Acinetobacter Immune Responses in Multiple Sclerosis Etiopathogenetic Role and Its Possible Use As a Diagnostic Marker

NEUROLOGICAL REVIEW

SECTION EDITOR: DAVID E. PLEASURE, MD Acinetobacter Immune Responses in Multiple Sclerosis Etiopathogenetic Role and Its Possible Use as a Diagnostic Marker

Alan Ebringer, MD; Lucy Hughes, PhD; Taha Rashid, MBChB; Clyde Wilson, PhD

ultiple sclerosis (MS) is the most common cause of neurologic disability among young people. The etiology of MS is controversial, but immune responses are considered to somehow be involved. The diagnosis of MS depends on a combination of various clinical and laboratory features, but apart from some myelin-neuronal autoantibody profilesM or oligoclonal bands in the cerebrospinal fluid no other serologic diagnostic test or marker has yet been discovered. However, the presence of antibodies to Acinetobacter species in MS patients opens the possibility of developing a composite laboratory diagnostic marker, the myelin-Acinetobacter- neurofilament index. Whether Acinetobacter is the triggering agent of MS remains to be determined, but the measurement of anti-Acinetobacter antibodies could be used as a marker of disease activity. To evaluate this, prospective randomized controlled studies should be performed with MS patients, especially in the early stages of the disease. Arch Neurol. 2005;62:33-36

Multiple sclerosis (MS) is a common de- screening studies performed in outbred myelinating disease of the central nervous populations of Northern European origin system, usually affecting young people and was successful in identifying genetic mark- characterized pathologically by scattered ers that show statistically significant link- areas of inflammation with or without axo- age with the disease. It has been suggested nal damage, involving particularly the white that many different chromosomal regions matter. Multiple sclerosis has a world- are likely to be involved in MS.4 wide distribution, and as many as 1 mil- lion people are affected by the disease. The prevalence of MS shows a latitudinal effect Evidence of Environmental Factors in that it is higher in Scandinavian coun- 1 tries than in the tropics. The reverse is ob- Various studies have supported the role of served in the Southern hemisphere, where environmental or microbial factors in the MS is 7 times more common in Tasmania pathogenesis of MS: (1) a relatively low and Southern New Zealand than in tropi- 2 concordance rate for monozygotic twins cal Queensland in populations coming pre- (approximately 25%), dizygotic twins (ap- dominantly from Anglo-Celtic stock. This proximately 5%), and siblings (approxi- latitudinal effect could be linked to greater mately 3%) of patients with MS5; (2) re- prevalence of respiratory infections dur- ports of an outbreak of MS in the Faroe ing autumn and winter. Islands, which occurred after the start of World War II6; (3) increase in the prob- ETIOPATHOGENESIS ability of developing MS with migration rate from low- to high-risk areas, espe- Evidence of Genetic Factors cially among those in younger age groups7;

3 and (4) the occurrence of remissions and In a recent review, none of the 3 large exacerbations in MS as a common hall- mark of the disease and a possible result Author Affiliations: Division of Life Sciences, Infection and Immunity Group, of fluctuations in the exposure to certain Waterloo Campus, King’s College, London, England. triggering environmental factors.

(REPRINTED) ARCH NEUROL / VOL 62, JAN 2005 WWW.ARCHNEUROL.COM 33

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 Microorganisms Possibly Involved in the in patients with Alzheimer disease and dementia, when com- Etiopathogenesis of MS pared with controls.

