Early Programming of T Cell Populations Responding to Bacterial Infection

Roberto Mercado, Sujata Vijh, S. Elise Allen, Kristen Kerksiek, Ingrid M. Pilip and Eric G. Pamer
J Immunol December 15, 2000, 165 (12) 6833-6839; DOI: https://doi.org/10.4049/jimmunol.165.12.6833

Abstract

The duration of infection and the quantity of Ag presented in vivo are commonly assumed to influence, if not determine, the magnitude of T cell responses. Although the cessation of in vivo T cell expansion coincides with bacterial clearance in mice infected with Listeria monocytogenes, closer analysis suggests that control of T cell expansion and contraction is more complex. In this report, we show that the magnitude and kinetics of Ag-specific T cell responses are determined during the first day of bacterial infection. Expansion of Ag-specific T lymphocyte populations and generation of T cell memory are independent of the duration and severity of in vivo bacterial infection. Our studies indicate that the Ag-specific T cell response to L. monocytogenes is programmed before the peak of the innate inflammatory response and in vivo bacterial replication.

L;-2qisteriamonocytogenes is a Gram-positive bacterium that causes severe disease in immunocompromised and pregnant individuals (1). A murine model of systemic L. monocytogenes infection has been used to dissect innate and adaptive immune responses to intracellular bacterial infection (2). Neutrophils and NK cells play a critical role in early control of infection (3, 4, 5, 6), while CD4 and CD8 T cells are implicated in clearance of infection and development of long-term immunity (7, 8). In addition, Abs specific for L. monocytogenes have recently been shown to influence the kinetics of in vivo infection (9, 10).

Upon cellular infection, L. monocytogenes escapes the phagosomal vacuole, entering the host cell cytosol where it divides and locomotes by polymerizing actin (11). Bacterial proteins secreted by L. monocytogenes into the host cell cytosol are degraded into peptides for presentation by MHC class I molecules to CD8 T lymphocytes (12, 13). Several L. monocytogenes-derived peptide epitopes that stimulate CD8 T cell responses during murine infection have been identified (14). The expansion and contraction of CD8 T cell responses specific for four different L. monocytogenes-derived epitopes were recently shown to be remarkably synchronous (15). Coordinate regulation of T cell responses to these different epitopes was surprising; previous studies had shown that the epitopes are present in infected cells in remarkably different amounts and have very different binding stabilities with the MHC class I molecule H2-Kd (12). This result suggested that the duration of T cell expansion following L. monocytogenes infection is not determined by the duration of in vivo Ag presentation.

The factors that determine the magnitude of T cell responses following viral or bacterial infection remain incompletely defined. A comfortable but unproven assumption is that the magnitude of T cell responses is determined by the amount of Ag and the duration of its presentation. Ample evidence, however, suggests that Ag quantity and stability do not determine the magnitude of T cell responses (15, 16, 17, 18). In the L. monocytogenes model, the immunodominant T cell response is specific for a peptide that is far less prevalent than two other peptides that elicit substantially smaller T cell responses (16, 19). Similarly, the correlation between Ag quantity and the size of the T cell response has been broken in studies of the murine T cell response to influenza virus infection (17).

In this report, we investigate the effect of severity and duration of in vivo infection on expansion of Ag-specific CD8 T cells specific for L. monocytogenes. We found that the rate and duration of in vivo T cell proliferation following priming is independent of the duration of infection or the quantity of Ag present in vivo. Our results suggest that pathogen-specific T lymphocytes are programmed during the first day of infection and subsequently undergo proliferation and differentiation into effector T cells without further calibration by the progressing inflammatory response.

Materials and Methods

Mice and bacteria

BALB/c mice were obtained from The Jackson Laboratory (Bar Harbor, ME). BALB/c Thy1.1-congenic mice were provided by Charles Surh (The Scripps Research Foundation, La Jolla, CA). L. monocytogenes strain 10403S was provided by Daniel Portnoy (University of California, Berkeley, CA).

Flow cytometry and tetramer generation

H2-Kd tetramers complexed with listeriolysin O (LLO)591–99 and p60449–457 were generated as previously described (15). Epitope-specific T cell populations were detected with PE-conjugated, tetrameric MHC-peptide complexes and concurrently stained for other surface molecules using directly conjugated mAbs as described previously (15). Briefly, after blocking with unconjugated streptavidin (0.5 mg/ml; Molecular Probes, Eugene, OR) and Fc block (PharMingen, San Diego, CA), ∼1–2 × 106 splenocytes were incubated in FACS staining buffer (PBS (pH 7.45), 0.5% BSA, and 0.02% sodium azide) for 1 h on ice in the presence of saturating concentrations of tetramer reagents (0.25–0.5 mg/ml) and the various mAbs. Cells were subsequently washed three times in staining buffer and then fixed in 1% paraformaldehyde/PBS (pH 7.45). Flow cytometry was performed using a FACScalibur, and data were further analyzed with CellQuest software (Becton Dickinson, Mountain View, CA). In some cases (see Fig. 7⇓), dead cells were excluded by staining with ethidium monoazide bromide (Molecular Probes) at a concentration of 1.25 mg/ml, added during staining, with exposure to light during the last 10 min of staining. The following mAbs were used (all obtained from PharMingen): Cy-Chrome-conjugated anti-CD8α (clone 53-6.7), FITC-conjugated anti-CD62L (clone MEL-14), and FITC-conjugated anti-CD4 (clone H129.19).

