Ty21a Typhoid Vaccine Serovar Typhi Strain Salmonella Enterica

Home , Ty21a, Typhoid vaccine

Characterization of CD8+ Effector T Cell Responses in Volunteers Immunized with Salmonella enterica Serovar Typhi Strain Ty21a Typhoid Vaccine This information is current as of September 27, 2021. Rosângela Salerno-Goncalves, Marcela F. Pasetti and Marcelo B. Sztein J Immunol 2002; 169:2196-2203; ; doi: 10.4049/jimmunol.169.4.2196

http://www.jimmunol.org/content/169/4/2196 Downloaded from

References This article cites 27 articles, 8 of which you can access for free at: http://www.jimmunol.org/content/169/4/2196.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 27, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Characterization of CD8؉ Effector T Cell Responses in Volunteers Immunized with Salmonella enterica Serovar Typhi Strain Ty21a Typhoid Vaccine1

Rosaˆngela Salerno-Goncalves, Marcela F. Pasetti, and Marcelo B. Sztein2

Salmonella enterica serovar Typhi (S. typhi) strain Ty21a remains the only licensed attenuated typhoid vaccine. Despite years of research, the identity of the protective immunological mechanisms elicited by immunization with the Ty21a typhoid vaccine remains elusive. The present study was designed to characterize effector T cell responses in volunteers immunized with S. typhi strain Ty21a typhoid vaccine. We determined whether immunization with Ty21a induced specific CTL able to lyse S. typhi-infected cells and secrete IFN-␥, a key effector molecule against intracellular pathogens. We measured the functional activity of these CTL by a 51Cr-release assay using 8-day restimulated PBMC from Ty21a vaccinees as effector cells and S. Typhi-infected autologous PHA-activated PBMC as target cells. Most vaccinees exhibited consistently increased CD8-mediated lysis of targets by postimmunization PBMC when compared with preimmunization levels. We also developed an IFN-␥ ELISPOT assay to quantify the fre- Downloaded from quency of IFN-␥ spot-forming cells (SFC) in PBMC from Ty21a vaccinees using an ex vivo system. Significant increases in the fre- as compared ,(0.010 ؍ quency of IFN-␥ SFC following immunization (mean ؎ SD, 393 ؎ 172; range 185–548 SFC/106 PBMC; p with preimmunization levels, were observed. IFN-␥ was secreted predominantly by CD8؉ T cells. A strong correlation was recorded -p < 0.001). In conclusion, this work consti ,0.910 ؍ between the cytolytic activity of CTL lines and the frequency of IFN-␥ SFC (r2 -tutes the first evidence that immunization of volunteers with Ty21a elicits specific CD8؉ CTL and provides an estimate of the fre /quency of CD8؉ IFN-␥-secreting cells induced by vaccination. The Journal of Immunology, 2002, 169: 2196–2203. http://www.jimmunol.org

almonella enterica serovar Typhi (S. typhi), the causative from S. typhi infection remains unclear. There is considerable ev- agent of typhoid fever, is a human-restricted intracellular idence that host resistance to many intracellular bacteria such as S Gram-negative bacterium that infects both phagocytic and Listeria monocytogenes (8, 9), and Mycobacteria (10, 11) is nonphagocytic cells (1, 2). Worldwide, typhoid fever affects ϳ16 strongly inﬂuenced by CMI responses. We have previously dem- million individuals annually with 600,000 deaths (3). In Asia and onstrated the presence of speciﬁc CTL and IFN-␥ production to S. northeast Africa, the appearance of S. typhi showing resistance to typhi Ags in volunteers immunized with attenuated S. typhi strain

many antibiotics has become an important public-health problem CVD 908 and suggested that these might be important effector by guest on September 27, 2021 (4). Therefore, an improved prophylactic vaccine to prevent ty- mechanisms in resistance to S. typhi infection (12, 13). phoid fever is urgently needed. Despite years of research, there is little information on the pro- Although much is known regarding the immune responses elic- tective immunological mechanisms elicited by oral immunization ited by Salmonella typhimurium in the murine model (1, 5), which with S. typhi strain Ty21a typhoid vaccine (the only licensed at- results in a typhoid-like disease, little is known about the protec- tenuated live vaccine) which has been shown to have a variable tive immune responses to S. typhi infection in humans. Because of rate of protection depending on the formulation used and the num- the narrow restriction of S. typhi for human hosts, deﬁnitive studies ber and spacing of the doses administered (2). The purpose of this in humans are desirable3 (6). Results from studies in typhoid pa- study was to determine whether immunization with Ty21a typhoid tients and vaccine trials with attenuated S. typhi indicate that Abs vaccine elicited two key T cell-mediated effector mechanisms, i.e., appear to be involved in protection against S. typhi (2, 6, 7). How- speciﬁc CTL able to lyse S. typhi-infected targets and production ever, the role of cell-mediated immunity (CMI)4 in protection of IFN-␥ in response to stimulation with S. typhi-infected cells. Materials and Methods Center for Vaccine Development, Department of Pediatrics, University of Maryland Subjects School of Medicine, Baltimore, MD 21201 Five healthy volunteers, between 24 and 41 years of age, recruited from the Received for publication February 7, 2002. Accepted for publication June 12, 2002. Baltimore-Washington area and University of Maryland at Baltimore cam- The costs of publication of this article were defrayed in part by the payment of page pus (Baltimore, MD), participated in this study. They were immunized with charges. This article must therefore be hereby marked advertisement in accordance four spaced doses of 2Ð6 ϫ 109 CFU of Ty21a at an interval of 48 h with 18 U.S.C. Section 1734 solely to indicate this fact. between doses (14). Healthy volunteers underwent leukapheresis following 1 This work was supported by Grant R01 AI36525 from the National Institute of standard procedures before and between 4 and 37 mo (mean ϭ 16 mo) after Allergy and Infectious Diseases, National Institutes of Health (to M.B.S.).. ingestion of the vaccine. PBMC were then isolated by density gradient 2 Address correspondence and reprint requests to Dr. Marcelo B. Sztein, Center for centrifugation and cryopreserved in liquid N2. Demographic characteristics Vaccine Development, University of Maryland School of Medicine, 685 West Bal- of volunteers, as well as their HLA haplotypes, are shown in Table I. timore Street, Health Sciences Facility 480, Baltimore, MD 21201, E-mail address: Volunteers were without any antibiotic treatment and had normal blood [email protected] counts at the times of leukapheresis. Before the leukapheresis procedures 3 M. F. Pasetti, M. M. Levine, and M. B. Sztein. Animal models paving the way for were performed, the purpose of this study was explained to volunteers and clinical trials of attenuated Salmonella typhi live oral vaccines and live vectors. Sub- they signed informed consents. mitted for publication. Preparation of target/stimulator cells 4 Abbreviations used in this paper: CMI, cell-mediated immunity; rhIL-2, recombi- ϫ 6 ␮ nant human IL-2; B-LCL, EBV-transformed lymphoblastoid B cell lines; wt, wild Blasts were obtained by incubating 5Ð10 10 PBMC with 1 g/ml type; MOI, multiplicity of infection; CSA-1, Salmonella common structural Ags; PHA-L (Sigma-Aldrich, St. Louis, MO) for 24 h in RPMI 1640 (Life ECD, PE-Texas Red; ALPC, allophycocyanin; SFC, spot-forming cells. Technologies, Grand Island, NY) supplemented with 100 U/ml penicillin,

