Novel brain reactive autoantibodies: Prevalence in systemic lupus erythematosus and association with psychoses and seizures

Journal of Neuroimmunology 169 (2005) 153 – 160 www.elsevier.com/locate/jneuroim

Novel brain reactive autoantibodies: Prevalence in systemic lupus erythematosus and association with psychoses and seizures S.K. Tin a, Q. Xu b, J. Thumboo a,b, L.Y. Lee c, C. Tse a, K.Y. Fong a,b,* a

Department of Rheumatology and Immunology, Singapore General HospitalOutram Road, Singapore 169608, Singapore b Department of Medicine, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore c Department of Clinical Research, Singapore General Hospital, Outram Road, Singapore 169608, Singapore Received 4 June 2005; accepted 25 July 2005

Abstract Autoantibodies can cause neuropsychiatric manifestations in lupus patients by altering the physiological function of neuronal cells. In this study, we identified Brain Reactive Autoantibodies (BRAAs) against murine neuronal membrane proteins (M.W. 27.5 and 29.5 kD) and found them correlating with psychosis and/or seizures in lupus patients. They were specific to neuronal membrane tissues of mammalian origin and are significantly associated with psychosis and/or seizures ( p < 0.0001). These membrane proteins mass spectrometry profiles did not match to any published protein sequences. These BRAAs may play important roles in the pathophysiology of neuropsychiatric lupus. D 2005 Elsevier B.V. All rights reserved. Keywords: Neuropsychiatric lupus; Brain reactive autoantibodies; Psychosis; Seizures

1. Introduction Systemic Lupus Erythematosus (SLE) is a chronic and potentially fatal autoimmune disease which attacks many organ systems during its disease course. Significant organs involved in SLE are the kidneys (lupus nephritis) and the brain (neuropsychiatric lupus) (Cameron, 1999; West et al., 1995). It was reported that between 31% and 70% of SLE patients has significant neuropsychiatric (NP) manifestations in the course of their disease (West et al., 1995; Kaell et al., 1986; Futrell et al., 1992; Sibley et al., 1992; Rood et al., 1999). The spectrum of NP manifestation in SLE includes both neurologic and psychiatric features, many of which can be subjective to both patients and clinicians. They vary from overt neurologic dysfunctions due to psychoses, seizures, to subtle abnormalities in neurocognitive functions such as * Corresponding author. Department of Rheumatology and Immunology, Singapore General Hospital Outram Road, Singapore 169608, Singapore. Tel.: +65 6321 4059; fax: +65 6220 7765. E-mail address: [email protected] (K.Y. Fong). 0165-5728/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2005.07.015

memory, intellect. (Hanly and Hong, 1993). They cannot be easily differentiated because lupus itself, drugs used in lupus management or other associated pathological conditions may be responsible for the neuropsychiatric manifestations. Currently the diagnosis of NP-lupus has to depend largely on exclusion of other causes, as the pathogenesis is yet unclear (Hanson et al., 1992). More than 100 different autoantibodies have been found in both human and animal lupus studies (Sherer et al., 2004). Among these autoantibodies, some may become ‘‘pathogenic’’ against self-antigens and result in (a) pathological change in the target organ and/or (b) binding of autoantibodies to target antigen that may lead to altered physiological function of target antigens (Tung, 1994). Pathogenic roles of many autoantibodies are not well defined in SLE. Prevalence of brain reactive autoantibodies detected in sera and cerebrospinal fluid (CSF) in both human and animal studies were reported (Hanson et al., 1992; Bluestein and Zvaifler, 1976; Wilson et al., 1979; Toh and Mackay, 1981; Hoffman et al., 1987; Moore, 1992; Zameer and Hoffman, 2001). However, not all brain reactive autoantibodies are responsible for NP-lupus, i.e. they are not

154

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

necessarily of pathogenic importance (Khin and Hoffman, 1993). Of these autoantibodies, some pathogenic autoantibodies may target neuronal tissue and lead to neuropsychiatric manifestations. These autoantibodies can bind to neuronal surface membranes (Koren et al., 1992) and cause injury via a direct toxic effect, or they disrupt the physiological functions of neuronal tissue when bound, but without causing cell death. In preliminary results, we identified novel antibodies present in the sera of lupus patients that were reactive to neuronal membrane antigens. We then proceeded to study these autoantibodies in an unselected group of SLE patients and evaluate whether they are associated with neuropsychiatric manifestations in lupus patients.