IFN-γ and TNF assays

Spleens were removed from naive BALB/c mice, L. monocytogenes-infected mice that had received ampicillin for 48 h beginning 1 day after infection and infected mice that were not treated with antibiotics. Spleens were macerated into ice-cold PBS containing 0.01% Triton X-100 and centrifuged at 10,000 × g for 10 min. The amounts of murine IFN-γ and TNF in the spleen supernatants were determined by sandwich ELISA using OptEIA kits from PharMingen, following the manufacturer’s protocols.

CTL assays

CD8 T cells were enriched from spleens by negative selection using magnetically activated cell sorting (MACS, Miltenyi, Germany) with anti-rat IgG microbeads and anti-CD4 (GK1.5) and anti-MHC class II (TIB120). Target cells were labeled with 51Cr, coated with 10−6 M LLO91–99, and incubated in the presence of enriched CD8 T cells at an E:T ratio of 33:1. After 4 h of incubation, the released 51Cr was quantified and expressed as the percent specific lysis, as previously described (20).

Generation of TCR-transgenic mice

VDJ regions of the TCR β- and α-chain genes were PCR amplified from genomic DNA of CTL clone WP11.12, which is specific for the L. monocytogenes- derived, subdominant epitope p60449–457 (14). The β-chain VDJ region was amplified with plaque-forming units of polymerase (Stratagene, La Jolla, CA) using the sense Vβ8.2 primer 5′-CGCTCGAGGGAAGCATGGGCTCCAGGCTCTTCTTCGTG-3′ and the antisense Jβ1.1 primer 5′-TCTTATCGATTGCACATCAGAATACAGATACTCG-3′. The resulting fragment was digested with XhoI and ClaI and cloned into the TCR β cassette (44). The TCR α-chain VJ region was amplified using the sense Vα8 primer 5′-CGCCCGGGATGAAC-ATGCGTCCTGTCACCTGCTCAGTT-3′ and the antisense Jα20 primer 5′-GCGGCGGCCGTCCCTACTCTGTCCTTATGAG-3′ that include XmaI and NotI restriction sites, respectively. This fragment was digested with XmaI and NotI and cloned into the TCR α cassette. Plasmids were screened by restriction mapping for the presence of the complete TCR β or α gene. The TCR β gene insert was separated from the prokaryotic fragment of the vector by digestion with KpnI and the TCR α gene insert by digestion with SalI. These DNA fragments were purified and comicroinjected into fertilized oocytes of (C57BL6 × SJL)F1 mice at the Yale transgenic mouse facility. Founders were identified by PCR screening of DNA purified from mouse tails and were bred for 10 generations onto the BALB/cJ background.

Adoptive transfer of CFSE-labeled splenocytes

Splenocytes were isolated from WP11.12 TCR-transgenic mice and 1 × 107 cells were labeled in 1 ml of 5 μM CFSE in PBS at 37°C for 10 min. Cells were washed with RPMI 1640/10% FCS and resuspended in PBS for i.v. inoculation into recipient mice. Recipients were inoculated with 30 million CFSE-labeled splenocytes and infected 1 day later with 2000 L. monocytogenes or left uninfected.

Results

In previous studies of the CD8 T cell response to L. monocytogenes infection, we found that the efficiency of Ag presentation could vary over a 5-fold range without affecting the size of the T cell response (21). This result suggests that the number of MHC class I-restricted epitopes presented in vivo does not determine the magnitude of the T cell response. To obtain further evidence for this concept, we infected mice with different doses of L. monocytogenes and measured the peak CD8 T cell response to the immunodominant epitope LLO91–99 using MHC class I tetramers. The size of the LLO91–99- specific T cell response was similar in mice infected with either 1,000, 5,000 or 25,000 live L. monocytogenes (Fig. 1⇓). This result is striking because the severity of infection and the amount of Ag present in vivo are markedly different in these settings (22). This result suggests that neither the amount of Ag presented in vivo nor the severity of the inflammatory response induced by these infections significantly influences the rate at which LLO91–99-specific T lymphocytes expand in vivo.

FIGURE 1.
FIGURE 1.

The magnitude of the LLO91–99-specific T cell response is similar in mice receiving different doses of L. monocytogenes. BALB/c mice were infected i.v. with 1,000 (upper panel), 5,000 (middle panel), or 25,000 (lower panel) L. monocytogenes and 7 days later splenocytes were isolated and stained with mAbs specific for CD8α and CD62L and H2-Kd tetramers complexed with LLO91–99. Dot plots show tetramer staining on the vertical axis and CD62L staining on the horizontal axis for CD8-gated T lymphocytes. The percentage of CD8 T cells that are tetramer positive and CD62Llow is indicated in the left upper quadrant of each plot. Each plot is representative of three mice infected with the different doses of L. monocytogenes.

To investigate the role of the duration of bacterial infection upon the activation and expansion of L. monocytogenes-specific CD8 T cells, we used antibiotics to terminate in vivo infection at various times following inoculation. Ampicillin is a bacteriocidal antibiotic that impairs bacterial cell wall synthesis. Addition of 2 mg/ml ampicillin to the drinking water of mice eliminated viable bacteria from spleens of L. monocytogenes-infected mice within 12 h of antibiotic initiation (Fig. 2⇓A). Cultures of liver also demonstrated complete elimination of viable bacteria (results not shown). In contrast, untreated mice had viable L. monocytogenes isolated from the spleen and liver for up to 7 days following infection. Thus, ampicillin therapy rapidly eliminates bacterial infection and dramatically alters the in vivo Ag load. To determine whether the inflammatory response was also attenuated in antibiotic-treated mice, we measured the amounts of TNF and IFN-γ in spleens of mice that were treated for 48 h with ampicillin or left untreated. As shown in Fig. 2⇓, B and C, 48 h after initiation of antibiotics (3 days following bacterial inoculation), the amounts of TNF and IFN-γ in the spleen are indistinguishable from uninfected control mice. In contrast, L. monocytogenes-infected untreated mice have elevated levels of these two cytokines. This result indicates that early antibiotic treatment of mice infected with L. monocytogenes down-regulates the in vivo inflammatory response in addition to eliminating viable bacteria.

