The E enhancer controls the generation of CD4–CD8– ?TCR-expressing T cells that can give rise to different lineages of ? T cells
http://www.100md.com
《实验药学杂志》
1 Dana-Farber Cancer Institute and 2 Children's Hospital, Harvard Medical School, Boston, MA 02115
3 University of Pennsylvania School of Medicine, Philadelphia, PA 19104
4 Department of Medicine, The University of Chicago, Chicago, IL 60637
CORRESPONDENCE Harald von Boehmer: harald_von_boehmer@dfci.harvard.edu
It is well established that the pre–T cell receptor for antigen (TCR) is responsible for efficient expansion and differentiation of thymocytes with productive TCR? rearrangements. However, Ptcra- as well as Tcra-targeting experiments have suggested that the early expression of Tcra in CD4–CD8– cells can partially rescue the development of ? CD4+CD8+ cells in Ptcra-deficient mice. In this study, we show that the TCR E but not E enhancer function is required for the cell surface expression of ?TCR on immature CD4–CD8– T cell precursors, which play a crucial role in promoting ? T cell development in the absence of pre-TCR. Thus, ?TCR expression by CD4–CD8– thymocytes not only represents a transgenic artifact but occurs under physiological conditions.
Abbreviations used: DN, double negative; DP, double positive; SP, single positive; wt, wild type.
Intrathymic development can be divided into discrete stages at which thymocytes express distinct surface markers that include CD4, 8, 25, and 44. Cells that express neither CD4 nor CD8 are called double-negative (DN) cells that consist of CD25–CD44+ (DN1), CD25+CD44+ (DN2), CD25+CD44– (DN3), and CD25–CD44– (DN4) subsets (1). Thymus development is also characterized by sequential rearrangement and expression of TCR genes (2). Rearrangement of variable and TCR gene segments begins at the DN2 stage and, if productive, results in CD4–CD8– (DN) cells that express TCR on the cell surface. Rearrangement of variable TCR? gene segments sets in slightly later and, also if productive, results in the surface expression of the pre-TCR consisting of a TCR? chain that is covalently associated with the pre-TCR chain and noncovalently associated with CD3 signal-transducing molecules (3, 4). The pre-TCR is expressed at rather low levels on DN3 and DN4 cells (5). Pre-TCR–expressing DN cells undergo several rounds of division before the rearrangement of variable TCR gene segments sets in at the late DN4 and early CD4+CD8+ double-positive (DP) stage. This results in the surface expression of ?TCRs at the expense of the pre-TCR because TCR chains generally compete favorably with pre-TCR for TCR? chains. TCR?-expressing DP cells then undergo positive or negative selection by intrathymic peptide–MHC complexes (6).
The temporal order of TCR V gene segment rearrangement has suggested the following scenario: DN4 cells that contain productive Tcrg and Tcrd rearrangements express TCR on the cell surface and, thereby, become functionally mature T cells that are ready to leave the thymus. Pre-TCR–expressing cells will eventually become DP thymocytes expressing an ?TCR, the specificity of which determines their further developmental fate. This simple scheme of T cell development was found not to be valid when T cell development was analyzed in Ptcra-deficient mice. In contrast to CD3-deficient and Tcrb plus Tcrd double-deficient mice that contain none or very few DP thymocytes, Ptcra-deficient mice were shown to harbor strongly reduced but still considerable numbers of DP thymocytes (7). Further analysis in Ptcra,Tcrd as well as Ptcra,Tcra double-deficient mice then revealed that both an early expressed TCR as well as an early expressed ?TCR could still rescue the development of DP thymocytes in the absence of pre-TCR (8). In the case of ?TCR in Ptcra–/–,Tcrd–/– mice, this resulted in thymocytes of which >95% harbored TCR? chains (i.e., DP cells that were selected by an early expressed ?TCR and therefore contained in-frame Tcrb rearrangements). In the case of an early expressed TCR in Ptcra–/–,Tcra–/– mice, only 15% of DP cells contained TCR? chains. Thus, these experiments indicated that receptors other than the pre-TCR could relieve DN3 cells from a development block resulting in the production of DP cells and that there must be a rearrangement of TCR V gene segments at the DN3 or earlier stages of T cell development.
The temporal order of TCR V gene segment rearrangement appears especially important with regard to the Tcra and Tcrd locus. This is where the Tcra locus is embedded in the Tcra locus and where the early rearrangement of V gene segments results in the formation of TCR chains, whereas late V rearrangement is accompanied by the deletion of the Tcrd locus and generation of Tcra genes (9). Two different enhancer elements have been invoked in the control of rearrangement and expression of the Tcra/Tcrd locus: the E enhancer located in the J-C intron and the E enhancer 4 kb downstream of C (10–13). Both enhancers have been deleted by homologous recombination. In E–/– animals, the development of ? T cells appeared to proceed normally with the exception that there was a considerable reduction of thymic and peripheral T cells. In this context, E-deficient alleles exhibited a substantial reduction of Tcrd gene rearrangements. E–/– mice contained normal numbers of DP cells but reduced numbers of CD4+CD8– and CD4–CD8+ single-positive (SP) cells as a result of an almost complete block in V to J rearrangements. The E enhancer also controls levels of ? as well as TCR expression, as evident by reduced levels of TCR and TCR transcripts in E–/– mice. These results indicate that E functions in DN thymocytes to promote Tcrd gene rearrangement and gene expression but not J accessibility to the V(D)J recombinase. On the other hand, E functions in DP thymocytes to promote TCR gene rearrangement via J accessibility and controls both Tcrd and Tcra gene expression. However, because TCR chains were still present in E–/– mice, it was hypothesized that in the absence of E, the E enhancer or other elements might have a role in promoting rearrangement, possibly via the promotion of J accessibility and/or expression of some V segments (11). This would account for the TCR chains with limited diversity (mostly V2) that were expressed in peripheral lymphoid tissue in E-deficient mice.
In this study, we have tested the hypothesis that a low level of V to J rearrangements is controlled by the E enhancer and occurs in DN thymocytes earlier than the bulk of E-controlled V to J rearrangements, thus leading to the expression of ?TCR in DN cells, which permits some of these cells to enter the ? lineage of DP thymocytes.
RESULTS
Early ?TCR-expressing DN cells
Analyzing total thymocytes and DN thymocytes from wild-type (wt) C57BL/6 mice, we find that 5% express TCR? chains on the cell surface of which 10% are paired with V2-containing TCR chains (Fig. 1 A ). Because a subset of NKT cells (14) has the TCR?+ CD4–CD8– phenotype, NK1.1 cells were excluded by using NK1.1 antibodies for depletion as well as by the analysis of NKT cell–deficient CD1–/– animals (Fig. 2 ). The results show that V+,TCR?+ cells belong to a distinct, non–NKT cell subset of DN cells. In addition, we found TCR?+ DN T cells in secondary lymphoid organs in both wt and CD1- deficient animals (Fig. 2).
Figure 1. V2-enriched early ? T cells in wt and Ptcra–/– mice. (A) TCR? versus V2 FACS profile in thymi of wt and Ptcra–/– thymocytes. DN (CD4–CD8–) cells were electronically gated after the exclusion of CD4+, CD8+, and NK1.1 cells. (B) A CD4 versus CD8 profile of Ptcra–/– thymi. (C and D) Staining of Ptcra–/– thymocytes (total or CD4–CD8–) with TCR? and V2 antibodies. Numbers in quadrants indicate the percentages of cells in that quadrant.
Figure 2. CD4–CD8– ? T cells in the thymus and lymphoid periphery. (A) CD4,CD8 versus NK1.1 staining in the thymus of wt and CD1–/– mice (top). TCR? versus V2 analysis of NK1.1– CD4–CD8– thymocytes (bottom). (B and C) Identification of peripheral CD4–CD8– TCR?+ cells in the spleen and lymph nodes using TCR?, V2, CD4, and CD8 antibody labeling. Numbers in quadrants indicate the percentages of cells in that quadrant.
