Commentary

How activated receptors couple to G proteins

Heidi E. Hamm*

Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37232-6600

protein-coupled receptors (GPCRs) vide several new approaches to these ques- Gare involved in the control of every tions and important new information aspect of our behavior and physiology. about the active conformation of rhodop- This is the largest class of receptors, with sin and how it contacts the G protein several hundred GPCRs identified thus (13–15, 35). far. Examples are receptors for hormones Rhodopsin signal transduction in rods such as calcitonin and luteinizing hor- and cones underlies our ability to see both mone or neurotransmitters such as sero- in dim light (rod vision) and in color (cone tonin and dopamine. G protein-coupled vision). Different rhodopsins absorb light receptors can be involved in pathological maximally at different light wavelengths, processes as well and are linked to numer- and on activation they activate rod or cone ous diseases, including cardiovascular and transducins. Transducins activate rod and mental disorders, retinal degeneration, cone cGMP phosphodiesterases, causing cancer, and AIDS. More than half of all rapid light-activated cGMP breakdown, drugs target GPCRs and either activate or resultant closure of cGMP-sensitive chan- inactivate them. Binding of specific li- nels, and photoreceptor cell hyperpolar- gands, such as hormones, neurotransmit- ization and inhibition of photoreceptor ters, chemokines, lipids, and glycopro- neurotransmitter release. The study of teins, activates GPCRs by inducing or visual signal transduction has provided stabilizing a new conformation in the recep- many firsts. The major breakthroughs in tor (1, 2). Activated receptors (R*) can then receptor structure, including the primary activate heterotrimeric G proteins (com- sequence (9, 10), the tertiary structure (8), posed of ␣.GDP, ␤, and ␥ subunits) on the and the conformational changes required inner surface of the cell membrane (3–5). for activation (11), all have first come COMMENTARY GPCRs have a common body plan with from studies of rhodopsin and then have Fig. 1. Orientation of rhodopsin, transducin, and seven transmembrane helices. The intra- been verified to varying degrees in other G the membrane. Refined rhodopsin structure is cellular loops that connect these helices protein-coupled receptors. The visual G from ref. 36, and Gt is from ref. 23. Models are form the G protein-binding domain re- protein transducin was, in addition, the based on the crystal structures and are to scale. The viewed by refs. 5–7. How do GPCRs ac- first heterotrimeric G protein whose struc- carboxyl-terminal residues after S316 are not tivate G proteins and cause such specific ture was solved (12). shown. The orientation of Gt with respect to rho- responses in cells? What are the triggering Khorana’s group, in collaboration with dopsin and the membrane is based on the charge and hydrophobicity of the surface, the known rho- changes in GPCRs on agonist binding? Wayne Hubbell’s laboratory, has pio- dopsin-binding sites on Gt, and the sites of lipida- How do they fold, and what causes mis- neered the use of site-directed Cys mu- tion of G␣ and G␤␥ (23). folding in so many genetic diseases? All of tagenesis to place reporters (13) and these unanswered questions in the field crosslinkers (14, 15) in specific sites for depend on detailed structural informa- more detailed structural understanding of environment, which can complement crys- tion. Recently, the first high-resolution rhodopsin. Loewen et al. (13) use 19F- tallographic findings. In addition, they structure of a GPCR, rhodopsin, the vi- trifluoroethylthio groups to derivatize demonstrate a new method for examining sual light receptor, was solved by the several sets of cysteines at particular sites tertiary contacts in proteins by using so- groups of Palczewski, Okada, Stenkamp, in rhodopsin’s first, second, and third cy- lution NMR, which is applicable to mem- ␣ and Miyano (8). This structure reveals a toplasmic loops and helix 8 (previously brane proteins. This method will be wealth of information about how retinal is designated intracellular loop 4 before the particularly powerful in examining light- bound and how the rhodopsin ground crystal structure showed its helical na- induced conformational changes in the 19 state is stabilized. It also shows that crit- ture). They then use F nuclear Over- cytoplasmic surface, which almost cer- ical residues for G protein activation hauser effects between the fluorine labels tainly underlie G protein activation. Ex- (E134, R135) are buried and inaccessible to analyze distances between them. All of tensive mutagenesis and biochemical ex- to the rod photoreceptor G protein, trans- the 19F labels on single cysteines have periments in rhodopsin as well as a variety ducin (Gt). However, this does not resolve distinct chemical shifts, but when pairs of of other G protein-coupled receptors sug- the question of how an activated receptor cysteines are labeled, there are upfield or gest that receptor activation by ligand activates a G protein, because the struc- downfield shifts, suggesting proximity be- binding causes changes in the relative ori- ture of the inactive receptor was solved, tween residues 139 and 251 on rhodopsin’s entations of transmembrane helices 3 and leaving open the question of the activation second and third loops and between res- mechanism and the structure of the active idues 65 and 316 on the first loop and ␣ receptor. Thus, new structural approaches helix 8. These measures of distances pro- See companion articles on pages 4872, 4877, 4883, and are needed to address these questions. vide information on the structure of rho- 4888. Four papers from Khorana’s group pro- dopsin’s cytoplasmic face in the native *E-mail: [email protected].