2. Materials and methods 2.1. Human subjects and sera collection One hundred unselected SLE patients, fulfilling the 1997-updated ACR criteria for classification of SLE were recruited into the study (Hochberg, 1997). One hundred and thirty non-lupus patients (20 rheumatoid arthritis (RA), 20 osteoarthritis (OA), 20 ankylosing spondylitis (AS), 20 psoriatic arthritis (PsA) and 50 primary anti-phospholipid syndrome (APS)) were also recruited, and all of them satisfied current diagnostic criteria for their corresponding diseases. One hundred and two healthy controls, recruited from among healthcare workers and medical students, were also included in this study. Five milliliters of peripheral venous blood in plain tubes were obtained after informed consent. The venepunctures were timed to coincide with review visits at the Rheumatology Clinics. They were processed to separate the sera from clotted blood using standard protocol. All sera were kept in 80 -C freezer until further experiments. A standard protocol was used to record lupus patients’ clinical data, ACR criteria and neuropsychiatric manifestations by chart review. This study was approved by the hospital’s institutional review board. 2.2. Animal selection Two 8-week-old female BALB/c mice purchased from the Animals Holding Unit, National University of Singapore, were used to extract membrane proteins from brain, heart, liver, kidney and spleen. The female Wistar rat’s brain was a gift from Ms. Irene Kee of Department of Experimental Surgery, Singapore General Hospital. The brains of pig, chicken, duck, quail, frog and fish (Indian Mackerel—Rastrelliger kanagurta) were purchased from local markets. Human brain membrane protein (normal adult male) was purchased from BioChain, USA.

2.3. Membrane protein extraction All chemicals were purchased from Sigma Aldrich, unless otherwise mentioned. The frozen/fresh tissue was thawed and cut into small pieces, followed by homogenization using mortar and pestle in 1  PBS containing a cocktail of proteinase inhibitors (5 Ag/ml leupeptin, 5 Ag/ml approtinin and 1 mM phenyl methyl sulfonyl fluoride). Brain lysates were further homogenized using syringe and needles method (starting from 19 G and gradually reduced in bore sizes to 27 G). The water-insoluble pallet was separated by centrifugation at 12,000 g for 10 min at 4 -C. Repeated washings with 1  PBS containing cocktail of protein inhibitors were done until proteins were not detected in wash buffer using BioRad protein assay. The water-insoluble protein was dissolved in 1  PBS containing 6 M urea together with 1 mM phenyl methyl sulfonyl fluoride and incubated on shaker for 15 min at room temperature. The urea-soluble protein was separated using centrifugation at 12,000 g for 30 min at 20 -C. The protein concentration was measured using BioRad protein assay and aliquoted proteins were kept in 80 -C freezer until future use. The membrane proteins were extracted from different mouse tissues (brain, heart, liver, kidney and spleen) using the protocol described. The same protocol was also used to extract neuronal membrane proteins from different species (rat, pig, duck, chicken, quail, frog and fish). 2.4. Immunoblotting The sodium dodecyl sulphate – polyacrylamide gel electrophoreses (SDS – PAGE) were prepared using analytical comb or preparative comb according to the needs of experiments. The membrane proteins (20 Ag proteins per well for analytical comb or 200 Ag proteins per well for preparative comb) were denatured and separated in 20% SDS –PAGE. Upon completion of protein separation, the polyacrylamide gels were either visualized with Coomassie blue or silver stain or transferred to Hybond nitrocellulose membrane (Amshersham). Coomassie Brilliant Blue R-250 (BioRad) and SilverPlus silver staining kit (BioRad) were compatible with Mass Spectrometry (MS) analysis. The electrophoretic transfer of separated proteins to nitrocellulose membrane was performed at 100 constant Volts for 2 h using buffer transfer tank in cold room. The nitrocellulose membrane, on which the proteins were transferred, was incubated in blocking solution (3% scanned milk powder in 1  TBS-T; 1  TBS containing 0.1% Tween 20) on shaker for 1 h at room temperature to block non-specific reactions prior to sample incubation. Depending on the purpose of experiment, nitrocellulose membrane with proteins separated using analytical combs were incubated with samples while those with proteins separated using preparative combs were cut into strips and

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

155

Table 1 Demographic data of controls, lupus and non-lupus patient cohort

2.6. 2-D gel electrophoresis

Lupus (n = 100) Female : male ratio Mean age (range) Race

Prior to 2-D gel electrophoresis, the water-insoluble pallet was dissolved in 2-DE sample solubilization buffer containing 6 M urea, 50 mM DTT, 4% CHAPS, 0.2% carrier amphyolytes (1 : 2 Bio-lyte 7/9:Bio-lyte 8/10), 0.0002% Bromophenol Blue. Protein concentration was determined by BioRad RC DC protein assay. One hundred micrograms of brain membrane proteins each were applied onto a pair of pH 7 –10, 7 cm ReadyStrip IPTG strip (BioRad) during passive rehydration step. The first dimension (Isoelectric Focusing) was done for 8000 V-h at 15 -C on Protean IEF cell (BioRad). Upon completion of first dimensional gel electrophoresis, the strips were equilibrated in SDS-buffers prior to second dimensional gel electrophoresis. The second dimensional separation was done in 20% SDS –PAGE. The paper wick soaked with mouse brain membrane proteins was added next to IPG strip when the strip was loaded onto SDS – PAGE gel. Upon completion of second dimensional gel electrophoresis, one was visualized with silver staining and the other gel was used for immunoblotting with BRAA-positive serum. Immunoblotting was performed according protocol described earlier. The immunoblot-matched gel spots were punched from silver stained polyacrylamide gel and sent for peptide mass finger-printing (PMF) using MALDI-TOF-MS (matrixassisted laser desorption ionization-Time of Flight-Mass Spectrometry) at Agenica Research Pte Ltd, Singapore.

