Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Mar 9.
Published in final edited form as: J Immunol. 2008 Sep 15;181(6):3804–3810. doi: 10.4049/jimmunol.181.6.3804

MyD88 Plays a Critical T Cell-Intrinsic Role in Supporting CD8 T Cell Expansion during Acute Lymphocytic Choriomeningitis Virus Infection1

Adeeb H Rahman *,2, Weiguo Cui †,2, David F LaRosa *, Devon K Taylor *, Jidong Zhang *, Daniel R Goldstein ‡, E John Wherry §, Susan M Kaech †,3, Laurence A Turka *,3,4
PMCID: PMC2835404  NIHMSID: NIHMS181523  PMID: 18768833

Abstract

During acute lymphocytic choriomeningitis virus (LCMV) infection, CD8 T cells rapidly expand and differentiate into effectors that are required for viral clearance. The accumulation of activated T cells is greatly reduced in mice lacking the adaptor molecule MyD88. Although MyD88 has generally been considered to indirectly regulate adaptive immune responses by controlling inflammatory cytokine production and Ag presentation in innate immune cells, in this study, we identify an unappreciated cell-intrinsic role for MyD88 in LCMV-specific CD8 T cells. Using reciprocal adoptive transfer models and bone marrow chimeras, we show that Myd88−/− CD8 T cells are defective in their clonal expansion in response to LCMV infection, independent of their environment. Furthermore, we show that while MyD88 is dispensable for initial activation and division of LCMV-specific CD8 T cells during the early stages of viral infection, MyD88-dependent signals are critical for supporting their survival and sustained accumulation.

During the early stages of a viral infection, cells of the innate immune stem act to directly combat the pathogen (1), and also serve to initiate an adaptive immune response (2). Under the direction of inflammatory signals from the innate immune system, CD8 T cells rapidly expand and differentiate into effector cells, which play a major role in viral clearance. The recognition of viral components is one of the first steps in this process and is largely mediated by innate pattern recognition receptors (3). TLRs comprise a major family of these receptors. The TLR family currently consists of 13 members that detect a broad range of microbial-derived ligands (4). MyD88 is an adaptor protein that is required for signal transduction through most TLRs, as well as the IL-1R family (5).

MyD88 has been shown to be important in host defense as MyD88-deficient mice have increased susceptibility to a number of eukaryotic, bacterial, and viral pathogens (6 – 8). The immune impairments associated with MyD88-deficiency have generally been attributed to its importance in cells of the innate immune system, which serve both as the initial sensors of, and effectors against, microbial infections as well as initiators of subsequent adaptive immune responses (2). Yet, MyD88 is also expressed by T cells and we, and others, have shown that TLR ligands can directly costimulate multiple subsets of T cells to enhance survival, proliferation, and effector functions in vitro (9 –12). In addition, we have recently shown that T cell expression of MyD88 is required for resistance to the protozoan pathogen, Toxoplasma gondii (13). However, the specific role of MyD88 in T cells during a physiological immune response remains unclear. In particular, it is not known whether MyD88-dependent pathways contribute to initial T cell activation, effector cell differentiation, proliferation, survival, or the maintenance of memory cells following infection.

Lymphocytic choriomeningitis virus (LCMV)5 is the prototypic murine arenavirus and has been extensively characterized as a murine infection model (14). Acute LCMV infection elicits a dramatic expansion of CD8 T cells in C57BL/6 mice, and the availability of reagents allowing the detection of Ag-specific effector cells make this an ideal model for examining in vivo T cell responses to a natural, viral pathogen.

Myd88−/− mice are known to mount dramatically reduced CD8 T cell responses following LCMV infection, resulting in impaired viral clearance (15, 16). In response to LCMV infection, APCs from Myd88−/− mice produce significantly lower levels of type I IFNs and other proinflammatory cytokines (16), which, in turn, have been shown to be important in regulating T cell expansion (17, 18). Ag presentation of LCMV-derived peptides may also be impaired (19). Although these defects in the innate immune response may be expected to contribute to the impaired adaptive immune response to LCMV, they do not entirely account for the reduced effector CD8 T cell response in Myd88−/− mice (15). In the present study, we directly show that T cell expression of MyD88 is critical for optimal CD8 T cell responses to LCMV. Specifically, we demonstrate that MyD88 is dispensable for early T cell division and effector differentiation, but plays an unappreciated, T cell-intrinsic role in supporting the survival and accumulation of LCMV-specific CD8 T cells during clonal expansion.

Materials and Methods

Mice and LCMV infection

Mice were used in accordance with University of Pennsylvania and Yale University Institutional Animal Care and Use Committee guidelines. Wild-type (WT) C57BL/6, IL1r1−/− and IL18r1−/− mice were purchased from The Jackson Laboratory. Myd88−/− and P14 TCR-transgenic mice have been previously described (20, 21) and were backcrossed at least eight times onto the C57BL/6 background. Myd88−/− P14 TCR-transgenic mice were generated by interbreeding the two strains. All strains were back-crossed onto congenic Thy1.1/1, Thy1.1/2, and Thy1.2/2 backgrounds. Mice were injected with 2 × 105 PFU of LCMV-Armstrong i.p. to initiate acute infection.

Adoptive transfers

For all adoptive transfers, CD8 T cells were purified from spleens and lymph nodes of indicated naive donor mice using MACS technology (Miltenyi Biotec) and CFSE labeled where indicated. Purified CD8 T cells were transferred into naive, Thy1 congenic recipients via retro-orbital injection. For polyclonal experiments, 1 × 107 CD8 T cells were transferred. For P14 experiments, 2 × 104 cells were transferred. For cotransfer experiments, 1 × 106 P14 cells comprising a 1:1 mixture of WT (Thy1.1/1.2) and Myd88−/− (Thy1.2/1.2) were transferred into WT (Thy1.1/1.1) recipients. Recipients were infected 18 h following adoptive transfers.

