Inositol (1,4,5) trisphosphate 3 kinase B controls positive selection of T cells and modulates Erk activity

Abstract

The mechanisms governing positive selection of T cells in the thymus are still incompletely understood. Here, we describe a N-ethyl-N-nitrosourea induced recessive mouse mutant, Ms. T-less, which lacks T cells in the peripheral blood because of a complete block of thymocyte development at the CD4⁺CD8⁺ stage. Single nucleotide polymorphism mapping and candidate gene sequencing revealed a nonsense mutation in the inositol (1,4,5) trisphosphate 3 kinase B (Itpkb) gene in Ms. T-less mice. Accordingly, Ms. T-less thymocytes do not show detectable expression of Itpkb protein and have drastically reduced basal inositol (1,4,5) trisphosphate kinase activity. Itpkb converts inositol (1,4,5) trisphosphate to inositol (1,3,4,5) tetrakisphosphate, soluble second messengers that have been implicated in Ca²⁺ signaling. Surprisingly, Ca²⁺ responses show no significant differences between wild type (WT) and mutant thymocytes. However, extracellular signal-regulated kinase (Erk) activation in response to suboptimal antigen receptor stimulation is attenuated in Ms. T-less thymocytes, suggesting a role for Itpkb in linking T cell receptor signaling to efficient and sustained Erk activation.

The development of mature T cells is a tightly regulated process that has been studied extensively at the cellular and molecular level. Lymphoid precursors destined to become T cells arrive in the thymus from the bone marrow, where they face a gauntlet of checkpoints to determine their ultimate fate (reviewed in ref. 1). Briefly, T cell development in the thymus can be followed by the expression of the two T cell receptor (TCR) coreceptors CD4 and CD8. Thymocytes at the earliest developmental stage are CD4^–CD8^– double negative cells. After successful rearrangement of the TCR β chain, they undergo rapid proliferation and begin expressing both coreceptors simultaneously, thereby entering the CD4⁺CD8⁺ double positive (DP) stage. At this stage, the TCR α chain is rearranged and expressed on the cell surface to form a functional receptor. In addition, DP cells face a fate decision to either become mature CD4⁺ or CD8⁺ single positive T cells or to die. This fate decision is directed by the avidity and affinity of the TCR on DP thymocytes to self peptides presented by MHC class I or class II molecules (reviewed in ref. 2). Cells that recognize the peptide–MHC complex with high or no affinity die by apoptosis during negative selection or by neglect, respectively. Cells that recognize the peptide–MHC complex with intermediate affinity are positively selected to mature into CD4⁺ or CD8⁺ T cells. The molecular events dictating this differentiation process are an area of intense investigation (reviewed in refs. 3 and 4).

In an attempt to find mediators of immune function, we are conducting a forward genetics screen in mice by using N-ethyl-N-nitrosourea (ENU) (5). Here, we describe Ms. T-less, a recessive mouse mutant that lacks peripheral T cells because of a nearly complete block of T cell development at the DP stage. Single-nucleotide polymorphism mapping and candidate gene sequencing revealed a nonsense mutation in the Itpkb gene. The mutant allele encodes a N-terminally truncated protein, which lacks the lipid kinase domain. We used various antibodies but were unable to detect expression of Itpkb in thymocytes from mutant mice.

Itpkb, also known as inositol (1,4,5) 3 kinase B, converts inositol (1,4,5) trisphosphate (IP3) to inositol (1,3,4,5) tetrakisphosphate (IP4) (6). IP3 is a critical mediator of TCR induced Ca²⁺ release from internal stores (7). Several studies suggest roles for IP4 in calcium signaling in nonlymphoid cells, possibly by modulating the levels of IP3 (8–10).

Mammals express three Itpk isoforms: Itpka, Itpkb, and Itpkc (6, 11, 12). Itpka and Itpkb are regulated through the binding of Ca²⁺/calmodulin. Disruption of the brain-enriched Itpka gene results in minor enhancements of long-term potentiation in the CA1 region of the hippocampus; yet no other major defects have been noted in these mice (13). This mild phenotype may reflect functional redundancy with Itpkb, which shows an overlapping expression pattern (Fig. 5, which is published as supporting information on the PNAS web site). However, Itpkb is also enriched in lymphoid tissues. Itpkc shows a broader tissue expression pattern.

