*2.3. Di*ff*erent Patterns of Distribution Of IL-17A, IFN*γ *And IL-4 CD4*<sup>+</sup> *T-Lymphocytes Are Observed in MTX Responder and Nonresponder Ra Patients*

In new-onset DMARD-naïve patients, we investigated the expression of IL-17A, IFNγ and IL-4 by circulating CD4+T-lymphocytes before and during the first 6 months of MTX treatment. We stratified the patients into two groups, which were defined according to the clinical response to MTX treatment attained after 6 months of treatment. There were 31 and 16 patients who met the criteria for responders and nonresponders to MTX, respectively. The patients were studied in parallel with 29 HCs. There were no significant differences in the number or frequency of circulating CD4+T-lymphocytes between MTX nonresponder and MTX responder RA patients and HCs at baseline (CD4+T cells: 821.99 <sup>±</sup> 187.91 vs. 1101.54 ± 260.11 cells/μL and 33.07 ± 3.46 vs. 36.22 ± 2.73%, responder vs. nonresponder patients, respectively).

Under basal conditions, MTX nonresponder patients showed a significantly increased number of CD4+IL-17A+T-lymphocytes with respect to responder patients, which could be explained by an expansion of the CD4+IL-17A<sup>+</sup>TN and CD4+IL-17A<sup>+</sup>TCM subsets (Figure 5b). After 6 months of treatment, there were no differences in the numbers of CD4+IL-17A+T-lymphocytes between nonresponder and responder RA patients, but the numbers of CD4+IL-17A<sup>+</sup>TN remained significantly increased in nonresponders. During treatment, MTX responder patients did not show significant modifications in the number of CD4+IL-17A+T-lymphocytes.

Under basal conditions, the number of CD4+IFNγ+T-lymphocytes was significantly increased in MTX nonresponder patients with respect to responders, which was due to an increase in the CD4+IFNγ<sup>+</sup>TN, CD4<sup>+</sup>IFNγ<sup>+</sup>TCM and CD4+IFNγ+TEM subset numbers, which were significantly reduced after 6 months of MTX treatment. In MTX responder patients, there were no significant modifications of the CD4+IFNγ+T-lymphocytes numbers during the 6 months of treatment follow-up (Figure 5a).

There were no significant differences in CD4+IL-4+T-lymphocyte numbers between MTX responder and nonresponder patients under basal conditions. However, after 6 months of treatment, there was a significant increase in the number of CD4+IL-4+T-lymphocytes as well as in the CD4+IL-4<sup>+</sup>TCM and CD4+IL-4<sup>+</sup>TEM subsets in nonresponder patients. In contrast, MTX responder RA patients showed a significant reduction in numbers of CD4+IL-4+T-lymphocytes and CD4+IL-4<sup>+</sup>TCM subsets (Figure 5c).

**Figure 5.** IFNγ, IL-4 and IL-17A-producing CD3+CD4<sup>+</sup>, Tnaïve, TCM, TEM and TE CD4<sup>+</sup> T-lymphocyte numbers in RA patients according to MTX response. Note: (**a**–**c**) Data represent numbers (nº/μL) of IFNγ, IL-4 and IL-17A-producing CD3+CD4<sup>+</sup>, TNaïve, TCM, TEM and TE CD4<sup>+</sup> T-lymphocytes according to the MTX response in ( ) responder and ( ) nonresponder RA patients. The dotted line represents the mean value recorded in healthy controls ( ). All values are expressed as the mean cell numbers (nº/μL) ± S.E.M. \*, *p* < 0.05 for responder or nonresponder RA patients vs. healthy controls; †, *p* < 0.05 for responders vs. nonresponders, σ *p* < 0.05 for 6 months of follow-up time vs. baseline. (**d**) Dot plots represent the percentages of IFNγ+, IL-17A<sup>+</sup> and IL-4<sup>+</sup> expression by CD4<sup>+</sup> T-lymphocytes after in vitro PMA stimulation in two representative situations: a nonresponder and a responder RA patient at baseline before MTX treatment.