Bone marrow chimeras

Mixed bone marrow chimeras were generated by reconstituting lethally irradiated (1000 rads) WT mice (Thy1.2) with a 1:1 mixture of WT (Thy1.2) and Myd88−/− (Thy1.1) T cell-depleted bone marrow. Twelve weeks were allowed for immune reconstitution before mice were used for experiments.

Flow cytometry and tetramer staining

Allophycocyanin-conjugated MHC class I tetramers of H-2Db complexed with LCMV GP33– 41, NP396 – 404, and GP276 –286 were prepared as previously described (21). All Abs were purchased from BD Biosciences. Splenocytes were stained as previously described (21) and analyzed using a FACSCanto flow cytometer (BD Biosciences).

Ex vivo restimulation and intracellular cytokine staining

Three × 106 splenocytes were cultured in the absence or presence of gp33–41, np396 – 404, or a pooled peptide mixture of np396 – 404, np205–212, np166 –175, np235–243, gp33– 41, gp276 –286, gp118 –125, gp92–101, and gp70 –77. These pooled peptides account for nearly all known LCMV epitopes, thereby providing a measure of total LCMV-specific CTL responses (22). Cells were stimulated for 6 h at 37°C in the presence of BD GolgiStop. Following staining of surface Ags as indicated above, cells were stained for intracellular cytokines using the BD Cytofix/Cytoperm kit (BD Pharmingen) according to the manufacturer’s instructions.

Determining proliferation by BrdU incorporation

Mice were given 1 mg of BrdU i.p. at 12 h intervals before sacrifice, for the duration indicated in figure legends. Following staining of surface Ags as described above, cells were stained for BrdU expression using the FITC BrdU flow kit (BD Pharmingen) according to the manufacturer’s instructions.

Assessment of cell viability

Splenocytes were examined either directly ex vivo or following 20 h of culture in RPMI 1640 with 10% serum in the absence of additional peptides or cytokines. Cells were incubated with 0.5 μg/ml fluorescein-conjugated V-D-FMK (R&D Systems) for 30 min at 37°C to determine total caspase activity. Following staining of surface Ags, cells were resuspended in 200 μl of annxein binding buffer with 4 μl of annexin V PE (BD Pharmingen).

Statistical analysis

Values of p were calculated using a Student’s t test (two-tailed distribution, equal variance) with p < 0.05 taken as statistically significant.

Results

Myd88−/− mice mount greatly reduced CD8 T cell responses to LCMV

Consistent with earlier reports (15, 16), we found that at the peak of the adaptive response, 8 days after acute LCMV-Armstrong (Arm) infection, Myd88−/− mice had greatly reduced numbers of CD8 T cells in their spleens in comparison to WT C57BL/6 mice (Fig. 1A). Characterization of these cells with H2-Db tetramers for known immunodominant LCMV epitopes revealed a reduced percentage of LCMV-specific Myd88−/− CD8 T cells (Fig. 1B), which corresponded to a reduced percentage of Ag-specific IFN-γ producing CTLs (Fig. 1C). Thus, the absence of IFN-γ producing cells in Myd88−/− mice, reported earlier (15), primarily reflects a reduced frequency of LCMV-specific CD8 T cells at the peak of the response, rather than a failure to develop effector functions.

FIGURE 1.

FIGURE 1

Open in a new tab

The size of the CD8 T cell response against LCMV is reduced in Myd88−/− mice. WT and Myd88−/− C57BL/6 mice were infected with LCMV-Arm and splenocytes were purified 8 days after infection. A, CD8 T cells were enumerated using Abs against CD8α and CD44. Data represent mean ± SD for at least 10 mice. B, Gated CD8 T cells were analyzed with the indicated LCMV-specific tetramers. C, Splenocytes were restimulated ex vivo with the indicated LCMV-derived peptides to analyze Ag-specific IFN-γ production by CD8 T cells. D, LCMV-specific CD8 T cells in the spleens of infected WT, Myd88−/−, IL1r1−/−, IL18r1−/−, and ICE−/− were enumerated using the indicated H2-Db tetramers. Data represent mean ± SD of at least four mice and p values of total tetramer-positive cell numbers relative to WT are indicated.

It has recently been shown that the profoundly reduced T cell response to LCMV in Myd88−/− mice is not recapitulated by deficiencies in any individual TLR pathway (16). Consistent with this, we found no reduction in LCMV-specific CTLs in the absence of TLR2 or TLR9 expression (data not shown). MyD88 also plays a role in signaling downstream of the IL-1R family, and the cytokines using these receptors, IL-1 and IL-18, have been shown to influence T cell survival and effector functions (20, 23). We therefore investigated whether these pathways, rather than TLRs, might regulate the CD8 T cell response to LCMV. LCMV infection induced considerable expansion of LCMV-specific CD8 T cells in IL1r1−/−, IL18r1−/−, and caspase-1-deficient (ICE−/−) mice, which are unable to produce functional IL-1β or IL-18 (Fig. 1D). Although the spleens of IL18r1−/− mice contained significantly lower numbers of tetramer-specific CD8 T than those of WT mice, the CD8 T cell response was far larger than that observed in Myd88−/− mice. These data suggest that the importance of MyD88 during LCMV infection cannot be entirely accounted for by either IL-1R- or IL-18R-dependent signaling pathways. In light of these findings, we focused on cells from Myd88−/− mice in subsequent experiments.

