Human Cytomegalovirus miR-US5-2 Downregulation of GAB1 Regulates Cellular Proliferation and UL138 Expression through Modulation of Epidermal Growth Factor Receptor Signaling Pathways

Abstract

Regulation of epidermal growth factor (EGF) receptor (EGFR) signaling is critical for the replication of human cytomegalovirus (HCMV) as well as latency and reactivation in CD34⁺ hematopoietic progenitor cells. HCMV microRNAs (miRNAs) provide a means to modulate the signaling activated by EGF through targeting components of the EGFR signaling pathways. Here, we demonstrate that HCMV miR-US5-2 directly downregulates the critical EGFR adaptor protein GAB1 that mediates activation and sustained signaling through the phosphatidylinositol 3-kinase (PI3K) and MEK/extracellular signal-regulated kinase (ERK) pathways and cellular proliferation in response to EGF. Expression of HCMV UL138 is regulated by the transcription factor early growth response gene 1 (EGR1) downstream of EGFR-induced MEK/ERK signaling. We show that by targeting GAB1 and attenuating MEK/ERK signaling, miR-US5-2 indirectly regulates EGR1 and UL138 expression, which implicates the miRNA in critical regulation of HCMV latency.IMPORTANCE Human cytomegalovirus (HCMV) causes significant disease in immunocompromised individuals, including transplant patients. HCMV establishes latency in hematopoietic stem cells in the bone marrow. The mechanisms governing latency and reactivation of viral replication are complex and not fully understood. HCMV-encoded miRNAs are small regulatory RNAs that reduce protein expression. In this study, we found that the HCMV miRNA miR-US5-2 targets the epidermal growth factor receptor (EGFR) adaptor protein GAB1 which directly affects downstream cellular signaling pathways activated by EGF. Consequently, miR-US5-2 blocks the EGF-mediated proliferation of human fibroblasts. Early growth response gene 1 (EGR1) is a transcription factor activated by EGFR signaling that regulates expression of HCMV UL138. We show that miR-US5-2 regulates UL138 expression through GAB1-mediated downregulation of the signaling pathways that lead to EGR1 expression. These data suggest that miR-US5-2, through downregulation of GAB1, could play a critical role during reactivation from latency by reducing proliferation and UL138 expression.

Keywords: GAB1; UL138; cytomegalovirus; miRNA.

FIG 1

GAB1 is a target of HCMV miR-US5-2. (A) The GAB1 3′ UTR or the same 3′ UTR lacking the miR-US5-2 target site was cloned into a Dual-Luciferase reporter vector and transfected into HEK293T cells along with negative-control (Neg) miRNA or miR-US5-2 mimic. 24 h later, cells were lysed, and luciferase expression was measured. Experiments were performed in triplicate. MUT, mutant. (B to D) HEK293T cells (B), normal human dermal fibroblasts (NHDF) (C), and human aortic endothelial cells (hAEC) (D) were transfected with negative-control miRNA, miR-US5-2 mimic, or a GAB1 siRNA. 48 h later, cells were lysed and subjected to immunoblotting for GAB1 and GAPDH. GAB1 band intensity was calculated using ImageJ software and compared to GAPDH band intensity. The ratio of GAB1 band intensity to GAPDH band intensity was set to 1 for the Neg sample, and each subsequent sample ratio is presented as a multiplier of the value corresponding to the Neg time point. (E) HEK293T cells were transfected with negative-control miRNA, miR-US5-2 or a GAB1 siRNA, or expression vectors expressing the miR-US5-2 hairpin or an shRNA targeting GAB1. 48 h later, cells were lysed and immunoblotted as described for panels B to D. (F) HEK293T cells were transfected as described for panel E, and RNA was harvested 48 h later. GAB1 mRNA expression levels were determined using qRT-PCR and normalized to 18S expression levels. Experiments were performed in triplicate. Data are presented as standard errors of the means. *, P < 0.05 (as determined by two-tailed two-sample t test).

FIG 2

HCMV miR-US5-2 and downregulation of GAB1 attenuate MEK/ERK and PI3K signaling. (A) HEK293T cells were transfected with an SRE luciferase reporter construct along with negative-control miRNA, miR-US5-2 mimic, or a GAB1 siRNA. After 24 h, cells were serum starved for 4 h and treated with EGF (5 ng/ml) for an additional 4 h or left untreated. Cells were lysed, and luciferase expression was measured. Experiments were performed in triplicate, and data are presented relative to results from negative-control transfected cells treated with EGF. Data are presented as standard errors of the means. *, P < 0.05 (as determined by two-tailed two-sample t test). (B) NHDF were transfected with negative-control miRNA, miR-US5-2 mimic, or a GAB1 siRNA. After 48 h, cells were serum starved for 4 h and then stimulated with EGF (0.05 nM). Protein lysates were harvested at the indicated times and subjected to immunoblotting for phosphorylated and total MEK and AKT as well as GAB1 and GAPDH. Band intensity of p-AKT and p-MEK samples was measured using ImageJ software, and results are presented as a ratio of the band intensity of p-AKT or p-MEK to total AKT and total MEK, respectively. The ratio for the Mock time = 0 time point was set to a value of 1, and each subsequent ratio is presented as a multiplier of the reference sample.

