Permissiveness of human embryonal fibroblasts for coxsackieviruses B3. Investigations on virus genetic markers in vitro and localization of virus receptor distribution by immunogold replica technique

The viral genetic background determines the outcome of coxsackievirus B3 infection in outbred NMRI mice

The reasons for the different outcome of coxsackievirus B3 (CVB3)-induced heart disease in humans are not well understood. Since there are no experimental data on the course of disease after infection with genetically different CVB3 in a natural variable population until now, we studied the outcome of virus infection in outbred NMRI mice after inoculation of genetically different CVB3 variants. Adult male mice were inoculated with seven closely related CVB3 variants. The histopathological changes of heart and pancreas tissue, antibody induction, virus titers, and persistence of viral positive- as well as negative-strand RNA in spleen and heart tissue were compared at day 7 or day 28 after infection to detect prerequisites and predictive factors for chronic myocarditis. Six CVB3 variants infected NMRI mice. CVB3 infection (i) did not induce detectable myocardial injury, (ii) caused signs of healing up acute myocarditis or (iii) ongoing chronic myocarditis. Neither IgG antibody responses nor the extent of destruction of exocrine pancreatic tissue or viral RNA load in spleen did correlate with myocardial histopathology. In contrast, a high persistent viral RNA load in heart tissue specimens was characteristic for mice developing chronic myocarditis. The results of the present study corroborate high viral load in the acute stage of myocarditis and high amounts of persisting CVB3 RNA in heart tissue as predictive marker of chronic myocarditis. The outcome of CVB3-induced heart disease in outbred NMRI mice depends strongly on the viral genetic background. In particular an important role of viral capsid proteins is suggested.

N- and 6-O-sulfated heparan sulfates mediate internalization of coxsackievirus B3 variant PD into CHO-K1 cells

Recently, it was demonstrated that the coxsackievirus B3 variant PD (CVB3 PD) is able to infect coxsackievirus-adenovirus receptor (CAR)-lacking cells by using heparan sulfates (HS) as additional receptors (A. E. Zautner, U. Korner, A. Henke, C. Badorff, and M. Schmidtke, J. Virol. 77:10071-10077, 2003). For this study, competition experiments with growth factors binding to known HS sequences as well as with specifically desulfated heparins were performed with Chinese hamster ovary cells (CHO-K1) to determine the structural requirements of HS for interaction with CVB3. Hepatocyte growth factor interacting with HS sequences containing [IdUA-GlcNSO(3)(6OSO(3))](n), but not basic fibroblast growth factor binding to [HexUA-GlcNSO(3)-HexUA-GlcNSO(3)-IdUA(2OSO(3))](n), was shown to compete effectively with CVB3 PD for cell surface HS. Whereas unmodified heparin and 2-O-desulfated heparin strongly inhibited the CVB3 PD-induced cytopathic effect, the antiviral activity was markedly reduced after N-, O- and 6-O-desulfation of heparin. Taken together, these results indicate that 6-O- and N-sulfation of GlcNAc of HS is crucial for HS interaction with CVB3 PD and that the disaccharide [IdUA-GlcNSO(3)(6OSO(3))](n) is involved in viral binding. Results from experiments with various inhibitors of endocytic pathways suggest that HS-mediated virus internalization is pH dependent. Despite the fact that CVB3 PD initiates infection about four times slower by making use of HS as a receptor than by using CAR, the time required for a complete viral life cycle in Chinese hamster ovary cells was independent of the utilized receptor.

Susceptibility of coxsackievirus B3 laboratory strains and clinical isolates to the capsid function inhibitor pleconaril: antiviral studies with virus chimeras demonstrate the crucial role of amino acid 1092 in treatment

OBJECTIVES

At present, most promising compounds to treat enterovirus-induced diseases are broad-spectrum capsid function inhibitors which bind into a hydrophobic pocket in viral capsid protein 1 (VP1). Coxsackievirus B3 (CVB3) Nancy was the only prototypic enterovirus strain shown to be pleconaril-resistant. This study was designed to better understand the polymorphism of the hydrophobic pocket in CVB3 laboratory strains and clinical isolates and its implications for treatment with the capsid function inhibitor pleconaril.

