Hippo signaling pathway regulates Ebola virus transcription and egress

Abstract

Filovirus-host interactions play important roles in all stages of the virus lifecycle. Here, we identify LATS1/2 kinases and YAP, key components of the Hippo pathway, as critical regulators of EBOV transcription and egress. Specifically, we find that when YAP is phosphorylated by LATS1/2, it localizes to the cytoplasm (Hippo "ON") where it sequesters VP40 to prevent egress. In contrast, when the Hippo pathway is "OFF", unphosphorylated YAP translocates to the nucleus where it transcriptionally activates host genes and promotes viral egress. Our data reveal that LATS2 indirectly modulates filoviral VP40-mediated egress through phosphorylation of AMOTp130, a positive regulator of viral egress, but more surprisingly that LATS1/2 kinases directly modulate EBOV transcription by phosphorylating VP30, an essential regulator of viral transcription. In sum, our findings highlight the potential to exploit the Hippo pathway/filovirus axis for the development of host-oriented countermeasures targeting EBOV and related filoviruses.

Fig. 1. Egress of EBOV is controlled by the Hippo pathway effector YAP/TAZ.

a YAP differentially regulates EBOV VP40 VLP egress in a phosphorylation-dependent manner. eVP40 VLP budding assay at 24 h.p.t. in HEK 293 cells expressing eVP40 alone, or with exogenous flag-tagged WT-YAP, or YAP mutants S127A or S127D. b eVP40 VLP budding assay at 24 h.p.t. in WT and Y/T-dKO cells. c Representative confocal images of WT and Y/T-dKO cells showing eVP40 (green) and endogenous YAP (magenta). Scale bar = 10 µm. Three times experiments were repeated independently with similar results. d eVP40 VLP budding assay at 24 h.p.t. in WT or Y/T-dKO cells, or Y/T-dKO cells expressing exogenous WT-YAP, or YAP mutants S127A or S127D. e Working model showing Hippo “ON” vs. Hippo “OFF” conditions and differential regulation of VP40 VLP egress by YAP/TAZ. f Schematic diagram showing the EBOV infection protocol in WT and Y/T-dKO cells. g Representative images and quantification of EBOV-GFP (MOI = 0.05) infected WT and Y/T-dKO cells at 24- and 48- h.p.i. The infection efficiency was determined as a ratio of GFP-positive (infected) cells to total cell nuclei. h Representative images and quantification (data were normalized to the infection efficiency in corresponding HEK 293 cells) of Vero cells infected with the virus isolated from the supernatants of WT and Y/T-dKO cells in panel (g). Scale bar = 200 µm. Data in (a, b) and (d) are presented as mean ± SD (n = 3). Data in (g, h) are presented as mean ± SD (n = 8). Statistical analyses were performed by one-way ANOVA in (a, d), or by two-tailed unpaired t-test in (b, g, h). Source data are provided as a Source Data file.

Fig. 2. YAP/TAZ transcriptional activity is required for EBOV infectivity and egress.

a–c Gene Ontology (GO) enrichment analysis of significantly downregulated DEGs in Y/T-dKO cells clustered in three sub-ontologies as Cellular Component, CC (a), Biological Process, BP (b), and Molecular Function, MF (c), the top15 enriched GO terms in each sub-ontologies are shown based on P values and gene counts. P values are based on a two-sided test. d Heatmap showing the top 20 significantly downregulated DEGs in Y/T-dKO cells clustered in the GO term of Cell projection organization, in descending order by fold change. e Representative confocal images of WT and Y/T-dKO cells showing eVP40 (green), endogenous F-actin (Phalloidin, red), and endogenous YAP (magenta). Scale bar = 5 µM. Three times experiments were repeated independently with similar results. f Heatmap showing significantly downregulated DEGs in Y/T-dKO cells clustered in the GO term of Virus entry into the host cell, in descending order by fold change. g Representative confocal images showing internalized EBOV VP40 + GP VLPs (green) in WT and Y/T-dKO cells. Scale bar = 10 µM. Three times experiments were repeated independently with similar results. h Quantification of entry efficiency of VSV pseudotypes encoding the luciferase reporter gene in WT and Y/T-dKO cells. Data in (h) were presented as mean ± SD (n = 6). Statistical analyses were performed by a two-tailed unpaired t-test. Source data are provided as a Source Data file.