WT CD8 T cells expand and develop effector functions in MyD88-deficient hosts

The reduced levels of type I IFNs and other proinflammatory cytokines in Myd88−/− mice during LCMV infection (16) suggest that the MyD88-deficient innate immune compartment may not effectively support CD8 T cell expansion (17). However, given that MyD88-dependent pathways can also directly regulate T cell functions, the impaired CD8 T cell response in Myd88−/− mice could potentially reflect an intrinsic role for MyD88 in T cells during LCMV infection. To examine the relative contribution of MyD88 in the innate immune compartment, naive WT CD8 T cells were transferred into congenic, Myd88−/− recipients, which were then infected with LCMV. We found that WT CD8 T cells mounted a reduced response to LCMV infection in Myd88−/− hosts compared with their response in WT controls, resulting in 2-fold fewer transferred cells 8 days after infection (Fig. 2A). This corresponded to a smaller percentage of LCMV-specific CTLs as detected by tetramer staining (Fig. 2B). Interestingly, there was a more marked reduction in the percentage of np396-specific CTLs than gp33-specific CTLs within the WT donor population in Myd88−/− hosts. Although the number of LCMV-specific cells in the donor population was significantly reduced, these cells produced IFN-γ following ex vivo restimulation (Fig. 2, C and D). Thus, consistent with earlier reports using TCR transgenic T cells (15), the absence of MyD88 in the host environment does not appear to impair the differentiation of naive CD8 T cells into functional effector cells. Most importantly, while the donor WT T cells did not expand as greatly in Myd88−/− hosts as in WT hosts, they still expanded far better than endogenous Myd88−/− CD8 T cells, as shown by a greater relative percentage of tetramer-specific cells (Fig. 2B). Therefore, as suggested by earlier findings (15), the absence of MyD88 in the host environment did not entirely impair the polyclonal CD8 T cell response to LCMV, indicating that the innate immune system does not entirely account for the reduced CD8 T cell response in Myd88−/− mice.

FIGURE 2.

FIGURE 2

Open in a new tab

WT CD8 T cells are able to expand and develop effector functions in response to LCMV in MyD88-deficient hosts. One × 107 naive WT Thy1.1 CD8 T cells were adoptively transferred into Thy1.2 congenic WT or Myd88−/− recipients, which were left uninfected or infected with LCMV 18 h later. Splenocytes were examined 8 days after infection. A, The number of adoptively transferred cells per spleen was enumerated using Abs against Thy1.1, CD8α, and CD44. Data represent mean ± SD of at least five mice. B, The percentage of tetramer-specific cells in the donor and host CD8+ populations was compared. C and D, Splenocytes were restimulated with the indicated LCMV-derived peptides and the frequency (C) and number (D) of IFN-γ-producing CTLs in the donor populations was determined. p < 0.05 for all values relative to corresponding WT controls.

Myd88−/− CD8 T cells do not effectively expand in WT hosts

To more directly examine the potential contribution of MyD88 in T cells during LCMV infection, we performed the reciprocal adoptive transfer of purified CD8 T cells from naive Myd88−/− mice into congenic WT recipients. Despite the provision of a WT APC compartment and proinflammatory cytokine milieu, we found that Myd88−/− donor CD8 T cells did not expand as effectively as WT CD8 T cells in response to LCMV, resulting in ~10-fold fewer donor cells 8 days after infection (Fig. 3A). This corresponded to a reduced percentage of LCMV-specific CTLs, as detected by tetramer staining (Fig. 3B). Although LCMV-specific cells in the Myd88−/− donor population produced IFN-γ following ex vivo restimulation (Fig. 3C), the overall number of IFN-γ-producing CTLs was greatly reduced (Fig. 3D). This suggests that MyD88 regulates LCMV-specific T cell expansion rather than the development of effector functions.

FIGURE 3.

FIGURE 3

Open in a new tab

Myd88−/− CD8 T cells do not effectively expand in response to LCMV infection despite transfer to a WT host. One × 107 naive Thy1.1 WT or Myd88−/− CD8 T cells were adoptively transferred into Thy1.2 congenic WT recipients, which were left un-infected or infected with LCMV 18 h later. Splenocytes were examined 8 days after infection. A, The number of adoptively transferred cells per spleen was enumerated using Abs against Thy1.1, CD8α, and CD44. Data represent mean ± SD of at least 5 mice. B, The percentage of tetramer-specific cells in the donor and host CD8+ populations was compared. C and D, Splenocytes were restimulated with the indicated LCMV-derived peptides and the frequency (C) and number (D) of IFN-γ-producing CTLs in the donor populations was determined. p < 0.01 for all values relative to corresponding WT controls.

The impaired LCMV response of Myd88−/− T cells does not result from developmental defects

The adoptive transfer experiments above suggested a T cell-intrinsic role for MyD88 during acute LCMV infection. However, it was possible that T cell development in a MyD88-deficient environment could cause persistent functional impairments, resulting in diminished expansion of Myd88−/− LCMV-specific CD8 T cells despite transfer into a WT environment. To address this possibility, we generated mixed bone marrow chimeras using a 1:1 ratio of congenic WT and Myd88−/− bone marrow. Twelve weeks after bone marrow reconstitution, these chimeric mice contained a comparable proportion of WT and Myd88−/− CD8 T cells (Fig. 4A, left panel), but following LCMV infection, the MyD88−/− CD8 T cell compartment did not expand as effectively as the WT compartment, resulting in a 6-fold decrease in the relative proportion of Myd88−/− CD8 T cells (Fig. 4A, right panel). The Myd88−/− population contained a similarly reduced percentage of LCMV-specific IFN-γ-producing CD8 T cells (Fig. 4B). Therefore, despite developing under identical conditions, and having access to the same APC compartment and cytokines during LCMV infection, Myd88−/− T cells did not expand as effectively as WT T cells.