To investigate whether DN ?TCR-expressing cells are pre-TCR selected as their DP (CD4+CD8+) or SP (CD4+CD8– and CD4–CD8+) counterparts, thymocytes from Ptcra–/– mice were analyzed (Fig. 1 B). Although Ptcra–/– mice were previously shown to be deficient in NKT cells (15), we used NK1.1 antibodies in the depletion procedure that yielded DN thymocytes. In Ptcra–/– mice, 6–7% of DN thymocytes express TCR? chains on the cell surface, and, in this particular experiment, 40–50% of DN cells with TCR? proteins coexpressed V2,TCR chains (Fig. 1 C). In several experiments, the proportion of V2+ cells among TCR?+ DN cells in Ptcra–/– mice was variable and ranged between 10 and 50% (Fig. 4; see Fig. 7).
Of interest was the observation that V2 is not only contained in ? but also in TCRs in DN thymocytes from Ptcra–/– mice (Fig. 1 C), a notion consistent with recent findings that V2 gene segments can join to both J and DJ sequences (16). To exclude that the existence of TCR?+,V2+ DN cells is the result of V2 segment pairing with TCR diversity joining elements (16), a similar analysis was conducted in Ptcra–/–,Tcrd–/– double-deficient mice. About 7% of CD4–CD8– cells were found to express TCR? proteins on the cell surface, and 25% of the TCR? chains were paired with V2-containing TCR chains, suggesting that these TCR?-expressing cells are bona fide ? T cells that do not require pre-TCR selection (Fig. 3 A ).
Figure 3. Phenotype and developmental potential of early ? T cells. (A) V2 versus TCR? staining of Ptcra–/–,Tcrd–/– thymocytes. (B) Phenotypic analysis of CD4–CD8– and CD4–CD8– TCR?+ cells using CD44 and CD25 antibodies. (C) Embryonic day 14.5 fetal (Rag-1–/–) thymic organ culture of CD4–CD8– TCR?+ NK1.1– Ptcra–/– cells. Cells were cultured for 7 d and were stained with V2, TCR?, CD4, and CD8 antibodies. A CD4 versus CD8 staining of V2+ donor cells is also shown (bottom). Numbers in quadrants indicate the percentages of cells in that quadrant.
Further phenotypic analysis of the TCR?+ thymocytes showed that almost all "early" TCR?-expressing CD4–CD8– cells are CD25 negative and, thus, belong to the DN4 (CD25–CD44–) subset (Fig. 3 B). The developmental potential of early ?TCR-expressing DN cells was then addressed by culturing purified LY5.2+ CD4–CD8– NK1.1– TCR?+ Ptcra–/– thymocytes together with embryonic thymi from LY5.1+ in Rag1–/– donors in fetal thymic organ cultures. After 7 d of culture, 40% of the donor-derived cells have up-regulated the expression of CD4,CD8 coreceptors. Moreover, >90% of the cultured thymocytes retain the surface expression of TCR?, and 20% of them are V2+ (Fig. 3 C). Not all cells up-regulate CD4 and CD8, however, and a substantial fraction of cells expressing high levels of V2-containing ?TCRs remain CD4–CD8–. It is likely that these ?TCR-expressing DN cells normally exit the thymus because CD4–CD8– ?TCR+ cells can be detected in the lymph nodes and spleen of adult mice (Fig. 2), and it was shown in TCR? transgenic mice that DN cells with the transgenic TCR can accumulate in peripheral lymphoid tissue (6). Thus, TCR?-expressing DN cells can give rise to both immature DP and mature SP (CD4/8) T cells as well as DN ? T cells that do not enter the DP ? lineage (17).
The E enhancer controls late but not early development of ? T cells in Ptcra–/– mice
To address the role of the described TCR enhancers on early TCR expression, we generated Ptcra–/– mice with a targeted deletion of the E locus (Ptcra–/–,E–/–). Ptcra–/–, E–/– and Ptcra–/–,E+/+ littermate mice contain similar numbers of thymocytes (Fig. 4 A ). However, there is a clear reduction of CD4+CD8– SP thymocytes in the Ptcra–/–, E–/– mice because of limited TCR diversity and/or expression levels (Fig. 4 B). This is supported by the staining of either all thymocytes or only DN thymocytes with a combination of TCR? and Va2 antibodies: although total thymocytes from Ptcra–/– single-deficient mice contain <10% TCR?+,V2+ cells, among TCR?+ cells, this proportion is much higher (>30%) in Ptcra–/–,E–/– double-deficient mice. In DN thymocytes from Ptcra–/–,E+/+ mice, TCR?+,V2+ cells represent 10% of all TCR?+ cells, whereas in Ptcra–/–,E–/– mice, TCR?+,V2+ cells represent 50%. These data indicate that E is not required for the early rearrangement and expression of Tcra genes (Fig. 4 C). The data show that normally E predominantly contributes to the rearrangement and expression of TCR V gene segments other than V2 gene segments. Consistent with this notion, the spleen of Ptcra–/–,E–/– mice contains almost exclusively V2+,TCR?+ cells, whereas only 10% of V2+,TCR?+ cells among TCR?+ cells are found in the spleen of Ptcra–/–,E+/+ mice (Fig. 4 D). These observations are in line with earlier observations in E–/– mice (11) showing that TCR?+ cells in the spleen of E–/– mice express almost exclusively V2 (Fig. 4 D).
Figure 4. E controls late but not early development of ? T cells. (A) Absolute cell numbers of Ptcra–/– and Ptcra–/–,E–/– littermates. Error bars represent SD. (B) CD4 versus CD8 profiles of littermates belonging to all indicated genotypes. (C and D) V2 versus TCR? antibody labeling in total DN thymocytes (C) and spleen cells (D). Numbers in quadrants indicate the percentages of cells in that quadrant.
DN cells but not mature T cells in E–/–-deficient mice express V2-negative Tcra genes
The data in Fig. 4 suggest that DN cells in E–/– mice can express V genes other than V2 and that the V2 dominance in the periphery of E–/– mice is established at a later developmental stage. This issue was addressed in more detail by analyzing the expression of TCR V2 genes and other TCR V genes (V3, V8, and V11) in Ptcra–/– and Ptcra–/–,E–/– mice at various stages of development. As shown in Fig. 5 B , NKT cell–depleted CD4– and CD8– (DN) cells from both Ptcra–/– and Ptcra–/–,E–/– mice do express V genes that stain with a cocktail of V3, 8, and 11 antibodies. Although the proportion of cells stained with the cocktail versus V2-positive cells remains about the same in CD4+CD8+ cells of Ptcra–/– mice, it drastically decreases in DP cells of E–/– mice such that the vast majority of cells expresses V2 (Fig. 5 A). This trend is also evident in peripheral T cells from E–/– but not wt mice in which virtually all TCR?+ cells express V2 (Fig. 4 D). Thus, these data indicate that the predominance of V2 expression in peripheral T cells of E–/– mice is not caused by the fact that the E enhancer only allows for V2 rearrangement at the DN stage of T cell development. Instead, this is likely the result of the fact that V2 gene segments are able to sustain a sufficiently high expression of Tcra genes in the absence of the E enhancer such that only V2-positive ? T cells can be positively selected and maintained in peripheral lymphoid tissue.
Figure 5. Immature but not mature T cells from E–/– mice express diverse TCR chains. (A) Staining of thymocytes from Ptcra–/–,E+/+ and Ptcra–/–,E–/– mice with CD4 and CD8 (top), and staining of CD4+CD8+ TCR?+ cells with V3 + V8 + V11 antibodies versus V2 antibodies (middle). Bottom panel shows isotype controls for the staining in the middle panel. (B) Staining of NK1.1– CD4–CD8– TCR? thymocytes from Ptcra–/–,E+/+ and Ptcra–/–,E–/– mice with V3 + V8 + V11 antibodies versus V2 antibodies (top) and with isotype controls (bottom). Numbers in quadrants indicate the percentages of cells in that quadrant.
