T Cell Homeostasis Is Controlled by IL-7 and IL-15 Together with Subset-Specific Factors
http://www.100md.com
免疫学杂志 2005年第8期
Abstract
Among T cell subsets, T cells uniquely display an Ag receptor-based tissue distribution, but what defines their preferential homing and homeostasis is unknown. To address this question, we studied the resources that control T cell homeostasis in secondary lymphoid organs. We found that and T cells are controlled by partially overlapping resources, because acute homeostatic proliferation of T cells was inhibited by an intact T cell compartment, and both populations were dependent on IL-7 and IL-15. Significantly, to undergo acute homeostatic proliferation, T cells also required their own depletion. Thus, T cell homeostasis is maintained by trophic cytokines commonly used by other types of lymphoid cells, as well as by additional, as yet unidentified, -specific factors.
Introduction
T cells expressing TCR constitute a significant fraction of lymphocytes in secondary lymphoid organs and blood, and predominate in mucosa and epithelia of various tissues (reviewed in Refs. 1, 2, 3, 4). In support of the role played by these cells in both innate and adaptive immunity, mice lacking T cells show a variety of abnormalities, including enhanced susceptibility to bacterial and viral infections, higher incidence of tumor development, defective wound healing, airway hyperresponsiveness, and increased inflammatory reactions (reviewed in Refs. 3 and 5). T cells with pro- or anti-inflammatory functions have also been implicated in various experimental and spontaneous models of autoimmunity (reviewed in Ref. 6).
A unique feature of T cells is the correlation between TCR structure and tissue localization. In mice, for example, V3 is primarily expressed by T cells found in the skin, V4 is mostly expressed in the female reproductive tract and the tongue, V1 and V2 predominate in lymph nodes (LNs), 5 spleen, and blood, whereas V5 defines T cells in the intestinal epithelium (1). Intriguingly, the various T cell subsets are not produced simultaneously, but rather in sequential waves during defined periods of fetal, neonatal, or adult life, a phenomenon apparently regulated by specific programs of preferential V gene rearrangements. The fact that T cell subsets produced early in life persist in the adult and can even become the dominant lymphocyte populations in the corresponding tissues implies the existence of specific homeostatic mechanisms that define the size of these cell subsets by controlling preferential homing, survival, and/or expansion. However, the factors that govern the homeostasis of T cell populations have thus far not been explored.
Lymphocyte homeostasis is a fundamental process by which the overall T and B cell pool sizes are maintained at remarkably constant levels despite the changes throughout life, including age-associated declines in lymphogenesis, expansions and contractions of Ag-engaged cell populations, conversion of selected clonotypes into long-lived memory cells, and loss of cellular functions through anergy, exhaustion, or senescence. Several studies, primarily focusing on T cells, have recently provided molecular and cellular explanations of lymphocyte homeostasis regulation (reviewed in Refs. 7, 8, 9, 10, 11). Under lymphocyte-sufficient conditions, survival (without proliferation) of naive CD4+ and CD8+ T cells depends on at least two independent signals, i.e., TCR interaction with self-peptide/MHC ligands and availability of IL-7. In contrast, survival of memory CD4+ and CD8+ T cells is MHC independent, but requires IL-7. Furthermore, the continuous stem cell-like self-renewal of memory T cells, termed "basal homeostatic proliferation", that assures the long-term maintenance of the memory T cell pool is dependent on IL-7 (and, to a lesser extent, on IL-15 and MHC recognition) for memory CD4+ T cells and on IL-15 (and, to a lesser extent, on IL-7) for memory CD8+ T cells. Additional studies showed that, under conditions of lymphopenia, T cells proliferate to reconstitute a nearly normal lymphocyte pool, a process termed "acute homeostatic proliferation". This proliferation requires self-peptide/MHC recognition and IL-7 for naive CD4+ and CD8+ T cells, whereas proliferation of memory CD4+ T cells requires either MHC or IL-7 and proliferation of memory CD8+ T cells needs either IL-7 or IL-15. Similar approaches have been used to define the homeostatic requirements for NK cells, NKT cells, and B cells. For homeostatic expansion in lymphopenic mice, NK cells require IL-15 (12), NKT cells require IL-15 (and less IL-7) but not the Ag-presenting CD1d molecule (13), and B cells require Btk-mediated signals but not IL-7 (14).
In the present study, we sought to define the homeostatic requirements of T cells. We show that, like all other major lymphocyte types, T cells from secondary lymphoid organs undergo homeostatic expansion after adoptive transfer into lymphopenic recipients. We also demonstrate that homeostatic T cell proliferation requires depletion of both and T cell populations, signaling from either IL-7 or IL-15, as well as additional T cell-specific factors.
Materials and Methods
Mice
C57BL/6 (B6, Thy1.2+ CD45.2+), B6.PL (Thy1.1+), and B6.Ly5a (CD45.1+) congenic mice were obtained from the breeding facility of The Scripps Research Institute. B6.RAG-1–/–, B6.TCR–/–, B6.TCR–/–, and B6.2M–/– mice were purchased from The Jackson Laboratory. B6.I-A–/–, B6.IL-7–/–, B6.IL-15–/–, and double-deficient B6.IL-7–/–IL-15–/– mice have been previously described (15). All mice were maintained under specific pathogen-free conditions and all experimental protocols were approved by the Institutional Animal Care Committee.
Donor cells
Peripheral T cells were enriched by negative selection using panning- and magnetic bead-based procedures. Briefly, single cell suspensions were prepared from spleen and inguinal, axillary, brachial, and cervical LN of B6.PL or B6.Ly5a mice. Splenocytes were purified by density gradient centrifugation on Lympholyte-M (Cedarlane Laboratories) and pooled with LN cells. The resulting cell suspensions were first incubated (1 h at room temperature) in flasks coated with anti-mouse IgG and anti-mouse IgM Abs (Caltag Laboratories). Unbound cells were then incubated (45 min, rotating at 4°C) with rat Abs (BD Pharmingen) specific for mouse CD4 (RM4-5), CD8 (53-6.7), CD45R (RA3-6B2), and MHC class II (M5/114.15.2). After washing with DMEM-2% FCS, cells were incubated (45 min, rotating at 4°C) with magnetic beads coated with anti-rat Ig (BioMag; Qiagen). Bead-coated cells were removed using a magnet (Advanced Magnetics), and the unbound cells were washed in DMEM-2% FCS and purified by Lympholyte-M gradient centrifugation. As assessed by flow cytometry, T cells were typically enriched to 40–50%.
Adoptive cell transfers
Donor cells consisting of either enriched T cells or total LN cells were stained with CFSE (Molecular Probes). Briefly, cells were washed in PBS-0.1% BSA, resuspended to 107 cells/ml in prewarmed (37°C) PBS-0.1% BSA containing 10 μM CFSE, incubated for 10 min at 37°C, and washed twice with cold DMEM-20% FCS and DMEM. Aliquots of 1.5–3 x 106 cells were injected i.v. into either unmanipulated or sublethally irradiated mice (exposed to 600 rad whole body irradiation 1 day before). At the indicated time points, mice were sacrificed, and LN and spleen cells were harvested and analyzed by flow cytometry. In some experiments, recipient mice were injected with recombinant mouse IL-7 (R&D Systems; 1 μg per day s.c.) for five days, starting before cell transfusion.
FACS analysis
mAbs to mouse Thy1.1, CD45.1, TCR, CD44, CD25, CD69, CD62L, IFN-, TCRV2, CD127, CD27, CD122, and streptavidin, either biotinylated or conjugated to PE, PerCP or APC, were all purchased from BD Pharmingen. For surface staining, cells were sequentially incubated with various combinations of Abs or streptavidin, washed, and analyzed on a four-color FACSCalibur (BD Biosciences). Calibration and color compensation were performed using fluorescent beads and single color controls per standard procedures (BD Biosciences). For intracellular staining of IFN-, cells were incubated with brefeldin A to inhibit intracellular protein transport and subsequently stained with appropriate Abs to surface molecules, fixed in formaldehyde, resuspended in saponin buffer containing 1 μg of anti-IFN- Ab, and analyzed on a FACSCalibur (BD Biosciences).
Immunohistochemistry
LNs and spleen from 6- to 8-wk-old B6 mice were frozen in Tissue-Tek compound (Sakura Finetek). Sections (6 μm) were acetone-fixed, air-dried, rehydrated in wash buffer (PBS supplemented with 0.7% FCS, 0.07% sodium-azide and 0.1% Tween 20) and stained for 30 min with FITC- or PE-conjugated Abs to TCR, TCR, or CD45R. After 20 min washing, sections were fixed in PBS-1% paraformaldehyde, mounted (Dako mounting medium; DakoCytomation), and examined with a Zeiss axiovert microscope and Spot 32 software.
Results
T cells undergo homeostatic expansion in sublethally irradiated mice
To study T cell homeostasis, peripheral T cells were isolated from LN and spleen of unmanipulated B6.PL (Thy1.1+) or B6.Ly5a (CD45.1+) mice, stained with CFSE, and transferred into either nonirradiated or sublethally irradiated B6 (Thy1.2+, CD45.2+) recipients. Because cell sorting using TCR-specific Abs tends to activate target cells, enrichment of T cells was accomplished by negative selection. Seven days after transfer into nonirradiated recipients, no evidence of cell division was observed, with all donor T cells uniformly retaining high levels of CFSE (Fig. 1a). In contrast, T cell homeostatic proliferation was detected in hosts rendered T cell-deficient by sublethal irradiation, as indicated by multiple peaks of decreasing CFSE intensity (Fig. 1, a and b). Thus, 60% of donor T cells recovered 7 days after transfer had undergone one to four cell divisions, with no significant differences between LN and spleen. After 14 days, the frequency of T cells that had divided increased to 86%, with >20% of the donor cells exhibiting undetectable CFSE levels, which typically indicates at least seven cell divisions.