Surprisingly, we did not detect any significant effects on calcium responses in TCR-stimulated CD4⁺CD8⁺ T cells from Ms. T-less mice. Instead, we found a specific defect in the activation of extracellular signal-regulated kinase (Erk), a critical mediator of positive selection (1). This result identifies Itpkb as a unique link between the TCR and the Ras mitogen-activated protein kinase (MAPK) pathway, which is essential for T cell development.

Materials and Methods

Mice. All mice used in this study were between 6 and 12 weeks of age. ENU mutagenized C57BL/6 mice were generated as described (14). Mice were maintained by backcrossing affected animals to C57BL/6 and housed in the Genomics Institute of the Novartis Research Foundation Specific Pathogen Free animal facility. All procedures were approved by the Genomics Institute of the Novartis Research Foundation Institutional Animal Care and Use Committee.

Flow Cytometry. Single cell suspensions of thymus, lymph node, or spleen were stained with FITC-, phycoerythrin-, peridinin chlorophyll protein-, and allophycocyanin-conjugated antibodies against B220, TCRβ, CD4, CD8, CD3, CD69, CD44, CD45.1, and CD45.2 (Pharmingen and eBioscience, San Diego). Cells were analyzed by flow cytometry on a FACSCalibur flow cytometer (Becton Dickinson). Acquisition and analysis were performed with cellquest (Becton Dickinson) and flowjo (TreeStar, Ashland, OR) software.

Analysis of Ca²⁺ Responses. The protocol for measuring intracellular calcium levels by flow cytometry was derived from L. B. Dustin (15). Thymocytes at 10 × 10⁶/ml were labeled in DMEM + 10 mM Hepes with 2 μM Fura red (Molecular Probes), 1 μM Fluo-4 (Molecular Probes), and 0.2% Pluronic (Molecular Probes) for 30 min at room temperature. Cells were washed twice in DMEM + 10 mM Hepes with 1% FCS and rested for 20 min in the dark. For stimulation, labeled cells were incubated with biotinylated αCD3 and αCD4 (Pharmingen and eBioscience) on ice for 15 min, washed, then resuspended with prewarmed streptavidin in Hanks' balanced salt solution (HBSS) with EGTA or CaCl₂, and analyzed by flow cytometry. Ca²⁺ mobilization was determined ratiometrically as described in ref. 15. For single-cell calcium imaging, we adapted a protocol described in ref. 16. Briefly, thymocytes were isolated as above, labeled with 1 μM Fura-2 (Molecular Probes) and 0.2% Pluronic, stained with biotinylated αCD3 and αCD4, and adhered to poly(L-lysine)-treated coverslips. Cells were stimulated by perfusion of streptavidin and imaged on an inverted microscope (Nikon) under 40× magnification with an UV light source. Images were acquired over time at 340 nm and 380 nm, and the A₃₄₀/A₃₈₀ ratio was used to determine the relative intracellular Ca²⁺ concentration.

IP3 Kinase Activity Assay. Itpk activity in thymocyte lysates were performed essentially as described in ref. 17. Briefly, ³H-inositol was added to whole cell lysates. After incubation, the various inositol polyphosphates formed were resolved by thin layer chromatography. Itpk activity was determined by measuring IP3 conversion to IP4.

Immunoblotting and Northern Analysis. Itpkb was immunoprecipitated from whole thymocyte extracts with an antibody against the N-terminal region of rat Itpkb (Santa Cruz Biotechnology). Precipitate eluates or whole cell lysates from sorted DP thymocytes were separated by SDS/PAGE, transferred to nitrocellulose, probed with the N-terminal antibody, and developed by enhanced chemiluminescence (ECL, Amersham Pharmacia). For Northern blot analysis, RNA was isolated from whole thymocytes or sorted DP thymocytes, separated on a denaturing formaldehyde-agarose gel, transferred to nylon, and hybridized with a radioactive probe against an internal portion of the Itpkb transcript.

