Conditional deletion of PTEN in peripheral T cells augments TCR-mediated activation but does not abrogate CD28 dependency or prevent anergy induction

Abstract

PTEN is thought to play a critical role in T cell activation by negatively regulating the PI3K signaling pathway important for cellular activation, growth, and proliferation. To directly eliminate PTEN in post-thymic T cells for studies of functional effects, we utilized CAR Tg x PTEN^flox/flox mice which enabled gene deletion using a Cre adenovirus in vitro. These mice were also immunized to generate stable Th1 clones that could have PTEN deleted when desired. PTEN-deleted T cells exhibited enhanced IL-2 production, proliferation, and Akt phosphorylation upon TCR/CD28 engagement, whereas T cell survival was not potentiated. Gene expression profiling revealed a small subset of induced genes that were augmented upon PTEN deletion. However, PTEN-deficient T cells still required CD28 costimulation for IL-2 production and remained susceptible to anti-CD3-induced anergy. The absence of PTEN within the CD8 T cell compartment led to markedly increased cytolytic activity following an allogeneic MLR in vitro, without increasing autologous MLR activity. Our results indicate that deletion of PTEN can augment the activation of post-thymic T cells but does not mediate CD28-independence or anergy resistance. Nonetheless, PTEN inhibition may be a viable target for immune potentiation due to increased cytokine production by activated CD4⁺ cells, and increased cytotoxicity by CD8⁺ T cells.

Introduction

PTEN protein (phosphatase and tensin homolog) is a phosphatase that plays a key role in the regulation of cellular survival and proliferation. Implicated as a tumor suppressor gene, PTEN gene loss leads to augmented cell survival (1) and it is frequently mutated or epigenetically silenced in hereditary and sporadic cancer, including T cell acute lymphoblastic leukemia/lymphoma (2). PTEN acts by dephosphorylating PIP3 to generate PIP2 and thus negatively regulates the PI3 kinase signaling pathway. The PI3 pathway is critical for cell growth, survival, and motility signaling in numerous cell types (3). Within the T cell lineage, PTEN has been reported to negatively regulate TCR and CD28 signaling, is up-regulated upon activation as a negative feedback mechanism (4), plays a role in CD4 and CD8 T cell development (5–7), affects regulatory T cell development (8), and appears to be involved in PD-1 and CTLA-4 inhibitory signaling (9).

Because of alterations in thymic development that can occur when signaling molecules are conditionally deleted using Lck-Cre or CD4-Cre Tg mice (10, 11) it has become desirable to develop strategies to delete genes directly in post-thymic T cells, to determine functional effects directly within the peripheral T cell compartment without disturbing thymic selection. We have recently developed such a method by crossing mice transgenic for the Coxsackie and adenovirus receptor (CAR) in the T cell lineage with mice bearing LoxP-targeted gene alleles, enabling specific gene deletion using a Cre adenovirus in vitro (12). Using this strategy in the current study, we have investigated the functional effects of PTEN deletion in primary T cells and Th1 clones. We find that PTEN deletion does lead to a decreased TCR signaling threshold for T cell activation, augments cytokine production, and allows for increased CTL activity in vitro. However, deletion of PTEN in peripheral T cells did not abrogate the need for CD28 and did not prevent anergy induction.

Materials and Methods

Mice and T cells

PTENflox/flox mice were a gift from Dr. Tak Mak of the Ontario Cancer Institute (10) and were crossed with Coxsackie and adenovirus receptor transgenic (CAR Tg) mice expressing the extracellular domain of the CAR under control of an Lck promoter/CD2 enhancer (13). The resultant C57BL6/CAR-Tg/PTEN^flox/flox mice (or CAR Tg x PTEN^flox/flox), were homozygous for the PTEN/loxp sequence. All mice were maintained under specific pathogen-free conditions in a barrier facility at the University of Chicago according to approved protocols and NIH guidelines. The ovalbumin (OVA)-specific CAR Tg x PTEN^flox/flox Th1 clone was previously described (12). T cell clones were maintained by weekly passage with OVA, IL-2 and syngeneic APCs (irradiated B6 splenocytes) as reported (14). Unless otherwise noted, T cells were cultured in complete DMEM media supplemented with 10% FCS (5% FCS for Th1 clones), penicillin, streptomycin, MOPS, 2-ME, and nonessential amino acids in an 8% CO2 incubator at 37°C.

