The Akt/mTOR pathway is a key driver of murine CD4+ T cell differentiation, and induction of regulatory T (Treg) cells results from low TCR signal strength and low Akt/mTOR signaling. However, strong TCR signals induce high Akt activity that promotes Th cell induction. Yet, it is unclear how Akt controls alternate T cell fate decisions. We find that the strength of the TCR signal results in differential Akt enzymatic activity. Surprisingly, the Akt substrate networks associated with T cell fate decisions are qualitatively different. Proteomic profiling of Akt signaling networks during Treg versus Th induction demonstrates that Akt differentially regulates RNA processing and splicing factors to drive T cell differentiation. Interestingly, heterogeneous nuclear ribonucleoprotein (hnRNP) L or hnRNP A1 are Akt substrates during Treg induction and have known roles in regulating the stability and splicing of key mRNAs that code for proteins in the canonical TCR signaling pathway, including CD3ζ and CD45. Functionally, inhibition of Akt enzymatic activity results in the dysregulation of splicing during T cell differentiation, and knockdown of hnRNP L or hnRNP A1 results in the lower induction of Treg cells. Together, this work suggests that a switch in substrate specificity coupled to the phosphorylation status of Akt may lead to alternative cell fates and demonstrates that proteins involved with alternative splicing are important factors in T cell fate decisions.

This article is featured in In This Issue, p.379

CD4+ regulatory T (Treg) cells prevent autoimmunity and control immunopathology during immune responses. The TCR binds to peptide–MHC complexes, initiating a signaling cascade that determines the fate of the T cell. The Akt/mTOR pathway plays a critical role in determining CD4+ T cell fate (13). The induction of Treg cells is inversely correlated with the degree of Akt/mTOR signaling (13). A low degree of Akt/mTOR signaling is necessary for Treg stability and function (46). Akt/mTOR signaling is tightly regulated in T cells, and several feedback loops control the degree of Akt/mTOR activation, one of which involves phosphatase and tensin homolog (PTEN), the transcription factors Foxp3 and FoxO1, and mTORC2 (7, 8).

Akt is a serine threonine kinase that regulates cellular processes including cell proliferation and cellular metabolism (9, 10). Akt activity is controlled by phosphorylation at Ser473 and Thr308. When Treg cells are activated, only Akt Thr308 is phosphorylated, and phosphorylation of both Thr308 and Ser473 sites results in the loss of Treg suppressive function (4). It appears that Treg function depends on the ability to modulate Akt activity and that Akt may regulate other key proteins important for the differentiation and function of either Th or Treg cells. There are >100 known Akt substrates (11), yet the full network of substrates phosphorylated by Akt during Th or Treg differentiation remains largely unexplored.

To better define how Akt functions in T cell differentiation, we identified Akt substrates during the induction of Th and Treg cells via mass spectrometry. Remarkably, phosphorylated Akt substrates differed during Treg versus Th cell induction, and this was associated with different patterns of Akt phosphorylation under both conditions. RNA processing factors were major Akt targets during both Th and Treg induction, including heterogeneous nuclear ribonucleoprotein (hnRNP) L and hnRNP A1. We determined that RNA splicing of TCR signaling proteins, such as CD3ζ and CD45, was regulated by Akt and essential for the Th versus Treg cell fate choice. Knockdown of hnRNP L or hnRNP A1 altered the TCR signal and changed the ratios of Th versus Treg cells induced. Together, these results reveal that there are distinct Akt signaling networks that drive Th versus Treg induction, and identify RNA splicing factors as determinants of CD4+ T cell fate decisions.

C57BL/6 mice were purchased from The Jackson Laboratory. C57BL/6-Foxp3tm1Flv/J (Foxp3-RFP reporter) mice were a kind gift from Dr. D. Vignali (University of Pittsburgh). All of the mice were housed in a pathogen-free facility at the University of Pittsburgh. Mice were handled under Institutional Animal Care and Use Committee–approved guidelines in accordance with approved protocols.

CD4+ T cells were isolated as previously described (7). For the mass spectrometry and biochemical experiments, CD4+ T cells were isolated from C57BL/6 spleens using a CD4+ negative selection kit (Miltenyi Biotec). Induction of Treg and Th cells was performed as previously described (7). To generate Treg cells, we activated freshly isolated CD4+ T cells with 0.25 μg/ml plate-bound anti-CD3 mAb with 1 μg/ml soluble anti-CD28 mAb. To generate Th cells, we activated freshly isolated CD4+ T cells with 1.0 μg/ml plate-bound anti-CD3 mAb with 1 μg/ml soluble anti-CD28 mAb. In some experiments, natural Treg (nTreg; CD4+ Foxp3-RFP+) and naive T (CD4+ Foxp3-RFP) cells were sorted using a FACSAria from the spleen of Foxp3-RFP reporter mice.

Five million CD4+ T cells were activated in Treg or Th conditions in the presence or absence of an Akt inhibitor (iAkt; Akt1/2 10 μM). After 15 min of activation, cells were lysed in 500 μl of ice-cold buffer containing 0.5% (v/v) Nonidet P-40, 1% (v/v) Triton X-100, 150 mM NaCl, 50 mM Tris-HCl [pH 7.4], 1 mM EDTA, and 1 mM EGTA. The lysis buffer also contained protease and phosphatase inhibitors (1 mM Na3VO4, 1 mM aprotinin, 1 mM leupeptin, 1 mM pepstatin, 10 mM NaF, 1 mM PMSF, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate). The lysate was sonicated, centrifuged, and incubated with an Ab specific to phosphorylated Akt substrate motif RXRXX (phospho-Thr/phospho-Ser) (Cell Signaling Technologies) at a ratio of 1 μl of Ab to 50 μl of cell lysate at 4°C for 12 h. The Ab–Akt substrate complexes were captured by protein A–coated beads. As a negative control, CD4+ T cell lysate not treated with the Akt substrate Ab was incubated with the protein A beads. Beads were washed two times with lysis buffer and two times in PBS. The captured Akt substrates were eluted from the beads with a urea buffer (8 M urea and 0.1 M Tris/HCl [pH 8.5]). To generate tryptic fragments, we used the filter-aided sample preparation method (12). All of the immunoprecipitates (IP) were done in duplicate.

Bottom-up proteomics experiments were conducted with either a Thermo-Finnegan Velos-Orbitrap (Velos Orbi) or Q-Exactive (QE) mass spectrometer coupled to a Waters nanoAcquity ultra-HPLC system via a nanoelectrospray ionization source. Between 1 and 5 μl of a sample containing tryptic peptides was injected on a Waters Acquity BEH 300 C18 75 μm internal diameter × 100-mm-long column packed with 1.7-μm particles running a linear gradient of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B) at 1 μl/min as follows: 98:2 for 10 min, 92:8 at 12 min, 65:35 at 65 min, 20:80 at 68 min until 75 min, then return to initial conditions at 76 min and held there to 90 min. Mass spectra were acquired in the Orbitrap at 60,000 resolving power, and tandem mass spectra were then generated for the top 10 (with the Velos Orbi) or 12 (with the QE) most abundant ions with charge states ranging between two and five. Fragmentation of selected peptide ions was achieved via collisionally induced dissociation at normalized collision energy of 35 eV in either the Velos mass analyzer region of the Velos Orbi or in the C-trap section of the QE. Proteome Discoverer 1.4 software using either Mascot or Sequest search engines with a decoy search at a 1% false discovery rate was used to identify proteins present in each sample by matching tandem mass spectra with peptides expected for proteins in the SwissProt database.

