of January 26, 2015.This information is current as

HIV-1 Env in a Dose-Dependent Mannerthe Cellular Immune Response AgainstVirus Attenuates the Vector and Enhances IL-12 Delivery from Recombinant Vaccinia

Zavala and Mariano EstebanRodríguez, Juan R. Rodríguez, Gen-Ichiro Sano, Fidel M. Magdalena Gherardi, Juan C. Ramirez, Dolores

http://www.jimmunol.org/content/162/11/67241999; 162:6724-6733; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/162/11/6724.full#ref-list-1

, 32 of which you can access for free at: cites 55 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

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

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1999 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on January 26, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

IL-12 Delivery from Recombinant Vaccinia Virus Attenuatesthe Vector and Enhances the Cellular Immune ResponseAgainst HIV-1 Env in a Dose-Dependent Manner1

M. Magdalena Gherardi,2* Juan C. Ramirez,2* Dolores Rodrıguez,* Juan R. Rodrıguez,*Gen-Ichiro Sano,† Fidel Zavala,† and Mariano Esteban3*

To develop vaccination strategies against HIV-1 infection aimed to specifically enhance the cell-mediated immunity (CMI), wehave engineered vaccinia virus (VV) recombinants expressing HIV-1 Env (rVVenv) and murine IL-12 (rVVlucIL-12) genes orcoexpressing both genes (rVVenvIL-12). In mice inoculated with rVVlucIL-12 there is a rapid clearance of the virus, and thiscorrelates with the induction of high levels of IL-12 and IFN-g in serum and spleen early after infection. Enzyme-linked immu-nospot analysis of mice inoculated with rVVlucIL-12, revealed a nearly 2-fold increase in the number of specific anti-VV CD81

T cells compared with that in mice given control rVV, and the serum Ab response was biased in favor of a Th1 response. Anenhancement of about 2-fold in the number of anti-gp160 IFN-g-secreting CD81 T cells was observed in mice inoculated withrVVenvIL-12, when a dose of 13 107 PFU/mouse was used, but this enhancement was not observed when mice were given 53

107 PFU. This variation with virus dosage was confirmed in mice immunized simultaneously with different multiplicities of rVVexpressing singly theenvor IL-12 genes. The highest specific CMI was obtained in mice coadministered a low dose (23 104 PFU)of rVVlucIL-12 and 1 3 107 PFU of rVVenv. Our findings provide evidence for specific enhancement of the CMI to HIV-1 Envby the differential expression of IL-12 andenvgenes delivered from VV recombinants. This approach can be of wide vaccinationinterest as a means to improve immune responses to other Ags.The Journal of Immunology,1999, 162: 6724–6733.

L ive-based vectors represent a promising area of vaccineresearch, and probably the best studied vector is vacciniavirus (VV),4 the prototype member of the poxvirus fam-

ily, that was successfully used as a live vaccine to eradicate small-pox (1). This virus represents a good candidate for vaccinationpurposes because of its broad host range and the ability to generaterecombinant viruses (rVV) that express a variety of foreign Ags(2). Moreover, rVV have been proven effective in field animalvaccination programs (3). When rVV containing HIV-1 and SIVgenes were tested in different vaccination schedules on simianhosts, specific cellular and humoral immunities, both systemic (4)and mucosal (5), were elicited. The above findings suggestthatVV-based vaccination approaches using highly attenuated strainsmight be a promising prophylactic strategy against HIV-1 infection.

The understanding of the immune response generated duringHIV-1 infection has revealed that neutralizing Abs arise, but maynot be critical in limiting viral replication due to genetic variabilityof the virus (6). In addition, different vaccination approaches insimian animal models or in human trials failed to elicit protectiveneutralizing Abs (7). On the contrary, several studies have empha-sized the importance of CTL in combating HIV-1 infection andcontrolling the development of AIDS. CTL activity coincides withthe early containment of HIV-1 replication (7) and correlates withstable clinical status and low virus load in chronically infectedindividuals (8–10). Moreover, high anti-HIV CTL responses arefrequently observed in healthy women repeatedly exposed to thevirus during unprotective sexual practices (11) and have also beendemonstrated in uninfected infants born to HIV-1-positive mothers(12). In the SIV infection model, protection after vaccination hasbeen correlated with a strong CTL response over other immuno-logical parameters (13). Consequently, of a spectrum of varioushost immune responses, induction of cell-mediated immunity(CMI) might be an important requirement in an effective candidateHIV-1 vaccine.

The development of an effective CMI response after vaccinationrests on an extensive scope of factors, among which cytokinespresent during the priming could play a critical role. Two differentsubsets of CD41 T lymphocytes, Th1 and Th2, differing in thepattern of cytokines produced, have been described to be crucial inthe generation of a cellular or an Ab immune response, respec-tively. Different lines of evidence showed that the early decisiontoward Th1- or Th2-type immune response is mainly dependent onthe balance between IL-12 (which favors a Th1 response) and IL-4(which favors a Th2 response). Among the main functional fea-tures of IL-12 are 1) it potentiates cytokine production, particularlyIFN-g, in T lymphocytes and NK cells; 2) it acts as a growth factorfor preactivated T and NK cells; and 3) it is involved in the

*Department of Molecular and Cellular Biology, Centro Nacional de Biotecnologia,Consejo Superior de Investigaciones Cientificas, Campus Universidad Autonoma,Madrid, Spain; and†Department of Medical and Molecular Parasitology, New YorkUniversity Medical Center, New York, NY 10010

Received for publication December 22, 1998. Accepted for publication March 16, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by Grant 08.6/0020/97 from the Comunidad Autonoma deMadrid, Grant SAF98-0056 from the Comision Interministerial de Ciencia y Tecno-logia from Spain, and postdoctoral fellowships from Consejo Nacional de Investiga-ciones Cientıficas y Tecnicas de Argentina (to M.M.G.), and Comunidad Autonomade Madrid, Spain (to J.C.R.).2 M.M.G. and J.C.R. contributed equally in the realization of this work.3 Address correspondence and reprint requests to Dr. Mariano Esteban, Centro Na-cional Biotecnologia, Campus Cantoblanco, 28049 Madrid, Spain. E-mail address:mesteban@cnb.uam.es4 Abbreviations used in this paper: VV, vaccinia virus; CMI, cell-mediated immunity;HA, hemagglutinin; hpi, hours postinoculation; IRES, internal ribosomal entry sitesequence; dpi, days postinfection; HPRT, hypoxanthine phosphoribosyltransferase;PBS-T, PBS plus 0.05% Tween-20; ELISPOT, enzyme-linked immunospot.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

generation of CTLs and in the activation of cytotoxicity in bothCD81 T and NK cells. In addition, IL-12 has a prominent role inthe generation of Th1 cells and the optimal differentiation of CTLs(14). Thus, to induce strong and stable CMI responses againstHIV-1 infection, the use of vectors delivering cytokines capable oftriggering a Th1 response in conjunction with appropriate Ags is aencouraging approach. In this regard, different vaccination strate-gies with IL-12 delivered as a soluble product, expressed fromDNA vectors or viral vectors, have provided evidence for enhance-ment of CTL responses that correlated with regression of tumors(15), resolution of autoimmune diseases (16), and protectionagainst various intracellular pathogens (17) and against the devel-opment of murine AIDS (18).

In addition to the promising prophylactic and therapeutic ad-vantages of the use of live vectors delivering immunomodulators,the potential capability of the encoded cytokine to modulate thevector pathogenicity might be a desirable property, especially inthe case of live virus-based vaccines by decreasing the risk ofside-effects during virus infections. Indeed, different recombinantvirus-encoding cytokines have been described as immunologicaltools for dissecting the in vivo functions of cytokines in antiviralimmunity. Expression of cytokines from rVV have been reportedto have a profound effect on viral infection (19), but their inductionand their involvement in promoting specific immune responses toAgs are poorly characterized events. Hence, analysis of the effectof cytokines on the virus vector itself should be explored to un-derstand the implications for the Ag-specific immune responseselicited when both Ag and cytokine are delivered from a live-basedvector.

In this investigation we have defined the antiviral and immuno-logical roles of IL-12 when expressed from VV in the absence andthe presence of HIV-1 Env. Our findings demonstrate that rVVexpressing IL-12 genes are safe vectors, since VV replication isseverely compromised through the induction of IFN-g. In immu-nized mice IL-12 expression drives a Th1-type response againstboth the vector and theenvgene product, leading to an enhancedcellular immune response and a biased serum Ab response in favorof IgG2a subclass. Moreover, we show that the dosage of the rVVexpressing IL-12 andenvgenes plays an essential role in the en-hancement of the cellular immune response against the HIV-1gp160 protein, and that by coexpressing IL-12 andenv genes indifferent ratios it is possible to trigger a desirable cellular immuneresponse to the HIV-1 Ag.