Adenoviral transduction of CAR Tg T cells

The generation of the adenoviral vectors containing the gene expression unit of cre recombinase (adeno-Cre) or without a coding cDNA (adeno-EV) and the protocol for transduction of peripheral T cells and Th1 clones was previously described (12). For adenoviral transduction, peripheral CAR Tg x PTEN^flox/flox T cells (total, CD4⁺, or CD8⁺) were isolated from splenocytes by negative selection with MACS antibody cocktails and magnetic beads (Miltenyi Biotec). Transduced CAR Tg x PTEN^flox/flox Th1 clones were rested overnight and then passaged under normal conditions, and 9 days later clones were harvested for experiments. Primary T cells were transduced, rested overnight, then cultured 8 days at 10⁶ cells/ml in complete medium. As naïve T cells are kept alive in vivo through IL-7 signaling, we supplemented the media with 1ng/ml of IL-7 (R&D Systems) to prevent primary mouse T cells from dying precipitously in vitro (15). This allows time for gene deletion before being used for further experiments, as we have described previously (12). For primary T cell experiments we considered the possibility that PTEN-deletion might impact on IL-7 signaling and skew the population of T cells that survive. We conducted control experiments contrasting adeno-EV or adeno-Cre treated splenic T cells immediately after transduction against those after the week long rest with IL-7 to allow for gene deletion and PTEN protein degradation. We found no skewing of the population toward increased CD4 or CD8 numbers. PTEN deletion also had no substantial effect on CD127 surface expression; did not alter central memory, effector, or naïve cell phenotypes; and did not yield a different number of T cells after IL-7 co-culture (data not shown).

At the start of all experiments, 1×10⁶ cells were selected from each experimental condition (Adeno-Cre or Adeno-EV treated CAR Tg x PTEN^flox/flox T cells) and lysed for western blot analysis of the PTEN protein, thus confirming Cre-mediated deletion.

Western blotting

Western blotting was conducted as previously described (16). For primary antibody incubation the antibodies were diluted into TBST+5% BSA. The following Abs were utilized: anti-total-ERK1/2 (Zymed Laboratories Inc.), anti-PTEN (138G6; Cell Signaling Technology), anti-Cre (Novagen), anti-total Akt (11E7;Cell Signaling Technology), and anti-phosphorylated (p)-Akt (C31E5E; Cell Signaling Technology). Quantification was conducted using ImageJ v1.46 software (http://rsbweb.nih.gov/ij/) and a paired t-test was performed for pAkt;tAkt ratio in PTEN deleted compared to wildtype with the indicated conditions.

Flow Cytometry

Antibodies against the following molecules coupled to the indicated fluorochromes were utilized: BD PharMingen (BD) or eBiosciences (eBio): PE anti–CD3 (2C11; BD PharMingen), PE anti–CD62L (MEL14; BD PharMingen), APC anti-CD44 (1M7; BD PharMingen). In general, 10⁶ cells were blocked with the anti-FcR mAb 2.4G2, stained with the indicated antibodies or appropriate isotype controls for 15 minutes at 4°C and then washed and resuspended for FACS analysis. Flow cytometry was performed on the FACScanto cytometer using BD FACSDiva software. Data analysis was performed using FlowJo software.

T cell stimulation and anergy induction

T cells were activated with beads (Dynal) coated overnight at 4 degrees with anti-CD3 (145-2C11) at 1 μg/ml and anti-CD28 (PV1) at 1 μg/ml unless otherwise indicated. For stimulation, T cells were incubated with the antibody-coated beads in 96 well flat bottom plates, in triplicate, at 1×10⁵ cells/well in 200 μl of complete medium (or 24 well plates at 1×10⁶ cells/well in 1 ml for gene array experiments) for the indicated time periods at 37°C at a 5:1 bead to T cell ratio. Previously we have shown that concentrations of anti-CD3 mAb at 0.1–1.0 μg/ml give optimal responses corresponding to the level of functional output and signaling events seen with antigen/APC stimulation, and at higher levels anti-CD3 can actually be inhibitory (17). For anergy induction, first Th1 CAR Tg x PTEN^flox/flox cells were made anergic by initially stimulating for 24 hours with plate-bound anti-CD3, then were collected and allowed to ‘rest’ for 24 h in culture medium alone as described (18). Cells were then collected, washed and re-stimulated with anti-CD3 + anti-CD28 coated beads for functional analysis.

Cytokine ELISA

For cytokine concentration analysis, culture supernatants were removed from wells at the indicated time points. Mouse IL-2 and IFN-γ antibody pairs were obtained from BD and cytokine production was measured by ELISA using NUNC Maxisorp 96-well plates according to the manufacturer’s protocol. Cytokine concentrations were determined with the Softmax PRO data analysis program (Molecular Devices). Three identical experimental wells were tested and analyzed for each condition, and results are expressed as mean ± SD with the data shown representative of 3 replicated experiments.