FIGURE 1. Peripheral T cells undergo homeostatic expansion in lymphopenic recipients. Donor T cells isolated from LN and spleen of B6.Ly5a (CD45.1+) or B6.PL (Thy1.1+) mice, were labeled with CFSE and adoptively transferred into either unmanipulated or sublethally irradiated (600 rad), allelically different (CD45.2+, Thy1.2+) B6 recipients. Control mice were injected with LN cells containing CD4+ and CD8+ T cells. At day 7 or 14 posttransfer, LN and spleen cells were stained for either donor (CD45.1 or Thy1.1) T cells or donor CD4+ and CD8+ T cells and analyzed by FACS. a, Analysis of donor T cells present in the LNs of recipient mice, 7 or 14 days after infusion. Shown are a representative dot plot of live-gated cells from nonirradiated hosts (top panel), and CFSE profiles of gated donor T cells after 7 days in a nonirradiated recipient (second panel), or after 7 days (third panel) or 14 days (forth panel) in irradiated recipients. Similar data were obtained by analyzing donor T cells present in the spleen of recipient mice. b, Percentage of donor T cells in the various cell divisions as determined by CFSE profile analysis. Data represent average ± SD. , Nonirradiated hosts, 7 days posttransfer (n = 5); , irradiated hosts, 7 days posttransfer (n = 16); , irradiated hosts, 14 days posttransfer (n = 7). c, Comparison of homeostatic expansions of vs T cells. Data are percentages (average ± SD) of donor cells in the various cell divisions as determined by CFSE profile analysis 7 days after transfer into irradiated hosts. , T cells (n = 16); , CD4+ (n = 4) or CD8+ (n = 5) T cells.
Comparisons with T cell subsets at day 7 posttransfer indicated a similar frequency of proliferating cells between T cells and CD4+ T cells, although T cells had more cells in divisions 2, 3, and 4 (Fig. 1c, top panel). Moreover, although more CD8+ T cells had proliferated (88%), the overall CFSE profiles of cells in divisions 1 through >7 were similar to T cells, suggesting comparable proliferation kinetics (Fig. 1c, bottom panel).
During homeostatic proliferation, conventional T cells acquire a memory-like phenotype characterized by increased expression of CD44, but not CD69 and CD25, and loss of CD62L (9). Similar phenotypic changes were observed with T cells (Fig. 2). Before transfer, most T cells expressed low levels of CD44, CD25, and CD69, and approximately one-third exhibited high levels of CD62L (Fig. 2a). At day 14 posttransfer, CD44 expression was also low for T cells that had not divided (CFSEhigh), but progressively increased during the first three cell divisions, remaining high thereafter (Fig. 2b, top panels). Expression of CD25 and CD69 remained low after transfer, without significant changes between undivided and divided cells. However, as previously reported for purified CD4+CD25– T cells (16), a slight increase in the expression of CD25 was observed in the transferred T cells. Expression of CD62L was high in T cells that had divided one to five times, but was drastically reduced in cells that had undergone more divisions, with kinetics similar to T cells. In contrast, no changes were observed in the frequency of IFN--producing cells, as determined by intracellular staining before and after homeostatic expansion (data not shown).
FIGURE 2. T cells acquire a memory-like phenotype during homeostatic expansion. a, Expression of CD44, CD25, CD69, and CD62L by T cells freshly isolated from LNs of unmanipulated (nonlymphopenic) B6.PL mice. Numbers represent the percentage of cells expressing high levels of CD44 and CD62L, or the geometric mean of fluorescence intensity (GMFI) of CD25 and CD69 expression (n = 6). b, Changes in activation-marker expression during T cell homeostatic proliferation in lymphopenic irradiated mice. Peripheral T cells from B6.PL mice were labeled with CFSE and transfused into irradiated B6 recipients. After 14 days, donor T cells present in the LNs of recipient mice were analyzed by FACS. Expression profiles of CD44, CD25, CD69, and CD62L on gated donor T cells are shown as histograms (left column) or as dot plots as a function of CFSE intensity and cell division (right column).
Additional changes were revealed by expression analysis of TCR V2, one of the two dominant V in lymphoid organs (1). Before transfer, V2+ T cells represented 27.4 ± 1.7% of LN T cells (n = 5). However, 14 days after transfer into irradiated hosts, V2+ T cells represented 39.5 ± 3.5% of nondivided (CFSEhigh) T cells and 50.2 ± 5.0% of cells that had divided up to five times, but only 13.5 ± 2.1% of T cells in division 6, and 10.5 ± 3.5% of cells that divided more than seven times (n = 2). These results suggest that V2+ T cells exhibit slower proliferation rates compared with V2– (presumably V1+) T cells.
T cells compete with T cells during homeostatic proliferation
The above experiments indicated that peripheral T cells undergo homeostatic expansion in lymphopenic recipients, but not in lymphocyte-sufficient hosts, suggesting competition with other cells for space, ligands, or trophic cytokines. To characterize the cell types competing with T cells for homeostasis-controlling factors, T cells were adoptively transferred into mutant recipients that selectively lacked specific lymphocyte subsets. T cells proliferated in nonirradiated syngeneic RAG-1–/– mice lacking both and T cells (Fig. 3, upper panels). Because RAG-1–/– mice retain NK cells, these results suggest that, similar to T cells, but unlike NK cells (12), T cells are not significantly inhibited by NK cells during homeostatic proliferation. In addition, although more T cells proliferated in RAG-1–/– than in irradiated B6 mice (77 vs 60%), CFSE profiles indicated similar cell division kinetics in these recipients. This contrasted with T cells, which showed a dramatic increase in proliferation rates in RAG-1–/– compared with irradiated B6 hosts (Fig. 3, right upper panel).
FIGURE 3. T cell homeostatic expansion requires depletion of T cells. T cells from LNs and spleen of B6.Ly5a mice were labeled with CFSE and transfused into nonirradiated RAG-1–/– (top row, n = 3) or TCR–/– (bottom row, n = 3) congenic mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3–5/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data are shown as profiles of CFSE intensity on gated donor T cells (left column) or as percentage (average ± SD) of cells in various cell divisions as determined by CFSE profile analysis (right column).
We next examined whether selective depletion of T cells in TCR–/– mice is sufficient to initiate homeostatic proliferation of T cells. As expected, control CD4+ and CD8+ T cells did not proliferate in nonirradiated TCR–/– recipients, most likely due to inhibition by host T cells. Surprisingly, T cells also failed to proliferate when transfused in these recipients (Fig. 3, lower panels). In contrast, both and T cells proliferated in TCR–/– mice rendered broadly lymphopenic by sublethal irradiation (data not shown). Thus, T cells are sufficient to completely inhibit homeostatic proliferation of T cells in TCR–/– hosts.
Homeostatic proliferation of T cells is MHC independent
The experiments with TCR–/– recipients suggested that and T cell populations are controlled by overlapping homeostatic mechanisms. Homeostasis of T cells depends on TCR interactions with MHC/peptide-ligands and/or availability of cytokines (9). Although MHC-deficient mice exhibit normal T cell numbers (17), T cell subsets specific for either MHC class II or nonclassical MHC molecules have been identified (18, 19, 20, 21). To determine whether MHC molecules play any role in T cell homeostatic proliferation, adoptive transfer experiments were performed using 2-microglobulin–/– (2M–/–) and I-A–/– mice (Fig. 4). As expected, control CD4+ T cells proliferated normally in irradiated 2M–/– mice, but less efficiently in I-A–/– mice, while the converse was observed for CD8+ T cells (Fig. 4 and data not shown). In contrast, T cells showed virtually identical homeostatic proliferation kinetics in 2M–/–, I-A–/–, and wild-type B6 mice. These results clearly indicate that homeostatic proliferation of most peripheral T cells is not dependent on MHC recognition.
FIGURE 4. T cell homeostatic expansion is MHC independent. T cells from LNs and spleen of B6.Ly5a or B6.PL mice were labeled with CFSE and transfused into sublethally irradiated (600 rad) congenic 2M–/– (top row, n = 4), I-A–/– (bottom row, n = 3) or wild-type recipient mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data represent profiles of CFSE intensity on gated donor T cells (left column) or percentages (average ± SD) of cells in various cell divisions as determined by CFSE profile analysis (middle and right columns). Comparisons are made between percentage of cells dividing in wild-type B6 recipients (wild type (WT), ) vs 2M–/– or I-A–/– (knockout (KO), ) recipients.
T cells require either IL-7 or IL-15 for homeostatic proliferation
We next examined the possibility that inhibition of T cell homeostatic expansion by T cells could be explained by competition for IL-7 and IL-15, the main controllers of T cell homeostasis. Both IL-7 and IL-15 are known to play significant roles in T cell development and/or tissue localization (22, 23, 24, 25, 26, 27, 28), but their effect on the overall T cell homeostasis has not been studied. Like conventional T cells, most T cells in LN and spleen expressed significant levels of the IL-7R chain CD127 (Fig. 5a). In addition, a small proportion of T cells expressed higher levels of CD127 and CD44 in both LNs (3–23%, average 9.0 ± 7.1%, n = 6) and spleen (3–16%, average 6.9 ± 5.7%, n = 6). A similar cell population was almost undetectable among T cells of unmanipulated B6 mice (0.7 ± 0.2% in LNs and 0.9 ± 0.2% in spleen, n = 6), and may define an activated subset of Ag-experienced T cells (29). In support of this idea, CD127highCD44high T cells expressed low levels of CD62L and CD27 (Fig. 5a). Most T cells also expressed the IL-15R chain (CD122) at levels similar to those exhibited by most T cells (Fig. 5b). In addition, a subset of both T cells (12.9 ± 6.4% in LNs and 19.9 ± 6.7% in spleen, n = 6) and T cells (4.8 ± 1.4% in LNs and 8.1 ± 1.8% in spleen, n = 6) expressed high levels of CD122, intermediate levels of CD44, and high levels of CD62L (Fig. 5b and data not shown). Previous studies with T cells indicated that this subset corresponds to memory-phenotype CD8+ T cells (29, 30).
