Introduction

Amyotrophic lateral sclerosis (ALS) is a relentlessly progressive and universally fatal disease, which is the most common phenotype of motor neuron disease (MND). It is characterized by the selective degeneration of upper and lower motor neurons within the brain and spinal cord. The clinical manifestations of ALS are heterogeneous regarding age and site of disease onset, progression rate, and survival time. Efficacious treatments to significantly slow the progression of ALS are still lacking. The etiology of ALS has not been completely clarified despite extensive research. Epidemiological investigation revealed that about 10% of total ALS cases were familial ALS, of which 20% were linked to a point mutation of Cu/Zn superoxide dismutase (SOD1) [1]. The remaining 90% are sporadic ALS without apparent genetic abnormalities. Numerous environmental and occupational factors have been associated with ALS, including exposure to metals, chemicals, pesticides, and unhealthy lifestyles, such as smoking and excessive physical exercise [1]. It is becoming increasingly evident that various cellular and molecular processes mediate the neurodegeneration of ALS, including glutamatergic excitotoxicity, RNA and protein metabolism, mitochondrial dysfunction, oxidative stress, and immune response [2,3,4,5].

Understanding how the immune system participates in the pathogenesis of ALS has attracted substantial attention. Both peripheral and central immune systems are activated in the progression of ALS, which leads to a chronic proinflammatory microenvironment [6]. The chronic proinflammatory microenvironment is characterized by activated resident microglia and astrocytes in the central nervous system (CNS) and proinflammatory lymphocytes, monocytes, and mast cells in the periphery [5]. Antibody production is the most widely recognized role of B lymphocytes, which are key regulators in the inflammatory process [7]. Several autoantibodies have been identified in the serum and cerebrospinal fluid (CSF) of ALS patients and showed a correlation with disease progression [8]. These antibodies may serve as potential biomarkers for monitoring the ALS disease progression and diagnostics. A reliable and quantifiable biomarker will not only serve as an indicator of disease severity and prognosis, but also help to increase our understanding of ALS pathogenesis and facilitate the assessment of the response to pharmacologic intervention.

The aim of this review is to summarize the presence and clinical significance of autoantibodies in ALS. We begin with a brief presentation of immune dysregulation in ALS, followed by an overview of various autoantibodies in ALS.

Immune dysregulation in ALS

Cumulative data demonstrated active participation of the immune system in the pathogenesis of ALS [9]. The local immune system of CNS mainly comprises the activation of microglia and astrocytes, while peripheral immune system comprises the innate and adaptive immune system. Dysregulation of both central and peripheral immune systems has been reported in ALS [5]. A growing number of studies have also revealed the breakdown of blood-brain barrier and blood-spinal cord barrier in ALS patients, which led to functional crosstalk between peripheral immune cells and CNS [6].

In CNS, microglia and astrocytes play dual roles at different disease stages of ALS progression. At the onset/early stage, both microglia and astrocytes present neuroprotective function via secreting neurotrophic and anti-inflammatory mediators, while at the terminal stage, the activated microglia and astrocytes shift to a pro-inflammatory phenotype and aggravate neuron damage [6, 10]. Dysregulation of innate immune system components has been indicated and regarded as hallmarks for ALS disease monitor or progression to some extent, such as monocytes and macrophages [11]. So were the components of adaptive immune system in ALS patients. Elevated levels of classical complement pathway (C1q and C4) and downstream complement components (C3 and C5b-9) were found in the spinal cord and motor cortex of ALS patients [12]. Mice injected with IgG purified from the sera of ALS patients led to the gradual loss of spinal motor neurons, along with decreasing muscle strength in the limbs [13]. Besides, elevated senescent and late memory T and B lymphocytes showed a correlation with fast-progressing ALS and bulbar involvement [14]. Antibodies generation is the main way for B lymphocytes to regulate immune system. Furthermore, multiple kinds of antibodies have been detected in the sera or CSF of ALS patients. In the next part, we will give a detailed description of the autoantibodies detected in ALS patients.

The presence and clinical significance of autoantibodies in ALS

We performed a literature search until 31 August, 2023 in PubMed, with the following terms ("motor neuron disease" OR "MND" OR "amyotrophic lateral sclerosis" OR "ALS") AND ("antibody" OR "antibodies" OR "autoantibody" OR "autoantibodies"). Titles and abstracts were screened, and relevant full-text articles were retrieved. The summary of the results from the literature search is presented in Fig. 1.

