1. Background

Gangliosides, glycosphingolipids containing one or more sialic acid residues, and paranodal/nodal molecules are critical targets of immune attacks in immune-mediated neuropathies. Autoantibodies targeting these structures cause nerve injury in nodo-paranodal regions, characterized by distinctive pathophysiological mechanisms, as well as electrophysiological and pathological findings, collectively termed nodo-paranodopathy [1]. Sulfatide, a glycolipid distinct from gangliosides due to its lack of sialic acid residues, is synthesized from galactocerebroside, a simple glycosphingolipid, via sulfation of the hydroxyl group on the third carbon of galactose [2]. Sulfatide is predominantly localized in the Golgi apparatus, lysosomes, and cell membranes and is abundant in the nervous system, kidneys, gastrointestinal tract, pancreatic islets, trachea, and various cancers and tumors. It plays roles in physiological and pathological processes involving the nervous system, immune response, hemostasis/thrombosis, diabetes, and bacterial/viral infections [3]. In the nervous system, sulfatide is primarily found in oligodendrocytes and Schwann cells, enriched in the noncompact myelin sheath [4]. In peripheral nerves, it is distributed along the myelin internode and concentrated at the paranodal loops of terminal heminodes [5]. Sulfatide contributes to the differentiation of myelinating cells and the maintenance of paranodal glia–axon junctions and signaling [3, 4]. In sulfatide-deficient mice, the glia–axon junction at the paranode, along with the clustering of Na + channels at nodes and K+ channels at juxtaparanodes, becomes disorganized [6], although the underlying mechanism remains unclear. Anti-sulfatide antibodies have been detected in autoimmune peripheral neuropathies such as Guillain–Barré syndrome (GBS), chronic inflammatory demyelinating polyradiculoneuropathy (CIDP), and multifocal motor neuropathy, as well as in IgM monoclonal gammopathy [7–9]. In one study, 85% of anti-sulfatide IgM-positive patients also exhibited elevated IgM antibody titers against myelin-associated glycoprotein (MAG) [7]. Since both sulfatide and MAG are predominantly localized in the myelin paranodal region [5, 10], this finding supports the hypothesis that anti-sulfatide /MAG antibodies contribute to nodo-paranodopathy. Nevertheless, the clinical, electrophysiological, and pathological features of anti-sulfatide antibody-positive autoimmune neuropathy remain poorly defined.

In this study, we reviewed the clinical and electrophysiological features as well as the treatment response of the anti-sulfatide positive patients with or without antiganglioside antibodies in two hospitals. Furthermore, the electrophysiological findings of paranodopathy mediated by anti-sulfatide and anti-NF155 antibodies were compared.

2. Methods

3. Results

3.1. Accompanied Antiganglioside Antibodies in Anti-Sulfatide Positive Neuropathy

Among the 34 patients with antisulfatide-positive neuropathy, 32 were adult-onset cases, while 2 patients presented symptoms during adolescence (Figure 1a). The specific antibodies detected in each patient and their clinical diagnoses are summarized in Table 1. In adult-onset cases, 13 patients had only anti-sulfatide antibodies in serum or CSF, whereas 19 patients also had accompanying antiganglioside antibodies (Figure 1b). The most frequently co-occurring antiganglioside antibody was anti-GM1 (Figure 1c,d). Other commonly observed antibodies included anti-GQ1b, anti-GM3, and anti-GD1b. Additional antiganglioside antibodies detected in anti-sulfatide -positive patients included anti-GM2, anti-GM4, anti-GD1a, anti-GD2, anti-GD3, anti-GT1a, and anti-GT1b. Thirteen patients tested positive exclusively for anti-sulfatide antibodies.

Table 1. Antibodies and diagnosis of anti-sulfatide -positive patients.

