- Email: [email protected]
Splenic CD4+and CD8+T Cells from Influenza Immune Mice Concurrently Producein VitroIL2, IL4, and IFN-γ
CELLULAR IMMUNOLOGY ARTICLE NO.
170, 222–229 (1996)
0155
Splenic CD4/ and CD8/ T Cells from Influenza Immune Mice Concurrently Produce in Vitro IL2, IL4, and IFN-g ROBERTO FALCHETTI,*,1 GIULIA LANZILLI,* IDA A. CASALINUOVO, ROBERTA GAZIANO, ANNA TERESA PALAMARA, PAOLO DI FRANCESCO, GIAMPIETRO RAVAGNAN,* AND ENRICO GARACI Department of Experimental Medicine and Biochemical Sciences, University of Rome, Tor Vergata, Rome, Italy; and *Institute of Experimental Medicine, CNR, Rome, Italy Received September 22, 1995; accepted March 1, 1996
The cytokine responses exerted by virus-primed spleen T cells upon in vitro restimulation were studied. Spleen cells obtained from mice injected intraperitoneally with A/PR8 (H1N1) influenza virus (PR8) were restimulated in vitro with UV-inactivated PR8 virus. The percentage of both CD4/ and CD8/ T cells producing IL2, IL4, or IFN-g was assayed at the single cell level by flow cytometric analysis of intracytoplasmic cytokine content. In parallel, the levels of the different cytokines in spleen cell culture supernatants were quantitated by enzyme linked immunosorbent assay. The results showed that in vitro virus restimulation of immune spleen cells induced the concurrent increase, in both CD4/ and CD8/ T cells, of the frequency of IL2-, IFN-g-, and IL4-producing cells. The frequency of IFN-g-producing T cells was found to be significantly higher in CD8/ T cells. Significant levels of the three cytokines were also detected in the culture supernatants. These data suggest that both CD4/ and CD8/ T cells play an important role in cytokine response to virus infection and that the synthesis and secretion of antiviral and regulatory cytokines is not mutually exclusive either between or within the two T cell subsets. The results of the experiments also indicated that the virus restimulation did not induce a dominant type 1 or type 2 cytokine response. q 1996 Academic Press, Inc.
INTRODUCTION Cytokine production by T cells plays an important role in the generation of immune response to infection. Exposure to different nonviral pathogens can stimulate at least two patterns of cytokine production by CD4positive (CD4/) T-helper cells (1, 2). Responses that are characterized by the predominant secretion of interleukin (IL) 2, interferon-g (IFN-g), and lymphotoxins are 1 To whom correspondence should be addressed at Institute of Experimental Medicine, CNR, Viale C. Marx, 43, 00137 Rome, Italy.
222
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
CI 8145
/
6c10$$$$$1
classified as T-helper-1 (Th1)-type responses and generally occur during infections with intracellular bacteria and parasites. In contrast, responses following nematode infections and characterized by a predominant production of IL4, IL5, IL9, IL10, and IL13 are classified as T-helper-2 (Th2)-type responses. It has recently been suggested (3) that the two types of cytokine production be termed type 1 and type 2, respectively, since some of the immunoregulatory cytokines are also produced during infection either by other populations of T cells or by non-T cells. Less precise information has been gathered on cytokine responses exerted by T cells to viral infections. Studies (4, 5) performed using different models of viral infection showed that both type 1 and type 2 cytokines can be expressed by T cells, although locally a preferential production of a certain type of cytokine can be found. Recent results (6, 7) obtained on influenza virusinfected mice showed, for instance, differences in the pattern of cytokine production by T cells from bronchoalveolar lavage (BAL), lung tissue, and mediastinal lymph node (MLN). An interesting aspect concerning the study of T cell cytokine response to viral infection is the determination of the cytokine-producing cell immunophenotypes, which is the determination of the pattern of cytokine production not only by CD4/ but also by CD8/ T cells, as this latter subset of T cells plays a central role in antiviral immunity (8, 9). The early events of the response to viral infections are, in fact, associated with the activation of specific class I-restricted cytotoxic T lymphocyte (CTL) responses and CD8/ T-cell expansion (10). At this stage of immune response different cytokines, i.e., IL2, IL4, IFN-g, and IL10, possibly promoting activation and regulation of virus-specific CTL responses, are expressed in lymphoid tissues. Recent evidence suggests that alloantigen-stimulated CD8/ mouse spleen cells can secrete these cytokines (11) and that CD8/ T cells secreting IL2, IL5, IL10, and IFNg are present in MLN, BAL, or lung tissue of influenza virus-infected mice (6, 7). It is therefore possible that CD8/ T cells are the source of the cytokines required for
05-17-96 06:33:19
cia
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
the early phase of immune response, which thus would be independent from CD4/ T cell cytokine release. On the other hand, studies on the role of class II-restricted CD4/ T cells indicated that virus-primed CD8/ cells did not give a significant CTL response in the absence of CD4/ T cells or of their produced factors, thus suggesting that the antiviral CTL response is dependent upon Thelper cells (12). The pattern of virus-induced CD4/ T cell cytokine production was examined by different laboratories. In particular, experiments using ELISPOT assay showed (6) that both freshly isolated and virusrestimulated CD4/ T cells obtained from MLN of influenza virus-infected mice produced IL2, IL10, and IFNg. Interestingly, these cells, differently from BAL cells obtained from animals similarly treated (13), did not produce significant amounts of IL4. The greatest part of the data on both virus-induced CD4/ and CD8/ T cell cytokine production was generated by restimulating virus-specific T cell clones or separated T cell subsets obtained by fluorescence-activated sorting or selective killing. In consideration of the fact that some type 1 and type 2 cytokines can exert cross-regulatory (and in particular cross-inhibitory) activities (14, 15), the pattern of the virus-primed CD4/ and CD8/ T cell cytokine production should preferably be evaluated in an experimental situation in which both populations are contemporaneously present. To this aim we studied, at the single cell level, virusinduced cytokine production by CD4/ and CD8/ T cells under experimental conditions ensuring the contemporaneous presence of the different immunocompetent cell populations at the moment of restimulation with the virus of the immune cells. A model system was established in which unseparated spleen cells obtained from mice injected intraperitoneally (ip) with PR8 influenza virus were restimulated in vitro with UV-inactivated PR8 virus. The percentage of both CD4/ and CD8/ T cells producing the single cytokines (e.g., IL2, IL4, and IFN-g) was evaluated by simultaneous flow cytometric analysis of cellular intracytoplasmic cytokine content and surface markers. By this method both the frequency of cytokine-producing cells within the two subsets and the type 1–type 2 profile were analyzed. In parallel, the levels of the different cytokines in the culture supernatants were quantitated by ELISA. The results showed that (i) upon in vitro restimulation, virus-primed CD4/ and CD8/ T cells concurrently produced the same cytokines, i.e., IL2, IL4, and IFNg; (ii) the virus did not induce a dominant type 1 or type 2 response; and (iii) no evident cross-inhibitory activity on cytokine production was exerted by the two T cell subsets.
