Molecular Mechanism of Porcine Epidemic Diarrhea Virus Cell Tropism

Abstract

In the 21st century, several human and swine coronaviruses (CoVs) have emerged suddenly and caused great damage to people's lives and property. The porcine epidemic diarrhea virus (PEDV), leading to enormous economic losses to the pork industry and remains a large challenge. PEDV showed extensive cell tropism, and we cannot ignore the potential risk of cross-species transmission. However, the mechanism of adaptation and cell tropism of PEDV remains largely unknown and in vitro isolation of PEDV remains a huge challenge, which seriously impedes the development of vaccines. In this study, we confirmed that the spike (S) protein determines the adaptability of PEDV to monkey Vero cells and LLC-PK1 porcine cells, and isolated exchange of S1 and S2 subunits of adaptive strains did not make PEDV adapt to cells. Further, we found that the cellular adaptability of rCH/SX/2016-S_HNXP depends on S1 and the first half of S2 (S3), and the 803L and 976H of the S2 subunit are critical for rCH/SX/2016-S1_HNXP+S3_HNXP adaptation to Vero cells. These findings highlight the decisive role of PEDV S protein in cell tropism and the potential role of coronaviruses S protein in cross-species transmissibility. Besides, our work also provides some different insight into finding PEDV receptors and developing PEDV and other coronaviruses vaccines. IMPORTANCE CoVs can spill from an animal reservoir into a naive host to cause diseases in humans or domestic animals. PEDV results in high mortality in piglets, which has caused immense economic losses in the pork industry. Virus isolation is the first step in studying viral pathogenesis and developing effective vaccines. However, the molecular mechanism of PEDV cell tropism is largely unknown, and isolation of endemic PEDV strains remains a major challenge. This study confirmed that the S gene is the decisive gene of PEDV adaptability to monkey Vero cells and porcine LLC-PK1 cells by the PEDV reverse genetics system. Isolated exchange of S1 and S2 of adaptive strains did not make PEDV adapt to cells, and the 803L and 976H of S2 subunit are critical for rCH/SX/2016-S1_HNXP+S3_HNXP adaptation to Vero cells. These results illustrate the decisive role of PEDV S protein in cell tropism and highlight the potential role of coronaviruses S protein in cross-species transmissibility. Besides, our finding also provides some unique insight into identifying PEDV functional receptors and has guiding significance for developing PEDV and other coronavirus vaccines.

Keywords: cellular tropism; coronavirus; porcine epidemic diarrhea virus; reverse genetic analysis; spike.

FIG 1

PEDV isolation. Vero cells were infected with PEDV. (A) Cells were fixed at 36 hpi, and immunofluorescence analysis was performed with anti-PEDV N (green) antibodies. (B) CPE was observed under a light microscope. (C) The viral titer was determined at 5-passage intervals.

FIG 2

Construction and rescue of infectious PEDV rCH/SX/2015 and rCH/SX/2016 from the cDNA clone. (A) Schematic illustration of constructing rCH/SX/2015 and rCH/SX/2016. (B) Vero cells were infected with CH/SX/2015, rCH/SX/2015, and rCH/SX/2016. Infected cells were fixed at 36 hpi and immunolabeled with Rhodamine (TRITC) AffiniPure Goat Anti-Mouse IgG (H+L). Nuclei were labeled with DAPI (blue). (C) WB identified the rescue of rCH/SX/2015 and rCH/SX/2016. (D) The rCH/SX/2015 RNA of the three-passage (P3) virus stock was extracted and reverse transcribed. Then the marker mutation of rCH/SX/2015 was identified by sequencing. The recombinant plasmid rCH/SX/2016 was also identified by sequencing. (E) Vero cells in 12-well plates were infected with CH/SX/2015 and rCH/SX/2015. The supernatant was harvested at 6, 12, 24, 36, and 48 hpi and titrated on Vero cells. (F) Representative images of the plaque morphologies of CH/SX/2015 and rCH/SX/2015.

FIG 3

Adaptation of rCH/SX/2016-S₂₀₁₅ to Vero cells. (A) Schematic illustration of constructing rCH/SX/2015-S₂₀₁₆ and rCH/SX/2016-S₂₀₁₅. (B) Vero cells were infected with rCH/SX/2015, rCH/SX/2015-S₂₀₁₆, and rCH/SX/2016-S₂₀₁₅. Infected cells were fixed at 36 hpi and immunolabeled with Rhodamine (TRITC) AffiniPure Goat Anti-Mouse IgG (H+L). Nuclei were labeled with DAPI (blue). (C) WB identified the rescue of rCH/SX/2015, rCH/SX/2015-S₂₀₁₆, and rCH/SX/2016-S₂₀₁₅. (D) Vero cells in 12-well plates were infected with rCH/SX/2015 and rCH/SX/2016-S₂₀₁₅. The supernatant was harvested at 6, 12, 24, 36, and 48 hpi and titrated on Vero cells. (E) Representative images of the plaque morphologies of rCH/SX/2015 and rCH/SX/2016-S₂₀₁₅. Error bars indicate standard deviations. The level of significance was expressed as *, P < 0.05, **, P < 0.01 or ***, P < 0.001.

