Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

The PI3K p110δ Regulates Expression of CD38 on Regulatory T Cells

  • Daniel T. Patton,

    Affiliation Laboratory of Lymphocyte Signalling and Development, Babraham Institute, Cambridge, United Kingdom

    ⨯
  • Marcus D. Wilson,

    Affiliation Laboratory of Lymphocyte Signalling and Development, Babraham Institute, Cambridge, United Kingdom

    ⨯
  • Wendy C. Rowan,

    Affiliation Tool Monoclonal Antibody Group, GlaxoSmithKline Research and Development, Stevenage, United Kingdom

    ⨯
  • Dalya R. Soond,

    Affiliation Laboratory of Lymphocyte Signalling and Development, Babraham Institute, Cambridge, United Kingdom

    ⨯
  • Klaus Okkenhaug

    klaus.okkenhaug@bbsrc.ac.uk

    Affiliation Laboratory of Lymphocyte Signalling and Development, Babraham Institute, Cambridge, United Kingdom

    ⨯

Abstract

The PI3K pathway has emerged as a key regulator of regulatory T cell (Treg) development and homeostasis and is required for full Treg-mediated suppression. To identify new genes involved in PI3K-dependent suppression, we compared the transcriptome of WT and p110δD910A Tregs. Among the genes that were differentially expressed was the gene for the transmembrane cyclic ADP ribose hydrolase CD38. Here we show that CD38 is expressed mainly by a subset of Foxp3+CD25+CD4+ T cells originating in the thymus and on Tregs in the spleen. CD38high WT Tregs showed superior suppressive activity to CD38low Tregs, which failed to upregulate CD73, a surface protein which is important for suppression. However, Tregs from heterozygous CD38+/− mice were unimpaired despite lower levels of CD38 expression. Therefore, CD38 can be used as a marker for Tregs with high suppressive activity and the impaired Treg function in p110δD910A mice can in part be explained by the failure of CD38high cells to develop.

Citation: Patton DT, Wilson MD, Rowan WC, Soond DR, Okkenhaug K (2011) The PI3K p110δ Regulates Expression of CD38 on Regulatory T Cells. PLoS ONE 6(3): e17359. https://doi.org/10.1371/journal.pone.0017359

Editor: Jacques Zimmer, Centre de Recherche Public de la Santé (CRP-Santé), Luxembourg

Received: December 22, 2010; Accepted: January 27, 2011; Published: March 1, 2011

Copyright: © 2011 Patton et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The work described in this manuscript was funded by the BBSRC and GlaxoSmithKline. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: Klaus Okkenhaug is a paid consultant for GSK. Klaus Okkenhaug is employed by the BBSRC and has received research funding from the BBSRC, the Wellcome Trust and GSK. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.

Introduction

The role of regulatory T cells (Tregs) in preventing systemic autoimmunity and to limit inflammation is well established. CD4+Foxp3+ T which develop from CD4+CD8+ T cell precursors in the thymus are referred to as natural Tregs [1], [2]. Induced Tregs develop from Foxp3−CD4+ T cells in the peripheral immune organs in presence of low concentrations of antigen or TGF-β [3], [4], [5], [6]. Tregs play a critical role in limiting the responses of not only other T cells, but also B cells and components of the innate immune system to antigen and/or inflammatory stimuli. Several mechanisms have been proposed as to how Treg function [7]. The expression of CTLA-4 is essential for Treg function by a mechanism thought to involve the suppression of APCs [8], [9]. Tregs also express high levels of CD25 which may consume available IL-2 thus depriving T helper cells of this cytokine [10]. CD39 and CD73 expressed by Tregs generate adenosine which has an immunosuppressive effect on Th cells [11]. Tregs also mediate immunosuppression in different physiological contexts by secreting the anti-inflammatory cytokines including IL-10, IL-35 and TGF-β [9], [10], [12], [13], [14], [15], [16].

The Class I PI3K enzymes phosphorylate the D3-position of Phosphatidylinositol PtdIns(4,5)P2 to generate PtdIns(3,4,5)P3 which in turn is bound by proteins such as Pdk1, Akt and Itk that contain a pleckstrin homology domain [17]. Four catalytic isoforms of Class I PI3K are expressed in T cells: p110α, p110β, p110γ and p110δ [17]. p110α, p110β and p110δ form heterodimers with SH2-domain containing p85, p55 or p50 regulatory subunits whereas p110γ is bound by a p101 or p84 regulatory subunit. In T cells antigen, costimulatory and cytokine receptors activate p110δ, whereas p110γ is activated by chemokine receptors [18]. We have previously shown that Treg development, differentiation and function are altered in p110δD910A mice which possess a kinase-dead mutant of p110δ [19]. Treg development in the thymus was enhanced whereas there were fewer Tregs in the peripheral organs. Importantly, p110δD910A Tregs were impaired in their capacity to suppress the proliferation of responder CD4+ T cells, secreted reduced levels of IL-10 and failed to suppress inflammation of the colon [19]. Moreover, p110δD910A mice were resistant to infection by Leishmania major and this was attributed to defective Treg expansion and recruitment to the site of infection [20]. However, despite their impaired function, p110δD910A Tregs express similar levels of Foxp3, CD25 and CTLA-4 [19]. Since IL-10 is not essential for all Treg-dependent functions, this leaves open the question of the precise nature of the suppressive mechanism that is defective in p110δD910A Tregs. Deletion of the p85α and p85β PI3K regulatory isoforms in T cells resulted in a reduction in Tregs in the spleen and development of a Sjogren's-syndrome-like disease; however, whether this is linked to Treg-deficiency has not been determined [21]. More recently, Pdk1 has been shown to be essential for Treg function, but not for Treg development, which is consistent with this being and important signaling protein downstream of p110δ [22].