FIGURE 4.

FIGURE 4

Open in a new tab

Myd88−/− CD8 T cell defects do not arise from development in a MyD88-deficient environment. Lethally irradiated WT recipients were reconstituted with a 1:1 mixture of WT and Myd88−/− bone marrow and infected with LCMV. Splenocytes were examined 8 days after infection. A, The ratio of WT to Myd88−/− CD8 T cells was compared in infected and uninfected chimeras. B, Splenocytes were restimulated with the indicated peptides to examine LCMV-specific IFN-γ production by WT and Myd88−/− CD8 T cells. Data are representative of four mice and p < 0.05 for all values relative to WT.

Myd88−/− P14 transgenic T cells do not expand effectively in response to LCMV infection but develop effector functions and differentiate into long-lived memory cells

Our results thus far strongly suggested that MyD88 plays a T cell intrinsic role in regulating the response to LCMV infection. However, because these experiments examined naturally occurring LCMV-specific CD8 T cell clones in a polyclonal response, it remained possible that Myd88−/− T cells developed a different TCR repertoire, which subsequently influenced their response to infection. To circumvent potential repertoire differences in the polyclonal Myd88−/− T cell population, we crossed Myd88−/− mice with P14-TCR transgenic mice, whose T cells recognize the LCMV-derived gp33 epitope. Following transfer into WT recipients, Myd88−/− P14 cells expanded in response to LCMV infection, but the number of Myd88−/− P14 cells at peak expansion was 15-fold less than WT P14 (Fig. 5A), consistent with the reduced number of LCMV-specific CD8 T cells in the polyclonal repertoire of Myd88−/− mice. Although we noted some differences in their relative tissue distributions, the percentage of Myd88−/−, relative to WT, P14 cells was also reduced in the peripheral lymph nodes, peritoneum, and liver of infected mice, (Fig. 5B). The limited number Myd88−/− P14 cells present at day 7 postinfection appeared to be functionally competent and produced IFN-γ and TNF-α at a similar frequency and level to WT P14 cells (Fig. 5C). In addition, we found no noticeable differences in the expression of the co-stimulatory receptors CD28, CD27, OX-40, ICOS, and 41BB, or the inhibitory receptors PD-1, CTLA-4, and Fas (data not shown). Thus, consistent with our results with polyclonal CD8 T cell transfers, MyD88 appears to regulate LCMV-specific CD8 T cell expansion but not the development of effector functions.

FIGURE 5.

FIGURE 5

Open in a new tab

Myd88−/− P14 transgenic T cells do not expand effectively during LCMV infection but develop effector functions and differentiate into long-lived memory cells. Myd88−/− or WT P14 cells were transferred into individual WT recipients, which were infected 18 h later. A, On day 7 and day 35 post infection the numbers of WT and Myd88−/− P14 cells per spleen were enumerated. Data represent mean ± SD of nine mice and p < 0.01 for both time points. B, The percentage of WT and Myd88−/− P14 cells was determined in spleens, lymph nodes, liver and peritoneum of day 6 infected mice. Data represent mean ± SD of three mice and p values of Myd88−/− relative to WT are indicated. C, Splenocytes from day 7-infected mice were restimulated with gp33– 41 and IFN-γ and TNF-α production by donor WT and Myd88−/− P14 cells was compared. Data are representative of five mice and differences were not statistically significant. D, LCMV-infected recipients were longitudinally bled and the percentage of WT and Myd88−/− P14 cells was determined relative to total CD8 T cells. p < 0.01 for all time points. E, Expression of KLRG-1 and IL-7Rα by WT and Myd88−/− P14 cells in day 7 and day 35 infected spleens. Data are representative of five mice and differences were not statistically significant.

Although overall expansion of the Myd88−/− P14 cells was considerably reduced in response to LCMV, serial bleeds showed that these cells expanded and contracted with similar kinetics to WT P14 cells (Fig. 5D). Both WT and Myd88−/− P14 cells differentiated into memory cells, and while the ultimate size of the Myd88−/− P14 memory population present 35 days post infection was substantially smaller than the WT population, their relative proportion was similar to that at day 7 (Fig. 5, A and B). Thus, the magnitude of contraction was proportional between the two cell populations, suggesting that the absence of MyD88 does not affect the survival or maintenance of precursors of the memory CD8 T cell population. In support of this hypothesis, the expression of killer cell lectin-like receptor subfamily G, member 1 (KLRG1) and IL-7R by the WT and Myd88−/− P14 cells indicated that the relative proportions of KLRG1highIL7Rlow short-lived effector cells and KLRG1lowIL7Rhigh memory precursor effector cells (24) were similar at day 7 post infection, and that the majority of the cells present at day 35 exhibited the KLRG1lowIL7Rhigh memory phenotype (Fig. 5E). These data suggest that MyD88 plays a critical, T cell intrinsic role in supporting CD8 T cell expansion during the early phase of Ag-driven proliferation, but is not involved in the contraction and subsequent maintenance of resting memory cells.

