Abstract
Objective: To characterize the specificity of anti-GAD65 antibodies in patients with stiff person syndrome (SPS), quantify antibody titers, and examine antibody production within the CNS.
Methods: The authors studied 18 patients with SPS and positive serum immunoreactivity to gamma-aminobutyric acid (GABA)-ergic neurons. The reactivity of serum and CSF to purified GAD antigen was examined by Western blots, and the anti-GAD65 antibody titers in serum and CSF were quantified by ELISA and compared with 70 disease controls (49 with other autoimmune disorders and 11 with insulin-dependent diabetes mellitus). The intrathecal synthesis of anti-GAD65 IgG was calculated, and the functional significance of the antibodies was examined by measuring the GABA levels in the CSF.
Results: The serum and CSF of all selected patients with SPS had high anti-GAD65 titers (from 7.0 to 215 μg/mL in serum and from 92 to 2500 ng/mL in CSF) and immunoreacted strongly with recombinant GAD65 on Western blots and with GABA-ergic neurons on rat cerebellum. Among controls, only the serum of eight patients with insulin-dependent diabetes mellitus had low anti-GAD65 antibody titers (from 200 to 1760 ng/mL) but no reactivity to recombinant GAD65. The CSF showed oligoclonal IgG bands in 10 (67%) of 15 patients and an increased anti-GAD65-specific IgG index in 11 (85%) of 13. The mean level of GABA in the CSF was lower in patients with SPS than in controls.
Conclusions: In patients with SPS, there is marked intrathecal antibody response against neuronal GAD65 epitopes, indicating a clonal B cell activation in the CNS. Anti-GAD65 antibodies at high titers, when confirmed with immunoblots, are highly specific for SPS and appear to impair GABA synthesis.
Stiff person syndrome (SPS) is a rare and disabling disorder characterized by muscle rigidity and episodic spasms involving the axial and limb musculature.1-4⇓⇓⇓ The cause of SPS is unknown, but an autoimmune pathogenesis is suspected based on the following: presence of antibodies against GAD65 (anti-GAD65), the rate-limiting enzyme for the synthesis of gamma-aminobutyric acid (GABA)5,6⇓; association with other autoimmune disorders and autoantibodies3,4⇓; and response to immunother-apies.7-9⇓⇓ The significance and specificity of anti-GAD65 antibodies in SPS has been questioned10 because these antibodies have been seen in some other disorders, particularly insulin-dependent diabetes mellitus (IDDM), which often coexists with SPS. Because SPS appears to have a central origin,4 determination of these antibodies in the CSF is expected to be informative in the pathogenesis of the disease.
To explore the specificity of anti-GAD65 antibodies in the CSF and substantiate the autoimmunity of SPS, we examined the immunoreactive pattern of serum and CSF IgG to GABA-ergic neurons by immunocytochemistry and to linear epitopes of GAD65 by immunoblotting; quantified the anti-GAD65 antibody titers by ELISA and compared the titers with those of patients with other autoimmune neuromuscular disorders and IDDM; examined the intrathecal IgG anti-GAD65 antibody synthesis; and measured the GABA level in the CSF in a few representative patients.
Methods.
Patients.
We examined the sera of 18 patients with SPS selected from a total of 26 patients referred to the Neuromuscular Diseases Section of the National Institute of Neurological Disorders and Stroke. Patient selection was biased because of the need to be positive for anti-GAD antibody. All patients fulfilled the following clinical criteria of SPS3,4⇓: insidious onset of muscular rigidity in the limbs and axial (trunk) muscles, prominent in the abdominal and thoracolumbar paraspinals with difficulty in turning or bending; continuous cocontraction of agonist and antagonist muscles with inability to relax, as confirmed clinically and electrophysiologically1-4⇓⇓⇓; episodic spasms superimposed on the rigidity and precipitated by unexpected noises, tactile stimuli, or emotional upset; absence of any other neurologic disease that could explain stiffness and rigidity; absence of any systemic disorder, including cancer, chronic pain syndromes, or psychiatric illness; and positive immunoreactivity to GABA-ergic neurons by immunocytochemistry, performed as described below.
Immunocytochemistry.
Immunocytochemical study was performed on frozen sections of rat cerebellum using serum and CSF as described.3 Immunostaining of GABA-ergic neurons and terminals was detected by an indirect immunofluorescence technique using fluorescein-conjugated goat antihuman IgG.
Anti-GAD antibody titers determined by ELISA in serum and CSF.