Table I. Summary data on the characteristics of Ty21a vaccinees participating in this study

Immunized Volunteers

TS RH MP SD MR

Age (years)a 34 34 29 24 41 Sex Male Female Female Female Female Date of leukapheresis Before immunization May-98 Jan-99 Aug-96 Aug-97 Jul-96 After immunization Sep-98 May-00 Dec-96 Sep-00 Aug-98 HLA class I Agsb A*0301, 2901 A*0201, 68012 A*0201, 2902 A*0101, 0302 A*0202, 6901 B*0705, 5501 B*4402, 44XX B*3901, 44031 B*0801, 3502 B*4402, 5001 HLA class II Agsb DRB1*0408, 1401 DRB1*11011, 0701 DRB1*16011, 0701 DRB1*03011, 11041 DRB1*0701

a Numbers represent the age of volunteers before immunization. b Performed by PCR-sequence-speciﬁc oligonucleotide probes.

100 ␮g/ml streptomycin, 50 ␮g/ml gentamicin, 2 mM L-glutamine, 2.5 mM in various combinations and analyzed using an Epics Elite ESP flow sodium pyruvate, 10 mM HEPES buffer, and 10% heat-inactivated FBS cytometer/cell sorter system. (complete RPMI). PHA-activated PBMC were then washed three times Downloaded from with RPMI 1640, and cultured in complete RPMI supplemented with 20 Preparation of effector cells IU/ml recombinant human IL-2 (rhIL-2) (Boehringer Mannheim, Mann- For cytotoxic assays, we used as effector cells both ex vivo- and in vitro- heim, Germany) for 5Ð6 days. expanded PBMC from immunized volunteers. In vitro-expanded effector EBV-transformed lymphoblastoid B cell lines (B-LCL) were established cells were obtained using a modification of a previously described tech- from PBMC isolated from Ty21a vaccinees following standard procedures niques (17). Briefly, PBMC were cocultured with stimulator cells at an (13, 15). B-LCL were maintained in culture in complete RPMI or cryo- effector to stimulator cell ratio of 7:1 in complete RPMI supplemented with preserved until used in the experiments. 60 IU/ml of rhIL-2 for 7Ð8 days. Stimulator cells consisted of autologous http://www.jimmunol.org/ blasts infected with S. typhi and were gamma-irradiated (4000 rad) as de- Infection of target/stimulator cells by S. typhi scribed above. Aliquots of effector cells were used for phenotypic analysis PHA-stimulated PBMC (henceforth called “blasts”), were incubated in by flow cytometry. Effector cells were stained with mAbs to CD3, CD4, ␣␤ RPMI (without antibiotics) for3hat37¡C, 5% CO , in the absence or CD8, CD56, TCR -FITC (clone WT31; BD Immunocytometry Sys- 2 ␥␦ presence of wild-type S. typhi strain ISP1820 (wt S. Typhi) (obtained from tems), and TCR -ALPC (clone B1.1; BD Immunocytometry Systems) Dr. J. Nataro, Center for Vaccine Development, University of Maryland conjugated to the appropriate fluorochromes and analyzed by five-color School of Medicine, Baltimore, MD) at a different multiplicity of infection flow cytometry. ␥ (MOI). In control experiments, inactivated S. Typhi (licensed heat-inacti- For IFN- ELISPOT assays, PBMC from immunized volunteers were vated phenol-preserved typhoid vaccine, typhoid vaccine USP; Wyeth- used ex vivo as effector cells. In some experiments, PBMC were fraction- Ayerst Pharmaceuticals, Marietta, PA) was added at a MOI of 10:1. After ated into CD4-, CD8-, or NK-depleted subsets using anti-CD4 and CD8 by guest on September 27, 2021 exposure to S. Typhi, cells were washed and incubated overnight at 37¡C, immunomagnetic beads (Dynal Biotech, Great Neck, NY) or anti-NK mi- crobeads (Miltenyi Biotec, Auburn, CA). Alternatively, purified cell pop- 5% CO2, in complete RPMI containing 20 IU/ml of rhIL-2. The following day, cells were gamma-irradiated (4000 rad) and used as stimulators to ulations were obtained by flow cytometric cell sorting following staining Ͼ expand effector cells for CTL assays or in ELISPOT assays. Alternatively, with the appropriate mAbs. Cell populations were 90% pure as deter- infected and uninfected blasts or B-LCL were labeled with 200 ␮Ci of mined by flow cytometric analysis. 51 sodium chromate ( Cr) (Amersham Pharmacia Biotech, Piscataway, NJ) Cytotoxic (chromium release test) and competitive inhibition for1hat37¡C, 5% CO2, washed three times, and used as targets in CTL assays. assays Cytotoxicity was determined by a 4-hr 51Cr-release assay as previously Analysis of intracellular and surface S. typhi Ags described (17). Spontaneous release was determined from wells containing Intracellular staining was performed following standard techniques (16). medium alone; maximum release was determined from wells to which 2% Briefly, blasts infected with S. typhi, or not infected, were removed from Triton-X (Sigma-Aldrich) was added. All cultures were set-up in quadru- Ϫ culture after3hofincubation, washed with PBS and fixed with PBS/4% plicate. Lysis (%) was calculated as follows: ((experimental release Ϫ ϫ formaldehyde (Polysciences, Warrington, PA). After 30 min at room tem- spontaneous release)/(maximum release spontaneous release) 100). perature, cells were washed with PBS and incubated for 10 min with 150 Specific cytotoxicity mediated by effector cells was calculated by subtract- ␮l PBS/1% BSA/0.5% saponin (permeabilization buffer; Sigma-Aldrich). ing the lysis of uninfected targets from the lysis of S. typhi-infected targets. Ͼ Optimal concentrations of a FITC-labeled polyclonal Ab recognizing Sal- The cut-off for positive responses in CTL assays was defined as 10% monella common structural Ags (CSA-1; Kirkegaard & Perry Laboratories, specific lysis above the mean specific lysis of effector PBMC when cul- Gaithersburg, MD) or a mouse IgG1-FITC isotype control (BD Immuno- tured with not-infected target cells as previously described (18). cytometry Systems, San Jose, CA) were then added and the mixture was Competitive inhibition studies were conducted by measuring the spe- incubated for 30 min at room temperature. Anti-CSA-1 is an Ab broadly cific lysis of labeled target cells by a fixed number of effector cells in reactive to Salmonella that was purified by affinity chromatography from a the presence of varying numbers of unlabeled target cells as previously pool of serum from goats immunized sequentially with different Salmo- described (17). nella strains, including S. typhi. Cells were then washed twice with per- Analysis of intracellular levels of IFN-␥ meabilization buffer, and once with PBS. Samples were analyzed with an Epics Elite ESP flow cytometer cell sorter system (Beckman Coulter, Identification of the effector cell populations secreting IFN-␥ following Miami, FL). exposure to S. typhi-infected targets was performed by multicolor flow S. typhi Ag expression was also determined on infected and not cytometry. To exclude from analysis target cells that might secrete IFN-␥, infected CD4ϩ or CD8ϩ T cells, NK, B cells, and macrophages within uninfected or S. typhi-infected target cells were labeled with a mAb to the blast populations by multicolor surface staining after 16 h of CD45-ALPC (a leukocyte common surface Ag marker; clone IM2473; incubation. To this end, blasts were stained with CSA-1-FITC (Kirkeg- Beckman Coulter). CD45-stained target cells were then cocultured with aard & Perry Laboratories) or mAbs to CD3-FITC (clone UCHT1; effector cells at an E:T cell ratio of 1:7. Effector cells cultured without Beckman Coulter), CD4-PE-Texas Red (ECD) (clone SFCl12T4011; target cells or with anti-CD3/CD28 beads (0.6 ␮l/ml; Dynal Biotech) were Beckman Coulter), CD8-PerCP (clone SK1; BD Immunocytometry used as negative and positive controls, respectively. After 16 h of incuba- Systems), CD14-allophycocyanin (ALPC) (clone MoP9; BD Immuno- tion at 37¡C, cytokine secretion was blocked by the addition of brefeldin A cytometry Systems), CD19-PerCP (clone SJ25-C1; BD Immunocytom- (Sigma-Aldrich) at a final concentration of 10 ␮g/ml for 5Ð6 h. Cells were etry Systems), CD56-PE (clone B159; BD Immunocytometry Systems) then harvested, washed with PBS, and surface-stained with CD56-FITC, 2198 CD8 RESPONSES IN Ty21a VACCINEES

