Identifying Functional Hotspot Residues for Biased Ligand Design in G- Protein-Coupled Receptors

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Title Page

Identifying Functional Hotspot Residues for Biased Ligand Design in G- protein-coupled Receptors

Anita K. Nivedha, Christofer S. Tautermann, Supriyo Bhattacharya, Sangbae Lee, Paola Downloaded from Casarosa, Ines Kollak, Tobias Kiechle, and Nagarajan Vaidehi

Department of Molecular Immunology, Beckman Research Institute of the City of Hope, molpharm.aspetjournals.org

1500, E. Duarte Road, Duarte, CA – 91010. (A.N., S.B., S.L., N.V.)

Dept. Medicinal Chemistry, Boehringer Ingelheim Pharma GmbH & Co. KG,

Birkendorfer Straße 65, 88397 Biberach, Germany (C.T.) at ASPET Journals on October 2, 2021 Dept. Imm. and Resp. Diseases Research, Boehringer Ingelheim Pharma GmbH & Co.

KG, Birkendorfer Straße 65, 88397 Biberach, Germany (I.K., T.K.)

Corp. Dept. Bus. Development & Licensing, C.H. Boehringer Sohn, Binger Str. 173,

55216 Ingelheim, Germany (P.C.)

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Running Title Page

Running Title: Identifying Functional Hotspots for Biased Ligand Design

Corresponding Author:

Nagarajan Vaidehi Department of Molecular Immunology, Downloaded from Beckman Research Institute of the City of Hope, 1500, E. Duarte Road Duarte, CA – 91010.

Phone: 626-301-8408 molpharm.aspetjournals.org Fax: 626-301-8186

Number of text pages: 32

Number of tables: 0 at ASPET Journals on October 2, 2021 Number of figures: 6 Number of references: 35 Number of words in the Abstract: 233 Number of words in the Introduction: 927 Number of words in the Discussion: 507 List of Non-Standard Abbreviations used: β2AR, β-2 Adrenergic Receptor; BAI, β-arrestin Interface residues; BS, Binding Site residues; EC, Extracellular Region residues; GPI, G-protein Interface residues; κOR, κ- Opioid Receptor; PR: Allosteric Pipeline Ratio

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Abstract G-protein-coupled receptors (GPCRs) mediate multiple signaling pathways in the cell, depending on the agonist that activates the receptor and multiple cellular factors. Agonists that show higher potency to specific signaling pathways over others are known as “biased agonists” and have been shown to have better therapeutic index. Although biased agonists are desirable, their design poses several challenges to date. The number of assays to

identify biased agonists seems expensive and tedious. Therefore, computational methods Downloaded from that can reliably calculate the possible bias of various ligands ahead of experiments and provide guidance will be both cost and time effective. In this work, using the mechanism molpharm.aspetjournals.org of allosteric communication from the extracellular region to the intracellular transducer protein coupling region in GPCRs, we have developed a computational method, to calculate ligand bias ahead of experiments. We have validated the method for several β-

arrestin biased agonists in β2-adrenergic receptor, serotonin receptors 5HT1B and 5HT2B at ASPET Journals on October 2, 2021 and for G-protein biased agonists in the κ-opioid receptor. Using this computational method, we also performed a blind prediction followed by experimental testing and showed that the agonist carmoterol is β-arrestin biased in β2-adrenergic receptor. Additionally, we have identified amino acid residues in the biased agonist binding site in both β2-adrenergic and κ-opioid receptors that are involved in potentiating the ligand bias. We call these residues as “functional hotspots” and they can be used to derive pharmacophores to design biased agonists in GPCRs.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Introduction

G-protein-coupled receptors (GPCRs) are a superfamily of signaling proteins of critical importance to cellular function, and which comprise the largest group of drug targets. The binding of endogenous agonists to their target GPCRs results in coupling to a variety of intracellular transducer proteins including various heterotrimeric G-proteins and

β-arrestins, that mediate different downstream signaling pathways. Certain GPCR agonists Downloaded from elicit differential responses to different signaling pathways and are known as “biased agonists”, and the phenomenon is called “functional selectivity” or “biased signaling”

(Urban et al., 2006; Rajagopal et al., 2010; Rasmussen et al., 2011; Wisler, Xiao, AR molpharm.aspetjournals.org

Thomsen, et al., 2014). The ability of biased agonists to cause differential signaling resulting in more focused cellular changes could translate to lower unfavorable side-effects of a drug. This property of biased agonists can markedly improve their therapeutic index when compared to unbiased or balanced agonists (Mailman, 2007; Neve, 2009; Violin et at ASPET Journals on October 2, 2021 al., 2014). Design of biased agonists remains a daunting challenge due to the complexity of cellular and structural factors that contribute to the bias. Therefore, an understanding of the structural underpinnings of biased signaling, of both G-protein bias as well as -arrestin bias of GPCRs is needed to design biased agonists for GPCRs.