Average disease duration (range) Non-lupus (n = 130) Female : male ratio Mean age (range) Race

Control (n = 102) Female : male ratio Mean age (range) Race

9:1 39 years (19 – 71 years) Chinese 81 Malay 14 Indian 5 11 years (1 – 20 years) 1:1 52 years (21 – 81 years) Chinese 91 Malay 10 Indian 29 2.7 : 1 35 years (21 – 41 years) Chinese 82 Malay 14 Indian 6

Non-lupus patients include 20 rhuematoid arthritis (RA) patients, 20 osteoarthritis (OA) patients, 20 ankylosing spondylitis (AS) patients, 20 psoriatic arthritis (PsA) patients and 50 primary anti-phospholipid syndrome (APS) patients.

incubated with samples separately. The samples, diluted at 1 : 200 in blocking solution, were incubated with membrane on shaker overnight at 4 -C. Membranes were washed by 1  TBS-T thrice with 10 min soaking while shaking between washes. The membranes were further incubated with goat anti-human Ig conjugated with Alkaline Phosphatase (Chemicon) diluted in blocking buffer 1 : 10,000 on shaker for 1 h at room temperature. The membranes were, after incubation, washed again with 1  TBS-T thrice with 10 min, shaking between washes. The 5-Bromo-4-Chloro3V-Indolyphosphate p-Toluidine/Nitro-Blue Tetrazolium Chloride (BCIP/NBT) substrate (Zymed) was used to visualise the positive protein bands on the membranes. When optimal colour developed, the substrate was removed, membranes dried and documented. 2.5. Detection of BRAAs A total of 100 lupus patients, 130 non-lupus patients and 102 healthy individuals were screened for BRAAs using membrane proteins extracted from mouse brain as antigens. The immunoblotting screening was performed using 20% SDS – PAGE for higher resolution. The Western blot analyses were repeated at least twice to confirm the results. One positive serum and one negative serum were selected as positive and negative test controls for further immunoblotting experiment using membrane proteins extracted from different mouse tissues (liver, heart, spleen and kidney). These two representative samples were again used as test controls in immunoblotting experiment using brain membrane proteins extracted from different animal species (rat, pig, chicken, duck, quail, frog, fish) together with human brain membrane proteins.

2.7. Peptide mass fingerprinting Peptide mass fingerprinting was done using the MALDITOF-MS analysis. Briefly, the silver-stained gel spots were destained with freshly prepared destaining buffer (5 Al of 30 mM potassium ferricyanide and 5 Al of 100 mM sodium thiosulfate). The gel spots were washed properly and incubated overnight with trypsin (Promega). The trypsincleaved peptides released by sonication were desalted by ZipTip A-C18 (Millipore). The peptides together with 1 Al of matrix mix containing 5 mg/ml Alpha-Cyano-4-Hydroxycinnamic Acid (CHCA) and 5 mg/ml 2,5-dihydroxybenzoic acid (DHB) were loaded onto MALDI plate. The MS Table 2 ACR criteria presentation of lupus patients at diagnosis (n = 100) No

ACR criteria

No of pts

(%)

1 2 3 4 5 6 7 8 9 10 11

Antinuclear antibody Immunologic disorder Hematologic disorder Arthritis Renal disorder Malar rash Photosensitivity Oral ulcers Serositis Neurologic disorder Discoid rash

97 95 81 77 53 41 31 29 16 10 8

(97) (95) (81) (77) (53) (41) (31) (29) (16) (10) (8)

156

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

Table 3 Neuropsychiatric syndromes recorded in SLE cohort (n = 100) according to the 1999 ACR Ad hoc committee on neuropsychiatric lupus nomenclature

Table 4 BRAA in controls, lupus and non-lupus groups

Neuropsychiatric syndromes

Control group Lupus patients* Non-lupus patients

No of patients [n (%)]

Total Acute confusional state Acute inflammatory demyelinating polyradiculoneuropathy Anxiety disorder Aseptic meningitis Autonomic disorder Cerebrovascular disease Cognitive dysfunction Demyelinating syndrome Headache Mononeuropathy (single/mutiplex) Mood disorders Movement disorder (chorea) Myasthenia gravis Myelopathy Neuropathy, cranial Plexopathy Polyneuropathy Psychosis* Seizures and seizure disorders* Psychosis* and/or seizures*

27 (27) 0 – 0 –

Chi-square test and Fisher’s exact test were used to analyze data to calculate p value. A value of p < 0.05 was considered statistically significant.