Proliferation Assay

For the [³H]thymidine incorporation assay, cells were stimulated in 96 well plates with antibody coated beads as described. After 24 or 48 hours of incubation, wells were pulsed with 1 μCi of tritiated thymidine, incubated for a further 8–16 h, then frozen until harvested. The wells were harvested for the determination of [³H]thymidine incorporation using a Packard Filtermate Harvester and TopCount-NXT (PerkinElmer Life Sciences). Three identical wells were analyzed for each condition, and results (in cpm) are expressed as mean ± SD with the data shown representative of 3 replicated experiments.

Quantitative RT-PCR

Reactions were performed in 25 μl volumes and carried out in 96-well optical plates. An ABI 7700 thermal cycler (Applied Biosystems) was used for amplification. Each gene was evaluated in duplicate using a specific primer/probe set purchased from Applied Biosystems and labeled with FAM dye. CT of a particular gene was normalized against the CT of 18s. Data were analyzed using SDS Software (Applied Biosystems).

Mixed Lymphocyte Reaction (MLR)

MLR stimulation to generate effector cells for proliferation and CTL analysis was adapted from a previously published protocol (19). Total T cells, or separate CD4⁺ and CD8⁺ fractions, were purified from spleens by negative selection with antibodies and magnetic beads from Miltenyi (MACS) according to the manufacturer’s protocol. In 24- (or 96-) well tissue culture places, responder cells were plated at 1 ×10⁶/well (or 5 ×10⁴/well), along with stimulator cells consisting of allogeneic T cell–depleted irradiated (5,000 rad) splenocytes at 1 × 10⁶/well (or 5 ×10⁴/well). For proliferation assays, 96-well plates were used and thymidine was added after 72 hours, for overnight incorporation, each condition performed in triplicate wells. For chromium-release assays, 24-well plates were used and after 5 days, the resulting cells were collected and counted prior to exposure to target cells at equivalent proportions.

⁵¹Cr-release assay

Chromium-release assays were performed as previously described (20). Briefly, ⁵¹Cr-labeled targets (2×10³) were plated with PTEN deleted or control, CD4⁺ and CD8⁺ T cells at varying concentrations a 96-well V-bottom plate (ICN Biomedicals), in triplicate. After 4 h of incubation at 37°C, 25 μl of supernatant was transferred to a LumaPlate-96 (PerkinElmer Life Sciences) and allowed to dry overnight. Plates were then counted using the TopCount-NXT (PerkinElmer Life Sciences). Percent specific lysis was calculated using standard methods.

Gene array analysis

The indicated T cell populations were stimulated as described. At the times indicated, total RNA was isolated by TRIzol reagent protocol (Invitrogen), followed by RNeasy Mini column purification (QIAGEN) according to manufacturer’s instructions. RNA integrity was evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies). The concentration/purity was determined using a NanoDrop 1000 (Thermo Scientific). All RNA samples used for hybridization had an OD260/280 and OD260/230 ratio >1.8 and RIN (RNA Integrity Number) > 8.0. Total RNA was processed to cDNA synthesis, cRNA synthesis, fragmentation, and hybridization to expression arrays according to Genechip Expression analysis technical manual (Affymetrix, Inc). Briefly, 2 ug of total RNA was used to synthesize double-stranded cDNA using the Genechip Expression 3′-Amplification one cycle cDNA Synthesis kit (Affymetrix). First strand cDNA synthesis was primed with a oligo(dT)24 primer that contains T7 promoter sequences. From the cDNA purified by Genechip sample cleanup module (Affymetrix), biotin-labeled antisense RNA (cRNA) was synthesized using Genechip Expression 3′ amplification IVT labeling kit (Affymetrix). After cleanup of cRNA with Genechip sample cleanup module, 20 μg of cRNA was fragmented in fragmentation buffer for 35 min at 94°C. The fragmented cRNA (12 μg) was hybridized to Affymetrix MG 430 2.0 expression arrays for 16 h at 45°C and 60 rpm in an affymetrix hybridization oven 640. Arrays were washed and stained with Streptavidin Phycoerythrin and the fluorescent signal was amplified using a biotinylated antibody solution in an Affymetrix Fluidics Station 450 according to the Affymetrix GeneChip protocol. The arrays were scanned using the Affymetrix Gene Chip Scanner 3000 7G. CEL intensity files generated by Gene Chip Operating Software v. 1.4 (MicroArray Suite 5.0) were used to extract and analyze data using dChip software (Harvard).