FIGURE 2.
FIGURE 2.

Oral ampicillin therapy rapidly eliminates viable L. monocytogenes and inflammatory cytokines from infected mice. A, BALB/c mice were infected i.v. with 2000 live L. monocytogenes 10403s and divided into two groups: the first did not receive antibiotics (○) and the second group received drinking water supplemented with 2 mg/ml ampicillin, beginning 24 h after bacterial inoculation (•). Spleens were harvested from three mice per group at the indicated time points, homogenized, and the number of viable L. monocytogenes was determined by plating on brain-heart infusion plates. The mean number of bacteria per spleen at the indicated time points was determined and plotted. B, BALB/c mice were infected with L. monocytogenes and either left untreated or treated with ampicillin 24 h after infection. Seventy-two hours after infection, spleens were harvested and the amount of IFN-γ was determined by ELISA as described in Materials and Methods. Cytokine levels were also determined for spleens from uninfected mice. The plotted values represent the means for three mice and the SD is indicated. C, The level of TNF was determined in spleens as described in B.

We next measured CD8 T cell responses to the dominant H2-Kd-restricted LLO91–99 epitope 7 days after primary infection of mice with a sublethal dose of L. monocytogenes (Fig. 3⇓). Mice were divided into four groups that either received no antibiotics or were treated with ampicillin at 0, 12, or 24 h after bacterial inoculation, and the LLO91–99-specific T cell response was measured 7 days after infection. Mice that received antibiotics immediately following L. monocytogenes inoculation had essentially undetectable LLO91–99-specific T cell responses. Mice that received ampicillin 12 h after infection mounted LLO91–99-specific T cell responses of intermediate size, whereas mice that received ampicillin 24 h after infection had nearly normal LLO91–99-specific T cell responses when compared with untreated mice. These results indicate that the magnitude of the LLO91–99-specific response correlates with the duration of in vivo bacterial growth during the first 24 h after infection.

FIGURE 3.
FIGURE 3.

The magnitude of the LLO91–99-specific T cell response is similar in mice undergoing unrestricted or antibiotic-attenuated L. monocytogenes infection. BALB/c mice were infected with 2000 L. monocytogenes and either treated with ampicillin, as described in the legend to Fig. 2⇑, immediately (0 h), 12 h, or 24 h after inoculation or left untreated (upper panel). Seven days after bacterial inoculation, splenocytes were prepared and stained with mAbs specific for CD8α and CD62L and with H2-Kd tetramers complexed with LLO91–99. The dot plots show tetramer staining on the vertical axis and CD62L staining on the horizontal axis of live CD8-gated T cells. The percentage of activated, LLO91–99-specific T cells is indicated in the left upper quadrant of each plot. Each plot is representative of three mice in each treatment group.

To determine whether the kinetics of T cell expansion are altered in mice in which bacterial growth and the inflammatory response have been attenuated by antibiotic therapy, we measured the LLO91–99-specific T cell response in the presence and absence of antibiotics at various times after infection (Fig. 4⇓A). Although the absolute number of LLO91–99-specific T cells is lower when ampicillin is administered 24 h after L. monocytogenes inoculation, reflecting the smaller spleen size in antibiotic-treated mice, the peak frequency of the Ag-specific T cell response occurs 7–9 days after infection in both cases. This finding indicates that the kinetics of L. monocytogenes-specific T cell expansion and contraction is neither dictated by the duration of infection nor by the associated inflammatory response following the first day of bacterial infection.

FIGURE 4.
FIGURE 4.

Early attenuation of bacterial infection does not decrease the duration of T cell expansion or the acquisition of effector function. A, The number of LLO91–99-specific CD8 T cells was determined in BALB/c mice that were untreated (□) or treated with ampicillin 24 h after L. monocytogenes inoculation (•). Spleens were harvested from three mice per group at 3, 7, 9, and 14 days after infection, stained, and analyzed flow cytometrically as described in B, and the mean number of LLO91–99-specific cells per spleen for each time point is plotted. B, BALB/c mice were infected with L. monocytogenes and either treated with ampicillin 24 h after infection or left untreated. CD8 T cells were enriched from splenocytes 7 days after infection and assayed for cytolytic activity against 51Cr-labeled P815 target cells. The percent specific lysis in the presence of LLO91–99 (▪) or in the absence (□) is plotted at an E:T ratio of 33:1.

Ag-specific T lymphocytes measured by tetramer staining 7 days after bacterial infection are predominantly effector T cells, as determined by the rapid induction of IFN-γ production following Ag exposure and direct ex vivo cytolytic activity (15, 16). To test whether LLO91–99-specific CD8 T cells in mice following early termination of infection have acquired effector function, we assayed enriched CD8 T cells isolated from ampicillin-treated and untreated mice for direct cytolytic activity on LLO91–99-coated target cells. Fig. 4⇑B demonstrates that the ex vivo cytolytic activity toward LLO91–99-coated target cells was similar in mice that had undergone attenuated infection and mice that underwent the normal course of infection. Thus, prolonged infection and inflammation are not necessary for L. monocytogenes-specific T cells to acquire effector function.