Fig. 1
figure 1

Summary of the results from the literature search. The literature search was until 31 August, 2023 in the database Pubmed. The literature search was conducted using terms ("motor neuron disease" OR "MND" OR "amyotrophic lateral sclerosis" OR "ALS") AND ("antibody" OR "antibodies" OR "autoantibody" OR "autoantibodies"). Titles and abstracts were screened, and full-text articles were retrieved

Totally twenty-nine kinds of autoantibodies have been detected in the body fluids of ALS patients. We divided these autoantibodies into the following categories according to different antigens: paraneoplastic antibodies, neuron-related antibodies, peripheral nerve-related antibodies, neuromuscular junction-related antibodies, muscle-related antibodies, and pathophysiology-related antibodies (Table 1). We will comprehensively describe the positive rate and clinical significance of different autoantibodies in ALS patients.

Table 1 Autoantibodies reported in ALS

Paraneoplastic antibodies

Paraneoplastic antibodies are strongly associated with both paraneoplastic neurologic syndromes and cancers [68]. Several kinds of paraneoplastic antibodies have been detected in the body fluid specimens of ALS patients. In an MND cohort of 145 patients, none of the sera revealed high anti-neuronal antigen (HuD/Yo/Ri/CV2/CRMP5/Ma2/Amphiphysin) reactivity, while five sera showed a very weak reactivity [15]. However, another MND cohort revealed that 9% (13/138) of patients had positive paraneoplastic antibodies to VGCC (n=4), striated muscle (n=4), VGKC (n=3), GAD65 (n=3) and gAChR (n=2) [16]. Donaldson et al. conducted a study in 405 MND patients, which reported that one or more cation channel antibodies (VGKC, VGCC, gAChR) were detected in 6.9% of patients, mostly at low titers. Furthermore, the presence of cation channel antibodies exerted no effects on disease progression in MND patients [17]. Victoria et al. investigated the presence of VGKC antibodies in ALS patients compared to that in a cohort of peripheral nervous system disorders. The abnormal antibody level (VGKC antibody titer≥100pmol/L) was more common in ALS, while it failed to reach statistical significance (16/54 vs. 6/41, p=0.08). None of ALS patients with abnormal antibody titer demonstrated clinical or electrodiagnostic evidence of myokymia or neuromyotonia [18]. Godani et al. explored the prevalence of VGKC antibody in 20 MND patients, which reported that 25% (5/20) of patients had VGKC-complex antibody>100pmol/L. Furthermore, patients with slow-progression MND showed a higher prevalence of VGKC-complex antibodies than those with a typical course [19]. Yang et al. reported an ALS patient with serum antibodies against both Sry-like high mobility group box 1 (SOX1) and glutamic acid decarboxylase 65 (GAD65). However, immunotherapy failed to alleviate symptoms [20].

Neuron-related antibodies

Anti-neurofilament antibody

Neurofilaments (NF) are the major components of the neuronal cytoskeleton, which can be divided into neurofilament heavy chain (NfH), neurofilament medium chain (NfM), and neurofilament light chain (NfL). NfL concentration has been regarded as a potential marker of neuronal injury in various neurodegenerative diseases. Increased CSF NfL concentration has been found to correlate with disease severity and progression in ALS [69].

The presence of anti-NF antibody has been found in the serum and/or CSF of ALS patients. In 1995, an ALS patient with anti-NF antibody has been reported [34]. A sporadic ALS cohort revealed an elevated percentage (24.7%, 21/85) of serum anti-NF antibody compared to healthy controls (6.1%, 6/98) and unrelated neurological disease controls (12.6%, 10/79). Surprisingly, the level of anti-NF antibody was significantly correlated with a slow rate of progression [21]. Fialová et al. measured IgG antibodies against NfL and NfM by ELISA in paired CSF and serum samples from 38 ALS patients and 20 controls [22]. The level of anti-NfL antibody in serum was significantly elevated in ALS patients, while serum levels of anti-NfM antibody was only significantly elevated in bulbar-onset subgroup of ALS. There were no significant differences in CSF levels of anti-NfL and anti-NfM antibodies between ALS patients and controls. Furthermore, serum anti-NfL/CSF anti-NfM levels and ALS Functional Rating Scale (ALSFRS) showed a weak correlation.

