor signaling pathways, and expression of moleculesthat directly interfere with caspase activation. Attenuated pox-viruses such as MVA retain some antiapoptotic genes that areessential for their replication. MVA carries E3L, a double-stranded RNA binding protein that inhibits protein kinase Ractivation (14). Deletion of E3L generates an MVA that ishighly apoptotic in human cells and that fails to replicate inCEFs (21, 28). Recently, the F1L gene product has also beenshown to contribute to the ability of MVA to inhibit apoptoticresponses (21). The abundance of antiapoptotic factors pro-duced by poxviruses suggests that for viral replication to suc-ceed, it is essential for the viruses to potently disable cellularapoptotic responses. In the context of a vaccine vector, how-ever, complete avoidance of apoptosis may not be desirable.Apoptosis may enhance immune responses through the forma-tion of apoptotic bodies and the uptake and presentation ofantigen through the alternative class I antigen presentationpathway (43). Enhanced apoptosis induction by MVA com-pared to that by vaccinia virus in human DC has been proposed

FIG. 9. Infection of human DC and in vitro antigen presentation. (A) Expression of HIV Env and Gag proteins in DC loaded with vCP205 (left)or MVA205 (right), as assessed by flow cytometry. The percentage of cells infected at an MOI of 5 is shown by detection of the respective vector.Data shown are the means for six experiments plus standard deviations. (B to D) IFN-� ELISPOT analysis of T-cell responses following coculturewith DC infected with vCP205, MVA205, or a vector controls (CPpp and MVAp580, respectively). Results from experiments with blood cells fromthree representative donors are shown. DC were infected at an MOI of 5 unless noted otherwise.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

as an explanation for its greater immunogenicity, presumablythrough enhanced antigen cross-presentation (11). Apoptosisinduction is also likely to play an important role in the immu-nogenicity of avipoxvirus-based vectors. vCP1452 was designedto enhance the immunogenicity of canarypox virus ALVACvectors through increased antigen expression and avoidance ofcellular apoptosis (20). In cell culture systems, this vector suc-cessfully diminished apoptosis and resulted in greater levels ofHIV antigen expression. Cellular immune responses elicited byvCP1452 in human trials were disappointing, however, andwere substantially lower than those seen in earlier trials em-ploying vCP205 (6, 23). While there are other plausible expla-nations for the apparently lower level of responses generatedby vCP1452, it is possible that the reduced apoptosis generatedin host cells infected by vCP1452 contributed to the reductionin immunogenicity seen in human trials with this vaccine can-didate vector. Thus, effective antigen production in the hostmust be considered together with the ability to elicit apoptosisas a key factor contributing to vector-induced immunity.

The fact that gag/pro and env gene expression dramaticallyincreased vector-induced apoptosis in our study is a novelfinding that should be considered in strategies to modify theimmunogenicity of poxvirus-based vectors. Perhaps this effectshould not be surprising, because numerous studies have doc-umented the proapoptotic effects of HIV gp120 (2, 3, 27, 30,32, 41). The effects of expression of the Gag protein on apop-tosis are less certain, but we have observed apoptotic nuclearchanges in mammalian cells upon expression of the Gag pro-tein (unpublished observations). We found that the proapop-totic effects of gag and env gene expression in either vectorwere profound, making the more subtle differences seen inMVA and canarypox virus empty vectors less important. Thesefindings highlight the important role that the genetic insertmay play in altering apoptosis. For the design of the ideal levelof apoptosis by a recombinant vaccine vector, both the vectorbackbone and the insert effects must be considered. We notethat insertion sites differed between the two vectors and thatindirect effects on neighboring genes could have contributedsomewhat to our findings.