The role of PI3Ks in Treg development and function has been further emphasized by the identification of Foxo transcription factor binding sites in the Foxp3 promoter and by the observation that Treg development is impaired in mice with a T cell-specific deletion of Foxo1 and Foxo3 [23], [24], [25]. PI3Ks regulate Foxo activity via Akt, which phosphorylates Foxo proteins leading to their sequestration in the cytoplasm [26]. Therefore, inhibition of PI3K could lead to enhanced Foxo-dependent Foxp3 expression and may help explain why there are more Foxp3 cells in the thymus of p110δ-deficient mice. Whether Foxo plays a similar role in peripheral Treg is less clear given the reduced proportions of Treg in the spleen and lymph nodes of p110δD910A mice. Moreover, CD4Cre-Foxo1 mice had more Tregs in the peripheral organs, despite reduced proportions in the thymus. The serine/threonine kinase mTOR integrates signals for the PI3K and MAP-kinase pathways in T cells [27]. Surprisingly, the mTOR inhibitor rapamycin enhances differentiation of Treg cells by a mechanism that has yet to be fully understood, but which may involve Foxp3-dependent upregulation of Pim2 [28], [29], [30]. Hence, the PI3K pathway can affect Treg numbers positively or negatively, depending in part on their stage of development and anatomical context.

To gain a more complete understanding of the role of p110δ-dependent transcriptional regulation in Treg development and function we compared the transcriptome of p110δD910A Treg with WT Treg, and found reduced expression of the gene for the transmembrane cyclic ADP ribose hydrolase CD38. Sorted CD38high Treg showed superior suppressive capacity to CD38low Treg. However, CD38+/− heterozygous Treg showed normal suppression in vitro suggesting that reduced CD38 expression per se is insufficient to abolish Treg activity, but rather identifies CD38high Tregs as a population of highly suppressive Treg that fails to develop in p110δD910A mice.

Methods

Mice

CD38−/−, p110δD910A, and RAG2−/− have been previously described [31], [32], [33] and were maintained on the C57BL/6 (B6) background. Congenic B6.SJL mice (in which the CD45.1 allele from the SJL strain has been backcrossed onto the B6 genetic background) were originally purchased from Taconic. All experimental protocols had been approved by the UK Home Office and local ethical review (PPL 80/1809 and 80/2248).

Reagents

The following antibodies were purchased from eBioscience or Becton Dickinson: CD4 (GK1.5), CD8 (53-5.7), CD25 (PC61.5), CD38 (clone 90), CD45.1 (A20), CD45.2 (clone 104), CD73-PE, CD90.2 (53.2.1), CTLA4 (UCH10-4B9), Foxp3 (FJK-16s), GITR (DTA-1), ICOS (15F9). Anti-CD3ε (clone 145-2C11) was prepared in-house.

Microarray experiments

RNA from 2×106 lymph node CD4+CD25+ cells from 6–8 mice was isolated using Trizol, biotinylated, fragmented and hybridised to Affymetrix GeneChip Mouse Genome 430 2.0 mouse arrays according to the manufacturer's protocols (array service provided by Geneservice). Three separate preparations of RNA from each genotype were compared using GeneSpring (cut-off: 2 fold difference, p<0.01). Microarray data has been submitted to ArrayExpress with the accession number E-MEXP-2955.

Quantitative real-time PCR (qRT-PCR)

cDNA was synthesized from RNA purified as above using the Superscript II kit (Invitrogen). PCR was performed using SYBR green PCR mastermix (Applied Biosystems) and a Chromo4 machine (Bio-Rad). Primer sequences are given in Table S1.

Bone Marrow Chimeras

RAG2−/− mice were irradiated with 20 gy and reconstituted with 5×106 cells from a 1∶1 mixture of either WT∶B6.SJL or p110δD910A∶B6.SJL bone marrow. After eight weeks, the mice were dissected and spleen cells analyzed by FACS.