Akt substrates identified during Th and Treg induction were analyzed using the Ingenuity software package (Qiagen). A core analysis was performed using default settings to identify pathways that were targeted by Akt, which were used to bin Akt targets into functional categories. Further analysis was performed using the PANTHER classification system (13). The p values were calculated by the Ingenuity software package using the right-tailed Fisher exact test, and the recommended p value cutoff for significance was <0.05.

Western blot analysis was performed with Abs purchased from Cell Signaling Technology: phosphor-Zap70 (2701), total phospho-tyrosine (p-Tyr-1000), Akt (C67E7), phospho-308 Akt (D25E6), phospho-473 Akt (C67E7), phosphor-Rictor (D30A3; Cell Signaling), Rictor (53A2), Akt substrate (23C8D2), FoxO1 (C29H4), hnRNP A1 (D21H11), and Actin (13E5). The hnRNP L Ab (H-78) was purchased from Santa Cruz Biotechnology.

CD4+ T cells were stained with the following mAbs: anti–CD3-PE (BD), anti–CD4-PerCP5.5 (eBioscience), anti–CD25-Pe-Cy7 (eBioscience), anti-Foxp3-Pacific Blue (eBioscience), anti-CD45RB FITC (eBioscience), and anti–phospho-S6 (Ser235) Alexa-647 (Cell Signaling Technology) using buffers from eBioscience. The stained cells were analyzed on an LSR II flow cytometer, and data were analyzed with the FlowJo software package.

Total RNA was isolated from CD4+ T cells with the RNeasy Plus Mini Kit (Qiagen). Reverse transcription reactions were performed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). A CD45 splicing assay was performed as previously described (14). In brief, primers were designed to flank exons 4–6 of CD45 (primer 1: 5′-GGCAACACCTACACCCA-3′, primer 2: 5′-GCTTGCAGGCCCAGA-3′). PCRs (30 cycles) were performed using the goTaq DNA polymerase (Promega). PCRs were resolved on a 1% agarose gel, SeaKem LE Agarose (Lonza), stained with ethidium bromide (Sigma) and imaged on a Red imaging system (Protein system).

Total RNA was isolated and reverse transcribed as described above. Primer pairs were designed to look at the splice variant usage for Lck, Fyn Tec, and CD247 in resting and activated CD4+ T cells (Table I). Quantitative PCR (qPCR) was performed on a StepOne Plus real-time PCR system with the SYBR Green Master Mix (Applied Biosystems) from three independent experiments. All results are reported, unless otherwise stated, as fold change (2−ΔΔCT) relative to resting T cells.

Small interfering RNA (siRNA) kits (Origene) were used to knock down the expression of hnRNP A1 and hnRNP L in mouse primary CD4+ T cells. The nucleofector kits for mouse T cells (Lonza) and Amaxa nucleofector were used to introduce siRNAs into murine CD4+ T cells. To measure the knockdown efficiency, we performed both qPCR analysis and Western blot analysis in triplicate.

All statistics were calculated in the GraphPad Prism 7 software package. All analyses used the two-way ANOVA analysis with Bonferroni postanalysis correction.

We previously identified the dominant role of Ag dose in the induction and expansion of Treg cells (3). In these studies, we showed that low-dose Ag favored the induction/expansion of Treg cells in the islet-specific BDC2.5 and in two OVA-specific TCR transgenic lines, OTII and DO11.10 (3), and in T cells stimulated by anti-CD3 (3, 8). The cells generated by these cultures had strong suppressive function in vitro (3) and in vivo (15). Signaling via the Akt/mTOR pathway influences CD4+ T cell differentiation; low levels favor Treg induction and high levels favor Th induction (13). We recently described several important feedback loops that control the degree of Akt/mTOR activation in CD4 T cells. These involve the lipid phosphatase PTEN, the transcription factors Foxp3 and FoxO1, and Akt (7). Importantly, phosphorylation of the transcription factor FoxO1 by Akt was critical for the observed reduction of PTEN levels seen after high-dose stimulation (7). These studies prompted us to examine how TCR signal strength influenced the phosphorylation status of Akt. CD4+ T cells isolated from C57BL/6 mice were stimulated with increasing doses of plate-bound anti-CD3 in the presence of anti-CD28. Western blot analysis of Ser473 phosphorylation revealed a sharp threshold for Ser473 phosphorylation (Fig. 1A, 1B). Western blot analysis revealed that phosphorylation of Thr308 occurred during both high- (1 μg/ml) and low-dose (0.25 μg/ml) stimulation. Phosphorylation of Ser473 was observed only under high-dose conditions (Fig. 1C, 1E).

FIGURE 1.

TCR signal strength controls Akt phosphorylation status and substrate specificity. (A and B) Western blot of p-Ser473 and total Akt in CD4 T cells stimulated with the indicated concentrations of anti-CD3. A representative blot (A) and the mean ± SEM of three independent experiments (B) are shown. (C) The phosphorylation status of Thr308 and Ser473 on Akt and total Akt was examined by Western blot on induced Treg (low, 0.25 μg/ml) and Th (high, 1 μg/ml) cells activated for the indicated time points. (D and E) Densitometry was performed to quantitate the levels of p-Thr308 and p-Ser473 observed from the Western blot analysis in (C) from three independent experiments. (F) The phosphorylation status of Rictor Thr1135 was monitored during low- (0.25 μg/ml) and high-dose (1 μg/ml) stimulation by Western blot. Total levels of Rictor and Actin are also shown. These experiments are representative of three independent experiments and quantitated by densitometry in (G). Purified CD4+ T cells were activated under Treg (H) or Th (I) induction conditions for the indicated time points, and Western blot analysis was performed with an Ab specific for the phospho-Akt substrate motif. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 1.

TCR signal strength controls Akt phosphorylation status and substrate specificity. (A and B) Western blot of p-Ser473 and total Akt in CD4 T cells stimulated with the indicated concentrations of anti-CD3. A representative blot (A) and the mean ± SEM of three independent experiments (B) are shown. (C) The phosphorylation status of Thr308 and Ser473 on Akt and total Akt was examined by Western blot on induced Treg (low, 0.25 μg/ml) and Th (high, 1 μg/ml) cells activated for the indicated time points. (D and E) Densitometry was performed to quantitate the levels of p-Thr308 and p-Ser473 observed from the Western blot analysis in (C) from three independent experiments. (F) The phosphorylation status of Rictor Thr1135 was monitored during low- (0.25 μg/ml) and high-dose (1 μg/ml) stimulation by Western blot. Total levels of Rictor and Actin are also shown. These experiments are representative of three independent experiments and quantitated by densitometry in (G). Purified CD4+ T cells were activated under Treg (H) or Th (I) induction conditions for the indicated time points, and Western blot analysis was performed with an Ab specific for the phospho-Akt substrate motif. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Close modal

The lack of Ser473 phosphorylation on Akt under low-dose conditions could be because of reduced mTORC2 activity, which is the only known kinase to phosphorylate this site (16). The activity of mTORC2 is decreased by phosphorylation of the Thr1135 residue on the rapamycin-insensitive companion of mTOR (Rictor) subunit (17). Western blot analysis revealed that Rictor Thr1135 was constitutively phosphorylated in unstimulated CD4+ T cells (Fig. 1F, 1G). Thr1135 remained phosphorylated under low-dose conditions (Fig. 1C, 1D), but it was significantly reduced during high-dose stimulation (Fig. 1D).