METHODS

Pleconaril susceptibility was determined in cytopathic effect-inhibitory, plaque reduction or virus yield assays. Sequence analysis of the genome region coding for VP1 and/or subsequent alignment of amino acids lining the hydrophobic pocket of five CVB3 laboratory strains and 20 clinical isolates were carried out. Virus chimeras and computational analysis were used to prove the role of amino acid 1092.

RESULTS AND CONCLUSIONS

Despite high conservation of pocket amino acids, polymorphism was detected at positions 1092, 1094 and 1180. Neither Pro-1094-->Thr nor Val-1180-->Ile altered efficacy of pleconaril treatment. But the amino acid at position 1092 was strongly associated with susceptibility of CVB3 to the capsid inhibitor. Whereas leucine was involved in resistance, isoleucine and valine were detected in pleconaril-susceptible CVB3. Results from antiviral assays with hybrid viruses demonstrate the crucial role of amino acid 1092 in pleconaril susceptibility. A resistant cDNA-generated CVB3 became pleconaril-susceptible after accepting parts from the genome region encoding Ile-1092 into its capsid. Computational analysis suggests that conformational changes in the hydrophobic pocket occur when leucine is substituted for isoleucine or valine and that this change leads to susceptibility to pleconaril.

Heparan sulfates and coxsackievirus-adenovirus receptor: each one mediates coxsackievirus B3 PD infection

Amino acid exchanges in the virus capsid protein VP1 allow the coxsackievirus B3 variant PD (CVB3 PD) to replicate in decay accelerating factor (DAF)-negative and coxsackievirus-adenovirus receptor (CAR)-negative cells. This suggests that molecules other than DAF and CAR are involved in attachment of this CVB3 variant to cell surfaces. The observation that productive infection associated with cytopathic effect occurred in Chinese hamster ovary (CHO-K1) cells, whereas heparinase-treated CHO-K1 cells, glucosaminoglycan-negative pgsA-745, heparan sulfate (HS)-negative pgsD-677, and pgsE-606 cells with significantly reduced N-sulfate expression resist CVB3 PD infection, indicates a critical role of highly sulfated HS. 2-O-sulfate-lacking pgsF-17 cells represented the cell line with minimum HS modifications susceptible for CVB3 PD. Inhibition of virus replication in CHO-K1 cells by polycationic compounds, pentosan polysulfate, lung heparin, and several intestinal but not kidney HS supported the hypothesis that CVB3 PD uses specific modified HS for entry. In addition, recombinant human hepatocyte growth factor blocked CVB3 PD infection. However, CAR also mediates CVB3 PD infection, because this CVB3 variant replicates in HS-lacking but CAR-bearing Raji cells, infection could be prevented by pretreatment of cells with CAR antibody, and HS-negative pgsD-677 cells transfected with CAR became susceptible for CVB3 PD. These results demonstrate that the amino acid substitutions in the viral capsid protein VP1 enable CVB3 PD to use specific modified HS as an entry receptor in addition to CAR.

Attachment of coxsackievirus B3 variants to various cell lines: mapping of phenotypic differences to capsid protein VP1

The coxsackievirus B3 (CVB3) strain Nancy P establishes a persistent carrier-state infection without visible cytopathic effect in primary human fibroblasts (HuFi H), whereas the derivative variant PD induces a complete lysis of the cell monolayer. To define the molecular basis of this exceptional growth property, the complete genomes of both viruses were sequenced and compared to all published sequences of CVB3. As a result, six unique amino acid substitutions in the VP1 capsid protein were observed. Via hybrid virus construction, the lytic phenotype was transferred to a nonlytic cDNA-generated CVB3. Mapping experiments indicate that the presence of amino acid residues K78, A80, A91, and I92 in VP1 is sufficient to induce "lytic" infections in HuFi H cells. Binding assays demonstrate that CVB3 Nancy P preferentially binds to the human coxsackievirus-adenovirus receptor (CAR), while PD exhibits a very weak interaction with CAR but strong binding to the decay accelerating factor (DAF). These results suggest that the mutated amino acid residues in VP1 are involved in receptor recognition/binding. Moreover, the lytic replication of CVB3 PD and the hybrid virus in various nonpermissive rodent cell lines indicates that cell surface molecules other than CAR and DAF may be involved in attachment of this variant to cell surfaces.