Immunogenetic Studies in Patients With GAD-Positive Stiff-Person Syndrome Reveal Novel Lymphocytic Genes and KLK10-Gene Variants
Abstract
Background and Objectives
The aim of this study was to identify genetic markers and immunologic characteristics of glutamic acid decarboxylase (GAD) antibody–positive patients with stiff-person syndrome (SPS).
Methods
We conducted systemic immunogenetic studies in 11 GAD-positive patients: 8 with sporadic SPS and 3 from a three-generation family with very high GAD-ab titers but diverse symptomatology (one with GAD-epilepsy and SPS and 2 only with diabetes), by performing complete immunologic profile and whole-exome sequencing analysis.
Results
Two genes expressed in immune and neuronal tissues were identified: the ORAI1 that codes for a calcium release–activated channel protein with a role in the activation of T lymphocytes and the LILRA4 that encodes an IgG-like cell surface protein expressed in plasmacytoid dendritic cells. An important finding was the identification of 7 genetic polymorphisms in the novel Kallikrein 10 (KLK10) gene, shared by all 9 typical patients with SPS, as verified by Sanger sequencing, but not in the 2 GAD-positive family members with diabetes or the GAD-negative controls. To further verify these findings, Sanger sequencing was performed in 10 more patients with SPS and 15 autoimmune controls collectively confirmed that among a total of 39 tested samples, 95% of the 19 patients with SPS were homozygous or heterozygous for all 7 KLK10 variants while 90% of the 20 controls had the wild type or were heterozygous. KLK10 is a peptidase expressed in the choroid plexus epithelium and neuroendocrine organs and participates in the initiation of systemic inflammatory responses and immune-modulated disorders through proteolytic cascades.
Discussion
KLK10 is a novel and potentially key genetic marker in patients with SPS that can contribute to disease pathogenesis by altering protease activity or the expression of neuron-to–immune cell signaling facilitating GAD autoimmunity. Along with the 2 newly identified immune-related genes, KLK10 is likely an interplay between genetic predisposition and immune dysregulation, necessitating the need to explore their significance as susceptibility disease factors and possibly as novel therapeutic targets.
Introduction
Stiff-person syndrome (SPS) is part of the glutamic acid decarboxylase (GAD)–spectrum autoimmune neuronal excitability syndromes, referred to as GAD antibody–spectrum disorders (GAD-SDs), that also include autoimmune epilepsy, cerebellar ataxia, limbic encephalitis, myoclonus, and nystagmus,1-7 each presenting as a standalone entity or with overlapping symptomatology. All have in common the presence of high-titer antibodies to GAD, mainly expressed in the CNS and the pancreatic β-cells, which catalyzes the conversion of the excitatory neurotransmitter l-glutamate to the inhibitory gamma-aminobutyric acid (GABA). Up to 35% of patients with GAD-SDs, and more specifically the SPS which is the most common, also have type 1 diabetes (TD1) while 80% of patients with TD1 have low GAD antibody titers, implying a generalized disturbed self-tolerance epitomized by GAD antigen–specific pathogenic mechanisms.8 Patients with SPS or GAD-SDs frequently have other coexistent systemic autoimmune diseases or significant genetic predisposition for autoimmunity including specific HLA haplotypes in other family members,8-13 suggesting the possibility of a poorly understood genetic or inherited cause of immune tolerance breakdown..
We now report studies in 11 GAD-positive patients, 8 with typical sporadic SPS and 3 from three-generation family members with very high GAD antibody titers (one with SPS and 2 with diabetes), by performing complete immunologic profile and whole-exome sequencing (WES) analysis. We investigated genetic predispositions that underlie autoimmunity in patients with sporadic and familial SPS to identify polymorphisms, genetic variants, and susceptibility factors contributing to the pathogenesis of various SPS phenotypes. We report that specific susceptibility factors provide insights into driving the molecular mechanisms of GAD-positive SPS.
Methods
Participants
Patients were recruited in the Neuroimmunology Unit, Department of Pathophysiology, of the Faculty Medicine, National and Kapodistrian University of Athens (NKUA). Archived serum and DNA samples from 8 Americans/White patients (4 men and 4 women) with sporadic GAD-positive SPS without any other known autoimmunity, previously recruited at NIH by the PI (MCD, and previously published14,15), were tested following an MTA agreement between NIH and NKUA. Clinical data were collected under established clinical protocols by the PI (MCD). Serum and peripheral blood mononuclear cells (PBMCs) were collected and stored in −80°C following standard protocols and written informed consent according to the Declaration of Helsinki and the local ethical rules. Blood samples were stored at −20°C (EDTA blood/serum) or −80°C (PBMCs). DNA was extracted from PBMCs for the WES, and IgG was isolated from all patients' serum samples using the Thermo Scientific Melon Gel IgG purification kit. Serum samples were tested for GAD antibodies by ELISA (Euroimmun).
