Molecular Mechanisms of Pre-T Cell Receptor-induced Survival*

En route to maturing as T cell receptor (TCR) αβ-expressing cells, the development of thymocytes is contingent on expression of a pre-TCR complex comprising a TCRβ chain paired with a surrogate TCRα chain, pre-Tα (pTα). The pre-TCR has been proposed to promote cell survival, proliferation, differentiation, and lineage commitment. However, the precise molecular mechanisms governing this variety of effects remain elusive. Here, we present a cellular system designed to biochemically dissect signals elicited upon pre-TCR expression. Using the T cell line 4G4 stably transfected with one of the two known pTα isoforms or selective pTα deletion mutants and TCRβ, we were able to observe that expression of a functional pre-TCR complex is sufficient to control the levels of surface Fas protein, the stimulation of mitogen-activated and stress-regulated kinases, and the activation status of the p53 antioncogene. We demonstrate that this regulation has a major impact on the expression of important regulators of apoptosis, such as Bcl-2 family members, and the cell cycle, such as p21WAF. Furthermore, we show here that cells expressing a functional pre-TCR are more resistant to different types of DNA damage-induced apoptosis and that these effects are contingent on an intact cytoplasmic tail of pTα. We finally propose that the presence of a functional pre-TCR complex triggers many intracellular pathways capable of driving and ensuring thymocyte survival in the presence of DNA damage.

Development of immature thymocytes is contingent on passage through at least two major checkpoints: positive selection mediated by interactions of TCR␣␤ with proteins of the major histocompatibility complex, and the ␤-selection point mediated by major histocompatibility complex-independent signaling through the pre-TCR 1 (1)(2)(3)(4)(5). The most immature thymocytes are double negative (DN) for CD4 and CD8 expression and progress through four subsets defined by CD44 and CD25 expression: DN I (CD25 Ϫ CD44 ϩ ), DN II (CD25 ϩ CD44 ϩ ), DN III (CD25 ϩ CD44 lo ), and DN IV (CD25 lo CD44 lo ) (6). From these arise CD25 lo CD44 lo CD3 lo CD4 Ϫ CD8 ϩ immature single positive cells that develop into cells double positive (DP) for CD4 and CD8 expression (7), from which mature TCR␣␤(ϩ) thymocytes are selected. The major changes associated with TCR␤ gene rearrangement occur at the transition from DN III to DN IV cells, precisely when the pre-TCR is expressed. However, whether the presence of pre-TCR is sufficient to induce the concomitant occurrence of these many events remains still an unresolved matter.
Like TCR␣␤, the pre-TCR is a multicomponent signaling complex comprising a TCR␤ chain paired with at least one of two isoforms of a pT␣ molecule both associated with CD3 chains, specifically ⑀, either ␥ or ␦, and to some extent (8 -13). The pre-TCR spontaneously clusters and associates with signaling molecules such as p56lck, CD3 molecules, and zap-70 via sequestration in lipid rafts even in the absence of any extracellular ligand (14), consistent with the idea that the main role for the pre-TCR is to facilitate pairing with the CD3 complex. However, unlike TCR␣␤, the pre-TCR has an additionally extended cytoplasmic tail encoded by the pT␣ gene. Within the tail two proline-rich motifs can be identified by sequence similarity to motifs in the cytoplasmic tail of human CD2 that mediate binding to CD2BP2, an adaptor molecule involved in intracellular signaling (8,16). Heretofore, the importance of the intracytoplasmic region of the pre-TCR has been uncertain, but very recently the expression of pT␣ mutants in retrovirally transduced T cell precursors and cell lines showed that the pT␣ cytoplasmic tail, in particular the proline-rich domain, plays a crucial role in pre-TCR signal transduction (17).
