Intramolecular allosteric communication in D2 receptor revealed by evolutionary amino acid covariation

Yun-Min Sunga, Angela D. Wilkinsb, Gustavo J. Rodrigueza, Theodore G. Wensela,1, and Olivier Lichtargea,b,1

aVerna and Marrs Mclean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX 77030; and bDepartment of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030

Edited by Brian K. Kobilka, Stanford University School of Medicine, Stanford, CA, and approved February 16, 2016 (received for review August 19, 2015) The structural basis of allosteric signaling in G -coupled led us to ask whether ET could also uncover couplings among receptors (GPCRs) is important in guiding design of therapeutics protein sequence positions not in direct contact. and understanding phenotypic consequences of genetic variation. ET estimates the relative functional sensitivity of a protein to The Evolutionary Trace (ET) algorithm previously proved effective in variations at each residue position using phylogenetic distances to redesigning receptors to mimic the ligand specificities of functionally account for the functional divergence among sequence homologs distinct homologs. We now expand ET to consider mutual informa- (25, 26). Similar ideas can be applied to pairs of sequence positions tion, with validation in GPCR structure and dopamine D2 receptor to recompute ET as the average importance of the couplings be- (D2R) function. The new algorithm, called ET-MIp, identifies evolu- tween a residue and its direct structural neighbors (27). To measure tionarily relevant patterns of amino acid covariations. The improved the evolutionary coupling information between residue pairs, we predictions of structural proximity and D2R mutagenesis demon- present a new algorithm, ET-MIp, that integrates the mutual in- strate that ET-MIp predicts functional interactions between residue formation metric (MIp) (5) to the ET framework. We used dopa- pairs, particularly potency and efficacy of activation by dopamine. mine D2 receptor (D2R), a target of drugs for neurological and Remarkably, although most of the residue pairs chosen for mutagen- psychiatric diseases (28), to test whether ET-MIp could elucidate the esis are neither in the binding pocket nor in contact with each other, allosteric functional communications from amino acid covariation many exhibited functional interactions, implying at-a-distance cou- patterns and resolve the evolutionary distance at which the allosteric pling. The functional interaction between the coupled pairs corre- pathways of D2R homologs are sufficiently conserved to detect res- lated best with the evolutionary coupling potential derived from idue−residue coupling signatures. D2R is expressed in the central sequences rather than with broader sets of GPCR nervous system and responds to dopamine, the major catecholamine sequences. These data suggest that the allosteric communication re- neurotransmitter. Canonical D2R signaling is effected by G class sponsible for dopamine responses is resolved by ET-MIp and best i/o G , which regulate ion channels (29, 30), MAPK kinases discerned within a short evolutionary distance. Most double mutants restored dopamine response to wild-type levels, also suggesting that (31), phospholipase C (32), and inhibition of (33). D1 class receptors (D1R and D5R) have lower affinities for tight regulation of the response to dopamine drove the coevolution – and intramolecular communications between coupled residues. Our dopamine (34 36) and activate adenylyl cyclase through Gs class G approach provides a general tool to identify evolutionary covariation proteins. To characterize allosteric communication between patterns in small sets of close sequence homologs and to translate them into functional linkages between residues. Significance