WES and Statistical Analysis of the Data
WES was performed in all 12 patients, including the three-generation family members with high GAD-abs described further. The 8 patients with sporadic SPS (4 men and 4 women) were from the NIH series, and the other 4 included the 3 GAD-positive family described further and the brother of the index patient (patient 7, eTable 1) who had no GAD-ab and served as normal family control. Genomic DNA was purified from PBMC samples from all 12 individuals at the BGI Europe Genomic Center in Denmark. Sequencing was performed on the DNBseqTM NGS platform with libraries prepared using the BGI V4 kit and 100× coverage with PE100 sequencing. Reads were aligned with reference genome center GRCh38/hg38 using BWA-0.7.17 (r1188)16 producing binary alignment map (BAM) files. BAM files were duplicate marked with Picard tools. DeepVariant 1.317 was used to call variants using a model that is best suited for Illumina WES data. Resulting variant call format (VCF) files (one per each sample) were merged using GLNexus18 producing a cohort VCF file. OpenCRAVAT was then used to annotate the resulting cohort VCF file with a variety of functional databases (1,000 Genomes, gnomAD v3.0, Ensembl, Enrichr, Gene Ontology, CGD: Clinical Genomic Database, ClinGen Gene, ClinVar, DGIdb: The Drug Interaction Database, ENCODE_TFBS, Human Phenotype Ontology, Pharm GKB).
Variant Prioritization
Variants were prioritized according to 3 different strategies taking into consideration the rareness, quality, and biological input, using specialized filters:
1.
Common variants found to be shared by all individuals (n = 11) with GAD antibodies (SPS and non-SPS), categorized by type of variant
2.
Common missense variants found to be shared by all individuals (n = 11) with GAD antibodies (SPS and non-SPS) that are in coding genes
3.
Common variants found to be shared by all patients with SPS (n = 9)
Sanger Sequencing and Genotyping
Sanger sequencing was performed by CeMIA SA, Larissa, Greece, initially in 14 samples using custom designed primers for each single-nucleotide polymorphism (SNP) detected in KLK10 gene by WES to validate the discovered allelic variants in the WES analysis. The trace files of the sequencer were exported in ABI sequencer data file format (.ab1). The genomic sequence of KLK10 primary transcript was downloaded from Ensembl.19 To perform SNP genotyping, SnapGene20 was used in Janurary 2021 and again in 2022 to align the trace files with the reference genomic region. The samples used for the Sanger sequencing analysis were at first DNA samples extracted from PBMCs from 9 patients with stiff-person syndrome, 2 patients with autoimmune diabetes (T1D and LADA), and 3 healthy donors. After analyzing the first group of tested samples, Sanger sequencing was also performed in 10 additional GAD-positive patients with SPS (19 in total) and 15 additional autoimmune controls, 12 with MAG-autoimmune neuropathy with monoclonal gammopathy of unknown significance (MGUS) and 3 with inclusion body myositis.
Immunologic and Immunogenetic Studies in 6 Members of a Unique GAD-Positive Family
This family was included in the genetic sequencing for GAD-specific controls because 3 members in 3 generations had very high GAD antibody titers but only the index patient in mid-20s had SPS-SD (patient 7, eTable 1). The patient presented in 2012 with epilepsy and very high GAD-ab, but during our close follow-up (by MCD), the patient evolved in 5 years into typical SPS with truncal stiffness, spasms, and excitability, fulfilling SPS criteria.8,14 The patient's GAD-ab titers repeatedly tested were 2.5–4.6 × 106 IU/mL (2,500.000–4,600.000 by ELISA). The patient's father (patient 4), in late 50s, presented with TD1 and pernicious anemia with very high GAD-ab titers of 1.8 × 106 (1,800.000) ΙU/mL but without neurologic disease, and the paternal grandparent (patient 2), now in late 80s, presented with LADA, thyroiditis, and early signs of Alzheimer disease with very high GAD-ab titers of 0.7 × 106 (700,000) ΙU/mL (eTable 1 and eFigure 1). Because such high GAD antibody titers are not observed in patients with diabetes, we also checked their GAD-ab specificity with (a) immunohistochemistry in sagittal adult mouse brain sections; (b) cell-based assay in HEK293T cells transfected with whole GAD gene (cDNA clones from Origene), followed by anti-IgG1/2/3/4 secondary antibodies (Thermo Fischer); and (c) luciferase immunoprecipitation system (LIPS) assay for epitope mapping, as previously validated for GAD-positive SPS,15,21 measured by light emission (LU, light units) transformed to a log10 scale and color-coded. HLA haplotype identification in all family members (n = 6) was also performed with PCR sequence–specific primers with low-resolution analysis at 2 digits' level.
