Lower Frequency of CD62Lhigh and Higher Frequency of TNFR2+ Tregs Are Associated with Inflammatory Conditions in Type 1 Diabetic Patients

Diabetes type 1 is a chronic autoimmune disease in which insulin-producing cells are gradually destroyed by autoreactive T cells. Human regulatory cells play important role in controlling autoimmunity, and their qualitative or quantitative dysfunctions may result in ineffective suppression of autoreactive T cells. CD62L is a surface molecule that plays role in homing capabilities of Tregs, and only cells with high expression of CD62L have high suppressive potential. Tregs are also characterized by the constant expression of TNFR2. The frequency of Tregs carrying TNFR2 is higher in inflammatory conditions. We investigated blood regulatory T cells with CD62L expression and regulatory T cells expressing TNFR2 in type 1 diabetic patients. We found differences in these populations when comparing to healthy individuals. We propose that these may be associated with inflammatory conditions that are present in patients with type 1 diabetes. The lower percentage of Tregs and Treg CD62Lhigh may contribute to ineffective suppression of proinflammatory cytokines production during type 1 diabetes.


Introduction
Diabetes type 1 is a chronic autoimmune disease that is characterized by the destruction of insulin-producing cells. It has been shown that this disorder involves failure in immune regulation [1,2]. NOD mice and patients with type 1 diabetes have deficiencies in at least two T-cell populations with regulatory properties: NKT and CD4 + CD25 + [3][4][5]. Human CD4 + CD25 + regulatory T cells, that are characterized by high expression of CD25 molecule-comprise 5-10% of peripheral CD4 + T cells [6,7] and are characterized by the upregulation of the IL-2receptor alfa-chain (CD25) and expression of the transcription factor Foxp3 [8,9]. Some data suggest that regulatory T cells are most effective in preventing priming of naive cells [10]. This requires their presence in lymph nodes. The receptor that homes to lymph nodes is CD62L molecule (L-selectin). Lower expression of CD62L was observed on CD3 + lymphocytes from diabetic type 1 patients in comparison to their healthy counterparts [11]. In addition, the low percentage of CD4 + CD25 + CD62L + and their poor efficacy in preventing autoimmunity was observed in NOD mouse [12][13][14][15]. Another surface molecule which seems to be important for regulatory T cells is the TNF receptor type 2 (TNFR2). Its constant expression was observed on Tregs [16,17]. It was also shown that exposure to TNF enhanced the percentage of Tregs expressing TNFR2 and downmodulated their suppressive activity at the same time [17]. Chronic inflammation present in autoimmune environment may impair the number and/or function of Tregs, which then are not able to prevent the activity of inflammatory cells. As inflammation proceeds, the risk for vascular complications increases [18,19].
Here, we demonstrate that patients with type 1 diabetes have lower percentage of peripheral blood regulatory T cells with CD62L expression and higher percentage of regulatory T cells expressing TNFR2 than healthy individuals.   We propose that these differences may be associated with inflammatory conditions that are present in patients with type 1 diabetes. was approved by The Ethics Committee of The Medical University of Gdańsk. Clinical characteristics of the patients are presented in Table 1.

Determination of TNF Level.
Serum level of TNF was measured by immunoenzymatic ELISA method (Quantikine High Sensitivity Human by R&D Systems Inc, USA) according to the manufacturer protocol. Minimum detectable concentrations were determined by the manufacturer as 0.12 pg/mL. Data are shown as mean ± standard deviation.

Statistical Analysis.
All statistical analyses were performed using Statistica 8.0 (StatSoft, Inc USA). The differences between the groups were calculated with the nonparametric U-Mann Whitney tests. Multiple regression analysis was performed to determine if CD4 + CD25 high T cell level depends on age and disease duration in diabetic group. Spearman's correlations were used to compare cell frequencies with analyzed parameters. P values less than.05 were considered statistically significant.