223
Mice were negative for antibodies to mycoplasma, murine hepatitis virus, and Sendai virus by ELISA. Virus A/PR8 (H1N1) influenza virus (PR8) was grown in the allantoid cavities of 10-day-old embryonated chicken eggs. Virus titer was expressed in hemoagglutinating units (HAU)/ml. UV-inactivated PR8 was prepared by exposing virus suspensions to UV light (40 W, 254 nm, 8-cm distance) for 5 min. Immunization Mice were injected ip with 75 HAU/animal of PR8 daily for 3 consecutive days. Cell Cultures Mice were sacrificed 2 days after the last virus injection (fifth day of immunization) by cervical dislocation. Spleen cell suspensions from immunized or naive animals (injected ip with virus-free appropriately diluted allantoic fluid, following immunization timing) were cultured in 24-well plates (Becton–Dickinson, Lincoln Park, NJ) at 1 1 107/ml (2 ml/well) in complete culture medium (CCM), consisting of RPMI 1640 supplemented with 10% fetal calf serum (FCS, Flow Laboratories, Rickmansworth, UK), 1% L-glutamine (Flow Laboratories), and 1% penicillin–streptomycin (Flow Laboratories). Spleen cells were cultured alone or with UV-inactivated PR8 at the concentration of 250 HAU/ ml. Cell cultures were incubated at 377C in humidified 5% CO2 . At the indicated time points culture supernatants or cells were removed and used for successive evaluations. Cytokine Determination Concentrations of IFN-g, IL2, and IL4 in the culture supernatants were determined by two-site sandwich ELISA according to the method previously described (16) using rat anti-mouse IFN-g mAb (PharMingen, San Diego, CA; clone XMG1.2), rat anti-mouse IL-2 mAb (PharMingen; clone JES6-1A12), and rat antimouse IL4 mAb (PharMingen; clone BVD4-1D11) as coat and biotinylated rat anti-mouse IFN-g mAb (PharMingen; clone XMG1.2), biotinylated rat anti-mouse IL2 mAb (PharMingen; clone JES6-5H4), and biotinylated rat anti-mouse IL4 mAb (PharMingen; clone BVD6-24G2) as detecting antibodies. Standards of rIFN-g, rIL2, and rIL4 purchased from PharMingen were used as reference.
MATERIALS AND METHODS
Detection of Cytokine-Producing Cells by Intracellular Immunofluorescence Staining
Animals Male BALB/c mice (5–7 weeks old) purchased from Charles River Italia (Calco, Como, Italy) were used.
To determine the frequency of cytokine-producing CD4/ and CD8/ T cells, the method described by Sander et al. (17) with minor modifications (16) was
AID
cia
CI 8145
/
6c10$$$$$2
05-17-96 06:33:19
AP: Cell Immuno
224
FALCHETTI ET AL.
used. Briefly, after culture, 106 cells fixed in 4% paraformaldehyde were stained by incubating them at room temperature (RT) with rat anti-mouse cytokine mAb in PBS/BSA/NaN3 containing 0.3% saponin (Sigma, St. Louis, MO). The following rat anti-mouse cytokine mAb were used: anti-mouse IL-2 mAb (PharMingen; clone JES6-1A12, staining concentration 20 mg/ml), antimouse IL-4 mAb (PharMingen; clone BVD4-1D11, staining concentration 20 mg/ml), and anti-mouse IFNg mAb (PharMingen; clone XMG1.2, staining concentration 20 mg/ml). As negative controls some aliquots of cell suspensions were incubated with an irrelevant isotype-matched mAb. As second-step Ab, a fluorescein (FITC)-conjugated immunopure goat anti-rat IgG (H / L) (Pierce, Rockford, IL) was used. To determine the immunophenotype of the cytokine-producing cells, samples were subsequently surface stained with rat anti-mouse CD4 mAb (PharMingen; clone RM4-5, staining concentration 1 mg/million cells) and rat antimouse CD8 mAb (PharMingen; clone 53-6.7, staining concentration 1 mg/million cells) coupled to phycoerythrin (PE). A FACScan flow cytometer (Becton–Dickinson, San Jose, CA) was used to analyze each sample. Signals from light scatter channels were collected in linear mode, and signals from the fluorescence-detecting PMTs were collected as logarithmic signals. Ten thousand events were analyzed for each test and signals from all parameters were captured as listmode data and analyzed by FACScan research software. Statistical Analysis Data are expressed as means { SEM. Statistical significance was determined using Student’s t test. RESULTS PR8 Immune Spleen Cells Secrete both Type 1 (IL2 and IFN-g) and Type 2 (IL4) Cytokines upon in Vitro Restimulation with the Virus In a first series of experiments, to control both the pattern and the peak time of in vitro virus-induced cytokine secretion, spleen cells obtained from mice injected ip with PR8 virus were cultured for 24 or 48 hr in the presence of UV-inactivated PR8. Cultures of both naive spleen cells stimulated with PR8 or of immune spleen cells lacking the virus were also established. At the end of incubation time the supernatants were removed and the levels of IL2, IL4, and IFN-g were measured by ELISA. Figure 1 shows that after 24 hr of incubation, the levels in PR8-stimulated immune spleen cell supernatants of the three tested cytokines were significantly higher than those of unstimulated cells. At 48 hr (Fig. 1) IL2 levels markedly decreased, whereas the levels of both IL4 and IFN-g remained unchanged. No significant production of cytokines was found in cultures of naive spleen cells stimulated with PR8 (data not shown). These results
AID
CI 8145
/
6c10$$$$$2
05-17-96 06:33:19
FIG. 1. In vitro production of cytokines by influenza virus-primed spleen cells. The levels of IL2 (A), IFN-g (B), and IL4 (C) were determined by ELISA in culture supernatants of PR8-immunized animalderived spleen cells cultured for 24 and 48 hr in the presence or absence of UV-inactivated PR8. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. *P õ 0.05; **P õ 0.01; ***P õ 0.001 (Student’s t test).
showed that in vitro virus-restimulated immune spleen cells were able to produce significant amounts of both type 1 and type 2 cytokines and that the peak level was already reached at 24 hr. Both CD4/ and CD8/ PR8 Immune T Cells Produce IL2, IL4, and IFN-g upon in Vitro Restimulation with Virus The production of cytokines by in vitro virus-restimulated immune CD4/ and CD8/ T cells was analyzed at the single cell level. Through the use of flow cytome-
cia
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
225
FIG. 2. Flow cytometric analysis of the percentage of IL2-, IFN-g-, and IL4-producing CD4/ splenocytes. The frequency of cytokineproducing cells was evaluated on PR8-immunized animal-derived spleen cells both unstimulated (A) or restimulated in vitro with PR8 (B). Samples were analyzed by first gating on the CD4-positive population (Y axis). The threshold for detection of positive IL2-, IFN-g-, and IL4-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with an isotype-specific antibody (C). The percentage of specific IL2-, IFN-g-, and IL4-positive lymphocytes (A, B) was calculated by subtracting the isotype control (1%) from the specific anti-IL2, IFN-g, and IL4 signal. X axis, FL1, log10 IL2, IFN-g, or IL4 signal; Y axis, FL2, log10 CD4 signal.