The clinical significance of anti-NF antibodies has also been explored. Puentes et al. measured antibodies and immune complexes against NfL, NfM, NfH, and poly-(GP)-(GR) dipeptide repeats in serum from the ALS Biomarkers cohort (n=107), the phenotype–genotype biomarker cohort (n=129) and in normal controls (n=140). The results revealed a significantly higher concentration of anti-NfH and NfL immune complexes in ALS, especially in those with a faster progressing rate. The longitudinal study among different time points suggested that increasing levels of anti-NF antibodies and immune complexes were observed in faster-progressing ALS [24]. Furthermore, Puentes et al. reported that higher plasma anti-NF level was suggestive of the advanced stage in a cohort of 73 ALS [23].

Plasma anti-NF antibody level was significantly increased in ALS and varied with the disease progression stage. The determination of anti-NF antibody levels in plasma could be a potential disease-monitoring biomarker for ALS.

Anti-IgLON5 antibody

IgLON5 is the fifth member of the IgLON family, which belongs to the immunoglobulin superfamily of neuronal cell adhesion molecules [70]. IgLON5 plays a key role in neuroplasticity/neurogenesis and the maintenance of blood-brain barrier integrity [71, 72]. The IgLON5 antibody was first detected in patients with the sleep-breathing disorder. Anti-IgLON5 disease is a rare autoimmune encephalitis with anti-IgLON5 antibodies in serum and/or CSF, which was first described in 2014 [73]. However, a growing spectrum of clinical manifestations is being recognized in association with anti-IgLON5 autoimmunity, including recent reports of MND-like phenotype.

Sista et al. presented 4 cases with seropositive IgLON5-IgG (3 possible ALS and 1 definite ALS), manifesting as ALS-like phenotype with parasomnias, vocal cord dysfunction, or hyperkinetic movements. Furthermore, the team evaluated the positive rate of IgLON5-IgG by indirect immunofluorescence assay and cell-based assay (CBA) in a cohort of 109 probable or definite ALS patients. None experienced with laryngeal dysfunction, rapid eye movement sleep behavior disorder (RBD) and involuntary movement disorder. Meanwhile, all were IgLON5-IgG seronegative [25]. The first anti-IgLON5 seropositive patient with MND-like phenotype was reported in 2018 [26]. The patient also presented with prominent sleep abnormalities and involuntary movement. Werner et al. reported 5 anti-IgLON5-seropositive cases with bulbar MND-like phenotype, mostly accompanied by sleep-related breathing disorders, parasomnias, and laryngeal dysfunction [27].

Anti-IgLON5 antibody could be detected in MND patients, especially cases with laryngeal dysfunction, sleep-related breathing disorders, parasomnias or involuntary movement disorder. From the perspective of neuropathology, it is not surprising that MND-like phenotype is associated with anti-IgLON5 disease as neuron-specific tau deposits. The accumulation of hyperphosphorylated tau preferentially involved the hypothalamus, and more severely the tegmental nuclei of the brainstem with a cranio-caudal gradient of severity until the upper cervical cord [74].

Anti-Transglutaminase 6 antibody

Tissue transglutaminase 6 (TG6) was abundantly expressed in the septal region, basal ganglia, hypothalamus, and brainstem [75]. Anti-TG6 antibody (IgA and/or IgG) was detected positive in 62% of patients with gluten ataxia [76], a neurological manifestation of Celiac disease. Several independent cases with an initial diagnosis of ALS that ultimately were identified as Celiac disease have been reported [77,78,79,80]. Gadoth et al. detected the Celiac disease-related antibodies and anti-TG6 antibody in ALS patients. Compared to the healthy control [4.3% (5/115)], the TG6 IgA antibody was detected positive in 15.3% (23/150) of ALS patients [28]. However, there were no significant differences in the clinical presentation between anti-TG6 IgA antibody seropositive and seronegative patients [28]. In contrast to IgA, anti-TG6 IgG antibody positivity or concentration did not differ significantly from healthy control. The study indicated that ALS related to gluten sensitivity might occur in a subgroup of patients and that anti-TG6 IgA antibody might be a potential marker for identifying gluten-sensitive patients. The authors recommended that a strict gluten-free diet might be therapeutically indicated in early-identified ALS patients with gluten sensitivity.

Peripheral nerve-related antibodies

Anti-Gangliosides antibodies

Gangliosides are sialic acid containing glycosphingolipids, which are expressed on the surface of all vertebrate cells and particularly abundant in mammalian nerve tissue. According to the unique structure, gangliosides are named differently by the number of sialic acid residues. G at the beginning of each ganglioside name indicates the belonging to the ganglio-series of glycosphingolipids. A, M, D, and T indicate the presence of zero, one, two, and three sialic acid residues, respectively [81]. As reported, more than 90% of the brain gangliosides were constituted by the same four structures (GM1, GD1a, GD1b, and GT1b) [82]. Gangliosides play an important role in maintaining the stability and regeneration of axon, regulating synaptic plasticity and cellular differentiation.

