Differences in peripheral myelin antigen-specific T cell responses and T memory subsets in atypical versus typical CIDP

Background Chronic inflammatory demyelinating polyneuropathy (CIDP) is presented by a large heterogeneity of clinical phenotypes. Around 50% of patients suffer from typical CIDP and show better therapy response than atypical variants. The goal of our study was to search for cellular immunological differences in typical versus atypical CIDP in comparison to controls. Methods We evaluated 26 (9 typical, 17 atypical) patients with mainly active-unstable CIDP using clinical and immunological examinations (enzyme-linked immunospot assay ELISPOT, fluorescence-activated cell sorting FACS) in comparison to 28 healthy, age-matched controls (HC). Typical or atypical CIDP measurements were compared with HC using Kruskal-Wallis test. Results Atypical CIDP patients showed increased frequencies of T cell subsets, especially CD4+ effector memory T cells (TEM) and CD4+ central memory T cells (TCM) as well as a tendency of higher T cell responses against the peripheral myelin antigens of PMP-22, P2, P0 and MBP peptides compared to typical CIDP. Searching for novel auto-antigens, we found that T cell responses against P0 180-199 as well as MBP 82-100 were significantly elevated in atypical CIDP patients vs. HC. Conclusions Our results indicate differences in underlying T cell responses between atypical and typical CIDP characterized by a higher peripheral myelin antigen-specific T cell responses as well as a specific altered CD4+ memory compartment in atypical CIDP. Larger multi-center studies study are warranted in order to characterize T cell auto-reactivity in atypical CIDP subgroups in order to establish immunological markers as a diagnostic tool.


Background
Chronic inflammatory demyelinating polyneuropathy (CIDP) is the most common autoimmune peripheral neuropathy but remains a rare disease with a prevalence of 0.8-8.9 cases per 100.000 [1,2]. The disorder causes severe disability in more than 50% of the patients in a chronic-progressive course [1]. Diagnosis can be difficult given the heterogeneity of CIDP phenotypes. About 50% of the patients suffer from so-called atypical variants including Distal Acquired Demyelinating Polyneuropathy (DADS) in 25-35% of the cases, Multifocal Acquired Demyelinating Sensory And Motor Polyneuropathy (MADSAM) in 15% and rare variants such as pure sensory CIDP (10-13%), pure motor CIDP (<10%) and focal CIDP (2%) [3]. These CIDP subtypes are likely to differ with respect to underlying pathomechanisms and may necessitate different treatment approaches.
Despite recent progress, the underlying immunopathogenetic mechanisms remain poorly understood [4]. Both humoral as well as cellular immune responses are likely to play a role in the induction of autoimmune neuroinflammation, which leads to demyelination and axonal degeneration [4][5][6][7].
Peripheral myelin antigens are promising auto-antigens in CIDP pathogenesis. Recently, we demonstrated higher frequencies of auto-reactive IFN-γ responses directed against the peripheral myelin antigens PMP-22 and P2 in treatment naïve patients who responded subsequently well to intravenous immunoglobulin (IVIG) treatment. Clinical improvement under IVIG-treatment correlated with the reduction of antigen-specific responses against PMP-22 and P2 [8].
Myelin basic protein (MBP) is a major constituent of the myelin sheath in the central and peripheral nervous system [12]. Whereas it has been established as an immunodominant auto-antigen for demyelination in the immunopathogenesis of Multiple Sclerosis (MS) its auto-reactive potential in CIDP remains elusive [13].
Here, we hypothesize that autoreactive myelin-specific T cell responses as well as T cell memory subsets differ between atypical and typical manifestations of CIDP.

Patients
We evaluated 26 CIDP patients using clinical and immunological (enzyme-linked immunospot assay ELISPOT, fluorescence-activated cell sorting FACS) examinations in comparison to 28 healthy, age-matched controls. CIDP patients who met the diagnostic criteria of European Federation of Neurological Sciences (EFNS) 2010 were divided into "typical" vs. "atypical" according to EFNS 2010 [20]. Therapy response was defined as an improvement of ≥2 in Medical Research Council (MRC) sum score in 2 different muscle groups, an improvement of ≥1 in Inflammatory Neuropathy Cause and Treatment (INCAT) score (excluding changes in arm function from 0 to 1) or alternatively an improvement of ≥50% of the walking distance as described previously [8]. Patients and controls were recruited in the outpatient clinic of the Department of Neurology, Charité University Medicine Berlin.