Regulatory T cell purification and co-culture experiments

Tregs cells were purified using Miltenyi magnetic beads and by FACS to greater than 98% purity. CD4+CD25− cells were purified by negative selection using magnetic beads. In some experiments, the CD4+CD25− cells were stained using 2 µM CFSE (Molecular Probes) for seven minutes at room temperature. For APC preparations, splenocytes were depleted of T cells by labeling with an anti-Thy1.2 antibody (Sigma) and lysing the T cells using Rabbit LowTox-M complement (Cedarlane, Burlington, Ontario, Canada). The remaining cells were layered over Lympholyte-M (Cedarlane) and centrifuged, with the cells from the interface layer irradiated and used as APCs. 105 CD25− responder cells and 105 APCs were added to wells of a 96 U-well plate along with 1µg/ml anti-CD3ε. CD25+ regulatory cells were added at ratios of 1-1 to 1-32 to the CD25− cells. All experiments were performed in RPMI-1640 media (Invitrogen, Paisley, UK) containing 5% FCS (Biosera, Sussex, UK), 1% Penicillin-Streptomycin (Sigma) and 5×10−5 M 2-mercaptoethanol (Sigma). After three days the cells were labeled with anti-CD45.1-PE, anti-CD4-APC, anti-CD90.2-PE and anti-CD45.2 APC-eFluor780 and analyzed on a LSRII flow cytometer (BD) in a buffer containing 2 µg/ml 7-AAD (Molecular Probes). The division history of CD4+CD90.2+CD45.1+CD45.2−7AAD− responder cells was analyzed using Flowjo v8.8.6 (TreeStar, Stanford, US) and the division index (mean divisions per divided cell) plotted. Alternatively, proliferation was measured by 3H-thymidine incorporation.

Differentiation of Treg from CD4+CD25− T cells

105 CD4+CD25− cells were placed in culture with 105 APCs and stimulated with one or more of: 1µg/ml anti-CD3e (2C11), 20 ng/ml TGF-β1 (Peprotech or eBioscience), all-trans retinoic acid in ethanol (ATRA, Sigma) or with ethanol or DMSO at equivalent concentrations used as a vehicle controls. 3µM PI103 (pan-class I and mTOR inhibitor, Calbiochem) and 3µM IC87114 (p110δ selective inhibitor, synthesized in-house) were used to inhibit PI3K. After three days, the cells were stained with CD73-PE, Foxp3-Alexa647, CD38-FITC and analyzed by flow cytometry.

Results

CD38 is one of a limited number of differentially expressed transcripts in p110δ D910A Tregs

We have previously shown that p110δD910A CD4+CD25+ cells are impaired in their ability to suppress the responses of conventional T cells [19]. To identify differentially expressed transcripts that may lead to this impairment, cDNA from p110δD910A and wild type (WT) Treg cells was hybridized to Affymetrix arrays. 125 out of 45,002 probe sets were significantly (p<0.01 and 2-fold different) different between WT and p110δD910A (Fig. 1A and Table S2). Of these, 27 belonged to a set of 603 Treg signature probe sets described by Hill et al. [34] (Fig. 1B). However, the expression of well-characterized Treg-associated genes, including Foxp3, CD25, CTLA-4 was unaltered, suggesting that the CD4+CD25+ T cells from p110δD910A mice genuinely belong to the Treg lineage (Fig. 1B).

thumbnail
Download:
Figure 1. Genome-wide expression-profiling of WT and p110δD910A CD4+CD25+cells.

CD4+CD25+ cells were isolated from WT and p110δD910A mice and RNA expression analysed by gene arrays. A. Shown is data from all 45,002 probe sets plotted as expression in WT Treg against expression in p110δD910A Treg (n = 3 for each genotype). Black dots represent genes that were significantly differentially expressed between WT and p110δD910A (greater than 2-fold difference and p<0.01). B. Differentially expressed genes from A were then compared against a previously published set of Treg signature genes [34]. The numbers of probe sets belonging in each section of the Venn diagram are shown. C. Comparison of expression data from selected probe sets found to be significantly different in A with qRT-PCR performed on RNA from separate preparations of WT and p110δD910A Treg.

https://doi.org/10.1371/journal.pone.0017359.g001

We next performed qRT-PCR analysis of WT and p110δD910A Tregs. Although most genes showed a similar pattern of differential expression, only seven genes were confirmed to be expressed at two-fold higher or lower levels in p110δD910A Tregs compared to WT Tregs by this method (Stard5, IL4, Btbd11, Plac8, Penk1, Cd38 and Rab6b) (Fig. 1C). As CD38 has been previously described as a marker of T cells with regulatory function [35] and because CD38−/− mice bred to the NOD background are more susceptible to develop diabetes [36], we investigated its role in Treg biology and its regulation by PI3K.

We first defined when CD38 was expressed during T cell development. Few CD4−CD8− (double negative, DN), CD4+CD8+ (double positive, DP), CD8+CD4− (CD8 single positive (CD8 SP) or CD4+CD8−Foxp3− (CD4 SP Foxp3−) cells from the thymus of WT or p110δD910A mice expressed CD38 (Fig. 2A). However, high levels of CD38 were expressed on a proportion of WT CD4 SP Foxp3+ cells (Fig. 2A and 2B). In contrast, significantly fewer CD4 SP Foxp3+ cells from p110δD910A mice expressed high levels of CD38 (Fig. 2A and 2B). The level of CD38 expression was also higher on WT Tregs than on p110δD910A Tregs from the spleen (Fig. 2C). Foxp3−CD4+ T cells of both genotypes expressed similar and low levels of CD38, suggesting that the reduced CD38 expression in p110δD910A mice is limited to Tregs.

thumbnail
Download:
Figure 2. Expression of CD38 by WT and p110δD910A CD4+Foxp3+ Treg cells.