An advantage of viral vectors is their ability to enter mam-malian cells and endogenously produce relevant intracellularproteins for subsequent antigen presentation. Viral vectorcharacteristics that allow efficient entry, early protein produc-tion, and sufficiently prolonged cell survival to allow antigenpresentation might be important, especially in the context ofdirect infection of professional APCs. In this direct head-to-head comparison of two different viral backbones with identi-cal gene inserts, we found that expression levels of HIV pro-teins were directly correlated to ELISPOT responses in an invitro assay of antigen presentation. Despite similar infectionrates, very different expression levels of relevant antigens wereobserved within myeloid DC. Therefore, simple “targeting” ofAPCs with viral vectors may be insufficient to prime the im-mune response. We propose that longer cell survival of DCinfected with MVA but not canarypox virus allowed sustainedtransgene expression and successful antigen presentation andrecognition by T cells to occur and that somewhat delayedapoptosis may have been beneficial. Whether DC serve as theprimary APCs (direct priming) or serve as a protein/peptidesource for indirect priming (cross-priming), the amount of

antigen/protein produced is likely to be important. Under-standing and balancing the dynamics of vector insert expres-sion and vector-induced apoptosis will be important to opti-mize future live vector HIV vaccine development.

These cell culture-based results provide the basis for direct,head-to-head immunogenicity comparisons of avipoxvirus andMVA vectors bearing identical inserts in animal models. It willbe interesting to determine whether the differences in antigenproduction and presentation documented in this study usingcell culture models bear a direct relationship to the cellularimmune responses in animal models and in humans.

ACKNOWLEDGMENTS

This work was supported by grants R01 AI52007 and R21AI65312(P.S.). This work was supported in part by the cooperative agreementDAMD17-98-2-8007 between the U.S. Army Medical Research andMateriel Command, the Henry M. Jackson Foundation for the Ad-vancement of Military Medicine, and the Military Infectious DiseaseResearch Program.

We are grateful to investigators at Aventis for reagents and adviceand to Linda Wyatt and Bernard Moss for key reagents for generatingMVA recombinant viruses.

The views and opinions expressed herein are those of the authorsand do not purport to reflect the official policy or position of theDepartment of Defense.

REFERENCES

1. Amara, R. R., F. Villinger, J. D. Altman, S. L. Lydy, S. P. O’Neil, S. I.Staprans, D. C. Montefiori, Y. Xu, J. G. Herndon, L. S. Wyatt, M. A.Candido, N. L. Kozyr, P. L. Earl, J. M. Smith, H. L. Ma, B. D. Grimm, M. L.Hulsey, J. Miller, H. M. McClure, J. M. McNicholl, B. Moss, and H. L.Robinson. 2001. Control of a mucosal challenge and prevention of AIDS bya multiprotein DNA/MVA vaccine. Science 292:69–74.

2. Amendola, A., G. Lombardi, S. Oliverio, V. Colizzi, and M. Piacentini. 1994.HIV-1 gp120-dependent induction of apoptosis in antigen-specific human Tcell clones is characterized by ‘tissue’ transglutaminase expression and pre-vented by cyclosporin A. FEBS Lett. 339:258–264.

3. Bagetta, G., M. T. Corasaniti, W. Malorni, G. Rainaldi, L. Berliocchi, A.Finazzi-Agro, and G. Nistico. 1996. The HIV-1 gp120 causes ultrastructuralchanges typical of apoptosis in the rat cerebral cortex. Neuroreport 7:1722–1724.

4. Barouch, D. H., S. Santra, M. J. Kuroda, J. E. Schmitz, R. Plishka, A.Buckler-White, A. E. Gaitan, R. Zin, J. H. Nam, L. S. Wyatt, M. A. Lifton,C. E. Nickerson, B. Moss, D. C. Montefiori, V. M. Hirsch, and N. L. Letvin.2001. Reduction of simian-human immunodeficiency virus 89.6P viremia inrhesus monkeys by recombinant modified vaccinia virus Ankara vaccination.J. Virol. 75:5151–5158.

5. Belshe, R. B., G. J. Gorse, M. J. Mulligan, T. G. Evans, M. C. Keefer, J. L.Excler, A. M. Duliege, J. Tartaglia, W. I. Cox, J. McNamara, K. L. Hwang,A. Bradney, D. Montefiori, and K. J. Weinhold. 1998. Induction of immuneresponses to HIV-1 by canarypox virus (ALVAC) HIV-1 and gp120 SF-2recombinant vaccines in uninfected volunteers. AIDS 12:2407–2415.

