The Epstein-Barr virus-encoded G protein-coupled receptor BILF1 hetero-oligomerizes with human CXCR4, scavenges Gαi proteins, and constitutively impairs CXCR4 functioning

Abstract

Cells express distinct G protein-coupled receptor (GPCR) subtypes on their surface, allowing them to react to a corresponding variety of extracellular stimuli. Cross-regulation between different ligand-GPCR pairs is essential to generate appropriate physiological responses. GPCRs can physically affect each other's functioning by forming heteromeric complexes, whereas cross-regulation between activated GPCRs also occurs through integration of shared intracellular signaling networks. Human herpesviruses utilize virally encoded GPCRs to hijack cellular signaling networks for their own benefit. Previously, we demonstrated that the Epstein-Barr virus-encoded GPCR BILF1 forms heterodimeric complexes with human chemokine receptors. Using a combination of bimolecular complementation and bioluminescence resonance energy transfer approaches, we now show the formation of hetero-oligomeric complexes between this viral GPCR and human CXCR4. BILF1 impaired CXCL12 binding to CXCR4 and, consequently, also CXCL12-induced signaling. In contrast, the G protein uncoupled mutant BILF1-K(3.50)A affected CXCL12-induced CXCR4 signaling to a much lesser extent, indicating that BILF1-mediated CXCR4 inhibition is a consequence of its constitutive activity. Co-expression of Gα(i1) with BILF1 and CXCR4 restored CXCL12-induced signaling. Likewise, BILF1 formed heteromers with the human histamine H(4) receptor (H(4)R). BILF1 inhibited histamine-induced Gα(i)-mediated signaling by H(4)R, however, without affecting histamine binding to this receptor. These data indicate that functional cross-regulation of Gα(i)-coupled GPCRs by BILF1 is at the level of G proteins, even though these GPCRs are assembled in hetero-oligomeric complexes.

FIGURE 1.

Detection of BILF1 and CXCR4 heterodimers by BRET. A and B, HEK293T cells were transfected with BBS-BILF1 or CXCR4 BRET fusion constructs. After 48 h, receptor surface expression (RSE) levels were quantified for BBS-BILF1 and CXCR4 by [¹²⁵I]αBTX binding or ELISA, respectively, on whole cells. Measured Rluc (gray bars) and Rluc8 (white bars) luminescence (A) and eYFP (gray bars) and mVenus (white bars) fluorescence (B) are normalized for RSE. C and D, HEK293T cells were transfected with fixed amounts of the BRET donor construct BBS-BILF1-Rluc8 (C) and CXCR4-Rluc8 (D) DNA and increasing amounts of BRET acceptor constructs BBS-BILF1-mVenus (○), CXCR4-mVenus (●), and GABA_B2-YFP (□) DNA. The BRET ratio is defined as [BRET/luminescence] in the presence of acceptor constructs minus [BRET/luminescence] in the absence of acceptor constructs. BRET ratio was plotted as a function of the increasing acceptor/donor ratio as quantified by fluorescence and luminescence, respectively. Data are means ± S.E. of representative experiments performed at least three times in triplicate. Significant differences (p < 0.0001) were determined using a t test and are indicated by asterisks. Curves were fitted using nonlinear regression, assuming a single binding site. au, arbitrary units.

FIGURE 2.

Detection of BILF1 and CXCR4 heterodimers by BiLC and BiFC. A, HEK293T cells were co-transfected with 10 ng of BBS-BILF1-L1 and 1–50 ng of CXCR4-L2 DNA/10⁶ cells (○) or 10 ng of BBS-BILF1-L2 and 1–50 ng of CXCR4-L1 (●). B, HEK293T cells were co-transfected with 10 ng of CXCR4-L1 and 1–50 ng of BBS-BILF1-L2 (○) or 10 ng of CXCR4-L2 and 1–50 ng of BBS-BILF1-L1 (●). Luminescence was measured after 48 h and plotted as a function of BBS-BILF1 and CXCR4 RSE as determined by [¹²⁵I]αBTX binding and ELISA, respectively. C and D, HEK293T cells were co-transfected with the split BiLC constructs BBS-BILF1-L1 and BBS-BILF1-L2 or BBS-BILF1-Rluc8. The RSE of the BBS-BILF1 fusion proteins was determined by [¹²⁵I]αBTX binding (C), and luminescence was measured in parallel (D). E–H, HEK293T cells were co-transfected with 10 ng of BBS-BILF1-V1 and CXCR4-V2 DNA/10⁶ cells and cultured on poly-l-lysine-coated coverslips. After 48 h, the BBS-BILF1 protomer was labeled with TMR-αBTX as depicted in the schematic (E). Reconstituted mVenus (green (F)), TMR-αBTX labeling (red (G)), and co-localization (yellow (H)) were analyzed using an Olympus FSX100 fluorescence microscope. Nuclei were stained with 4′,6-diamidino-2-phenylindole (blue). Data are means ± S.E. of representative experiments performed at least three times in triplicate. rlu, relative light units.