allostery | G protein-coupled receptors | residue covariation | Characterizing relationships among protein structure, function, Evolutionary Trace and evolution requires understanding the evolutionary con- straints on each constituent residue of a protein. Previous dentifying residues that coevolved to maintain or acquire fit- studies have shown that structural information can be re- Iness properties is critical for understanding protein structure, trieved from evolutionary residue covariation in protein fami- function, and evolution (1). Previous studies have shown that lies. However, whether the evolutionary history in protein covarying residue pairs, those that exhibit correlated amino acid sequences informs on functional interactions between non- changes in large multiple sequence alignments, tend to form adjacent residues has been unclear. Here, we developed a structural contacts (2–7), enhancing predictions of protein 3D method that uses evolutionary amino acid covariation to infer BIOPHYSICS AND structures (8–11). Covariation can also involve distal residues, functionally coupled residue pairs in the dopamine D2 receptor. but the function of these at-a-distance couplings is elusive and We discovered functional coupling between residue pairs that COMPUTATIONAL BIOLOGY has been attributed to background noise, alternative protein have coevolved mainly to control responses to dopamine and conformations, or subunit interactions of protein homooligomers maintain them at wild-type levels. Our findings demonstrate the (5, 7, 12). Alternately, distal covarying residue pairs could in- possibility of extracting the networks of intramolecular allosteric dicate allosteric couplings (6, 13–18). communication from evolutionary residue covariation patterns. The possibility of capturing intramolecular allosteric communi- Author contributions: Y.S., A.D.W., T.G.W., and O.L. designed research; Y.S. and A.D.W. cation by amino acid covariation analysis of protein family se- performed research; Y.S., A.D.W., and G.J.R. contributed new reagents/analytic tools; Y.S., quences has not been extensively explored. Nonproximal A.D.W., T.G.W., and O.L. analyzed data; and Y.S., A.D.W., T.G.W., and O.L. wrote thermodynamic coupling between correlated residue pairs was the paper. noted in 274 PDZ domains (14), but the relationship to allo- The authors declare no conflict of interest. stery is still debated (19, 20). It may be that distinctive allosteric This article is a PNAS Direct Submission. mechanisms, even among close homologs, limit the extraction Freely available online through the PNAS open access option. of allosteric couplings from sequences (13). Our previous identi- 1To whom correspondence may be addressed. Email: [email protected] or twensel@ fication of residues important for allosteric signaling within G bcm.edu. protein-coupled receptors (GPCRs) using Evolutionary Trace (ET) This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. (21–24) and strong conservation of some of the residues implicated 1073/pnas.1516579113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1516579113 PNAS | March 29, 2016 | vol. 113 | no. 13 | 3539–3544 Downloaded by guest on September 26, 2021 covarying pairs of residues ranked as important by ET (ET residue functional allostery in D2R, tuned to a specific ligand and sig- pairs), we examined functional coupling for ligand binding affinities naling bias, a more restricted alignment may be best. Accord- and downstream Gi activation induced by agonist-stimulated D2R. ingly, multiple alignments were tested (Class A, bioamine, dopamine, and D2Rs) and yielded distinct coupling scores Results (Table S2). ET-MIp Identified Pairs of Residue Positions with Evolutionary Covariation Patterns. We hypothesized that accounting for spe- Functional Interactions Between Covarying ET Residue Pairs Maintained cies divergence would improve the detection of functionally Dopamine Response at WT Level. To test whether the selected ET coupled residues over covariation analyses that ignore phyloge- residue pairs were functionally coupled, we first compared the netic information. ET-MIp adds the mutual information metric effects of single and double mutations on dopamine efficacy using (5) to ET to keep track of the phylogenetic distance at which a a fluorescence-based assay to study Gi activation induced by pair of residues vary (Fig. 1A; see Materials and Methods and agonist-stimulated D2R (Fig. 2; see Materials and Methods and Supporting Information for details). In ∼2,500 Class A GPCR Supporting Information for details). For five pairs (V83L2.53/ transmembrane (TM) domain sequences, we found that residue V91S2.61, M117F3.35/Y199F5.48, I48T1.46/F110W3.28,V152A4.42/ 4.61 3.42 5.54 pairs with high ET-MIp scores were more enriched for direct L171P , and N124H /T205M ), activation of Gi in re- contacts in a reference structure (PDB 2RH1) compared with sponse to dopamine was unexpectedly decreased or restored to a results obtained with leading alternative methods (5, 37, 38) (Fig. near-wild-type (WT) level in the double mutants, even though 1B), showing that GPCR phylogenetic information improves the one or both of the constituent single mutants showed signifi- coupling signal. Preliminary analysis of other protein families cantly enhanced response (Fig. 2 A and C). These results in- suggests that this result may be fairly general. This opens the dicate that covarying ET residue pairs help maintain dopamine possibility that ET-MIp also detects functionally relevant co- responses at the WT level. A trend was that functional coupling variation among structurally distant yet coupled residue pairs. was more apparent in pairs with high evolutionary coupling po- tential when calculated using sequences only from dopamine Residue and Sequence Selection. To test predicted couplings ex- receptors (Fig. 2C). For example, the loss-of-function mutation perimentally, we selected 10 covarying ET residue pairs in D2R L379F6.41 is rescued by T205M5.54, with which it has a strong in which one or both of the residues were involved in allosteric evolutionary coupling potential, but not by N124H3.42,withwhich pathways of D2R ligand responses (23), plus two more pairs. the coupling potential is weaker (Fig. 2 B and C). To estimate the Most of these are predicted to be functionally important by ET epistatic effect of double mutations, we used four standard models and alter function upon mutation (23). These pairs cover a range (product, logarithmic, minimal, and additive interaction models) of ET-MIp coupling scores and involve structurally distant resi- (39, 40). Except for the minimal model, the five residue pairs that dues (except T205M5.54/L379F6.41; see Table S1) whose cou- were functionally coupled to maintain WT dopamine responses plings, if any, would be allosteric. To probe the role of functional yielded higher epistasis scores, and the epistasis scores correlated coupling in discriminating between dopamine and serotonin better with the evolutionary coupling scores when calculated using during evolution, ET residues in D2R were mutated to the input sequences made up only of dopamine receptors (Fig. 2C and corresponding residues in the closely related 5-HT2A serotonin Figs. S1–S4). This suggests that the allosteric communication re- receptor (5-HT2AR), so only sites at which 5-HT2AR and D2R sponsible for dopamine response is a unique evolutionary signa- differ were included. As single mutations, these substitutions still ture of dopamine receptors. Compared with direct-coupling allow for a functional receptor (23). In addition to mutation analysis (DCA), Frobenius norm (FN), and direct information selection, we considered the choice of sequences used for cal- (DI) algorithms (37) (Tables S3 and S4), these ET-MIp results culating scores, because this can strongly impact ET-MIp cou- correlated better with experimental epistasis scores (Table S5). plings. For structural contacts, ET-MIp can be applied to Class A To assess the effect of functional coupling on dopamine po- GPCR sequences because they are all structurally similar. For tency, dose–response curves were generated to derive EC50 val- ues for WT and mutant D2Rs (Fig. 2 D and E and Table S6). We used relative ln(EC50) values, defined as the difference between ln(EC50 mutant)andln(EC50 WT), to approximate free energy 48-110 205-379 1.0 changes, because potency reflects differences in ligand binding DRD2 I F T L 110 affinity and/or activation kinetics, either of which must be de- I F T L 0.8 DRD3 termined by a free energy term. To assess nonadditivity, we DRD4 V L M V 48 0.6 ADA2C T Y MM compared the sum of relative ln(EC50) values of the single mu- T Y M I ADA2B 0.4 ET-MIp tants with the relative ln(EC50) value of the corresponding ADA2A T W M I DCA-FN double mutants (Fig. 2F), except when mutants showed almost DCA-DI DRD1 T W MM 0.2