Materials and MethodsViruses and cells

The VV recombinants employed in this study derived from the wild-typeWR strain, rVVluc (expressing the luciferase andb-galactosidase genes)and rVVenv (expressing the entire env gene of HIV-1 strain IIIB andb-ga-lactosidase gene), have been described previously (20, 21). The recombi-nant viruses rVVlucIL-12 (expressing the luciferase gene and p35 and p40IL-12 subunits), its control rVVlucHA- (expressing the luciferase geneand insertional inactivated in the HA gene), as well as rVVenvIL-12 andrVVenvHA- were generated for this study, and their construction is de-scribed below. Viruses were grown in HeLa cells, titrated in African greenmonkey kidney BSC-40 cells, and purified as described previously (22).

Engineering of the VV recombinant viruses

The cDNAs coding for both IL-12 subunits (p35 and p40) linked by aninternal ribosomal entry site sequence (IRES) were isolated from plasmidpBS-IL-12 by digestion with the restriction enzymesEcoRI andBamHI.The DNA fragment containing the complete IL-12 sequence (p35-IRES-p40) was blunt ended by treatment with the large fragment of theEsche-richia coli DNA polymerase I (Klenow) and cloned into theSmaI site ofthe VV insertion vector pJR101. As a result of this cloning strategy, weisolated a plasmid, pJR101-IL-12, that contains the IL-12 (p35-IRES-p40cassette) genes under the control of a VV synthetic early/late promoter e/l

(23), theE. coli b-glucuronidase marker gene under the control of the VVearly/late promoter p7.5, and all these sequences flanked by regions fromVV hemagglutinin (HA) gene. Double rVV were prepared by infectingBSC-40 cells with either the VVenv or the VVluc recombinant virus andtransfecting them with the plasmid pJR101-IL-12. Cell cultures were har-vested at 48 h postinoculation (hpi), and the double-recombinant viruseswere selected after plaque assay by the addition of X-Gluc to the agaroverlay (24). By this procedure, rVV containing the HIV-1env gene(VVenvIL-12) or the luciferase gene (VVlucIL-12) into the TK region andthe IL-12 cassette into the HA locus were isolated. After three rounds ofselection, viruses were purified following standard procedures. A simi-lar strategy was followed to generate control viruses VVlucHA- orVVenvHA-, but in this case, transfection was performed with empty VVinsertion plasmid pJR101. In Fig. 1A is shown a schematic representationof the different rVV constructed for this study. In Fig. 1B is shown IL-12expression in extracts from rVV BSC-40-infected cells by Western blotanalysis. An IL-12 bioassay was also performed with the same samples,indicating that IL-12 expressed from rVV was bioactive (data not shown).

Immunizations of mice and serum sample collection

BALB/c mice (H-2d; 6–8 wk old) were immunized i.p. with different doses(indicated as PFU) of the different rVV in 200ml of sterile PBS. Fourteendays after virus inoculation, blood was obtained from the retro-orbitalplexus by a heparinized capillary tube, collected in an Eppendorf tube, andcentrifuged, and serum was obtained and stored at220°C.

Measurement of luciferase activity in mice tissues

Replication of rVV in different mouse tissues was followed by a highlysensitive luciferase assay, previously described (20). Different groups of

FIGURE 1. Characterization of the different rVV generated.A, Schemeof the rVV genomes. The genes coding for luciferase (Luc) and HIV-1 Env(Env) proteins were inserted into the thymidine kinase (TK) locus of theVV genome (M). The DNA cassette containing the genes coding for IL-12was inserted into the HA locus (f) to generate the double recombinants,rVVlucIL-12 and rVVenvIL-12. p11, p7.5, and e/l represent different VVpromoters.B, Western blot analysis of extracts from cells infected with thedifferent rVV. Monolayers of BSC-40 cells were infected (10 PFU/cell)with the indicated rVV. The proteins were fractionated by SDS-PAGEunder reducing conditions, transferred to nitrocellulose paper, and reactedwith an anti-gp120 rabbit polyclonal Ab (left panel) or an anti-IL-12 p40rat mAb (right panel). Ab reactivity was detected by immunoperoxidasestaining using standard procedures.

6725The Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

mice received an i.p. inoculation with 53 107 PFU/mouse of the recom-binant viruses: rVVluc, rVVlucHA2, or rVVlucIL-12. At various timespostinoculation animals were sacrificed, and spleens, livers, and ovarieswere resected, washed with sterile PBS, weighed, and stored at270°C.Then, tissues from individual mice were homogenized in luciferase extrac-tion buffer (300ml/spleen extract and 100ml/ovary extract) containing 1%Triton X-100, 25 mM glycylglycine (pH 7.8), 15 mM MgSO4, 4 mMEGTA, 1 mM DTT, 1 mM PMSF, 100mg/ml soybean trypsin inhibitor,and 10mg/ml leupeptin. The luciferase activity was measured in the pres-ence of luciferin and ATP using a Lumat LB 9501 Berthold luminometer(Berthold, Nashua, NH), and it was expressed as relative luciferase unitsper milligram of protein. Protein content in tissue extracts was measuredemploying the bicinchoninic acid protein assay reagent kit (Pierce,Rockford, IL).

RNA extraction

Total mRNA was purified from aseptically removed spleens. Relativelyidentical small pieces of spleens from three mice per group were pooledand homogenized in extraction buffer using an Ultraturrax T8 mechanicalhomogenizer (Janke & Kunkel, Staufen, Germany). Clear lysates weresource for mRNA purification using QuickPrep Micro mRNA purifi-cation kit (Pharmacia, Uppsala, Sweden) following the instructions of themanufacturer.

Amplification of mRNA by RT-PCR

Semiquantitative RT-PCR was performed on mRNA using the SuperScriptOne-Step RT-PCR System (Life Technologies, Gaithersburg, MD). TheRT conditions used were 30 min at 50°C followed by a denaturing step at94°C for 2 min, followed by 30 (for hypoxanthine phosphoribosyltrans-ferase, HPRT) or 40 (for IFN-g and IL-12) cycles. The number of cycleswas adjusted for every pair of primers to get a linear range during thereactions. Cycling conditions were 94°C for 30 s, 60°C (58°C for IL-12)for 30 s, and 68°C for 1.5 min, followed by a final extension step at 68°Cfor 5 min. Primers sequences were: for HPRT, 59-GTTGGATACAGGCCAGACTTTGTTG-39 (sense) and 59-GATTCAACTTGCGCTCATCTTAGGC-39 (antisense); for IFN-g, 59-TGAACGCTACACACTGCATCTTGG-39 (sense) and 59-CGACTCCTTTTCCGCTTCCTGAG-39(anti-sense); and for IL-12 p40 subunit, 59-CTCACATCTGCTGCTCCACAA-39 (sense) and 59-CTCCTTCATCTTTTCTTTCTT-39(antisense).PCR products were analyzed by ethidium bromide staining after electro-phoresis on 1.2% agarose gels.

Ab measurements by ELISA

ELISA was used to determine the presence of Abs against VV Ags inserum samples. The VV Ags employed to coat 96-well flat-bottom platesat a concentration of 1mg/ml consisted of envelope proteins extracted frompurified virions, as described previously (25). VV Ags were suspended incarbonate buffer, pH 9.6, plated at 50ml/well, and incubated overnight at4°C. Afterward, the contents of the wells were discarded and washed threetimes with PBS plus 0.05% Tween-20 (PBS-T), and blocking buffer (bo-rate-buffered saline with 1% BSA, 1 mM EDTA, and 0.05% Tween-20)was added at 100ml/well and incubated for 1 h at37°C. The plates werewashed once with PBS-T, and samples diluted in blocking buffer wereadded in a volume of 100ml/well and incubated for 1 h at 37°C. Then,plates were washed three times before the detection Ab was added. Per-oxidase-conjugated goat anti-mouse IgG, IgG1, or IgG2a (Southern Bio-technology Associates, Birmingham, AL) Abs were diluted 1/1000,1/1500, and 1/2000 respectively in blocking buffer and incubated for 1 h at37°C. After washing the plates three times with PBS-T, the wells werereacted with the peroxidase substrateO-phenylendiamine dihydrochloride(Sigma, St. Louis, MO). After 10–15 min of incubation at room temper-ature, the reaction was stopped by adding 2 N H2SO4, and the absorbancevalues were measured at 492 nm on a Labsystems Multiskan Plus platereader (Chicago, IL).