A. Expression of CD38 on DN, DP and CD4 and CD8 single-positive cells from the thymus of WT and p110δD910A mice. Mean percentages of the cells within the CD38+ gate are for WT (bold typeface) and p110δD910A mice (italic). B. Summary of statistics shown in A (n = 3 for each genotype). C. Expression of CD38 on Foxp3+ and Foxp3− cells from the spleen of WT and p110δD910A mice. D. Representative FACS plots of CD38 on Foxp3+CD4+ T cells from the spleens of WT∶B6.SJL or p110δD910A∶B6.SJL bone marrow chimeras. E. Mean expression of CD38 on Foxp3+ and Foxp3− cells from WT∶B6.SJL or p110δD910A∶B6.SJL bone marrow chimeras, n = 4 for WT∶B6.SJL and n = 3 for p110δD910A∶B6.SJL.

https://doi.org/10.1371/journal.pone.0017359.g002

To determine if the expression of CD38 on p110δD910A Treg is governed by signaling within the T cells themselves or by an extrinsic factor, competitive bone-marrow chimeras were generated. In these experiments, we mixed WT or p110δD910A (both CD45.2+) bone marrow cells with bone marrow cells from B6.SJL CD45.1+ congenic mice and injected these mixture into lethally irradiated RAG2−/− mice to generate WT∶B6.SJL and p110δD910A∶B6.SJL chimeras, respectively. After eight weeks, the expression of CD38 on CD4+Foxp3+ Treg from the spleens of these chimeric mice was determined. In the spleens of WT∶B6.SJL chimeras, CD4+Foxp3+ Treg cells from both donors showed identical expression of CD38 (Fig. 2D and 2E). However, in p110δD910A∶B6.SJL chimeras, the CD45.2+ cells showed a lower expression of CD38 (Figs. 2D and 2E). The failure to express substantial amounts of CD38 is therefore due to an intrinsic defect within the p110δD910A Tregs.

CD38 defines a highly suppressive subset of Tregs

To determine if the level of CD38 expression correlates with suppressive ability, CD4+CD25+CD38high or CD4+CD25+CD38low cells were sorted from B6 mice and co-cultured with CFSE labeled B6.SJL responder cells. After three days CD38high Treg cells suppressed CD45.1+ responder cells proliferation better than did CD38low Treg (Fig. 3A). No difference in the survival of CD38low Treg at the end of the experiment was observed (data not shown), suggesting that equivalent numbers of Tregs were available for suppression throughout the experiment. CD38 ligation has been described to result in selective induction of CD73 expression [37], which in turn is critical for Treg-mediated suppression [11]. In co-culture experiments, CD38high Tregs upregulated CD73 at higher Treg∶T ratios but did not affect the CD4+CD25− responder cells (Fig. 3B and 3C). We also confirmed that the level of CD73 expression was higher on the CD38high subset of Tregs than on the CD38low subset directly ex vivo (Fig. 3C). This effect appears specific to CD73, as expression of CTLA-4 and Granzyme B was similar on CD38high and CD38low cells (Fig. 3D). Hence, we have defined two new distinct sub populations of Tregs, one that is CD38high and able to up-regulate CD73 and suppress the responses of other T cells and CD38low Treg which cannot upregulate CD73.

thumbnail
Download:
Figure 3. In vitro suppression of T cell proliferation by CD38high and CD38low Treg.

A. Proportion of undivided CD4+CD25− after three days culture with CD38high or CD38low CD4+CD25+ Tregs. B. Expression of CD73 by CD38high and CD38low Tregs after suppression. C. Expression of CD73 by CD4+CD25− cells after suppression by either CD38high or CD38low Treg. D. Expression of CD38 versus CD73, CTLA-4 and Granzyme B on CD4+CD25+Foxp3+ cells from WT and p110δD910A Treg cells.

https://doi.org/10.1371/journal.pone.0017359.g003

Reduced expression of CD38 on p110δD910A Treg is insufficient to cause altered Treg development

CD38−/− mice have reduced numbers of CD4+CD25+Foxp3+ cells in the spleen, as previously reported [36] (Fig. 4B). However, the absence of active p110δ lead to a reduction, but not complete loss of CD38 in Tregs (Fig. 2C), similar to the expression of CD38 in thymus and spleen of CD38+/− heterozygous mice (fig 3A). To determine if altered number of Tregs in p110δD910A mice is related to their lower expression of CD38, we compared the proportions of Tregs in the thymus and spleen of WT, CD38+/−, CD38−/− and p110δD910A mice (Fig 4A and 4B). Unlike in p110δD910A mice, the proportions of Tregs in the spleen and thymus were not altered in CD38+/− mice. Therefore, altered CD38 expression on p110δD910A Tregs is not sufficient to explain the reductions in the numbers or function of Treg cells in p110δD910A mice. Moreover, CD38+/− Tregs suppressed the proliferation of responder cells as well as WT Treg did, under conditions where p110δD910A Treg showed minimal suppression (Fig. 4C).

thumbnail
Download:
Figure 4. Intact development and function of Treg development in CD38+/− mice.

