GPCRs may not pair up as often as thought

Some purported pairings are actually just random interactions between molecules, a new report claims

Charles Q. Choi
Nov 5, 2006
The largest known family of proteins, G-protein coupled receptors (GPCRs), may not pair up as often as is widely thought, scientists report in Nature Methods. The controversial new results suggest prior studies may have mistaken random interactions for true dimerizations. There is "no doubt that a large number of labs do fall into the type of traps described in the manuscript because they do not have the relevant technical expertise," said Graeme Milligan of the University of Glasgow in Scotland, who did not participate in this study. "I think many of the best people in the field do consider that a lot of relatively poor papers have been produced, particularly in studies examining hetero-dimerization. Of course, key experiments now require the presence of dimers to be shown unequivocally in native tissues," he told The Scientist in an email. Despite this, "I feel that this manuscript fails to consider the large number of other approaches that have provided support for GPCR dimerization," Milligan said. "Many pretty good papers have used a range of techniques. One of our own studies used five different approaches, each based on a different biochemical/biophysical principle."GPCRs are key to communication between cells and physiological senses, and are often the principal targets of therapeutic drugs. For years, scientists thought GPCRs worked alone, but recent evidence suggested they often pair up as homodimers or heterodimers or both.The principal method by which scientists now investigate how GPCRs form oligomers is bioluminescence resonance energy transfer (BRET), which couples putative donors with luciferase and acceptors with green fluorescent protein (GFP) and then looks for any transfer of energy in the form of light between paired molecules. BRET analyses are often performed at the highest expression levels available to get clear results. They also frequently use just one ratio between donor and acceptor. Lead coauthor John James, of the University of Oxford in England, and colleagues speculated that some pairings seen via BRET were actually just random interactions between molecules.Senior coauthor Simon Davis and his colleagues at the University of Oxford designed two sets of experiments to test this possibility. In one, the combined expression levels of both donors and acceptors were kept constant, and the donor-acceptor ratio varied. In this set of experiments, the level of random interactions between molecules stayed constant as ratios rose and fell above a certain threshold level of expression, while true pairing levels fluctuated. In the second set of experiments, the combined expression levels varied, while the donor-acceptor ratios stayed the same. The level of random interactions then fluctuated with the expression level, while the rate of true pairing did not."We were very surprised to find that all proteins, even known monomeric proteins, gave substantial BRET values. This suggested that proteins that are confined to the plasma membrane are relatively crowded, with non-specific interactions common," James told The Scientist via email.The team also found that beta-2 adrenergic receptor (ß2AR), which belongs to rhodopsin-like GPCR Family A, had "no evidence for homo- or heterodimerization," James added. This contrasts with the findings of several previous papers, mainly from the lab of Michel Bouvier at the University of Montreal, "which found ß2AR readily homo- and heterodimerized," James said. "Conversely, in our hands, the known GPCR heterodimer GABAßR1/2 behaved as expected, strengthening our conclusions."These results should prompt reevaluation of suspected dimers of GPCRs, "especially between hetero-GPCRs within Family A," according to Kenneth Jones at Lundbeck Research in Paramus, New Jersey, who was not a coauthor. "Many in the field, including myself, have begun to take it for granted that all members of this family of receptors 'live' in the dimerized state. It may be that dimerization does widely occur, but it is a short-lived event.""If certain GPCRs dimerized transiently, they could be doing so as part of the signal transduction process, related either to G-protein activation, desensitization, or internalization. It would be important to understand such transient dimerization to better understand signal transduction, or possibly protein trafficking," Jones told The Scientist via email.However, the results of the new study have already sparked controversy."The premise of this paper, that we need good quantitative assessments of studies trying to look at dimerization, is good. But we continue to believe in the interpretation of many studies, either BRET or FRET (fluorescence resonance energy transfer), that says many GPCRs, if not all, form dimers or even higher structure oligomers," Bouvier, who did not participate in this study, told The Scientist."In this paper, they do seem to completely disregard all the other evidence of other kinds of studies supporting dimerization of class A GPCRs," Bouvier said. "Ironically, a paper that just appeared from Krzysztof Palczewski in PNAS crystallized active rhodopsin, a class A GPCR, and it came out as a dimer."Bouvier also noted that James and his colleagues employed second-generation BRET analysis, which has a better signal-to-noise ratio than first-generation BRET but "is 100 times less sensitive." This forces higher expression levels of proteins to measure signals. "We think they may have therefore expressed too many receptors, thus favouring random collisions that may supersede the signal originating from specific dimers," Bouvier explained. "You need to quantify the amount of receptors expressed in molar terms."Charles Q. Choi Links within this article:T.P. Sakmar. "Twenty Years of The Magnificent Seven," The Scientist, January 17, 2005 James et al. "A rigorous experimental framework for detecting protein oligomerization using bioluminescence resonance energy transfer," Nature Methods, advance online publication, November 5, 2006 DOI: 10.1038/nmeth978 http://www.nature.comGraeme Milligan Wilson, G. Wilkinson, G. Milligan. "The CXCR1 and CXCR2 receptors form constitutive homo- and heterodimers selectively and with equal apparent affinities," J. Biol. Chem. 280: 28663-28674, August 5, 2005. Kling. "GPCRs: Researchers Disprove the Single Polypeptide Theory," The Scientist, October 29, 2001 Davis Wells. "Signaling for Survival," The Scientist, December 8, 2000 Bouvier Salom et al. "Crystal structure of a photoactivated deprotonated intermediate of rhodopsin," PNAS 103:16123-8, October 21, 2006.