Myd88−/− P14 T cells proliferate in response to LCMV but do not accumulate

Our results thus far show greatly reduced clonal expansion of Myd88−/− CD8 T cell in response to LCMV. T cell numbers at peak expansion reflect the combined effects of proliferation and cell survival, thus we sought to examine which of these processes involved MyD88. Following LCMV infection in Myd88−/− mice, the percentage of CD8 T cells that incorporated BrdU during a 24 h labeling period at the peak of the adaptive response was much smaller than in WT controls (Fig. 6A). This was consistent with the reduced proportion of LCMV-specific CTLs in the CD8 T cell population of Myd88−/− mice (Fig. 1). Interestingly, however, examination of the tetramer-specific cells revealed that the small population of LCMV-specific Myd88−/− CD8 T cells incorporated BrdU comparably to WT CD8 T cells. This suggested that LCMV-specific T cells in Myd88−/− mice proliferated at a normal rate, implying that their dramatically lower numbers reflect impaired survival.

FIGURE 6.

FIGURE 6

Open in a new tab

Myd88−/− CD8 T cells proliferate normally but exhibit reduced survival. A, WT and Myd88−/− mice were infected and given a 24 h BrdU-pulse before sacrifice on day 8. BrdU incorporation by tetramer-specific CD8 T cells was examined. Differences in the percentage of BrdU+ cells among tetramer-specific populations were not statistically significant between WT and Myd88−/− mice. B and C, CFSE-labeled WT and Myd88−/− P14 cells were cotransferred into WT recipients, which were infected with LCMV 18 h later. At the indicated time points, division profiles (B) and frequency (C) of gated WT and Myd88−/− P14 cells were compared. D, WT and Myd88−/− P14 cells were cotransferred as in C. Following a 12 h BrdU pulse, recipients were sacrificed at the indicated time points and BrdU incorporation by donor WT and Myd88−/− P14 cells was compared. Differences in the percentage of BrdU+ WT and Myd88−/− P14 cells were not statistically significant at any of the time points examined. E, WT and Myd88−/− P14 cells were cotransferred as in C. Splenocytes were isolated 4 days following infection and the viability of gated WT and Myd88−/− P14 cells was assessed by scatter properties, annexin V staining and the detection of total active caspases. Cells were examined either immediately (left panels) or following a 20 h culture in the absence of exogenous peptides or cytokines (right panels). Data are representative of five mice. p < 0.01 for the percentage of live Myd88−/− P14 cells relative to WT P14 cells as determined by scatter properties following 20 h in culture.

To more comprehensively examine cell proliferation, we co-transferred CFSE-labeled WT and Myd88−/− P14 cells into congenic hosts to examine early proliferation in response to LCMV infection. No spontaneous proliferation was observed upon transfer, because both WT and Myd88−/− P14 cells remained CFSEhigh in uninfected recipients (Fig. 6B). At 72 h following LCMV infection, both WT and Myd88−/− P14 cells proliferated rapidly and underwent multiple rounds of division, diluting CFSE beyond the level of detection by 96 h, suggesting that WT and Myd88−/− P14 cells proliferate comparably in response to LCMV. Although division profiles appeared similar, the relative proportion of Myd88−/− to WT P14 cells decreased over time, corresponding to reduced accumulation of Myd88−/− T cells (Fig. 6, B and C). The reduced numbers of Myd88−/− P14 cells became most apparent by day 4 post infection, at which point proliferation could no longer be resolved by CFSE dilution. As it was possible that Myd88−/− T cells proliferated normally for the first 7– 8 rounds of division but subsequently stopped proliferating earlier than WT T cells, we used BrdU to compare the proportion of transferred WT and Myd88−/− P14 cells that were actively cycling at later time points. We found that a similar percentage of the adoptively transferred WT and Myd88−/− P14 cells incorporated BrdU during a 12 h labeling period on days 4, 5, and 6 following LCMV infection (Fig. 6D). Thus, the capacity for proliferation between the two populations appeared to be similar throughout the period during which the number of WT P14 cells greatly exceeded the number of Myd88−/− P14 cells.

To directly compare the survival of transferred Myd88−/− with WT P14 cells following LCMV infection, we assessed their viability using light scatter properties, annxein V staining and caspase activation. Surprisingly, we observed no significant differences in the viability of WT and Myd88−/− P14 cells directly ex vivo at any of the time points examined (Fig. 6E, left panels and data not shown), which could reflect the rapid clearance of apoptotic cells in vivo. However, when splenocytes were isolated from infected mice during the period of T cell expansion post infection and cultured for 20 h ex vivo we found significantly reduced survival of the Myd88−/− P14 cells (Fig. 6E, right panels and data not shown). This reduced survival resulted in a decline in the relative ratio of Myd88−/− to WT P14 cells during the culture period, consistent with the reduced accumulation of Myd88−/− P14 cells seen in vivo (data not shown). Together, these data indicate that the reduced number of Myd88−/− T cells following LCMV infection primarily reflects impaired survival, rather than impaired proliferation.

Discussion

Acute LCMV infection ordinarily elicits robust expansion of CD8 T cells, which play an important role in viral clearance. This expansion is critically dependent on the presence of MyD88. In the present study, we confirm that MyD88 plays an extrinsic role in regulating T cell expansion, as evidenced by reduced expansion and a shift in immunodominance during the response of WT CD8 T cells to LCMV infection in a Myd88−/− host. This likely reflects reduced production of type I IFNs and other proinflammatory cytokines by Myd88−/− APCs (16), which have been shown to be important in regulating T cell expansion (17). However, we also extend earlier observations (15) by using polyclonal and TCR transgenic adoptive transfers and mixed bone marrow chimeras to show that MyD88 plays a critical, T cell-intrinsic role in supporting the accumulation of Ag specific CD8 T cells during the response to acute LCMV infection.