Next, we investigated the expression and phosphorylation of STAT-1, STAT-3 and STAT-6 transcription factors on CD4+T-lymphocytes from MTX responder and nonresponder patients. Under basal conditions and at 3 and 6 months, there were no significant differences in either the percentage of phosphorylation or in the total protein in the different CD4+T-lymphocytes between both groups of patients (Figure 6). However, there were significant differences in the total expression of STAT-1 and STAT-3 and STAT-6 phosphorylation by CD4+T-lymphocytes from MTX nonresponder patients between basal conditions and after six months of treatment.

**Figure 6.** STAT-1, STAT-3 and STAT-6 phosphorylation in CD4<sup>+</sup> T-lymphocytes of RA patients according to MTX response. Note: Data represent the mean florescence intensity (MFI) and percentage (%) of total and phosphorylated proteins, respectively, on CD3+CD4<sup>+</sup>, TNaïve, TCM, TEM, and TE CD4<sup>+</sup> T-lymphocytes according to the MTX response in ( ) responder and ( ) nonresponder RA patients. The dotted line represents the mean value recorded in healthy controls ( ). All values are expressed as the mean cell numbers ± S.E.M. †, *p* < 0.05 for responders vs. nonresponders, σ *p* < 0.05 for 6 months of follow-up time vs. baseline.

Finally, we also investigated the serum levels of IL-17A, IFNγ and IL-4 in RA patients before and during the initial 6 months of treatment. Significantly increased levels of IL-17A were detected in MTX nonresponders with respect to responders under basal conditions and during the six months of treatment follow-up. Bot, MTX responder and nonresponder patients showed similar basal levels and significant reductions in IFNγ levels during the 6 months of MTX treatment. IL-4 serum levels were significantly increased in MTX nonresponders with respect to MTX responders before and during the 6 months of treatment (Figure 7).

**Figure 7.** IFNγ, IL-4 and IL-17A serum levels in RA patients according to MTX response. Note: Data represent the mean value of serum levels of IFNγ, IL-4 and IL-17A in ( ) nonresponder, ( ) responder RA patients and ( ) healthy controls. All values are expressed as the mean serum levels ± S.E.M.\*, *p* < 0.05 for responders or nonresponders vs. healthy controls; †, *p* < 0.05 for responders vs. nonresponders, σ *p* < 0.05 for 3 or 6 months of follow-up time vs. baseline.

#### **3. Discussions**

In this paper, we have shown that new-onset DMARD-naïve RA patients have abnormally functioning circulating CD4+T-lymphocytes with an expansion of the CD4+IL-17A+, CD4+IFNγ<sup>+</sup> and CD4+IL-17A+IFNγ+T subsets. This functional bias of the CD4<sup>+</sup> T cell population is associated with increased intracellular STAT-1 and STAT-3 stimulation and increased circulating levels of IFNγ and IL-17A. Furthermore, the pattern of IL-17+, IFNγ<sup>+</sup> and IL-4<sup>+</sup> CD4+T-lymphocytes production detected in new-onset DMARD-naïve RA patients could be modified by MTX treatment, and two different behaviors were identified in responders and nonresponders.