try, the frequency of the CD4/ or CD8/ T cells containing intracytoplasmic IL2, IL4, or IFN-g was evaluated. Spleen cells obtained from PR8-immunized animals were cultured for 24 hr in the presence or absence of the virus. As control, spleen cells obtained from naive animals were also incubated with or without PR8. At the end of incubation time the frequencies of both CD4/ and CD8/ T cells expressing IL2, IL4, or IFN-g were determined by immunofluorescence, staining fixed and saponin-permeabilized cells for either CD4 or CD8 surface molecule, and intracytoplasmic IL2, IL4, or IFN-g. As paraformaldehyde fixation increased autofluorescence in nonlymphocyte populations, analysis of the samples was performed by first
gating on CD4/ or CD8/ T cells. Previous experiments (16) showed that no significant cytokine immunofluorescence is present in fixed but not saponin-permeabilized cells, thus excluding the possibility that in our system significant fluorescent signals could be due to receptor-bound FITC-labeled cytokines on the surface of target cells. Irrelevant isotype-specific antibody was used as a control for nonspecific binding of the anticytokine monoclonal antibodies. As described in a previous paper (16), the threshold for detection of positive cytokine-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with isotype-specific antibodies (Figs. 2C and 3C). The percentage of specific cytokine-positive cells was cal-
AID
cia
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
226
FALCHETTI ET AL.
FIG. 3. Flow cytometric analysis of the percentage of IL2-, IFN-g-, and IL4-producing CD8/ splenocytes. The frequency of cytokineproducing cells was evaluated on PR8-immunized animal-derived spleen cells both unstimulated (A) or restimulated in vitro with PR8 (B). Samples were analyzed by first gating on the CD8-positive population (Y axis). The threshold for detection of positive IL2-, IFN-g-, and IL4-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with an isotype-specific antibody (C). The percentage of specific IL2-, IFN-g-, and IL4-positive lymphocytes (A, B) was calculated by subtracting the isotype control (1%) from the specific anti-IL2, IFN-g, and IL4 signal. X axis, FL1, log10 IL2, IFN-g, or IL4 signal; Y axis, FL2, log10 CD8 signal.
culated by subtracting the isotype control from the specific anti-cytokine signal. Results of the analysis of cytokine production at the single cell level showed that at this cutoff level, only a slight and not significant percentage of cytokine-positive cells was observed in both unstimulated and PR8-stimulated naive spleen cells as well as in unstimulated immune spleen cells, either in the CD4/ (Fig. 4) or in the CD8/ T cell (Fig. 5) subset. In contrast, the in vitro restimulation of immune T cells with PR8 induced a statistically significant increase, versus unstimulated cells, in the percentage of IL2-, IL4-, and IFN-g-producing cells in both CD4/ (Figs. 2 and 4) and CD8/ T cell
(Figs. 3 and 5) subsets. Within the single immune cell subset, analysis of the frequency of the cells producing the different cytokines on virus restimulation showed that the percentage of IL2- and IL4-producing cells was the same in CD4/ and CD8/ T cell subsets. The percentage of IFN-g-producing cells was statistically ( P õ 0.05) higher in the CD8/ T cell subset. The results also showed that the percentage of total spleen T cells positive for the single cytokine was quite similar, although slightly higher values were obtained for IL2and IFN-g-positive T cells (11.6 and 11.0%, respectively) in comparison to IL4-positive T cells (8.6%). Finally, the immunophenotypic analysis (Fig. 6) of
AID
cia
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
227
aration, the distribution of both cytokine-producing CD4/ and CD8/ T cells generated by the restimulation with influenza virus of unseparated immune spleen cells. This provided conditions for the determination of possible cross-regulatory activities exerted by the T cell subsets through the different cytokines they produce. By this method we found that in vitro restimulation of immune spleen cells with influenza virus induced an increase, versus unstimulated spleen cells, in the proportion of both CD4/ and CD8/ T cells positive for IL2, IL4, or IFN-g. No significant increase in cytokine-producing T cells could be detected in virus-stimulated cultures of naive cells. These results indicate that, upon virus restimula-
FIG. 4. Frequency of cytokine-producing CD4/ T cells, determined by flow cytometry analysis, in in vitro restimulated spleen cells. The percentage of IL2- (A), IFN-g- (B), and IL4- (C) producing cells was evaluated on both naive and immune animal-derived spleen cells restimulated (hatched bars) or not (open bars) with UV-inactivated PR8. Percentage specific cytokine-labeled cells was calculated by subtracting the isotype control (1%) from the specific anti-cytokine signal. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. ***P õ 0.001 (Student’s t test).
splenic T cells showed a slight but statistically significant increase in the percentage of CD8/ T cells in mice injected with PR8 when compared to naive animals, paralleled by a decrease of the percentage of CD4/ T cells. No significant changes in the proportion of both CD4/ and CD8/ T cells were observed in either naive or immune spleen cells after the 24-hr in vitro restimulation with PR8 (data not shown). DISCUSSION The combined analysis by flow cytometry of intracytoplasmic cytokine production and surface antigen expression allowed us to define, without previous cell sep-
AID
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
FIG. 5. Frequency of cytokine-producing CD8/ T cells, determined by flow cytometry analysis, in in vitro restimulated spleen cells. The percentage of IL2- (A), IFN-g- (B) and IL4- (C) producing cells was evaluated on both naive and immune animal-derived spleen cells restimulated (hatched bars) or not (open bars) with UV-inactivated PR8. Percentage specific cytokine-labeled cells was calculated by subtracting the isotype control (1%) from the specific anti-cytokine signal. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. ***P õ 0.001 (Student’s t test).
cia
AP: Cell Immuno
228
FALCHETTI ET AL.
