INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Multiple sclerosis (MS) is the most common demyelinating disease of the central nervous system, affecting approximately 80,000 individuals in the United Kingdom and 400,000 in the United States. It is generally thought to be an autoimmune disease that is triggered by an infectious agent, possibly through a molecular mimicry mechanism of induction (1), similar to the situation occurring in the pathology of rheumatic fever (8).

A number of different microbial sequences, including Pseudomonas peptides, are capable of activating myelin basic protein (MBP)-specific T-cell clones from MS patients (19), thereby indicating that several microorganisms may carry mimicry epitopes of myelin. This could explain the difficulty of linking the immunopathogenesis of MS to a single microbial agent, suggesting that several such bacteria could be involved in initiating the autoimmune process (6).

It was decided to investigate MS patients for possible immune responses to pathogens that carry molecular mimicry sequences resembling brain components, such as Acinetobacter species and Pseudomonas aeruginosa. We have identified an amino acid sequence similarity between the encephalitogenic peptide of bovine MBP (5) and a sequence present in 4-carboxymuconolactone decarboxylase of Acinetobacter calcoaceticus (3). A similar sequence was also found in -carboxymuconolactone decarboxylase of P. aeruginosa (Swissprot).

The aims of this study were to determine whether MS patients have elevated levels of antibodies to five different strains of Acinetobacter as well as to P. aeruginosa, Escherichia coli, MBP, and neurofilaments, compared to cerebrovascular-accident (CVA) patients and control healthy blood donors. A further study was carried out, incorporating an inflammatory non-MS disease control, viral encephalitis, and a new set of control subjects, to assess whether MS patients have elevated levels of immunoglobulin G (IgG) antibodies to A. calcoaceticus 16904.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Serum samples. Sera from 26 MS patients (nine males and 17 females; mean age, 42 years; range, 29 to 55 years) were obtained from the Institute of Neurology at the Hospital for Nervous Diseases, London, United Kingdom. Diagnosis was made according to the Poser criteria (13). Benign MS patients are designated by infrequent exacerbation with full recovery. Relapsing remitting MS patients are those patients who have more frequent exacerbations followed by partial or complete remission. Serum samples were obtained from relapsing remitting patients during remission and exacerbation. Secondary progressive MS patients are defined as patients who continue to deteriorate without remission following an initial relapsing remitting course of disease. Primary progressive MS is defined as continuous deterioration without remission from the onset of disease. The clinical features of the MS patients are summarized in Table 1. In addition, serum samples were obtained from 20 patients in the Department of Geriatric Medicine at University College Hospital, who had suffered from a unilateral hemiplegia due to a CVA (10 males and 10 females; mean age, 80.5 years; range, 69 to 94 years) and 10 patients with viral encephalitis (eight males and two females; mean age, 38 years; range, 3 to 66 years) attending the National Hospital for Neurology and Neurosurgery. Sera from 25 subjects attending the London Blood Donor services were used as healthy controls (12 males and 13 females; mean age, 40.6 years; range, 22 to 67 years). A further set of sera from 29 healthy control subjects attending the London Blood Donor services was used in the viral encephalitis study (15 males and 14 females; mean age, 43 years; range, 19 to 66 years).

Preparation of bacteria. Cultures of Acinetobacter sp. strain 11171, Acinetobacter sp. strain 19004, Acinetobacter junii 17908, Acinetobacter lwoffii 5866 and Acinetobacter radioresistens (sp12) were provided by the Public Health Laboratory, Nottingham, United Kingdom. A. calcoaceticus (NCIMB 16904) was obtained from the National Collections of Industrial and Murine Bacteria Ltd. (Aberdeen, Scotland). The Department of Microbiology at King's College London provided P. aeruginosa (NCTC 8203) and E. coli (NCTC 9002).

ELISA. ELISAs were carried out as previously described (9). Briefly, aliquots of 200 µl of bacterial suspension or bovine MBP or neurofilaments (25 µg/ml; Sigma), diluted in a 0.05 M carbonate buffer (pH 9.6), were adsorbed onto a 96-well flat-bottom polystyrene microELISA plate (Dynatech) overnight at 4°C. Plates were washed three times for 5 min in PBS containing 0.05% (vol/vol) Tween 20 (Sigma) (washing and incubation buffer) and were blocked with PBS containing 0.1% bovine serum albumin (Sigma) for 1 h at 37°C. The washing procedure was repeated, and 200 µl of test or control serum, diluted 1 in 200 in incubation buffer, was added to the wells in duplicate and incubated for 1 h at 37°C. Plates were washed three times in washing buffer, and 200 µl of IgM, IgG, or IgA rabbit anti-human conjugate with horseradish peroxidase (Dako), diluted 1 in 500 in incubation buffer, was added and incubated for 1 h at 37°C. The washing procedure was repeated, and 200 µl of substrate solution2,2'-azinobis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) (0.5 mg/ml; Sigma) in citrate phosphate buffer, pH 4.1, containing 0.98 mM H₂O₂ (Sigma)was added to each well. The plates were developed in the dark, at room temperature, for 25 min. The reaction was stopped by addition of 100 µl of sodium fluoride (2 mg/ml; Sigma) solution. Absorbances were measured on a microtiter plate reader (Dynatech MR606) at 630 nm. All studies were carried out under code, in that the tester did not know which sera were test or control.