E controls V expression in DN cells
Because the analysis of Ptcra–/–,E–/– mice has shown that E is not an enhancer element required for early TCR? expression and pre-TCR–independent T cell development (8, 11), we have focused our attention on the E enhancer by generating and studying animals that lack both Ptcra as well as the E enhancer element (Ptcra–/–,E–/– mice). When compared with Ptcra–/–,E+ thymi, it was noted that Ptcra–/–,E–/– thymi contain much-reduced numbers of thymocytes (Fig. 6 A) that are severely deficient in DP and SP cells (Fig. 6 B ). Also, the number of TCR-expressing cells is reduced even though some T cells are still present. Within the DN compartment, there is a more complete block at the DN3 stage of development in Ptcra–/–,E–/– versus Ptcra–/–,E+ mice (Fig. 6, C and D).
Figure 6. E is essential for pre-TCR–independent T cell development. (A) Absolute cell numbers of Ptcra–/– and Ptcra–/–,E–/– littermates. 4–8-wk-old mice were analyzed. Error bars represent SD. (B and C) CD4 versus CD8 (B) and TCR? versus TCR (C) profiles of littermate thymi. (D) Analysis of the DN compartment using CD25 and CD44 antibodies. Numbers in quadrants indicate the percentages of cells in that quadrant.
Ablation of E in Ptrcra–/– mice has a profound effect on the percentage and absolute number of DN ?TCRs as well as of V2-expressing cells, as TCR?+,V2+ thymocytes are virtually absent (<0.25%; Fig. 7, A and B ). However, there was a small number of TCR?-expressing cells among total thymocytes in some of the Ptcra–/–,E–/– mice (Fig. 7), suggesting some remaining level of TCR rearrangement in the Ptcra–/–, E–/– thymi (perhaps mediated by the E enhancer in cells) that was "rescued" by the expression of TCR at the DN developmental stage (8).
Figure 7. Absence of V2-expressing cells in the thymi of Ptcra–/–,E–/– mice. TCR? versus V2 staining of either total thymocytes (A) or CD4–CD8– cells (B) of littermate mice analyzed at 4 wk of age. Numbers in quadrants indicate the percentages of cells in that quadrant.
Thus, in the absence of Ptcra, the E enhancer has a crucial role in the early assembly of TCR chains as well as TCR chains in DN3 and/or earlier stages of T cell development (i.e., TCR chains that are required for the formation of and ?TCR in the absence of pre-TCR). The ?TCRs (and TCRs) can rescue some developmental progression beyond the DN3 stage in Ptcra-deficient mice, which appropriately explains the incomplete developmental block in Ptcra–/– versus CD3–/– or Tcrb–/– x Tcrd–/– mice (i.e., mice that cannot assemble any TCR on the cell surface).
DISCUSSION
The results obtained in mice with a combined deficiency of the pre-TCR chain plus the E or E enhancer provide an explanation for the observation that development of DN thymocytes into ? lineage DP cells in Ptcra–/–,Tcrd–/– double-deficient mice can be rescued by ?TCR. The data indicate that E can promote early V to J rearrangements, which results in the expression of an ?TCR on the surface of DN4 thymocytes. The early expression of an ?TCR was previously considered to represent a transgenic artifact caused by the too early expression of TCR chains in TCR transgenic mice (18), but, as shown here, it also occurs under physiological conditions. Our experiments and earlier experiments in TCR transgenic mice show that some but not all of the ?TCR CD4–CD8– cells can become CD4+CD8+ cells (8, 18). This developmental pathway may make only a limited contribution to the generation of DP ? lineage cells in pre-TCR–competent mice not only because of the paucity of TCR chains in DN cells of normal mice but also because TCR is a bad "surrogate" for pre-TCR. Indeed, we found that under competitive conditions, pre-TCR is far more effective in the generation of DP cells than ?TCR (19).
The fate of the DN cells that express ?TCR on the cell surface and do not become DP thymocytes needs to be further evaluated. These cells apparently can leave the thymus because they can be detected in the spleen and lymph nodes of normal (Fig. 2) as well as in exaggerated numbers in TCR transgenic mice. Such cells were previously shown to acquire functional maturity (i.e., respond with proliferation and cytokine production to TCR ligation) and to acquire CD8 expression when antigenically stimulated. In fact, this unusual subset of T cells was analyzed in TCR transgenic mice many years ago, and it was concluded that the early expression of an ?TCR could mimic signals generated when thymocytes express TCR and become functionally mature (18, 20). Of interest is that such cells can express an autoreactive TCR but are not deleted because of the lack of coreceptors and can accumulate in secondary lymphoid tissue of mice, as shown in various TCR transgenic models (21). Some of these cells exhibit an activated phenotype and have an as yet undefined role in the immune system.
Recent data have, in fact, shown that the extensive accumulation of CD8 cells in the gut of TCR transgenic mice (22) to a large extent depends on the premature expression of the transgenic TCR in DN cells because "on time" expression of the same Tcra transgene does not result in the strong accumulation of CD8 cells in the gut (23). Another possibility is that thymic DN ?TCR+ cells represent precursors of some peripheral regulatory T cells. Perhaps these cells are akin to both mouse and human ?TCR+ CD3+ NK.1.1– CD4–CD8– DN Regulatory T cells that can suppress antigen-specific immune responses mediated by CD8+ and CD4+ T cells through a process that requires cell to cell contact and Fas–FasL interactions (24). In this regard, it is important to point out that the ?TCR+ DN studied here are different from NK1.1+ DN T cells and from ?TCR+ DN cells studied by others who concluded that all TCR?+ DN cells were derived from DP precursors (25); either this generalization is wrong or the fate mapping approach used by the authors is not valid. However, it is clear that the TCR?+ DN cells in Ptcrd–/–, Tcrd–/– mice are involved in the rescue of development rather than being derived from DP cells (8).
The results shown in this study also provide an adequate explanation for earlier observations in E-deficient mice that exhibited a TCR repertoire limited to TCR,V2 chains. In this context, on E-deleted alleles, other cis-acting elements such as E and/or V2 promoters were hypothesized to promote either only V2 to J rearrangements or V to J rearrangements. These rearrangements involve a diverse array of V segments with assembled V2J complexes expressed in a much higher proportion in the absence of E. Our results indicate that in DN cells, E can direct a low level of V to J rearrangements, possibly via promoting J accessibility that results in the expression of a variety of different V gene segments. The early expressed ?TCRs allow the development of some DP thymocytes with TCRs containing mostly V2+,TCR chains. In the absence of E, only early VJ rearrangements with a high proportion of V2 would continue to be expressed in DP thymocytes and mature T cells, perhaps because their promoters do not require E activity to drive gene expression. This pathway of differentiation observed in E-deficient mice might be invisible in wt mice because of continual Tcra rearrangement in DP thymocytes that will swamp out the E-initiated V rearrangements and, thereby, lead to a much more diverse ?TCR repertoire.
MATERIALS AND METHODS
Mice.
Mice were kept in the sterile facilities of The University of Chicago, Dana-Farber Cancer Institute, and Children's Hospital. Animal protocols were approved by the Institutional Animal and Use Committees of these institutes. C57BL/6, Rag1–/–, and Tcrd–/– mice (also on the C57BL/6 background) were purchased from Jackson ImmunoResearch Laboratories. C57BL/6 CD1–/– were provided by A. Bendelac (The University of Chicago, Chicago, IL). E–/– and E–/– mice were generated in the laboratory of Frederick W. Alt (10, 11). Ptcra–/– mice were described previously (7).
Flow cytometric analysis and cell sorting.
