Simultaneous Th1-Type Cytokine Expression Is a Signature of Peritoneal CD4+ Lymphocytes Responding to Infection with Listeria monocytogenes
http://www.100md.com
免疫学杂志 2005年第13期
Abstract
The robust murine response to infection with Listeria monocytogenes makes an excellent model to study the functional development of immune cells. We investigated the cellular immune response to i.p. infection using intracellular cytokine staining to identify Ag-specific lymphocytes. CD4+ peritoneal exudate cells obtained 10 days postinfection predominantly coexpressed TNF-, IFN-, and IL-2 after polyclonal or Ag stimulation. A population of cells simultaneously making TNF- and IFN- was also detected but at a lower frequency. By following the kinetics of the response to Listeria, we found that CD4+ lymphocytes coexpressing TNF- and IFN- dominated on day 6 postinfection and then declined. From days 10–27, TNF-+IFN-+IL-2+ (triple-positive) was the most prevalent cytokine phenotype, and the frequency steadily declined. These characteristic cytokine expression patterns were observed in both primary and secondary responses to Listeria infection and developed even when infection was terminated with antibiotic treatment. A cytokine-assisted immunization procedure resulted in both double- and triple-positive cells, but the clear predominance of triple-positive cells required Listeria infection. Triple-positive cells were preferentially noted in the peritoneal cavity tissue site; spleen cells displayed a predominant population of double-positive T cells (TNF-+IFN-+). We speculate that the appearance of triple-positive cells represents a functionally significant subset important in host defense at nonlymphoid tissue sites.
Introduction
Listeria monocytogenes is a Gram-positive, intracellular bacterium that causes illness in susceptible populations, such as pregnant, immunocompromised, and elderly individuals. It is also an excellent model organism used to study the complex and coordinated immune response by mice to a pathogen (1, 2). Clinically, Listeria gains access to its host orally, although other routes of infection, including i.p. and i.v., are used experimentally. Listeria infects and replicates inside macrophages and hepatocytes before spreading to other cells via actin propulsion (1). Within hours after infection, neutrophils are recruited to phagocytose and destroy infected cells; they are later replaced by macrophages (3). T cells and NK cells also have roles in the initial immune response to Listeria (4, 5). Between 7–10 days postinfection (p.i.),3 Listeria-specific effector T cells peak, eliminating remaining infected cells, and then retract, leaving long-lived memory cells (6, 7). Both CD4+ and CD8+ T cells contribute to bacterial clearance through destruction of infected cells and cytokine secretion, which directs macrophage activation and granuloma formation (8, 9, 10, 11).
The immune response to Listeria is finely coordinated through the actions of cytokines. Mice injected with neutralizing Abs to TNF- had increased susceptibility to Listeria and decreased listericidal activity by peritoneal macrophages (12). In mice treated with neutralizing IFN- Ab (13) or genetically lacking IFN- (14) or its receptor (15), typically sublethal doses of Listeria resulted in death. TNF- and IFN- produced during the initial days of Listeria infection function cooperatively to activate macrophages and organize granulomas that ultimately result in the elimination of bacterium and infected cells (16). This proinflammatory state promotes Th1 lymphocyte development, characterized by the expression of TNF-, IFN-, and/or IL-2 by CD4+ Ag-specific T cells. Functionally, these cells and their secreted products are important for efficient dendritic cell activation and subsequent development of effective memory CD8+ T cells (17, 18, 19). IL-2 plays multiple roles in the immune response, including T cell clonal expansion and contraction and T regulatory cell function (20, 21). Differential IL-2 expression by CD4+ T cells is a result of heterogeneous activation by dendritic cells (22, 23), and IL-2 expression may be a characteristic of highly differentiated T cells. Coexpression of TNF- and/or IFN- by IL-2-producing CD4+ T cells may also reflect activation heterogeneity and implies that simultaneous secretion of three cytokines with divergent actions may be a hallmark of fully differentiated lymphocytes. Cytokine coexpression patterns may be either random or deliberate. If consistent deliberate patterns are observed, the conditions that determine cytokine heterogeneity can be defined, and the potential functional significance of concomitant delivery of cytokine mixtures can be addressed.
The goal of this paper was to follow Listeria-specific CD4+ lymphocyte development based on cytokine secretion patterns. To do this, we used intracellular cytokine staining to examine the simultaneous expression of three Th1 cytokines on a single-cell level. This novel approach identified TNF-+IFN-+IL-2+ (triple-positive) CD4+ peritoneal lymphocytes that were generated after polyclonal or Ag stimulation. The data reveal defined cytokine expression patterns among Ag-specific T cells that were unique to the time after infection and the tissue site examined. Cytokine patterns were a signature of in vivo differentiation, rather than in vitro stimulation conditions. Functionally, triple-positive cells may be important in protective immunity. The novel characterization of Listeria-specific cells based on combinatorial cytokine expression is a useful approach to explore heterogeneity among responding T cell populations.
Materials and Methods
Mice and injections
C57BL/6 mice, obtained from Charles River Laboratories and The Jackson Laboratory, were housed in filter-topped microisolator cages in a specific pathogen-free facility and were used at 8–20 wk of age. The institutional animal care and use committee approved all experimental procedures. L. monocytogenes wild-type strain 43251 (American Type Culture Collection) was grown overnight in brain-heart infusion broth (BHI; Difco) at 37°C and washed twice in PBS before i.p. injection. Bacterial concentrations were determined by measuring the OD and were confirmed by plating on BHI agar plates. A sublethal dose of 2 x 104 bacteria was used for primary infections, and secondary infections were 2 log higher. For some studies, groups of mice were treated with ampicillin (2 mg/ml) in their drinking water to truncate infection. Treatment was started 24 h after Listeria injection, and fresh ampicillin-containing water was administered every 2–3 days. Spleen cells were cultured on BHI agar plates to ensure bacterial clearance 24 h after ampicillin treatment (data not shown).
For peptide immunization studies, mice were immunized i.p. with three doses of rIL-12 (0.5 μg/mouse/dose) and listeriolysin O190–201 (LLO190–201) peptide (30 μM/mouse/dose). LLO190–201 (LLO 190) is an immunodominant, MHC class II-restricted peptide (NEKYAQAYPNVS) derived from LLO (24). Successive IL-12 plus peptide injections induce protective immunity against subsequent live Listeria challenge (25).
Cell preparation and culture
Peritoneal exudate cells (PECs) were harvested by lavage with cold HBSS containing 0.06% BSA, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.1 M HEPES buffer, 0.04% sodium bicarbonate, and 10 U/ml heparin. PECs from each group of mice (two to eight mice per group, as indicated in the figure legends for each experiment) were pooled. Single-cell suspensions were made from excised and pooled spleens using a glass homogenizer, followed by lysis of erythrocytes with ammonium chloride buffer. Cells from both sites were washed and resuspended for culture in RPMI 1640 supplemented with 10% FCS, 5 x 10–5 M 2-ME, 0.5 mM sodium pyruvate, 10 mM HEPES buffer, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine. For some experiments, T cells from PECs or spleen suspensions were negatively selected using T cell enrichment columns according to the manufacturer’s instructions (R&D Systems). During culture, enriched T cells were treated identically to their unenriched counterparts. Effective depletion of APCs was evidenced by the 90% reduction in cytokine response upon stimulation of enriched T cells with peptide Ag (data not shown).