Linking GPCR conformation to cellular function – There are several factors that contribute to bias signaling by an agonist-GPCR pair in cells. (a) Structural features of the agonist-

GPCR-G-protein or agonist-GPCR--arrestin complex (Shukla et al., 2014), and (b) cell- specific and tissue-specific factors (Zhou and Bohn, 2014). Designing biased agonists is a challenge due to lack of structural information on how an agonist-GPCR pair selectively couples to specific G-proteins or -arrestins. Moreover, since GPCRs are dynamic with

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 various possible active state conformations, the structural ensemble of the agonist-GPCR- transducer complex is one of the key determinants driving the biased signaling (Shukla et al., 2014). The selectivity of GPCRs to trigger distinct intracellular signaling profiles is very ligand-dependent. Spectroscopic studies have now established that ligands differentially stabilize a range of GPCR conformations (Yao et al., 2006; Kahsai et al.,

2011; Liu et al., 2012; Nygaard et al., 2013). Therefore, knowledge about how the agonist affects the agonist-GPCR-G-protein structural ensemble will significantly aid the design Downloaded from of biased agonists.

Challenges in Biased Ligand Identification and Design: Given the huge challenges in molpharm.aspetjournals.org crystallography and NMR in determining structural ensembles of agonist-GPCR-effector complexes, computational methods can greatly aid in delineating the structural factors that determine agonist bias and its outcome. Despite a wealth of biochemical and biophysical at ASPET Journals on October 2, 2021 studies on inactive conformations, there is paucity of structural information on active conformations of GPCRs. More importantly the dynamics of the agonist-GPCR-effector complex is one of the major factors leading to a conformational ensemble, which governs the functional selectivity of biased signaling. A better understanding of the structural basis of G protein/-arrestin selection will provide for a more precise targeting of GPCRs with pharmaceuticals and also inform ongoing structure-based drug discovery efforts. Hence, our goal in this work is to identify structural elements in the GPCR that confer functional selectivity either to the G-protein signaling pathway or to the -arrestin signaling pathway.

Our outstanding question is how do the various biased agonists influence the role of these structural determinants in biasing the signaling to effect -arrestin signaling or the G- protein signaling? Here we have developed a computational method to calculate the ligand

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 bias ahead of experiments, that would aid the design of biased agonists. The two-part goal in this work is (a) to develop a method to compute the ligand bias ahead of experiments, and (b) to identify the residues in the agonist binding site that potentiate the ligand mediated bias. We define these residues as the “functional hotspots”. To this end we chose systems that cover a wider base of bias systems as described below.

The human β2 adrenergic receptor (2AR) and angiotensin receptor AT1R have Downloaded from been studied extensively for discovery of -arrestin biased ligands (Violin et al., 2014;

Wisler et al., 2014). Therefore, robust experimental data on the quantitative ligand bias factor calculated using the Black-Leff model (Black and Leff, 1983) is available for these molpharm.aspetjournals.org receptors. We chose 2AR for our study, since the three-dimensional crystal structure of the fully active Gs-protein bound state is available and this constitutes a robust test case for the computational method. The experimental values of the ligand bias factor have been at ASPET Journals on October 2, 2021 calculated using cAMP secondary messenger assays for assessing the G-protein coupling potency for the 2AR and using TANGO or other -arrestin recruitment assays for calculating the ligand potency towards -arrestin mediated signaling pathways (Rajagopal et al., 2011; Weiss et al., 2013; Zhou et al., 2013). There has been a surge in efforts to develop G-protein biased opioid receptor agonists that serve as better analgesics with reduced side effects coming from receptor desensitization (Soergel et al., 2014; Luttrell et al., 2015; Stahl et al., 2015). We tested three agonists, two of which showed different extent of G-protein bias for κ-opioid receptor (κOR) (Zhou et al., 2013). We chose to study the

κOR instead of the μOR since the active state structure of κOR is not available and we wanted to test our computational method using a homology model of the receptor that makes the testing more realistic.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

The serotonin receptors 5HT1B and 5HT2B have been crystallized with the same agonist ergotamine, yet ergotamine shows -arrestin bias in 5HT2B and not in 5HT1B

(Wacker et al., 2013). We chose to study the system to find out which residues in 5HT2B cause the bias in ergotamine. This example would provide a validation for the same ligand showing different bias in two related receptors. Another similar example is the bias behavior of epinephrine in the triple mutant T68F, Y132G, Y219A 2AR, denoted here as Downloaded from 2ARTYY. This mutant was designed using evolutionary trace analysis to identify residues that could cause bias in 2ARTYY (Shenoy et al., 2006). Thus, choosing the variety of

different systems described above provides a test for the robustness of our computational molpharm.aspetjournals.org method. To summarize, we have studied 2AR and its biased mutant 2ARTYY with several agonists, 5HT1B and 5HT2B and κOR with three agonists including two G-protein biased agonists. These systems are also listed in Table S1 of the Supplemental Information for at ASPET Journals on October 2, 2021 convenience.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Materials and Methods

A New Computational Method for Calculating Ligand Bias

-1 The experimentally measured ligand bias factor depends on the ligand affinity (KA ) and its efficacy (τ) towards a specific signaling pathway. The bias factor is calculated with respect to a reference ligand using the Black and Leff operational model (Kenakin et al.,