Anti-CD4 (L3T4), CD8 (53–6.7), CD25 (3C7), CD44 (IM7), NK1.1 (PK136), TCR? (H57-597), TCR (GL3), Va2 (B20.1), Va11 (RR8-1), and Va3 (RR3-16) mAbs were purchased from BD Biosciences. The Va8 (CTVA8) antibody was purchased from CALTAG. These mAbs were directly coupled to FITC, PE, Cy-crome, APC, or biotin. Surface marker expression on thymocytes and peripheral T cells was visualized using a FACScalibur (Becton Dickinson) and analyzed with FlowJo (Tree Star) and CellQuest software (Beckton Dickinson). Cell sorting was performed using Mo-Flo (DakoCytomation) and FACS-Aria (Becton Dickinson) sorters.
Fetal thymic organ culture.
Thymi were cultured as described previously (26). In brief, (LY5.2+) TCR?+ CD4–CD8– NK1.1– cells were FACS purified from thymi of Ptcra–/– animals. Lineage-negative (CD3, CD8, Mac-1, NK1.1, Gr-1, Ter-119, and CD19) bone marrow progenitors were also used as a reconstitution control. Isolated embryonic day 14.5 thymi from Rag1 mice (expressing LY5.1) were initially incubated in Terasaki plates (Nunc), subsequently applied on Transwell (Nunc) porous filters, and incubated in Iscove's modified Dulbecco's medium supplemented with 10% fetal calf serum (Sigma-Aldrich) plus penicillin, streptomycin, and mercapto-ethanol. Cultures were maintained at 37°C for 7 d, after which LY5.2+ cells were stained and analyzed by FACS analysis.
Acknowledgments
We would like to thank A. Bendelac for the CD1–/– animals, F. Meng for expert technical assistance, R. Duggan and J. Marvin for cell sorting, and I. Apostolou for advice and technical expertise.
I. Aifantis was supported by the National Institutes of Health (NIH) grant R01 CA105129 and the V Foundation for Cancer Research. K. Sawai is supported by a University of Chicago Molecular Biology Training grant. H. von Boehmer was supported by NIH grants R01 AI45846 and R01 AI47281. F.W. Alt was supported by NIH grant R01 AI20047 and is a Howard Hughes Medical Institute Investigator.
References
Godfrey, D.I., and A. Zlotnik. 1993. Control points in early T-cell development. Immunol. Today. 14:547–553.
Rodewald, H.R., and H.J. Fehling. 1998. Molecular and cellular events in early thymocyte development. Adv. Immunol. 69:1–112.
Groettrup, M., K. Ungewiss, O. Azogui, R. Palacios, M.J. Owen, A.C. Hayday, and H. von Boehmer. 1993. A novel disulfide-linked heterodimer on pre-T cells consists of the T cell receptor beta chain and a 33 kd glycoprotein. Cell. 75:283–294.
Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H.J. Fehling, and H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science. 266:1208–1212.
Aifantis, I., V.I. Pivniouk, F. Gartner, J. Feinberg, W. Swat, F.W. Alt, H. von Boehmer, and R.S. Geha. 1999. Allelic exclusion of the T cell receptor ? locus requires the SH2 domain-containing leukocyte protein (SLP)-76 adaptor protein. J. Exp. Med. 190:1093–1102.
von Boehmer, H. 2004. Selection of the T cell repertoire: receptor- controlled checkpoints in T cell development. Adv. Immunol. 84:201–238.
Fehling, H.J., A. Krotkova, C. Saint-Ruf, and H. von Boehmer. 1995. Crucial role of the pre-T-cell receptor alpha gene in development of alpha beta but not gamma delta T cells. Nature. 375:795–798. (published erratum appears in Nature. 1995. 378:419).
Buer, J., I. Aifantis, J.P. DiSanto, H.J. Fehling, and H. von Boehmer. 1997. Role of different T cell receptors in the development of pre-T cells. J. Exp. Med. 185:1541–1547.
Chien, Y.H., M. Iwashima, K.B. Kaplan, J.F. Elliott, and M.M. Davis. 1987. A new T-cell receptor gene located within the alpha locus and expressed early in T-cell differentiation. Nature. 327:677–682.
Monroe, R.J., B.P. Sleckman, B.C. Monroe, B. Khor, S. Claypool, R. Ferrini, L. Davidson, and F.W. Alt. 1999. Developmental regulation of TCR delta locus accessibility and expression by the TCR delta enhancer. Immunity. 10:503–513.
Sleckman, B.P., C.G. Bardon, R. Ferrini, L. Davidson, and F.W. Alt. 1997. Function of the TCR alpha enhancer in alphabeta and gammadelta T cells. Immunity. 7:505–515.
Hernandez-Munain, C., B.P. Sleckman, and M.S. Krangel. 1999. A developmental switch from TCR delta enhancer to TCR alpha enhancer function during thymocyte maturation. Immunity. 10:723–733.
Balmelle, N., N. Zamarreno, M.S. Krangel, and C. Hernandez-Munain. 2004. Developmental activation of the TCR alpha enhancer requires functional collaboration among proteins bound inside and outside the core enhancer. J. Immunol. 173:5054–5063.
Bendelac, A., M.N. Rivera, S.H. Park, and J.H. Roark. 1997. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535–562.
Eberl, G., H.J. Fehling, H. von Boehmer, and H.R. MacDonald. 1999. Absolute requirement for the pre-T cell receptor alpha chain during NK1.1+ TCRalphabeta cell development. Eur. J. Immunol. 29:1966–1971.
Jouvin-Marche, E., C. Aude-Garcia, S. Candeias, E. Borel, S. Hachemi-Rachedi, H. Gahery-Segard, P.A. Cazenave, and P.N. Marche. 1998. Differential chronology of TCRADV2 gene use by alpha and delta chains of the mouse TCR. Eur. J. Immunol. 28:818–827.
von Boehmer, H., I. Aifantis, O. Azogui, J. Feinberg, C. Saint-Ruf, C. Zober, C. Garcia, and J. Buer. 1998. Crucial function of the pre- T-cell receptor (TCR) in TCR beta selection, TCR beta allelic exclusion and alpha beta versus gamma delta lineage commitment. Immunol. Rev. 165:111–119.
von Boehmer, H., J. Kirberg, and B. Rocha. 1991. An unusual lineage of ? T cells which contain autoreative cells. J. Exp. Med. 174:1001–1008.
Borowski, C., X. Li, I. Aifantis, F. Gounari, and H. von Boehmer. 2004. Pre-TCR and TCR are not interchangeable partners of TCR? during T lymphocyte development. J. Exp. Med. 199:607–615.
Bruno, L., H.J. Fehling, and H. von Boehmer. 1996. The ?TCR can replace the TCR in the development of lineage cells. Immunity. 5:343–352.
Terrence, K., C.P. Pavlovich, E.O. Matechak, and B.J. Fowlkes. 2000. Premature expression of T cell receptor (TCR)? suppresses TCR gene rearrangement but permits development of lineage T cells. J. Exp. Med. 192:537–548.
Rocha, B., H. von Boehmer, and D. Guy-Grand. 1992. Selection of intraepithelial lymphocytes with CD8 alpha/alpha co-receptors by self-antigen in the murine gut. Proc. Natl. Acad. Sci. USA. 89:5336–5340.
Baldwin, T.A., M.M. Sandau, S.C. Jameson, and K.A. Hogquist. 2005. The timing of TCR expression critically influences T cell development and selection. J. Exp. Med. 202:111–121.
Ford, M.S., K.J. Young, Z. Zhang, P.S. Ohashi, and L. Zhang. 2002. The immune regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J. Exp. Med. 196:261–267.
Egawa, T., G. Eberl, I. Taniuchi, K. Benlagha, F. Geissmann, L. Hennighausen, A. Bendelac, and D.R. Littman. 2005. Genetic evidence supporting selection of the V14i NKT cell lineage from DP thymocyte precursors. Immunity. 22:705–716.