Based on the analysis of cells on either side of the DN III to DN IV transition, the pre-TCR has been hypothesized to regulate thymocyte survival, proliferation, differentiation, and lineage commitment (18 -21). Several classes of molecules implicated in the regulation of apoptosis were noted to change as DN III cells moved on to become DN IV cells. First, the prototypic death receptor, Fas, shows very poor expression in DN cells compared with the high levels found in DP thymocytes (22). Second, as recently reported (23), changes were noted in the DNA binding activity of NF-B, a known inhibitor of proapoptotic signaling from death receptor pathways. In mice transgenic for a luciferase gene driven by a NFB-responsive element, luciferase activity was significantly greater in DN IV cells than in DN III and dropped precipitously in DP cells (23). Third, apoptosis induced by DNA damage requires an intact p53 antioncogene in thymocytes (24) and occurs via transcriptional activation of the cyclin-dependent kinase inhibitor p21 WAF (25). A lack of p53 in a CD3␥-deficient background impairs cell death in DN thymocytes and partially rescues the block in pre-TCR cell differentiation caused by this pre-TCR defect (26).
The analysis of mutant mice expressing variant forms of either the pre-TCR or of molecules putatively associated with it support the fact that only thymocytes that succeed in generating a functional TCR␤ chain selectively survive through the transition from DN III to DN IV. However, despite the elegance of these in vivo experiments, there is little direct evidence showing a cause and effect relationship between the pre-TCR expression and its proposed anti-apoptotic function. This is of some concern because multiple regulators of thymocyte survival and fate are expressed in vivo (e.g. Notch and Interleukin-7), any of which may be primarily responsible for changes observed in cells as across the DN III to DN IV transition (27,28). In addition, the very low levels of expression of the pre-TCR in thymocytes make it very difficult to detect and even more difficult to biochemically characterize this molecule. All of these difficulties emphasize the need for additional, complementary approaches to study pT␣ function.
In this report we undertake one such new approach by using exogenous expression of the pre-TCR genes to examine the influence that this molecule exerts on some important regulators of programmed cell death (apoptosis) and thymocyte differentiation. Here, we describe the development and characterization of a variety of stably transfected clones expressing different pre-TCR chains in a T cell line able to express a functional pre-TCR complex upon transfection of the pT␣ and TCR␤ chains (11). The capacity of the pre-TCR to regulate the expression of Fas, Bcl-2 family members, and the activity of key molecules such as p53 and stress-regulated kinases is analyzed here. We also test in this system the effect of pre-TCR expression in the regulation of cell death induced by DNA damage. Altogether, the results presented in this study allow us to conclude that expression of a complete and functional pre-TCR complex is able per se to regulate multiple signals that bring about the inactivation of p53, activation of mitogen-activated protein kinases/stress-activated protein kinases, and downregulation of pro-death gene products such as Fas, p21, and Bax.

EXPERIMENTAL PROCEDURES
DNAs, Transfections, and Generation of Stable Clones-4G4 cells were electroporated with DNAs encoding a TCR␤ chain alone or together with HA-tagged pT␣ a and pT␣ b constructs and were selected in G418 (Invitrogen), as described (11). Clones transfected with either one isoform of the pT␣ chain alone, without cotransfection of the TCR␤ chain, expressed the pT␣ mRNA as could easily be detected by reverse transcription-PCR. However, we were unable to detect any pT␣ protein in those selected clones either by FACS or by Western blot, thus indicating that the pre-T␣ chain does not form a stable protein complex when devoid of TCR␤, consistent with what has been described before (29). A truncated form of pT␣ b , named ⌬P1P2, in which the last cytoplasmic 16 amino acids containing two proline-rich regions were deleted, was PCR-generated with the following primers: 5Ј-AATA-GATCTCTACCATCAGGGGAATCT-3Ј (containing a BglII site) and 5Ј-AATCCGCGGCTACTGGAGGTGCTGGCCCGC-3Ј (containing a SacII site). The product was cloned into pGEM-T-Easy (Promega), from which it was subcloned in frame using BglII/SacII into the expression vector pDisplay that includes an N-terminal HA tag (Invitrogen) (13).