Abs for flow cytometry
The following reagents were purchased from BD Pharmingen: RM4-5-PerCP (anti-CD4), MP6-XT22-FITC (anti-TNF-), XMG1.2-PE (anti-IFN-), and JES6-5H4-allophycocyanin (anti-IL-2). Isotype controls for these Abs were purchased as follows: rat IgG-FITC from Southern Biotechnology Associates and rat IgG1-PE, rat IgG2a-PerCP, and rat IgG2b-allophycocyanin from BD Pharmingen.
Detection of intracellular cytokines in individual lymphocyte populations by flow cytometry
Various stimuli were used during a 5-h culture period to induce cytokine production. Polyclonal, or nonspecific, stimulation was provided by either PMA (10 ng/ml) and ionomycin (1 μM; Calbiochem) or plate-bound anti-CD3 (145-2C11; 5 μg/ml) and soluble anti-CD28 (37.51; 10 μg/ml) purchased from BD Pharmingen. For Ag-specific stimulation, cells were cultured with either heat-killed Listeria (HKLM; 107/ml) or LLO 190 (30 μg/ml). Brefeldin A (10 μg/ml; Sigma-Aldrich) was added for the last 4 h to inhibit cytokine secretion and enhance intracellular detection (26). Dose-response curves and stimulation time courses were conducted to optimize the detection of cytokine-expressing CD4+ lymphocytes (data not shown).
For analysis by flow cytometry, 1–2 x 106 cells were incubated for 30 min at 4°C with PerCP-conjugated mAb to cell surface markers to identify individual lymphocyte populations, then were washed twice with FACS wash buffer (PBS, 3% FCS, and 0.1% sodium azide). Cells were incubated for 15 min at room temperature with 50 μl of fixation medium (Fix and Perm kit; Caltag Laboratories) and washed once. Predetermined optimal concentrations of fluorochrome-conjugated anti-cytokine Abs diluted in permeabilization buffer (Fix and Perm kit) were added for 15 min at room temperature, followed by two washes and resuspension in FACS wash buffer. Background fluorescence was <0.5% after incubation with Ig isotype controls conjugated to fluorochromes (data not shown). Cells were analyzed on a FACSCalibur flow cytometer (BD Biosciences) by gating on the lymphocyte population, as defined by forward scatter/side scatter parameters.
A logical gating scheme was used to identify CD4+ lymphocytes that were expressing a cytokine or combinations of cytokines. The data from each sample were initially plotted on a forward scatter by cytokine (TNF-, IFN-, or IL-2) dot plot. A region (R) was drawn around those cells that were positive for the desired cytokine, i.e., R1 for TNF-+ cells, R2 for IFN-+ cells, and R3 for IL-2+ cells. Under the gating menu in CellQuest (BD Biosciences), eight logical gates were assigned as follows: TNF-+, R1 (not R2 or R3); IFN-+, R2 (not R1 or R3); IL-2+, R3 (not R1 or R3); TNF-+IFN-+, R1*R2 (not R3); TNF-+IL-2+, R1*R3 (not R2); IFN-+IL-2+, R2*R3 (not R1); TNF-+IFN-+IL-2+, R1*R2*R3; and cytokine negative, not R1, R2, or R3. Each possible combination was arbitrarily assigned a color, as shown in Fig. 2. Additionally, only those cells that fell in a CD4+ gate (R4) as well as a lymphocyte gate (R5; as determined by characteristic forward scatter by side scatter appearance) are displayed in Fig. 2. As a consequence of this gating scheme, the same data are shown repeatedly, but with respect to three different parameters, i.e., the same TNF-+ cells are shown in all three plots. Examining all three plots is informative with regard to the intensity of staining (cytokine expression), i.e., height on the y-axis, for each cytokine.
Results
Parameters of tracking Listeria-specific T cells via intracellular cytokine staining
Intracellular cytokine staining (ICC) is a robust method for identifying Ag-specific cells and studying them on a per cell, rather than a bulk population, level. We chose to investigate PECs because the peritoneal cavity provides a model tissue site to quantify Listeria-specific lymphocytes after i.p. inoculation. C57BL/6 mice were infected i.p. with a nonlethal (2 x 104 CFU/ml) dose of Listeria and were killed 10 days p.i. at the peak of the T cell response. Cytokine production by PECs after in vitro stimulation is shown in Fig. 1. CD4+ lymphocytes from naive or infected mice did not spontaneously produce the Th1 cytokines TNF-, IFN-, or IL-2, as shown by the lack of response after incubation with only medium (no Ag). Stimulation with either a polyclonal (PMA and ionomycin) or listerial (HKLM or peptide) Ags was needed to elicit Th1 cytokines. Listeria infection increased (2- to 3-fold) the response of peritoneal T cells to polyclonal stimulation with PMA and ionomycin, possibly because more activated cells are present in infected mice. After PMA and ionomycin stimulation, the range (mean ± SD) for TNF-+ cells over 10 experiments was 36.8–94.3% (76.5 ± 16.8%), that for IFN-+ cells was 40–90.3% (75.5 ± 14.6%), and that for IL-2+ cells was 28.8–76.5% (55.0 ± 17.2%). In vivo infection was necessary to elicit a response to either HKLM or LLO 190 peptide. Cells from uninfected control mice were unresponsive to Ag simulation, because frequencies of cytokine+ cells were at background levels (0.6%). The Ag specificity of the anti-Listeria response was evidenced by the minimal response to other bacteria products. For example, when PECs from Listeria-infected mice were stimulated with either heat-killed Salmonella or lipoteichoic acid (a component of the Gram-positive cell wall), frequencies of cytokine+CD4+ lymphocytes were at background levels (data not shown).
We did not detect IL-4- or IL-10-producing cells by ICC after either polyclonal or listerial Ag stimulation above background levels on day 10 p.i. (data not shown). Although Th2-type cytokine-expressing cells have been noted (27), such observations were made with ELISPOT analysis of different tissues and at different times after infection. Collectively, our data indicate that both in vivo infection and in vitro restimulation are necessary to elicit cytokine production by CD4+ PECs, and that this response is Ag specific and Th1 biased.