FIGURE 5. T cell homeostatic expansion requires either IL-7 or IL-15. a, Expression of IL-7R (CD127) vs CD44, CD62L or CD27 on gated T cells (left column) or T cells (right column) freshly isolated from LNs of unmanipulated (nonlymphopenic) B6.Ly5a mice. Data are representative of two independent experiments (n = 3 per experiment). Similar profiles were obtained with spleen cells. b, Expression of IL-15R (CD122) vs CD44. c, T cell homeostatic proliferation is inhibited in mice lacking both IL-7 and IL-15. T cells from LN and spleen of B6.Ly5a mice were labeled with CFSE and transfused into either IL-7–/–, IL-15–/–, or IL-7–/–IL-15–/– mice, or into IL-7–/–IL-15–/– mice injected with IL-7 (1 μg before transfusion, then 1 μg/day for 5 days). All mice were sublethally irradiated (600 rad) one day before transfusion. At day 7, LN and spleen cells were analyzed by FACS. Representative CFSE profiles of gated donor T cells in the LNs are shown.
To directly establish the role of IL-7 and IL-15 in T cell homeostatic proliferation, adoptive transfer studies were performed using recipients lacking one or both these cytokines. T cells proliferated in sublethally irradiated IL-7–/– recipients at least as efficiently as in irradiated B6 recipients (Fig. 5c). In fact, a small population of CFSE-negative T cells was detected in the LNs of IL-7–/– recipients, and this could reflect faster proliferation rates of IL-7-independent cells in these severely lymphopenic mice. Likewise, T cells showed normal homeostatic proliferation upon adoptive transfer into irradiated IL-15–/– mice. In contrast, homeostatic proliferation of T cells was barely detectable in irradiated IL-7–/–IL-15–/– hosts (Fig. 5c).
In view of the fundamental roles played by IL-7 and IL-15 in the development of several cell types, including T cells, B cells, NK cells, and NKT cells, and considering the interdependence of lymphocyte cellularity and lymphoid tissue development, it is conceivable that lack of homeostatic expansion in the absence of these cytokines is not due to a direct effect on T cells, but rather to a developmental defect resulting in the absence of specific cell populations required for this function. To exclude this possibility, we tested whether T cell homeostatic proliferation could be corrected in IL-7–/–IL-15–/– hosts through short-term provision of exogenous IL-7 (Fig. 5c). Indeed, homeostatic expansion of T cells was restored if the double-deficient recipients were treated with rIL-7 (1 μg/day, starting on the day of transfusion). Thus, to undergo lymphopenia-induced homeostatic expansion, T cells require either IL-7, as most T cells, or IL-15, as memory CD8+ T cells.
Requirement of T cell-specific factors for homeostatic expansion
If IL-7 and IL-15 were the only factors controlling T cell homeostasis, T cells would be expected to expand in nonirradiated TCR–/– recipients lacking T cells. Availability of IL-7 and IL-15 in these mice was demonstrated by the fact that adoptively transferred T cells extensively proliferated, at rates similar as in RAG-1–/– mice (Fig. 6). Remarkably, however, no evidence of homeostatic expansion was observed for T cells in nonirradiated TCR–/– mice (Fig. 6). Because T cells proliferated in RAG-1–/– hosts (Fig. 3), these results indicate that, in addition to T cells, T cells also restrain acute homeostatic expansion of T cells. One possibility for the T cell-mediated inhibition of homeostatic proliferation is that these cells occupy specific niches that need to be accessible during this process. However, as previously reported in humans (31), immunohistochemical analysis of mouse spleen and LNs showed that T cells do not segregate into defined zones (data not shown). Thus, it appears that the size of the T cell pool in lymphoid organs is defined by availability of cytokines commonly used by other lymphoid cells (i.e., IL-7 and IL-15) as well as by additional T cell-specific factors.
FIGURE 6. T cell homeostatic expansion requires depletion of T cells. T cells from LNs and spleen of B6.Ly5a mice were labeled with CFSE and transfused into nonirradiated TCR–/– (n = 4) congenic mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3–5/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data are shown as profiles of CFSE intensity on gated donor T cells, CD4+ T cells, or CD8+ T cells (a) or as percentage (average ± SD) of cells in various cell divisions (b) as determined by CFSE profile analysis.
Discussion
In this study, we evaluated mechanisms potentially involved in the homeostatic control of T cell populations. We found that spleen and LN T cells survive, but do not proliferate, after transfusion into lymphocyte-sufficient recipients. Similarly, no T cell expansion was observed after transfer into mice lacking either or T cells. In contrast, reductions in the cellularity of both and T cell compartments promoted T cell homeostatic proliferation, but only in mice expressing either IL-7 or IL-15. Thus, in secondary lymphoid organs, T cell homeostasis is controlled, in part, by the size of both the T cell pool, which defines availability of specific cytokines, and the T cell pool, which defines availability of additional T cell ligands. Together with previous reports documenting lymphopenia-induced homeostatic expansion of T cells (9), NKT cells (13), NK cells (12), and B cells (14), the present findings provide further support to the notion that the lymphocyte repertoire is not static, but subject to continuous dynamic changes, partially determined by the symbiotic relationship and interdependence among the various cell types with regard to overlapping resources available in limited supply.
The physiological significance of lymphopenia-induced homeostatic proliferation remains a matter of speculation. Clearly, it occurs in several clinical conditions, such as in lymphodepleted cancer patients during reconstitution with bone marrow or peripheral T cells. However, it is also possible that episodes of lymphopenia targeting subpopulations of cells in defined tissues are common during normal life, as a result of exposure to infectious agents, cytotoxic compounds, radiation or apoptosis-inducing signals. Considering that, during homeostatic expansion, T cell clonotypes with increased affinity for highly represented ligands have a selective advantage (32, 33, 34) and acquire a preactivated phenotype (9), it is conceivable that homeostatic proliferation is a mechanism that evolved, in part, to allow a rapid resetting of the lymphocyte repertoire for faster and more efficient immune responses. In support of this possibility, we (35) and others (36, 37) have shown that homeostatic expansion concurrent with immunization generates T cell populations enriched for CD8+ effectors with enhanced anti-tumor activities. Here, we report that T cells acquire several activation markers during homeostatic expansion, but whether this principle can be used to more efficiently exploit the anti-tumor potential of these cells remains to be investigated. An additional situation in which homeostatic proliferation may be part of normal physiology is during ontogeny, when the first waves of thymus-derived T cells begin populating secondary lymphoid organs. Indeed, a recent study using bone marrow and thymic-graft models showed that early thymic emigrants expressing TCR proliferate in neonatal mice in a way regulated by the interaction with self-peptide/MHC and by the size of the peripheral T cell pool (38). Our study suggests that T cells, known to be among the first to be produced early in life, are also likely to proliferate in the T cell-devoid periphery of neonates, and identifies IL-7 and IL-15 as possible modulators of such expansion.
Unlike NK cells (12), T cells strongly proliferated in nonirradiated RAG-1–/– recipients, indicating no (or limited) inhibition by NK cells. In contrast, lack of homeostatic expansion in TCR–/– mice suggested that T cells must compete with T cells for homeostasis controlling factors. To investigate the mechanisms underlying such competition, we evaluated the involvement of MHC/peptide ligands, IL-7 and IL-15, i.e., the main modulators of T cell homeostasis. Consistent with the previous observation that T cell populations are not decreased in MHC-deficient mice (17), we found normal homeostatic proliferation in 2M–/– and I-A–/– recipients. Although we cannot exclude the possibility that T cell subpopulations reactive, for example, with T10/T22 or class II MHC (18, 19, 20, 21), were impaired in these transfers, the results indicated that homeostatic proliferation of most peripheral T cells is MHC-independent and, hence, that inhibition by T cells cannot be explained by competition for TCR-ligands. In contrast, lack of homeostatic proliferation in double-deficient IL-7–/–IL-15–/– mice, as opposed to normal expansion in single-deficient (IL-7–/– or IL-15–/–) mice and in IL-7–/–IL-15–/– mice treated with rIL-7, suggested that T cell homeostatic expansion requires either IL-7 (like most T cells) or IL-15 (like memory CD8+ T cells). Other examples of homeostatic proliferation inhibition among functionally different lymphocyte subsets, such as NK, NKT, and memory CD8+ T cells have also been interpreted as reflecting overlapping cytokine requirements (13). Thus, whereas inhibition of T cell homeostatic expansion by T cells is consistent with their consumption of IL-7 and IL-15, lack of inhibition by the IL-15-dependent NK cells is likely due to the small number of NK cells, to the fact that T cells can use IL-7 when IL-15 is not available, or to different localization of and NK cells.
It is of interest that, whereas T cells inhibited T cell homeostatic expansion in TCR–/– mice, the opposite (i.e., inhibition of T cell expansion by T cells) was not observed in TCR–/– mice. One possibility could be that T cells developing in TCR–/– mice are defective, as reported for T cells of TCR–/– mice (39). However, unlike their counterpart in TCR–/– mice, T cells of TCR–/– mice exhibited an apparently normal gene expression profile (39), and inhibited homeostatic expansion of adoptively transferred T cells (this study). A more likely explanation for this unidirectional inhibition is that the larger T cell population may efficiently deplete IL-7 and IL-15, whereas the smaller T cell pool leaves sufficient levels of these factors to allow proliferation of T cells.
A significant finding in this study was that the control of the T cell pool size cannot be solely explained on the basis of competition with T cells for survival- and proliferation-promoting factors. Indeed, lack of T cells in TCR–/– mice was not associated with a significant expansion of endogenous T cells (data not shown). Moreover, unlike T cells, transfused T cells did not expand in TCR–/– hosts despite availability of IL-7 and IL-15. Thus, additional elements acting in conjunction with IL-7 and IL-15 seem to define T cell niches, including other cytokines, chemokines, adhesion molecules, or ligands for the TCR.