Our results demonstrate that the clonal expansion of Myd88−/− CD8+ T cells in response to LCMV is greatly reduced. The ultimate number of CTLs at the peak of expansion reflects the combined effects of proliferation and cell survival. BrdU uptake revealed that WT and Myd88−/− Ag-specific CD8 T cells cycled at comparable rates during the polyclonal T cell response to LCMV, suggesting that the reduced numbers of Myd88−/− CD8 T cells do not appear to result from differences in proliferation. Similarly, CFSE and BrdU labeling showed that Myd88−/− P14 T cells proliferated at comparable rates to WT P14 cells throughout their period of clonal expansion, yet fail to accumulate. Consistent with this reduced accumulation, Myd88−/− P14 cells in the spleens of infected mice displayed significantly greater levels of spontaneous death following in vitro culture. Thus, it appears that naive CD8 T cells can be activated and induced to proliferate and differentiate into effector cells during the early stages of the antiviral response in the absence of MyD88. However, MyD88-dependent signals are critical in supporting their subsequent survival and accumulation.

The relative tissue distribution of Myd88−/− P14 cells differed somewhat from that of WT P14 cells following LCMV infection, with a greater proportion of the Myd88−/− cells found in the lymph nodes and peritoneum. This could suggest a role for MyD88 in regulating T cell migration. However, the percentage of Myd88−/− P14 cells was consistently lower than WT in all tissues examined, indicating that the reduced numbers of Myd88−/− cells in the spleens of LCMV infected mice did not simply reflect their sequestration in other tissues. It is also unlikely that Myd88−/− T cells die due to their preferential lysis by direct virus infection, because LCMV-Arm is not known to infect CD8 T cells, nor is infection generally cytopathic (25, 26). Furthermore, it does not appear that MyD88-dependent signals confer an indirect survival effect by altering expression of costimulatory or inhibitory molecules. Thus, the primary importance of MyD88 in T cells during LCMV infection seems to be directly related to a MyD88-dependent survival pathway.

The kinetics of the differences in the expansion of WT and Myd88−/− P14 cells to LCMV suggest that MyD88-dependent survival signals are required during the initial phase of Ag-dependent proliferation. The reduced numbers of Myd88−/− P14 population that are present at the peak of the acute response appear to contract and differentiate into a long lived effector population similarly to WT cells. Furthermore, no differences are observed in the expression of KLRG1 and IL7R, which have recently been described as markers allowing the discrimination of short-lived effector cells and long lived memory precursors (24). Hence, it appears that the MyD88-dependent signals are no longer required for CD8 T cell homeostasis when the cells progress beyond the period of Ag-driven proliferation. In this context, it is notable that the phenotype of the Myd88−/− P14 CD8 T cells following LCMV infection is strikingly similar to that of IFNαβR−/− T cells (17). Thus, like the signals induced by type I IFNs, MyD88-dependent pathways may contribute a third signal that supports the expansion of activated T cells (18).

A role for MyD88 in controlling T cell survival and expansion may partially explain our recent finding that T cell expression of MyD88 is needed for resistance to T. gondii (13). However, it is notable that the relative expansion of Myd88−/− T cells does not appear to be as severely reduced following T. gondii infection (data not shown). Thus, the MyD88 dependent pathways that regulate T cell survival may be linked to inflammatory signals that are specific to certain infections, such as LCMV. However, the specific identity of these inflammatory signals remains unclear.

We, and others, have shown that TLR2 and TLR9 agonists can directly enhance survival and proliferation of activated murine T cells in vitro, in a MyD88-dependent fashion (9, 11, 27). However, we found that Tlr2−/− and Tlr9−/− mice did not display the dramatically reduced numbers of LCMV-specific CD8 T cells found in Myd88−/− mice (data not shown). Surprisingly, the spleens of Tlr2−/− mice contained significantly more LCMV-specific CD8 T cells than WT mice (data not shown), which could potentially relate to the reduced numbers of regulatory T cells found in these mice (28). Furthermore, it has been shown that mice deficient in TLR3, TLR4, and TLR8 mount normal CD8 T cell responses to LCMV, and while LCMV-specific CTL numbers are slightly reduced in mice lacking both TLR7 and TLR9, they do not recapitulate the phenotype of Myd88−/− mice (16, 29). Similarly, we found no reduction in LCMV-specific CD8 T cells in the spleens of IL1r1−/− mice and somewhat reduced numbers in IL18r1−/− and ICE−/− mice, which still greatly exceeded those found in Myd88−/− mice.

Thus, while MyD88 plays a critical role in T cells, the upstream initiators of MyD88-dependent signaling pathways during LCMV infection remain unclear, as has been noted in other infection models (13). One possible explanation for this is that there is considerable redundancy and cooperation between different MyD88-dependent pathways. It may also be possible that MyD88-dependent viral recognition may occur through novel, as yet unidentified TLRs. A third, nonmutually exclusive, possibility is that the importance of MyD88 in CD8 T cells during LCMV infection is linked to a function outside of its traditional role in TLR signaling (30, 31). In support of this latter possibility, although costimulation of T cells with TLR agonists induces clear up-regulation of Bcl-xL in vitro (9, 11), we did not detect any significant differences in the expression of this prosurvival factor by intracellular staining between WT and Myd88−/− P14 cells following LCMV infection (data not shown).