T cell proliferation assays

Lymphocytes were removed from spleens by passing tissues through asterile mesh to obtain cell suspensions. Cells were suspended in completemedium (RPMI 1640 supplemented with 10% FCS, 2 mML-glutamine,and 10 mM 2-ME). RBC in preparations of spleen cells were lysed with 0.1M ammonium chloride buffer. Splenocytes were cultured in triplicate (106

cells/well) in 96-well microtiter flat-bottom plates and stimulated with pu-rified VV previously inactivated by UV light at 1mg/ml, purified gp160protein (Intracel, Cambridge, MA; 1mg/ml), or Con A (1mg/ml; Sigma).Plates were incubated for 3 days at 37°C in 5% CO2. After this incubationperiod, proliferation assays were conducted by labeling the cells with

[3H]thymidine (1 mCi/well) for 18 h. Following automated harvesting,[3H]thymidine incorporation was measured by liquid scintillation counting.Cytokine levels in culture supernatants were determined after 48 h (IL-10)or 72 h (IFN-g) of incubation. Supernatants from triplicate cultures werepooled and stored at270°C until performing the assay.

Evaluation of cytokines levels by ELISA

Cytokine levels in culture supernatants and sera were determined byELISA using the appropriate combination of Abs from Genzyme (Cam-bridge, MA). Briefly, 96-well flat-bottom plates were coated with 100ml ofanti-cytokine Abs diluted in the buffer specified by the manufacturer andincubated overnight at 4°C. The wells were then washed with PBS-T andcoated with PBS containing 1% BSA at 37°C for 2 h. Serial 2-fold dilutionsof supernatants or sera and adequate dilutions of standard cytokines wereadded in duplicate and incubated at 37°C for 1–2 h. The wells were thenwashed with PBS-T and incubated with the specific biotinylated anti-cytokine Ab diluted in PBS-T with 1% BSA for 1–2 h. After three or fourwashes, wells were incubated with HRP-conjugated streptavidin for 15 minat 37°C and developed with TMB reagent (Sigma). The reaction stoppedwith 2 N SO4H2, and the absorbance values were measured at 450 nm.

Evaluation of CD81 T cells by the ELISPOT assay

The ELISPOT assay to detect Ag-specific CD81 T cells was performed aspreviously described (26). Briefly, 96-well nitrocellulose plates werecoated with 8mg/ml of anti-mouse IFN-g mAb R4–6A2 (PharMingen,San Diego, CA) in 75ml of PBS. After overnight incubation at roomtemperature, wells were washed three times with RPMI 1640, and 100mlof complete medium supplemented with 10% FCS was added to each well.Afterward, the plate was incubated at 37°C for at 1 h. Spleen cells (de-pleted of RBC) from different groups of mice were added in triplicate at2-fold dilutions. P815 cells (a mastocytome cell line that expresses onlyMHC class I molecules) were used as APC. When the number of specificCD81 T cell anti-VV Ags was evaluated, P815 cells (107 cells/ml) wereinfected at a multiplicity of infection of 5 PFU/cell, and at 4.5 hpi, cellswere washed and treated with mitomycin C (30mg/ml; Sigma). When thenumber of CD81 IFN-g-secreting cells specific for the V3 loop epitope ofthe HIV-1 Env protein was evaluated, P815 cells were pulsed with 1026 Mof the synthetic peptide GPGRATVTI (9 Env) or RGPGRAFVTI (10 Env)and treated with mitomycin C as described above. After several washeswith culture medium, 105 P815 cells were added to each well. As a control,P815 cells not pulsed with the peptide or uninfected but treated undersimilar conditions were used. Plates were incubated for 24 h in a 37°Cincubator with a 5% CO2 atmosphere, washed extensively with PBS-T, andincubated overnight at 4°C with a solution of 2mg/ml of biotinylatedanti-mouse IFN-g mAb XMG1.2 (PharMingen) in PBS-T. Thereafter,plates were washed with PBS-T, and 100ml of peroxidase-labeled avidin(Sigma) at a 1/800 dilution in PBS-T was added to each well and incubatedat room temperature. One hour later, wells were washed with PBS-T andPBS. The spots were developed by adding 1mg/ml of the substrate 3,39-diaminobenzidine tetrahydrochloride (Sigma) in 50 mM Tris-HCl, pH 7.5,containing 0,015% hydrogen peroxide. Then spots were counted with theaid of a stereomicroscope.

IL-12 bioassay

Biologically active IL-12 was measured in supernatants from rVV express-ing IL-12 BSC-40-infected cells and in serum samples from inoculatedmice. The indicator WEHI 279 cells (European Cell Culture Collection,Salisbury, U.K.), a mouse B cell lymphoma, was used to titrate the bio-active IL-12 as previously described (27). Briefly, flat-bottom 96-welldishes were coated with a rat mAb against mouse IL-12 (nonneutralizingC15 rat mAb, Genzyme) and incubated overnight at 4°C. Afterward, plateswere washed and blocked with filtered PBS-T with 1% BSA for 1 h at37°C. Dilutions of samples and standard recombinant mouse IL-12 (Gen-zyme) were added and stored at 37°C for 4 h. Freshly prepared splenocyteswere added at 106 cells/well and cultured for 48 h in the presence of 50U/ml of rIL-2. Culture supernatants were mixed with 104 cells/well ofexponentially growing WEHI 279 cells and cultured for an additional 72 h.IFN-g inhibition of indicator cells growth was measured by the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide colorimetric assay.Absorbance values were introduced in the standard curve and transformedto IL-12 units according to the manufacturer’s data (8000 U/mg).

CD81 T cell purification

CD81 cells were purified using MACS High Gradient Magnetic SeparationColumns VS1 (Miltenyi Biotec, Bergisch Gladbach, Germany) for positiveselection from whole splenic populations following the manufacturer’s

6726 IL-12 AND IMMUNE RESPONSES TO VV AND HIV-1 Env

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

procedure. Cells were labeled with CD8a (Ly2.53–6.7) Micro Beads(Miltenyi Biotec), and 108 cells/ml were loaded in the column. A fractionof the whole population or positive and negative eluates were formalde-hyde fixed and labeled with fluorescence-specific anti-CD4 and anti-CD8Abs for cell sorting by flow cytometry (FACS, Becton Dickinson, Moun-tain View, CA). The sorted profiles were used to evaluate the accuracy ofthe purification and quantitate the ELISPOT assay performed with thosepopulations. Less than 3% of CD81 cells were present in the CD81-de-pleted fraction.

ResultsEffects of IL-12 expression on VV replication and viruspersistence in mice

Little information is available on the effect of IL-12 on VV infec-tion; hence, our first goal was to evaluate the effect of IL-12 on thecapability of the viral vector to infect and persist in target tissues.Thus, we studied the replication of a WR-based rVV that expressesthe IL-12 gene (rVVlucIL-12) by measuring the activity of thecoexpressed luciferase reporter gene. As the IL-12 expression cas-sette was inserted into the HA gene, and as it is known that inac-tivation of this gene reduces VV infectivity in vivo (28), we em-ployed as controls two luciferase expression viruses rVVlucHA2

and rVVlucHA1 with an inactivated or an intact HA gene, respec-tively. To evaluate in vitro whether IL-12 expression could haveany effect on VV replication, BSC40 cells were infected (10 PFU/cell) with the three different VV recombinants (rVVlucIL-12,rVVlucHA-, and rVVluc) and at different times postinfection (5,18, and 24 hpi) luciferase activity was measured in cell extracts.No significant differences in luciferase activity were observedamong the different rVV (data not shown), indicating that IL-12expression had no effect on viral replication. To evaluate in vivothe effect of IL-12 expression on VV replication, groups of micewere i.p. inoculated with a single dose of 53 107 PFU/mice ofrVVluc, rVVlucHA 2, or rVVlucIL-12, and three animals per dayand group were sacrificed on days 1, 2, 3, 4, and 7 postinoculation(dpi). Luciferase activity in spleen and ovaries of individual sam-ples were measured, and results are depicted in Fig. 2. At 1 dpi,similar luciferase values were detected in spleen samples frommice inoculated with either rVVlucIL-12 or rVVlucHA2, andthese were about 50-fold lower than those found after infection

with rVVluc. However, at 2 dpi luciferase levels fell 100-fold inanimals given the IL-12-expressing virus, and no infectious viruswas detected by plaque assay in this group (not shown). By thistime, levels of replication of rVVlucHA2 or rVVluc control vi-ruses remained essentially stable. At 3 dpi only background lucif-erase activity was measured in all mice given the rVVlucIL-12,whereas low, but still measurable, luciferase activity was detectedin mice inoculated with control viruses. The kinetics of therVVlucHA2 virus were parallel to those of control rVVluc virus,but infection in the spleen was resolved later in animals inoculatedwith rVVluc virus.