6. Belshe, R. B., C. Stevens, G. J. Gorse, S. Buchbinder, K. Weinhold, H.Sheppard, D. Stablein, S. Self, J. McNamara, S. Frey, J. Flores, J. L. Excler,M. Klein, R. E. Habib, A. M. Duliege, C. Harro, L. Corey, M. Keefer, M.Mulligan, P. Wright, C. Celum, F. Judson, K. Mayer, D. McKirnan, M.Marmor, and G. Woody. 2001. Safety and immunogenicity of a canarypox-vectored human immunodeficiency virus type 1 vaccine with or withoutgp120: a phase 2 study in higher- and lower-risk volunteers. J. Infect. Dis.183:1343–1352.

7. Cao, H., P. Kaleebu, D. Hom, J. Flores, D. Agrawal, N. Jones, J. Serwanga,M. Okello, C. Walker, H. Sheppard, R. El-Habib, M. Klein, E. Mbidde, P.Mugyenyi, B. Walker, J. Ellner, and R. Mugerwa. 2003. Immunogenicity ofa recombinant human immunodeficiency virus (HIV)-canarypox vaccine inHIV-seronegative Ugandan volunteers: results of the HIV Network for Pre-vention Trials 007 Vaccine Study. J. Infect. Dis. 187:887–895.

8. Carroll, M. W., and B. Moss. 1995. E. coli beta-glucuronidase (GUS) as amarker for recombinant vaccinia viruses. BioTechniques 19:352–356.

9. Carroll, M. W., and B. Moss. 1997. Host range and cytopathogenicity of thehighly attenuated MVA strain of vaccinia virus: propagation and generationof recombinant viruses in a nonhuman mammalian cell line. Virology 238:198–211.

10. Cebere, I., L. Dorrell, H. McShane, A. Simmons, S. McCormack, C. Schmidt,C. Smith, M. Brooks, J. E. Roberts, S. C. Darwin, P. E. Fast, C. Conlon, S.Rowland-Jones, A. J. McMichael, and T. Hanke. 2006. Phase I clinical trial



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

safety of DNA- and modified virus Ankara-vectored human immunodefi-ciency virus type 1 (HIV-1) vaccines administered alone and in a prime-boostregime to healthy HIV-1-uninfected volunteers. Vaccine 24:417–425.

11. Chahroudi, A., D. A. Garber, P. Reeves, L. Liu, D. Kalman, and M. B.Feinberg. 2006. Differences and similarities in viral life cycle progression andhost cell physiology after infection of human dendritic cells with modifiedvaccinia virus Ankara and vaccinia virus. J. Virol. 80:8469–8481.

12. Chen, X., M. T. Rock, J. Hammonds, J. Tartaglia, A. Shintani, J. Currier, B.Slike, J. E. Crowe, Jr., M. Marovich, and P. Spearman. 2005. Pseudovirionparticle production by live poxvirus human immunodeficiency virus vaccinevector enhances humoral and cellular immune responses. J. Virol. 79:5537–5547.

13. Cox, W. I., J. Tartaglia, and E. Paoletti. 1993. Induction of cytotoxic Tlymphocytes by recombinant canarypox (ALVAC) and attenuated vaccinia(NYVAC) viruses expressing the HIV-1 envelope glycoprotein. Virology195:845–850.

14. Davies, M. V., H. W. Chang, B. L. Jacobs, and R. J. Kaufman. 1993. The E3Land K3L vaccinia virus gene products stimulate translation through inhibi-tion of the double-stranded RNA-dependent protein kinase by differentmechanisms. J. Virol. 67:1688–1692.

15. de Bruyn, G., A. J. Rossini, Y. L. Chiu, D. Holman, M. L. Elizaga, S. E. Frey,D. Burke, T. G. Evans, L. Corey, and M. C. Keefer. 2004. Safety profile ofrecombinant canarypox HIV vaccines. Vaccine 22:704–713.

