Regulatory Cell Populations in Relapsing-Remitting Multiple Sclerosis (RRMS) Patients: Effect of Disease Activity and Treatment Regimens

Multiple sclerosis (MS) is a demyelinating disease of the central nervous system (CNS) of autoimmune etiology that results from an imbalance between CNS-specific T effector cells and peripheral suppressive mechanisms mediated by regulatory cells (RC). In this research, we collected blood samples from 83 relapsing remitting MS (RRMS) patients and 45 healthy persons (HC), to assess the sizes of their RC populations, including CD4+CD25highFoxp3+ (nTregs), CD3+CD4+HLA−G+, CD3+CD8+CD28−, CD3+CD56+, and CD56bright cells, and how RC are affected by disease activity (acute phase or remission) and types of treatment (methylprednisolone, interferon, or natalizumab). In addition, we isolated peripheral blood mononuclear cells (PBMC) and cultured them with peptides mapping to myelin antigens, to determine RC responsiveness to autoantigens. The results showed decreased levels of nTregs in patients in the acute phase ± methylprednisolone and in remission + natalizumab, but HC levels in patients in remission or receiving interferon. Patients + interferon had the highest levels of CD3+CD4+HLA−G+ and CD3+CD8+CD28− RC, and patients in the acute phase + methylprednisolone the lowest. Patients in remission had the highest levels of CD3+CD56+, and patients in remission + natalizumab the highest levels of CD56bright cells. Only nTregs responded to autoantigens in culture, regardless of disease activity or treatment. The highest suppressive activity was exhibited by nTregs from patients in remission. In conclusion, in RRMS disease activity and type of treatment affect different RC populations. nTregs respond to myelin antigens, indicating that it is possible to restore immunological tolerance through nTreg induction.


Introduction
Multiple sclerosis (MS) is a chronic autoimmune/inflammatory disease of the central nervous system (CNS) that results in the demyelination of neurons leading to axonal loss and the accumulation of disability [1]. MS usually affects young adults aged between 20 and 40 years; women are affected at least twice as often as men [2]. The course of MS can follow four clinical patterns that include relapsing remitting MS (RRMS, which accounts for 80%-90% of MS cases at onset), tissue inflammation [31]. Natalizumab has been approved for the prophylactic treatment of RRMS patients who have failed to respond to first-line therapies or have severe, breakthrough disease [32,33]. Although natalizumab has been shown to affect immune cell responses in more ways than just preventing infiltration of T-cells into the CNS, very little is known regarding its effects on Tregs.
In this work we assessed the frequencies of peripheral blood CD4 + CD25 + Foxp3 + (nTregs), CD3 + CD4 + HLA − G + , CD3 + CD56 + (iNKT cells), CD8 + CD28 − (CD8 + Tregs) and CD56 bright cells in RRMS patients and healthy controls, to determine the effects of disease activity (acute or stable disease) and treatment regimens (methylprednisolone, interferon, natalizumab) thereof. In addition, we tested regulatory cell responses to various antigenic peptides mapping to myelin epitopes, to assess their function in patients at different states of the disease and under different therapies.

CD4 + CD25 high Foxp3 + T-Cells (nTregs) in RRMS
The levels of nTregs in peripheral blood of MS patients and HC, are shown in Figure 1. nTreg levels were lower in acute-phase patients with no treatment (AP-noRx) or under methylprednisolone (AP-MP), compared to all other patient groups and HC. nTregs were restored to normal levels in acute-phase patients under interferon β treatment (AP-IFN), whereas patients in remission under no treatment (Rem-noRx) showed intermediate levels between HC and AP-noRx. Patients in remission under natalizumab (Rem-NATA) had nTreg levels comparable to those of AP-noRx and AP-MP patients ( Figure 1D).

Other Regulatory Cell Populations in RRMS
The levels of the other regulatory cell populations studied in the peripheral blood of MS patients and HC, are shown in Figures 2-4. The percentage of CD3 + CD4 + HLA − G + T-cells was higher in AP-IFN patients compared to all other groups, with the difference reaching statistical significance between AP-IFN and HC, AP-MP, and Rem-NATA patients ( Figure 2).
The CD3 + CD8 + CD28 − T-cells (CD8 + Tregs) showed the same pattern, with the AP-IFN patients having the highest levels of CD8 + Tregs in all groups ( Figure 3). The opposite occurred with CD3 + CD56 + cells (iNKT cells), with the AP-IFN patients showing the lowest frequency ( Figure 4). The frequency of CD56 bright cells was higher in all patient groups compared to HC, but only the difference between Rem-NATA patients and HC was statistically significant ( Figure 4).

