Articles https://doi.org/10.1038/s41589-018-0115-3

Allosteric mechanisms underlie GPCR signaling to SH3-domain proteins through arrestin

Fan Yang1,2,3,4,9, Peng Xiao1,5,9, Chang-xiu Qu1,9, Qi Liu1,2,3,9, Liu-yang Wang6, Zhi-xin Liu1, Qing-tao He1,4, Chuan Liu4, Jian-ye Xu1, Rui-rui Li1, Meng-jing Li1, Qing Li3, Xu-zhen Guo2, Zhao-ya Yang1,4, Dong-fang He1, Fan Yi7, Ke Ruan8, Yue-mao Shen6, Xiao Yu3, Jin-peng Sun 1,4* and Jiangyun Wang2*

Signals from 800 G-protein-coupled receptors (GPCRs) to many SH3 domain-containing proteins (SH3-CPs) regulate impor- tant physiological functions. These GPCRs may share a common pathway by signaling to SH3-CPs via agonist-dependent arrestin recruitment rather than through direct interactions. In the present study, 19F-NMR and cellular studies revealed that downstream of GPCR activation engagement of the receptor-phospho-tail with arrestin allosterically regulates the specific conformational states and functional outcomes of remote β​-arrestin 1 proline regions (PRs). The observed NMR chemical shifts of arrestin PRs were consistent with the intrinsic efficacy and specificity of SH3 domain recruitment, which was controlled by defined propagation pathways. Moreover, in vitro reconstitution experiments and biophysical results showed that the receptor– arrestin complex promoted SRC kinase activity through an allosteric mechanism. Thus, allosteric regulation of the conforma- tional states of β​-arrestin 1 PRs by GPCRs and the allosteric activation of downstream effectors by arrestin are two important mechanisms underlying GPCR-to-SH3-CP signaling.

xtracellular signals received by G-protein-coupled receptors stabilization with an antibody or through extensive modification of (GPCRs) induce structural rearrangements in their cytoplasmic arrestin and its receptor26,27. Eregions, which are subsequently recognized by transducer G Alternatively, fluorescence spectroscopy and 19F-NMR have suc- proteins and in turn regulate the concentration of intracellular mes- cessfully been used to observe structural rearrangements in GPCR sengers1–6. Activated GPCRs are then phosphorylated by a group of signaling11,30–34. unnatural amino acid incorporation techniques GPCR kinases (GRKs)7–9, leading to the recruitment of a different combined with 19F-NMR, fluorescence spectroscopy and cellular type of transducer, arrestin, which then initiates another wave of approaches to study the mechanisms underlying arrestin-mediated cellular signaling10–16. In general, even a single type of GPCR can GPCR signaling to SH3 domain-containing proteins (SH3-CPs)35,36. initiate a broad range of physiological processes through arrestin Consequently, we not only determined the specificity of different engagement by scaffolding different downstream effectors11,14,17–22. PRs of arrestin in connecting different receptors to downstream Meanwhile, cellular studies suggest that conformational changes SH3-CPs and the mechanisms regulating their conformational in arrestin occur after GPCR activation, and arrestin retains its states through an allosteric mechanism, but also found that the active conformation even after dissociating from receptors11,23–25. active arrestin conformation promotes the activation of down- Consistent with these observations, recent crystallographic studies stream kinases containing SH3 domains by disrupting its autoin- have revealed that either the binding of arrestin to a phospho-recep- hibitory structural organization. tor C-tail or the direct fusion of a receptor to arrestin induces signif- icant structural rearrangements in arrestin26,27. Based on the current Results paradigm, receptor–arrestin complexes display two distinct confor- Arrestin mediates GPCR–SH3-CP coupling via three proline mations: (1) the ‘core’ or ‘snuggly’ conformation, which is impor- regions. Among the 825 GPCR sequences, 753 GPCRs (91.3%) do tant for desensitization of G-protein signaling, and (2) the ‘tail’ not contain the classic PR recognized by SH3-CPs (Supplementary or ‘hanging’ conformation, which facilitates G-protein activation Figs. 1–3; see also Supplementary Note 1). Although 72 GPCRs or G-protein-independent signaling10,15,28. Thus, conformational bearing the PR motifs were potentially able to directly interact states are an important aspect of arrestin function11,26,27,29. Despite with SH3-CPs, the remaining GPCRs might connect to SH3- this substantial research progress, the mechanisms by which con- CPs through other receptor-interacting proteins. We therefore formational states directly contribute to arrestin-mediated signal- postulated that many GPCRs might share a common pathway ing remain unclear, as high-resolution structures of active arrestin connected to a landscape of SH3-CPs via arrestin scaffolding, conformations have only been captured using crystallography via as most activated GPCRs bind to arrestin, which encodes three

1Key Laboratory Experimental Teratology of the Ministry of Education and Department of Biochemistry and Molecular Biology, Shandong University School of Medicine, Jinan, Shandong, China. 2Institute of Biophysics, Chinese Academy of Sciences, Beijing, Chaoyang district, Beijing, China. 3Key Laboratory Experimental Teratology of the Ministry of Education and Department of Physiology, Shandong University School of Medicine, Shandong, China. 4Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Peking University, Key Laboratory of Molecular Cardiovascular Science, Ministry of Education, Beijing, China. 5Key Laboratory of Chemical Biology, Ministry of Education, School of Pharmaceutical Science, Shandong University, Jinan, Shandong, China. 6Department of Molecular Genetics and Microbiology, School of Medicine, Duke University, Durham, NC, USA. 7Department of Pharmacology, Shandong University School of Medicine, Jinan, China. 8Laboratory for Physical Sciences at the Microscale and School of Life Sciences, University of Science and Technology of China, Hefei Anhui, China. 9These authors contributed equally: Fan Yang, Peng Xiao, Chang-xiu Qu, Qi Liu. *e-mail: [email protected]; [email protected]

876 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles type I PRs that are potential SH3-CP docking sites (Fig. 1b and and 11a). These results suggest that the phospho-barcode in the Supplementary Fig. 4a). receptor’s C-tail plays a critical role in arrestin-mediated GPCR– To test this hypothesis, we selected a panel of GPCRs that lack a SH3-CP coupling11,37. PR but exhibit diverse arrestin recruitment and G-protein subtype coupling characteristics (Supplementary Fig. 4b) and examined The phospho-barcode encodes functional PR conformations. their ability to recruit several known SH3-CPs through arrestin Based on the crystallographic and NMR results, the phospho- after agonist stimulation. We co-transfected plasmids encoding C-tails of β​2AR and V2R lie along both the surface of the three ele- HA-β​-arrestin 1, Flag-tagged GPCRs and different SH3-CPs into ments (β​-strand I, α-helix​ I and the C terminus) and the finger-loop HEK293 cells. After agonist stimulation, the β​-arrestin-1-mediated region of β​-arrestin 1 in the hanging conformation of the recep- recruitment of SH3-CPs was detected by co-immunoprecipitation tor–arrestin complex10,11,15,27,28,38. Importantly, all three functional (co-IP). Importantly, regardless of the G-protein subtype coupling PRs were located between 28 Å and 36 Å away from these recep- or arrestin-binding avidity, each receptor used arrestin to recruit tor-phospho-binding sites of β​-arrestin 1 (Supplementary Fig. 9a). 1–3 of the 5 SH3-CPs tested (Fig. 1a and Supplementary Fig. 5). Thus, allosteric regulation of β​-arrestin 1 occurs between β​-arrestin We then determined the specificity of the PR regions of β​-arres- 1’s receptor-phospho-binding sites and its remote PR regions. tin 1 in mediating receptor–SH3-CP interactions using co-IP We next incorporated the unnatural amino acid 3,5-difluoroty- experiments with different PR mutations. The results showed that rosine (F2Y) in specific positions of β-arrestin​ 1 and used 19F-NMR all three PRs in β​-arrestin 1 participated in SH3-CP recruitment to detect the conformational changes in the three PRs in response to (Supplementary Figs. 6 and 7). The PR 1 (P1) region played an different types of receptor phospho-C-tail stimuli (Supplementary important role in most of these interactions, except for the inter- Figs. 9 and 11a,b). These synthesized receptor phospho-C-tails action between SRC and SSTR2. The PR 2 (P2) region specifically induced the formation of active arrestin conformations that largely interacted with SRC by mediating its connections to β​2AR and mimicked the hanging conformation of the receptor–arrestin com- SSTR2 but not to V2R (Supplementary Fig. 7c). The PR 3 (P3) plex10,15,28. We used the F2Y technique because F2Y-incorporated region selectively bound to GRB2 (Fig. 1b and Supplementary arrestin causes minimal perturbations to the overall structure Figs. 4c, 6 and 7). In conclusion, all three PRs of β​-arrestin 1 are and provides a more significant chemical shift than traditional potential SH3-CP docking sites, each with different receptor and bromo-4(trifluoromethyl)acetanilide (BTFA)-labeled cysteine- downstream effector selectivity. negative arrestin (Supplementary Fig. 10a–c). The increase in the 19F-NMR signal was due to the sensitivity of the phenolic F2Y The phospho-barcode mediates β-arrestin-1–SH3-CP coupling. oxygen in response to environmental alterations (Supplementary Following GPCR activation, two distinct elements, the agonist- Fig. 10a–c). All F2Y-incorporated β​-arrestin 1 probes were func- occupied receptor seven-transmembrane (7-TM) core and the tional (Supplementary Fig. 10d–f). Intriguingly, the phospho-V2R- phospho-barcode localized to the C-tail or intracellular loops, have C-tail (V2Rpp) and the GRK6pp-β​2AR-CT induced prominent been shown to bind arrestin10,11,28,36,37. We examined the roles of dif- chemical shifts, whereas GRK2Bpp-β2AR-CT​ produced small ferent GRK subtypes using the prototype β​2AR as a model to assess but significant downfield-shifted resonances at the F87 and L123 the contribution of the receptor-phospho barcode. Overexpression positions, representing structural rearrangements in P1 and P2 of of GRK6, but not GRK2, effectively promoted arrestin-mediated β​-arrestin 1 (Supplementary Figs. 9c and 11a,b). In contrast, only the β​2AR–SH3-CP coupling in the co-IP assay (Fig. 1c–e). V2R phospho-C-tail induced a conformational change at the P3 site We then used the β​2AR mutant β​2ARΔICL3, which exhibits a (R177F2Y) (Supplementary Fig. 9c). Notably, GRK6pp-β​2AR-CT decreased ability to form the core interaction while maintaining and the phospho-V2R-C-tail did not induce any chemical shifts at the tail contacts10, and the β2AR​ C-terminal-specific phospho-site the F87, L123 or R177 positions in phospho-peptide-binding-defi- mutants S355A and E369A to further delineate the contributions cient mutants of β-arrestin​ 1 (Supplementary Fig. 12a–e). of the phospho-barcode and core interactions (Supplementary For the hanging conformation, previous studies have suggested Fig. 8a)11,28. In HEK293 cells, isoproterenol (ISO) stimulation sig- that the V2R and β​2AR phospho-C-tails bind to seven phosphate- nificantly increased the recruitment of β-arrestin​ 1 to the β​2ARΔICL3, binding pockets along the arrestin N terminus, with V2Rpp β​2AR-S355A and β​2AR-E369A mutants, although the levels were occupying all seven phosphate-binding sites, GRK6pp-β​2AR-CT slightly decreased compared to wild-type β​2AR (Supplementary showing a phospho-binding pattern of sites 1–5, and GRK2Bpp- Fig. 8b,c). We therefore used the sequential co-immunoprecipitation β​2AR-CT showing a phospho-binding pattern of sites 1, 4, 6, and method to characterize the interactions between SH3-CPs and the 7 (Supplementary Fig. 9b)11,15,26. For the snuggly conformation, the β​2AR–β-arrestin​ 1 complexes14. After ISO stimulation, wild-type or phospho-interaction pattern of sites 3, 4, 5, 6, and 7 is preserved and various mutants of β2AR​ expressed in HEK293 cells were initially named according to the V2R-C-tail interaction pattern, whereas purified with M1-Flag beads and eluted with Flag peptide, and the interactions at phospho sites 1–2 might be replaced with receptor– β​2AR–β-arrestin​ 1 complex was then further immunoprecipitated 7-TM core interactions. Therefore, we speculated that the observed with HA-agarose (Supplementary Fig. 8d). Although β​2ARΔICL3 phospho-patterns of the receptor-phospho-C-tails mostly reflect activation did not significantly decrease the association of SRC the hanging conformation. and FYN with β​-arrestin 1, the β​2AR phospho-site mutants abol- Based on our NMR results, the failure of GRK2Bpp-β​2AR-CT or ished these interactions (Fig. 1f,g and Supplementary Fig. 8d). The GRK6pp-β2AR-CT​ to induce a conformational change at R177F2Y β​2AR S355A mutant significantly reduced the interaction between indicated that phospho-binding sites 2–3 are important for the PLCγ​ and β​-arrestin 1 in response to ISO stimulation. Moreover, allosteric coupling of the phospho-receptor C-tail to the structural the E369A mutation of β2AR,​ which is a potential common interac- rearrangements at P3 in β-arrestin​ 1. Consistent with our hypoth- tion site for both the ‘hanging’ and ‘snuggly’ conformations of the esis, a phospho-peptide containing phospho-binding sites 1, 2, and GRK6-phosphorylated-β​2AR–β​-arrestin 1 complex, completely 3 (V2Rpp-2/3) induced a significant downfield shift at R177F2Y eliminated the interaction between β​-arrestin 1 and PLCγ​ (Fig. 1h). (Supplementary Fig. 9c). The coupling of phosphate-binding sites to Moreover, the inclusion of the GRK6-phosphorylated β​2AR C-tail PRs was more apparent at P1 and P2 because the mutation of E369 (GRK6pp-β2AR-CT),​ but not that of the GRK2-phosphorylated in GRK6pp-β2AR-CT​ to non-acidic residues eliminated phospho- C-tail (GRK2Bpp-β2AR-CT),​ biochemically recapitulated the site 5 interactions and completely abolished the chemical shifts at recruitment of the SH3 domain to active arrestin based on in vitro F87F2Y and L123F2Y (Supplementary Fig. 9d). Therefore, binding reconstitution and a GST-pull-down assay (Supplementary Figs. 8e at receptor-phospho-site-S5 triggered a conformational change in

Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology 877 Articles NaTuRe CHeMIcal BIoloGy

Agonist a b M3R, D2R … β2AR,SSTR2, V2R … >3 2–3 1.5–2 ns *** P *** *** P P P P P P P P *** P P ICL3 C-terminal P P P *** tail P ‘Snugly’ 4 *** *** ‘Hanging’ ** 3 ** *** β-Arr1 ** Recruitment Recruitment 2 HCK SRC P1 P2 P3 1 FYN Increased folds PLC Grb2 0 γ SRC SRC Grb2 D2R β2AR M3R SSTR2 V2R PLCγ FYN Grb2

c SRC deFYN PLCγ ## 6 ## 6 6 *** *** ## *** 4 4 4 *** *** ***

2 ns 2 ns 2 ns ### ### ##

Folds compared to basal

Folds compared to basal

Folds compared to basal 0 0 0 ISO ––+++– ISO ––++– + ISO –––+++ +pcDNA3.0 +GRK6 +GRK2 +pcDNA3.0 +GRK6+GRK2 +pcDNA3.0 +GRK6 +GRK2

fgns SRC FYN h PLCγ ns 4 *** 4 4 ns *** *** *** *** *** 3 3 3 # ns ns ns ns 2 2 2 ** ### ### ### ### ns

1 1 1 ###

Folds compared to basal

Folds compared to basal

Folds compared to basal 0 0 0 ISO –+––++– + ISO –––++++– ISO –––+++ – + WT ΔICL3 S355A E369A WT ΔICL3 S355AE369A WT ΔICL3 S355A E369A i jk 4 2 4 2 R = 0.90 R = 0.90 P2 GRK6pp GRK6pp P1 P3 3 3 P89 P121 F87 V2Rpp P120 L123 P91 P124 P178 V2Rpp P88 P175 2 GRK6pp-pS356S 2 GRK6pp-pS355S R177 P180 V2Rpp-2/3 GRK6pp-E369A Allosteric GRK6pp-pS356S 54° regulation GRK2Bpp 1 1 GRK2Bpp S5 S3 GRK6pp-E369A V2Rpp-2/3 S2

Pull down (folds to control) GRK6pp-pS355S Pull down (folds to control) 0 0 N domain C domain 00.4 0.81.2 1.6 0 0.20.4 0.60.8 1.0 Δ chemical shift (p.p.m.) Δ chemical shift (p.p.m.)

Fig. 1 | GPCRs signal to SH3-CPs via the 3 proline regions of β-arrestin 1. a, Activation of selected GPCRs recruited SH3-CPs to β-arrestin​ 1, which was examined using Co-IP. The bar graph presents quantitative analyses of interactions between β​-arrestin 1 and different SH3-CPs after the activation of different receptors by their corresponding agonists (see Methods and Supplementary Fig. 5 for further details). Data are presented as means ±​ s.d. from three independent experiments (n =​ 3). *P <​ 0.05; **P <​ 0.01; ***P <​ 0.005 (agonist-treated cells and PBS buffer-treated cells), one-way ANOVA. b, Graphic representation of the specificity of β​-arrestin 1 PRs in mediating GPCR–SH3-CPs coupling. c–e, Overexpression of GRK6 promotes β​-arrestin- 1-mediated SH3-CP recruitment, whereas GRK2 overexpression exerted significant opposite effects. The results of quantitative analyses of Co-IP between β​-arrestin 1 and SH3-CPs are shown as bar graphs, and representative western blots are shown in Supplementary Fig. 4d. f–h, Effects of the ‘snuggly’ conformation mutant (β​2ARΔICL3) and the ‘hanging’ conformation mutants (the S355A and E369A mutants of β​2AR) on β​-arrestin-1-mediated SH3-CP recruitment. Quantitative analyses of sequential co-IP for HEK293 cells transfected with Flag-β2AR​ (or its mutants), HA-β​-arrestin 1 and different SH3-CPs (see Methods and Supplementary Fig. 8d for further details). For c–e and f–h, values are presented as the mean ± s.d.​ of three independent experiments (n = 3).​ Significant differences between ISO-treated and untreated cells (**P <​ 0.01; ***P <​ 0.005; ns, no significant difference), different β​2AR mutants and wild-type β​2AR, or cells transfected with the selected GRK and pcDNA3.0 (##P <​ 0.01; ###P < 0.005)​ were analyzed by one-way ANOVA. i, A best-fit linear correlation of the receptor phospho-C-tail-induced chemical shifts at the F87F2Y site with the binding efficacy of the SRC SH3 domain at the P1 site of β​-arrestin 1 (Supplementary Fig. 13b). j, Correlation of the receptor phospho-C-tail-induced chemical shifts at the L123F2Y site with the binding efficacy of SRC SH3 domain at the P2 site of β​-arrestin 1 (Supplementary Fig. 13c). In i and j the error bar along the y axis for each point indicates the deviation of the pull-down assay results derived from three independent experiments (n =​ 3). k, Structural representation of the allosteric regulation between specific phospho-binding sites and remote PRs (PDB: 4JQI).

878 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

ae 17° β-Arr1 N domain β-Arr1 N domain P2(PxxP) L123 P120 Core III I119 Y21 V8 β-Arr1 P121 P36 P36 β-Arr1 I119V34 N domain 6 V40 V37V37 β Core I N domain Core II F87 Core II 5 F27 L22 β P114 Core II L100 L22 L166

Core III L104 Core I P1(PxxP) I105 L108 10 Loop 6 P89 β Y113 F9 4 β2 F9 β GRK6pp-S5 Loop8 7 β1 53° GRK6pp-S5 β C domain β3 β2 Loop3 Loop 7 β1

b F87F2Y V40E F87F2Y P114E F87F2Y f L123F2Y P36E L123F2Y V34R L123F2Y

No peptide No peptide

–0.79 –0.19 –0.13 –0.76 –0.02 –0.20

Δ shift 0.60 Δ shift 0.66 GRK6pp Δ shift 0.56 Δ shift 0.74 GRK6pp

–132 –138 –132 –138 –132 –138 –132 –138 –132 –138 –132 –138 19F-NMR (p.p.m.) 19F-NMR (p.p.m.)

cgP114F2Y V40E P114F2Y V40F2Y V34F2Y P36E V34F2Y P36F2Y

No peptide No peptide

–0.51 –0.24 –0.28 –0.13 GRK6pp Δ shift 0.01 GRK6pp Δ shift 0.02

–132 –138 –132 –138 –132 –138 –132 –138 –132 –138 –132 –138 19F-NMR (p.p.m.) 19F-NMR (p.p.m.)

dh

P2(PxxP)

P1 (PXXP) Y21

V40 V8 V34 GRK6pp-S5 P36 P114 GRK6pp-S5 F9

Inactive Inactive Active Active

Fig. 2 | Residue contacts govern the propagation pathway from phospho-binding site 5 to conformational changes in the P1 and P2 region. a, Cartoon representation of β​-arrestin 1 residue contacts between phospho-binding site 5 and the P1 region at the residue level (right) and secondary structure level (left) (PDB: 4JQI). b, Effects of the mutation of key residue contacts (V40E and P114E) on the 19F-NMR spectra of β​-arrestin 1 F87F2Y before and after GRK6pp-β2AR-CT​ (GRK6pp) binding. c, Effect of the mutation of key upstream residue contacts (V40E) on the 19F-NMR spectra of β​-arrestin 1 P114F2Y. d, Visualization of the propagation pathway (in the V8-Y21-V40-P114-P1 region) mediating the conformational change observed in the P1 region of β​-arrestin 1 in response to phospho-site-5 binding. Active (PDB: 4JQI, colored in golden yellow) and inactive (PDB: 1G4M, colored in gray) crystal structures of β​-arrestin 1 were compared. e, Cartoon illustrating of β​-arrestin 1 residue contacts between phospho-binding site 5 and the P2 region at the residue and secondary structure levels (PDB code 4JQI). f, Effects of the mutation of key residue contacts (V34R and P36E) on the 19F-NMR spectra of β​-arrestin 1 L123F2Y before and after GRK6pp-β​2AR-CT (GRK6pp) binding. g, Effect of the mutation of key upstream residue contacts (P36E) on the 19F-NMR spectra of β​-arrestin 1 V34F2Y before and after GRK6pp-β​2AR-CT (GRK6pp) binding. h, Visualization of the propagation pathway (F9-P36- V34-P2 region) relating phospho-site-5 binding to the conformational change observed in the P2 region of β-arrestin​ 1. Active (PDB: 4JQI, colored in golden yellow) and inactive (PDB: 1G4M, colored in gray) crystal structures of β​-arrestin 1 were compared.

both P1 and P2 (Supplementary Fig. 9b–d). Additionally, the chem- β​-arrestin 1 PR mutants, as determined by GST pull-down assays ical shifts at F87F2Y and L123F2Y observed in response to the bind- (R2 =​ 0.9 and R2 =​ 0.9, respectively; Fig. 1i,j and Supplementary ing of different receptor phospho-C tails correlated with the efficacy Fig. 13a–c). This correlated linear trend indicated that arrestin of the interaction between the SRC SH3 domain and specific receives and converts information from specific receptor-phosphate

Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology 879 Articles NaTuRe CHeMIcal BIoloGy

a F87F2Y b L123F2Y c L177F2Y d N375F2Y 300 +GRK6pp +GRK6pp 50 +GRK6pp 50 +GRK6pp *** 80 * ns +GRK6pp+SRC +GRK6pp+SRC +GRK6pp+SRC ** +GRK6pp+SRC 40 40 60 200 ns ns HZ) 30 * 30 *** 40 ns ns 20 20

100 *

Full width half half width Full

Δ HZ) ( maxima

Full width half width Full

Full width half width Full

Δ HZ) (

## maxima

Full width half width Full

Δ HZ) ( maxima

Δ ( maxima maxima 20 *** ### ns 10 10 ### 0 0 0 0 021435 021435 02143 5 021435

Cr(III) (mM) Cr(III) (mM) Cr(III) (mM) Cr(III) (mM)

f e 4 R2 = 0.94 +GRK6pp SRC P2 P1 GRK6pp P1 P2 3 P3 PLCγ FYN

L123 L123 P3 F87 + F87 2 R177 R177

C-tail Pull down N375 C-tail N375 YES

(folds to control ) 1

0 0 0.20.4 0.6 Δ chemical shift (p.p.m.)

