Gene Expression Profiles of Human Phosphotyrosine Phosphatases Consequent to Th1 Polarisation and Effector Function

Phosphotyrosine phosphatases (PTPs) constitute a complex family of enzymes that control the balance of intracellular phosphorylation levels to allow cell responses while avoiding the development of diseases. Despite the relevance of CD4 T cell polarisation and effector function in human autoimmune diseases, the expression profile of PTPs during T helper polarisation and restimulation at inflammatory sites has not been assessed. Here, a systematic analysis of the expression profile of PTPs has been carried out during Th1-polarising conditions and upon PKC activation and intracellular raise of Ca2+ in effector cells. Changes in gene expression levels suggest a previously nonnoted regulatory role of several PTPs in Th1 polarisation and effector function. A substantial change in the spatial compartmentalisation of ERK during T cell responses is proposed based on changes in the dose of cytoplasmic and nuclear MAPK phosphatases. Our study also suggests a regulatory role of autoimmune-related PTPs in controlling T helper polarisation in humans. We expect that those PTPs that regulate T helper polarisation will constitute potential targets for intervening CD4 T cell immune responses in order to generate new therapies for the treatment of autoimmune diseases.


Introduction
CD4 T cells are important components of adaptive immune responses. During antigen stimulation, T cells polarise towards a type of effector cell specialised in controlling different sorts of infections by secreting different cytokines: Effector T helper 1 (Th1) secretes IFN and is specialised against intracellular pathogens, Th2 secretes IL-4 and is specialised against helminths, and Th17 secretes IL-17 and is specialised against extracellular bacterial and fungi. Despite having a crucial role in the immunity against pathogens, helper T cells are also involved in immune system-related diseases, including allergies and autoimmune pathologies. It is well established that Th2 responses mediate allergy and, currently, major efforts are directed to understand the pathological balance of Th1, Th2, and Th17 polarisation in autoimmune diseases [1][2][3][4].
In humans, protein tyrosine phosphatases (PTPs) constitute a family of more than 100 enzymes that regulate the phosphorylation state of molecular components of signalling networks. The folding of the PTP domain classifies PTPs in four classes: class I, containing the classical nonreceptor and receptor PTPs (NRPTPs and RPTPs, respectively) and the dual specific phosphatases (DSPs) [5]; class II, containing the low molecular weight PTP (LM-PTP); class III, containing cell division cycle-25 PTPs (CDC25s); and class IV, containing the eyes absent PTPs (EYAs) [6]. Catalytic activity of classes I to III is based on a Cysteine residue, while in the case of class IV it is based on an Aspartic acid residue [5,6]. Despite their important role in balancing phosphorylation levels, it is becoming clear that they also regulate intracellular signalling by mechanisms not dependent on the phosphatase activity, including the competition for the binding of inhibitors, like in the case of phosphatase of regenerating liver-1 (PRL-1) [7], the control of the spatial regulation of nonphosphorylated substrates, like in the case of MAPK phosphatases (MKPs) [8], and the control of the catalytic activity of other PTPs, like in the case of noncatalytic myotubularins (MTMs) [9]. These mechanisms underscore the relevance of the dose and the spatial 2 Journal of Immunology Research regulation of PTPs in the signalling networks that control cell responses.
Lymphocytes express around 60 to 70 genes coding for PTPs [10][11][12] and the significance of the above-mentioned regulatory mechanisms for the immune responses by human CD4 T cells has been barely established. Studying these mechanisms is needed in order to understand how CD4 T cells achieve normal immune responses while preventing diseases. In this regard, the critical role of some classical PTPs in lymphocyte activation and the association of genetic variants to autoimmune disease have been described [13,14]. Nonetheless the dose and the regulatory role of the majority of DSPs and class II to IV PTPs (for simplicity called in this study nonclassical PTPs or NCs) in T helper polarisation and effector function have not been studied. Here, we characterise the expression profile of the genes coding for these groups of PTPs in human naïve CD4 T cells, during the polarisation to Th1 effector cells and in response to PKC stimulation and cytosolic raise of Ca 2+ . Our data suggest that changes in the dose of MAPK phosphatases (MKPs) might dramatically affect the regulation of the MAPK module during T cell polarisation and stimulation at the inflammatory sites. Gene expression changes found in our study suggest the existence of previously nonnoted regulators of Th1 polarisation and effector functions and, consequently, potential targets for the manipulation of CD4 T cell immune responses in future research directed to obtain therapies for the treatment of autoimmune diseases.