To determine whether differential Akt phosphorylation altered the substrates targeted by Akt, we performed a time course of low- and high-dose stimulation followed by Western blot analysis with an Ab that specifically recognizes the phosphorylated Akt substrate motif (XRXRXXS/TX) (18). Substrates appeared at 15 min and persisted in both high- and low-dose conditions (Fig. 1H, 1I). There were clear qualitative differences in Akt substrates phosphorylated during low- or high-dose stimulation (Fig. 1H, 1I).

To identify proteins phosphorylated by Akt during high- and low-dose stimulation, we used a mass spectrometric approach (Fig. 2A). The anti–phospho-(Ser/Thr) Akt substrate Ab (18) was used to IP phosphorylated Akt substrates from CD4+ T cells stimulated with low- or high-dose anti-CD3 for 15 min in the presence or absence of an iAkt (7). This analysis was repeated twice, and 99 unique proteins were identified after removal of those that occurred with the iAkt or control beads, those that were identified in only one of the two experiments, and those that did not contain an Akt consensus substrate motif as determined by the Scansite server (19).

FIGURE 2.

Proteomic and pathway analysis of Akt signaling during Treg versus Th induction. (A) An immunoprecipitation workflow was developed to isolate phospho-Akt substrates during high- and low-dose TCR stimulation, and the Akt substrates were identified with mass spectrometry. (B) From the mass spectrometry analysis, we identified 50 Akt substrates unique to Treg (low) induction, 31 Akt substrates unique to Th (high) induction, and 18 shared Akt substrates. (C) Analysis performed with the Ingenuity software package identified that similar processes were targeted by Akt during both low (upper panel) and high dose (lower panel). (D) A pathway analysis of the Akt substrates identified during low-dose (Treg, checkered bars) and high-dose (Th, solid bars) stimulation was performed in the Ingenuity software package. (E) Akt substrates were placed into functional classes overrepresented in the Treg (checkered bars) or Th (solid bars) datasets. p < 0.05.

FIGURE 2.

Proteomic and pathway analysis of Akt signaling during Treg versus Th induction. (A) An immunoprecipitation workflow was developed to isolate phospho-Akt substrates during high- and low-dose TCR stimulation, and the Akt substrates were identified with mass spectrometry. (B) From the mass spectrometry analysis, we identified 50 Akt substrates unique to Treg (low) induction, 31 Akt substrates unique to Th (high) induction, and 18 shared Akt substrates. (C) Analysis performed with the Ingenuity software package identified that similar processes were targeted by Akt during both low (upper panel) and high dose (lower panel). (D) A pathway analysis of the Akt substrates identified during low-dose (Treg, checkered bars) and high-dose (Th, solid bars) stimulation was performed in the Ingenuity software package. (E) Akt substrates were placed into functional classes overrepresented in the Treg (checkered bars) or Th (solid bars) datasets. p < 0.05.

Close modal

Of the 99 identified Akt substrates, only 18 were shared between the high- and low-dose T cell activation conditions. Thirty-one proteins were unique to high-dose stimulation conditions, and 50 were unique to low-dose stimulation conditions (Fig. 2B). A pathway analysis using the Ingenuity and KEGG databases demonstrated that key processes, including metabolism, TCR signaling, and RNA processing, were targeted by Akt during high- and low-dose stimulation (Fig. 2C). Pathways including glycolysis, Tec kinase signaling, and the TCA cycle were differentially targeted by Akt during high-dose stimulation (Fig. 2D). Interestingly, RNA processing and splicing were overrepresented biological processes targeted by Akt during both high- and low-dose stimulation (Fig. 2E).

Specific proteins targeted by Akt under conditions of Treg or Th induction were distinct (Fig. 3A, Supplemental Tables I, II). Some Akt substrates in both Th and Treg induction included hnRNPs A2B1, enolase, and Hspa8 (Fig. 3A). Akt substrates unique to low-dose stimulation included STAT1, aconitase, hnRNP A1, and hnRNP L. Akt substrates unique to high-dose stimulation included the mitochondrial ADP/ATP carrier (SLC25A5) and LCP-1 (Fig. 3A). Further work will be needed to validate many of these substrates, and we concentrated our efforts on the RNA processing factors hnRNP L and hnRNP A1. Western blot analysis confirmed that both hnRNP L and hnRNP A1 were immunoprecipitated using the Akt substrate motif Ab under low-dose TCR stimulation and not under high-dose TCR stimulation (Fig. 3B, 3C). Western blot analysis for FoxO1 revealed that FoxO1 was immunoprecipitated only under high-dose stimulation consistent with previous studies (7) (Fig. 3B, 3C).

FIGURE 3.

Identification of specific Akt substrates phosphorylated during Th or Treg induction. (A) Some of the Akt substrates identified in low- (left circles) and high-dose (right circles) conditions. Akt substrates that are shared between low- and high-dose conditions are listed at the overlap between the left (low) and right (high) circles. (B) Western blot analysis was performed on the proteins bound to the Akt substrate Ab (beads) versus the amount of protein remaining in solution (Sup.) for hnRNP L, hnRNP A1, and FoxO1 from cells stimulated with high (H) or low (L) dose anti-CD3. (C) Densitometry was performed to determine the ratio of the proteins bound to the Akt substrate Ab (beads) versus the amount of protein remaining in solution from three independent experiments (hnRNP L and hnRNP A1) or two independent experiments (FoxO1). A two-way ANOVA analysis with Bonferroni posttest was performed. ***p < 0.001, ****p < 0.0001.

FIGURE 3.

Identification of specific Akt substrates phosphorylated during Th or Treg induction. (A) Some of the Akt substrates identified in low- (left circles) and high-dose (right circles) conditions. Akt substrates that are shared between low- and high-dose conditions are listed at the overlap between the left (low) and right (high) circles. (B) Western blot analysis was performed on the proteins bound to the Akt substrate Ab (beads) versus the amount of protein remaining in solution (Sup.) for hnRNP L, hnRNP A1, and FoxO1 from cells stimulated with high (H) or low (L) dose anti-CD3. (C) Densitometry was performed to determine the ratio of the proteins bound to the Akt substrate Ab (beads) versus the amount of protein remaining in solution from three independent experiments (hnRNP L and hnRNP A1) or two independent experiments (FoxO1). A two-way ANOVA analysis with Bonferroni posttest was performed. ***p < 0.001, ****p < 0.0001.

Close modal

RNA processing and splicing proteins were a predominant subgroup of Akt substrates in activated T cells (Fig. 3A). A classic example of a protein regulated by alternative splicing in T cells is the CD45 phosphatase (Fig. 4A) (20). Expression of the CD45RB isoform has been used to distinguish Th cell subsets such that CD45RBhigh cells are T effector cells (21), whereas low expression of CD45RB is a marker of Treg cells (22).

FIGURE 4.

Akt regulates the splicing of proteins involved in TCR signaling. (A) Diagram depicting the splicing variants for CD45. (B) Flow cytometric analysis of the expression of the CD45RB isoform on induced Treg cells (iTregs), nTregs, and induced Th cells (iTh). (C) A PCR-based assay was used to monitor the splicing of CD45 under low- or high-dose conditions with and without an iAKT. Results are representative of two independent experiments. (D) The levels of specific splice isoforms of Lck (LCKA or LCKB), (E) Fyn (FynA or FynB), and (F) Tec (Tec_C or Tec_A) were monitored by qPCR at 12 h of high or low TCR stimulation with (+ I) and without the iAKT inhibitor. (G) Naive CD4+ T cells (white) and nTregs (black) were isolated from a Foxp3-RFP reporter mouse, and the levels of Tec isoforms were measured by qPCR. (D–F) Results represent the mean ± SEM of three independent experiments. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ****p < 0.0001.

