Murine cytomegalovirus (CMV) M33 and human CMV US28 receptors exhibit similar constitutive signaling activities

Abstract

Cellular infection by cytomegalovirus (CMV) is associated with very early G-protein-mediated signal transduction and reprogramming of gene expression. Here we investigated the involvement of human CMV (HCMV)-encoded US27, US28, and UL33 receptors as well as murine CMV-encoded M33 transmembrane (7TM) receptors in host cell signaling mechanisms. HCMV-encoded US27 did not show any constitutive activity in any of the studied signaling pathways; in contrast, US28 and M33 displayed ligand-independent, constitutive signaling through the G protein q (Gq)/phospholipase C pathway. In addition, M33 and US28 also activated the transcription factor NF-kappaB as well as the cyclic AMP response element binding protein (CREB) in a ligand-independent, constitutive manner. The use of specific inhibitors indicated that the p38 mitogen-activated protein (MAP) kinase but not the extracellular signal-regulated kinase 1/2-MAP kinase pathway is involved in M33- and US28-mediated CREB activation but not NF-kappaB activation. Interestingly, UL33-the HCMV-encoded structural homologue of M33-was only marginally constitutively active in the Gq/phospholipase C turnover and CREB activation assays and did not show any constitutive activity in the NF-kappaB pathway, where M33 and US28 were highly active. Hence, CMVs appear to have conserved mechanisms for regulating host gene transcription, i.e., constitutive activation of certain kinases and transcription factors through the constitutive activities of 7TM proteins. These data, together with the previous identification of the incorporation of such proteins in the viral envelope, suggest that these proteins could be involved in the very early reprogramming of the host cell during viral infection.

FIG. 1.

Dendrogram of CMV-encoded chemokine receptors based on their amino acid identities. This phylogenetic tree was generated by Francois Talabot, Serono Pharmacological Research Institute, by using Clustal W 1.5 alignments of the full amino acid sequences followed by analysis with the Distance program (Genetics Computer Group, Madison, Wis.). The length of each branch reflects the identity between the receptors. Tree View (http://taxonomy.zoology.gla.ac.uk/rod/rod.html) was used for graphic presentation. Asterisks indicate members of the human CMV receptor family.

FIG. 2.

Inositol phosphate (PI) accumulation induced by US27, US28, UL33, and M33 receptor activities. Gene dose experiments with increasing concentrations of US27 (white bar), US28 (black bar), UL33 (grey bar), and M33 (hatched bar) receptor DNAs were performed with transiently transfected COS-7 cells. One hundred percent corresponds to the basal activity of cells transfected with 20 μg of the empty expression vector pTEJ8. Data are means and standard errors of the means for three experiments carried out in quadruplicate.

FIG. 3.

Cellular localization of US28, US27, UL33, and M33 in COS-7 cells. COS-7 cells were transiently transfected with US28-EGFP (A), US27-YFP (B), UL33-EGFP (C), or M33-EGFP (D). US28-EGFP (A) and US27-YFP (B) were mostly located on intracellular vesicles in the perinuclear regions of the cells. UL33-EGFP (C) was distributed both intracellulary and on the cell surface. M33-EGFP (D) was found predominantly on the cell membrane surface. Shown are representative fluorescence microscopy images of the distribution of the GFP- and YFP-tagged receptors taken with a ×100 oil immersion objective. Bar, 10 μm

FIG. 4.

Gene-dose-dependent induction of NF-κB activity by US28, US27, UL33, and M33. Induction of NF-κB-luciferase was determined in transiently transfected COS-7 cells (35,000 cells/well). Cells were cotransfected with 50 ng of NF-κB-luciferase vector together with increasing amounts of either pTEJ8 vector DNA (white bars) as a control or US28 (A), US27 (B), UL33 (C), or M33 (D) DNA (black bars). Shown are representative results from at least four independent experiments performed in quadruplicate. Data are means and standard errors of the means. RLU were measured with a Packard TopCounter (5 s/well).

FIG. 5.

Induction of CREB activity by US28, US27, UL33, and M33. Activation of CREB-luciferase was determined in transiently transfected COS-7 cells (35,000 cells/well). Cells were cotransfected with a CREB-luciferase vector mixture (for details, see Materials and Methods) together with increasing amounts of either vector DNA (white bars) as a control or US28 (A), US27 (B), UL33 (C), or M33 (D) DNA (black bars). Shown are representative results from at least four independent experiments performed in quadruplicate. Data are means and standard errors of the means. RLU were measured with a Packard TopCounter (5 s/well).

FIG. 6.

Modulation of US28-induced inositol phosphate (PI) turnover and CREB activity by various CC chemokines and CX₃CL. (A) Effects of a 100 nM concentration of the chemokine domain of fractalkine/CX₃CL1 (black bar) and RANTES/CCL5, MIP-1α/CCL3, MIP-1β/CCL4, MCP-1/CCL2, MCP-3/CCL7, and vMIP-II/vCCL2 (grey bars) on the basal inositol phosphate turnover in COS-7 cells transfected with US28. The asterisk indicates a statistically significant difference (P < 0.05; unpaired t test) from the maximally obtained basal activity (control, white bar). Experiments were carried out in duplicate (n = 5). (B) Effect of RANTES/CCL5 (filled squares) or the chemokine domain of fractalkine/CX₃CL (open triangles) on the basal inositol phosphate turnover (filled circles) in COS-7 cells transiently transfected with the indicated amounts of US28 receptor DNA. The basal activity of 30 μg of US27 is indicated by an open circle. One hundred percent corresponds to the maximally obtained basal activity, and 0% corresponds to the background activity obtained with 30 μg of empty expression vector pTEJ8. (C) COS-7 cells were transiently transfected with increasing amounts of US28 receptor DNA and 50 ng of CREB-luciferase cDNA. The cells were exposed to either 100 nM RANTES/CCL5 (filled squares) or 100 nM fractalkine/CX₃CL1 (open triangles) immediately following transfection. Basal constitutive activity of US28 is represented by filled circles, and the activity in mock-transfected cells is represented by open circles. Shown are representative results from at least four independent experiments performed in quadruplicate. Data in all three panels are means and standard errors of the means.

FIG. 7.

Effects of selected MAP kinase inhibitors and the Gαi inhibitor PTX on receptor-mediated constitutive activation of CREB and NF-κB. COS-7 cells were transiently transfected with 40 ng of US28 (A and D), M33 (B and E), or UL33 (C) receptor DNA and 50 ng of either CREB-luciferase (A, B, and C) or NF-κB-luciferase (D and E) reporter DNA. Immediately following transfection, cells were incubated with a 1 μM concentration of the selective p38 MAP kinase inhibitor SB202190 (grey bars) or a 30 μM concentration of the ERK1/2-MAP kinase inhibitor PD98059 (black bars). Likewise, transiently transfected COS-7 cells were treated with PTX (100 ng/ml) for 24 h (hatched bars). Data are shown as percentages of the data for the control (100%), whereby the control represents the maximal constitutive activation of the respective reporter in the presence of 40 ng of the respective receptor. The asterisk indicates a statistically significant difference (P < 0.05; unpaired t test) from the maximally obtained basal activity (control, white bar). Data are means and standard errors of the means for at least four independent experiments performed in quadruplicate.