CD4+ regulatory and effector/memory T cell subsets profile motor dysfunction in Parkinson's disease

Abstract

Animal models and clinical studies have linked the innate and adaptive immune system to the pathology of Parkinson's disease (PD). Despite such progress, the specific immune responses that influence disease progression have eluded investigators. Herein, we assessed relationships between T cell phenotype and function with PD progression. Peripheral blood lymphocytes from two separate cohorts, a discovery cohort and a validation cohort, totaling 113 PD patients and 96 age- and environment-matched caregivers were examined by flow cytometric analysis and T cell proliferation assays. Increased effector/memory T cells (Tem), defined as CD45RO+ and FAS+ CD4+ T cells and decreased CD31+ and α4β7+ CD4+ T cells were associated with progressive Unified Parkinson's Disease Rating Scale III scores. However, no associations were seen between immune biomarkers and increased age or disease duration. Impaired abilities of regulatory T cells (Treg) from PD patients to suppress effector T cell function was observed. These data support the concept that chronic immune stimulation, notably Tem activation and Treg dysfunction is linked to PD pathobiology and disease severity, but not disease duration. The association of T cell phenotypes with motor symptoms provides fresh avenues for novel biomarkers and therapeutic designs.

Figures

Fig. 1

Gating strategy for flow cytometric analysis of PBMC and CD4+ T cell and Teff frequency.a Representative flow cytometric scatter plots used for data collection. The CD4+ T cell population was identified by high expression of CD4 and low side scatter (left panel). Treg and Teff were identified within the CD4+ T cell population as CD25+CD127− and CD25+CD127+ (right panel). b The percentage of CD4+ lymphocytes for Cohort B. CD4+ data (b) are expressed as means ± SEM, and significant differences between CD4+ T cell means were determined by Mann–Whitney test for 63 caregivers and 71 PD patients where *p ≤ 0.05

Fig. 2

CD4+ T cell and Teff phenotypes are associated with UPDRS-III score. Flow cytometric data of caregivers and PD patients from Cohort B were binned into 4 groups based on UPDRS-III scores: caregivers (CG, n = 61), 1–20 (n = 25), 21–30 (n = 28), and ≥31 (n = 16). The percentages of CD4+ T cells expressing CD45RO (a), α4β7 (b), FAS (c), and CD31 (d) in each group were associated with UPDRS-III group (p < 0.05). Percentages of CD45RO+ (e) and CD31+ (f) Teff and the percentages of α4β7+ (g) and CD31+ (h) Treg in each group were associated with UPDRS-III group (p < 0.05). Data are the percent-positive of T cells with medians (horizontal lines). Significant differences among groups were determined by Kruskal-Wallis nonparametric ANOVA (CD45RO, α4β7 and FAS) or by general linear model (CD31), and pair-wise comparisons were determined by either Dunn’s or Bonferroni adjustments for multiple comparisons (CD31) where * p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001

Fig. 3

Treg from PD patients are dysfunctional while naïve T cells (nT) and Teff show no alterations in proliferative capacity. aTreg, nT and Teff were identified within the CD4+ T cell population as CD25hiCD127-, CD25-CD127+, and CD25+CD127+, respectively (top, center dot plot). CFSE-labeled CD4+CD25− allogeneic Tresp were co-cultured with serially-diluted Treg. By flow cytometric analysis, CFSE was measured to determine the percentage of Tresp that proliferated, and the percentage of CD25+ Tresp was measured as an indication of activation (histograms, left panels). The proliferative capacity of isolated CFSE-labeled Teff and nT was measured by flow cytometric analysis of CFSE (histograms, right panels). b Inhibition of proliferation of Tresp by Treg at ratios of 1:1 (n = 28), 1:0.5 (n = 26), 1:0.25 (n = 28) and 1:0.125 (n = 25) of Tresp to Treg, demonstrating that Treg from PD patients had decreased inhibitory capacity at the 1:0.125 dilution (p = 0.006). c CD25 expression was not altered at any dilution. d The percentage of proliferating Teff did not differ (p < 0.05) for PD patients (n = 34) compared to caregivers (n = 32). e The percentage of proliferating nT did not differ (p < 0.05) for PD patients (n = 16) compared to caregivers (n = 15). Data are expressed as the percentage of proliferating cells out of all CFSE+ events. Significant differences among groups were determined by Kruskal-Wallis nonparametric ANOVA, and pair-wise comparisons determined by Dunn’s multiple comparison’s post-hoc analysis (b, c) or by Mann–Whitney test (d, e)

Fig. 4

Correlation difference networks of caregivers and PD patients. Differences network interactions are demonstrated by the correlation difference network, where edge width and opacity reflects the correlation difference score (n = 37 CG, 46 PD). Correlations with p-value >0.0005 were not considered statistically significant, and edges outside that threshold were discarded