Martin, C.H., I. Aifantis, M.L. Scimone, U.H. von Andrian, B. Reizis, H. von Boehmer, and F. Gounari. 2003. Efficient thymic immigration of B220+ lymphoid-restricted bone marrow cells with T precursor potential. Nat. Immunol. 4:866–873.(Iannis Aifantis1,4, Craig)
3 University of Pennsylvania School of Medicine, Philadelphia, PA 19104
4 Department of Medicine, The University of Chicago, Chicago, IL 60637
CORRESPONDENCE Harald von Boehmer: harald_von_boehmer@dfci.harvard.edu
It is well established that the pre–T cell receptor for antigen (TCR) is responsible for efficient expansion and differentiation of thymocytes with productive TCR? rearrangements. However, Ptcra- as well as Tcra-targeting experiments have suggested that the early expression of Tcra in CD4–CD8– cells can partially rescue the development of ? CD4+CD8+ cells in Ptcra-deficient mice. In this study, we show that the TCR E but not E enhancer function is required for the cell surface expression of ?TCR on immature CD4–CD8– T cell precursors, which play a crucial role in promoting ? T cell development in the absence of pre-TCR. Thus, ?TCR expression by CD4–CD8– thymocytes not only represents a transgenic artifact but occurs under physiological conditions.
Abbreviations used: DN, double negative; DP, double positive; SP, single positive; wt, wild type.
Intrathymic development can be divided into discrete stages at which thymocytes express distinct surface markers that include CD4, 8, 25, and 44. Cells that express neither CD4 nor CD8 are called double-negative (DN) cells that consist of CD25–CD44+ (DN1), CD25+CD44+ (DN2), CD25+CD44– (DN3), and CD25–CD44– (DN4) subsets (1). Thymus development is also characterized by sequential rearrangement and expression of TCR genes (2). Rearrangement of variable and TCR gene segments begins at the DN2 stage and, if productive, results in CD4–CD8– (DN) cells that express TCR on the cell surface. Rearrangement of variable TCR? gene segments sets in slightly later and, also if productive, results in the surface expression of the pre-TCR consisting of a TCR? chain that is covalently associated with the pre-TCR chain and noncovalently associated with CD3 signal-transducing molecules (3, 4). The pre-TCR is expressed at rather low levels on DN3 and DN4 cells (5). Pre-TCR–expressing DN cells undergo several rounds of division before the rearrangement of variable TCR gene segments sets in at the late DN4 and early CD4+CD8+ double-positive (DP) stage. This results in the surface expression of ?TCRs at the expense of the pre-TCR because TCR chains generally compete favorably with pre-TCR for TCR? chains. TCR?-expressing DP cells then undergo positive or negative selection by intrathymic peptide–MHC complexes (6).
The temporal order of TCR V gene segment rearrangement has suggested the following scenario: DN4 cells that contain productive Tcrg and Tcrd rearrangements express TCR on the cell surface and, thereby, become functionally mature T cells that are ready to leave the thymus. Pre-TCR–expressing cells will eventually become DP thymocytes expressing an ?TCR, the specificity of which determines their further developmental fate. This simple scheme of T cell development was found not to be valid when T cell development was analyzed in Ptcra-deficient mice. In contrast to CD3-deficient and Tcrb plus Tcrd double-deficient mice that contain none or very few DP thymocytes, Ptcra-deficient mice were shown to harbor strongly reduced but still considerable numbers of DP thymocytes (7). Further analysis in Ptcra,Tcrd as well as Ptcra,Tcra double-deficient mice then revealed that both an early expressed TCR as well as an early expressed ?TCR could still rescue the development of DP thymocytes in the absence of pre-TCR (8). In the case of ?TCR in Ptcra–/–,Tcrd–/– mice, this resulted in thymocytes of which >95% harbored TCR? chains (i.e., DP cells that were selected by an early expressed ?TCR and therefore contained in-frame Tcrb rearrangements). In the case of an early expressed TCR in Ptcra–/–,Tcra–/– mice, only 15% of DP cells contained TCR? chains. Thus, these experiments indicated that receptors other than the pre-TCR could relieve DN3 cells from a development block resulting in the production of DP cells and that there must be a rearrangement of TCR V gene segments at the DN3 or earlier stages of T cell development.
The temporal order of TCR V gene segment rearrangement appears especially important with regard to the Tcra and Tcrd locus. This is where the Tcra locus is embedded in the Tcra locus and where the early rearrangement of V gene segments results in the formation of TCR chains, whereas late V rearrangement is accompanied by the deletion of the Tcrd locus and generation of Tcra genes (9). Two different enhancer elements have been invoked in the control of rearrangement and expression of the Tcra/Tcrd locus: the E enhancer located in the J-C intron and the E enhancer 4 kb downstream of C (10–13). Both enhancers have been deleted by homologous recombination. In E–/– animals, the development of ? T cells appeared to proceed normally with the exception that there was a considerable reduction of thymic and peripheral T cells. In this context, E-deficient alleles exhibited a substantial reduction of Tcrd gene rearrangements. E–/– mice contained normal numbers of DP cells but reduced numbers of CD4+CD8– and CD4–CD8+ single-positive (SP) cells as a result of an almost complete block in V to J rearrangements. The E enhancer also controls levels of ? as well as TCR expression, as evident by reduced levels of TCR and TCR transcripts in E–/– mice. These results indicate that E functions in DN thymocytes to promote Tcrd gene rearrangement and gene expression but not J accessibility to the V(D)J recombinase. On the other hand, E functions in DP thymocytes to promote TCR gene rearrangement via J accessibility and controls both Tcrd and Tcra gene expression. However, because TCR chains were still present in E–/– mice, it was hypothesized that in the absence of E, the E enhancer or other elements might have a role in promoting rearrangement, possibly via the promotion of J accessibility and/or expression of some V segments (11). This would account for the TCR chains with limited diversity (mostly V2) that were expressed in peripheral lymphoid tissue in E-deficient mice.
In this study, we have tested the hypothesis that a low level of V to J rearrangements is controlled by the E enhancer and occurs in DN thymocytes earlier than the bulk of E-controlled V to J rearrangements, thus leading to the expression of ?TCR in DN cells, which permits some of these cells to enter the ? lineage of DP thymocytes.
RESULTS
Early ?TCR-expressing DN cells
Analyzing total thymocytes and DN thymocytes from wild-type (wt) C57BL/6 mice, we find that 5% express TCR? chains on the cell surface of which 10% are paired with V2-containing TCR chains (Fig. 1 A ). Because a subset of NKT cells (14) has the TCR?+ CD4–CD8– phenotype, NK1.1 cells were excluded by using NK1.1 antibodies for depletion as well as by the analysis of NKT cell–deficient CD1–/– animals (Fig. 2 ). The results show that V+,TCR?+ cells belong to a distinct, non–NKT cell subset of DN cells. In addition, we found TCR?+ DN T cells in secondary lymphoid organs in both wt and CD1- deficient animals (Fig. 2).
Figure 1. V2-enriched early ? T cells in wt and Ptcra–/– mice. (A) TCR? versus V2 FACS profile in thymi of wt and Ptcra–/– thymocytes. DN (CD4–CD8–) cells were electronically gated after the exclusion of CD4+, CD8+, and NK1.1 cells. (B) A CD4 versus CD8 profile of Ptcra–/– thymi. (C and D) Staining of Ptcra–/– thymocytes (total or CD4–CD8–) with TCR? and V2 antibodies. Numbers in quadrants indicate the percentages of cells in that quadrant.
Figure 2. CD4–CD8– ? T cells in the thymus and lymphoid periphery. (A) CD4,CD8 versus NK1.1 staining in the thymus of wt and CD1–/– mice (top). TCR? versus V2 analysis of NK1.1– CD4–CD8– thymocytes (bottom). (B and C) Identification of peripheral CD4–CD8– TCR?+ cells in the spleen and lymph nodes using TCR?, V2, CD4, and CD8 antibody labeling. Numbers in quadrants indicate the percentages of cells in that quadrant.