Flow
Cytometry. Th1 polarisation was assessed by analysing the production of IFN in response to stimulation with PMA and Ionomycin. Th1 cells were stimulated as explained in Section 2.1 in the presence of 5 g/mL Brefeldin A (Sigma Aldrich, USA). Cells were then washed, fixed with 4% paraformaldehyde (Sigma Aldrich, USA) permeabilized with 0,1% saponin (Sigma Aldrich, USA), and stained with anti-IFN antibody (BD pharmingen, USA). Flow cytometry data were collected using a FacsCalibur flow cytometer (BD Biosciences, USA). (qPCR). RNA was extracted from naïve, Th1, and Th1-PI cells using the Absolutely RNA Microprep Kit (Agilent Technologies, USA) and the RNA integrity was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, USA). 2 g of RNA was used to synthesize cDNA with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA). Expression profiles were obtained by qPCR implemented with TaqMan Low Density Arrays (TLDA, ThermoFisher, USA) in a 7900HT Fast Real-Time PCR System (Applied Biosystems, USA). Genes with CT values under 33 were considered to be nonexpressed. Delta (D)CT was calculated by using as housekeeping gene the average of CT values obtained for the genes 18S and GNB2L1. RQ was calculated using the 2 −ΔΔCT method [48].

Data and Statistical
Analysis. An agglomerative hierarchical tree was implemented in MATLAB (The Mathworks, Inc, USA) by using Euclidean metrics and the ward method. Statistical analysis was implemented in GraphPad Prism version 5.04 (GraphPad Software, USA). Changes in mRNA levels between different conditions (Naïve, Th1, and Th1-PI) obtained for each donor were analysed with a paired -test. A consistent change in the expression level was considered when it was obtained a difference in expression equal or higher than 1,5 CTs in the majority of donors analysed and a value under 0.1 in the -test. In each data and statistical analysis, only those donors were used in which all the conditions that wanted to be compared were obtained ( = 3 donors in analysis of Table 2

Th1
Polarisation of Human Naïve CD4 T Cells. Human naïve CD4 T cells were isolated and polarised to Th1 conditions as detailed in materials and methods. The polarisation was confirmed by the production of IFN . The majority of cells in the population produced IFN in response to phorbol esters and Ionomycin (PI) treatment (Figure 1(a)). The induction of the Th1 master regulator transcription factor Tbet was also corroborated (Figure 1(b)). PKC activation and cytosolic raise of Ca 2+ , induced by PI treatment, mimic antigen stimulation via the T cell receptor (TCR) during T cell effector functions at inflammatory sites. By using this in vitro model, we studied the amount of mRNA, which indicates the dose of each PTP in naïve and Th1 effector cells.

Expression Profile of PTPs Associated with Human CD4 T Cells.
We included in this study all NCs given the unknown function of the majority of them in T cells and some classical RPTPs and NRPTPs, due to their regulatory role in the signalling downstream the TCR and cytokine receptors and in the dynamics of the cellular machinery or due to their  (Table 1). An agglomerative hierarchical tree and statistical analysis were used to characterise the obtained expression profiles (see materials and methods). Consistent with the high number of PTP-coding genes that are expressed in the lymphoid compartment in mice [11], the mRNA of all 14 classical PTPs and 55 out of 65 NCs was detectable in the analysed human naïve and Th1 cells. This fact underscored the relevant role of the family of PTPs in Th1 polarisation and function. Agglomerative hierarchical tree revealed clusters of PTPs, which share not only the expression level and profile during Th1 polarisation but also, in some cases, related functions ( Figure 2). The change in the expression profile during Th1 polarisation might be indicative of previously nonnoted regulators of T cell activation/ polarisation. . PTPRJ, PTPN6, and PTPN7, which are known to regulate intracellular phosphotyrosine levels during T cell activation [18,19,28,30,31], were upregulated with Th1 polarisation ( Table 2). Consistent with our data, upregulation of PTPRJ in response to TCR stimulation has been previously described [18]. Interestingly, mouse Shp1 limits IL-4 signals and then controls abnormal skewing to Th2 polarisation [49]. Therefore, our data indicate that SHP1 might also have a role in balancing Th1/Th2 polarisation in human CD4 T cells.