Peripheral myelin antigens
ELISPOT assay was performed using peptides of seven peripheral myelin antigens and CEF as positive control for T cell responses (Table 1). CEF is a peptide pool containing 23 MCH class 1 restricted viral antigens [21]. Peripheral myelin antigens were provided by Dr. R. Volkmer, Institute of Medical Immunology, Charité University Medicine Berlin. CEF was provided by JPT Peptide Technologies GmbH, Berlin.

Cryopreservation of Peripheral Blood Monocytes (PBMC)
To evaluate T cell responses efficiently we preserved PBMC in liquid nitrogen over a maximum of 6 months. Blood was sampled in CPT tubes for ELISPOT and in EDTA tubes for flow cytometry. PBMC were isolated within 2 h after venipuncture by 1500 g centrifugation for 20 min. After washing, we diluted the PBMC at a concentration of 2x10 7 cells/ml in freezing medium A (60% FCS; 40% RPMI, Biochrom, Berlin, Germany) at 4°C. The same volume of freezing medium B (20% DMSO, 80% FCS) at 4°C was added before cell suspensions were transferred into cryovials (Sarstedt, Nürnbrecht, Germany) and set in one at 4°C prechilled Nalgene Cryogenic Freezing Container (Fisher Scientific, Hannover, Germany) which was placed in −80°C overnight. After 12-24 h, cryovials were transferred into liquid nitrogen tanks for storage until ELISPOT.
ELISPOT IFN-γ ELISPOT assay in this study was performed on human PBMC as previously described [8]. We plated 4×10 5 cells/well in triplicates for each antigen and positive (CEF) or negative control (medium). CEF, a peptide pool containing viral antigens functioning as a positive control for T cell responses, was added at 9 μg/ml [21].

Statistics
All statistical tests were performed using GraphPadPrism 6.0 software. The study was assessed as an exploratory analysis. Typical or atypical CIDP measurements were compared with healthy, age-matched controls (HC) using Kruskal-Wallis test followed by post-hoc unpaired t-test or Mann-Whitney-test when p < 0.05. For group differences with regard to sex, prior treatment, disease activity and therapy response, Fisher's exact test was used. For age and INCAT score, unpaired t-test was used. For time since diagnosis, Mann-Whitney-test was used. Level of significance was defined as p < 0.05 for all comparative tests.
T cell IFN-γ-responses to P0 180-199 and MBP 82-100 were elevated in CIDP patients compared to healthy controls T cell responses against the peripheral myelin antigens, P0 180-199 and MBP 82-100 were measured by IFN-γ ELISPOT in a cohort of 26 CIDP patients. Due to spontaneous IFN-γ-production, 6 patients (1 typical, 5 atypical) were excluded for further ELISPOT analysis.

Atypical CIDP variants have significantly higher levels of Cd4+ memory T cells
Frequencies of T cells (p < 0.01) and CD4+ T cells (p < 0.001) were higher in patients with atypical CIDP variants in comparison to typical CIDP patients (Fig. 3a).