Analysis of Transfectants for Pre-TCR Components-Expression of transfected proteins was analyzed by Western blot and by intracellular flow cytometry, as described (11). For Western blots, 10 7 cells were collected by centrifugation and resuspended in lysis buffer as described previously (30); subsequently the protein content of each sample was measured (DC protein assay; Bio-Rad). For detection of pT␣, 600 g of total cell lysates were immunoprecipitated with anti-HA antibodies (12CA.5; Roche Molecular Biochemicals). The precipitates were eluted off beads by boiling, separated in 15% SDS-PAGE in parallel with prestained markers, and transferred to polyvinylidene fluoride membranes (Immobilon P; Millipore). Proteins containing a HA epitope were detected with another anti-HA antibody (HA.11; Covance) followed by peroxidase-conjugated secondary antibody (Cappel) and developed with a chemiluminescent method (ECL; Amersham Biosciences). For detection of TCR␤, the lysates were resolved by PAGE, transferred to poly-vinylidene fluoride membranes, and directly detected with an antibody directed against the C terminus of TCR␤ (catalog number 1579 from Santa Cruz Biotechnology). For intracellular flow cytometry, the cell suspensions were fixed in 1% paraformaldehyde for 10 min, washed in phosphate-buffered saline, and resuspended in a 0.3% saponin (Sigma) buffer for 10 min. Further staining steps were carried out in 0.1% saponin buffer. The antibodies used were phycoerythrin-conjugated H57-597 and fluorescein isothiocyanate-conjugated 12CA.5. The cells were washed in phosphate-buffered saline and analyzed immediately on a FACS Calibur. Data analysis was performed using CellQuest.
Induction of UV-induced Apoptosis in Pre-TCR Transfectants-Susceptibility to apoptosis in 4G4-derived clones was determined as follows. The cells were grown on poly-D-lysine-treated coverslips for 1-2 days before being subjected to UV irradiation (120 Jul/m 2 ) as described previously (30). The cells were then maintained in a serum-free, G418free medium for 6 h and fixed for 30 min in 4% paraformaldehyde in phosphate-buffered saline. After permeabilization in 2% Triton X-100 and blocking for 1 h in 4% bovine serum albumin in phosphate-buffered saline, the TUNEL reaction was performed using an in situ cell death detection kit (Roche Molecular Biochemicals) following the manufacturer's instructions, except that the labeling reaction was performed at 25°C instead of at 37°C. After DAPI staining, the cells were mounted on coverslips and counted using a Zeiss Axioplan2 fluorescence microscope. The percentage of apoptotic cells was calculated by counting total (DAPI-stained) and apoptotic (TUNEL-positive) cells.
Fas expression was measured by flow cytometry using the Jo-2 antibody (22) (anti-FAS-phycoerythrin (BD-Pharmingen)) and expressed in arbitrary units reflecting mean fluorescence intensities. All of the data plots showed unimodal distribution of Fas expression.
Analysis of Transfectants for Stimulation of Mitogen-and Stressactivated Kinases-Total cell lysates were prepared, and protein concentration was measured (DC protein assay; Bio-Rad) as described above. Each sample (400 g) was immunoprecipitated with anti-JNK (BD-Pharmingen) or anti-p38␣ antibodies (Santa Cruz Biotechnology). Kinase activity was assayed using, in anti-JNK precipitates, 4 g of glutathione S-transferase-c-Jun (1-79). To measure the activation of p38 an antibody anti-phospho-p38 (1:1000, Cell Signaling) was used. Total kinase levels in the different clones were assessed by Western blot analysis using anti-JNK (BD-Pharmingen) or anti-p38 (Santa Cruz Biotechnology). As a positive control, the clones were treated with a calcium ionophore (1 M ionomycin; Calbiochem) and a phorbol ester (100 ngr/ml 12-O-tetradecanoylphorbol-13-acetate; Calbiochem) for p38 activation or 1 M NaCl for JNK stimulation 30 min prior to cell lysis.