CD4-ECD, and CD8-PerCP mAbs. After washing, cells were stained for vested, washed three times with RPMI containing gentamicin (100 intracellular IFN-␥ content using a protocol similar to the one described ␮g/ml) to kill extracellular bacteria, and the presence of intracel- above for the analysis of intracellular S. typhi Ags, in this case using an lular S. typhi Ags was examined immediately using the intracel- anti-IFN-␥-PE mAb (clone 4S.B3; BD Immunocytometry Systems). In these experiments, stained cells were analyzed by flow cytometry by first lular staining procedure described in Materials and Methods. Al- gating on the CD45Ϫ cell population to exclude target cells, followed by ternatively, cells were incubated for an additional 13 h in complete electronic gating on the CD3ϪCD56ϩ, CD3ϩCD4ϩ,orCD3ϩCD8ϩ pop- RPMI containing gentamicin (100 ␮g/ml) before being examined ϩ ulations for the determination of IFN-␥ cells in each of them. In other for surface expression of S. typhi Ags. We observed that after 3 h experiments, effectors were stained with a combination of CD3-FITC, ϳ CD56-PE, CD4-ECD, CD8-PerCP-Cy5.5 (clone RPA-T8; Becton Dickin- of incubation 90% of blasts exposed to S. typhi at various MOI son), and IFN-␥ ALPC mAbs. In these experiments, cells were analyzed by expressed S. typhi Ags intracellularly (Fig. 1, B–D). In contrast, no sequential gating on CD3ϩ, CD8ϩ, CD4Ϫ, and CD56Ϫ and the results were significant expression of S. typhi Ags was observed in the cyto- ϩ reported as % IFN-␥ cells in this population. plasm of uninfected cells (mean ϭ 1.9%, Fig. 1A). Similarly, Ͼ IFN-␥ ELISPOT assay 90% of infected cells were found to express S. typhi Ags on their cell membrane following 16 h of incubation in the presence of The frequency of IFN-␥-secreting cells was quantified by using a modified gentamicin (Fig. 1, F–H), whereas only a small proportion of un- IFN-␥ ELISPOT assay. Anti-human IFN-␥ mAb (5 ␮g/ml, clone 2G1; Endogen, Woburn, MA) was diluted in coating buffer (PBS/0.4 M NaOH, infected cells (Fig. 1E) or cells exposed to inactivated S. typhi (Fig. 2 mM EDTA), and 100 ␮l/well were added to dry MultiScreen-HA filtra- 1I) stained positively (range 1Ð6%) in several repeat experiments. tion plates (MAHA S4510; Millipore, Bedford, MA). After overnight in- These results suggest that expression of S. typhi Ags on the cell cubation at 4¡C, the wells were washed with wash buffer (PBS/ surface is dependent on bacterial cell invasion. The observed ki-

␮ Downloaded from 0.05%Tween 20) and unoccupied sites were blocked with 200 l/well netics of expression of bacterial Ags is similar to that reported by PBS/5% BSA for2hatroom temperature. After washing, 200 ␮l/well complete medium was added and incubated at room temperature. After 1 h, us (13) and others (19) using similar culture conditions. the complete RPMI was discarded and stimulator-effector cells at a 1:7 To determine whether a particular T cell population was pref- ratio were seeded in 200 ␮l/well and incubated in a humidiﬁed 37¡C, 5% erentially infected and expressed S. typhi Ags, infected blasts were CO2 incubator. Effector cells cultured without stimulator cells or with stained with mAbs to CD3, CD4, and CD8 in addition to the anti-S. CD3/CD28 beads (0.6 ␮l/ml; Dynal Biotech), were used as negative and typhi CSA-1 polyclonal Ab. As can be observed in Fig. 1, I–K, the positive controls, respectively. Target cells uninfected or infected with S. ϩ ϩ ϩ ϩ typhi without effector cells were also used as controls. After 16 h of un- vast majority of CD3 CD4 , as well as CD3 CD8 , expressed S. http://www.jimmunol.org/ disturbed incubation, wells were washed and incubated for2hatroom typhi Ags on the cell membrane. In the experiment shown, the temperature with 100 ␮l/well of biotin-labeled anti-human IFN-␥ mAb mean percentages of S. typhi-expressing cells among the different ␮ (clone B133.5; Endogen) (2 g/ml in PBS/1% BSA, ELISPOT buffer). PBMC subpopulations were 79% of CD3ϩ T cells, 86% of CD4ϩ After washing, 100 ␮l/well of avidin-peroxidase (Sigma-Aldrich), diluted T cells, and 82% of CD8ϩ T cells. No statistical differences were 1/400 in ELISPOT buffer, were added and incubated at room temperature ϩ ϩ for 2 h. Wells were then washed and 50 ␮l/well of the detection substrate observed among CD4 and CD8 cell populations expressing S. TrueBlue (Kirkegaard & Perry Laboratories) were added. After 15 min., typhi Ags following S. typhi infection in the various repeat exper- wells were washed with distilled water and allowed to dry. Spots were iments with cells from three different donors ( p ϭ 0.472). Other enumerated using a stereomicroscope. Net frequencies of IFN-␥ spot-form- populations, such as NK cells, B cells, and macrophages, were