Percentage of Foxp3+ cells and mean linear fluorescence (MLF) of anti-CD38 antibody staining on CD4+Foxp3+ in the thymus (A) and spleen (B) of WT, CD38+/−, CD38−/− and p110δD910A mice. (C) Comparison of suppression of CD4+CD25− proliferation by WT, p110δD910A or CD38+/− Tregs.

https://doi.org/10.1371/journal.pone.0017359.g004

CD38 is induced by ATRA on Treg cells

Expression of CD38 is regulated by several factors including the retinoic acid receptor which binds to the CD38 promoter [38], [39]. Since all-trans retinoic acid (ATRA) has been described to play a role on Treg biology [40], we investigated its role in CD38 expression on regulatory T cells. Tregs were induced from CD4+CD25−Foxp3− T cells by stimulating the cells with anti-CD3 in the presence of TGF-β with or without ATRA. The pan-PI3K inhibitor PI103, the p110δ-specific inhibitor IC87114 or DMSO vehicle control were also added to mimic the p110δD910A genotype pharmacologically. After three days of culture, TGF-β induced Foxp3 which was blocked by the PI3K inhibitors. PI103 had a more potent effect suggesting that PI3K isoforms other than p110δ contribute to Treg formation in vitro. ATRA had little effect on the proportion of Foxp3+ cells produced either on its own or in combination with PI103 or IC87114 (Fig. 5A). We next examined the expression of CD38 and CD73 on TGF-β induced Treg cells. ATRA had no effect on the proportion of Foxp3+ Treg developing, but enhanced the level of CD38 to the same level found in IC87114-inhibited T cells (Fig 5B). Hence, ATRA enhances CD38 expression on Treg. Curiously, IC87114 blocked Treg induction, yet enhanced CD38 and CD73 expression on the few Tregs that were induced (Fig. 5C). These results indicate that ATRA may contribute to increased CD38 and CD73 expression independently of p110δ activity on Treg. Moreover, p110δ activity appears not to be required CD38 expression on induced Treg.

thumbnail
Download:
Figure 5. Induction of CD38 on induced Treg cells is mildly enhanced by IC87114 and ATRA.

TGF-β-dependent conversion of CD4+CD25− cells in presence of different concentrations of ATRA and the PI3K inhibitors IC87114 or PI103. (A) Percentage of Foxp3+ cells after 3 days of stimulation with anti-CD3, TGF-β and indicated drugs. (B) Mean expression of CD38 on Foxp3+ cells induced in A. (C) Mean expression of CD73 on Foxp3+ cells induced in A.

https://doi.org/10.1371/journal.pone.0017359.g005

Discussion

The mechanism that underlies the impaired capacity of p110δD910A Treg is incompletely understood. We therefore sought to identify other genes that may help explain the reduced function of p110δD910A Treg. We found 83 probe sets that showed greater than two-fold increased expression in p110δD910A Treg and 42 probe sets that showed reduced expression. Given the importance of Foxo transcription factors for Treg development and the key role p110δ plays in the regulation of [30], we anticipated that Foxo target genes would be over-represented among the genes that were expressed at higher levels in p110δD910A Tregs. However, there were no candidates among the over expressed genes that are known to us to be Foxo targets. One possibility is that such genes might be more readily identified in arrays from acutely activated Tregs in which Foxo may be more completely excluded from the nucleus. Another consideration is that while PI3Ks are important regulators of Foxo transcription factors, many downstream signaling pathways controlled by PI3Ks regulate post-transcriptional, metabolic, cytoskeletal and other relevant cellular events [41]. Hence, inhibition of PI3K signaling can have profound effects on cellular function without directly affecting the rates of transcription of selected genes. It follows that some of the changes we did note in mRNA abundance may not reflect direct transcriptional control by Akt, Foxo and other proteins immediately linked to PI3K activity, but rather reflect broader changes in cell biology with knock-on effects on the transcription of selected genes.

The expression of CD38 by p110δD910A Tregs was shown to be reduced by the gene arrays, qRT-PCR and most importantly, at the level of protein expression. CD38 is a transmembrane glycoprotein which has both enzymatic activity and which acts as a receptor. The extracellular domain of CD38 acts on NAD(P) to generate a wide range of products including cyclic adenosine diphosphate ribose (cADPR), nicotinic acid-adenine dinucleotide phosphate (NAADP) and nicotinamide [42], [43], [44]. cADPR can act on ryanodine receptors on the endoplasmic reticulum to stimulate Ca2+ release; however its exact role is uncertain as it is generated extracellularly and cannot readily penetrate the membrane. It has been suggested previously that the major role of CD38 is to limit the availability of NAD+ to the mono-ADP-ribosyltransferase ART2 [45]. ART2, in the presence of NAD+, ribosylates P2X7 resulting in the rapid apoptosis of CD4+CD25+ cells [46], [47], [48]. CD38−/− mice show enhanced development of autoimmune diabetes in NOD/Lt and this effect is dependent on expression of ART2 [36]. We were unable to isolate Tregs from CD38−/− mice, presumably because the cells die during the preparation of the cell suspensions [48]. However, we observed no decrease in the survival of sorted CD38low cells over CD38high cells, despite the reduced viability of CD38−/− Treg in vitro. Thus, minimal levels of CD38 expression appear to be sufficient to prevent apoptosis.