Analysis of the Effects of DNA Damage-To measure the response to genotoxic agents of the p53 pathway in each clone, 4G4 transfectants were grown to 10 6 cells/ml and then treated with 1 g/ml doxorubicin (Sigma) for several periods of time. A peak in p53 phosphorylation was found to lie at 6 h of treatment fading subsequently until undetectable after 16 h. After a 6-h treatment, the cells were lysed as described above, and each sample was immunoprecipitated with 1 g of anti-p53 antibody (Ab-1; Calbiochem) and subjected to Western blot analysis using anti-Ser(P) 15 p53 (1:1000; New England Biolabs) with appropriate secondary antibodies (1:5000; Cappel). Total p53, p21, and Bax levels in the different clones were quantified by Western blot analysis of 60 g of total cell lysate using anti-p53 (Ab-3; Calbiochem), anti-p21 WAF (SX118; Pharmingen), and anti-Bax (Santa Cruz Biotechnology). To quantify the content of Bcl-2 protein in the panel of clones, Western blots were performed using two different anti-Bcl-2 antibodies (Santa Cruz Biotechnology number sc-7382 and Pharmingen number 15021A). For apoptosis studies, the cells were treated with 1 g/ml doxorubicin (Sigma) for 72 h previous to staining with propidium iodide followed by FACS analysis. The percentage of dying cells was determined by electronic gating of the sub-G 1 population.

RESULTS
Characterization of Pre-TCR Transfectants-The T cell line 4G4 (11) was used to stably express several pre-TCR constructs. Different clones of transfectants were established by drug selection and expression of TCR␤ alone; TCR␤ plus either the large (pT␣ a ) or short (pT␣ b ) pT␣ isoforms; and TCR␤ plus a mutant form of pT␣ b (termed ⌬P1P2), in which two CD2-like proline motifs in the cytoplasmic tail were deleted was confirmed by Western blot (Fig. 1) and flow cytometry (not shown). Although the levels of expression varied slightly among different transfectant clones receiving the same cDNAs, the different gene products were readily detectable in every case ( Fig. 1 and data not shown). Thus, mutations of the pT␣ tail did not significantly affect expression or stability of pT␣ or TCR␤ in these cells.
Functional Regulation of Fas and Apoptosis by Pre-TCR Expression-With the aim to determine whether the pre-TCR expression could protect from apoptotic cell death, we developed a system where apoptosis was induced in the various transfectant clones upon irradiation with UV light. Interestingly, those transfectants receiving a complete set of pre-TCR components appeared more resistant to UV-induced apoptosis than those receiving either TCR␤ alone or cells expressing TCR␤ plus the pT␣ tail mutant ⌬P1P2 pT␣ b (Fig. 2). For example, for TCR␤-only transfectants, this procedure induced death in Ͼ80% of cells, whereas in several TCR␤ϩpT␣ transfectants, only ϳ20% of cells underwent apoptosis after UV treatment.
Susceptibility to this UV-induced cell death has been attributed to ligand-independent activation of Fas in T cells (31)(32)(33), and it can also be mediated by p53. Consistent with this data, those clones expressing a complete pre-TCR displayed partially reduced surface Fas expression (Fig. 3) as compared with parental cells or cells expressing the TCR␤ chain alone. Significantly, pre-TCR-induced changes in Fas levels were not measurably reduced in clones expressing the cytoplasmic tail mutant of pT␣ (Fig. 3), demonstrating a contribution of the pT␣ tail both to the regulation of surface Fas expression and to protection from UV-induced apoptosis.