Array normalization and expression value calculation were performed (21). Each analysis was done in duplicate utilizing different primary material treated with equivalent experimental conditions. Arrays were normalized at the probe cell level by the invariant set normalization method to allow for comparison of expression values computed using the model-based method (21). Measurement accuracy evaluated by standard error was used to compute 90% confidence intervals of fold changes in two sample and two-group comparisons (ie: EV to Cre comparison done twice and compared for consistency). Increased or decreased expression of genes by more than 2-fold (lower confidence bound) are presented.

The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus (22) and are accessible through GEO Series accession number GSE43936 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE43936).

Results

PTEN-deleted peripheral T cells exhibit increased Akt phosphorylation, proliferation, and cytokine production upon CD3/CD28 engagement but do not show increased survival

In order to understand the role of PTEN in T cell activation, we used a model which allows specific deletion of genes directly within peripheral T cells (12). Mice homozygous for conditionally targeted PTEN alleles (10) were interbred with mice transgenic for the coxsackie and adenovirus receptor (CAR-Tg) under the control of an Lck promoter/CD2 enhancer cassette (13, 23). This model system allows for deletion of targeted alleles using a Cre adenovirus in vitro, without a need to induce cell proliferation (12). To determine the effects of PTEN deletion in peripheral T cells, either splenic T cells or Th1 clones generated by immunization of these mice with ovalbumin (24) were employed. In both cases, transduction with Adeno-Cre led to near complete elimination of PTEN protein as detected by Western blot analysis, compared to transduction with an empty vector (Figure 1A–B). In all functional experiments, Western blot analysis was similarly performed to confirm efficient PTEN deletion.

Figure 1. Deletion of PTEN using adenoviral delivery of Cre recombinase in peripheral T cells leads to augmented TCR-mediated Akt phosphorylation.

Following adenoviral transduction of CAR Tg x PTEN^flox/flox T cells with Cre or empty vector (EV), Western blot revealed that (a) total splenic T cells and (b) CD4⁺ Th1 clones exhibited a significant diminution of PTEN protein expression as compared to the ERK positive control. A representative of six experiments for each is shown. (c) After transduction, Th1 clones were co-cultured for 30 minutes with anti-CD3 or anti-CD28 coated beads at the indicated concentrations and Western blot analysis was performed for phosphorylated versus total Akt and ERK. A representative of two independent experiments is shown.

Diminished PTEN levels would be expected to augment PI3K pathway activity. We examined the proximal downstream event of this pathway, Akt phosphorylation, in response to CD3/CD28 ligation. As shown in Figure 1C, a significant increase in Akt phosphorylation was indeed observed upon deletion of PTEN. The increase in pAkt:tAkt was not constitutive (p=0.6 for un-stimulated cells) but still required TCR ligation (p<0.005 for stimulated cells). Furthermore the fold induction of pAkt:tAkt in stimulated cells compared to un-stimulated cells was increased with PTEN deletion (p<0.005). This supports the notion that PTEN deletion alone is not sufficient for Akt activation. It is also noteworthy that the augmentation in Akt phosphorylation was seen with ligation of the TCR complex alone, without CD28 costimulation (p<0.05 for anti-CD3-stimulated cells), indicating that PTEN deletion is largely augmenting TCR-based signaling rather than costimulatory receptor effects. Erk phosphorylation was slightly increased in the experiment shown, however this was not reproducible across experiments (data not shown).

To begin to address the functional effects of PTEN loss, T cells were stimulated with anti-CD3/anti-CD28-coated beads and proliferation was assessed. As expected, PTEN-deleted splenic T cells and Th1 clones showed increased proliferation as measured by thymidine incorporation (Figures 2A–B). As increased Akt activation has been associated with augmented cell survival in some model systems (25), the percentage of apoptotic cells was also assessed. In fact, there was no detectable difference in the fraction of viable cells upon PTEN-deletion (Figures 2C–D), indicating that the increased thymidine incorporation observed was directly due to increased proliferation and not via decreased cell death.

Figure 2. Deletion of PTEN leads to increased TCR-induced proliferation without affecting apoptosis.

(a,b) Either CD4⁺ T cell clones (a) or total splenic T cells (b) from CAR Tg x PTEN^flox/flox mice were transduced with EV or Cre adenovirus as described in Materials and Methods. Following confirmation of PTEN deletion, transduced cells were washed and stimulated with anti-CD3 + anti-CD28 mAb-coated beads and thymidine incorporation was measured. A representative of three independent experiments for each is shown. Results are expressed as mean ± SEM. Two tailed t-test differences with p > 0.05 were considered to be insignificant, *p value < 0.05. (c,d) CD4⁺ T cell clones (c) or splenic T cells (d) from CAR Tg x PTEN^flox/flox mice were transduced with EV or Cre adenovirus and stimulation with anti-CD3+anti-CD28 mAb-coated beads was performed as above. Flow cytometry was performed for Annexin V and propidium iodide staining. A representative of two independent experiments is shown.

