In this study We found 88 DEPs (Fig. 2C), mainly associated with immune/ inflammation function, and Neurological disease (Fig. 3A,3B,3C). A panel of Alpha-1-antitrypsin (SERPINA1) and S100-A9 (S100A9) could achieve AUC 0.991. Finally, we verified the DPP4 protein with ELISA.
In the function analysis DEPs were mostly involved in immune/inflammatory-related activity (Chemotaxis of T lymphocytes、Systemic autoimmune syndrome、Chronic inflammatory demyelinating polyradiculoneuropathy) (Fig. 3C) .Inflammation is present in all subgroups of MS[21], The diagnostic hallmark of MS pathology is the presence of a large number of confluent demyelinating lesions in the gray matter and white matter of the CNS[22, 23].It is generally believed that inflammation causes demyelination damage and disrupts neuronal signal transmission in the affected area[24]. Demyelination can occur in the brain, cerebellum, brainstem, basal ganglia, spinal cord and other parts[22]. After the nerve fiber loses myelin sheath, there will be a series of motor and sensory disorders[25], and in our function annotation enriched to Urination disorder、Mild cognitive impairment、Neuromuscular disease, etc. which are consistent with previous studies(Supplementary Table S4, S5).
Studies on the immune system of MS patients indicate that MS is an autoimmune disease, T cells、 B cells and autoantibodies may be the most important factors in the pathogenesis of its immune system[26, 27] .And it is generally believed that the inflammatory process of MS is caused by the autoimmune cascade, including the main target of T cells (Th17 phenotype) is the self-myelin antigen[24] The autoimmune pathogenesis of MS has been confirmed in its animal model (EAE)[28].EAE is caused by the subject's subcutaneous injection of CNS myelin antigen in complete Freund's adjuvant or adoptive transfer of newly activated neural antigen/myelin specific CD4 + T helper cell 1 (Th1) or Th17 cells[28] .Therefore, self-reactive CD4 + T cells are sufficient to induce MS-like inflammatory demyelinating diseases. Of which Macrophages are the most prominent contributor to the pathological changes of MS, in the early stage of MS, the M1 phenotype of macrophages releases inflammatory factors to infiltrate the CNS, thereby promoting neuroinflammation and neurodegenerative development, in the later stage of MS, its M2 phenotype is mainly responsible for removing fragments of myelin and assisting in the repair of the CNS, which is a sign of MS[29, 30]. Complement deposition is found in the cortical gray matter of MS. Studies have confirmed that the genetic variation of early complement components is related to the retinal neurodegeneration of MS[31].
DEPs are mostly enriched in blood coagulation system-related pathways (Fig. 3B).Previous studies have shown that a variety of coagulation factors act as inflammatory driving factors in EAE[32], and a variety of coagulation factors are dysregulated in MS[33], suggesting that the coagulation system plays an important role in MS. DEPs are also enriched in Synaptogenesis Signaling Pathway. Synaptogenesis Signaling Pathway is becoming a key determinant of early neurodegeneration of MS[34]. Previous studies used immunohistochemical methods to study the degree of synapse loss in the spinal cord (18 patients with chronic MS, 8 healthy controls) after death. A large amount (58–96%) of synapse loss was detected in the entire spinal cord, and moderate (47%) loss of neurons in the anterior horn especially in demyelinating MS lesions. It indicates that synaptic damage in chronic MS may lead to the accumulation of disability[35]. Moreover, in animal models of MS, the loss of synaptic signals has been shown to be related to changes in the balance of inhibitory and excitatory neurotransmission caused by inflammation[36]. Moreover, many DEPS were correlated with energy metabolism, Glycolysis is another source of energy in the synapse[37],and MS patients can result into energy failure in mitochondria and cytosol[37].
In our study, from the ROC curve, We found a good predictive panel biomarkers with SERPINA1 and S100A9(Figure 4A). Previous studies have revealed the presence of SERPINA1 is a serine protease inhibitor, which not only has the ability to inhibit serine proteases, but also exerts anti-inflammatory and tissue protective effects independent of inhibiting proteases[38].SERPINA1 can modify the maturation of dendritic cells and can promote T cells to regulate and differentiate[38]. Past studies have shown that sustained levels of circulating SERPINA1 profoundly inhibit induction of clinical and histological signs of EAE[39]. It suggests that SERPINA1 may have the ability to interfere with the progression of MS[39]. SERPINA1 also enriched in our Inflammatory response, Progressive neurological disorder process.