2012). Since macroscopic properties such as ligand potency and coupling efficacy are still Downloaded from intractable computationally, we developed a method to calculate the ligand bias from atomic level properties of the structural ensemble of the agonist-GPCR complex that

potentiates the agonist bias. Agonist and G-protein binding to GPCRs leads to G-protein molpharm.aspetjournals.org activation. Such activation at a distance is potentiated by the allosteric communication between the ligand binding site and the G-protein coupling site in the fully active state of

GPCRs. We recently developed a computational method to calculate the strength of the allosteric communication pipelines from the extracellular region of the GPCR to the at ASPET Journals on October 2, 2021 intracellular regions where the G-protein or -arrestin couples to the receptor. In this work we hypothesize that the strength of the allosteric communication pipelines from the extracellular (EC) region passing through the residues in the ligand binding site (ligand shown in blue spheres in Figure 1) to the residues in the G-protein and -arrestin coupling interface in the GPCRs is related to the ligand potency and its coupling efficacy. Figure 1 shows the scheme used here for calculating the ligand bias in GPCRs. The allosteric communication pipelines that connect to the G-protein interface is shown in pink sticks and those to the β-arrestin interface is shown in green sticks in Figure 1. We defined the ligand bias as the ratio of the strength of the allosteric communication pipelines to the G- protein pathway to that of the -arrestin signaling pathway for a test ligand to that of the

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 same quantity for the reference ligand. We use a reference ligand in these calculations so that these quantities can be directly compared to the experimental bias factor.

Computational Methods

The computational methods to calculating the allosteric communication pipelines in

GPCRs involve the following steps (i) preparation of the GPCR structure with agonists for the molecular dynamics (MD) simulations, and (ii) performing MD simulations in explicit Downloaded from membrane bilayer and water. The list of receptors used in this work is shown in

Supplemental Table S1. The two-dimensional representation of the agonists used in this molpharm.aspetjournals.org study is shown in Supplemental Figure S1. The details of the receptor structure preparation, docking of the agonists and the MD simulations are provided in the Supplemental

Information file. We performed brute force MD simulations with no constraints for wild type β2AR with several ligands listed in Supplemental Table S1, β2ARTYY mutant with at ASPET Journals on October 2, 2021 epinephrine, serotonin receptors 5HT1B and 5HT2B with ergotamine and homology model of κOR with three different agonists bound.

Allosteric Pipeline Analysis:

The MD simulation trajectories were used to calculate the allosteric communication pipelines for each agonist-GPCR pair using the Allosteer computational method

(Bhattacharya and Vaidehi, 2014b; c; Bhattacharya et al., 2016; Vaidehi and Bhattacharya,

2016) that we have developed previously. The details of the Allosteer computational method, for calculating the allosteric communication pipelines in GPCRs have been described in references (Bhattacharya and Vaidehi, 2014b; c). Here we provide a brief

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 description of the method as used in this work. Using the MD trajectories, we first calculate the correlated movement between residues in the EC surface of the receptor and the residues in G-protein or -arrestin coupling interface. Here we have used the trajectories from the five 200ns of MD simulations collected to a total of 1μs of MD simulations, to calculate the mutual information in torsion angle space between all pairs of residues in each receptor-ligand complex. Using the mutual information, we then calculate the shortest Downloaded from pathway from the EC region to the G-protein or -arrestin coupling surfaces that maximizes the correlated movement. This method is called Allosteer (Bhattacharya and

Vaidehi, 2014b). Using the Allosteer method we then calculated the allosteric molpharm.aspetjournals.org communication pipelines starting from the residues in the extracellular (EC) loops (defined in the section titled “Residue group definitions” below) passing through the residues in the agonist binding site to the residues in the -arrestin or G-protein interacting interfaces. The at ASPET Journals on October 2, 2021 list of residues in each of these regions is given in the Supplemental Table S2. For each

MD run, the top 10% of allosteric pathways, ranked by the total mutual information were used for further calculation of the ligand bias. The strength of an allosteric communication pipeline is the number of overlapping allosteric communication pathways contained in the pipeline (Bhattacharya and Vaidehi, 2014b; Vaidehi and Bhattacharya, 2016).

Residue Groups definitions:

Here we list the residues that define various regions in the GPCRs. These residues were included in the calculation of the allosteric communication pipelines.

Ligand Binding Site residues (BS): The residues located within 5Å of the agonist binding site in the GPCR and contacting the agonist in more than 40% of the snapshots from the

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

MD simulation trajectories were chosen as the binding site residues for that agonist. The super set of binding site residues for all agonists to a particular receptor was listed by combining the residues list for all the agonists to that receptor studied here. This superset of residues was used to calculate the allosteric communication pipeline from the EC loops going through the ligand binding site to the G-protein or β-arrestin coupling interface. The list of residues in the ligand binding site used in this study is shown in Supplemental Table

S2. Downloaded from

β-arrestin Interface residues (BAI) and G-protein Interface residues (GPI): The list of residues in the GPCR interacting with the G-protein was obtained from Supplementary molpharm.aspetjournals.org

Table 1 of the crystal structure paper of β2AR-Gs protein complex (Rasmussen et al.,

2011). The PDB ID of this crystal structure is 3SN6. The list of residues interacting with the β-arrestin interface was obtained by aligning the β2AR structure from PDB ID: 3SN6 to the crystal structure of rhodopsin bound to visual arrestin (PDB ID: 5DGY) (Zhou et al., at ASPET Journals on October 2, 2021

2016) and identifying the residues in the receptor that are within 5Å of visual arrestin in the aligned structure. The κOR structure was aligned to that of β2AR and the BAI and GPI residues were translated for the κOR based on the alignment of residues and corresponding

BW numbers. The list of residues in the GPI and BAI used in this study are shown in

Supplemental Table S2.

