Diacylglycerol kinase α activity promotes survival of CD4⁺ 8⁺ double positive cells during thymocyte development

Abstract

The diacylglycerol kinases (DGK) form a family of isoenzymes that catalyse the conversion of diacylglycerol (DAG) to phosphatidic acid (PA), both powerful second messengers in the cell. DGKα is expressed in brain, peripheral T cells and thymocytes and has been shown to translocate to the nuclear matrix upon T-cell receptor (TCR) engagement. Here, we show that high level expression of DGKα is induced following a signal transmitted through the pre-TCR and the protein tyrosine kinase, lck. Activity of DGKα contributes to survival in CD4⁺ 8⁺ (DP) thymocytes as pharmacological inhibition of DGK activity results in death of this cell population both in cell suspension and thymic explants. DGKα promotes survival in these thymocytes through a Bcl-regulated pathway. A consequence of inhibition of DGKα is the specific down-regulation of Bcl-xl, whereas in transgenic mice that over-express Bcl-2, death induced by the inhibitor is partially blocked. Thus we report a novel activity of DGKα in survival of thymocytes immediately after entry into the DP stage in development.

Introduction

During T cell development in the thymus double negative (DN) thymocytes differentiate into double positive (DP) cells, which then become mature CD4⁺ single positive (SP) or mature CD8⁺ SP cells. Progression from DN to DP depends upon successful rearrangement of a T-cell receptor (TCR) β chain gene (TCRβ) and expression of the pre-TCR complex, whereas the transition from DP to SP cell depends on successful rearrangement of a TCRα chain gene and expression of a mature αβ TCR.¹ DP cells make up at least 80% of thymocytes and provide a large pool of cells that will undergo positive and negative selection of the TCR repertoire.²

Transition from DN to DP cell involves delivery of a survival signal, proliferation, differentiation and allelic exclusion of the TCRβ gene locus and is dependent upon expression of the pre-TCR. Thymocyte development in mice unable either to express the pre-TCR or to signal through the pre-TCR, is completely or partially blocked at this DN stage.³ The downstream events mediated by the pre-TCR are dependent upon expression of the src-family tyrosine kinases, lck and fyn.⁴ DP thymocyte development is severely compromised and SP development is arrested when lck expression or activity is diminished during thymopoiesis.⁵

Recently another family of kinases, the diacylglycerol kinases (DGK) have been shown to be expressed in T cells.^6,7 The DGK regulate lipid-induced signal transduction in cells by the conversion of diacylglycerol (DAG) to phosphatidic acid (PA), both powerful second messengers (for review see 8). DAG is best known as an activator of the protein kinase C (PKC) pathway, but is also implicated in a number of other important signalling pathways such as activation of vav⁹ and rat sarcoma–guanyl releasing protein (RAS-GRP)¹⁰ as well as controlling the synthesis of major phospholipids and triacylglycerols.¹¹ PA, the end product of this catalytic reaction, can stimulate DNA synthesis^12,13 modulate the activity of Raf-1¹⁴, p21 activated protein kinase-1 (PAK 1)¹⁵ PKC and Ras-GTPase activating protein (Ras-GAP)¹⁶ and mediate interleukin-2 (IL-2)-induced proliferation of cytotoxic T lymphocyte (CTLL-2) cells.¹⁷ PA has also been shown to have an antagonistic effect on ceramide induction of poly (ADP-ribose) polymerase (PARP) proteolysis and preventing ceramide induction of apoptosis.¹⁸ Inhibition of DGKα activity has been studied using the chemical inhibitor, R59949. This pharmacological reagent was initially described as an inhibitor of DAG phosphorylation in intact cells¹⁹ and was subsequently used to investigate DAG-mediated signalling events in a wide range of tissues.^17,20,21 Specificity of this inhibitor for particular DGK isoforms was demonstrated recently in a study that showed that R59949 binds directly to the catalytic domain of the calcium dependent DGK isoform, DGKα. The one other DGK isoform isolated from thymus to date, DGKζ²² is reported to be calcium independent and was not inhibited by R59949.²³

The DGK form a family of isoenzymes, each with its own tissue specificity and distinctive structure.²⁴ The α isoform is expressed in brain, peripheral T cells and thymocytes, and translocates from the cytosol to the nuclear matrix on TCR engagement in rat thymocytes.⁷ Here, we show that high level expression of the DGKα gene occurs subsequent to pre-TCR- and lck-mediated signalling. Inhibition of the activity of this gene by a chemical inhibitor, R59949, resulted in death of the DP thymocyte population. Within 4·5 hr of inhibition of DGKα activity by R59949, expression of the anti-apoptotic gene, bcl-xl was down-regulated. In transgenic mice over-expressing the closely related molecule Bcl-2, in DP thymocytes, R59949-induced death is partially blocked. Thus we report a novel activity of DGKα in maintenance of cell survival in thymocytes immediately after entry into the DP stage in development.