True positive rate True MIp no response to dopamine. Unlike the three residue pairs with the DRD5 T W MM 205 379 nMI 0.0 lowest evolutionary coupling potential, the six pairs with the A F MM 5HT1A 0.0 0.2 0.4 0.6 0.8 1.0 highest evolutionary coupling potential were nonadditive—a 5HT2A T W M F False positive rate functional coupling that presumably fine-tunes the sensitivity to Fig. 1. ET-MIp decodes evolutionary correlations between residue posi- dopamine (Fig. 2F). Here again, deviation from additivity cor- tions. (A) Examples of residue positions displaying covariation patterns in the related most highly with the evolutionary coupling potential context of evolutionary trees. For simplicity, only residues from human se- derived from the dopamine receptor sequence set (Fig. 2G and quences are shown. Covarying residue pairs were mapped onto the D3R Fig. S5), and, overall, ET-MIp correlated better with non- structure (PDB 3PBL). Some have structural contacts (blue spheres), whereas additivity than DCA–FN and DCA–DI (Table S7). others are distant in the structure (red spheres). These positions are hy- pothesized to be functionally coupled during evolution. (B) Receiver oper- Rescue Effect on Serotonin Activation Was Observed for Some D2R ating characteristic curves comparing the performance of ET-MIp and other Double Mutants. Some of the studied ET residues were involved mutual information (MI)-based methods, MIp (5), normalized MI (nMI) (38), in discriminating against G activation induced by serotonin and DCA algorithms, FN and DI (37), in identifying residues in contact (within 16 stimulation of D2R (23). To test whether the covarying ET 6 Å) in the structure of the β2 (PDB 2RH1). Each method was applied to aligned sequences of ∼2,500 Class A GPCRs. The areas under residue pairs are functionally coupled with respect to regulating the curves, which indicate the accuracy of the prediction, are 0.75 for ET- agonist specificity, we compared the effects of single and double 1.46 MIp, 0.68 for DCA–FN, 0.65 for DCA–DI, 0.65 for MIp, and 0.56 for nMI. mutations on serotonin responses. The single mutants I48T ,

3540 | www.pnas.org/cgi/doi/10.1073/pnas.1516579113 Sung et al. Downloaded by guest on September 26, 2021 A D2R (WT) B D2R(WT) L379F C 7 ** ** Evolutionary Coupling Potential 8 V152A 3 N124H N124H/L379F ** L171P T205M T205M/L379F high low 6 V152A/L171P 2 6 4 ** ** Dopamine Dopamine 1 * 2 5 ** * * 0 0 *

Activation by dopamine 0 20 40 60 80 100 Activation by dopamine 0 20 40 60 80 100 4 * * * Time (s) Time (s) ** * * * D E * * * ** D2R (WT) D2R (WT) 3 * * 120 120 * F202L I105K * 100 Y213I 100 I195F * F202L/Y213I I105K/I195F 2 * ** * 80 80 ** * * * * ** 60 60 Relative activation by dopamine 40 40 1 * 20 20 * ** ** ** 0 0 0 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 I48T Percent of maximal response Percent of maximal response V83L V91S

I195F Y213I I105K

L379F L379F F202L L171P Y199F V191L L387C V152A S409N N124H N124H T205M T205M M117F M117F Log [Dopamine], M Log [Dopamine], M S193G C385M F110W

( EC50 ( EC50

F mutant A mutant B V83L/V91S I48T/F110W I105K/I195F

ln ( +ln ( F202L/Y213I

N124H/L379F T205M/L379F V152A/L171P V191L/S409N M117F/Y199F

EC50 EC50 M117F/L387C WT WT N124H/T205M S193G/C385M EC50mutant AB ( ln ( EC50WT * G H 3.23 5 7.36 I105K S409N 4.61 2.0 R 2.61 L171P ** = 0.762 V91S 4 * M117F/Y199F V191L5.40 F110W3.28 1.5 5.42 3 N124H/T205M V83L/V91S V83L2.53 S193G 3.35 L387C6.49 * * I48T/F110W M117F 2 * 1.0 5.44 C385M6.47 I195F F202L/Y213I 1.46 5.48 V191L/S409N I48T Y199F 1 0.5 V152A/L171P 5.51 M117F/L387C N124H3.42 F202L 5.54 0 I105K/I195F L379F6.41 T205M 0.0 4.42 (observed or additive ratio) Deviation additivity from V152A

-1 5.62 ln 20 40 60 80 100 Y213I Evolutionary coupling potential V83L/V91S I48T/F110W I105K/I195F F202L/Y213I V152A/L171P V191L/S409N M117F/Y199F M117F/L387C N124H/T205M (V83L) + (V91S) (I105K) + (I195F) (I48T) + (F110W) (F202L) + (Y213I) (V152A) + (L171P) (V191L) + (S409N) (M117F) + (Y199F) (M117F) + (L387C) (N124H) + (T205M)

Fig. 2. Functional coupling between covarying ET residue pairs was observed at Gi activation by dopamine. (A and B) Examples of D2R activation curves for covarying ET residue pairs. Membrane potential changes induced by Gi activation in response to dopamine stimulation of D2Rs were detected with the membrane potential assay. HEK293 cells stably expressing TRPC4β were transiently transfected with negative control [pcDNA3.1(+)], WT, or mutant D2R plasmids, loaded with potential-sensing dye, and stimulated with 10 μM dopamine after baseline fluorescence was read for 30 s. Each trace was normalized by receptor surface expression after subtraction of the signal of negative control cells. (C) Maximal activation of mutant D2Rs, normalized to WT (bars indicate mean ± SEM, n = 3–8). Mutants were compared with WT using one-sample t test against WT ≡ 1 (asterisks directly above bars), and mutants within a covarying group were compared with each other using one-way ANOVA followed by Bonferroni’s multiple comparison test (asterisks with brackets) (**P < 0.001; *P < 0.05). Bars are color-coded according to the evolutionary coupling potential predicted using amino acid sequences sharing >35% identity with D2R; see Table