Standard Protocol Approvals, Registrations, and Patient Consents
Patients were recruited in the Neuroimmunology Unit, Department of Pathophysiology, of the Faculty Medicine, NKUA, as mentioned earlier. Archived serum and DNA samples were tested following an MTA agreement between NIH and NKUA. Clinical data were collected under established clinical protocols by the PI (MCD). Serum samples and PBMCs were collected and stored in −80°C following standard protocols and written informed consent according to the Declaration of Helsinki and the local ethical rules.
Data Availability
The data that support the findings of this study are available from the corresponding author, on reasonable request.
Results
Patient Characteristics
DNA samples of 11 patients with high-titer GAD antibodies and 1 healthy control member of the studied family were sequenced and included in the WES analysis. Patients' age ranged from 22 to 68 years. Eight patients (73%) were Americans/White and 3 Greeks/White (27%); 6 patients were female (55%), and 5 were male (45%). All patients had high GAD antibody titers. Nine of them (82%) had SPS; 1 (9%) had TD1 and pernicious anemia; and 1 (9%) had LADA, thyroiditis, and dementia. The complete patient characteristics are shown in Supplementary data (eTable 1).
WES Analysis Results: Filtering and Variant Prioritization
Variants Shared by All Patients With GAD-ab: Overrepresentation of ORAI1, IL7R, RYR3, HRC, and LILRA4 Genes
The necessary filters, applied for the variant annotation, revealed 52,494 variants. First, we focused on the 278 common variants found to be shared by all patients with GAD antibodies (n = 11, SPS and non-SPS). Most SNPs resided in intron regions (46.4%), followed by synonymous variants (22.3%), missense variants (20.5%), 5′ prime UTR variants (3.3%), lnc-RNA variants (2.9%), and other types of variants (Figure 1A). To dissect the strongest possible effect of the different SNPs in a gene, we focused on the nonsynonymous variants (missense and splice-site variants) in coding regions and the possible changes that ensue in the protein level. A total of 58 variants of 45 genes (data not shown) were used to identify categories of overrepresented genes by performing KEGG pathway analysis and gene ontology (GO) analysis using the Enrichr platform. The most overrepresented pathways revealed using KEGG pathway analysis across the patients with GAD antibodies, were primary immunodeficiency (p value = 3.6*10−3) with implicated genes the ORAI1 and IL7R and calcium signaling pathway (p value = 1.88*10−2) with implicated genes the HRC, ORAI1, and RYR3 (Figure 1B). Another interesting pathway that was highlighted was the B-cell receptor signaling pathway, with LILRA4 being the implicated gene.
(A) Type of variant pie chart in all patients with anti-GAD antibodies: type of variant in a total of 278 genetic variants that were identified to be common among all patients with anti-GAD antibodies (n = 11). (B) KEGG pathway analysis in all patients with anti-GAD antibodies: presented are the pathways that are highlighted in nonsynonymous variants (missense and splice-site variants) and in coding regions (58 variants). (C, D) Important genes in each pathway are pointed out. Volcano plot for GO Biological Process for all patients with anti-GAD-abs: some biological functions found to be important among the shared variants, according to GO Biological Process, match the KEGG pathway analysis results. Regulation of inorganic and calcium transport, the relaxation of the cardiac muscle, and the regulation of synapses are a few of them. GAD = glutamic acid decarboxylase.
The 58 common missense and splice-site variants found in all patients with GAD antibodies were also categorized by GO Biological Process. The biological functions found to be of high importance among the shared variants matched the KEGG pathway analysis at some level. Regulation of immune-related procedures, such as interferon-alpha production and synapse organization; regulation of inorganic and calcium transport; and the relaxation of the cardiac muscle are a few of them. The most overrepresented implicated genes matching the KEGG pathway analysis were ORAI1, RYR3, HRC, and LILRA4 (Figure 1, C and D).
Variants Shared Only by All Patients With SPS: The Revelation of KLK10 Gene
To address in the WES analysis whether there is any evidence of a single gene, or a cluster of genes, implicated in the pathogenesis of SPS, the next variant prioritization strategy was targeted on the common variants shared entirely by all patients with both SPS and GAD antibodies (n = 9). This analysis revealed 36 variants organized in 26 genes (data not shown) with a single emerged gene, the KLK10 (Figure 2).
Diagram represents the overlap between patients with anti-GAD antibodies only, patients with SPS and anti-GAD antibodies, and the presence of KLK10 gene of interest. GAD = glutamic acid decarboxylase; SPS = stiff-person syndrome.