ing cells (SFC) were calculated by using the following formula: ((number by guest on September 27, 2021 of SFC in effector cell populations incubated with S. typhi-infected tar- present in very small numbers in blast cell populations, making it gets) Ϫ (number of SFC in effector cell populations incubated with unin- difficult to establish whether these cell populations express S. typhi fected targets ϩ number of SFC in cultures containing S. typhi-infected Ags on their cell membrane following S. typhi infection. Taken target cell populations alone)). The cut-off for positive responses in IFN-␥ 6 together, these results demonstrated the ability of S. typhi to effi- ELISPOT assays (175 SFC/10 PBMC) was established by calculating the ϩ ϩ ϩ ϩ average frequency of IFN-␥-producing cells/106 PBMC when cultured ciently infect a large percentage of CD3 CD4 and CD3 CD8 with not-infected target cells ϩ 3 SE. blasts. Based on these findings, a MOI of 20:1 and an incubation period of 16 h were chosen for the infection of blasts in subsequent Statistical analysis studies. Comparisons between the expression of S. typhi Ags in the different stimulator/target cell subpopulations were performed by one-way Induction of CTL activity by immunization with Ty21a typhoid ANOVA tests. Comparisons of cytotoxicity levels and IFN-␥ produc- vaccine tion by ELISPOT before and after immunization were performed by Student’s t tests. Linear regression analysis was used to correlate the To determine whether immunization with the Ty21a typhoid vac- results obtained by CTL and IFN-␥ ELISPOT assays. All tests were cine elicits the appearance of effector CTL able to lyse S. typhi performed using SigmaStat software (version 2.03; SSPS, Chicago, IL). Ag-expressing blasts, PBMC from Ty21a typhoid vaccinees were Values of p Ͻ 0.05 were considered significant. used ex vivo (i.e., immediately after isolation) as effectors in 51Cr- release cytotoxicity assays. These experiments showed that ex vivo Results PBMC from four volunteers were unable to lyse autologous blasts Expression of S. typhi Ags in PHA-stimulated PBMC infected with S. typhi (data not shown). We next evaluated the We have previously demonstrated the presence of specific CTL ability of PBMC from these volunteers to lyse S. typhi Ag-ex- responses in volunteers immunized with attenuated S. typhi strain pressing blasts following an 8-day in vitro expansion with irradi- CVD 908 and suggested that this might be an important effector ated autologous blasts infected with S. typhi at a PBMC-stimulator mechanism in resistance to S. typhi infection (13). To determine ratio of 7:1. This ratio was selected based on preliminary experi- whether immunization with the licensed typhoid vaccine attenu- ments showing that it generated the highest numbers of effectors ated strain Ty21a also elicits the appearance in circulation of CTL with the highest viability, as indicated by their ability to exclude able to lyse infected targets, we modified the CTL assay previously trypan blue (data not shown). As shown in Fig. 2A, in experiments described (13) by using autologous S. typhi-infected blasts as tar- using preimmunization PBMC as effector cells, low levels of spe- gets instead of B-LCL. As a preliminary step in the generation of cific CTL activity were observed in three of four volunteers at the appropriate target cells, we evaluated whether wt S. typhi was able 10:1 and 3:1 E:T ratios. Significant increases in CTL activity fol- to infect blasts and express Ags on their cell membranes. To this lowing immunization were observed in three of four volunteers end, blasts were incubated with medium alone or with wt S. typhi (Fig. 2A). Although small increases in CTL activity were observed for3hatdifferent MOI. After this incubation, cells were har- in the remaining volunteer (MP), these increases did not reach The Journal of Immunology 2199

FIGURE 1. Expression of intracellular (3 h) and surface (16 h) S. typhi Ags. Blasts were incubated with medium alone, with wt S. typhi strain ISP1820 at 5:1, 20:1, and 40:1 MOI (A–H) or with inactivated (heat-pheno- lized) S. Typhi at a MOI of 10:1. Num- bers correspond to the percentage of positive cells in the indicated regions in each histogram. The percentage of T cell populations expressing S. typhi Ags on the cell membrane after 16 h of Downloaded from culture is shown in I–K. http://www.jimmunol.org/

statistical significance (Fig. 2A). No specific cytotoxic activity was To identify the T cell subpopulation responsible for the ob- observed when uninfected autologous blasts or blasts exposed to served lysis of autologous S. typhi-infected targets, CD3ϩCD8ϩ inactivated S. typhi were used as targets in these assays (Fig. 2B), CD56Ϫ or CD3ϩCD8ϪCD56Ϫ cell populations were isolated from indicating that infection is necessary to enable killing of targets by PBMC obtained from Ty21a-immunized volunteers by positive effector CTL. flow cytometric sorting. Purity of the sorted cell populations was In two Ty21a-immunized volunteers from whom sufficient cells routinely Ͼ90% CD3ϩCD8ϩ for positively sorted populations, as by guest on September 27, 2021 were available, similar CTL results were observed with PBMC determined by flow cytometric analysis. Results demonstrated that expanded in vitro with S. typhi-infected autologous blasts and eval- CTL activity against autologous S. typhi-infected blasts was me- uated against S. typhi-infected autologous B-LCL or blasts (data diated by CD3ϩCD8ϩCD56Ϫ T cells, while no significant activity not shown). These results are consistent with our previous obser- was observed in cultures containing CD3ϩCD8ϪCD56Ϫ cells vations that oral immunization of volunteers with attenuated S. (Fig. 4). To evaluate whether the CD3ϩCD8ϩCD56Ϫ CTL effector typhi strain CVD 908 induces specific cytotoxicity against S. typhi- populations express TCR␣␤ or TCR␥␦, expanded PBMC were infected autologous B-LCL (13). stained with mAbs to CD3, CD4, CD8, TCR␣␤, and TCR␥␦ and The specificity of CTL effector cells was further investigated by analyzed by five-color flow cytometry. Results indicated that vir- competitive inhibition assays using standard techniques (17). As tually all CD3ϩCD8ϩ express TCR␣␤ while Ͻ1% expressed shown in Fig. 3, lysis of 51Cr-labeled autologous S. typhi-infected TCR␥␦ (data not shown). Taken together, these findings demon- blasts was inhibited in a dose-dependent fashion by unlabeled au- strate that CD3ϩCD8ϩCD56Ϫ TCR␣␤ϩ CTL effectors isolated tologous S. typhi-infected blasts. In contrast, no inhibition was from volunteers immunized with S. typhi strain Ty21a are respon- observed when unlabeled autologous uninfected blasts were added sible for the killing of autologous S. typhi-infected targets. to the cultures. Induction of IFN-␥-secreting effector T cells Characterization of the effector CTL population We next examined whether immunization with Ty21a elicits the As a first step to identify the effector cell population responsible appearance of effector cells able to secrete IFN-␥ in response to for CTL activity, PBMC obtained from Ty21a-immunized volun- stimulation with S. typhi-infected blasts, as well as the frequency teers following an 8-day in vitro expansion with S. typhi-infected of these effector cells. To this end, we developed an ex vivo IFN-␥ autologous blasts were used in CTL assays unfractionated or fol- ELISPOT assay using PBMC from immunized volunteers as ef- lowing depletion of NK cells using anti-CD56-coated magnetic fectors and autologous blasts infected with wt S. typhi as stimula- beads. Flow cytometric analysis of the depleted cell populations tors. Significant increases in the net frequency of IFN-␥ SFC were consistently showed that they contained Ͻ1.5% CD56ϩ NK cells. observed in four of five volunteers following immunization (Fig. Depletion of NK cells did not affect the ability of in vitro-expanded 5). Increases in the net frequency of IFN-␥ SFC following immu- PBMC from these volunteers to lyse autologous S. typhi-infected nization as compared with preimmunization levels ranged from blasts, indicating that NK cells are not responsible for the observed 185Ð548 SFC/106 PBMC (mean Ϯ SD, 393 Ϯ 172). CTL activity in unfractionated PBMC populations (data not In control cultures, the presence of IFN-␥-producing cells in shown). Of note, depletion of NK cells decreased to a considerable effector PBMC incubated with uninfected blasts (mean Ϯ SD, extent the observed background cytotoxicity (10Ð25% at 10:1 E:T 220 Ϯ 119; range 0Ð403 SFC/106 PBMC) was more frequent than ratio). when effectors were incubated in the presence of medium alone 2200 CD8 RESPONSES IN Ty21a VACCINEES