In conclusion, our study clearly demonstrates that MyD88 plays a critical, CD8 T cell-intrinsic role during an in vivo immune response against a viral pathogen. Although MyD88 expression in CD8 T cells during LCMV infection appears to be dispensable for initial activation, proliferation, and differentiation into fate committed subsets of effector T cells, it is required for the survival and accumulation of CTLs during the phase of Ag driven proliferation. This suggests that MyD88 may play a role in regulating T cell-responses to inflammatory stimuli that are induced during LCMV infection, and deliver prosurvival signals to CTLs during clonal expansion. The nature of the inflammatory stimuli that link MyD88 to T cell survival remain unclear and raises the possibility that MyD88 may function outside of its traditional role in TLR signaling during LCMV infection. Together, our findings broaden the importance of MyD88 in T cells and may have significant implications for understanding the signals that control CD8 T cell survival during inflammatory immune responses.

Acknowledgments

We thank Hao Shen, Haina Shin, Erika Pearce, Allison Crawford, Joanna Dispirito, and Padmini Jayaraman for technical assistance and insightful discussion.

Footnotes

5

Abbreviations used in this paper: LCMV, lymphocytic choriomeningitis virus; WT, wild type; KLRG1, killer cell lectin-like receptor subfamily G.

Disclosures

The authors have no financial conflict of interest.

1

This work was supported by National Institutes of Health Grants AI-062789 (to L.A.T.), AI-070153 (to D.F.L.), AI-071309 (to E.J.W.), HHSN26620050030C (to E.J.W.), and AI-066232-01 (to S.M.K.); Burroughs-Wellcome Fund 1004313 (to S.M.K.); and the Commonwealth Universal Research Enhancement Program, Pennsylvania Department of Health (to E.J.W.).