Effects of Pre-TCR Expression on DNA Damage-mediated
Events-The presence of double-stranded DNA breaks caused by V(D)J recombination has been considered the source of the apoptotic response that leads to depletion of entire populations of thymocytes (24). This critical event occurs early in the ␤-selection checkpoint and can be reversed by expression of a TCR␤ chain paired with the pT␣ gene product. However, exactly how pre-TCR expression promotes protection from DNA damageinduced apoptosis is still an unresolved question. In an attempt to answer this question, we performed a set of experiments using chemotherapeutic agents as opposed to UV irradiation to avoid any Fas-mediated effects. Pre-TCR-expressing clones were treated with doxorubicin, a drug known to cause severe DNA damage through inhibition of topoisomerase II (34). Upon doxorubicin treatment, transfectants expressing a full set of pre-TCR components displayed high levels of protection relative to those receiving either TCR␤ alone or cells receiving TCR␤ plus the pT␣ tail mutant (Fig. 4). For example, for TCR␤-only transfectants the death rate was between 80 and 95% of cells, whereas less than 30% of the TCR␤ϩpT␣ transfectants were dead. These data demonstrated that expression of pre-TCR is sufficient per se to effectively protect cells from DNA damage-induced apoptosis.
It has been suggested that the pre-TCR regulates progression through the DNA damage checkpoint characteristic of the DN to DP transition by inactivating p53 (26). With the aim to biochemically assess whether pre-TCR could somehow regulate p53 levels or activity, we set out to determine whether the total levels of p53 and two well established p53-regulated genes, namely p21 WAF (35) and Bax (36), were altered in representative clones expressing TCR␤ alone or TCR␤ plus any of the pT␣ isoforms and TCR␤ plus ⌬P1P2 pT␣ b . As shown in Fig. 5A, the total content of p21 WAF or Bax protein was markedly reduced in those clones expressing either isoform of the pT␣. In sharp contrast, the total amount of a pivotal pro-survival gene product, Bcl-2, was increased in those particular clones (Fig. 5C). Essentially the same result was obtained using a second anti-Bcl-2 antibody (15021A, BD-Pharmingen). The levels of p53 protein were very similar in all clones tested, indicating that possible changes in the amount of total p53 do not seem to be the cause of the differences observed in p21 WAF and Bax expression. Of note, clones expressing the ⌬P1P2 mutant of the pT␣ b plus TCR␤ did not display any detectable changes in p21 WAF , Bcl-2, or Bax when compared with TCR␤-expressing control cells. These results pointed at the possibility that the p53 transcriptional activity, and not total p53 levels, was somehow decreased in pT␣-expressing cells.
To further challenge this hypothesis, we needed to test the functionality of the p53 protein itself. With that purpose, we took advantage of the availability of sensitive anti-phospho-p53 antibodies that allowed us to test the activation status of the p53 protein in situ. In particular, phosphorylation of Ser 15 in The migration of each protein is indicated by an arrow. Essentially the same result was obtained in three independent experiments. B, expression of Bcl2 protein in a panel of clones as analyzed by Western Blot with specific anti-Bcl2 antibodies. A band was detected at ϳ26 kDa corresponding to Bcl2 and is indicated by an arrow. The figure is representative of three different experiments. C, selected transfectants were treated (ϩ) or not (Ϫ) with doxorubicin for 6 h, and the amount of total protein required to normalize for the same total levels of p53 was subjected to immunoprecipitation with an anti-p53 antibody (Ab-1). After electrophoresis, the proteins were transferred to a polyvinylidene fluoride membrane and blotted using an anti-phospho-p53 (Ser 15 ). To verify equal loading of p53, the same membrane was stripped and subsequently reblotted using an anti-p53 antibody as described under "Experimental Procedures." The migration of the p53 protein is indicated by an arrow. The autoradiograph shown is representative of three independent experiments. p53 has been demonstrated to increase the ability to transactivate p21 WAF upon DNA damage (37) and represents a prototypical marker for p53 activation upon DNA damage. In cells treated with doxorubicin, an inhibitor of topoisomerase II, we could observe a clear phosphorylation of endogenous p53 in cells expressing TCR␤ that were used as a control when the same total levels of p53 were loaded into the gel (Fig. 5C). However, we detected differences in the activation status of p53 in response to doxorubicin in the different 4G4-derived clones. The amount of phosphorylated p53 was reduced in cells expressing pT␣ b plus TCR␤ as compared with cells expressing only TCR␤, although this difference was even more evident in cells expressing pT␣ a plus TCR␤, where very low levels of phosphorylated p53 could be detected (Fig. 5C). We also noticed that transfectants expressing the ⌬P1P2 pT␣ b mutant showed normal levels of p53 phosphorylation. This observation highlights again the importance of the cytoplasmic domain of the pT␣ for certain biochemical effects of the pre-TCR and further corroborates that expression of a functional pre-TCR protein leads to defective phosphorylation of p53 upon DNA damage.