FIGURE 4. Speciﬁc cytotoxicity of S. typhi-infected blasts is mediated by CD8ϩ effector T cells. Unfractionated PBMC, CD3ϩCD8ϩCD56Ϫ or CD3ϩCD8ϪCD56Ϫ sorted cell populations were used as effector cells against S. typhi-infected autologous blast targets. Bars represent the mean percent-speciﬁc cytotoxicity observed for each target at an E:T cell ratio of 10:1. Error bars show the SD of triplicate cultures. The dashed line represents the cut-off for positive 51Cr-release assays. Downloaded from

teers whether infection of blasts with live S. typhi is required to induce IFN-␥ production. Only stimulators exposed to live S. typhi, but not those incubated with inactivated S. typhi, elicited IFN-␥ production as determined by ELISPOT (data not shown).

␥ http://www.jimmunol.org/ FIGURE 2. Ability of PBMC effectors to lyse S. typhi-infected autolo- Characterization of the IFN- -secreting effector cell population gous target cells. A, The ability of 8-day restimulated PBMC obtained pre- We next investigated which cell populations in PBMC from im- or postimmunization to lyse autologous blast targets infected with live S. munized volunteers secrete IFN-␥ when stimulated with S. typhi- p ϭ 0.10 infected autologous blasts. To this end, PBMC were negatively ,ءء p Ͻ 0.05 and ,ء .typhi as described in Materials and Methods by Student’s t test. B, The ability of PBMC obtained from volunteers depleted of CD8ϩ, CD4ϩ, or NK cells using immunomagnetic postimmunization to lyse targets incubated with medium, inactivated, or beads. Flow cytometric analysis of the depleted cell populations live S. typhi. Bars represent the mean percent-specific lysis observed for indicated that virtually all NK and CD4ϩ cells and ϳ80% of the each volunteer at the indicated E:T cell ratios. Error bars show the SD of ϩ quadruplicate cultures. The dashed line represents the cut-off for positive CD8 cells present in PBMC populations were removed by this 51 procedure (data not shown). Depleted cell populations were stim- Cr-release assays. by guest on September 27, 2021 ulated with autologous blasts infected or not with S. typhi. Results indicate that CD8 depletion abrogated the presence of IFN-␥-se- (virtually none detected; data not shown). In positive control wells creting cells in response to S. typhi-infected autologous blasts, (effector PBMC incubated with CD3/CD28 beads), the frequency whereas depletion of NK cells did not significantly alter the num- ␥ 6 of IFN- -producing cells ranged from 2,693 to 3,860 SFC/10 ber of IFN-␥-secreting cells observed in unfractionated PBMC Ϯ Ϯ PBMC (mean SD, 3182 451). In three of the volunteers for (Fig. 6). In contrast, CD4-depleted populations exhibited increased whom sufficient cells were available, increased numbers of SFC/ numbers of IFN-␥-secreting SFC, likely the result of increased 6 10 PBMC following immunization in response to S. typhi-in- proportions of CD8ϩ T cells in this population. These results sug- fected blasts were observed when assayed in three independent gest that CD8ϩ T lymphocytes are the cells within the PBMC experiments (data not shown). We also evaluated in these volun- population primarily responsible for IFN-␥ secretion in response to stimulation with autologous S. typhi-infected cells.

FIGURE 5. Frequency of IFN-␥-producing cells before and after im- FIGURE 3. Analysis of CTL activity by competitive inhibition assay. munization in the presence of autologous blasts infected with S. typhi. Net Effector cells expanded in vitro for 8 days were tested for lysis of 51Cr- frequencies of IFN-␥ SFC were assessed by an IFN-␥ ELISPOT assay labeled S. typhi-infected autologous blast targets (hot targets) at an E:T using ex vivo PBMC from immunized volunteers as effectors and autolo- ratio of 40:1 in the absence or presence of varying numbers of unlabeled gous blasts infected with wt S. typhi as stimulators. Net frequencies of autologous target cells (cold targets). The cold targets were autologous IFN-␥ SFC were calculated as described in Materials and Methods. The blasts uninfected (F) or infected with S. typhi (f). These results are rep- dashed line represents the cut-off for positive ELISPOT assays determined .p Ͻ 0.05 by Student’s t test ,ء .resentative of one of two experiments with cells from different donors. as described in Materials and Methods The Journal of Immunology 2201