Ovaries are target tissues of VV infection on the mouse, andviral clearance from this organ is delayed with respect to that fromspleen or liver (29). As observed in Fig. 2 multiplication of rVV inovaries was extended for the three viruses analyzed compared withthat in the spleen. In this organ, differences between rVVlucHA2

and rVVlucIL-12 were more pronounced than those in spleen. At1 dpi, luciferase activity in ovaries was 50-fold lower in animalsinoculated with rVV expressing IL-12 than in animals inoculatedwith the control virus (rVVlucHA2). By 2 dpi, luciferase activityreached a peak in both rVVlucHA2- and rVVluc-inoculatedgroups, whereas 100-fold lower activity was detected in mice in-fected with rVVlucIL-12. In all groups, luciferase expression inovaries declined by day 3 and was greatly reduced by 4 dpi inanimals receiving IL-12. A slower decrease in luciferase levelswas observed in mice given control viruses, especially in thosereceiving rVVluc.

The results shown in Fig. 2 demonstrate that IL-12 expressionduring VV infection impairs virus multiplication and promotes arapid clearance of the virus from infected tissues, suggesting thatdelivery of IL-12 ensures attenuation of VV along the infectiousprocess in mice.

Induction of IL-12 and IFN-g in vivo after inoculation of micewith a rVV expressing IL-12

To determine whether the decreased infectivity observed forrVVlucIL-12 in mice could be correlated with induction of IFN-gexpression, we next analyzed the levels of IL-12 and IFN-gin serum samples from mice inoculated with rVVlucIL-12 orrVVlucHA2. Due to the rapid clearance observed for VV afterIL-12 expression (Fig. 2), we examined the levels of these cyto-kines at early times postinfection. Thus, serum samples werecollected every 6 h during the first dpi and daily thereafter until 7dpi. IL-12 was measured both by ELISA against the p40 subunit(data not shown) and by determination of bioactive p70 het-erodimeric form (Fig. 3A, upper panel), rendering comparable re-sults. The maximum amount of IL-12 was found at 6 hpi in theinoculated groups. Nevertheless, at least 5-fold higher levels werefound in samples from rVVlucIL-12-immunized mice than in an-imals inoculated with the control virus. Levels of IL-12 in thecontrol group fell into the detection limit of the assay (,0.64U/ml) beyond 12 hpi and were the same as those found in naivemice (not shown). The higher levels of IL-12 (ranging from 480–240 U/ml) were found during the first dpi in samples from miceinoculated with rVVlucIL-12, but IL-12 was still detected at 3 and4 dpi. This result clearly demonstrates that there is a rapid induc-tion of IL-12 upon infection with rVV expressing this cytokine,and significant levels are still present during the clearance periodof the virus (see Fig. 2).

High levels of IFN-g, evaluated by ELISA, were detected inserum from mice inoculated with IL-12 expressing rVV. The max-imum level of IFN-g appeared, with a lag of 12 h with respect toIL-12 production, around 18 hpi in rVVlucIL-12-immunized mice.High levels were still present at 2 dpi, falling to background levels

FIGURE 2. Expression of IL-12 by rVV severely compromises virusreplication in mice. BALB/c mice were inoculated i.p. with 53 107 PFU/animal of rVVluc (M ), rVVlucHA2 (E), or rVVlucIL-12 (F), and ovariesand spleens were collected on the indicated days. The extent of virus geneexpression was evaluated by the luciferase assay as described inMaterialsand Methods. Luciferase activity in the different homogenate tissue sam-ples was measured, and values were related to the amount of proteinpresent in tissue extracts (relative luciferase units (RLU) per milligrams ofprotein). Background levels found in control uninfected tissues were 3 and4 log10 (RLU per milligrams of protein) for spleen and ovaries, respec-tively. Results represent mean values from samples of three animals perday and group with the SD. Similar results were obtained in two indepen-dent experiments.

6727The Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

by 4 dpi, while in the control group detectable levels were onlyfound at 6 and 12 hpi (Fig. 3A, lower panel).

The kinetics of mRNA expression for the inducible p40 subunitof IL-12 and for IFN-g in the spleen were analyzed by semiquan-titative RT-PCR (Fig. 3B). Induction of IL-12 mRNA was ob-served at 6 hpi in mice inoculated with rVVlucIL-12, increasedabout 10 times by 18 hpi and declined by 1 dpi, whereas in miceinoculated with the control rVVlucHA2 virus, low but consistentamounts of p40 mRNA were present at 18 hpi. Levels of IFN-gmRNA in the spleens of mice inoculated with the rVVluc-IL12peaked at 24 hpi and were detectable until 2 dpi (Fig. 3B and datanot shown). The induction of IFN-g mRNA was clearly detected at18 hpi in mice inoculated with the control rVV, and the levels were.10 times lower than those in mice inoculated with rVVlucIL-12.Under the conditions of the assay, spleens from naive noninocu-lated mice did not reveal detectable mRNAs for the two cytokines.

Evaluation of the anti-VV immune response elicited in miceinoculated with rVVlucIL-12

The findings presented in Figs. 2 and 3 clearly reveal that IL-12has a profound effect on the replication of VV and that this cyto-kine potentiates IFN-g production. As both cytokines could have amajor impact on the modulation of host immune responses, ournext approach was to analyze the role of IL-12 in specific immuneresponses against the VV vector.

CD81 IFN-g-secreting T cells against VV

As IL-12 has the capacity to augment cell-mediated immune re-sponses (30, 31), and as CD81 CTL responses are involved in theresolution of infection by poxviruses (32, 33), we first evaluatedthe CD81 T cell immune response elicited against VV followingexpression of IL-12 in mice inoculated with rVVlucIL-12. Weemployed a modification (34) of the ELISPOT assay that quanti-fies the number of specific anti-VV MHC class I-restricted IFN-g-secreting cells. Groups of three or four mice were i.p. inoculatedwith 5 3 107 PFU/mice of rVVlucIL-12 or rVVluc-HA2, and at 7or 14 dpi the number of anti-VV IFN-g-secreting CD81 T cellswas determined. As shown in Fig. 4A, at 7 dpi the numbers ofspecific IFN-g-secreting CD81 T cells in spleens from animalsinoculated with rVVlucIL-12 (16416 104/106 cells) were signif-icantly different (p , 0.05) with respect to those in mice inocu-lated with the control virus (9646 64/106 cells). This represents1.7-fold more MHC I class-restricted anti-VV IFN-g-secretingcells in animals inoculated with rVVlucIL-12 virus than in thecontrol group. As expected, by 14 dpi the number of specific anti-VV CD81 T cell IFN-g-secreting cells decreased in the two groupsof animals (Fig. 4A), but differences between the groups weremaintained. The number of specific CD81 T cells quantified afterpurification revealed no differences when the total population wascompared with the CD81-selected fraction (Fig. 4B). These ob-servations confirm that in the ELISPOT assay shown in Fig. 4A theMHC class I-restricted cell population responsible for IFN-g se-cretion was mainly CD81 T cells.

Pattern of cytokine secretion in splenocytes after VV Agrestimulation

We next investigated the influence of IL-12 expressed from VV onthe pattern of cytokines expressed by T cells in vitro after Agrestimulation. Two groups of mice were immunized i.p. with 53107 PFU of rVVlucIL-12 or control rVVlucHA2, and 14 days latersplenocytes from both groups of mice were restimulated in vitrowith UV-inactivated VV. As shown in Fig. 4C, high levels ofIFN-g (Th1-type cytokine) were found in splenocyte cultures re-stimulated with VV in both groups of animals, with a slight in-crease in rVVlucIL-12 with respect to control virus. However, lev-els of IL-10 (Th2-type cytokine) were significantly decreased insupernatants of splenocytes from rVVlucIL-12 compared withcontrols, suggesting that IL-12 was suppressing an antiviral Th2type of response, rather than enhancing the CD41 Th1 response.

Effect of IL-12 delivered by rVV on systemic Ab responseto VV Ags

In murine systems it has been shown that Th2 cytokines favor theinduction of IgG1 subclass Abs, whereas IgG2a subclass Abs areinduced in a context of Th1 cytokines (35). Thus, we next evalu-ated anti-VV IgG subclasses in sera, 2 wk after inoculation of micewith rVVlucIL-12 or rVVlucHA2. As shown in Fig. 5, there wereno major differences in levels of specific IgG or IgG2a Abs be-tween the groups, while in rVVlucIL-12-inoculated mice, anti-VVIgG1 subclass Abs were greatly reduced compared with the levelsfound in control mice. Thus, the ratio of IgG2a/IgG1 Abs (Fig. 5,

FIGURE 3. Kinetics of IL-12 and IFN-g expression following inocula-tion of mice with rVVIL-12. Groups of mice were inoculated i.p. with 53107 PFU/animal of rVVlucHA2 (open symbols) or rVVlucIL-12 (filledsymbols), and at the indicated time points sera and spleens were obtainedfrom three mice per group.A, Evaluation of serum levels of IFN-g (byELISA) and bioactive IL-12 (by a bioassay).B, Levels of IL-12 and IFN-gmRNAs in spleens as determined by RT-PCR. The expected sizes of thePCR products are indicated. Levels of HPRT mRNA were used as aninternal control. The data shown are representative of two independentexperiments. n, naive mice.

