Portal de Periódicos da CAPES

Phosducin-like protein acts as a molecular chaperone for G protein βγ dimer assembly

2005; Springer Nature; Volume: 24; Issue: 11 Linguagem: Inglês

10.1038/sj.emboj.7600673

1460-2075

Georgi L. Lukov, Ting Hu, Joseph N. McLaughlin, Heidi E. Hamm, Barry M. Willardson,

Glycosylation and Glycoproteins Research

Article5 May 2005free access Phosducin-like protein acts as a molecular chaperone for G protein βγ dimer assembly Georgi L Lukov Georgi L Lukov Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Ting Hu Ting Hu Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Joseph N McLaughlin Joseph N McLaughlin Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA Search for more papers by this author Heidi E Hamm Heidi E Hamm Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA Search for more papers by this author Barry M Willardson Corresponding Author Barry M Willardson Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Georgi L Lukov Georgi L Lukov Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Ting Hu Ting Hu Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Joseph N McLaughlin Joseph N McLaughlin Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA Search for more papers by this author Heidi E Hamm Heidi E Hamm Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA Search for more papers by this author Barry M Willardson Corresponding Author Barry M Willardson Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA Search for more papers by this author Author Information Georgi L Lukov1, Ting Hu1, Joseph N McLaughlin2, Heidi E Hamm2 and Barry M Willardson 1 1Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT, USA 2Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA *Corresponding author. Department of Chemistry and Biochemistry, Brigham Young University, C210 BNSN, Provo, UT 84602, USA. Tel.: +1 801 422 2785; Fax: +1 801 422 0153; E-mail: [email protected] The EMBO Journal (2005)24:1965-1975https://doi.org/10.1038/sj.emboj.7600673 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Phosducin-like protein (PhLP) is a widely expressed binding partner of the G protein βγ subunit dimer (Gβγ). However, its physiological role is poorly understood. To investigate PhLP function, its cellular expression was blocked using RNA interference, resulting in inhibition of Gβγ expression and G protein signaling. This inhibition was caused by an inability of nascent Gβγ to form dimers. Phosphorylation of PhLP at serines 18–20 by protein kinase CK2 was required for Gβγ formation, while a high-affinity interaction of PhLP with the cytosolic chaperonin complex appeared unnecessary. PhLP bound nascent Gβ in the absence of Gγ, and S18–20 phosphorylation was required for Gγ to associate with the PhLP-Gβ complex. Once Gγ bound, PhLP was released. These results suggest a mechanism for Gβγ assembly in which PhLP stabilizes the nascent Gβ polypeptide until Gγ can associate, resulting in membrane binding of Gβγ and release of PhLP to catalyze another round of assembly. Introduction Heterotrimeric G proteins mediate a wide array of hormonal, neuronal and sensory signals that control numerous physiological processes ranging from cardiac rhythm (Rockman et al, 2002) to psychological behavior (Gainetdinov et al, 2004) to vision (Arshavsky et al, 2002). G protein signaling is initiated by the binding of a ligand to the extracellular face of a G protein-coupled receptor (GPCR), resulting in a change in the packing of the seven-transmembrane α-helices found in all GPCRs. This conformational change activates the G protein on the intracellular surface of the receptor by initiating an exchange of GDP for GTP on the G protein α subunit (Gα). GTP binding causes Gα to dissociate from the G protein βγ subunit complex (Gβγ). Both Gα·GTP and Gβγ control the activity of effector enzymes and ion channels that determine the intracellular concentration of second messengers (cyclic nucleotides, inositol phosphates, Ca2+ and K+), which in turn orchestrate the cellular response to the stimulus. Phosducin-like protein (PhLP) is a widely expressed member of the phosducin gene family that is believed to participate in G protein