Majority of cytokine-producing CD4+ peritoneal lymphocytes obtained 10 days after i.p. Listeria infection coexpress three Th1-type cytokines
We questioned whether individual cells were expressing TNF-, IFN-, and IL-2 alone or in combination by costaining for three cytokines during ICC. Mice were infected, and PECs were obtained and stimulated as described in Fig. 1 before staining cells simultaneously for TNF-, IFN-, and IL-2. Eight possible combinations, or cytokine phenotypes, were possible, as indicated in Fig. 2. The predominant cytokine phenotype after polyclonal or Ag-specific stimulation in PECs from infected mice was TNF-+IFN-+IL-2+ or triple-positive, followed by TNF-+IFN-+. Other combinations, including TNF-+IL-2+ and IFN-+IL-2+, as well as cells expressing only TNF-, only IFN-, or only IL-2 were rarely detected after Ag stimulation. Cells expressing IL-2 were usually (i.e., >97%) positive for both TNF- and IFN-, implying that their expression is a prerequisite for IL-2. The hierarchy of cytokine phenotypes was similar between PMA plus ionomycin and Ag stimulation from infected mice. The ratio of triples to doubles was consistent and stable over more than nine experiments (range, 1.12–5.21; mean ± SD, 3.02 ± 1.51). In contrast to cells from Listeria-infected mice, PECs from uninfected animals expressed a broader array of cytokine phenotypes in response to PMA plus ionomycin stimulation (Fig. 2B). Infection focused cytokine expression such that cytokines were expressed in a restricted, ordered, and defined manner. The relative frequencies of doubles and triples remained stable upon culture for 3 days in vitro (data not shown). These data suggest that simultaneous expression of three Th1-type cytokines is a characteristic of in vivo-activated, Listeria-specific T cells, and that in vivo infection is necessary to focus cytokine phenotype expression.
Cytokine-expressing CD4+ lymphocytes arise in a distinct order after infection
To better understand the acquisition of cytokine expression by CD4+ lymphocytes after infection, we examined the kinetics of the response to Listeria by PECs. We measured both overall cytokine expression (as in Fig. 1) and cytokine phenotypes (as in Fig. 2) expressed by Listeria-specific CD4+ T cells by collecting PECs at various time points after infection. Cytokine expression was a dynamic and ordered process and was determined by time after in vivo infection (Fig. 3A). Considering overall frequencies for each cytokine, TNF- producing cells as well as IFN- producing cells were readily detected by day 6 p.i. at nearly identical frequencies. In contrast, IL-2-producing cells were rare until day 10 p.i. (Fig. 3A), suggesting that IL-2 is regulated separately, but not necessarily independently, from TNF- and IFN-. To dissect the order of cytokine phenotype appearance after infection, we also examined simultaneous cytokine expression by both absolute and relative frequencies. Relative frequencies are informative when absolute frequencies of cytokine phenotypes, such as at early or later time points, are low. TNF-/IFN--coproducing cells were detected and peaked on day 6 p.i. and thereafter declined (Fig. 3B). In relation to total cytokine-expressing cells, their frequency remained stable between days 10–27 (Fig. 3C). TNF-+IFN-+IL-2+ cells were infrequent 6 days p.i., peaked on day 10, and steadily declined through the duration of the experiment into the memory phase (day 27; Fig. 3B).
Maintenance of cytokine phenotypes after secondary Listeria infection
A secondary, or recall, response to Listeria could potentially elicit different cytokine phenotypes or alter the hierarchy in responding CD4+ peritoneal lymphocytes. To test this possibility, mice were infected with Listeria as described previously, allowed to recover for at least 3 wk, and rechallenged i.p. with a bacterial dose 2 log greater than the primary dose, and T cells were collected after 6 days. The magnitude of the Ag-specific secondary response was greater than that of the primary dose (Fig. 4B), implying that a higher frequency of CD4+ T cells was recruited during a memory response. There was an increase in both TNF-+IFN-+IL-2+ and TNF-+IFN-+ PECs after PMA plus ionomycin or Ag restimulation (Fig. 4A), but the hierarchy of phenotypes remained intact. Overall, the cytokine signature appears to be a stable characteristic of Ag-specific CD4+ peritoneal lymphocytes.
Minimal in vivo conditions necessary to induce highly differentiated cytokine phenotypes
Based on the time course of the response, triple-cytokine-expressing CD4+ T cells may represent a later stage in the differentiation or localization of Ag-specific cells. Therefore, if CD4+ T cell development were a preprogrammed ordered event, then a similar frequency of triple-positive cells would be detected with a truncated infection. However, if T cell differentiation were driven by continued presence of bacteria, then cytokine phenotypes would be altered. Mice were infected i.p., and one group was given antibiotic treatment 24 h after infection to truncate Listeria infection. Treated mice were found to be free of splenic Listeria 24 h after beginning treatment and were protected from subsequent Listeria infection (28) (data not shown). On day 10 p.i., PECs were stimulated and stained as stated previously. We found no difference in the frequency of TNF-+IFN-+IL-2+ CD4+ peritoneal lymphocytes between treated and untreated mice after polyclonal stimulation (Fig. 7). Other cytokine phenotypes were detected in similar frequencies between the two groups. Control and antibiotic-treated mice also had similar frequencies of Ag-specific cells after LLO 190 stimulation. We conclude that a brief infection (24 h) is sufficient to initiate a programmed differentiation resulting in a precise cytokine signature.
Discussion
Listeria infection in mice is a common model used to study the parameters and regulation of the immune response during an infection. The dynamics of CD4+ and CD8+ T cells, organ-restricted responses, phenotypic heterogeneity, and T cell memory have all been elucidated using this model (6, 7, 27, 31). The current study contributes to the existing dataset concerning CD4+ T cells by carefully defining activation requirements for and combinatorial expression of three Th1-type cytokines by individual cells.
This novel ex vivo characterization of lymphocyte differentiation suggests a highly ordered program of cytokine appearance resulting in particular patterns of cytokine coexpression by T cells. The patterns observed for peritoneal T cells were independent of the magnitude of the response (Figs. 2 and 4) or the duration of infection (Fig. 7) as well as activation conditions in vitro (Figs. 2, 5, and 6) and could be partially recapitulated with a minimal immunization scheme (Fig. 8). After infection, the overall frequencies of TNF-- or IFN--expressing cells were similar, and the two cytokines were coexpressed after polyclonal or Ag stimulation. Their coexpression appears to be a prerequisite for IL-2 expression, because nearly 100% of peritoneal T cells expressing IL-2 after either polyclonal or Ag stimulation were triple positive (Fig. 2). In contrast, T cells from naive, uninfected mice expressed a much wider array of cytokine combinations compared with cells from infected mice (Fig. 2). This focusing was apparent in Ag-reactive T cells as well as in a large fraction of the polyclonally activated bystander T cells. Collectively, these data show that lymphocyte differentiation driven by infection results in highly focused cytokine expression patterns, with double- and triple-positive cells being the predominant CD4+ peritoneal lymphocytes. The hierarchy of cytokines expression patterns determined by differentiation in vivo remained constant even upon culture in vitro for 3 days (data not shown).