The Pre-TCR Controls the Activation of Stress-regulated Kinases-Several studies have established a very important role for mitogen-and stress-activated kinases in the regulation of thymocyte development and survival (38). To gain a mechanistic understanding on how the pre-TCR might contribute to the regulation of this type of pathway, 4G4 transfectants were examined for changes in the status of two such protein kinases, JNK and p38␣. Pre-TCR transfectants showed increased basal kinase activity of JNK. Most notable was the absence of any JNK activity in transfectants expressing the tail mutant of pT␣ plus TCR␤ (Fig. 6); indeed, JNK activities in these transfectants were lower than those found in cells receiving TCR␤ alone. By contrast, absolute levels of JNK protein were essentially the same in all clones (Fig. 6B). We next tested the ability of the pre-TCR to modulate a mitogen-activated kinase that has also been implicated in the regulation of thymocyte development (39, 40) namely p38. The basal activity of p38 appears to be elevated in the pT␣ b -expressing cells when compared with TCR␤ or pT␣ a transfectants that showed very low p38 activity (Fig. 6A). Also in this case, deletion of intracytoplasmic do-mains abolished the transduction of signals from the receptor to the p38 kinase.

DISCUSSION
The ␤ selection checkpoint is a critical step in lymphocyte development defined not by a single biochemical event but by a plethora of varied cellular changes affecting apoptosis regulators, cell cycle effectors, and other signaling pathways (reviewed in Ref. 41). These include, but may not be restricted to, the down-regulation of Fas (22), the activation of NF-B (23), the regulation of p38 (40), and the inhibition of p53 (26). All of these changes have been shown to occur during the time interval surrounding pre-TCR expression; however, no direct cause and effect relationship has been demonstrated that directly links the pre-TCR molecule with these intracellular events. To try to clarify whether these changes are the direct consequence of the expression of the pre-TCR molecule or simply simultaneous in time, we have developed and characterized T cell clones stably expressing different components of the pre-TCR complex. This system allowed us to explore in detail the signaling cascades that this molecule is able to launch and to measure defined biological consequences. Some of these transfectants have been already utilized to show that NF-B activation in stage III thymocytes is dependent on expression of pre-TCR (23). The findings reported here that pre-TCR can be functionally assayed in an heterologous system create the potential for delineating the pathway from pre-TCR to downstream effectors and for exploring the implication of defined pT␣ domains in such effects.