www.pnas.org͞cgi͞doi͞10.1073͞pnas.011099798 PNAS ͉ April 24, 2001 ͉ vol. 98 ͉ no. 9 ͉ 4819–4821 Downloaded by guest on September 26, 2021 6 (11, 16–18, 31). These changes are then thought to affect the conformation of G protein-interacting intracellular loops of the receptor and thus uncover previously masked G protein-binding sites on the second, third, and fourth cytoplasmic loops (19–21). Fig. 1 shows a view of the receptor–G protein complex. The activated receptor causes hetero- trimeric G protein activation (Fig. 1) by causing GDP release from its binding site on the G␣ subunit. GTP binds to G␣, and G␣-GTP has reduced affinity for G␤␥ and receptor; both G␣-GTP and G␤␥ are then Fig. 2. GRASP (http:͞͞trantor.bioc.columbia.edu͞grasp) views of the interacting surfaces between free to activate downstream effectors. G rhodopsin’s cytoplasmic face and Gt’s rhodopsin-interacting surface. Imaginary folding on the dotted line protein activation leads to activation of will dock the two molecules in the ‘‘best guess’’ complementary surface. The cytoplasmic face of rhodopsin is relatively small and has a distinct orientation, because rhodopsin is a transmembrane protein. Coordi- various second messenger systems and nates from the refined rhodopsin structure [Protein Data Bank (PDB) no. 1HZX (37)] and Gt [PDB entry intracellular responses, leading to physio- 1GOT (23)] were used in GRASP to examine complementary surfaces of the two molecules. The extreme logical responses of tissues and organisms. carboxyl-terminal residues of rhodopsin after S316 are not involved in G protein binding and occlude the In the inactive heterotrimeric state, GDP intracellular loops. These residues were removed from the GRASP view for clarity. This is the ground state is bound to the G␣ subunit. The three- of rhodopsin, and critical activating residues such as the ERY sequence are buried in the structure. The dimensional structures of the heterotri- loops making up the cytoplasmic face are somewhat disordered in the crystal structure. Still, there is an meric G proteins Gt and Gi (22, 23) show overall complementarity of shape between the sites of interaction that might already be used to guide the overall shape of the GDP-bound het- mutagenesis on G␣. Notice the overall charge complementarity and the more explicit charge comple- ␣ erotrimer and the residues on the surface mentarity between residues K341, K248, K141, and R147 on rhodopsin and D311 and E212 on G at the bottom of both molecules. Also notice that the deep pockets made up of the interhelical space in that can interact with other proteins. The ␣ ␤␥ ␣ rhodopsin (where the flexible carboxyl-terminal residues from G could bind?) and the cleft (where CIII amino terminal region of the subunit residues could fit?) may be complementary. and the carboxyl-terminal region of the ␥ subunit, which are both sites of lipid mod- ification, are relatively close together, sug- gions on G␣ are very close together in the alternative possibility is that the functional gesting a site of membrane attachment receptor-bound Gt. In the ground state of unit activating a G protein is a receptor (Fig. 1). Gt (1GOT), the closest approach is be- dimer. There is convincing evidence that a How does light-activated rhodopsin in- tween the side chains of Leu-32, Val-30, number of GPCRs do homo- and het- teract with transducin? Cai et al. (15) and and Ile-339 (Ͻ4 Å). This pocket of hydro- erodimerize (32). It is clear, therefore, Itoh et al. (14) used particular cysteines on phobic interactions couples the amino and that there is a compelling need for de- rhodopsin’s cytoplasmic face to crosslink carboxyl-terminal regions of G␣. Another tailed structural studies of the sort pio- transducin by using two different hetero- point of great