Patient no.	Serum			CSF			Diagnosis
	Anti-sulfatide		Antiganglioside abs	Anti-sulfatide		Antiganglioside abs
	IgG	IgM		IgG	IgM
1	+						AMSAN
2		+					Unclassified peripheral neuropathy
3				+			AMAN
4	+						Uncertain diagnosis
5				+			Anti-GQ1b syndrome
6					+		CIDP
7		+					CIDP
8		+					Unclassified peripheral neuropathy
9		+					CIDP
10		+					Unclassified peripheral neuropathy
11		+					AMSAN
12				+			AMSAN
13				+			Unclassified peripheral neuropathy
14		+	GM1				AIDP
15		+	GM1				Unclassified peripheral neuropathy
16		+	GM1				Unclassified peripheral neuropathy
17	+		GM1				HNPP
18	+		GM3				Uncertain diagnosis
19	+		GD2	+			Unclassified peripheral neuropathy
20			GD3	+			AMAN
21		+	GM1, GM2, and GM3				Uncertain diagnosis
22		+	GM1 and GD1b				AMSAN
23			GM1 and GD3	+			AMAN
24		+	GM1 and GM2, GD1b and GD2				CIDP
25	+	+	GM1, GM3, and GD1a	+			Uncertain diagnosis
26			GM1, GD1b, and GD3	+		GD3	AMAN
27		+	GM3, GM4, and GD3		+	GT1b, GD3, GM3, and GM4	CIDP
28			GQ1b	+			Anti-GQ1b syndrome
29		+	GQ1b				CMT suspected
30			GT1a and GQ1b	+		GT1a and GQ1b	Anti-GQ1b syndrome
31			GM1 and GD1b	+		GM2, GD1b, and GT1a	AMAN
32		+	GM2, GQ1b, and GT1a				Anti-GQ1b syndrome
33		+	GM1, GM3, GD1b, GT1b, and GQ1b				CIDP
34	+		GM4, GD1a, GT1a, GT1b, and GQ1b	+		GT1a and GQ1b	Anti-GQ1b syndrome

• Abbreviations: AIDP, acute inflammatory demyelinating polyneuropathy; AMAN, acute motor axonal neuropathy; AMSAN, acute motor sensory axonal neuropathy; CIDP, chronic inflammatory demyelinating polyradiculoneuropathy; CMT, Charcot–Marie–Tooth disease; HNPP, hereditary motor neuropathy with liability to pressure palsies.

3.2. Clinical and Laboratory Findings in Adult-Onset Patients

Patients with anti-sulfatide antibodies were categorized into seropositive (antibodies detected in serum and/or CSF) and CSF-positive (antibodies detected only in CSF). For clarity, the latter is referred to as the CSF-positive group. The demographic characteristics and clinical findings of adult-onset patients (n = 32) are summarized in Table 2. The average onset age was 55.22 ± 14.66 years among all patients, with no significant differences between groups. The male-to-female ratio was 18:14. Prodromal symptoms, including fever, respiratory, and gastrointestinal symptoms, were observed in approximately 30% of patients. Limb weakness was the most common symptom, affecting 26 patients (81.25%). Disturbance in superficial sensation (68.75%) was frequently observed, while only 16.67% of patients experienced deep sensory disturbances. Ataxia (37.04%), diplopia (25.00%), limb pain (25.00%), dizziness, and headache (21.88%) occurred occasionally. Less frequent symptoms included barylalia (15.63%), nausea and vomiting (12.50%), dyspnea, tremor, and cranial nerve involvement (other than ocular motor nerves) (6.25%). Only one patient (3.13%) reported autonomic dysfunction.

Table 2. Demographic and clinical characteristics of patients with anti-sulfatide antibody.

