Naive T lymphocytes recirculate through the lymph-vascular system and enter and exit lymphoid organs. Using mice expressing the photoconvertible fluorescence protein Kaede, we demonstrated that naive T cells seek to interact with endogenous Ags after migrating to the lymphoid organs. The interaction with endogenous Ags transiently induces CD69 expression on T cells, which prolongs retention in the lymphoid organs. Cells that fail to express CD69 or lose CD69 expression migrate to other lymphoid organs. Functionally, CD69+-naive CD4+ T cells exhibit faster and greater cytokine production than do CD69− naive CD4+ T cells. These results indicate that CD4+ T cells continuously migrate to interact with endogenous Ags, and such an interaction plays an important role in the Ag reactivity of naive CD4+ T cells.
Cells of hematopoietic origin migrate through the whole body to accomplish specific functions. Lymphocytes migrate continuously through secondary lymphoid organs and seldom migrate out to tissues in the steady state (1–5). However, upon infection with pathogens, a series of events take place to initiate immune response and to eliminate the pathogens. The initial phase of the response is mediated by the recruitment of cells of the innate immune system, such as neutrophils, macrophages, and dendritic cells, to the site of infection where they are activated by pathogens via pattern-recognition receptors (6). The activation of the cells by pathogens induces the expression of a variety of proteins necessary for the initial control of the pathogens and the stimulation of adaptive immune response. Activated APCs then migrate to lymphoid organs where they interact with T and B lymphocytes (6, 7). This step is critical for the generation of an Ag-specific adaptive immune response.
The interaction between innate and adaptive immune systems during infection in vivo is well coordinated by a complex series of events involving cell migration to and retention at different anatomical sites (8–10). One of the molecules involved in cell migration in vivo is CD69. The CD69 molecule is a type I C-type lectin expressed by activated T cells (11–13). CD69 is expressed on Ag-activated T cells, some T cells during homeostatic expansion in vivo, and positively selected CD4+ and CD8+ double-positive thymocytes (14). In all cases, the expression of CD69 by the cells is a transient event. It was shown that the downregulation of CD69 plays a critical role in the emigration of mature thymocytes to secondary lymphoid organs (15). In the periphery, the expression of CD69 molecules downregulates S1P1, leading to the inhibition of lymphocyte egress from the lymph node (LN) and the retention of lymphocytes in the LN (16–18). These results clearly demonstrate the mechanisms by which Ag-activated T cells accumulate in the draining LNs via interaction with cognate Ag-bearing dendritic cells for the induction of CD69 and the subsequent downregulation of S1P1. Recently, the molecular mechanisms of T cell retention within and egress from LNs were reported. Lymphocyte egress from LNs is promoted by S1P1 to overcome retention signals mediated by CCR7 and additional Gαi-coupled receptors (19). However, the biological significance of naive T cell retention in lymphoid organs and recirculation in the steady state in the maintenance of immune homeostasis is unclear.
Recently, we reported a novel system for monitoring cell migration in vivo by using photoconvertible protein Kaede transgenic (Tg) mice. This system allowed us to demonstrate that CD4+, CD8+, and B lymphocytes have distinct migration kinetics in vivo in the steady state (20). In this study, using Kaede-Tg mice, we analyzed the migration of CD4+ T cells in vivo in the context of their phenotypic and functional changes. We demonstrate that CD69 expression on CD4+ T cells is regulated by the interaction with endogenous Ags via surface Ag receptors and that naive and memory CD4+ T cells show clear differences in their capacity to interact with Ags in vivo. Naive and memory CD4+ T cells that fail to interact with Ags rapidly emigrate from the LN and move to other LNs to seek an opportunity to interact with the Ags. Cells that successfully interact with the Ags and express CD69 show prolonged retention in the LN. However, the expression of CD69 is transient, and the loss of CD69 expression results in emigration from the LN. Thus, CD4+ T cells are continuously migrating through lymphoid organs to seek endogenous Ags for interaction. This interaction transiently retains the cells in the local LNs and regulates cell migration and emigration. Furthermore, functional analysis of naive CD4+ T cells revealed that cells expressing CD69 exhibited significantly rapid and strong responses to the stimuli through TCR compared with the CD69− population. These results suggest that the interaction between naive CD4+ T cells and endogenous Ags enhances the functional activity of the T cells, which may play an important role in preparing the adaptive immune system to deal with pathogens expressing cognate Ags to T cells.