To confirm that CD8ϩ T lymphocytes are the cells that secrete IFN-␥ in response to stimulation with autologous S. typhi Ag- expressing targets, PBMC from immunized volunteers were cocultured with autologous blasts infected or not with S. typhi. After 16 h (the same time point at which the ELISPOT assay is assessed), the intracellular levels of IFN-␥ in the various cell populations were determined by flow cytometry. Preliminary experiments in which CD45 was used to exclude targets cells from analysis demonstrated that Ͻ2% of the cells remain CD45ϩ within the “live” lymphocyte gate (as defined in forward vs side light scatter cytograms) following the overnight stimulation with S. typhi-infected cells. Thus, CD45 staining was not used in subsequent experiments. Costaining for intracellular IFN-␥ and cell surface markers demonstrated that CD3ϩCD4ϪCD8ϩCD56Ϫ T cells rather than CD3ϩCD4ϩCD8ϪCD56Ϫ T cells or NK cells were the dominant sources of IFN-␥ in postimmunization effector PBMC populations. Increases of 2- to 5-fold in the frequency of ϩ ϩ Ϫ Ϫ ␥ FIGURE 7. Detection of intracellular IFN-␥ levels in response to S. CD3 CD8 CD4 CD56 cells expressing IFN- were observed ϩ ϩ Ϫ Ϫ in populations isolated from volunteers following immunization as typhi Ags in CD3 CD8 CD4 CD56 cells. PBMC from immunized vol- Downloaded from unteers were stimulated with targets cells incubated with medium, inacti- compared with preimmunization levels (Fig. 7). In contrast, ϩ Ϫ Ϫ Ϫ Ϫ ϩ vated, or live S. typhi for 16 h, stained for CD3, CD8, CD4, and CD56 CD4 CD8 CD56 or CD4 CD8 CD56 cell populations ex- surface markers and for intracellular IFN-␥ as described in Materials and ␥ Ͻ ϩ ϩ Ϫ pressing IFN- showed a variation of 2-fold (data not shown). Methods. Data are expressed as the percentage of CD3 CD8 CD4 ϩ ϩ Ϫ Ϫ ϩ Of note, increased numbers of CD3 CD8 CD4 CD56 IFN-␥ CD56Ϫ cells expressing IFN-␥. Similar results were obtained with an ad- effector cells (3.3%) were observed in volunteer TS (Fig. 7). No ditional volunteer (RH). increases in IFN-␥ production were observed when effectors were http://www.jimmunol.org/ incubated in the presence of stimulators exposed to inactivated S. typhi (Fig. 7) indicating that infection is required to trigger IFN-␥ production by effector cells. Discussion In this study, we described an in vitro model using S. typhi-ex- Correlation of CTL and IFN-␥-secreting effector T cell pressing autologous blasts to investigate CMI in volunteers immu- populations in Ty21a vaccinees nized with S. typhi strain Ty21a. Our data showed that oral im- To evaluate whether the frequency of cells that secrete IFN-␥ in munization with attenuated vaccine strain Ty21a induced potent ␥ ϩ ϩ response to stimulation with autologous S. typhi-infected cells cor- CTL responses as well as IFN- production by CD3 CD8 T by guest on September 27, 2021 relates with CTL activity, we compared the results obtained by cells in volunteers of different HLA-class I haplotypes. Moreover, ELISPOT with those obtained by 51Cr-release assays. A positive our data suggest that the Ty21a vaccine can induce cellular im- correlation was found between the frequency of IFN-␥-secreting mune responses lasting for at least 2 years after immunization cells and the cytotoxic activity of CTL effectors at E:T ratios of (volunteers MR and SD, Table I). 10:1 (r ϭ 0.954, p Ͻ 0.001) or at E:T ratios of 3:1 (r ϭ 0.892, p ϭ 0.003; Fig. 8). When we compared the cytolytic activity at an E:T ratio of 10:1 with the frequency of IFN-␥-secreting cells, as measured by intracellular IFN-␥ staining (r ϭ 0.751, p ϭ 0.085), or the frequencies of IFN-␥-secreting cells measured by ELISPOT with those measured by intracellular IFN-␥ staining (r ϭ 0.755, p ϭ 0.083), we found trends rather than statistical significant correla- tions, likely because of the small number of volunteers studied.

FIGURE 6. Cellular sources of IFN-␥-producing cells. PBMC from a Ty21a vaccinee were depleted of CD8ϩ, CD4ϩ,orNKϩ cells by negative selection using immunomagnetic beads and stimulated with autologous FIGURE 8. Correlation between the frequency of IFN-␥ SFC detected blasts infected or not with S. typhi. After 16 h, IFN-␥ SFC were detected by ELISPOT assays and cytolytic activity of CTL effectors at E:T ratios of using a modiﬁed ELISPOT assay. Net frequencies of IFN-␥ SFC were 10:1 (F; long dashed line) (r ϭ 0.954, p Ͻ 0.001) and 3:1 (E; short dashed calculated as described in Materials and Methods. Error bars show the SD line) (r ϭ 0.892, p ϭ 0.003) detected by 51Cr-release assay. Pre- and of triplicates cultures. These results are representative of those observed postimmunization PBMC from volunteers RH, TS, MR, and MP were used with two different donors evaluated in two independent experiments. for this analysis. 2202 CD8 RESPONSES IN Ty21a VACCINEES

The efficiency of the immune system to control infection by volunteers with Ty21a elicits IFN-␥ production in response to S. intracellular pathogens depends to a large extent on the ability of typhi flagella (16). activated effector cells to interact with infected host cells present- The effector CTL activity was mediated by CD3ϩCD8ϩ T cells, ing antigenic peptides in the context of HLA class I or II mole- but not by NK and CD4ϩ T cells. These findings are consistent cules. To investigate the role of CMI in S. typhi infection, we with our prior observations that B-LCL-expressing S. typhi pro- explored the use of S. typhi-infected autologous blasts as APC and teins can be lysed by specific CD8ϩ CTL from volunteers immu- as targets. Their use affords a more “physiological target cell pop- nized with attenuated S. typhi strain CVD 908 (13). Moreover, ulation” because they are primary cells and not cell lines like B- these results provide additional evidence to support the observa- LCL. Moreover, because autologous blasts are infected by wt S. tion that CTL effectors are able to recognize and kill both S. typhi- typhi, a significant feature of our system is its capacity to express infected autologous blasts and B-LCL suggesting that CTL recog- multiple epitopes of S. typhi in the context of various HLA mol- nition of S. typhi-infected B-LCL might not differ from that of ecules, thereby allowing CTL of different specificities to recognize blasts. infected cells. By using a MOI of 20:1, blasts could be infected A second mechanism by which activated effector T cells can with high efficiency, as demonstrated by flow cytometric assess- contribute to protection is the release of cytokines, among which ment of intracellular and surface S. typhi Ags, and maintained IFN-␥ is of paramount importance. Intracellular bacteria, including good viability following infection. Another important advantage is S. typhi (12, 16, 21), can stimulate IFN-␥ production by specificT its accessibility; blasts can be readily established from PBMC in cells, which in turn can lead to the recruitment and/or activation of only 1 wk instead of the ϳ4Ð6 wk required to establish B-LCL. the microbicidal activities of macrophages, neutrophils, and/or NK Downloaded from Blasts have been successfully used as target cells in viral systems cells, as well as increases in the expression of immunologically using human cells (17, 20). relevant molecules such as those involved in Ag presentation (e.g., One mechanism by which activated effector cells can contribute HLA-class I molecules) (22Ð24). In the present study, we have to protection is to lyse infected target cells. In this study, PBMC adapted an IFN-␥ ELISPOT assay to quantitate the frequency of were tested ex vivo against blasts expressing S. typhi Ags. We IFN-␥-producing S. typhi Ag-specific T cells. This assay is more found that the use of PBMC as effector cells ex vivo did not result sensitive and does not require the in vitro expansion step for the in any detectable lysis of S. typhi-infected targets. However, spe- detection of specific T cell activity. Results indicated that the stim- http://www.jimmunol.org/ cific expansion of PBMC with S. typhi-infected autologous blasts ulation of effector cells by blasts infected with S. typhi resulted in resulted in strong CTL activity toward S. typhi-infected cells in the generation of S. typhi Ag-specific CD8ϩ T cells able to pro- most volunteers. In these studies, we observed that infection of duce IFN-␥, whereas PBMC depleted of this population do not. target cells with live S. typhi is required for lysis by effector CTL. Interestingly, the observed increase in S. typhi Ag-specific T cell These results extend previous observations with PBMC from vol- frequencies following immunization (i.e., 85Ð548 SFC/106 unteers immunized orally with attenuated S. typhi strain CVD 908 PBMC) is similar to the increases in frequencies of IFN-␥-secret- which showed that the use of live bacteria is critical to observe ing specific T cells observed in volunteers naturally exposed to