A major evidence suggesting that TCR ligands may be involved in T cell homeostasis is the restricted TCR usage in defined tissues. Although the mechanisms responsible for this restriction remain unknown, the involvement of specific Ags seems plausible. For example, it was shown that in the skin of mice lacking the prototypic V3+ TCR, part of the substitute T cells seem to express a TCR similar in structure (and possibly Ag specificity) to the prototypic TCR, as suggested by shared reactivity with a clonotypic Ab (40). Additional evidence includes changes in V/V usage during fetal and neonatal life (41, 42), the abnormal expansion of T cells in mice homozygous for a mutated form of the TCR signaling-adaptor linker for activation of T cells (43), as well as cellular turnover differences between TCR-transgenic and polyclonal T cells (44). The present results indicating that T cells inhibit each other during homeostatic expansion and suggesting that the expressed TCR V affects homeostatic proliferation kinetics may provide further support for a role of the TCR in T cell homeostasis.
IL-7 and IL-15 were previously recognized as playing essential roles in the biology of T cells. IL-7 signaling was shown to be absolutely required for the initiation of TCR gene rearrangements (24). Consistent with this observation, T cells were absent in IL-7–/– and IL-7R–/– mice (22, 23), but could be restored through introduction of a rearranged TCR transgene (25, 45). Likewise, T cell generation was rescued in IL-7–/– mice upon grafting of an IL-7-expressing thymus (46). Additional BrdU labeling studies with TCR-transgenic IL-7–/– mice suggested that IL-7, although not required for T cell survival, may increase the life span of proliferating cells, whereas other cytokines, such as IL-15, may be important for driving this proliferation (25), a prediction consistent with our present results. In regard to IL-15, studies indicated reductions in T cell subsets in IL-15–/–, IL-15R–/–, and IL-15R–/– mice (26, 47, 48). It was also shown that lack of V3+ dendritic epidermal T cells in IL-15–/– mice could not be corrected by adoptive transfer of wild-type thymocytes (27), and that a V3+ transgenic TCR could restore this cell subset in IL-7R–/– mice, but not in IL-15R–/– mice (45), further suggesting that IL-15 signals are redundant for T cell maturation, but required for localization in the skin. More recent studies using mixed bone marrow-chimera indicated that development of intraepithelial T cells requires IL-15R expression by bone marrow-derived cells, and IL-15 and IL-15R expression by parenchymal cells, consistent with a model in which IL-15 is trans-presented by IL-15R-expressing cells to IL-15R/c-expressing cells (28). The present findings extend this information revealing additional functions for IL-7 and IL-15 in T cell biology.
Considerable evidence indicates that, like other cells of the innate immune system, T cells are part of the first line of defense to infection and play central regulatory roles in the maintenance of tissue integrity. T cells also share several characteristics with memory T cells, particularly within the CD8 subset. As shown by phenotypic characterization, T cells appear constitutively activated and gene profile analysis indicated high expression of several cytolytic effector molecules (reviewed in Ref. 3). The present results on the homeostasis requirements provide additional evidence for the resemblance of T cells with cells of both the innate and adaptive immune systems.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Vanessa Clifton, Marieke Svoboda, and Matthew Haynes for technical assistance and M. Kat Occhipinti-Bender for editorial assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported in part by U.S. Public Health Service Grants (AR32103, AR39555, AI32751, AI52257) and the Department of Defense Breast Cancer Research Program (W81XWH-04-1-0454). This is publication number 17277-IMM from the Department of Immunology of The Scripps Research Institute.
2 R.B. and D.W. contributed to this article equally.
3 Address correspondence and reprint requests to Dr. Roberto Baccala or Dr. Argyrios N. Theofilopoulos, Department of Immunology, IMM3, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. E-mail addresses: rbaccala{at}scripps.edu or argyrio{at}scripps.edu
4 Current address: Genentech, 1 DNA Way, South San Francisco, CA 94080.
5 Abbreviations used in this paper: LN, lymph node; 2M, 2-microglobulin.
Received for publication October 28, 2004. Accepted for publication January 4, 2005.
References
Allison, J. P., W. L. Havran. 1991. The immunobiology of T cells with invariant antigen receptors. Annu. Rev. Immunol. 9:679.
Hayday, A. C.. 2000. cells: a right time and a right place for a conserved third way of protection. Annu. Rev. Immunol. 18:975.
Hayday, A., E. Theodoridis, E. Ramsburg, J. Shires. 2001. Intraepithelial lymphocytes: exploring the third way in immunology. Nat. Immunol. 2:997.
Carding, S. R., P. J. Egan. 2002. T cells: functional plasticity and heterogeneity. Nat. Rev. Immunol. 2:336.
Jameson, J., D. Witherden, W. L. Havran. 2003. T-cell effector mechanisms: and CD1d-restricted subsets. Curr. Opin. Immunol. 15:349
Yin, Z., J. Craft. 2000. T cells in autoimmunity. Springer Semin. Immunopathol. 22:311.
Marrack, P., J. Bender, D. Hildeman, M. Jordan, T. Mitchell, M. Murakami, A. Sakamoto, B. C. Schaefer, B. Swanson, J. Kappler. 2000. Homeostasis of TCR+ T cells. Nat. Immunol. 1:107.
Surh, C. D., J. Sprent. 2002. Regulation of naive and memory T-cell homeostasis. Microbes Infect. 4:51.
Jameson, S. C.. 2002. Maintaining the norm: T-cell homeostasis. Nat. Rev. Immunol. 2:547.
Schluns, K. S., L. Lefrancois. 2003. Cytokine control of memory T-cell development and survival. Nat. Rev. Immunol. 3:269.
Seddon, B., R. Zamoyska. 2003. Regulation of peripheral T-cell homeostasis by receptor signalling. Curr. Opin. Immunol. 15:321.
Prlic, M., B. R. Blazar, M. A. Farrar, S. C. Jameson. 2003. In vivo survival and homeostatic proliferation of natural killer cells. J. Exp. Med. 197:967.
Matsuda, J. L., L. Gapin, S. Sidobre, W. C. Kieper, J. T. Tan, R. Ceredig, C. D. Surh, M. Kronenberg. 2002. Homeostasis of V14i NKT cells. Nat. Immunol. 3:966.
Cabatingan, M. S., M. R. Schmidt, R. Sen, R. T. Woodland. 2002. Naive B lymphocytes undergo homeostatic proliferation in response to B cell deficit. J. Immunol. 169:6795.
Tan, J. T., B. Ernst, W. C. Kieper, E. LeRoy, J. Sprent, C. D. Surh. 2002. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J. Exp. Med. 195:1523.
Gavin, M. A., S. R. Clarke, E. Negrou, A. Gallegos, A. Rudensky. 2002. Homeostasis and anergy of CD4+CD25+ suppressor T cells in vivo. Nat. Immunol. 3:33.
Grusby, M. J., H. Auchincloss, Jr, R. Lee, R. S. Johnson, J. P. Spencer, M. Zijlstra, R. Jaenisch, V. E. Papaioannou, L. H. Glimcher. 1993. Mice lacking major histocompatibility complex class I and class II molecules. Proc. Natl. Acad. Sci. USA 90:3913
Matis, L. A., A. M. Fry, R. Q. Cron, M. M. Cotterman, R. F. Dick, J. A. Bluestone. 1989. Structure and specificity of a class II MHC alloreactive T cell receptor heterodimer. Science 245:746.
Schild, H., N. Mavaddat, C. Litzenberger, E. W. Ehrich, M. M. Davis, J. A. Bluestone, L. Matis, R. K. Draper, Y. H. Chien. 1994. The nature of major histocompatibility complex recognition by T cells. Cell 76:29.
Ito, K., L. Van Kaer, M. Bonneville, S. Hsu, D. B. Murphy, S. Tonegawa. 1990. Recognition of the product of a novel MHC TL region gene (27b) by a mouse T cell receptor. Cell 62:549.
Crowley, M. P., A. M. Fahrer, N. Baumgarth, J. Hampl, I. Gutgemann, L. Teyton, Y. Chien. 2000. A population of murine T cells that recognize an inducible MHC class Ib molecule. Science 287:314.
Moore, T. A., U. von Freeden-Jeffry, R. Murray, A. Zlotnik. 1996. Inhibition of T cell development and early thymocyte maturation in IL-7–/– mice. J. Immunol. 157:2366.
Lodolce, J. P., D. L. Boone, S. Chai, R. E. Swain, T. Dassopoulos, S. Trettin, A. Ma. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669.
De Creus, A., K. Van Beneden, F. Stevenaert, V. Debacker, J. Plum, G. Leclercq. 2002. Developmental and functional defects of thymic and epidermal V3 cells in IL-15-deficient and IFN regulatory factor-1-deficient mice. J. Immunol. 168:6486.
Schluns, K. S., E. C. Nowak, A. Cabrera-Hernandez, L. Puddington, L. Lefrancois, H. L. Aguila. 2004. Distinct cell types control lymphoid subset development by means of IL-15 and IL-15 receptor expression. Proc. Natl. Acad. Sci. USA 101:5616.
Kaech, S. M., J. T. Tan, E. J. Wherry, B. T. Konieczny, C. D. Surh, R. Ahmed. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191.
Zhang, X., S. Sun, I. Hwang, D. F. Tough, J. Sprent. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591
Ye, S. K., K. Maki, H. C. Lee, A. Ito, K. Kawai, H. Suzuki, T. W. Mak, Y. Chien, T. Honjo, K. Ikuta. 2001. Differential roles of cytokine receptors in the development of epidermal T cells. J. Immunol. 167:1929.
Laky, K., L. Lefrancois, U. von Freeden-Jeffry, R. Murray, L. Puddington. 1998. The role of IL-7 in thymic and extrathymic development of TCR cells. J. Immunol. 161:707.