All studied patients with SPS shared the same 7 variants in the KLK10 gene, 4 of them in the coding regions and 3 missense variants (each variant is represented by an arrow in Figure 3B). The structural representation of the enzyme shows that S50 and L149 are in 2 unstructured loops on the surface of KLK10, at the opposite side of the active site. S50 is located at the hinge between the crystallographically solved part of the enzyme and the predicted helical structure (Figure 3C).
(A) The KLK locus resides on the long arm of chromosome 19. The 15 KLK genes are tightly clustered. The classical KLK genes (KLK1, KLK2, and KLK3) are represented by red arrows while KLK4–KLK15 by gray arrows and the ψ1 pseudogene by a yellow one. (B) KLK10 gene structure: all KLK genes consist of 5 exons and 4 introns with a conserved intron phase pattern (I, II, I, 0). The positions of the codons for the active-site catalytic residues are highly conserved and highlighted in red, with histidine (H) codon near the end of coding exon 2, aspartic acid (D) codon in the middle of coding exon 3, and serine (S) codon near the start of coding exon 5. Black boxes represent noncoding exons. The 7 variants found in all patients with SPS are represented with multicolor vertical lines. The blue and yellow ones are in missense variants, the purple ones are in synonymous variants, and the green ones are in introns (C, D). Visual representation of human KLK10 protein structure. Ser50Ala and Leu149Pro substitutions are highlighted in yellow and blue, respectively. The amino acids that constitute the catalytic triad (His86, Asp137, and Ser229) are highlighted in red. (C) Position of S50 and L149 in relation to the active site. (D) Position of S50 in relation to the predicted helical structure.
Validation of the Discovered KLK10 Variants With Sanger Sequencing and Genotyping
Sanger sequencing was performed using custom designed primers for each SNP detected in KLK10 gene by WES, to confirm the polymorphisms discovered in the WES analysis. Sanger sequencing was initially performed in a total of 14 samples, the 12 used in the WES analysis as described above and 2 more family control individuals with no GAD antibodies: the mother and the grandfather of the index patient (Patients 3 and 1, respectively, Figure 4).
All SPS samples share the same 7 variants presented either as heterozygous or homozygous for each variant. The non-SPS samples, 2 with autoimmune diabetes (T1D and LADA) and 3 healthy donors, presented either the wild type of the SNPs or as heterozygous for each variant. SNP = single-nucleotide polymorphism; SPS = stiff-person syndrome.
Sanger sequencing and genotyping analysis verified that all patients with SPS share the same 7 variants in the KLK10 gene, 4 of them being in coding regions and 3 missense variants, being either heterozygous or homozygous for each variant. Among the 5 non-SPS samples, including those with TD1 and LADA and healthy family members, 4 (80%) had the wild-type variant and one (the mother of the patient with SPS) was heterozygous for all the variants (Figure 4).
Based on these early findings, we performed Sanger sequencing in 25 additional samples in 10 more GAD-positive patients with SPS (19 in total) and 15 immune controls (12 with MAG neuropathy and MGUS and 3 with myositis) in a total of 39 samples (with the 5 healthy or non-SPS family members mentioned earlier). These samples verified that all 19 patients with SPS share the same 7 variants in the KLK10 gene with 10 patients (52.6%) being homozygous for all variants, 8 (42.1%) being heterozygous, and one having the wild-type variant (5.2%). Among the total of 20 non-SPS control samples (including those with TD1 and LADA, the 3 healthy donors, and the 15 patients with autoimmune diseases), 45% had the wild-type variants and 45% were heterozygous for all variants, while 2 patients with MGUS were homozygous.
Immunologic Profile in the 6 Family Members
Purified IgGs from all 5 family members showed GAD staining in the molecular layer of the cerebellum only in the 3 family members with GAD-ab; the GAD-IgG from Patient 8 (index patient, with SPS and autoimmune epilepsy) initially showed a general neuropil staining in the hippocampus (Figure 5), cortex, striatum, and the fourth ventricle (data not shown), but by the time the patient's disease progressed to typical SPS, the staining pattern transitioned resembling the one of Patients 2 and 4 (Figure 5), which is unique for GAD-ab-positive patients having only diabetes. The fading staining in the hippocampus when transitioned to SPS is not, however, related to reduction of antibody titers because the patient's GAD-ab titers became even stronger when evolved to SPS, from 4.2 × 106 IU/mL initially to 4.6 × 106. In GAD gene–transfected HEK293T cells, IgG1 was their predominant GAD-IgG subclass including the 2 with only diabetes (Figure 6A). Serum samples of all family members (n = 6) were tested for identification of GAD epitope mapping using the LIPS assay to show immunoreactivity to GAD domains, the D1 that represents the amino-terminal domain (1–285 nt), D2 that represents the PLP domain with the enzyme core (286–1,331 nt), and fragment D3 that represents the carboxy-terminal domain (1,332–1,858 nt)15,21 (Figure 6B). 8 serum samples of the patients with SPS showed immunoreactivity with the amino-terminal domain while the Patients 2 and 4 with GAD-ab and LADA or T1D showed immunoreactivity with the carboxy-terminal domain (Figure 6C). The PCR sequence–specific primers identified a rare haplotype for the HLA II antigens; Patient 8 was identical to the patient's neurologically healthy brother (Patient 7) while Patients 1 and 2 (the paternal grandparents) were possibly homozygotes for the HLA-A genomic locus for the HLA-A*02 allele. Of interest, the very rare haplotype for the HLA II antigens, previously found in multiple GAD-positive families,12 was also observed in the present family with the HLA-DRB1*15 DQB1*05, inherited from Patient 2 (grandparent) to the son (Patient 4 with T1D) and grandchildren (Patients 7 and 8).