To investigate whether DN ?TCR-expressing cells are pre-TCR selected as their DP (CD4+CD8+) or SP (CD4+CD8– and CD4–CD8+) counterparts, thymocytes from Ptcra–/– mice were analyzed (Fig. 1 B). Although Ptcra–/– mice were previously shown to be deficient in NKT cells (15), we used NK1.1 antibodies in the depletion procedure that yielded DN thymocytes. In Ptcra–/– mice, 6–7% of DN thymocytes express TCR? chains on the cell surface, and, in this particular experiment, 40–50% of DN cells with TCR? proteins coexpressed V2,TCR chains (Fig. 1 C). In several experiments, the proportion of V2+ cells among TCR?+ DN cells in Ptcra–/– mice was variable and ranged between 10 and 50% (Fig. 4; see Fig. 7).
Of interest was the observation that V2 is not only contained in ? but also in TCRs in DN thymocytes from Ptcra–/– mice (Fig. 1 C), a notion consistent with recent findings that V2 gene segments can join to both J and DJ sequences (16). To exclude that the existence of TCR?+,V2+ DN cells is the result of V2 segment pairing with TCR diversity joining elements (16), a similar analysis was conducted in Ptcra–/–,Tcrd–/– double-deficient mice. About 7% of CD4–CD8– cells were found to express TCR? proteins on the cell surface, and 25% of the TCR? chains were paired with V2-containing TCR chains, suggesting that these TCR?-expressing cells are bona fide ? T cells that do not require pre-TCR selection (Fig. 3 A ).
Figure 3. Phenotype and developmental potential of early ? T cells. (A) V2 versus TCR? staining of Ptcra–/–,Tcrd–/– thymocytes. (B) Phenotypic analysis of CD4–CD8– and CD4–CD8– TCR?+ cells using CD44 and CD25 antibodies. (C) Embryonic day 14.5 fetal (Rag-1–/–) thymic organ culture of CD4–CD8– TCR?+ NK1.1– Ptcra–/– cells. Cells were cultured for 7 d and were stained with V2, TCR?, CD4, and CD8 antibodies. A CD4 versus CD8 staining of V2+ donor cells is also shown (bottom). Numbers in quadrants indicate the percentages of cells in that quadrant.
Further phenotypic analysis of the TCR?+ thymocytes showed that almost all "early" TCR?-expressing CD4–CD8– cells are CD25 negative and, thus, belong to the DN4 (CD25–CD44–) subset (Fig. 3 B). The developmental potential of early ?TCR-expressing DN cells was then addressed by culturing purified LY5.2+ CD4–CD8– NK1.1– TCR?+ Ptcra–/– thymocytes together with embryonic thymi from LY5.1+ in Rag1–/– donors in fetal thymic organ cultures. After 7 d of culture, 40% of the donor-derived cells have up-regulated the expression of CD4,CD8 coreceptors. Moreover, >90% of the cultured thymocytes retain the surface expression of TCR?, and 20% of them are V2+ (Fig. 3 C). Not all cells up-regulate CD4 and CD8, however, and a substantial fraction of cells expressing high levels of V2-containing ?TCRs remain CD4–CD8–. It is likely that these ?TCR-expressing DN cells normally exit the thymus because CD4–CD8– ?TCR+ cells can be detected in the lymph nodes and spleen of adult mice (Fig. 2), and it was shown in TCR? transgenic mice that DN cells with the transgenic TCR can accumulate in peripheral lymphoid tissue (6). Thus, TCR?-expressing DN cells can give rise to both immature DP and mature SP (CD4/8) T cells as well as DN ? T cells that do not enter the DP ? lineage (17).
The E enhancer controls late but not early development of ? T cells in Ptcra–/– mice
To address the role of the described TCR enhancers on early TCR expression, we generated Ptcra–/– mice with a targeted deletion of the E locus (Ptcra–/–,E–/–). Ptcra–/–, E–/– and Ptcra–/–,E+/+ littermate mice contain similar numbers of thymocytes (Fig. 4 A ). However, there is a clear reduction of CD4+CD8– SP thymocytes in the Ptcra–/–, E–/– mice because of limited TCR diversity and/or expression levels (Fig. 4 B). This is supported by the staining of either all thymocytes or only DN thymocytes with a combination of TCR? and Va2 antibodies: although total thymocytes from Ptcra–/– single-deficient mice contain <10% TCR?+,V2+ cells, among TCR?+ cells, this proportion is much higher (>30%) in Ptcra–/–,E–/– double-deficient mice. In DN thymocytes from Ptcra–/–,E+/+ mice, TCR?+,V2+ cells represent 10% of all TCR?+ cells, whereas in Ptcra–/–,E–/– mice, TCR?+,V2+ cells represent 50%. These data indicate that E is not required for the early rearrangement and expression of Tcra genes (Fig. 4 C). The data show that normally E predominantly contributes to the rearrangement and expression of TCR V gene segments other than V2 gene segments. Consistent with this notion, the spleen of Ptcra–/–,E–/– mice contains almost exclusively V2+,TCR?+ cells, whereas only 10% of V2+,TCR?+ cells among TCR?+ cells are found in the spleen of Ptcra–/–,E+/+ mice (Fig. 4 D). These observations are in line with earlier observations in E–/– mice (11) showing that TCR?+ cells in the spleen of E–/– mice express almost exclusively V2 (Fig. 4 D).
Figure 4. E controls late but not early development of ? T cells. (A) Absolute cell numbers of Ptcra–/– and Ptcra–/–,E–/– littermates. Error bars represent SD. (B) CD4 versus CD8 profiles of littermates belonging to all indicated genotypes. (C and D) V2 versus TCR? antibody labeling in total DN thymocytes (C) and spleen cells (D). Numbers in quadrants indicate the percentages of cells in that quadrant.
DN cells but not mature T cells in E–/–-deficient mice express V2-negative Tcra genes
The data in Fig. 4 suggest that DN cells in E–/– mice can express V genes other than V2 and that the V2 dominance in the periphery of E–/– mice is established at a later developmental stage. This issue was addressed in more detail by analyzing the expression of TCR V2 genes and other TCR V genes (V3, V8, and V11) in Ptcra–/– and Ptcra–/–,E–/– mice at various stages of development. As shown in Fig. 5 B , NKT cell–depleted CD4– and CD8– (DN) cells from both Ptcra–/– and Ptcra–/–,E–/– mice do express V genes that stain with a cocktail of V3, 8, and 11 antibodies. Although the proportion of cells stained with the cocktail versus V2-positive cells remains about the same in CD4+CD8+ cells of Ptcra–/– mice, it drastically decreases in DP cells of E–/– mice such that the vast majority of cells expresses V2 (Fig. 5 A). This trend is also evident in peripheral T cells from E–/– but not wt mice in which virtually all TCR?+ cells express V2 (Fig. 4 D). Thus, these data indicate that the predominance of V2 expression in peripheral T cells of E–/– mice is not caused by the fact that the E enhancer only allows for V2 rearrangement at the DN stage of T cell development. Instead, this is likely the result of the fact that V2 gene segments are able to sustain a sufficiently high expression of Tcra genes in the absence of the E enhancer such that only V2-positive ? T cells can be positively selected and maintained in peripheral lymphoid tissue.
Figure 5. Immature but not mature T cells from E–/– mice express diverse TCR chains. (A) Staining of thymocytes from Ptcra–/–,E+/+ and Ptcra–/–,E–/– mice with CD4 and CD8 (top), and staining of CD4+CD8+ TCR?+ cells with V3 + V8 + V11 antibodies versus V2 antibodies (middle). Bottom panel shows isotype controls for the staining in the middle panel. (B) Staining of NK1.1– CD4–CD8– TCR? thymocytes from Ptcra–/–,E+/+ and Ptcra–/–,E–/– mice with V3 + V8 + V11 antibodies versus V2 antibodies (top) and with isotype controls (bottom). Numbers in quadrants indicate the percentages of cells in that quadrant.
