INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

T-cell receptor (TCR)- T cells are present in only small numbers in peripheral lymphoid tissues but respond against infection by intracellular bacteria such as Mycobacterium tuberculosis (24), Listeria monocytogenes (43), Salmonella choleraesuis (9), and other pathogens (16, 26). The contribution of the T cells to protection against infection with intracellular bacteria has been tested in mice depleted of the cells. We have previously reported that pretreatment with anti-TCR- monoclonal antibody (MAb) impaired the host defense early after infection with L. monocytogenes (17). Mice rendered deficient in T cells by homologous recombination of the TCR- chain gene showed an impaired host defense against M. tuberculosis (29). Thus, T cells may play important roles in the host defense against infection by intracellular parasites. On the other hand, TCR--deficient mice showed exaggerated intestinal damage after oral infection with Eimeria verformis, suggesting that some T cells, such as intraepithelial T cells, play a role in resolution of the inflammatory process (45). T cells may be heterogeneous in function during the course of infectious diseases.

T cells are reported to respond to various bacterial products, such as tetanus toxoid (28), staphylococcal enterotoxin A (46), heat shock protein 65 (HSP65) (4), and isopentenyl pyrophosphate (34, 56), through a TCR-dependent mechanism. On the other hand, it has been reported that a significant fraction of the T-cell population is stimulated by lipopolysaccharide (LPS) (30, 41, 51), a cell wall component of gram-negative bacteria, through an apparently TCR-independent mechanism. We have recently found by using TCR--deficient mice that T cells play an important role in the priming of macrophages for tumor necrosis factor alpha (TNF-) production in response to LPS (38). Takada et al. have reported prominent increases in T cells in the peritoneal cavities of some strains of mice infected with Escherichia coli, a gram-negative extracellular bacterium (55). Protection against extracellular bacteria is thought to depend mainly on neutrophils and antibody (Ab) (58). Therefore, it is of interest to elucidate whether LPS-stimulated T cells are involved in the host defense against infection with E. coli.

Interleukin-15 (IL-15) is a novel cytokine that uses  and  chains of IL-2 receptor for signal transduction and shares many properties with, despite having no sequence homology to, IL-2 (14, 15). IL-2 is produced mainly by activated T cells, whereas IL-15 is produced by a wide variety of tissues, including placenta, skeletal muscle, kidney, and macrophages, upon stimulation with LPS (15, 54). IL-15 has stimulatory activities for natural killer (NK) cells, T cells, and B cells (1, 5, 6, 15). We have recently reported that T cells appearing after Salmonella infection or in intestinal intraepithelial lymphocytes can proliferate in response to exogenous IL-15 or IL-15 derived from infected macrophages (18, 39). A significant number of T cells, which emerge at the early stage of infection well before the appearance of IL-2-producing T cells, may preferentially use IL-15 from stimulated macrophages as a growth factor. However, the role of IL-15 in the host defense against bacterial infection remains to be elucidated.

In the present study, to elucidate the roles of T cells and IL-15 in protection against infection, we examined the host defense against E. coli infection in mice depleted of T cells or IL-15. Mice depleted of T cells by TCR- gene targeting showed exaggerated bacterial growth after E. coli infection. Administration of an anti-IL-15 MAb inhibited the appearance of T cells after infection and impaired the host defense against E. coli. The implications of these findings for the roles of T cells and IL-15 in the host defense against E. coli infection are discussed.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Mice. C3H/HeJ, C3H/HeN, and C57BL/6 mice were purchased from Japan SLC (Shizuoka, Japan). Eight- to 10-week-old female mice were used for the experiments. TCR-^/ mice, which lack the TCR- gene, have previously been described (23). Briefly, chimeric mice were produced by injecting ES clones into C57BL/6 mice. A homogeneous TCR-^+/ population was established by backcrossing  heterozygotes to C57BL/6 mice more than five times. The resultant heterozygotes (TCR-^+/) were bred to obtain the TCR-^/ homozygotes. Mice were housed under specific-pathogen-free conditions and offered food and water ad libitum.

