Previous Article | Next Article ![]()
Clinical and Diagnostic Laboratory Immunology, May 2005, p. 677-679, Vol. 12, No. 5
1071-412X/05/$08.00+0 doi:10.1128/CDLI.12.5.677-679.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Section of Immunobiology, Dept. of Ophthalmology, Ludwig-Maximilians-University, Munich, Germany
Received 5 November 2004/ Returned for modification 11 January 2005/ Accepted 2 March 2005
|
|
|---|
|
|
|---|
![]() View larger version (38K): [in a new window] |
FIG. 1. Arthritogenicity of mimotope peptides. Peptide sequences are shown with their amino acid homologies or similarities with the HLA-B27 peptides B27PD, B2702PA, and B27PB and/or their autoantigen mimotopes (PDSAg or Ker333) underlined. Left brackets mark the respective mimotope pairs, B27PD/PDSAg and B2702PA/Ker333. Prop, Propionibacterium freudenreichii pyrophosphate-fructose 6-phosphate 1-phosphotransferase (amino acids [aa] 159 to 168); B27PD, HLA-B (aa 125 to 138); PDSAg, retinal S antigen (aa 341 to 354); Rota, rotavirus vp4 (aa 591 to 601); Rick, Rickettsia prowazekii hypothetical protein RP006 (aa 245 to 255); RRR, human retinoid-related receptor alpha (aa 5 to 14); B2702PA, HLA-B27 (aa 60 to 80); Ker333, cytokeratin 5 (aa 333 to 353); CLP, chicken cartilage link protein (aa 183 to 193); Shig, Shigella flexneri flagellin (aa 108 to 121); Bor, Borrelia burgdorferi RNase H (aa 117 to 128); B27PB, HLA-B27 (aa 109 to 118); Kleb, Klebsiella pneumoniae nitrogenase cofactor synthesis protein (aa 220 to 229); CMV, human herpesvirus 5 open reading frame UL136 (aa 88 to 97). Bars show clinical arthritis scores and standard errors from groups of 4 to 38 rats. The incidence of arthritis is shown at the right side of the diagram. Solid horizontal lines separate groups of related mimotopes (PDSAg/B27PD; B2702PA/Ker333; and B27PB, Kleb, and CMV). Dashed lines within these groups separate subgroups of mimotopes (e.g., B27PD and Prop versus PDSAg, Rota, and Rick).
|
Uveitis was induced only after immunization with peptides PDSAg and Rota and, in a very mild form, with B27PD (Fig. 2). No other morbidities or abnormalities were observed by physical and morphological examination of the animals, except for arthritis in the cases shown in Fig. 1. Complete Freund's adjuvant immunization alone was neither arthritogenic nor uveitogenic (data not shown). Of those peptides, which represent mimotopes of PDSAg and B27PD (Rota, Rick, and Prop), only Rota induced arthritis in limbs, digits, and tail spines in five of eight rats (Fig. 1). We found a significant correlation of uveitis and arthritis severity in Rota-immunized rats (Fig. 2) (Spearman rank correlation, r = 0.826; P = 0.015).
![]() View larger version (25K): [in a new window] |
FIG. 2. Uveitis and arthritis in rats. Concomitant appearance of arthritis (open symbols; clinical score) and uveitis (closed symbols; histological score) in rats immunized with mimotopes Rota (circles), PDSAg (squares), and B27PD (triangles) is shown. A scale for arthritis scores is shown at the left y axis; one for uveitis scores is shown at the right. Dotted lines link the respective arthritis and uveitis scores of single animals.
|
C- and/or N-terminal truncation of HLA-B27-specific peptide B2702PA resulted in peptides B27PA and B2702PAS with increased arthritogenicity, suggesting that trimming might favor the presentation of highly pathogenic epitopes with low major histocompatibility complex (MHC)-binding capacity over the presentation of strongly MHC-binding but minor pathogenic epitopes. Peptide Ker333 and its truncated form, KerS, show similar arthritogenicities, indicating that KerS represents the predominant epitope within that 20-mer peptide.
The mimotopes of the arthritogenic HLA-B27 peptides B27PA, B2702PA, and B2702PAS and the cytokeratin peptides Ker333, RRR, CLP, Shig, and Bor were derived either from other autoantigens (RRR and CLP) (Fig. 1) or from bacteria (Shig and Bor). All mimotopes were arthritogenic, except for Bor, although Bor is highly homologous to the pathogenic peptide KerS but has no putative C-terminal anchor for binding (7) to rat MHC class II (12), which might explain the lack of pathogenicity in rats. Nevertheless, infections with Borrelia burgdorferi are known to be followed by Lyme disease with joint affection in humans.
Infections with Shigella flexneri cause gastroenteritis and, sometimes, reactive arthritis (15); peptide Shig from its flagellin was arthritogenic in our rat model.
The sequence of CLP from chicken cartilage link protein differs from rat and human sequences, and thus it cannot be regarded as a real "autoantigen" in our model, although it was highly arthritogenic. ROR
(retinoid-related orphan receptor alpha), the founder protein of RRR, is a nuclear receptor expressed in cartilage (1) and bone (5). Targeting the activation of ROR
(9) was previously described as a potential treatment for arthritis (6).
The HLA-B27-derived peptide B27PB was highly arthritogenic in Lewis rats following subcutaneous immunization. However, testing proliferation of peripheral blood lymphocytes from patients with ankylosing spondylitis or rheumatoid arthritis did not reveal B27PB-specific responses (4). Kleb was derived from Klebsiella pneumoniae, which can cause urinary tract infections and pneumonia. This pathogen is described as a potential causative agent for ankylosing spondylitis (3). Peptide Kleb has only three major amino acid differences from B27PB (Y5, >G; A9, >Y; and Y10, >G), but is significantly (P = 0.045 by the Mann-Whitney test) less arthritogenic.
Peptide CMV differs from B27PB in four amino acids (Y1, Y4, S9, and W10) but completely lacks pathogenicity. The restriction element for B27PB and its potential mimotopes Kleb and CMV is unknown; therefore, we cannot speculate about anchor positions that facilitate or impede MHC binding. The joint-specific mimotope of T cells reactive to B2702PA and its derivatives B27PA and B2702PAS is postulated as being cytokeratin, represented by peptide Ker333. The joint-specific autoantigens mimicked by peptides Rota, B27PB, and Kleb finally targeted by the arthritogenic immune responses are unknown.
Experimental autoimmune uveitis and arthritis in the Lewis rat reflect the human situation with respect to immune responses to the uveitogenic peptides PDSAg, B27PD, and Rota (10, 14) as well as to peptide B27PA (4). In vivo we observed pathogenic effects by induction of autoimmune diseases (arthritis and/or uveitis) in rats; in vitro we found cross-reactive T-cell responses (PDSAg and Rota [10]; B2702PA and Ker333 [11]).
We have shown that cross-reactive immune responses can be highly promiscuous, because certain autoantigen peptides can be mimicked by more than one mimotope of viral, bacterial, or even self origin. This indicates that an autoimmune response must not originate from a single mimotope of a certain infectious agent and is probably not even directed to only a single autoantigen. Multiple cross-reactivities between tissue-specifically expressed autoantigens might explain the effect on multiple tissues or organs by some autoimmune diseases.
This work was supported by the Deutsche Forschungsgemeinschaft SFB 571 and the Fördergesellschaft zur Behandlung von Autoimmunerkrankungen e.V.
|
|
|---|
and
and retinoid Z receptor in cartilage. PPAR
activation modulates the effects of interleukin-1ß on rat chondrocytes. J. Biol. Chem. 275:12243-12250.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»