Statistical analysis. The mean OD units of control groups (CVA and healthy blood donors) were compared with the mean OD of the 26 MS patients, using a one-tail Student's t test, and 95% confidence limits of control groups were calculated. Pearson's correlation coefficient (r) was also calculated using the statistical package Prism3.0 (GraphPad Software).

Western blot analysis. Electrophoresis was performed according to the method of Laemmli (10). Briefly, prepared samples (10 µl) and a set of molecular weight markers (5 µl) (RPN800; Amersham) were loaded and run on a 10% SDS-polyacrylamide gel at 100 V for 2 h in electrolyte buffer (0.025 M Tris, 0.2 M glycine, and 0.1% SDS, pH 8.3) and the gel was transferred to nitrocellulose membrane at 30 V for 16 h in transfer buffer (20 mM Tris, 0.15 M glycine, pH 8.3, containing 20% [vol/vol] methanol) (Protean III cell; Bio-Rad Laboratories). All incubation steps were carried out at room temperature with gentle shaking. After transfer the membrane was cut into individual strips and incubated in 2% Marvel-Tris-buffered saline (TBS) for 2 h. The strips were washed five times for 5 min in TBS-Tween and then individually incubated with patient or control serum for 2 h. Serum was diluted at 1 in 200 in 2% Marvel-TBS. The strips were then washed as before and incubated for 2 h with rabbit anti-human immunoglobulin (IgG, IgM, and IgA) conjugate with horseradish peroxidase (Dako), diluted 1 in 1,000 in 2% Marvel-TBS. Strips were washed and developed using a chemiluminescent detection assay (Amersham).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

IgA anti-Acinetobacter antibodies. Levels of IgA antibodies to Acinetobacter sp. strain 11171 (P < 0.0001), Acinetobacter sp. strain 19004 (P < 0.0001), A. junii 17908 (P < 0.01), A. lwoffii 5866 (P < 0.0001), and A. radioresistens (P < 0.0001) in MS patients were significantly higher than those in the healthy control group.

View larger version (26K): [in this window] [in a new window]   FIG. 1.   Levels of IgA, IgM, and IgG antibodies (mean + standard error [error bars]) to Acinetobacter sp. strain 11171 (a), Acinetobacter sp. strain 19004 (b), A. junii 17908 (c), A. lwoffii 5866 (d), A. radioresistens (e), P. aeruginosa (f), and E. coli (g), in sera from 26 MS patients, 20 CVA patients, and 25 healthy blood donors. Symbols: , controls; , CVA; , MS.

IgG anti-Acinetobacter antibodies. Levels of IgG antibodies to Acinetobacter sp. strain 11171 (P < 0.0001), Acinetobacter sp. strain 19004 (P < 0.0001), A. junii 17908 (P < 0.0001), A. lwoffii 5866 (P < 0.0001), and A. radioresistens (P < 0.0001) were significantly elevated in MS patients compared to healthy controls.

View larger version (16K): [in this window] [in a new window]   FIG. 2.   IgG antibody levels (mean + standard error [error bars]) to A. calcoaceticus 16904 and E. coli in sera from 26 MS patients, 10 viral encephalitis patients, and 29 healthy blood donors.

IgM anti-Acinetobacter antibodies. Levels of IgM antibodies to Acinetobacter sp. strain 11171 (P < 0.0001), Acinetobacter sp. strain 19004 (P < 0.0001), A. junii 17908 (P < 0.0001), A. lwoffii 5866 (P < 0.0001), and A. radioresistens (P < 0.0001) were significantly higher in MS patients than in the healthy control group.

Antibodies to P. aeruginosa Levels of IgA antibodies to Pseudomonas were shown to be significantly higher in MS patients than in the control group (P < 0.001) and CVA patients (P < 0.05). There was also a significant elevation in the level of IgA anti-Pseudomonas antibodies in the CVA group (P < 0.05) compared to controls.