Materials and Methods
Kaede-Tg mice (BDF1mixed background) (20), backcrossed to C57BL/6 mice for more than eight generations, were used for experiments. OVA-specific DO11.10 TCR-Tg Rag2−/− mice were generated by crossing DO11.10 TCR-Tg mice (21) (a kind gift from Dr. Kenneth M. Murphy, Washington University School of Medicine, St. Louis. MO) with BALB/c Rag2−/− mice (purchased from Taconic Farms, Germantown, NY). The Tg mice expressing pigeon cytochrome c-specific and I-Ek–restricted 2B4 TCR (22) with a Rag2−/− background were described previously (23). MHC class II knockout (KO) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) (24). C57BL/6 and BALB/c mice were purchased from CREA Japan (Tokyo, Japan). The mice were maintained under specific pathogen-free (SPF) conditions, and experiments were carried out in compliance with institutional guidelines.
Cells from peripheral LNs (PLNs) and spleen were transferred i.v. to recipient mice. In some experiments, cells were labeled with 10 μM CFSE for 10 min before the transfer. For the transfer of CD69+ and CD69− T cells, LN cells were stained with biotin–anti-CD69 mAb, followed by anti-biotin MACS beads. Then, CD69+ and CD69− cells were sorted by MACS (Miltenyi Biotec, Auburn, CA) before transfer. For experiments using CD4++ cells were depleted from purified cells by staining with anti-CD11c, followed by anti-allophycocyanin MACS beads. Ten million purified CD4+ cells (class II+ cells and CD11c+ cells were not detected, and the purity of CD4+ cells was >98%) were transferred to WT and MHC KO mice.
Flow cytometry analysis
Mice were anesthetized and positioned on their right side. The abdomen was opened through a left subcostal incision, and the cisterna chyli was identified as a small cream-colored sac located dorsal to the left renal LN. A 30-gauge needle was inserted into the cisterna chyli, and lymph fluid was harvested.
Photoconversion of inguinal LN was carried out with a method described previously (20). Briefly, Kaede-Tg mice were anesthetized, and the inguinal LN was visualized through an incision made on the abdominal skin. After covering the surrounding tissue with aluminum foil, the LN was exposed to violet light (95 mW/cm2 with a 436-nm band pass filter with Spot UV curing equipment [SP500], Usio, Tokyo, Japan) with continuous instillation of warmed PBS. The wound was closed with sutures.
CD4++ T cells were stained with anti-CD4, anti-CD25, anti-CD62L, and anti-CD69 Abs, as described above. CD69+CD25−CD62L+CD44midlowCD4+ and CD69−CD25−CD62L+CD44midlowCD4+ cells were sorted by JSAN (Bay Bioscience) (purity of each population was >95%). Purified cells (5 × 104 or 5 × 105 cells) were stimulated with plate-coated anti-CD3 mAb (2C11, 10 μg/ml) in a final volume of 200 μl DMEM containing 5% FCS in flat-bottom microtiter plates, and the supernatant was harvested at indicated times. Cytokine concentrations in the supernatants were measured with a Cytometric Bead Array system (BD Bioscience).
Induction of CD69 expression on TCR-Tg naive CD4+ T cells by endogenous Ags
OVA-specific CD4+ T cells from DO11.10 TCR-Tg Rag2−/− mice (21) showed no signs of activation in vivo and expressed a high level of CD62L and medium to low levels of CD44 (Supplemental Fig. 1). However, a small, but significant, percentage of cells expressed CD69, which is known to be an early activation marker of T cells (11, 12) (Fig. 1A). These DO11.10 T cells were labeled with CFSE and cotransferred with the same number of CFSE-labeled WT CD4+ T cells into normal BALB/c mice. Four days after the transfer, the cells were analyzed for cell-cycle progression and the expression of CD69. DO11.10 and WT CD4+ T cells did not show significant proliferation in normal BALB/c mice, as reported previously (25). However, a large fraction of DO11.10 T cells (>90%) expressed CD69 when small numbers (1 × 104 and 1 × 105 cells) of cells were transferred. The percentage of CD69+ cells decreased to ∼60% when a large number of cells (1 × 106 cells) was transferred. The percentage of WT CD4+ T cells expressing CD69 was stable, regardless of the number of cells used for the transfer (Fig. 1B). Thus, these results suggest that the induction of CD69 expression is a unique feature of DO11.10 T cells and is not a result of the cell-transfer experiment.