As early as the 1980s, abnormal gangliosides composition in ALS has been reported, along with reports of unusual gangliosides and additional complex found in CSF, brains and spinal cords of ALS patients [83]. Rapport et al. conducted a postmortem study suggested that marked aberrations in brain ganglioside profiles were present in 17 of 21 patients with ALS [84]. The aberrations were detected both in motor cortex and in non-motor regions such as frontal, temporal, and parahippocampal gyrus cortex. However, Dawson failed to detect the major quantitative difference of the ganglioside between spinal cords from 9 patients with clinically diagnosed ALS and 9 normal spinal cords [85]. Furthermore, Dodge et al. reported increased levels of GM3 and GM1 in spinal cords of ALS patients and SOD1 (G93A) transgenic mouse model of ALS [86]. These results suggested that gangliosides might be important participants in ALS pathogenesis and merited further analysis as potential drug targets. However, clinical trials of exogenous bovine gangliosides in ALS treatment yielded inconclusive results [87,88,89,90].

Furthermore, anti-gangliosides antibodies have also been detected in serum or CSF of ALS patients. From a theoretical perspective, antibodies against gangliosides are able to inhibit nerve regeneration, synaptic plasticity, neurotransmission, and axonal growth, which are crucial for the function of the nervous system. The presence and significance of anti-gangliosides antibodies in ALS have been investigated. Cobanet al. reported that anti-gangliosides antibodies were detected in 2 out of 35 (5.7%) ALS patients [29]. Kollewe et al. determined IgG/IgM antibodies to the six gangliosides (asialoGM1 (GA1), GM1, GM2, GD1a, GD1b, GQ1b) in serum of 84 ALS patients. Anti-gangliosides antibodies were seropositive in 22 ALS patients (26.2%). In detail, IgG and IgM antibodies were respectively detected in 9/84 (10.7%) and 15/84 (17.9%) of ALS patients. There was no significant correlation between age, gender, onset site or survival and anti-ganglioside-positive/-negative titers in ALS patients [8].

Anti-GM1 antibody is the mostly detected anti-ganglioside antibody in ALS patients. GM1 is enriched at paranodal regions of Ranvier nodes in myelinated axons. As early as 1988, Pestronk et al. reported that polyclonal anti-GM1 IgM antibody was present in 42 of 74 (57%) ALS patients [30]. Furthermore, anti-GM1 antibody was particularly common in patients with prominent lower motor neuron signs (41/59; 69%) [30]. Shy et al. [31] and Lamb et al. [33] found a similarly higher positive rate (59% and 44%, respectively) of anti-GM1 antibody in MND patients. However, relatively lower positive rate of anti-GM1 was reported in other ALS cohorts. Adams et al. [32] and Taylor et al. [36] reported that anti-GM1 antibody presented in 9/43 (21%) or 1/121 (0.8%) ALS patients. The discrepancies might be ascribed to a wide variation in sensitivity of detection methods and possible bias of patient selection. Haggiag S. et al traced the titers changes of anti-GM1 in three ALS patients. Interestingly, they found that the originally negative anti-GM1 antibody became positive during follow-up [35].

In addition to anti-GM1 antibody, the presence of other anti-gangliosides antibodies in ALS patients has also been investigated. Pestronk et al. reported that anti-GD1a antibody was detected in 42 of 73 (58%) patients with ALS [37]. Furthermore, the anti-GD1a antibody was prevalent when upper motor neuron sign was prominent [37]. Niebroj-Dobosz et al. investigated the titers of anti-AGM1 antibody and anti-sulfatide antibody in serum and CSF of an ALS cohort [38]. In the serum of 103 ALS patients, elevated titers of anti-AGM1 and anti-sulfatide were present in 18% (32/103) and 11% (11/103), respectively [38]. Meanwhile, increased titers of anti-AGM1 and anti-sulfatide were present 15% (12/79) in sera and 8% (6/79) in CSF, respectively [38]. Furthermore, seropositive ALS patients with anti-GM2, anti-GD2, anti-GalNAc-GD1a, or anti-GQ1b antibody have also been reported [39,40,41].