Expression of Classical
Our analysis uncovered a relation between the upregulated expression and the function of some PTPs. For example, PTPN9, which regulates the fusion of secretory vesicles with the plasma membrane [32,33], shared cluster with the myotubularin MTMR2 ( Figure 2), a known regulator of endosomal dynamics (see below). The higher gene expression found in Th1 cells ( Figure 2 and Table 2) suggests an important role of these PTPs in endosomal dynamics during the immune responses of Th1 cells. Another example was found inside the group of low expressed PTPs: PTPN18 and PTPN13 were found upregulated in the same cluster ( Figure 2). Both PTPs have been suggested to regulate (PTPN18) or associate with (PTPN13) the actin cytoskeleton [50,51]. Interestingly, they also shared cluster with the DSP   DUSP4  PTEN  PTPRA  PTPN22  PTPN2  MTM1  MTMR1  MTMR10  MTMR12  PTPN12  PTPN4  CDC14A  PTP4A1  DUSP11  MTMR6  MTMR3  MTMR14  DUSP2  SSH2  DUSP1  CDC25B  PTPRJ  PTP4A2  DUSP7  PTPN6  PTPN7  DUSP16  PTPN1  STYX  MTMR4  SBF1  PTPRC  CDC14B  DUSP18  EPM2A  PTPRK  DUSP19  PTPN13  SSH3  PTPN18  TENC1  TNS1  TPTE2  CDC25A  CDKN3  DUSP22  MTMR2  PTPN9  DUSP23  DUSP5  DUSP14  DUSP3  FAN1  DUSP6  MTMR8  PTP4A3  PTPDC1  STYXL1  SBF2  DUSP10  DUSP12  RNGTT  SSH1  EYA3  MTMR9  DUSP8   0  5  5   10 20 20 35 * * * * * * * * * * * * * * * * * * * * * Th1 Naïve Figure 2: Agglomerative hierarchical tree of the gene expression patterns in naïve and Th1 cells. Numbers below the tree indicate the distance among gene patterns. Hitmap represents the average DCT obtained for each gene in both conditions and 3 donors. The calibration bar is shown between 5 and 20 DCTs. Green and red squares point to clusters of upregulated and downregulated genes, respectively. Asterisks indicate those genes whose expression levels were considered to significantly change, as detailed in Table 2, and explained in materials and methods. Clusters are indicated of high, middle, and low expression.
Slingshot-3 (SSH3), another regulator of the actin cytoskeleton. Thus, the function of PTPN18 and PTPN13 in the actin dynamics subjacent T cell activation should be investigated. Consistent with our data, these PTPs have also been found to be expressed in T cells of mice [11]. Interestingly, a regulatory role of PTPN13 in Th1 and Th2 polarisation has also been proposed [35].
The genes PTPN1, PTPN2, PTPN4, PTPN12, PTPN22, PTPRA, and PTPRK were not regulated with Th1-polarising conditions, although the function of some of them in TCR or cytokine signalling has been described ( Table 1).

Expression of NCs.
We did not detect mRNA of the atypical MKPs DUSP13, DUSP15, DUSP26, and DUSP27,  DUSP1  DUSP2  DUSP4  DUSP5  DUSP6  DUSP7  DUSP8  DUSP10  DUSP16  STYXL1  SSH1  SSH2  SSH3  CDC14A  CDC14B  CDKN3  CDC25A  CDC25B  CDC25C  MTM1  MTMR1  MTMR2  MTMR3  MTMR4  MTMR6  MTMR8  MTMR9  MTMR10  MTMR11  MTMR12  MTMR14 Average DCT change −6,00 −4,00 −2,00 * * * * * * * * * * * * * * the classical MKP DUSP9, the tensin homolog TPTE, the eyes absent EYA1, EYA2, and EYA4, and the myotubularin MTMR7. The reported expression of the Eya1, Eya2, and Eya3 mouse orthologs [11] suggests a different requirement of this group of PTPs in mice and humans. The same might apply for the myotubularin MTMR7, which seems to have a regulatory role in Th polarisation in mice [53]. The expression of tensin homologs was very low and only PTEN was highly expressed particularly in Th1 cells (Figure 2). Although very lowly expressed, TPTE2 was consistently downmodulated with Th1polarising conditions ( Figure 2 and Table 2). Nonexpressed MKPs in our study matched previous data in mice [11]. 22 NCs were found in the group of highly expressed PTPs ( Figure 2). Substantial changes in expression levels associated with Th1 polarisation were found in genes coding for regulators of the phosphorylation state of phosphoinositides (MTMs), the MAPK signalling module (MKPs), the actin cytoskeleton (Slingshots or SSHs), and the cell cycle (CDC25s and CDKN3) ( Table 2). Interestingly, the role of some of these enzymes, including MTMR1, MTMR2, DUSP7, DUSP8, DUSP23, STYXL1, and CDC25B, has not been studied in T cells. To investigate the expression profile of these different groups of PTPs during the effector functions of Th1 cells at inflammatory sites, we analysed the change in expression levels induced by the PI treatment ( Figure 3).