CD31 (also known as PECAM) is a ligand for CD38 and in a parallel study, we found that CD31−/− Treg show reduced suppressive ability, suggesting that a potentially complex role for CD31–CD38 interactions in Treg function [49], . CD38 is found within rafts in close association with LAT and the intracellular domain can directly bind the SH2 domain of Lck [51]. Ligation of CD38 results in translocation of several important signaling proteins to those rafts, including SOS and p85. Hence, the possibility that CD38 transmits signals in Treg also needs to be considered.

Our results suggest that expression of CD38 on Treg cells is controlled, probably indirectly, by p110δ during the development of Treg in the thymus. Since Treg suppression potential correlated with CD38 expression levels in WT cells, but was unaffected on CD38+/− Treg, which expressed lower levels of CD38, we speculate that CD38 expression per se may not directly affect the potency of Tregs, but rather correlates with a yet-unidentified factor which promotes Treg-mediated suppression. Further research is required to fully understand the molecular basis for the impaired suppressive activity of p110δD910A Tregs. Nevertheless, our work identifies CD38 as a marker that may be used to purify highly suppressive Treg. This may be of use in clinical preparation of human Tregs, for instance, where isolation of Treg is made more challenging by the lack of cell surface proteins that unequivocally identify Tregs. Moreover, further mining of the dataset presented here may provide new leads in efforts to map genes that facilitate Treg-mediated suppression.

PI3K inhibitors are currently being developed for a variety of indications. Indeed, clinical trials have recently been initiated with CAL-101, an inhibitor that selectively blocks p110δ activation [52]. A potential consideration has been that inhibition of Treg may be detrimental for the treatment of autoimmune diseases although perhaps beneficial in the context of anti-cancer therapies. If the impaired ability of p110δ deficient Tregs to suppress is primarily due to a developmental lesion (as evidenced by the CD38low phenotype originating in the thymus), then the current results leave open the possibility that acute therapeutic inhibition of p110δ will not necessarily have an adverse effect on Treg function.

Supporting Information

Table S1.

qRT-PCR primers used in this study. The primers listed were used to determine expression of genes identified to be increased or decreased more than two fold by gene array analysis.

https://doi.org/10.1371/journal.pone.0017359.s001

(DOC)

Table S2.

Genes that were increased or decreased more than two-fold. The genes listed in this table were increased or decreased at least two fold in p110δD910A Tregs relative to WT Tregs. Column 1 shows the Affymetrix gene probe identifier. Column 2 shows the p value (only genes with p<0.01 are included in this table). The genes are sorted according to the log difference in gene expression. Genes that were more highly expressed in p110δD910A Tregds are listed in green and those that were expressed at lower levels in red.

https://doi.org/10.1371/journal.pone.0017359.s002

(XLS)

Acknowledgments

The authors would like to thank Simon Andrews for help with analysis of the microarray and the CD38 promoter, Adam Hales and Lewis Brown for animal husbandry, Juliet Emery, Amy MacQueen and Michelle Janas for help and advice.

Author Contributions

Conceived and designed the experiments: DTP WCR KO. Performed the experiments: DTP MDW DRS. Analyzed the data: DTP MDW DRS KO. Wrote the paper: DTP DRS KO.