Generally speaking, various kinds of anti-ganglioside antibodies have been detected in ALS patients. However, the presence of anti-gangliosides antibodies is of limited significance for ALS patients, despite few studies demonstrating that anti-GM1 antibody was associated with lower motor neuron signs, anti-GD1a antibody was related to upper motor neuron signs [30, 37].

Anti-sulfoglucuronyl paragloboside antibody

Sulfoglucuronyl paragloboside (SGPG) is a unique glycolipid presented in both peripheral nerve and vascular endothelial cells, which are important structures of the blood-brain barrier and blood-nerve barrier. E-election and P-selections are biomarkers of activated endothelial cells. Anti-SGPG antibody could be detected in 34.7% (25/72) of ALS patients [42]. Furthermore, most anti-SGPG seropositive patients were lower limbs-onset [42]. Another study suggested that anti-SGPG antibody was detected in 28% (7/25) of ALS patients [43]. Interestingly, ALS patients with positive anti-SGPG antibody presented with a higher level of serum soluble E-selectin, which was regarded as a biomarker of activated endothelial cells [43]. In a large-size ALS cohort, serum anti-SGPG antibody was found in 13.3% (15/113) of ALS patients [44]. More importantly, multiple logistic regression analysis suggested that anti-SGPG presence was positively correlated with age and negatively correlated with ALSFRS [44]. Hence, serum anti-SGPG antibody might represent a potential diagnostic biomarker of ALS.

Anti-neurofascin 186 (NF186) antibody

Neurofascin, a family of cell adhesion molecules, is critical for the formation and maintenance of Ranvier nodes [91]. As an isoform of neurofascin, NF186 is located in the axonal initial segment of Ranvier nodes. Owing to the expression site of NF186, anti-NF186 can be detected in autoimmune nodopathy, multiple sclerosis, and multifocal motor neuropathy.

The first case of ALS patient with serum anti-NF186 positivity was reported in 2023. The female patient presented with progressive weakness and amyotrophy of the upper limbs, and then gradually aggravated and presented upper motor neuron signs. Anti-NF186 antibody was positive at a titer of 1:100 throughout the whole course of the disease despite the treatment with intravenous immunoglobulins [45]. The absence of upper motor neuron signs in early stage made it easier to misdiagnose as peripheral neuropathy, especially when the anti-NF186 antibody was positive. However, the exact role of anti-NF186 antibody in ALS remains unknown. Further studies with large sample sizes are needed to evaluate the percentage and clinical significance of anti-NF186 antibody in ALS.

Neuromuscular junction-related antibodies

Signal transduction at the neuromuscular junction (NMJ) is impaired in various NMJ disorders. Signaling molecules including acetylcholine (ACh)-acetylcholine receptor (AChR), lipoprotein-related protein 4 (LRP4), agrin, etc. ensure the efficient signal transduction at the NMJ. Mounting evidence has suggested that ALS was characterized by progressive loss of motor neurons and degradation of NMJ. Pathological changes of the NMJ were reported to occur before the onset of clinical symptoms, which supported the dying-back hypothesis of motor neurons in ALS [92, 93]. Reasonably, anti-AChR, anti-LRP4, and anti-agrin antibodies were detected in the body fluid of ALS patients.

Anti-AChR antibody

The AChR is expressed at the postsynaptic membrane, which promotes muscle excitation after binding with the ACh released from the presynaptic membrane. As early as 1980, anti-AChR antibody was first reported in 9 of 68 (13.2%) ALS patients. Interestingly, the seven anti-AChR seropositive ALS patients with high antibody titers (1.4-50nM, cutoff: <0.25nM) all experienced modified-neurotoxin therapy (snake venom). The remaining two patients had low antibody titers (0.39-0.54nM, cutoff: <0.25nM) [46]. Ashizawa et al. reported that anti-AChR antibody was positive in 9/102 (8.8%) MND patients [47]. Besides, Okuyama et al. reported a 73-year-old ALS female with anti-AChR antibody. It was worth noting that the level of anti-AChR antibody in serum displayed a weak fluctuation during ALS progression [94].

Furthermore, a few cases with a concomitant diagnosis of ALS and myasthenia gravis (MG) have been reported [95,96,97,98,99]. Retrospectively, Del Mar Amador et al. identified six cases (total n=4757, 0.15%) with ALS-MG coexistence in a French cohort during twelve years [100]. Similarly, 0.75% (5/671) of ALS patients were also affected by MG in an Italian cohort over six years [101]. In another Swedish nationwide register-based study, the percentage of concomitant diagnosis of MG in ALS patients was 0.48% (17/3561) [102]. The underlying mechanism of ALS-MG overlap is still elusive. It is speculated that this association may be triggered by immunological mechanisms and alterations in the NMJ.