signaling by virtue of its ability to bind the Gβγ dimer with high affinity (Miles et al, 1993; Thibault et al, 1997; Savage et al, 2000; Schroder and Lohse, 2000). Phosducins were originally thought to downregulate G protein pathways by sequestering Gβγ from its interaction with Gα (Bauer et al, 1992; Lee et al, 1992; Yoshida et al, 1994). However, the results of recent studies have not been consistent with this putative role. Specifically, disruption of the PhLP1 gene in the chestnut blight fungus Cryphonectria parasitica (Kasahara et al, 2000) and in the soil amoeba Dictyostelium discoideum (Blaauw et al, 2003) yielded the same phenotype as the disruption of the Gβ gene. Moreover, PhLP deletion blocked G protein signaling in Dictyostelium (Blaauw et al, 2003). In another study, the duration of opiate desensitization was prolonged in mice in which PhLP expression in the brain was inhibited by antisense oligonucleotide treatment (Garzon et al, 2002). All of these observations are the exact opposite of what would be expected if PhLP were a negative regulator. As a result, they have led to the conclusion that PhLP must be a positive regulator of G protein signaling. Insight into possible ways in which PhLP might facilitate G protein function has come from the observation that PhLP interacts with the cytosolic chaperonin complex (CCT), an essential molecular chaperone that mediates the folding of actin, tubulin and other proteins into their native structures (McLaughlin et al, 2002b). PhLP was shown to interact with CCT as a regulator and not as a folding substrate. In addition, the cryoelectron microscopic structure of the PhLP-CCT complex (Martín-Benito et al, 2004) shows that PhLP binds CCT at the top of the CCT apical domains positioned above the folding cavity in a manner analogous to prefoldin, a CCT cochaperone that binds nascent actin polypeptide chains and delivers them to CCT for folding (Martin-Benito et al, 2002). Coupling these findings with the fact that yeast Gβ (Ho et al, 2002) and other proteins with seven β-propeller structures similar to Gβ (Valpuesta et al, 2002; Camasses et al, 2003) interact with CCT suggests that PhLP might function as a chaperone for the folding of Gβ. To test this notion, the effects of small interfering RNA (siRNA)-mediated inhibition of PhLP expression in human cell lines on G protein signaling, Gβγ expression and assembly of nascent Gβγ dimers were determined, as were the effects of overexpression of PhLP and several PhLP variants lacking either protein kinase CK2 (CK2) phosphorylation sites, Gβγ binding or CCT binding. The results show that PhLP is required for the formation of the Gβγ complex, and they outline a mechanism by which PhLP catalyzes Gβγ dimer assembly. Results Cellular depletion of PhLP inhibits Gβ expression and G protein signaling To gain insight into the physiological function of PhLP, siRNA was used to block its expression in HeLa cells. Two different siRNA sequences were prepared and are referred to as PhLP-A and PhLP-B siRNA. Cells were transfected with either of these siRNAs, or a control siRNA targeting lamin A/C, and PhLP protein expression was determined by immunoblotting. PhLP-A siRNA was modestly effective, inhibiting PhLP protein expression by 50% (Figure 1A), while PhLP-B siRNA was much more effective, blocking PhLP expression by 90%. The control lamin A/C siRNA had no effect on PhLP expression when compared to mock-transfected cells, yet it inhibited lamin A/C expression by 90%. Likewise, the PhLP siRNAs had no effect on lamin A/C expression, indicating that the siRNAs were acting specifically to reduce their target mRNAs. Figure 1.siRNA-mediated depletion of PhLP inhibits Gβ expression and G protein signaling. (A) HeLa S3 cells were treated with the indicated siRNAs and then assayed after 96 h. Protein expression of PhLP, Gβl and lamin A/C were determined by immunoblotting. PhLP and Gβ1 band intensities were quantified and expressed as a percentage of the lamin A/C sample. Bars and symbols in each panel represent the average±s.e. from three separate experiments. Statistical significance relative to the lamin A/C control was determined by a paired t-test (*P<0.01). (B) Changes in intracellular Ca2+ in living cells were determined by measuring fluorescence from a Ca2+-sensitive dye using a FlexStation plate reader. Histamine (50 nM) was added at time zero and measurements were taken at the times indicated. Fluorescence data were normalized to the signal prior to histamine addition. (C) Levels of Gβ1 and GAPDH mRNA were determined by Northern blotting. Band intensities were quantified, normalized to that of GAPDH and expressed as a percentage of the lamin A/C sample. Download figure Download PowerPoint Based on the lack of G protein signaling upon PhLP deletion in single-celled organisms (Kasahara et al, 2000; Blaauw et al, 2003), it was reasonable to suspect that siRNA-mediated depletion of PhLP would adversely affect G protein signaling in HeLa cells. As a first step to investigate this idea, the expression of the Gβ1 subunit, the most widely and abundantly expressed Gβ subunit, was determined in PhLP-depleted cells. PhLP-A siRNA had no detectible effect on Gβ1 expression in HeLa cells. However, the more potent PhLP-B siRNA consistently decreased Gβ1 expression by 40%, demonstrating that removal of 90% of the PhLP from the cell somehow inhibited endogenous Gβ1 expression. Such a decrease in Gβ1 levels would be expected to impact G protein signaling. This notion was investigated by measuring the change in intracellular Ca2+ in HeLa cells in response to histamine. Histamine receptors initiate a classical Gq-mediated cascade, which results in an influx of Ca2+ into the cytosol (Bootman et al, 1997). PhLP-B siRNA transfection caused a 60% reduction in histamine-induced Ca2+ transient compared to lamin A/C siRNA or mock-transfected cells (Figure 1B). PhLP-A siRNA treatment also caused a modest but reproducible decrease in Ca2+ influx. Thus, it appears that PhLP acts as a positive regulator of G protein signaling in HeLa cells by contributing to the cellular expression of Gβ1. The PhLP siRNA-mediated impairment of Gβ1 expression could potentially result from a loss of PhLP function at any level of the expression process from gene transcription to protein degradation. To assess pretranslational events, the effect of PhLP depletion on Gβ1 mRNA levels was determined by Northern blotting. Treatment of HeLa cells with either of the PhLP siRNAs caused no change in the levels of Gβ1 mRNA compared to lamin A/C or mock-transfected controls (Figure 1C), indicating that PhLP acts translationally or post-translationally to promote Gβ1 expression. PhLP is required for Gβγ dimer assembly Two recent studies have suggested that PhLP might be involved in post-translational regulation of Gβ folding or Gβγ assembly (Blaauw et al, 2003; Martín-Benito et al, 2004). This idea was based on several observations including the similarity of the PhLP-CCT structure to another CCT cochaperone, prefoldin (Martín-Benito et al, 2004), and on the mislocalization of Gβ-GFP and Gγ-GFP to the cytosol when the PhLP1 gene was deleted in Dictyostelium (Blaauw et al, 2003). These observations lead to an examination of the effect of siRNA-mediated PhLP depletion on the expression of the Gβγ dimer. HEK-293 cells were chosen for this experiment because they readily overexpress Gβ and Gγ from plasmid vectors. Cells were siRNA treated and then cotransfected 24 h later with Gβ1 and N-terminally hemaggluttinin (HA)-tagged Gγ2. The HA-Gγ2 was immunoprecipitated 72 h later and the co-immunoprecipitate was immunoblotted with anti-Gβ1 and anti-HA antibodies to determine the amount of Gβγ complex formed. PhLP-A and -B siRNA treatment decreased PhLP expression in these cells by 25 and 75%, respectively, compared to lamin A/C controls (Figure 2A). These siRNAs also decreased the amount of Gβγ complex by 35 and 75%, respectively (Figure 2A). The close correlation between the inhibition of PhLP expression and the decrease in Gβγ levels in a second human cell line further demonstrates the need for PhLP in Gβγ expression. Figure 2.siRNA-mediated depletion of PhLP inhibits Gβγ dimer assembly. (A) HEK-293 cells were treated with the indicated siRNA and transiently transfected with Gβl and HA-Gγ2 and then assayed after 96 h. Levels of overexpressed