References

  1. 1. Fontenot JD, Gavin MA, Rudensky AY (2003) Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 4: 330–336.
  2. 2. Hori S, Nomura T, Sakaguchi S (2003) Control of regulatory T cell development by the transcription factor Foxp3. Science 299: 1057–1061.
  3. 3. Chen W, Jin W, Hardegen N, Lei KJ, Li L, et al. (2003) Conversion of peripheral CD4+CD25− naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J Exp Med 198: 1875–1886.
  4. 4. Fantini MC, Becker C, Monteleone G, Pallone F, Galle PR, et al. (2004) Cutting edge: TGF-beta induces a regulatory phenotype in CD4+CD25− T cells through Foxp3 induction and down-regulation of Smad7. J Immunol 172: 5149–5153.
  5. 5. Daniel C, Wennhold K, Kim HJ, von Boehmer H (2010) Enhancement of antigen-specific Treg vaccination in vivo. Proc Natl Acad Sci U S A 107: 16246–16251.
  6. 6. Gottschalk RA, Corse E, Allison JP (2010) TCR ligand density and affinity determine peripheral induction of Foxp3 in vivo. J Exp Med 207: 1701–1711.
  7. 7. Shevach EM (2009) Mechanisms of foxp3+ T regulatory cell-mediated suppression. Immunity 30: 636–645.
  8. 8. Wing K, Onishi Y, Prieto-Martin P, Yamaguchi T, Miyara M, et al. (2008) CTLA-4 control over Foxp3+ regulatory T cell function. Science 322: 271–275.
  9. 9. Schmidt EM, Wang CJ, Ryan GA, Clough LE, Qureshi OS, et al. (2009) Ctla-4 controls regulatory T cell peripheral homeostasis and is required for suppression of pancreatic islet autoimmunity. J Immunol 182: 274–282.
  10. 10. Pandiyan P, Zheng L, Ishihara S, Reed J, Lenardo MJ (2007) CD4+CD25+Foxp3+ regulatory T cells induce cytokine deprivation-mediated apoptosis of effector CD4+ T cells. Nat Immunol 8: 1353–1362.
  11. 11. Kobie JJ, Shah PR, Yang L, Rebhahn JA, Fowell DJ, et al. (2006) T regulatory and primed uncommitted CD4 T cells express CD73, which suppresses effector CD4 T cells by converting 5′-adenosine monophosphate to adenosine. J Immunol 177: 6780–6786.
  12. 12. Maynard CL, Harrington LE, Janowski KM, Oliver JR, Zindl CL, et al. (2007) Regulatory T cells expressing interleukin 10 develop from Foxp3+ and Foxp3− precursor cells in the absence of interleukin 10. Nat Immunol 8: 931–941.
  13. 13. Asseman C, Mauze S, Leach MW, Coffman RL, Powrie F (1999) An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J Exp Med 190: 995–1004.
  14. 14. Collison LW, Workman CJ, Kuo TT, Boyd K, Wang Y, et al. (2007) The inhibitory cytokine IL-35 contributes to regulatory T-cell function. Nature 450: 566–569.
  15. 15. Green EA, Gorelik L, McGregor CM, Tran EH, Flavell RA (2003) CD4+CD25+ T regulatory cells control anti-islet CD8+ T cells through TGF-beta-TGF-beta receptor interactions in type 1 diabetes. Proc Natl Acad Sci U S A 100: 10878–10883.
  16. 16. Bynoe MS, Viret C (2008) Foxp3+CD4+ T cell-mediated immunosuppression involves extracellular nucleotide catabolism. Trends Immunol 29: 99–102.
  17. 17. Okkenhaug K, Vanhaesebroeck B (2003) PI3K in lymphocyte development, differentiation and activation. Nat Rev Immunol 3: 317–330.
  18. 18. Garcon F, Patton DT, Emery JL, Hirsch E, Rottapel R, et al. (2008) CD28 provides T-cell costimulation and enhances PI3K activity at the immune synapse independently of its capacity to interact with the p85/p110 heterodimer. Blood 111: 1464–1471.
  19. 19. Patton DT, Garden OA, Pearce WP, Clough LE, Monk CR, et al. (2006) Cutting edge: the phosphoinositide 3-kinase p110 delta is critical for the function of CD4+CD25+Foxp3+ regulatory T cells. J Immunol 177: 6598–6602.
  20. 20. Liu D, Zhang T, Marshall AJ, Okkenhaug K, Vanhaesebroeck B, et al. (2009) The p110delta isoform of phosphatidylinositol 3-kinase controls susceptibility to Leishmania major by regulating expansion and tissue homing of regulatory T cells. J Immunol 183: 1921–1933.
  21. 21. Oak JS, Deane JA, Kharas MG, Luo J, Lane TE, et al. (2006) Sjogren's syndrome-like disease in mice with T cells lacking class 1A phosphoinositide-3-kinase. Proc Natl Acad Sci U S A 103: 16882–16887.
  22. 22. Park SG, Mathur R, Long M, Hosh N, Hao L, et al. (2010) T Regulatory Cells Maintain Intestinal Homeostasis by Suppressing gammadelta T Cells. Immunity 33: 791–803.
  23. 23. Harada Y, Elly C, Ying G, Paik JH, DePinho RA, et al. (2010) Transcription factors Foxo3a and Foxo1 couple the E3 ligase Cbl-b to the induction of Foxp3 expression in induced regulatory T cells. J Exp Med 207: 1381–1391.
  24. 24. Ouyang W, Beckett O, Ma Q, Paik JH, DePinho RA, et al. (2010) Foxo proteins cooperatively control the differentiation of Foxp3+ regulatory T cells. Nat Immunol 11: 618–627.
  25. 25. Kerdiles YM, Stone EL, Beisner DL, McGargill MA, Ch'en IL, et al. (2010) Foxo transcription factors control regulatory T cell development and function. Immunity 33: 890–904.
  26. 26. Hedrick SM (2009) The cunning little vixen: Foxo and the cycle of life and death. Nat Immunol 10: 1057–1063.
  27. 27. Salmond RJ, Emery J, Okkenhaug K, Zamoyska R (2009) MAPK, phosphatidylinositol 3-kinase, and mammalian target of rapamycin pathways converge at the level of ribosomal protein S6 phosphorylation to control metabolic signaling in CD8 T cells. J Immunol 183: 7388–7397.