Kennedy, M. K., M. Glaccum, S. N. Brown, E. A. Butz, J. L. Viney, M. Embers, N. Matsuki, K. Charrier, L. Sedger, C. R. Willis, et al 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771
Ohteki, T., H. Yoshida, T. Matsuyama, G. S. Duncan, T. W. Mak, P. S. Ohashi. 1998. The transcription factor interferon regulatory factor 1 (IRF-1) is important during the maturation of natural killer 1.1+ T cell receptor-/+ (NK1+ T) cells, natural killer cells, and intestinal intraepithelial T cells. J. Exp. Med. 187:967(Roberto Baccala, Deborah )
Among T cell subsets, T cells uniquely display an Ag receptor-based tissue distribution, but what defines their preferential homing and homeostasis is unknown. To address this question, we studied the resources that control T cell homeostasis in secondary lymphoid organs. We found that and T cells are controlled by partially overlapping resources, because acute homeostatic proliferation of T cells was inhibited by an intact T cell compartment, and both populations were dependent on IL-7 and IL-15. Significantly, to undergo acute homeostatic proliferation, T cells also required their own depletion. Thus, T cell homeostasis is maintained by trophic cytokines commonly used by other types of lymphoid cells, as well as by additional, as yet unidentified, -specific factors.
Introduction
T cells expressing TCR constitute a significant fraction of lymphocytes in secondary lymphoid organs and blood, and predominate in mucosa and epithelia of various tissues (reviewed in Refs. 1, 2, 3, 4). In support of the role played by these cells in both innate and adaptive immunity, mice lacking T cells show a variety of abnormalities, including enhanced susceptibility to bacterial and viral infections, higher incidence of tumor development, defective wound healing, airway hyperresponsiveness, and increased inflammatory reactions (reviewed in Refs. 3 and 5). T cells with pro- or anti-inflammatory functions have also been implicated in various experimental and spontaneous models of autoimmunity (reviewed in Ref. 6).
A unique feature of T cells is the correlation between TCR structure and tissue localization. In mice, for example, V3 is primarily expressed by T cells found in the skin, V4 is mostly expressed in the female reproductive tract and the tongue, V1 and V2 predominate in lymph nodes (LNs), 5 spleen, and blood, whereas V5 defines T cells in the intestinal epithelium (1). Intriguingly, the various T cell subsets are not produced simultaneously, but rather in sequential waves during defined periods of fetal, neonatal, or adult life, a phenomenon apparently regulated by specific programs of preferential V gene rearrangements. The fact that T cell subsets produced early in life persist in the adult and can even become the dominant lymphocyte populations in the corresponding tissues implies the existence of specific homeostatic mechanisms that define the size of these cell subsets by controlling preferential homing, survival, and/or expansion. However, the factors that govern the homeostasis of T cell populations have thus far not been explored.
Lymphocyte homeostasis is a fundamental process by which the overall T and B cell pool sizes are maintained at remarkably constant levels despite the changes throughout life, including age-associated declines in lymphogenesis, expansions and contractions of Ag-engaged cell populations, conversion of selected clonotypes into long-lived memory cells, and loss of cellular functions through anergy, exhaustion, or senescence. Several studies, primarily focusing on T cells, have recently provided molecular and cellular explanations of lymphocyte homeostasis regulation (reviewed in Refs. 7, 8, 9, 10, 11). Under lymphocyte-sufficient conditions, survival (without proliferation) of naive CD4+ and CD8+ T cells depends on at least two independent signals, i.e., TCR interaction with self-peptide/MHC ligands and availability of IL-7. In contrast, survival of memory CD4+ and CD8+ T cells is MHC independent, but requires IL-7. Furthermore, the continuous stem cell-like self-renewal of memory T cells, termed "basal homeostatic proliferation", that assures the long-term maintenance of the memory T cell pool is dependent on IL-7 (and, to a lesser extent, on IL-15 and MHC recognition) for memory CD4+ T cells and on IL-15 (and, to a lesser extent, on IL-7) for memory CD8+ T cells. Additional studies showed that, under conditions of lymphopenia, T cells proliferate to reconstitute a nearly normal lymphocyte pool, a process termed "acute homeostatic proliferation". This proliferation requires self-peptide/MHC recognition and IL-7 for naive CD4+ and CD8+ T cells, whereas proliferation of memory CD4+ T cells requires either MHC or IL-7 and proliferation of memory CD8+ T cells needs either IL-7 or IL-15. Similar approaches have been used to define the homeostatic requirements for NK cells, NKT cells, and B cells. For homeostatic expansion in lymphopenic mice, NK cells require IL-15 (12), NKT cells require IL-15 (and less IL-7) but not the Ag-presenting CD1d molecule (13), and B cells require Btk-mediated signals but not IL-7 (14).
In the present study, we sought to define the homeostatic requirements of T cells. We show that, like all other major lymphocyte types, T cells from secondary lymphoid organs undergo homeostatic expansion after adoptive transfer into lymphopenic recipients. We also demonstrate that homeostatic T cell proliferation requires depletion of both and T cell populations, signaling from either IL-7 or IL-15, as well as additional T cell-specific factors.
Materials and Methods
Mice
C57BL/6 (B6, Thy1.2+ CD45.2+), B6.PL (Thy1.1+), and B6.Ly5a (CD45.1+) congenic mice were obtained from the breeding facility of The Scripps Research Institute. B6.RAG-1–/–, B6.TCR–/–, B6.TCR–/–, and B6.2M–/– mice were purchased from The Jackson Laboratory. B6.I-A–/–, B6.IL-7–/–, B6.IL-15–/–, and double-deficient B6.IL-7–/–IL-15–/– mice have been previously described (15). All mice were maintained under specific pathogen-free conditions and all experimental protocols were approved by the Institutional Animal Care Committee.
Donor cells
Peripheral T cells were enriched by negative selection using panning- and magnetic bead-based procedures. Briefly, single cell suspensions were prepared from spleen and inguinal, axillary, brachial, and cervical LN of B6.PL or B6.Ly5a mice. Splenocytes were purified by density gradient centrifugation on Lympholyte-M (Cedarlane Laboratories) and pooled with LN cells. The resulting cell suspensions were first incubated (1 h at room temperature) in flasks coated with anti-mouse IgG and anti-mouse IgM Abs (Caltag Laboratories). Unbound cells were then incubated (45 min, rotating at 4°C) with rat Abs (BD Pharmingen) specific for mouse CD4 (RM4-5), CD8 (53-6.7), CD45R (RA3-6B2), and MHC class II (M5/114.15.2). After washing with DMEM-2% FCS, cells were incubated (45 min, rotating at 4°C) with magnetic beads coated with anti-rat Ig (BioMag; Qiagen). Bead-coated cells were removed using a magnet (Advanced Magnetics), and the unbound cells were washed in DMEM-2% FCS and purified by Lympholyte-M gradient centrifugation. As assessed by flow cytometry, T cells were typically enriched to 40–50%.
Adoptive cell transfers
Donor cells consisting of either enriched T cells or total LN cells were stained with CFSE (Molecular Probes). Briefly, cells were washed in PBS-0.1% BSA, resuspended to 107 cells/ml in prewarmed (37°C) PBS-0.1% BSA containing 10 μM CFSE, incubated for 10 min at 37°C, and washed twice with cold DMEM-20% FCS and DMEM. Aliquots of 1.5–3 x 106 cells were injected i.v. into either unmanipulated or sublethally irradiated mice (exposed to 600 rad whole body irradiation 1 day before). At the indicated time points, mice were sacrificed, and LN and spleen cells were harvested and analyzed by flow cytometry. In some experiments, recipient mice were injected with recombinant mouse IL-7 (R&D Systems; 1 μg per day s.c.) for five days, starting before cell transfusion.
FACS analysis
mAbs to mouse Thy1.1, CD45.1, TCR, CD44, CD25, CD69, CD62L, IFN-, TCRV2, CD127, CD27, CD122, and streptavidin, either biotinylated or conjugated to PE, PerCP or APC, were all purchased from BD Pharmingen. For surface staining, cells were sequentially incubated with various combinations of Abs or streptavidin, washed, and analyzed on a four-color FACSCalibur (BD Biosciences). Calibration and color compensation were performed using fluorescent beads and single color controls per standard procedures (BD Biosciences). For intracellular staining of IFN-, cells were incubated with brefeldin A to inhibit intracellular protein transport and subsequently stained with appropriate Abs to surface molecules, fixed in formaldehyde, resuspended in saponin buffer containing 1 μg of anti-IFN- Ab, and analyzed on a FACSCalibur (BD Biosciences).
Immunohistochemistry
LNs and spleen from 6- to 8-wk-old B6 mice were frozen in Tissue-Tek compound (Sakura Finetek). Sections (6 μm) were acetone-fixed, air-dried, rehydrated in wash buffer (PBS supplemented with 0.7% FCS, 0.07% sodium-azide and 0.1% Tween 20) and stained for 30 min with FITC- or PE-conjugated Abs to TCR, TCR, or CD45R. After 20 min washing, sections were fixed in PBS-1% paraformaldehyde, mounted (Dako mounting medium; DakoCytomation), and examined with a Zeiss axiovert microscope and Spot 32 software.
Results
T cells undergo homeostatic expansion in sublethally irradiated mice
To study T cell homeostasis, peripheral T cells were isolated from LN and spleen of unmanipulated B6.PL (Thy1.1+) or B6.Ly5a (CD45.1+) mice, stained with CFSE, and transferred into either nonirradiated or sublethally irradiated B6 (Thy1.2+, CD45.2+) recipients. Because cell sorting using TCR-specific Abs tends to activate target cells, enrichment of T cells was accomplished by negative selection. Seven days after transfer into nonirradiated recipients, no evidence of cell division was observed, with all donor T cells uniformly retaining high levels of CFSE (Fig. 1a). In contrast, T cell homeostatic proliferation was detected in hosts rendered T cell-deficient by sublethal irradiation, as indicated by multiple peaks of decreasing CFSE intensity (Fig. 1, a and b). Thus, 60% of donor T cells recovered 7 days after transfer had undergone one to four cell divisions, with no significant differences between LN and spleen. After 14 days, the frequency of T cells that had divided increased to 86%, with >20% of the donor cells exhibiting undetectable CFSE levels, which typically indicates at least seven cell divisions.