Isolated IgGs from all family members were tested in sagittal mouse brain sections. (A) Cerebellum: GAD staining pattern of the molecular layer (indicated by the white arrow) was present in all family members. (B) Hippocampus: Patient 8 had strong immunoreactivity in the hippocampus when she presented only with autoimmune epilepsy and staining faded by the time the patient evolved into typical SPS despite her antibody titers being even stronger (4.2 × 106 IU/mL initially and 4.6 × 106 when evolved to SPS). No other member was immunoreactive. a: IgG of Patient 8 from when she had autoimmune epilepsy, b: IgG of Patient 8 from when she evolved into SPS, c: IgG of Patient 4, d: IgG of Patient 2, mol: molecular layer, gr: granular layer. DG: dentate gyrus; GAD = glutamic acid decarboxylase; SPS = stiff-person syndrome.
(A) IgG subclasses in transfected HEK293T cells with GAD gene. IgG1 and IgG2 subclasses are present in all patients with anti-GAD antibodies, but one (patient 4). (B) Fragmentation of GAD gene in N-terminal domain, PLP-domain, and C-terminal domain for the LIPS assay. (C) Heatmap for epitope mapping of GAD gene in all family members. S1–8: Patient 1–Patient 8; E: epilepsy; S: SPS, C: control. GAD = glutamic acid decarboxylase; SPS = stiff-person syndrome.
Discussion
Genetic analysis of patients with GAD-positive sporadic SPS, triggered by a GAD-positive family with diverse symptomatology, revealed 2 relatively important genes, the ORAI1 and the LILRA4, related to novel pathways for cytokine production and synaptic organization. The ORAI1 is a calcium-selective ion channel that codes for calcium release–activated calcium channel protein-1 that plays an important role in the activation of T lymphocytes22; its role is highlighted by the loss-of-function variant of ORAI1 causing severe combined immunodeficiency in humans. The LILRA4 gene encodes an immunoglobulin-like cell surface protein, preferentially expressed in plasmacytoid dendritic cells, which is rapidly downregulated by interleukin-3 (IL-3). The LILRA4 gene is one of the 19 highly related genes that form a leukocyte immunoglobulin-like receptor gene cluster (LRC) at chromosomal region 19q13.4.23 Because ORAI1 and LILRA4 are expressed in immune and neuronal tissues, such as whole blood, white blood cells, lymph node, brain (cortex/cerebellum) and spinal cord,24 it is important to explore whether altered expression of any of these genes is implicated in the generation of GAD antibodies or susceptibility to develop GAD-positive SPS.
The most exciting observation of the genetic analysis on all studied patients, however, is the uncovering of variant polymorphisms in the KLK10 gene relevant to the clinicogenetic phenotype, as confirmed by the Sanger sequencing, which verified that 95% of the patients with SPS share the same 7 variants in the KLK10 gene. Four of these variants are localized in the coding regions and the other 3 are missense variants, being either heterozygous or homozygous for each variant.
Analysis performed on KLKL10 gene variants in the general population, as per 1,000 Genomes data set and gnomAD3, shows that, individually, the frequency of each variant is 0.5–0.7. Although, all 7 variants may seem individually common, it is their co-occurrence as a combination of the 7 variants in the same individuals that exerts a synergistic or additive effect on biological pathways relevant to SPS that may be contributing to the disease process. The observation that among the 19 patients with SPS, 52.6% were homozygous for all KLK10 variants and 42% were heterozygous, compared with the 20 non-SPS controls where 45% had the wild type and 45% were heterozygous, suggests that the specific combination of KLK10 variants seems to represent the haplotype that predisposes to SPS development. Testing more samples from healthy donors and patients with other neurologic autoimmunity is, however, essential for more extensive validation.
