Human otic progenitor cell models of congenital hearing loss reveal potential pathophysiologic mechanisms of Zika virus and cytomegalovirus infections

Abstract

Congenital hearing loss is a common chronic condition affecting children in both developed and developing nations. Viruses correlated with congenital hearing loss include human cytomegalovirus (HCMV) and Zika virus (ZIKV), which causes congenital Zika syndrome. The mechanisms by which HCMV and ZIKV infections cause hearing loss are poorly understood. It is challenging to study human inner ear cells because they are encased in bone and also scarce as autopsy samples. Recent advances in culturing human stem cell-derived otic progenitor cells (OPCs) have allowed us herein to describe successful in vitro infection of OPCs with HCMV and ZIKV, and also to propose potential mechanisms by which each viral infection could affect hearing. We find that ZIKV infection rapidly and significantly induces the expression of type I interferon and interferon-stimulated genes, while OPC viability declines, at least in part, from apoptosis. In contrast, HCMV infection did not appear to upregulate interferons or cause a reduction in cell viability, and instead disrupted expression of key genes and pathways associated with inner ear development and function, including Cochlin, nerve growth factor receptor, SRY-box transcription factor 11, and transforming growth factor-beta signaling. These findings suggest that ZIKV and HCMV infections cause congenital hearing loss through distinct pathways, that is, by inducing progenitor cell death in the case of ZIKV infection, and by disruption of critical developmental pathways in the case of HCMV infection.

Importance: Congenital virus infections inflict substantial morbidity and devastating disease in neonates worldwide, and hearing loss is a common outcome. It has been difficult to study viral infections of the human hearing apparatus because it is embedded in the temporal bone of the skull. Recent technological advances permit the differentiation of otic progenitor cells (OPCs) from human-induced pluripotent stem cells. This paper is important for demonstrating that inner ear virus infections can be modeled in vitro using OPCs. We infected OPCs with two viruses associated with congenital hearing loss: human cytomegalovirus (HCMV), a DNA virus, or Zika virus (ZIKV), an RNA virus. An important result is that the gene expression and cytokine production profiles of HCMV/ZIKV-infected OPCs are markedly dissimilar, suggesting that mechanisms of hearing loss are also distinct. The specific molecular regulatory pathways identified in this work could suggest important targets for therapeutics.

Keywords: Zika virus; hearing loss; human cytomegalovirus; inner ear; model system; organoid; virus infection.

Fig 1

Differentiation of hiPSC into otic progenitor cells. (A) Schematic summary of the differentiation protocol for deriving the otic progenitor cells from hiPSCs. (B) A bright-field view of SK8-A hiPSCs (d0), otic placodal progenitor cells (d14), and otic progenitor cells (d20). Scale bars, 100 µm. (C) Quantitative reverse transcription polymerase chain reaction (RT-qPCR) data showing expression of otic-specific marker transcripts and control transcripts expressed as fold change of d20 OPCs relative to SK8-A hiPSCs (d0). Data were normalized to control for the amount of RNA isolated from SK8-A hiPSCs (d0). n = 3 biological samples, 3 technical repeats; *P < 0.05, ***P < 0.001, ****P < 0.0001; mean ± SEM. Panels (D–F) Representative images of the expression of otic lineage markers by immunocytochemistry in OPCs derived from UCSD112i-2–11 (D, F) and SK8-A (E) 20 days post differentiation. Scale bars, 50 µm. At least three biological samples and three technical repeats. Abbreviations: d, days; DBZ, difluoro-benzeneacetamide; OCT4, octamer-binding transcription factor 4; SOX2, sex-determining region Y-box 2; GATA3, GATA binding protein 3; PAX2, paired box gene 2; POU4F3, POU class 4 homeobox 3; ATOH1, atonal homolog 1; CXCL1, C-X-C motif chemokine ligand 1; DARC, duffy antigen/chemokine receptor; COCH, cochlin; OPG, osteoprotegerin.

Fig 2

ZIKV and HCMV infect human iPSC-derived otic progenitor cells. All statistical comparisons are to mock-infected cell values. Panels A and E: imaging of OPCs infected by HCMV or ZIKV at 48 hpi and 96 hpi, respectively. X-axis labels in A–E represent imaging with (i) immunofluorescent anti-actin antibody (F-actin), (ii) GFP-tagged CMV or immunofluorescent anti-ZIKV envelope protein antibody; (iii) DAPI (4′,6-diamidino-2-phenylindole), and (iv) merge of the signals. HI, heat-inactivated virus. Panels B and F: quantification of virus-infected cells at 48 hpi and 96 hpi; panels C and G: quantification of cell viability using the CellTiter-Glo assay; panels D and H: quantification of apoptosis activity activity using the Caspase-Glo assay. All experiments are representative of two independent experiments with each OPC cell line using n = 3 replicates per experiment. For all Caspase and CellTiter-Glo experiments, the same number of cells were plated in each well prior to infecting, and samples were normalized based on luciferase activity of the mock condition.

Fig 3

RNAseq and cytokine analysis of ZIKV-infected SK8-A and UCSD112i-2–11 OPCs. Panels A–D: volcano plots showing differentially expressed genes from OPCs infected with heat-inactivated control ZIKV at 24 and 48 hpi (A and B) or infectious ZIKV (C and D), as compared to mock-infected OPCs. Panels E and F: gene set enrichment analysis (GSEA) of ZIKV-infected OPCs at 24 hpi and 48 hpi, respectively. Panel G: Luminex cytokine analysis of supernatants from ZIKV-infected OPCs at 24 and 48 hpi. Results are displayed as a log2-fold change relative to mock-infected cells.

Fig 4

RNAseq and cytokine analysis of HCMV-infected SK8-A and UCSD112i-2–11 OPCs. Panels A–D: volcano plots showing differentially expressed genes from OPCs infected with heat-inactivated control HCMV (A and B) and parallel infectious CMV (C and D) at 48 and 96 hpi as compared to mock-infected cells. Panels E and F: gene set enrichment analysis (GSEA) of HCMV-infected OPCs at 48 hpi and 96 hpi, respectively. Panel G: Luminex cytokine analysis of supernatants from HCMV-infected OPCs at 48 and 96 hpi. Results are displayed as a log2-fold change relative to mock-infected cells.