tion, immune CD4/ and CD8/ T cells concurrently produce the same pattern of antiviral and regulatory cytokines and that the synthesis and secretion of these cytokines are not mutually exclusive either within or between the two T cell subsets. This evidently does not exclude the presence of cytokine-mediated crossregulatory activities of T cell subsets on their secretory functions. Moreover, we do not know if there is a coproduction by the single cells of the three cytokines or if each one is produced by a different cell. The finding that either T cell subset might produce the same immune regulatory cytokines could explain the persistence of antiviral responses following the depletion of one but not both cell subsets (10, 18). The presence of significant proportions of both CD4/ and CD8/ T cells positive for IL2, IL4, and IFN-g strongly suggests that influenza virus restimulation did not induce dominant type 1 or type 2 cytokine response. This was confirmed by the detection of significant levels of both cytokine types in the culture supernatants. Thus, the combination of the two responses appears to be the more appropriate in controlling virus infection. Our data are only partially in agreement with the results of previous studies (6, 7, 19) concerning the virus-induced production of cytokines by T cells obtained from different sources. In one study (6), the production of cytokines was tested on MLN cells obtained from mice infected with influenza virus by intranasal (i.n.) route. Experiments were performed by ELISPOT assay on both FACS-sorted, freshly isolated and restimulated MLN cells. The results indicated that both CD4/ and CD8/ T cell subsets contained IL2- and IFN-g-producing cells. In contrast to our results, the frequency of
IL4-producing cells within the two subsets was extremely low, if any, and no significant levels of IL4 were detected in culture supernatants of virus-restimulated cells. A similar pattern of cytokine production was found in another study (7) in which the virus was, likewise, administered i.n. but cytokine production was induced by stimulating immune MLN cells with antiCD3 monoclonal antibody. The same study also reported an absence of IL4 production by lung tissue T cells obtained from immune animals. The differences between these and our data could be due to the different experimental conditions used, namely ip versus i.n. virus administration, restimulation by UV-inactivated virus versus virus-infected irradiated splenocytes or anti-CD3, and determination of cytokine-producing cells by flow cytometry versus ELISPOT assay or mRNA expression. Alternatively, the differences between MLN and spleen T cells in the production of IL4 could be due to different effector activities of the different T cell compartments, perhaps determined by local conditions (4). In this case, a possible explanation could be related to differences in the availability of antigen and antigen-presenting cells (APC). This is suggested by the observation that BAL cells from influenza virus-infected animals, differently from MLN cells, produce IL4 (6, 13). One can, therefore, hypothesize that higher signal levels are necessary for the production of IL4 than, for example, of IL2 and that the differences in cytokine production could depend on the fact that, following infection, BAL cells are exposed to larger amounts of antigen and APC than MLN cells (6, 20). Our immunization protocol supports this hypothesis. In facts, it supplies spleen cells with large amounts of antigen and APC which induce a pattern of cytokine production, namely of IL4, similar to that of BAL cells. It must finally be noted that the frequency of IL4-producing cells was the same in CD4/ and CD8/ T cell subsets. Production of IL4 by spleen T cells suggests that these cells might play a different role from that of MLN cells in the response against virus infection (such as the regulation of the production of particular immunoglobulin classes or subclasses). With regard to the role of CD4/ and CD8/ T cells in the IFN-g response to virus immunization, we found that upon in vitro restimulation the frequency of IFNg-producing CD8/ T cells (CD8//IFN-g/ T cells) was significantly higher than that of CD4//IFN-g/ T cells. This is in accordance with results obtained by different authors (21–23) suggesting that antiviral activity is exerted by CD8/ T cells via an IFN-g-dependent mechanism. A slight but statistically significant increase in the percentage of total CD8/ T cells in PR8-treated animals versus untreated animals was also observed. On the other hand, our data also showed a significant virus restimulation-induced increase in the frequency of CD4//IFN-g/ T cells. As suggested by other authors (24, 25) who obtained similar results by using vaccinia
AID
cia
FIG. 6. Immunophenotypic analysis of splenic T cells of naive mice or mice immunized with influenza virus. The immunophenotypes were determined by FACS analysis and expressed as percentages of total lymphocytes. Mean values { SEM of at least six individual mice are shown. Statistical analysis was performed comparing immunized with naive mice. **P õ 0.01; ***P õ 0.001 (Student’s t test).
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
virus as antigen, the IFN-g produced by CD4/ T cells could play a regulatory role (alone or in combination with other factors produced by the same cells), in the production of IFN-g by CD8/ T cells. Influenza-specific CD4/ T cells were also found, however, to exert a direct specific cytolytic activity on virus-infected cells (26), although it is not clear if this activity is IFN-g-dependent (20, 27). In conclusion, we have shown that the in vitro restimulation with influenza virus of immune spleen cells induces the concurrent production by CD4/ and CD8/ T cells of IL2, IL4, and IFN-g. This observation provides further evidence that both CD4/ and CD8/ T cell subsets play an important role in the cytokine response to virus infection. Furthermore, the finding that cytokine response was not type 1- or type 2-dominated suggests that the combination of the two responses may be the more effective in controlling virus infection or that a redundancy in the immune response may occur upon influenza virus immunization. ACKNOWLEDGMENTS This work was supported by Italian National Research Council (CNR) Projects FATMA and CT04 and by a grant from ‘‘Centro Studi Ministero della Sanita`,’’ Rome, Italy.
REFERENCES 1. Mosmann, T. R., and Coffman, R. L., Adv. Immunol. 46, 111, 1989. 2. Paul, W. E., and Seder, R. A., Cell 76, 241, 1994. 3. Clerici, M., and Shearer, G. M., Immunol. Today 15, 575, 1994. 4. Biron, C. A., Curr. Opin. Immunol. 6, 530, 1994. 5. Carding, S. R., Allan, W., McMickle, A., and Doherty, P. C., J. Exp. Med. 177, 475, 1993. 6. Sarawar, S. R., and Doherty, P. C., J. Virol. 68, 3112, 1994.
AID
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
229
7. Baumghart, N., Brown, L., Jackson, D., and Kelso, A., J. Virol. 68, 7575, 1994. 8. Ada, G. L., and Jones, P. D., Curr. Top. Microbiol. Immunol. 128, 1, 1986. 9. Doherty, P. C., Br. Med. Bull. 41, 7, 1985. 10. Allan, W., Tabi, Z., Cleary, A., and Doherty, P. C., J. Immunol. 144, 3980, 1990. 11. Sad, S., Marcotte, R., and Mosmann, T. R., Immunity 2, 271, 1995. 12. Palladino, G., Scherle, P. A., and Gerhard, W., J. Virol. 65, 6071, 1991. 13. Sarawar, S. R., Carding, S. R., Allan, W., McMickle, A., Fujiashi, K., Kiyono, H., McGhee, J. R., and Doherty, P. C., Reg. Immunol. 5, 142, 1993. 14. Mosmann, T. R., and Kevin, K. W., Immunol. Today 12, A49, 1991. 15. Gajewsky, T. F., and Fitch, F. W., J. Immunol. 140, 4245, 1988. 16. Falchetti, R., Di Francesco, P., Lanzilli, G., Gaziano, R., Casalinuovo, I., Ravagnan, G., and Garaci, E., Cell. Immunol. 164, 57, 1995. 17. Sander, B., Andersson, J., and Andersson, U., Immunol. Rev. 119, 65, 1991. 18. Eichelberger, M., Allan, W., Zijristra, M., Jaenisch, R., and Doherty, P. C., J. Exp. Med. 174, 875, 1991. 19. Sarawar, S. R., Sangster, M., Coffman, R. L., and Doherty, P. C., J. Immunol. 153, 1246, 1994. 20. Eichelberger, M. C., Wang, M., Allan, W., Webster, R. G., and Doherty, P. C., J. Gen. Virol. 72, 1695, 1991. 21. Ruby, J., and Ramshaw, I., Lymphokine Cytokine Res. 10, 353, 1991. 22. Klavinskis, L. S., Geckeler, R., and Oldstone, M. B. A., J. Gen. Virol. 70, 3317, 1989. 23. Ramsay, A. J., Ruby, J., and Ramshaw, I. A., Immunol. Today 14, 155, 1993. 24. Carpenter, E. A., Ruby, J., and Ramshaw, I. A., J. Immunol. 152, 2652, 1994. 25. Ramshaw, I., Ruby, J., Ramsay, A., Ada, G., and Karupiah, G., Immunol. Rev. 127, 157, 1992. 26. Taylor, P. M., Esquival, F., and Askonas, B. A., Int. Immunol. 2, 323, 1990. 27. Graham, M. B., Dalton, D. K., Giltinan, D., Braciale, V. L., Stewart, T. A., and Braciale, T. J., J. Exp. Med. 178, 1725, 1993.