Antibodies to E. coli. No significant differences in the levels of IgA, IgM, or IgG anti-E. coli antibodies were observed between any of the groups tested (Fig. 1g and 2).

Antibodies to brain components. (i) MBP. Significantly elevated levels of IgA to MBP were observed in MS sera compared to controls (P < 0.0001) and CVA sera (P < 0.001). However, no significant difference was observed between the control and CVA group, when measuring IgA antibody levels.

View larger version (54K): [in this window] [in a new window]   FIG. 3.   Levels of IgA, IgM, and IgG antibody (mean + standard error [error bars]) to Acinetobacter sp. strain 11171 (a), A. radioresistens (b), MBP (c), and neurofilaments (d) in sera from 26 MS patients, 20 CVA patients, and 25 healthy blood donors.

(ii) Neurofilaments. Levels of IgA antibody to neurofilaments were shown to be significantly higher in MS patients than in controls (P < 0.01) and CVA patients (P < 0.0001). No significant difference between the control and CVA groups was seen in the level of antineurofilament IgA antibodies.

Measurement of serial dilutions. Levels of total immunoglobulin (IgG, IgM, and IgA) to A. radioresistens (Fig. 4a), Acinetobacter sp. strain 11171 (Fig. 4b), and P. aeruginosa (Fig. 4c) are shown for a high- and low-titer MS serum and a control sample (M39).

View larger version (20K): [in this window] [in a new window]   FIG. 4.   Serial doubling dilutions (mean ± standard error [error bars]) of high- and low-titer MS sera and a control sample against A. radioresistens (a), Acinetobacter sp. strain 11171 (b), and P. aeruginosa (c).

Western blot analysis. The specificity of the serum antibodies was evaluated by Western blotting using A. radioresistens, Acinetobacter sp. strain 11171, and P. aeruginosa antigens. The high-titer MS serum (MS080) showed prominent reactivity to a 15- to 28.6-kDa region of both A. radioresistens and Acinetobacter sp. strain 11171, and a further band of reactivity can be seen with a 9.1-kDa protein of Acinetobacter sp. strain 11171. The control serum (M39) showed much weaker reactivity to proteins of Acinetobacter sp. Reactivity to a number of proteins of P. aeruginosa can be seen in both the MS and control patients. The MS patient demonstrated prominent reactivity to a 78-kDa protein and an 8.5-kDa protein of P. aeruginosa (Fig. 5).

View larger version (131K): [in this window] [in a new window]   FIG. 5.   Immunoblotting antigenic bands developed with serum samples from an MS patient (MS080) (MS) and a control healthy blood donor (M39) (Con) against A. radioresistens (lanes A and B), Acinetobacter sp. strain 11171 (lanes C and D), and P. aeruginosa (lanes E and F).

Correlation coefficient analysis. The correlation coefficient (r) was calculated between all strains of Acinetobacter tested and P. aeruginosa. There was a significant positive correlation between the IgM levels of A. junii 17908 and P. aeruginosa (r = +0.831; P < 0.0001). A significant positive correlation was also seen between the IgM levels of Acinetobacter sp. strain 19004 and P. aeruginosa (r = +0.819; P < 0.0001). This was also the case for levels of IgA antibodies between these two organisms (r = +0.407; P < 0.01). Both Acinetobacter sp. strain 19004 and A. junii 17908 also showed significant positive correlations with both MBP and neurofilaments, with regards to IgM and IgG levels (P < 0.0001). A significant positive correlation was seen between the IgM levels of P. aeruginosa and both MBP (r = +0.851; P < 0.0001) and neurofilaments (r = +0.693; P < 0.0001). A significant positive correlation was also seen between the IgA (r = +0.399; P < 0.05), IgM (r = +0.663; P < 0.001) and IgG (r = +0.772; P < 0.0001) levels of MBP and neurofilaments.

MAN index. The myelin-Acinetobacter-neurofilament (MAN) index was calculated using the results from the IgG antibody assays. The formula used was as follows: (MBP OD × 10) (Acinetobacter strain OD/Pseudomonas OD × 10) (neurofilament OD × 10). All MS patients had values above the 99.9% confidence limit of the controls when the MAN index was calculated using A. lwoffii 5866, Acinetobacter sp. strain 11171, A. radioresistens, Acinetobacter sp. strain 19004, and A. junii 17908. Only 88.5% of MS patients were shown to have a MAN index above the 99.9% confidence limit, when calculating with P. aeruginosa antibodies. In all cases only one control was shown to lie above the 99.9% confidence limit. Seven CVA patients were shown to have values above the 99.9% confidence limit when the MAN index was calculated using Acinetobacter sp. strain 11171. For all other strains of Acinetobacter and P. aeruginosa more than 12 CVA patients had MAN indices above the 99.9% confidence limit. No significant differences were seen in the MAN indices between the different MS groups. Interestingly, when calculating the MAN index using A. radioresistens all MS patients could be distinguished from the two control groups. When removing the Acinetobacter parameter, the MAN index could distinguish MS patients from healthy controls but not from the CVA group (Fig. 6). When the MAN index was calculated with IgA and IgM antibodies, the differences between the MS patients and control groups were not as significant (data not shown).