CD69 is known as an early marker of T cell activation by cognate Ag stimulation (11, 12), and the cell number-dependent decrease in CD69+ DO11.10 T cells after the transfer prompted us to test the possibility of competition among DO11.10 T cells for the interaction with endogenous Ags. To this end, 1 million CFSE-labeled DO11.10 T cells were transferred into normal BALB/c mice, either alone or with an excess number (1 × 107 cells) of nonlabeled DO11.10 cells as competitors. Four days after the transfer, the expression of CD69 on the CFSE-labeled DO11.10 T cells was analyzed. More than 60% of transferred DO11.10 T cells expressed CD69 in the absence of a competitor, similar to the result in Fig. 1B. The percentage of CD69+ cells decreased to 19% in the presence of a large number of nonlabeled DO11.10 competitor cells (Fig. 1C). In contrast, the presence of 1 × 107 WT T cells did not affect the CD69 expression of CFSE-labeled DO11.10 T cells, suggesting that the inhibition of CD69 expression occurred in a DO11.10 T cell-specific manner. Nonlabeled DO11.10 T competitor cells expressed the same percentage of CD69 as CFSE-labeled DO11.10 T cells (19%). Thus, the results suggest that the induction of CD69 expression on transferred DO11.10 T cells is a result of the interaction with endogenous Ags. However, the available Ags (either the number of cells expressing endogenous Ags or the available niche for the interaction) are limited; thus, DO11.10 T cells compete with each other for interaction with the Ags to express CD69.
Pigeon cytochrome c-specific CD4+ T cells from 2B4 TCR-Tg Rag2−/− mice were tested in the same manner as B10.BR recipient mice. When 1 × 105 2B4 T cells were transferred, 21% of the cells became positive for CD69 expression, which was significantly greater than the percentage of CD69+ T cells (5%) prior to the transfer (Supplemental Fig. 2). When the number of transferred T cells was increased (1 × 106 cells), the percentage of CD69+ 2B4 T cells decreased to 6.2%, under identical conditions. These results demonstrate that in DO11.10 and 2B4 T cells, the interaction with endogenous Ags induces CD69 expression, but the available Ags may differ for DO11.10 T cells in BALB/c mice and 2B4 T cells in B10.BR mice.
CD69 expression in normal CD4+ T cells
CD69 was expressed in a small fraction of CD4+ T cells from LN and spleen, and very few, if any, CD4+ T cells in circulation in peripheral blood and lymph expressed CD69 (Fig. 2A). The CD69+ CD4+ T cell fraction in LN contained more CD62Lnegative/low, CD44high, CD25+, CCR7low, and Foxp3+ cells than did the CD69− cell fraction (Supplemental Fig. 3). These results indicate that CD69+ CD4+ T cells contain more memory and regulatory T cell populations. However, a significant number of CD25−CD62LhighCD44mid naive CD4+ T cells also expressed CD69 (Fig. 2B). Thus, CD69 expression cannot act as a marker for any particular subpopulation of CD4+ T cells.
We then analyzed whether the heterogeneity of CD4+ T cells in the expression of CD69 represents subpopulations of CD4+ T cells that are capable of interacting with endogenous Ags, as demonstrated by T cells from TCR-Tg mice. CD4+ T cells from LN cells were sorted into CD69+ and CD69− populations, labeled with CFSE, and transferred to normal C57BL/6 mice i.v. Four days after the transfer, in the mice that received CD69− CD4+ T cells, 10% of the transferred CD4+ T cells were positive for CD69, which is the same level as that of the host CD4+ T cell population (12%). In the mice that received CD69+ CD4+ T cells, a large fraction of T cells lost the expression of CD69, and 24% remained positive for CD69 expression (Fig. 2C). These results clearly indicate that the expression of CD69 on CD4+ T cells in vivo under nonimmunized conditions is unstable, similar to the transient expression of CD69 in Ag-activated T cells.