3. Results 3.1. Demographic data of lupus and non-lupus patients

2.8. Calculation of molecular weights The known molecular weights of prestained protein ladder (Fermentus) were plotted against corresponding distance from upper edge of the polyacrylamide gel in semi-log (log – linear) graph. Then the approximate expected molecular 4

102 90 130

2.9. Statistics

analysis was done by the curved field reflectron instrument Axima CFR-plus (Kratos, Shimadzu). The MALDI mass spectra were matched with Swissport and/or NCBI protein databases for known protein identification.

3

– 10 –

weights of brain membrane protein were calculated using linear regression. Similar observations were made in 10 Western blot results. The mean theoretical molecular weights of the two neuronal autoantigens were calculated.

(*Psychosis and seizure were neurological disorder defined by 1997updated ACR criteria for classification of SLE.) Some patients presented with more than one neuropsychiatric syndrome.

2

Negative

*p < 0.0007.

0 – 2 (2) 0 – 2 (2) 6 (6) 0 – 12 (12) 0 – 0 – 0 – 0 – 0 – 0 – 0 – 0 – 4 (4) 6 (6) 10 (10)

1

n = 102 n = 100 n = 130

Positive

5

6

There are 102 healthy control individuals, 100 lupus patients and 130 non-lupus patients (Table 1). The lupus patients had the following ACR criteria: abnormal titers of antinuclear antibodies (97%), immunological disorder (95%), hematological disorder (81%), arthritis (77%) and renal disorder (53%) followed by the remaining ACR criteria (8– 41%) (Table 2). Neuropsychiatric involvement in SLE patients were found in 10 lupus patients as defined by 1997-updated ACR criteria for classification of SLE (psychosis/seizure). Four had psychoses and six had seizure disorders. The numbers with neuropsychiatric symptoms went up to 27% according to the 1999 ACR Ad Hoc Committee on 7

8

9

P

N

100 kD 73 kD 54 kD

35 kD

24 kD

29.5 kD 27.5 kD

16 kD

Fig. 1. Immunoblotting results of BRAAs in sera of lupus, non-lupus and healthy controls; 1 – 3Ynon-lupus patients; 4 – 7Ylupus patients; 8 – 9Ycontrols; PYpositive serum; NYnegative serum.

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

→

Br

Liv

Ht

Sp

→

→

negative serum

M

Kid

M

157

→

positive serum Br

Liv

Ht

Sp

Kid

73 kD

35 kD

24 kD Fig. 2. Immunoblotting results of BRAA against membrane protein extracted from different mouse tissues; MYprotein marker; BrYmembrane proteins of mouse brain; LivYmembrane proteins of mouse liver; HtYmembrane proteins of mouse heart; SpYmembrane proteins of mouse spleen; KidYmembrane proteins of mouse kidney; Positive serumYserum representing positive samples; Negative serumYserum representing negative samples.

Neuropsychiatric Lupus Nomenclature (19 neuropsychiatric syndromes) (Table 3). 3.2. Detection of BRAAs in sera of controls, lupus and non-lupus patients We found BRAAs present in 10 (10%) out of 100 unselected lupus patients and none (0%) of the 102 healthy individuals. To determine whether these antibodies are present in other rheumatic diseases, we tested 130 non-lupus rheumatic patients. Positive and negative controls were added in all immunoblotting experiments. The 130 non-lupus patients comprising 20 RA, 20 OA, 20 AS, 20 PsA and 50APS did not show presence of BRAAs (Fig. 1). Of the 50 primary APS patients, 20 of them had cerebral involvement, e.g., cerebral thrombosis or transient ischaemic attacks. Thus BRAAs reacting with mice neuronal membrane proteins were found only in lupus patients, not in the healthy individuals and non-lupus rheumatic patients ( p < 0.0007) (Table 4). 3.3. Tissue and species specificity of BRAA The immunoblotting experiment was carried out using membrane protein lysates from different mouse tissues

M-Br 35 kD

H-Br

R-Br

(heart, liver, spleen and kidney). It was noted that BRAAs react only with membrane protein lysates from mouse brain tissue (Fig. 2). The same experiment was carried out using neuronal membrane protein lysates of different animals. It is interesting to note that BRAAs target neuronal membrane protein lysates of mammalian origin and did not react with those of other species such as avian, fish and amphibian (Fig. 3). The experiment using human neuronal membrane proteins as antigen produced similar bands as that of mouse neuronal membrane proteins. Random samples (n = 5) comprising positive and negative samples showed concordance between human and mouse antigens. 3.4. Peptide mass fingerprinting of brain antigens reacting with BRAA We used 2-D gel electrophoresis to separate the BRAAs. The matched protein spots of silver stained SDS –PAGE were sent for peptide mass fingerprinting using MALDI-TOF-MS analysis (Fig. 4). The trypsincleaved peptides mass spectra were matched against public protein databases (NCBI and Swissport). It was noted that these membrane proteins did not match to any published protein sequences.