References

  • 1.Thoma-Uszynski S, Stenger S, Takeuchi O, Ochoa MT, Engele M, Sieling PA, Barnes PF, Rollinghoff M, Bolcskei PL, Wagner M, et al. Induction of direct antimicrobial activity through mammalian toll-like receptors. Science. 2001;291:1544–1547. doi: 10.1126/science.291.5508.1544. [DOI] [PubMed] [Google Scholar]
  • 2.Iwasaki A, Medzhitov R. Toll-like receptor control of the adaptive immune responses. Nat Immunol. 2004;5:987–995. doi: 10.1038/ni1112. [DOI] [PubMed] [Google Scholar]
  • 3.Kawai T, Akira S. Innate immune recognition of viral infection. Nat Immunol. 2006;7:131–137. doi: 10.1038/ni1303. [DOI] [PubMed] [Google Scholar]
  • 4.Akira S. Mammalian Toll-like receptors. Curr Opin Immunol. 2003;15:5–11. doi: 10.1016/s0952-7915(02)00013-4. [DOI] [PubMed] [Google Scholar]
  • 5.Medzhitov R, Preston-Hurlburt P, Kopp E, Stadlen A, Chen C, Ghosh S, Janeway CA., Jr MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol Cell. 1998;2:253–258. doi: 10.1016/s1097-2765(00)80136-7. [DOI] [PubMed] [Google Scholar]
  • 6.Scanga CA, Aliberti J, Jankovic D, Tilloy F, Bennouna S, Denkers EY, Medzhitov R, Sher A. Cutting edge: MyD88 is required for resistance to Toxoplasma gondii infection and regulates parasite-induced IL-12 production by dendritic cells. J Immunol. 2002;168:5997–6001. doi: 10.4049/jimmunol.168.12.5997. [DOI] [PubMed] [Google Scholar]
  • 7.Scanga CA, Bafica A, Feng CG, Cheever AW, Hieny S, Sher A. MyD88-deficient mice display a profound loss in resistance to Mycobacterium tuberculosis associated with partially impaired Th1 cytokine and nitric oxide synthase 2 expression. Infect Immun. 2004;72:2400–2404. doi: 10.1128/IAI.72.4.2400-2404.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Lund J, Sato A, Akira S, Medzhitov R, Iwasaki A. Toll-like receptor 9-mediated recognition of Herpes simplex virus-2 by plasmacytoid dendritic cells. J Exp Med. 2003;198:513–520. doi: 10.1084/jem.20030162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Gelman AE, Zhang J, Choi Y, Turka LA. Toll-like receptor ligands directly promote activated CD4+ T cell survival. J Immunol. 2004;172:6065–6073. doi: 10.4049/jimmunol.172.10.6065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Gelman AE, LaRosa DF, Zhang J, Walsh PT, Choi Y, Sunyer JO, Turka LA. The adaptor molecule MyD88 activates PI-3 kinase signaling in CD4+ T cells and enables CpG oligodeoxynucleotide-mediated costimulation. Immunity. 2006;25:783–793. doi: 10.1016/j.immuni.2006.08.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cottalorda A, Verschelde C, Marcais A, Tomkowiak M, Musette P, Uematsu S, Akira S, Marvel J, Bonnefoy-Berard N. TLR2 engagement on CD8 T cells lowers the threshold for optimal antigen-induced T cell activation. Eur J Immunol. 2006;36:1684–1693. doi: 10.1002/eji.200636181. [DOI] [PubMed] [Google Scholar]
  • 12.Imanishi T, Hara H, Suzuki S, Suzuki N, Akira S, Saito T. Cutting edge: TLR2 directly triggers Th1 effector functions. J Immunol. 2007;178:6715–6719. doi: 10.4049/jimmunol.178.11.6715. [DOI] [PubMed] [Google Scholar]
  • 13.LaRosa DF, Stumhofer JS, Gelman AE, Rahman AH, Taylor DK, Hunter CA, Turka LA. T cell expression of MyD88 is required for resistance to Toxoplasma gondii. Proc Natl Acad Sci USA. 2008;105:3855–3860. doi: 10.1073/pnas.0706663105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Oldstone MB. Biology and pathogenesis of lymphocytic choriomeningitis virus infection. Curr Top Microbiol Immunol. 2002;263:83–117. doi: 10.1007/978-3-642-56055-2_6. [DOI] [PubMed] [Google Scholar]
  • 15.Zhou S, Kurt-Jones EA, Mandell L, Cerny A, Chan M, Golenbock DT, Finberg RW. MyD88 is critical for the development of innate and adaptive immunity during acute lymphocytic choriomeningitis virus infection. Eur J Immunol. 2005;35:822–830. doi: 10.1002/eji.200425730. [DOI] [PubMed] [Google Scholar]
  • 16.Jung A, Kato H, Kumagai Y, Kumar H, Kawai T, Takeuchi O, Akira S. Lymphocytoid choriomeningitis virus activates plasmacytoid dendritic cells and induces a cytotoxic T-cell response via MyD88. J Virol. 2008;82:196–206. doi: 10.1128/JVI.01640-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kolumam GA, Thomas S, Thompson LJ, Sprent J, Murali-Krishna K. Type I interferons act directly on CD8 T cells to allow clonal expansion and memory formation in response to viral infection. J Exp Med. 2005;202:637–650. doi: 10.1084/jem.20050821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Curtsinger JM, Valenzuela JO, Agarwal P, Lins D, Mescher MF. Type I IFNs provide a third signal to CD8 T cells to stimulate clonal expansion and differentiation. J Immunol. 2005;174:4465–4469. doi: 10.4049/jimmunol.174.8.4465. [DOI] [PubMed] [Google Scholar]
  • 19.Chen M, Barnfield C, Naslund TI, Fleeton MN, Liljestrom P. MyD88 expression is required for efficient cross-presentation of viral antigens from infected cells. J Virol. 2005;79:2964–2972. doi: 10.1128/JVI.79.5.2964-2972.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Adachi O, Kawai T, Takeda K, Matsumoto M, Tsutsui H, Sakagami M, Nakanishi K, Akira S. Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity. 1998;9:143–150. doi: 10.1016/s1074-7613(00)80596-8. [DOI] [PubMed] [Google Scholar]
  • 21.Wherry EJ, Blattman JN, Murali-Krishna K, van der Most R, Ahmed R. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J Virol. 2003;77:4911–4927. doi: 10.1128/JVI.77.8.4911-4927.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Masopust D, Murali-Krishna K, Ahmed R. Quantitating the magnitude of the lymphocytic choriomeningitis virus-specific CD8 T-cell response: it is even bigger than we thought. J Virol. 2007;81:2002–2011. doi: 10.1128/JVI.01459-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Li W, Kashiwamura S, Ueda H, Sekiyama A, Okamura H. Protection of CD8+ T cells from activation-induced cell death by IL-18. J Leukocyte Biol. 2007;82:142–151. doi: 10.1189/jlb.0706431. [DOI] [PubMed] [Google Scholar]
  • 24.Joshi NS, Cui W, Chandele A, Lee HK, Urso DR, Hagman J, Gapin L, Kaech SM. Inflammation directs memory precursor and short-lived effector CD8+ T cell fates via the graded expression of T-bet transcription factor. Immunity. 2007;27:281–295. doi: 10.1016/j.immuni.2007.07.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Homann D, McGavern DB, Oldstone MB. Visualizing the viral burden: phenotypic and functional alterations of T cells and APCs during persistent infection. J Immunol. 2004;172:6239–6250. doi: 10.4049/jimmunol.172.10.6239. [DOI] [PubMed] [Google Scholar]
  • 26.King CC, de Fries R, Kolhekar SR, Ahmed R. In vivo selection of lymphocyte-tropic and macrophage-tropic variants of lymphocytic choriomeningitis virus during persistent infection. J Virol. 1990;64:5611–5616. doi: 10.1128/jvi.64.11.5611-5616.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Komai-Koma M, Jones L, Ogg GS, Xu D, Liew FY. TLR2 is expressed on activated T cells as a costimulatory receptor. Proc Natl Acad Sci USA. 2004;101:3029–3034. doi: 10.1073/pnas.0400171101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Netea MG, Sutmuller R, Hermann C, Van der Graaf CA, Van der Meer JW, van Krieken JH, Hartung T, Adema G, Kullberg BJ. Toll-like receptor 2 suppresses immunity against Candida albicans through induction of IL-10 and regulatory T cells. J Immunol. 2004;172:3712–3718. doi: 10.4049/jimmunol.172.6.3712. [DOI] [PubMed] [Google Scholar]
  • 29.Edelmann KH, Richardson-Burns S, Alexopoulou L, Tyler KL, Flavell RA, Oldstone MB. Does Toll-like receptor 3 play a biological role in virus infections? Virology. 2004;322:231–238. doi: 10.1016/j.virol.2004.01.033. [DOI] [PubMed] [Google Scholar]
  • 30.Altemeier WA, Zhu X, Berrington WR, Harlan JM, Liles WC. Fas (CD95) induces macrophage proinflammatory chemokine production via a MyD88-dependent, caspase-independent pathway. J Leukocyte Biol. 2007;82:721–728. doi: 10.1189/jlb.1006652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Sun D, Ding A. MyD88-mediated stabilization of interferon-γ-induced cytokine and chemokine mRNA. Nat Immunol. 2006;7:375–381. doi: 10.1038/ni1308. [DOI] [PubMed] [Google Scholar]

RESOURCES