Erk Activation. Thymocytes were incubated with biotinylated αCD3 and/or αCD4 for 30 min at 4°C with rotation, followed by stimulation with prewarmed streptavidin or phorbol 12-myristate 13-acetate at the indicated time points. Stimulation was stopped by adding ice-cold PBS. Cells were then washed and lysed. Protein lysates were subjected to SDS/PAGE, transferred to nitrocellulose, probed with antibodies to Erk1/Erk2 or phospho-Erk1/Erk2 (Cell Signaling Technology), and detected by enhanced chemiluminescence.

Supporting Materials and Methods. Further information can be found in Supporting Materials and Methods, which is published as supporting information on the PNAS web site.

Results

Identification of an ENU Mouse Mutant with No Peripheral Blood T Cells. By using ENU mutagenesis to generate mice with defects in lymphoid development, we identified Ms. T-less, a mutant with a specific lack of peripheral T cells (Fig. 1A). The decrease in CD3⁺ cells encompassed both CD4⁺ and CD8⁺ T cells. There were no obvious alterations in other peripheral blood cells analyzed, including B220⁺ B cells. Mutant mice displayed this phenotype in a manner representative of a recessive trait and were obtained at expected Mendelian frequencies (data not shown). They showed no gross physical abnormalities or overt behavioral defects (data not shown).

Fig. 1.

Ms. T-less mice display a paucity of peripheral T cells and a CD4⁺CD8⁺ block in T cell development. (A) Peripheral blood lymphocytes from mutant and control mice on a C57BL/6J background were stained with antibodies to CD3, B220, CD4, and CD8. The scatter plots show lymphocyte subpopulations as % of total lymphocytes. Thymocytes from WT (wt) and mutant (mut) mice were stained with antibodies against CD4 and CD8 to follow T cell development (B) and with antibodies to activation markers, including CD69, CD3, and TCRβ (C). The histograms in C are gated on CD4⁺CD8⁺ thymocytes. (D) Spleens from WT and mutant mice were stained with antibodies against CD4 and CD8 to analyze the peripheral T cell compartment.

Ms. T-less Mice Display a Block in T Cell Development at the CD4⁺CD8⁺ Stage. To determine the cause of the lack of T cells in the peripheral blood, we analyzed T cell development in the thymus. We found a nearly complete block at the CD4⁺CD8⁺ DP stage (Fig. 1B). The DP cells from mutant mice do not efficiently up-regulate activation markers, such as CD69; nor do they increase cell surface expression of CD3 or the TCR β chain (Fig. 1C). The lack of CD69⁺, CD3^hi, and TCRβ^hi cells in the DP population suggests that Ms. T-less thymocytes are unable to respond to or properly translate signals emanating from the TCR. Thymic cellularity of age- and sex-matched mutant mice is slightly larger than that of WT mice, suggesting that there are no gross proliferative defects in Ms. T-less thymocytes (Fig. 6, which is published as supporting information on the PNAS web site). Concurrent with a block of T cell development at the DP stage, development of γδ T cells appears unaffected (data not shown). γδ T cells differentiate from αβ T cells before the DP stage. Therefore, the developmental block in Ms. T-less mice is specific for the αβ T cell compartment, likely resulting from a specific defect in thymic selection rather than a generalized impairment of T cell development.

The spleen and lymph nodes of Ms. T-less mice do not contain significant numbers of CD4⁺ T cells, although a small population of these cells do accumulate in older mice (Fig. 1D and data not shown). Expression of the TCR β chain and high levels of CD44 (data not shown) suggest that these cells may have expanded in a manner reminiscent of homeostatic proliferation to fill a lymphopenic environment (18, 19).