Extracellular Region residues (EC): Residues comprising the extracellular loops of the receptor in addition to those residues forming two turns in the extracellular region of each transmembrane helix formed the set of EC residues. The EC loop residues included in the calculation of allosteric communication pipelines for the GPCRs studied here are also given in Supplemental Table S2.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Definition of the Computational ligand bias:

Using the MD trajectories we calculated the allosteric communication pipelines of every agonist-GPCR complex to the G-protein and -arrestin interface. We then calculated the ratio of the strength of the allosteric communication pipeline for the -arrestin coupling interface to that of the G-protein coupling interface residues. We did the same calculation Downloaded from for all agonists including the reference agonist, which is usually a balanced agonist, used in the experimental assays for the bias factor calculations. The computational ligand bias

values is defined as molpharm.aspetjournals.org

푃푅 Computational Ligand Bias, β = 푙푖푔 − 1 --- (1) 푃푅푟푒푓

Where PRlig is the ratio of strength of the allosteric communication pipeline for given agonist-receptor pair for the -arrestin to the G-protein coupling interfaces, and PRref is at ASPET Journals on October 2, 2021 the same for a reference agonist.

Testing the robustness of the method to calculate the Ligand bias

The computational ligand bias shown in equation (1) above, is a ratio that is obtained by calculating the strength of the allosteric communication pipelines to the -arrestin and to the G-protein coupling interfaces. The error in the calculation of the strength of the allosteric pipelines is dependent on (i) the number of snapshots used from the MD trajectories and (ii) the simulation time length of each MD simulation. The dependence on the number of snapshots stored in the MD simulations is because the state of the system has to be captured often enough to enable the recording of all significant, meaningful correlations. However, recording snapshots at extremely frequent intervals would increase

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 the probability of registering spurious correlations and dependencies, which in turn will introduce noise into the bias calculations. The second factor which is the length of the MD simulations is important because, longer MD simulations will capture collective correlated motions of torsion angles in proteins. However, a trade-off has to be made between the cost of computational resources and accuracy of ligand bias calculated here. Therefore, we examined the optimum number of snapshots to be stored and length of MD simulations to reduce the significant errors in the ligand bias calculations. Thus, we calculated the Downloaded from variation in the standard error in the pipeline strength as a function of these two variables.

The standard error in the above calculations was analyzed with the aim of choosing optimal molpharm.aspetjournals.org values for both (Figures S2 and S3). Upon analysis of the standard error values, storing snapshots from MD simulations at every 20ps timesteps in a 200ns MD simulation (10,000 snapshots) was chosen as the optimum frame rate and 200ns as the simulation time length for the calculations. For more details on these tests we refer the reader to the Supplemental at ASPET Journals on October 2, 2021

Information.

Experimental Methods

β-Arrestin recruitment assay and cAMP Detection Assay

To test the predictivity of the computational methods, 3 compounds (isoproterenol, formoterol and carmoterol) have been tested in β-arrestin assay as well as cAMP assay on human β2-adrenoceptors. Isoproterenol and formoterol were purchased from Sigma-

Aldrich - Germany, and carmoterol was synthesized using the protocol described in the

US Patent US2010/0113790 A1 (Kankan et al., 2010). For β-arrestin assay the

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

PathHunter technology was chosen and for cAMP the AlphaScreen technology was employed.

β-Arrestin recruitment assay

The PathHunter® β-Arrestin Assay system (DiscoveRx) was used to determine β-arrestin recruitment to human β2-adrenoceptors. CHO-K1 cells stably expressing beta2- adrenoceptors fused to ProLink™ and the β-arrestin-enzyme acceptor were seeded onto Downloaded from opaque 384 well plates at 5.000 cells per well in DMEM/F12 (1:1) without phenol red supplemented with 1% heat inactivated foetal bovine serum. Cells were maintained at molpharm.aspetjournals.org 37°C, 5% CO2, humidified atmosphere overnight. The next day cells were stimulated in growth medium with different concentrations of agonists at 37°C for 60, 30, 15 and 5 minutes. Arrestin recruitment, resulting in β-galactosidase enzyme complementation was detected by use of the PathHunter Flash detection Kit (DiscoveRx). Luminescence was at ASPET Journals on October 2, 2021 read on an Envision plate reader (PerkinElmer Life and Analytical Sciences). We performed n=13 experiments for isoproterenol, n=9 for formoterol and n=4 for carmoterol.

cAMP Detection Assay

Changes in intracellular cAMP levels were determined by AlphaScreen technology

(PerkinElmer Life and Analytical Sciences) in CHO-K1 cells stably expressing beta2- adrenoceptors following manufacturers` instructions. Cells in suspension (15.000 cells per well) were stimulated in opaque 384 well plates with respective agonists at different concentrations in Hanks' buffered saline solution supplemented with 5 mM HEPES, 0.1% bovine serum albumin, and 500 mM 3-isobutyl-1-methylxanthine for 30 min at room temperature. Cells were lysed by using Alphascreen reagents. After 2 h, plates were read

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 on an Envision plate reader (PerkinElmer Life and Analytical Sciences). The concentration of cAMP in the samples was calculated from a standard curve. We performed n=12 experiments for isoproterenol, formoterol and carmoterol.

