Transcriptomic analysis of DENV-2-infected human dermal fibroblasts identified potential mechanisms that suppressed ZIKV replication during sequential coinfection

Abstract

Dengue virus (DENV) and Zika virus (ZIKV) are closely related flaviviruses which are transmitted by the same species of mosquitoes. Due to overlapping geographic distributions and transmission vectors, cases of DENV-ZIKV coinfection have been reported. However, the impact of coinfection on disease outcomes remains unclear. In this study, an in vitro model of DENV-ZIKV coinfection was developed using the primary human dermal fibroblasts (HDFs). The interaction between DENV-2 and ZIKV during sequential coinfection revealed that prior DENV-2 infection significantly suppressed ZIKV RNA accumulation in the culture supernatant. Transcriptomic profile in response to DENV-2 infection suggested three hypothetical pathways that potentially interfere with ZIKV replication. The first mechanism is prior DENV infection drove HDFs into an antiviral state through upregulation of genes involving innate immune response pathways, including PRR signaling, type I and type II IFN signaling, ISG activity, and cytokine/chemokine activity. This state significantly enhanced resistance to subsequent ZIKV infection in both infected cells and uninfected neighboring cells. The second potential pathway is inhibition of viral entry. This was supported by DENV-2-infected HDFs significantly suppressed expression of ZIKV receptor and reduced expression of genes involving in clathrin-mediated endocytosis. This can interfere with entry of ZIKV into host cells. The last possible mechanism is driving cells into cell cycle arrest, as DENV-2 infection downregulated genes related to cell cycle progression, which may hinder ZIKV replication. These findings partly unfold the interplay between DENV and ZIKV at the entry site which may explain the disease outcome of DENV-ZIKV coinfection.

Keywords: Coinfection; Dengue virus; Human dermal fibroblasts; Superinfection exclusion; Transcriptomic profiling; Zika virus.

Fig. 1

Viral RNA accumulation during DENV-2-ZIKV coinfection in HDF cells. HDF cells were coinfected with DENV-2 (MOI of 5) and ZIKV (MOI of 10). The RNA levels of DENV-2 and ZIKV were measured by qRT-PCR at 24, 48, and 72 hpi. Compared to the monoinfected cells, ZIKV RNA levels were significantly reduced (A), while DENV-2 RNA levels in coinfected cells were not significantly different from those in cells infected with DENV-2 alone (B). Data were presented as mean ± SEM from three independent experiments and analyzed using Student’s t-test (p < 0.05). * indicated significant differences between monoinfection and simultaneous/sequential coinfection, while ** indicated significant differences between simultaneous and sequential coinfections

Fig. 2

Global transcriptome and DEGs of HDFs in response to DENV-2 infection. (A) PCA of HDFs in response to DENV-2 or mock infection. (B and C) To identify DEGs, data were analyzed using DESeq2 in the iDEP web application by comparing DENV-2-infected samples with mock controls at 16 and 24 hpi. Significant DEGs were selected with an FDR-adjusted p-value ≤ 0.05 and │log2FoldChange│≥ 1. (B) The DEGs were expressed as Volcano plots. Red dots represented significant DEGs, gray dots represented non-DEGs, green dots represented genes with │log2FoldChange│≥ 1 but p-value > 0.05, and blue dots represented genes with p-value ≤ 0.05 but │log2FoldChange│< 1. (C) The identified DEGs were revealed as heatmap between DENV-2- or mock-infected HDFs

Fig. 3

Functional enrichment analysis of total DEGs in DENV-2-infected HDFs at 24 hpi. Total DEGs at 24 hpi, encompassing both upregulated and downregulated DEGs, was subjected to Gene Ontology (GO) enrichment analysis. The top 10 enriched pathways were displayed for biological processes (A), cellular components (B), and molecular functions (C). Colored dots represented distinct significant enrichment (p ≤ 0.05). Dot size represented the number of genes

Fig. 4

Clusters of upregulated DEGs potentially implicated in viral interference. The significantly upregulated DEGs at 24 hpi, selected based on an FDR-adjusted p-value ≤ 0.05 and log2FoldChange ≥ 1, were input into the STRING database to generate PPI networks and visualized in Cytoscape. Highly interconnected clusters were identified with MCODE, including module 1 (A), module 2 (B), and module 3 (C)

Fig. 5

Clusters of downregulated DEGs potentially involved in viral interference. The significantly downregulated DEGs at 24 hpi, selected based on an FDR-adjusted p-value ≤ 0.05 and log2FoldChange ≤ -1, were input into the STRING database to generate PPI networks and visualized in Cytoscape. Highly interconnected clusters were identified with MCODE, including module 1 (A), module 2 (B), and module 3 (C)

Fig. 6

Validation of gene expression levels and biological activities in DENV-2-infected HDFs. Validation of gene expression levels using qRT-PCR. Total RNAs were extracted from DENV-2-infected cells and reverse-transcribed into cDNAs. The expression levels of upregulated transcripts (A) and downregulated transcripts (B) were validated using qRT-PCR, using human GAPDH as the reference gene. Data were presented as mean ± SEM from three independent experiments. (C) HDFs were treated with UV-inactivated DENV-2 HDF supernatant for 24 h, followed by ZIKV infection at an MOI of 5 for 24 h. ZIKV infectivity was quantified using flow cytometry. (D-F) Production levels of IFN-β, IFN-γ, and CXCL10 in DENV-2-infected HDF supernatant were quantitated using ELISA. Data were presented as mean ± SEM from three independent experiments and analyzed using Student’s t-test (p < 0.05)