6728 IL-12 AND IMMUNE RESPONSES TO VV AND HIV-1 Env

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

right panel) in mice inoculated with rVVlucHA- was 1.7, and thisratio was increased 13.5 times in sera from mice inoculated withrVVlucIL-12. These findings indicate that IL-12 expressed fromrVV modulates the humoral immune response by down-regulatingspecific IgG1 subclass Ab (Th2 cytokine) production, rather thanby promoting up-regulation of the IgG2a subclass (Th1 cytokine),a process resembling in vitro restimulation assays (see Fig. 4C).

Coexpression of IL-12 and HIV-1 Env by rVV increases thecellular immune response against gp160 in a dose-dependentmanner

To assess whether IL-12 delivered by VV might be effective inpotentiating the cellular immune response against the Env proteinof HIV-1, we first evaluated the IL-12 action when the Ag and thecytokine were coexpressed from the same virus vector. To thisaim, groups of mice were immunized i.p. with either 53 107 or1 3 107 PFU/mouse of rVVenvHA- (that expresses the complete

envof HIV-1 strain IIIB) or rVVenvIL-12 (coexpressingenvandIL-12 genes). Fourteen days after immunization, we evaluated thenumbers of specific splenic IFN-g-secreting CD81 T cells, usingP815 cells pulsed with a CD81 T cell peptide specific for the V3loop of env. Immunization with 13 107 PFU of rVVenvIL-12induced about 2-fold higher number of splenic Env-specific IFN-g-secreting CD81 T cells with respect to spleen cells from miceinoculated with control virus (p , 0.01; Fig. 6A). However, thisimmune response was 3-fold lower than that in control when micewere inoculated with the higher dose (53 107 PFU) of rVVs (Fig.5A). Indeed, levels of IFN-g and IL-10 measured 14 dpi after Agin vitro restimulation of splenocytes from mice inoculated with1 3 107 PFU of rVVenvIL-12 were 3-fold higher and 10-foldlower, respectively, than those found in control samples, indicatingthat an increase in the Th1-type immune response was occurring atthe low dose (13 107 PFU) of infection with the IL-12-expressingrVV (data not shown).

To understand the dose-dependent action of IL-12 deliveredfrom rVV we attempted to establish a relationship among the virusdose, serum levels of IL-12 and IFN-g, and the extent of viralclearance during short times after rVV inoculation. To this aim,groups of four mice were inoculated i.p. with either 53 107 or 13107 PFU/mouse of rVVlucHA- or rVVlucIL-12, which as live vec-tors are phenotypically undistinguishable from rVV expressingEnv, but allowed measurement of VV replication following lucif-erase activity in tissues. Levels of IL-12 in serum at 12 hpi were10- or 50-fold higher in mice inoculated with 13 107 or 5 3 107

PFU of rVVlucIL-12, respectively, than in the corresponding con-trol groups (Fig. 6B). However, levels of IFN-g induced were 10-fold higher in mice inoculated with 53 107 PFU of rVVlucIL-12than in control immunized mice, but lower levels of IFN-g (20–120 pg/ml) were present in mice inoculated with 13 107 PFU ofeither control or IL-12-expressing rVV viruses. Interestingly, bythe second dpi 50-fold lower luciferase activity was found in miceinoculated with 53 107 PFU of rVVlucIL-12 compared with thatin control infected mice (Fig. 2), while at this time luciferase levelswere identical in mice given 13 107 PFU of either rVVlucHA- orrVVlucIL-12 (data not shown).

FIGURE 4. Cellular immune response elicited against VV upon expression of IL-12.A, Determination by the ELISPOT assay of the number ofIFN-g-secreting CD81 T cells specific for VV Ags. Groups of four mice were immunized i.p. with 53 107 PFU/animal of rVVlucHA- (M) or rVVlucIL-12(f), and 7 or 14 days later spleen cells were employed as responder cells in the ELISPOT assay using P815 cells infected with VV as targets. Bars representthe mean number6 SD for individual samples (at 7 days) or the average of three pooled samples6 SD (at 14 days) from triplicate cultures. Data arerepresentative of at least two independent experiments performed at 7 and 14 days after immunization. Data from mice inoculated with the different rVVwere significant (p,p , 0.01). B, Number of IFN-g-secreting CD81 T cells specific for VV Ags in total splenocytes, in CD81 selected T cells, or inCD81-depleted T cells 14 days after immunization with rVVlucHA-. The mean6 SD were calculated from data from triplicate cultures of pooled samples.C, Pattern of cytokine secretion in supernatants of spleen cells after Ag restimulation. Mice were inoculated as inA with rVVlucHA- (M) or rVVlucIL-12(f). Fourteen days later splenocytes were cultured in vitro with VV Ag (UV inactivated) at 1mg/ml, and supernatants were collected and cytokinesdetermined at 48 h (IL-10) or 72 h (IFN-g) by ELISA. Three different experiments were conducted, and the results of one representative experiment areshown.

FIGURE 5. Humoral immune response against VV Ags following im-munization with rVVlucHA- or rVVlucIL-12. Fourteen days after micewere inoculated i.p. with 53 107 PFU/mouse of the different rVV asdescribed in Fig. 3, blood was obtained, and sera were tested for specificAbs. Anti-vaccinia IgG, IgG1, and IgG2a Ab titers are referred to as theinverse log2 dilution of sera that gave an absorbance at 492 nm of.0.1.Values represent the mean6 SD for individual samples corresponding tofour or five mice per group.

6729The Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

The results presented in Fig. 6 showed that IL-12 levels can becontrolled by the dose of rVV expressing IL-12 inoculated, whichseems to be critical for the extent of cellular immune responsesto Env.

Enhancement of the immune response to HIV-1 Env bydelivering Env and IL-12 from two different rVV vectors

To more accurately explore the dose-dependent action of IL-12 onHIV-1 Env, we tried to modulate the immune response by deliv-ering Env and IL-12 from two different vectors. Groups of fourmice were inoculated i.p. with 13 107 PFU/mouse of rVVenvHA-alone or in combination with different doses (23 104, 2 3 105,and 23 106 PFU) of rVVlucIL-12 (as a source of IL-12), and 14days later the number of specific splenic CTLs was determined byELISPOT (Fig. 7A). The results obtained revealed that coadmin-istration of 23 104 PFU of the rVV expressing IL-12 with 13 107

PFU of rVVenvHA- increased by 3 times (p , 0.001) the numberof specific IFN-g-secreting CD81 T cells. In contrast, groups of

mice inoculated with higher doses of the rVV expressing IL-12(2 3 105 and 2 3 106 PFU) showed no significant differences( p . 0.2) with respect to the control group inoculated only withrVVenvHA-. To further investigate the IL-12 enhancement of cel-lular activity we performed Th cell proliferation assays 2 wk afterimmunization with splenocytes from immunized mice. Fig. 7B(left panel) shows similar stimulation indexes in spleen cells frommice immunized with rVVenv alone (13 107 PFU) or with thesame dose of rVV env and 23 106 PFU of rVVlucIL-12. How-ever, a nearly 3-fold increase in specific T cell proliferation activ-ity appeared in the mice receiving the lower dose of rVVlucIL-12virus (23 104 PFU). Since IFN-g production by CD41 T cells isthe most reliable indicator of a Th1 phenotype, we also measuredthe levels of IFN-g secreted in stimulated spleen cells. As shownin Fig. 7B (right panel) higher levels of IFN-g expression corre-lated with the higher T cell proliferation. The findings shown in

FIGURE 6. Enhancement of specific cellular immune response againstthe HIV-1 Env protein is dependent of the viral dose of rVVenvIL-12.A,Data represent the numbers of IFN-g-secreting CD81 T cells specific forthe V3 loop epitope of the HIV-1 Env protein. Splenocytes from four miceimmunized i.p. with the indicated dose of the rVVenvHA- or rVVenvIL-12were pooled 14 days after immunization, and the number of specific IFN-g-secreting CD81 T cells was determined after coculture with P815 cellscoated with the specific peptide (9 Env) by ELISPOT assay. Bars representthe mean6 SD for triplicate cultures.B, Data represent IL-12 and IFN-gserum levels at 12 h postinoculation at the indicated doses. Pooled serafrom four mice inoculated in each case were used to measure IL-12 orIFN-g by bioassay or ELISA, respectively. Results are the mean6 SD oftriplicate measurements.