Studies of T cell functional heterogeneity have noted that highly activated cells can simultaneously express multiple cytokines. CD4+ and CD8+ T cells generated after Listeria or LCMV were double positive for TNF- and IFN- (7, 32, 33). We found that CD8+ T cells were double positive for IFN- and TNF- (data not shown), whereas activated CD4+ T cells also expressed IL-2 and were triple positive. Several previous studies have noted that IL-2-expressing cells represent maximal functional differentiation within an activation hierarchy of cytokine expression (22, 33, 34). IL-2 expression has been linked to long term interactions with dendritic cells (23) and/or degree of T cell ligation (35), ultimately providing a competitive survival advantage for memory cells (36). Additional signals by IL-7 have also been noted to be important in memory cell generation and survival (37, 38). Memory T cells are enriched in nonlymphoid tissues (39), including the peritoneal cavity, which contains the highest frequency of triple-positive cells (Fig. 5).
Anatomical heterogeneity of cytokine expression may be influenced by the site of infection, the tissue type, and the cellular migration route (Figs. 5 and 6) (27, 31, 40, 41). Our results indicate that peritoneal T cells were more likely to be triple positive than splenic T cells (Figs. 5 and 6). After i.p. inoculation, Listeria traffics rapidly to the spleen and liver, resulting in a systemic infection, but initial encounters between bacteria and macrophages may initiate a program of differentiation that would determine functional capabilities. For example, i.p. infection dramatically alters myeloid cell migration, development, and activation (42). It also results in dramatic polyclonal activation of bystander T cells (Fig. 2) and NK cells. Additionally, because the peritoneal cavity is a repository for activated or effector memory cells (39), inflammatory cytokines initiated by infection can have a strong activating effect on many cells (43, 44).
Heterogeneity may also be due to rapid reversible adaptation to the microenvironment. Although the peritoneal cavity appears to be part of the regular extralymphoid circulation route with T cells only transiently present (50% turnover/day; data not shown), the heightened frequency of triple-positive cells is striking at this site. In a recent study, T cell surface marker expression was found to change upon migration into the peritoneal cavity (41). It is possible that TNF-+IFN-+ cells leave the spleen and gain IL-2 expression in the peritoneal cavity or that triple-positive cells rapidly leave the spleen and become enriched in the peritoneal cavity. In either case, our studies clearly revealed that cytokine phenotypes were associated with the organ/tissue site. It will be interesting to determine whether cells from other tissue sites, such as lamina propria or lung, selective for memory cells, predominantly coexpress TNF-, IFN-, and IL-2. One intriguing outcome is that different tissue sites have characteristic cytokine expression signatures, reflecting specialized functions unique to these sites.
The timing of cytokine expression, particularly IL-2, after infection is noteworthy, because it may strongly influence various outcomes, such as clonal expansion vs contraction or activation of regulatory T cells. Ag-specific, double-positive T cells (TNF-+IFN-+) appear on day 6, but triple-positive cells, expressing IL-2, were not detected in the peritoneal cavity until 10 days after infection. The predominance of IL-2-expressing T cells (triple-positive cells) corresponds to the beginning of a clonal contraction or regulatory phase of the immune response. The late appearance of IL-2 expression is consistent with the role of IL-2 in sensitizing activated cells to activation-induced cell death (45) as well as in promotion of T regulatory cell function (21, 46). This putative induction of regulatory cells may provide important safeguards against autoimmune responses after the striking polyclonal bystander activation that accompanies infection.
The predominance of double- and triple-positive cells may also reflect an important functional capability and raises the question of the possible value of coordinated delivery of cytokines. In other words, what is the significance of generating one cell that makes three cytokines with divergent effects rather than three cells each making one cytokine? One possibility is that the individual cytokines may have a cooperative effect and enhance the activity of either cytokine delivered alone. This may be especially relevant in situations where cytokine-mediated intercellular communication occurs in a paracrine synaptic manner. For example, the local delivery and the resulting signaling through both the TNF-R and IFN-R pathways may be required for optimal macrophage activation and bactericidal function (16). Additionally, a balance may be obtained between opposing affects to regulate the response (47, 48). For example, IFN- promotes IL-12 expression (49), whereas TNF- inhibits IL-12 production (M. Zakharova and H. Ziegler, manuscript in preparation) (50). Balanced control may prevent immunopathological inflammation (48) and provide appropriate T cell responses and granuloma formation needed for control of infection (48). Finally, it may be simply more efficient to both generate and down-regulate one cell making three cytokines rather than three cells each making a single cytokine. In this way, a balance between efficiency and functional specialization may be maintained by particular patterns of T cell differentiation. Defining T cell differentiation by cytokine signatures should continue to provide greater understanding of this important process.
Acknowledgments
We thank Edna Scott and Walter Valesky for excellent technical assistance, and Maria Zakharova for helpful discussions.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health, National Institute of Arthritis and Infectious Disease Grant AI034065.
2 Address correspondence and reprint requests to Dr. H. Kirk Ziegler, Department of Microbiology and Immunology, Emory University School of Medicine, 1510 Clifton Road, Atlanta, GA 30322. E-mail address: ziegler@microbio.emory.edu
3 Abbreviations used in this paper: p.i., postinfection; BHI, brain-heart infusion broth; HKLM, heat-killed Listeria monocytogenes; ICC, intracellular cytokine staining; LLO, listeriolysin O; PEC, peritoneal exudate cell.
Received for publication February 14, 2005. Accepted for publication April 12, 2005.
References
Kaufmann, S. H.. 1993. Immunity to intracellular bacteria. Annu. Rev. Immunol. 11: 129-163.
Milon, G.. 1997. Listeria monocytogenes in laboratory mice: a model of short-term infectious and pathogenic processes controllable by regulated protective immune responses. Immunol. Rev. 158: 37-46.
North, R. J., P. L. Dunn, J. W. Conlan. 1997. Murine listeriosis as a model of antimicrobial defense. Immunol. Rev. 158: 27-36.
Ladel, C. H., C. Blum, S. H. Kaufmann. 1996. Control of natural killer cell-mediated innate resistance against the intracellular pathogen Listeria monocytogenes by /T lymphocytes. Infect. Immun. 64: 1744-1749.
Skeen, M. J., H. K. Ziegler. 1993. Induction of murine peritoneal /T cells and their role in resistance to bacterial infection. J. Exp. Med. 178: 971-984.
Busch, D. H., E. G. Pamer. 1999. T lymphocyte dynamics during Listeria monocytogenes infection. Immunol. Lett. 65: 93-98
Schiemann, M., V. Busch, K. Linkemann, K. M. Huster, D. H. Busch. 2003. Differences in maintenance of CD8+ and CD4+ bacteria-specific effector-memory T cell populations. Eur. J. Immunol. 33: 2875-2885.