MTMs dephosphorylate the position 3 of phosphatidylinositol phosphate (PIP) molecules PI(3)P and PI(3,5)P 2 , making them important regulators of the endosomal compartment, the cytoskeleton, and ion channels [9]. Eight out of 12 MTMs were found in the group of highly expressed PTPs (Figure 2), which suggests that they finely tune PIP levels for a proper function of these components of the cellular machinery in T cells. For example, MTMR14 controls the activity of the Ryanodine Receptor (RyR), which is essential for the homeostasis of Ca 2+ [9]. RyR is needed for Ca 2+ signalling and proper IL-2 production and proliferation of activated T cells [54]. MTMR14 shared cluster with MTMR6 ( Figure 2), which has been described to control naïve CD4 T cell activation by inhibiting the KCa3. 1K + channel, which is essential for Ca 2+ influx [55]. Consistently, a slightly higher mRNA level was found in naïve T cells than in Th1 cells. Other MTM of this cluster was MTMR3, which, along with MTMR6, has been described to regulate autophagy [9], an essential process for the metabolic changes required during T cell activation [56]. Among the highly expressed MTMs, only MTMR1 was found significantly upregulated ( Table 2) and shared cluster with MTM1 (Figure 2), which was also weakly upregulated. MTM1 interacts with desmin [57] and might have an important role in cytoskeleton dynamics in T cells.
We also found phosphatase death (PD) MTMs expressed in CD4 T cells, including the highly expressed SBF1 (MTMR5), MTMR10, and MTMR12, the middle expressed MTMR9, and the inducible with Th1 polarisation MTMR11 ( Figure 2 and Table 2). PD MTMs have been found to physically interact and increase the catalytic activity of MTMs [9]. For example, MTMR9 regulates the activity of MTMR6 [9] and MTMR8, a previously described regulator of the PI3K/AKT pathway in zebrafish [58]. MTMR9 might constitute an important regulator of CD4 T cell polarisation as has been proposed in mice [53]. The PI treatment upregulated MTMR11 levels in Th1 cells (Figure 3), suggesting a previously unknown regulatory role in effector functions at inflammatory sites. The MTMs regulated by MTMR11 have not been described.
Inside the group of middle expressed PTPs, MTMR2 was found upregulated by Th1-polarising conditions and PI treatment (Table 2 and Figure 3). These data suggest a previously unknown role of this MTM in T cells during Th1 immune responses. MTMR2 catalytic activity is increased by the PD MTMs SBF1, MTMR12, and SBF2 (MTMR13). Although none of these MTMs were found upregulated upon polarisation, the expression level of SBF1 and MTMR12 seems to be enough to allow this regulatory mechanism to take place in naïve and Th1 cells. MTMR2 dephosphorylates PI(3,5)P, which regulates membrane homeostasis and endosomal transport and has been described to interact with Disc large-1 (Dlg-1), which is involved in polarised membrane trafficking [59]. Recently, it has been proposed that ezrin controls tubulin cytoskeleton dynamics, immunological synapse organization, and NFAT activation by interacting with Dlg-1 [60]. Thus, it is tempting to speculate that MTMR2 has a regulatory role in the dynamics of the endosomal compartment during the activation of Th1 cells at inflammatory sites.
