Nucleotide sequence of an attenuated mutant of coxsackievirus B3 compared with the cardiovirulent wildtype: assessment of candidate mutations by analysis of a revertant to cardiovirulence.

BACKGROUND: Coxsackievirus B3 (CVB3) causes myocarditis in the SWR (H2q) mouse model and persistence of CVB3 in myocardium disposes to the development of dilated cardiomyopathy. An attenuated strain of CVB3 has been isolated, sequenced and several candidate mutations for attenuation identified. Derivation of a revertant to cardiovirulence allows the significance of these mutations to be assessed. OBJECTIVES: To ascertain which candidate mutation(s) determine(s) the attenuated phenotype. STUDY DESIGN: A revertant to cardiovirulence was isolated following passage through severe combined immunodeficient disease (SCID) mouse heart. The 5'-non-translated region (NTR) and region coding for capsid proteins were sequenced and compared to the wildtype and attenuant. RESULTS: There are five candidates for attenuation: (1) A-G at base 580 in the 5'-NTR; (2) A-T at base 690 in the 5'-NTR; (3) CG-GC at bases 1401/2 (Thr to Ser at amino acid 151 in VP2); (4) AA-GT at bases 2691/2 (Lys to Ser at amino acid 80 in VP1); (5) A-G at base 2916 (Asp to Gly at amino acid 155 in VP1). It was shown previously that mutations at 580, 690 and 2691/2 are not important in attenuation. Additionally, there are three novel mutations in the coding region of the revertant and one in the 5'-NTR which are unlikely to be relevant for attenuation as they are not present in the attenuant. Of nucleotide changes seen at 1401/2 and 2916 in the attenuant, only 2916 reverts to the wildtype sequence and so is a strong candidate for a determinant of attenuation.

Coxsackievirus B3-induced myocarditis. Characterization of stable attenuated variants that protect against infection with the cardiovirulent wild-type strain.

Coxsackievirus B3 (CVB3) is the enterovirus most frequently involved in human myocarditis or dilated cardiomyopathy. Attenuated variants were derived from a cardiovirulent CVB3 reactivated from a sequenced, full-length cDNA clone. The prophylactic potential of these variants was assessed in SWR/Ola (H-2q) mice. Animals immunized with attenuated variants of CVB3 were protected from myocarditis when challenged subsequently with the cardiovirulent wild-type virus. In contrast to nonimmunized controls, the wild-type virus was not isolated from myocardium of protected mice, nor was viral RNA detected in myocardium by reverse transcription nested polymerase chain reaction. Specific antibody to CVB3 was demonstrated by virus neutralization assay and by indirect immunofluorescence. The attenuated phenotype of one variant, p14V-1, remained stable throughout 20 consecutive passages in SWR mice and induced a markedly lower level of autoantibody against mouse cardiac myosin heavy chain than the cardiovirulent wild type. These data demonstrate that attenuated strains protect against CVB3-induced myocarditis in mice, that the attenuated phenotype is stable, and that they do not persist in myocardium nor induce a significant level of anti-heart anti-body against myosin heavy chain. These attenuants may be the basis of a live vaccine against CVB3 in the prevention of enteroviral heart muscle disease.

A single amino acid substitution in the capsid protein VP1 of coxsackievirus B3 (CVB3) alters plaque phenotype in Vero cells but not cardiovirulence in a mouse model.

We previously described a large plaque attenuant (p14V-1) derived from a cardiovirulent Coxsackievirus B3 (CVB3) and showed that there were no major determinants of either attenuation or plaque phenotype in the 5' nontranslated region (5'NTR). Part of the region encoding the last 124 amino acids of VP3 and the first 106 amino acids of VP1 of the attenuant was then sequenced and compared to the wild-type. Three nucleotide changes were found in the VP1 coding region: a silent single base change at nucleotide position 2467 (C to U) and a double-base change at position 2690-1 (AA to GT), which leads to a change from lysine to serine at amino acid position 80. This mutation maps to the begining of B-C loop of the three-dimensional structure of VP1 of CVB3, where a distinct surface projection is formed. Two infectious chimeric cDNA clones were constructed, based on a cardiovirulent cDNA construct. In one construct, the 5'NTR and the VP3-VP1 region were from p14V-1 and in the other, only the VP3-VP1 region was from this attenuant. Both chimeric viruses produced large plaques on Vero cell monolayers, similar to p14V-1 but larger than the prototypic cardiovirulent virus. In vivo experiments showed that both chimeric viruses induced myocarditis in a murine model, similar to wild-type virus. We conclude that mutation serine-80 in capsid protein VP1 of p14V-1 is a determinant of the large plaque phenotype but is not responsible for attenuation.

The immune response to and expression of cross-reactive retroviral gag sequences in autoimmune disease.

To examine the immune response to retroviral gag sequences in autoimmune disease, we measured antibody levels to synthetic peptides representing the major epitopes on HTLV-1 p19 gag and a homologous sequence on the endogenous retrovirus, HRES-1, in sera from 121 patients with autoimmune disease and 52 healthy controls. In the absence of HTLV-1 antibodies, using a conventional diagnostic assay, significantly elevated levels of antibodies to the HTLV-1 peptide were found in 23% of multiple sclerosis and 20% of anti-Sm antibody positive systemic lupus erythematosus patients. Elevated antibody levels to HRES-1 were found in 32% of Sjögren's syndrome and 19% of multiple sclerosis patients. Evidence of reactivity with both HTLV-1 and HRES-1 was found in human sera and cross-reactivity demonstrated with affinity purified rabbit anti-peptide antibodies. Expression of HRES-1, detected by antibodies and Northern blots, was found in lymphoblastoid cells, salivary gland biopsy sections and salivary gland epithelial cells in culture. This study confirms previous reports of low levels of anti-retroviral gag antibodies in autoimmune disease. The cross-reactions support the concept that reports of elevated HTLV-1 antibodies may be due to an endogenous agent such as HRES-1. The expression of HRES-1 salivary gland may explain its antigenicity in a small proportion of Sjögren's syndrome patients as well as suggesting mechanisms whereby it may contribute to the chronic inflammation of autoimmune disease.