To determine whether memory T cell generation was affected by early termination of bacterial infection, we measured LLO91–99-specific T cell populations 5 days after reinfection of mice that had been immunized 1 mo earlier with 2000 L. monocytogenes followed by either no antibiotic therapy or oral ampicillin at 0, 12, or 24 h after bacterial inoculation (Fig. 5⇓). All four groups of mice survived a lethal challenge of 100,000 L. monocytogenes, demonstrating that even mice immediately treated with antibiotics following infection had acquired protective immunity. The memory CD8 T cell response to LLO91–99 was identical in mice that received either no antibiotics or ampicillin 12–24 h after infection (Fig. 5⇓), indicating that progressive bacterial infection during the CD8 T cell expansionary phase is not required for memory generation. As might be expected, the markedly diminished primary response to LLO91–99 in the setting of immediate antibiotic therapy following infection resulted in an attenuated memory response to this epitope.

FIGURE 5.
FIGURE 5.

Memory CD8 T cell responses are not diminished by early attenuation of bacterial infection. BALB/c mice were infected and treated with ampicillin as described in Fig. 3⇑, and reinfected with 100,000 live L. monocytogenes 35 days later. Five days after reinfection, splenocytes were stained with mAbs specific for CD8 and CD62L and with H2-Kd/LLO91–99 tetramers. The percentage of CD8 T cells that are activated and LLO91–99 specific is plotted for the four different treatment groups.

The finding that the frequency of LLO91–99-specific CD8 T cells is similar in untreated mice and those treated with ampicillin 24 h after infection was surprising since the amount of Ag that is present in vivo in these two circumstances is markedly different. To increase the range of bacterial inocula that can be used to infect mice, since the LD50 of BALB/c mice for L. monocytogenes is ∼5 × 104, we treated infected mice with ampicillin 24 h after inoculation with either 10,000, 100,000, or 1,000,000 live L. monocytogenes (Fig. 6⇓). Remarkably, the magnitude of the LLO91–99-specific T cell responses was similar even though the doses of Ag differed by a factor of 100. Thus, in agreement with the findings presented in Fig. 1⇑, exceeding the threshold for complete activation of available naive T cells does not result in further increases in the magnitude of the T cell response. Previous studies from our laboratory have shown that increasing the presentation of epitope above the threshold required for T cell priming does not enhance T cell responses (21). At the extreme end, it has been shown that presentation of massive amounts of Ag induces clonal T cell exhaustion (23). The amount of Ag that is presented during murine infection with L. monocytogenes most likely does not approach the amount that is required to induce clonal exhaustion. Our results show that the size of the T cell response is not significantly different following inoculation with a nearly 1000-fold dose range of viable bacteria (see Figs. 1⇑ and 6⇓), demonstrating a remarkable degree of independence between T cell responses and in vivo Ag presentation and infection-induced inflammation.

FIGURE 6.
FIGURE 6.

Infection of BALB/c mice with a 500-fold range of L. monocytogenes dosages result in similar LLO91–99-specific T cell responses. BALB/c mice were infected i.v. with 2,000, 10,000, 100,000, or 1,000,000 live bacteria and treated with ampicillin 24 h later. Seven days after bacterial inoculation, splenocytes were stained and analyzed by flow cytometry. The percentage of CD8 T cells specific for LLO91–99 is indicated in the left upper quadrant of each plot.

Although the magnitude of the LLO91–99-specific T cell response was maximal when bacterial replication was terminated 24 h after infection, earlier administration of antibiotics resulted in significantly smaller T cell responses (Fig. 3⇑). This finding suggests that, during the first 24 h of infection, the magnitude of the primary T cell response to L. monocytogenes infection correlates with the duration of in vivo bacterial replication. One explanation for this finding is that the number of naive T cells that are primed during the first day of infection is time dependent. This would suggest that ∼24 h of viable bacterial infection are required for most naive T cells to come into contact with APCs. An alternative explanation for differing T cell response magnitudes in the setting of early antibiotic therapy is that the rate or extent of T cell proliferation differs. Thus, T cell populations primed by very short exposure to Ag, such as when antibiotic administration begins 12 h after bacterial inoculation, may undergo fewer in vivo divisions than T cells exposed to 24 h of infection. To differentiate between these two possibilities, we generated transgenic mice expressing the TCR for the L. monocytogenes-derived, H2-Kd-restricted p60449–457 epitope. In contrast to LLO91–99, p60449–457 is a subdominant epitope which elicits a nearly undetectable T cell response following primary L. monocytogenes infection (15, 16). However, p60449–457 is generated with great efficiency and is the most prevalent epitope presented by H2-Kd in L. monocytogenes-infected cells (19). Although p60449–457 is highly prevalent in cells infected with live bacteria, it is rapidly lost from infected cells because the epitope dissociates from H2-Kd with a half-life of ∼1 h, whereas LLO91–99 has a half-life that exceeds 6 h (24). Thus, in the presence of antibiotics that inhibit L. monocytogenes, p60449–457 is rapidly lost (24). Fig. 7⇓ demonstrates that TCR-transgenic mice contain a normal proportion of CD8 T cells (Fig. 7⇓, upper panel) but a slightly diminished proportion of CD4 T cells. Staining with H2-Kd/p60449–457 tetramers shows that in excess of 90% of CD8 T cells from these mice stain with the p60449–457 tetramers (Fig. 7⇓, lower panel), and that most of the Ag-specific cells express high levels of CD62L.