Human tissue kallikrein-related peptidases (KLKs) constitute a subgroup of 15 secreted trypsin-like or chymotrypsin-like serine proteases.25,26 KLK1 and 14 kallikrein-related peptidases (KLK2–KLK15) belong to a highly homologous family whose genes are located on chromosome 19q13.4 comprising the largest group of genes encoding proteases in the human genome (Figure 3A).25,27 Of special interest is that KLKs are clustered in the same genetic region as LILRA4; KLK10 is located in 19q13.41 and LILRA4 in 19q13.42.
Disruption of the balanced tissue-specific regulation of KLKs has been linked to several heterogeneous conditions including neurodegeneration, anxiety, inflammatory skin diseases, and cancer.28,29 There is emerging evidence that KLKs participate in the initiation of systemic inflammatory responses and immune-modulated disorders through proteolytic cascades and have a role in the development of proinflammatory responses mediated by the adaptive immune system with integral roles in T-cell recall response to viral antigens and monocyte activation.29-31 KLK10 is also expressed in peripheral nerves, neuroendocrine organs, and the choroid plexus epithelium, where altered forms of the protein might lead to disruption of ventricular barriers and contribute to disruption of blood-brain barrier, neurodegeneration, or inflammatory processes.32 The molecular location, however, of the noted polymorphisms cannot predict what effects they might have on the function or the expression pattern of the enzyme.
A recently conducted GWAS analysis from the large German consortium highlighted the role of PRKCB, a factor in cell death and survival, and of the SMARCA4, involved in stem cell renewal and proliferation, both biological pathways participating in immunity and neural function, implicated in Rh blood group cluster and neuronal development.33 These are relevant to our GWAS analysis because both studies collectively support that genomic loci implicated in immune-mediated processes may play a role in GAD autoimmunity.
Regarding the family immunogenetics, the index patient (Patient 8) presented with autoimmune epilepsy but evolved into typical SPS with very high GAD-ab titers8,14; 2 other family members, her father with TD1 and paternal grandmother with LADA, despite having equally high GAD-ab titers and almost identical immunocytochemical and epitope specificity pattern seen only in neurologically symptomatic patients with SPS-SD,8,14 did not have or develop any neurologic syndromes on repeated examination over a 4-year period. These findings support the view that GAD-abs on their own are not pathogenic for developing SPS but other antibodies or immune factors, such as T cells play a role, as previously discussed.1-5 Of interest, the switch of the staining pattern in the index patient, from cerebellar and hippocampal neuronal staining during her “epilepsy phase,” into predominantly “cerebellar” staining during her SPS phase (Figure 5) suggests that distinct epitope-specific cellullar, humoral, or immunogenetic factors may play a role in directing the manifestation of the predominant clinical phenotype.
The KLK10 gene variants within the family are also of interest because Patient 3, the healthy mother of the index patient with SPS (Patient 8), shares with her the same heterozygous variant profile while the patient's ancestors, her paternal grandmother (Patient 2) and father (Patient 4) who presented with high GAD-ab titers and DM1 or LADA but without neurologic disease, have the wild type of KLK10 variant. These gene family variants suggest that the heterozygous variant profile of the mother has been inherited to the daughter (Patient 8) and triggered the prominent SPS phenotype in conjunction with additional mechanisms, like their sharing the ORAI1, RYR3, HRC, and LILRA4 autoimmunity genes.
The HLA genetic predisposition of GAD autoimmunity in the family is also of interest. In the only previously reported study of a single family,12 the very rare haplotypes HLA-DRB1*15:01; DQA1*01:02; DQB1*05:02 were detected in 2 relatives with GAD neurologic syndromes; of interest, the same very rare haplotype HLA-DRB1*15 DQB1*05 was also shared by all members in our family with very high GAD-ab titers even without any neurologic manifestation. The nondetection of this haplotype in the other large studies of sporadic cases,33 where alternative HLA haplotypes but without genome-wide significance were detected,34 suggests that strong genetic predisposition with implications in GAD autoimmunity is mainly conferred by genomic regions outside the classical HLA alleles.
The main study limitations are the small number of DNA samples tested at the time this work began 6 years ago, the disease rarity, and the need to check more autoimmune neurologic diseases and healthy controls, an effort that has now restarted by Sanger sequencing 19 more patients that strengthened the early results but needs to be expanded. In summary, despite these limitations, the data suggest that in addition to 2 genes ORAI1 and LILRA4 implicated in autoimmunity, KLK10 seems to be a hotspot for polymorphisms/variants in patients with SPS, as confirmed by the validation cohort of all studied patients and controls where KLK10 was selectively deep-sequenced. Further structure-function experiments are in process exploring whether KLK10 has a role in immune dysregulation, possibly in conjunction with LILRA4 gene, or in a downstream pathogenetic process related to blood-brain barrier disruption or neuronal degeneration; coloclalization of GAD with KLK10 and possible effects of KLK10 on GABAergic neurotransmission are also in the exploratory phase.