Anti-agrin and anti-LRP4 antibody

Agrin is released by motor neurons. LRP4 is located at the postsynaptic membrane of the NMJ and motor neurons of the brain and spinal cord. Agrin-LRP4-MuSK signaling plays a critical role in promoting AChR clustering and muscle excitation [103].

Several studies have demonstrated the presence of anti-agrin antibody and anti-LRP4 antibody in ALS patients. Rivner et al. found that 9 of 65 (13.8%) ALS patients were positive with anti-agrin antibody [48]. Furthermore, anti-agrin seropositive ALS patients were slightly younger than seronegative ALS patients [48]. As for anti-LRP4 antibody, it was detected serological positive in 23.4% (24/104) of patients from two sporadic ALS cohorts (Greek and Italian). Meanwhile, CSF anti-LRP4 antibody was positive in 85.7% (6/7) of anti-LRP4 seropositive ALS patients, while none of 17 CSF samples from anti-LRP4 seronegative ALS patients was positive [49]. However, no significant differences in clinical patterns were revealed between the anti-LRP4-positive and anti-LRP4-negative ALS patients [49]. Tüzün et al. reported an approximate positive rate (23.5%, 4/17) of anti-LRP4 in ALS patients and confirmed that serum anti-LRP4 antibody in ALS patients was able to interact with brain neurons [51]. Rivner et al. reported a relatively lower percentage (9.8%, 8/82) of anti-LRP4 antibody in an American ALS cohort [48]. Lei et al. investigated the presence of anti-LRP4 antibody in a Chinese ALS cohort and explored the correlation between anti-LRP4 and repetitive nerve stimulation study [52]. Serum anti-LRP4 antibody was detected by cell-based assay in 5.4% (3/56) of Chinese ALS patients and all three seropositive patients had a positive repetitive nerve stimulation response, while 50.9% (27/53) of the seronegative patients also had impaired neuromuscular transmission [52]. Interestingly, Takahashi et al. reported two anti-LRP4 seropositive probable ALS patients (the Awaji criteria) with myasthenic symptoms that were partially improved by immunotherapies [50].

The discrepancy in the positive rate of anti-LRP4 among different cohorts might be ascribed to the differences in detection assays and patient demographics. Further studies should be performed to investigate the percentage of anti-LRP4 antibody in different ethnic groups and to determine its pathogenic significance.

Muscle-related antibodies

Anti-cytosolic 5'-nucleotidase 1A (cN1A) antibody

Autoantibody recognizing cN1A has been recognized as a biomarker for the diagnosis of inclusion body myositis (IBM) [104]. Furthermore, in vitro and in vivo studies suggested that anti-cN1A antibody might exert an influence on protein degradation in myofibers [105]. Liewluck et al. reported 2 MND patients with seropositive anti-cN1A antibody [53]. The significance of anti-cN1A antibody in MND remains elusive. Margotta et al. investigated the impact of skeletal muscle dysregulation on the phenotypic characteristics in ALS transgenic mice [106]. The results suggested that immune-mediated myogenesis played a pivotal role in skeletal muscle homeostasis and promoted a slow disease progression in ALS transgenic mice [106].

Pathophysiology-related antibodies

Although the detailed mechanisms remain unclear, the pathophysiological processes in ALS are roughly classified into four major parts: impaired RNA metabolism, altered proteostasis or autophagy, cytoskeletal or trafficking defects, and mitochondrial dysfunction [107]. Besides, the effects of environmental exposure cannot be neglected [108]. So far, several pathophysiology-related antibodies have been reported in ALS patients.

Anti-Fas antibody

Fas is a type I membrane protein, which belongs to the tumor necrosis factor receptor family. It can combine with anti-Fas antibody to transmit the programmed cell death signal. Serum anti-Fas antibody was detected by ELISA in 8 of 31 ALS (26%). Besides, anti-Fas antibody in serum of ALS patients could induce neuronal apoptosis [54]. In another cohort, anti-Fas antibody was detected in 25% (13/52) sporadic ALS and 22% (2/9) familial ALS patients [55]. However, there was no statistical correlation between antibody level and disease duration or severity [55].

Anti-annexin V antibody

Annexins belong to a family of multi-functional membrane and Ca(2+)-binding proteins, which play a pivotal role in various cellular activities. The elevated level of anti-annexin V antibody has been reported in several autoimmune diseases [109]. In an ALS cohort, positive anti-annexin V antibody was detected in 16% (4/25) of the CSF samples and 8% (2/25) of the serum samples, respectively [56].