Materials and methods

All results shown are representative of a minimum of three experiments.

Reverse transcriptase–polymerase chain reaction (RT–PCR)

RNA and cDNA were isolated and synthesized from thymocytes as previously described.²⁵ PCR reactions were carried out on a Stratagene Robocycler gradient 96 PCR machine using 35 cycles, 96°/1 min, 58°/1 min, 72°/1 min.

HPRT forward: GCTGGTGAAAAGGACCTCT,HPRT reverse: CACAGGACTAGAACACCTGC.DGKα forward: GTCTACTGCTACTTCACCCTCC,DGKα reverse: GAT GTT CAA TAC CGC AAT GCC.Bcl-xl forward: TCGCTCGCCCACATCCCAGCTTCACATAACCCC,Bcl-xl reverse: CCACCAACAAGACAGGCT.

All reactions were resolved on a 1% agarose gel.

Flow cytometry and antibodies

Azide-free anti-CD3 monoclonal antibody (mAb) was obtained from Pharmingen (San Diego, CA). Thymocytes were stained with combinations of anti-CD4–phycoerythrin (PE), anti-CD8α–PE, annexin-V–fluoroscein isothiocyanate (FITC; Pharmingen) and anti-CD4-Tricolor (TC), anti-CD8α-TC (Caltag, UK). Staining was carried out as previously described.²⁶ Fluorescence-activated cell sorting (FACS) analysis was performed using a Facscan (Becton Dickinson, San Jose, CA). Thymic subsets were prepared by FACS on an ALTRA (Coulter). Annexin-V staining was carried out using an annexin-V–FITC apoptosis detection kit (Pharmingen). Cells were stained with propidium iodide by incubation in 500 µl 0·1% Triton-X and 100 µl of 50 µg/ml propidium iodide (PI; Pharmingen) and immediately FACS analysed.

Mice

BALB/c mice were maintained in the CNB animal house, Madrid or purchased from B & K Universal Ltd (Grimston, UK) and maintained in the Central Biomedical Services Unit at Imperial College, London. Rag 2^–/–, lck^–/–, TCR β^–/–. TCR α^–/– mice were purchased from Jackson Laboratories (Bar Harbor, ME) and were maintained in individually ventilated cages. Timed matings were performed as described.²⁶ Adult mice were 6–10 weeks of age.

Immunoprecipitation and DGK assay

DGKα was isolated from mouse thymocytes (approx. 10⁶ cells) using a polyclonal anti-DGKα antibody prepared against the glutathione-S-transferase (GST)-murine DGKα catalytic domain (M. A. Sanjuan and I. Merida, manuscript in preparation). DGKα activity was assessed by incubation of the immunoprecipitated enzyme with [γ³²P]ATP and DAG and treated as previously described (Flores et al. 1996).⁶ When indicated the R59949 DGK inhibitor (Calbiochem, La Jolla, CA; 266788) was added to the reaction 10 min before the DAG at a final concentration of 1 µm.

Fetal thymus organ culture (FTOC) and cell culture

Fetal thymuses were dissected and cultured on Millipore filters (8 µm pore size) in Dulbecco's modified Eagle's minimal essential medium (DMEM) supplemented with 10% fetal calf serum (FCS). Cultures were incubated at 37° in 5% CO_2. R59949 from Calbiochem was prepared in dimethylsulphoxide (DMSO) and added directly to the medium. Thymocytes in suspension were incubated at 5×10⁶/ml in the same media as above.