In particular, the role of the cytoplasmic tail of the pT␣, unique among TCR chains, has been proposed in the past (41). However, complementation studies in pT␣-deficient mice have so far produced ambiguous results (Refs. 13 and 17 and references therein) that have lead to some controversy in the field. Only recently, a study by Aifantis et al. (17) by expression of different pT␣ mutants in retrovirally transduced T cell precursors and cell lines showed that the pT␣ cytoplasmic tail, in particular the proline-rich domain, plays a crucial role in pre-TCR signaling and T cell development. In our report we progress one step further and elucidate the molecular basis for this FIG. 6. Regulation of JNK and p38 pathways by expression of pre-TCR components. A complete panel of stable transfectants expressing pre-TCR components was analyzed for total levels and activity of p38␣ (A) and JNK (B). Total kinase protein levels of JNK and p38 were determined by Western blot on 60 g of total cell lysates as described under "Experimental Procedures." The JNK activity was determined by incorporation of phosphate into recombinant glutathione S-transferase-c-Jun (1-79) (GST-cjun), a specific JNK substrate, and p38 activation by using an anti-phospho-p38 antibody. Note the differential activation of these kinases by expression of the various pre-T␣ isoforms. As a positive control, the clones were treated with a calcium ionophore (ionomycin) together with a phorbol ester (Ca 2ϩ ϩ 12-O-tetradecanoylphorbol-13-acetate (TPA)) for p38 activation or subjected to osmotic stress (1 M NaCl) for JNK stimulation for 30 min prior to cell lysis. W Blot, Western blot. biological effect by demonstrating an essential role of the pT␣ tail in pre-TCR-mediated survival and signaling. We also establish here that the pT␣ tail is indeed required for at least some of the biological functions elicited by the pre-TCR.
Also using the cellular system described here, we were able to challenge the influence that pre-TCR expression has in the expression of Fas protein. In support of some involvement of Fas in DN thymocyte development, thymocytes from severe combined immunodeficiency mice, which cannot ordinarily survive ␤-selection, show improved DN to DP maturation when the severe combined immunodeficiency mutation is bred onto a Fas deficiency (42). In this regard, it is worth mentioning that the analysis of thymocyte subsets showed that the decrease in Fas expression observed in DN cells was attributable primarily to subsets I and IV; Fas levels were high in DN II and DN III but showed a specific decline on DN IV cells (22), just after the pre-TCR checkpoint. These results argue that a selective decline in Fas levels during ␤-selection would be necessary for a correct DN to DP transition and raise the possibility that the pre-TCR may somehow regulate this phenomenon. Using the heterologous system described here, we were able to demonstrate that there is a selective drop in the amount of Fas upon expression of any of the two full-length pT␣ proteins together with a TCR␤ chain. Furthermore, there seems to be a correlation between the levels of Fas expressed in different pre-TCR clones and the susceptibility of these clones to apoptosis induced by engagement of Fas. Our observations are in full agreement with the in vivo results mentioned above and demonstrate that there is a direct correlation between expression of the pre-TCR and the levels of Fas and further suggest that the presence of pre-TCR is sufficient to achieve a down-regulation of the Fas molecule and protect cells from Fas-mediated apoptosis.
Apoptosis induced by DNA damage is p53-dependent in thymocytes (24). In vivo, severe combined immunodeficiency ϫ p53-deficient mice are permissive for the generation of DP thymocytes, and there seems to be a correlation between inactivation of the p53 gene and progression to the DP stage (43,44). The N terminus of the p53 protein has been implicated in recruitment of important p53 regulators such as Mdm2 and corepressors (45). Indeed, phosphorylation of Ser 15 causes dissociation of p53 from Mdm2 (46) and allows for the association of coactivators of its transcriptional activity (Ref. 45 and references therein). We show here that N-terminal phosphoryation of p53 in response to DNA damage is clearly defective in pT␣/ TCR␤-expressing cells but not in cells expressing only TCR␤ or a tail-less form of pT␣ b and TCR␤. We may argue that, when either one isoform of the pT␣ is present, p53 is being constitutively inactivated, even in the presence of double-stranded DNA breaks that in vivo are caused by V(D)J recombination (24). If this were the case, targets of p53 transcriptional activation such as p21 WAF and Bax proteins would be reduced in pT␣-transfectants. In fact, as described in the present study, expression of pT␣ seems per se to be able to down-regulate the amount of p21 WAF and Bax in our system. At the same time it up-regulates the amount of other apoptosis regulators such as the pro-survival protein Bcl-2.