P-Br

A

24 kD

M-Br | R-Br | P-Br | H-Br | CH-Br | D-Br | Q-Br | Fg-Br | Fi-Br 35 kD

B

24 kD Fig. 3. Immunoblotting results of BRAA against brain membrane proteins extracted from different animal species; ((A) mammalian species brain lysates; (B) Mammalian, Avian, Amphibian and Fish brain lysates) M-BrYmouse brain membrane proteins; R-BrYrat brain membrane proteins; P-BrYpig brain membrane proteins; H-BrYhuman brain membrane proteins; CH-BrYchicken brain membrane proteins; D-BrYduck brain membrane proteins; Q-BrYquail brain membrane proteins; Fg-BrYFrog brain membrane proteins; Fi-BrYFish brain membrane proteins.

158

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

→ 35 kD

M-Br separated by iso-electric points

→ M-Br

B

24 kD

A

Fig. 4. Immunoblotting results of BRAA against mouse brain membrane proteins separated by 2-D gel electrophoresis. This experiment was done for a pair of 2-D gel strips with a guide lane using mouse brain membrane proteins (M-Br). One gel was processed to immunoblotting using BRAA positive serum. Another gel was visualised using silver plus stain. The protein spots of silver-stained similar gel matching to A and B were punched, and processed for trypsin-cleaved peptide mass finger printing using MALDI-TOF-MS.

3.5. Correlation of BRAAs with psychosis and/or seizures in lupus patients The BRAAs were found to be highly associated with psychoses and seizures. 6 (60%) out of 10 BRAA-positive lupus patients had psychoses and/or seizures while only 4 (2%) of 90 BRAA-negative lupus patients had psychoses and/or seizures ( p < 0.0001) (Table 5).

4. Discussion Three classifications for neurological and psychiatric disorders of SLE have been published; the first consists of 12 diagnostic criteria proposed by The Ad Hoc Neuropsychiatric Lupus Workshop Group (Singer and Denburg, 1990), the second consists of a list of 29 lupus related CNS descriptors ranked according to diagnostic weightage, (Rood et al., 1999) and the last is the nomenclature system comprising 19 lupus related CNS syndromes with detailed definitions (ACR Ad Hoc Committee, 1999. However all reports indicated that the classification, nomenclature and definitions are designed to facilitate clinical research and multicentre trials in order to generate unique diagnostic criteria for NP lupus. A single mechanism may not be adequate to explain the pathogenesis of different neuropsychiatric manifestations. It is fair to say that sub-sets of NP-lupus may have different pathogenesis that can lead to specific manifestations. We chose psychosis and seizure in our study because both of them are categorized under neurologic disorder in 1997updated ACR criteria for classification of SLE and secondly they are easily identified and diagnosed by rheumatologists compared to other neuropsychiatric manifestations, e.g., mild cognitive deficits.

There is evidence that supports ‘‘pathogenicity’’ conferred by autoantibodies in autoimmune diseases, e.g., myasthenia gravis and anti-phospholipid syndrome. In myasthenia gravis, specific acetylcholine receptor autoantibodies lead to altered physiological state of neuromuscular end-plate resulting in the reduction in impulse transmission (Engel, 1984). Similarly, antiphospholipid antibodies target the cardiolipin-h2 glycoprotein I complex that result in thrombosis or trophoblast dysfunction (Sebire et al., 2002). Epilepsy has also been reported to be associated with the antiphospholipid syndrome (Shoenfeld et al., 2004). Many autoantibodies have been associated with neuropsychiatric lupus. Among autoantibodies targetting normal cellular components in human and animal studies, autoantibodies associated with NP-lupus are brain cross-reactive lymphocytotoxic antibodies (Bluestein and Zvaifler, 1976), anti-ribosomal P protein antibodies (Bonfa et al., 1987), anticardiolipin antibodies (Harris and Pierangeli, 1997) and antigangliosides antibodies (Hirano et al., 1980). These antibodies are not specific to lupus. They were detected in other diseases and even in normal controls (Iverson, 1996; Martinez et al., 1992). Similar to lymphocytotoxic antibodies cross-reacting with neuronal tissue, anti-ribosomal P protein antibodies were reported to cross-react with plasma membrane of human tumor cells (neuroblastoma and hepatoma) and human fibroblasts (Koren et al., 1992). Shoenfeld et al. hypothesized that anticardiolipin antibodies of APS patients, after intrathecal passive transfer, bind to brain tissue of mouse, causing cognitive dysfunctions (Shoenfeld et al., 2003). Thus, it is possible that specific autoantibodies in SLE patients may have cross activity with neuronal tissues, resulting in neuropsychiatric manifestation. The strongest association at present is reported with antiribosomal P protein antibodies. These antibodies target the cytoplasmic ribosomes, which are reported to have molecular weights of 38, 19 and 17 kD (Koren et al., 1992). The BRAAs, we identified, are different from anti-ribosomal P proteins antibodies because they target antigens with different molecular weights (27.5 and 29.5 kD) and different tissues. BRAAs identified in our study were targeting against neuronal membrane proteins of mammalian species, not to avian, amphibian and fish. We therefore postulate that BRAA-targeted neuronal membrane proteins (27.5 and 29.5 kD) are conserved proteins across the mammalian species. Table 5 Correlation of BRAAs with psychoses/seizures in lupus patients