Results

Expression of DGKα during thymocyte development

We assessed expression of the DGKα gene by performing RT–PCR on RNA prepared from sorted CD25⁺ DN, DP, CD4SP and CD8SP thymocyte populations. High level expression of DGKα RNA was detected in DP cells and in SP cells; however, DGKα RNA expression was below detection levels in the CD25⁺ DN population (Fig. 1a). RT–PCR analysis of DGKα expression in thymocytes isolated from pairs of mice deficient in TCRα, TCRβ, Rag 2, and lck, revealed that DGKα is expressed at normal levels in TCRα^–/– mice, but was below detection levels in Rag 2^–/– mice and was barely detectable in lck^–/– mice and TCRβ^–/– mice (Fig. 1b). The low level expression of DGKα in lck^–/– mice and TCRβ^–/– mice was only detectable by Southern blot hybridization (data not shown).

Figure 1.

DGKα RNA Expression is up-regulated following pre-TCR signalling (a) CD25⁺ DN, DP, CD4SP and CD8SP thymocyte populations were sorted by flow cytometry. RNA was isolated and used as a template for RT–PCR. In each experiment levels of expression of HPRT and DGKα were analysed in the same tube, each RNA species is indicated by an arrow. (b) Thymocytes were isolated from a panel of knockout mice and used as a template for RT–PCR. Two mice were analysed for each gene knockout. The quantity of template used was normalized according to levels of expression of HPRT. Products were visualized by ethidium bromide staining of an agarose gel. (c) Rag 2^–/– mice were injected intraperitoneally with anti-CD3 mAb. After 10 days thymocytes were isolated and analysed for expression of CD4 and CD8D. Expression of DGKα was analysed by RT–PCR in thymocytes isolated from WT (a) anti-CD3 treated Rag 2^–/– (b and c) and Rag 2^–/– control (d). Amount of cDNA used was normalized according to HPRT expression. (e) Expression of DGKα was analysed by RT–PCR in thymocytes isolated from Rag 2^–/– adult thymic explants treated with or without anti-CD3 for 24 hr. In this experiment HPRT and DGKα expression were analysed simultaneously. Each band is indicated. (f) Cell surface expression of CD25 and CD8 was assessed by FACS analysis on control Rag^–/– thymic explants and Rag^–/– explants treated with anti-CD3 for 24 hr.

These results suggest that high-level DGKα gene expression occurs following pre-TCR signalling. To test this we induced differentiation to DP thymocyte in Rag 2^–/– mice by injection of anti-CD3 mAb allowing us to assess the consequences of a signal delivered through the pre-TCR.²⁷ Thymocytes were isolated 10 days after injection, stained with antibodies against CD4 and CD8 and analysed for expression of DGKα RNA (Fig. 1c, d). Thymocyte development had progressed from the DN to the DP stage following injection of anti-CD3 (Fig. 1c). Associated with this thymic differentiation was an up-regulation in expression of the DGKα gene observed by RT–PCR (Fig. 1d). To show that expression of DGKα is an early consequence of signalling through the pre-TCR we performed RT–PCR on thymus explants obtained from Rag^–/– mice treated for 24 hr with anti-CD3 mAb (Fig. 1e). Expression of DGKα was below levels of detection in untreated Rag^–/– thymus explants as observed previously (Fig. 1b). However, after only 24 hr in culture expression of DGKα was detectable. Analysis of cell surface CD25 and CD8 expression (Fig. 1f) demonstrated that DGKα expression was induced upon pre-TCR signalling and before cells had differentiated to the DP stage.

DGKα activity is immunoprecipitated from thymocytes and inhibited by R59949

To confirm that the DGKα protein is present and active in thymocytes we immunoprecipitated DGKα from thymocyte lysates using a specific polyclonal antibody. DGK activity in the immunoprecipitate was then determined by incubation of the immunoprecipitate with the DGK substrate, DAG, together with [γ³²P]ATP. Generation of [γ³²P]PA was assessed by thin layer chromatography and autoradiography. DGKα activity was clearly detectable in thymocyte lysates (Fig. 2a). Complete inhibition of the activity of DGKα was achieved by addition of the chemical inhibitor, R59949¹⁹ 10 min prior to addition of DAG. Thus we demonstrate DGKα activity in thymocytes and that this activity is inhibited by R59949.

Figure 2.