CTL activity (13). Of note, low levels of lysis of S. typhi-infected intracellular pathogens (e.g., CMV, M. tuberculosis,influenza, by guest on September 27, 2021 targets by preimmunization CTL were observed at a 10:1 E:T ratio EBV, and Plasmodium falciparum (25Ð28) or following vaccina- in some volunteers. However, postimmunization levels of CTL tion (29, 30). activity against S. typhi-infected targets were consistently above In this report, we showed that CD8ϩ T cells isolated from Ty21a this background lysis. This background lysis may be the conse- vaccinees are not only able to lyse S. typhi-infected blasts, but also quence of cross-reactivity with other Gram-negative bacteria. Pre- are potent producers of IFN-␥ and demonstrated a significant as- vious studies in the murine model have shown that epitopes shared sociation between S. typhi-specific CTL and IFN-␥ production. among Salmonella species, as well as closely related Gram-nega- These findings strongly support the involvement of cells with CTL tive bacteria, can be recognized by a same CTL population (19). activity in IFN-␥ production. A close association between the fre- Alternatively, this background can be the result of pre-exposure to quency of IFN-␥-secreting cells and CTL activity also has been S. typhi before immunization with Ty21a. Of note, the volunteer reported in other systems (31). However, our results do not ex- with the highest preimmunization CTL activity (volunteer TS) is a clude that different subpopulations of CD8ϩ T cells might mediate native from Kenya who immigrated to the U.S. as a teenager, different effector functions, i.e., cytolytic activity or cytokine pro- increasing the likelihood of previous exposure to Salmonella. To duction, acting in concert and complementing each other. evaluate whether volunteer TS might have been previously ex- In summary, this work constitutes the first evidence that immu- posed to S. typhi or cross-reactive Ags, we measured IFN-␥ pro- nization of humans with Ty21a typhoid vaccine elicits specific duction by PBMC isolated before and after immunization from CTL and provides an estimate of the frequency of IFN-␥-secreting three of the volunteers (TS, MR, and RH) in response to soluble S. cells following oral vaccination. Although our data do not provide typhi flagella Ag. We observed that PBMC obtained before im- a direct indication that CD8ϩ-mediated CTL and IFN-␥ responses munization from TS, but not those from two other volunteers, se- play a critical role in protection from typhoid fever, their presence creted significant amounts of IFN-␥ in response to S. typhi flagella following vaccination with Ty21a suggests that these effector (R. Salerno-Gonçalves and M. B. Sztein, unpublished observa- mechanisms might indeed be involved in controlling S. typhi in- tions). These results, together with the observation of increased infection. Similarly, we have previously reported that immunization tracellular IFN-␥ levels by preimmunization PBMC from volunteer with attenuated strains of S. typhi elicits the appearance in circu- TS following exposure to S. Typhi-infected blasts (Fig. 7), further lation of CD8ϩ, MHC class I-restricted, CTL effector cells capable support the hypothesis that this individual has indeed been previ- of killing autologous S. typhi-infected targets, as well as sensitized ously exposed to S. typhi or cross-reactive Ags. All volunteers, in- lymphocytes that proliferate and produce IFN-␥ in response to cluding TS, exhibited significant increases in IFN-␥ production to stimulation with S. typhi Ags, suggesting that CMI responses may S. typhi flagella following immunization (R. Salerno-Gonçalves and play a role in limiting the progression of typhoid infection (12, 13). M. B. Sztein, unpublished observations). Moreover, these results In addition to these CMI responses, it is likely that serum and confirm and extend previous observations that oral immunization of mucosal Ab responses also contribute to protection. Results from The Journal of Immunology 2203 a field study in volunteers vaccinated with the typhoid Ty21a vac- lymphoproliferative responses in volunteers orally immunized with attenuated cine strain suggested a correlation between increased serum levels vaccine strains of Salmonella typhi. J. Infect. Dis. 170:1508. 13. Sztein, M. B., M. K. Tanner, Y. Polotsky, J. M. Orenstein, and M. M. Levine. of IgG O Ab and protective efficacy (7). Based on this study, 1995. Cytotoxic T lymphocytes after oral immunization with attenuated vaccine although serum O Abs are not believed to be the main effector strains of Salmonella typhi in humans. J. Immunol. 155:3987. immune mechanism in protection to S. typhi infection, increased 14. Levine, M. M., C. Ferreccio, P. Abrego, O. S. Martin, E. Ortiz, and S. Cryz. 1999. Duration of efficacy of Ty21a, attenuated Salmonella typhi live oral vaccine. levels of IgG to S. typhi LPS have been proposed as a surrogate Vaccine 17:S22. marker of protection. Finally, the fact that the licensed parenteral 15. Walker, B. D. 1990. HIV-1-Specific Cytotoxic T Lymphocytes. Stockton Press, Vi polysaccharide vaccine elicits serum Vi Abs indicates that al- New York. 16. Wyant, T. L., M. K. Tanner, and M. B. Sztein. 1999. Salmonella typhi flagella are most certainly protection with this vaccine is mediated by means potent inducers of proinflammatory cytokine secretion by human monocytes. In- of serum Vi Abs (32). fect. Immun. 67:3619. Direct evidence for the role of CMI in protection from S. typhi 17. Salerno-Goncalves, R., W. Lu, A. Achour, and J. M. Andrieu. 1998. Lysis of CD4ϩ T cells expressing HIV-1 gag peptides by gag-specific CD8ϩ cytotoxic T infection might be provided in clinical trials in which CMI re- cells. Immunol. Lett. 64:71. sponses are correlated with protection from exposure to wt S. typhi 18. Cox, J. H. 1998. HIV-1-specific cytotoxic T-cell assays. In HIV Protocols. in healthy volunteers or individuals immunized with attenuated N. Michael and J. H. Kim, eds. Humana Press, Totawa. 19. Lo, W. F., H. Ong, E. S. Metcalf, and M. J. Soloski. 1999. T cell responses to strains of S. typhi vaccine candidates. Gram-negative intracellular bacterial pathogens: a role for CD8ϩ T cells in immunity to Salmonella infection and the involvement of MHC class Ib molecules. Acknowledgments J. Immunol. 162:5398. 20. Fang, S. H., B. L. Chiang, M. H. Wu, H. Iba, M. Y. Lai, P. M. Yang, D. S. Chen, We thank the volunteers who allowed us to perform this study. We also and L. H. Hwang. 2001. Functional measurement of hepatitis C virus core-spe- thank Dr. Marcelo Fernandez-Vinã (Naval Medical Research Center/ cific CD8ϩ T-cell responses in the livers or peripheral blood of patients by using Downloaded from Georgetown University) for HLA Ag typing; Dr. Myron M. Levine for autologous peripheral blood mononuclear cells as targets or stimulators. J. Clin. critical review of the manuscript; Dr. Bernadette McConnell and the staff Microbiol. 39:3895. 21. Viret, J. F., D. Favre, B. Wegmuller, C. Herzog, J. U. Que, S. J. Cryz, Jr., and from the Blood Bank of University of Maryland hospital for their help in A. B. Lang. 1999. Mucosal and systemic immune responses in humans after collecting leukapheresis specimens; and Regina Harley for excellent tech- primary and booster immunizations with orally administered invasive and non- nical assistance in flow cytometric determinations and sorting. invasive live attenuated bacteria. Infect. Immun. 67:3680. 22. Harty, J. T., and M. J. Bevan. 1999. Responses of CD8ϩ T cells to intracellular