Various microbial agents have been implicated in the etio- UPPER RESPIRATORY TRACT AND PARANASAL pathogenesis of MS. The role of viruses8 has been exten- SINUSES AS THE SOURCE OF INFECTIONS IN MS sively studied, but no clear consensus has emerged from these investigations. Viruses such as Epstein-Barr virus The most likely source for the entry of any triggering mi- and human herpesvirus 6, which have been implicated crobial factor in MS is through the upper respiratory tract in MS, are ubiquitous in the environment, but others have (URT). Several lines of evidence support this possibility. failed to confirm a link between human herpesvirus 6 and First, sinusitis is present in many MS patients, and the rate MS.9 Some researchers have found a link between the of MS exacerbations during the sinusitis attacks was found gram-negative Chlamydia pneumoniae bacteria and MS,10 to be doubled.17 Furthermore, using magnetic resonance but others have failed to confirm these associations.11 The imaging of the nasal sinuses, 53% of MS patients had evi- role of other bacteria, such as Acinetobacter species and dence of sinusitis.18 Second, in one study,19 the main caus- Pseudomonas aeruginosa, has only been studied recently ative agents involved in sinusitis were found to be Aci- but with some encouraging results. netobacter, Pseudomonas, and Staphylococcus aureus.Ina subsequent study involving antral tap and endoscopi- Acinetobacter and MS cally directed tissue culture performed on acute sinusitis patients, 37% and 33% of the isolates from the antral tap Myelin sequences known to produce experimental aller- and the endoscopically directed tissue culture, respec- gic encephalomyelitis in guinea pigs, an animal model tively, were Acinetobacter bacteria.20 Third, in another ret- of MS,12 were examined using the GenBank (NCBI, rospective study,21 more than 50% of MS patients had a Bethesda, Md) and SwissProt (TrEMBL, Geneva, Swit- history of repeated respiratory tract infections during their zerland) databases for molecular mimicry between brain childhood. Fourth, clinically manifest infections predomi- tissues and microbes. The ubiquitous saprophytic mi- nantly of the URT were observed to be followed by more crobe, Acinetobacter, was found to possess such a se- attacks of exacerbations in patients with MS.22 quence.13 These results suggest that MS could be triggered and Antibody levels against 5 strains of Acinetobacter sp, perpetuated following repeated attacks of subclinical or P aeruginosa, and Escherichia coli were investigated in 26 overt infections in the URT or paranasal sinuses. Further- English patients with MS and compared with 20 pa- more, viral infections of the URT and paranasal sinuses, tients with cerebrovascular accidents and 25 healthy con- which occur more frequently in autumn and winter, could trol. There were significant elevations in the levels of an- provide a suitable biological milieu for the secondary tibodies of IgM, IgG, and IgA classes against all bacterial growth of a saprophytic microbe such as Acinetobacter. agents except E coli in patients with MS when compared with those with cerebrovascular accidents and healthy MOLECULAR MIMICRY AND MS controls.14 The elevations in these antibodies were found to be more prominent against Acinetobacter calcoaceti- Molecular mimicry or cross-reactivity has been pro- cus, Acinetobacter 11171, and Acinetobacter lwoffii strains, posed as the main pathogenetic mechanism for develop- and in some cases their levels were reaching titers of up ment of autoimmune diseases, such as rheumatic fever, to 1:6400. Sydenham chorea, rheumatoid arthritis, and ankylosing In a more recent study15 performed on serum samples spondylitis.23 The molecular mimicry hypothesis is based taken from the same group of patients and controls in- on the demonstration of autoimmunity, involving mo- cluded in the previous study, antibody levels against mim- lecular and/or immunologic cross-reactivity between the icking peptide from Acinetobacter (PϽ.001), Pseudomo- putatively causative microbes and autoantigens, in- nas (PϽ.001), myelin basic protein (MBP) (PϽ.001), and creased antibodies against these microbial and autoan- myelin oligodendrocyte glycoprotein (MOG) (PϽ.001) tigenic molecules, and the binding of these microbial were found to be elevated in patients with MS when com- cross-reactive antibodies to the targeted tissues with re- pared with those with cerebrovascular accidents or healthy sultant immune-mediated cytotoxic tissue damage.24 controls. Antiserum raised in mice against Acinetobacter- mimicking peptides were found to be significantly in- Evidence of Autoimmunity in MS hibited by peptides from the MBP or Pseudomonas microbe. Furthermore, MOG peptides were found to inhibit Autoantibodies to many myelin-neuronal antigens have antibodies against the mimicking sequences present in been found in patients with MS. For example, the levels Acinetobacter, but no inhibition was observed when hu- of antibodies to MBP,25 MOG,26 and neurofilament14,27 were man papillomavirus peptides were used as controls. found to be elevated in patients with MS when com- In another study,16 serum samples from Austrian pa- pared with control groups. Some of these antibodies were tients with MS or other neurologic diseases and healthy con- even found to be predictive of the development of clini- trols were screened against 3 strains of Acinetobacter spp. cally definite MS after a first demyelinating event.28 Fur- There were significant elevations of total antibodies against thermore, levels of anti-MBP or anti-MOG antibodies were A calcoaceticus, A lwoffii, and A 11171 bacterial strains in reported to be elevated in the cerebrospinal fluid and/or patients with MS and those with some other neurologic dis- isolated from the plaque lesions in the central nervous eases, such as sporadic Creutzfeldt-Jakob disease, but not system of MS patients.29 In studies30 that involved ani-