Migration of CD4+ T cells and regulation of CD69 expression
The total lack of CD69+ CD4+ T cells in the circulation is fully compatible with the notion that CD69 regulates lymphocyte emigration from the LN (11, 12), and only CD69− T cells emigrate from the LN. Using a recently established mouse line expressing the photoconvertible fluorescence protein Kaede (20), we investigated the relationship between cell migration and the expression of CD69 in vivo, with minimum manipulation of the mice. Inguinal LNs were exposed to violet light, which caused all cells in the inguinal LNs to express red Kaede protein (20) (Fig. 3A). Cells in the photoconverted inguinal LNs and connecting axillary LNs were analyzed 24 h after the photoconversion. In this analysis, nonphotoconverted cells in the inguinal LNs are recent migrants from other lymphoid tissues, and photoconverted cells are the remaining cells in the inguinal LNs from the time of photoconversion. In the axillary LNs that are connected directly to the inguinal LNs via the lymphatic system (20), photoconverted cells are recent migrants from the inguinal LNs. In this analysis, CD4+ T cells were also separated into naive and memory T cells based on their surface expression of CD62L and CD44 molecules (Fig. 3B). As shown in Fig. 3C, naive CD4+ T cells in the inguinal LNs were rapidly replaced; 24 h after the photoconversion, only 16% of the cells remained in the inguinal LNs. At this time, the cells that emigrated from the inguinal LNs could be found in the axillary LNs (8.1%). The percentage of cells expressing CD69 in the remaining population (photoconverted cells) was significantly greater than that of the recently immigrated naive CD4+ T cell population (nonphotoconverted cells; 26% versus 5.9%). Naive CD4+ T cells that migrated to the axillary LNs contained a percentage of CD69+ cells that was similar to the naive T cell population present in the LNs. These results suggest that CD69 expression promotes T cell retention in LNs by downregulating the migratory response to S1P1 in the steady state. To show that the modulation of CD69 expression is functionally significant for the migration of naive CD4+ T cells, we evaluated the migration activity of CD69− and CD69+ CD4+ T cells to sphingosine-1-phosphate in a Transwell assay. As shown in Supplemental Fig. 4, CD69− and CD69+ CD4+ T cells migrated to CCL19 at a comparable level. However, CD69−, but not CD69+, CD4+ T cells migrated to sphingosine-1-phosphate. These results indicate that CD69 expression is of functional significance for the migration of naive CD4+ T cells in the nonimmunized condition.
In the case of memory CD4+ T cells, replacement in the initial 24 h was significantly slower than that of naive CD4+ T cells (84% for naive T cells and 51% for memory T cells) (Fig. 3C). Again, the remaining cells contained a significantly greater percentage (57%) of CD69+ cells than did the cells that immigrated to the LNs (14%). In axillary LNs, photoconverted memory CD4+ T cells from the inguinal LN and nonphotoconverted CD4+ T cells had similar percentages of CD69+ cells (22% versus 29%). These results demonstrate that naive and memory CD4+ T cell subsets have distinct migratory kinetics, and the expression of CD69 in each population is regulated differently.
The above kinetic analysis revealed a clear relationship between migration/emigration of CD4+ T cells and the expression of CD69. As shown in Fig. 4A, naive CD4+ T cells emigrated from the inguinal LN much faster than did memory T cells. This difference in emigration from the photoconverted LNs coincided with the migration of CD4+ T cells to the connecting ipsilateral axillary LNs. The migration of naive CD4+ T cells to ipsilateral axillary LNs peaked at 6 h after photoconversion, whereas the migration of memory T cells was delayed, reaching its peak at 12 h after photoconversion. The percentage of CD69-expressing cells gradually increased in the remaining CD4+ T cell population in the inguinal LN (Fig. 4B, left panel). This finding, which is in agreement with those of previous studies (16–18), indicates that CD69 plays a critical role in the retention of T cells in the LN, and cells expressing this molecule preferentially remain in the photoconverted inguinal LNs. At the same time, the percentage of CD69-expressing cells also increased in the photoconverted CD4+ T cells in the ipsilateral axillary LN, which had migrated from the inguinal LN (Fig. 4B, right panel).