The cytoskeleton regulators SSH2 and SSH3 were found downregulated and upregulated, respectively, by Th1polarising conditions ( Table 2). These data underscore the relevance of SSH3 for the cytoskeleton rearrangements during Th1 polarisation or effector function. Despite both proteins being downmodulated by PI treatment (Figure 3 expressed SSH1, in the intracellular signalling operating during antigen-induced T cell stimulation is not completely understood [61]. Among the 9 out of 10 classical MKPs [8] whose expression was detected, DUSP1, DUSP2, DUSP4, DUSP7, and DUSP16 were found in the group of highly expressed PTPs (Figure 2). Nuclear DUSP1 has been found to be required for proper T cell activation [40] and nuclear/cytoplasmic DUSP16 has been proposed to be downregulated and upregulated during Th1 and Th2 polarisation, respectively, being, in mice, a regulator of the Th1/Th2 balance [41]. Consistently, the expression levels of DUSP16 were found downmodulated during Th1-polarising conditions ( Table 2), suggesting a role in balancing human T helper differentiation or in initial T cell immune responses by naïve cells. The cytoplasmic ERKspecific MKP DUSP7 was upregulated with Th1-polarising conditions and shared cluster with the SHP1 and the ERK regulator He-PTP (Table 2 and Figure 2). Interestingly, its function and regulatory mechanism in T cells are not known, and our data might indicate a role of DUSP7 in Th1 polarisation or effector function. The nuclear ERKspecific MKPs DUSP2 and DUSP4 have been described to interact with phosphorylated and dephosphorylated ERK and were upregulated by the PI treatment ( Figure 3). Thus, they might be involved in both ERK inactivation and/or accumulation of the dephosphorylated form in the nucleus (see below). DUSP5, DUSP6, DUSP8, and DUSP10 were found inside the group of middle expressed PTPs. Nuclear DUSP5 has been proposed to modulate T cell development and activation [62] and Cytosolic DUSP6 has been proposed to decrease T cell sensitivity by dephosphorylating ERK and, consequently, inhibiting the Lck-ERK positive feedback loop established upon strong T cell stimulation [63]. Interestingly, DUSP5 shared cluster with genes upregulated with polarisation ( Figure 2) and both MKPs were upregulated by the PI treatment ( Figure 3). The function of the p38-and JNK-specific DUSP8 and DUSP10 in T cell immune responses is not known. DUSP8 was found substantially more abundant in naïve than in Th1 cells ( Figure 2) and was upregulated with PI treatment (Figure 3). DUSP10 was not regulated by Th1-polarising conditions or PI treatment.
The group of MKPs are characterised by sharing MAPK substrates. For example, ERK is dephosphorylated by 13 different MKPs. Interestingly, it has been proposed that the spatial distribution of dephosphorylated MAPKs is regulated by the binding of MKPs, such as the nucleocytoplasmic location of ERK by nuclear DUSP5 and cytoplasmic DUSP6, and the accumulation of ERK in the nucleus by DUSP2, DUSP4, and DUSP5 [64][65][66]. Thus, the coordinated expression levels of MKPs are expected to be essential for the proper function of the MAPK signalling module. Dephosphorylated MAPKs can have functions nondependent on the kinase activity, for example, as transcription factors [8], and they bind with high affinity and increase the activity of MKPs [67]. Thus, more than simply having a role in down-modulating the response of the module, MKPs regulate the subcellular localisation and crosstalk of MAPKs [8,65]. In this regard, changes in the dose, as may be achieved by regulating their expression levels, may enable MKPs to compete with other molecules for the binding of MAPKs.
Consistent with this complex scenario, Th1 polarisation and PI treatment of Th1 cells induced dramatic changes in the expression profile of ERK-directed MKPs and, consequently, remarkable differences were found between naïve and Th1 restimulated cells (Th1-PI) (Figures 4(a) and 4(b)). While the relative expression of nuclear DUSP1 and DUSP2 was dominant in naïve cells, in Th1 restimulated cells there was a clear dominance of partners of dephosphorylated ERK, including DUSP2, DUSP4, and DUSP5 in the nucleus, and DUSP6 in the cytoplasm. These data suggest that interactions between MKPs and dephosphorylated ERK might accumulate ERK in the cytoplasm and/or the nucleus of restimulated Th1 cells, while in naïve cells nuclear translocation and transient phosphorylation of ERK should dominate the response (Figure 4(c)). DUSP7 was also more abundant in restimulated Th1 cells (Figure 4(b)). Whether it can also bind dephosphorylated ERK should be investigated. Thus, functions nonrelated to its kinase activity of dephosphorylated ERK in the cytoplasm and the nucleus of effector T cells at inflammatory sites and the regulatory role of MKPs in the spatial distribution of dephosphorylated ERK should be investigated. Dynamic compartmentalisation of MAPK signalling module by MKPs warrants future research. Some atypical MKPs were found regulated by Th1polarising conditions ( Table 2). The JNK-and ERK-specific DUSP22 and DUSP23 were found upregulated in our analysis, suggesting an important role in controlling intracellular signalling during Th1 polarisation or effector function. DUSP22 has been proposed to regulate TCR signalling [43] while the function of DUSP23 in T cell activation is not known. The JNK-and p38-specific DUSP13 and DUSP21 (whose substrates are unknown) were repressed with polarisation. The function of these proteins in T cell biology is also unknown.