The Defect in T Cell Development Is Inherent to the T Cells. The DP block of T cell development in Ms. T-less mice is reminiscent of the phenotype seen in mice lacking both MHC I and MHC II proteins (20, 21). However, we found no major differences in MHC protein expression between WT and mutant animals (data not shown). One possibility is that an unknown ligand necessary for differentiation into mature CD4⁺ or CD8⁺ T cells is presented on the thymic epithelium and lacking in Ms. T-less mice. To address this issue, we performed bone marrow reconstitution experiments into lethally irradiated, B6.SJL hosts. The irradiation depletes the hosts of their complement of hematopoietic cells and precursors but keeps the thymic epithelium intact. We observed a profound block of T cell development at the DP stage in hosts reconstituted with Ms. T-less bone marrow (Fig. 7, which is published as supporting information on the PNAS web site). In addition, no T cells were present in the periphery, although B cells reconstituted efficiently (data not shown). Bone marrow from WT mice exhibited normal T cell development in the host thymus. WT bone marrow reconstitution into lethally irradiated Ms. T-less hosts exhibited normal T cell development in the thymus (data not shown). These findings indicate that the developmental block in T cell maturation is intrinsic to the developing thymocytes and does not depend on ligands on, or signals from, the thymic epithelium.

Ms. T-less Mice Harbor a Nonsense Mutation in Itpkb. To determine the genetic lesion underlying the Ms. T-less phenotype, mutant mice on a C57BL/6 background were crossed to WT 129SvJ mice. Single-nucleotide polymorphism genotyping of multiple phenotypically mutant or WT F2 offspring revealed a perfect phenotype–genotype correlation for a 2-megabase interval distal on chromosome 1 (Fig. 8A, which is published as supporting information on the PNAS web site, and ref. 22). Analysis of this region did not reveal any obvious candidate genes known to be involved in thymic development. Thus, we examined the expression status of most known or predicted genes in the region by using the Genomics Institute of the Novartis Research Foundation Gene Expression Atlas (http://expression.gnf.org) (23). We found that the Itpkb transcript accumulates in both murine and human lymphoid tissues, especially the thymus (Fig. 5 B and C). Sequencing of this candidate gene revealedaTtoA transversion at position 596 of the transcript, changing the codon encoding cysteine 199 to a stop (Fig. 8B). The mutant transcript encodes an N-terminally truncated Itpkb protein lacking most of its structure, including domains that are involved in targeting and regulation, as well as the catalytic domain. Immunoblot analyses of lysates from sorted CD4⁺CD8⁺ DP cells or of immunoprecipitates from whole thymocyte extracts revealed that Ms. T-less thymocytes lack full-length Itpkb protein (Fig. 2A). Expression of Itpkb RNA, however, is quite abundant in sorted mutant DP thymocytes (Fig. 2B). In agreement with these data, extracts from Ms. T-less thymocytes showed an ≈50% reduction of Itpk activity compared with WT extracts (Fig. 2C). This residual activity could reflect low-level thymic expression of other Itpk isoforms (6, 11, 12, 24). Our data suggest that lack of full-length Itpkb protein expression and concomitant reduction of Itpk activity in thymocytes underlies the defect in T cell development in Ms. T-less mice.

Fig. 2.

Ms. T-less thymocytes lack functional Itpkb protein. (A) Lack of full-length Itpkb protein in Ms. T-less thymocytes. (Left) Thymocytes were immunoprecipitated with an antibody against Itpkb, followed by SDS/PAGE and immunoblot analysis. (Right) Whole-cell lysates from sorted CD4⁺CD8⁺ cells were analyzed for the presence of Itpkb protein. All immunoblots were probed with an antibody against an N-terminal peptide of Itpkb. (B) Northern blot analysis of Itpkb expression in sorted WT and mutant CD4⁺CD8⁺ thymocytes with an internal probe (Upper). (Lower) Expression of GAPDH as loading control. (C) Reduced IP3 kinase activity in Ms. T-less thymocytes. Extracts from WT (black bars) and mutant (white bars) thymocytes were incubated with ³H-inositol for the indicated times, after which inositol polyphosphates were separated by thin layer chromatography and quantified. Shown are IP4 levels as the percent of total ³H-inositol converted. The rate constants of the reaction, k_obs, which describe the rate of conversion of IP3 to IP4, for WT and mutant extracts are 0.02 min^–1 and 0.01 min^–1, respectively.