Demographics	Seropositive (n = 22)	Only CSF-positive (n = 10)	Total (n = 32)
Age of onset (years, mean ± SD)	55.41 ± 15.71	54.80 ± 14.33	55.22 ± 14.66
Sex ratio (male/female)	13:9	5:5	18:14
Symptoms
Prodromal symptoms	6/22 (27.27%)	4/10 (40.00%)	10/32 (31.25%)
Disturbance in superficial sensation	16/22 (72.73%)	6/10 (60.00%)	22/32 (68.75%)
Disturbance in deep sensation	3/21 (14.29)	2/8 (25.00%)	5/30 (16.67%)
Limb weakness	16/22 (72.73%)	10/10 (100%)	26/32 (81.25%)
Ataxia	8/20 (40.00%)	2/7 (28.57%)	10/27 (37.04%)
Diplopia	7/22 (31.82%)	1/10 (10.00%)	8/32 (25.00%)
Other cranial nerve involvement	2/22 (9.09%)	0/10 (0.00%)	2/32 (6.25%)
Barylalia	4/22 (18.18%)	1/10 (10.00%)	5/32 (15.63%)
Autonomic dysfunction	1/22 (4.55%)	0/10 (0.00%)	1/32 (3.13%)
Limb pain	3/22 (13.64%)	5/10 (50.00%)	8/32 (25.00%)
Nausea and vomiting	4/22 (18.18%)	0/10 (0.00%)	4/32 (12.50%)
Dyspnea	0/22 (0.00%)	2/10 (20.00%)	2/32 (6.25%)
Dizziness and headache	6/22 (27.27%)	1/10 (10.00%)	7/32 (21.88%)
Tremor	2/22 (9.09%)	0/10 (0.00%)	2/32 (6.25%)
Diagnosis
AIDP	1/22 (4.55%)	0/10 (0.00%)	1/32 (3.13%)
AMAN ∗	0/22 (0.00%)	5/10 (50.00%)	5/32 (15.63%)
AMSAN	3/22 (13.64%)	1/10 (10.00%)	4/32 (12.50%)
Anti-GQ1b syndrome	3/22 (13.64%)	2/10 (20.00%)	5/32 (15.63%)
CIDP	5/22 (22.73%)	1/10 (10.00%)	6/32 (18.75%)
Unclassified peripheral neuropathies	6/22 (27.27%)	1/10 (10.00%)	7/32 (21.88%)
Uncertain diagnosis	4/22 (18.18%)	0/10 (0.00%)	4/32 (12.50%)
Comorbidity
DM	5/22 (22.73%)	2/10 (20.00%)	7/32 (21.88%)
Tumor	1/22 (4.55%)	1/10 (10.00%)	2/32 (6.25%)
MND	2/22 (9.09%)	0/10 (0.00%)	2/32 (6.25%)
MSA	1/22 (4.55%)	0/10 (0.00%)	1/32 (3.13%)

• Abbreviations: AIDP, acute inflammatory demyelinating polyneuropathy; AMAN, acute motor axonal neuropathy; AMSAN, acute motor sensory axonal neuropathy; CIDP, chronic inflammatory demyelinating polyradiculoneuropathy; DM, diabetes mellitus; MND, motor neuron disease; MSA, multiple systemic atrophy.
•  ^∗p < 0.05 between seropositive and only CSF-positive group.

To clarify the role of anti-sulfatide antibodies while minimizing the potential confounding effects of antiganglioside antibodies, we compared clinical characteristics between patients with and without antiganglioside antibodies. No significant differences were observed (Figure 2a, Table S1). Diagnoses varied among the patients, including CIDP (18.75%), anti-GQ1b syndrome (15.63%), and acute motor axonal neuropathy (AMAN) (15.63%). Four patients were diagnosed with acute motor-sensory axonal neuropathy (AMSAN), and one had acute inflammatory demyelinating polyneuropathy (AIDP). Seven patients presented with peripheral neuropathy symptoms that could not be classified due to unknown etiology. Four patients with uncertain diagnoses exhibited diplopia, but central nervous system infections could not be excluded. Notably, the frequency of AMAN was significantly higher in the CSF-positive group (p = 0.001). Furthermore, diagnoses appeared related to the isotype of anti-sulfatide antibody, as AMAN and anti-GQ1b syndrome were found only in patients with anti-sulfatide IgG, while AIDP and CIDP occurred exclusively in patients with anti-sulfatide IgM (Table S2).