Lupus patients + BRAA (n = 10) Lupus patients BRAA (n = 90)

Psychoses/ seizures+

Psychoses/ seizures

6(60%)a 4(2%)b

4(40%) 86(98%)

p < 0.0001. a Lupus patients with BRAAs: 4 psychoses and 2 seizures. b Lupus patients without BRAA: 4 seizures.

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

Recently, two autoantibodies have been described as associated with NP lupus. Antibodies to triosephosphate isomerase (TPI) was reported to be associated with NP lupus (Watanabe et al., 2004). TPI is a highly conserved 29 kD glycolytic enzyme that catalyses the interconversion of dihydroxyacetone phosphate and d-glyceraldehyde 3-phosphate. Deficiency of TPI causes haemolytic anemia and neurological disorders (Schneider, 2000). TPI, though quite similar in molecular weight to one of the 2 novel BRAAs identified by us, has recently been found to be associated with osteoarthritis and advocated as a diagnostic marker for OA (Xiang et al., 2004). Our BRAAs are not present in OA patients. N-methyl-d-aspartate (NMDA) receptors are responsible for the majority of excitatory synaptic transmission in the central nervous system. Antibodies to NMDA receptor types NR2a or NR2b, have been reported as being associated with NP lupus, specifically to depressed mood, decreased short-time memory and learning (Omdal et al., 2005). These 2 receptor subunits differ from our BRAAtargeted proteins in molecular weights. They are about 180 kD in molecular weight (Stephenson, 2001). In this study, BRAAs targeted against 27.5 and 29.5 kD neuronal membrane proteins of mammalian origin are highly associated with psychosis and seizures. Six out of 10 BRAA-positive lupus patients had psychoses (manic features) or seizures. Four of the 10 patients with BRAAs did not have neuropsychiatric manifestations. There may be several reasons for this scenario. Arbuckle et al. (2003) presented a retrospective study on development of autoantibodies before clinical onset of SLE. They found that at least one SLE autoantibodies (anti-nuclear antibodies, anti dsDNA antibodies, anti-phospholipid antibodies, anti-Ro antibodies, anti-La antibodies and anti nuclear ribonucleoprotein antibodies) was present in 88 percentage of lupus patients before the clinical diagnosis of SLE. Hence these 4 patients with positive BRAAs need to be followed up prospectively to see whether they develop neuropsychiatric features later on. At the same time, the possibility exists of subsets of BRAAs being responsible for pathogenic damage in specific neuronal tissue. Similar examples can be seen in sub-population of dsDNA antibodies which are responsible of lupus nephritis (Rekvig et al., 2004). Four BRAA-negative patients presented with psychosis or seizures in our study. We have repeated the blots of these samples to confirm the results. There may be two possibilities; firstly the detection method did not detect all BRAAs. Vincent and Newsom-Davis (1985) reported that a modified protocol could detect more anti-acetylcholine receptor (AChR) antibodies, which were undetectable in the normal protocol. Secondly other autoantibodies may be the cause of these psychoses and seizures. As autoantibodies are plentiful in SLE, it is possible that some other pathogenic autoantibodies may be present in NP-lupus. According to peptide mass fingerprinting results, the 27.5 and 29.5 kD neuronal membrane proteins do not match to available protein sequences from NCBI, Swissport protein

159

databases and Human Protein Reference Database (http:// www.hprd.org). We are confident that these are novel proteins. Currently there are no single diagnostic test or investigation to establish NP lupus. The diagnosis of NP-lupus is difficult and depends on the exclusion of other causes of CNS manifestations such as metabolic, cardiovascular or even infectious causes. Thus the diagnosis of NP-lupus depends on a series of tests (series of hematological, biochemical, radiological and psychological tests). Serial measurements of acetylcholine receptor antibodies are useful in monitoring disease progression as well as the effect of treatment in Myasthenia gravis (Besinger et al., 1983; Somnier, 1993). Likewise, our novel BRAAs may play such roles. The autoantibodies against neuronal membrane protein can also provide a better understanding of the pathogenesis of NP-lupus when more specific animal experiments are done. We feel that these 2 novel BRAA-targeted neuronal membrane proteins can fill gaps in the clinical management of neuropsychiatric lupus. Studies incorporating the prospective monitoring of these autoantibodies in a defined cohort of lupus patients will allow us to better understand their diagnostic and prognostic values.