Normal Ca²⁺ Responses in Ms. T-less Thymocytes. Itpkb converts IP3 to IP4. IP3 is a well characterized second messenger involved in calcium signaling (25). TCR ligation leads to activation of phospholipase Cγ, which hydrolyzes phosphatidylinositol (4,5) bisphosphate (PIP2) to diacylglycerol and IP3. The augmentation of intracellular IP3 levels triggers the release of Ca²⁺ from internal stores by means of IP3 receptors (26). In Jurkat cells, Itpk activity and IP4 production are elevated during TCR stimulation (27). Thus, Itpkb could serve to limit TCR-induced Ca²⁺ mobilization through conversion of IP3 to IP4. However, it has also been postulated that IP4 can potentiate IP3 signaling through the specific inhibition of a 5′-phosphatase that hydrolyzes IP3 to inositol (1,4) bisphosphate (10). Thus, formation of IP4 could also affect Ca²⁺ mobilization positively. We therefore investigated whether defects in Ca²⁺ signaling might underlie the developmental defect in Ms. T-less thymocytes.

As shown in Fig. 3A, bulk Ca²⁺ responses to stimulation through the TCR or with thapsigargin, a chemical that depletes internal Ca²⁺ stores and bypasses the requirement for IP3 generation, are similar between WT and mutant thymocytes in the absence or presence of external Ca²⁺. Thus, Ms. T-less thymocytes have no major defects in internal Ca²⁺ release or external Ca²⁺ influx. We next investigated the Ca²⁺ responses of individual cells to TCR stimulation. Thymocytes undergoing positive selection display dramatic oscillations of intracellular Ca²⁺ levels that have a periodicity on the order of seconds (16). Cells that do not receive signals for positive selection do not show the same magnitude of oscillations at the single-cell level. In HeLa cells, Itpk activity and IP4 have been implicated in the modulation of histamine-induced Ca²⁺ oscillations (28). Surprisingly, single-cell imaging of Fura-2 labeled thymocytes did not reveal significant differences in magnitude or periodicity of TCR-induced Ca²⁺ oscillations between WT and mutant mice (Fig. 3B). Taken together, these data suggest that defects in Ca²⁺ signaling are unlikely to underlie the profound defect in T cell development observed in Ms. T-less mice.

Fig. 3.

Normal Ca²⁺ responses but impaired Erk activation after TCR stimulation in Ms. T-less thymocytes. (A) Calcium responses of bulk DP thymocyte populations from stimulated WT (green lines) or mutant (orange lines) and unstimulated WT (red lines) or mutant (blue lines) mice were analyzed by flow cytometry. Cells were either stimulated with αCD3/αCD4 followed by streptavidin crosslinking (Upper) or thapsigargin (Thaps) (Lower), and responses were measured in the absence (Left) or presence (Right) of exogenous Ca²⁺. The arrows indicate the times when streptavidin, thapsigargin, or ionomycin (Iono) were added. (B) Single-cell measurements of intracellular Ca²⁺ concentrations over time. WT and mutant thymocytes were labeled with Fura-2, and single cells were imaged after stimulation with αCD3/αCD4/streptavidin, followed by addition of ionomycin. Relative intracellular Ca²⁺ concentrations were determined ratiometrically. (Upper) Several traces superimposed. (Lower) Traces of a single cell. (C) Thymocytes from WT and Ms. T-less mutant mice were isolated and stimulated as indicated at the indicated time points. Activated Erk1 and Erk2 were detected with an antibody specific to the phosphorylated forms of those proteins. Total Erk1 and Erk2 were also detected to determine equal loading of the samples.

Defective Erk Activation in Ms. T-less Thymocytes: A Role for Itpkb in Ras Signaling. The lack of an overt effect on Ca²⁺ responses in Ms. T-less thymocytes led us to consider other mechanisms of how Itpkb could control T cell selection. Because IP3 levels do not seem to be affected in these mice (unpublished observations), we addressed putative mechanisms involving IP4. The protein GT-Pase-activating protein (GAP)1^IP4BP has been shown to bind IP4 with high affinity and specificity in vitro (29). GAP1^IP4BP is a protein that stimulates the small GTPase Ras to convert GTP to GDP, rendering Ras inactive. Several studies have demonstrated essential roles for the Ras pathway in T cell development. Functional inactivation of the Ras activator Ras guanine nucleotide-releasing protein (RasGRP) (30), Ras (31), or components of the MAPK pathway downstream of Ras (32–35) all affect maturation of DP thymocytes and positive selection. Therefore, GAP1^IP4BP could be an important component that connects Itpkb-mediated IP4 production to Ras activation in T cell development.