The increased cell proliferation observed after PTEN deletion suggested that IL-2 production may have been increased. Therefore, cytokine production was assessed in response to CD3/CD28 ligation. As shown in Figure 3A, IL-2 production was indeed augmented in PTEN-deleted Th1 clones, and a modest increase in IFN-γ production was also observed (Figure 3B). Increased cytokine secretion was accompanied by augmented cytokine mRNA expression (Figures 3C–D), arguing for potentiation at the level of gene transcription. Kinetic analysis revealed that the augmentation of IL-2 mRNA expression was detected as early as 3 hours following CD3/CD28 stimulation (Figure 3E), consistent with the notion that PTEN predominantly negatively regulates proximal TCR-mediated signaling events.

Figure 3. PTEN-deleted T cells reveal increased cytokine production which is reflected at the mRNA level.

PTEN-deleted versus control Th1 clones were stimulated with anti-CD3+anti-CD28 mAb-coated beads. (a, b) Supernatant was collected at 24 hours for IL-2 and IFN-γ quantification by ELISA. (c, d) Cells were collected 6 hours after stimulation and mRNA was isolated for IL-2 and IFN-γ mRNA quantification. A representative of three individual experiments is shown for each. (e) Analysis of IL-2 mRNA induction by qRT-PCR over time in CD4⁺ T cell clones. A representative of three independent experiments is shown. Results are expressed as mean ± SEM. Two tailed t-test differences with p > 0.05 were considered to be insignificant, *p value < 0.05, **p value < 0.005.

PTEN deletion in peripheral T cells lowers the TCR threshold for activation, but does not abrogate the need for CD28 costimulation and does not prevent anergy induction

It has previously been suggested that the PTEN-PI3K axis represents a signaling event predominantly regulated by CD28 costimulation (26). We therefore examined closely whether deletion of PTEN directly in peripheral T cells appeared to bypass a need for CD28 ligation. A dose titration of anti-CD3 mAb was performed with or without the addition of anti-CD28 mAb. As shown in Figure 4A–B, a shift in the dose-response curve was observed with PTEN-deleted T cells. Approximately 10-fold less anti-CD3 mAb was required in PTEN-deleted T cells to generate comparable levels of IL-2 production to non-deleted cells for both primary splenic T cells (Figure 4A) and Th1 clones (data not shown). A similar shift in dose titration was observed for Th1 helper production of IFN-γ (Figure 4B), and also for the dose of anti-CD3 mAb required to induce proliferation (Figure 4C–D). However, it is important to note that CD28 ligation was still required to induce substantial IL-2 production, even when PTEN was deleted (Figure 4a and 4e). Therefore, deletion of PTEN directly in post-thymic T cells did not bypass the requirement for CD28 costimulation.

Figure 4. Following PTEN deletion, the threshold for TCR-mediated signaling is lowered yet CD28 dependency and anergy susceptibility are not affected.

(a) PTEN-deleted versus control T cells were stimulated with the indicated concentrations of anti-CD3 +/− anti-CD28 mAb. IL-2 production (a) from total peripheral T cells and IFN-γ production (b) from Th1 clones were assessed by ELISA. In parallel experiments, proliferation was measured in (c) total peripheral T cells and (d) CD4⁺ Th1 clones clones. (e) Following anergy induction using plate bound anti-CD3 antibody, Th1 cells were re-stimulated with anti-CD3/anti-CD28 mAb-coated beads and IL-2 production was assessed by ELISA. Representatives of three independent experiments are shown for each. Results are expressed as mean ± SEM. Two tailed t-test differences with p > 0.05 were considered to be insignificant, *p value <0.05, **p value <0.005.

We next examined whether PTEN deletion in peripheral T cells would prevent the ability of those cells to be capable of anergy induction. This was examined in Th1 clones, which represents the classical cellular model of T cell anergy (27–29). Anergy was induced using immobilized anti-CD3 mAb, as we and others have reported previously (12, 30). As shown in Figure 4e, Th1 clones were equivalently rendered hyporesponsive to CD3/CD28 ligation under anergizing conditions, whether or not PTEN was deleted. Cells in both conditions remained viable as indicated by their ability to secrete IL-2 in response to PMA and ionomycin. Therefore, deletion of PTEN directly in post-thymic T cells failed to prevent anergy induction in vitro.