Black and Leff Model used for calculating the bias factor

The concentration response data for each agonist at the G protein and -arrestin signaling assays were fitted to the Black-Leff model of agonism (Black and Leff, 1983) Downloaded from

n n [A] t (Em - Basal) Response- Basal = n n n [A] t +([A]+ KA ) … (1), molpharm.aspetjournals.org where [A] is the agonist concentration, τ is equal to RT/KE (RT: receptor density, KE: intrinsic agonist efficacy), n is the transducer slope, and Em is the maximal response of the system. The transduction coefficient for each agonist for a given pathway was calculated as log(τ/KA) (Kenakin and Christopoulos, 2013). For fitting the data, the Black-Leff at ASPET Journals on October 2, 2021 equation was recast to a different form according to (Tschammer et al., n.d.) and log(τ/KA) was directly obtained from the fit. Em was estimated as the maximum value of the signaling response for a given assay. Setting the value of n=1 gives a good fit for all the dose- response data. In practice, n was allowed to vary within a very narrow range (0.9-1.2) to account for statistical variability. The fitting was performed using the Genetic Algorithm

(GA) module in MATLAB (operational model fitting of GraphPad Prism did not always find a solution for all datasets). Instead of starting from a single initial guess of the solution,

10,000 initial guesses were randomly generated within a prescribed range. The provided

-15 range for KA was 10 to 1, while for log(τ/KA) it was 0 to 15 (range for n is stated earlier).

Using the different initial guesses, the algorithm converged to a solution within the

-8 provided tolerance limit of 10 . For each agonist and a given pathway, Δlog(τ/KA) was

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

calculated as log(τ/KA)- log(τ/KA)ref, where log(τ/KA)ref is the value for the reference agonist isoproterenol. ΔΔlog(τ/KA) was then calculated as Δlog(τ/KA)arrestin- Δlog(τ/KA)G protein. Thus,

ΔΔlog(τ/KA) bias factor is given by 10 . To estimate the variability in the calculated log(τ/KA)

2 values, we calculated Sij for agonist i and pathway j as:

nj 2 1 2 Sij = å(yijk - ymean ) n -1 ij k=1 … (2) where k denotes the number of experiments and y corresponds to the log(τ/KA) for the Downloaded from agonist/pathway pair. The total variability of the estimates Spooled is given by:

2 3 S2 åi=1å j=1 ij S = molpharm.aspetjournals.org pooled df error … (3) where dferror is the degree of freedom given by:

2 3 dferror = å å (nij -1) i=1 j=1 … (4) The 95% confidence levels for the calculated log(τ/KA) values are given by: at ASPET Journals on October 2, 2021 c.l. = log(τ/KA)±T(dferror,0.975) x (standard error) … (5),

where standard error is given by:

1 standard error = S pooled n ij …(6), and T corresponds to a two-tailed t-test with 95% confidence. For calculating the confidence levels for ΔΔlog(τ/KA), we used the same method (eqn. 5), except the standard error is given by:

… (7), where n refers to the number of experiments for the given pathway, and nref is the number of experiments for the reference ligand.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Results Calculation of Ligand Bias Factor using Black-Leff operational model for carmoterol and formoterol:

We have measured the ligand bias factor of carmoterol and formoterol using isoproterenol as the reference ligand. Using the concentration dependent curves measured using the cAMP assay for G-protein activation and PathHunter assay for β-arrestin recruitment as detailed in the Methods section, we have calculated the ligand bias factor Downloaded from for carmoterol and formoterol with reference to isoproterenol in 2AR. We used the Black-

Leff model to obtain bias factors as described in the Methods section. The calculated bias molpharm.aspetjournals.org factors are plotted in Figure 2A and also shown in Supplemental Table S3. Our results show that carmoterol has a two-fold higher bias factor towards the -arrestin signaling pathway compared to formoterol in 2AR.

at ASPET Journals on October 2, 2021

Comparison of calculated ligand bias to experimental bias factor:

Figures 2A and 2B show comparison of computed ligand bias to the experimental ligand bias factor for various β-arrestin and G-protein biased agonists for the 2AR and

κOR respectively. In one set of experiments epinephrine was used as a reference ligand to calculate the bias factor from experimental titration curves (Rajagopal et al., 2011), while isoproterenol was used as a reference ligand in another set of experimental measurements

(Weiss et al., 2013). Supplemental Figures S4A and S4B show the same comparisons for