+FYN +SRC +PLCγ

P2 SRC PLCγ FYN P2 SRC P2 P2 P3 L123 P3 P3 L123 P3 P1 F87 P1 L123 P1 L123 F87 P1 F87 F87 R177 R177 R177 R177 N375 C-tail N375 N375 C-tail C-tail N375 C-tail

Fig. 3 | 19F-NMR spectra and paramagnetic titration revealed structural alterations in the β-arrestin 1 PRs in response to the binding of SRC or different SH3 domains. a–d, Paramagnetic titration experiments of F87F2Y (a), L123F2Y (b), L177F2Y (c) and N375F2Y (d) sites in response to GRK6pp-β​2AR-CT (GRK6pp) and SRC (see Methods and Supplementary Figs. 18–21 for further details). Data represent the mean ± s.d.​ of three independent experiments (n = 3).​ Statistical differences were determined by One-way ANOVA. *P <​ 0.05; **P <​ 0.01; ***P <​ 0.005, GRK6pp incubation was compared with control vehicles; ##P <​ 0.01; ###P < 0.005,​ SRC incubation was compared with non-SRC incubation; ns, no significant differences. e, A best-fit linear correlation of different SH3 domain-induced chemical shifts at the F87F2Y site with the binding avidity of the SRC SH3 domain at the P1 site of β-arrestin​ 1. The error bar along the y axis for each point indicates the deviation of the pull-down assay results between three independent experiments (n = 3)​ (Supplementary Fig. 22i). f, A cartoon illustration of structural rearrangements in β​-arrestin 1 occurring in response to the receptor phospho-C-tail interaction and subsequent SRC or SH3 domain binding. binding into explicit structural responses. Taken together, our data receptor (Supplementary Fig. 14b,c). The structural integrity of revealed allosteric coupling between specific receptor-phospho- these mutants was further verified by circular dichroism (CD) binding sites and functionally relevant arrestin conformational spectroscopy (Supplementary Fig. 14d). Although GRK6pp-β​2AR- states (specifically, phospho-binding site 5 coupling to P1 and P2 CT stimulation caused a significant downfield shift (0.79 p.p.m.) at and phospho-binding sites 2–3 to the P3 region of β​-arrestin 1). The the F87F2Y position, the V40E and P114E mutations significantly magnitude of the 19F-NMR chemical shift reflected the intrinsic effi- decreased this chemical shift by 0.60 and 0.66 p.p.m., respectively. cacy of β-arrestin​ 1 for recruiting downstream SH3-CPs (Fig. 1i–k). Therefore, the V40E and P114E mutations markedly decreased the GRK6pp-β​2AR-CT-induced conformational changes that Precise allosteric propagation path connects to P1 conforma- occurred at the β​-arrestin 1 P1 region (Fig. 2b). Moreover, V40E tion. The coupling between receptor-phospho-site S5 binding and and P114E specifically disrupted the recruitment of the SRC SH3 the structural rearrangement of the P1 and P2 regions prompted us domain to β​-arrestin 1 in response to GRK6pp-β​2AR-CT but had to determine the underlying signal propagation pathway. Phospho- no significant effect on the interaction with clathrin after V2Rpp binding site S5 connects to the P1 region of arrestin via three hydro- binding via GST-pull-down assay (Supplementary Fig. 15a–d). phobic core regions (Fig. 2a). In response to V2Rpp binding, only In contrast, mutating these knobs to hydrophobic residues (for a slight reorganization of β-arrestin​ 1 residue contacts has been example, V40W and P40W) or mutating other residues proximal observed using crystallography26, including the relaxation of core to the connecting region (for example, L108 and Y113) had no sig- 1 and core 3, as well as the assembly of core 2 (Supplementary nificant effect on the β​-arrestin-1–SRC SH3 domain association Fig. 14a). These small changes observed in crystal structures are (Supplementary Fig. 15e–j). attributed to the lowest energy structural states, which are often We next monitored the conformational changes in V40 and captured by crystallography. P114 of β​-arrestin 1 in response to GRK6pp-β​2AR-CT stimu- We mutated the connecting knobs V40 or P114 to the polar lation. Incubation with the GRK6pp-β​2AR-CT caused signifi- residue E, which maintains its ability to associate with the active cant downfield shifts at both the V40F2Y and P114F2Y positions

880 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

a SRC SRC+β-Arr1 SRC+β-Arr1 Anti-pY416 55 kDa +p- 2AR 10 β ### Anti-SRC 55 kDa 8 *** Anti-β-Arr1 55 kDa ### 6 *** Anti-Flag 55 kDa *** *** 4 *** Anti-pS355/356 55 kDa 2

Time (min) 0103060010 30 60 0103060 basal over folds pY416 0 β-Arr1 ––––++++++++ 0103060 0103060 0103060 p-β2AR –––– ––––++++ Time (min)

bc SRC SRC+β-Arr1 SRC+β-Arr1 SRC+β-Arr1 P1P2 5 +p-β2AR +p-β2AR 300 ### *** 10 ) 240 4

m ### K / 8

cat 180 *** 3 *** 6 120 ns 2 *** 4

ns

k (

RR-SRC RR-SRC 60 1 2

Folds compared to basal to compared Folds

s toward activity Enzyme 0 0 basal over folds pY416 0 0030 30 030 030 SRC β-Arr1 β-Arr1 Time (min)

SRC+ SRC+ -Arr12AR P1P2 +p-β2AR β +p- SRC+β

Fig. 4 | A phospho-receptor/arrestin complex promoted the activation of SRC. a, Time course of the SRC tyrosine phosphorylation level at the pY416 site. Representative western blots from three independent experiments are shown on the left. The tyrosine phosphorylation level of SRC were qualified by densitometric analysis and expressed as fold above basal in the right panel. Data represent the mean ± s.d.​ of three independent experiments (n =​ 3). Statistical differences were determined by one-way ANOVA. ***P <​ 0.005; β​-arrestin 1 or GRK6-phosphorylated-β​2AR–β​-arrestin 1 complex treated were compared to control vehicles; ###P < 0.005;​ the GRK6-phosphorylated-β​2AR–β​-arrestin 1 complex treated was compared to β​-arrestin 1 alone treated. (Full blots shown in Supplementary Fig. 39d). b, The effects of β​-arrestin 1 or the GRK6-phosphorylated β​2AR–β​-arrestin 1 complex on the SRC kinase 48 activity. The kinetic parameters (kcat/Km) are expressed as a bar graph (see Methods and Supplementary Fig. 25c) . Values are means ±​ s.d. from three independent experiments (n =​ 3). Statistical differences were determined by one-way ANOVA. ***P <​ 0.005; ns, no significance; β​-arrestin 1, GRK6- phosphorylated-β​2AR–β​-arrestin 1 complex or GRK6-phosphorylated-β​2AR–β​-arrestin 1 P1P2 mutation complex treated were compared to control solutions. c, The effects of P1–P2 region of β​-arrestin 1 on the GRK6-phosphorylated-β​2AR–β​-arrestin 1 complex regulated SRC autophosphorylation at the pY416 site. Representative western blots from three independent experiments are shown in Supplementary Fig. 25a. Values are means ±​ s.d. from three independent experiments (n =​ 3). ***P <​ 0.005; ns, no significant difference (addition of control solution and β​-arrestin 1, GRK6-phosphorylated- β​2AR–β-arrestin​ 1 complex or GRK6-phosphorylated-β​2AR–β​-arrestin-1–P1P2 complex at 30 min); ###P < 0.005​ (addition of β​-arrestin 1 and GRK6-phosphorylated-β​2AR–β​-arrestin 1 complex at 30 min), one-way ANOVA.

(Fig. 2c). Moreover, the V40E mutation completely abolished the We then examined the conformational changes at V34 and P36 chemical shift at the P114F2Y position in response to GRK6pp-β​ in β​-arrestin 1. In response to GRK6pp-β2AR-CT​ stimulation, both 2AR-CT stimulation (Fig. 2c). Based on these results, the binding the P36F2Y and V34F2Y positions exhibited significant downfield of GRK6pp-β​2AR-CT to phospho-binding-site 5 in β​-arrestin 1 shifts (Fig. 2g). Moreover, the P36E mutant completely eliminated induced a conformational change in the P1 region through a propa- the chemical shift at the V34F2Y position after GRK6pp-β​2AR-CT gating path of V40→​P114→​F87 (Fig. 2d). administration (Fig. 2g). Therefore, the binding of GRK6pp-β​2AR- CT to phospho-binding site 5 of β​-arrestin 1 instigated a conforma- Allosteric path connecting to the P2 region. Diverging by tional change in the P2 region through a propagating path of P36→​ approximately 50 degrees compared to the P1 region, the P2 region V34→​F123 (Fig. 2h). Taken together, these results provide direct is connected to receptor-phospho-site-S5 binding site through the biophysical and biochemical evidence that explicit pathways exist β​3 sheet and loop 3 of β​-arrestin 1 (Figs. 1k and 2e). Hydrophobic to control functionally relevant arrestin conformations via allosteric interactions dominate the connection between phospho-site 5 regulation by receptors (Fig. 2d,h). and the P2 region, which also contains three hydrophobic cores (Supplementary Fig. 16e). Disruption of these hydrophobic cores The recognition of PR conformations by downstream effectors. with the V34R or P36E mutation significantly decreased the We next used 19F-NMR spectroscopy to investigate the confor- downfield shifts at the P2 probe L123F2Y position and decreased mational states of PRs after the formation of the ternary receptor SRC recruitment in response to GRK6pp-β2AR-CT​ engagement -phospho-tail–β​-arrestin-1–SH3-CP complex. Specifically, com- (Fig. 2f and Supplementary Fig. 16a–d). In contrast, these muta- plexes consisting of F2Y-incorporated β-arrestin​ 1, SRC and the tions had no significant effect on V2Rpp-induced clathrin bind- receptor phospho-tail (in a 1:1:3 molar ratio) were reconstituted ing to arrestin, their recruitment to the active β​2AR or the CD in vitro and subjected to 19F-NMR analysis. Binding of the SRC spectrum, suggesting their structural integrity (Supplementary induced significant upfield shifts at F87F2Y and L123F2Y but not Figs. 16a–d and 14c–e). at R177F2Y (Supplementary Fig. 17). We examined the effect of the

Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology 881 Articles NaTuRe CHeMIcal BIoloGy

a TA-bimane HC C N E N R L L N L KPN O O KPN RS mRNA N N N H N3 H N N C C KPN O Cu(I) O HC C AUC N AUC UAG N N O O KPNtRNA Ribosome SRC-164KPN SRC-164KPN-TAb

b 164KPN-TAb/397W ### d 164KPN-TAb/397W+pβ2AR β5-FlAsH-αD 2.0 164KPN-TAb/397W+β-Arr1 2.0 *** 164KPN-TAb/397W+pβ2AR+β-Arr1 164KPN-TAb/397W+pβ2AR+β-Arr1 P1P2 Rluc 1.5 * 1.5 ns ns

1.0

to control

to control 1.0

Folds compared

Folds compared 0.5 β-Arr1 –++ –– 0.5 p 2AR 460 480 500 520 β ––+ ++ β-Arr1 P1P2 –––+ – Wavelength (nm) βB-FlAsH-αA

c 93KPN-TAb/289W ### e 93KPN-TAb/289W+pβ2AR B-FlAsH- A 5-FlAsH- D 93KPN-TAb/289W+ -Arr1 β α β α 2.5 β 0.010 93KPN-TAb/289W+pβ2AR+β-Arr1 2.5 *** 93KPN-TAb/289W+pβ2AR+β-Arr1 P1P2 2.0 2.0 0.000 1.5 * 1.5 ns ns –0.010 to control 1.0

to control

Δ net BRET net 1.0 Folds compared

Folds compared –0.020 0.5 *** β-Arr1 –++ –– β-Arr1-WT 0.5 –0.030 ### p 2AR ––+ ++ β-Arr1-P1P2 460 480 500 520 β ### β-Arr1-P114E Wavelength (nm) β-Arr1 P1P2 ––––+

Fig. 5 | A phospho-receptor–arrestin complex promoted disassembly of the autoinhibitory domains of SRC. a, Schematic flowchart for the development of fluorescent probes to study SRC conformational changes via the incorporation of 2-amino-6-(((3-azidopropoxy)carbonyl)amino)hexanoic acid (KPN) in SRC and successive chemical labeling (see Methods for further details) RS, aminoacyl-tRNA synthetase. b,c, Fluorescence spectra for TAb-labeled 164KPN-TAb/397 W (b) and 93KPN-TAb/289 W (c) SRC mutants in the presence of control solutions (black), GRK6-phosphorylated β​2AR (pβ​2AR, golden yellow), β​-arrestin 1 (β-Arr1,​ blue), GRK6-phosphorylated β​2AR–β​-arrestin 1 complex (pink), or GRK6-phosphorylated β​2AR–β​-arrestin-1–P1P2 mutant (β-Arr1​ P1P2) complex (green). Representative fluorescence spectra with the signal normalized to 100% for the SRC-N164KPN/N397W (b) or SRC-E93KPN/T289W (c) are shown on left panels (see also Supplementary Fig. 26c,d). The maximal responses of each spectrum was quantified and depicted as bar graph on the right panel. Data are presented as mean ±​ s.d. from five independent experiments (n =​ 5). Statistical differences between control solutions and GRK6-phosphorylated β​2AR, β​-arrestin 1, GRK6 phosphorylated-β​2AR–β​-arrestin 1 complex or GRK6-phosphorylated β​2AR–β​ -arrestin-1–P1P2 mutant complex (*P <​ 0.05; ***P <​ 0.005; ns, not significant), and between wild-type GRK6 phosphorylated-β​2AR–β​-arrestin 1 complex and GRK6-phosphorylated β​2AR–β​-arrestin-1–P1P2 mutant complex (###P < 0.005)​ were determined by one-way ANOVA. d, Schematic representation of the SRC-FlAsH constructs used for intramolecular BRET assays (depicted according to the crystal structure; PDB code 2SRC). e, Quantification of ISO evoked intramolecular BRET signal changes in SRC. Data are presented as the mean ±​ s.d. from three independent experiments (n =​ 3). Statistical differences between control solutions and wild-type β​-arrestin 1 (***P <​ 0.005), and between wild-type β​-arrestin 1 and the β​-arrestin-1–P1P2 mutant or β​-arrestin-1–P2-P114E mutant (###P < 0.005),​ were determined by one-way ANOVA.

paramagnetic relaxation agent chromium acetylacetone (Cr(iii)) to complex (Fig. 3d and Supplementary Fig. 21). Thus, SRC recog- further characterize the environment of F87F2Y and L123F2Y. The nized specific PR conformational states in β-arrestin​ 1 but not dis- signal-to-noise (S/N) ratios for 19F-NMR spectra at different F2Y lodgement of the C-tail. incorporation sites were determined (Supplementary Figs. 18–21). We next investigated the structural rearrangements in the Specifically, although the 19F-NMR spectra revealed that all three PRs in response to different SH3 domain interactions. Whereas F2Y sensors located in different PRs were more susceptible to line all three SH3 domains induced significant upfield shifts at the broadening by Cr(iii) after incubation with GRK6pp-β​2AR-CT, F87F2Y site, only the SRC SH3 domain stimulated a significant only two specific sites in P1 and P2, including F87F2Y and L123F2Y upfield shift at the L123F2Y site (Supplementary Fig. 22a–h). in the ternary GRK6pp-β​2AR-CT–β-arrestin-1–SRC​ complex, were Importantly, the extent of the chemical shift produced by the less accessible to solvent (Fig. 3a–c and Supplementary Figs. 18–20). binding of different SH3 domains correlated with the avidity of In contrast, although the N375F2Y site, a hallmark 19F-NMR probe their interactions with GRK6pp-β​2AR-CT-activated β​-arrestin 1, that reports on the conformational change at the arrestin C-tail, as determined by GST pull-down assays (R2 =​ 0.92; Fig. 3e and was dislodged in response to receptor interactions and was sensi- Supplementary Fig. 22i). In conclusion, the 19F-NMR upfield tive to line broadening by Cr after clathrin association11, significant chemical shifts at the PRs were good indicators of both the occur- solvent-accessible protection at this site was not observed after the rence of SH3 domain binding and avidity in response to SH3 formation of the ternary GRK6pp-β​2AR-CT–β​-arrestin-1–SRC domain binding (Fig. 3e,f).