S2 for values. (D and E) Dopamine dose–response curves for Gi activation were generated with the membrane potential assay as in A. Examples of nonadditive and additive effects of double mutations on the potency of dopamine are shown in D and E, respectively. (F) Dopamine EC50 values for WT and mutant D2Rs were determined by dose–response curves (details in Table S6) and used for the log-additive analysis. Results represent mean ± SEM (n = 3–7; **P < 0.001; *P < 0.05; independent two-tailed Student’s t tests). The color scale is as in C.(G) Evolutionary coupling potential, predicted as described for C, is plotted against

deviation from additivity jlnðEC50mutant A=EC50 WT Þ + lnðEC50mutant B=EC50 WT Þ − lnðEC50mutant AB=EC50 WT Þj using EC50 data shown in F. The two were highly cor- related (Pearson’s R = 0.762, P = 0.017). (H) Positions of the covarying ET residues, shown as spheres (Cα atoms), mapped onto the structure of D3R (PDB 3PBL). Different colors indicate different groups of covarying ET residue pairs. BIOPHYSICS AND COMPUTATIONAL BIOLOGY Y213I5.62, T205M5.54, L387C6.49, and I105K3.23 had enhanced Nonadditivity of Free Energy Changes upon Ligand Binding Was responses to serotonin, indicating that these positions in D2R Observed at Some D2R Double Mutants. To investigate whether participate in discriminating against Gi activation induced by covarying ET residue pairs interact to control ligand affinity, spi- 1.46 serotonin (Fig. 3C). A rescue effect was observed at I48T / perone competition binding experiments determined Ki for both F110W3.28 and N124H3.42/T205M5.54, even though F110W3.28 dopamine and serotonin and measured the energetic perturbations and N124H3.42 alone abolished activation by serotonin, sug- by calculating the Gibbs free energy change (ΔΔG0)(Table S8). In gesting functional coupling in controlling the specificity for D2R contrast to EC50, these measurements on whole cell membranes activation in these cases (Fig. 3 A−C). Given the small number of reflect low-affinity non-G protein-coupled binding. Significant dif- 0 0 0 functional coupling cases found, we infer that discriminating ferences between (ΔΔG A + ΔΔG B)andΔΔG AB observed for against serotonin is not the main functional role of the covarying F202L5.51/Y213I5.62 and T205M5.54/L379F6.41 revealed nonadditive ET residue pairs studied here. Moreover, this finding suggests effects on both dopamine (Fig. 4 A and C) and serotonin (Fig. 4 that the allosteric communication responsible for specificity of B and D) binding, indicating functional coupling in regulating receptor activation may vary across a protein family, making receptor−ligand binding affinity. Strikingly, F202L5.51/Y213I5.62 it difficult for ET-MIp to extract such a pattern from a broad and T205M5.54/L379F6.41 are outside the ligand binding pocket, sequence input. and yet alter receptor−ligand interaction (Fig. 2H), indicating

Sung et al. PNAS | March 29, 2016 | vol. 113 | no. 13 | 3541 Downloaded by guest on September 26, 2021 The most striking observations are that covarying ET residue A 10 D2R (WT) B 10 D2R (WT) I48T N124H pairs work together to modulate efficacy and potency of dopa- 8 F110W 8 T205M mine in D2R, and that the main function encoded by the allo- I48T/F110W N124H/T205M 6 6 steric communication is to maintain the dopamine response at 4 Serotonin 4 Serotonin the WT level and fine-tune sensitivity to dopamine. ET-MIp 2 2 identified functional couplings that regulate receptor−ligand in- 0 0 teractions through residue pairs outside the ligand binding pocket,

Activation by serotonin Activation by serotonin underscoring the allosteric nature of the pathways extending from 0 20 40 60 80 100 0 20 40 60 80 100 Time (s) Time (s) the ligand binding site. Covarying ET residue pairs control do- C Evolutionary Coupling Potential pamine responsiveness tightly, possibly reflecting an evolutionary need to maintain distinct dopamine affinities among dopamine high low – * * receptors (D1, D2, D3, D4, and D5) (34 36), each activating di- * * verse downstream signaling pathways. Coupling scores calculated 8 * * * at varying evolutionary depth indicate that coevolution coincides * ** 7 with functional separation of the subfamilies fairly late in evolu- * ** ** tion. The results of cosubstitution suggest that deviation in either 6 * direction from this highly tuned response conferred selective * evolutionary disadvantage. 5 There have been few studies of the functional relationships – 4 between covarying residues (14, 16 18, 41). Our results reveal * functional coupling of the covarying ET residues in D2R, and the * 3 ** synergistic or antagonistic features of the coupling. For example, * ** * F202L5.51 and Y213I5.62 have a synergistic effect on dopamine * 2 * * efficacy, whereas the effects on dopamine potency are antago- *

Relative activation by serotonin nistic (Fig. 2 C and F), suggesting that different interactions 1 * govern efficacy and potency. The observation that covarying ET * * * * ** * * ** ** * ** * ** ** * ** **** residue pairs mediate ligand-specific functional interactions 0 * ** ** ** without contacting the ligand directly supports the previous proposal of a conformational filter in D2R (23). I48T V83L V91S I195F Y213I I105K F202L L379F L379F Y199F V191L L171P L387C

V152A S409N Previous studies have proposed models of molecular switches T205M T205M N124H N124H M117F S193G M117F F110W C385M associated with receptor activation, such as the ionic lock (D/E)RY V83L/V91S I48T/F110W I105K/I195F