5HT receptors and 2ARTYY mutant. As seen in Figures 2A and 2B the computed ligand bias recapitulates the rank ordering of the ligand bias factors for all the agonists in all the four receptors for both β-arrestin and G-protein biased agonists. It is important to note that

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 the ligand bias ordering has been captured for the G-protein biased agonists where we used a homology model for κOR. The magnitude of computed ligand bias shown in Figure 2A is an order of magnitude lower than the experimental ligand bias factor. This is because we have not computed the macroscopic property of the ligand bias factor as measured by experiments. Instead we have computed the relative strength of the allosteric communication pipelines for the test ligand to that of the reference ligand and this is a molecular factor that potentiates and leads to change in ligand potency and efficacy that Downloaded from manifests as ligand bias. Thus, the computational method provides a rapid method to identify biased ligands ahead of experiments that is useful in designing other biased molpharm.aspetjournals.org ligands.

The agonists epinephrine, isoproterenol and salbutamol are considered balanced agonists to 2AR (Rajagopal et al., 2011) while formoterol, BI-167107 and carvedilol are arrestin-biased, with carvedilol acting as an inverse agonist to G-protein (Wisler et al., at ASPET Journals on October 2, 2021

2007). In the study by Lefkowitz and coworkers, the bias factors of epinephrine, isoproterenol, formoterol and salbutamol was measured, and the authors concluded that formoterol was an arrestin-biased agonist, while the remaining ligands failed to exhibit any considerable bias towards either arrestin or G-protein (Rajagopal et al., 2011). As shown in Figure 2A, in agreement with experiments, formoterol is more biased towards β-arrestin compared to isoproterenol or salbutamol in 2AR. Similarly, the super agonist BI-167107 is more biased towards β-arrestin compared to isoproterenol. Carvedilol is a weak β- arrestin biased agonist, and a G-protein antagonist, and therefore, experimental bias factor values could not be obtained. However, the computational ligand bias values indicate that carvedilol is indeed β-arrestin biased. As a blind test to our method using MD simulations

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 and calculation of ligand bias, we further predicted that carmoterol is a β-arrestin biased ligand compared to isoproterenol. We then performed the cAMP and the PathHunter assays as described in the Results section and calculated the bias factors shown in Supplemental

Table S3. As predicted, carmoterol is indeed β-arrestin-biased with reference to isoproterenol.

In κOR (Figure 2B), we correctly predict that the agonists 1.1 and 2.1 are G-protein biased compared to U69593. The experimental results (Zhou et al., 2013) showed that Downloaded from using the ligand U69593 as the reference agonist, probe 1.1 and probe 2.1 are G-protein biased, with probe 1.1 being a stronger G-protein biased agonist compared to 2.1. molpharm.aspetjournals.org

Theoretical ligand bias calculations were obtained for these 3 ligand-receptor systems, and the values were found to be in excellent agreement with the experimental bias factor values, not only indicating the G-protein bias of both probes with respect to U69593, but also reflecting the greater G-protein bias for probe 1.1 compared to probe 2.1. at ASPET Journals on October 2, 2021

Comparison of the computational ligand bias for ergotamine in 5HT1B and 5HT2B as well as epinephrine in the 2AR and 2ARTYY mutant shows good agreement with the experimental bias factors reported (Supplemental Figures S4). The serotonin receptors are an example of a system with different receptor serotypes in complex with the same ligand, and one receptor system being neutral while the other is -arrestin-biased. The computational ligand bias values were in agreement with experimental bias factor values.

There is no quantitative experimental bias factor available for the 2ARTYY mutant.

The above results for several systems and different types of agonists show the robustness and the reliability of our method in predicting the ligand bias ahead of experiments. As described in detail in the Methods section, the confidence level for the

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 prediction of ligand bias is 90%. Thus, the theoretical ligand bias calculations were validated against experimental bias factor values, and were found to be clearly indicative of whether a preference for either G-protein and/or -arrestin exists.

Identifying functional hotspot residues in the ligand binding site that contribute to the bias:

Figures 3A and 3B show the comparison of the shape of the ligand binding sites in Downloaded from epinephrine (neutral agonist) and formoterol (β-arrestin biased agonist) bound to the fully active state of 2AR respectively as modeled in this work. For fair comparison, this figure

shows the same view in the same receptor orientation for both the agonists. Unlike molpharm.aspetjournals.org epinephrine, the binding site of formoterol extends to the extracellular loop 2 (ECL2) region of the receptor. The residues D1133.32, F193ECL2, S2035.42 and S2045.43 although present in both the ligand binding sites, show different side chain conformations at ASPET Journals on October 2, 2021 (Supplemental Figure S5) in formoterol bound 2AR compared to the epinephrine bound

2AR. Thus, changes in the rotameric states of these residues reshape the common regions of the cavities forming the ligand binding site for both epinephrine and formoterol. The allosteric communication pipelines from the extracellular loops passing through the residues in the formoterol binding site to the β-arrestin coupling interface are stronger in formoterol-2AR complex as shown in Figure 3B than in the epinephrine-2AR complex.