cia
AP: Cell Immuno
170, 222–229 (1996)
0155
Splenic CD4/ and CD8/ T Cells from Influenza Immune Mice Concurrently Produce in Vitro IL2, IL4, and IFN-g ROBERTO FALCHETTI,*,1 GIULIA LANZILLI,* IDA A. CASALINUOVO, ROBERTA GAZIANO, ANNA TERESA PALAMARA, PAOLO DI FRANCESCO, GIAMPIETRO RAVAGNAN,* AND ENRICO GARACI Department of Experimental Medicine and Biochemical Sciences, University of Rome, Tor Vergata, Rome, Italy; and *Institute of Experimental Medicine, CNR, Rome, Italy Received September 22, 1995; accepted March 1, 1996
The cytokine responses exerted by virus-primed spleen T cells upon in vitro restimulation were studied. Spleen cells obtained from mice injected intraperitoneally with A/PR8 (H1N1) influenza virus (PR8) were restimulated in vitro with UV-inactivated PR8 virus. The percentage of both CD4/ and CD8/ T cells producing IL2, IL4, or IFN-g was assayed at the single cell level by flow cytometric analysis of intracytoplasmic cytokine content. In parallel, the levels of the different cytokines in spleen cell culture supernatants were quantitated by enzyme linked immunosorbent assay. The results showed that in vitro virus restimulation of immune spleen cells induced the concurrent increase, in both CD4/ and CD8/ T cells, of the frequency of IL2-, IFN-g-, and IL4-producing cells. The frequency of IFN-g-producing T cells was found to be significantly higher in CD8/ T cells. Significant levels of the three cytokines were also detected in the culture supernatants. These data suggest that both CD4/ and CD8/ T cells play an important role in cytokine response to virus infection and that the synthesis and secretion of antiviral and regulatory cytokines is not mutually exclusive either between or within the two T cell subsets. The results of the experiments also indicated that the virus restimulation did not induce a dominant type 1 or type 2 cytokine response. q 1996 Academic Press, Inc.
INTRODUCTION Cytokine production by T cells plays an important role in the generation of immune response to infection. Exposure to different nonviral pathogens can stimulate at least two patterns of cytokine production by CD4positive (CD4/) T-helper cells (1, 2). Responses that are characterized by the predominant secretion of interleukin (IL) 2, interferon-g (IFN-g), and lymphotoxins are 1 To whom correspondence should be addressed at Institute of Experimental Medicine, CNR, Viale C. Marx, 43, 00137 Rome, Italy.
222
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
CI 8145
/
6c10$$$$$1
classified as T-helper-1 (Th1)-type responses and generally occur during infections with intracellular bacteria and parasites. In contrast, responses following nematode infections and characterized by a predominant production of IL4, IL5, IL9, IL10, and IL13 are classified as T-helper-2 (Th2)-type responses. It has recently been suggested (3) that the two types of cytokine production be termed type 1 and type 2, respectively, since some of the immunoregulatory cytokines are also produced during infection either by other populations of T cells or by non-T cells. Less precise information has been gathered on cytokine responses exerted by T cells to viral infections. Studies (4, 5) performed using different models of viral infection showed that both type 1 and type 2 cytokines can be expressed by T cells, although locally a preferential production of a certain type of cytokine can be found. Recent results (6, 7) obtained on influenza virusinfected mice showed, for instance, differences in the pattern of cytokine production by T cells from bronchoalveolar lavage (BAL), lung tissue, and mediastinal lymph node (MLN). An interesting aspect concerning the study of T cell cytokine response to viral infection is the determination of the cytokine-producing cell immunophenotypes, which is the determination of the pattern of cytokine production not only by CD4/ but also by CD8/ T cells, as this latter subset of T cells plays a central role in antiviral immunity (8, 9). The early events of the response to viral infections are, in fact, associated with the activation of specific class I-restricted cytotoxic T lymphocyte (CTL) responses and CD8/ T-cell expansion (10). At this stage of immune response different cytokines, i.e., IL2, IL4, IFN-g, and IL10, possibly promoting activation and regulation of virus-specific CTL responses, are expressed in lymphoid tissues. Recent evidence suggests that alloantigen-stimulated CD8/ mouse spleen cells can secrete these cytokines (11) and that CD8/ T cells secreting IL2, IL5, IL10, and IFNg are present in MLN, BAL, or lung tissue of influenza virus-infected mice (6, 7). It is therefore possible that CD8/ T cells are the source of the cytokines required for
05-17-96 06:33:19
cia
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
the early phase of immune response, which thus would be independent from CD4/ T cell cytokine release. On the other hand, studies on the role of class II-restricted CD4/ T cells indicated that virus-primed CD8/ cells did not give a significant CTL response in the absence of CD4/ T cells or of their produced factors, thus suggesting that the antiviral CTL response is dependent upon Thelper cells (12). The pattern of virus-induced CD4/ T cell cytokine production was examined by different laboratories. In particular, experiments using ELISPOT assay showed (6) that both freshly isolated and virusrestimulated CD4/ T cells obtained from MLN of influenza virus-infected mice produced IL2, IL10, and IFNg. Interestingly, these cells, differently from BAL cells obtained from animals similarly treated (13), did not produce significant amounts of IL4. The greatest part of the data on both virus-induced CD4/ and CD8/ T cell cytokine production was generated by restimulating virus-specific T cell clones or separated T cell subsets obtained by fluorescence-activated sorting or selective killing. In consideration of the fact that some type 1 and type 2 cytokines can exert cross-regulatory (and in particular cross-inhibitory) activities (14, 15), the pattern of the virus-primed CD4/ and CD8/ T cell cytokine production should preferably be evaluated in an experimental situation in which both populations are contemporaneously present. To this aim we studied, at the single cell level, virusinduced cytokine production by CD4/ and CD8/ T cells under experimental conditions ensuring the contemporaneous presence of the different immunocompetent cell populations at the moment of restimulation with the virus of the immune cells. A model system was established in which unseparated spleen cells obtained from mice injected intraperitoneally (ip) with PR8 influenza virus were restimulated in vitro with UV-inactivated PR8 virus. The percentage of both CD4/ and CD8/ T cells producing the single cytokines (e.g., IL2, IL4, and IFN-g) was evaluated by simultaneous flow cytometric analysis of cellular intracytoplasmic cytokine content and surface markers. By this method both the frequency of cytokine-producing cells within the two subsets and the type 1–type 2 profile were analyzed. In parallel, the levels of the different cytokines in the culture supernatants were quantitated by ELISA. The results showed that (i) upon in vitro restimulation, virus-primed CD4/ and CD8/ T cells concurrently produced the same cytokines, i.e., IL2, IL4, and IFNg; (ii) the virus did not induce a dominant type 1 or type 2 response; and (iii) no evident cross-inhibitory activity on cytokine production was exerted by the two T cell subsets.