Microorganisms and reagents. E. coli (ATTC 26; American Type Culture Collection, Rockville, Md.) grown in brain heart infusion broth (Difco Laboratories, Detroit, Mich.) was washed repeatedly, resuspended in phosphate-buffered saline (PBS), and stored at 80°C in small aliquots until used. LPS derived from E. coli O26:B6 or Salmonella typhimurium was purchased from Sigma Chemical Co. (St. Louis, Mo.). Mitomycin C (MMC) was purchased from Kyowa Hakko Kogyo Co. (Tokyo, Japan).

Preparation of lymphocytes. Mice were intraperitoneally (i.p.) inoculated with E. coli at a dose of one-fifth the 50% lethal dose (LD₅₀) (10⁸ CFU/mouse) in 1.0 ml of PBS on day 0. Peritoneal exudate cells (PEC) were harvested on days 0, 1, 3, 5, and 7 after inoculation by centrifugation at 110 × g for 5 min, washed twice, and resuspended at optimal concentrations in RPMI 1640 medium (GIBCO, Grand Island, N.Y.) supplemented with 10% serum. Smear specimens for differential counts were stained with Giemsa solution. PEC were spread on plastic plates and incubated for 1 h in a CO₂ incubator at 37°C to obtain nonadherent cells. For liver lymphocytes, fresh liver was immediately perfused with sterile Hanks balanced salt solution through the portal vein to wash out all remaining peripheral blood and then meshed with a stainless steel mesh. After the coarse pieces were removed by centrifugation at 50 × g for 1 min, the cell suspensions were again centrifuged, resuspended in 8 ml of 45% Percoll (Sigma), and layered on 5 ml of 67.5% Percoll. The gradients were centrifuged at 600 × g at 20°C for 20 min. Lymphocytes at the interface were harvested and washed twice with Hanks balanced salt solution.

Bacterial growth. Mice were inoculated i.p. with 10⁸ CFU of E. coli in 1.0 ml of PBS. The peritoneal contents were lavaged with 3 ml of PBS and harvested after gentle massage. Samples were serially diluted with PBS. The livers and spleens were removed and separately placed in homogenizers containing 5 ml of PBS. Samples were spread on Tripto-Soya agar (Nissui Pharmaceutical, Tokyo, Japan) plates, and colonies were counted after incubation for 24 h at 37°C.

Flow cytometry. Non-plastic-adherent PEC and liver lymphocytes were incubated with saturating amounts of FITC-, PE-, and biotin-conjugated Abs for 30 min at 4°C. To detect biotin-conjugated MAb, cells were stained with Red-613-conjugated streptavidin after incubation with a primary MAb. Cells were analyzed with a FACScan flow cytometer (Becton Dickinson, San Jose, Calif.). The liver lymphocytes were carefully gated by forward and side light scattering. The data were analyzed with FACScan Research software (Becton Dickinson).

Proliferation assay. The T cells were purified by cell sorting with an EPICS ELITE (Coulter, Hialeah, Fla.) electric cell sorter from the non-plastic-adherent cells and liver lymphocytes on day 3 after E. coli infection. The purity of sorted cells was more than 97% (data not shown). Ninety-six-well tissue culture plates were incubated overnight at 4°C with 100 µg of anti-TCR- MAb per ml. The plates were then washed thoroughly and incubated for 1 h at 37°C with RPMI 1640 medium containing 10% fetal calf serum. The sorted T cells (5 × 10⁴/well) were incubated in the 96-well plates for 48 h with or without 1, 10, or 100 µg of LPS per ml in the presence or absence of MMC-treated spleen cells (3 × 10⁴/well). During the last 8 h of incubation, 1.0 µCi of [³H]thymidine/well was added. The cells were then harvested, and the amount of [³H]thymidine incorporated was determined by scintillation counting.

Cytokine assays. The IL-2, IL-4, and gamma interferon (IFN-) levels in supernatants were determined by enzyme-linked immunosorbent assay (ELISA). ELISA for IFN- was performed in triplicate with Genzyme (Cambridge, Mass.) MAb according to the manufacturer's instructions, and ELISAs for IL-2 and IL-4 were performed with Biotrak MAbs (Amersham, Buckinghamshire, England).