However, it is interesting to note that the allosteric communication pipelines to the G- protein interface are equally strong in both epinephrine and formoterol (Figure 3A). Both epinephrine and formoterol show strong G-protein communication pipelines in the fully active state of the receptor and the allosteric pipeline strength to the β-arrestin coupling interface is comparatively weaker in epinephrine than in formoterol. This is also true in the

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

experimental assays (Rajagopal et al., 2011), where Rajagopal et al show that the Emax for formoterol and epinephrine are similar for the cAMP response while that for -arrestin recruitment of lower for epinephrine than for formoterol. To date there is no known biased agonist that shows strong preference for β-arrestin and not to G-protein in 2AR. It should be noted that we have used the fully active G-protein bound conformation of 2AR for all these calculations and the allosteric pipeline calculations are sensitive to the Downloaded from conformational state of the receptor (Bhattacharya and Vaidehi, 2014a).

One of the challenges in designing biased agonists is the lack of structural details

of biased agonist binding sites that could aid Structure Activity Relationship (SAR) studies. molpharm.aspetjournals.org

The binding site of biased agonists looks similar to that of balanced agonist from crystal structures (Warne et al., 2012). Therefore, it would be beneficial to the SAR studies, if we could delineate the residues in the biased agonist binding site that contribute to the ligand at ASPET Journals on October 2, 2021 bias. Towards this goal, we identified the residues in the biased agonist binding site that show significant contribution to the allosteric communication pipelines of either -arrestin or G-protein coupling interface, by calculating the differences in the strengths of the allosteric communication pipelines to the BAI and to the GPI for the biased and unbiased agonists. Figures 4A and 4B show the residues within 5Å of formoterol and epinephrine binding sites respectively. The larger circles show sustained contact with the ligand. There are many residues that are common to both the epinephrine and formoterol binding sites.

As seen in Figure 4A, the four residues T1103.29, F193ECL2, S2045.43, and Y3087.35 show significant contribution to the allosteric communication pipeline to the -arrestin interface than to the G-protein pipelines with reference to epinephrine shown in Figure 4B. These residues are the functional hotspot residues that confer -arrestin bias to formoterol-2AR

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 pairing. Carmoterol bound 2AR also shows T1103.29, F193ECL2 and S2045.43 as functional hotspots in its binding site. Thus, three functional hotspot residues (T1103.29, F193ECL2 and

S2045.43) are common to both formoterol and carmoterol. This shows that GPCRs might have specific residues that show functional selectivity to signaling pathways in the receptor. The 3D view showing the relative orientation of these residues in the formoterol binding site is shown in the Supplemental Figure S6. To identify which chemical moieties Downloaded from in formoterol contribute to bias, we identified the moieties in formoterol that contact the functional hotspot residues in its binding site. The functional hotspots T1103.29 and Y3087.35

show strong contact with the 4-methoxyphenyl-propanyl group attached to the protonated molpharm.aspetjournals.org amine (Figure S6). This moiety that contacts the functional hotspots are absent in epinephrine. Therefore, extending the agonist beyond the protonated amine group and involving contacts with T1103.29 and Y3087.35 could yield biased agonists in 2AR.

at ASPET Journals on October 2, 2021

Identifying functional hotspot residues in the G-protein biased agonist binding site in κOR:

Figure 5A and 5B show the residues within 5Å of the G-protein biased κOR agonist probe

1.1 and that of the unbiased agonist U69593, respectively. U69593 predominantly nested in the region between TM2, TM3 and TM7 (Figure 5B). On the other hand, the G-protein biased agonist 1.1 is more flexible in the binding site and made contacts with residues in

TM2, TM3, TM5, TM6 as shown in Figure 5A. The residues Y652.64, D843.32, K1735.39,

I2406.55, and Y2587.35 are functional hotspot residues in the 1.1 biased agonist binding site that contribute significantly to the strength of the allosteric communication pipeline to the

G-protein interacting interface compared to the -arrestin interacting interface. The residues Y652.64, D843.32, I2406.55, and Y2587.35 are common functional hotspot residues to

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 both 1.1 and 2.1. The residues D843.32 and Y652.64 are common to both 1.1 and U69593 binding sites but the residues K1735.39, I2406.55, and Y2587.35 are unique to 1.1. The residues D843.32 and Y652.64 also show a difference in the side chain rotamer conformations when bound to U69593 or probe 1.1 (Supplemental Figure S7). Y2587.35A mutation in the

κOR has a strong deleterious effect on the potencies of agonists U69593, dynorphin A,

1xx and salvinorin A (Vardy et al., 2013). Mutation of residues D843.32 to Ala/Asn and

Y652.64 to Ala/Phe, negatively affected the affinity and potency of the receptor in the Downloaded from presence of U69593, 1xx and salvinorin A (Vardy et al., 2013). In summary, we find that the G-protein biased agonists in κOR displayed a rather rigid side-chain behavior, whereas, molpharm.aspetjournals.org the -arrestin biased agonists, such as formoterol in 2AR displayed a more flexible side chain rotamer distribution. Supplemental Figure S8 shows the 3D orientation of the probe

1.1 bound to κOR.