16. Egan, M. A., W. A. Pavlat, J. Tartaglia, E. Paoletti, K. J. Weinhold, M. L.Clements, and R. F. Siliciano. 1995. Induction of human immunodeficiencyvirus type 1 (HIV-1)-specific cytolytic T lymphocyte responses in seronega-tive adults by a nonreplicating, host-range-restricted canarypox vector(ALVAC) carrying the HIV-1 MN env gene. J. Infect. Dis. 171:1623–1627.

17. Evans, T. G., M. C. Keefer, K. J. Weinhold, M. Wolff, D. Montefiori, G. J.Gorse, B. S. Graham, M. J. McElrath, M. L. Clements-Mann, M. J. Mulli-gan, P. Fast, M. C. Walker, J. L. Excler, A. M. Duliege, and J. Tartaglia.1999. A canarypox vaccine expressing multiple human immunodeficiencyvirus type 1 genes given alone or with rgp120 elicits broad and durableCD8� cytotoxic T lymphocyte responses in seronegative volunteers. J. In-fect. Dis. 180:290–298.

18. Everett, H., and G. McFadden. 2002. Poxviruses and apoptosis: a time to die.Curr. Opin. Microbiol. 5:395–402.

19. Fang, Z. Y., I. Kuli-Zade, and P. Spearman. 1999. Efficient human immu-nodeficiency virus (HIV)-1 Gag-Env pseudovirion formation elicited frommammalian cells by a canarypox HIV vaccine candidate. J. Infect. Dis.180:1122–1132.

20. Fang, Z. Y., K. Limbach, J. Tartaglia, J. Hammonds, X. Chen, and P.Spearman. 2001. Expression of vaccinia E3L and K3L genes by a novelrecombinant canarypox HIV vaccine vector enhances HIV-1 pseudovirionproduction and inhibits apoptosis in human cells. Virology 291:272–284.

21. Fischer, S. F., H. Ludwig, J. Holzapfel, M. Kvansakul, L. Chen, D. C. Huang,G. Sutter, M. Knese, and G. Hacker. 2006. Modified vaccinia virus Ankaraprotein F1L is a novel BH3-domain-binding protein and acts together withthe early viral protein E3L to block virus-associated apoptosis. Cell DeathDiffer. 13:109–118.

22. Franchini, G., S. Gurunathan, L. Baglyos, S. Plotkin, and J. Tartaglia. 2004.Poxvirus-based vaccine candidates for HIV: two decades of experience withspecial emphasis on canarypox vectors. Expert Rev. Vaccines 3:S75–S88.

23. Goepfert, P. A., H. Horton, M. J. McElrath, S. Gurunathan, G. Ferrari, G. D.Tomaras, D. C. Montefiori, M. Allen, Y. L. Chiu, P. Spearman, J. D. Fuchs,B. A. Koblin, W. A. Blattner, S. Frey, M. C. Keefer, L. R. Baden, and L.Corey. 2005. High-dose recombinant canarypox vaccine expressing HIV-1protein, in seronegative human subjects. J. Infect. Dis. 192:1249–1259.

24. Gupta, K., M. Hudgens, L. Corey, M. J. McElrath, K. Weinhold, D. C.Montefiori, G. J. Gorse, S. E. Frey, M. C. Keefer, T. G. Evans, R. Dolin, D. H.Schwartz, C. Harro, B. Graham, P. W. Spearman, M. Mulligan, and P.Goepfert. 2002. Safety and immunogenicity of a high-titered canarypox vac-cine in combination with rgp120 in a diverse population of HIV-1-uninfectedadults: AIDS Vaccine Evaluation Group protocol 022A. J. Acquir. ImmuneDefic. Syndr. 29:254–261.

25. Hanke, T., R. V. Samuel, T. J. Blanchard, V. C. Neumann, T. M. Allen, J. E.Boyson, S. A. Sharpe, N. Cook, G. L. Smith, D. I. Watkins, M. P. Cranage,and A. J. McMichael. 1999. Effective induction of simian immunodeficiencyvirus-specific cytotoxic T lymphocytes in macaques by using a multiepitopegene and DNA prime-modified vaccinia virus Ankara boost vaccinationregimen. J. Virol. 73:7524–7532.