882 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

‘Snugly’

R* ? ‘Hanging’ β-Arr1*

R* Receptor phospho-barcode

P1 P114 Path

Agonist GRK6-mediated V40 Allostericpath β-Arr1*

binding phosphorylation β-Arr1* SH3-CPs

V8 recruitment P2 SH3-CPs P3 NMR chemical shift

Allosteric regulation of specific arrestin conformations through an explicit path R R*

Allosteric activation ‘Hanging’ ‘Snugly’ or -Arr1* R* R* β SH3 Activation

-Arr1* SH3 β Disassemble SH2 β-Arr1* Kinase SH2 Kinase

Allosteric activation of SH3-containing kinases by active arrestin

Fig. 6 | Schematic description of the allosteric mechanism underlying arrestin-mediated GPCR functions at two levels. At the first level, the phospho- barcode of the receptor (R) engages a precise pathway to modulate specific conformational states of remote proline regions of β-arrestin​ 1, which defines the efficacy for SH3-domain-containing protein recruitment (upper panel). At the second level, active arrestin engages with the SH3 domain of SH3- domian-containing kinases, thus disrupting the autoinhibitory conformation of the kinase, leading to increased downstream kinase activity (lower panel). Agonist-bound receptors are indicated as R*.

Upregulation of SRC activity by active β-arrestin 1. Many SH3- kinase domains form a highly organized, inactive structure40,41. The CPs, such as SRC, FYN and phospholipase C gamma 1 (PLCγ​), tethering of these ordered, inactive structures occurs via interac- have enzymatic activities14. However, the mechanisms by which tions between the SH3 domain and the linker regions between the GPCR–arrestin complexes regulate these proteins’ enzymatic activi- SH2 and the kinase domains. Both the P1 and P2 regions of active ties remain poorly understood. We therefore co-infected insect cells arrestin are protected from solvents after SRC association (Fig. 3a,b with a baculovirus containing both β​2AR and GRK6, administered and Supplementary Figs. 18 and 19). Therefore, the PRs of active agonist and purified proteins using M1-Flag beads. There were β​-arrestin 1 might interact with the SRC SH3 domain, inhibiting higher levels of β​2AR phosphorylated at pS355 and pS356 (with its interaction with the SH2-kinase linker, thereby resulting in SRC respect to stoichiometry) with the in vivo phosphorylation approach activation via the disassembly of its regulatory domains. than with the in vitro method (Supplementary Figs. 23a–c, 24 and We used site-specific fluorescence labeling to investigate SRC 25b). Specific phosphorylation at pS355/pS356 was verified by mass activation by GPCR–arrestin complexes and test this hypoth- spectrometry (Supplementary Fig. 23c). esis. We developed a genetic code expansion system by incorpo- We first analyzed levels of SRC autophosphorylation at the Y416 rating 2-amino-6-(((3-azidopropoxy)carbonyl)amino)hexanoic site, a hallmark for SRC activation39. Whereas addition of β​-arrestin 1 acid (KPN) into the desired site, which was followed by a click to SRC increased SRC autophosphorylation at pY416, pre-incubation reaction with 2,3,6-trimethyl-5-((methyl(prop-2-ynyl)amino) of β​-arrestin 1 with GRK6-phosphorylated β​2AR followed by the methyl)-tetrahydropyrazolo[1,2-a]pyrazole-1,7-dione (TA-bimane) addition of the receptor–arrestin complex to SRC markedly increased (Supplementary Fig. 26a). The azide group in KPN enabled site- autophosphorylation (Fig. 4a). Moreover, pre-incubating arrestin specific labeling of the TA-bimane (Fig. 5a and Supplementary with SRC slightly increased the kcat/Km of SRC toward the peptide Fig. 26a,b). We next chose two pairs of interactions: (1) N164KPN/ substrate by approximately two-fold, whereas the engagement of SRC N397W, which indicates the distance between the SH2 domain and with the GRK6-phosphorylated-β​2AR–arrestin complex increased the large lobe of the kinase domain, and (2) E93KPN/T289W, which the kcat/Km by approximately four-fold (Fig. 4b and Supplementary indicates the interaction between the SH3 domain and the small lobe Fig. 25c). Importantly, pre-incubation with GRK6pp-β​2AR-CT reca- of the kinase domain (Supplementary Fig. 27a). Mass spectrometry pitulated SRC activation by the phosphorylated receptor, although data confirmed KPN incorporation (Supplementary Fig. 26b). The with lower efficacy (Supplementary Fig. 25d). Mutations in the P1 labeling of N164KPN or E93KPN of β-arrestin​ 1 with TA-bimane and P2 regions of β​-arrestin 1 abolished the activation of the SRC produced a fluorescent bimane spectra with maximal emission at by either the receptor–arrestin complex or GRK6pp-β​2AR-CT 480 nm that underwent significant fluorescence quenching in the (Fig. 4c and Supplementary Fig. 25a, d). These data provided direct N397W-SRC or T289W-SRC proteins (Supplementary Fig. 26c,d), biochemical evidence that the active arrestin conformation induced respectively. These observations were consistent with the inactive by the GPCR coordinates SRC activation through its PRs. SRC crystal structure, in which the paired residues are located in close proximity (6 Å for N164-N397 and 7 Å for E293-T289; PDB Active β-arrestin 1 allosterically engages SRC activation. Crystal code 2SRC). After incubation with β-arrestin​ 1, the N164KPN/ structures of SRC family kinases indicate that their SH3, SH2 and N397W and E93KPN/T289W mutants exhibited 15% and 18%

Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology 883 Articles NaTuRe CHeMIcal BIoloGy increases in fluorescence intensity, respectively, thus indicating be common to both the hanging and the snuggly receptor–arres- the partial disassembly of the regulatory domain of inactive SRC tin conformations. Binding of the receptor-phospho-S5 to β​-arres- (Fig. 5b,c). Pre-incubation of full-length, GRK6 phosphorylated tin 1 is associated with the conformational states of both P1 and β​2AR with β​-arrestin 1 markedly weakened the interaction between P2, which are precisely controlled by defined activation paths. For the SH2 or SH3 domains and the large or small lobe of the SRC example, the connection of the β1​ sheet and loop 6 to the β​4 and kinase domain, as indicated by a further 170% and 220% increases β​7 sheets propagates the signal from the phospho-receptor-S5 bind- in the fluorescence intensity of N164KPN/N397W and E93KPN/ ing site to the conformational states of the P1 region of β​-arrestin 1 T289W, respectively (Fig. 5b,c). The observed conformational (Fig. 2a,d). Therefore, our results identified one of the mechanisms change in SRC was abolished either by removing β​-arrestin 1 from underlying the specificity and efficacy of GPCR and SH3-CP cou- the reconstituted system or by mutating both the P1 and P2 regions pling achieved through arrestin-mediated pathways. of β​-arrestin 1, confirming the essential role of PRs in mediat- At the second level, for which we used SRC as a model, we pro- ing SRC activation (Fig. 5b,c). These data indicated that GRK6- vided direct biophysical and cellular evidence that arrestin signals phospho-β​2AR–arrestin complexes modulate SRC activation in an to downstream kinases using allosteric mechanisms (Figs. 5d,e and allosteric manner by alleviating inhibition mediated by the SH2 and 6 and Supplementary Fig. 25e). Arrestin is known to mediate GPCR SH3 domains (Supplementary Fig. 26). functions by scaffolding downstream effectors to receptor–arrestin We next examined the arrestin-mediated conformational change complexes. Researchers have never explored whether active arrestin in SRC in a cellular context using intramolecular bioluminescence instigates downstream signaling through an allosteric mechanism. resonance energy transfer (BRET) with fluorescein arsenical hair- Here, we established a method to purify GRK6-phosphorylated pin (FlAsH)-labeled SRC probes carrying an N-terminal Renilla β​2AR, which enabled in vitro biochemical and functional analy- luciferase (rLuc) tag (Fig. 5d,e and Supplementary Fig. 27a–c)24,25. ses of arrestin activated by specific phosphorylated receptors These N-rLuc-SRC-FlAsH biosensors were designed to measure (Supplementary Fig. 23). After the binding of an agonist-occupied BRET between the fluorescence donor at the N terminus of SRC phosphorylated GPCR, the PR of β​-arrestin 1 assumes an active and a fluorescein arsenical acceptor specifically located at the intra- conformational state that subsequently engages with SH3 domains domain of the SRC proteins. β2AR​ activation by ISO increased and disassembles the compact regulatory domains of SH3-domain- the intermolecular BRET between the β​-arrestin-1-YFP and rLuc containing kinases such as SRC, thus markedly increasing kinase wild-type SRC and between different N-rLuc-SRC-FlAsH biosen- activity (Fig. 6). The active arrestin conformation has recently been sors, consistent with the co-IP results (Fig. 1a and Supplementary shown to persist, even after receptor disengagement25. Therefore, Fig. 27b). Intriguingly, β​2AR activation led to decreased intramo- in addition to assembling the receptor–downstream effector super lecular BRET in rLuc-SRC-β​5-FlAsH-αD​ but had no significant complex, the allosteric regulation of downstream signaling effectors effect on rLuc-SRC-β​B-FlAsH-α​A (Fig. 5d,e), suggesting that the by active arrestin might be a more general and necessary mecha- kinase domain shifted away from the N-terminal SH3 domain of nism underlying GPCR-regulated signaling than has been previ- SRC. Notably, these conformational changes were abolished by ously considered. mutating either the P1 and P2 regions of β​-arrestin 1 or the P2 Here, by exploiting specific β2AR​ mutations and using sequen- region and P114E, the key residue governing the allosteric path tial co-IP methods, the phospho-barcode of the receptor ‘hanging’ responsible for the P1 conformational change (Fig. 5d,e). Thus, the conformation was shown to potentially contribute to the β​-arrestin- BRET data confirmed that the observed conformational changes in 1-mediated β​2AR coupling to SRC, FYN and PLCγ​. Consistently, SRC were mediated by arrestin and its specific allosteric regulation the P2 region of β-arrestin​ 1 is located close to the rhodopsin–arres- by the receptor (Fig. 5d,e). tin interface in the ‘snuggly’ conformation of the rhodopsin–visual arrestin complex27. The binding of SRC to the P2 region may steri- Discussion cally clash with this ‘snuggly’ conformation and require substantial GPCR signaling encompasses three components: the receptor, rearrangements. However, direct evidence for this prediction is still the transducer G protein or arrestin, and downstream effectors. lacking. Therefore, an EM study of the agonist–β​2AR–arrestin–SRC Although GPCRs are a classic model for studying allosteric mecha- complex or a TROSY NMR spectrum obtained by simultaneously nisms42, investigations of the allosteric regulation of transducers labeling β​2AR and arrestin could help researchers delineate how are still relatively new3–6,43. For example, the molecular efficacy specific receptor snuggly and hanging conformations contribute to of different calcitonin receptor (CTR) ligands directly correlates receptor–arrestin–SH3-CP coupling. Moreover, both the P1 and P2 with distinct G-protein conformations via a potential allosteric regions of β-arrestin​ 1 contribute to the formation of the ternary mechanism1, and helix H5 has been identified as one of the key β​2AR/arrestin/SRC complex17,44,45. Consistently, our 19F-NMR spec- determinants of Gα​ in mediating the allosteric regulation of G pro- tra revealed that both F87F2Y and L123F2Y in β​-arrestin 1 were teins by GPCRs3. However, the mechanisms by which functionally less solvent accessible in response to SRC binding. In contrast, the relevant conformations of β​-arrestins are precisely regulated by SRC kinase domain was reported to directly interact with β​-arrestin receptors remain elusive. Our results provide direct experimental in another study20. Cross-linking experiments or structural studies evidence that allosteric regulation is a crucial mechanism under- could provide a more accurate explanation. Additionally, most of lying arrestin-mediated GPCR functions on at least two levels our structural analyses were based on the labeling of a single residue (Fig. 6). The first level lies in the distinct, functionally related con- in β​-arrestin 1 and 19F-NMR analyses. The use of methyl TROSY formational states of arrestin directed by phosphorylated GPCR and 3Q/3Q spectroscopy could provide more detailed information barcodes. Allosteric coupling between selective receptor-phospho- about the allosteric regulation of arrestins in future studies46,47. site binding and the specific conformational states of three PRs Taken together, these data show that an allosteric regulatory located within β​-arrestin 1 determine the selectivity and intrinsic mechanism participates in GPCR–SH3-CP coupling through arres- efficacy of downstream SH3-CP recruitment (Figs. 1b,i–k and 6). tin at two levels. Specific allosteric pathways were identified that Notably, recent works have identified two functionally important connected specific receptor binding sites to conformational changes conformational modes of the receptor–arrestin complex, namely, in remote PRs of arrestin. Moreover, active arrestin engages with the snuggly and hanging conformations10,15,28. Although the use of SH3 domain-containing kinases, such as SRC, and improves their the synthesized receptor-phospho-C-tail in this study primarily kinase activities through an allosteric mechanism. These results reflected the hanging conformation of the receptor–arrestin com- will elicit future interest in determining the specificity and working plex, the identified receptor-phospho-S5–arrestin interaction may mechanisms of arrestin-mediated GPCR functions.