FIGURE 3.

Detection of BILF1 and CXCR4 hetero-oligomers by BiLC/BiFC-BRET. A, HEK293T cells were co-transfected with 10 ng of DNA from the BiLC constructs BBS-BILF1-L1 and CXCR4-L2 (1:1) and 0–950 ng of DNA from the BiFC constructs BBS-BILF1-V1 and CXCR4-V2 (○) or CXCR4-V1 and BBS-BILF1-V2 (●) (1:1)/10⁶ cells. B, BiLC constructs CXCR4-L1 and BBS-BILF1-L2 in combination with BiFC constructs BBS-BILF1-V1 and CXCR4-V2 (○) or CXCR4-V1 and BBS-BILF1-V2 (●). After 48 h, fluorescence, luminescence, and BRET were measured. The BRET ratio was plotted as a function of the increasing reconstituted acceptor (BiFC)/donor (BiLC) ratio as quantified by fluorescence and luminescence, respectively. Data are representative of at least three independent experiments each performed in triplicate.

FIGURE 4.

BILF1 inhibits CXCL12 binding to CXCR4. HEK293T cells were co-transfected with 50 ng of CXCR4 and 0, 2.5 or 25 ng of BBS-BILF1 or BBS-BILF1-K^3.50A DNA/10⁶ cells. A, DNA dose-dependent BBS-BILF1 and BBS-BILF1-K^3.50A expression was shown by [¹²⁵I]αBTX binding to membrane preparations. B, CXCR4 expression was determined in parallel by ELISA. C, [¹²⁵I]CXCL12 binding on the same membrane set. Data are means ± S.E. from three independent experiments, each performed in triplicate. Statistical differences between CXCR4 in the absence or presence of BBS-BILF1-(K^3.50A) were determined using a one-way ANOVA followed by a Bonferroni's multiple comparison test (*, p < 0.05; **, p < 0.001).

FIGURE 5.

BILF1 inhibits CXCL12-induced G protein activation by CXCR4. HEK293T cells were transfected with 50 ng of CXCR4 and/or 20 ng of BBS-BILF1 DNA or BBS-BILF1-K^3.50A DNA/10⁶ cells. A, CXCL12-induced (0.1 μm) binding of [³⁵S]GTPγS was measured in membranes derived from cells co-transfected with BBS-BILF1 and CXCR4 (CO) or from cells expressing either BBS-BILF1 or CXCR4 that were mixed prior to membrane preparation (MIX) in response to a 1-h incubation with buffer (black bars), 0.1 μm CXCL12 (gray bars), or 0.1 μm CXCL12 and 10 μm AMD3100 (white bars). Mock-transfected cells were used as control. B, binding of [³⁵S]GTPγS to membranes expressing the indicated receptor constructs was determined in response to vehicle treatment or 0.1 μm CXCL12. To emphasize changes in ligand-induced signaling, [³⁵S]GTPγS binding was normalized to vehicle treatment for each of the membrane preparations and consequently is shown as fold basal. Data are means ± S.E. from three independent experiments, each performed in triplicate. Statistical differences were determined using a one-way ANOVA followed by a Bonferroni's multiple comparison test (*, p < 0.05; **, p < 0.001).

FIGURE 6.