26. Hel, Z., J. Nacsa, W. P. Tsai, A. Thornton, L. Giuliani, J. Tartaglia, and G.Franchini. 2002. Equivalent immunogenicity of the highly attenuated pox-virus-based ALVAC-SIV and NYVAC-SIV vaccine candidates in SIVmac251-infected macaques. Virology 304:125–134.

27. Herbein, G., U. Mahlknecht, F. Batliwalla, P. Gregersen, T. Pappas, J.Butler, W. A. O’Brien, and E. Verdin. 1998. Apoptosis of CD8� T cells ismediated by macrophages through interaction of HIV gp120 with chemokinereceptor CXCR4. Nature 395:189–194.

28. Hornemann, S., O. Harlin, C. Staib, S. Kisling, V. Erfle, B. Kaspers, G.

Hacker, and G. Sutter. 2003. Replication of modified vaccinia virus Ankarain primary chicken embryo fibroblasts requires expression of the interferonresistance gene E3L. J. Virol. 77:8394–8407.

29. Ignatius, R., M. Marovich, E. Mehlhop, L. Villamide, K. Mahnke, W. I. Cox,F. Isdell, S. S. Frankel, J. R. Mascola, R. M. Steinman, and M. Pope. 2000.Canarypox virus-induced maturation of dendritic cells is mediated by apop-totic cell death and tumor necrosis factor alpha secretion. J. Virol. 74:11329–11338.

30. Kapasi, A. A., G. Patel, N. Franki, and P. C. Singhal. 2002. HIV-1 gp120-induced tubular epithelial cell apoptosis is mediated through p38-MAPKphosphorylation. Mol. Med. 8:676–685.

31. Koopman, G., C. P. Reutelingsperger, G. A. Kuijten, R. M. Keehnen, S. T.Pals, and M. H. van Oers. 1994. Annexin V for flow cytometric detection ofphosphatidylserine expression on B cells undergoing apoptosis. Blood 84:1415–1420.

32. Lawson, V. A., K. A. Silburn, P. R. Gorry, G. Paukovic, D. F. Purcell, A. L.Greenway, and D. A. McPhee. 2004. Apoptosis induced in synchronizedhuman immunodeficiency virus type 1-infected primary peripheral bloodmononuclear cells is detected after the peak of CD4� T-lymphocyte loss andis dependent on the tropism of the gp120 envelope glycoprotein. Virology327:70–82.

33. Marovich, M. A., J. R. Mascola, M. A. Eller, M. K. Louder, P. A. Caudrelier,R. El-Habib, S. Ratto-Kim, J. H. Cox, J. R. Currier, B. L. Levine, C. H. June,W. B. Bernstein, M. L. Robb, B. Schuler-Thurner, R. M. Steinman, D. L.Birx, and S. Schlesinger-Frankel. 2002. Preparation of clinical-grade recom-binant canarypox-human immunodeficiency virus vaccine-loaded humandendritic cells. J. Infect. Dis. 186:1242–1252.

34. Montefiori, D. C., J. T. Safrit, S. L. Lydy, A. P. Barry, M. Bilska, H. T. Vo,M. Klein, J. Tartaglia, H. L. Robinson, and B. Rovinski. 2001. Induction ofneutralizing antibodies and Gag-specific cellular immune responses to an R5primary isolate of human immunodeficiency virus type 1 in rhesus macaques.J. Virol. 75:5879–5890.

35. Moss, B. 1992. Poxvirus expression vectors. Curr. Top. Microbiol. Immunol.158:25–38.

36. Moss, B. 1994. Replicating and host-restricted non-replicating vaccinia virusvectors for vaccine development. Dev. Biol. Stand. 82:55–63.

37. Najera, J. L., C. E. Gomez, E. Domingo-Gil, M. M. Gherardi, and M.Esteban. 2006. Cellular and biochemical differences between two attenuatedpoxvirus vaccine candidates (MVA and NYVAC) and role of the C7L gene.J. Virol. 80:6033–6047.