  28. 28. Basu S, Golovina T, Mikheeva T, June CH, Riley JL (2008) Cutting edge: Foxp3-mediated induction of pim 2 allows human T regulatory cells to preferentially expand in rapamycin. J Immunol 180: 5794–5798.
  29. 29. Battaglia M, Stabilini A, Roncarolo MG (2005) Rapamycin selectively expands CD4+CD25+FoxP3+ regulatory T cells. Blood 105: 4743–4748.
  30. 30. Merkenschlager M, von Boehmer H (2010) PI3 kinase signalling blocks Foxp3 expression by sequestering Foxo factors. J Exp Med 207: 1347–1350.
  31. 31. Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, et al. (1992) RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68: 855–867.
  32. 32. Cockayne DA, Muchamuel T, Grimaldi JC, Muller-Steffner H, Randall TD, et al. (1998) Mice deficient for the ecto-nicotinamide adenine dinucleotide glycohydrolase CD38 exhibit altered humoral immune responses. Blood 92: 1324–1333.
  33. 33. Okkenhaug K, Bilancio A, Farjot G, Priddle H, Sancho S, et al. (2002) Impaired B and T cell antigen receptor signaling in p110delta PI 3-kinase mutant mice. Science 297: 1031–1034.
  34. 34. Hill JA, Feuerer M, Tash K, Haxhinasto S, Perez J, et al. (2007) Foxp3 transcription-factor-dependent and -independent regulation of the regulatory T cell transcriptional signature. Immunity 27: 786–800.
  35. 35. Read S, Mauze S, Asseman C, Bean A, Coffman R, et al. (1998) CD38+ CD45RB(low) CD4+ T cells: a population of T cells with immune regulatory activities in vitro. Eur J Immunol 28: 3435–3447.
  36. 36. Chen J, Chen YG, Reifsnyder PC, Schott WH, Lee CH, et al. (2006) Targeted disruption of CD38 accelerates autoimmune diabetes in NOD/Lt mice by enhancing autoimmunity in an ADP-ribosyltransferase 2-dependent fashion. J Immunol 176: 4590–4599.
  37. 37. Peola S, Borrione P, Matera L, Malavasi F, Pileri A, et al. (1996) Selective induction of CD73 expression in human lymphocytes by CD38 ligation: a novel pathway linking signal transducers with ecto-enzyme activities. J Immunol 157: 4354–4362.
  38. 38. Gao Y, Camacho LH, Mehta K (2007) Retinoic acid-induced CD38 antigen promotes leukemia cells attachment and interferon-gamma/interleukin-1beta-dependent apoptosis of endothelial cells: implications in the etiology of retinoic acid syndrome. Leuk Res 31: 455–463.
  39. 39. Drach J, McQueen T, Engel H, Andreeff M, Robertson KA, et al. (1994) Retinoic acid-induced expression of CD38 antigen in myeloid cells is mediated through retinoic acid receptor-alpha. Cancer Res 54: 1746–1752.
  40. 40. Mucida D, Park Y, Kim G, Turovskaya O, Scott I, et al. (2007) Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 317: 256–260.
  41. 41. Hawkins PT, Anderson KE, Davidson K, Stephens LR (2006) Signalling through Class I PI3Ks in mammalian cells. Biochem Soc Trans 34: 647–662.
  42. 42. Cosker F, Cheviron N, Yamasaki M, Menteyne A, Lund FE, et al. (2010) The ecto-enzyme CD38 is a nicotinic acid adenine dinucleotide phosphate (NAADP) synthase that couples receptor activation to Ca2+ mobilization from lysosomes in pancreatic acinar cells. J Biol Chem 285: 38251–38259.
  43. 43. Lund FE (2006) Signaling properties of CD38 in the mouse immune system: enzyme-dependent and -independent roles in immunity. Mol Med 12: 328–333.
  44. 44. Lund FE, Cockayne DA, Randall TD, Solvason N, Schuber F, et al. (1998) CD38: a new paradigm in lymphocyte activation and signal transduction. Immunol Rev 161: 79–93.
  45. 45. Krebs C, Adriouch S, Braasch F, Koestner W, Leiter EH, et al. (2005) CD38 controls ADP-ribosyltransferase-2-catalyzed ADP-ribosylation of T cell surface proteins. J Immunol 174: 3298–3305.
  46. 46. Aswad F, Kawamura H, Dennert G (2005) High sensitivity of CD4+CD25+ regulatory T cells to extracellular metabolites nicotinamide adenine dinucleotide and ATP: a role for P2X7 receptors. J Immunol 175: 3075–3083.
  47. 47. Aswad F, Dennert G (2006) P2X7 receptor expression levels determine lethal effects of a purine based danger signal in T lymphocytes. Cell Immunol 243: 58–65.
  48. 48. Hubert S, Rissiek B, Klages K, Huehn J, Sparwasser T, et al. (2010) Extracellular NAD+ shapes the Foxp3+ regulatory T cell compartment through the ART2-P2X7 pathway. J Exp Med 207: 2561–2568.
  49. 49. Deaglio S, Aydin S, Grand MM, Vaisitti T, Bergui L, et al. (2010) CD38/CD31 interactions activate genetic pathways leading to proliferation and migration in chronic lymphocytic leukemia cells. Mol Med 16: 87–91.
  50. 50. Ma L, Mauro C, Cornish GH, Chai JG, Coe D, et al. (2010) Ig gene-like molecule CD31 plays a nonredundant role in the regulation of T-cell immunity and tolerance. Proc Natl Acad Sci U S A 107: 19461–19466.
  51. 51. Munoz P, Navarro MD, Pavon EJ, Salmeron J, Malavasi F, et al. (2003) CD38 signaling in T cells is initiated within a subset of membrane rafts containing Lck and the CD3-zeta subunit of the T cell antigen receptor. J Biol Chem 278: 50791–50802.
  52. 52. Lannutti BJ, Meadows SA, Herman SE, Kashishian A, Steiner B, et al. (2011) CAL-101, a p110{delta} selective phosphatidylinositol-3-kinase inhibitor for the treatment of B-cell malignancies, inhibits PI3K signaling and cellular viability. Blood 117: 591–594.