FIGURE 7.
FIGURE 7.

CD8 T cells from WP TCR-transgenic mice are specific for p60449–457. Upper panel, Splenocytes from a WP TCR-transgenic mouse, backcrossed for 10 generations onto the BALB/c background, were stained with mAbs specific for CD8α and CD4. The dot plot demonstrates CD8 staining of the vertical axis and CD4 staining on the horizontal axis of live cells. The percentages of positively staining cells is indicated in the quadrants. Lower panel, Splenocytes from a transgenic mouse were stained with mAbs specific for CD8 and CD62L and with H2-Kd tetramers complexed with p60449–457. The dot plot shows tetramer staining on the vertical axis and CD62L staining on the horizontal axis for live CD8-gated T cells. The percentages of tetramer staining CD8 T cells is indicated in the upper quadrants.

To measure T cell proliferation in mice undergoing normal or antibiotic attenuated bacterial infection, transgenic CD8 T cells were stained with CFSE, a fluorescent dye that is incrementally diluted with each cell division (25), and injected into naive BALB/c recipient mice. Fig. 8⇓ demonstrates that transgenic T cells transferred into mice were readily detectable and maintained their high CFSE content 8 days later (Fig. 8⇓A). Much larger p60449–457-specific CD8 T cell populations were detected in recipients of transgenic cells that were subsequently infected with L. monocytogenes (Fig. 8⇓B). These expanded p60449–457 T cell populations were CFSE negative, indicating that they had undergone extensive in vivo division. We next transferred CFSE-labeled TCR-transgenic T cells (expressing Thy1.2) into Thy1.1-congenic BALB/c mice. One day later, mice were infected with L. monocytogenes and then treated with ampicillin at 0, 12, or 24 h after bacterial inoculation. Splenocytes were harvested 5 days after infection, and CFSE staining of Thy1.2-positive, p60449–457-specific T cells was measured by flow cytometry. Similar to the findings in our investigation of the endogenous LLO91–99-specific T cell response, we found that the size of the p60449–457-specific T cell response is dramatically influenced by the duration of live bacterial infection during the first 24 h after inoculation of L. monocytogenes. Mice that received ampicillin 24 h after inoculation had maximal CD8 T cell responses to p60449–457, whereas mice that received ampicillin 12 h after inoculation had smaller T cell responses (Fig. 8⇓B). Importantly, although the responses in the setting of early ampicillin therapy are smaller, the responding T cells have all undergone extensive in vivo division since they have lost all CFSE staining. Interestingly, 5 days after L. monocytogenes infection, p60449–457-specific T cells fall into one of two categories: those that have undergone extensive division (>8–10 divisions) and those that have not divided at all. Thus, it appears that cells that are activated undergo extensive division, whereas others, which presumably received either no or a subthreshold exposure to Ag, do not divide. It is noteworthy that undivided, p60449–457-specific T cells persist in mice that received no antibiotics even though they were exposed to extensive in vivo bacterial infection. This result indicates that the extensive bacterial growth that occurs between 24 h and 7 days after inoculation does not result in progressive priming of naive T cells, suggesting that in vivo T cell priming is temporally restricted.

FIGURE 8.
FIGURE 8.

CD8 T cells stimulated by transient bacterial infection undergo extensive in vivo proliferation following bacterial clearance. A, Splenocytes were harvested from WP TCR-transgenic mice and labeled with CFSE as described in Materials and Methods. A total of 30 × 106 labeled splenocytes (equivalent to 4 × 106 CD8 T cells) were transferred into naive BALB/c mice and left uninfected for 8 days (left panel), or infected with 2000 L. monocytogenes 24 h after T cell transfer (right panel). Eight days after splenocyte transfer, recipient splenocytes were harvested and stained with mAb specific for CD8 and with H2-Kd tetramers complexed with p60449–457. The dot plots demonstrate tetramer staining on the vertical axis and CFSE fluorescence on the horizontal axis of live CD8-gated T cells. The percentage of CD8 T cells that are specific for p60449–457 and that have lost CFSE staining is indicated in the left upper quadrant of the plots. B, Splenocytes were harvested from p60449–457 TCR-transgenic mice, labeled with CFSE, and transferred into BALB/c Thy1.1-congenic mice. Mice were left either uninfected (upper panel) or infected with 2000 L. monocytogenes and treated with ampicillin immediately, 12 h, or 24 h after infection or left untreated. Five days after bacterial infection, recipient splenocytes were harvested and stained with mAbs specific for Thy 1.2 and CD8α and with H2-Kd tetramers complexed with p60449–457. The histograms plot live, CD8+, Thy1.2+, and tetramer-positive cells and measure the intensity of CFSE staining on the horizontal axis. The increments indicated on the upper aspect of each plot indicate the fluorescence intensity anticipated with sequential T cell divisions. The number of p60449–457-specific T cells per 100,000 CD8 T cells is indicated for the indicated region in each histogram.

Discussion

Our results indicate that CD8 T cells are activated during the first 24 h of bacterial infection and can undergo extensive division and acquire effector functions in the absence of progressive infection. Curtailing infection with the bactericidal antibiotic ampicillin following the first day of infection, or infecting with a broad range of bacterial doses, does not effect the magnitudes of the CD8 T cell responses to the dominant LLO91–99 or the subdominant p60449–457 epitopes. This suggests that once T cells are activated during the first day of infection, the amount of Ag that is present during T cell expansion does not determine the magnitude of the T cell burst. These results support the notion that, following T cell activation, the expansion and contraction of Ag-specific T cells is Ag independent.