V gene segment usage analysis. Total RNA was extracted by the acid-guanidium-phenol-chloroform method from T cells purified by cell sorting. cDNA synthesis and PCR were performed as described by Saiki et al. (47) with a cDNA cycle kit (Invitrogen Corp., San Diego, Calif.). RNA was primed with either 20 pmol of  chain C region (C) primers (5' CTT ATG GAG GAT TTG TTT CAG C 3') or 6.7 pmol of  chain J region (J) primers (5' TTG GTT CCA CAG TCA CTT GG 3') in 21-µl reaction mixtures for reverse transcription. The PCR was performed with a PCR thermal cycler (Takara Corp., Tokyo, Japan). PCR cycles were run for 1 min at 94°C, 1 min at 54°C, and 30 s at 72°C. Before the first cycle, a denaturation step for 7 min at 94°C was included, and after 35 cycles, the extension was prolonged for 4 min at 72°C. The 5' V primers were as follows: V1/2, 5' ACA CAG CTA TAC ATT GGT AC 3'; V2, 5' CGG CAA AAA ACA AAT CAA CAG 3'; V4, 5' TGT CCT TGC AAC CCC TAC CC 3'; V5, 5' TGT GCA CTG GTA CCA ACT GA 3'; V6, 5' GGA ATT CAA AAG AAA ACA TTG TCT 3'; V7, 5' AAG CTA GAG GGG TCC TCT GC 3'; V1, 5' ATT CAG AAG GCA ACA ATG AAA G 3'; V2, 5' AGT TCC CTG CAG ATC CAA GC 3'; V3, 5' TTC CTG GCT ATT GCC TCT GAC 3'; V4, 5' CCG CTT CTC TGT GAA CTT CC 3'; V5, 5' CAG ATC CTT CCA GTT CAT CC 3'; V6, 5' TCA AGT CCA TCA GCC TTG TC 3'; V7, 5' CGC AGA GCT GCA GTG TAA CT 3'; and V8, 5' AAG GAA GAT GGA CGA TTC AC 3' (9).

Expression of cytokine genes. C3H/HeJ and C3H/HeN mice were killed 3 days after i.p. inoculation with E. coli (10⁸ CFU/mouse). Extraction of total RNA from sorted T cells and cDNA synthesis were performed as described above. PEC were spread in plastic plates and incubated for 1 h in a CO₂ incubator at 37°C. After nonadherent cells were washed away with PBS, adherent cells were used in in vitro experiments. Of the adherent cells, >95% were macrophages, as assessed by morphological findings. Extraction of total RNA from macrophages and cDNA synthesis were performed as described above. Serial dilutions of total RNA were primed with 20 pmol of random primer in 21-µl reaction mixtures for reverse transcription. Synthesized cDNAs were amplified by PCR with primers derived from the murine cDNA. The specific primers were as follows: IL-2 sense, 5' TGA TGG ACC TAC AGG AGC TCC TGA G 3'; IL-2 antisense, 5' GAG TCA AAT CCA GAA CAT GCC GCA G 3'; IL-4 sense, 5' CGA AGA ACA CCA CAG AGAGTG AGC T 3'; IL-4 antisense, 5' GAC TCA TTC ATG GTG CAG CTT ATC G 3'; IL-6 sense, 5' TGG AGT CAC AGA AGG AGT GGC TAA G 3'; IL-6 antisense, 5' TCT GAC CAC AGT GAG GAA TGT CCA C 3'; IL-10 sense, 5' TAC CTG GTA GAA GTG ATG CC 3'; IL-10 antisense, 5' CAT CAT GTA TGC TTC TAT GC 3'; IL-12 sense, 5' GGA GAC CCT GCC CAT TGA ACT 3'; IL-12 antisense, 5' CAA CGT TGC ATC CTA GGA TCG 3'; IL-15 sense, 5' GTG ATG TTC ACC CCA GTT GC 3'; IL-15 antisense, 5' TCA CAT TCT TTG CAT CCA GA 3'; IFN- sense, 5' AGC GGC TGA CTG AAC TCA GAT TGT AG 3'; IFN- antisense, 5' GTC ACA GTT TTC AGC TGT ATA GGG 3'; TNF- sense, 5' GGC AGG TCT ACT TTG GAG TCA TTG C 3'; TNF- antisense, 5' ACA TTC GAG GCT CCA GTG AAT TCG G 3'; transforming growth factor (TGF-) sense, 5' CTT TAG GAA GGA CCT GGG TT 3'; and TGF- antisense, 5' CAG GAG CGC ACA ATC ATG TT 3'.