884 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

Methods 32. Gurevich, V. V. & Gurevich, E. V. Te molecular acrobatics of arrestin Methods, including statements of data availability and any asso- activation. Trends Pharmacol. Sci. 25, 105–111 (2004). 33. Kim, M. et al. Conformation of receptor-bound visual arrestin. Proc. Natl ciated accession codes and references, are available at https://doi. Acad. Sci. USA 109, 18407–18412 (2012). org/10.1038/s41589-018-0115-3. 34. Liu, J. J., Horst, R., Katritch, V., Stevens, R. C. & Wthrich, K. Biased signaling pathways in β​2-adrenergic receptor characterized by 19F-NMR. Science 335, Received: 15 January 2018; Accepted: 3 July 2018; 1106–1110 (2012). Published online: 17 August 2018 35. Taniguchi, K. et al. A gp130-Src-YAP module links infammation to epithelial regeneration. Nature 519, 57–62 (2015). 36. Rahmeh, R. et al. Structural insights into biased G protein-coupled receptor References signaling revealed by fuorescence spectroscopy. Proc. Natl Acad. Sci. USA 1. Furness, S. G. et al. Ligand-dependent modulation of G protein conformation 109, 6733–6738 (2012). alters drug efcacy. Cell 167, 739–749.e11 (2016). 37. Yang, Z. et al. Phosphorylation of G protein-coupled receptors: from the 2. Flock, T. et al. Selectivity determinants of GPCR-G-protein binding. Nature barcode hypothesis to the fute model. Mol. Pharmacol. 92, 201–210 (2017). 545, 317–322 (2017). 38. Hirsch, J. A., Schubert, C., Gurevich, V. V. & Sigler, P. B. Te 2.8 A crystal 3. Flock, T. et al. Universal allosteric mechanism for Gα​ activation by GPCRs. structure of visual arrestin: a model for arrestin’s regulation. Cell 97, Nature 524, 173–179 (2015). 257–269 (1999). 4. Venkatakrishnan, A. J. et al. Diverse activation pathways in class A 39. Brown, M. T. & Cooper, J. A. Regulation, substrates and functions of src. GPCRs converge near the G-protein-coupling region. Nature 536, Biochim. Biophys. Acta 1287, 121–149 (1996). 484–487 (2016). 40. Xu, W., Harrison, S. C. & Eck, M. J. Tree-dimensional structure of the 5. Isogai, S. et al. Backbone NMR reveals allosteric signal transduction networks tyrosine kinase c-Src. Nature 385, 595–602 (1997). in the β​1-adrenergic receptor. Nature 530, 237–241 (2016). 41. Sicheri, F., Moaref, I. & Kuriyan, J. Crystal structure of the Src family 6. Sounier, R. et al. Propagation of conformational changes during μ​-opioid tyrosine kinase Hck. Nature 385, 602–609 (1997). receptor activation. Nature 524, 375–378 (2015). 42. Changeux, J. P. & Christopoulos, A. Allosteric modulation as a 7. Komolov, K. E. et al. Structural and functional analysis of a β​2-adrenergic unifying mechanism for receptor function and regulation. Cell 166, receptor complex with GRK5. Cell 169, 407–421.e16 (2017). 1084–1102 (2016). 8. Homan, K. T. & Tesmer, J. J. Structural insights into G protein-coupled 43. DeVree, B. T. et al. Allosteric coupling from G protein to the agonist-binding receptor kinase function. Curr. Opin. Cell Biol. 27, 25–31 (2014). pocket in GPCRs. Nature 535, 182–186 (2016). 9. Reiter, E. & Lefowitz, R. J. GRKs and β​-arrestins: roles in receptor silencing, 44. Bourquard, T. et al. Unraveling the molecular architecture of a G protein- trafcking and signaling. Trends Endocrinol. Metab. 17, 159–165 (2006). coupled receptor/β-arrestin/Erk​ module complex. Sci. Rep. 5, 10760 (2015). 10. Kumari, P. et al. Functional competence of a partially engaged GPCR- ​ β 45. Fessart, D. et al. Src-dependent phosphorylation of β​2-adaptin dissociates the -arrestin complex. Nat. Commun. 7, 13416 (2016). β​-arrestin-AP-2 complex. J. Cell Sci. 120, 1723–1732 (2007). 11. Yang, F. et al. Phospho-selective mechanisms of arrestin conformations and 19 46. Ruschak, A. M. & Kay, L. E. Proteasome allostery as a population shif functions revealed by unnatural amino acid incorporation and F-NMR. between interchanging conformers. Proc. Natl Acad. Sci. USA 109, Nat. Commun. 6, 8202 (2015). E3454–E3462 (2012). 12. Dong, J.-H. et al. Adaptive activation of a stress response pathway improves 47. Tugarinov, V., Sprangers, R. & Kay, L. E. Probing side-chain dynamics in learning and memory through Gs and β​-arrestin-1-regulated lactate the proteasome by relaxation violated coherence transfer NMR spectroscopy. metabolism. Biol. Psychiatry 81, 654–670 (2017). J. Am. Chem. Soc. 129, 1743–1750 (2007). 13. Lefowitz, R. J. & Shenoy, S. K. Transduction of receptor signals by 48. Barker, S. C. et al. Characterization of pp60c-src tyrosine kinase activities β​-arrestins. Science 308, 512–517 (2005). using a continuous assay: autoactivation of the enzyme is an intermolecular 14. Liu, C. H. et al. Arrestin-biased AT1R agonism induces acute catecholamine autophosphorylation process. Biochemistry 34, 14843–14851 (1995). secretion through TRPC3 coupling. Nat. Commun. 8, 14335 (2017). 15. Shukla, A. K. et al. Visualization of arrestin recruitment by a G-protein- coupled receptor. Nature 512, 218–222 (2014). Acknowledgements 16. Wang, W., Qiao, Y. & Li, Z. New insights into modes of GPCR activation. We thank D.-S. Li for stimulating discussions and critical reading of the manuscript. Trends Pharmacol. Sci. 39, 367–386 (2018). We thank S.-S. Zang and X-H. Liu of the Core Facility of Protein Research, Institute of 17. Luttrell, L. M. et al. β​-Arrestin-dependent formation of β​2 adrenergic Biophysics, Chinese Academy of Sciences, for their help in the NMR data collection, receptor-Src protein kinase complexes. Science 283, 655–661 (1999). analysis and valuable discussion. We thank J. Jia and S. Sun for their technical assistance 18. Hara, M. R. et al. A stress response pathway regulates DNA damage through in flow cytometry analysis. We thank X. Ding from the Laboratory of Proteomics, Core β​2-adrenoreceptors and β​-arrestin-1. Nature 477, 349–353 (2011). Facilities for Protein Science, at the Institute of Biophysics (IBP), Chinese Academy 19. Xiao, K. et al. Functional specialization of β​-arrestin interactions revealed by of Sciences (CAS), for her help with the mass spectrometry analysis. We thank R.J. proteomic analysis. Proc. Natl Acad. Sci. USA 104, 12011–12016 (2007). Lefkowitz (Duke University) for giving the constructs of Flag-β2AR,​ Flag-V2R, Flag- 20. Miller, W. E. et al. β​-Arrestin1 interacts with the catalytic domain of the SSTR2, β​-arrestin-1-YFP, GRK6-YFP and GRK2-YFP. We thank Z. Yang, Y.-S. He, X.-L. tyrosine kinase c-SRC. Role of β​-arrestin1-dependent targeting of c-SRC in Fu and L. Chen for participating in the collection and analysis of this large sequence receptor endocytosis. J. Biol. Chem. 275, 11312–11319 (2000). library of GPCRs. We thank an anonymous scientist who helped in design of the 21. Tobin, A. B., Butcher, A. J. & Kong, K. C. Location, location, location…​ receptor–arrestin–SRC complex formation strategy and contributed significantly to this site-specifc GPCR phosphorylation ofers a mechanism for cell-type-specifc project. We acknowledge support from the National Key Basic Research Program of signalling. Trends Pharmacol. Sci. 29, 413–420 (2008). China (2015CB856203 to J.-Y.W.), the National Natural Science Foundation of China 22. Peterson, Y. K. & Luttrell, L. M. Te diverse roles of arrestin scafolds in G (81773704 and 31470789 to J.-P.S., 21325211 to J.-Y.W., 31700692 to P.X.), the Shandong protein-coupled receptor signaling. Pharmacol. Rev. 69, 256–297 (2017). Natural Science Fund for Distinguished Young Scholars (JQ201517 to J.-P.S.), the 23. Shukla, A. K. et al. Distinct conformational changes in β​-arrestin report Fundamental Research Fund of Shandong University (2014JC029 to X.Y.), the National biased agonism at seven-transmembrane receptors. Proc. Natl Acad. Sci. USA Science Fund for Distinguished Young Scholars (81525005 to F.Y.) and the Program for 105, 9988–9993 (2008). Changjiang Scholars and Innovative Research Team in University (IRT13028). 24. Lee, M. H. et al. Te conformational signature of β-arrestin2​ predicts its trafcking and signalling functions. Nature 531, 665–668 (2016). 25. Nuber, S. et al. β-Arrestin​ biosensors reveal a rapid, receptor-dependent Author contributions activation/deactivation cycle. Nature 531, 661–664 (2016). J.-P.S. conceived the idea for allosteric regulatory mechanism of arrestin and its 26. Shukla, A. K. et al. Structure of active β​-arrestin-1 bound to a G-protein- engagement with effectors. J.-Y.W. conceived all chemical biology experiments and coupled receptor phosphopeptide. Nature 497, 137–141 (2013). synthesis route. J.-P.S. conceived that majority of GPCR members connect to SH3- 27. Kang, Y. et al. Crystal structure of rhodopsin bound to arrestin by domain-containing proteins through arrestins. J.-Y.W. brought up the idea for F2Y femtosecond X-ray laser. Nature 523, 561–567 (2015). and KPN. J.-P.S, F. Yang and P.X. designed the key experiments to dissect the allosteric 28. Cahill, T. J. III et al. Distinct conformations of GPCR-β​-arrestin complexes propagation pathways. J.-P.S., J.-Y.W. and X.Y. designed most of the experiments. F. Yang mediate desensitization, signaling, and endocytosis. Proc. Natl Acad. Sci. USA and K.R. collected and analyzed the 19F-NMR data. P.X., Q. Liu. and F. Yang purified 114, 2562–2567 (2017). β​2AR and GRK6 proteins. P.X., C.-X.Q and F. Yang performed SRC kinase activity assays. 29. Kim, Y. J. et al. Crystal structure of pre-activated arrestin p44. Nature 497, Q. Liu Synthesized TA-bimane and performed fluorescence spectroscopy assays. Z.- 142–146 (2013). X.L. performed BRET assays. F. Yang, C.-X.Q., C.L., R.-R.L., Y.-M.S. Z.-Y.Y and X.-Z.G. 30. Manglik, A. et al. Structural insights into the dynamic process of synthesized F2Y and purified F2Y incorporated β​-arrestin 1 proteins. F. Yang, C.L. and β​2-adrenergic receptor signaling. Cell 161, 1101–1111 (2015). Q.-T.H. purified GST-SRC-FL, GST-SH3-CPs and GST-clathrin proteins. F. Yang, P.X. 31. Hanson, S. M. et al. Structure and function of the visual arrestin oligomer. C.-X.Q., Q.-T.H. and Q. Li. performed GST-pull down and CO-IP assays. F. Yang and EMBO J. 26, 1726–1736 (2007). Q.-T.H. performed circular dichroism assays. F. Yang, F. Yi and Q. Li performed Trypsin

Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology 885 Articles NaTuRe CHeMIcal BIoloGy

digestion and MS/MS assays. L.-Y.W., P.X., C.-X.Q., J.-Y.X. and D.-F.H., analyzed the SH3 Additional information binding motif (proline region) across GPCR family members. J.-P.S., J.-Y.W. and X.Y. Supplementary information is available for this paper at https://doi.org/10.1038/ supervised the overall project design and execution. J.-P.S. participated in data analysis s41589-018-0115-3. and interpretation. J.-P.S. and J.-Y.W. wrote the manuscript. All of the authors have seen Reprints and permissions information is available at www.nature.com/reprints. and commented on the manuscript. Correspondence and requests for materials should be addressed to J.-p.S. or J.W. Competing interests Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in The authors declare no competing interests. published maps and institutional affiliations.