Ladel, C. H., S. Daugelat, S. H. Kaufmann. 1995. Immune response to Mycobacterium bovis bacille Calmette Guerin infection in major histocompatibility complex class I- and II-deficient knock-out mice: contribution of CD4 and CD8 T cells to acquired resistance. Eur. J. Immunol. 25: 377-384.(Molly M. Freeman and H. K)
The robust murine response to infection with Listeria monocytogenes makes an excellent model to study the functional development of immune cells. We investigated the cellular immune response to i.p. infection using intracellular cytokine staining to identify Ag-specific lymphocytes. CD4+ peritoneal exudate cells obtained 10 days postinfection predominantly coexpressed TNF-, IFN-, and IL-2 after polyclonal or Ag stimulation. A population of cells simultaneously making TNF- and IFN- was also detected but at a lower frequency. By following the kinetics of the response to Listeria, we found that CD4+ lymphocytes coexpressing TNF- and IFN- dominated on day 6 postinfection and then declined. From days 10–27, TNF-+IFN-+IL-2+ (triple-positive) was the most prevalent cytokine phenotype, and the frequency steadily declined. These characteristic cytokine expression patterns were observed in both primary and secondary responses to Listeria infection and developed even when infection was terminated with antibiotic treatment. A cytokine-assisted immunization procedure resulted in both double- and triple-positive cells, but the clear predominance of triple-positive cells required Listeria infection. Triple-positive cells were preferentially noted in the peritoneal cavity tissue site; spleen cells displayed a predominant population of double-positive T cells (TNF-+IFN-+). We speculate that the appearance of triple-positive cells represents a functionally significant subset important in host defense at nonlymphoid tissue sites.
Introduction
Listeria monocytogenes is a Gram-positive, intracellular bacterium that causes illness in susceptible populations, such as pregnant, immunocompromised, and elderly individuals. It is also an excellent model organism used to study the complex and coordinated immune response by mice to a pathogen (1, 2). Clinically, Listeria gains access to its host orally, although other routes of infection, including i.p. and i.v., are used experimentally. Listeria infects and replicates inside macrophages and hepatocytes before spreading to other cells via actin propulsion (1). Within hours after infection, neutrophils are recruited to phagocytose and destroy infected cells; they are later replaced by macrophages (3). T cells and NK cells also have roles in the initial immune response to Listeria (4, 5). Between 7–10 days postinfection (p.i.),3 Listeria-specific effector T cells peak, eliminating remaining infected cells, and then retract, leaving long-lived memory cells (6, 7). Both CD4+ and CD8+ T cells contribute to bacterial clearance through destruction of infected cells and cytokine secretion, which directs macrophage activation and granuloma formation (8, 9, 10, 11).
The immune response to Listeria is finely coordinated through the actions of cytokines. Mice injected with neutralizing Abs to TNF- had increased susceptibility to Listeria and decreased listericidal activity by peritoneal macrophages (12). In mice treated with neutralizing IFN- Ab (13) or genetically lacking IFN- (14) or its receptor (15), typically sublethal doses of Listeria resulted in death. TNF- and IFN- produced during the initial days of Listeria infection function cooperatively to activate macrophages and organize granulomas that ultimately result in the elimination of bacterium and infected cells (16). This proinflammatory state promotes Th1 lymphocyte development, characterized by the expression of TNF-, IFN-, and/or IL-2 by CD4+ Ag-specific T cells. Functionally, these cells and their secreted products are important for efficient dendritic cell activation and subsequent development of effective memory CD8+ T cells (17, 18, 19). IL-2 plays multiple roles in the immune response, including T cell clonal expansion and contraction and T regulatory cell function (20, 21). Differential IL-2 expression by CD4+ T cells is a result of heterogeneous activation by dendritic cells (22, 23), and IL-2 expression may be a characteristic of highly differentiated T cells. Coexpression of TNF- and/or IFN- by IL-2-producing CD4+ T cells may also reflect activation heterogeneity and implies that simultaneous secretion of three cytokines with divergent actions may be a hallmark of fully differentiated lymphocytes. Cytokine coexpression patterns may be either random or deliberate. If consistent deliberate patterns are observed, the conditions that determine cytokine heterogeneity can be defined, and the potential functional significance of concomitant delivery of cytokine mixtures can be addressed.
The goal of this paper was to follow Listeria-specific CD4+ lymphocyte development based on cytokine secretion patterns. To do this, we used intracellular cytokine staining to examine the simultaneous expression of three Th1 cytokines on a single-cell level. This novel approach identified TNF-+IFN-+IL-2+ (triple-positive) CD4+ peritoneal lymphocytes that were generated after polyclonal or Ag stimulation. The data reveal defined cytokine expression patterns among Ag-specific T cells that were unique to the time after infection and the tissue site examined. Cytokine patterns were a signature of in vivo differentiation, rather than in vitro stimulation conditions. Functionally, triple-positive cells may be important in protective immunity. The novel characterization of Listeria-specific cells based on combinatorial cytokine expression is a useful approach to explore heterogeneity among responding T cell populations.
Materials and Methods
Mice and injections
C57BL/6 mice, obtained from Charles River Laboratories and The Jackson Laboratory, were housed in filter-topped microisolator cages in a specific pathogen-free facility and were used at 8–20 wk of age. The institutional animal care and use committee approved all experimental procedures. L. monocytogenes wild-type strain 43251 (American Type Culture Collection) was grown overnight in brain-heart infusion broth (BHI; Difco) at 37°C and washed twice in PBS before i.p. injection. Bacterial concentrations were determined by measuring the OD and were confirmed by plating on BHI agar plates. A sublethal dose of 2 x 104 bacteria was used for primary infections, and secondary infections were 2 log higher. For some studies, groups of mice were treated with ampicillin (2 mg/ml) in their drinking water to truncate infection. Treatment was started 24 h after Listeria injection, and fresh ampicillin-containing water was administered every 2–3 days. Spleen cells were cultured on BHI agar plates to ensure bacterial clearance 24 h after ampicillin treatment (data not shown).
For peptide immunization studies, mice were immunized i.p. with three doses of rIL-12 (0.5 μg/mouse/dose) and listeriolysin O190–201 (LLO190–201) peptide (30 μM/mouse/dose). LLO190–201 (LLO 190) is an immunodominant, MHC class II-restricted peptide (NEKYAQAYPNVS) derived from LLO (24). Successive IL-12 plus peptide injections induce protective immunity against subsequent live Listeria challenge (25).
Cell preparation and culture
Peritoneal exudate cells (PECs) were harvested by lavage with cold HBSS containing 0.06% BSA, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.1 M HEPES buffer, 0.04% sodium bicarbonate, and 10 U/ml heparin. PECs from each group of mice (two to eight mice per group, as indicated in the figure legends for each experiment) were pooled. Single-cell suspensions were made from excised and pooled spleens using a glass homogenizer, followed by lysis of erythrocytes with ammonium chloride buffer. Cells from both sites were washed and resuspended for culture in RPMI 1640 supplemented with 10% FCS, 5 x 10–5 M 2-ME, 0.5 mM sodium pyruvate, 10 mM HEPES buffer, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine. For some experiments, T cells from PECs or spleen suspensions were negatively selected using T cell enrichment columns according to the manufacturer’s instructions (R&D Systems). During culture, enriched T cells were treated identically to their unenriched counterparts. Effective depletion of APCs was evidenced by the 90% reduction in cytokine response upon stimulation of enriched T cells with peptide Ag (data not shown).