DGKα activity is inhibited by R59949 and R59949 induces death in DP thymic suspension cultures. (a) DGKα was immunoprecipitated from thymocyte lysates using a polyclonal antibody raised against murine DGKα. Enzyme activity was determined with and without addition of R59949. An autoradiogram of the thin layer chormatography separation of the [³²P]PA formed in the phosphorylation assay is shown. Migration of the PA standard is shown. (b) Thymocytes were cultured in suspension with 5 µm R59949 or DMSO in serum-free media for 2, 4 and 6 hr. The reaction was terminated by addition of 10% FCS and cultures maintained for a total of 16 hr. Cells were stained with CD4–PE, CD8–TC and annexin-V–FITC. Mean annexin-V staining for three independent experiments is shown for DP cells treated with R59949 or DMSO. (c) The degree of specific R59949-induced apoptosis was calculated by subtracting the percentage of annexin-V staining following DMSO treatment from the percentage of R59949 annexin-V staining after R59949 treatment for each time point. Values were calculated for DN, DP, CD4SP and CD8SP subsets and the mean level of annexin-V–FITC staining for three experiments is shown.

Survival of DP thymocytes is impaired by inhibiting DGKα activity

To study the consequence of inhibition of DGKα activity in thymocytes, cell suspensions were treated in vitro with 5 µm R59949 or DMSO for 2, 4 and 6 hr. As R59949 is inactivated by FCS, incubations were carried out in the absence of serum, and each reaction was terminated by the addition of 10% FCS to the culture. All thymocytes were maintained in culture for a total of 18 hr. To determine the degree of apoptosis of each thymic subset, cells were stained for expression of CD4, CD8 and the apoptotic marker, annexin-V. Annexin-V staining in the DP revealed an increase in apoptosis in the presence of R59949 (Fig. 2b). DMSO did not increase apoptosis. Degree of apoptosis specifically induced by R59949 in each subset was calculated by subtracting the percentage of annexin-V positive cells in control cultures from that in the R59949 treated cultures (Fig. 2c). R59949 inhibited survival in the DP cells, but not in SP cells and to a lesser extent in DN thymocytes. The lack of apoptosis in the SP subsets could reflect the ability of these cells to use additional survival pathways, perhaps those induced by cytokines. In support of this idea R59949 has been shown to block IL-2-induced cell cycle progression in the mature mouse T-cell line, CTLL-2 with no effect on cell viability.¹⁷ The slight increase in R59949-induced death in the DN subset may reflect a low level of DGKα expression in the more mature CD25⁻ DN thymocyte subset. Thus, incubation with R59949 causes a loss in survival of DP thymocytes in cell suspension.

lck^–/– thymocytes are less susceptible to R59949-induced death

RT–PCR analysis showed that lck^–/– mice express very little DGKα in the thymus (Fig. 1b). Although lck is an important mediator of pre-TCR signalling the presence of DP thymocytes in lck^–/– mice suggests that lck is not essential for the DN to DP transition. About 60% of thymocytes in lck^–/– mice are DP cells (Fig. 3a and ²⁸). Thus we were able to test the effect of R59949 on DP thymocytes isolated from lck^–/– mice supporting specificity of the inhibitor. Thymocyte suspensions from wild type (WT) and lck^–/– mice were treated in vitro with R59949 (Fig. 3b). R59949 induced approximately 2·5 times more cell death in WT DP cells than in the lck^–/– DP cells. Some increase in death was observed in the lck^–/– thymocytes, and this may reflect the very low level of DGKα in these cells. We reasoned that if DGK is required to maintain survival in DP thymocytes, then the reduced level of DGKα observed in lck^–/– mice should be reflected by a reduced ability to survive. DP thymocytes from a WT and lck^–/– mouse were sorted on the basis of CD4 and CD8 expression and cultured in vitro in the presence of 10% FCS (Fig. 3c) for 8 and 24 hr. Following this time period, cells were stained with PI and FACS analysed. Some degree of spontaneous cell death occurred in the double positive thymocyte population isolated from WT mice. The percentage of cells falling in the live gate decreased from 98 to 44%. However DP cells isolated from lck^–/– mice died more quickly, with only 31% of cells falling in the live gate after 24 hr and a clear apoptotic peak after only 8 hr. These results support the specificity of R59949 in inhibiting DGKα activity and show that thymocytes isolated from a mouse deficient for lck expression survive less well in suspension culture.