interest is that with neered by Khorana’s group (13), as well as bifunctional crosslinking reagents, one crosslinkers on Cys-240 of the third intra- the studies using site-directed spin label- photoactivatable and one chemically pre- cellular loop, not much crosslinking was ing done in collaboration with Hubbell’s activated. They could crosslink transducin seen to G␤␥. Again, there is good evi- group (11, 16) to understand the nature of from a variety of different sites on the dence in the literature that receptors do the activated receptor–G protein complex. second and third intracellular loops with directly contact the G␤␥ subunit (28–30), Similar site-directed spin-labeling work each of these reagents, consistent with and recent studies implicate a different on the conformation of Gt when it is in current expectations that these loops are part of the receptor’s cytoplasmic surface: the complex is also needed. Of course, the involved in G protein binding. In the case ␣ helix 8, previously called intracellular crystal structure of the complex is eagerly where the crosslinkers derivatized Cys-240 loop 4 (20, 21). awaited! on the third intracellular loop, the sites on As seen in Fig. 2, the receptor-facing Rhodopsin also serves as a model re- transducin that had been crosslinked were analyzed by using mass spectroscopy to surface of Gt is large with respect to the ceptor for folding diseases of GPCRs (33), identify the insertion site. Interestingly, cytoplasmic surface of rhodopsin. A real because a large number of mutations the two crosslinkers yielded different in- puzzle in thinking about receptor–G pro- throughout rhodopsin’s primary sequence sertion points on transducin. The photo- tein contact is that in the ground states of have been found to cause retinitis pigmen- activatable reagent inserted in two sites at both rhodopsin and Gt (Fig. 2), the known tosa, a progressive retinal degenerative the carboxy terminus of the ␣ subunit, at points of contact are not all achievable, disease (34) resulting from misfolding and residues 342–345 at the extreme C termi- and therefore it is likely that large con- improper targeting of rhodopsin. In their nus (350 is the last residue in G␣t) and formational changes occur in both pro- fourth paper in the series, Hwa et al. (35) within the ␣4-␤6 loop at residues 310–313. teins to produce the active complex. For again used mass spectroscopic analysis to Both of these sites had been know recep- example, the carboxyl-terminal region is identify the cysteines involved in disulfide tor-binding sites (refs. 24–27; reviewed in shown to contact the third intracellular formation. Conserved cysteines at the ex- refs. 3 and 4), and the findings are impor- loop (14, 15), in agreement with Kostenis tracellular border of the molecule in the tant independent evidence of these con- and Wess (31), but the carboxy terminus large family of GPCRs take part in disul- tacts by using a different methodology. of G␣ also contacts ␣ helix 8 of rhodopsin fide formation and participate in the fold- The chemically preactivated crosslinker, (20, 21), which is close by (Fig. 2). How- ing process. In wild-type rhodopsin, the which typically inserts into the uncharged ever, the carboxyl terminus of G␥ also disulfide is between Cys-110 and Cys-187. ␧-amino groups of lysines, derivatized res- interacts with ␣ helix 8 of rhodopsin (20, Several rhodopsin mutants that are known idues 19–28 at the amino terminal helix of 21). Asn-343 on ␣ and Glu-66 on ␥ are to fold improperly were revealed to form ␣t. These data suggest that amino and removed from each other by 42 Å in the a disulfide bond between Cys-185 and carboxyl-terminal receptor-binding re- crystal structure of heterotrimeric Gt! An Cys-187. Formation of the improper di-