Fig. 3. Modulation of YAP activity regulates VP40-mediated VLP and live EBOV egress.

a, b eVP40 VLP budding assay at 24 h.p.t. in Huh7 cells that were mock-treated, treated with 5 or 20 ng/ml EGF (a), or treated with 1 or 5 µM Verteporfin (b). c, d mVP40 VLP budding assay at 24 h.p.t. in Huh7 cells that were mock-treated, treated with 5 or 20 ng/ml EGF (c), or treated with 1 or 2.5 µM Verteporfin (d). e Representative confocal images showing eVP40 (green) and endogenous YAP (magenta) in Huh7 cells that were mock-treated, treated with 20 ng/ml EGF, or treated with 5 µM Verteporfin. Scale bar = 5 µm. Three times experiments were repeated independently with similar results. f Heatmap showing the top 20 significantly downregulated DEGs in Y/T-dKO cells clustered in the GO term of Growth factor binding and activity, in descending order by fold change. g Schematic diagram showing the EBOV infection and treatment protocol in MDMs. h, i Quantification of virus infectivity and production in EBOV-GFP infected (MOI = 0.2) human MDMs pre-treated (h), or post-treated (i) with Verteporfin at 24 h.p.i. Virus production from MDMs was titrated on Vero cells and was normalized to the infection efficiency in corresponding MDMs. Data in (a–d) and (h, i) are presented as mean ± SD (n = 3). Statistical analyses were performed by one-way ANOVA in (a–d) or two-way ANOVA in (h, i). Source data are provided as a Source Data file.

Fig. 4. LATS2, but not LATS1, modulates AMOTp130 to inhibit EBOV and MARV VP40 VLP egress.

a, b eVP40 (a) and mVP40 (b) VLP budding assay at 24 h.p.t. in HEK 293 cells expressing VP40 alone, or with exogenous myc-tagged LATS1 or LATS2. c Schematic diagram showing the LATS kinase-mediated key phosphorylation sites within the consensus motifs HXRXXS of AMOTp130 and YAP. d Co-IP assay with myc-tagged LATS1 or LATS2 and endogenous AMOTp130 and YAP in HEK 293 cells. e Representative confocal images of HEK 293 cells expressing eVP40 (green), AMOTp130 (red), and myc-tagged LATS1 or LATS2 (magenta). Scale bar = 10 µm. f eVP40 VLP budding assay at 24 h.p.t. in HEK 293 cells expressing eVP40 alone, or with constant amounts of exogenous LATS2, and increasing amounts of AMOTp130. g eVP40 VLP budding assay at 24 h.p.t. in HEK 293 cells expressing eVP40 alone, or with exogenous WT LATS2 or LATS2 mutants T1041A or T1041E. h Representative confocal images of shAMOT HEK 293 cells expressing eVP40 (green), and exogenous WT AMOTp130, or AMOTp130 mutants S175A or S175D (red), and endogenous F-actin (Phalloidin, cyan). Scare bar = 10 µm. Three times experiments (d–h) were repeated independently with similar results. i, j eVP40 (i) and mVP40 (j) VLP budding assay and quantification in WT, LATS1-KO, or LATS2-KO cells. Data in (a, b) and (i, j) are presented as mean ± SD (n = 3). Statistical analyses were performed by one-way ANOVA. Source data are provided as a Source Data file.

Fig. 5. LATS1 and LATS2 modulate EBOV infectivity and egress.

a Schematic diagram of the live EBOV-GFP infection protocol. b, c Representative images and quantification of EBOV-GFP (MOI = 0.05) infected WT, LATS1-KO, LATS2-KO, and LATS1/2-dKO cells at 24- (b) and 48- (c) h.p.i. The infection efficiency was determined by a ratio of GFP-positive cells to total cell nuclei. d, e Representative images and quantification (data was normalized to the infection efficiency in b and c, respectively) of Vero cells infected with the supernatants isolated from WT, LATS1-KO, LATS2-KO, and LATS1/2-dKO cells in panels (b) and (c). Scale bar = 200 µm. f Schematic diagram of the EBOV minigenome assay. g–i EBOV minigenome transcription determined by quantification of eGFP reporter (g) or luciferase reporter (h) in WT, LATS1-KO, LATS2-KO, and LATS1/2-dKO cells. Detection of the indicated viral RNP proteins in HEK 293 cells expressing the EBOV minigenome system (i). Three times experiments were repeated independently with similar results. Data in (b–e, n = 7) and (h, n = 10) are presented as mean ± SD. Statistical analyses were performed by one-way ANOVA. Source data are provided as a Source Data file.