Among the 32 adult-onset anti-sulfatide -positive patients, comorbidities included diabetes mellitus (21.88%), motor neuron disease (MND) (6.25%), thyroid or lung tumors (6.25%), and multiple system atrophy (MSA) (3.13%). Lumbar puncture and CSF routine examinations were conducted in 29 patients, with albumin-cytological dissociation observed in 18. This dissociation was present in 80% of CSF-positive patients but only about half of seropositive patients (Table 3). Serum immunofixation electrophoresis was performed on 18 patients, but none exhibited monoclonal gammopathy. Rheumatic antibodies and factors were screened in 28 patients, with 10 testing positive for rheumatoid factors and antibodies (including antinuclear antibodies (ANAs) and anticardiolipin antibodies (ACAs)). Among these 10 patients, eight patients had serum ANA, while two exhibited elevated rheumatoid factors, one of whom also had glucose-6-phosphate isomerase and ACA. In the CSF-positive group, only one patient (12.50%) tested positive for rheumatic antibodies, which was lower than in the seropositive group (Table 3).

Table 3. Laboratory findings of patients with anti-sulfatide antibody.

Laboratory findings	Seropositive, n/N (%)	Only CSF-positive, n/N (%)	Total, n/N (%)	p value
CSF albumincytological dissociation	10/19 (52.63%)	8/10 (80.00%)	18/29 (62.07%)	0.234
Blood monoclonal gamma globulin	0/13 (0.00%)	0/5 (0.00%)	0/18 (0.00%)	—
Blood rheumatic antibodies and factors	9/20 (55.00%)	1/8 (12.50%)	10/28 (35.71%)	0.194

• Note: N: number of cases received electrophysiology examination; n: number of efficacious cases.
• Abbreviation: CSF, cerebrospinal fluid.

4. Discussion

This study highlights the underrecognized clinical importance of anti-sulfatide antibodies across various neuropathies. We find that anti-sulfatide antibody may exist in varied peripheral neuropathies, including AMAN, AMSAN, Anti-GQ1b syndrome, CIDP, and unclassified peripheral neuropathy. The patients with anti-sulfatide antibody mainly complained of limb weakness (81.25%), superficial sensation disturbance (68.75%), and ataxia (37.04%) no matter with antiganglioside antibodies or without. Electrophysiological examination shows that anti-sulfatide antibody may result in both axonal and demyelinating damage. Compared with anti-NF155 IgG4-seropositive and seronegative CIDP patients, anti-sulfatide antibody positive patients have shorter distal motor latency and faster motor and sensory nerve conduction velocities. Most patients responded to the IVIG and corticosteroid treatment. A recent study focused on pediatric GBS revealed that GBS spectrum disorders with positive anti-sulfatide antibodies present a distinct clinical phenotype, with shorter time to peak serving as an independent predictor of poor prognosis [11].

In our cohort, there were 10 patients with anti-sulfatide antibody only in CSF-positive. This prompted us to consider whether CSF anti-sulfatide antibodies entered the CNS through blood-nerve barrier (BNB) leakage or were synthesized intrathecally. Previous studies on anti-sulfatide IgG in acquired immunodeficiency syndrome (AIDS) suggested that CSF anti-sulfatide antibodies are probably synthesized intrathecally by infiltrating peripheral blood cells and are associated with peripheral neurological disorders in AIDs patients [12]. In fact, the blood-CSF barrier in the spinal nerve roots consists of meningeal cells in the pia mater, endothelial cells, and the basement membrane in capillary walls [13]. It is suggested that, in GBS, swelling nerve roots may cause venular congestion and result in disturbance of microcirculation. The anatomic changes of BNB in pathological conditions facilitated the entrance of mononuclear cells into nerves, as well as the leak of plasma protein. Here, CSF anti-sulfatide positive patients are prone to have CSF albumin-cytological dissociation. In addition, 3 of the 10 patients were accompanied by antiganglioside antibodies in their CSF samples, while in the other 7 patients, anti-sulfatide antibodies were the only autoantibody detected in CSF. Thus, CSF anti-sulfatide antibodies detected in patients with peripheral neuropathy are more likely to be synthesized intrathecally.