FIG 4

Adaptation of rCH/SX/2015-S_HNXP to Vero and LLC-PK1 cells. (A) Schematic illustration of constructing rCH/SX/2015-S_HNXP. (B) Vero cells were infected with rCH/SX/2015 and rCH/SX/2015-S_HNXP. Infected cells were fixed at 36 h postinfection and immunolabeled with Fluorescein (FITC)-AffiniPure Goat Anti-Mouse IgG (H+L). Nuclei were labeled with DAPI (blue). (C) WB identified the rescue of rCH/SX/2015 and rCH/SX/2015-S_HNXP. (D) Vero cells in 12-well plates were infected with rCH/SX/2015, HNXP, and rCH/SX/2015-S_HNXP. The supernatant was harvested at 6, 12, 24, 36, and 48 hpi and titrated on Vero cells. (E) Representative images of the plaque morphologies of mock, rCH/SX/2015, HNXP, and rCH/SX/2015-S_HNXP. (F) LLC-PK1 cells were infected with rCH/SX/2015, HNXP, and rCH/SX/2015-S_HNXP. Infected cells were fixed at 36 hpi and immunolabeled with Fluorescein (FITC)-AffiniPure Goat Anti-Mouse IgG (H+L). Nuclei were labeled with DAPI (blue). Error bars indicate standard deviations. The level of significance was expressed as *, P < 0.05, **, P < 0.01 or ***, P < 0.001.

FIG 5

Schematic illustration of construction of rCH/SX/2015-S1₂₀₁₆, rCH/SX/2015-S2₂₀₁₆ and rCH/SX/2016-S1₂₀₁₅, rCH/SX/2016-S2₂₀₁₅. S1:1-729aa, S2:730-1387aa.

FIG 6

Adaptation of rCH/SX/2016-S1_HNXP+S3_HNXP to Vero cells. (A) Schematic illustration of the construction of rCH/SX/2016-S_HNXP, rCH/SX/2016-S1_HNXP, rCH/SX/2016-S2_HNXP, rCH/SX/2016-S1_HNXP+S3_HNXP, and rCH/SX/2016-S1_HNXP+S4_HNXP. S:1-1387aa, S1:1-729aa, S2:730-1387aa, S1_HNXP+S3_HNXP:1-976aa, S1_HNXP+S4_HNXP:1-729aa and 977-1387aa. (B) Vero cells were infected with rCH/SX/2016-S_HNXP, rCH/SX/2016-S1_HNXP, rCH/SX/2016-S2_HNXP, rCH/SX/2016-S1_HNXP+S3_HNXP, and rCH/SX/2016-S1_HNXP+S4_HNXP. Cells were harvested at 24 and 48 hpi. Then RT-qPCR was performed to detect the relative expression of mRNA of PEDV N protein (C). Vero cells were infected with rCH/SX/2016-S_HNXP and rCH/SX/2016-S1_HNXP+S3_HNXP. Infected cells were fixed at 36 h postinfection and immunolabeled with Fluorescein (FITC)-AffiniPure Goat Anti-Mouse IgG (H+L). Nuclei were labeled with DAPI (blue).

FIG 7

Growth characteristics of rCH/SX/2016-S1_HNXP+S3_HNXP. (A) Vero cells in 12-well plates were infected with rCH/SX/2016-S1_HNXP+S3_HNXP and rCH/SX/2016-S_HNXP. The supernatant was harvested at 24, 36, 48, 60, and 72 hpi and titrated on Vero cells. (B) MARC145 cells were infected with rCH/SX/2016-S_HNXP and rCH/SX/2016-S1_HNXP+S3_HNXP at an MOI of 0.01. Cells were harvested at 12, 24, 36, and 48 hpi. Then RT-qPCR was performed to detect the relative expression of mRNA of PEDV N protein. (C) Representative images of the plaque morphologies of rCH/SX/2016-S_HNXP and rCH/SX/2016-S1_HNXP+S3_HNXP. Error bars indicate standard deviations. The level of significance was expressed as *, P < 0.05, **, P < 0.01 or ***, P < 0.001.

FIG 8

Adaptation of rCH/SX/2016-S1_HNXP+S3_P803L, rCH/SX/2016-S1_HNXP+S3_Y976H, and rCH/SX/2016-S1_HNXP+S3_P803L+Y976H chimeric strains to Vero cells. (A) Schematic illustration of constructing rCH/SX/2016-S1_HNXP+S3_P803L, rCH/SX/2016-S1_HNXP+S3_Y976H and rCH/SX/2016-S1_HNXP+S3_P803L+Y976H. (B) Schematic illustration of the structure of the PEDV S gene. SP: signal peptide, FP: the fusion peptide, HR1: heptad repeat 1, HR2: heptad repeat 2, TM: transmembrane domain. (C) Sequence analysis of rCH/SX/2016-S1_HNXP+S3_P803L, rCH/SX/2016-S1_HNXP+S3_Y976H, and rCH/SX/2016-S1_HNXP+S3_P803L+Y976H plasmids. (D) Vero cells were infected with rCH/SX/2016-S_HNXP, rCH/SX/2016-S1_HNXP, rCH/SX/2016-S1_HNXP+S3_HNXP, rCH/SX/2016-S1_HNXP+S3_P803L, rCH/SX/2016-S1_HNXP+S3_Y976H, and rCH/SX/2016-S1_HNXP+S3_P803L+Y976H. Cells were harvested at 24 hpi. Then RT-qPCR was performed to detect the relative expression of mRNA of PEDV N protein.

FIG 9

Schematic illustration of the molecular mechanism of PEDV cell tropism. The S genes of cell-adapted PEDV strains impart the cellular adaptation of cell-unadapted PEDV strains. The S genes of cell-unadapted PEDV strains abolish the cellular adaptation of cell-adapted PEDV strains. The cellular adaptability of PEDV depends on the cooperation between S1 and S2 (or the first half of S2). We speculate that the first half of S2 (S3) might involve the binding of S1 to receptors or in membrane adsorption mediated by the FP, in the process of which the 803L and 976H perhaps played an important role. FP: the fusion peptide, HR1: heptad repeat 1, HR2: heptad repeat 2, 6-HB: six-helix bundle.