bacteria. Curr. Opin. Immunol. 11:89. http://www.jimmunol.org/ References 23. Fruh, K., and Y. Yang. 1999. Antigen presentation by MHC class I and its 1. Eisenstein, T. K. 1999. Mucosal immune defense: the Salmonella typhimurium regulation by interferon-␥. Curr. Opin. Immunol. 11:76. model. In Intracellular Bacterial Vaccine Vectors. Y. Paterson, ed. Wiley-Liss, 24. Raupach, B., and S. H. Kaufmann. 2001. Immune responses to intracellular bac- New York, p. 51. teria. Curr. Opin. Immunol. 13:417. 2. Levine, M. M. 1994. Typhoid fever vaccines. In Vaccines. S. A. Plotkin and 25. Lalvani, A., R. Brookes, S. Hambleton, W. J. Britton, A. V. Hill, and E. A. Mortimer, eds. W.B. Saunders, Philadelphia, p. 597. A. J. McMichael. 1997. Rapid effector function in CD8ϩ memory T cells. J. Exp. 3. Ivanoff, B., M. M. Levine, and P. H. Lambert. 1994. Vaccination against typhoid Med. 186:859. fever: present status. Bull. W. H. O. 72:957. 26. Gonzalez, J. M., K. Peter, F. Esposito, I. Nebie, J. M. Tiercy, A. Bonelo, 4. Levine, M. M., and O. S. Levine. 1997. Influence of disease burden, public M. Arevalo-Herrera, D. Valmori, P. Romero, S. Herrera, et al. 2000. HLA- perception, and other factors on new vaccine development, implementation, and A*0201 restricted CD8ϩ T-lymphocyte responses to malaria: identification of continued use. Lancet 350:1386. new Plasmodium falciparum epitopes by IFN-␥ ELISPOT. Parasite Immunol 5. Mittrucker, H. W., and S. H. Kaufmann. 2000. Immune response to infection with 22:501. by guest on September 27, 2021 Salmonella typhimurium in mice. J. Leukocyte Biol. 67:457. 27. Pathan, A. A., K. A. Wilkinson, R. J. Wilkinson, M. Latif, H. McShane, 6. Levine, M. M., C. O. Tacket, J. E. Galen, E. M. Barry, F. Noriega, and G. Pasvol, A. V. Hill, and A. Lalvani. 2000. High frequencies of circulating M. B. Sztein. 1997. Progress in development of new attenuated strains of Sal- IFN-␥-secreting CD8 cytotoxic T cells specific for a novel MHC class I-restricted monella typhi as live oral vaccines against typhoid fever. In New Generation Mycobacterium tuberculosis epitope in M. tuberculosis-infected subjects without Vaccines. M. M. Levine, G. C. Woodrow, J. B. Kaper, and G. S. Cobon, eds. disease. Eur. J. Immunol. 30:2713. Marcel Dekker, New York, p. 437. 7. Levine, M. M., C. O. Tacket, and M. B. Sztein. 2001. Host-Salmonella interac- 28. Brice, G. T., N. L. Graber, S. L. Hoffman, and D. L. Doolan. 2001. Expression of the chemokine MIG is a sensitive and predictive marker for antigen-specific, tion: human trials. Microbes Infect. 3:1271. ␥ ␥ 8. Lenz, L. L., and M. J. Bevan. 1997. CTL responses to H2ÐM3-restricted Listeria genetically restricted IFN- production and IFN- -secreting cells. J. Immunol. epitopes. Immunol. Rev. 158:115. Methods 257:55. 9. Pamer, E. G., A. J. Sijts, M. S. Villanueva, D. H. Busch, and S. Vijh. 1997. MHC 29. Wang, R., J. Epstein, F. M. Baraceros, E. J. Gorak, Y. Charoenvit, D. J. Carucci, R. C. Hedstrom, N. Rahardjo, T. Gay, P. Hobart, et al. 2001. Induction of CD4ϩ class I antigen processing of Listeria monocytogenes proteins: implications for ϩ dominant and subdominant CTL responses. Immunol. Rev. 158:129. T cell-dependent CD8 type 1 responses in humans by a malaria DNA vaccine. 10. Sieling, P. A., D. Chatterjee, S. A. Porcelli, T. I. Prigozy, R. J. Mazzaccaro, Proc. Natl. Acad. Sci. USA 98:10817. T. Soriano, B. R. Bloom, M. B. Brenner, M. Kronenberg, P. J. Brennan, et al. 30. Lopez, J. A., C. Weilenman, R. Audran, M. A. Roggero, A. Bonelo, J. M. Tiercy, F. Spertini, and G. Corradin. 2001. A synthetic malaria vaccine elicits a potent 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science ϩ ϩ 269:227. CD8 and CD4 T lymphocyte immune response in humans: implications for 11. Winter, N., M. Lagranderie, S. Gangloff, C. Leclerc, M. Gheorghiu, and vaccination strategies. Eur. J. Immunol. 31:1989. ϩ B. Gicquel. 1995. Recombinant BCG strains expressing the SIVmac251nef gene 31. Butz, E. A., and M. J. Bevan. 1998. Massive expansion of antigen-specific CD8 induce proliferative and CTL responses against nef synthetic peptides in mice. T cells during an acute virus infection. Immunity 8:167. Vaccine 13:471. 32. Klugman, K. P., I. T. Gilbertson, H. J. Koornhof, J. B. Robbins, R. Schneerson, 12. Sztein, M. B., S. S. Wasserman, C. O. Tacket, R. Edelman, D. Hone, D. Schulz, M. Cadoz, and J. Armand. 1987. Protective activity of Vi capsular A. A. Lindberg, and M. M. Levine. 1994. Cytokine production patterns and polysaccharide vaccine against typhoid fever. Lancet 2:1165.