F202L/Y213I – N124H/L379F T205M/L379F V152A/L171P V191L/S409N M117F/Y199F

M117F/L387C (42 47), transmission switch (48), and tyrosine toggle switch N124H/T205M S193G/C385M (NPxxY) (44–47, 49). The discovery of functionally coupled Fig. 3. A rescue effect on serotonin responses was observed in some of the covarying ET residue pairs provides insight into the allosteric covarying ET residue pairs. (A and B) Membrane potential changes induced pathways connecting ligand binding, molecular switches, and G by Gi activation in response to serotonin stimulation of D2Rs were measured protein coupling in D2R. Based on the positions of the covarying and analyzed as in Fig. 2 A and B.(C) Maximal activation by serotonin of mutant D2Rs was normalized to that of WT, which was defined as 1. Bars, ET residue pairs mapped onto the structure of D3R (50), we can statistics, and color-coding are as described for Fig. 2C. classify the coupling mechanisms into three categories. (i) The covarying ET residues are far from each other (Table S1), with one at or near the ligand binding site and the other close to the their involvement in allosteric pathways linking ligand binding molecular switches, e.g., I48T1.46/F110W3.28.F1103.28 is at and receptor conformational change. Further, the free energy the orthosteric binding pocket (50, 51), and I481.46 is in direct change demonstrated that ET-MIp was able to detect energetic contact with D802.50, which interacts with the transmission switch coupling for ligand binding at both proximal (3.6 Å) and long- and NPxxY motif through water molecules in molecular dy- distance (13.2 Å) residue pairs, even when located far away from namics simulations of β1 and β2 adrenergic receptors (52), sug- the ligand binding site. Because the double mutations F202L5.51/ gesting a role for this pair in coupling ligand binding to switch Y213I5.62 and T205M5.54/L379F6.41 rendered both dopamine and residues. (ii) Both of the covarying ET residues are at or near the serotonin binding energetically more favorable (Fig. 4 C and D), ligand binding site, with one close to the 3–7lockswitch we did not find evidence for functional coupling responsible for (D1143.32–Y4167.43), which forms a link between TM3 and TM7 the specificity of ligand binding. that breaks upon receptor activation in (47, 50, 53). These include V83L2.53/V91S2.61 and M117F3.35/Y199F5.48.A 2.53 7.43 G Protein-Specific Effects. Our new assay for Gi activation let us side-chain rotamer of V83L can interact with Y416 3.35 compare mutational effects on D2R activation of Gi vs. G16 (23), (Fig. S6A) and an inward-facing side-chain rotamer of M117F and revealed several striking contrasts. For serotonin activation, could interact with D1143.32, if TM2 and TM3 move apart from 3.42 N124H greatly decreased maximal activation of Gi (Fig. 3C) each other (Fig. S6B). Their coupling may contribute to propa- but significantly increased maximal activation of G16. For maximal gation of the dopamine binding signal to the 3–7 lock. (iii) Both dopamine activation, T205M5.54, F110W3.28,I48T1.46,andV91S2.61 of the covarying ET residues are far from the ligand binding all displayed activation of G16 close to that of WT (23), but acti- pocket but close to the molecular switches or G protein- 5.51 5.62 3.42 vated the physiological partner of D2R, Gi,atlevelsrangingfrom binding region. These include F202L /Y213I , N124H / 2.5-fold to nearly fourfold higher than WT (Fig. 2C), implying T205M5.54,andT205M5.54/L379F6.41.F2025.51 is part of the somewhat different coupling mechanisms for these two G proteins. transmission switch, and the side chain of the mutant F202L5.51 can contact the side chain of I1223.40, which is part of the transmission Discussion switch (48) (Fig. S6C). Y2135.62 interacts with V2155.64,whichwas ET-MIp predicts functionally coupled residue pairs by weighing found to participate in receptor−G protein interactions (44, 46) mutual variations by the phylogenetic depth of the associated (Fig. S6C). N124H3.42, T205M5.54, and L379F6.41 may interact with evolutionary divergences. Experiments in D2R suggest the the residues making up the hydrophobic barrier in TM2, TM3, and method can reveal allosteric communication between distant and TM6 (47, 54). In addition, T2055.54 and L3796.41 are in direct contact evolutionarily important residue positions. with P2015.50 and F3826.44, respectively, of the transmission switch

3542 | www.pnas.org/cgi/doi/10.1073/pnas.1516579113 Sung et al. Downloaded by guest on September 26, 2021 D2R (WT) D2R (WT) continuous network of contacts leading from the transmission A B 5.54 6.41 T205M T205M switch to the G protein. The observation that T205M /L379F 120 120 L379F L379F are coupled in ligand binding is consistent with bidirectional com- 100 T205M/L379F 100 T205M/L379F 80 80 munication between ligand binding and G protein activation. The 60 60 idea that coupled residues are links in a chain of allosteric inter- 40 40 actions is supported by the observation that intervening resi- 20 20 dues contacting them tend to have high coupling scores (Fig. S7 H] spiperonebinding H] spiperonebinding 3 0 3 0 and Table S9). However, these residues are identical in D2R and % [ -8 -7 -6 -5 -4 -3 -2 -1 % [ -7 -6 -5 -4 -3 -2 -1 5-HT2AR and could not be tested in our paradigm. Log [Dopamine], M Log [Serotonin], M ThesidechainofM117F3.35 points toward the putative choles- C ∆∆G0 + ∆∆G0 Evolutionary Coupling Potential terol binding site (55, 56) and may enhance cholesterol interac- mutant A mutant B 5.48 − ∆∆G0 tions; Y199F points toward the putative receptor dimerization mutant AB high low interface(57,58).V1524.42 contacts cholesterol-contacting position 6 I1564.46 (56), and L1714.61 points toward the dimerization interface. 5.51 5.62 4 ThesidechainsofbothF202L and Y213I may lie at the * same interface. The functional coupling observed may reflect ef- 2 fects on modulation by cholesterol (59) and GPCR dimerization * 5.62 (57, 60). The equivalent residue to Y213 in activated β2 ad- 0 renergic receptor (46) stabilizes the “outward” movement of -2 TM6 through a van der Waals contact, consistent with effects of Y213I5.62 on G protein coupling. -4 Overall, the results support the idea that covariation patterns are signatures preserved in protein sequences during evolution