FIGURE 4.

Akt regulates the splicing of proteins involved in TCR signaling. (A) Diagram depicting the splicing variants for CD45. (B) Flow cytometric analysis of the expression of the CD45RB isoform on induced Treg cells (iTregs), nTregs, and induced Th cells (iTh). (C) A PCR-based assay was used to monitor the splicing of CD45 under low- or high-dose conditions with and without an iAKT. Results are representative of two independent experiments. (D) The levels of specific splice isoforms of Lck (LCKA or LCKB), (E) Fyn (FynA or FynB), and (F) Tec (Tec_C or Tec_A) were monitored by qPCR at 12 h of high or low TCR stimulation with (+ I) and without the iAKT inhibitor. (G) Naive CD4+ T cells (white) and nTregs (black) were isolated from a Foxp3-RFP reporter mouse, and the levels of Tec isoforms were measured by qPCR. (D–F) Results represent the mean ± SEM of three independent experiments. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ****p < 0.0001.

Close modal

hnRNP L and hnRNP A1 regulate splicing of CD45 mRNA (23, 24) and are phosphorylated by Akt only under low-dose conditions (Fig. 3A, 3B). Flow cytometric analysis of CD45RB expression revealed that nTreg cells had the lowest CD45RB abundance (Fig. 4B). Induced Treg cells predominantly expressed low CD45RB levels. However, induced Th cells were mostly CD45RB high (Fig. 4B). PCR analysis revealed that, initially, CD45RO was the predominant isoform expressed by unactivated murine CD4+ T cells (Fig. 4C). Activation under both low- and high-dose conditions resulted in the increase of total CD45 transcription and the production of the CD45RB isoform (Fig. 4C). Pretreatment of the cells with iAkt before activation induced larger isoforms of CD45, including CD45RB (Fig. 4C). In addition, treatment with the iAkt also appeared to increase total CD45. These results support the role of Akt in the regulation CD45 splicing.

Previous CLIPseq data demonstrated that, in human T cells, hnRNP L binds to many mRNA transcripts encoding proteins in the T cell signaling pathway, including CD45, Fyn, Tec, Lck, and CD247 (25). Based on these reports, qPCR-based assays were designed (Table I) to monitor the splicing of mRNA encoding key T cell signaling proteins.

Table I.
Primer sequences used in alternative splicing analysis
IsoformForward PrimerReverse Primer
LCK A 5′-GCCTCTGAGCTGACGATCT-3′ 5′-CACACGTCAATGTTCTCCATCC-3′ 
LCK B 5′-GGCTAGGGAGCATCATGTGAATA-3′ 5′-TCCAGTCATCTTCAGGGTTTG-3′ 
FYN A 5′-CTGTCACAAAGGGATGCCA-3′ 5′-TGATCAACTGCAGGGATTCTC-3′ 
FYN B 5′-TTCATCAAGTTGTACCCCACAAA-3′ 5′-AGCCACACTTCAGCGAAAC-3′ 
TEK A 5′-TCCCAAGTAATTACGTCACAGG-3′ 5′-GCTGTTCTGTTTGCTTCTG-3′ 
TEK C 5′-GATGGTGTCATTCCCTGTCAA-3′ 5′-TGCCCAGAATTTAGGATGGTATT-3′ 
CD3ζ 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-CATATGCAGGGCATCATAGGT-3′ 
CD3θ 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-TCTTATGTCGGCAGGCTTTG-3′ 
CD3η 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-ACATCTCCTTCCTCTCCTGTAG-3′ 
Total CD247 5′-GGA TCC CAA ACT CTG CTA CTT-3′ 5′-CTGTAGGCT TCT GCC ATC TT-3′ 
IsoformForward PrimerReverse Primer
LCK A 5′-GCCTCTGAGCTGACGATCT-3′ 5′-CACACGTCAATGTTCTCCATCC-3′ 
LCK B 5′-GGCTAGGGAGCATCATGTGAATA-3′ 5′-TCCAGTCATCTTCAGGGTTTG-3′ 
FYN A 5′-CTGTCACAAAGGGATGCCA-3′ 5′-TGATCAACTGCAGGGATTCTC-3′ 
FYN B 5′-TTCATCAAGTTGTACCCCACAAA-3′ 5′-AGCCACACTTCAGCGAAAC-3′ 
TEK A 5′-TCCCAAGTAATTACGTCACAGG-3′ 5′-GCTGTTCTGTTTGCTTCTG-3′ 
TEK C 5′-GATGGTGTCATTCCCTGTCAA-3′ 5′-TGCCCAGAATTTAGGATGGTATT-3′ 
CD3ζ 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-CATATGCAGGGCATCATAGGT-3′ 
CD3θ 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-TCTTATGTCGGCAGGCTTTG-3′ 
CD3η 5′-CAGAAAGACAAGATGGCAGAAG-3′ 5′-ACATCTCCTTCCTCTCCTGTAG-3′ 
Total CD247 5′-GGA TCC CAA ACT CTG CTA CTT-3′ 5′-CTGTAGGCT TCT GCC ATC TT-3′ 

Alternatively spliced variants of Lck, Tec kinase, and Fyn (Supplemental Fig. 1) were examined. A 2-fold increase in the expression of the Lck-B (26) and FynA (27) isoforms was observed in T cells stimulated with low-dose anti-CD3 for 24 h (Fig. 4D, 4E). The increase in both the Lck-B and FynA isoforms (Supplemental Fig. 1A, 1B) was Akt dependent because the presence of iAkt inhibited the observed increase (Fig. 4D, 4E). Tec kinase, which regulates IL-2 and IL-4 gene transcription (28), also underwent alternative splicing during T cell differentiation. The PH domain is critical for Tec activity (29), and splicing generates an isoform, Tec_C, that lacks the PH domain (Supplemental Fig. 1C) (30). During low-dose stimulation, the levels of Tec_C increased >20-fold but remained constant during high-dose stimulation (Fig. 4F). These changes were Akt dependent because cells pretreated with iAkt failed to increase expression of the Tec_C isoform under low-dose conditions (Fig. 4F). To determine whether Akt-regulated splicing of Tec kinase induced by T cell activation was maintained in fully differentiated Treg cells, we sorted naive CD4+ T cells and nTreg cells from the spleen of Foxp3 reporter mice and examined the expression of Tec isoforms. There was no observed difference in expression of Tec isoforms between naive T cells and nTreg cells (Fig. 4G).

Another target for alternative splicing is CD247, which has a key function in the TCR/CD3 complex (31, 32). There are four splice variants of CD247 (33, 34), which include CD3η, CD3ζ, CD3θ, and CD3ι. Both CD3η and CD3θ lack the third intracellular ITAM, which could modulate signaling (Fig. 5A) (31, 32). The CD3ζ, CD3θ, and CD3η isoforms of CD247 were detectable during both low- and high-dose stimulation (Fig. 5B, 5D). During high-dose stimulation, CD3ζ mRNA expression increased 20-fold over the initial level, and this increase was Akt dependent (Fig. 5B–D). Conversely, during low-dose stimulation, CD3η and CD3θ mRNA expression increased by 20- and 4-fold, respectively (Fig. 5D).

FIGURE 5.