Gene expression profiling reveals a small subset of induced genes influenced by PTEN expression

Our results thus far indicated that conditional deletion of PTEN in peripheral T cells led to a marked augmentation of IL-2 production and proliferation in response to CD3/CD28 ligation. It was therefore of interest to determine the spectrum of transcripts regulated by PTEN using gene expression profiling. To this end, CAR Tg x PTEN^flox/flox Th1 clones were either treated with adeno-Cre or adeno-EV, then stimulated for 6 hours with anti-CD3/anti-CD28 mAb-coated beads. We found that deletion of PTEN resulted in augmented expression of a remarkably limited set of transcripts after activation (Figure 5 and Table 1). The top two differentially expressed genes were CDK19 and the aryl hydrocarbon receptor nuclear translocator 2, which could theoretically influence cell proliferation. The third most differentially expressed gene was IL-2, which was already defined through analysis of candidates. Most other differentially expressed genes were near the 2-fold cutoff threshold. These results suggest that the major role of PTEN in terms of influencing gene expression in peripheral CD4⁺ T cells is likely regulation of IL-2 production and proliferation.

Figure 5. Gene array analysis of activated CD4⁺ T cell clones with or without PTEN deletion reveals a narrow set of genes regulated by PTEN.

CAR Tg x PTEN^flox/flox Th1 cells were transduced with either Adeno-EV or Adeno-Cre. Rested cells were stimulated with anti-CD3+anti-CD28 mAb-coated beads for 6 hours and gene expression profiling was performed using the Affymetrix platform. Data for the entire panel of genes from two completely independent experiments were averaged and are plotted for the two conditions. Large squares indicate statistically significant differences in gene expression between groups, which are listed in Table 1.

Table 1.

Genes with 2 fold (lower confidence bound) increased or decreased expression, 6 hours following stimulation, in PTEN deleted T cells (Adeno-Cre) as compared to wildtype T cells (Adeno-EV).

Gene	Accession number	Fold Change in mRNA transcripts PTEN deleted-Control
Cdc2l6	BB510904	5.83
Arnt2	BQ174321	4.07
IL2	AF065914	3.96
CRTAM	NM_019465	3.69
Lymphotactin, Xcl1	NM_008510	3.02
Siglec5	AF293371	2.79
Helicard	AY075132	2.7
Esm1	BC020038	2.54
Slam7	AK016183	2.49
lymphoid-restricted membrane protein	NM_008511	2.37
Hectd2	AV256030	2.28
nuclear protein 1	NM_019738	2.2
Pdzk1ip1	BC013542	2.13
Sulf2	AK008108	2.08
Sez6	BB079338	2.07
Slc41a1	BF134253	2.06
GTP cyclohydrolase 1	NM_008102	2.05
Kcna4	BB131475	−2.01
Herc4	BB053466	−2.02
Ccr1	AV231648	−2.05
integrin alpha 4	BB284583	−2.06
Rrm2b	BB702377	−2.07
Musculin	NM_010827	−2.11
interferon inducible GTPase 1	BM239828	−2.2
glutaminase	BB543271	−2.3
Ccar1	AI503765	−2.31
ameloblastin	NM_009664	−2.46
Tmtc3	BB749399	−2.58
Pabpc1	AK005009	−2.59
nephroblastoma overexpressed gene	X96585	−276.
Ptar1	AK006988	−2.86

PTEN deletion in CD8 T cells yields markedly augmented cytolytic activity in vitro

The PI3K pathway is thought to be an important signaling event in CD8⁺ T cell function as well, and its blockade has been proposed as an immunosuppressive strategy to prevent alloreactivity activity after transplantation. Conversely, mechanisms to augment signaling through the PI3K axis could improve CD8⁺ T cell-based immunity against tumors or infectious agents. To examine the effect of PTEN deletion in CD8⁺ T cell function, we performed an allogeneic MLR using total splenic T cells from CAR Tg x PTEN^flox/flox mice transduced with adeno-Cre (PTEN deleted) or adeno-EV (control). Stimulators were T cell-depleted splenocytes from either allogenic DBA/2 mice or autologous C57BL/6 mice as a control. Thymidine incorporation was measured, and PTEN deletion led to a significant increase in proliferation against allogeneic stimulator cells as compared to control cells (Figure 6a).

Figure 6. Allogeneic MLR reveals increased proliferation and cytolytic activity in PTEN-deleted cells.