For detection and quantitation of anti-GAD65 antibodies, we used ELISA kits (Boehringer, Mannheim, Germany). At least three independent experiments in duplicate wells for each of the 18 SPS sera and 14 CSF were performed, and an average value for each sample was obtained. Antibody concentration was calculated based on the standard curve, according to the manufacturer’s instructions, by plotting the average absorbancies on the y-axis against the standard concentration of purified anti-GAD65 provided on the x-axis. The intraassay coefficient of variation was on average 9%. Serum samples with values between 32 and 50 ng/mL decreased within the borderline area of the testing system; samples with concentration more than 50 ng/mL were considered positive for anti-GAD65 antibodies, and samples with concentration less than 32 ng/mL were considered negative. Dilutions, 1:500 for the serum and 1:125 for the CSF, were used. We tested serum from 18 and CSF from 14 patients with SPS who not only fulfilled the strict clinical criteria outlined above but also had positive immunostaining on GABA-ergic neurons on the rat cerebellum and serum from 49 patients with various autoimmune neuromuscular disorders, including 15 with MG, 17 with polymyositis, 8 with chronic inflammatory demyelinating polyradiculoneuropathy, 7 with dermatomyositis, and 2 with HTLV-I associated myelopathy. The CSF from eight of these patients was also examined. We also tested serum from 10 patients with nonspecific muscle spasms or cramps and serum from 11 patients with IDDM.
CSF studies: oligoclonal bands, IgG index, and intrathecal anti-GAD65 synthesis.
The CSF from these patients was examined for total immunoglobulins, immunofixation electrophoresis, IgG index, oligoclonal bands, and intrathecal synthesis of IgG-specific anti-GAD65 antibodies. For the latter, the serum albumin–to–CSF albumin ratio was determined first as an indicator of the integrity of the blood–brain barrier (normal, >130). The index for intrathecal synthesis of IgG (IgG index) was calculated according to the formula of Link,11 as modified:12
The index is considered normal when it is below 0.67.11,12⇓
The index for intrathecal synthesis of IgG against GAD65 antibodies (anti-GAD65-specific IgG index) was calculated using the values obtained with the ELISA, based on the following formula in paired serum and spinal fluid specimens:
The final concentration of antibodies was obtained after normalization of the different dilutions to the same standard curve. Values higher than the IgG index, particularly higher than 1, are strong indicators for intrathecal synthesis of antibody-specific IgG.11,12⇓
GABA level in the CSF.
The GABA level was measured by conventional chromatographic analysis in eight representative patients and several controls, as described.13
Western blotting with serum and CSF.
Rat brain was washed by addition of phosphate-buffered saline (pH 7.4) and disposal of the supernatant twice. The tissue was then transferred to a Dounce homogenizer, prechilled on ice, and suspended in 20 volumes of lysis buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% (w/v) sodium dodecyl sulfate, 2 mM EGTA, 10% glycerol, 25 μg/mL phenylmethylsulfonyl fluoride, 100 μM leupeptin, and 10 μg/mL aprotinin. The tissue suspension was incubated in lysis buffer for 15 minutes, then homogenized and centrifuged at 100,000× g for 45 minutes at 4 °C. The supernatant was kept for further analysis.
A volume from rat brain extract corresponding to 20 or 200 μg protein, or an equal volume containing 10 μL of purified recombinant GAD65 protein (catalog no. 1556240, Boehringer) was processed on 10% SDS-polyacrylamide gel and transferred onto a polyvinylidene difluride membrane (Millipore, Bedford, MA) for immunoblotting. Sera and CSF from patients with SPS and control subjects (diluted from 1: 200 to 1:2000 for the serum and from 1:1 to 1:25 for the CSF) or mouse monoclonal antibodies against GAD65 (1:5000; Chemicon, Temecula, CA) were applied. Antibody binding was detected in a direct labeling method with peroxidase-conjugated polyclonal antibodies against human or mouse Ig (Amersham International, Buckinghamshire, UK). Enhanced chemiluminescence reagent (Amersham International) was used for the detection of the immunoreactive bands.
Results.
Immunocytochemistry with the patients’ sera and CSF.
Based on the preselection criteria, the sera from all patients with SPS immunoreacted positively with GABA-ergic neurons and GABA nerve terminals (figure 1). An identical pattern was also obtained with pooled CSF from six of these patients.
Figure 1. Immunofluorescent staining of a rat cerebellum section (20 μm) with serum or CSF from patients with stiff-person syndrome (1:1000 dilution for the serum, 1:25 dilution for CSF; 2-day incubation). The pattern of immune staining corresponds to the well-known distribution of the GABA-ergic system in rat brain. Heavy labeling is observed around granule cells in the granule cell layer, Purkinje cells, and in the molecular layer. (original magnification, ×270).
Anti-GAD65 antibody titers in sera and CSF determined by ELISA.