CD4+T-lymphocytes play a critical role in the pathogenesis of RA [3–6,8,28]. Heterogeneous results, demonstrating increased, unchanged or reduced numbers and/or percentages of Th1, Th2 and Th17 CD4+T-lymphocyte subsets in the circulation of RA patients have been reported [19–23,29–33]. This variability may be explained by different nonmutually exclusive mechanisms, including disease duration, previous and active DMARD and immunosuppressor treatments, concomitant diseases, the genetic and epidemiological backgrounds of the patients, cohort size, and the methodologies used to record different immune system variables. To minimize these potential interferences with the mechanisms directly associated with RA pathophysiology, we focused herein on a clinically homogeneous population of new-onset DMARD-naïve patients. Our data revealed increased numbers of circulating CD4+IFNγ<sup>+</sup> and CD4+IL-17A+T-lymphocytes in new-onset DMARD-naïve RA patients, but normal CD4+IL-4+T-lymphocytes. The frequency of Th17 cells was also increased, but the percentages of Th1 and Th2 lymphocytes were similar to those found in HCs. The differences in the results obtained using both methods of quantification indicate that the analyses of these CD4+T-lymphocytes subsets require the simultaneous study of numbers and percentages in RA patients. The numbers of circulating Th1, Th2 or Th17 cells appear to have special potential pathogenic relevance since they are a main source of IFNγ, IL-4 and IL-17A secretion [7,8]. In agreement with these cellular findings, the serum levels of IL-17A and IFNγ were increased, but those of IL-4 were normal in new-onset DMARD-naïve RA patients. Increased circulating IL-17A and IFNγ levels have been described in patients with early RA [34,35]. Interestingly, the numbers of circulating Th1, Th2 and Th17 cells showed differences under basal conditions and/or during MTX treatment between responder and nonresponder patients. Moreover, in agreement with a previous report, we found an expanded number of double IL-17A+IFNγ+CD4+T-lymphocytes in RA patients [23]. These different observations may contribute to understanding the established confusion concerning the normality or alteration of circulating Th1, Th2 and Th17 subsets in RA patients. Taken together, these findings improve knowledge of the involvement of CD4+T-lymphocytes in the early clinical stages of RA patients. Furthermore, this CD4+T-lymphocyte disturbance in RA patients cannot be ascribed to a

single Th subset since both Th1 and Th17 were expanded with increased levels of circulating IFNγ and IL-17A cytokines.

In addition to the pattern of cytokine production, CD4+T-lymphocytes are a heterogeneous population with different stages of differentiation/activation and patterns of circulation and tissue infiltration [11–15]. Interestingly, the number of CD4<sup>+</sup>TN able to express IL-17A<sup>+</sup> and IFNγ+IL-17A<sup>+</sup> was increased in new-onset DMARD-naïve RA patients. These data suggest an abnormal bias of nonantigen activated CD4+T-lymphocytes from these patients toward IL-17A production, which is also observed in antigen-promoted TCM CD4+T-lymphocytes. The relevance of the predisposition and acquired activation of CD4+T-lymphocytes to express cytokines is supported by the observation of the opposite results with respect to IFNγ production in naïve RA patients. The increasing numbers of IFNγ-producing CD4+T-lymphocytes were mainly focused in the CD4<sup>+</sup>TEM and CD4<sup>+</sup>TE subsets in new-onset DMARD-naïve RA patients. Different mechanisms might be involved in these different functional findings, including the intrinsic/genetic characteristics of the patients, the activating microenvironment and preferential extra-vascular tissue migration, such as the inflamed joints in naïve RA patients. It has been proposed that the CD14<sup>+</sup>highCD16<sup>+</sup> monocyte subset participates in the expansion of Th17 T-lymphocytes in RA patients [36,37]. There is also evidence supporting the relevance of the cytokine microenvironment in the differentiation of naïve T-lymphocytes into Th1, Th2 and Th17 subsets [7]. Furthermore, precursors such as CD4+CD161+T-lymphocytes may differentiate into either Th1 or TH17 lymphocytes based on the presence of IL-1γ and IL-23 or TGFγ [7]. It is possible that this plasticity might be involved in the observed expansion of CD4<sup>+</sup>TN able to express IL-17A<sup>+</sup> and IFNγ+IL-17A<sup>+</sup> in new-onset DMARD-naïve patients with RA. In contrast, the selective overexpansion of IFNγ in CD4<sup>+</sup>TEM and CD4<sup>+</sup>TE lymphocytes suggests the occurrence of antigen stimulation in the IL-12 microenvironment.