As reported, expression of Bcl-2 protein was low-to-moderate in double negative cells and declined to negligible levels in DP cells, prior to re-expression in positively selected single positive cells (47)(48)(49)(50). The decline in Bcl-2 expression as DN cells matured to DP cells was evident in DN stage IV cells, which expressed significantly lower levels than the preceding DN III cells. We believe the possibility that Bcl-2 is the sole protein for rescue from different types of apoptotic stimuli is somewhat unlikely based on in vivo data, particularly for pre-TCR-medi-ated rescue. On the one hand Bcl-2 overexpression is not sufficient to prevent cell death in pre-TCR-deficient T cells (26,28). On the other hand, it is very likely that p53 induces death via a Bcl-2-insensitive pathway, as pointed out by Green and Schuler (51). First, p53 can trigger cell death through Fas expression (52). Second, Fas-deficient ϫ severe combined immunodeficiency thymocytes can develop normally (42). Third, Bcl-2 expression fails to block Fas-mediated apoptosis in the thymus (53,54). Also, the down-regulation of Fas indicates that pre-TCR selection inhibits apoptosis by both Bcl-2-regulated and Bcl-2-independent pathways. Given that ␤-selection appears to involve multiple regulators of apoptosis, it would seem unlikely that complete rescue of pre-TCR deficiency could be achieved by regulation of a single signaling cascade.
Proteins belonging to the stress-activated protein kinase and mitogen-activated protein kinase families have long been known to play a pivotal role in the development and differentiation of thymocytes (38). In particular, the consequence of CD2-driven JNK activation in T cells is inhibition of Fas expression and rescue of T cells from apoptosis (55), similar to that shown in this study for the pre-TCR. Interestingly, although both isoforms of pT␣ trigger the activation of cascades leading to the phosphorylation of c-Jun, only the short form, pT␣ b , is able transduce signals that activate p38, a member of the mitogen-activated protein kinase family of proteins essential for the transition from DN to DP cells. Is important to mention here that expression of this short pT␣ is enough in our system to trigger p38 activation, even in the absence of any exogenous stimulation. Indeed, a p38 activity is induced in thymocytes without any apparent requirement of extracellular factors (56). The data presented here that presence of a functional pre-TCR is enough to drive p38 activation may provide a valuable explanation for the high p38 activity detected intrathymically. Although p38 is strictly required for differentiation of immature thymocytes (39), constitutive activation of the p38 pathway leads to excessive thymocyte proliferation and impaired thymic development (40). In line with these data, our results show that the presence of the long pT␣ isoform is enough to down-regulate p38 activity to levels below those of control and TRC␤-expressing cells. These results allow us to argue that the exquisite regulation of p38 activity that seems to be required for normal T cell development in vivo might be achieved by the selective expression of one specific pT␣ isoform, a hypothesis that will need to be challenged in vivo.
Based on the results obtained in this study, we propose that pre-TCR may affect cell survival via effects on Fas and p53 in addition to the reported regulation of NF-B and cell differentiation through the control of mitogen-regulated kinases, all of which are regulated across the ␤-selection point. This proposal accommodates the possibility that late DN thymocytes are vulnerable both to death by DNA damage caused by V(D)J recombination events and to death induced via Fas-type receptors. In fact, thymocytes from mice lacking functional Fas ligand (gld) show normal sensitivity to apoptosis transduced by p53 (15), thus demonstrating that Fas-and p53-mediated apoptosis are independent processes in thymocytes. In p53-deficient mice, the survival defects caused by pre-TCR absence were remarkably restored (26); thus p53 appears to be a very distal component of the pre-TCR pathway. This pleiotropy of signals elicited by expression of a single receptor complex may serve in vivo to block the various apoptotic stimuli to which thymocytes are subjected during the ␤-selection checkpoint, thus ensuring the viability and growth of key cellular populations and helping guarantee a successful thymic development.