Fig. 6. LATS1/2 kinases phosphorylate Ser29 of eVP30 to regulate EBOV transcription.

a LATS kinase targets the Serine within the consensus HXRXXS which is present in EBOV VP30. b Western blot to detect and quantify eVP30 Ser29 phosphorylation levels in WT, LATS1-KO, LATS2-KO, and LATS1/2-dKO cells. Ser29 phosphorylation intensity was normalized to the total level of eVP30. c RT-qPCR to detect and quantify EBOV GP genome equivalent (GE) kinetics in 1, 5, 10 h.p.i. windows in WT and LATS1/2-dKO cells infected with EBOV-GFP (MOI = 0.01). d, e EBOV minigenome assay (d) and Western blot to detect eVP30 Ser29 phosphorylation (e) in WT and LATS1/2-dKO cells, or LATS1/2-dKO cells expressing exogenous LATS1 or LATS2. f Quantification of EBOV-GFP (MOI = 0.05) infected WT, LATS1/2-dKO, and LATS1/2-dKO cells trans-complemented with exogenous LATS1 or LATS2 at 48 h.p.i. g–h Co-IP assay with eNP and exogenous myc-tagged LATS1 or LATS2 (g), or endogenous LATS (h) in HEK 293 cells. i Co-IP assay showing the association of eNP and eVP30 with endogenous LATS in HEK 293 cells. j, k Representative confocal images of Huh7 cells expressing myc-tagged LATS1/2 (green) and flag-tagged eVP30 (magenta) in the absence (j) or presence (k) of eNP (red). Insert panels in (k) showing the colocalization of LATS1/2, eVP30, and eNP in IBs. Scale bar = 10 µm in (j) and 5 µm in (k). Three times experiments (g–k) were repeated independently with similar results. l Detection of eVP30 Ser29 phosphorylation and the EBOV minigenome assay in HEK 293 cells that were mock-treated or treated with 2.5 or 5 µM LATS kinase inhibitor TDI-011536. m Detection of eVP30 Ser29 phosphorylation and the EBOV minigenome assay in HEK 293 cells that mock-treated, treated with 5 µM TDI-011536, treated with 5 µM SRPK kinase inhibitor SPHINX31, or in combination. n Quantification of virus production from EBOV-GFP (MOI = 0.2) infected MDMs pre-treated or post-treated with TDI-011536 at 24 h.p.i. Data in (b, c, n = 4), (d, l, n = 8), (f, m, n = 9), and (n, n = 3) are presented as mean ± SD. Statistical analyses were performed by one-way ANOVA or two-way ANOVA in (c). Source data are provided as a Source Data file.

Fig. 7. Loss of YAP/TAZ and LATS1/2 kinases impairs EBOV propagation in 3D spheroids.

a Schematic diagram showing the generation of 3D spheroids and the EBOV infection protocol. b Representative images of mock and EBOV-GFP (10³ and 10⁴ I.U.) infected 3D spheroids of WT, YAP/TAZ-dKO, and LATS1/2-dKO cells at 48 h.p.i. Scale bar = 300 µm. c, d RT-qPCR to detect and quantify viral genome equivalent (GE) from the spheroids infected with 10³ (c) and 10⁴ (d) I.U. of EBOV-GFP. Data were presented as mean ± SD (n = 8). Statistical analyses were performed by one-way ANOVA. Source data are provided as a Source Data file.

Fig. 8. Working model showing the intersection of the Hippo pathway and multiple stages of the EBOV lifecycle.

a Target genes of YAP/TAZ regulate the composition and function of the ECM/PM. b LATS1/2 kinases interact with NP and phosphorylate VP30 to regulate EBOV transcription. c Involvement of LATS1/2, YAP/TAZ, as well as AMOT in regulating EBOV/MARV egress.