The serum from all patients with SPS had high anti-GAD65 antibody titers ranging from 7.5 to 215 μg/mL (positive values are greater than 50 ng/mL) and optical densities ranging from 2.0 to 3.2 at serum dilution of 1:500 (figure 2). Sera from 8 of the 11 patients with IDDM also had detectable anti-GAD65 antibodies, but their titers at a dilution of 1:500 were 50 times lower than the titers of patients with SPS, ranging from 0 to 1,760 ng/mL, and optical densities ranging from 0.2 to 0.9 (figure 2). Among the other 58 patients with autoimmune diseases, 2 of 15 patients with MG and two patients with dermatomyositis had low anti-GAD65 antibody titers ranging from 100 to 265 ng/mL and optical densities less than 0.3, at a dilution of 1:500 (figure 2). None of the patients with nonspecific muscle stiffness or other autoimmune neuromuscular diseases (NDS) had positive anti-GAD65 antibodies.
Figure 2. Comparison of anti-GAD65 antibody titers between patients with stiff person syndrome, insulin-dependent diabetes mellitus (IDDM), and other autoimmune neuromuscular disorders. Note the higher titers in patients with stiff person syndrome than in patients with IDDM or control subjects.
In the CSF, anti-GAD65 antibody titers were detected in all patients with SPS and ranged from 92 to 2,500 ng/mL (table). The CSF of eight disease controls did not contain anti-GAD65 antibodies.
Spinal fluid studies in patients with stiff person syndrome
Serum and CSF immunoreactivity to recombinant GAD65.
By immunoblotting, the serum and CSF from patients with SPS, but not with IDDM, recognized a 65-kilodalton protein in brain homogenates corresponding to GAD65, as confirmed with the monoclonal antibody against GAD65 (figure 3, left panel). The linear epitope of recombinant GAD65, immunoreacted only with the sera or CSF of patients with SPS who had high anti-GAD65 titers and optical densities greater than 1 (figure 3, right panel). None of the patients with IDDM or the other autoimmune controls immunoreacted with purified GAD65. Pooled IDDM serum with the highest anti-GAD65 antibody concentration of up to 1.5 μg/mL did not react with purified recombinant GAD65 (figure 3, lane 1 right panel). In contrast, SPS serum, when diluted to as low as 1.4 μg/mL of anti-GAD65 antibody titers to match the mean titers of IDDM specimens, continued to react strongly to the same recombinant GAD65 protein, indicating unique specificity of SPS sera to recognize epitopes of GAD65.
Figure 3. Western blot analysis of autoantibodies against GAD65 from the serum and CSF of patients with stiff person syndrome or control subjects. Left: Strong immunoreactivity to a 65-kilodalton protein of rat brain extracts is seen with the serum from a representative patient with stiff person syndrome (lane 2). Mouse monoclonal antibodies against GAD65 recognizes the same autoantigen (arrow, lane 1). The molecular weight of protein standards is shown on the right. The same, though weaker, immunoreactivity was obtained with CSF. Right: Strong immunoreactivity to purified recombinant GAD65 (arrow, lane 2) is seen with the serum from a patient with stiff person syndrome or with pooled CSF from six patients with stiff person syndrome. No reactivity was seen with pooled serum from patients with insulin-dependent diabetes mellitus (lane 1).
CSF oligoclonal bands and intrathecal synthesis of anti-GAD65 antibodies.
The blood–CSF barrier and the IgG index were within the normal range (<0.67) in all patients except one (patient 12, table), who had a slight increase. Based on the specific anti-GAD65 IgG index, there was evidence of de novo intrathecal IgG synthesis of anti-GAD65 antibodies in 11 (85%) of 13 patients (table). In eight (61%) of them, there was marked intrathecal synthesis, but in three (13%) of them, the synthesis was at the lowest cutoff point (table). Oligoclonal IgG bands were detected in 10 (67%) of 15 patients. These IgG bands are probably anti-GAD65-specific, as suggested by the strong immunoreactivity to rat GABA-ergic neurons (figure 1) and to purified GAD65 obtained with pooled CSF from six of them (figure 3).
As indicated in the table, the mean values of the total GABA level in the CSF of eight representative patients was 3.0 nmol/mL (range, 1.6 to 4.3 nmol/mL), compared with a mean of 5.8 ± 1 nmol/mL in the CSF of control subjects. The level of GABA in the CSF did not inversely correlate with the CSF anti-GAD65 antibody titers. However, the number of CSF samples in which the GABA level was examined was rather small to make definitive correlations.
Discussion.
