Severe pneumonia and pathogenic damage in human airway epithelium caused by Coxsackievirus B4

Abstract

Coxsackievirus B4 (CVB4) has one of the highest proportions of fatal outcomes of other enterovirus serotypes. However, the pathogenesis of severe respiratory disease caused by CVB4 infection remains unclear. In this study, 3 of 42 (7.2%, GZ-R6, GZ-R7 and GZ-R8) patients with severe pneumonia tested positive for CVB4 infection in southern China. Three full-length genomes of pneumonia-derived CVB4 were sequenced and annotated for the first time, showing their high nucleotide similarity and clustering within genotype V. To analyze the pathogenic damage caused by CVB4 in the lungs, a well-differentiated human airway epithelium (HAE) was established and infected with the pneumonia-derived CVB4 isolate GZ-R6. The outcome was compared with that of a severe hand-foot-mouth disease (HFMD)-derived CVB4 strain GZ-HFM01. Compared with HFMD-derived CVB4, pneumonia-derived CVB4 caused more intense and rapid disruption of HAE polarity, leading to tight-junction barrier disruption, loss of cilia, and airway epithelial cell hypertrophy. More pneumonia-derived CVB4 were released from the basolateral side of the HAE than HFMD-derived CVB4. Of the 18 cytokines tested, only IL-6 and IL-1b secretion significantly increased on bilateral sides of HAE during the early stage of pneumonia-derived CVB4 infection, while multiple cytokine secretions significantly increased in HFMD-derived CVB4-infected HAE. HFMD-derived CVB4 exhibited stronger neurovirulence in the human neuroblastoma cells SH-SY5Y than pneumonia-derived CVB4, which is consistent with the clinical manifestations of patients infected with these two viruses. This study has increased the depth of our knowledge of severe pneumonia infection caused by CVB4 and will benefit its prevention and treatment.

Keywords: Coxsackievirus B4; full-length genome; hand-foot-mouth disease (HFMD); human airway epithelium; neurovirulence; pathogenic damage; severe pneumonia.

Figure 1.

The genome organization (A) and percent identities of VP1/genome sequences (B) among pneumonia-derived/HFMD-derived coxsackievirus B4 isolates and prototype J.V.B. Benschoten. The genome is indicated by the black horizontal line marked at 2000-nt intervals. Untranslated regions (UTR) at both ends are designated by gray boxes; the dark red arrow indicates the coding sequence; while light red arrows designate mature peptides. Percent identities are showed as VP1/genome.

Figure 2.

Phylogenetic analysis of genome of coxsackievirus B4 isolates. Three pneumonia-derived coxsackievirus B4 isolates in this research were subjected to genome sequencing analysis to determine their phylogenetic relationships using the neighbour-joining method with 1000 bootstrap replicates implemented in MEGA 11 software. For reference, taxon names include genome type, corresponding GenBank accession number, and country of isolation, strain name, and year of isolation. Strains of coxsackievirus B4 isolated from pneumonia patients in this study are marked with “▴”, the HFMD-derived strain GZ-HFM01 is marked with “●”, a previous reported pneumonia-derived strain with complete ORF is marked with “▾” and the prototype strain is marked with “▪”.

Figure 3.

Plaque formation (A), plaque size distribution (B) and the proliferation curve (C) of the pneumonia-derived CVB4 GZ-R6 and the reference HFMD-derived coxsackievirus B4 GZ-HFM01 infection in SH-SY5Y cells. Plaque plates were incubated and stained with crystal violet for 3 days in SH-SY5Y cells in six-well culture plates. Plaque sizes are shown as mean values with SEM and performed by Mann – Whitney U test. Virus genome copies were quantified by qPCR at different hours post-infection (n = 3). Statistical analysis was conducted by ANOVA. ns, not significant; ****, p < 0.0001.

Figure 4.

Pneumonia-derived and HFMD-derived coxsackievirus B4 infection in HAE were persistent and caused cytopathogenic effects. GZ-R6 and GZ-HFM01 infected HAE from the apical surface. At the indicated days p.i., the proliferation curves of the two viruses at the apical side (A) and basolateral side (B). At 8 days p.i., coxsackievirus B4-infected HAE membranes taken from the bottom of the inserts were embedded in paraffin, sectioned, and stained using hematoxylin and eosin (C). The transepithelial electrical resistance (TEER) of mock – and coxsackievirus B4-infected HAE was measured at 8 d.p.i. (D). Data are shown as mean values with SEM (n = 4). Statistical analysis was performed by ANOVA test. ns, not significant; ***, p < 0.001.

Figure 5.

Immunofluorescence analysis of the tight junction protein ZO-1 and the cilia marker α-tubulin during pneumonia-derived coxsackievirus B4 GZ-R6 and HFMD-derived coxsackievirus B4 GZ-HFM01 infection of HAE. Mock – and CVB4-infected HAE cultures at the indicated days post-infection (d.p.i.) were co-stained with anti-CVB4 and anti-ZO-1 antibodies (A), or co-stained with anti-coxsackievirus B4 and anti-α-tubulin antibodies (B). Confocal images were taken at a magnification of ×40. Nuclei were stained with DAPI (blue).

Figure 6.

Cytokines secretion of HAE with pneumonia-derived coxsackievirus B4 GZ-R6 and HFMD-derived coxsackievirus B4 GZ-HFM01 infection. IL-6 and IL-1b secretion increased in both apical and basolateral sides rapidly in 2 to 4 d.p.i. in HAE with GZ-R6 infection comparing with mock – and GZ-HFM01 infection, and then decreased immediately (A); cytokines increased significantly in the basolateral but not apical side in GZ-R6 infected HAE, while stronger secretion from both sides occurred in GZ-HFM01-infected HAE, especially in the basolateral direction (B); both sides had no obvious differences in HAE with GZ-R6 infection but increased in GZ-HFM01 infection (C). Statistical analysis compared with mock-infection was performed by ANOVA test. *, p < 0.05; ** p < 0.01; ***, p < 0.001; ****, p < 0.0001.