Figure 3.

lck^–/– thymocytes are less susceptible to R59949-induced death and survive less well in suspension culture. (a) Thymocytes were isolated from a WT and lck^–/– mouse and analysed for cell surface expression of CD4 and CD8 using flow cytometry. (b) Thymocytes from WT and lck^–/– mice were cultured in the presence of 5 µm R59949 or the equivalent DMSO for 2, 4, 6 and 9 hr. The reaction was terminated by addition of 10% FCS and cultures maintained for 16 hr. Apoptosis was determined by staining for expression of CD4, CD8 and annexin-V. Percentage increase apoptosis over background was calculated as in (c) Thymocytes isolated from an adult lck^–/– and WT mouse control mouse were cultured in the presence of 10% FCS for 8 and 24 hr. Following which time period apoptosis was determined by PI staining. The figure shown for each time-point represents the percentage of cells that are still viable.

Inhibition of DGKα blocks survival of DP thymocytes in FTOC

As thymocyte development has been shown to depend on signals provided by the thymus epithelium²⁹ we tested the effect of R59949 in FTOC. E14.5 FTOC were cultured in the presence of 10% FCS and 10 µm R59949 or DMSO for 6 days. In the FTOC experiments, R59949 was used at a higher concentration because of the fact that it is difficult to culture FTOC without serum. The cultures were transferred to fresh R59949 every 24 hr. At the end of this period, thymocytes were stained for CD4, CD8 and annexin-V expression (Fig. 4a). Annexin-V staining revealed that R59949 specifically induced death in the DP population. DMSO treatment caused some apoptosis of DP cells, leading to 17·4% annexin-V staining, whereas R59949 resulted in 43% annexin-V staining in the DP cells. This R59949-induced apoptosis of DP cells was reflected by a reduction in the percentage of DP from 75% after treatment with DMSO to 67% after R59949 treatment. Analysis of cell number showed that the reduction in the percentage of DP cells in the R59949 cultures was accompanied by a 60% reduction in cell number related to DMSO control. These results confirm that DGKα activity contributes to the survival of thymocytes at the DP stage in development.

Figure 4.

R59949 induces apoptosis in DP thymocytes in FTOC but does not block transition from DN to DP stage in thymic development. (a) E14.5 FTOC were cultured in the presence of 10% FCS and 10 µm R59949 or the equivalent DMSO for a period of 6 days. FTOC were transferred to fresh R59949/DMSO every 24 hr. At the end of this period, thymocytes were isolated and stained for CD4, CD8 and annexin-V expression. CD4–PE and CD8–TC staining is shown in the upper panel, and annexin-V–FITC of the DP population is shown below. (b) E17.5 Rag 2^–/– FTOC were cultured with 10% FCS and 10 µg/ml anti-CD3 mAb for 48 hr. FTOC were then transferred to the same medium supplemented with 10 µm R59949 or the equivalent DMSO and analysed 72 hr later. Thymocyte suspensions were stained with CD4–PE and annexin–FITC. CD4–PE staining is shown with and without anti-CD3, following treatment with either R59949 or DMSO. Annexin-V staining is shown for the CD4–PE-positive population.

Having established a role for DGKα in survival signals mediated through the pre-TCR we determined if this enzyme could also have an effect on differentiation as the two pathways may be linked.³⁰ We tested the ability of R59949 to block thymic differentiation induced by anti-CD3 mAb in Rag 2^–/– FTOC. Day 17·5 Rag 2^–/– FTOC were cultured in the presence of 10 µg/ml anti-CD3 mAb for 48 hr. FTOC were then transferred to the same media supplemented with 10 µm R59949 and analysed 72 hr later. Thymocytes were then stained with CD4-PE and annexin-V–FITC (Fig. 4b). Progression from DN to DP stage in development was induced in Rag 2^–/– FTOC in the presence of the anti-CD3 mAb. CD4 expression increased from 2·2 to 68% following addition of anti-CD3 (Fig. 4b, DMSO control). As no CD4SP cells are produced in the Rag 2^–/– FTOC, CD4 staining is an indicator of DP cells. R59949 did not inhibit the anti-CD3 induced production of DP cells. The percentage CD4⁺ cells increased from 5 to 50%. However, staining of thymocytes with annexin-V–FITC revealed that only 16% of thymocytes were apoptotic in the presence of DMSO, whereas 29·5% of thymocytes were apoptotic in the presence of R59949. These results indicate that R59949, does not block differentiation from the DN to the DP stage in thymocyte development, but induces apoptosis in these thymocytes soon after entry into the DP stage.