References ACR Ad Hoc Committee on Neuropsychiatric Lupus Nomenclature, 1999. The American College of Rheumatology nomenclature and case definitions for neuropsychiatric lupus syndromes. Arthritis Rheum. 42 (4), 599 – 608. Arbuckle, M.R., McClain, M.T., Rubertone, M.V., Scofield, R.H., Dennis, G.J., James, J.A., Harley, J.B., 2003. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N. Engl. J. Med. 349 (16), 1526 – 1533. Besinger, U.A., Toyka, K.V., Homberg, M., Heininger, K., Hohlfeld, R., Fateh-Moghadam, A., 1983. Myasthenia gravis: long-term correlation of binding and bungarotoxin blocking antibodies against acetylcholine receptors with changes in disease severity. Neurology 33 (10), 1316 – 1321. Bluestein, H.G., Zvaifler, N.J., 1976. Brain-reactive lymphocytotoxic antibodies in the serum of patients with systemic lupus erythematosus. J. Clin. Invest. 57 (2), 509 – 516. Bonfa, E., Golombek, S.J., Kaufman, L.D., Skelly, S., Weissbach, H., Brot, N., Elkon, K.B., 1987. Association between lupus psychosis and antiribosomal P protein antibodies. N. Engl. J. Med. 317 (5), 265 – 271. Cameron, J.S., 1999. Lupus nephritis. J. Am. Soc. Nephrol. 10 (2), 413 – 424. Engel, A.G., 1984. Myasthenia gravis and myasthenic syndromes. Ann. Neurol. 16 (5), 519 – 534. Futrell, N., Schultz, L.R., Millikan, C., 1992. Central nervous system disease in patients with systemic lupus erythematosus. Neurology 42 (9), 1649 – 1657. Hanly, J.G., Hong, C., 1993. Antibodies to brain integral membrane proteins in systemic lupus erythematosus. J. Immunol. Methods 161, 107 – 118. Hanson, V.G., Horowitz, M., Rosenbluth, D., Spiera, H., Puszkin, S.J., 1992. Systemic lupus erythematosus patients with central nervous system involvement show autoantibodies to a 50-kD neuronal membrane protein. J. Exp. Med. 176 (2), 565 – 573.

160

S.K. Tin et al. / Journal of Neuroimmunology 169 (2005) 153 – 160

Harris, E.N., Pierangeli, S., 1997. Antiphospholipid antibodies and cerebral lupus. Ann. N. Y. Acad. Sci. 823, 270 – 278. Hirano, T., Hashimoto, H., Shiokawa, Y., Iwamori, M., Nagai, Y., Kasai, M., Ochiai, Y., Okumura, K., 1980. Antiglycolipid autoantibody detected in the sera from systemic lupus erythematosus patients. J. Clin. Invest. 66 (6), 1437 – 1440. Hochberg, M.C., 1997. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus [letter]. Arthritis Rheum. 40, 1725. Hoffman, S.A., Arbogast, D.N., Ford, P.M., Shucard, D.W., Harbeck, R.J., 1987. Brain-reactive autoantibody levels in the sera of ageing autoimmune mice. Clin. Exp. Immunol. 70 (1), 74 – 83. Iverson, G.L., 1996. Are antibodies to ribosomal P proteins a clinically useful predictor of neuropsychiatric manifestations in patients with systemic lupus erythematosus? Lupus 5 (6), 634 – 635. Kaell, A.T., Shetty, M., Lee, B.C., Lockshin, M.D., 1986. The diversity of neurologic events in systemic lupus erythematosus. Prospective clinical and computed tomographic classification of 82 events in 71 patients. Arch. Neurol. 43 (3), 273 – 276. Khin, N.A., Hoffman, S.A., 1993. Brain reactive monoclonal autoantibodies: production and characterisation. J. Neuroimmunol. 44, 137 – 148. Koren, E., Reichlin, M.W., Koscec, M., Fugate, R.D., Reichlin, M., 1992. Autoantibodies to the ribosomal P proteins react with a plasma membrane-related target on human cells. J. Clin. Invest. 89 (4), 1236 – 1241. Martinez, X., Tintore, M., Montalban, J., Ordi, J., Vilardell, M., Codina, A., 1992. Antibodies against gangliosides in patients with SLE and neurological manifestations. Lupus (5), 299 – 302. Moore, P.M., 1992. Evidence for bound antineuronal antibodies in brains of NZB/W mice. J. Neuroimmunol. 38, 147 – 154. Omdal, R., Brokstad, K., Waterloo, K., Koldingsnes, W., Jonsson, R., Mellgren, S.I., 2005. Neuropsychiatric disturbances in SLE are associated with antibodies against NMDA receptors. Eur. J. Neurol. 12 (5), 392 – 398. Rekvig, O.P., Kalaaji, M., Nossent, H., 2004. Anti-DNA antibody subpopulations and lupus nephritis. Autoimmun. Rev. 3 (2), 1 – 6. Rood, M.J., Breedveld, F.C., Huizinga, T.W., 1999. The accuracy of diagnosing neuropsychiatric systemic lupus erythematosus in a series of 49 hospitalized patients. Clin. Exp. Rheumatol. 17 (1), 55 – 61. Schneider, A.S., 2000. Triosephosphate isomerase deficiency: historical perspectives and molecular aspects. Bailliere’s Best Pract. Res. Clin. Haematol. 13 (1), 119 – 140. Sebire, N.J., Fox, H., Backos, M., Rai, R., Paterson, C., Regan, L., 2002. Defective endovascular trophoblast invasion in primary antiphospholipid antibody syndrome-associated early pregnancy failure. Hum. Reprod. 17 (4), 1067 – 1071. Sherer, Y., Gorstein, A., Fritzler, M.J., Shoenfeld, Y., 2004. Autoantibody explosion in systemic lupus erythematosus: more than 100 different antibodies found in SLE patients. Semin. Arthritis Rheum. 34 (2), 501 – 537.