E controls V expression in DN cells
Because the analysis of Ptcra–/–,E–/– mice has shown that E is not an enhancer element required for early TCR? expression and pre-TCR–independent T cell development (8, 11), we have focused our attention on the E enhancer by generating and studying animals that lack both Ptcra as well as the E enhancer element (Ptcra–/–,E–/– mice). When compared with Ptcra–/–,E+ thymi, it was noted that Ptcra–/–,E–/– thymi contain much-reduced numbers of thymocytes (Fig. 6 A) that are severely deficient in DP and SP cells (Fig. 6 B ). Also, the number of TCR-expressing cells is reduced even though some T cells are still present. Within the DN compartment, there is a more complete block at the DN3 stage of development in Ptcra–/–,E–/– versus Ptcra–/–,E+ mice (Fig. 6, C and D).
Figure 6. E is essential for pre-TCR–independent T cell development. (A) Absolute cell numbers of Ptcra–/– and Ptcra–/–,E–/– littermates. 4–8-wk-old mice were analyzed. Error bars represent SD. (B and C) CD4 versus CD8 (B) and TCR? versus TCR (C) profiles of littermate thymi. (D) Analysis of the DN compartment using CD25 and CD44 antibodies. Numbers in quadrants indicate the percentages of cells in that quadrant.
Ablation of E in Ptrcra–/– mice has a profound effect on the percentage and absolute number of DN ?TCRs as well as of V2-expressing cells, as TCR?+,V2+ thymocytes are virtually absent (<0.25%; Fig. 7, A and B ). However, there was a small number of TCR?-expressing cells among total thymocytes in some of the Ptcra–/–,E–/– mice (Fig. 7), suggesting some remaining level of TCR rearrangement in the Ptcra–/–, E–/– thymi (perhaps mediated by the E enhancer in cells) that was "rescued" by the expression of TCR at the DN developmental stage (8).
Figure 7. Absence of V2-expressing cells in the thymi of Ptcra–/–,E–/– mice. TCR? versus V2 staining of either total thymocytes (A) or CD4–CD8– cells (B) of littermate mice analyzed at 4 wk of age. Numbers in quadrants indicate the percentages of cells in that quadrant.
Thus, in the absence of Ptcra, the E enhancer has a crucial role in the early assembly of TCR chains as well as TCR chains in DN3 and/or earlier stages of T cell development (i.e., TCR chains that are required for the formation of and ?TCR in the absence of pre-TCR). The ?TCRs (and TCRs) can rescue some developmental progression beyond the DN3 stage in Ptcra-deficient mice, which appropriately explains the incomplete developmental block in Ptcra–/– versus CD3–/– or Tcrb–/– x Tcrd–/– mice (i.e., mice that cannot assemble any TCR on the cell surface).
DISCUSSION
The results obtained in mice with a combined deficiency of the pre-TCR chain plus the E or E enhancer provide an explanation for the observation that development of DN thymocytes into ? lineage DP cells in Ptcra–/–,Tcrd–/– double-deficient mice can be rescued by ?TCR. The data indicate that E can promote early V to J rearrangements, which results in the expression of an ?TCR on the surface of DN4 thymocytes. The early expression of an ?TCR was previously considered to represent a transgenic artifact caused by the too early expression of TCR chains in TCR transgenic mice (18), but, as shown here, it also occurs under physiological conditions. Our experiments and earlier experiments in TCR transgenic mice show that some but not all of the ?TCR CD4–CD8– cells can become CD4+CD8+ cells (8, 18). This developmental pathway may make only a limited contribution to the generation of DP ? lineage cells in pre-TCR–competent mice not only because of the paucity of TCR chains in DN cells of normal mice but also because TCR is a bad "surrogate" for pre-TCR. Indeed, we found that under competitive conditions, pre-TCR is far more effective in the generation of DP cells than ?TCR (19).
The fate of the DN cells that express ?TCR on the cell surface and do not become DP thymocytes needs to be further evaluated. These cells apparently can leave the thymus because they can be detected in the spleen and lymph nodes of normal (Fig. 2) as well as in exaggerated numbers in TCR transgenic mice. Such cells were previously shown to acquire functional maturity (i.e., respond with proliferation and cytokine production to TCR ligation) and to acquire CD8 expression when antigenically stimulated. In fact, this unusual subset of T cells was analyzed in TCR transgenic mice many years ago, and it was concluded that the early expression of an ?TCR could mimic signals generated when thymocytes express TCR and become functionally mature (18, 20). Of interest is that such cells can express an autoreactive TCR but are not deleted because of the lack of coreceptors and can accumulate in secondary lymphoid tissue of mice, as shown in various TCR transgenic models (21). Some of these cells exhibit an activated phenotype and have an as yet undefined role in the immune system.
Recent data have, in fact, shown that the extensive accumulation of CD8 cells in the gut of TCR transgenic mice (22) to a large extent depends on the premature expression of the transgenic TCR in DN cells because "on time" expression of the same Tcra transgene does not result in the strong accumulation of CD8 cells in the gut (23). Another possibility is that thymic DN ?TCR+ cells represent precursors of some peripheral regulatory T cells. Perhaps these cells are akin to both mouse and human ?TCR+ CD3+ NK.1.1– CD4–CD8– DN Regulatory T cells that can suppress antigen-specific immune responses mediated by CD8+ and CD4+ T cells through a process that requires cell to cell contact and Fas–FasL interactions (24). In this regard, it is important to point out that the ?TCR+ DN studied here are different from NK1.1+ DN T cells and from ?TCR+ DN cells studied by others who concluded that all TCR?+ DN cells were derived from DP precursors (25); either this generalization is wrong or the fate mapping approach used by the authors is not valid. However, it is clear that the TCR?+ DN cells in Ptcrd–/–, Tcrd–/– mice are involved in the rescue of development rather than being derived from DP cells (8).
The results shown in this study also provide an adequate explanation for earlier observations in E-deficient mice that exhibited a TCR repertoire limited to TCR,V2 chains. In this context, on E-deleted alleles, other cis-acting elements such as E and/or V2 promoters were hypothesized to promote either only V2 to J rearrangements or V to J rearrangements. These rearrangements involve a diverse array of V segments with assembled V2J complexes expressed in a much higher proportion in the absence of E. Our results indicate that in DN cells, E can direct a low level of V to J rearrangements, possibly via promoting J accessibility that results in the expression of a variety of different V gene segments. The early expressed ?TCRs allow the development of some DP thymocytes with TCRs containing mostly V2+,TCR chains. In the absence of E, only early VJ rearrangements with a high proportion of V2 would continue to be expressed in DP thymocytes and mature T cells, perhaps because their promoters do not require E activity to drive gene expression. This pathway of differentiation observed in E-deficient mice might be invisible in wt mice because of continual Tcra rearrangement in DP thymocytes that will swamp out the E-initiated V rearrangements and, thereby, lead to a much more diverse ?TCR repertoire.
MATERIALS AND METHODS
Mice.
Mice were kept in the sterile facilities of The University of Chicago, Dana-Farber Cancer Institute, and Children's Hospital. Animal protocols were approved by the Institutional Animal and Use Committees of these institutes. C57BL/6, Rag1–/–, and Tcrd–/– mice (also on the C57BL/6 background) were purchased from Jackson ImmunoResearch Laboratories. C57BL/6 CD1–/– were provided by A. Bendelac (The University of Chicago, Chicago, IL). E–/– and E–/– mice were generated in the laboratory of Frederick W. Alt (10, 11). Ptcra–/– mice were described previously (7).
Flow cytometric analysis and cell sorting.