MLR was performed with control versus PTEN-deleted total peripheral T cell or separated CD4⁺ and CD8⁺ T cells as indicated in the Materials and Methods. (a) Proliferation after MLR of total peripheral T cells was assessed on day 3. Results are expressed as mean ± SEM. Two tailed t-test differences with p > 0.05 considered to be insignificant, *p value < 0.05. A representative of three independent experiments is shown. (b) Cytolytic activity was assessed by Chromium-release assay against allogeneic targets, following an MLR in which PTEN deletion was performed in either CD4⁺ T cells or CD8⁺ T cells. A representative of three independent experiments is shown. Results are expressed as mean with error bars showing standard deviation. Repeated measures one way ANOVA was performed, pairing results for each responder:target ratio across groups. Dunnett’s post test was used to compare each curve against the curve for the control wild type T cells (CD4-EV + CD8-EV) with p > 0.05 considered to be insignificant, **p value < 0.001.

It was of interest to determine the effects of PTEN deletion on the development of cytotoxicity and whether the effect of PTEN deletion was within the CD4⁺ or CD8⁺ T cell compartment. To examine this possibility, CD4⁺ and CD8⁺ T cell populations were first purified from CAR Tg x PTEN^flox/flox mice and then separately transduced with either adeno-Cre or adeno-EV. An MLR was performed and CTL activity was assessed against chromium-labeled P815 (H-2^d) tumor cells. As shown in Figure 6b, a markedly increased level of cytolytic activity was observed upon deletion of PTEN, and this effect was completely recapitulated when PTEN deletion was performed selectively within the CD8⁺ T cell compartment. Thus, in addition to augmenting cytokine production and proliferation by CD4⁺ T cells, PTEN deletion can potentiate cytolytic activity by CD8⁺ T cells.

Discussion

In this report we show that deletion of PTEN directly in peripheral T cells having undergone normal thymic development decreases the activation threshold for TCR-mediated signaling. PTEN deletion in peripheral T cells retains dependency on CD28 co-stimulatory signaling for activation and does not generate an anergy-resistant phenotype. Nonetheless, PTEN deletion augmented cytokine production and proliferation in CD4⁺ T cells, and additionally potentiated cytolytic activity by CD8⁺ T cells. These functional effects were associated with augmentation in activation of the PI3K pathway as evidenced by increased TCR-induced phospho-Akt. Thus, developing pharmacologic strategies to potentiate the PI3K pathway in T cells could serve as an attractive immune potentiating platform.

Previously published reports of T cell PTEN deletion during thymic development required analysis of T cells while the mice remained young, as lymphoproliferation, autoimmunity, and lymphoma ultimately developed by 16 weeks. The differences between those results and our current data are most likely due to the timing of PTEN deletion during T cell development, although a direct comparison will have to be performed for definitive conclusions to be drawn. In preliminary experiments, we have adoptively transferred peripherally PTEN-deleted polyclonal T cells into RAG2^−/− recipient mice and have found no evidence of autoimmunity or lymphoma out to 6 months. Further work in this area will benefit from a strategy to selectively deplete PTEN in peripheral T cells in vivo, without a need for cumbersome adoptive transfer. We are currently developing a tamoxifen-regulated Cre Tg mouse driven by an Lck-CD2 promoter/enhancer cassette which should facilitate answering this question in more detail in the future.

Buckler and colleagues previously reported that PTEN deletion earlier in thymic development led to mature T cells that produced significant levels of IL-2 in a CD28-independent manner and that were relatively resistant to anergy induction (31). As mentioned above, the most likely explanation for this contrasting result is the differential timing of Cre-mediated deletion during T cell development. Cre expression and subsequent gene deletion utilizing Lck-Cre or CD4-Cre occurs during both positive and negative selection, and there are clear examples of paradoxical findings in the literature. For example, the phosphatase SHP2 has been implicated as a key negative regulatory molecule for TCR signaling using in vitro models, and is thought to mediate inhibitory effects of PD-1 and possibly CTLA-4 (32, 33). However, conditional deletion of SHP2 in thymocytes results in blunted TCR signaling and altered selection (34). Deletion of PTEN in the thymus could generate thymic emigrees that have an altered threshold for TCR-mediated signaling and are hyper-responsive to TCR ligation, including self peptide MHC-complexes that give rise to autoimmunity. In the future it would be of interest to perform comparative studies regarding the timing of PTEN deletion at differing stages of thymic development.