(REPRINTED) ARCH NEUROL / VOL 62, JAN 2005 WWW.ARCHNEUROL.COM 34

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 Overt or Subclinical Infections by Acinetobacter and Pseudomonas Acinetobacter Pseudomonas Bacterial Species in the Paranasal Sinuses or Upper Respiratory Tract With or Without Prior Viral Infection 3-OACT-A 4-CMLD γ-CMLD

90 Leu-Tyr-Arg-Ala- 42 Ser-Arg-Phe-Ala- 42 Thr-Arg-His-Ala- Formation of Antibodies to the 4-Carboxy-Muconolactorne Decarboxylase and Gly-Lys 95 Try-Gly 47 Try-Gly 47 3-Oxoadipate Coenzyme A Transferase Enzymes of Acinetobacter and/or γ-Carboxy-Muconolactone Decarboxylase Enzyme of Pseudomonas Bacteria 50 Leu-Tyr-Arg-Asp- 114 Ser-Arg-Phe-Ser- 114 Ser-Arg-Phe-Ser- in the Local Lymph Nodes Gly-Lys 55 Try-Gly 119 Try-Gly 119

MOG MBP MBP Crossing of the Blood-brain Barrier by These Cross-reactive Antibodies, Particularly of IgG Class Central Nervous System White Matter

Binding of These Antibodies to MBP, MOG, and Neurofilament and Figure 1. Schematic molecular similarities between Acinetobacter and Possibly Other as Yet Unidentified Cross-reactive Neuronal Antigens Pseudomonas bacterial enzymes and brain antigens. 4-CMLD indicates in the White Matter of Brain, Spinal Cord, or Optic Nerves 4-carboxy-muconolactone decarboxylase; 3-OACT,3-oxoadipate coenzyme A transferase; and ␥-CMLD, ␥-carboxy-muconolactone decarboxylase. Amino acids and their locations are also shown: Leu indicates leucine; Tyr, tyrosine; Activation of the Complement and Other Inflammatory Cascades With Arg, arginine; Ala, alanine; Gly, glycine; Lys, lysine; Asp, asparagine; Ser, Local Productions of Cytotoxic Enzymes and Cytokines and Recruitment serine; Phe, phenylalanine; Try, tryptophan; Thr, threonine; and His, of Other Immune Cells in Those Individuals With High Titers of Antibodies histidine.