Altered CD247 splicing regulates TCR signaling. (A) CD247, more commonly referred to as CD3ζ, has four splice variants. (BD) Purified CD4+ T cells were activated under high- and low-dose conditions, and qPCR was used to monitor the levels of CD247 isoforms [η (B), θ (C), and ζ (D)] during differentiation. (D) Data are also presented for CD4+ T cells skewed under Th conditions and treated with iAkt (Th + I). Results shown are normalized to the total level of CD247 mRNA and represent the mean ± SEM of three independent experiments. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ****p < 0.0001. (E) Purified CD4+ T cells were stimulated with low- (L) or high (H)-dose stimulation for the indicated times. Western blotting was performed for total tyrosine phosphorylation. (F) Purified CD4+ T cells were activated for 38 h under Treg (0.25 μg/ml anti-CD3) or Th (1 μg/ml anti-CD3) skewing conditions, rested for 4 h, and restimulated with 1 μg/ml of plate-bound anti-CD3 Ab and 1 μg/ml soluble anti-CD28 Ab. Western blot analysis was performed for total tyrosine phosphorylation and actin at the indicated time points. The blot shown is representative of three independent experiments. (G) Naive CD4+ T cells and nTregs were isolated from a Foxp3-RFP reporter mouse, and the levels of the CD247 isoforms were measured by qPCR. The results shown are representative of two independent experiments.

FIGURE 5.

Altered CD247 splicing regulates TCR signaling. (A) CD247, more commonly referred to as CD3ζ, has four splice variants. (BD) Purified CD4+ T cells were activated under high- and low-dose conditions, and qPCR was used to monitor the levels of CD247 isoforms [η (B), θ (C), and ζ (D)] during differentiation. (D) Data are also presented for CD4+ T cells skewed under Th conditions and treated with iAkt (Th + I). Results shown are normalized to the total level of CD247 mRNA and represent the mean ± SEM of three independent experiments. A two-way ANOVA analysis with Bonferroni posttest was performed. *p < 0.05, **p < 0.01, ****p < 0.0001. (E) Purified CD4+ T cells were stimulated with low- (L) or high (H)-dose stimulation for the indicated times. Western blotting was performed for total tyrosine phosphorylation. (F) Purified CD4+ T cells were activated for 38 h under Treg (0.25 μg/ml anti-CD3) or Th (1 μg/ml anti-CD3) skewing conditions, rested for 4 h, and restimulated with 1 μg/ml of plate-bound anti-CD3 Ab and 1 μg/ml soluble anti-CD28 Ab. Western blot analysis was performed for total tyrosine phosphorylation and actin at the indicated time points. The blot shown is representative of three independent experiments. (G) Naive CD4+ T cells and nTregs were isolated from a Foxp3-RFP reporter mouse, and the levels of the CD247 isoforms were measured by qPCR. The results shown are representative of two independent experiments.

Close modal

Altered splice variant usage of signaling molecules may be one of the factors that alter the threshold required for T cell activation between Th cells compared with Treg cells (26, 27). To test whether signaling was altered during Treg differentiation, we activated T cells under Th (1 μg/ml anti-CD3) or Treg (0.25 μg/ml anti-CD3) conditions for 36 h; cells were rested for 4 h and restimulated with a high dose of anti-CD3. During the initial stimulation, both the Treg and Th induction conditions resulted in similar tyrosine phosphorylation at early time points, although this was poorly sustained with low-dose anti-CD3 (Fig. 5E). Upon restimulation with high dose, anti-CD3 cells destined to become Treg cells exhibited a much-reduced level of total tyrosine phosphorylation compared with cells destined to become Th cells (Fig. 5F).

To determine whether Akt-regulated splicing induced by T cell activation was maintained in fully differentiated Treg cells, we sorted naive CD4+ T cells and nTreg cells from the spleen of Foxp3 reporter mice and examined the expression of CD247 isoforms. Naive T cells predominantly express the CD3ζ isoform, whereas nTreg cells preferentially express the CD3θ isoform and little CD3ζ (Fig. 5G).

hnRNP proteins influence alternative splicing in T cells, including hnRNP L and hnRNP A1 (35). In humans, hnRNP A1 synergizes with hnRNP L to block splicing of exons 4, 5, and 6 in CD45 (24). hnRNP L and hnRNP A1 were both phosphorylated by Akt during low-dose stimulation (Fig. 3A, 3B), suggesting that their activity could be important for Treg differentiation. We used an siRNA approach to knock down hnRNP A1 and hnRNP L to examine their roles in T cell differentiation. The siRNA knockdown resulted in 1- and 200-fold reduction in the expression of hnRNP A1 and hnRNP L mRNAs (Fig. 6A) and a 7- to 15-fold reduction in hnRNP A1 and hnRNP L protein (Fig. 6B, 6C). Knockdown of hnRNP A1 in resting CD4+ T cells resulted in the production of larger CD45 isoforms compared with the scrambled control (Fig. 6D), similar to that observed when the enzymatic activity of Akt was inhibited (Fig. 4C). Notably, the CD45RB isoform is more abundant with knockdown of hnRNP A1.

FIGURE 6.

hnRNP A1 or hnRNP L are required for optimal Treg induction. (A) The mRNA expression levels of hnRNP L and hnRNP A1 were measured by qPCR 24 h after electroporation with the respective siRNAs. (B and C) The protein abundance was measured by Western blotting (B), and the protein levels were quantified by densitometry from two independent experiments (C). (D) Knockdown of hnRNP A1 resulted in altered CD45 splicing in resting CD4+ T cells relative to a scrambled control, as monitored by PCR analysis of exons 4–6 and gel electrophoresis 24 h after electroporation. (EG) CD4+ T cells were electroporated with siRNA targeting hnRNP A1 or hnRNP L, rested for 24 h, and activated with either a low or high TCR signal. (E) At 12 h postactivation, flow cytometry was used to monitor the expression of CD45RB and phosphorylation of S6 under Th and Treg induction conditions in cells treated with the scrambled or hnRNP A1 targeting siRNA. (F and G) Cells treated with scrambled, hnRNP A1 (F), or hnRNP L (G) siRNA were gated on CD3+CD4+ T cells. The CD25 and Foxp3 markers were used to track the induction of Th (CD3+CD4+Foxp3CD25+) and Treg (CD3+CD4+Foxp3+CD25+) cells. (H) The percentage of CD4+ Treg or Th cells was plotted averaged over four independent experiments. A two-way ANOVA analysis with Tukey posttest was performed. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 6.

hnRNP A1 or hnRNP L are required for optimal Treg induction. (A) The mRNA expression levels of hnRNP L and hnRNP A1 were measured by qPCR 24 h after electroporation with the respective siRNAs. (B and C) The protein abundance was measured by Western blotting (B), and the protein levels were quantified by densitometry from two independent experiments (C). (D) Knockdown of hnRNP A1 resulted in altered CD45 splicing in resting CD4+ T cells relative to a scrambled control, as monitored by PCR analysis of exons 4–6 and gel electrophoresis 24 h after electroporation. (EG) CD4+ T cells were electroporated with siRNA targeting hnRNP A1 or hnRNP L, rested for 24 h, and activated with either a low or high TCR signal. (E) At 12 h postactivation, flow cytometry was used to monitor the expression of CD45RB and phosphorylation of S6 under Th and Treg induction conditions in cells treated with the scrambled or hnRNP A1 targeting siRNA. (F and G) Cells treated with scrambled, hnRNP A1 (F), or hnRNP L (G) siRNA were gated on CD3+CD4+ T cells. The CD25 and Foxp3 markers were used to track the induction of Th (CD3+CD4+Foxp3CD25+) and Treg (CD3+CD4+Foxp3+CD25+) cells. (H) The percentage of CD4+ Treg or Th cells was plotted averaged over four independent experiments. A two-way ANOVA analysis with Tukey posttest was performed. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To determine the function of hnRNP A1 or hnRNP L in CD4+ T cell differentiation, we treated cells with hnRNP A1 or hnRNP L siRNA for 24 h and activated them under low- and high-dose conditions. At 12 h postactivation in cells treated with hnRNP A1 siRNA, CD45RB was elevated in both high- and low-dose stimulation (Fig. 6E). In addition, knockdown of hnRNP A1 resulted in increased phosphorylation of S6 under both high- and low-dose stimulation (Fig. 6E). Phospho-S6 abundance inversely correlated to the induction of Treg cells (3, 15), suggesting that hnRNP A1 is important for Treg induction.