Patients with SPS, but not control subjects, have high IgG anti-GAD65 antibody titers detected by ELISA. Because the serum and CSF IgG of the same patients immunostain also GABA-ergic neurons and recognize the GAD65 epitope by Western blots, high titers of anti-GAD65 antibodies confer specificity for SPS. As determined by their concurrent quantification in the serum and CSF, these antibodies are synthesized intrathecally and seem to impair the in situ synthesis of GABA. The findings indicate that in patients with SPS, there is an ongoing autoimmune response in the CNS compartment and provide a possible explanation for the clinical symptomatology.
The pathogenic role of anti-GAD65 antibodies in SPS has been questioned10,14⇓ because GAD is a cytoplasmic antigen and the few autopsy reports from the brains of these patients lack histologic changes15 or show mild, mostly nonspecific findings.16,17⇓ However, recent data suggest that GAD easily becomes membrane-associated to synaptic vesicles, through complex formation with the heat shock protein 70 family18 and can be recognized by the immune system. Inhibition of GAD at the synaptic vesicles may affect GABA because there is a functional link between GABA synthesis and vesicular GABA transport at the nerve terminals.18 Furthermore, the serum IgG anti-GAD65 antibodies from patients with SPS, but not from patients without such antibodies, inhibit GAD activity in vitro and impair the synthesis of GABA in crude rat cerebellar extracts without causing structural changes in the GABA-ergic neurons.19,20⇓ Consequently, if the anti-GAD65-specific antibodies are produced in the brain parenchyma and immunoreact in situ with the GABA-ergic neurons, they may also cause dysfunction of these neurons in vivo. The present observations support this notion and provide evidence that in SPS, the anti-GAD65 antibodies are not passively present in the CSF as markers of the autoimmune process, but they are synthesized intrathecally presumably by B cells active in the CNS compartment. The in situ produced anti-GAD65 antibodies appear to inhibit the function of GAD and impair the synthesis of GABA in vivo, as supported by the following: reduced GABA level in the brain compartment, shown by our magnetic resonance spectroscopy studies21; reduced GABA level in the CSF of the eight patients measured in this study; and hyperexcitability of the motor cortex caused by impaired GABA-ergic transmission, as shown by the electrophysiologic studies we have performed in the same patients.22
Because GABA is the brain’s predominant inhibitory neurotransmitter and 25 to 35% of all synapses are GABA-ergic, a reduction of GABA level as the result of anti-GAD antibodies can explain the noted muscle hyperactivity. The clinical improvement of several patients with SPS after immunotherapies,7-9,23,24⇓⇓⇓⇓ the lack of persistent focal neurologic signs other than increased tone, and the reduction of GABA level in the brain without any structural changes in the MRI21 support the view that SPS is a functional rather than a structural disorder with an ongoing immune response in the CNS. Whether anti-GAD65 antibody titers correlate with disease severity or change after successful immunotherapies remains to be determined.
The strong association of SPS with HLA-DR alleles3,25⇓ suggests that in patients of a certain immunogenetic background, T lymphocytes may be activated during the initiation of the disease by GAD peptide fragments bound to specific loci of the major histocompatibility complex (MHC) molecule. In genetically predisposed individuals, enzymes such as GAD may be targets of an autoimmune disease when trivial structural defects arising during organ and tissue ontogenesis allow enzyme leakage to the surface and attract lymphocytes generating an autoantibody response.26 A new such disease with autoimmune similarities to SPS is the autoimmune polyendocrine syndrome in which a new autoantigen, tryptophan hydroxylase, the enzyme responsible for serotonin synthesis, has been found to cause gastrointestinal dysfunction.26
In contrast to IDDM and other autoimmune disorders, high anti-GAD65 antibody titers detected by ELISA and confirmed by Western blots, are specific for SPS because they recognize distinct epitopes of GAD65, as reported previously.27 This is also supported by the series of dilutions we performed in the serum of patients with SPS to match the low concentration of anti-GAD65 antibodies seen in IDDM. The only other CNS disorder in which high anti-GAD65 antibody titers appear to be pathogenic or produced intrathecally is in patients with cerebellar ataxia and therapy-resistant epilepsy,28,29⇓ presumably because these antibodies gain access in the GABA-rich cerebellar neurons. Whether anti-GAD65 antibodies selectively affect only certain GABA-ergic pathways or there are additional target antigens against inhibitory neurotransmitters remains to be determined.
Acknowledgments
Acknowledgment
The authors thank Dr. Elinor Ben-Menachem, Institute of Clinical Neuroscience, Sahlgrenska University Hospital, University of Göteborg, Sweden for measuring the level of GABA in the CSF specimens.
- Received March 1, 2001.
- Accepted April 24, 2001.
References
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.