Abs for flow cytometry
The following reagents were purchased from BD Pharmingen: RM4-5-PerCP (anti-CD4), MP6-XT22-FITC (anti-TNF-), XMG1.2-PE (anti-IFN-), and JES6-5H4-allophycocyanin (anti-IL-2). Isotype controls for these Abs were purchased as follows: rat IgG-FITC from Southern Biotechnology Associates and rat IgG1-PE, rat IgG2a-PerCP, and rat IgG2b-allophycocyanin from BD Pharmingen.
Detection of intracellular cytokines in individual lymphocyte populations by flow cytometry
Various stimuli were used during a 5-h culture period to induce cytokine production. Polyclonal, or nonspecific, stimulation was provided by either PMA (10 ng/ml) and ionomycin (1 μM; Calbiochem) or plate-bound anti-CD3 (145-2C11; 5 μg/ml) and soluble anti-CD28 (37.51; 10 μg/ml) purchased from BD Pharmingen. For Ag-specific stimulation, cells were cultured with either heat-killed Listeria (HKLM; 107/ml) or LLO 190 (30 μg/ml). Brefeldin A (10 μg/ml; Sigma-Aldrich) was added for the last 4 h to inhibit cytokine secretion and enhance intracellular detection (26). Dose-response curves and stimulation time courses were conducted to optimize the detection of cytokine-expressing CD4+ lymphocytes (data not shown).
For analysis by flow cytometry, 1–2 x 106 cells were incubated for 30 min at 4°C with PerCP-conjugated mAb to cell surface markers to identify individual lymphocyte populations, then were washed twice with FACS wash buffer (PBS, 3% FCS, and 0.1% sodium azide). Cells were incubated for 15 min at room temperature with 50 μl of fixation medium (Fix and Perm kit; Caltag Laboratories) and washed once. Predetermined optimal concentrations of fluorochrome-conjugated anti-cytokine Abs diluted in permeabilization buffer (Fix and Perm kit) were added for 15 min at room temperature, followed by two washes and resuspension in FACS wash buffer. Background fluorescence was <0.5% after incubation with Ig isotype controls conjugated to fluorochromes (data not shown). Cells were analyzed on a FACSCalibur flow cytometer (BD Biosciences) by gating on the lymphocyte population, as defined by forward scatter/side scatter parameters.
A logical gating scheme was used to identify CD4+ lymphocytes that were expressing a cytokine or combinations of cytokines. The data from each sample were initially plotted on a forward scatter by cytokine (TNF-, IFN-, or IL-2) dot plot. A region (R) was drawn around those cells that were positive for the desired cytokine, i.e., R1 for TNF-+ cells, R2 for IFN-+ cells, and R3 for IL-2+ cells. Under the gating menu in CellQuest (BD Biosciences), eight logical gates were assigned as follows: TNF-+, R1 (not R2 or R3); IFN-+, R2 (not R1 or R3); IL-2+, R3 (not R1 or R3); TNF-+IFN-+, R1*R2 (not R3); TNF-+IL-2+, R1*R3 (not R2); IFN-+IL-2+, R2*R3 (not R1); TNF-+IFN-+IL-2+, R1*R2*R3; and cytokine negative, not R1, R2, or R3. Each possible combination was arbitrarily assigned a color, as shown in Fig. 2. Additionally, only those cells that fell in a CD4+ gate (R4) as well as a lymphocyte gate (R5; as determined by characteristic forward scatter by side scatter appearance) are displayed in Fig. 2. As a consequence of this gating scheme, the same data are shown repeatedly, but with respect to three different parameters, i.e., the same TNF-+ cells are shown in all three plots. Examining all three plots is informative with regard to the intensity of staining (cytokine expression), i.e., height on the y-axis, for each cytokine.
Results
Parameters of tracking Listeria-specific T cells via intracellular cytokine staining
Intracellular cytokine staining (ICC) is a robust method for identifying Ag-specific cells and studying them on a per cell, rather than a bulk population, level. We chose to investigate PECs because the peritoneal cavity provides a model tissue site to quantify Listeria-specific lymphocytes after i.p. inoculation. C57BL/6 mice were infected i.p. with a nonlethal (2 x 104 CFU/ml) dose of Listeria and were killed 10 days p.i. at the peak of the T cell response. Cytokine production by PECs after in vitro stimulation is shown in Fig. 1. CD4+ lymphocytes from naive or infected mice did not spontaneously produce the Th1 cytokines TNF-, IFN-, or IL-2, as shown by the lack of response after incubation with only medium (no Ag). Stimulation with either a polyclonal (PMA and ionomycin) or listerial (HKLM or peptide) Ags was needed to elicit Th1 cytokines. Listeria infection increased (2- to 3-fold) the response of peritoneal T cells to polyclonal stimulation with PMA and ionomycin, possibly because more activated cells are present in infected mice. After PMA and ionomycin stimulation, the range (mean ± SD) for TNF-+ cells over 10 experiments was 36.8–94.3% (76.5 ± 16.8%), that for IFN-+ cells was 40–90.3% (75.5 ± 14.6%), and that for IL-2+ cells was 28.8–76.5% (55.0 ± 17.2%). In vivo infection was necessary to elicit a response to either HKLM or LLO 190 peptide. Cells from uninfected control mice were unresponsive to Ag simulation, because frequencies of cytokine+ cells were at background levels (0.6%). The Ag specificity of the anti-Listeria response was evidenced by the minimal response to other bacteria products. For example, when PECs from Listeria-infected mice were stimulated with either heat-killed Salmonella or lipoteichoic acid (a component of the Gram-positive cell wall), frequencies of cytokine+CD4+ lymphocytes were at background levels (data not shown).
We did not detect IL-4- or IL-10-producing cells by ICC after either polyclonal or listerial Ag stimulation above background levels on day 10 p.i. (data not shown). Although Th2-type cytokine-expressing cells have been noted (27), such observations were made with ELISPOT analysis of different tissues and at different times after infection. Collectively, our data indicate that both in vivo infection and in vitro restimulation are necessary to elicit cytokine production by CD4+ PECs, and that this response is Ag specific and Th1 biased.