Discussion
In the present study, typical CIDP differed from the group of atypical variants. Here, we found a stronger activated immune system in patients suffering from atypical variants of CIDP defined by a trend towards increased peripheral myelin antigen-specific (PMP-22, P0 180-199, MBP 82-100) T cell responses associated with a specific altered CD4+ memory compartment of increased CD4+ TEM and CD4+ TCM counts in the blood. Further we detected elevated T cell responses against antigens P0 180-199 and MBP 82-100 in CIDP patients which have not described before.
We confirmed or previous findings that changes of the T memory compartment is a common finding especially  in untreated patients [8,17], which is in contrast to Sanvito and colleagues who showed no differences in T cell subpopulation [23]. In the present study, we detected elevated TEM and TCM primarily in atypical CIDP patients. Clinical experience and studies suggest that typical CIDP patients respond better to therapy than atypical CIDP variants, especially DADS [24], which is in line with our presented data showing that 100% (9/9) of typical compared to 47% (8/17) of atypical CIDP patients were therapy-responders. The reason for different treatment responses of CIDP subtypes remains unknown. The higher specific immune responses against myelin-derived peptides in atypical compared to typical variants may be a cause for the lower treatment-responses. Likewise, the increased immune reactivity in atypical CIDP patients could result from insufficient treatment.
Recently, it has been demonstrated that CIDP patients show a diminished pro-regenerative function of Schwann cells leading to the axonal loss and therefore incomplete clinical recovery after treatment which is probably caused by inflammatory mediators [25]. Thus, differences in immune responses between typical and atypical CIDP we have demonstrated might also influence Schwann cell function resulting in different treatment responses and long-term outcome. The INCAT score was significantly lower in atypical cases. However, there was no difference in the time since diagnosis so that a longer disease course and hence pronounced disability and/or altered immune response is not the cause of this difference. Yet, we included mainly atypical case with mild motoric disability (6 patients with sensory CIDP [35.3%] and 5 patients with DADS [29.4%]) who are less often dependent on walking aids leading to lower INCAT disability scores.
Since we included mainly clinically unstable patients who had partly received treatment before, we are not able to answer this question at present. Based on previous results of reduced CD4+ memory subsets in GS-treated patients [17], it might be further argued that GS treatment may be efficient for this patient group. In contrast to Sanvito et al. [26], we identified higher IFN-γ responses to P2 and PMP22 peptides which have been more pronounced in the atypical compared to the typical CIDP subgroup. A higher number and proportion of atypical patients might explain this discrepancy as well as the fact that we included mainly clinically unstable and newly diagnosed patients.
Up to now, only little is known about the role of MBP 82-100 in the pathogenesis of CIDP even though MBP has been detected as part of the myelin sheath of peripheral nerves. Nevertheless, there is long-standing evidence that MBP 82-100 can induce neuroinflammation in autoimmune diseases [27]. Glatirameracetat, known antagonist of MBP 82-100 specific T cell receptor and part of MS therapy has been demonstrated to alleviate symptoms also in EAN-mice [28,29]. Here, we demonstrated significantly elevated MBP 82-100 specific T cell responses in CIDP patients, again primarily in patients with atypical manifestations.
There is growing evidence for the autoimmune potential of antigens which are derived from non-compact myelin of the nodal/paranodal region such as neurofascin 155 of 186 leading to antibody response in distinct subgroups of CIDP or multifocal motoric neuropathy (MMN) [30][31][32][33]. Thus, antigenic targets derived from both compact and non-compact myelin leading to humeral and/or cellular immune response may define underlying immune mechanism of different clinical phenotypes of CIDP.
Several limitations may have affected our results. First, our clinically heterogeneous group of atypical patients was too small to distinguish between subgroups of atypical CIDP, which would be necessary to characterize atypical subtypes and to define specific cut offs for our immunological parameters. Second, differences in gender and INCAT score between typical and atypical CIDP patients might have influenced our immunological findings. Third, we aimed to recruit treatment-naïve patients in activeunstable stages of the disease. However, only 46% of patients (12/26) were treatment-naïve at enrollment. Previous immunosuppressive and -modulating therapy might have influenced our immunological findings.

Conclusions
Higher myelin-antigen specific T cell responses together with elevated T cell memory subsets were found in atypical compared to typical CIDP patients suggesting different patterns of immune responses in clinically distinctive CIDP subgroups. Myelin as well as nodal/paranodal proteins might serve as candidate autoantigens to establish robust immune markers for CIDP subtype differentiation. Given the clinical diversity of CIDP a larger cohort study is warranted in order to establish those markers with reliable cut-off values. The authors thank Sandra Bauer, Sonya Becker and Maik Stein for technical assistance.

Funding
The study was funded by a research grant from Grifols and supported by the Deutsche Forschungsgemeinschaft = German Research Foundation (NeuroCure Cluster of Excellence, Exc. 257). The funder had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Availability of data and materials
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
Author's contributions MS recruited the patients, performed the experiments, analyzed and interpreted the data, and wrote the manuscript. JD recruited patients and analyzed the data. CM performed the experiments and revised the manuscript. AM analyzed the data and revised the manuscript. JK designed the study, was involved in ethical approval, recruited the patients, analyzed the data, revised the manuscript and supervised the study. She is the guarantor of the study. All authors gave their final approval to the study to be published and agree to be accountable for all aspects of the work.
Competing interests J. Klehmet and A. Meisel have received personal compensation outside the submitted work for activities with Grifols, Octapharma and CSL Behring. M. Staudt, JM. Diederich, C. Meisel report no relevant financial activities outside the submitted work from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years, no other relationships or activities that could appear to have influenced the submitted work disclosures. Parts of this work has been presented as a poster at the Inflammatory Neuropathy Consortium (INC) in Glasgow, UK in June 2016.

Consent for publication Not applicable.
Ethics approval and consent to participate Our study was approved by the local ethics committee (Ethikkommission Charité Universitätsmedizin Berlin, EA 1/025/11) and conducted according to the Declaration of Helsinki. All patients gave written informed consent to the study.