Effect of Culture with Peptides on Regulatory Cell Function
Peripheral blood mononuclear cells (PBMC) were isolated from RRMS patients (AP-noRx, n = 5; AP-MP, n = 5; AP-IFN, n = 6; Rem-noRx; n = 5, Rem-NATA, n = 14) and controls (HC, n = 19) and cultured in the presence or absence of peptides mapping to myelin antigens (see M and M), to assess which RC populations responded to autoantigens. It was observed that only nTregs responded to the peptides by proliferation and cytokine secretion, whereas the other RC populations studied did not respond to the peptides (data not shown).
To delineate the effects of disease state and treatment regimens on nTreg responses, the numbers of CD4 + CD25 − and CD4 + CD25 + cells were measured in cultures ± peptides, to calculate the ratio of effector to suppressor cells, as well as the percentage of CD4 + CD25 + Foxp3 + nTregs. The net change of effector/suppressor ratios between cultures ± peptides was also calculated, to assess any shift towards an effector or suppressor phenotype. Positive changes reflected an effector shift whereas negative a suppressor shift ( Figure 5, left column). These changes were accompanied by corresponding changes in cytokine ratios, with an increase in anti-inflammatory cytokines when a suppressor shift was observed ( Figure 5, right column).
Effector/suppressor ratios in cultures-peptides showed a reverse trend to that of nTregs in peripheral blood ( Figure 6). The higher ratios were observed in AP-noRx, AP-MP, and Rem-NATA patients, and the lower in AP-IFN patients ( Figure 6A). This observation was confirmed by a strong inverse correlation between effector/suppressor ratios and CD4 + CD25 + Foxp3 + T-cells (r = −0.64, p < 0.0001) ( Figure 6B).
Overall, all groups of RRMS patients displayed a suppressive phenotype during co-cultures with one or more antigenic peptides, similarly to the HC group. Peptides PEP2 and PEP3 were relatively more tolerogenic for HC subjects and PEP1 and PEP3 for RRMS patients (Figure 7).
Overall, all groups of RRMS patients displayed a suppressive phenotype during co-cultures with one or more antigenic peptides, similarly to the HC group. Peptides PEP2 and PEP3 were relatively more tolerogenic for HC subjects and PEP1 and PEP3 for RRMS patients (Figure 7).
When the percentages of co-cultures with a suppressive change in effector/suppressor ratio were calculated, no statistical differences between patient groups and HC were found, with the exception of the Rem-noRx patient group that showed a more than two-fold higher percentage of suppressive cultures compared to the other patient groups and HC ( Figure 8).                 When the percentages of co-cultures with a suppressive change in effector/suppressor ratio were calculated, no statistical differences between patient groups and HC were found, with the exception of the Rem-noRx patient group that showed a more than two-fold higher percentage of suppressive cultures compared to the other patient groups and HC (Figure 8).