Majority of cytokine-producing CD4+ peritoneal lymphocytes obtained 10 days after i.p. Listeria infection coexpress three Th1-type cytokines
We questioned whether individual cells were expressing TNF-, IFN-, and IL-2 alone or in combination by costaining for three cytokines during ICC. Mice were infected, and PECs were obtained and stimulated as described in Fig. 1 before staining cells simultaneously for TNF-, IFN-, and IL-2. Eight possible combinations, or cytokine phenotypes, were possible, as indicated in Fig. 2. The predominant cytokine phenotype after polyclonal or Ag-specific stimulation in PECs from infected mice was TNF-+IFN-+IL-2+ or triple-positive, followed by TNF-+IFN-+. Other combinations, including TNF-+IL-2+ and IFN-+IL-2+, as well as cells expressing only TNF-, only IFN-, or only IL-2 were rarely detected after Ag stimulation. Cells expressing IL-2 were usually (i.e., >97%) positive for both TNF- and IFN-, implying that their expression is a prerequisite for IL-2. The hierarchy of cytokine phenotypes was similar between PMA plus ionomycin and Ag stimulation from infected mice. The ratio of triples to doubles was consistent and stable over more than nine experiments (range, 1.12–5.21; mean ± SD, 3.02 ± 1.51). In contrast to cells from Listeria-infected mice, PECs from uninfected animals expressed a broader array of cytokine phenotypes in response to PMA plus ionomycin stimulation (Fig. 2B). Infection focused cytokine expression such that cytokines were expressed in a restricted, ordered, and defined manner. The relative frequencies of doubles and triples remained stable upon culture for 3 days in vitro (data not shown). These data suggest that simultaneous expression of three Th1-type cytokines is a characteristic of in vivo-activated, Listeria-specific T cells, and that in vivo infection is necessary to focus cytokine phenotype expression.
Cytokine-expressing CD4+ lymphocytes arise in a distinct order after infection
To better understand the acquisition of cytokine expression by CD4+ lymphocytes after infection, we examined the kinetics of the response to Listeria by PECs. We measured both overall cytokine expression (as in Fig. 1) and cytokine phenotypes (as in Fig. 2) expressed by Listeria-specific CD4+ T cells by collecting PECs at various time points after infection. Cytokine expression was a dynamic and ordered process and was determined by time after in vivo infection (Fig. 3A). Considering overall frequencies for each cytokine, TNF- producing cells as well as IFN- producing cells were readily detected by day 6 p.i. at nearly identical frequencies. In contrast, IL-2-producing cells were rare until day 10 p.i. (Fig. 3A), suggesting that IL-2 is regulated separately, but not necessarily independently, from TNF- and IFN-. To dissect the order of cytokine phenotype appearance after infection, we also examined simultaneous cytokine expression by both absolute and relative frequencies. Relative frequencies are informative when absolute frequencies of cytokine phenotypes, such as at early or later time points, are low. TNF-/IFN--coproducing cells were detected and peaked on day 6 p.i. and thereafter declined (Fig. 3B). In relation to total cytokine-expressing cells, their frequency remained stable between days 10–27 (Fig. 3C). TNF-+IFN-+IL-2+ cells were infrequent 6 days p.i., peaked on day 10, and steadily declined through the duration of the experiment into the memory phase (day 27; Fig. 3B).
Maintenance of cytokine phenotypes after secondary Listeria infection
A secondary, or recall, response to Listeria could potentially elicit different cytokine phenotypes or alter the hierarchy in responding CD4+ peritoneal lymphocytes. To test this possibility, mice were infected with Listeria as described previously, allowed to recover for at least 3 wk, and rechallenged i.p. with a bacterial dose 2 log greater than the primary dose, and T cells were collected after 6 days. The magnitude of the Ag-specific secondary response was greater than that of the primary dose (Fig. 4B), implying that a higher frequency of CD4+ T cells was recruited during a memory response. There was an increase in both TNF-+IFN-+IL-2+ and TNF-+IFN-+ PECs after PMA plus ionomycin or Ag restimulation (Fig. 4A), but the hierarchy of phenotypes remained intact. Overall, the cytokine signature appears to be a stable characteristic of Ag-specific CD4+ peritoneal lymphocytes.
Minimal in vivo conditions necessary to induce highly differentiated cytokine phenotypes
Based on the time course of the response, triple-cytokine-expressing CD4+ T cells may represent a later stage in the differentiation or localization of Ag-specific cells. Therefore, if CD4+ T cell development were a preprogrammed ordered event, then a similar frequency of triple-positive cells would be detected with a truncated infection. However, if T cell differentiation were driven by continued presence of bacteria, then cytokine phenotypes would be altered. Mice were infected i.p., and one group was given antibiotic treatment 24 h after infection to truncate Listeria infection. Treated mice were found to be free of splenic Listeria 24 h after beginning treatment and were protected from subsequent Listeria infection (28) (data not shown). On day 10 p.i., PECs were stimulated and stained as stated previously. We found no difference in the frequency of TNF-+IFN-+IL-2+ CD4+ peritoneal lymphocytes between treated and untreated mice after polyclonal stimulation (Fig. 7). Other cytokine phenotypes were detected in similar frequencies between the two groups. Control and antibiotic-treated mice also had similar frequencies of Ag-specific cells after LLO 190 stimulation. We conclude that a brief infection (24 h) is sufficient to initiate a programmed differentiation resulting in a precise cytokine signature.
Discussion
Listeria infection in mice is a common model used to study the parameters and regulation of the immune response during an infection. The dynamics of CD4+ and CD8+ T cells, organ-restricted responses, phenotypic heterogeneity, and T cell memory have all been elucidated using this model (6, 7, 27, 31). The current study contributes to the existing dataset concerning CD4+ T cells by carefully defining activation requirements for and combinatorial expression of three Th1-type cytokines by individual cells.
This novel ex vivo characterization of lymphocyte differentiation suggests a highly ordered program of cytokine appearance resulting in particular patterns of cytokine coexpression by T cells. The patterns observed for peritoneal T cells were independent of the magnitude of the response (Figs. 2 and 4) or the duration of infection (Fig. 7) as well as activation conditions in vitro (Figs. 2, 5, and 6) and could be partially recapitulated with a minimal immunization scheme (Fig. 8). After infection, the overall frequencies of TNF-- or IFN--expressing cells were similar, and the two cytokines were coexpressed after polyclonal or Ag stimulation. Their coexpression appears to be a prerequisite for IL-2 expression, because nearly 100% of peritoneal T cells expressing IL-2 after either polyclonal or Ag stimulation were triple positive (Fig. 2). In contrast, T cells from naive, uninfected mice expressed a much wider array of cytokine combinations compared with cells from infected mice (Fig. 2). This focusing was apparent in Ag-reactive T cells as well as in a large fraction of the polyclonally activated bystander T cells. Collectively, these data show that lymphocyte differentiation driven by infection results in highly focused cytokine expression patterns, with double- and triple-positive cells being the predominant CD4+ peritoneal lymphocytes. The hierarchy of cytokines expression patterns determined by differentiation in vivo remained constant even upon culture in vitro for 3 days (data not shown).