dopamine binding (kJ/ mol) -6 and reflect functional interactions important for fitness-confer- Change in Gibbs free energy upon Change in Gibbs energy free -8 ring properties. They open the possibility of improving structure- based drug design by accounting for intramolecular allosteric communication. The involvement of covarying ET residue pairs in allosteric pathways linking ligand binding, molecular switches, and G protein coupling offers the potential to reengineer allo- I105K/I195F I48T/F110W F202L/Y213I N124H/L379F V152A/L171P V191L/S409N T205M/L379F M117F/Y199F M117F/L387C N124H/T205M steric pathways of receptor activation. The divergent effects on (I105K) + (I195F) (I48T) + (F110W) (F202L) + (Y213I) G and G (23) activation observed in some mutants suggest that (V191L) + (S409N) (N124H) + (L379F) (V152A) + (L171P) (T205M) + (L379F) i 16 (M117F) + (Y199F) (M117F) + (L387C) (N124H) + (T205M) covarying ET residue pairs play a role in governing receptor D * 10 preference for downstream effectors. Thus, this work also serves 8 as a starting point for studies on bias in activation of effectors controlled by covarying ET residue pairs, and on the interpretation 6 of genome variations. 4 Materials and Methods

2 ** The key methods are briefly described here. For full details, please see Supporting Information. 0 Evolutionary Trace and MI Analysis. To identify evolutionarily coupled residue

serotonin binding (kJ/ mol) -2 pairs, we integrated MI into the ET framework as follows: Change in Gibbs free energy upon Change in energy Gibbs free

-4 XN 1 Xn ETðMIpði, jÞÞ = MIpgði, jÞ. n=1 n g=1

The alignment is broken up into subalignments g according to the phylo- I48T/F110W I105K/I195F F202L/Y213I genetic tree with N nodes. Our measure of mutual information, MIp (5), is T205M/L379F V152A/L171P V191L/S409N N124H/L379F M117F/Y199F M117F/L387C N124H/T205M computed for all possible residue pairs i and j for each subalignment selected (I48T) + (F110W) (I105K) + (I195F) (F202L) + (Y213I) (T205M) + (L379F) (V152A) + (L171P) (N124H) + (L379F) (V191L) + (S409N) (M117F) + (Y199F) (M117F) + (L387C)

(N124H) + (T205M) by the phylogenetic tree. The performance of ET-MIp in contact prediction

was compared with other methods as described in Supporting Information, BIOPHYSICS AND Fig. 4. Nonadditive effects on Gibbs free energy change upon ligand binding. 3 using an alignment of 2,500 Class A GPCRs to predict interresidue contacts in (A and B)The[ H]spiperone binding to WT or mutant D2Rs was detected in the COMPUTATIONAL BIOLOGY the structure of the β2 adrenergic receptor (PDB 2RH1). To identify covarying presence of various concentrations of competing dopamine or serotonin. Ex- residues in D2R, BLAST (Basic Local Alignment Search Tool) analysis of D2R amples of competition binding curves for dopamine and serotonin are shown in was first performed against the Uniref90 sequence database (61). To identify A and B, respectively. Nonlinear regression analysis was performed with the homologs, protein sequences were filtered by protein length and sequence one-site model to determine the IC50, which was then used in the Ki.(C and D) identity (>35%, >42%, >50%), where each alignment was respectively made Δ 0 Values for free energy change ( G ) upon dopamine or serotonin binding were up of all dopamine receptors, dopamine D2 and D3 receptors, and only D2 Δ 0 = = −1· −1 = derived from G RT lnKi,whereR 8.314 J K mol and T 298.15 K. receptors. The bioamine and Class A GPCR alignments were described pre- The mutational effects on free energy change upon binding to dopamine viously (23). (C)orserotonin(D) were examined by calculating ΔΔG0 of each D2R mutant as ΔΔG0 = ΔG0 − ΔG0 . Results of the log-additive analysis are mutant mutant WT Membrane Potential Assay. G activation induced by agonist-stimulated D2R expressed as mean ± SEM (n = 3–6; **P < 0.001; *P < 0.05; independent two- i triggers the opening of TRPC4β channels in HEK293 cells, leading to mem- tailed Student’s t tests). Color-coding of the bars is as described for Fig. 2C. brane potential changes (62). Details are given in Supporting Information.

ACKNOWLEDGMENTS. We thank Melina A. Agosto and Rhonald Lua for β (48) (Fig. S6D). A backbone hydrogen bond links L379F6.41 and constructive suggestions, and Michael X. Zhu for providing the TRPC4 -expressing 6.37 6.37 HEK293 cells. This work was supported by NIH Grants R01-GM066099, R01- the G protein interacting residue L375 (V331 in D3R) (46) EY011900, R01-EY007981, R01-GM079656, and T90-DK070109; by National Sci- 5.54 6.41 (Fig. S6E). Thus, T205M and L379F form the lynchpin in a ence Foundation Grant DBI-1356569; and by Welch Foundation Grant Q-0035.