mal models of MS, such as experimental allergic encepha- Multiple Sclerosis lomyelitis, MS-like lesions were detected and characterized by local infiltrations of immune-mediated cells and products, when these animals were inoculated with the Figure 2. Proposed sequential pathogenetic events in the development of myelin-neuronal antigens. Patients with MS have report- multiple sclerosis. MBP indicates myelin basic protein; MOG, myelin edly responded to the currently used immunomodula- oligodendrocyte glycoprotein. tory and anti-inflammatory therapeutic agents, especially interferon beta 1 products31 and corticosteroids. toxicity in the same way that has been shown in patients with Sydenham (rheumatic) chorea.33 These events Evidence of Molecular Similarities Between could eventually result in multiple sites of demyelin- Acinetobacter and Pseudomonas Bacteria and Brain ations with or without axonal degenerations. Antigens PROPOSAL FOR A NEW LABORATORY Acinetobacter and Pseudomonas enzymes show molecu- DIAGNOSTIC MARKER IN MS lar similarities to certain brain antigens (Figure 1). The enzyme 4-carboxy-muconolactone decarboxylase pos- Apart from the criteria for the diagnosis of MS by Poser et sesses an amino acid sequence that is similar to MBP,13 al,34 other diagnostic criteria, which have been recom- and protocatechuate 3,4-dioxygenase has a sequence ho- mended by neurologists from the United States and Eu- mologous to neurofilament.32 The 3-oxoadipate coen- rope, are all based on a combination of clinical and para- zyme A transferase enzyme of the same bacteria pos- clinical features.35 The main clinical evidence is based on sesses amino acid sequences that are similar to those the objective findings of dissemination in time (relapses) present in the MOG brain tissue antigens.15 In this latter and space (different locality) of the clinical presentations study, it was also shown that the enzyme ␥-carboxy- typical of MS, whereas the nonclinical criteria are mainly muconolactone decarboxylase of the Pseudomonas mi- based on the following: (1) magnetic resonance imaging crobes possesses amino acid residues that are homolo- evidence of multiple neurologic lesions of different sizes gous to those present in the MBP molecules. and locations; (2) identification of IgG oligoclonal bands in the cerebrospinal fluid; and (3) delayed but well- Possible Pathogenetic Pathway in MS Involving preserved visual evoked potentials. Acinetobacter Species These paraclinical criteria are helpful in aiding the diagnosis of clinically probable cases of MS, but they re- It is proposed that antibodies against Acinetobacter and quire expensive facilities. Furthermore, these diagnos- possibly Pseudomonas microorganisms are produced as tic criteria cannot identify all cases of MS because of the the result of URT or paranasal sinus infections by these clinical variability of the disease. Hence, the search for microbes (Figure 2). In view of the existing molecular the development of a less invasive and reproducible test similarities between microbes and brain antigens, these is of crucial importance. cross-reactive antibodies, which are mostly of the IgG iso- We propose that a combination of elevated titers of type, could cross the blood-brain barrier. The binding of antibodies against Acinetobacter, MBP, and neurofila- these antibodies to the myelin-neuronal antigens, such ment could be used for such a diagnostic test. The myelin- as MBP, MOG, and neurofilaments, when present in high Acinetobacter-neurofilament (MAN) index, which has been titers would activate complement and other inflamma- described previously,14 was calculated using the optical tory cascades, thereby producing demyelination through density (OD) readings from the IgG antibody deter- the process of antibody-dependent, cell-mediated cyto- mined by enzyme-linked immunosorbent assay. The for-