FIG 3

HCMV miR-US5-2 and downregulation of GAB1 affect proliferation of human fibroblasts. (A) NHDF were transfected with negative-control miRNA, miR-US5-2 mimic, or a GAB1 siRNA. 48 h later, cells were trypsinized, counted, and replated at a density of 5,000 cells/well in a 24-well plate in complete media (+FBS), in media lacking FBS (-FBS), or in media lacking FBS but with the inclusion of exogenous EGF (5 ng/ml) (-FBS +EGF). After an additional 48, 96, or 144 h, total viable cells were counted (n = 2). *, P < 0.05 (as determined by two-tailed two-sample t test). (B) At 48 h posttransfection, 2 × 10⁵ cells from the experiment performed as described for panel A were lysed and subjected to immunoblotting for GAB1 and GAPDH. Band intensity was calculated using ImageJ software. The ratio of GAB1 to GAPDH band intensities was set to 1 for the Neg sample, and each subsequent sample ratio is presented as a multiplier of the Neg time point.

FIG 4

HCMV miR-US5-2 affects UL138 expression during viral infection through targeting GAB1 for downregulation. (A) HEK293T cells were transfected with an EGR-1 luciferase reporter construct along with negative-control miRNA, miR-US5-2 mimic, and GAB1 or EGR-1 siRNA. After 24 h, cells were serum starved for 4 h and treated with EGF (5 ng/ml) for an additional 4 h or left untreated. Cells were lysed, and luciferase expression was measured. Experiments were performed in triplicate, and data are presented relative to results from negative-control transfected cells treated with EGF. Data are presented as standard errors of the means. *, P < 0.05 (as determined by two-tailed two-sample t test). (B) NHDF were transfected and treated as described for panel A, with EGF (5 ng/ml) being added for 1 h. RNA was harvested, and qRT-PCR was used to analyze EGR-1 expression normalized to 18S. *, P < 0.05 (as determined by two-tailed two-sample t test). (C) NHDF were infected with the indicated viruses at a multiplicity of infection (MOI) of 3 for 2 and 4 days, after which time protein lysates were harvested and subjected to immunoblotting for GAB1, HCMV IE2, and GAPDH. Band intensities were calculated using ImageJ software. The ratio of GAB1 to GAPDH band intensities was set to 1 for the Mock 2 day postinfection (dpi) sample, and each subsequent sample ratio is presented as a multiplier of the Mock time point. The experiment was performed in duplicate. (D) Quantitation of GAB1 protein levels compared to GAPDH from 2 Western blots using ImageJ software as described for panel C. (E) NHDF were infected as described for panel C, cell supernatants were harvested at the indicated time points, and titers were determined on NHDF. (F) NHDF were infected as described for panel C, and protein lysates were harvested at the indicated times and subjected to immunoblotting for HCMV UL138, IE2, and GAPDH. Band intensities were calculated using ImageJ software. The ratio of UL138 to GAPDH band intensities was set to 1 for the WT 2 dpi sample, and each subsequent ratio is presented as a multiplier of the WT 2 dpi time point. The experiment was performed in triplicate. (G) Quantitation of UL138 expression compared to GAPDH from 3 Western blots using ImageJ software as described for panel D. (H) NHDF were transfected with negative-control miRNA, miR-US5-2 mimic, or siRNAs targeting GAB1 or EGR1 followed by infection with WT or ΔEGR1 viruses. At 48 h later, infected cells were serum starved and treated with EGF (5 ng/ml) for 2 h, after which time protein lysates were obtained and immunoblotted for UL138, IE2, and GAPDH. Band intensities were calculated using ImageJ software. The ratio of UL138 to GAPDH band intensities was set to 1 for negative-control transfected WT samples, and each subsequent ratio was presented as a multiplier of the negative-control transfected WT sample. Experiment was performed in duplicate. (I) Quantitation of UL138 expression compared to GAPDH from 2 Western blots using ImageJ software as described for panel G.

FIG 5

HCMV regulation of EGFR signaling pathways. A proposed model for the interactions of HCMV proteins and miRNAs with the EGFR signaling pathways is presented. Our data demonstrate that miR-US5-2 downregulation of GAB1 dampens MEK/ERK and PI3K signaling pathways, which ultimately affects the expression of EGR1 and UL138. We propose that interference with the EGFR-EGR1-UL138 signaling loop is a component of the switch from latent to lytic replication.