A Campylobacter jejuni gene associated with immune ... - Nature

© 2001 Nature Publishing Group http://medicine.nature.com

LETTERS TO THE EDITOR


A Campylobacter jejuni gene associated with immune-mediated neuropathy To the editor—Campylobacter jejuni is the most frequent cause of bacterial diarrhea, which leads to the Guillain–Barré syndrome (GBS) or the Miller Fisher syndrome (MFS) in 1 in every 1000 infections1. Neuropathy is probably triggered by molecular mimicry between C. jejuni lipooligo- or lipopolysaccharides (LOS or LPS) and ganglioside epitopes on peripheral nerve tissue. Biochemical studies indeed revealed that C. jejuni LOS contains ganglioside-like structures with sialic acid as an essential ingredient2. Most C. jejuni strains involved in GBS or MFS isolated in the United States and Japan are serotype O:19 and are genetically closely related3. This hampers the search for bacterial factors involved in the development of GBS or MFS based on genetic comparisons among strains. Recently, however, a group of phenotypically and genetically unrelated GBS/MFS-associated C. jejuni strains was described4. This provided the opportunity to search for the differential presence of genes putatively involved in the

biosynthesis of the gangliosides-mimicking epitopes with emphasis on the LOSmodifying sialic acid transferaseencoding gene cstII (ref. 11). When the GBS/MFS-associated strains were compared with the control strains, no significant difference in the occurrence of GM1, GQ1b or other ganglioside epitopes in the bacterial LOS was observed (see Table 1). The immune detection of a GQ1b epitope in the bacterial LOS moiety of GBS/MFS-related strains is strongly associated with the occurrence of an anti-GQ1b response in the patients. In contrast, PCR–RFLP (restriction-fragment length polymorphism) analysis of cstII gene PCR product revealed some sequence heterogeneity and, more importantly, that this particular gene was frequently missing from the C. jejuni genomes (Table 1). In 12 of 34 strains analyzed, the PCR was negative, which was corroborated by Southern-blot analysis (data not shown). There was no apparent difference in gene frequency between strains from GBS/MFS patients versus the

control strains (13/18 versus 9/16). However, 8/8 (100%) strains displaying a GQ1b epitope have cstII, whereas among the GQ1b-negative strains only 14/26 (54%) are cstII positive (P = 0.03). This indicates that the cstII sialic acid transferase is a necessary determinant for the synthesis of GQ1b-like epitopes, but in itself not sufficient to produce the epitopes and, consequently, trigger GBS or MFS. The PCR–RFLP patterns did not associate with any of the serological responses or the nature of the disease caused by the C. jejuni strain involved. For Haemophilus influenzae, for instance, sialylation of LPS is considered a virulence factor defining serum resistance9. It was recently shown that sialylation of LOS cores affects serum resistance in C. jejuni10. The observations of GQ1b epitopes in the C. jejuni LOS, the serological response in the patients and the ubiquitous presence of the cstII gene indicate that the activity of the bi-functional sialic acid transferase is necessary to synthesize a GQ1b-like epitope. This

Table 1 Presence of cstII among GBS/MFS-associated and control C. jejuni strains

GBS/MFS-associated strains

Control strains

Strain

Ganglioside epitopes GM1 GQ1b Other

cstII analysis pres. AluI DraI

GB1 GB4 GB11 GB21 GB2 GB3 F13 F14 GB18 GB19 GB17 GB16 MF6 MF20 MF7 MF8 GB5 GB15

+ + + + + + + + + + + + + + – – – –

– – + + + + – – + + + + + + + + + –

– – – – – – – – – – + + + + + + – –

– – – – + + + + + + + + + + + + + –

AA AA AA AA

AA DA BB DA BB EB CC AA BB

Strain

Ganglioside epitopes GM1 GQ1b Other

cstII analysis pres. AluI DraI

Pen1 Pen2 Pen3 Pen4 Pen19 Pen23 Pen35 Pen36 Pen64 98–623 98–624 98–652 1033 1039 1040 1087

+ + – + + – – ± – – – + + – – +

– – – + + + + + – + – + + + – –

– – – – – – – + – – – + – – – –

+ + – ± + + + + + – + + + – – –

AA AA CC FD AA FE GB BB FE

Strains displayed a unique genotype, only F13 and F14 are closely related. The ganglioside epitopes in the LOS and the anti-ganglioside responses in the GBS/MFS patients were determined using monoclonal antibodies and defined serum samples5–8. DNA was isolated using the Wizard Genomic kit (Promega, Madison, Wisconsin). PCR for the cstII gene used a primer set based on the cstII DNA sequence of C. jejuni OH4384 (ref. 11) (5′-ATGAAAAAAGTTATTATTGCTGGAAATG-3′and 5′-TTATTTTCCTTTGAAATAAT-GCTTTATTC-3′). Variability in cstII was assessed with AluI and DraI RFLP analysis. Associations were verified with Instat (Graphpad Software, San Diego, California) using 2×2 contingency tables. Fisher’s exact tests determined two sided P-values (P < 0.05 considered significant). GB: Guillain–Barré associated isolate; MF: Miller Fisher associated isolate, F: isolate from GB family member. The collection of control strains consists of Penner type strains and isolates from enteritis patients from the same geographical locale and isolation period as the GBS and MFS strains. Note that all MFS-derived strains are GQ1b-positive.