886 Nature Chemical Biology | VOL 14 | SEPTEMBER 2018 | 876–886 | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

Methods ATP at 25 °C for 1 h. Iterative ATP was added for every half hour to promote the Antibodies and reagents. Te monoclonal anti-GST (2622), His-Tag Mouse mAb kinase reaction to a saturating limit. The phosphorylation level of the β​2AR at 355 356 (2366), and Phospho-Src Family (Tyr416) (2101 S) antibodies were purchased GRK6-specific phosphorylation site (pS pS ) was examined by western blotting from Cell Signaling Technology. Te c-src antibody (sc-130124), a Fyn antibody with specific antibody. For clarity, samples were treated with PNGase F for 30 min (sc-51598), PLCγ1​ antibody (sc-7290),GRB2 antibody (sc-255), p-β​2-AR (Ser355/ at room temperature (25° C) to minimize the glycosylation. Ser356)-R antibody (sc-16719-R), and HA-probe antibody (sc-7392) were from Santa Cruz. Te monoclonal anti-Flag M2 antibody (F3165) and HA-antibody- In vivo phosphorylation of β2AR by GRK6. The high-titered viruses encoding conjugated agarose were purchased from Sigma. Glutathione-Sepharose 4B β​2AR and GRK6 were co-infected in Sf9 cells. After 48 h infection, the cells (45-000-139) were from GE Healthcare. Ni-NTA agarose were from Termo were stimulated with BI-167107 at 37 °C for 15 min. The cells were lysed in Fisher scientifc. Quinpirole, isoproterenol, acetylcholine, somatostatin, l-lactic homogenization buffer containing 10 mM HEPES, pH 7.5, 100 mM NaCl, 1% (w/v) dehydrogenase from rabbit muscle, Phospho(enol)pyruvic acid trisodium salt DDM, 0.02% (w/v) CHS, and protease and phosphatase inhibitor cocktail. The hydrate and pyruvate kinase from rabbit muscle were purchased from Sigma phosphorylated β​2AR were purified by M1-Flag beads and size-exclusion column Aldrich. Flag M1 antibody was produced by Flag-M1 hybridoma cell and purifed as described in the procedure of purification of wild-type β​2AR. Phosphorylation by Protein A/G beads. NADH and disodium salt were from Roche, adenosine of β2AR​ was examined by western blotting. The PNGase F was used to minimize 5′​-triphosphate (ATP) was from BBI Life Sciences and N-methyl-N-prop-2- the glycosylation. ynylamine was from J&K Scientifc; monobromobimane was from Termo Fisher scientifc; V2Rpp was from Tufs University core facility; GRK2pp, GRK6pp, Expression and purification of β-arrestin 1 WT and F2Y incorporated Proteins. GRK6pp–pS355S, GRK6pp–pS356S, GRK6pp–E369A, V2Rpp-2/3 and RR-SRC See Supplementary Note 1 (in supplementary Methods section, ‘‘Expression were purchased from China Peptides Co., Ltd. All of the other reagents were and purification of β​-arrestin 1 WT and F2Y incorporated proteins’’) for from Sigma. further details.

Cell culture. HEK293 cells were cultured in DMEM (Invitrogen) medium with Expression and purification KPN incorporated proteins and Click reaction. 10% FBS (Invitrogen) in 37 °C incubator with 5% CO2. Sf9 cells were cultured in For the expression of SRC-KPN proteins, pEVOL-KPNRS was co-transformed ESF 921 culture medium (Expression system) at 27 °C. with different GST-SRC TAG mutants into BL21 (DE3) E. coli cells. 6 L transformed cells were induced with 0.02% (w/v) l-arabinose, 0.3 mM IPTG Constructs. The pcDNA3.0-Flag-β​2AR, pcDNA3.0-Flag-V2R, pcDNA3.0-Flag- and 1 mM KPN at an OD 600 nm of 1.0. After growing at 18 °C for 16 h, the cells SSTR2, β​-arrestin-1-YFP, pcDNA3.0-GRK6-YFP and pcDNA3.0-GRK2-YFP were harvested and resuspended in lysis buffer (20 mM Tris–HCl, pH 8.0 and constructs were generous gifts from R.J. Lefkowitz at Duke University. Wild-type 150 mM NaCl) and then lysed by sonication. After centrifugation, the supernatant cDNAs of M3R, D2R, FYN, PLCγ​1, HCK, Grb2 and human SRC were subcloned was incubated with 2 mL glutathione-Sepharose 4B resin at 4 °C for 2 h. The beads into EcoRI/XhoI sites of pcDNA3.0 vector. Full-length wild-type cDNA of bovine were washed for three times, and the bound protein was eluted with 10 mM GSH. β​-arrestin 1 were subcloned into NdeI/XhoI sites of pET-22b vector with The concentrated protein was subjected to gel filtration using a Superdex 200 C-terminal 6×​His tag as previously described11. Wild-type human β​2AR was column (GE Healthcare) according to the manufacturer’s instructions. subcloned into EcoRI/XhoI sites of pFastbac1 vector (Invitrogen) with Flag epitope The KPN-incorporated SRC was subsequently labeled with TA-bimane tag at N terminus. Wild-type human GRK6 was constructed into the same vector (2,3,6-trimethyl-5-((methyl(prop-2-ynyl)amino)methyl)-tetrahydropyrazolo[1,2-a] with C terminus 10×​His tag. The gene encoding Renilla luciferase was linked to pyrazole-1,7-dione) using copper(i)-catalyzed click reaction. Click reactions were the N terminus of pcDNA3.0-SRC by overlapping PCR. Two luciferase c-SRC- carried out in tubes in 150 μ​L solution containing 5 μM​ SRC-KPN, 2 mM sodium FlAsH BRET reporters were constructed by inserting a cDNA sequence encoding ascorbate, 50 μ​M CuSO4, and 100 μ​M TA-bimane, with 20 mM pH 7.0 HEPES the FlAsH motif CCPGCC immediately following amino acid residues 166 or 340 buffer (150 mM NaCl)/DMSO (99:1) for 5 h at 25 °C. of human SRC, using Quikchange mutagenesis kit (Stratagene). DNA encoding residues 1–494 of human clathrin, full length-SRC, Grb2, c-SRC-SH3 domain, Expression and purification of GST-clathrin, GST-SRC and GST-SH3-CPs. FYN-SH3 domain, and PLCγ​-SH3 domain were subcloned into pGEX-6P-1 vector The expression and purification of GST-clathrin, GST-SRC-full length (human) with an N-terminal GST tag for E. coli expression and GST pull-down assays. All or GST-SH3-CPs proteins follow the same procedures as previously described11,14. point mutations were generated using Quikchange mutagenesis kit (Stratagene). The plasmid encoding GST-tagged protein was transformed into BL21 (DE3) All constructs and mutations were verified by DNA sequencing. E. coli cells, and 1 to 6 L cells were cultured, induced by 0.3 mM IPTG at 18 °C for 16 h. After centrifugation at 4,000 r.p.m., the cell pellet was collected and Computational analysis of SH3 binding motif (PRR region) across GPCR resuspended in GST purification buffer containing 25 mM Tris–HCl at pH 8.0, family members. A total of 825 GPCR sequences were downloaded from Uniprot 150 mM NaCl, 0.5% (v/v) Triton X-100, 5% (v/v) glycerol, 2 mM EDTA, and 1 mM and subjected to further analysis. The SH3 domain is able to interact with its ligand DTT, and then subjected to cell cracker or sonication. After centrifugation, the through a polyproline rich region (PRR). Here, we followed definition of class supernatant was incubated with 1 mL of glutathione-Sepharose 4B resin at 4 °C for I motif as [R/K/Q/M/E]XXPXXP and class II motif as PXXPX[R/K]49. Protein 2 h. The beads were washed three times, and the bound protein was eluted with sequences were aligned using MUSCLE software50. An in-house python program 10 mM GSH. The protein was concentrated to 1~5 mg/ml and further purified was developed to search class I and II motifs against intracellular loop 3 (ICL3) and by gel filtration using a Superdex 75 column (GE Healthcare) according to the the C terminus of each GPCR member. Phylogenetic relationship of GPCRs were manufacturer’s instructions. reconstructed using the Neighbor-joining (NJ) method implemented in the MEGA v7.0 program51. The phylogenetic tree and domain representation was visualized Quantitative densitometry analysis for western blots. Quantitative densitometry using ETE3 software52. We built Neighbor-Joining trees from individual sets of analysis of the integrated band intensity of western blots was performed routinely multiple alignment and then merged these trees to give the dendrogram clustering using ImageJ as previously described11. For clarity, all samples tested for the representation of the human GPCRs. presence of Flag-tagged GPCRs were treated with PNGase F for 30 min at room temperature to minimize the glycosylation effects. Synthesis of TA-bimane and F2Y. See Supplementary Note 1 (in supplementary Methods sections ‘‘Synthesis of TA-bimane’’ and ‘‘Synthesis of F2Y’’) for NMR experiments. To examine the phospho-receptor-C-tail induced β​-arrestin further detail. 1 conformational change, 30~60 μ​M β​-arrestin 1 F2Y protein probe was mixed with or without a three-fold phospho-receptor-C-tail fragment and then incubated

Expression and purification of β2AR. Expression and purification of β2AR​ was in binding buffer (20 mM Tris–HCl, pH 8.0, 150 mM NaCl, 10% (v/v) D2O) with done according to methods previously described7,15,34,53–55. See Supplementary Note 1 end-to-end rotation at room temperature for 30 min. To examine β​-arrestin 1 (in supplementary Methods section ‘‘Expression and Purification of β​2AR’’) for conformational change after SRC-SH3, FYN-SH3, PLCγ​-SH3 or YES-SH3 binding, further details. an equal molar of each protein was added into β​-arrestin 1 (or phospho-receptor- C-tail fragment/β​-arrestin 1 complex). After incubation at room temperature for Expression and purification of GRK6 from insect cells. GRK6 was expressed in 30 min, the protein samples were subjected to 19F-NMR experiments. Sf9 insect cells and purified as previously described56, except that Co-NTA resin For paramagnetic relaxation experiments, 0, 1, 3 or 5 mM paramagnetic (Talon Superflow, Clontech) was used for 10×​His-tagged recombinant protein relaxation agent chromium acetylacetone (Cr (iii)) was mixed with different purification. See Supplementary Note 1 (in supplementary Methods section β​-arrestin 1 mutants (F87F2Y, L123F2Y, R177F2Y and N375F2Y). The protein ‘‘Expression and Purification of GRK6 from insect cells’’) for further details. samples were subjected to 19F-NMR experiments. The 19F-NMR spectrum was then collected after subsequently adding the GRK6pp, full length SRC protein or In vitro phosphorylation of β2AR by GRK6. β​2AR and GRK6 were expressed different SH3 domains. separately in Sf9 insect cells and purified homogenous individually. The purity of 19F NMR spectra were recorded on a Bruker AVANCE 600 spectrometer the protein was verified by electrophoresis and Coomassie blue staining. For in fitted with a QCI -F cryoprobe at 25 °C. The 90 degree pulse is 11 μ​s. Data were vitro phosphorylation, 1 μ​M β​2AR protein was incubated with 3 μ​M GRK6 protein processed using 20-Hz Lorentzian line broadening and were referenced to the in buffer containing 20 mM HEPES, PH 7.5, 200 mM NaCl, 4 mM MgCl2 and 5 mM internal TFA standard (−​76.5 p.p.m.).