FIGURE 1. Peripheral T cells undergo homeostatic expansion in lymphopenic recipients. Donor T cells isolated from LN and spleen of B6.Ly5a (CD45.1+) or B6.PL (Thy1.1+) mice, were labeled with CFSE and adoptively transferred into either unmanipulated or sublethally irradiated (600 rad), allelically different (CD45.2+, Thy1.2+) B6 recipients. Control mice were injected with LN cells containing CD4+ and CD8+ T cells. At day 7 or 14 posttransfer, LN and spleen cells were stained for either donor (CD45.1 or Thy1.1) T cells or donor CD4+ and CD8+ T cells and analyzed by FACS. a, Analysis of donor T cells present in the LNs of recipient mice, 7 or 14 days after infusion. Shown are a representative dot plot of live-gated cells from nonirradiated hosts (top panel), and CFSE profiles of gated donor T cells after 7 days in a nonirradiated recipient (second panel), or after 7 days (third panel) or 14 days (forth panel) in irradiated recipients. Similar data were obtained by analyzing donor T cells present in the spleen of recipient mice. b, Percentage of donor T cells in the various cell divisions as determined by CFSE profile analysis. Data represent average ± SD. , Nonirradiated hosts, 7 days posttransfer (n = 5); , irradiated hosts, 7 days posttransfer (n = 16); , irradiated hosts, 14 days posttransfer (n = 7). c, Comparison of homeostatic expansions of vs T cells. Data are percentages (average ± SD) of donor cells in the various cell divisions as determined by CFSE profile analysis 7 days after transfer into irradiated hosts. , T cells (n = 16); , CD4+ (n = 4) or CD8+ (n = 5) T cells.
Comparisons with T cell subsets at day 7 posttransfer indicated a similar frequency of proliferating cells between T cells and CD4+ T cells, although T cells had more cells in divisions 2, 3, and 4 (Fig. 1c, top panel). Moreover, although more CD8+ T cells had proliferated (88%), the overall CFSE profiles of cells in divisions 1 through >7 were similar to T cells, suggesting comparable proliferation kinetics (Fig. 1c, bottom panel).
During homeostatic proliferation, conventional T cells acquire a memory-like phenotype characterized by increased expression of CD44, but not CD69 and CD25, and loss of CD62L (9). Similar phenotypic changes were observed with T cells (Fig. 2). Before transfer, most T cells expressed low levels of CD44, CD25, and CD69, and approximately one-third exhibited high levels of CD62L (Fig. 2a). At day 14 posttransfer, CD44 expression was also low for T cells that had not divided (CFSEhigh), but progressively increased during the first three cell divisions, remaining high thereafter (Fig. 2b, top panels). Expression of CD25 and CD69 remained low after transfer, without significant changes between undivided and divided cells. However, as previously reported for purified CD4+CD25– T cells (16), a slight increase in the expression of CD25 was observed in the transferred T cells. Expression of CD62L was high in T cells that had divided one to five times, but was drastically reduced in cells that had undergone more divisions, with kinetics similar to T cells. In contrast, no changes were observed in the frequency of IFN--producing cells, as determined by intracellular staining before and after homeostatic expansion (data not shown).
FIGURE 2. T cells acquire a memory-like phenotype during homeostatic expansion. a, Expression of CD44, CD25, CD69, and CD62L by T cells freshly isolated from LNs of unmanipulated (nonlymphopenic) B6.PL mice. Numbers represent the percentage of cells expressing high levels of CD44 and CD62L, or the geometric mean of fluorescence intensity (GMFI) of CD25 and CD69 expression (n = 6). b, Changes in activation-marker expression during T cell homeostatic proliferation in lymphopenic irradiated mice. Peripheral T cells from B6.PL mice were labeled with CFSE and transfused into irradiated B6 recipients. After 14 days, donor T cells present in the LNs of recipient mice were analyzed by FACS. Expression profiles of CD44, CD25, CD69, and CD62L on gated donor T cells are shown as histograms (left column) or as dot plots as a function of CFSE intensity and cell division (right column).
Additional changes were revealed by expression analysis of TCR V2, one of the two dominant V in lymphoid organs (1). Before transfer, V2+ T cells represented 27.4 ± 1.7% of LN T cells (n = 5). However, 14 days after transfer into irradiated hosts, V2+ T cells represented 39.5 ± 3.5% of nondivided (CFSEhigh) T cells and 50.2 ± 5.0% of cells that had divided up to five times, but only 13.5 ± 2.1% of T cells in division 6, and 10.5 ± 3.5% of cells that divided more than seven times (n = 2). These results suggest that V2+ T cells exhibit slower proliferation rates compared with V2– (presumably V1+) T cells.
T cells compete with T cells during homeostatic proliferation
The above experiments indicated that peripheral T cells undergo homeostatic expansion in lymphopenic recipients, but not in lymphocyte-sufficient hosts, suggesting competition with other cells for space, ligands, or trophic cytokines. To characterize the cell types competing with T cells for homeostasis-controlling factors, T cells were adoptively transferred into mutant recipients that selectively lacked specific lymphocyte subsets. T cells proliferated in nonirradiated syngeneic RAG-1–/– mice lacking both and T cells (Fig. 3, upper panels). Because RAG-1–/– mice retain NK cells, these results suggest that, similar to T cells, but unlike NK cells (12), T cells are not significantly inhibited by NK cells during homeostatic proliferation. In addition, although more T cells proliferated in RAG-1–/– than in irradiated B6 mice (77 vs 60%), CFSE profiles indicated similar cell division kinetics in these recipients. This contrasted with T cells, which showed a dramatic increase in proliferation rates in RAG-1–/– compared with irradiated B6 hosts (Fig. 3, right upper panel).
FIGURE 3. T cell homeostatic expansion requires depletion of T cells. T cells from LNs and spleen of B6.Ly5a mice were labeled with CFSE and transfused into nonirradiated RAG-1–/– (top row, n = 3) or TCR–/– (bottom row, n = 3) congenic mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3–5/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data are shown as profiles of CFSE intensity on gated donor T cells (left column) or as percentage (average ± SD) of cells in various cell divisions as determined by CFSE profile analysis (right column).
We next examined whether selective depletion of T cells in TCR–/– mice is sufficient to initiate homeostatic proliferation of T cells. As expected, control CD4+ and CD8+ T cells did not proliferate in nonirradiated TCR–/– recipients, most likely due to inhibition by host T cells. Surprisingly, T cells also failed to proliferate when transfused in these recipients (Fig. 3, lower panels). In contrast, both and T cells proliferated in TCR–/– mice rendered broadly lymphopenic by sublethal irradiation (data not shown). Thus, T cells are sufficient to completely inhibit homeostatic proliferation of T cells in TCR–/– hosts.
Homeostatic proliferation of T cells is MHC independent
The experiments with TCR–/– recipients suggested that and T cell populations are controlled by overlapping homeostatic mechanisms. Homeostasis of T cells depends on TCR interactions with MHC/peptide-ligands and/or availability of cytokines (9). Although MHC-deficient mice exhibit normal T cell numbers (17), T cell subsets specific for either MHC class II or nonclassical MHC molecules have been identified (18, 19, 20, 21). To determine whether MHC molecules play any role in T cell homeostatic proliferation, adoptive transfer experiments were performed using 2-microglobulin–/– (2M–/–) and I-A–/– mice (Fig. 4). As expected, control CD4+ T cells proliferated normally in irradiated 2M–/– mice, but less efficiently in I-A–/– mice, while the converse was observed for CD8+ T cells (Fig. 4 and data not shown). In contrast, T cells showed virtually identical homeostatic proliferation kinetics in 2M–/–, I-A–/–, and wild-type B6 mice. These results clearly indicate that homeostatic proliferation of most peripheral T cells is not dependent on MHC recognition.
FIGURE 4. T cell homeostatic expansion is MHC independent. T cells from LNs and spleen of B6.Ly5a or B6.PL mice were labeled with CFSE and transfused into sublethally irradiated (600 rad) congenic 2M–/– (top row, n = 4), I-A–/– (bottom row, n = 3) or wild-type recipient mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data represent profiles of CFSE intensity on gated donor T cells (left column) or percentages (average ± SD) of cells in various cell divisions as determined by CFSE profile analysis (middle and right columns). Comparisons are made between percentage of cells dividing in wild-type B6 recipients (wild type (WT), ) vs 2M–/– or I-A–/– (knockout (KO), ) recipients.
T cells require either IL-7 or IL-15 for homeostatic proliferation
We next examined the possibility that inhibition of T cell homeostatic expansion by T cells could be explained by competition for IL-7 and IL-15, the main controllers of T cell homeostasis. Both IL-7 and IL-15 are known to play significant roles in T cell development and/or tissue localization (22, 23, 24, 25, 26, 27, 28), but their effect on the overall T cell homeostasis has not been studied. Like conventional T cells, most T cells in LN and spleen expressed significant levels of the IL-7R chain CD127 (Fig. 5a). In addition, a small proportion of T cells expressed higher levels of CD127 and CD44 in both LNs (3–23%, average 9.0 ± 7.1%, n = 6) and spleen (3–16%, average 6.9 ± 5.7%, n = 6). A similar cell population was almost undetectable among T cells of unmanipulated B6 mice (0.7 ± 0.2% in LNs and 0.9 ± 0.2% in spleen, n = 6), and may define an activated subset of Ag-experienced T cells (29). In support of this idea, CD127highCD44high T cells expressed low levels of CD62L and CD27 (Fig. 5a). Most T cells also expressed the IL-15R chain (CD122) at levels similar to those exhibited by most T cells (Fig. 5b). In addition, a subset of both T cells (12.9 ± 6.4% in LNs and 19.9 ± 6.7% in spleen, n = 6) and T cells (4.8 ± 1.4% in LNs and 8.1 ± 1.8% in spleen, n = 6) expressed high levels of CD122, intermediate levels of CD44, and high levels of CD62L (Fig. 5b and data not shown). Previous studies with T cells indicated that this subset corresponds to memory-phenotype CD8+ T cells (29, 30).