NATURE MEDICINE • VOLUME 7 • NUMBER 7 • JULY 2001

752




eventually leads to an anti-GQ1b immune response in the host and consequent neurological symptoms. Here we present the first example of a bacterial determinant associated with the pathogenesis of post-infectious acute immunemediated neuropathy. Although the circumstantial evidence is strong, it remains to be delineated whether this association translates into a causal relationship between cstII gene activity and GBS/MFS.

ALEX VAN BELKUM1, NICOLE VAN DEN BRAAK1, PEGGY GODSCHALK1, WIM ANG1,2, BART JACOBS2, MICHEL GILBERT3, WARREN WAKARCHUK3, HENRI VERBRUGH1 & HUBERT ENDTZ1 1 2

Medical Microbiology & Infectious Diseases and Neurology and Immunology, Erasmus University

Medical Center Rotterdam, Rotterdam, the Netherlands 3 Institute of Biological Sciences, National Research Council of Canada, Ottawa, Ontario, Canada Email: [email protected] 1. Nachamkin, I., Allos, B.M. & Ho, T. Campylobacter species and Guillain–Barré syndrome. Clin. Microbiol. Rev. 11, 555–567 (1998). 2. Moran, A.P., Appelmelk, B.M. & Aspinall, G.O. Molecular mimicry of host structures by lipopolysaccharides of Campylobacter and Helicobacter spp.: Implications in pathogenesis. J. Endotoxin Res. 3, 521–531 (1996). 3. Kuroki, S. et al. Campylobacter jejuni strains from patients with Guillain–Barré syndrome belong mostly to penner serogroup 19 and contain β-N-acetylglucosamine residues. Ann. Neurol. 33, 243–247 (1993). 4. Endtz, H.P. et al. Molecular characterization of C. jejuni from patients with Guillain–Barré and Miller Fisher syndrome. J. Clin. Microbiol. 38, 2297–2301 (2000). 5. Hartung, H.P., Van der Meché, F.G.A. & Pollard, J.D. Guillain–Barré syndrome, CDIP and other

chronic immune-mediated neuropathies. Curr. Opin. Neurol. 11, 497–513 (1998). 6. Ang, C.W. et al. Structure of C. jejuni lipopolysaccharides determines anti-ganglioside specificity and clinical features of Guillain–Barré and Miller Fisher patients (In: PhD Thesis). (Erasmus University Rotterdam, Rotterdam, the Netherlands, 2001). 7. Sheikh, K.A. et al. C. jejuni lipopolysaccharides in Guillain–Barré syndrome: Molecular mimicry and host susceptibility. Neurology 51, 371–378 (1998). 8. Ang, C.W. et al. C. jejuni lipopolysaccharides from Guillain–Barré syndrome patients induce IgG antiGM1 antibodies in rabbits. J. Neuroimmunol. 104, 133–138 (2000). 9. Hood, D.W. et al. Sialic acid in the lipopolysaccharide of H. influenzae: strain distribution, influence on serum resistance and structural characterization. Mol. Microbiol. 33, 679–692 (1999). 10. Guerry, P. et al. Sialylation of lipooligosaccharide cores affects immunogenicity and serum resistance of C. jejuni. Infect. Immun. 68, 6656–6662 (2000). 11. Gilbert, M. et al. Biosynthesis of ganglioside mimics in C. jejuni OH4384: identification of the glycosyltransferase genes, enzymatic synthesis of model compounds, and characterization of nanomole amounts by 600-MHz 1H and 13C NMR analysis. J. Biol. Chem. 275, 3896–3906 (2000).

Animal cloning experiments still banned in Italy To the editor—A news story in the May issue of your journal that reports on worldwide legislation regarding human cloning (Nature Med., 7, 518; 2001) gives the impression that Italy has one of the most open and advanced policies in the world. This is far from the truth. In fact, an ‘ordinanza’ (legislative power used by the Ministry for urgent and temporary matters) has been in place since March 1997 banning “any experiment targeted directly or indirectly to human and animal cloning.” This rule permits only the cloning of transgenic animals or endangered species assuming that cloning is a routine technique that does

not require experimentation1. In 1999, when our laboratory announced the cloning of Galileo (a bull obtained from blood cells2) we were charged with a criminal offence3, a verdict that was overturned six months later by a High Court Judge in Cremona who ruled that the ban on cloning was illegal. Although the scientific community had high hopes that the government would change its position on the matter— Health Minister Umberto Veronesi created a Commission headed by Renato Dulbecco that returned an open-minded report on cloning at the end of last year—nothing has changed. The ‘ordi-

nanza’ banning cloning it is still in place and any scientist embarking on cloning experiments in Italy (animal and human) could still face criminal charges.

CESARE GALLI & GIOVANNA LAZZARI Laboratorio di Tecnologie della Riproduzione Consorzio Incremento Zootecnico Cremona, Italy Email: [email protected] 1. Gazzetta Ufficiale. n 55, 7 March (1997). 2. Galli, C. et al. Mammalian leukocytes contain all the genetic information necessary for the development of a new individual. Cloning 1, 161–170 (1999). 3. Simini, B. Italian scientist investigated after animal cloning experiment. Lancet 354, 1365 (1999).


We want to hear from you! Nature Medicine is the forum for the latest, best, and most original biomedical research, news, and opinion. As such, we welcome letters from readers wishing to address topics reported on in previous issues, or subjects of interest to the biomedical research community at large. Letters should be brief and concise (no more than 500 words), and sent to Nature Medicine, 345 Park Avenue South, New York NY 10010, USA, or sent by fax (212.683.5751) or email to [email protected].

NATURE MEDICINE • VOLUME 7 • NUMBER 7 • JULY 2001

753