Discussion
The role of Tregs in the pathogenesis of MS and other autoimmune diseases, is investigated by many groups, mainly because of their putative therapeutic potential [8][9][10]. In earlier years, the only marker used to differentiate Tregs from effector CD4 + T-cells was the constitutive expression of CD25 (IL-2Rα chain) [34], although it was already recognized that the pool of CD4 + CD25 + Tregs could include recently activated Teffs. However, CD4 + T-cells that express high levels of CD25 (CD25 high ) exhibit the highest suppressive activity and represent a small percentage (2%-4%) of CD4 + T-cells [34]. Earlier studies of CD4 + CD25 high Tregs in MS, showed similar frequencies of these cells in the peripheral blood between MS patients and healthy controls [17][18][19]28], but reported reduced suppressive activity using various functional assays [18,19,28]. In the CSF, compared to blood, CD4 + CD25 high Tregs were reported to be similar [19], or elevated [28] in MS patients. Subsequent studies used additional markers to characterize nTregs from the CD4 + CD25 high pool, the most important of which was the expression of Foxp3. Foxp3 is considered a specific marker for nTregs, although Teffs and Tr1 cells can transiently express Foxp3 upon activation [13], through STAT-5 signaling cytokines [12].
Other markers used for the characterization of Tregs include CD39 (an ectoenzyme that degrades ATP to AMP), CTLA-4 (Cytotoxic T-lymphocyte antigen 4, a CD28-family receptor) and GITR (Glucocorticoid-induced tumor necrosis factor receptor, a member of the TNF receptor superfamily) [23,35]. Huan et al. [20] reported identical levels of CD4 + CD25 + Tregs in RRMS patients and HC, but reduced expression of Foxp3 in RRMS patients. Reduced levels of CD4 + CD25 high Foxp3 + Tregs in the blood of RRMS (but not SPMS) patients, with an associated increase of Tregs in the CSF, were also observed by Venken et al. [22]. Michel et al. [36] reported that the CD4 + CD25 high pool of T-cells was depleted of cells expressing IL-7 receptor α-chain (CD127), a marker present on activated T-cells but not on Tregs. CD4 + CD25 high CD127 low Tregs exhibited similar suppressive functions in RRMS patients and HC. Dalla Libera et al. [23] reported decreased numbers of Tregs (defined by CD25, Foxp3, CD39, CTLA-4, and GITR expression) in RRMS patients during remission, which were restored to normal levels in the acute phase, concluding that Tregs are not involved in causing clinical attacks, but retain functionality and are increased during acute phase to restore homeostasis. Bjerg et al. [37] studied RRMS patients in remission and found that CD4 + CD25 high CD127 low Foxp3 + Tregs were higher than HC in the group of patients with less severe disease (lower EDSS score and shorter disease duration) and lower in patients with more severe disease.
Other studies addressed the issue of the effect of immunomodulating and immunosuppressive treatments on Tregs. One study with IFNβ-1α-treated MS patients, showed increased numbers of CD4 + CD25 high Tregs (CTLA-4 + and GITR + ) and functional enhancement after six months of treatment [38]. Other studies with RRMS patients treated with IFNβ (β-1α or β-1β), found improved frequency and function of CD4 + CD25 + Foxp3 + Tregs [27,28]. Natalizumab treatment was reported to have no effect on CD4 + CD25 high Foxp3 + Tregs frequency or activity [21,39], with the exception of one case study that reported increased numbers and restored function of nTregs in a patient treated with natalizumab after relapsing following stem cell transplantation [40]. Treatment with glucocorticoids after an acute attack was reported to induce an increase in CD4 + CD25 high [29] and Foxp3 + Treg frequency and function [30].
The marked diversity in the observations between various studies regarding population sizes, disease characteristics, comparators, Treg identification markers, and methodology used for Treg isolation and functional assays, make a direct comparison of the results very difficult. Our results showed decreased levels of CD4 + CD25 high Foxp3 + Tregs in patients in the acute phase (except when under IFNβ treatment) and an up-regulation during remission (except when under natalizumab). This is in contrast to the observations by Dalla Libera et al. [23] who reported lower levels during stable disease and restoration to control levels during acute clinical attacks. However, the authors did not clarify how many of their patients were under prophylactic treatment, if any, and they also used mRNA expression of Foxp3 and other Treg markers within the CD4 + T-cells, which are also expressed by activated Teffs and Tr1 cells [13]. Haas et al. [19] found decreased suppressive activity of CD4 + CD25 high Tregs in their RRMS patients in the acute phase. In a minor subset of patients entering remission, suppressive function of Tregs displayed an increase, though not statistically significant. The majority of the studies included patients in remission, so there was no acute phase arm to compare [17,18,20,22,37]. Reduced frequency of nTregs during the acute phase can be explained by their migration to sites of inflammation in the CNS, as previously suggested [22,41,42].
Our finding of increased frequency of nTregs under IFNβ treatment is consistent, since most studies verify it [22,24,[26][27][28]38,43]. The fact that our patients were in the acute phase of the disease indicates that IFNβ treatment per se increases nTreg numbers regardless of disease activity. In addition, the effector/suppressor cell ratio of the patients under IFNβ treatment was similar to the HC group (cf. Figure 6) and their nTregs retained their function (cf. Figure 8). Nevertheless, these events were not by themselves sufficient to protect these patients from relapsing, perhaps because higher numbers of nTregs and/or nTregs with higher suppressive potential are needed, compared to HC, to ameliorate the disease by counteracting the activity of the autoreactive clones.
We also found that treatment with natalizumab does not restore the frequency of nTregs, a finding corroborated by others [21,39]. Patients in remission under natalizumab had also a significantly higher effector/suppressor cell ratio compared to HC (cf. Figure 6), probably reflecting the inhibition of Teff infiltration into the CNS through the BBB, although their nTregs retained their function (cf. Figure 8). It appears that in this group of patients peripheral blood nTreg levels, per se, are not reflective of disease activity and remission is due to natalizumab activity [31] and, perhaps, the effect of other RC populations that were elevated (see below).
Regarding the effect of glucocorticoids (methylprednisolone), we did not find any increase in nTregs compared to untreated acute phase patients, as indicated by others [29,30]. The reduced frequency of nTregs in patients in the acute phase under methylprednisolone treatment is, perhaps, due to nTreg migration to sites of inflammation in the CNS [22,41,42] that is reversed when the patients enter remission. Again, in these patients nTregs retained their function (cf. Figure 8).
In our study we also examined the frequencies of other RC phenotypes, which may play a role in regulating autoimmune diseases. HLA − G + and CD8 + Tregs were elevated mainly in the AP-IFN group, whereas iNKT cells were markedly down-regulated in the same group. HLA − G + Tregs have been recently identified as a novel subset of naturally, thymus-derived regulatory T-cells, found in the blood and sites of inflammation to modulate inflammatory responses [14,44]. CD8 + Tregs are also important in immune regulation and treatment with glatiramer acetate, another first-line therapy for RRMS, has been shown to enhance their function [45]. Natalizumab-treated patients in our study, while having lower levels of Foxp3 + Tregs, showed higher levels of HLA − G + and CD8 + Tregs, as well as CD56 bright NK-cells, a subset of NK-cells with suppressive activity [46], the expansion of which is proposed as a mechanism of action of a new therapy for RRMS with a monoclonal antibody to CD25 (daclizumab) [47].
Our results from the functional assays showed that only nTregs, isolated from RRMS patients and HC, exhibited suppressive activity when cultured with various MOG or MBP peptides. This does not mean that the other RC populations studied in this work are not functional, merely that they do not respond to this type of antigenic stimulation. Under these culture conditions, nTregs from RRMS patients exhibited suppressive activity equal to nTregs isolated from HC or, as in the case of nTregs from patients in remission without therapy, higher than in HC. These results indicate that the reduction in nTreg suppressive activity reported in earlier studies [18,19] can be attributed to the type of stimulus used for the functional assays, and that it is possible to restore immunological tolerance through nTreg induction when suitable antigenic stimuli are employed.