The impact of hnRNP A1 or hnRNP L knockdown on CD4+ T cell differentiation was analyzed. At 48 h postactivation in high-dose conditions, similar numbers of Th cells were observed in cells treated with hnRNP A1 and the scrambled siRNA (Fig. 6F, 6H). Under low-dose conditions, hnRNP A1 knockdown resulted in >60% of the cells becoming Th versus 21% in the control sample. In addition, hnRNP A1 knockdown resulted in significantly decreased induction of Treg cells relative to the control. Similarly, hnRNP L knockdown also resulted in significantly increased Th cell generation and decreased Treg cell induction (Fig. 6G, 6H).

In this article, we identified distinct Akt substrates during the differentiation of CD4+ T cells into Th or Treg cell subsets. Alternate signaling networks were associated with differential regulation of the phosphorylation status of Akt. Of interest was the regulation of splicing factors by Akt, and we observed that the splicing of multiple mRNAs encoding key TCR signaling molecules was Akt dependent. Two proteins known to regulate splicing in T cells, hnRNP A1 and hnRNP L, were Akt substrates during Treg induction. Knockdown of either hnRNP A1 or hnRNP L suppressed Treg induction by elevating Akt/mTOR signaling. Together, this work provides mechanistic insight into how Akt initiates different differentiation programs to drive alternate CD4+ T cell fate decisions.

Previous work correlated the phosphorylation status of Akt with regulating its enzymatic activity (36). Thr308 is phosphorylated by PDK1 (9), which is directly activated by PIP3 downstream of TCR signaling, and is dephosphorylated by protein phosphatase 2A (37). Ser473 is phosphorylated by mTORC2 (16) and dephosphorylated by PHLPP (38). For maximal enzymatic activity, it is thought that Akt should be phosphorylated at both Thr308 and Ser473, and that if only one of the sites is phosphorylated, enzymatic activity is much reduced (36). The phosphorylation of FoxO1 by Akt requires that both Thr308 and Ser473 be phosphorylated on Akt (4, 7, 39). Intriguingly, human CD4+ CD25+ T cells stimulated with anti-CD3 are defective in phosphorylation of the Ser473 site on Akt, whereas Thr308 is robustly phosphorylated. Introduction of an acidic residue at the Ser473 site resulted in constitutive activation of Akt in human Treg cells, which abrogated their suppressive activity and restored the ability of Akt to exclude FoxO1 from the nucleus (4).

Low-dose stimulation was marked with suppressed phosphorylation of Ser473 on Akt and normal levels of Thr308 phosphorylation, whereas high-dose stimulation induced robust phosphorylation of both Ser473 and Thr308. The ability of mTORC2 to phosphorylate Ser473 is controlled by Thr1135. This inhibitory site, Thr1135, on the Rictor subunit of mTORC2 remained phosphorylated under low-dose stimulation. Conversely, high-dose stimulation resulted in dephosphorylation of Thr1135, which correlated with elevated levels of Ser473 phosphorylation. The mechanism of mTORC2 activation in T cells is not well established. One possibility is that TCR signaling generates PIP3 to activate mTORC2, which was observed in HEK293 T cells (40). In this model, PIP3 would be higher during high-dose stimulation to activate mTORC2 and lower in low-dose conditions to weakly activate mTORC2. This would allow for elevated PIP3 levels, heightened mTORC2 activity, and increased phosphorylation of Akt Ser473.

A surprising result is the essential role of Akt-dependent mRNA splicing in CD4+ T cell differentiation. Notably, Treg induction resulted in the reduction of the CD3ζ isoform of CD247 and the emergence of the CD3η and CD3θ isoforms. Both the CD3η and CD3θ isoforms lack the third ITAM domain, which could impact downstream signaling events. T cells with more of the CD3η and CD3θ isoforms exhibited reduced early TCR signaling, suggesting that splicing of CD247 is used to tune the level of TCR signaling during T cell differentiation. Previous studies demonstrated that the three CD3ζ ITAMs are critical for optimal T cell activation (41). In addition, defects in CD3ζ signaling correlated with disease pathogenesis in patients with systemic lupus erythematosus (42). In nTreg cells, the CD3θ isoform dominated, suggesting that the alternative splicing observed during Treg induction remains after differentiation. This may provide an explanation for the reduced TCR signaling in Treg cells, which was recently investigated (43). This study examined the role of several Treg-specific negative regulators, such as CTLA-4, PD-1, CD5, and DUSP4, but these were not found to be responsible for the reduced signaling in Treg cells (43). Together, these results suggest that splicing regulated by Akt functions to rewire the canonical TCR signaling pathway to drive alternate CD4+ T cell differentiation programs.

During low-dose stimulation, both hnRNP A1 and hnRNP L were phosphorylated by Akt (44, 45). There is a defined Akt phosphorylation site on hnRNP A1 at Ser199, and this has been shown to influence hnRNP A1 function in cell lines (44). In the case of hnRNP L, there are two potential Akt phosphorylation sites, and a recent study showed that Akt phosphorylation of hnRNP L influenced the alternative splicing of caspase 9 in cell lines (45). Knockdown of either hnRNP A1 or hnRNP L reduced the capacity of CD4+ T cells to differentiate into Treg cells and favored Th induction. A recent study of LckCre+hnRNP Lfl/fl mice demonstrated a striking defect in peripheral T cell numbers that was associated with defects in thymocyte migration to the periphery (46). This study identified multiple changes in alternative splicing of proteins involved in T cell signaling and chemotaxis (46), and showed an increase in splicing to the CD45RB isoform, which we also observed in our study. Although it is well accepted that induction of transcription factors including Foxp3 drives T cell differentiation programs, our results suggest that splicing is a key driver of differentiation. HnRNP L binds >2000 mRNAs, and it is likely that regulation of splicing factors by Akt results in different complements of protein isoforms during Th and Treg differentiation.

Work presented in this article identified novel Akt substrates differentially phosphorylated during the induction of Th and Treg cells from naive CD4+ cells and demonstrated that Akt activity was functionally important for regulating splicing programs. Future work will be aimed at validating the novel Akt substrates identified in this study. The phosphorylation status of Akt was different based on whether T cells were stimulated with high or low dose, and this influenced Akt activity and substrate specificity. In conclusion, this work provides novel insight into CD4+ T cell differentiation by identifying unique Akt substrates during Treg and Th induction, defining how the phosphorylation status of Akt regulates substrate specificity, and identifying that Akt-mediated phosphorylation of RNA processing factors plays a major role in determining CD4+ T cell fate.

