Patient recruitment and clinical data

Patients who received a sural nerve biopsy were recruited from January 2006 to May 2023 at the CIDP outpatient Department and Department of Neuropathology at the Charité-Universitätsmedizin Berlin, Germany. Study inclusion required patients above the age of 18 years, independent of disease duration and severity, to meet the national and European guidelines for diagnosis of CIDP according to EAN/PNS criteria [44]. In accordance with these criteria, patients with confirmed paranodal auto-antibodies (anti-NF155, anti-CNTN1, and anti-Caspr1) were excluded, as such cases are classified as autoimmune nodopathies rather than CIDP. Due to diagnostic doubts or refractory status, patients received a sural nerve biopsy. To minimize treatment-related effects on complement deposition, we excluded patients who had received corticosteroids or intravenous immunoglobulin (IVIG) within 6 weeks prior to sural nerve biopsy. This interval was chosen based on published recommendations for the assessment of complement activity in serum and plasma [1], as it allows for normalization of transient complement suppression following these treatments. The treatment-free interval was verified through detailed review of medical records.

DC served a total of 8 sural nerves with a morphological diagnosis of chronic mildly active axonal damage without overt signs of any inflammatory alterations. These patients were classified as idiopathic neuropathy (IPN) as an extended diagnostic approach according to the EAN/PNS guidelines, which did not reveal any underlying cause for the neuropathy. In addition, one sural nerve biopsy with clinical manifestation of hereditary neuropathy (HNP) and a genetic test confirming the diagnosis was also examined for complement deposition. Moreover, we included two non-disease control (NDC), which received a sural nerve biopsy to exclude potential somatic cause of neuropathic symptoms. One NDC was recruited from the Department of Neurology, Friedrich-Baur Institute, LMU Munich, Germany.

Patients were examined by a team of peripheral nerve specialists. Socio-demographics and current medication were documented. The medical research council sum scale (MRC-SS) was assessed to measure muscle strength at the impairment level [25]. For the MRC-SS the following muscles were tested on both sides: shoulder abduction, elbow flexion, wrist extension, index finger extension, hip flexion, knee extension, ankle dorsiflexion, and extension of the big toe. A sum score of 80 indicates normal muscle strength. The adjusted inflammatory neuropathy cause and treatment disability score (INCAT-DS) was used to assess the clinical disability in daily arm and leg mobility, which has evolved as the most established primary outcome in clinical trials [21, 30]. In addition, we assessed CSF parameters, comorbidities, treatment regime (corticosteroids, IVIg, plasmapheresis, and immunosuppressive therapies) and treatment response and further categorized the patients according to their CIDP disease activity status (CDAS) into active disease status with relapsing or chronic progressive disease course or remission [18].

Sural nerve specimens

After shock cryofixation and diagnostic processing all sural nerve specimens had been cryopreserved at − 80 °C before analysis according to the predefined standard operating procedure of the Department of Neuropathology at the Charité-Universitätsmedizin, Berlin, Germany [34, 35]. For mass spectrometry-based analysis, sural nerve samples were transferred on dry ice to the Leibniz-Institute for Analytical Sciences–ISAS–e.V. and stored at − 80 °C prior to sample processing. Sural nerve biopsies without any evidence of pathological alterations by histology, semithin studies, fiber teasing, and ultrastructural analysis were defined as non-diseased control.

Histological and immunohistochemical analysis

Routine stains, immunohistochemical-, and double-immunofluorescence reactions were performed as previously described [34, 35]. The following antibodies were used for staining procedures: C5b-9 (Dako, aE11, 1:200), CD8 (Dako, C8/144B, 1:100), CD45 (Dako, 2B11, 1:400), CD68 (Dako, EBM11, 1:100). The presence of C5b-9 was further visually quantified in high, medium and low staining intensity (examples given in eFigure 1). CD8 + T cells were graded for none (0–4 cells/10 HPF), few (5–12 cells/10 HPF) or multiple (> 12 cells/10 HPF). HPFs, based on the microscope used and the respective oculars (Olympus WH10x-H/22) ≙ 0.16 mm2. CD68 + macrophages were graded semi-quantitatively for few and diffusely distributed (endoneurium); grade 1, multiple and increased but diffusely distributed; grade 2, or many and clustering with T cell accumulation; grade 3. HPFs, based on the microscope used and the respective oculars (Olympus WH10x-H/22) ≙ 0.16 mm2. Antibodies were detected with the immunoperoxidase method. All staining procedures were performed in the same laboratory on a Benchmark XT immunostainer (Ventana, Tucson, AZ) [34].

Real-time qPCR

RNA extraction from the nerve tissue samples, reverse transcription, and quantitative PCR reactions were performed as previously described [35]. In short, total RNA was extracted from sural nerve specimens, and cDNA was synthesized using the High-Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA). For qPCR reactions, 10 ng of cDNA was used. For subsequent analysis, the QuantStudio 6 Flex System (Applied Biosystems) was used with the following running conditions: 95 °C 0:20, 95 °C 0:01, and 60 °C 0:20, 45 cycles (values above 40 cycles were defined as not expressed). All targeted transcripts were run as triplicates. The TaqMan® Gene Exp Assay (Life Technologies/ThermoFisher) are as follows: APRIL/TNFS13 Hs00182565_m1, BAFF/TNFSF13B Hs00198106_m1, C1QA Hs00706358_s1, C3 Hs00163811_m1, C4A Hs00246758_m1, C5 Hs01004342_m1, C6 Hs01110040_m1, C8A Hs00175098_m1, C9 Hs01036216_g1, TNFA Hs00174128_m1.