Statistical analysis. The statistical significance of the data was determined by the Student t test. A P value of less than 0.05 was taken as significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Kinetics of bacterial growth in organs after i.p. inoculation with E. coli. C3H/HeN and C3H/HeJ mice were inoculated i.p. with E. coli at a dose of 10⁸ CFU (one-fifth the LD₅₀)/mouse, and the kinetics of bacterial growth in the peritoneal cavity, liver, and spleen were examined. As shown in Fig. 1, the numbers of bacteria in the peritoneal cavity and spleen had decreased by day 3 of infection in both mouse strains. However, on day 2, the number of bacteria in C3H/HeN mice was significantly smaller than that in C3H/HeJ mice (P < 0.05). There was no difference in the bacterial number in the liver between the strains of mice at any stage after E. coli infection.

View larger version (13K): [in this window] [in a new window]   FIG. 1.   Kinetics of bacterial growth in the peritoneal cavities, livers, and spleens of C3H/HeJ and C3H/HeN mice after i.p. inoculation with 108 CFU of E. coli. Means and standard errors for five mice are shown. An asterisk indicates a significant difference from the value for C3H/HeN mice (P < 0.05).

Kinetics of T cells in the peritoneal cavity and liver after E. coli infection. To examine the cell influx in the peritoneal cavity and liver after E. coli infection, we analyzed the kinetics of PEC from C3H/HeJ and C3H/HeN mice inoculated i.p. with 10⁸ CFU of E. coli. The numbers of polymorphonuclear leukocytes (PMN) and macrophages in the peritoneal cavity were smaller in C3H/HeJ than in C3H/HeN mice on day 3 after E. coli infection (PMN, [2.5 ± 0.6] × 10⁶ versus [32.3 ± 5.5] × 10⁶; macrophages, [18.3 ± 3.6] × 10⁶ versus [87.5 ± 16.0] × 10⁶ [n = 5]). The absolute numbers of lymphocytes did not differ significantly for the two strains at any stage after E. coli infection (data not shown). Flow cytometry analysis of the expression of CD3 and TCR- was carried out with the nonadherent PEC and liver lymphocytes from both strains of mice on days 0, 1, 3, 5, and 7 of infection. A representative result from three independent experiments is shown in Fig. 2A. T cells in the nonadherent PEC of C3H/HeN mice increased, constituting more than 30% of all cells on day 3 after E. coli inoculation, whereas the percentage of T cells in the PEC of C3H/HeJ mice was only 4.0% at this stage of infection. No significant difference in the proportions of T cells in livers of C3H/HeJ and C3H/HeN mice was observed at any stage of infection (data not shown). The proportion of T cells in the peritoneal cavity changed little (from 10.9 to 11.6%) and that in the liver decreased (from 22.2 to 14.9%) after E. coli infection in C3H/HeN mice (data not shown). The kinetics of the absolute numbers of peritoneal and liver T cells after i.p. E. coli inoculation are shown in Fig. 2B. The absolute number of T cells in the peritoneal cavity was significantly increased 3 days after E. coli infection in C3H/HeN mice compared with C3H/HeJ mice (P < 0.05), while no significant difference in the numbers of T cells in the livers of C3H/HeN and C3H/HeJ mice was detected.

View larger version (24K): [in this window] [in a new window]   View larger version (13K): [in this window] [in a new window]   FIG. 2.   Kinetics of peritoneal T cells after i.p. E. coli inoculation. C3H/HeJ or C3H/HeN mice were inoculated with 108 CFU of E. coli (one-fifth the LD50) on day 0. (A) Non-plastic-adherent PEC were stained with anti-CD3 and anti-TCR- MAbs. The number in each panel indicates the percentage of T cells in whole nonadherent peritoneal cells. (B) Kinetics of the absolute numbers of peritoneal and liver T cells after i.p. inoculation with E. coli. , C3H/HeJ mice; , C3H/HeN mice. The number of T cells was calculated from the percentage of the cells among nonadherent PEC or liver lymphocytes. Means and standard errors for five mice are shown. Asterisks indicate significant differences from the values for C3H/HeJ mice (P < 0.05).