Glossary
- BAM
- binary alignment map
- GABA
- gamma-aminobutyric acid
- GAD
- glutamic acid decarboxylase
- GAD-SD
- GAD antibody–spectrum disorder
- LIPS
- luciferase immunoprecipitation system
- MGUS
- monoclonal gammopathy of unknown significance
- PBMCs
- peripheral blood mononuclear cells
- SPS
- stiff-person syndrome
- TD1
- type 1 diabetes
- WES
- whole-exome sequencing
Acknowledgment
The authors thank the SPS Warriors for providing a gift fund to Thomas Jefferson University for Prof. Dalakas' research on SPS. The authors also thank Kalliopi Adam (Department of Immunology and Histocompatibility, General Hospital of Athens “Laiko”) for performing the HLA haplotype analysis. Prof. Dalakas additionally thanks all his patients with SPS-SD that he has been following for years at the NIH, Thomas Jefferson University, and University of Athens for their participation in the present and other immunological or therapeutic studies.
Supplementary Materials
Supplements
- Download
- 173.42 KB
References
1.
Dalakas MC. Stiff-person syndrome and related disorders-diagnosis, mechanisms and therapies. Nat Rev Neurol. 2024;20(10):587-601.
2.
Graus F, Saiz A, Dalmau J. GAD antibodies in neurological disorders—insights and challenges. Nat Rev Neurol. 2020;16(7):353-365.
3.
Martinez-Hernandez E, Arino H, McKeon A, et al. Clinical and immunologic investigations in patients with stiff-person spectrum disorder. JAMA Neurol. 2016;73(6):714-720.
4.
Balint B, Bhatia KP. Stiff person syndrome and other immune-mediated movement disorders—new insights. Curr Opin Neurol. 2016;29(4):496-506.
5.
Tsiortou P, Alexopoulos H, Dalakas MC. GAD antibody-spectrum disorders: progress in clinical phenotypes, immunopathogenesis and therapeutic interventions. Ther Adv Neurol Disord. 2021;14:17562864211003486.
6.
Saiz A, Blanco Y, Sabater L, et al. Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: diagnostic clues for this association. Brain. 2008;131(Pt 10):2553-2563.
7.
McKeon A, Robinson MT, McEvoy KM, et al. Stiff-man syndrome and variants: clinical course, treatments, and outcomes. Arch Neurol. 2012;69(2):230-238.
8.
Dalakas MC, Fujii M, Li M, McElroy B. The clinical spectrum of anti-GAD antibody-positive patients with stiff-person syndrome. Neurology. 2000;55(10):1531-1535.
9.
Burns TM, Jones HR, Phillips LH 2nd, Bugawan TL, Erlich HA, Lennon VA. Clinically disparate stiff-person syndrome with GAD65 autoantibody in a father and daughter. Neurology. 2003;61(9):1291-1293.
10.
Kuchling J, Shababi-Klein J, Numann A, Gerischer LM, Harms L, Pruss H. GAD antibody-associated late-onset cerebellar ataxia in two female siblings. Case Rep Neurol. 2014;6(3):264-270.
11.
Xiao Z, Shan J, Huang X, et al. Familial case reports of stiff-person syndrome. Clin Neurophysiol. 2015;126(12):2408-2409.
12.
Belbezier A, Joubert B, Montero-Martin G, et al. Multiplex family with GAD65-Abs neurologic syndromes. Neurol Neuroimmunol Neuroinflamm. 2018;5(1):e416.
13.
Muniz-Castrillo S, Ambati A, Dubois V, et al. Primary DQ effect in the association between HLA and neurological syndromes with anti-GAD65 antibodies. J Neurol. 2020;267(7):1906-1911.
14.
Dalakas MC, Li M, Fujii M, Jacobowitz DM. Stiff person syndrome: quantification, specificity, and intrathecal synthesis of GAD65 antibodies. Neurology. 2001;57(5):780-784.
15.
Burbelo PD, Groot S, Dalakas MC, Iadarola MJ. High definition profiling of autoantibodies to glutamic acid decarboxylases GAD65/GAD67 in stiff-person syndrome. Biochem Biophys Res Commun. 2008;366(1):1-7.
16.
Li H, Durbin R. Fast and accurate short read alignment with burrows-wheeler transform. Bioinformatics. 2009;25(14):1754-1760.
17.
Poplin R, Chang PC, Alexander D, et al. A universal SNP and small-indel variant caller using deep neural networks. Nat Biotechnol. 2018;36(10):983-987.
18.
Yun T, Li H, Chang PC, Lin MF, Carroll A, McLean CY. Accurate, scalable cohort variant calls using DeepVariant and GLnexus. Bioinformatics. 2021;36(24):5582-5589.