Shoenfeld, Y., Lev, S., Blatt, I., Blank, M., Font, J., von Landenberg, P., Lev, N., Zaech, J., Cervera, R., Piette, J.C., Bertolaccini, M.L., Hughes, G.R., Youinou, P., Meroni, P.L., Pengo, V., Alves, J.D., Tincani, A., Szegedi, G., Lakos, G., Sturfelt, G., Jonsen, A., Koike, T., Sanmarco, M., Ruffatti, A., Ulcova-Gallova, Z., Praprotnik, S., Rozman, B., Lorber, M., Chapman, J., van-Breda-Vriezman, P.J., Damoiseaux, J., 2004. Features associated with epilepsy in the antiphospholipid syndrome. J. Rheumatol. 31 (7), 1344 – 1348. Shoenfeld, Y., Nahum, A., Korczyn, A.D., Dano, M., Rabinowitz, R., Beilin, O., Pick, C.G., Leider-Trejo, L., Kalashnikova, L., Blank, M., Chapman, J., 2003. Neuronal-binding antibodies from patients with antiphospholipid syndrome induce cognitive deficits following intrathecal passive transfer. Lupus 2 (6), 436 – 442. Sibley, J.T., Olszynski, W.P., Decoteau, W.E., Sundaram, M.B., 1992. The incidence and prognosis of central nervous system disease in systemic lupus erythematosus. J. Rheumatol. 19 (1), 47 – 52. Singer, J., Denburg, J.A., 1990. Diagnostic criteria for neuropsychiatric systemic lupus erythematosus: the results of a consensus meeting. The Ad Hoc Neuropsychiatric Lupus Workshop Group. J. Rheumatol. 17 (10), 1397 – 1402. Somnier, F.E., 1993. Clinical implementation of anti-acetylcholine receptor antibodies. J. Neurol. Neurosurg. Psychiatry 56 (5), 496 – 504. Stephenson, F.A., 2001. Subunit characterization of NMDA receptors. Curr. Drug Targets 2 (3), 233 – 239. Toh, B., Mackay, I.R., 1981. Autoantibody to a novel neuronal antigen in systemic lupus erythematosus and in normal human sera. Clin. Exp. Immunol. 44 (3), 555 – 559. Tung, K.S., 1994. Mechanism of self-tolerance and events leading to autoimmune disease and autoantibody response. Clin. Immunol. Immunopathol. 73 (3), 275 – 282. Vincent, A., Newsom-Davis, J., 1985. Acetylcholine receptor antibody as a diagnostic test for myasthenia gravis: results in 153 validated cases and 2967 diagnostic assays. J. Neurol. Neurosurg. Psychiatry 48 (12), 1246 – 1252. Watanabe, H., Seino, T., Sato, Y., 2004. Antibodies to triosephosphate isomerase in patients with neuropsychiatric lupus. Biochem. Biophys. Res. Commun. 321 (4), 949 – 953. West, S.G., Emlen, W., Wener, M.H., Kotzin, B.L., 1995. Neuropsychiatric lupus erythematosus: a 10-year prospective study on the value of diagnostic tests. Am. J. Med. 99 (2), 153 – 163. Wilson, H.A., Winfield, J.B., Lahita, R.G., Koffler, D., 1979. Association of IgG anti-brain antibodies with central nervous system dysfunction in systemic lupus erythematosus. Arthritis Rheum. 22 (5), 458 – 462. Xiang, Y., Sekine, T., Nakamura, H., Imajoh-Ohmi, S., Fukuda, H., Nishioka, K., Kato, T., 2004. Proteomic surveillance of autoimmunity in osteoarthritis: identification of triosephosphate isomerase as an autoantigen in patients with osteoarthritis. Arthritis Rheum. 50 (5), 1511 – 1521. Zameer, A., Hoffman, S.A., 2001. Immunoglobulin binding to brain in autoimmune mice. J. Neuroimmunol. 120, 10 – 18.