38. Pincus, S., J. Tartaglia, and E. Paoletti. 1995. Poxvirus-based vectors asvaccine candidates. Biologicals 23:159–164.

39. Pitisuttithum, P. 2005. HIV-1 prophylactic vaccine trials in Thailand. Curr.HIV Res. 3:17–30.

40. Pogo, B. G., S. M. Melana, and J. Blaho. 2004. Poxvirus infection andapoptosis. Int. Rev. Immunol. 23:61–74.

41. Pugliese, O., M. Boirivant, and M. Viora. 1997. Apoptosis induction byhuman immunodeficiency virus type 1 (HIV-1) gp120 peptides. Viral Immu-nol. 10:95–102.

42. Ranasinghe, C., J. C. Medveczky, D. Woltring, K. Gao, S. Thomson, B. E.Coupar, D. B. Boyle, A. J. Ramsay, and I. A. Ramshaw. 2006. Evaluation offowlpox-vaccinia virus prime-boost vaccine strategies for high-level mucosaland systemic immunity against HIV-1. Vaccine 24:5881–5895.

43. Reimann, J., and R. Schirmbeck. 1999. Alternative pathways for processingexogenous and endogenous antigens that can generate peptides for MHCclass I-restricted presentation. Immunol. Rev. 172:131–152.

44. Russell, N. D., B. S. Graham, M. C. Keefer, M. J. McElrath, S. G. Self, K. J.Weinhold, D. C. Montefiori, G. Ferrari, H. Horton, G. D. Tomaras, S.Gurunathan, L. Baglyos, S. E. Frey, M. J. Mulligan, C. D. Harro, S. P.Buchbinder, L. R. Baden, W. A. Blattner, B. A. Koblin, and L. Corey. 2007.Phase 2 study of an HIV-1 canarypox vaccine (vCP1452) alone and incombination with rgp120: negative results fail to trigger a phase 3 correlatestrial. J. Acquir. Immune Defic. Syndr. 44:203–212.

45. Sancho, M. C., S. Schleich, G. Griffiths, and J. Krijnse-Locker. 2002. Theblock in assembly of modified vaccinia virus Ankara in HeLa cells revealsnew insights into vaccinia virus morphogenesis. J. Virol. 76:8318–8334.

46. Singh-Sandhu, D., and J. Tartaglia. 2003. Poxviruses as immunizationvehicles, p. 1297–1316. In S. A. Plotkin (ed.), Vaccines, 4th ed. Saunders,Philadelphia, PA.

47. Tartaglia, J., M. E. Perkus, J. Taylor, E. K. Norton, J. C. Audonnet, W. I.Cox, S. W. Davis, J. van der Hoeven, B. Meignier, M. Riviere, et al. 1992.NYVAC: a highly attenuated strain of vaccinia virus. Virology 188:217–232.

48. Tartaglia, J., S. Pincus, and E. Paoletti. 1990. Poxvirus-based vectors asvaccine candidates. Crit. Rev. Immunol. 10:13–30.

49. Taylor, J., B. Meignier, J. Tartaglia, B. Languet, J. Van der Hoeven, G.Franchini, C. Trimarchi, and E. Paoletti. 1995. Biological and immunogenicproperties of a canarypox-rabies recombinant, ALVAC-RG (vCP65) in non-avian species. Vaccine 13:539–549.

50. Wyatt, L. S., B. Moss, and S. Rozenblatt. 1995. Replication-deficient vacciniavirus encoding bacteriophage T7 RNA polymerase for transient gene ex-pression in mammalian cells. Virology 210:202–205.



EN

N S

TA

TE

UN

IVhttp://jvi.asm

.org/D

ownloaded from

http://jvi.asm.org/

Vaccine Vectors Ankara-Human Immunodeficiency Virus Virus- and ...

Documents

Transcript of Vaccine Vectors Ankara-Human Immunodeficiency Virus Virus- and ...