PGK1 Hs99999906_m1 was included as an internal control to normalize the relative expression of the targeted transcripts. Gene expression was illustrated by the fold-change values compared with that in NDCs.

Lysate generation and processing for proteomic deep mapping

The entire nerve sample was lysed in 200 µl of 50 mM Tris–HCl buffer (pH 7.8), containing 5% SDS and cOmplete ULTRA protease inhibitor (Roche), using the Bioruptor® (Diagenode) for 10 min with 30-s on/off cycles for a total of 10 cycles at 4 °C. An additional sonication step was performed with an ultrasonic probe (30 s, alternating 1 s on and 1 s off, amplitude 40%) to ensure thorough lysis. This was followed by centrifugation at 20,000 g for 15 min at 4 °C. The protein concentration in the resulting supernatant was measured using a BCA assay following the manufacturer's instructions. Disulfide bonds were reduced by adding 10 mM TCEP at 37 °C for 30 min, and free sulfhydryl groups were alkylated using 15 mM IAA at room temperature, in the dark, for 30 min. For proteolysis, 100 µg of protein from each sample was processed using the S-Trap protocol (Protifi) with a 20:1 protein-to-trypsin ratio. The trypsin digestion step was carried out for 2 h at 42 °C and stopped by acidifying the sample with formic acid to achieve a pH below 3.0. Samples were dried using a Speedvac (Thermo Fisher Scientific, Waltham, MA, USA) and dissolved in 0.1% TFA to achieve a 0.5 µg/µl concentration.

All hydrolyzed samples were checked for completeness of digestion after desalting through monolithic column separation (PepSwift monolithic PS-DVB PL-CAP200-PM, Dionex) on an Ultimate 3000 HPLC system (Dionex, Germering, Germany) via direct injection of 0.5 μg of sample. A binary gradient (solvent A: 0.1% TFA, solvent B: 0.08% TFA, 84% ACN) was applied, transitioning from 5% to 12% B over 5 min, followed by 12–50% B over 15 min, at a flow rate of 2.2 μL/min and 60 °C. UV detection was performed at 214 nm (doi.org/https://doi.org/10.1016/j.jprot.2011.11.016).

Mass spectrometry-based proteomic profiling

All samples were analyzed using an UltiMate 3000 RSLC nano-UHPLC system coupled to a QExactive HF mass spectrometer, with 1 µg of peptide used for each analysis. Initially, the samples were transferred to a 75 µm × 2 cm, 100 Å, C18 pre-column at a 10 µl/min flow rate for 20 min. Separation was then performed on a 75 µm × 50 cm, 100 Å, C18 main column at a 250 nl/min flow rate. The separation utilized a linear gradient composed of solution A (99.9% water, 0.1% formic acid) and solution B (84% acetonitrile, 15.9% water, 0.1% formic acid), with a pure gradient length of 120 min, transitioning from 3% to 45% solution B. The gradient profile was as follows: 3% solution B for the first 20 min, 3–35% over 120 min, followed by three washing steps at 95% solution B, each lasting 3 min. After the final wash, the system was equilibrated for 20 min. For MS survey scans, the following settings were used: MS was operated in data-dependent acquisition mode (DDA) with full MS scans from 300 to 1600 m/z (resolution 60,000) with the polysiloxane ion at 371.10124 m/z as lock mass. Maximum injection time was set to 120 ms. The automatic gain control (AGC) was set to 1E6. For fragmentation, the 15 most intense ions (above the threshold ion count of 5E3) were chosen at a normalized collision energy (nCE) of 27% in each cycle, following each survey scan. Fragment ions were acquired (resolution 15,000) with an AGC of 5E4 and a maximum injection time of 50 ms. Dynamic exclusion was set to 15 s.

Bioinformatic analysis of protein data

Z-scoring was performed with the raw abundance values obtained and normalized by the Proteome Discoverer 2.5 software (Thermo Scientific) to create and visualize the heatmap. Subsequently, the received data were filtered for complement proteins. Only proteins that could be identified with at least two unique peptides were considered reliable identifications and retained. After filtering the data, the visualization was displayed in a heatmap using orange data mining software. Clustering was performed based on similarity with hierarchical clustering on Euclidean distances and with average linkage of the individual samples. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the data set identifier PXD056286 [33].

Statistical analysis

Descriptive statistics are reported as means and standard deviations (SDs), medians and interquartile ranges (IQRs) for continuous variables, and absolute and relative frequencies for nominal data. All statistical analyses were performed using GraphPad Prism V10.2.2.

Mann–Whitney tests were performed to compare levels of complement components between patients and controls. Their correlations with clinical disease parameters were analyzed using Spearman's rank correlation coefficient. A two-tailed p value < 0.05 was considered statistically significant.