Nature Chemical Biology | www.nature.com/naturechemicalbiology Articles NaTuRe CHeMIcal BIoloGy

Fluorescence spectroscopy. Steady state fluorescence was measured by a Varioskan were added into the mixture and the mixture was underwent end-to-end flash (Thermo Scientific) instrument with full wavelength scanning mode at 25 °C. rotation at 4 °C for 2 h. The GST beads were collected by centrifuge and washed Typically, 150 μ​L samples containing 1 μ​M labeled SRC KPN-TAb in a MicroFluor with lysis buffer for four times. After removing the supernatant, the samples 96-well plate were excited at 390 nm, and the emitted fluorescence was measured were resuspended in 50 μ​L 2XSDS loading buffer and boiled for 10 min before from 450 to 520 nm using a 2 nm step size. Each data point was integrated for western blot. 0.2 s. To determine the effect in response to β​-arrestin1 binding in the absence/ presence of activated β​2AR, we took the spectra after 1 h incubation of SRC with Western blot quantification analysis for arrestin-mediated complex formation. the equivalent amounts of β​-arrestin 1 only or β​-arrestin 1 in plus of three-fold The western blot quantification analysis has been successfully used to assess of GRK6-phosphorylated β​2AR stabilized with 10 μ​M BI-167107. Spectra were the DR5:DISC stoichiometry, to measure the SV proteins and lipids, and to analyzed using the GraphPad Prism 5. determine the enzyme catalytic activity of MKP3 and STEP58–60. To determine the concentration of the β​-arrestin 1 or its mutant in the GST-SRC-SH3 containing Confocal microscopy. Confocal microscopy experiments were performed as complex, a calibration curve was established first (Supplementary Fig. 13a) previously described12. The plasmids encoding β​-arrestin-1-YFP WT or mutants and quantitative analysis of western blots were then performed11,58. Protein were transiently co-transfected with Flag-β​2AR in HEK293 cells. 24 h after samples alongside a serial concentration of purified recombinant His-β​-arrestin transfection, the cells were seeded onto fibronectin-coated, 35-mm, glass-bottom 1 proteins within the linear range were subjected to SDS–PAGE in precast 12% plates. After 8 h starvation, the cells were stimulated with isoproterenol (10 μ​M) polyacrylamide gels (Bio-Rad), which was followed by transferring and blocking. at 37 °C for 10 min. The translocation of β​-arrestin-1-YFP was monitored by Blots were incubated with anti His-Tag Mouse mAb antibody (Cell signaling Olympus FV1200 confocal microscopy. technology, #2366) and HRP goat anti mouse IgG secondary antibodies, and the integrated band intensities were detected and quantified using ImageJ. A linear Trypsin digestion and MS/MS analysis. The GRK6-phosphorylated β​2AR and standard regression curve of integrated band intensity versus the concentration of SRC-164 KPN proteins were subjected to electrophoresis, and the protein band β​-arrestin 1 standard was generated. The amount of β​-arrestin 1 in complex was cut into small plugs. After dehydrating in acetonitrile for 10 min, the gel was samples were assessed in comparison to the standard curve. dried in a Speedvac (Labconco) for about 15 min. Disulfide bonds were reduced by dithiothreitol (DTT) and subsequently alkylated by 40 mM IAA, 25 mM NH4HCO3 Circular dichroism. Far-UV circular dichroism (CD) spectra were acquired for 45 min in the dark. The sample was then digested by trypsin at 37 °C for 12 h between 200 and 260 nm on a Chirascan Plus CD instrument (Applied and the reaction was stopped by 1% (v/v) formic acid. Digested samples were Photophysics, UK) at 25 °C in a 0.1-mm path-length thermostated quartz cuvette. purified, desalted and redissolved in 30 μ​L 50% (v/v) acetonitrile /0.1% (v/v) Spectra of 4 μ​M β​-arrestin 1 WT, P114E, V40E, V34R, P36E and P1P2 mutations trifluoroacetic acid buffer before MS/MS analysis. MS data were analyzed by were measured in buffer containing 20 mM Tris, 150 mM NaCl, pH 8.0. Bioworks 3.2 software. For determination of the phosphorylated sites in β​2AR, samples were treated with PNGase F and immediately followed by electrophoresis Intramolecular FlAsH BRET assay. HEK293 cells were co-transfected with Flag-β​ and tandem mass spectrometry analysis. 2AR, Rluc-SRC FlAsH and HA-arrestin. 24 h after transfection, the cells were seeded into clear-bottom 96-well plates at a density of 100,000 cells per well. After Co-immunoprecipitation (Co-IP). The recruiting of SH3-CPs by β​-arrestin 1 another 24 h incubation at 37 °C, the cells were washed with 600 μ​L BRET buffer after specific GPCR activation was examined by Co-IP experiments. HEK293 (146 mM NaCl, 4.2 mM KCl, 0.5 mM MgCl2, 1 mM CaCl2, 10 mM HEPES, pH 7.4, cells were co-transfected with HA-β​-arrestin 1 (or HA-β​-arrestin 1 PRR mutants), 0.1% glucose) and incubated with FlAsH-EDT2 labeling reagent from TC-FlAsH Flag-tagged GPCRs (D2R, β​2AR, M3R, SSTR2 or V2R) and different SH3-CPs II In-Cell Tetracysteine Tag Detection Kit (Thermo Fisher Scientific Inc, Waltham, (SRC, FYN, PLCγ​, Grb2 or HCK). 48 h after transfection, the cells were starved MA) at a final concentration of 2.5 μ​M at 37 °C for 1 h. FlAsH-labeled cells were for 8 h and stimulated with 10 μ​M corresponding agonists for 15 min (quinpirole washed twice with 1×​ BAL buffer from the detection kit and stimulated with for D2R; Isoproterenol for β2AR;​ acetylcholine for M3R; somatostatin for 10 μ​M isoproterenol or vehicle at 37 °C for 10 min. 5 μM​ of luciferase substrate SSTR2 and arginine vasopressin for V2R). Transfected HA-β​-arrestin 1 was coelenterazine H was added before reading BRET in a Mithras LB940 microplate immunoprecipitated by HA-antibody-conjugated agarose, and β​-arrestin-1- reader (Berthold Technologies) equipped with 485 nm excitation and 530 nm mediated recruitment of SRC, FYN, PLCγ​, Grb2 or HCK, and the presence of the emission filters. The BRET signal was determined as the ratio of the light emitted receptor in the ternary complex was detected by western blotting using specific by FlAsH-labeled biosensors and the light emitted by Rluc-tagged biosensors. SRC, FYN, PLCγ​, HCK or Flag antibodies. Receptor samples were treated with The Δnet​ change in intramolecular BRET ratio was calculated by background PNGase F for deglycosylation. subtracting the BRET ratio measured for FlAsH-labeled cells stimulated with To examine the role of GRK in β​-arrestin-1-mediated SH3-CP enlistment, vehicle only. HEK293 cells were co-transfected with Flag-β​2AR, HA-β​-arrestin 1, different SH3-domain containing proteins (SH3-CP: SRC, FYN or PLCγ​) and selected Intermolecular BRET assay. Flag-β​2AR, arrestin-YFP and Rluc-SRC (or its GRK or control plasmids. 48 h after transfection, the cells were stimulated with corresponding FlAsH mutants) were co-transfected into HEK293 cells. 24 h after 10 μ​M isoproterenol for 15 min. HA-β-arrestin​ 1 was immunoprecipitated by transfection, the cells were seeded into clear-bottom 96-well plates at a density of HA-antibody-conjugated agarose and the formation of the β​-arrestin-1–SH3-CP 100,000 cells per well. After another 24 h incubation at 37 °C, the cells were washed complex was detected by specific SRC, FYN, or PLCγ​ antibodies. thrice with 600 μ​L PBS and stimulated with 10 μ​M Isoproterenol or vehicle at 37 °C for 10 min. 5 μM​ of luciferase substrate coelenterazine H was added before Sequential-co-immunoprecipitation (Co-IP). HEK293 cells were co-transfected reading BRET in a Mithras LB940 microplate reader (Berthold Technologies) with Flag-β​2AR (or Flag-β​2ARΔICL3, Flag-β​2AR-S355A, Flag-β​2AR-E369A equipped with 485 nm excitation and 530 nm emission filters. The BRET signal mutants), HA-β​-arrestin 1 and different SH3-CPs (SRC, FYN or PLCγ​). 48 h after was determined as the ratio of the light emitted by YFP biosensors and the light transfection, the cells were starved for 8 h and stimulated with 10 μ​M ISO for emitted by Rluc-tagged biosensors. The Δ​ net change in intermolecular BRET ratio 15 min. Cells were washed with cold PBS three times and then collected in cold was calculated by background subtracting the BRET ratio measured for YFP cells lysis buffer. The cell lysates were centrifuged for 30 min at 12,000 g after 40 min of stimulated with vehicle only. end-to-end rotation at 4 °C. The protein complexes were first immunoprecipitated by anti-Flag M1 agarose and then eluted with 0.2 mg/mL Flag peptide and 5 mM Statistical analysis. When needed, a one-way ANOVA test was performed using EDTA. The complexes were then immunoprecipitated by HA-antibody-conjugated the analysis software GraphPad Prism. For all experiment, the number of replicates agarose. The association of SH3-CPs was detected by specific SRC, FYN, or and P value cutoffs are described in the respective figure legends. Error bars are PLCγ​ antibodies. shown for all data points with replicates as a measure of variation with the group.

SRC kinase activity. The kinetic assays for SRC were performed using RR-SRC Reporting Summary. Further information on experimental design is available in peptide as the substrate and the spectrophotometric analysis of the oxidation the Nature Research Reporting Summary linked to this article. of NADPH as established previously. The decrease rate of absorbance at 340 nm is coupled to the steady-state SRC kinase phosphorylation rate of RR-SRC peptide. Data availability. The authors declare that data supporting the finding of this The kinetic parameters were determined by fitting the initial rate to the study are available within the article and its Supplementary Information. Full blots 57,58 Michaelis–Menten equation to obtain Km and kcat . See Supplementary Note 1 are shown in Supplementary Figs. 28–39. Additional data are available from the (in supplementary Methods section, "SRC kinase activity") for further details. corresponding author upon reasonable request.

GST pull-down assay. Binding assays of clathrin or different GST-SH3-CPs with β​-arrestin 1 were performed as previously described11,14. 2 μ​M wild-type or mutant References β​-arrestin 1 was first mixed with specific phospho-receptor-C-tail fragment and 49. Carducci, M. et al. Te protein interaction network mediated by human SH3 incubated in binding buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM EDTA, domains. Biotechnol. Adv. 30, 4–15 (2012). 1 mM DTT) at 25 °C for 30 min. After incubation, 2 μ​M GST-clathrin or GST-SH3- 50. Edgar, R. C. MUSCLE: multiple sequence alignment with high accuracy and CPs was added and then incubated for another 1 h. Subsequently, 10 μ​L GST beads high throughput. Nucleic Acids Res. 32, 1792–1797 (2004).

Nature Chemical Biology | www.nature.com/naturechemicalbiology NaTuRe CHeMIcal BIoloGy Articles

51. Kumar, K. V., Modi, K. D. & Kalra, S. Mighty journals, mega trials, and 56. Kim, T. H. et al. Te role of ligands on the equilibria between modest points. Indian J. Endocrinol. Metab. 20, 578–579 (2016). functional states of a G protein-coupled receptor. J. Am. Chem. Soc. 135, 52. Huerta-Cepas, J., Serra, F. & Bork, P. ETE 3: reconstruction, analysis, 9465–9474 (2013). and visualization of phylogenomic data. Mol. Biol. Evol. 33, 57. Li, H. et al. Crystal structure and substrate specifcity of PTPN12. Cell Rep. 1635–1638 (2016). 15, 1345–1358 (2016). 53. Rasmussen, S. G. et al. Crystal structure of the human β​2 adrenergic 58. Li, R. et al. Molecular mechanism of ERK dephosphorylation by striatal- G-protein-coupled receptor. Nature 450, 383–387 (2007). enriched protein tyrosine phosphatase. J. Neurochem. 128, 315–329 (2014). 54. Liang, Y.-L. et al. Phase-plate cryo-EM structure of a class B GPCR-G-protein 59. Majkut, J. et al. Diferential afnity of FLIP and procaspase 8 for complex. Nature 546, 118–123 (2017). FADD’s DED binding surfaces regulates DISC assembly. Nat. Commun. 5, 55. Loudon, R. P. & Benovic, J. L. Expression, purifcation, and characterization 3350 (2014). of the G protein-coupled receptor kinase GRK6. J. Biol. Chem. 269, 60. Kim, Y., Rice, A. E, & Denu, J. M. Intramolecular dephosphorylation of ERK 22691–22697 (1994). by MKP3. Biochemistry 42, 15197–15207 (2003).

Nature Chemical Biology | www.nature.com/naturechemicalbiology nature research | reporting summary

Corresponding author(s): Jin-peng Sun, Jiang-yun Wang

Reporting Summary Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parameters When statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main text, or Methods section). n/a Confirmed The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section. A description of all covariates tested A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and code Policy information about availability of computer code Data collection The following softwares were used: topspin 3.5, Bioworks 3.2, SoftMax pro 5.4.1.

Data analysis The following softwares were used: MUSCLE, MEGA v7.0, Bioworks 3.2 , SoftMax pro 5.4.1, WinCoot v0.8.6, Pymol v1.7, ImageJ, GraphPad Prism 7, Adobe Illustrator CS6 ,Microsoft Excel 2016,topspin 4.0,ETE3

For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

Data April 2018 Policy information about availability of data All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A list of figures that have associated raw data - A description of any restrictions on data availability

The authors declare that data supporting the finding of this study are available within the article and its Supplementary Information. Additional data are available from the corresponding author upon reasonable request.

1 nature research | reporting summary Field-specific reporting Please select the best fit for your research. If you are not sure, read the appropriate sections before making your selection. Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study design All studies must disclose on these points even when the disclosure is negative. Sample size The manuscript did not contain animal studies. For western, enzymatic study and fluorescence quenching experiment, no statistical method was used to predetermine sample size, because the experimental methods used in this article are classic and highly applied, according to the convention, after 3 to 5 times repeated in parallel, it is sufficient to obtain scientific conclusions,when the result is consistent.

Data exclusions No data were excluded from the analysis.

Replication Each individual experiment was repeated independently with similar results at least three times, except that experiments shown in Figures 5b and 5c were repeated for five times.