In situ upregulation of CD69 in T cells migrating to LN
The increase in the number of CD69+ cells in the population that migrated to the axillary LN suggests that CD69 is induced by the in situ interaction between T cells and endogenous Ags. However, it is possible that cells expressing CD69 migrate to the same LN with delayed kinetics and accumulate in a time-dependent manner. To distinguish between these two possibilities, we treated mice with the S1P1 agonist, FTY720 (18, 26), which inhibits the emigration of CD4+ T cells from LNs 6 h after inguinal LN photoconversion, and we analyzed cell migration and upregulation of CD69 molecules 24 h after photoconversion (Fig. 5). Six hours after photoconversion, 42% of naive CD4+ T cells in the inguinal LNs were replaced, and, at the same time, 16% of naive CD4+ T cells in the axillary LNs were from the photoconverted inguinal LNs. Among the cells that migrated to the axillary LNs, 2.2% expressed CD69. In parallel experiments, inguinal LNs were photoconverted, and mice were treated with FTY720 6 h after photoconversion. Cell migration and CD69 expression were analyzed 24 h after photoconversion. The percentage of photoconverted naive CD4+ T cells in the inguinal LNs (58%) was similar to that of cells analyzed 6 h after photoconversion but without FTY720 treatment (58%). There also was no increase in the percentage of photoconverted naive CD4+ T cells in the axillary LNs treated with FTY720 (16% versus 16%). These results indicate that cell migration is efficiently inhibited by treatment with FTY720 after the initial migration in the first 6 h. When cell migration was inhibited 6 h after photoconversion, the percentage of photoconverted naive CD4+ T cells expressing CD69 in the axillary LNs increased from 2.2% to 6.1%. These findings strongly indicate that cells migrating to the axillary LNs do not express CD69, although the expression is induced in the new LN environment by interaction with endogenous Ags.
In the case of memory CD4+ T cells, only 19% of the cells were replaced in the inguinal LNs in the initial 6 h after photoconversion. FTY720 treatment also inhibited the additional migration of memory CD4+ T cells, similar to naive T cells. CD69 expression on memory CD4+ T cells increased from 7.4% at 6 h to 19% at 24 h after photoconversion, with no apparent cell migration between 6 and 24 h. Thus, memory CD4+ T cells also increase CD69+ populations in situ in the LN.
Induction of CD69 expression on CD4+ T cells and retention of CD4+ T cells in LN are sustained by interaction with endogenous Ags
We have suggested that CD69 expression on CD4+ T cells is induced by interaction with endogenous Ags, and cells that interact with endogenous Ags are retained in lymphoid organs. We tried to confirm the role of endogenous Ags in T cell sequestration into lymphoid organs. Endogenous Ag presentation to CD4+ T cells was thought to be generated by the MHC class II + self-peptide complex. CD4+ T cells were purified from Kaede-Tg mice and transferred to WT and MHC class II KO mice. Four days later, inguinal LNs were photoconverted; 12 h later, CD69 expression levels and the remaining transferred CD4+ T cells in the photoconverted LNs were analyzed. As shown in Fig. 6, 19% of transferred CD4+ T cells in WT mice expressed CD69. However, CD69 expression on the transferred CD4+ T cells was almost completely downregulated when the cells were transferred to MHC class II KO mice. Consistent with CD69 expression, 40% of the transferred CD4+ T cells remained in the WT host. However, only 14% of the cells remained in the photoconverted LNs in the MHC class II KO host, indicating that CD4+ T cells leave LNs more rapidly in the absence of TCR-self-peptide–MHC interactions. Thus, these results indicated that endogenous Ags play a crucial role in CD69 expression and the sequestration of CD4+ T cells into lymphoid organs in the steady state.