FIGURE 7. A combined immunization scheme improved the IL-12-me-diated enhancement of cellular immune response against the HIV-1 EnvAg. A, Numbers of IFN-g-secreting CD81 T cells specific for the V3 loopepitope of the HIV-1 Env protein. Splenocytes from four mice immunizedwith the indicated doses of the rVVenv or rVVIL-12 viruses were pooled14 days after immunization, and the number of specific IFN-g-secretingCD81 T cells was determined after coculture with P815 cells coated withthe specific peptide (10 Env) by ELISPOT assay. Bars represent themean 6 SD for triplicate cultures.B, Two weeks after immunization,spleens from mice of the different groups were collected, and their lym-phocytes were isolated. These cells were then tested for T cell proliferationby stimulation with either gp160 protein (1mg/ml) or medium, whichserved as a negative control. After 72 h of culture [3H]thymidine (1mCi/well) was added, and 18 h later cells were harvested. To theleft, the graphrepresents the relative specific T cell proliferative response against gp160,calculated as relative stimulation index (counts per minute in the combinedimmunization group/counts per minute in the control group (rVVenvHA2).On theright, are the levels of IFN-g in supernatants from the same lym-phocyte cultures, as described above, after 72 h of incubation in the pres-ence of gp160.

6730 IL-12 AND IMMUNE RESPONSES TO VV AND HIV-1 Env

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

Fig. 7 established that the dose of 23 104 PFU of rVVlucIL-12significantly increased the cellular immune response against thegp160 Ag delivered by 13 107 PFU of rVVenv HA2, augmentingthe numbers of specific CD81 T cells and specific Th1 cellresponse.

DiscussionThe potential of VV as a vector for vaccine purposes and the prom-ising previous observations in different vaccination schedules withrVV expressing HIV-1 and SIV genes (4, 13, 36, 37), encouragefurther studies to improve the vaccination strategies against HIV-1based on VV vectors. In this regard, a critical point related to theuse of live viral vaccines is to develop approaches to attenuate thevector without losing the immunogenicity of the expressed Ag.One way to circumvent this problem and increase the imunoge-nicity of the recombinant product is the local coexpression of theAg and immunostimulating cytokines. In this study we show in themouse model that delivery of IL-12 from rVV has a marked effecton clearance of the virus from tissues, as previously described forother cytokines expressed from VV (19, 38). However, by modu-lating the levels of expression of IL-12 and Ags it is possible toachieve enhanced cellular immune responses against VV and aforeign Ag such as HIV-1 Env. In addition, the immune responsecan be modulated to a Th1 type, as observed by the ratio of IgG2a/IgG1 subtype and elevated levels of IFN-g in serum.

To our knowledge this is the first study of IL-12 induction dur-ing VV infection in mice. We have shown that IL-12 is producedafter VV infection as early as 6 hpi, and this is followed by anincrease in IFN-g at 12 hpi (Fig. 3). This finding, observed at theprotein level in serum samples and at the mRNA accumulation inspleen cells, clearly shows that VV per se induces a Th1-typeimmune response. Furthermore, when IL-12 is expressed from therVV, both the levels and time course of IL-12 and IFN-g accu-mulation are greatly enhanced following viral infection.

The rVV expressing the IL-12 was attenuated, and it was elim-inated from target tissues at earlier times than control virus. By thesecond and third dpi the levels of replication of rVV-IL-12 inspleen and ovaries were 100-fold lower than those in mice inoc-ulated with the corresponding control virus. However, minor dif-ferences at the anatomical level were observed in mice inoculatedwith the rVV expressing IL-12. Splenomegaly was apparent in70–80% of the mice inoculated with 53 107 PFU of rVVlucIL-12, and this was associated with a marked increase in levels ofIFN-g in serum (data not shown), with no consequences on mousesurvival during the course of the experiment. These results agreewith previous findings after in vivo administration of rIL-12 (39).

Data obtained with in vitro restimulation of splenocytes fromimmunized mice revealed that a Th1-type immune response is elic-ited upon infection with VV. Moreover, after delivery of IL-12from rVV, a down-regulation of a Th2-type response is triggered(Fig. 4C). In this regard, we have found that VV infection elicitsan Ab response biased toward the isotype IgG2a, and the IgG2a/IgG1 ratio is increased 13.5 times relative to that for control VVwhen the IL-12 gene is expressed from rVV. Hence, our findingsclearly demonstrate that IL-12 expression from rVV steers a potentTh1 response following inoculation in mice.

Th1-type immune responses characterized by production ofIFN-g are documented to occur during viral infections, pointingout the important antiviral role played by this cytokine as a first-line defense mechanism of the organism. A number of studies haveobserved IL-12 induction in mice upon viral infection. Expressionof IL-12 has been demonstrated in mice early after infection(12–24 hpi) with RNA (murine hepatitis virus, lactate dehydroge-

nase-elevating virus, influenza virus) and DNA (adenovirus,HSV-1, and murine CMV) viruses, at both mRNA and proteinlevels (40–42). Studies on experimental infection in mice withdifferent viruses correlated the IL-12 effect on antiviral immunitywith the IFN-g induction (43, 44) and activation of the CTL re-sponse (45). However, alternative pathways of IFN-g inductionduring viral infection independent of IL-12 have been reported(46). In mice inoculated with VV, it has been shown that the anti-VV activity of IL-12 is abolished in IFN-gR2/2 mice (19). More-over, a drastic impairment of VV control results from neutraliza-tion of IFN-g or from the functional deficiency of the IFN-g genein knockout mice (47). Thus, it is well established that IFN-g isinvolved in the clearance of VV infection. In view of our findings,we propose that the rapid elimination of rVV expressing IL-12 inmice, compared with control virus infection, is mediated by theinduction of IFN-g, acting as a first antiviral nonspecific immuneresponse of the host. Inhibition of VV replication by IFN-g isprobably mediated by the production of nitric oxide, as we havepreviously reported that treatment of macrophages with this cyto-kine potently induce nitric oxide, and this correlates with inhibitionof VV replication (48). Furthermore, we have shown that inducibleexpression of nitric oxide synthase by rVV leads to inhibition ofVV DNA replication and induction of apoptosis (49, 50).

As documented here, IL-12 expression by VV has the capacityto modulate the antiviral immune response of the host, which wasrevealed by the nearly 2-fold increase in the number of anti-VVCD81 IFN-g-secreting cells at 7 and 14 days after inoculationcompared with that in controls. Although studies in other viralmodel systems showed that IL-12 can promote unspecific expan-sion of CD81 lymphocytes that can control viral infection (51),here we demonstrate that a reduction in VV titers after expressionof IL-12 correlated with an increased number of specific CD81 Tcells. In concordance with our results, expression from VV of IL-4,a cytokine that down-regulates Th1 responses, inhibits the devel-opment of mature antiviral CTL (52). Furthermore, mucosal de-livery of IL-12 from rVV was effective in restoring the antiviralCTL activity in a murine model of allergic airway disease (28).

Different lines of evidence support the current opinion that in-duction of CMI may be an important requirement for any candi-date vaccine for HIV-1. In this investigation we have examined thepotential enhancement of cellular immunity against HIV-1 by im-munomodulating the specific immune response to an HIV-1 Agthrough the codelivery of IL-12 and the HIV-1 Env protein. Wefound that immunization of mice with rVV coexpressing bothgenes enhanced the cellular immune response against Env whenthe rVV was administered at a dose of 13 107 PFU/animal, whilea higher dose reversed the effect. We found that the level of IFN-gproduced compromises the effectiveness of the CMI elicited, asthat parameter is critical in the velocity of the resolution of the VVinfection. However, expression of IL-12 to levels above those in-duced as a consequence of the VV infection is required to triggera strong cellular anti-Env immune response. These findings sug-gested that to achieve an enhancement of the cellular immune re-sponse to HIV-1 Env, critical Ag and IL-12 expression levelsshould exist. Indeed, in the SIV macaque model a direct correla-tion has been shown between the ability of attenuated SIV to rep-licate in the host and the degree of protection that was conferred(53). In this regard, Orange et al. (51) showed that IL-12 dosesrequired to promote protective, but not detrimental, responsesmight vary extremely in the context of different infections or im-mune responses.