Studies of T cell functional heterogeneity have noted that highly activated cells can simultaneously express multiple cytokines. CD4+ and CD8+ T cells generated after Listeria or LCMV were double positive for TNF- and IFN- (7, 32, 33). We found that CD8+ T cells were double positive for IFN- and TNF- (data not shown), whereas activated CD4+ T cells also expressed IL-2 and were triple positive. Several previous studies have noted that IL-2-expressing cells represent maximal functional differentiation within an activation hierarchy of cytokine expression (22, 33, 34). IL-2 expression has been linked to long term interactions with dendritic cells (23) and/or degree of T cell ligation (35), ultimately providing a competitive survival advantage for memory cells (36). Additional signals by IL-7 have also been noted to be important in memory cell generation and survival (37, 38). Memory T cells are enriched in nonlymphoid tissues (39), including the peritoneal cavity, which contains the highest frequency of triple-positive cells (Fig. 5).
Anatomical heterogeneity of cytokine expression may be influenced by the site of infection, the tissue type, and the cellular migration route (Figs. 5 and 6) (27, 31, 40, 41). Our results indicate that peritoneal T cells were more likely to be triple positive than splenic T cells (Figs. 5 and 6). After i.p. inoculation, Listeria traffics rapidly to the spleen and liver, resulting in a systemic infection, but initial encounters between bacteria and macrophages may initiate a program of differentiation that would determine functional capabilities. For example, i.p. infection dramatically alters myeloid cell migration, development, and activation (42). It also results in dramatic polyclonal activation of bystander T cells (Fig. 2) and NK cells. Additionally, because the peritoneal cavity is a repository for activated or effector memory cells (39), inflammatory cytokines initiated by infection can have a strong activating effect on many cells (43, 44).
Heterogeneity may also be due to rapid reversible adaptation to the microenvironment. Although the peritoneal cavity appears to be part of the regular extralymphoid circulation route with T cells only transiently present (50% turnover/day; data not shown), the heightened frequency of triple-positive cells is striking at this site. In a recent study, T cell surface marker expression was found to change upon migration into the peritoneal cavity (41). It is possible that TNF-+IFN-+ cells leave the spleen and gain IL-2 expression in the peritoneal cavity or that triple-positive cells rapidly leave the spleen and become enriched in the peritoneal cavity. In either case, our studies clearly revealed that cytokine phenotypes were associated with the organ/tissue site. It will be interesting to determine whether cells from other tissue sites, such as lamina propria or lung, selective for memory cells, predominantly coexpress TNF-, IFN-, and IL-2. One intriguing outcome is that different tissue sites have characteristic cytokine expression signatures, reflecting specialized functions unique to these sites.
The timing of cytokine expression, particularly IL-2, after infection is noteworthy, because it may strongly influence various outcomes, such as clonal expansion vs contraction or activation of regulatory T cells. Ag-specific, double-positive T cells (TNF-+IFN-+) appear on day 6, but triple-positive cells, expressing IL-2, were not detected in the peritoneal cavity until 10 days after infection. The predominance of IL-2-expressing T cells (triple-positive cells) corresponds to the beginning of a clonal contraction or regulatory phase of the immune response. The late appearance of IL-2 expression is consistent with the role of IL-2 in sensitizing activated cells to activation-induced cell death (45) as well as in promotion of T regulatory cell function (21, 46). This putative induction of regulatory cells may provide important safeguards against autoimmune responses after the striking polyclonal bystander activation that accompanies infection.
The predominance of double- and triple-positive cells may also reflect an important functional capability and raises the question of the possible value of coordinated delivery of cytokines. In other words, what is the significance of generating one cell that makes three cytokines with divergent effects rather than three cells each making one cytokine? One possibility is that the individual cytokines may have a cooperative effect and enhance the activity of either cytokine delivered alone. This may be especially relevant in situations where cytokine-mediated intercellular communication occurs in a paracrine synaptic manner. For example, the local delivery and the resulting signaling through both the TNF-R and IFN-R pathways may be required for optimal macrophage activation and bactericidal function (16). Additionally, a balance may be obtained between opposing affects to regulate the response (47, 48). For example, IFN- promotes IL-12 expression (49), whereas TNF- inhibits IL-12 production (M. Zakharova and H. Ziegler, manuscript in preparation) (50). Balanced control may prevent immunopathological inflammation (48) and provide appropriate T cell responses and granuloma formation needed for control of infection (48). Finally, it may be simply more efficient to both generate and down-regulate one cell making three cytokines rather than three cells each making a single cytokine. In this way, a balance between efficiency and functional specialization may be maintained by particular patterns of T cell differentiation. Defining T cell differentiation by cytokine signatures should continue to provide greater understanding of this important process.
Acknowledgments
We thank Edna Scott and Walter Valesky for excellent technical assistance, and Maria Zakharova for helpful discussions.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health, National Institute of Arthritis and Infectious Disease Grant AI034065.
2 Address correspondence and reprint requests to Dr. H. Kirk Ziegler, Department of Microbiology and Immunology, Emory University School of Medicine, 1510 Clifton Road, Atlanta, GA 30322. E-mail address: ziegler@microbio.emory.edu
3 Abbreviations used in this paper: p.i., postinfection; BHI, brain-heart infusion broth; HKLM, heat-killed Listeria monocytogenes; ICC, intracellular cytokine staining; LLO, listeriolysin O; PEC, peritoneal exudate cell.
Received for publication February 14, 2005. Accepted for publication April 12, 2005.
References
Kaufmann, S. H.. 1993. Immunity to intracellular bacteria. Annu. Rev. Immunol. 11: 129-163.
Milon, G.. 1997. Listeria monocytogenes in laboratory mice: a model of short-term infectious and pathogenic processes controllable by regulated protective immune responses. Immunol. Rev. 158: 37-46.
North, R. J., P. L. Dunn, J. W. Conlan. 1997. Murine listeriosis as a model of antimicrobial defense. Immunol. Rev. 158: 27-36.
Ladel, C. H., C. Blum, S. H. Kaufmann. 1996. Control of natural killer cell-mediated innate resistance against the intracellular pathogen Listeria monocytogenes by /T lymphocytes. Infect. Immun. 64: 1744-1749.
Skeen, M. J., H. K. Ziegler. 1993. Induction of murine peritoneal /T cells and their role in resistance to bacterial infection. J. Exp. Med. 178: 971-984.
Busch, D. H., E. G. Pamer. 1999. T lymphocyte dynamics during Listeria monocytogenes infection. Immunol. Lett. 65: 93-98
Schiemann, M., V. Busch, K. Linkemann, K. M. Huster, D. H. Busch. 2003. Differences in maintenance of CD8+ and CD4+ bacteria-specific effector-memory T cell populations. Eur. J. Immunol. 33: 2875-2885.
Ladel, C. H., S. Daugelat, S. H. Kaufmann. 1995. Immune response to Mycobacterium bovis bacille Calmette Guerin infection in major histocompatibility complex class I- and II-deficient knock-out mice: contribution of CD4 and CD8 T cells to acquired resistance. Eur. J. Immunol. 25: 377-384.(Molly M. Freeman and H. K)