S100A9 is a calcium-zinc-binding protein that plays a prominent role in the regulation of inflammation and immune response[40], past studies have shown that S100A9 induces the activation of microglia by activating the NF-kB signaling pathway, and promotes the production of pro-inflammatory factors, thereby aggravating the damage of oligodendrocyte precursor cells (OPCs), thereby affecting the damage of MS disease[41]. And in two independent EAE models, it was found that the protein expression was significantly different during the disease[42], and enriched in our Chronic inflammatory disorder、Progressive neurological disorder process etc. It further confirms the important value of S100A9 as a potential plasma marker of MS.
We used ELISA to perform follow-up verification of DPP4-dipeptidyl peptidase 4 (DPP4) protein, DPP4 is a serine protease, also known as the lymphocyte surface protein CD26, which plays a central role in T cell immunity [43]. In a variety of autoimmune diseases, the expression and activity of CD26 in serum have changed, suggesting that it may be involved in the pathogenesis of autoimmune diseases[43]. The study have found that in the magnetic resonance imaging of patients with MS subtypes, there was a correlation between changes in the frequency of CD26-expressing T cells and disease activity, which was consistent with the function of CD26 to regulate the activation state of T cells[44]. These data indicated that CD26 is associated with the expression of human CD4 + T cells in MS disease activity. So, CD26 has attracted great interest as a potential target for MS immunity therapy. A single synthetic CD26 inhibitor was successfully used to treat an animal model of MS[20].Importantly, studies have shown that CD26 inhibitor combination therapy can reduce the risk of complex autoimmune diseases (RA, IBD, MS, and SLE)[45]. This study used ELISA to verify that CD26 was consistent with our mass spectrometry results[19], confirming that DPP4 may be used as an inhibitor in clinical treatment of MS diseases. However, before advocating any clinical application, it is very important to define and optimize the molecular mechanism of such drugs, and we need further research.
Comparison Of Our Study, Ms-blood, And Ms-csf Proteome
In our study, we summarized our study with previous blood proteomics research, and compared it with past years of CSF proteomics. (Fig. 5A 、5B、Supplementary Table S5).
The blood-brain barrier is (or is) interrupted in most MS patients, the CNS protein substances released into the circulation can be assessed by proteomics analysis [46].However, due to the limited clinical utility of cerebrospinal fluid biomarkers, the lumbar puncture process for obtaining cerebrospinal fluid is invasive and associated with risks(19), And, it was not always feasible to collect CSF, especially for individuals and children who had never developed symptoms[47].Instead, blood can be routinely collected for the discovery of early protein biomarkers, and plasma contains proteins from almost all cell types in the organism(20), indicating that blood can be a viable alternative biological sample for patients with MS.
For MS-CSF proteomics: as shown in (Figure 5A, Supplementary Table S5)CSF proteins were more remarkably involved in the CNS process and MS ༈central nervous system cancer, Astrocytoma, Multiple sclerosis, Morphology of nervous system, Degeneration of neurons).In the pathway analysis in ༈Figure 5B༉CSF proteins were more remarkably involved in the Axonal Guidance Signaling、Synaptogenesis Signaling Pathway et al. Axonal degeneration as a cause of irreversible neurological impairment during MS[48].This result reflects that the CSF protein may better reflect the pathological characteristics of MS.
Overall ,for MS-Blood and MS-CSF proteomics:(Figure 5A, Supplementary Table S5), the protein pathways/functions of the blood enrichment were aggressively concentrated on the immune/inflammation-related activity (Systemic autoimmune syndrome、Dendritic Cell Maturation 、Chronic inflammatory demyelinating polyradiculoneuropathy),which consisting with our study (Fig. 5A、5B).
For the MS-Blood enrichment: Compared to MS-CSF proteomics, the central nervous system functions associated with MS disease are also altered (Glioblastoma). Also, the blood protein enrichment is significantly involved in immune/inflammatory pathways (macrophage activation process, systemic inflammatory response syndrome, etc.). These immune/inflammatory and neurological disease pathways may also reflect pathological changes in MS(Fig. 5A、5B). This result reflects that the blood protein also could reflect the pathological characteristics of MS.