To address activation of the Ras pathway in Ms. T-less thymocytes, cells were stimulated with either a suboptimal TCR signal by using αCD3 alone or with a maximal TCR signal by using a combination of αCD3 and αCD4 antibodies (Fig. 3C). Ras activation leads to the activation and phosphorylation of the MAPKs Erk1 and Erk2. By using immunoblot analysis, we found a significant impairment of Erk1 and Erk2 activation in response to suboptimal (αCD3 alone) stimulation in Ms. T-less thymocytes. Optimal stimulation conditions (αCD3 and αCD4) or stimulation with the diacylglycerol analog phorbol 12-myristate 13-acetate, which can localize RasGRP1 to the plasma membrane to allow for Ras activation (36, 37), elicited normal levels of Erk1 and Erk2 activation. These data demonstrate that Ms. T-less thymocytes are unable to efficiently activate Erk1 and Erk2 under moderate stimulation conditions. Thus, the block during positive selection in Ms. T-less thymocytes may reflect critical roles for Itpkb and its product IP4 as regulators of TCR-induced Ras activation.

Discussion

By using ENU mutagenesis, we have identified Ms. T-less, a mouse mutant with a paucity of peripheral T cells resulting from a block of αβ T cell development at the CD4⁺CD8⁺ DP stage with a phenotype suggestive of impaired positive selection. This defect is intrinsic to the developing T cells and can be attributed to a loss-of-function mutation in the gene Itpkb. Although Itpkb converts IP3, an important second messenger that mediates Ca²⁺ responses to TCR stimulation, into IP4, which has been implicated in modulating Ca²⁺ oscillations in HeLa cells, we found no obvious changes in IP3 levels or Ca²⁺ responses in mutant thymocytes stimulated by the TCR. These findings corroborate previous reports that nonspecific Itpk inhibition in Jurkat cells with adriamycin did not affect TCR-induced Ca²⁺ mobilization (38). We have, however, identified a significant defect in Erk activation after suboptimal stimulation of Ms. T-less thymocytes, which may be a direct consequence of insufficient Ras activation.

Our data substantiate and suggest a mechanistic explanation for the similar T cell developmental phenotype resulting from disruption of the Itpkb gene, which was reported by Pouillon et al. (39) while this manuscript was in preparation. The authors found similar defects in positive selection and unimpaired Ca²⁺ signaling in mice lacking Itpkb, even on transgenic TCR backgrounds.

Based on our finding of impaired TCR-induced Erk activation in Ms. T-less thymocytes, we propose a model of Ras activation that is regulated by Itpkb through the production of IP4 (Fig. 4). In naïve thymocytes, Ras is kept in the inactive GDP-bound state by the action of RasGAPs, in this case GAP1^IP4BP, which hydrolyzes Ras-GTP to Ras-GDP. GAP1^IP4BP is membrane-associated through an interaction of its pleckstrin homology (PH) domain with membrane bound PIP2. This same PH domain can also bind water-soluble IP4; thus, the balance between membrane-bound PIP2 and soluble IP4 may determine the localization of GAP1^IP4BP (29, 40). High-level expression of GAP1^IP4BP is restricted to lymphoid subsets (Fig. 9, which is published as supporting information on the PNAS web site). In thymocytes undergoing selection, TCR stimulation activates phospholipase Cγ, which hydrolizes PIP2 to diacylglycerol and IP3. Diacylglycerol recruits the Ras-activating RasGRP1 to the plasma membrane, which activates Ras through an exchange of GDP for GTP and initiates the MAPK cascade, leading to Erk activation (41). At the same time, hydrolysis of PIP2 weakens the interaction of GAP1^IP4BP with the plasma membrane. Accumulation of IP3 triggers Ca²⁺ release from intracellular stores and subsequent Ca²⁺ influx, allowing for Ca²⁺-dependent activation of Itpkb, possibly through Ca²⁺/calmodulin and Ca²⁺/calmodulin-dependent kinase II (42, 43). Itpkb converts IP3 to IP4, which may sequester GAP1^IP4BP in the cytoplasm by competition with the already low levels of PIP2 for binding to its PH domain. Sequestration of GAP1^IP4BP in the cytoplasm allows unhindered Ras activation at the plasma membrane. In Ms. T-less mutant thymocytes that lack Itpkb, a severe reduction in the generation of IP4 could presumably prevent the efficient removal of GAP1^IP4BP from the plasma membrane and thereby prevent sufficient Ras and Erk activation to allow for positive T cell selection.