Proliferation and cytokine production of T cells in the peritoneal cavity induced by E. coli infection. To investigate the functions of T cells induced by E. coli infection, we first examined the expression of cytokine genes in freshly isolated T cells from C3H/HeN and C3H/HeJ mice infected with E. coli 3 days previously by cell sorting with an electric cell sorter. The T cells in the peritoneal cavities of E. coli-infected mice expressed high levels of mRNAs specific for IFN-, TNF-, and TGF- but not IL-2, IL-4, or IL-6 (Fig. 3).

View larger version (24K): [in this window] [in a new window]   FIG. 3.   Expression of cytokine mRNAs in peritoneal T cells sorted from C3H/HeJ and C3H/HeN mice infected with E. coli 3 days previously. T cells were sorted from nonadherent PEC pooled from five mice of each strain, and total RNA was reverse transcribed into cDNA and amplified by PCR. The results are representative of those from three independent experiments.

View larger version (13K): [in this window] [in a new window]   FIG. 4.   Proliferative response and cytokine production of T cells from the peritoneal cavities or livers of C3H/HeN mice in the presence of LPS. (A) Purified populations of cells were incubated (5 × 104/well) in anti-TCR- MAb-coated 96-well plates for 48 h in the presence or absence of MMC-treated spleen cells (3 × 104/ml), with or without 10 µg of LPS per ml. During the last 8 h of incubation, 1.0 µCi of [3H]thymidine per well was added. The cells were then harvested, and the amount of [3H]thymidine incorporated was determined by scintillation counting. The data are representative of those from two separate experiments and are expressed as the means of triplicates ± standard deviations. Asterisks indicate significant differences from the values for the control (P < 0.05). (B) Purified T cells (5 × 104 cells) were cultured similarly in the presence or absence of MMC-treated spleen cells with or without LPS for 24 h at 37°C, and the culture supernatant was collected. The cytokine activity in the culture supernatant was tested for the presence of IFN- by ELISA. The data are representative of two separate experiments and are expressed as the means of triplicates ± standard deviations. Asterisks indicate significant differences from the values for the control (P < 0.05).

V and V gene expression in T cells in the peritoneal cavity and liver in mice infected with E. coli. To examine the V gene expression of the T cells in the peritoneal cavity and liver in C3H/HeN mice following E. coli infection, total RNA was extracted from T cells sorted from nonadherent PEC and livers of mice inoculated with E. coli 3 days previously, and V gene expression was analyzed by reverse transcription-PCR. As shown in Fig. 5, the T cells in PEC expressed the V6 and V1 genes preferentially, whereas the T cells in the livers of C3H/HeN mice expressed V1/2, V4, and a diversity of V genes. We further examined the V gene expression of the T cells in the peritoneal cavity after stimulation with LPS in vitro. T cells expressing the V6 and V1 genes were enriched after stimulation with LPS (Fig. 5).

View larger version (50K): [in this window] [in a new window]   FIG. 5.   V or V usages of T cells in PEC and liver on day 3 after E. coli infection. Total RNA extracted from T cells (5 × 104 cells) sorted from five C3H/HeN mice infected with E. coli 3 days previously or from T cells stimulated with LPS as described in the legend to Fig. 4 was reverse transcribed into cDNA and amplified by PCR with primers for C or C and various V or V segments, respectively. The Southern blot of  PCR products was hybridized with MNG6. The Southern blot of  PCR products was hybridized with an oligonucleotide probe for J1 or J2. The results are representative of those from three independent experiments.

Effects of T-cell-depletion on the eradication of bacteria in mice infected with E. coli. To confirm a protective role for T cells in E. coli infection, TCR-^/ mice with a C57BL/6 background were infected with E. coli (10⁸ CFU/mouse) and sacrificed on day 3. Control (TCR-^+/) mice showed an increase in T cells on day 3 after E. coli infection (data not shown). As shown in Fig. 6A, a significant increase in the number of E. coli cells was detected in the peritoneal cavities, livers, and spleens of TCR-^/ mice compared with control mice. The numbers of PMN, macrophages, and lymphocytes in the peritoneal cavities of TCR-^/ mice were smaller than those for control mice (Fig. 6B). These results suggest that an increase of T cells in the peritoneal cavity is one of the factors responsible for the eradication of E. coli.