Anti-High Mobility Group Box 1 (HMGB1) antibody

The damage-associated molecular pattern HMGB1 can initiate and perpetuate immune response in noninfectious inflammatory processes. Casula et al. found elevated HMGB1 levels in the spinal cord of ALS patients [110]. Further investigation suggested that serum level of anti-HMGB1 antibody in ALS was significantly higher than that in patients with Alzheimer's disease, Parkinson's disease, and healthy control subjects. More importantly, the level of anti-HMGB1 antibody was significantly correlated with disease severity [57].

Anti-LG72 antibody

G72 is a primate-specific gene, which has been regarded as a susceptibility gene that exerted significant functions in various neurodegenerative diseases, schizophrenia and major depression [111]. LG72, the longest G72 splice variant protein, induces the production of mitochondrial reactive oxygen species via interaction and aggregation with SOD1 [112]. The potency of serum anti-LG72 antibody as a biomarker for ALS diagnosis has been investigated [113]. The serum anti-LG72 antibody concentration was lower in ALS patients than in healthy controls and other neurodegenerative diseases. Furthermore, the concentration of anti-LG72 antibody did not differ significantly among subgroups of ALS patients [113]. The authors speculated that the anti-LG72 antibody was neutralized by endogenous LG72 in ALS patients and anti-LG72 antibody might serve as a surrogate biomarker for ALS.

Anti-gamma-synuclein antibody

Gamma-synuclein is a cytosolic protein, which is abundant in the perikarya, presynaptic terminals, and particularly within neuronal axons [114]. Motor neurons express higher levels of gamma-synuclein [115]. Gamma-synuclein pathological aggregation might contribute to ALS pathogenesis, which has been demonstrated by immunostaining on transverse spinal cord sections and sequential protein extraction from postmortem neural samples [116]. Further study investigated the presence of anti-gamma-synuclein antibody in ALS patients. Anti-gamma-synuclein antibody was positive in 4.9% (2/41) ALS patients, while the positive rate was 13.2% (5/38) in other neurological diseases group and 0% (0/19) in healthy control [58]. The authors concluded that anti-gamma-synuclein antibody was not a characteristic biomarker of ALS.

Anti-Proteasome subunit alpha type 7 (PSMA7) antibody

PSMA7 is an alpha-type subunit of the 20S proteasome core complex, which participates in protein degeneration via the ubiquitin-proteasome pathway (UPP). UPP is a pathway that plays an essential role in the regulation of antigen processing, apoptosis, and neural and muscular regeneration, which has been demonstrated to participate in ALS pathology [117]. The presence of anti-PSMA7 antibody has been investigated in ALS patients. Anti-PSMA7 antibody was positive in 38 of 71 ALS patients and significantly higher in ALS patients than in control group [59]. The anti-PSMA7 level was negatively correlated with ALS duration, which indicated that anti-PSMA7 positivity might be a disease-promoting factor in early-stage ALS [59]. Besides, anti-PSMA7 titers were positively related to the level of creatine kinase [59]. Thus, the results raised the possibility that anti-PSMA7 antibody might be a potential diagnostic marker for ALS. The authors concluded that anti-PSMA7 antibody might participate in ALS pathogenesis, possibly via its regulation of the UPP.

Anti-ß-actin antibody

Beat-actin (ACTB) is one of two non-muscle cytoskeletal forms of actin, which locates not only in the cytoskeleton sytosolic and compartment but also in the plasma membrane and extracellular space. The ACTB isoforms play a specific role in the spatial regulation of actin dynamics and stability in axons of developing motoneurons [118]. Anti-ACTB antibody was detected positive in 33 out of 70 (47.1%) ALS patients. The correlation analysis suggested that the level of anti-ACTB antibody was positively correlated with clinical disease stage and disease duration. Meanwhile, there was a negative correlation between the anti-ACTB antibody level and ALSFRS-R score [60]. The results indicated that anti-ACTB antibody might be a potential biomarker of ALS. However, the mechanism of anti-ACTB antibody production remains unclear. Further researches are needed to elucidate the clinical values and the underlying mechanisms of anti-ACTB antibody in ALS.