FIGURE 5. T cell homeostatic expansion requires either IL-7 or IL-15. a, Expression of IL-7R (CD127) vs CD44, CD62L or CD27 on gated T cells (left column) or T cells (right column) freshly isolated from LNs of unmanipulated (nonlymphopenic) B6.Ly5a mice. Data are representative of two independent experiments (n = 3 per experiment). Similar profiles were obtained with spleen cells. b, Expression of IL-15R (CD122) vs CD44. c, T cell homeostatic proliferation is inhibited in mice lacking both IL-7 and IL-15. T cells from LN and spleen of B6.Ly5a mice were labeled with CFSE and transfused into either IL-7–/–, IL-15–/–, or IL-7–/–IL-15–/– mice, or into IL-7–/–IL-15–/– mice injected with IL-7 (1 μg before transfusion, then 1 μg/day for 5 days). All mice were sublethally irradiated (600 rad) one day before transfusion. At day 7, LN and spleen cells were analyzed by FACS. Representative CFSE profiles of gated donor T cells in the LNs are shown.
To directly establish the role of IL-7 and IL-15 in T cell homeostatic proliferation, adoptive transfer studies were performed using recipients lacking one or both these cytokines. T cells proliferated in sublethally irradiated IL-7–/– recipients at least as efficiently as in irradiated B6 recipients (Fig. 5c). In fact, a small population of CFSE-negative T cells was detected in the LNs of IL-7–/– recipients, and this could reflect faster proliferation rates of IL-7-independent cells in these severely lymphopenic mice. Likewise, T cells showed normal homeostatic proliferation upon adoptive transfer into irradiated IL-15–/– mice. In contrast, homeostatic proliferation of T cells was barely detectable in irradiated IL-7–/–IL-15–/– hosts (Fig. 5c).
In view of the fundamental roles played by IL-7 and IL-15 in the development of several cell types, including T cells, B cells, NK cells, and NKT cells, and considering the interdependence of lymphocyte cellularity and lymphoid tissue development, it is conceivable that lack of homeostatic expansion in the absence of these cytokines is not due to a direct effect on T cells, but rather to a developmental defect resulting in the absence of specific cell populations required for this function. To exclude this possibility, we tested whether T cell homeostatic proliferation could be corrected in IL-7–/–IL-15–/– hosts through short-term provision of exogenous IL-7 (Fig. 5c). Indeed, homeostatic expansion of T cells was restored if the double-deficient recipients were treated with rIL-7 (1 μg/day, starting on the day of transfusion). Thus, to undergo lymphopenia-induced homeostatic expansion, T cells require either IL-7, as most T cells, or IL-15, as memory CD8+ T cells.
Requirement of T cell-specific factors for homeostatic expansion
If IL-7 and IL-15 were the only factors controlling T cell homeostasis, T cells would be expected to expand in nonirradiated TCR–/– recipients lacking T cells. Availability of IL-7 and IL-15 in these mice was demonstrated by the fact that adoptively transferred T cells extensively proliferated, at rates similar as in RAG-1–/– mice (Fig. 6). Remarkably, however, no evidence of homeostatic expansion was observed for T cells in nonirradiated TCR–/– mice (Fig. 6). Because T cells proliferated in RAG-1–/– hosts (Fig. 3), these results indicate that, in addition to T cells, T cells also restrain acute homeostatic expansion of T cells. One possibility for the T cell-mediated inhibition of homeostatic proliferation is that these cells occupy specific niches that need to be accessible during this process. However, as previously reported in humans (31), immunohistochemical analysis of mouse spleen and LNs showed that T cells do not segregate into defined zones (data not shown). Thus, it appears that the size of the T cell pool in lymphoid organs is defined by availability of cytokines commonly used by other lymphoid cells (i.e., IL-7 and IL-15) as well as by additional T cell-specific factors.
FIGURE 6. T cell homeostatic expansion requires depletion of T cells. T cells from LNs and spleen of B6.Ly5a mice were labeled with CFSE and transfused into nonirradiated TCR–/– (n = 4) congenic mice. Control mice were injected with LN cells containing CD4+ and CD8+ T cells (n = 3–5/group). After 7 days, donor T cells present in LNs and spleen of recipient mice were analyzed by FACS. Data are shown as profiles of CFSE intensity on gated donor T cells, CD4+ T cells, or CD8+ T cells (a) or as percentage (average ± SD) of cells in various cell divisions (b) as determined by CFSE profile analysis.
Discussion
In this study, we evaluated mechanisms potentially involved in the homeostatic control of T cell populations. We found that spleen and LN T cells survive, but do not proliferate, after transfusion into lymphocyte-sufficient recipients. Similarly, no T cell expansion was observed after transfer into mice lacking either or T cells. In contrast, reductions in the cellularity of both and T cell compartments promoted T cell homeostatic proliferation, but only in mice expressing either IL-7 or IL-15. Thus, in secondary lymphoid organs, T cell homeostasis is controlled, in part, by the size of both the T cell pool, which defines availability of specific cytokines, and the T cell pool, which defines availability of additional T cell ligands. Together with previous reports documenting lymphopenia-induced homeostatic expansion of T cells (9), NKT cells (13), NK cells (12), and B cells (14), the present findings provide further support to the notion that the lymphocyte repertoire is not static, but subject to continuous dynamic changes, partially determined by the symbiotic relationship and interdependence among the various cell types with regard to overlapping resources available in limited supply.
The physiological significance of lymphopenia-induced homeostatic proliferation remains a matter of speculation. Clearly, it occurs in several clinical conditions, such as in lymphodepleted cancer patients during reconstitution with bone marrow or peripheral T cells. However, it is also possible that episodes of lymphopenia targeting subpopulations of cells in defined tissues are common during normal life, as a result of exposure to infectious agents, cytotoxic compounds, radiation or apoptosis-inducing signals. Considering that, during homeostatic expansion, T cell clonotypes with increased affinity for highly represented ligands have a selective advantage (32, 33, 34) and acquire a preactivated phenotype (9), it is conceivable that homeostatic proliferation is a mechanism that evolved, in part, to allow a rapid resetting of the lymphocyte repertoire for faster and more efficient immune responses. In support of this possibility, we (35) and others (36, 37) have shown that homeostatic expansion concurrent with immunization generates T cell populations enriched for CD8+ effectors with enhanced anti-tumor activities. Here, we report that T cells acquire several activation markers during homeostatic expansion, but whether this principle can be used to more efficiently exploit the anti-tumor potential of these cells remains to be investigated. An additional situation in which homeostatic proliferation may be part of normal physiology is during ontogeny, when the first waves of thymus-derived T cells begin populating secondary lymphoid organs. Indeed, a recent study using bone marrow and thymic-graft models showed that early thymic emigrants expressing TCR proliferate in neonatal mice in a way regulated by the interaction with self-peptide/MHC and by the size of the peripheral T cell pool (38). Our study suggests that T cells, known to be among the first to be produced early in life, are also likely to proliferate in the T cell-devoid periphery of neonates, and identifies IL-7 and IL-15 as possible modulators of such expansion.
Unlike NK cells (12), T cells strongly proliferated in nonirradiated RAG-1–/– recipients, indicating no (or limited) inhibition by NK cells. In contrast, lack of homeostatic expansion in TCR–/– mice suggested that T cells must compete with T cells for homeostasis controlling factors. To investigate the mechanisms underlying such competition, we evaluated the involvement of MHC/peptide ligands, IL-7 and IL-15, i.e., the main modulators of T cell homeostasis. Consistent with the previous observation that T cell populations are not decreased in MHC-deficient mice (17), we found normal homeostatic proliferation in 2M–/– and I-A–/– recipients. Although we cannot exclude the possibility that T cell subpopulations reactive, for example, with T10/T22 or class II MHC (18, 19, 20, 21), were impaired in these transfers, the results indicated that homeostatic proliferation of most peripheral T cells is MHC-independent and, hence, that inhibition by T cells cannot be explained by competition for TCR-ligands. In contrast, lack of homeostatic proliferation in double-deficient IL-7–/–IL-15–/– mice, as opposed to normal expansion in single-deficient (IL-7–/– or IL-15–/–) mice and in IL-7–/–IL-15–/– mice treated with rIL-7, suggested that T cell homeostatic expansion requires either IL-7 (like most T cells) or IL-15 (like memory CD8+ T cells). Other examples of homeostatic proliferation inhibition among functionally different lymphocyte subsets, such as NK, NKT, and memory CD8+ T cells have also been interpreted as reflecting overlapping cytokine requirements (13). Thus, whereas inhibition of T cell homeostatic expansion by T cells is consistent with their consumption of IL-7 and IL-15, lack of inhibition by the IL-15-dependent NK cells is likely due to the small number of NK cells, to the fact that T cells can use IL-7 when IL-15 is not available, or to different localization of and NK cells.
It is of interest that, whereas T cells inhibited T cell homeostatic expansion in TCR–/– mice, the opposite (i.e., inhibition of T cell expansion by T cells) was not observed in TCR–/– mice. One possibility could be that T cells developing in TCR–/– mice are defective, as reported for T cells of TCR–/– mice (39). However, unlike their counterpart in TCR–/– mice, T cells of TCR–/– mice exhibited an apparently normal gene expression profile (39), and inhibited homeostatic expansion of adoptively transferred T cells (this study). A more likely explanation for this unidirectional inhibition is that the larger T cell population may efficiently deplete IL-7 and IL-15, whereas the smaller T cell pool leaves sufficient levels of these factors to allow proliferation of T cells.
A significant finding in this study was that the control of the T cell pool size cannot be solely explained on the basis of competition with T cells for survival- and proliferation-promoting factors. Indeed, lack of T cells in TCR–/– mice was not associated with a significant expansion of endogenous T cells (data not shown). Moreover, unlike T cells, transfused T cells did not expand in TCR–/– hosts despite availability of IL-7 and IL-15. Thus, additional elements acting in conjunction with IL-7 and IL-15 seem to define T cell niches, including other cytokines, chemokines, adhesion molecules, or ligands for the TCR.