It is possible that Ags produced by L. monocytogenes during the first 24 h of infection persist in a depot site for prolonged periods of time after infection. Furthermore, it is possible that secreted L. monocytogenes Ags are processed and presented in vivo long after the last viable bacterium has been killed by the host. Thus, it is not possible to state that Ag is absent in mice infected with low doses of L. monocytogenes or those treated with ampicillin at early time points following infection. In previous studies, however, we showed that the kinetics of the CD8 T cell response to peptides derived from three different L. monocytogenes Ags is similar (15). Therefore, if these Ags are driving the expansion and contraction of distinct CD8 T cell populations, it would indicate that these Ags are presented in vivo for similar periods of time. This seems highly unlikely, since Ags are synthesized by L. monocytogenes at vastly different rates, and their rate of degradation by mammalian cells is markedly different (26, 27). In the current study, we show that the size of T cell responses are similar in mice that are infected with nearly a 1000-fold range of L. monocytogenes doses. Additionally, we show that the size of Ag-specific T cell populations are similar in mice that have undergone normal or antibiotic-attenuated infection. It is highly unlikely that the duration of in vivo Ag presentation is similar in all of these circumstances. Thus, while it is possible that some Ags are present in vivo for much longer periods of time than the duration of viable infection, our results strongly argue that the duration and extent of T cell expansion are not simply determined by the presence or absence of cognate Ag.

At least two models can explain Ag-independent in vivo T cell expansion following bacterial infection. The first is that Ag-specific T cells are programmed during the first day of infection to undergo a given number of in vivo divisions. Once T cells are programmed, they are no longer dependent upon further Ag presentation. T cell proliferation following in vivo stimulation may be similar to the cellular programs or clocks that drive oligodendrocyte division and differentiation following exposure to growth factors and retinoic acid (28, 29). A second model to explain Ag-independent in vivo T cell proliferation is that an in vivo scaffold is generated during the first day of infection that promotes the expansion of effector and memory T cells. The scaffold would not depend upon continued in vivo infection to promote T cell expansion beyond the first day of infection. It is tempting to propose a structural scaffold that provides physical niches within lymphoid tissues for expanding Ag-specific T cell populations. An alternative idea, however, is that a soluble or cell-associated cytokine milieu that promotes lymphoid generation is generated during the first day of bacterial infection. TNF and related molecules have been implicated in the genesis of lymphoid tissues and likely play an important role in the generation of Ag-specific T cell populations (30, 31, 32). We suggest that the innate immune response to L. monocytogenes infection during the first day of infection creates a scaffold that determines the extent and duration of T cell expansion. The importance of innate immunity in determining T cell expansion kinetics is suggested by the finding that T cell responses to the same epitope presented in the context of infections by different pathogens results in T cell populations of markedly different magnitudes (33, 34).

The timing of in vivo T cell priming has been shown to dramatically influence the expansion of Ag-specific T lymphocytes (35). We extend this finding by showing that the size of the T cell response to two L. monocytogenes-derived epitopes is determined in the first day of infection and is tightly linked to the duration of live bacterial infection during the first 24 h after inoculation. Although we found that LLO91–99- and p60449–457-specific T cell responses are normal when ampicillin is administered 24 h after bacterial inoculation, administration of ampicillin 12 h after inoculation results in a significantly smaller LLO91–99-specific T cell response. This result is reminiscent of findings described by Sprent et al. (36) nearly 30 years ago when they demonstrated that systemic administration of Ag clears the thoracic duct lymph of Ag-specific T lymphocytes within 24–48 h. Thus, T lymphocyte circulation is sufficiently efficient that all Ag-reactive T cells come into contact with cognate Ag in a period as brief as 24 h. Our studies of LLO91–99- and p60449–457-specific responses also indicate that the maximal number of Ag-specific T cells can become activated with 24 h of infection. However, our studies transferring p60449–457-specific cells demonstrate that many residual Ag-specific T cells do not divide upon transfer. This result, while showing that the magnitude of a T cell response can be enhanced by increasing the precursor frequency of naive Ag- specific T cells, also suggests that the capacity to prime Ag-specific T cells is saturable. Saturation of the in vivo priming process has been previously suggested by the studies of Butz and Bevan (37) of lymphocytic choriomeningitis virus-specific T cell priming.

The finding that T cell expansion in response to L. monocytogenes infection is independent of prolonged in vivo Ag presentation is interesting in the context of recent studies of naive and memory T cell homeostasis. Naive T cells require MHC molecules for in vivo proliferation (38, 39, 40, 41) while memory T cells appear to undergo slow in vivo division without the need for any MHC-induced, TCR-mediated stimuli (42, 43). We do not know whether the expansion of Ag-specific T cells requires the presence of MHC class I molecules. It is possible that, upon activation, L. monocytogenes-specific T cells acquire proliferative characteristics that have been attributed to memory T cells, i.e., independence from TCR-mediated stimuli. Further studies will be required to answer this question.

Acknowledgments

We thank Christophe Viret and Derek Sant Angelo for excellent technical advice and discussions on the generation of TCR-transgenic mice. We also thank Dr. Charles Surh (The Scripps Research Institute) for providing us with Thy1.1-congenic BALB/c mice.

Footnotes

  • ↵1 This work was supported by National Institutes of Health Grants AI 39031 and AI42135. S.V. was supported by National Research Service Award individual Fellowship F32 AI09629-02.