Study Subjects
Eighty three patients with clinically definite MS of the relapsing-remitting type (RRMS), according to the 2005 revised McDonald's criteria [48], were included in the study. Historical data and neurologic examination were used to assess their disease course, current status, disease duration, and level of disability, according to the expanded disability status scale-EDSS [49]. The demographic and clinical characteristics of MS patients are presented in Table 1. The patients were divided into five groups based on the disease status (acute phase, AP, or remission, Rem) and the type of treatment they were receiving at the time their blood samples were drawn (no treatment, noRx; methylprednisolone, MP; interferon β, IFN or natalizumab, NATA). The groups were: (a) patients in the acute phase of the disease who were treatment-naive (AP-noRx, n = 13); (b) patients in the acute phase under treatment with methylprednisolone (AP-MP, n = 17); (c) patients in the acute phase under treatment with interferon β (AP-IFN, n = 12); (d) patients in remission receiving no treatment (Rem-noRx, n = 15); and (e) patients in remission under treatment with natalizumab (Rem-NATA, n = 26). To note, in patients with clinically definite RRMS the acute phase is defined as a clinical relapse, i.e., a new onset of a neurological dysfunction of the kind seen in MS that can be either a subjective report (symptom) or an objective observation (sign). Usually it is both, since a new symptom indicates a dysfunction that can be identified by the neurologist. The duration of the new symptom or sign has to be longer than 24 h, and usually lasts for days or weeks. It is not necessary to perform an MRI to verify the relapse; it is, per se, a clinical event.
Forty five age-and sex-matched, healthy individuals (Table 1), with no history of neurological or autoimmune disease and no concomitant signs or symptoms of infection or inflammation, served as a healthy control group (HC). All subjects signed an informed consent form before blood collection. The study protocol was approved by the Scientific Review Board and Ethics Committee of Patras University Hospital (Reg# 451/17.10.08, 17 October 2008). The Hospital abides by the Helsinki declaration on ethical principles for medical research involving human subjects.