Among cell cycle regulators, CDC14A and CDC25B were found in the group of highly expressed PTPs. Interestingly, CDC25B, a negative regulator of the cell cycle was upregulated by Th1-polarising conditions ( Table 2). CDC25A and CDKN3 constitute a cluster of strongly upregulated cell cycle regulators ( Figure 2) and CDC25C was induced with Th1-polarising conditions ( Table 2). These data suggest a relevant function of these enzymes for Th1 polarisation, proliferation, or effector function. Consistently, inhibition of T cell proliferation by PD-1 is mediated by suppression of CDC25A [42]. Whether there is a role of these cell cycle regulators during antigen-mediated T cell stimulation, as has been recently described for the CDC25B-specific kinase Aurora A [68], should be investigated.

Conclusions: Perspective on Autoimmune Diseases
The systematic analysis performed in this study reveals several genes coding for PTPs that are regulated during Th1 polarisation and restimulation of effector cells. Interestingly, the mRNA level of the majority of the regulated genes coding for NCs was increased during Th1 polarisation, which suggests a regulatory role of these PTPs in Th1 polarisation or effector function. By contrast downmodulated genes during polarising conditions might be involved in initial immune responses by naïve T cells. In general, changes in expression levels found during polarisation might indicate a role in achieving a healthy balance of T helper polarisation. Our data also suggest an important compartmentalisation of dephosphorylated ERK functions during the T cell responses at inflammatory sites. Finally, the obtained results also suggest the existence of PTPs that might regulate components of the cellular machinery (including the endosomal compartment, the cytoskeleton, and the activity of ion channels) during T cell immune responses. The regulatory role of these PTPs should be investigated. Regarding autoimmunity, the expression of PTPN6 has been found reduced in psoriatic T cells [29] and in blood cells of psoriatic arthritis [69]. Our data suggest that balanced nonpathological T helper polarisation requires a highly upregulated expression of SHP1 during the generation of IFN producing cells. By contrast, although the polymorphism rs1893217(C) in the PTPN2 gene has been associated with a decrease in the expression levels that promotes autoimmunity [23,70] and to a reduced response to IL-2 [26], the regulation of the expression levels of this gene does not seem to be necessary during the generation of IFN producing cells. Single nucleotide polymorphisms (SNPs) in nonregulated PTPs PTPN22 and PTPRC have been associated with human autoimmune diseases, including multiple sclerosis and autoimmune hepatitis in the case of PTPRC and type I diabetes, rheumatoid arthritis, and systemic lupus erythematosus in the case of PTPN22 [13,23,37,52].
In the mouse model the atypical DUSP22 has been proposed to inhibit TCR signalling and to control the development of experimental autoimmune encephalomyelitis [44]. Interestingly, low expression of DUSP22 in human T cells has been proposed to be a potential biomarker for systemic lupus erythematosus nephritis [45]. Thus, it seems that this protein is relevant in controlling T cell responses in order to prevent autoimmune diseases. Interestingly, overexpression of DUSP23 in CD4 T cells of patients with systemic lupus erythematosus has been described [46]. The upregulation of DUSP22 and DUSP23 found in our work by Th1-polarising conditions suggests a regulatory role of these enzymes during T helper polarisation and, consequently, their contribution to the control or development of pathological polarisation should be investigated. In general, our data encourage comparing the expression profile of PTPs in T cells of patients diagnosed with autoimmune diseases and healthy individuals. Some of the genes regulated during Th1 polarisation as assessed herein might also constitute biomarkers or might be involved in the pathogenesis or severity of autoimmune diseases.

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.