View larger version (16K): [in this window] [in a new window]   FIG. 6.   MAN index, calculated as follows: (MBP OD × 10) (Acinetobacter OD/Pseudomonas OD × 10) (neurofilament OD × 10) expressed in log10, in sera from 26 MS patients (B, benign; RR, relapsing remitting; ARR, acute relapsing remitting; 2P, secondary progressive; PP, primary progressive; Trans, transitional), 20 CVA patients, and 25 healthy blood donors. The dashed line represents the 99.9% confidence limit of controls.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Elevated levels of IgA, IgM, and IgG directed against Acinetobacter sp. strain 11171, A. lwoffii 5866, and A. radioresistens were found in the sera from MS patients compared to CVA patients or healthy controls. A similar observation was seen for Acinetobacter sp. strain 19004 and Acinetobacter sp. strain 17908, although there were no significant differences in the levels of IgA antibodies between the MS and CVA groups. A significant increase in levels of IgG to A. calcoaceticus 10694 was also shown in MS patients compared to healthy controls and viral encephalitis patients. Primary and secondary progressive MS patients tended to have higher levels of IgA antibodies to Acinetobacter sp. strain 11171 compared to other MS groups.

Molecular mimicry appears to be involved in MS (1, 7), but efforts to identify an etiological agent have not been successful. Seven viral and one bacterial peptides have been found to activate three of seven T-cell clones specific to MBP (amino acids 85 to 99) isolated from MS patients. The bacterial peptide identified was from phosphomannomutase protein in P. aeruginosa (19). In this study we have looked at potential respiratory pathogens, including Acinetobacter and P. aeruginosa, which carry sequences that mimic brain components. An amino acid sequence homology between a known encephalitogenic myelin peptide (5) and the enzyme 4-carboxymuconolactone decarboxylase has been found in both Acinetobacter spp. and P. aeruginosa (3).

Autoantibodies to neuronal components, including MBP (17), myelin oligodendrocyte protein (14), myelin-associated glycoprotein (16), proteolipid protein (18), and neurofilaments (11) have previously been reported in MS. IgG antimyelin antibodies, but to a lesser extent IgM and IgA, have also previously been demonstrated in 88% of sera from healthy controls by indirect immunofluorescence assays (4). Increased levels of IgA, IgM, and IgG to both bovine MBP and neurofilaments acting as heterospecific antigens were observed in MS patients compared to controls and the CVA group. There is 92% homology between bovine and human MBP, but almost complete amino acid identity over the most immunogenic epitopes (5). If the presence of anti-MBP and antineurofilament autoantibodies occurs as a result of previous brain damage, antibodies to these brain components should also be found in CVA patients as well as MS patients. However, the results of this study indicate that the MS sera but not the CVA sera had significantly elevated levels of these autoantibodies compared to controls. The pathology of MS is usually thought of as being confined to the white matter, but the observation that autoantibodies to neurofilaments were seen in MS patients and not controls indicates that gray matter of the brain may also be involved in the disease. This confirms previous observations indicating the presence of gray matter pathology in MS patients using magnetic resonance imaging (12) and studies demonstrating the presence of antibodies to the 69-kDa neurofilament protein (11).

We have used our results from the ELISAs measuring IgG antibodies against MBP, Acinetobacter, P. aeruginosa, and neurofilaments to obtain a MAN antibody index. This parameter made it possible for us to identify all the MS patients, and therefore it could offer the possibility of a diagnostic test, but prospective studies will be required to establish the validity of the MAN index.

The agent responsible for autoantibodies in MS is still unclear, but this study does show that MS patients have probably been exposed to Acinetobacter spp. and P. aeruginosa. It is likely that infection with these microorganisms, which carry epitopes resembling brain antigens, could lead to activation of auto-reactive cells and production of autoantibodies, which may cause tissue damage and thus lead to the development of a neurological autoimmune disease, such as MS. However, it must be considered that in the ELISA we are detecting anti-MBP antibodies that are cross-reacting with Acinetobacter spp. Longitudinal studies are necessary to determine whether exposure to Acinetobacter antigens precedes or follows pathological changes. Further work is required to evaluate the role, if any, of Acinetobacter spp. and P. aeruginosa in MS.