Cells and Cultures
Whole blood samples (10 mL) were collected from MS patients and HC in heparinized BD Vacutainers (BD, Plymouth, UK). Peripheral blood mononuclear cells (PBMC) were isolated by ficoll (BIOCHROM AG, Berlin, Germany) density gradient centrifugation. The cells were cultured for 72 h in RPMI 1640 medium (GIBCO BRL, Gaithersburg, MD, USA), containing 10% fetal bovine serum and 1% penicillin/streptomycin at a concentration of 10 6 cells/mL, in the presence or absence of peptides PEP1-4 or cP7 (see Section 4.3), at a concentration of 10 pg/mL/10 6 cells. At the end of the culture period, PBMC were harvested and the culture supernatants were collected for the determination of cytokine concentrations (see Section 4.5). PBMC were washed with PBS and their numbers and phenotypes were determined by flow cytometry.

Peptides Mapping to MOG and MBP Myelin Antigens
The peptides used for the functional assays are shown in Table 2. Peptides PEP1-4 are MOG  antigenic epitopes prepared as described [50]. Peptide cP7 is an MBP 87-99 antigenic epitope prepared as described [51]. Table 2. Peptides used for the functional assays.

Peptide
Description Sequence For intracellular staining of Foxp3, the cells were labeled with CD4-FITC and CD25-PE antibodies and were stained intracellularly for Foxp3, after fixation and permeabilization. Whole blood samples were treated with BD Pharm Lyse buffer 1× (BD) for 15 min, after staining, to lyse red blood cells.
Flow cytometry was performed on an EPICS Coulter XL-MCL Flow Cytometer (BC). At least 20,000 events were acquired for extracellular and 100,000 events for intracellular staining. Data analysis was performed using the FlowJo V7.5 software (Tree Star Inc., Ashland, OR, USA).

Cytokines Measured in Culture Supernatants
Both inflammatory (IFN-γ, TNF-α, and IL-17A) and anti-inflammatory (IL-4 and IL-10) cytokines were measured in the supernatants of PBMC cultured with or without peptides. Measurement of cytokine levels was performed on a BD FACSArray Bioanalyzer using the cytometric bead array (CBA) assay (human Th1/Th2/Th17 Cytokine Kit, BD Biosciences). The ratio of anti-inflammatory/inflammatory cytokines was calculated in all cultures and the net change was determined to assess the shift towards a suppressor or effector phenotype after culture with each peptide.

Statistical Analysis
Comparisons between different groups of patients and controls were performed using the one-way analysis of variance (ANOVA) or the Kruskal-Wallis test, depending on the normality of the distribution of values. When the null hypothesis of the Kruskal-Wallis or ANOVA test was rejected, the Mann-Whitney test with Bonferroni correction was employed for the pairwise comparisons of the groups. The p values were calculated two-tailed and in all cases considered statistically significant if p was ≤0.05. Data were analyzed using the GraphPad Prism v.5.03 (San Diego, CA, USA).

Conclusions
In conclusion, in RRMS patients, different regulatory cell populations show a varied distribution according to the phase of the disease and treatment regimens. nTreg levels were lower in the patients in the acute phase of the disease before therapy or under methylprednisolone treatment, but were restored to HC-levels in patients in the acute phase receiving interferon. Patients in remission receiving no treatment had HC-levels of nTregs, whereas under natalizumab-treatment their nTreg levels were decreased to acute phase levels. Patients receiving interferon had the highest levels of CD3 + CD4 + HLA − G + and CD8 + Tregs, whereas patients in the acute phase receiving methylprednisolone the lowest. Patients in remission had the highest levels of iNKT cells and patients in remission under natalizumab treatment the highest levels of CD56 bright cells. nTregs, but not the other regulatory cells studied, exhibited suppressive activity when cultured with at least one myelin peptide antigen, i.e., they retained their function when they recognized their cognate antigen.