We have also optimized the immunization procedure using therVV expressing IL-12. Thus, we found that delivering Env Ag and

6731The Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

IL-12 genes from individual VV vectors can lead to an enhance-ment of the specific CMI to Env by using different doses of the tworVV. For this concern it is noticeable that the dose-response effectof the IL-12 on the CMI elicited is observed either using doublerVV (Fig. 6) or delivering the Ag and the IL-12 from separatevectors (see Fig. 7). However the optimal dosages required aredifferent in each case; thus, an IL-12-mediated enhancement of theCMI anti-gp120 is observed at 107 PFU of the double rVV (ex-pressing IL-12 andenv genes; Fig. 6,left panel), while in themixing experiment this effect was observed with 23 104 PFU ofthe IL-12-expressing rVV but not when 23 106 PFU was used(Fig. 7A). These empirical data cannot be attributable to differ-ences in IL-12 expression from the different rVV (see Fig. 1B), butprobably reflect the fact that nonidentical infectious process aretaking place in each case, involving mechanisms that have a crit-ical role in the balance between the amount of IL-12 and the Ag.The finding that the dosage of rVVIL-12 plays a critical role in thegeneration of an effective Th1-type immune response against therecombinant Ag is important, especially when considering the im-plementation of vaccination strategies based on rVV. In a recentvaccine trial with NYVAC-SIV recombinants in macaques, it wasfound that although the addition of rNYVAC-IL-12 enhanced theCMI, it did not appear to influence the outcome of SIV challenge(54). It is possible that in this particular experimental animalmodel, the levels of IL-12 expressed by rVV were not optimal forproviding the qualitative and/or quantitative modulation of the im-mune response to influence the vaccine efficacy, as only one doseof the cytokine delivering recombinant virus was assayed. At-tempts to enhance the CMI response against HIV-1 Ags by coad-ministration of IL-12 and plasmids encoding HIV-1 genes havebeen described in various immunization strategies based on DNAvaccines (55–57), and increases in both specific CTL and Th1proliferative activities were obtained (58). In our study we dem-onstrate similar IL-12 effects when the cytokine is delivered fromVV vector, with the advantage that IL-12 is produced only tran-siently, thus diminishing the undesirable side effects derived fromthe expression of the cytokine for long periods of time, as expectedafter DNA immunization.

In conclusion, in this investigation we have designed protocolsof immunization based on rVV that increased significantly the cel-lular immune response to HIV-1 Env. We have also characterizedthe replication of rVV in tissues and how the IL-12 cytokine mod-ulates the immune response to VV Ags. The feasibility of practicalstrategies able to enhance the immune response to an Ag deliveredby rVV together with IL-12 and to steer the response toward adesired arm of the immune system, cellular or humoral, shouldprovide a practical means to improve vaccination against patho-gens, such as HIV-1, in which some immune responses may beprotective and others detrimental.

AcknowledgmentsWe thank Victoria Jimenez for excellent technical assistance.

References1. World Health Organization. 1980.The Global Eradication of Smallpox: Final

Report of the Global Commission for the Certification of Smallpox Eradication.History of International Public Health, No. 4. World Health Organization,Geneva.

2. Moss, B. 1996. Genetically engineered poxviruses for recombinant gene expres-sion, vaccination, and safety.Proc. Natl. Acad. Sci. USA 93:11341.

3. Brochier, B., F. Costy, and P. P. Pastoret. 1995. Elimination of fox rabies fromBelgium using recombinant vaccinia-rabies vaccine: an update.Vet. Microbiol.46:269.

4. Hirsch, V. M., T. R. Fuerst, G. Sutter, M. W. Carroll, L. C. Yang, S. Goldstein,M. Piatak, W. R. Elkins, W. G. Alvord, D. Montefiori, et al. 1996. Patterns ofviral replication correlate with outcome in simian immunodeficiency virus-in-

fected macaques: effect of prior immunization with a trivalent SIV vaccine inmodified vaccinia virus Ankara.J. Virol. 70:3741.

5. Belyakov, I. M., L. S. Wyatt, J. D. Adlers, P. Earl, C. D. Pendelton, B. L. Kelsall,W. Strober, B. Moss, and J. A. Berzofsky. 1998. Induction of mucosal cytotoxicT-lymphocyte response by intrarectal immunization with a replication-deficientrecombinant vaccinia virus expressing human immunodeficiency virus 89.6 en-velope protein.J. Virol. 72:8264.

6. Wyatt, R., and J. Sodrosky. 1998. The HIV-1 envelope glycoproteins: fusogens,antigens, and immunogens.Science 280:1884.

7. Connor, R. I., B. T. Korber, B. S. Graham, B. H. Hahn, D. D. Ho, B. D. Walker,A. U. Neuman, S. H. Vermund, J. Mestecky, S. Jackson, et al. 1998. Immuno-logical and virological analyses of persons infected by human immunodeficiencyvirus type 1 while participating in trials of recombinant gp120 subunits vaccines.J. Virol. 72:1552.

8. Koup, R. A., J. T. Safrit, Y. Cao, C. A. Andrews, G. Mcleod, W. Borkowsky,C. Fathing, and D. D. Ho. 1994. Temporal association of cellular immune re-sponses with the initial control of viremia in primary human immunodeficiencyvirus type 1 syndrome.J. Virol. 68:4650.

9. Musey, L., J. Hughes, T. Schacker, T. Shea, L. Corey, and M. J. McElrath. 1997.Cytotoxic-T-cell responses, viral load, and disease progression in early humanimmunodeficiency virus type 1 infection.N. Engl. J. Med. 337:1267.

10. Rowland-Jones, S., R. Tan, and A. J. McMichael. 1997. The role of cellularimmunity in protection against HIV infection.Adv. Immunol. 65:277.

11. Rowland-Jones, S., J. Sutton, K. Ariyoshi, T. Dong, F. Gotch, S. McAdam,D. Whitby, S. Sabally, A. Gallimore, T. Corrah, et al. 1995. HIV-1 specificcytotoxic T-cells in HIV-exposed but uninfected Gambian women.Nat. Med.1:59.

12. Buseyne, F., M. Bugard, J. P. Teglas, E. Bui, C. Rouzioux, M. J. Mayaux,S. Blanche, and Y. Riviere. 1998. Early HIV-specific cytotoxic T lymphocytesand disease progression in children born to HIV-infected mothers.AIDS Res.Hum. Retroviruses 16:1435.

13. Gallimore, A., M. Cranage, N. Cook, N. Almond, J. Bootman, E. Rud, P. Silvera,M. Dennis, T. Corcoran, J. Stott, et al. 1995. Early suppression of SIV replicationby CD81 nef-specific cytotoxic T cells in vaccinated macaques.Nat. Med.1:1167.

14. Trinchieri, G. 1994. Interleukin 12: a cytokine produced by antigen-presentingcells with immunoregulatory functions in the generation of T-helper cells type 1and cytotoxic lymphocytes.Blood 84:4008.

15. Tsung, K., J. B. Meko, G. R. Peplinski, Y. L. Tsung, and J. A. Norton. 1997.IL-12 induces T-helper 1-directed antitumor response.J. Immunol. 153:1997.

16. Liblau, R. S., S. M. Singer, and H. O. McDevitt. 1995. Th1 and Th2 CD41 T cellsin the pathogenesis of organ-specific autoimmune diseases.Immunol. Today 16:34.

17. Trinchieri, G. 1998. Immunobiology of interleukin-12.Immunol. Res. 17:269.18. Gazzinelli, R. T., N. A. Giese, and H. C. Morse III. 1994. In vivo treatment with

interleukin 12 protects mice from immune abnormalities observed during murineacquired immunodeficiency syndrome (MAIDS).J. Exp. Med. 180:2199.

19. Ramshaw, I. A., A. J. Ramsay, G. Karupiah, M. S. Rolph, S. Mahalingam, andJ. C. Ruby. 1997. Cytokines and immunity to viral infections.Immunol. Rev.159:119.

20. Rodrıguez, J. F., D. Rodrıguez, J. R. Rodrıguez, E. B. McGowan, andM. Esteban. 1988. Expression of the firefly luciferase gene in vaccinia virus: ahighly sensitive gene marker to follow virus dissemination in tissues of infectedanimals.Proc. Natl. Acad. Sci. USA 85:1667.

21. Rodrıguez, D., J. R. Rodrıguez, J. F. Rodrıguez, D. Trauber, and M. Esteban.1989. Highly attenuated vaccinia virus mutants for the generation of safe recom-binants viruses.Proc. Natl. Acad. Sci. USA 86:1287.

22. Dallo, S., and M. Esteban. 1987. Isolation and characterization of attenuatedmutants of vaccinia virus.Virology 159:408.

23. Chakrabarti, S., J. R. Sisler, and B. Moss. 1997. Compact, synthetic, vacciniavirus early/late promoter for protein expression.BioTechniques 23:1094.

24. Carroll, M. W., and B. Moss. 1995.E. coli b-glucoronidase (GUS) as a markerfor recombinant vaccinia viruses.BioTechniques 19:352.

25. Rodrıguez, J. R., C. Risco, J. L. Carrascosa, M. Esteban, and D. Rodrıguez. 1997.Characterization of early stages in vaccinia virus membrane biogenesis: impli-cations of the 21-kilodalton protein and a newly identified 15-kilodalton envelopeprotein.J. Virol. 71:1821.