Peripheral
Blood CD4 + CD25 high , CD4 + CD25 high CD62L high , and CD4 + CD25 high TNFR2 + T Cells. The multiple regression analysis showed no association between the peripheral blood level of CD4 + CD25 high regulatory T cells and age as well as disease duration in diabetic group (β = [−0.04] and [−0.37], resp., P = .2). Children with type 1 diabetes were characterized by lower number and percentage of the CD4 + CD25 high T cells ( Table 2, P = .00084, P = .00026), lower number and percentage of the CD4 + CD25 high CD62L high ( Table 2, P = .00012, P = .0000) than their healthy counterparts. When numbers of the CD4 + CD25 high TNFR2 + T cells were compared, the opposite effect was revealed. Tregs expressing TNFR2 were more frequent in DM1 group; however, this was true only when the percentage of cells was taken into account ( Table 2, P = .0000).
As to the expression of CD62L and TNFR2 in CD4 + CD25 high CD62L high and CD4 + CD25 high TNFR2 + cells, respectively, flow cytometric analysis indicated that there was a significant difference in the MFI of these molecules between analyzed groups. The expression of CD62L was higher in control group (Figure 2, P = .0001), while the expression of TNFR2 was higher in DM1 group (Figure 2, P = .0004).

The Relationship between CD4 + CD25 high Regulatory T-Cell Subpopulation Frequencies and Serum TNF Level in
Diabetic Type 1 Children. We then studied the association between serum level of TNF and the percentage of peripheral blood regulatory T cells in analyzed DM1 subjects. We found that diabetic type 1 children with higher percentage of CD4 + CD25 high TNFR2 + cells among peripheral blood CD4 + CD25 high had higher serum level of TNF ( Figure 5, r = 0.53; P < .05). When analyzing CD4 + CD25 high and CD4 + CD25 high CD62L high subpopulations, we did not find any association between frequency of these cells and TNF serum level.

Discussion
Some authors have previously suggested that patients with type 1 diabetes have lower frequency of regulatory T cells in peripheral blood [5,21]. Our work clearly revealed that children with type 1 diabetes are characterized by lower percentage and absolute number of CD4 + CD25 high regulatory T cells as compared with age-matched healthy individuals. These quantitative defects may be caused by the chronic inflammation, which is observed in diabetic type 1 patients [18,19,22] and was observed in our study group. The analyzed patients had high serum level of CRP and detectable level of TNF, which are indicators of chronic inflammatory response [18,19,22]. The strong correlation was observed in our study group; patients with lower CRP level had higher percentage of CD4 + CD25 high cells among peripheral blood lymphocytes than patients with higher CRP level. Some data concerning regulatory T cells have shown that these cells are susceptible to the effects mediated by the proinflammatory cytokine-TNF [23], which may impair their activity through TNFR2 signalization [17].Our results are consistent with the findings of Valencia et al., who demonstrated that CD4 + CD25 high Tregs from patients with active RA contained greater percentage of TNFR2 + cells compared with healthy controls [17]. Our studies gave similar results showing that the CD4 + CD25 high TNFR2 + T cells from DM1 patients had higher expression of TNFR2 than cells from the control group. Moreover, our studies revealed the correlation between percentage of CD4 + CD25 high TNFR2 + lymphocytes and serum level of HbA1c as well as CRP. Patients with poor metabolic control and higher CRP level had higher percentage of Tregs carrying TNFR2. These patients also produced more TNF than those with lower percentage of regulatory T cells expressing TNFR2. The observed greater CD4 + CD25 high TNFR2 + frequency in DM1 patients may reflect the greater sensitivity of this cell subset to inflammatory conditions and the action of TNF.
In view of the fact that in case of diabetes only subpopulation of Tregs that carries CD62L has high suppressive activity [13,15], we can suspect that the lower percentage of CD4 + CD25 high CD62L high cells among peripheral blood Tregs as well as the expression of CD62L on CD62L high Tregs in DM1 group may be associated with defect in the homing capabilities of this cell population, as CD62L is responsible for migration of Tregs to secondary lymphoid tissues [13]. In addition, when analyzing CD4 + CD25 high CD62L high subset, we found that patients with lower percentage of these cells had higher serum CRP level, in which elevated level may reflect an inflammatory state [24]. They also had higher HbA1c level which is an indicator of metabolic control in diabetic patients [25]. Higher HbA1c is associated with higher serum TNF level [26,27], so more intense inflammatory response. This may further cause disruption in regulatory T-cell population. The lower numbers of CD62L high Treg subpopulation in type 1 diabetic patients give weight to the importance of regulatory T-cell migration to secondary lymphoid tissues in order to control activity of inflammatory cells. The inflammatory conditions seen in patients with type 1 diabetes may modulate Treg subsets, such that they are not able to control inflammation. Further studies are needed to determine the suppressive activity of these cells in order to see if their function is disrupted.