4820 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.011099798 Hamm Downloaded by guest on September 26, 2021 sulfide would disallow folding into the improper surface delivery of regulatory normalities in disease-causing mutations proper conformation, providing a molec- molecules (36). of rhodopsin. The understanding of how ular explanation for the misfolding of rho- These papers show how useful site- receptors fold and how they interact with dopsin in mutant products of disease al- directed cysteine mutagenesis of rho- and activate G proteins and other regu- leles. Such a mechanism may be relevant dopsin is proving for the investigation latory proteins has enormous implica- in other diseases caused by misfolding and of the structure of normal rhodopsin tions for physiology, pathology, and drug and the molecular basis of folding ab- design.

1. Scheer, A. & Cotecchia, S. (1997) J. Recept. Signal 14. Itoh, Y., Cai, K. & Khorana, H. G. (2001) Proc. 26. Bae, H., Anderson, K., Flood, L. A., Skiba, N. P., Transduction Res. 17, 57–73. Natl. Acad. Sci. USA 98, 4883–4887. Hamm, H. E. & Graber, S. G. (1997) J. Biol. Chem. 2. Gether, U. (2000) Endocr. Rev. 21, 90–113. 15. Cai, K., Itoh, Y. & Khorana, H. G. (2001) Proc. 272, 32071–32077. 3. Bourne, H. R. (1997) Curr. Opin. Cell. Biol. 9, Natl. Acad. Sci. USA 98, 4877–4882. 27. Bae, H., Cabrera-Vera, T. M., Depree, K. M., 134–142. 16. Altenbach, C., Yang, K., Farrens, D. L., Farahba- Graber, S. G. & Hamm, H. E. (1999) J. Biol. Chem. 4. Hamm, H. E. (1998) J. Biol. Chem. 273, 669–672. khsh, Z. T. & Khorana, H. G. (1996) Biochemistry 274, 14963–14971. 5. Wess, J. (1997) FASEB J. 11, 346–354. 35, 12470–12478. 28. Kisselev, O., Ermolaeva, M. V. & Gautam, N. 6. Schoneberg, T., Schultz, G. & Gudermann, T. 17. Chen, S., Lin, F., Xu, M., Hwa, J. & Graham, R. M. (1994) J. Biol. Chem. 269, 21399–213402. (1999) Mol. Cell. Endocrinol. 151, 181–193. (2000) EMBO J. 19, 4265–4271. 29. Taylor, J. M., Jacob-Mosier, G. G., Lawton, R. G., 7. Strange, P. (1999) Biochem. Pharmacol. 7, 1081– 18. Jordan, B. A. & Devi, L. A. (1999) Nature (Lon- Vandort, M. & Neubig, R. R. (1996) J. Biol. Chem. 1088. don) 399, 697–700. 271, 3336–3339. 8. Palczewski, K., Kumasaka, T., Hori, T., Behnke, 19. Franke, R. R., Konig, B., Sakmar, T. P., Khorana, 30. Yasuda, H., Lindorfer, M. A., Woodfork, K. A., C. A. & Motoshima, H. (2000) Science 289, 739– H. G. & Hofmann, K. P. (1990) Science 250, Fletcher, J. E. & Garrison, J. C. (1996) J. Biol. 745. 123–125. Chem. 271, 18588–18495. 9. Ovchinnikov, Y. A., Abdulaev, N. G., Feigina, 20. Ernst, O., Meyer, C., Marin, E., Henklein, P. & Fu, 31. Kostenis, E, C. & Wess, J. (1997) Biochem. J. 36, M. Y., Artamonov, I. D. & Zolotarev, A. S. (1982) W. (2000) J. Biol. Chem. 275, 1937–1943. 1487–1495. Bioorg. Khim. 8, 1011–1014. 21. Marin, E., Krishna, A., Zvyaga, T., Isele, J. 32. Kunishima, N., Shimada, Y., Tsuji, Y., Sato, T. & 10. Hargrave, P. A., Mcdowell, J. H., Curtis, D. R., & Siebert, F. (2000) J. Biol. Chem. 275, 1930– Yamamoto, M. (2000) Nature (London) 407, 971– Wang, J. K. & Juszczak, E. (1983) Biophys. Struct. 1936. 977. 9, 235–244. 22. Wall, M. A., Coleman, D. E., Lee, E., Iniguez- 33. Spiegel, A. M. (1996) Annu. Rev. Physiol. 58, 11. Farrens, D. L., Altenbach, C., Yang, K., Hubbell, Lluhi, J. A. & Posner, B. A. (1995) Cell 83, 143–170. W. L. & Khorana, H. G. (1996) Science 274, 1047–1058. 34. Berson, E. L. (1996) Proc. Natl. Acad. Sci. USA 93, 768–770. 23. Lambright, D. G., Sondek, J., Bohm, A., Skiba, 4526–4528. 12. Noel, J. P., Hamm, H. E. & Sigler, P. B. (1993) N. P. & Hamm, H. E. (1996) Nature (London) 379, 35. Hwa, J,. Klein-Seetharaman, J. & Khorana, H. G. Nature (London) 366, 654–663. 311–319. (2001) Proc. Natl. Acad. Sci. USA 98, 4872–4876. 13. Loewen, M. C., Klein-Seetharaman, J., Get- 24. Kostenis, E., Conklin, B. R. & Wess, J. (1997) 36. Kuznetsov, G. & Nigam, S. K. (1998) N. Engl. manova, E. V., Reeves, P. J., Schwalbe, H. & Biochemistry 36, 1487–1495. J. Med. 339, 1688–1695. Khorana, H. G. (2001) Proc. Natl. Acad. Sci. USA 25. Onrust, R., Herzmark, P., Chi, P., Garcia, P. D. & 37. Teller, D. T., Behnke, C., Palczewsk, K. & Sten- 98, 4888–4892. Lichtarge, O. (1997) Science 275, 381–384. kamp, R. (2001) Curr. Top., in press. COMMENTARY

Hamm PNAS ͉ April 24, 2001 ͉ vol. 98 ͉ no. 9 ͉ 4821 Downloaded by guest on September 26, 2021