Randomization The manuscript did not contain any animal or behavior studies. Samples were not randomized for the experiments.

Blinding The manuscript did not contain animal studies. Investigators were not blinded to the group allocation during data collection and analysis. Experimenters operate according to the standard method, conducted a comprehensive analysis of all data, and obtain scientific conclusions. There was no personal preference for the experimental subjects and results.

Behavioural & social sciences study design All studies must disclose on these points even when the disclosure is negative. Study description Briefly describe the study type including whether data are quantitative, qualitative, or mixed-methods (e.g. qualitative cross-sectional, quantitative experimental, mixed-methods case study).

Research sample State the research sample (e.g. Harvard university undergraduates, villagers in rural India) and provide relevant demographic information (e.g. age, sex) and indicate whether the sample is representative. Provide a rationale for the study sample chosen. For studies involving existing datasets, please describe the dataset and source.

Sampling strategy Describe the sampling procedure (e.g. random, snowball, stratified, convenience). Describe the statistical methods that were used to predetermine sample size OR if no sample-size calculation was performed, describe how sample sizes were chosen and provide a rationale for why these sample sizes are sufficient. For qualitative data, please indicate whether data saturation was considered, and what criteria were used to decide that no further sampling was needed.

Data collection Provide details about the data collection procedure, including the instruments or devices used to record the data (e.g. pen and paper, computer, eye tracker, video or audio equipment) whether anyone was present besides the participant(s) and the researcher, and whether the researcher was blind to experimental condition and/or the study hypothesis during data collection.

Timing Indicate the start and stop dates of data collection. If there is a gap between collection periods, state the dates for each sample cohort.

Data exclusions If no data were excluded from the analyses, state so OR if data were excluded, provide the exact number of exclusions and the rationale behind them, indicating whether exclusion criteria were pre-established.

Non-participation State how many participants dropped out/declined participation and the reason(s) given OR provide response rate OR state that no participants dropped out/declined participation.

Randomization If participants were not allocated into experimental groups, state so OR describe how participants were allocated to groups, and if allocation was not random, describe how covariates were controlled. April 2018 Ecological, evolutionary & environmental sciences study design All studies must disclose on these points even when the disclosure is negative. Study description Briefly describe the study. For quantitative data include treatment factors and interactions, design structure (e.g. factorial, nested, hierarchical), nature and number of experimental units and replicates.

2 Describe the research sample (e.g. a group of tagged Passer domesticus, all Stenocereus thurberi within Organ Pipe Cactus National

Research sample nature research | reporting summary Monument), and provide a rationale for the sample choice. When relevant, describe the organism taxa, source, sex, age range and any manipulations. State what population the sample is meant to represent when applicable. For studies involving existing datasets, describe the data and its source.

Sampling strategy Note the sampling procedure. Describe the statistical methods that were used to predetermine sample size OR if no sample-size calculation was performed, describe how sample sizes were chosen and provide a rationale for why these sample sizes are sufficient.

Data collection Describe the data collection procedure, including who recorded the data and how.

Timing and spatial scale Indicate the start and stop dates of data collection, noting the frequency and periodicity of sampling and providing a rationale for these choices. If there is a gap between collection periods, state the dates for each sample cohort. Specify the spatial scale from which the data are taken

Data exclusions If no data were excluded from the analyses, state so OR if data were excluded, describe the exclusions and the rationale behind them, indicating whether exclusion criteria were pre-established.

Reproducibility Describe the measures taken to verify the reproducibility of experimental findings. For each experiment, note whether any attempts to repeat the experiment failed OR state that all attempts to repeat the experiment were successful.

Randomization Describe how samples/organisms/participants were allocated into groups. If allocation was not random, describe how covariates were controlled. If this is not relevant to your study, explain why.

Blinding Describe the extent of blinding used during data acquisition and analysis. If blinding was not possible, describe why OR explain why blinding was not relevant to your study. Did the study involve field work? Yes No

Field work, collection and transport Field conditions Describe the study conditions for field work, providing relevant parameters (e.g. temperature, rainfall).

Location State the location of the sampling or experiment, providing relevant parameters (e.g. latitude and longitude, elevation, water depth).

Access and import/export Describe the efforts you have made to access habitats and to collect and import/export your samples in a responsible manner and in compliance with local, national and international laws, noting any permits that were obtained (give the name of the issuing authority, the date of issue, and any identifying information).

Disturbance Describe any disturbance caused by the study and how it was minimized.

Reporting for specific materials, systems and methods

Materials & experimental systems Methods n/a Involved in the study n/a Involved in the study Unique biological materials ChIP-seq Antibodies Flow cytometry Eukaryotic cell lines MRI-based neuroimaging Palaeontology Animals and other organisms Human research participants

Unique biological materials

Policy information about availability of materials April 2018 Obtaining unique materials Describe any restrictions on the availability of unique materials OR confirm that all unique materials used are readily available from the authors or from standard commercial sources (and specify these sources).

Antibodies Antibodies used The following antibodies were used:anti-GST antibody (1: 1000, Cell Signaling

3 Technology,2622) , His-Tag Mouse mAb (1:1000,Cell Signaling Technology, 2366), Rabbit

Phospho-Src Family (Tyr416) (1:1000, Cell Signaling Technology,2101S). c-src antibody nature research | reporting summary (1:2000, Santa Cruz, sc-130124), Fyn antibody (1:1000, Santa Cruz, sc-51598) , PLCγ1 Antibody (1:1000, Santa Cruz,sc-7290), GRB2 Antibody (1:2000, Santa Cruz, sc-255), p-β2-AR (Ser355/Ser356)-R Antibody (1:1000, Santa Cruz sc-16719-R), and HA-probe Antibody (1:1000, Santa Cruz,sc-7392) .Monoclonal anti-Flag M2 antibody (1:1000, Sigma Aldrich, F3165).

Validation All antibodies are well characterized and were applied according to datasheet information details. Additional controls were included in the present study, wherever necessary. Antibodies validation : In Western Blot experiments, all antibodies produced specific signals only for positive controls.

Eukaryotic cell lines Policy information about cell lines Cell line source(s) HEK293 cells were obtained from Cell Resource Center of Shanghai Institute for Biological Sciences (Chinese Academy of Sciences, Shanghai, China). sf9 cells were purchased from Merckmillipore.

Authentication No further authentication were used.

Mycoplasma contamination All of the cell lines were tested regularly for mycoplasma contamination.

Commonly misidentified lines No commonly misidentified cell lines were used. (See ICLAC register)

Palaeontology Specimen provenance Provide provenance information for specimens and describe permits that were obtained for the work (including the name of the issuing authority, the date of issue, and any identifying information).

Specimen deposition Indicate where the specimens have been deposited to permit free access by other researchers.

Dating methods If new dates are provided, describe how they were obtained (e.g. collection, storage, sample pretreatment and measurement), where they were obtained (i.e. lab name), the calibration program and the protocol for quality assurance OR state that no new dates are provided. Tick this box to confirm that the raw and calibrated dates are available in the paper or in Supplementary Information.

Animals and other organisms Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research Laboratory animals For laboratory animals, report species, strain, sex and age OR state that the study did not involve laboratory animals.

Wild animals Provide details on animals observed in or captured in the field; report species, sex and age where possible. Describe how animals were caught and transported and what happened to captive animals after the study (if killed, explain why and describe method; if released, say where and when) OR state that the study did not involve wild animals.

Field-collected samples For laboratory work with field-collected samples, describe all relevant parameters such as housing, maintenance, temperature, photoperiod and end-of-experiment protocol OR state that the study did not involve samples collected from the field.

Human research participants Policy information about studies involving human research participants Population characteristics Describe the covariate-relevant population characteristics of the human research participants (e.g. age, gender, genotypic information, past and current diagnosis and treatment categories). If you filled out the behavioural & social sciences study design questions and have nothing to add here, write "See above."

Recruitment Describe how participants were recruited. Outline any potential self-selection bias or other biases that may be present and how

these are likely to impact results. April 2018

ChIP-seq Data deposition Confirm that both raw and final processed data have been deposited in a public database such as GEO. Confirm that you have deposited or provided access to graph files (e.g. BED files) for the called peaks.

4 For "Initial submission" or "Revised version" documents, provide reviewer access links. For your "Final submission" document,

Data access links nature research | reporting summary May remain private before publication. provide a link to the deposited data.

Files in database submission Provide a list of all files available in the database submission.

Genome browser session Provide a link to an anonymized genome browser session for "Initial submission" and "Revised version" documents only, to (e.g. UCSC) enable peer review. Write "no longer applicable" for "Final submission" documents. Methodology Replicates Describe the experimental replicates, specifying number, type and replicate agreement.

Sequencing depth Describe the sequencing depth for each experiment, providing the total number of reads, uniquely mapped reads, length of reads and whether they were paired- or single-end.

Antibodies Describe the antibodies used for the ChIP-seq experiments; as applicable, provide supplier name, catalog number, clone name, and lot number.

Peak calling parameters Specify the command line program and parameters used for read mapping and peak calling, including the ChIP, control and index files used.

Data quality Describe the methods used to ensure data quality in full detail, including how many peaks are at FDR 5% and above 5-fold enrichment.

Software Describe the software used to collect and analyze the ChIP-seq data. For custom code that has been deposited into a community repository, provide accession details.

Flow Cytometry Plots Confirm that: The axis labels state the marker and fluorochrome used (e.g. CD4-FITC). The axis scales are clearly visible. Include numbers along axes only for bottom left plot of group (a 'group' is an analysis of identical markers). All plots are contour plots with outliers or pseudocolor plots. A numerical value for number of cells or percentage (with statistics) is provided. Methodology Sample preparation Describe the sample preparation, detailing the biological source of the cells and any tissue processing steps used.

Instrument Identify the instrument used for data collection, specifying make and model number.

Software Describe the software used to collect and analyze the flow cytometry data. For custom code that has been deposited into a community repository, provide accession details.

Cell population abundance Describe the abundance of the relevant cell populations within post-sort fractions, providing details on the purity of the samples and how it was determined.

Gating strategy Describe the gating strategy used for all relevant experiments, specifying the preliminary FSC/SSC gates of the starting cell population, indicating where boundaries between "positive" and "negative" staining cell populations are defined. Tick this box to confirm that a figure exemplifying the gating strategy is provided in the Supplementary Information.

Magnetic resonance imaging Experimental design Design type Indicate task or resting state; event-related or block design.

Specify the number of blocks, trials or experimental units per session and/or subject, and specify the length of each trial Design specifications April 2018 or block (if trials are blocked) and interval between trials.

Behavioral performance measures State number and/or type of variables recorded (e.g. correct button press, response time) and what statistics were used to establish that the subjects were performing the task as expected (e.g. mean, range, and/or standard deviation across subjects).

5 Acquisition nature research | reporting summary Imaging type(s) Specify: functional, structural, diffusion, perfusion.

Field strength Specify in Tesla

Sequence & imaging parameters Specify the pulse sequence type (gradient echo, spin echo, etc.), imaging type (EPI, spiral, etc.), field of view, matrix size, slice thickness, orientation and TE/TR/flip angle.

Area of acquisition State whether a whole brain scan was used OR define the area of acquisition, describing how the region was determined.

Diffusion MRI Used Not used Preprocessing Preprocessing software Provide detail on software version and revision number and on specific parameters (model/functions, brain extraction, segmentation, smoothing kernel size, etc.).

Normalization If data were normalized/standardized, describe the approach(es): specify linear or non-linear and define image types used for transformation OR indicate that data were not normalized and explain rationale for lack of normalization.

Normalization template Describe the template used for normalization/transformation, specifying subject space or group standardized space (e.g. original Talairach, MNI305, ICBM152) OR indicate that the data were not normalized.

Noise and artifact removal Describe your procedure(s) for artifact and structured noise removal, specifying motion parameters, tissue signals and physiological signals (heart rate, respiration).

Volume censoring Define your software and/or method and criteria for volume censoring, and state the extent of such censoring. Statistical modeling & inference Model type and settings Specify type (mass univariate, multivariate, RSA, predictive, etc.) and describe essential details of the model at the first and second levels (e.g. fixed, random or mixed effects; drift or auto-correlation).

Effect(s) tested Define precise effect in terms of the task or stimulus conditions instead of psychological concepts and indicate whether ANOVA or factorial designs were used.

Specify type of analysis: Whole brain ROI-based Both

Statistic type for inference Specify voxel-wise or cluster-wise and report all relevant parameters for cluster-wise methods. (See Eklund et al. 2016)

Correction Describe the type of correction and how it is obtained for multiple comparisons (e.g. FWE, FDR, permutation or Monte Carlo). Models & analysis n/a Involved in the study Functional and/or effective connectivity Graph analysis Multivariate modeling or predictive analysis

Functional and/or effective connectivity Report the measures of dependence used and the model details (e.g. Pearson correlation, partial correlation, mutual information).

Graph analysis Report the dependent variable and connectivity measure, specifying weighted graph or binarized graph, subject- or group-level, and the global and/or node summaries used (e.g. clustering coefficient, efficiency, etc.).

Multivariate modeling and predictive analysis Specify independent variables, features extraction and dimension reduction, model, training and evaluation metrics. April 2018

6