That being said, we acknowledge some limitations of our present model as well. First, the adenovirus itself used for Cre transduction may alter T cell function, although we have shown previously that empty, non-replicating, adenoviral vectors do not affect the T cell functions analyzed in the assay systems used in this study (13, 23, 35). The use of an empty vector control allows for evaluation of Cre expression as the single variable tested. We also have done extensive analysis of the potential off-target effects of Cre on T cells (using CAR Tg T cells without a floxed allele) and have not detected any such effects in our in vitro assays (12, 16, 36) (and data not shown). Second, for gene deletion in primary T cells, co-culture with IL-7 was required to prevent cell death in vitro, similar to the requirement for IL-7R signaling for naïve T cell survival in vivo (15). Although PTEN deletion might affect IL-7R signaling, we did not find any skewing of T cell populations following this culture, there was no differential expression of the IL-7R in PTEN-deleted cells, and following withdrawal of IL-7 there was no expansion or contraction of PTEN deleted T cell population. It should be noted that any in vivo Cre system (such as CD4-Cre or Lck-Cre Tg mice) similarly would potentially impact on many different receptor-mediated signaling pathways, including cytokine receptors such as the IL-7R. Third, while PTEN protein absence is complete in the deleted Th1 clones, a very low level of residual PTEN protein persists in the primary T cell experiments (Figure 1B). Careful examination of Western blots from in vivo conditional knockout experiments reveal there often is a low level of residual protein detected (37, 38). Despite this low level of residual protein, we nonetheless observed a potent phenotype upon PTEN deletion: markedly augmented IL-2 production, proliferation, Akt phosphorylation, and cytolytic activity.

An implication of our findings is that CD28 costimulation relies on pathways other than augmentation of PI3K activity. It is known that CD28 ligation can augment several TCR signaling pathways, however transfection of CD28 mutants has indicated that the ability to activate PI3K is not necessary for IL-2 production and proliferation (6, 39). We typically have seen minimal augmentation of pAkt in conventional T cells with CD28 costimulation, but on the other hand have seen augmentation of Ras pathway signaling when suboptimal TCR ligation is used. In fact, we have recently reported evidence strongly implicating alterations in Ras pathway signaling that seem critical for the costimulatory effects of CD28. Ligation of CD28 promoted RasGRP translocation to the immunological synapse, and IL-2 production driven by CD3/CD28 costimulation was lost in RasGRP1^−/− mice. Moreover, introduction of active Ras could recapitulate all the major functional effects of CD28 engagement (increased IL-2 mRNA, increased mRNA stability, prevention of anergy, potentiation of T cell survival, and augmentation of glucose uptake) (40). Together, these results strongly argue that Ras pathway (rather than PI3K pathway) signaling constitutes a major functionally important signaling event in CD28 costimulation.

While we failed to detect any effect on T cell anergy upon PTEN deletion in Th1 cells, this is not a property of the model system itself. We recently have identified the transcription factor Egr2 as an important contributor to the anergic state. In an analogous fashion to the current report, we generated Th1 clones from Egr2^fl/fl x CAR Tg mice, which enabled Cre-mediated deletion of Egr2 in vitro. In fact, T cell anergy was largely prevented in that system (36). Use of this model also enabled identification of other key Egr2 transcription targets in anergy (41). These data not only support the utility of this model system, but additionally support the functional role for alternative signaling pathways as key for anergy besides PTEN.

It was striking to us that unbiased gene expression profiling on PTEN-deleted T cells revealed such a limited set of transcripts regulated by this signaling molecule. We primarily saw augmentation of IL-2 mRNA level transcription, which we had known already by examining it as a candidate target gene. Beyond effects on gene expression, PTEN deletion also markedly augmented cytolytic activity by CD8⁺ T cells, which is likely a post-translational effect. The increased cytolysis was not due to an increased number of T cells, rather this seems to be an increase in cytolytic activity directly. It is interesting to note that in other model systems, important changes in membrane ruffling and adhesion following conditional deletion of PTEN have been reported (42–44). Future study should focus on these mechanisms in CD8+ T cells. Together, these results suggest that the major function of PTEN is to negatively regulate IL-2 production, proliferation, and cytolytic effector function in activated peripheral T cells.

The increased effector function of T cells upon PTEN deletion argues that pharmacologic strategies to inhibit PTEN and/or augment PI3K activity may be attractive to consider as immune-potentiating drugs. While multiple small molecule inhibitors of T cell activation have been developed for immune suppression during settings of immune-mediated pathology, the rational development of immune potentiating drugs has in general lagged behind. The first example applied clinically and achieving FDA approval is the anti-CTLA-4 mAb ipilimumab (45). The significant expense of protein-based therapeutics makes it potentially attractive to pursue small molecule approaches. An additional theoretical advantage is the opportunity to discontinue such drugs in case immune-mediated adverse events arise, with potentially rapid resolution. Future development of such immune potentiating agents that phenocopy effects seen with deletion of PTEN or other negative regulatory factors in T cells seems warranted.