Anti-TAR DNA-binding protein 43 autoantibody

TAR DNA-binding protein 43 (TDP-43) is a highly conserved and essential DNA/RNA binding protein coded by the TARDBP gene. The neuropathological hallmark of ALS is the intracellular deposition of insoluble TDP-43 in degenerating motor neurons and glial cells. As normal blood components, naturally autoantibodies (NAbs) play a central role in clearing debris and maintaining homeostasis in multiple neurodegenerative diseases, including ALS [61, 119]. However, previous investigation on the role of anti-TDP-43 NAb in ALS have yielded inconsistent results. Nielsen et al. reported that anti-TDP-43 NAb level was significantly reduced in ALS patients and negatively correlated with disease severity [61]. However, Conti et al. and Simula et al. found elevated level of anti-TDP-43 NAb in different ALS cohorts [62, 63]. Besides, Ramachandran et al. failed to find significant differences in anti-TDP-43 NAb levels between ALS patients and controls [64]. Hence, the significance of anti-TDP-43 NAb in the diagnosis of ALS needs to be further investigated.

Anti-HERV-K antibody

Human endogenous retroviruses (HERVs) are genomic sequences of retroviral origin that constitute around 8% of the human genome. HERVs of the K family (HERV-K) is the most transcriptionally active subgroup among HERVs. The expression of HERV-K is modulated by TDP-43. Aberrant HERV-K expression has been identified in ALS [120]. The in vitro study revealed that HERV-K could regulate the function of peripheral blood immune cells in ALS patients, mainly via generating pro-inflammatory mediators [66]. In addition, the immune response against HERV-K has been investigated in ALS patients. Anti-HERV-K envelope surface 19-37 antibody was found in 81% (17/21) of serum and 86% (18/21) of CSF from ALS patients [65]. The level of anti-HERV-K envelope surface 19-37 antibody in serum and CSF was significantly correlated with disease severity [65]. A study from the same team revealed that anti-HERV-K antibody was seropositive in 78.2% (43/55, envelope surface 19-37 antibody) and 76.4% (42/55, envelope surface 109-126 antibody) of ALS patients [66]. Simula et al. founded that anti-HERV-K envelope surface 30-38 antibody was positive in 40% (18/45) of ALS patients [63]. In another cohort of 243 ALS patients, 55.14% (134/243) were anti-HERV-K seropositive (against envelope peptide VWVPGPTDDRCPAKPEEEG). More importantly, the level of anti-HERV-K in definite ALS (EL Escorial criteria) patients was lower than that in non-definite ALS patients. In addition, the lower level of anti-HERV-K was associated with a lower predicted and observed survival time [67]. Hence, anti-HERV-K antibody might be a potential biomarker for ALS diagnosis and exerts a protective role against ALS progression.

Conclusions

The majority of the clinical and pathological abnormalities in ALS can be ascribed to the injury of motor neurons in motor cortex, brainstem and spinal cord. However, the onset site of ALS remains elusive. Figuring out the relationship between upper and lower motor neuron dysfunction, particularly the site of disease onset, is critical in understanding of ALS pathogenesis. Dying-forward hypothesis proposed that the original site of ALS was cortico-fugal, and the pathogenesis of ALS was a dying forward process primarily starting in the corticomotoneuronal system. On the contrary, dying-back hypothesis supposed that lower motor neuron initially died, following which ALS process spread to upper motor neuron [121]. The positivity of antibodies indicated the impairment of the integrity in nervous system, including upper and lower motor neuron. However, whether the appearance of special antibodies might imply the early site of ALS onset remains to be investigated.

A number of studies have suggested that immune dysregulation might participate in ALS pathogenesis. Published cohort studies or case reports have shed light on the presence and clinical significance of autoantibodies in ALS patients. However, current research have several limitations. First, several autoantibodies directed against antigens can be found in ALS, however, published investigations showed widely differing proportions of positive antibodies in ALS patients. Possible explanations for the discrepancy of results include different detection assays, incubation temperature, purity of different proteins, different cut-off values and so on. In the future, autoantibodies should be tested by better standardized commercial tests (such as fixed cell-based assay, live cell-based assay and so on) and validated with a second technique (ELISA, tissue-based assay and so on). Second, the clinical significance of autoantibodies in ALS remains elusive. It is necessary to determine whether the presence of autoantibodies is associated with a particular clinical phenotype or different survival time. Last but not least, a still undetermined question is whether the autoantibodies are produced secondary to nervous system damage or the trigger of neuronal damage, and whether they are relevant to the ALS pathogenesis or represent an epiphenomenon. Regarding the pathogenicity, further in vivo and in vitro studies are needed to identify the exact role of autoantibodies in ALS.