Gβlγ2 dimer were determined by immunoprecipitating HA-Gγ2 and immunoblotting for Gβl and HA-Gγ2. The effect of siRNA treatment on PhLP expression in HEK-293 cells was determined by immunoblotting whole-cell extracts for PhLP. Band intensities were quantified and expressed as a percentage of the lamin A/C sample. Bars and symbols in each panel represent the average±s.e. from three or four separate experiments (*P<0.01). (B) The rate of nascent Gβlγ2 dimer formation was determined using a radiolabel pulse–chase assay. Cells were pulsed for 10 min with [35S]methionine followed by a chase with unlabeled methionine. Times indicate the sum of the pulse and chase periods. After the chase, Gβlγ2 dimers were immunoprecipitated with an antibody to HA-Gγ2. The dimers were resolved by Tris–Tricine–SDS–PAGE and radioactive protein bands were detected using a phosphorimager. Band intensities were quantified and molar ratios of Gβl to Gγ2 were calculated. Lines represent a nonlinear least-squares fit of the data to a first-order rate equation. Values for t1/2 are shown next to the graph. Download figure Download PowerPoint The data in Figure 1 suggest that the decreases in Gβ and Gγ expression were caused by effects of PhLP depletion on translation or post-translation events. To determine whether this was also the case in the overexpression system, the effect of PhLP depletion on overexpressed Gβ and Gγ mRNA levels were measured. No significant differences were observed (data not shown), confirming that PhLP depletion had little effect on pretranslational events. Given the observations suggesting a role for PhLP in Gβγ assembly (Blaauw et al, 2003; Martín-Benito et al, 2004), it seemed reasonable to explore the effects of PhLP depletion on this process. The rate of assembly of nascent Gβγ dimers was measured in a pulse-chase experimental format designed to detect newly synthesized proteins. HEK-293 cells that had been treated with PhLP-B siRNA and then transfected with Flag-Gβ1 and HA-Gγ2 were pulsed with [35S]methionine for 10 min, and then chased for the times indicated with excess unlabeled methionine. At the end of the chase period, the amount of Gβγ dimer formed was determined by immunoprecipitating the HA-Gγ and measuring the amount of co-immunoprecipitating [35S]-labeled Gβ. In lamin A/C siRNA-treated cells, there was a clear increase in [35S]Gβ in the Gγ immunoprecipitate as the chase time increased (Figure 2B). In contrast, there was almost no increase in co-immunoprecipitation of [35S]Gβ during the chase period in cells treated with the PhLP-B siRNA. In addition, the amount of Gγ synthesized during the pulse was reduced two-fold in the PhLP-B-treated cells. The observed decrease in [35S]Gβ co-immunoprecipitation in the PhLP-B-treated cells was clearly greater than the decrease in Gγ synthesis, especially at later time points in the chase period, suggesting that the rate of Gβγ assembly was also impaired in PhLP-depleted cells. To better assess this finding, the molar ratio of Gβ to Gγ was calculated at each time point during the chase period. A plot of the change in Gβ/Gγ ratio with time (Figure 2B) showed a significant decrease in the rate of assembly of Gβγ in PhLP-B siRNA-treated cells. The half-life for assembly of Gβγ in PhLP-depleted cells was ∼300 min compared to ∼60 min in lamin A/C siRNA-treated control cells. Similar results were observed when Gβ was immunoprecipitated. There was a two-fold reduction in Gβ synthesized and very little Gγ co-immunoprecipitated in PhLP-B siRNA-treated cells compared to the lamin A/C control (data not shown). Thus, these data support the idea that PhLP promotes the assembly of the Gβγ dimer. PhLP phosphorylation at serines 18–20 is required for Gβγ dimer assembly PhLP has been shown to be constitutively phosphorylated by CK2 at serines 18–20 (Humrich et al, 2003; Lukov et al, in preparation). Overexpression of a variant of PhLP in which these three residues were replaced by alanine (PhLP S18–20A) completely blocked the ability of overexpressed Gβγ to activate PLCβ in HEK-293 cells (Humrich et al, 2003). Subsequent experiments have shown that CK2 phosphorylation does not change the binding