(REPRINTED) ARCH NEUROL / VOL 62, JAN 2005 WWW.ARCHNEUROL.COM 35

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 mula used was as follows: MAN Index=(MBP OD ϫ 10) 6. Kurtzke JF, Hyllested K. Multiple sclerosis in the Faroe Islands, III: an alternative ϫ ϫ assessment of the three epidemics. Acta Neurol Scand. 1987;76:317-339. (Acinetobacter OD 10) (Neurofilament OD 10). 7. Dean G, Elian M. Age at immigration to England of Asian and Caribbean immi- The MAN index is calibrated by determining scores grants and the risk of developing multiple sclerosis. J Neurol Neurosurg Psychiatry. 1997;63:565-568. in MS patients and healthy controls. For an active MS 8. Gilden DH. Viruses and multiple sclerosis. JAMA. 2001;286:3127-3129. patient, the MAN index is expected to be above the 95% 9. Enbom M, Wang FZ, Fredrikson S, Martin C, Dahl H, Linde A. Similar humoral confidence intervals of the healthy controls. Further re- and cellular immunological reactivities to human herpesvirus 6 in patients with multiple sclerosis and controls. Clin Diagn Lab Immunol. 1999;6:545-549. search is needed to elaborate and modify this particular 10. Layh-Schmitt G, Bendl C, Hildt U, et al. Evidence for infection with Chlamydia diagnostic marker. For example, instead of using whole pneumoniae in a subgroup of patients with multiple sclerosis. Ann Neurol. 2000; 47:652-655. bacteria, Acinetobacter peptides, which carry the cross- 11. Chatzipanagiotou S, Tsakanikas C, Anagnostouli M, Rentzos M, Ioannidis A, Nico- reactive brain antigens, may give a clearer separation when laou C. Detection of Chlamydia pneumoniae in the cerebrospinal fluid of pa- compared with controls.15 tients with multiple sclerosis by combination of cell culture and PCR: no evidence for possible association. Mol Diagn. 2003;7:41-43. 12. Eylar EH, Caccam J, Jackson JJ, Westall FC, Robinson AB. Experimental allergic MONITORING TREATMENT IN MS encephalomyelitis. Science. 1970;168:1220-1223. 13. Ebringer A, Pirt SJ, Wilson C, Cunningham P, Thorpe C, Ettelaie C. Bovine spon- giform encephalopathy. Environ Health Perspect. 1997;105:1172-1174. The current therapeutic strategy in the management of 14. Hughes LE, Bonell S, Natt RS, et al. Antibody responses to Acinetobacter spp. and Pseudomonas aeruginosa in multiple sclerosis. Clin Diagn Lab Immunol. MS involves the use of immunomodulatory and immu- 2001;8:1181-1188. nosuppressive drugs. These modalities have been found 15. Hughes L, Smith PA, Bonell S, et al. Cross-reactivity between related sequences to be effective mainly in cases of relapsing-remitting forms found in Acinetobacter spp., Pseudomonas aeruginosa, myelin basic protein and 36 myelin oligodendrocyte glycoprotein in multiple sclerosis. J Neuroimmunol. 2003; but less so in the progressive forms of MS. The evi- 144:105-115. dence of the bacterial (Acinetobacter and Pseudomonas) 16. Allerberger F, Berger T, Reindl M, et al. Anti-Acinetobacter and anti-neuronal antibodies in Austrian patients with multiple sclerosis and other neurological diseases. involvements in the etiopathogenesis of MS suggests the Ann Neurol. In press. possibility of using antimicrobial therapy, and its use 17. Gay D, Dick G, Upton G. Multiple sclerosis associated with sinusitis: case- should be evaluated in prospective, longitudinal ran- controlled study in general practice. Lancet. 1986;1:815-819. 18. Jones RL, Crowe P, Chavda SV, Pahor AL. The incidence of sinusitis in patients domized controlled studies. with multiple sclerosis. Rhinology. 1997;35:118-119. In conclusion, the demonstration of the cross- 19. Bert F, Lambert-Zechovsky N. Sinusitis in mechanically ventilated patients and its role in the pathogenesis of nosocomial pneumonia. Eur J Clin Microbiol In- reactivity between Acinetobacter and Pseudomonas bacte- fect Dis. 1996;15:533-544. ria and brain antigens has raised the possibility of devel- 20. Casiano RR, Cohn S, Villasuso E, et al. Comparison of antral tap with endoscopi- oping a disease marker, the MAN index, which could cally directed nasal culture. Laryngoscope. 2001;111:1333-1337. 21. Lamoureux G, Lapierre Y, Ducharme G. Past infectious events and disease evo- predict the occurrence of relapses in patients with MS lution in multiple sclerosis. J Neurol. 