19.
Martin FJ, Amode MR, Aneja A, et al. Ensembl 2023. Nucleic Acids Res. 2023;51(D1):D933-D941.
21.
Fouka P, Alexopoulos H, Akrivou S, Trohatou O, Politis PK, Dalakas MC. GAD65 epitope mapping and search for novel autoantibodies in GAD-associated neurological disorders. J Neuroimmunol. 2015;281:73-77.
22.
Shaw PJ, Feske S. Regulation of lymphocyte function by ORAI and STIM proteins in infection and autoimmunity. J Physiol. 2012;590(17):4157-4167.
23.
Tavano B, Boasso A. Effect of immunoglobin-like transcript 7 cross-linking on plasmacytoid dendritic cells differentiation into antigen-presenting cells. PLoS One. 2014;9(2):e89414.
24.
Guzman R, Valente EG, Pretorius J, et al. Expression of ORAII, a plasma membrane resident subunit of the CRAC channel, in rodent and non-rodent species. J Histochem Cytochem. 2014;62(12):864-878.
25.
Prassas I, Eissa A, Poda G, Diamandis EP. Unleashing the therapeutic potential of human kallikrein-related serine proteases. Nat Rev Drug Discov. 2015;14(3):183-202.
26.
Borgono CA, Diamandis EP. The emerging roles of human tissue kallikreins in cancer. Nat Rev Cancer. 2004;4(11):876-890.
27.
Yousef GM, Chang A, Scorilas A, Diamandis EP. Genomic organization of the human kallikrein gene family on chromosome 19q13.3-q13.4. Biochem Biophys Res Commun. 2000;276(1):125-133.
28.
Filippou PS, Ren AH, Soosaipillai A, et al. Kallikrein-related peptidases protein expression in lymphoid tissues suggests potential implications in immune response. Clin Biochem. 2020;77:41-47.
29.
Filippou PS, Karagiannis GS, Musrap N, Diamandis EP. Kallikrein-related peptidases (KLKs) and the hallmarks of cancer. Crit Rev Clin Lab Sci. 2016;53(4):277-291.
30.
Emami N, Diamandis EP. New insights into the functional mechanisms and clinical applications of the kallikrein-related peptidase family. Mol Oncol. 2007;1(3):269-287.
31.
Sotiropoulou G, Pampalakis G. Kallikrein-related peptidases: bridges between immune functions and extracellular matrix degradation. Biol Chem. 2010;391(4):321-331.
32.
Mella C, Figueroa CD, Otth C, Ehrenfeld P. Involvement of kallikrein-related peptidases in nervous system disorders. Front Cell Neurosci. 2020;14:166.
33.
Strippel C, Herrera-Rivero M, Wendorff M, et al. A genome-wide association study in autoimmune neurological syndromes with anti-GAD65 autoantibodies. Brain. 2023;146(3):977-990.
34.
Thaler FS, Bangol B, Biljecki M, Havla J, Schumacher AM, Kumpfel T. Possible link of genetic variants to autoimmunity in GAD-antibody-associated neurological disorders. J Neurol Sci. 2020;413:116860.
Information & Authors
Information
Published In
Copyright
Copyright © 2025 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.
Publication History
Received: July 31, 2024
Accepted: December 10, 2024
Published online: February 11, 2025
Published in issue: March 2025
Disclosure
The authors report no relevant disclosures. Go to Neurology.org/NN for full disclosures.
Study Funding
The study was funded by the Neuroimmunology Unit National and Kapodistrian University of Athens, Greece (Chief Prof. Dalakas). Popianna Tsiortou was supported by the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT), under the HFRI PhD Fellowship grant (GA. no280).
Authors
Author Contributions
P. Tsiortou: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. H. Alexopoulos: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. K. Kyriakidis: major role in the acquisition of data; analysis or interpretation of data. M. Kosmidis: major role in the acquisition of data. C. Barba: major role in the acquisition of data. S. Akrivou: major role in the acquisition of data. I. Michalopoulos: major role in the acquisition of data; analysis or interpretation of data. P. Politis: major role in the acquisition of data; analysis or interpretation of data. M.C. Dalakas: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design.
Metrics & Citations
Metrics
Citation information is sourced from Crossref Cited-by service.
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
Cited By
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
The neurology.org payment platform is currently offline. Our technical team is working as quickly as possible to restore service.
If you need immediate support or to place an order, please call or email customer service:
- 1-800-638-3030 for U.S. customers - 8:30 - 7 pm ET (M-F)
- 1-301-223-2300 for customers outside the U.S. - 8:30 - 7 pm ET (M-F)
- [email protected]
We appreciate your patience during this time and apologize for any inconvenience.