The relevance of the different signals driving Th lymphocyte activation appears to be critical because the percentages of phosphorylation of the transcription factors STAT-1 and STAT-3 were increased in the four different CD4 differentiation/activation stages in the DMARD-naïve RA patients. It is possible to suggest that early RA is associated with an intrinsic CD4+IL-17A<sup>+</sup>TN differentiation. However, antigen pressure and cytokines favor Th1 differentiation with a predominance of CD4<sup>+</sup>TEM and CD4<sup>+</sup>TE lymphocyte activation. In addition, it has been proposed that Th17 cells are unstable and easily shift toward Th1 cells, named "non-classic Th1 cells", and they have been found in early RA with relevant pathogenic activity [34]. In cord blood or spondyloarthropathies, abnormal Th1 and Th17 differentiation have been postulated in response to IL-1β and IL-23 [38,39]. Our data showed the involvement of both Th1 and Th2 subsets in early RA patients; however, their pathogenic role remains to be elucidated.

Our data revealed a heterogeneous function of CD4+T-lymphocytes in the early stages of RA. Analysis of the basal characteristics of new-onset DMARD-naïve patients who did not achieve a clinical response to MTX showed a significant expansion of CD4+IFNγ<sup>+</sup> and CD4+IL-17A<sup>+</sup> TN and TCM and CD4+IFNγ<sup>+</sup> TEM and TE lymphocytes with respect to those circulating in responders. Interestingly, during the 6 months of follow-up, MTX nonresponders maintained increased numbers of CD4+IL-17A<sup>+</sup>TN cells. Nevertheless, a normalization of CD4+IFNγ<sup>+</sup> cell subset numbers was observed with a concomitant increase in CD4+IL-4<sup>+</sup>TCM and CD4+IL-4<sup>+</sup>TEM cells. In addition, the significantly increased levels of circulating IL-17A persisted, but those of IFNγ were normalized, during the 6 months of MTX treatment in nonresponders. Thus, it is possible to suggest that the persistence of Th17 polarization during MTX treatment is associated with a defective response to treatment with this drug in early RA patients. The relevance of this Th17 differentiation is also supported by the normal numbers of CD4+IFNγ+, CD4+IL-17A<sup>+</sup> and CD4+IL-4<sup>+</sup> subsets in MTX responder patients. These results support the idea that these Th17 might be important in RA pathogenesis and in the response to immunomodulator treatments. There were no significant differences in the levels and phosphorylation of the STAT-1, STAT-3 and STAT-6 proteins in the different CD4+T-lymphocytes between responders

and nonresponders. Interestingly, the different behaviors of the CD4+T-lymphocytes compartment in both groups of patients cannot be ascribed to different disease activities before starting MTX treatment.

The precise mechanism of action of MTX in RA patients remains obscure [40]. It may act by decreasing cell proliferation, enhancing the rate of apoptosis, increasing endogenous adenosine concentrations, or altering cytokine production [41,42]. However, MTX is not a general antiproliferative drug; indeed, it induces apoptosis only in highly activated immune system cells [42,43]. The present data indicate that MTX causes different regulatory effects on CD4+T-lymphocytes in responders and nonresponders. This different CD4+T-lymphocyte modulation cannot be ascribed to a differential effect on the levels and phosphorylation of STAT-1, STAT-3 and STAT-6 proteins. The absence of a clinical response to MTX does not, however, rule out a biological effect of the drug on patient CD4+T-lymphocytes. Furthermore, the progression of uncontrolled diseases may be related to the expansion of CD4+IL-17A<sup>+</sup>TN lymphocytes. Determining whether this is the case is impossible since it would be unethical to maintain patients with active RA without treatment. These results support the knowledge of the relevance of an early immunomodulation in a subset of new-onset RA patients. Future works have to investigate the potential value of CD4+T-lymphocytes parameters as biomarkers in new-onset DMARD-naïve RA patients.

#### **4. Materials and Methods**