Functional difference between CD69+ and CD69− CD4+ T cells
It was shown that T cells undergoing homeostatic expansion in lymphopenic mice change their surface phenotype (27–29) and acquire memory-like function (28–30). Thus, we tested whether there is a functional difference between CD69+ and CD69− naive CD4+ T cells (Fig. 7). Naive CD4+ T cells from normal mice were sorted into CD69+ and CD69− populations, and 5 × 104 cells of each population were stimulated with plate-bound anti-CD3 mAb. CD69+ CD4+ T cells exhibited rapid production of IL-2 and TNF-α, and the magnitude of cytokine production by the CD69+ population, especially IL-2, was significantly larger than that by CD69− T cells. CD69− T cells were capable of producing significant amounts of cytokines when a large number of cells (5 × 105 cells/well) were used for the assay. Thus, the use of a very small number of T cells for the assay revealed a kinetic difference in cytokine production between CD69+ and CD69− populations. We cannot exclude completely the possibility that the biotinylated anti-CD69 mAb cross-linked by streptavidin transmits some signals via the CD69 molecule. However, a previous study using highly purified T cells clearly indicated that the enhancement of plate-bound anti-CD3 T cell activation by anti-CD69 Ab requires cross-linking of anti-CD69 and other APC-derived signals (11). We stimulated sorted cells with plate-bound anti-CD3 Ab alone. Thus, we think that the high response of CD69+ cells compared with CD69− cells after TCR stimulation is an intrinsic characteristic and is not caused by signal transduction via CD69.
The interaction with endogenous Ags in the thymus plays a critical role in the selection of T lymphocytes. T cells that interact with endogenous Ags with high affinity are negatively selected. Immature thymocytes that interact with the endogenous Ag/MHC complex with certain affinity are selected to mature into CD4 or CD8 single-positive T cells that can recognize foreign Ag peptides presented by self-MHC. This selection using the endogenous Ag/MHC complex is essential to generate a T cell repertoire that can mount an immune response to foreign Ags without causing an autoimmune T cell response to endogenous Ags (31, 32).
The interaction between mature T cells and the endogenous Ag/MHC complex was initially demonstrated in transfer experiments using lymphopenic recipients. CD4+ and CD8+ T cells proliferated in vivo, when transferred to mice lacking T lymphocytes in class II and class I MHC-dependent manners, respectively (27, 33–37). Furthermore, the survival of T cells in vivo was shown to be dependent on the presence of MHC, and the reactivity of naive CD4+ T cells to foreign Ags in vitro was influenced by the interaction between T cells and class II MHC in vivo (28, 29, 38–40). At present, neither the nature of endogenous Ags interacting with T cells nor the precise biochemical events induced by this interaction is well characterized (28, 29). However, all of the results suggested that the function and maintenance of naive T cells in the periphery require the interaction between T cells and the endogenous Ag/MHC complex. We have demonstrated that DO11.10 T cells that recognize OVA-derived peptide in the context of I-Ad class II MHC interact with endogenous Ags in vivo and express the early activation marker CD69. The percentage of CD69+ DO11.10 T cells in vivo is inversely related to the number of cells transferred (Fig. 1B), and the expression of CD69 on DO11.10 T cells can compete by cells expressing the same TCR but not by T cells with diverse TCR expression (Fig. 1C). The same phenomenon was found in T cells expressing Tg TCR with different Ag specificities (Supplemental Fig. 2). The number of CD69+ DO11.10 T cells is relatively constant, regardless of the total number of DO11.10 T cells (Supplemental Table I). Thus, these results strongly indicate that T cells compete for limited endogenous Ag/MHC complexes in lymphoid organs for interaction via clonally distributed αβ TCR. The nature of endogenous Ags that interact with DO11.10 TCR is difficult to characterize. However, there is no difference in the induction of CD69 in DO11.10 T cells between germ-free recipients and recipients under normal SPF conditions (Supplemental Fig. 5), ruling out the involvement of Ags from gut bacterial flora. It is possible that positive selection in the thymus and activation of mature T cells in the periphery might share the Ags, because transient CD69 expression was observed in the positively selected thymocytes (14) and in T cells interacting with endogenous Ags in the peripheral lymphoid organs. The identification of endogenous Ags that interact with T cells expressing well-characterized TCR would provide new insights into the relationship between thymic selection and peripheral activation of T cells. Further experiments using new approaches are required.