Sung et al. PNAS | March 29, 2016 | vol. 113 | no. 13 | 3543 Downloaded by guest on September 26, 2021 1. Ortlund EA, Bridgham JT, Redinbo MR, Thornton JW (2007) Crystal structure of an 31. Luo Y, Kokkonen GC, Wang X, Neve KA, Roth GS (1998) D2 dopamine receptors ancient protein: Evolution by conformational epistasis. Science 317(5844):1544–1548. stimulate mitogenesis through pertussis toxin-sensitive G proteins and Ras-involved 2. Jones DT, Buchan DW, Cozzetto D, Pontil M (2012) PSICOV: Precise structural contact ERK and SAP/JNK pathways in rat C6-D2L glioma cells. J Neurochem 71(3):980–990. prediction using sparse inverse covariance estimation on large multiple sequence 32. Hernandez-Lopez S, et al. (2000) D2 dopamine receptors in striatal medium spiny alignments. Bioinformatics 28(2):184–190. neurons reduce L-type Ca2+ currents and excitability via a novel PLC[beta]1-IP3- 3. Nemoto W, Imai T, Takahashi T, Kikuchi T, Fujita N (2004) Detection of pairwise calcineurin-signaling cascade. J Neurosci 20(24):8987–8995. residue proximity by covariation analysis for 3D-structure prediction of G-protein- 33. Enjalbert A, Bockaert J (1983) Pharmacological characterization of the D2 dopamine coupled receptors. Protein J 23(6):427–435. receptor negatively coupled with adenylate cyclase in rat anterior pituitary. Mol – 4. Fuchs A, et al. (2007) Co-evolving residues in membrane proteins. Bioinformatics Pharmacol 23(3):576 584. 23(24):3312–3319. 34. Sunahara RK, et al. (1991) Cloning of the for a human dopamine D5 receptor – 5. Dunn SD, Wahl LM, Gloor GB (2008) Mutual information without the influence with higher affinity for dopamine than D1. Nature 350(6319):614 619. of phylogeny or entropy dramatically improves residue contact prediction. 35. Van Tol HH, et al. (1991) Cloning of the gene for a human dopamine D4 receptor with high affinity for the clozapine. Nature 350(6319):610–614. Bioinformatics 24(3):333–340. 36. Sokoloff P, Giros B, Martres MP, Bouthenet ML, Schwartz JC (1990) Molecular cloning 6. Burger L, van Nimwegen E (2010) Disentangling direct from indirect co-evolution of and characterization of a novel dopamine receptor (D3) as a target for neuroleptics. residues in protein alignments. PLOS Comput Biol 6(1):e1000633. Nature 347(6289):146–151. 7. Morcos F, et al. (2011) Direct-coupling analysis of residue coevolution captures native 37. Baldassi C, et al. (2014) Fast and accurate multivariate Gaussian modeling of protein contacts across many protein families. Proc Natl Acad Sci USA 108(49):E1293–E1301. families: Predicting residue contacts and protein-interaction partners. PLoS One 9(3): 8. Marks DS, et al. (2011) Protein 3D structure computed from evolutionary sequence e92721. variation. PLoS One 6(12):e28766. 38. Shannon CE (1997) The mathematical theory of communication. 1963. MD Comput 9. Hopf TA, et al. (2012) Three-dimensional structures of membrane proteins from ge- 14(4):306–317. nomic sequencing. Cell 149(7):1607–1621. 39. Mani R, St Onge RP, Hartman JL, 4th, Giaever G, Roth FP (2008) Defining genetic 10. Nugent T, Jones DT (2012) Accurate de novo structure prediction of large trans- interaction. Proc Natl Acad Sci USA 105(9):3461–3466. membrane protein domains using fragment-assembly and correlated mutation 40. Araya CL, et al. (2012) A fundamental protein property, thermodynamic stability, – analysis. Proc Natl Acad Sci USA 109(24):E1540 E1547. revealed solely from large-scale measurements of protein function. Proc Natl Acad Sci ł 11. Su kowska JI, Morcos F, Weigt M, Hwa T, Onuchic JN (2012) Genomics-aided structure USA 109(42):16858–16863. – prediction. Proc Natl Acad Sci USA 109(26):10340 10345. 41. Oliveira L, Paiva AC, Vriend G (2002) Correlated mutation analyses on very large se- 12. Weigt M, White RA, Szurmant H, Hoch JA, Hwa T (2009) Identification of direct res- quence families. ChemBioChem 3(10):1010–1017. − idue contacts in protein protein interaction by message passing. Proc Natl Acad Sci 42. Palczewski K, et al. (2000) Crystal structure of rhodopsin: A G protein-coupled re- USA 106(1):67–72. ceptor. Science 289(5480):739–745. 13. Livesay DR, Kreth KE, Fodor AA (2012) A critical evaluation of correlated mutation 43. Ballesteros JA, et al. (2001) Activation of the beta 2-adrenergic receptor involves algorithms and coevolution within allosteric mechanisms. Methods Mol Biol 796: disruption of an ionic lock between the cytoplasmic ends of transmembrane seg- 385–398. ments 3 and 6. J Biol Chem 276(31):29171–29177. 14. Lockless SW, Ranganathan R (1999) Evolutionarily conserved pathways of energetic 44. Choe HW, et al. (2011) Crystal structure of metarhodopsin II. Nature 471(7340): connectivity in protein families. Science 286(5438):295–299. 651–655. 15. Gloor GB, Martin LC, Wahl LM, Dunn SD (2005) Mutual information in protein mul- 45. Rasmussen SG, et al. (2011) Structure of a nanobody-stabilized active state of the β(2) tiple sequence alignments reveals two classes of coevolving positions. Biochemistry adrenoceptor. Nature 469(7329):175–180. 44(19):7156–7165. 46. Rasmussen SG, et al. (2011) Crystal structure of the β2 adrenergic receptor-Gs protein 16. Chakrabarti S, Panchenko AR (2009) Coevolution in defining the functional specificity. complex. Nature 477(7366):549–555. Proteins 75(1):231–240. 47. Standfuss J, et al. (2011) The structural basis of agonist-induced activation in consti- 17. Süel GM, Lockless SW, Wall MA, Ranganathan R (2003) Evolutionarily conserved tutively active rhodopsin. Nature 471(7340):656–660. networks of residues mediate allosteric communication in proteins. Nat Struct Biol 48. Deupi X, Standfuss J (2011) Structural insights into agonist-induced activation of 10(1):59–69. G-protein-coupled receptors. Curr Opin Struct Biol 21(4):541–551. 18. Halabi N, Rivoire O, Leibler S, Ranganathan R (2009) Protein sectors: Evolutionary 49. Fritze O, et al. (2003) Role of the conserved NPxxY(x)5,6F motif in the rhodopsin units of three-dimensional structure. Cell 138(4):774–786. ground state and during activation. Proc Natl Acad Sci USA 100(5):2290–2295. 19. Chi CN, et al. (2008) Reassessing a sparse energetic network within a single protein 50. Chien EY, et al. (2010) Structure of the human dopamine D3 receptor in complex with – domain. Proc Natl Acad Sci USA 105(12):4679–4684. a D2/D3 selective antagonist. Science 330(6007):1091 1095. 20. Tes¸ileanu T, Colwell LJ, Leibler S (2015) Protein sectors: Statistical coupling analysis 51. Wang C, et al. (2013) Structural basis for molecular recognition at serotonin receptors. – versus conservation. PLOS Comput Biol 11(2):e1004091. Science 340(6132):610 614. “ ” 21. Madabushi S, et al. (2004) Evolutionary trace of G protein-coupled receptors reveals 52. Vanni S, Neri M, Tavernelli I, Rothlisberger U (2009) Observation of ionic lock for- clusters of residues that determine global and class-specific functions. J Biol Chem mation in molecular dynamics simulations of wild-type beta 1 and beta 2 adrenergic receptors. Biochemistry 48(22):4789–4797. 279(9):8126–8132. 53. Trzaskowski B, et al. (2012) Action of molecular switches in GPCRs—Theoretical and 22. Shenoy SK, et al. (2006) beta-arrestin-dependent, G protein-independent ERK1/2 ac- experimental studies. Curr Med Chem 19(8):1090–1109. tivation by the beta2 adrenergic receptor. J Biol Chem 281(2):1261–1273. 54. Li J, Edwards PC, Burghammer M, Villa C, Schertler GF (2004) Structure of bovine 23. Rodriguez GJ, Yao R, Lichtarge O, Wensel TG (2010) Evolution-guided discovery and rhodopsin in a trigonal crystal form. J Mol Biol 343(5):1409–1438. recoding of allosteric pathway specificity determinants in psychoactive bioamine re- 55. Cherezov V, et al. (2007) High-resolution crystal structure of an engineered human ceptors. Proc Natl Acad Sci USA 107(17):7787–7792. beta2-adrenergic G protein-coupled receptor. Science 318(5854):1258–1265. 24. Kang HJ, Wilkins AD, Lichtarge O, Wensel TG (2015) Determinants of endogenous 56. Hanson MA, et al. (2008) A specific cholesterol binding site is established by the 2.8 A ligand specificity divergence among metabotropic glutamate receptors. J Biol Chem structure of the human β2-adrenergic receptor. Structure 16(6):897–905. 290(5):2870–2878. 57. Guo W, Shi L, Filizola M, Weinstein H, Javitch JA (2005) Crosstalk in G protein-coupled 25. Lichtarge O, Bourne HR, Cohen FE (1996) An evolutionary trace method defines receptors: Changes at the transmembrane homodimer interface determine activa- – binding surfaces common to protein families. J Mol Biol 257(2):342 358. tion. Proc Natl Acad Sci USA 102(48):17495–17500. 26. Katsonis P, Lichtarge O (2014) A formal perturbation equation between genotype 58. Guo W, Shi L, Javitch JA (2003) The fourth transmembrane segment forms the in- and phenotype determines the Evolutionary Action of protein-coding variations on terface of the dopamine D2 receptor homodimer. J Biol Chem 278(7):4385–4388. – fitness. Genome Res 24(12):2050 2058. 59. Zocher M, Zhang C, Rasmussen SG, Kobilka BK, Müller DJ (2012) Cholesterol increases 27. Wilkins AD, et al. (2013) Accounting for epistatic interactions improves the functional kinetic, energetic, and mechanical stability of the human β2-adrenergic receptor. Proc – analysis of protein structures. Bioinformatics 29(21):2714 2721. Natl Acad Sci USA 109(50):E3463–E3472. 28. Bonci A, Hopf FW (2005) The dopamine D2 receptor: New surprises from an old 60. Han Y, Moreira IS, Urizar E, Weinstein H, Javitch JA (2009) Allosteric communication friend. Neuron 47(3):335–338. between protomers of dopamine class A GPCR dimers modulates activation. Nat 29. Yan Z, Song WJ, Surmeier J (1997) D2 dopamine receptors reduce N-type Ca2+ cur- Chem Biol 5(9):688–695. rents in rat neostriatal cholinergic interneurons through a membrane-delimited, 61. Suzek BE, Huang H, McGarvey P, Mazumder R, Wu CH (2007) UniRef: Comprehensive protein-kinase-C-insensitive pathway. J Neurophysiol 77(2):1003–1015. and non-redundant UniProt reference clusters. Bioinformatics 23(10):1282–1288. 30. Einhorn LC, Oxford GS (1993) Guanine nucleotide binding proteins mediate D2 do- 62. Kang HJ, et al. (2014) Selectivity and evolutionary divergence of metabotropic glu- pamine receptor activation of a potassium channel in rat lactotrophs. J Physiol 462: tamate receptors for endogenous ligands and G proteins coupled to phospholipase C 563–578. or TRP channels. J Biol Chem 289(43):29961–29974.

3544 | www.pnas.org/cgi/doi/10.1073/pnas.1516579113 Sung et al. Downloaded by guest on September 26, 2021