View larger version (27K): [in this window] [in a new window]   FIG. 6.   Bacterial growth in the peritoneal cavities and spleens of TCR-/ mice after E. coli infection. TCR-/ mice and their littermate control mice were inoculated i.p. with 2 × 108 CFU of E. coli on day 0. (A) The numbers of E. coli CFU recovered from peritoneal cavities and spleens of infected mice on day 3 were determined by colony formation assay on tryptic soy agar. Values are means ± standard deviations for groups of five mice. Asterisks indicate significant differences from the values for control mice (P < 0.05). (B) Populations of PEC obtained from TCR-/ mice and control mice on day 3 after i.p. inoculation with E. coli. PMN, macrophages, and lymphocytes were judged by morphologic characteristics after staining with Giemsa solution. Values are means ± standard deviations for groups of five mice.

Involvement of IL-15 in the appearance of T cells after infection with E. coli. The T cells that appear during the course of infection with E. coli produce IFN-, but not IL-2, after LPS stimulation. We have previously shown that IL-15 produced by the macrophages is involved in the stimulation of the T cells during salmonellosis (39). We therefore examined whether IL-15 produced by the infected macrophages is involved in the stimulation of the T cells during E. coli infection. To determine whether IL-15 was induced in macrophages after E. coli infection, we tried to detect the IL-15 mRNA in the peritoneal macrophages of C3H/HeJ and C3H/HeN mice 3 days after E. coli infection. The levels of expression of the IL-15 gene in macrophages after E. coli infection are presented in Fig. 7. Consistent with previous reports (11, 36), macrophages of C3H/HeN mice expressed TNF- and IL-6 more abundantly after E. coli infection than did those of C3H/HeJ mice. The level of expression of IL-15 mRNA, especially the longer message, which is translated most efficiently (40), in macrophages after infection with E. coli was higher in C3H/HeN mice than in C3H/HeJ mice.

View larger version (24K): [in this window] [in a new window]   FIG. 7.   Expression of monokine mRNAs in peritoneal macrophages of mice infected with E. coli. The peritoneal macrophages were obtained from five C3H/HeJ or C3H/HeN mice 3 days after inoculation with E. coli (108 CFU/mouse). Total RNA extracted from the pooled macrophages was reverse transcribed into cDNA and amplified by PCR with each cytokine-specific primer. The results are representative of those from three independent experiments.

View larger version (28K): [in this window] [in a new window]   FIG. 8.   Effects of in vivo administration of anti-IL-15 MAb on recovery of bacteria from the peritoneal cavity and spleen and appearance of T cells in the peritoneal cavity after E. coli infection. C3H/HeN mice were inoculated i.p. with 200 µg of anti IL-15 MAb, or control Ab was injected i.p. 2 h before challenge with 2 × 108 CFU of E. coli. (A) Non-plastic-adherent PEC from infected mice on day 3 were stained with anti-TCR- and anti-TCR- MAbs. (B) The numbers of E. coli CFU recovered from the peritoneal cavities, livers, and spleens of infected mice on day 3 were determined by colony formation assay on tryptic soy agar. Values are means ± standard errors for groups of three mice. Asterisks indicate significant differences from the values for control mice (P < 0.05).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

It is generally accepted that the host defense against infection with E. coli, an extracellular bacterium, is almost exclusively dependent on neutrophils and Ab (58). We show here the possibility that T cells are also involved in the host defense against E. coli infection. T-cell numbers were remarkably increased in the peritoneal cavity after an i.p. infection with E. coli in LPS-responsive C3H/HeN mice but not in LPS-hyporesponsive C3H/HeJ mice. The T cells appearing in the peritoneal cavity after i.p. infection with E. coli produced a large amount of IFN- in the presence of LPS. Mice depleted of T cells by TCR- gene mutation showed an impaired host defense against E. coli. These results suggest that LPS-stimulated T cells help to protect against E. coli infection.