  • ↵2 R.M. and S.V. contributed equally to the work described in this manuscript.

  • ↵3 Current address: Henry M. Jackson Foundation for the Advancement of Military Medicine, 1600 East Gude Drive, Rockville, MD 20850.

  • ↵4 Address correspondence and reprint requests to Dr. Eric G. Pamer, Infectious Diseases Service, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021. E-mail address: pamere{at}mskcc.org

  • 5 Abbreviation used in this paper: LLO, listeriolysin O.

  • Received August 24, 2000.
  • Accepted September 21, 2000.

References

  1. ↵
    Gellin, B. G., C. V. Broome. 1989. Listeriosis. J. Am. Med. Assoc. 261: 1313
    OpenUrlCrossRefPubMed
  2. ↵
    Unanue, E. R.. 1997. Why listeriosis? A perspective on cellular immunity to infection. Immunol. Rev. 158: 5
    OpenUrlCrossRefPubMed
  3. ↵
    Unanue, E.. 1997. Studies in listeriosis show the strong symbiosis between the innate cellular system and the T-cell response. Immunol. Rev. 158: 11
    OpenUrlCrossRefPubMed
  4. ↵
    Rogers, H. W., E. R. Unanue. 1993. Neutrophils are involved in acute, nonspecific resistance to Listeria monocytogenes in mice. Infect. Immun. 91: 5090
    OpenUrl
  5. ↵
    Conlan, J. W., R. J. North. 1991. Neutrophil-mediated dissolution of infected host cells as a defense strategy against a facultative intracellular bacterium. J. Exp. Med. 174: 741
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Czuprynski, C. J., P. M. Henson, P. A. Campbell. 1985. Enhanced accumulation of inflammatory neutrophils and macrophages mediated by transfer of T cells from mice immunized with Listeria monocytogenes. J. Immunol. 134: 3449
    OpenUrlAbstract
  7. ↵
    Mombaerts, P., J. Arnoldi, F. Russ, S. Tonegawa, S. H. Kaufmann. 1993. Different roles of αβ and γδ T cells in immunity against an intracellular bacterial pathogen. Nature 365: 53
    OpenUrlCrossRefPubMed
  8. ↵
    Ladel, C. H., I. E. F. Flesch, J. Arnoldi, S. H. E. Kaufmann. 1994. Studies with MHC-deficient knock-out mice reveal impact of both MHC I- and MHC II-dependent T cell responses on Listeria monocytogenes infection. J. Immunol. 153: 3116
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Edelson, B. T., P. Cossart, E. R. Unanue. 1999. Paradigm revisited: antibody provides resistance to Listeria infection. J. Immunol. 163: 4087
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Ochsenbein, A. F., T. Fehr, C. Lutz, M. Suter, F. Brombacher, H. Hengartner, R. M. Zinkernagel. 1999. Control of early viral and bacterial distribution and disease by natural antibodies. Science 286: 2156
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Portnoy, D. A., T. Chakraborty, W. Goebel, P. Cossart. 1992. Molecular determinants of Listeria monocytogenes pathogenesis. Infect. Immun. 60: 1263
    OpenUrlFREE Full Text
  12. ↵
    Pamer, E. G., A. J. A. M. Sijts, M. S. Villanueva, D. H. Busch, S. Vijh. 1997. MHC class I antigen processing of Listeria monocytogenes proteins: implications for dominant and subdominant CTL responses. Immunol. Rev. 158: 129
    OpenUrlCrossRefPubMed
  13. ↵
    Harty, J. T., M. J. Bevan. 1999. Responses of CD8 T cells to intracellular bacteria. Curr. Opin. Immunol. 11: 89
    OpenUrlCrossRefPubMed
  14. ↵
    Busch, D. H., K. Kerksiek, E. G. Pamer. 1999. Processing of Listeria monocytogenes antigens and the in vivo T cell response to bacterial infection. Immunol. Rev. 172: 163
    OpenUrlCrossRefPubMed
  15. ↵
    Busch, D. H., I. M. Pilip, S. Vijh, E. G. Pamer. 1998. Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 8: 353
    OpenUrlCrossRefPubMed
  16. ↵
    Vijh, S., E. G. Pamer. 1997. Immunodominant and subdominant CTL responses to Listeria monocytogenes infection. J. Immunol. 158: 3366
    OpenUrlAbstract
  17. ↵
    Chen, W., L. C. Anton, J. R. Bennink, J. W. Yewdell. 2000. Dissecting the multifactorial causes of immunodominance in class I-restricted T cell responses to viruses. Immunity 12: 83
    OpenUrlCrossRefPubMed
  18. ↵
    Yewdell, J. W., J. R. Bennink. 1999. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 17: 51
    OpenUrlCrossRefPubMed
  19. ↵
    Sijts, A. J. A. M., A. Neisig, J. Neefjes, E. G. Pamer. 1996. Two Listeria monocytogenes CTL epitopes are processed from the same antigen with different efficiencies. J. Immunol. 156: 685
    OpenUrl
  20. ↵
    Pamer, E. G.. 1994. Direct sequence identification and kinetic analysis of an MHC class I-restricted Listeria monocytogenes CTL epitope. J. Immunol. 152: 686
    OpenUrlAbstract
  21. ↵
    Vijh, S., I. M. Pilip, E. G. Pamer. 1998. Effect of antigen processing efficiency on in vivo T cell response magnitudes. J. Immunol. 160: 3971
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Mackaness, G. B.. 1962. Cellular resistance to infection. J. Exp. Med. 116: 381