223
Mice were negative for antibodies to mycoplasma, murine hepatitis virus, and Sendai virus by ELISA. Virus A/PR8 (H1N1) influenza virus (PR8) was grown in the allantoid cavities of 10-day-old embryonated chicken eggs. Virus titer was expressed in hemoagglutinating units (HAU)/ml. UV-inactivated PR8 was prepared by exposing virus suspensions to UV light (40 W, 254 nm, 8-cm distance) for 5 min. Immunization Mice were injected ip with 75 HAU/animal of PR8 daily for 3 consecutive days. Cell Cultures Mice were sacrificed 2 days after the last virus injection (fifth day of immunization) by cervical dislocation. Spleen cell suspensions from immunized or naive animals (injected ip with virus-free appropriately diluted allantoic fluid, following immunization timing) were cultured in 24-well plates (Becton–Dickinson, Lincoln Park, NJ) at 1 1 107/ml (2 ml/well) in complete culture medium (CCM), consisting of RPMI 1640 supplemented with 10% fetal calf serum (FCS, Flow Laboratories, Rickmansworth, UK), 1% L-glutamine (Flow Laboratories), and 1% penicillin–streptomycin (Flow Laboratories). Spleen cells were cultured alone or with UV-inactivated PR8 at the concentration of 250 HAU/ ml. Cell cultures were incubated at 377C in humidified 5% CO2 . At the indicated time points culture supernatants or cells were removed and used for successive evaluations. Cytokine Determination Concentrations of IFN-g, IL2, and IL4 in the culture supernatants were determined by two-site sandwich ELISA according to the method previously described (16) using rat anti-mouse IFN-g mAb (PharMingen, San Diego, CA; clone XMG1.2), rat anti-mouse IL-2 mAb (PharMingen; clone JES6-1A12), and rat antimouse IL4 mAb (PharMingen; clone BVD4-1D11) as coat and biotinylated rat anti-mouse IFN-g mAb (PharMingen; clone XMG1.2), biotinylated rat anti-mouse IL2 mAb (PharMingen; clone JES6-5H4), and biotinylated rat anti-mouse IL4 mAb (PharMingen; clone BVD6-24G2) as detecting antibodies. Standards of rIFN-g, rIL2, and rIL4 purchased from PharMingen were used as reference.
MATERIALS AND METHODS
Detection of Cytokine-Producing Cells by Intracellular Immunofluorescence Staining
Animals Male BALB/c mice (5–7 weeks old) purchased from Charles River Italia (Calco, Como, Italy) were used.
To determine the frequency of cytokine-producing CD4/ and CD8/ T cells, the method described by Sander et al. (17) with minor modifications (16) was
AID
cia
CI 8145
/
6c10$$$$$2
05-17-96 06:33:19
AP: Cell Immuno
224
FALCHETTI ET AL.
used. Briefly, after culture, 106 cells fixed in 4% paraformaldehyde were stained by incubating them at room temperature (RT) with rat anti-mouse cytokine mAb in PBS/BSA/NaN3 containing 0.3% saponin (Sigma, St. Louis, MO). The following rat anti-mouse cytokine mAb were used: anti-mouse IL-2 mAb (PharMingen; clone JES6-1A12, staining concentration 20 mg/ml), antimouse IL-4 mAb (PharMingen; clone BVD4-1D11, staining concentration 20 mg/ml), and anti-mouse IFNg mAb (PharMingen; clone XMG1.2, staining concentration 20 mg/ml). As negative controls some aliquots of cell suspensions were incubated with an irrelevant isotype-matched mAb. As second-step Ab, a fluorescein (FITC)-conjugated immunopure goat anti-rat IgG (H / L) (Pierce, Rockford, IL) was used. To determine the immunophenotype of the cytokine-producing cells, samples were subsequently surface stained with rat anti-mouse CD4 mAb (PharMingen; clone RM4-5, staining concentration 1 mg/million cells) and rat antimouse CD8 mAb (PharMingen; clone 53-6.7, staining concentration 1 mg/million cells) coupled to phycoerythrin (PE). A FACScan flow cytometer (Becton–Dickinson, San Jose, CA) was used to analyze each sample. Signals from light scatter channels were collected in linear mode, and signals from the fluorescence-detecting PMTs were collected as logarithmic signals. Ten thousand events were analyzed for each test and signals from all parameters were captured as listmode data and analyzed by FACScan research software. Statistical Analysis Data are expressed as means { SEM. Statistical significance was determined using Student’s t test. RESULTS PR8 Immune Spleen Cells Secrete both Type 1 (IL2 and IFN-g) and Type 2 (IL4) Cytokines upon in Vitro Restimulation with the Virus In a first series of experiments, to control both the pattern and the peak time of in vitro virus-induced cytokine secretion, spleen cells obtained from mice injected ip with PR8 virus were cultured for 24 or 48 hr in the presence of UV-inactivated PR8. Cultures of both naive spleen cells stimulated with PR8 or of immune spleen cells lacking the virus were also established. At the end of incubation time the supernatants were removed and the levels of IL2, IL4, and IFN-g were measured by ELISA. Figure 1 shows that after 24 hr of incubation, the levels in PR8-stimulated immune spleen cell supernatants of the three tested cytokines were significantly higher than those of unstimulated cells. At 48 hr (Fig. 1) IL2 levels markedly decreased, whereas the levels of both IL4 and IFN-g remained unchanged. No significant production of cytokines was found in cultures of naive spleen cells stimulated with PR8 (data not shown). These results
AID
CI 8145
/
6c10$$$$$2
05-17-96 06:33:19
FIG. 1. In vitro production of cytokines by influenza virus-primed spleen cells. The levels of IL2 (A), IFN-g (B), and IL4 (C) were determined by ELISA in culture supernatants of PR8-immunized animalderived spleen cells cultured for 24 and 48 hr in the presence or absence of UV-inactivated PR8. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. *P õ 0.05; **P õ 0.01; ***P õ 0.001 (Student’s t test).
showed that in vitro virus-restimulated immune spleen cells were able to produce significant amounts of both type 1 and type 2 cytokines and that the peak level was already reached at 24 hr. Both CD4/ and CD8/ PR8 Immune T Cells Produce IL2, IL4, and IFN-g upon in Vitro Restimulation with Virus The production of cytokines by in vitro virus-restimulated immune CD4/ and CD8/ T cells was analyzed at the single cell level. Through the use of flow cytome-
cia
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
225
FIG. 2. Flow cytometric analysis of the percentage of IL2-, IFN-g-, and IL4-producing CD4/ splenocytes. The frequency of cytokineproducing cells was evaluated on PR8-immunized animal-derived spleen cells both unstimulated (A) or restimulated in vitro with PR8 (B). Samples were analyzed by first gating on the CD4-positive population (Y axis). The threshold for detection of positive IL2-, IFN-g-, and IL4-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with an isotype-specific antibody (C). The percentage of specific IL2-, IFN-g-, and IL4-positive lymphocytes (A, B) was calculated by subtracting the isotype control (1%) from the specific anti-IL2, IFN-g, and IL4 signal. X axis, FL1, log10 IL2, IFN-g, or IL4 signal; Y axis, FL2, log10 CD4 signal.
try, the frequency of the CD4/ or CD8/ T cells containing intracytoplasmic IL2, IL4, or IFN-g was evaluated. Spleen cells obtained from PR8-immunized animals were cultured for 24 hr in the presence or absence of the virus. As control, spleen cells obtained from naive animals were also incubated with or without PR8. At the end of incubation time the frequencies of both CD4/ and CD8/ T cells expressing IL2, IL4, or IFN-g were determined by immunofluorescence, staining fixed and saponin-permeabilized cells for either CD4 or CD8 surface molecule, and intracytoplasmic IL2, IL4, or IFN-g. As paraformaldehyde fixation increased autofluorescence in nonlymphocyte populations, analysis of the samples was performed by first
gating on CD4/ or CD8/ T cells. Previous experiments (16) showed that no significant cytokine immunofluorescence is present in fixed but not saponin-permeabilized cells, thus excluding the possibility that in our system significant fluorescent signals could be due to receptor-bound FITC-labeled cytokines on the surface of target cells. Irrelevant isotype-specific antibody was used as a control for nonspecific binding of the anticytokine monoclonal antibodies. As described in a previous paper (16), the threshold for detection of positive cytokine-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with isotype-specific antibodies (Figs. 2C and 3C). The percentage of specific cytokine-positive cells was cal-
AID
cia
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
226
FALCHETTI ET AL.