Similar to T cells, T cells secrete various cytokines and express cytolytic functions (27). Most T cells appearing after infection with intracellular bacteria are reported to produce Th1-type cytokines, in particular IFN- (2, 11, 12, 33, 57), whereas T cells during infection with a helminth, Nippostrongylus brasiliensis, preferentially produced Th2-type cytokines, mostly IL-4 (11). Furthermore, T cells, especially in the epithelium, produce TGF- for immunoregulation and/or immunoglobulin A production (7, 13, 53). We have previously observed, with mice depleted of T cells by treatment with anti-TCR- MAb, that T cells contribute to defense early after Listeria infection via IFN- production (17). Mice with mutated TCR- genes showed impaired TNF- production in response to LPS (38). Our results reveal that the T cells accumulating in the peritoneal cavity after E. coli infection expressed IFN- mRNA and produced a significant amount of IFN- in the presence of LPS under TCR triggering. We speculate that T cells produce IFN- in response to E. coli and their cell component LPS and activate macrophages which consequently eliminate the E. coli. We have previously reported that T cells appearing during the course of listeriosis produced macrophage chemotactic factor, in addition to IFN- (17). Our present results reveal that accumulation of PMN and macrophages in the peritoneal cavity following E. coli infection was impaired in TCR^/ mice. Therefore, it is also possible that T cells produce chemokines for neutrophils and monocytes in addition to IFN- and protect mice from E. coli infection. Although E. coli was completely eliminated in the peritoneal cavity by day 3,  T cells were still increased on days 5 and 7 after infection. Mukasa et al. have recently suggested that T cells expressing invariant V6 and V1 chains have a function in controlling influences on the host inflammatory responses (35). Our data indicated that the T cells also expressed mRNA specific for TGF-, which can modulate immune and inflammatory responses. T cells appearing after E. coli infection may have different functions at distinct time points during the course of E. coli infection. Further analysis of cytokine production by T cells is required to clarify their roles in E. coli infection.

It has been described that a significant number of T cells are stimulated by LPS through an apparently TCR-independent mechanism (30, 41, 51). Leclercq and Plum have reported that TCR V5 cells, which are preferentially present in the epidermis, are activated to produce cytokines upon interaction with LPS via TCR-independent pathways (30). Nitta et al. have reported that T cells in the peritoneal cavity proliferate in response to a TCR triggering in synergy with LPS (41). Similarly, we found that the T cells in the peritoneal cavity of E. coli-infected mice proliferate and produce a significant amount of IFN- in the presence of LPS when their TCRs are stimulated with anti-TCR MAb. Stimulation of the T cells by LPS was indeed accessory cell independent, excluding the possibility that LPS induced expression of ligands for TCR on accessory cells or production of growth factors from accessory cells. Thus, it appears that LPS may have a costimulatory activity for T-cell stimulation upon TCR triggering. The peritoneal T cells induced by E. coli infection expressed the V6 gene, which is expressed by T cells in the uterus and tongue, together with the V1 gene, rearranged to J2, similar to V5 T cells in the epidermis (22). All V6-J1 and V1-J2 mRNAs from the T cells we sequenced have no junctional diversity, similar to those from the T cells in the fetal thymus and uterus. On the other hand, the liver T cells expressed V1/2 and did not respond to LPS. Thus, it would appear that only T cells with particular V genes such as V5 and V1 or V6 and V1 are stimulated with LPS from gram-negative bacteria through a TCR-independent mechanism. Further analysis is required to clarify which receptor of the T cells recognizes LPS.