26. Miyahira, Y., K. Murata, D. Rodrıguez, J. R. Rodrıguez, M. Esteban,M. M. Rodrigues, and F. Zavala. 1995. Quantification of antigen specific CD81

T cells using an ELISPOT assay.J. Immunol. Methods 181:45.27. Hogan, S. P., P. S. Foster, X. Tan, and A. J. Ramsay. 1998. Mucosal IL-12 gene

delivery inhibits allergic airways disease and restores local antiviral immunity.Eur. J. Immunol. 28:413.

28. Shida, H., Y. Hinuma, M. Hatanaka, M. Morita, M. Kidokoro, K. Suzuky, T.,Maruyama, F. Takahashi, M. Sujimoto, R. Kitamura, et al. 1988. Effects andvirulences of recombinant vaccinia viruses derived from attenuated strains thatexpress the human T-cell leukemia virus type I envelope gene.J. Virol. 62:4474.

29. Karupiah, G., B. Coupar, I. Ramshaw, D. Boyle, R. Blanden, and M. Andrew.1990. Vaccinia virus-mediated damage of murine ovaries and protection by virus-expressed interleukin-2.Immunol. Cell Biol. 68:325.

30. Gately, M. K., A. G. Wolitzky, and P. M. Quinn. 1992. Regulation of humancytolytic lymphocyte responses by interleukin-12.Cell. Immunol. 143:127.

31. Bloom, E. T., and J. A. Horvath.1994. Cellular and molecular mechanisms of theIL-12 induced increased in allospecific murine cytolytic T-cell related decline inCTL. J. Immunol. 152:4242.

32. Huang, S., W. Hendriks, A. Althage, S. Hemmi, H. Bluethmann, R. Kamijo,J. Vilcek, R. M. Zinkernagel, and M. Auget. 1993. Immune response in mice thatlack the interferon-g receptor.Science 259:1742.

6732 IL-12 AND IMMUNE RESPONSES TO VV AND HIV-1 Env

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

33. Ruby, J., and I. Ramshaw. 1991. The antiviral activity of immune CD81 T cellsis dependent on interferon-g. Lymphokine Cytokine Res. 10:353.

34. Gherardi, M. M., and M. Esteban. 1999. Mucosal and systemic immune re-sponses induced after oral delivery of vaccinia virus recombinants.Vaccine 17:1074.

35. Snaper, C. M., and W. E. Paul. 1987. Interferon-g and B cell stimulatory factor-1reciprocally regulate Ig isotype production.Science 236:944.

36. Perkus, M. E., J. Tartaglia, and E. Paoletti. 1995. Poxvirus-based vaccine can-didates for cancer, AIDS, and other infectious diseases.J. Leukocyte Biol. 58:1.

37. Myagkikh, M., S. Alipanah, P. D. Markham, J. T. Tartaglia, E. Paoletti,R. C. Gallo, G. Franchini, and M. Robert-Guroff. 1996. Multiple immunizationswith attenuated poxviruses HIV type 2 recombinants and subunit boosts requiredfor protection of rhesus macaques.AIDS Res. Hum. Retroviruses 12:985.

38. Giavedoni, L. D., L. Jones, M. B. Gardner, H. L. Gibson, C. T. Lee, P. J. Barr,and T. Yilma. 1992. Vaccinia virus recombinants expressing chimeric proteins ofhuman immunodeficiency virus andg interferon are attenuated for nude mice.Proc. Natl. Acad. Sci. USA 89:3409.

39. Car, B. D., V. M. Eng, B. Schnyeder, M. LeHir, A. N. Shakhov, G. Woerly,S. Huang, M. Aguet, T. D. Anderson, and B. Ryffel. 1995. Role of interferongin interleukin 12-induced pathology in mice.Am. J. Pathol. 147:1693.

40. Coutelier, J. P., J. Van Broeck, and S. F. Wolf. 1995. Interleukin-12 gene ex-pression after viral infection in the mouse.J. Virol. 69:1955.

41. Kanangat, S., J. Thomas, S. Gangappa, J. Sam Babu, and B. T. Rouse. 1996.Herpes simplex virus type 1-mediated up-regulation of IL-12 (p40) mRNA ex-pression.J. Immunol. 156:1110.

42. Orange, J. S., and C. A. Biron. 1996. An absolute and restricted requirement forIL-12 in natural killer cell IFN-g production and antiviral defense.J. Immunol.156:1138.

43. Ozmen, L., M. Aguet, G. Trinchieri and G. Garotta. 1995. The in vivo antiviralactivity of interleukin-12 is mediated byg interferon.J. Virol. 69:8147.

44. Carr, J. A., J. Rogerson, M. J. Mulqueen, N. A. Roberts, and R. F. G. Booth.1997. Interleukin-12 exhibits potent antiviral activity in experimental herpesvirusinfections.J. Virol. 71:7799.

45. Monteiro, J., C. Harvey, and G. Trinchieri. 1998. Role of interleukin-12 in pri-mary influenza virus infection.J. Virol. 72:4825.

46. Schijns, V. E. C., B. L. Haagmans, C. M. H. Wierda, B. Kruithof,I. A. F. M. Heijnen, G. Alber, and M. C. Horzinek. 1998. Mice lacking IL-12develop polarized Th1 cells during viral infection.J. Immunol. 160:3958.

47. Ramsay, A. J., J. Ruby, and I. A. Ramshaw. 1993. The case of cytokines aseffector molecules in the resolution of virus infection.Immunol. Today 14:155.

48. Melkova, Z., and M. Esteban. 1994. Interferon-g severely inhibits DNA synthesisof vaccinia virus in a macrophage cell line.Virology 198:731.

49. Melkova, Z., and M. Esteban. 1995. Inhibition of vaccinia virus DNA replicationby inducible expression of nitric oxide synthase.J. Immunol. 155:5711.

50. Melkova, Z., S. B. Lee, D. Rodrıguez, and M. Esteban. 1997. Bc1–2 preventsnitric oxide-mediated apoptosis and poly (ADP-ribose) polymerase cleavage.FEBS Lett. 403:273.

51. Orange, J. S., S. F. Wolf, and C. A. Biron. 1994. Effects of L-12 on the responseand susceptibility to experimental viral infections.J. Immunol. 152:1253.

52. Sharma, D. P., A. J. Ramsay, D. J. Maguire, M. S. Rolph, and I. A. Ramshaw.1996. Interleukin-4 mediates down regulation of antiviral cytokine expressionand cytotoxic T-lymphocyte responses and exacerbates vaccinia infection in vivo.J. Virol. 70:7103.

53. Wyand, M. S., K. H. Manson, M. Garcia-Moll, D. Montefiori, and R. C. Dero-siers 1996. Vaccine protection by a triple deletion of simian immunodeficiencyvirus. J. Virol. 70:3724.

54. Benson, J., C. M. Chougnet, M. Robert-Guroff, D. Montefiori, P. Markham,G. Shearer, R. C. Gallo, M. Cranage, E. Paoletti, K. Limbach, et al. 1998. Re-combinant vaccine-induced protection against the highly pathogenic simian im-munodeficiency virus SIVmac251 dependence on route of challenge exposure.J. Virol. 72:4170.

55. Takashi, T., K. Hamajima, J. Fukushima, K. Q. Xin, N. Ishii, I. Aoki,Y. Ishigatsubo, K. Tani, S. Kawamoto, Y. Nitta, et al. 1997. Enhancement ofcell-mediated immunity against HIV-1 induced coinoculation of plasmid encodedHIV-1 antigen with plasmid expressing IL-12.J. Immunol. 158:4008.

56. Ahlers, J. D., Dunlop N., D. W. Alling, P. L. Nara, and J. A. Berzofsky. 1997.Cytokine in adjuvant steering of the immune response phenotype to HIV-1 vac-cine constructs.J. Immunol. 158:3947.

57. Kim, J. J., N. N. Triveldi, L. K. Nottingham, L. Morrison, A. Tsai, Y. Hu,S. Mahalingham, K. Dang, L. Ahn, N. K. Doyle, et al. 1998. Modulation ofamplitude and direction of in vivo immune responses by coadministration ofcytokine gene expression cassettes with DNA immunogens.Eur. J. Immunol.28:1089.

58. Kim, J. J., V. Ayyavoo, M. L. Bargarazzi, M. A. Chattergoon, K. Dang, B. Wang,J. D. Boyer, and D. B. Weiner. 1997. In vivo engineering of a cellular immuneresponse by coadministration of IL-12 expression vector with a DNA immuno-gen.J. Immunol. 158:816.

6733The Journal of Immunology

by guest on January 26, 2015http://w

ww

.jimm

unol.org/D

ownloaded from