FIG. 3. Flow cytometric analysis of the percentage of IL2-, IFN-g-, and IL4-producing CD8/ splenocytes. The frequency of cytokineproducing cells was evaluated on PR8-immunized animal-derived spleen cells both unstimulated (A) or restimulated in vitro with PR8 (B). Samples were analyzed by first gating on the CD8-positive population (Y axis). The threshold for detection of positive IL2-, IFN-g-, and IL4-producing cells was arbitrarily set at 1% above the negative control consisting of cells treated with an isotype-specific antibody (C). The percentage of specific IL2-, IFN-g-, and IL4-positive lymphocytes (A, B) was calculated by subtracting the isotype control (1%) from the specific anti-IL2, IFN-g, and IL4 signal. X axis, FL1, log10 IL2, IFN-g, or IL4 signal; Y axis, FL2, log10 CD8 signal.
culated by subtracting the isotype control from the specific anti-cytokine signal. Results of the analysis of cytokine production at the single cell level showed that at this cutoff level, only a slight and not significant percentage of cytokine-positive cells was observed in both unstimulated and PR8-stimulated naive spleen cells as well as in unstimulated immune spleen cells, either in the CD4/ (Fig. 4) or in the CD8/ T cell (Fig. 5) subset. In contrast, the in vitro restimulation of immune T cells with PR8 induced a statistically significant increase, versus unstimulated cells, in the percentage of IL2-, IL4-, and IFN-g-producing cells in both CD4/ (Figs. 2 and 4) and CD8/ T cell
(Figs. 3 and 5) subsets. Within the single immune cell subset, analysis of the frequency of the cells producing the different cytokines on virus restimulation showed that the percentage of IL2- and IL4-producing cells was the same in CD4/ and CD8/ T cell subsets. The percentage of IFN-g-producing cells was statistically ( P õ 0.05) higher in the CD8/ T cell subset. The results also showed that the percentage of total spleen T cells positive for the single cytokine was quite similar, although slightly higher values were obtained for IL2and IFN-g-positive T cells (11.6 and 11.0%, respectively) in comparison to IL4-positive T cells (8.6%). Finally, the immunophenotypic analysis (Fig. 6) of
AID
cia
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
227
aration, the distribution of both cytokine-producing CD4/ and CD8/ T cells generated by the restimulation with influenza virus of unseparated immune spleen cells. This provided conditions for the determination of possible cross-regulatory activities exerted by the T cell subsets through the different cytokines they produce. By this method we found that in vitro restimulation of immune spleen cells with influenza virus induced an increase, versus unstimulated spleen cells, in the proportion of both CD4/ and CD8/ T cells positive for IL2, IL4, or IFN-g. No significant increase in cytokine-producing T cells could be detected in virus-stimulated cultures of naive cells. These results indicate that, upon virus restimula-
FIG. 4. Frequency of cytokine-producing CD4/ T cells, determined by flow cytometry analysis, in in vitro restimulated spleen cells. The percentage of IL2- (A), IFN-g- (B), and IL4- (C) producing cells was evaluated on both naive and immune animal-derived spleen cells restimulated (hatched bars) or not (open bars) with UV-inactivated PR8. Percentage specific cytokine-labeled cells was calculated by subtracting the isotype control (1%) from the specific anti-cytokine signal. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. ***P õ 0.001 (Student’s t test).
splenic T cells showed a slight but statistically significant increase in the percentage of CD8/ T cells in mice injected with PR8 when compared to naive animals, paralleled by a decrease of the percentage of CD4/ T cells. No significant changes in the proportion of both CD4/ and CD8/ T cells were observed in either naive or immune spleen cells after the 24-hr in vitro restimulation with PR8 (data not shown). DISCUSSION The combined analysis by flow cytometry of intracytoplasmic cytokine production and surface antigen expression allowed us to define, without previous cell sep-
AID
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
FIG. 5. Frequency of cytokine-producing CD8/ T cells, determined by flow cytometry analysis, in in vitro restimulated spleen cells. The percentage of IL2- (A), IFN-g- (B) and IL4- (C) producing cells was evaluated on both naive and immune animal-derived spleen cells restimulated (hatched bars) or not (open bars) with UV-inactivated PR8. Percentage specific cytokine-labeled cells was calculated by subtracting the isotype control (1%) from the specific anti-cytokine signal. Results are expressed as means { SEM of at least six independent experiments. Statistical analysis was performed comparing PR8-stimulated cultures with unstimulated cultures. ***P õ 0.001 (Student’s t test).
cia
AP: Cell Immuno
228
FALCHETTI ET AL.