affinity of PhLP for Gβγ, but it does increase PhLP binding to CCT by three-fold (Lukov et al, in preparation). Coupling these findings with the observation that PhLP is required for Gβγ dimer assembly points to a role for CK2 phosphorylation in the regulation of Gβγ folding. To investigate this possibility, the effects of coexpression of PhLP S18–20A on Gβγ expression were measured by co-immunoprecipitation as in Figure 2A. The PhLP S18–20A variant inhibited Gβγ expression by approximately 70% compared to wild-type PhLP, while the empty vector control consistently showed 25% less Gβγ than wild-type PhLP (Figure 3A). This decrease in Gβγ expression was not attributable to a decrease in mRNA levels because Northern blot analyses showed no significant changes in overexpressed Gβ and Gγ mRNA when PhLP S18–20A was coexpressed (data not shown). Moreover, the effects of PhLP S18–20A overexpression appear to be a direct result of an inability to phosphorylate this site and were not caused by the alanine substitutions themselves because in the absence of CK2 phosphorylation, these substitutions had no effect on Gβγ or CCT binding (Lukov et al, in preparation). Figure 3.PhLP phosphorylation at serines 18–20 is required for Gβγ dimer assembly. (A) HEK-293 cells were transiently transfected with PhLP-myc, PhLP S18-20A or empty vector as indicated along with Gβ1 and HA-Gγ2 and then assayed after 48 h. Levels of overexpressed Gβlγ2 dimer were determined by immunoprecipitation as in Figure 2A. Bars represent the average±s.e. from six separate experiments (*P<0.01, **P<0.001). (B) The rate of nascent Gβlγ2 dimer formation was determined by a pulse–chase assay as in Figure 2B. Symbols represent the average±s.e. from three separate experiments. Download figure Download PowerPoint To further explore the role of CK2 phosphorylation, the effects of overexpression of the PhLP S18–20A variant on Gγ translation and Gβγ assembly were measured in the pulse–chase experimental format. Overexpression of wild-type PhLP increased the rate of Gβγ assembly substantially when compared to the empty vector control. The t1/2 for assembly was 12 min compared to 45 min for the control, nearly a four-fold increase (Figure 3B). In contrast, overexpression of the PhLP S18–20A variant caused a dramatic decrease in the rate of Gβγ assembly with a t1/2 of ∼180 min, more than 15-fold less than that of wild-type PhLP. The effects on Gγ translation were also very different. Wild-type PhLP had no effect, while PhLP S18–20A inhibited translation by 40%, similar to the decrease caused by PhLP depletion. These data confirm the role of PhLP as a positive regulator of Gβγ assembly and they show that CK2 phosphorylation at S18–20 is required for normal Gγ translation and Gβγ dimer assembly. Moreover, it appears that the overexpressed PhLP S18–20A interferes in some way with endogenous PhLP in performing these functions. High-affinity binding of PhLP to Gβγ but not to CCT is necessary for Gβγ assembly The findings that PhLP phosphorylation is necessary for Gβγ assembly and that phosphorylation increases the binding affinity of PhLP for CCT (Lukov et al, in preparation) suggest that the interaction between PhLP and CCT is necessary for assembly. This observation led to an analysis of the contribution of the PhLP–CCT interaction in Gβγ folding. A variant of PhLP with a greatly reduced binding affinity for CCT was prepared by substituting residues D132DEE with alanine. These residues have been shown to contribute considerably to CCT binding (Martín-Benito et al, 2004). This PhLP 132–135A variant had normal Gβγ binding properties, but it bound CCT poorly in co-immunoprecipitation experiments from cells overexpressing the variant (Figure 4A). In addition, two other variants were prepared that bound CCT normally, but had reduced binding to Gβγ. The first was a truncation of PhLP in which residues 1–75 were deleted (PhLP Δ1–75). This variant lacks Helix 1, which is known to make a substantial contribution to Gβγ binding (Gaudet et al, 1996), yet it retains regions known to interact with CCT (Martín-Benito et al, 2004). The second variant was a chimera in which residues 76–94 of PhLP