Anti-CD4 (L3T4), CD8 (53–6.7), CD25 (3C7), CD44 (IM7), NK1.1 (PK136), TCR? (H57-597), TCR (GL3), Va2 (B20.1), Va11 (RR8-1), and Va3 (RR3-16) mAbs were purchased from BD Biosciences. The Va8 (CTVA8) antibody was purchased from CALTAG. These mAbs were directly coupled to FITC, PE, Cy-crome, APC, or biotin. Surface marker expression on thymocytes and peripheral T cells was visualized using a FACScalibur (Becton Dickinson) and analyzed with FlowJo (Tree Star) and CellQuest software (Beckton Dickinson). Cell sorting was performed using Mo-Flo (DakoCytomation) and FACS-Aria (Becton Dickinson) sorters.
Fetal thymic organ culture.
Thymi were cultured as described previously (26). In brief, (LY5.2+) TCR?+ CD4–CD8– NK1.1– cells were FACS purified from thymi of Ptcra–/– animals. Lineage-negative (CD3, CD8, Mac-1, NK1.1, Gr-1, Ter-119, and CD19) bone marrow progenitors were also used as a reconstitution control. Isolated embryonic day 14.5 thymi from Rag1 mice (expressing LY5.1) were initially incubated in Terasaki plates (Nunc), subsequently applied on Transwell (Nunc) porous filters, and incubated in Iscove's modified Dulbecco's medium supplemented with 10% fetal calf serum (Sigma-Aldrich) plus penicillin, streptomycin, and mercapto-ethanol. Cultures were maintained at 37°C for 7 d, after which LY5.2+ cells were stained and analyzed by FACS analysis.
Acknowledgments
We would like to thank A. Bendelac for the CD1–/– animals, F. Meng for expert technical assistance, R. Duggan and J. Marvin for cell sorting, and I. Apostolou for advice and technical expertise.
I. Aifantis was supported by the National Institutes of Health (NIH) grant R01 CA105129 and the V Foundation for Cancer Research. K. Sawai is supported by a University of Chicago Molecular Biology Training grant. H. von Boehmer was supported by NIH grants R01 AI45846 and R01 AI47281. F.W. Alt was supported by NIH grant R01 AI20047 and is a Howard Hughes Medical Institute Investigator.
References
Godfrey, D.I., and A. Zlotnik. 1993. Control points in early T-cell development. Immunol. Today. 14:547–553.
Rodewald, H.R., and H.J. Fehling. 1998. Molecular and cellular events in early thymocyte development. Adv. Immunol. 69:1–112.
Groettrup, M., K. Ungewiss, O. Azogui, R. Palacios, M.J. Owen, A.C. Hayday, and H. von Boehmer. 1993. A novel disulfide-linked heterodimer on pre-T cells consists of the T cell receptor beta chain and a 33 kd glycoprotein. Cell. 75:283–294.
Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H.J. Fehling, and H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science. 266:1208–1212.
Aifantis, I., V.I. Pivniouk, F. Gartner, J. Feinberg, W. Swat, F.W. Alt, H. von Boehmer, and R.S. Geha. 1999. Allelic exclusion of the T cell receptor ? locus requires the SH2 domain-containing leukocyte protein (SLP)-76 adaptor protein. J. Exp. Med. 190:1093–1102.
von Boehmer, H. 2004. Selection of the T cell repertoire: receptor- controlled checkpoints in T cell development. Adv. Immunol. 84:201–238.
Fehling, H.J., A. Krotkova, C. Saint-Ruf, and H. von Boehmer. 1995. Crucial role of the pre-T-cell receptor alpha gene in development of alpha beta but not gamma delta T cells. Nature. 375:795–798. (published erratum appears in Nature. 1995. 378:419).
Buer, J., I. Aifantis, J.P. DiSanto, H.J. Fehling, and H. von Boehmer. 1997. Role of different T cell receptors in the development of pre-T cells. J. Exp. Med. 185:1541–1547.
Chien, Y.H., M. Iwashima, K.B. Kaplan, J.F. Elliott, and M.M. Davis. 1987. A new T-cell receptor gene located within the alpha locus and expressed early in T-cell differentiation. Nature. 327:677–682.
Monroe, R.J., B.P. Sleckman, B.C. Monroe, B. Khor, S. Claypool, R. Ferrini, L. Davidson, and F.W. Alt. 1999. Developmental regulation of TCR delta locus accessibility and expression by the TCR delta enhancer. Immunity. 10:503–513.
Sleckman, B.P., C.G. Bardon, R. Ferrini, L. Davidson, and F.W. Alt. 1997. Function of the TCR alpha enhancer in alphabeta and gammadelta T cells. Immunity. 7:505–515.
Hernandez-Munain, C., B.P. Sleckman, and M.S. Krangel. 1999. A developmental switch from TCR delta enhancer to TCR alpha enhancer function during thymocyte maturation. Immunity. 10:723–733.
Balmelle, N., N. Zamarreno, M.S. Krangel, and C. Hernandez-Munain. 2004. Developmental activation of the TCR alpha enhancer requires functional collaboration among proteins bound inside and outside the core enhancer. J. Immunol. 173:5054–5063.
Bendelac, A., M.N. Rivera, S.H. Park, and J.H. Roark. 1997. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535–562.
Eberl, G., H.J. Fehling, H. von Boehmer, and H.R. MacDonald. 1999. Absolute requirement for the pre-T cell receptor alpha chain during NK1.1+ TCRalphabeta cell development. Eur. J. Immunol. 29:1966–1971.
Jouvin-Marche, E., C. Aude-Garcia, S. Candeias, E. Borel, S. Hachemi-Rachedi, H. Gahery-Segard, P.A. Cazenave, and P.N. Marche. 1998. Differential chronology of TCRADV2 gene use by alpha and delta chains of the mouse TCR. Eur. J. Immunol. 28:818–827.
von Boehmer, H., I. Aifantis, O. Azogui, J. Feinberg, C. Saint-Ruf, C. Zober, C. Garcia, and J. Buer. 1998. Crucial function of the pre- T-cell receptor (TCR) in TCR beta selection, TCR beta allelic exclusion and alpha beta versus gamma delta lineage commitment. Immunol. Rev. 165:111–119.
von Boehmer, H., J. Kirberg, and B. Rocha. 1991. An unusual lineage of ? T cells which contain autoreative cells. J. Exp. Med. 174:1001–1008.
Borowski, C., X. Li, I. Aifantis, F. Gounari, and H. von Boehmer. 2004. Pre-TCR and TCR are not interchangeable partners of TCR? during T lymphocyte development. J. Exp. Med. 199:607–615.
Bruno, L., H.J. Fehling, and H. von Boehmer. 1996. The ?TCR can replace the TCR in the development of lineage cells. Immunity. 5:343–352.
Terrence, K., C.P. Pavlovich, E.O. Matechak, and B.J. Fowlkes. 2000. Premature expression of T cell receptor (TCR)? suppresses TCR gene rearrangement but permits development of lineage T cells. J. Exp. Med. 192:537–548.
Rocha, B., H. von Boehmer, and D. Guy-Grand. 1992. Selection of intraepithelial lymphocytes with CD8 alpha/alpha co-receptors by self-antigen in the murine gut. Proc. Natl. Acad. Sci. USA. 89:5336–5340.
Baldwin, T.A., M.M. Sandau, S.C. Jameson, and K.A. Hogquist. 2005. The timing of TCR expression critically influences T cell development and selection. J. Exp. Med. 202:111–121.
Ford, M.S., K.J. Young, Z. Zhang, P.S. Ohashi, and L. Zhang. 2002. The immune regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J. Exp. Med. 196:261–267.
Egawa, T., G. Eberl, I. Taniuchi, K. Benlagha, F. Geissmann, L. Hennighausen, A. Bendelac, and D.R. Littman. 2005. Genetic evidence supporting selection of the V14i NKT cell lineage from DP thymocyte precursors. Immunity. 22:705–716.
Martin, C.H., I. Aifantis, M.L. Scimone, U.H. von Andrian, B. Reizis, H. von Boehmer, and F. Gounari. 2003. Efficient thymic immigration of B220+ lymphoid-restricted bone marrow cells with T precursor potential. Nat. Immunol. 4:866–873.(Iannis Aifantis1,4, Craig)