Previous results showed that anti-sulfatide antibodies are elevated in CSF from multiple sclerosis (MS) patients, and that both IgM and IgG antibodies are present [14]. A sulfatide-reactive monoclonal antibody obtained from an MS patient selectively binds to the surface of rat oligodendrocytes, indicating a potential role of anti-sulfatide autoantibodies in CNS demyelination [15]. However, whether CSF anti-sulfatide antibodies cause CNS demyelination remains unclear. Recent studies indicated that anti-sulfatide antibodies are rarely found in CSF [16] and that pediatric anti-sulfatide antibody-positive GBS patients exhibited inflammatory demyelinating lesions shown by brain MRI [11]. So far, the clinical symptoms between seropositive and CSF-positive patients show no significant difference.

In 1992, Aotsuka et al. revealed that serum anti-sulfatide antibodies were present in autoimmune rheumatic disease patients with a frequency between 10% and 81%, varied among diseases and disease stages [17]. Serum anti-sulfatide antibody levels were associated with the levels of anti-double-strand DNA antibodies. Antiganglioside antibodies were also found to be associated with peripheral neuropathy in rheumatoid arthritis [18]. Further investigations carried out by Alpa et al. showed elevated anti-sulfatide IgM/IgG and anti-GM1 IgM/IgG in patients with small vessel ANCA-associated vasculitis and other multiorgan immune-mediated diseases [19]. However, anti-sulfatide IgM/IgG showed a weak correlation with anti-dsDNA antibodies and were not correlated to other serologic rheumatic markers including ANA, RF, ANCA, anti-SSA, and SSB antibodies. Interestingly, there is a small portion of patients who had anti-GM1 or anti-sulfatide antibodies but were absent of neuropathy. The above studies all demonstrated the co-occurrence of serologic anti-sulfatide antibodies and rheumatic markers in autoimmune rheumatic diseases, but none focused on neuropathy solely. Herein, we discovered that the anti-sulfatide antibody seropositive patients tend to have elevated serologic rheumatic antibodies (ANA and ACA) and rheumatic factors even though not previously diagnosed as connective tissue diseases. We suspect that the coexistence of anti-sulfatide and rheumatic antibodies is due to cross-reactivity between these antibodies. It is reported that anti-sulfatide antibodies possess cross-reactivity with various anionic molecules (dsDNA, sulfatide, etc.) [17]. In addition, sulfatides are targets for ACAs [20]. We thus hypothesize that the exposure of sulfatide at paranodal loops in neuropathy patients boosted the production of some rheumatic antibodies.

Diabetes is the most common cause of peripheral neuropathy. Our study shows that there were seven DM patients among anti-sulfatide positive patients. All of them were diagnosed as Type 2 diabetes mellitus (T2DM) preceding the onset of neuropathy. Among them, three patients had only anti-sulfatide antibodies, and the other four patients were accompanied with antiganglioside antibodies. A previous study showed that nearly a quarter of Type 1 diabetes mellitus (T1DM) patients had sera antiganglioside antibodies, most of which were IgG isotype [21]. Anti-sulfatide IgG was found positive in 88% of patients diagnosed as T1DM [22]. The presence of antiganglioside and anti-sulfatide antibodies in T1DM may result from nonspecific immune hyperactivation and cross-reactivity between peripheral nerve and pancreatic cells. So far, anti-sulfatide antibodies were not found to correlate with T2DM; however, sulfatide might be involved in T2DM [23], and autoimmune reactions may also contribute to the pathogenesis of T2DM [24]. Herein, another reason may be the disruptive role of T2DM which is in consistency with Appeltshauser’s finding [25] suggesting that DM may be a potential risk factor for predisposing to developing nodo-paranodopathy. In their study, the frequency of DM is significantly elevated in CIDP patients with antiparanodal (including neurofascin-155, pan-neurofascin, contactin-1–associated protein 1, and contactin-1) antibodies, especially in anti–contactin-1-seropositive patients.