Although LPS from E. coli is apparently involved in T-cell stimulation, the ligand for the TCR during E. coli infection is not known. It has been reported that V6 and V1 T cells expand at sites of inflammation in the absence of pathogen-derived antigens in Listeria infection and in Listeria-induced autoimmune orichitis (35, 44). This suggests that V6 and V1 T cells do not respond to foreign antigens but rather respond to a host-derived antigen that is conserved between the host and bacteria. In mice, a high proportion of T cells have been found to respond to unique peptides of mycobacterial and mammalian HSP65 (4). The HSP65-reactive T cells characteristically express V1 and V6 with junctional diversity (42). We have previously reported that the peritoneal T cells appearing during infection with intracellular bacteria such as S. choleraesuis (9), L. monocytogenes (17), and Mycobacterium bovis BCG (19) preferentially expressed V1 and V6 genes. However, the present study revealed that the T cells appearing in E. coli infection preferentially expressed V6 and V1 and thus differ from those capable of responding to HSP65. In fact, Takada et al. have reported that T cells from E. coli-infected mice did not proliferate in response to purified protein derivative or HSP65 derived from M. tuberculosis (55). A number of murine T-cell clones are reported to recognize major histocompatibility complex molecules or major histocompatibility complex-related gene products such as TL and Qa in a manner quite different from the antigen recognition shown by T cells (3, 21, 31, 48). Similarly, a herpesvirus protein was found to directly stimulate T cells independent of antigen processing and presentation (25, 50). Human T cells are stimulated by apparently nonproteinaceous low-molecular-weight ligands, including isopentenyl pyrophosphate, which represents a ubiquitous metabolite of various vitamins, lipids, and steroids in both prokaryotic and eukaryotic cells (8, 49, 56). Therefore, it is of interest to elucidate whether the T cells induced by E. coli recognize such unique antigens in a manner different from that of T cells.

Another notable finding is that IL-15 was involved in protection against E. coli infection. LPS-hyporesponsive C3H/HeJ mice carry the lps^d mutation on chromosome 4, and macrophages and B cells in these mice respond poorly to LPS (11, 36). Consistently, the macrophages induced by E. coli infection in C3H/HeJ mice showed an impaired expression of monokine genes such as those for TNF- and IL-6 compared with that in C3H/HeN mice. In correlation to the sensitivity of macrophages to LPS, T-cell numbers in the peritoneal cavity were remarkably increased after E. coli infection in LPS-responsive C3H/HeN mice but not in LPS-hyporesponsive C3H/HeJ mice. IL-15 promoter regions contained binding elements for LPS-inducible transcription factors such as NF-IL-6 and NF-B (60). Consistent with this finding, C3H/HeN mice infected with E. coli expressed higher levels of IL-15 mRNA, especially the longer transcript, than did C3H/HeJ mice infected with E. coli. We have recently found that IL-15 mRNA containing a longer alternative exon 5 is translated most efficiently among IL-15 mRNA isoforms (40). Administration of anti-IL-15 MAb inhibited, albeit partially, the increase in T cells after E. coli infection and impaired the host defense against E. coli. These results suggest that IL-15 is at least partly responsible for the increase in T cells in the peritoneal cavity after E. coli infection. T cells induced by E. coli infection did not produce IL-2, nor did IL-2-producing T cells appear in the peritoneal cavity after infection (data not shown). Thus, we speculate that IL-15 released from LPS-stimulated macrophages is responsible for the increase in T cells and consequently for the defense against E. coli infection in mice. We have previously reported that T cells proliferate to produce IFN- in response to IL-15 in vitro (18, 39). Therefore, IL-15 derived from LPS-stimulated macrophages may be responsible for local expansion of T cells preexisting in the peritoneal cavity in vivo. However, the V6/V1 T-cell subset is rare in the peritoneal cavity. T cells are stimulated with LPS in the absence of accessory cells. Our preliminary experiments revealed that addition of anti-IL-15 MAb in in vitro culture did not inhibit the T-cell proliferation in the presence of LPS. Taken together, IL-15 may not play a very important role in LPS-induced proliferation of T cells in situ. IL-15 is reported to have a strong chemotactic activity for T cells (32, 37). Therefore, we speculate that IL-15 released from LPS-stimulated macrophages may be more important in accumulation of T cells in the peritoneal cavity than in the expansion in vivo after E. coli infection. Anti-IL-15 administration only partially inhibited the appearance of the T cells. Skeen and Ziegler reported that peritoneal T cells proliferated in response to IL-1 and IL-7 (52). It has been reported that TNF- and IL-12 synergistically stimulate human T-cell proliferation (59). Although it cannot be ruled out that the amount of anti-IL-15 MAb was insufficient to cause inhibition in our experiments, it appears that the increase in T cells may be attributable in part to cytokines other than IL-15.

In conclusion, LPS-stimulated T cells play important roles in the host defense against E. coli infection. IL-15 released from LPS-stimulated macrophages may be involved in the accumulation of T cells during E. coli infection.