tion, immune CD4/ and CD8/ T cells concurrently produce the same pattern of antiviral and regulatory cytokines and that the synthesis and secretion of these cytokines are not mutually exclusive either within or between the two T cell subsets. This evidently does not exclude the presence of cytokine-mediated crossregulatory activities of T cell subsets on their secretory functions. Moreover, we do not know if there is a coproduction by the single cells of the three cytokines or if each one is produced by a different cell. The finding that either T cell subset might produce the same immune regulatory cytokines could explain the persistence of antiviral responses following the depletion of one but not both cell subsets (10, 18). The presence of significant proportions of both CD4/ and CD8/ T cells positive for IL2, IL4, and IFN-g strongly suggests that influenza virus restimulation did not induce dominant type 1 or type 2 cytokine response. This was confirmed by the detection of significant levels of both cytokine types in the culture supernatants. Thus, the combination of the two responses appears to be the more appropriate in controlling virus infection. Our data are only partially in agreement with the results of previous studies (6, 7, 19) concerning the virus-induced production of cytokines by T cells obtained from different sources. In one study (6), the production of cytokines was tested on MLN cells obtained from mice infected with influenza virus by intranasal (i.n.) route. Experiments were performed by ELISPOT assay on both FACS-sorted, freshly isolated and restimulated MLN cells. The results indicated that both CD4/ and CD8/ T cell subsets contained IL2- and IFN-g-producing cells. In contrast to our results, the frequency of
IL4-producing cells within the two subsets was extremely low, if any, and no significant levels of IL4 were detected in culture supernatants of virus-restimulated cells. A similar pattern of cytokine production was found in another study (7) in which the virus was, likewise, administered i.n. but cytokine production was induced by stimulating immune MLN cells with antiCD3 monoclonal antibody. The same study also reported an absence of IL4 production by lung tissue T cells obtained from immune animals. The differences between these and our data could be due to the different experimental conditions used, namely ip versus i.n. virus administration, restimulation by UV-inactivated virus versus virus-infected irradiated splenocytes or anti-CD3, and determination of cytokine-producing cells by flow cytometry versus ELISPOT assay or mRNA expression. Alternatively, the differences between MLN and spleen T cells in the production of IL4 could be due to different effector activities of the different T cell compartments, perhaps determined by local conditions (4). In this case, a possible explanation could be related to differences in the availability of antigen and antigen-presenting cells (APC). This is suggested by the observation that BAL cells from influenza virus-infected animals, differently from MLN cells, produce IL4 (6, 13). One can, therefore, hypothesize that higher signal levels are necessary for the production of IL4 than, for example, of IL2 and that the differences in cytokine production could depend on the fact that, following infection, BAL cells are exposed to larger amounts of antigen and APC than MLN cells (6, 20). Our immunization protocol supports this hypothesis. In facts, it supplies spleen cells with large amounts of antigen and APC which induce a pattern of cytokine production, namely of IL4, similar to that of BAL cells. It must finally be noted that the frequency of IL4-producing cells was the same in CD4/ and CD8/ T cell subsets. Production of IL4 by spleen T cells suggests that these cells might play a different role from that of MLN cells in the response against virus infection (such as the regulation of the production of particular immunoglobulin classes or subclasses). With regard to the role of CD4/ and CD8/ T cells in the IFN-g response to virus immunization, we found that upon in vitro restimulation the frequency of IFNg-producing CD8/ T cells (CD8//IFN-g/ T cells) was significantly higher than that of CD4//IFN-g/ T cells. This is in accordance with results obtained by different authors (21–23) suggesting that antiviral activity is exerted by CD8/ T cells via an IFN-g-dependent mechanism. A slight but statistically significant increase in the percentage of total CD8/ T cells in PR8-treated animals versus untreated animals was also observed. On the other hand, our data also showed a significant virus restimulation-induced increase in the frequency of CD4//IFN-g/ T cells. As suggested by other authors (24, 25) who obtained similar results by using vaccinia
AID
cia
FIG. 6. Immunophenotypic analysis of splenic T cells of naive mice or mice immunized with influenza virus. The immunophenotypes were determined by FACS analysis and expressed as percentages of total lymphocytes. Mean values { SEM of at least six individual mice are shown. Statistical analysis was performed comparing immunized with naive mice. **P õ 0.01; ***P õ 0.001 (Student’s t test).
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
AP: Cell Immuno
CYTOKINE PRODUCTION BY VIRUS-IMMUNE SPLEEN T CELLS
virus as antigen, the IFN-g produced by CD4/ T cells could play a regulatory role (alone or in combination with other factors produced by the same cells), in the production of IFN-g by CD8/ T cells. Influenza-specific CD4/ T cells were also found, however, to exert a direct specific cytolytic activity on virus-infected cells (26), although it is not clear if this activity is IFN-g-dependent (20, 27). In conclusion, we have shown that the in vitro restimulation with influenza virus of immune spleen cells induces the concurrent production by CD4/ and CD8/ T cells of IL2, IL4, and IFN-g. This observation provides further evidence that both CD4/ and CD8/ T cell subsets play an important role in the cytokine response to virus infection. Furthermore, the finding that cytokine response was not type 1- or type 2-dominated suggests that the combination of the two responses may be the more effective in controlling virus infection or that a redundancy in the immune response may occur upon influenza virus immunization. ACKNOWLEDGMENTS This work was supported by Italian National Research Council (CNR) Projects FATMA and CT04 and by a grant from ‘‘Centro Studi Ministero della Sanita`,’’ Rome, Italy.
REFERENCES 1. Mosmann, T. R., and Coffman, R. L., Adv. Immunol. 46, 111, 1989. 2. Paul, W. E., and Seder, R. A., Cell 76, 241, 1994. 3. Clerici, M., and Shearer, G. M., Immunol. Today 15, 575, 1994. 4. Biron, C. A., Curr. Opin. Immunol. 6, 530, 1994. 5. Carding, S. R., Allan, W., McMickle, A., and Doherty, P. C., J. Exp. Med. 177, 475, 1993. 6. Sarawar, S. R., and Doherty, P. C., J. Virol. 68, 3112, 1994.
AID
CI 8145
/
6c10$$$$$3
05-17-96 06:33:19
229
7. Baumghart, N., Brown, L., Jackson, D., and Kelso, A., J. Virol. 68, 7575, 1994. 8. Ada, G. L., and Jones, P. D., Curr. Top. Microbiol. Immunol. 128, 1, 1986. 9. Doherty, P. C., Br. Med. Bull. 41, 7, 1985. 10. Allan, W., Tabi, Z., Cleary, A., and Doherty, P. C., J. Immunol. 144, 3980, 1990. 11. Sad, S., Marcotte, R., and Mosmann, T. R., Immunity 2, 271, 1995. 12. Palladino, G., Scherle, P. A., and Gerhard, W., J. Virol. 65, 6071, 1991. 13. Sarawar, S. R., Carding, S. R., Allan, W., McMickle, A., Fujiashi, K., Kiyono, H., McGhee, J. R., and Doherty, P. C., Reg. Immunol. 5, 142, 1993. 14. Mosmann, T. R., and Kevin, K. W., Immunol. Today 12, A49, 1991. 15. Gajewsky, T. F., and Fitch, F. W., J. Immunol. 140, 4245, 1988. 16. Falchetti, R., Di Francesco, P., Lanzilli, G., Gaziano, R., Casalinuovo, I., Ravagnan, G., and Garaci, E., Cell. Immunol. 164, 57, 1995. 17. Sander, B., Andersson, J., and Andersson, U., Immunol. Rev. 119, 65, 1991. 18. Eichelberger, M., Allan, W., Zijristra, M., Jaenisch, R., and Doherty, P. C., J. Exp. Med. 174, 875, 1991. 19. Sarawar, S. R., Sangster, M., Coffman, R. L., and Doherty, P. C., J. Immunol. 153, 1246, 1994. 20. Eichelberger, M. C., Wang, M., Allan, W., Webster, R. G., and Doherty, P. C., J. Gen. Virol. 72, 1695, 1991. 21. Ruby, J., and Ramshaw, I., Lymphokine Cytokine Res. 10, 353, 1991. 22. Klavinskis, L. S., Geckeler, R., and Oldstone, M. B. A., J. Gen. Virol. 70, 3317, 1989. 23. Ramsay, A. J., Ruby, J., and Ramshaw, I. A., Immunol. Today 14, 155, 1993. 24. Carpenter, E. A., Ruby, J., and Ramshaw, I. A., J. Immunol. 152, 2652, 1994. 25. Ramshaw, I., Ruby, J., Ramsay, A., Ada, G., and Karupiah, G., Immunol. Rev. 127, 157, 1992. 26. Taylor, P. M., Esquival, F., and Askonas, B. A., Int. Immunol. 2, 323, 1990. 27. Graham, M. B., Dalton, D. K., Giltinan, D., Braciale, V. L., Stewart, T. A., and Braciale, T. J., J. Exp. Med. 178, 1725, 1993.
cia
AP: Cell Immuno