Sulfatide is a major component of the uncompacted myelin sheath in the peripheral nervous system [5]. It localizes at the paranodal loops and is involved in glial–axon interactions due to its essential role in the maintenance of the axo–glial junction and sodium/potassium ion channel clusters [4]. Other autoantigens at the paranodal loops include NF-155, Caspr1, and contactin 1, all of which mediate axo-glial attachment [26]. Though NF-155 and sulfatide both anchor on Schwann cells in paranodal loops, the clinical symptoms and electrophysiological changes caused by elevated anti-sulfatide and anti-NF-155 antibodies were not quite the same. The onset age of patients with anti-NF-155 antibody [27] was significantly younger (p = 0.0029) compared to anti-sulfatide antibody-positive patients. Most anti-NF155 antibody-positive patients were diagnosed as CIDP, while CIDP accounts for no more than 20% of anti-sulfatide antibody-positive patients. Recently, Martin-Aguilar et al. reported that anti-NF155 positive patients presented with a progressive (75%), sensory motor (87.5%), and symmetric distal predominant weakness in upper (97.2%) and lower extremities (94.5%), with tremor and ataxia (75%) [27]. Disturbances in deep sensation, ataxia, cranial nerve involvement, and tremor were more frequent in anti-NF155 antibody-positive patients [27–29]. In our anti-sulfatide positive cohort, the numbers of symptoms with tremor (6.25%) and ataxia (37.04%) are less than those of the anti-NF-155 positive patients. anti-sulfatide antibodies cause unique electrophysiological patterns when compared directly with those of anti-NF155 antibodies. Demyelination in motor and sensory nerves was more prominent in anti-NF-155 antibody-positive patients, as shown by electrophysiological data. Hence, though both NF-155 and sulfatide localize at paranodal loops, the clinical characteristics of patients with anti-sulfatide antibody differed from those with anti-NF-155 antibody-positive neuropathy.

The main limitations of our study arise from the small number of patients and its retrospective nature. First, due to the clinical significance of serum antiganglioside antibody in GBS, anti-sulfatide antibody and antiganglioside antibody detection in CSF were underestimated and therefore autoantibodies were not tested in all CSF samples. Also, the potential bias in data collection and lack of uniform testing may affect the validity and reliability of the results. Second, no patient was diagnosed with M-proteinemia in the current study, which is inconsistent with previous studies. This might be due to immunofixation electrophoresis not being commonly screened among patients with peripheral neuropathy. In addition, a previous study revealed that sulfatide-binding is a common property of antiganglioside antibodies [30]. Another limitation is that although autonomic dysfunction can be a clinical manifestation in peripheral neuropathies such as GBS, this study did not include assessments of autonomic electrophysiological parameters. Thus, valid methodology is crucial for the detection of anti-sulfatide and antiganglioside antibodies in peripheral neuropathy patients. A future prospective multicenter study with uniform testing for anti-sulfatide antibodies could be helpful to further reveal the role of anti-sulfatide antibody.

5. Conclusions

In conclusion, anti-sulfatide antibodies are associated with autoimmune peripheral neuropathy by causing axonal and demyelinating damage in adult patients. Though both sulfatide and NF155 localize at the paranodal loops, anti-sulfatide antibodies–related neuropathy causes a specific clinical spectrum and unique electrophysiological patterns.

Ethics Statement

This study was conducted according to a protocol approved by the Ethics Committee of the Qilu Hospital of Shandong University and the first affiliated hospital of Shandong First Medical University. All patients gave written informed consent to participate in the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding

This research is funded by the National Natural Science Foundation of China (Nos. 82101487 and 82171245), the Natural Science Foundation of Shandong Province, China (ZR2021QH161 and ZR2024MH051), and the Taishan Scholars Program of Shandong Province (202211318).