Survival pathways mediated by DGKα and the Bcl family are directly linked

The Bcl family of proteins regulate apoptosis in a wide variety of cell types.³¹ Both Bcl-2 and Bcl-x are involved in the prevention of apoptosis in thymocytes. Bcl-2 is expressed in DN and CD4 and CD8 SP subsets, whereas Bcl-xl and Bcl-Xγ are expressed in DP thymocytes. We analysed the expression of Bcl-xl and Bcl-xγ in thymocytes following inhibition of DGKα activity by R59949. Semiquantitiative RT–PCR was performed on thymocyte suspensions following treatment with 5 µm R59949 or the equivalent concentration of DMSO for 4·5 hr. PCR aliquots were removed at 20, 25, 30 and 35 cycles (Fig. 5a). A dramatic down-regulation of Bcl-xl expression was seen in thymocytes following R59949 treatment in comparison with the DMSO control. Expression of Bcl-xγ was difficult to detect but showed no obvious alteration in expression (data not shown). These results suggest that DGKα promotes survival in DP cells through a Bcl-xl mediated pathway.

Figure 5.

DGKα regulates thymocyte survival by a bcl-mediated pathway. (a) Thymocyte suspensions were incubated with 5 µm R59949 or the equivalent DMSO for 4·5 hr. RNA was then isolated and used as a template for cDNA synthesis. Expression of bcl-xl was determined in a semiquantitative PCR assay in which aliquots were removed at 20, 25, 30 and 35 cycles. For each experiment the amount of cDNA used was normalized according to HPRT expression. (b) Thymocytes from WT and Bcl-2 transgenic mice were cultured in the presence of 5 µm R59949 or the equivalent DMSO. Cells were cultured for 1·5, 3, 4·5 and 6 hr and the reaction stopped by addition of 10% FCS. Cells were cultured for a further 12 hr. Apoptosis was determined by staining for expression of CD4, CD8 and annexin-V. Percentage increase apoptosis over background was calculated as in Fig. 2.

As both Bcl-xl and Bcl-2 repress a common pathway of cell death³² we determined the effect of R59949 treatment on thymocytes isolated from Bcl-2 transgenic mice. Although Bcl-2 would not normally be expressed in DP thymocytes, introduction of the transgene under the control of the proximal lck promoter leads to expression in DP thymocytes in the transgenic mouse.³³ Thymocyte suspensions isolated from WT and Bcl-2 transgenic mice were incubated with 5 µm R59949 or the equivalent concentration of DMSO for 1·5, 3, and 4·5 hr. Each reaction was stopped by the addition of FCS and all reactions incubated for a total of 18 hr. Following this period cells were stained with anti-CD4, anti-CD8 and annexin-V–FITC. For each time-point the percentage change in apoptosis induced by R59949 was expressed as in Fig. 2(b) (Fig. 5b). We found that over-expression of Bcl-2 protected thymocytes from R59949-induced apoptosis as R59949 was significantly more effective at inducing apoptosis in WT thymocytes than in Bcl-2 transgenic thymocytes at all time-points analysed. These data suggest that DGKα contributes to the survival of DP thymocytes by regulation of expression of the antiapoptotic protein, Bcl-xl, and that the requirement for DGKα-mediated Bcl-xl up-regulation can be over-come by expression of transgenic Bcl-2 in DP cells.

Discussion

The maintenance of DP thymocyte survival during TCR repertoire selection is essential to the function of the thymus. Although apoptosis of DP thymocytes has been extensively studied, little is known about the control of survival. We show for the first time that the enzyme DGKα promotes this survival process in DP thymocytes. We show that expression of DGKα is up-regulated in the DP thymocytes following a signal through the pre-TCR and that in thymocytes isolated from Rag 2^–/– and TCRβ^–/– mice, both of which are incapable of expressing the pre-TCR, DGKα expression is below levels of detection. Thymocytes from mice deficient in p56^lck, a tyrosine kinase required for efficient pre-TCR signalling, also express reduced levels of DGKα.

Inhibition of DGKα activity using the chemical inhibitor, R59949 resulted in increased cell death in DP thymocytes, indicating that DGKα functions to promote cell survival, especially in the DP subset. Studies of the effect of this inhibitor on thymocytes that express reduced levels of DGKα demonstrate its specificity, as R59949 was much less effective at inducing increased death in lck^–/– thymocytes. Furthermore, thymocytes isolated from lck^–/– mice survive less well in suspension culture then those isolated from a WT mouse. Taken together, these data suggest that activity of DGKα may contribute to survival of those thymocytes that have successfully rearranged their TCRβ chain genes and express a pre-TCR. The pre-TCR signals thymocytes to survive, undergo allelic exclusion at the TCRβ gene locus, proliferate and differentiate into more mature DP cells. Numerous studies have revealed information about the nature of the signalling events controlling allelic exclusion, proliferation and differentiation. Less is known about the signalling events controlling survival in thymocytes subsequent to pre-TCR signalling.