BILF1 inhibits CREB activation by CXCR4 in response to CXCL12. A, CXCR4 RSE in HEK293T cells transfected with CXCR4 (50 ng/10⁶ cells), BBS-BILF1 (25 ng/10⁶ cells), and/or BBS-BILF1-K^3.50A (25 ng/10⁶ cells) as determined by ELISA. B, FSK-induced CREB activity was measured in parallel in the absence or presence of 0.1 μm CXCL12. To emphasize changes in ligand-induced signaling, CREB activity is normalized to vehicle treatment for each of the transfectants and consequently shown as fold basal. Data are means ± S.E. of representative experiments performed at least three times in triplicate. Significantly changes in CXCR4 RSE as compared with CXCR4-only cells are determined by a one-way ANOVA followed by a Bonferroni's multiple comparison test (A). Significant changes in CREB activity in the presence of CXCL12 as compared with mock cells were determined by the same statistical analysis (*, p < 0.05; **, p < 0.001).

FIGURE 7.

BILF1 inhibits CXCL12-induced CXCR4 signaling by scavenging Gα_i proteins. A, [¹²⁵I]CXCL12 binding to CXCR4-expressing membranes in the absence and presence of 3 μm GTPγS. B, FSK-induced CREB activity in cells expressing the indicated receptors in the absence or presence of co-transfected Gα_i1 proteins in response to vehicle or 0.1 μm CXCL12 treatment. CXCL12-modulated CREB activity is normalized to vehicle treatment for each of the transfectants and consequently shown as fold basal. C, BBS-BILF1 and CXCR4 RSE as determined by [¹²⁵I]αBTX binding or ELISA to whole cells, respectively. Data are means ± S.E. of representative experiments performed at least three times in triplicate. Significant differences compared with FSK-treated mock cells (gray bar (B)), unless otherwise specified, are indicated by asterisks as determined using a one-way ANOVA followed by a Bonferroni's multiple comparison test (*, p < 0.05; **, p < 0.001).

FIGURE 8.

BILF1 forms heteromers with H₄R and inhibits histamine-induced signaling without affecting binding. A–D, HEK293T cells were co-transfected with 50 ng of H₄R-V1 and BBS-BILF1-V2 DNA/10⁶ cells and cultured on poly-l-lysine-coated coverslips. After 48 h, the BBS-BILF1 protomer was labeled with TMR-αBTX as depicted in the schematic (A), and co-localization (yellow (D)) of reconstituted mVenus (green (B)) and TMR-αBTX (red (C)) was analyzed using an Olympus FSX100 fluorescence microscope. Nuclei were stained with 4′,6-diamidino-2-phenylindole (blue). E, FSK-induced CREB activity in cells in the absence or presence of 10 μm histamine (HA) 48 h after transfection with 50 ng of Myc-H₄R and/or 25 ng of BBS-BILF1 or BILF-K^3.50A DNA/10⁶ cells. Histamine-induced CREB activity is normalized to vehicle treatment for each of the transfectants and consequently shown as fold basal. F, [³H]histamine binding to membranes expressing the indicated receptors. G, [³H]histamine binding to membranes expressing H₄R in the presence of 10 μm histamine or increasing concentrations (0.3–10 μm) of GTPγS. The bar graphs represent the mean of pooled data from at least two experiments performed in triplicate. Error bars indicate S.E. Significant differences are indicated by asterisks, as determined using a one-way ANOVA followed by a Bonferroni's multiple comparison test (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

FIGURE 9.

Schematic representation of the inhibition of CXCR4 by the constitutively active BILF1. A, CXCL12 binding to CXCR4 is Gα_i protein-dependent (a) and results in receptor and G protein activation (b). B, constitutively active BILF1 heteromerizes with CXCR4 and may allosterically inhibit CXCL12 binding to CXCR4 (c). Moreover, BILF1 constitutively recruits and activates Gα_i proteins (d). This scavenging prevents Gα_i proteins from interacting with CXCR4, resulting in an inhibition of CXCL12 binding (e). C, the constitutively inactive mutant BILF1-K^3.50A is not able to cross-inhibit CXCL12 binding to CXCR4 upon heteromerization (f). BILF1-K^3.50A does not compete with CXCR4 for available Gα_i proteins (g), allowing Gα_i protein-dependent CXCL12 binding to CXCR4 (h). For the sake of clarity, receptor homomers and oligomers are not shown in this figure.