1983;230:81-90. and/or monitor response to therapy. However, further stud- 22. Buljevac D, Flach HZ, Hop WCJ, et al. Prospective study on the relationship between infections and multiple sclerosis exacerbations. Brain. 2002;125:952- ies need to be performed on greater numbers of MS pa- 960. tients to establish the sensitivity and specificity profiles and 23. Ebringer A, Wilson C. HLA molecules, bacteria and autoimmunity. J Med Microbiol. 2000;49:305-311. the degree of the reproducibility of the MAN index. 24. Wilson C, Rashid T, Tiwana H, et al. Cytotoxicity responses to peptide antigens in rheumatoid arthritis and ankylosing spondylitis. J Rheumatol. 2003;30:972- Accepted for Publication: January 28, 2004. 978. 25. Warren KG, Catz I. A correlation between cerebrospinal fluid myelin basic pro- Correspondence: Alan Ebringer, MD, Division of Life Sci- tein and anti-myelin basic protein in multiple sclerosis. Ann Neurol. 1987;21: ences, Infection and Immunity Group, King’s College Lon- 183-189. 26. Reindl M, Linington C, Brehm U, et al. Antibodies against the myelin oligoden- don, 150 Stamford St, London SE1, England (alan.ebringer drocyte glycoprotein and the myelin basic protein in multiple sclerosis and other @kcl.ac.uk). neurological diseases: a comparative study. Brain. 1999;122:2047-2056. Author Contributions: Acquisition of data: Ebringer, 27. Silber E, Semra YK, Gregson NA, Sharief MK. Patients with progressive multiple sclerosis have elevated antibodies to neurofilament subunit. Neurology. 2002; Hughes, Rashid, and Wilson. Analysis and interpretation 58:1372-1381. of data: Ebringer, Hughes, Rashid, and Wilson. Critical 28. Berger T, Rubner P, Schautzer F, et al. Antimyelin antibodies as a predictor of clinically definite multiple sclerosis after a first demyelinating event. NEnglJMed. revision of the manuscript for important intellectual con- 2003;349:139-145. tent: Ebringer, Hughes, Rashid, and Wilson. Statistical 29. Cross AH, Trotter JL, Lyons J. B cells and antibodies in CNS demyelinating disease. analysis: Ebringer, Hughes, Rashid, and Wilson. J Neuroimmunol. 2001;112:1-14. 30. Bernard CCA, Mandel T, Mackay IR. Experimental models of human autoim- Funding/Support: This work was supported by the Ameri- mune disease: overview and prototypes. In: Rose NR, Mackay IR, eds. The Au- can Friends of King’s College, London, England. toimmune Diseases. San Diego, Calif: Academic Press; 1992:47-106. 31. Revel M. Interferon-␤ in the treatment of relapsing-remitting multiple sclerosis. Pharmacol Ther. 2003;100:49-62. REFERENCES 32. Tiwana H, Wilson C, Pirt J, Cartmell W, Ebringer A. Autoantibodies to brain com- ponents and antibodies to Acinetobacter calcoaceticus are present in bovine spon- giform encephalopathy. Infect Immun. 1999;67:6591-6595. 1. Kurtzke JF. Multiple sclerosis: changing times. Neuroepidemiology. 1991;10:1-8. 33. Kirvan CA, Swedo SE, Heuser JS, Cunningham MW. Mimicry and autoantibody- 2. Miller DH, Hammond SR, McLeod JG, Purdie G, Skegg DC. Multiple sclerosis in mediated neuronal cell signalling in Sydenham chorea. Nat Med. 2003;9:914- Australia and New Zealand J Neurol Neurosurg Psychiatry. 1990;53:903-905. 920. 3. Hensiek AE, Sawcer SJ, Compston DA. Searching for needles in haystacks. Brain 34. Poser CM, Paty DW, Scheinberg L, et al. New diagnostic criteria for multiple scle- Res Bull. 2003;61:229-234. rosis: guidelines for research protocols. Ann Neurol. 1983;13:227-231. 4. Kenealy SJ, Pericak-Vance MA, Haines JL. The genetic epidemiology of multiple 35. McDonald WI, Compston A, Edan G, et al. Recommended diagnostic criteria for sclerosis. J Neuroimmunol. 2003;143:7-12. multiple sclerosis: guidelines from the International Panel on the diagnosis of 5. Willer CJ, Dyment DA, Risch NJ, Sadovnick AD, Ebers GC. Twin concordance multiple sclerosis. Ann Neurol. 2001;50:121-127. and sibling recurrence rates in multiple sclerosis. Proc Natl Acad SciUSA. 2003; 36. Noseworthy JH. Management of multiple sclerosis: current trials and future options. 100:12877-12882. Curr Opin Neurol. 2003;16:289-297.

(REPRINTED) ARCH NEUROL / VOL 62, JAN 2005 WWW.ARCHNEUROL.COM 36