We thank Matt Gable for technical assistance. We thank Dr. Greg Delgoffe and Dr. Larry Kane for careful reading of the manuscript.

This work was supported by National Institutes of Health Training Grant T32 AI089443 (to W.F.H.) and by a Competitive Medical Research Fund grant from the University of Pittsburgh Medical Center (to W.F.H.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

hnRNP

heterogeneous nuclear ribonucleoprotein

iAkt

Akt inhibitor

IP

immunoprecipitate

nTreg

natural Treg

PTEN

phosphatase and tensin homolog

QE

Q-Exactive

qPCR

quantitative PCR

siRNA

small interfering RNA

Treg

regulatory T.

1
Haxhinasto
,
S.
,
D.
Mathis
,
C.
Benoist
.
2008
.
The AKT-mTOR axis regulates de novo differentiation of CD4+Foxp3+ cells.
J. Exp. Med.
205
:
565
574
.
2
Sauer
,
S.
,
L.
Bruno
,
A.
Hertweck
,
D.
Finlay
,
M.
Leleu
,
M.
Spivakov
,
Z. A.
Knight
,
B. S.
Cobb
,
D.
Cantrell
,
E.
O’Connor
, et al
.
2008
.
T cell receptor signaling controls Foxp3 expression via PI3K, Akt, and mTOR.
Proc. Natl. Acad. Sci. USA
105
:
7797
7802
.
3
Turner
,
M. S.
,
L. P.
Kane
,
P. A.
Morel
.
2009
.
Dominant role of antigen dose in CD4+Foxp3+ regulatory T cell induction and expansion.
J. Immunol.
183
:
4895
4903
.
4
Crellin
,
N. K.
,
R. V.
Garcia
,
M. K.
Levings
.
2007
.
Altered activation of AKT is required for the suppressive function of human CD4+CD25+ T regulatory cells.
Blood
109
:
2014
2022
.
5
Huynh
,
A.
,
M.
DuPage
,
B.
Priyadharshini
,
P. T.
Sage
,
J.
Quiros
,
C. M.
Borges
,
N.
Townamchai
,
V. A.
Gerriets
,
J. C.
Rathmell
,
A. H.
Sharpe
, et al
.
2015
.
Control of PI(3) kinase in Treg cells maintains homeostasis and lineage stability.
Nat. Immunol.
16
:
188
196
.
6
Shrestha
,
S.
,
K.
Yang
,
C.
Guy
,
P.
Vogel
,
G.
Neale
,
H.
Chi
.
2015
.
Treg cells require the phosphatase PTEN to restrain TH1 and TFH cell responses.
Nat. Immunol.
16
:
178
187
.
7
Hawse
,
W. F.
,
R. P.
Sheehan
,
N.
Miskov-Zivanov
,
A. V.
Menk
,
L. P.
Kane
,
J. R.
Faeder
,
P. A.
Morel
.
2015
.
Cutting edge: differential regulation of PTEN by TCR, Akt, and FoxO1 controls CD4+ T cell fate decisions.
J. Immunol.
194
:
4615
4619
.
8
Miskov-Zivanov
,
N.
,
M. S.
Turner
,
L. P.
Kane
,
P. A.
Morel
,
J. R.
Faeder
.
2013
.
The duration of T cell stimulation is a critical determinant of cell fate and plasticity.
Sci. Signal.
6
:
ra97
.
9
Alessi
,
D. R.
,
S. R.
James
,
C. P.
Downes
,
A. B.
Holmes
,
P. R.
Gaffney
,
C. B.
Reese
,
P.
Cohen
.
1997
.
Characterization of a 3-phosphoinositide-dependent protein kinase which phosphorylates and activates protein kinase Balpha.
Curr. Biol.
7
:
261
269
.
10
Sarbassov
,
D. D.
,
S. M.
Ali
,
D. M.
Sabatini
.
2005
.
Growing roles for the mTOR pathway.
Curr. Opin. Cell Biol.
17
:
596
603
.
11
Vasudevan
,
K. M.
,
L. A.
Garraway
.
2010
.
AKT signaling in physiology and disease.
Curr. Top. Microbiol. Immunol.
347
:
105
133
.
12
Wiśniewski
,
J. R.
,
A.
Zougman
,
N.
Nagaraj
,
M.
Mann
.
2009
.
Universal sample preparation method for proteome analysis.
Nat. Methods
6
:
359
362
.
13
Mi
,
H.
,
A.
Muruganujan
,
J. T.
Casagrande
,
P. D.
Thomas
.
2013
.
Large-scale gene function analysis with the PANTHER classification system.
Nat. Protoc.
8
:
1551
1566
.
14
Wu
,
Z.
,
X.
Jia
,
L.
de la Cruz
,
X.-C.
Su
,
B.
Marzolf
,
P.
Troisch
,
D.
Zak
,
A.
Hamilton
,
B.
Whittle
,
D.
Yu
, et al
.
2008
.
Memory T cell RNA rearrangement programmed by heterogeneous nuclear ribonucleoprotein hnRNPLL.
Immunity
29
:
863
875
.
15
Turner
,
M. S.
,
K.
Isse
,
D. K.
Fischer
,
H. R.
Turnquist
,
P. A.
Morel
.
2014
.
Low TCR signal strength induces combined expansion of Th2 and regulatory T cell populations that protect mice from the development of type 1 diabetes.
Diabetologia
57
:
1428
1436
.
16
Sarbassov
,
D. D.
,
D. A.
Guertin
,
S. M.
Ali
,
D. M.
Sabatini
.
2005
.
Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex.
Science
307
:
1098
1101
.
17
Dibble
,
C. C.
,
J. M.
Asara
,
B. D.
Manning
.
2009
.
Characterization of Rictor phosphorylation sites reveals direct regulation of mTOR complex 2 by S6K1.
Mol. Cell. Biol.
29
:
5657
5670
.
18
Moritz
,
A.
,
Y.
Li
,
A.
Guo
,
J.
Villén
,
Y.
Wang
,
J.
MacNeill
,
J.
Kornhauser
,
K.
Sprott
,
J.
Zhou
,
A.
Possemato
, et al
.
2010
.
Akt-RSK-S6 kinase signaling networks activated by oncogenic receptor tyrosine kinases.
Sci. Signal.
3
:
ra64
.
19
Obenauer
,
J. C.
,
L. C.
Cantley
,
M. B.
Yaffe
.
2003
.
Scansite 2.0: proteome-wide prediction of cell signaling interactions using short sequence motifs.
Nucleic Acids Res.
31
:
3635
3641
.
20
Hermiston
,
M. L.
,
Z.
Xu
,
A.
Weiss
.
2003
.
CD45: a critical regulator of signaling thresholds in immune cells.
Annu. Rev. Immunol.
21
:
107
137
.
21
Morrissey
,
P. J.
,
K.
Charrier
,
S.
Braddy
,
D.
Liggitt
,
J. D.
Watson
.
1993
.
CD4+ T cells that express high levels of CD45RB induce wasting disease when transferred into congenic severe combined immunodeficient mice. Disease development is prevented by cotransfer of purified CD4+ T cells.
J. Exp. Med.
178
:
237
244
.
22
Powrie
,
F.
,
J.
Carlino
,
M. W.
Leach
,
S.
Mauze
,
R. L.
Coffman
.
1996
.
A critical role for transforming growth factor-beta but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RB(low) CD4+ T cells.
J. Exp. Med.
183
:
2669
2674