In normal mice, CD4+ T cell populations contain a significant number of CD69+ cells in the nonimmunized condition (Fig. 2A). Because the same percentage of CD69+ CD4+ T cells was found in the peripheral lymphoid organs (except in Peyer’s patches) of germ-free mice, the effect of gut bacteria seems to be negligible. However, CD69 expression on CD4+ T cells is not a stable phenotype. As shown in the transfer experiments, a fraction of the CD69− population became positive 4 d after transfer into normal recipients. In contrast, the CD69+ population readily lost the expression under the same transfer conditions (Fig. 2C). These results suggest that conventional CD4+ T cells, similar to DO11.10 T cells, also compete for interaction with endogenous Ags to induce the expression of CD69. This notion is further supported by the fact that no CD69 upregulation on normal CD4+ T cells (Fig. 5) or TCR-Tg T cells (30) was observed in cells that were transferred to MHC class II− recipients.
The use of Kaede-Tg mice revealed the relationship between cell migration and interaction with endogenous Ags without cell transfer. First, all circulating T cells in blood and lymph had the CD69− phenotype (Fig. 2A), and a combination of photoconversion of Kaede and the use of FTY720 demonstrated that the number of CD69-expressing cells increased in the new immigrant CD4+ T cell population, without a new influx of photoconverted cells in the LN (Fig. 5). Analysis of the emigration of CD4+ T cells from LNs also demonstrated that CD69-expressing cells remained in the LN much longer than did the CD69− population (Fig. 3). From these results, one can envision that lymphocytes migrate through the whole body in search of endogenous Ags to interact with in the lymphoid organs. In each lymphoid organ, T cells compete with each other for the limited endogenous Ags. Successful interaction with endogenous Ags induces the expression of CD69 molecules on T cells for prolonged retention in the lymphoid organs. However, CD69 expression is transient, and the loss of CD69 expression leads to emigration of the T cells. T cells that fail to interact with endogenous Ags in the LN emigrate rapidly through lymph or bloodstream to other lymphoid organs, where they continue their search for endogenous Ags.
The interaction with endogenous Ags induces CD69 expression for prolonged retention in the LN and changes the function of naive CD4+ T cells. It was demonstrated that the expansion of naive CD4+ T cells in the lymphopenic condition coincides with changes in their surface phenotype (27–29) and in response to stimulation through TCR (27–30). Our results indicate that even under physiological conditions, naive CD4+ T cells interact with endogenous Ags to change their phenotype and functions. The phenomenon observed in the lymphopenic condition may simply be an exaggeration of the continuous interaction between naive CD4+ T cells and endogenous Ags demonstrated in this study. It is difficult to evaluate the importance of the change in function of naive T cells in response to pathogens, although the increased responsiveness of naive CD4+ T cells induced by endogenous Ags may play an important role in preparing T cell response to foreign Ags, as suggested by a previous study with TCR-Tg mice (40). Further experiments with a pathogen-specific system are required to address this issue.
In the experiments using Kaede-Tg mice, naive and memory phenotype T cells had similar migration and emigration patterns based on CD69 expression. In both populations, only cells lacking CD69 emigrate from the lymphoid organs to other lymphoid organs, and some, but not all, T cells interact with endogenous Ags to express CD69 in situ. However, the percentage of cells expressing CD69 is much larger in the memory phenotype T cells than in the naive T cell population. This is reflected by the much slower emigration of memory-type T cells from the lymphoid organs compared with naive T cells. This difference may be a reflection of the difference in the threshold for activation by endogenous Ags between memory and naive T cells. We do not know the Ags responsible for the generation of memory phenotype T cells in nonimmunized mice under SPF and germ-free conditions. Nevertheless, it is possible that the continuous interaction with endogenous Ags plays an important role in the survival of memory T cells and the maintenance of immunological memory in the absence of original pathogens.
In summary, T cells move through lymphoid organs, interact with endogenous Ags, and change their function to maintain better reactivity to foreign Ags in the steady state. Our results are the first to demonstrate that the steady-state migration of T cells is not a passive movement prior to encountering foreign Ags; rather, T cells recirculate to search endogenous Ags in a limited niche. The role of this T cell–endogenous Ag interaction under pathogenic conditions, such as infection with pathogens and the initiation of autoimmune response, awaits further investigation.
Disclosures The authors have no financial conflicts of interest.
The online version of this article contains supplemental material.
Abbreviations used in this paper:
- lymph node
- peripheral lymph node
- specific pathogen-free
- wild type.
- Received December 8, 2009.
- Accepted February 18, 2010.
- Copyright © 2010 by The American Association of Immunologists, Inc.