A major evidence suggesting that TCR ligands may be involved in T cell homeostasis is the restricted TCR usage in defined tissues. Although the mechanisms responsible for this restriction remain unknown, the involvement of specific Ags seems plausible. For example, it was shown that in the skin of mice lacking the prototypic V3+ TCR, part of the substitute T cells seem to express a TCR similar in structure (and possibly Ag specificity) to the prototypic TCR, as suggested by shared reactivity with a clonotypic Ab (40). Additional evidence includes changes in V/V usage during fetal and neonatal life (41, 42), the abnormal expansion of T cells in mice homozygous for a mutated form of the TCR signaling-adaptor linker for activation of T cells (43), as well as cellular turnover differences between TCR-transgenic and polyclonal T cells (44). The present results indicating that T cells inhibit each other during homeostatic expansion and suggesting that the expressed TCR V affects homeostatic proliferation kinetics may provide further support for a role of the TCR in T cell homeostasis.
IL-7 and IL-15 were previously recognized as playing essential roles in the biology of T cells. IL-7 signaling was shown to be absolutely required for the initiation of TCR gene rearrangements (24). Consistent with this observation, T cells were absent in IL-7–/– and IL-7R–/– mice (22, 23), but could be restored through introduction of a rearranged TCR transgene (25, 45). Likewise, T cell generation was rescued in IL-7–/– mice upon grafting of an IL-7-expressing thymus (46). Additional BrdU labeling studies with TCR-transgenic IL-7–/– mice suggested that IL-7, although not required for T cell survival, may increase the life span of proliferating cells, whereas other cytokines, such as IL-15, may be important for driving this proliferation (25), a prediction consistent with our present results. In regard to IL-15, studies indicated reductions in T cell subsets in IL-15–/–, IL-15R–/–, and IL-15R–/– mice (26, 47, 48). It was also shown that lack of V3+ dendritic epidermal T cells in IL-15–/– mice could not be corrected by adoptive transfer of wild-type thymocytes (27), and that a V3+ transgenic TCR could restore this cell subset in IL-7R–/– mice, but not in IL-15R–/– mice (45), further suggesting that IL-15 signals are redundant for T cell maturation, but required for localization in the skin. More recent studies using mixed bone marrow-chimera indicated that development of intraepithelial T cells requires IL-15R expression by bone marrow-derived cells, and IL-15 and IL-15R expression by parenchymal cells, consistent with a model in which IL-15 is trans-presented by IL-15R-expressing cells to IL-15R/c-expressing cells (28). The present findings extend this information revealing additional functions for IL-7 and IL-15 in T cell biology.
Considerable evidence indicates that, like other cells of the innate immune system, T cells are part of the first line of defense to infection and play central regulatory roles in the maintenance of tissue integrity. T cells also share several characteristics with memory T cells, particularly within the CD8 subset. As shown by phenotypic characterization, T cells appear constitutively activated and gene profile analysis indicated high expression of several cytolytic effector molecules (reviewed in Ref. 3). The present results on the homeostasis requirements provide additional evidence for the resemblance of T cells with cells of both the innate and adaptive immune systems.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Vanessa Clifton, Marieke Svoboda, and Matthew Haynes for technical assistance and M. Kat Occhipinti-Bender for editorial assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported in part by U.S. Public Health Service Grants (AR32103, AR39555, AI32751, AI52257) and the Department of Defense Breast Cancer Research Program (W81XWH-04-1-0454). This is publication number 17277-IMM from the Department of Immunology of The Scripps Research Institute.
2 R.B. and D.W. contributed to this article equally.
3 Address correspondence and reprint requests to Dr. Roberto Baccala or Dr. Argyrios N. Theofilopoulos, Department of Immunology, IMM3, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. E-mail addresses: rbaccala{at}scripps.edu or argyrio{at}scripps.edu
4 Current address: Genentech, 1 DNA Way, South San Francisco, CA 94080.
5 Abbreviations used in this paper: LN, lymph node; 2M, 2-microglobulin.
Received for publication October 28, 2004. Accepted for publication January 4, 2005.
References
Allison, J. P., W. L. Havran. 1991. The immunobiology of T cells with invariant antigen receptors. Annu. Rev. Immunol. 9:679.
Hayday, A. C.. 2000. cells: a right time and a right place for a conserved third way of protection. Annu. Rev. Immunol. 18:975.
Hayday, A., E. Theodoridis, E. Ramsburg, J. Shires. 2001. Intraepithelial lymphocytes: exploring the third way in immunology. Nat. Immunol. 2:997.
Carding, S. R., P. J. Egan. 2002. T cells: functional plasticity and heterogeneity. Nat. Rev. Immunol. 2:336.
Jameson, J., D. Witherden, W. L. Havran. 2003. T-cell effector mechanisms: and CD1d-restricted subsets. Curr. Opin. Immunol. 15:349
Yin, Z., J. Craft. 2000. T cells in autoimmunity. Springer Semin. Immunopathol. 22:311.
Marrack, P., J. Bender, D. Hildeman, M. Jordan, T. Mitchell, M. Murakami, A. Sakamoto, B. C. Schaefer, B. Swanson, J. Kappler. 2000. Homeostasis of TCR+ T cells. Nat. Immunol. 1:107.
Surh, C. D., J. Sprent. 2002. Regulation of naive and memory T-cell homeostasis. Microbes Infect. 4:51.
Jameson, S. C.. 2002. Maintaining the norm: T-cell homeostasis. Nat. Rev. Immunol. 2:547.
Schluns, K. S., L. Lefrancois. 2003. Cytokine control of memory T-cell development and survival. Nat. Rev. Immunol. 3:269.
Seddon, B., R. Zamoyska. 2003. Regulation of peripheral T-cell homeostasis by receptor signalling. Curr. Opin. Immunol. 15:321.
Prlic, M., B. R. Blazar, M. A. Farrar, S. C. Jameson. 2003. In vivo survival and homeostatic proliferation of natural killer cells. J. Exp. Med. 197:967.
Matsuda, J. L., L. Gapin, S. Sidobre, W. C. Kieper, J. T. Tan, R. Ceredig, C. D. Surh, M. Kronenberg. 2002. Homeostasis of V14i NKT cells. Nat. Immunol. 3:966.
Cabatingan, M. S., M. R. Schmidt, R. Sen, R. T. Woodland. 2002. Naive B lymphocytes undergo homeostatic proliferation in response to B cell deficit. J. Immunol. 169:6795.
Tan, J. T., B. Ernst, W. C. Kieper, E. LeRoy, J. Sprent, C. D. Surh. 2002. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J. Exp. Med. 195:1523.
Gavin, M. A., S. R. Clarke, E. Negrou, A. Gallegos, A. Rudensky. 2002. Homeostasis and anergy of CD4+CD25+ suppressor T cells in vivo. Nat. Immunol. 3:33.
Grusby, M. J., H. Auchincloss, Jr, R. Lee, R. S. Johnson, J. P. Spencer, M. Zijlstra, R. Jaenisch, V. E. Papaioannou, L. H. Glimcher. 1993. Mice lacking major histocompatibility complex class I and class II molecules. Proc. Natl. Acad. Sci. USA 90:3913
Matis, L. A., A. M. Fry, R. Q. Cron, M. M. Cotterman, R. F. Dick, J. A. Bluestone. 1989. Structure and specificity of a class II MHC alloreactive T cell receptor heterodimer. Science 245:746.
Schild, H., N. Mavaddat, C. Litzenberger, E. W. Ehrich, M. M. Davis, J. A. Bluestone, L. Matis, R. K. Draper, Y. H. Chien. 1994. The nature of major histocompatibility complex recognition by T cells. Cell 76:29.
Ito, K., L. Van Kaer, M. Bonneville, S. Hsu, D. B. Murphy, S. Tonegawa. 1990. Recognition of the product of a novel MHC TL region gene (27b) by a mouse T cell receptor. Cell 62:549.
Crowley, M. P., A. M. Fahrer, N. Baumgarth, J. Hampl, I. Gutgemann, L. Teyton, Y. Chien. 2000. A population of murine T cells that recognize an inducible MHC class Ib molecule. Science 287:314.
Moore, T. A., U. von Freeden-Jeffry, R. Murray, A. Zlotnik. 1996. Inhibition of T cell development and early thymocyte maturation in IL-7–/– mice. J. Immunol. 157:2366.
Lodolce, J. P., D. L. Boone, S. Chai, R. E. Swain, T. Dassopoulos, S. Trettin, A. Ma. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669.
De Creus, A., K. Van Beneden, F. Stevenaert, V. Debacker, J. Plum, G. Leclercq. 2002. Developmental and functional defects of thymic and epidermal V3 cells in IL-15-deficient and IFN regulatory factor-1-deficient mice. J. Immunol. 168:6486.
Schluns, K. S., E. C. Nowak, A. Cabrera-Hernandez, L. Puddington, L. Lefrancois, H. L. Aguila. 2004. Distinct cell types control lymphoid subset development by means of IL-15 and IL-15 receptor expression. Proc. Natl. Acad. Sci. USA 101:5616.
Kaech, S. M., J. T. Tan, E. J. Wherry, B. T. Konieczny, C. D. Surh, R. Ahmed. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191.
Zhang, X., S. Sun, I. Hwang, D. F. Tough, J. Sprent. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591
Ye, S. K., K. Maki, H. C. Lee, A. Ito, K. Kawai, H. Suzuki, T. W. Mak, Y. Chien, T. Honjo, K. Ikuta. 2001. Differential roles of cytokine receptors in the development of epidermal T cells. J. Immunol. 167:1929.
Laky, K., L. Lefrancois, U. von Freeden-Jeffry, R. Murray, L. Puddington. 1998. The role of IL-7 in thymic and extrathymic development of TCR cells. J. Immunol. 161:707.
Kennedy, M. K., M. Glaccum, S. N. Brown, E. A. Butz, J. L. Viney, M. Embers, N. Matsuki, K. Charrier, L. Sedger, C. R. Willis, et al 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771
Ohteki, T., H. Yoshida, T. Matsuyama, G. S. Duncan, T. W. Mak, P. S. Ohashi. 1998. The transcription factor interferon regulatory factor 1 (IRF-1) is important during the maturation of natural killer 1.1+ T cell receptor-/+ (NK1+ T) cells, natural killer cells, and intestinal intraepithelial T cells. J. Exp. Med. 187:967(Roberto Baccala, Deborah )