    OpenUrlAbstract
  23. ↵
    Gallimore, A., A. Glithero, A. Godkin, A. C. Tissot, A. Pluckthun, T. Elliot, H. Hengartner, R. Zinkernagel. 1998. Induction and exhaustion of lymphocytic choriomeningitis virus-specific cytotoxic T lymphocytes visualized using soluble tetrameric major histocompatibility complex class I-peptide complexes. J. Exp. Med. 187: 1383
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Sijts, A. J. A., E. G. Pamer. 1997. Enhanced intracellular dissociation of major histocompatibility complex class I-associated peptides: a mechanism for optimizing the spectrum of cell surface-presented cytotoxic T lymphocyte epitopes. J. Exp. Med. 185: 1
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Lyons, A. B., C. R. Parish. 1994. Determination of lymphocyte division by flow cytometry. J. Immunol. Methods 171: 131
    OpenUrlCrossRefPubMed
  26. ↵
    Villanueva, M. S., P. Fischer, K. Feen, E. G. Pamer. 1994. Efficiency of antigen processing: a quantitative analysis. Immunity 1: 479
    OpenUrlCrossRefPubMed
  27. ↵
    Villanueva, M. S., A. J. A. M. Sijts, E. G. Pamer. 1995. Listeriolysin is processed efficiently into an MHC class I-associated epitope in Listeria monocytogenes-infected cells. J. Immunol. 155: 5227
    OpenUrlAbstract
  28. ↵
    Conlon, I., M. Raff. 1999. Size control in animal development. Cell 96: 235
    OpenUrlCrossRefPubMed
  29. ↵
    Durand, B., M. L. Fero, J. M. Roberts, M. C. Raff. 1998. p27Kip1 alters the response of cells to mitogen and is part of a cell-intrinsic timer that arrests the cell cycle and initiates differentiation. Curr. Biol. 8: 431
    OpenUrlCrossRefPubMed
  30. ↵
    Ruddle, N. H.. 1999. Lymphoid Neo-organogenesis. Immunol. Res. 19: 119
    OpenUrlCrossRefPubMed
  31. ↵
    Koni, P. A., R. Sacca, P. Lawton, J. L. Browning, N. H. Ruddle, R. A. Flavell. 1997. Distinct roles in lymphoid organogenesis for lymphotoxins α and β revealed in lymphotoxin β deficient mice. Immunity 6: 491
    OpenUrlCrossRefPubMed
  32. ↵
    Berger, D. P., D. Naniche, M. T. Crowley, P. A. Koni, R. A. Flavell, M. B. Oldstone. 1999. Lymphotoxin-β-deficient mice show defective antiviral immunity. Virology 260: 136
    OpenUrlCrossRefPubMed
  33. ↵
    Shen, H., J. F. Miller, X. Fan, D. Kolwyck, R. Ahmed, J. T. Harty. 1998. Compartmentalization of bacterial antigens: differential effects on priming of CD8 T cells and protective immunity. Cell 92: 535
    OpenUrlCrossRefPubMed
  34. ↵
    Rodrigues, E. G., F. Zavala, D. Eichinger, J. M. Wilson, M. Tsuji. 1997. Single immunizing dose of recombinant adenovirus efficiently induces CD8 T cell-mediated protective immunity against malaria. J. Immunol. 158: 1268
    OpenUrlAbstract
  35. ↵
    Bousso, P., J. P. Levraud, P. Kourilsky, J. P. Abastado. 1999. The composition of a primary T cell response is largely determined by the timing of recruitment of individual T cell clones. J. Exp. Med. 189: 1591
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Sprent, J., J. F. A. P. Miller, G. F. Mitchell. 1971. Antigen-induced selective recruitment of circulating lymphocytes. Cell. Immunol. 2: 171
    OpenUrlCrossRefPubMed
  37. ↵
    Butz, E. A., M. J. Bevan. 1998. Massive expansion of antigen-specific CD8+ T cells during an acute virus infection. Immunity 8: 167
    OpenUrlCrossRefPubMed
  38. ↵
    Viret, C., F. S. Wong, C. A. Janeway. 1999. Designing and maintaining the mature TCR repertoire: the continuum of self-peptide:self-MHC complex recognition. Immunity 10: 559
    OpenUrlCrossRefPubMed
  39. ↵
    Ernst, B., D. S. Lee, J. M. Chang, J. Sprent, C. D. Surh. 1999. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 11: 173
    OpenUrlCrossRefPubMed
  40. ↵
    Goldrath, A. W., M. J. Bevan. 1999. Low-affinity ligands for the TCR drive proliferation of mature CD8 T cells in lymphopenic hosts. Immunity 11: 183
    OpenUrlCrossRefPubMed
  41. ↵
    Nesic, D., S. Vukmanovic. 1998. MHC class I is required for peripheral accumulation of CD8 thymic emigrants. J. Immunol. 160: 3705
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Murali-Krishna, K., L. L. Lau, S. Sambhara, F. Lemonnier, J. Altman, R. Ahmed. 1999. Persistence of memory CD8 T cells in MHC class I-deficient mice. Science 286: 1377
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Swain, S. L., H. Hu, G. Huston. 1999. Class II-independent generation of CD4 memory T cells from effectors. Science 286: 1381
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Kouskoff, V., K. Signorelli, C. Benoist, D. Mathis. 1995. Cassette vectors directing expression of T cell receptor genes in transgenic mice. J. Immunol. Methods 180: 273
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section