.
23
Tong
,
A.
,
J.
Nguyen
,
K. W.
Lynch
.
2005
.
Differential expression of CD45 isoforms is controlled by the combined activity of basal and inducible splicing-regulatory elements in each of the variable exons.
J. Biol. Chem.
280
:
38297
38304
.
24
Chiou
,
N. T.
,
G.
Shankarling
,
K. W.
Lynch
.
2013
.
hnRNP L and hnRNP A1 induce extended U1 snRNA interactions with an exon to repress spliceosome assembly.
Mol. Cell
49
:
972
982
.
25
Shankarling
,
G.
,
B. S.
Cole
,
M. J.
Mallory
,
K. W.
Lynch
.
2014
.
Transcriptome-wide RNA interaction profiling reveals physical and functional targets of hnRNP L in human T cells.
Mol. Cell. Biol.
34
:
71
83
.
26
Rouer
,
E.
,
R.
Benarous
.
1992
.
Alternative splicing in the human lck gene leads to the deletion of exon 1′ and results in a new type II lck transcript.
Oncogene
7
:
2535
2538
.
27
Brignatz
,
C.
,
M. P.
Paronetto
,
S.
Opi
,
M.
Cappellari
,
S.
Audebert
,
V.
Feuillet
,
G.
Bismuth
,
S.
Roche
,
S. T.
Arold
,
C.
Sette
,
Y.
Collette
.
2009
.
Alternative splicing modulates autoinhibition and SH3 accessibility in the Src kinase Fyn.
Mol. Cell. Biol.
29
:
6438
6448
.
28
Yang
,
W. C.
,
D.
Olive
.
1999
.
Tec kinase is involved in transcriptional regulation of IL-2 and IL-4 in the CD28 pathway.
Eur. J. Immunol.
29
:
1842
1849
.
29
Yang
,
W.-C.
,
K. A.
Ching
,
C. D.
Tsoukas
,
L. J.
Berg
.
2001
.
Tec kinase signaling in T cells is regulated by phosphatidylinositol 3-kinase and the Tec pleckstrin homology domain.
J. Immunol.
166
:
387
395
.
30
Mano
,
H.
,
Y.
Yamashita
,
K.
Sato
,
Y.
Yazaki
,
H.
Hirai
.
1995
.
Tec protein-tyrosine kinase is involved in interleukin-3 signaling pathway.
Blood
85
:
343
350
.
31
Clayton
,
L. K.
,
A. C.
Diener
,
A.
Lerner
,
A. G.
Tse
,
S.
Koyasu
,
E. L.
Reinherz
.
1992
.
Differential regulation of T-cell receptor processing and surface expression affected by CD3 theta, an alternatively spliced product of the CD3 zeta/eta gene locus.
J. Biol. Chem.
267
:
26023
26030
.
32
Moulton
,
V. R.
,
G. C.
Tsokos
.
2010
.
Alternative splicing factor/splicing factor 2 regulates the expression of the ζ subunit of the human T cell receptor-associated CD3 complex.
J. Biol. Chem.
285
:
12490
12496
.
33
Lerner
,
A.
,
L.
D’Adamio
,
A. C.
Diener
,
L. K.
Clayton
,
E. L.
Reinherz
.
1993
.
CD3 zeta/eta/theta locus is colinear with and transcribed antisense to the gene encoding the transcription factor Oct-1.
J. Immunol.
151
:
3152
3162
.
34
Nocentini
,
G.
,
S.
Ronchetti
,
A.
Bartoli
,
G.
Testa
,
F.
D’Adamio
,
C.
Riccardi
,
G.
Migliorati
.
1995
.
T cell receptor iota an alternatively spliced product of the T cell receptor zeta gene.
Eur. J. Immunol.
25
:
1405
1409
.
35
Rothrock
,
C. R.
,
A. E.
House
,
K. W.
Lynch
.
2005
.
HnRNP L represses exon splicing via a regulated exonic splicing silencer.
EMBO J.
24
:
2792
2802
.
36
Alessi
,
D. R.
,
M.
Andjelkovic
,
B.
Caudwell
,
P.
Cron
,
N.
Morrice
,
P.
Cohen
,
B. A.
Hemmings
.
1996
.
Mechanism of activation of protein kinase B by insulin and IGF-1.
EMBO J.
15
:
6541
6551
.
37
Resjö
,
S.
,
O.
Göransson
,
L.
Härndahl
,
S.
Zolnierowicz
,
V.
Manganiello
,
E.
Degerman
.
2002
.
Protein phosphatase 2A is the main phosphatase involved in the regulation of protein kinase B in rat adipocytes.
Cell. Signal.
14
:
231
238
.
38
Gao
,
T.
,
F.
Furnari
,
A. C.
Newton
.
2005
.
PHLPP: a phosphatase that directly dephosphorylates Akt, promotes apoptosis, and suppresses tumor growth.
Mol. Cell
18
:
13
24
.
39
Bhaskar
,
P. T.
,
N.
Hay
.
2007
.
The two TORCs and Akt.
Dev. Cell
12
:
487
502
.
40
Gan
,
X.
,
J.
Wang
,
B.
Su
,
D.
Wu
.
2011
.
Evidence for direct activation of mTORC2 kinase activity by phosphatidylinositol 3,4,5-trisphosphate.
J. Biol. Chem.
286
:
10998
11002
.
41
van Oers
,
N. S. C.
,
P. E.
Love
,
E. W.
Shores
,
A.
Weiss
.
1998
.
Regulation of TCR signal transduction in murine thymocytes by multiple TCR ζ-chain signaling motifs.
J. Immunol.
160
:
163
170
.
42
Moulton
,
V. R.
,
G. C.
Tsokos
.
2015
.
T cell signaling abnormalities contribute to aberrant immune cell function and autoimmunity.
J. Clin. Invest.
125
:
2220
2227
.
43
Yan
,
D.
,
J.
Farache
,
M.
Mingueneau
,
D.
Mathis
,
C.
Benoist
.
2015
.
Imbalanced signal transduction in regulatory T cells expressing the transcription factor FoxP3.
[Published erratum appears in 2016 Proc. Natl. Acad. Sci. USA 113: E256.].
Proc. Natl. Acad. Sci. USA
112
:
14942
14947
.
44
Jo
,
O. D.
,
J.
Martin
,
A.
Bernath
,
J.
Masri
,
A.
Lichtenstein
,
J.
Gera
.
2008
.
Heterogeneous nuclear ribonucleoprotein A1 regulates cyclin D1 and c-myc internal ribosome entry site function through Akt signaling.
J. Biol. Chem.
283
:
23274
23287
.
45
Vu
,
N. T.
,
M. A.
Park
,
J. C.
Shultz
,
R. W.
Goehe
,
L. A.
Hoeferlin
,
M. D.
Shultz
,
S. A.
Smith
,
K. W.
Lynch
,
C. E.
Chalfant
.
2013
.
hnRNP U enhances caspase-9 splicing and is modulated by AKT-dependent phosphorylation of hnRNP L.
J. Biol. Chem.
288
:
8575
8584
.
46
Gaudreau
,
M.-C.
,
F.
Heyd
,
R.
Bastien
,
B.
Wilhelm
,
T.
Möröy
.
2012
.
Alternative splicing controlled by heterogeneous nuclear ribonucleoprotein L regulates development, proliferation, and migration of thymic pre-T cells.
J. Immunol.
188
:
5377
5388
.

The authors have no financial conflicts of interest.

Supplementary data