Functional analysis of human cytomegalovirus pUS28 mutants in infected cells

Abstract

The human cytomegalovirus (HCMV)-encoded viral G protein-coupled receptor pUS28 contributes to an array of biological effects, including cell migration and proliferation. Using FIX-BAC (bacterial artificial chromosome, derived from the HCMV clinical isolate VR1814) and lambda red recombination techniques, we generated HCMV recombinants expressing amino-terminally FLAG-tagged versions of wild-type pUS28 (FLAG-US28/WT), G-protein coupling deficient pUS28 (FLAG-US28/R129A) and chemokine-binding domain deficient pUS28 (FLAG-US28/DeltaN). Infection with the FLAG-US28/R129A virus failed to induce inositol phosphate accumulation, indicating that G-protein coupling is essential for pUS28 signalling to phospholipase C-beta (PLC-beta) during HCMV infection. The FLAG-US28/DeltaN virus induced about 80 % of the level of PLC-beta signalling induced by the FLAG-US28/WT virus, demonstrating that the N-terminal chemokine-binding domain is not required for pUS28-induced PLC-beta signalling in infected cells. The data presented here are the first to describe the functional analyses of several key pUS28 mutants in HCMV-infected cells. Elucidating the mechanisms by which pUS28 signals during infection will provide important insights into HCMV pathogenesis.

Fig. 1

Deletion and modification of US28 in the HCMV genome. A schematic representation of strategies used to manipulate US28 in the FIX-BAC genome is illustrated. Lambda phage red recombination was used to delete the US28 gene (ΔUS28, upper panel), introduce a FLAG-tagged US28 gene deleted for the chemokine binding domain located between amino acids 2 and 16 (FLAG–US28/ΔN, middle panel), introduce a FLAG-tagged wild-type US28 gene (FLAG–US28/WT, lower panel), or introduce a FLAG-tagged US28 gene containing a R129A mutation in the G-protein coupling motif (FLAG–US28/R129A, lower panel). The Kan or Kan/LacZ selectable markers were removed from the final constructs with lambda phage red or Flp recombinases as indicated.

Fig. 2

PCR analyses of HCMV US28 recombinants. FIX-BAC DNAs isolated from E. coli were used as templates in all PCR reactions. Primers with homology to 5′and 3′sequences flanking the US28 coding region were used to amplify the US28 locus (upper panel) by PCR. The primer with homology to the 5′ flanking sequence was used in combination with a FLAG-specific primer to verify the addition of an N-terminal FLAG epitope (middle panel). Amplification of the UL146 gene serves as a positive PCR control (lower panel).

Fig. 3

Kinetics of pUS28 expression in HCMV FLAG–US28/WT-infected cells. The pFLAG–US28 protein was immunoprecipitated from HCMV FLAG–US28/WT-infected HFFs (using anti-FLAG M2 agarose beads) and analysed by Western blotting using an α-FLAG polyclonal antibody. Overexposure of the blot enables pFLAG–US28 expression to be detected at 6 h post-infection (upper panels). Whole-cell lysates from the same time points were separated by SDS-PAGE and analysed by Western blotting using an α-IE1/IE2 antibody (lower middle panel) or an α-UL44 antibody (lower panel). Results shown are representative of three independent experiments.

Fig. 4

CCL5/RANTES binding to pFLAG–US28/R129A and pFLAG–US28/ΔN in infected cells. (a) Immunoprecipitation followed by Western blotting was performed as described in Fig. 3 to detect expression of the various forms of pFLAG–US28 encoded by HCMV parent, HCMV FLAG–US28/WT, HCMV ΔUS28, HCMV FLAG–US28/R129A and HCMV FLAG–US28/ΔN (upper panel). Whole-cell lysates from the same samples were subjected to Western blotting using an α-IE1/IE2 antibody (lower panel). Results shown are representative of three independent experiments. (b) HFFs infected as described in (a) were incubated with 28 pM [¹²⁵I]CCL5/RANTES in the absence or presence of 14 nM unlabelled RANTES to discriminate between specific and non-specific binding. The data shown represent specific binding of [¹²⁵I]CCL5/RANTES as assessed by liquid scintillation chromatography and are derived from three independent experiments performed in duplicate. (c) HFFs were infected with HCMV ΔUS28, HCMV FLAG–US28/WT, HCMV FLAG–US28/R129A or HCMV FLAG–US28/ΔN for 48 h. Surface expression of pFLAG–US28 on infected cells was detected by staining with FLAG-specific M2-biotin, followed by streptavidin-PE and analysed by FACS. The histograms shown are representative of at least four independent experiments performed in duplicate.

Fig. 5

Characterization of PLC-βsignalling in cells infected with HCMV FLAG–US28/WT or HCMV ΔUS28. HFFs were infected with increasing m.o.i. (0.03 to 3.0 p.f.u. per cell) using HCMV parent or HCMV FLAG–US28/WT (a), and HCMV FLAG–US28/WT or HCMV ΔUS28 (b). Medium containing 1 μCi ml⁻¹ (37 kBq ml⁻¹) [³H]myoinositol was added at 24 h post-infection and accumulated InsPs were determined at 48 h post-infection using anion-exchange chromatography. The data represent four independent experiments performed in duplicate and are presented as a ratio of accumulated inositol phosphates to total ³H incorporation.

Fig. 6

pUS28 requires the G-protein coupling motif, but not the chemokine binding domain for signalling through the PLC-β pathway in HCMV-infected cells. HFFs were infected with increasing m.o.i. (0.03 to 3.0 p.f.u. per cell) using HCMV FLAG–US28/WT or HCMV FLAG–US28/R129A (a), and HCMV FLAG–US28/WT or HCMV FLAG–US28/ΔN (b). Medium containing 1 μCi ml⁻¹ [³H]myoinositol was added at 24 h post-infection and accumulated InsPs were determined at 48 h post-infection using anion-exchange chromatography. The data represent at least four independent experiments performed in duplicate and are presented as a ratio of accumulated inositol phosphates to total ³H incorporation.