A number of factors have been described that control survival of DN thymocytes before pre-TCR mediated signalling such as Bcl-2³⁴ p53³⁰ and the small GTP-binding protein, Rho^35,36 However, the molecular events that mediate survival on pre-TCR signalling are unknown. Here we show that induction of DGKα gene expression is a consequence of pre-TCR signalling and that DGKα may function to promote survival of early DP cells.

Using the model system for pre-TCR signalling in which Rag^–/– thymocytes are induced to differentiate by anti-CD3 mAb²⁷ we determined the effect of inhibition of DGKα activity by R59949 on survival and differentiation following a signal through the pre-TCR. No apparent effect on differentiation was observed, suggesting that DGKα has no role in the process of differentiation, but serves to promote survival of thymocytes only once they have made the transition from DN to DP stage in development.

The signalling events downstream of DGKα activation are only partially understood and the increased cell survival may be due either to down-regulation of DAG or to up-regulation of PA expression. Both DAG and PA are powerful second messengers and we can not exclude the possibility that DAG and the subsequent activation of the PKC pathway may be involved in this process. Furthermore PKC can regulate the localization of DGK in the cell,³⁷ and conversely DGK can redistribute PKC from the membrane to the cytosol³⁸ introducing another degree of complexity to this process.

The best-documented factors involved in regulation of apoptosis during thymocyte development are the Bcl family of proteins. The founder member, bcl-2, is required for the survival signals transmitted by the IL-7 receptor during thymocyte development.^34,39 The related Bcl-x gene encodes several alternatively spliced isoforms that may enhance or diminish T-cell apoptosis. To date, two forms of Bcl-x, Bcl-xl and Bcl-xγ, have been shown to protect DP thymocytes against induction of apoptosis.^32,40 Bcl-xl has been shown to promote T-cell survival following CD28 signalling⁴¹ and over-expression in transgenic mice increases production of mature thymocytes³² without affecting clonal deletion.⁴² Expression of a second anti-apoptotic Bcl-x isoform, Bcl-xγ, is up-regulated in DP thymocytes following a signal through the TCR and was absent from thymocytes isolated from major histocompatibility complex class I/II double deficient mice. Conversely Bcl-xl expression was detectable in thymocytes from these double-deficient mice.⁴⁰ The data presented in this manuscript suggest that DGKα is involved in controlling this Bcl-mediated survival mechanism. As inhibition of DGKα activity leads to down-regulation of expression of Bcl-xl, we propose that DGKα activity may play a role in the maintenance of survival of immature DP thymocytes allowing them to provide a substrate cell for positive and negative selection. It is at this immature DP stage in development that thymocytes undergo rearrangement at the TCRα gene locus leading to expression of the mature αβ TCR at the cell surface allowing thymocytes to be either positively or negatively selected.

In summary, we show that DGKα is expressed in DP thymocytes following signalling events mediated through the pre-TCR and that activity of DGKα promotes survival of this DP subset in which it is expressed. Inhibition of activity by the chemical inhibitor, R59949, results in a dramatic loss of cell viability. This DGKα-mediated survival appears to involve regulation of expression of the Bcl-xl gene as levels of expression of this gene rapidly decline as the enzyme activity is blocked. Bcl-2 over-expression in DP cells protected them from R59949-induced death confirming the possibility of an interaction between these two signalling pathways. Thus we demonstrate a novel and important role for this molecule during thymocyte development.

Abbreviations

DAG

diacylglycerol

DGK

diacylglycerol kinase

DN

double negative

DP

double positive

FCS

fetal calf serum

FITC

fluoroscein isothiocyanate

FTOC

fetal thymic organ culture

PA

phosphatidic acid

PE

phycoerythrin

PI

propidium iodide

PKC

protein kinase C

RT–PCR

reverse transcriptase–polymerase chain reaction

SP

single positive

TCR

T-cell receptor

WT

wild type