were replaced with Pdc sequence. Previous binding experiments showed that this PhLP/Pdc 76–94 variant had reduced Gβγ binding but normal CCT binding (Martín-Benito et al, 2004). As expected, PhLP Δ1–75 bound Gβγ poorly while PhLP/Pdc 76–94 showed intermediate binding, significantly less than wild-type PhLP yet more than PhLP Δ1–75 (Figure 4A). Both of these variants bound CCT normally (Figure 4A). Figure 4.Interaction of PhLP with Gβγ but not with CCT is necessary for Gβγ dimer assembly. (A) HEK-293 cells were transiently transfected with PhLP-myc, PhLP 132–135A, PhLP Δ1–75, PhLP/Pdc 76–94, or empty vector as indicated along with Gβ1 and HA-Gγ2 and then assayed after 48 h. The binding of PhLP variants to Gβγ and CCT was measured by immunoprecipitating with an antibody to the C-terminal myc tag and immunoblotting for PhLP, Gβ or CCTε. Blots are representative of similar data from three experiments. (B) Levels of overexpressed Gβlγ2 dimer were determined by immunoprecipitation as in Figure 2A. Bars represent the average±s.e. from three separate experiments (*P<0.01, **P<0.001). (C) The rate of nascent Gβlγ2 dimer formation was determined by a pulse–chase assay as in Figure 2B. Symbols represent the average±s.e. from three separate experiments. Download figure Download PowerPoint The effects of coexpression of these PhLP variants on cellular levels of overexpressed Gβγ were measured as in Figure 2A. Surprisingly, coexpression of PhLP 132–135A enhanced expression of Gβγ by 20% compared to wild-type PhLP, while the opposite effect was observed with coexpression of PhLP Δ1–75, which dramatically blocked Gβγ expression by 90% (Figure 4B). PhLP/Pdc 76–94 coexpression also inhibited Gβγ expression but the effect was less striking, about 50% less than wild type. Northern blot analysis showed the mRNA levels of overexpressed Gβ and Gγ in these cells were the same as that found in cells coexpressing wild-type PhLP (data not shown), indicating that inhibition of Gβγ expression in cells coexpressing PhLP Δ1–75 or PhLP/Pdc 76–94 was not caused by decreases in mRNA levels. The effects of coexpression of these PhLP variants on Gγ translation and Gβγ assembly were also measured. PhLP 132–135A coexpression did not change the rate of Gγ translation and Gβγ dimer assembly when compared to wild-type PhLP (Figure 4C). On the other hand, the effect of PhLP Δ1–75 coexpression was striking, inhibiting translation of Gγ by 40% and completely blocking assembly. The effects of PhLP/Pdc 76–94 coexpression were more moderate, showing no effect on Gγ translation and reducing the rate of dimer assembly by four-fold. These results demonstrate that an interaction of PhLP with Gβγ is vital for assembly of the Gβγ dimer, and they suggest that high-affinity binding of PhLP to CCT is not necessary when PhLP is overexpressed. PhLP associates with the nascent Gβ polypeptide in the absence of Gγ To address the mechanism by which PhLP controls Gβγ assembly, the effects of combinatorial overexpression of each of the three G protein subunits on the ability of PhLP to form complexes with nascent Gβ or Gγ was assessed. PhLP was coexpressed with either Gβ1 alone, Gγ2 alone, Gβ1 and Gγ2 together or all three G protein subunits (Gαi3, Gβ1 and Gγ2) in HEK-293 cells, and the cells were pulsed with [35S]methionine for 30 min to label the nascent polypeptides. Complexes of newly synthesized proteins associated with PhLP, Gβ or Gγ were determined by co-immunoprecipitation. When all three G protein subunits were coexpressed together and PhLP was immunoprecipitated with an antibody to its C-terminal c-myc tag, significant amounts of nascent Gβ were found in the co-immunoprecipitate, but there was no nascent Gα or Gγ (Figure 5A, left four lanes). Gα was not expected to co-immunoprecipitate because it is known that Gα and PhLP compete for the same binding site on Gβ (Gaudet et al, 1996). However, it was very surprising not to find Gγ in the co-immunoprecipitate because PhLP has been shown to bind the Gβγ complex with moderately high affinity (Savage et al, 2000) and Gβ forms a very high-affinity complex with Gγ (Clapham and Neer, 1997). Similar results were seen when Gβ and Gγ or when Gβ alone were coexpressed, Gβ co-immunoprecipitated with PhLP without any detectible Gγ. Thus, the unanticipated conclusion from these data is that PhLP forms a complex with nascent Gβ in the absence of Gγ. Figure 5.PhLP binds nascent Gβ in the absence of Gγ and requires S18–20 phosphorylation for Gγ to associate with the complex. (A) HEK-293 cells were transiently cotransfected with wild-type PhLP-myc in combination with Gαi3, Flag-Gβ1 and/or HA-Gγ2 as indicated. After 48 h, the cells were pulsed with [35S]methionine for 30 min and extracts were immunoprecipitated using antibodies to the PhLP-myc, Flag-Gβ1 or HA-Gγ2 epitope tags. The co-immunoprecipitating proteins were resolved on 10% Tris–Glycine–SDS gels for PhLP, Gα and Gβ and 16.5% Tris–Tricine–SDS gels for Gγ. Radioactive protein bands were detected using the phosphorimager. The gel shown is representative of similar gels from three separate experiments. (B) HEK-293 cells were transiently transfected with PhLP-myc, PhLP S18–20A, PhLP 132–135A or PhLP Δ1–75 as indicated along with Flag-Gβ1- and HA-Gγ2-expressing vectors. After 48 h, the cells were pulsed with [35S]methionine for 30 min, extracts were immunoprecipitated and were analyzed as in panel A. Band intensities were quantified and the indicated molar ratios of binding partners were calculated as in Figure 2B. A representative gel is shown. Bar graphs represent the average molar ratios±s.e. from three separate experiments (*P<0.01, **P<0.001). Download figure Download PowerPoint A similar conclusion can be made from Gγ immunoprecipitation experiments. The same cell extracts were immunoprecipitated with an antibody to the HA-tag on the N-terminus of Gγ, resulting in co-immunoprecipitation of nascent Gβ and Gα, but not PhLP (Figure 5A, right four lanes). There was a minor band migrating just below the PhLP band that was observed in variable amounts in each of the four samples. This was a nonspecifically co-immunoprecipitating band since it was found in the sample expressing Gβ alone, which should have had no immunoprecipitate. When Gβ was immunoprecipitated from these same cell extracts, nascent PhLP, Gα and Gγ were co-immunoprecipitated whenever they were coexpressed (Figure 5A, middle four lanes), indicating that Gβ is in complexes with all three simultaneously. The composition of these complexes appears to be PhLP-Gβ, -Gβγ and -Gαβγ. PhLP phosphorylation at S18–20 is needed for Gγ to associate with the PhLP-Gβ complex To determine the role of CK2 phosphorylation, CCT binding and Gβγ binding in the assembly process, the PhLP variants deficient in these properties (PhLP S18–20A, PhLP 132–135A and PhLP Δ1–75, respectively) were coexpressed with Gβ and Gγ and their binding was measured as in Figure 5A. PhLP immunoprecipitation brought down detectible amounts of nascent Gβ in all of the samples (Figure 5B). However, plotting the ratio of Gβ to PhLP in the immunoprecipitates showed that there was significantly less Gβ relative to PhLP in the PhLP Δ1–75 sample, as expected from the decreased Gβγ binding of this variant. As observed in Figure 5A, there was no Gγ in any of the PhLP immunoprecipitates, confirming the observation that PhLP bound nascent Gβ in the absence of Gγ. In the Gβ immunoprecipitates, there was 60% less Gβ in the PhLP S18–20A or PhLP Δ1–75 samples compared to the wild type or PhLP 132–135A. The magnitude of this decrease cannot be attributed to a decrease in Gβ mRNA, so it appears that the decrease results from inhibition of Gβ translation in the presence of PhLP S18–20A or PhLP Δ1–75. Interestingly, there was no nascent Gγ in the Gβ immunoprecipitate when PhLP S18–20A and PhLP Δ1–75 were coexpressed, while Gγ was easily detected when PhLP and PhLP 132–135A were coexpressed. A plot of the Gγ to Gβ ratio showed that the lack of Gγ in the PhLP S18–20A and PhLP Δ1–75 samples was not merely a result of the decrease in Gβ expression, but was caused by a total inability of the nascent Gγ to associate with the Gβ in the presence of these PhLP variants. Consistent with this observation, there was no nascent Gβ in the Gγ immunoprecipitates when PhLP S18–20A and PhLP Δ1–75 were coexpressed, but there were detectib