Fig. 4.

A model for the role of Itpkb and its product IP4 in the regulation of Ras signaling to Erk in thymocytes. For details, see Discussion.

Our model provides an intriguing rationale for the defects in TCR-induced Erk activation and positive selection of Ms. T-less thymocytes caused by impaired Itpkb function. However, recent data suggest that the regulation of Ras activation may be more complex. First, T cells express another RasGAP, Ca²⁺-promoted Ras inactivation (CAPRI), in addition to GAP1^IP4BP (44). Second, a recent study suggests that in various cell types, including Jurkat cells, Ras exists both at the plasma membrane and on the surface of the Golgi compartment (45). However, only Golgi-associated Ras is activated in response to TCR stimulation by a mechanism involving RasGRP recruitment to the Golgi membrane in a phospholipase Cγ-dependent manner. In contrast, plasma membrane Ras is kept inactive through a Ca²⁺-mediated, PIP2/PIP3-independent plasma membrane recruitment of CAPRI. Thus, although both GAP1^IP4BP and CAPRI are likely negative regulators of Ras activation in T cells, their respective mechanisms of recruitment to membrane-bound Ras differ, and they are likely to be differentially affected by IP4. It will be interesting to investigate whether GAP1^IP4BP controls activation of the possibly more relevant Golgi-bound Ras in T cells.

It is not surprising that Ms. T-less mutant thymocytes exhibit a block in Erk activation only in response to suboptimal stimulation. Most genetic studies involving the Ras pathway have documented a dependence of T cell selection on the strength of the TCR signal, which modulates the level of Ras activation. RasGRP1^–/– mice exhibit a nearly complete block in positive selection (30), but in the context of a TCR transgene capable of providing sufficient signal strength, positive selection can occur normally (46). Additionally, transgenic mice expressing a dominant negative form of Ras (31) display impaired positive selection, a phenotype that is exacerbated when crossed to a transgenic mouse bearing a dominant negative form of MEK1, the MAPK kinase downstream of Ras activation (35). The additive effect of the two transgenes again points to the importance of the magnitude of Ras activation in the differentiation of immature thymocytes. Finally, genetic alteration of components of the Erk-MAPK signaling pathway (32–34) or chemical modulation of the Ras-Erk pathway (47, 48) can affect positive selection of T cells and alter lineage commitment. Collectively, these studies suggest that the amount of Ras activation is important for the developmental decisions made by T cells undergoing selection. In this regard, Itpkb and its product IP4 may serve to translate low-level TCR signaling that occurs during positive selection into levels of Ras-Erk activation that are sufficient to allow for thymocyte maturation.

IP4 is a precursor for several higher order inositol polyphosphates, some of which are generated in TCR-stimulated T cells (49, 50). Their physiological roles are largely undefined (9), but recent studies in yeast suggest functions in regulating chromatin remodeling (51, 52). Although we have no evidence for an involvement of IP4 in chromatin remodeling in thymocytes, further investigation is required. At present, it is more intriguing to focus on the relative roles of IP4 and RasGAPs in regulating T cell signaling through Ras. Regardless of the precise mechanism, this study clearly defines a critical role for Itpkb in the regulation of thymocyte maturation.