Thus, the functional hotspot residues in the ligand binding site identified by our at ASPET Journals on October 2, 2021 method can be used for (i) generating a pharmacophore for identifying newer -arrestin biased ligands for 2AR and (ii) for Structure-Activity-Relationship studies for enhancing or improving the bias factor.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Discussion

Using MD simulation trajectories, and the Allosteer method to calculate the allosteric communication pipelines, we have developed a computational method to calculate the ligand bias in GPCRs. The ligand bias is defined as the relative strength of allosteric communication pipelines starting from the residues in the EC loops passing through the residues in the agonist binding site to the residues in the -arrestin or G-protein Downloaded from interacting interface in the GPCR. We find that the calculated ligand bias shows the same ordering as the experimental bias factors. Using the computations we correctly predicted

that carmoterol is -arrestin biased in 2AR with respect to isoproterenol. We have molpharm.aspetjournals.org identified the functional hotspot residues in both 2AR and κOR as shown in Figures 4A and 5A respectively, that can be used in designing -arrestin or G-protein biased ligands.

Even to date, the identification and optimization of biased ligands is serendipity-driven. at ASPET Journals on October 2, 2021 Usually the starting points for ligand optimization are identified in finding campaigns, such as high-throughput screening, fragment-based screening, or virtual screening. The hit compounds are tested for their ligand bias experimentally and the promising starting points for further optimization have to show the desired biased profile to begin. The knowledge of functional hotspots can revolutionize this canonical hit-finding process. Using our computational method can be beneficial in two ways. (1) To identify biased agonists from virtual screening and, (2) using the functional hotpots to modify any hit ligand as a starting point for bias optimization. This allows us to pick starting points for lead optimization based on crucial parameters such as pharmacokinetic properties and off-target profiles rather than the bias of the initial hit. The ligand bias can be built in by modifications to contact the functional hotspots.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Another important sought-after tool in the pharmaceutical industry is to be able to assess if a particular target GPCR is amenable to bias or “biasable”, similar to assessing the “druggability” of a target protein. Therefore, we asked the question as to which of these two receptors 2AR and κOR, studied here is more “biasable”. Figures 6A and 6B show a heat map of the residue contribution to the differential -arrestin and G-protein allosteric communication pipelines in formoterol-bound active state of 2AR and 1.1-bound active Downloaded from state of κOR respectively. The heat map is an indication of the difference in the strength of allosteric communication by each residue toward -arrestin in 2AR and towards G-

protein in κOR. The more red the residue and thicker the size of the backbone cartoon in molpharm.aspetjournals.org

Figures 6A and 6B, the greater is its contribution to the ligand bias. We observe that there are large clusters of residues on TM5 and TM6 with high strength of allosteric communication towards G-protein coupling interface in κOR. Such strong functional at ASPET Journals on October 2, 2021 hotspot residues are absent in the 2AR. This could mean that κOR is probably more readily biasable compared to 2AR. In summary, our computational method to calculate the ligand bias and identify functional hotspot residues in GPCRs is extensible to other class A GPCRs, and will greatly aid biased ligand design for GPCRs.

Authorship Contributions

Participated in research design: Vaidehi, Tautermann, Nivedha

Conducted experiments: Nivedha, Lee and Bhattacharya conducted the simulations and calculations. Kollak and Kiechle conducted the experiments.

Performed data analysis: Nivedha, Bhattacharya, Vaidehi, Tautermann

Wrote or contributed to the writing of the manuscript: Nivedha, Vaidehi, Tautermann

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

References

Bhattacharya S, Salomon-Ferrer R, Lee S, and Vaidehi N (2016) Conserved Mechanism

of Conformational Stability and Dynamics in G-Protein-Coupled Receptors. J Chem

Theory Comput 12:5575–5584, American Chemical Society.

Bhattacharya S, and Vaidehi N (2014a) Differences in allosteric communication pipelines

in the inactive and active states of a GPCR. Biophys J 107:422–434, Biophysical Downloaded from Society.

Bhattacharya S, and Vaidehi N (2014b) Differences in Allosteric Communication

Pipelines in the Inactive and Active States of a GPCR. Biophys J 107:422–434. molpharm.aspetjournals.org

Bhattacharya S, and Vaidehi N (2014c) Mapping Allosteric Communication Pipelines in

GPCRs from Microsecond Timescale Molecular Dynamics Simulations. Biophys J

106:635a, Elsevier.

Black JW, and Leff P (1983) Operational Models of Pharmacological Agonism. Proc R at ASPET Journals on October 2, 2021

Soc London B Biol Sci 220.

Kahsai AW, Xiao K, Rajagopal S, Ahn S, Shukla AK, Sun J, Oas TG, and Lefkowitz RJ

(2011) Multiple ligand-specific conformations of the β2-adrenergic receptor. Nat

Chem Biol 7:692–700, Nature Publishing Group, a division of Macmillan Publishers

Kankan RN, Rao DR, Birari D, and Sawant AA (2010) Process for preparing isomers of

carmoterol, US, United States.

Kenakin T, and Christopoulos A (2013) Signalling bias in new drug discovery: detection,

quantification and therapeutic impact. Nat Rev Drug 12:205-16

Kenakin T, Watson C, Muniz-Medina V, Christopoulos A, and Novick S (2012) A

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Simple Method for Quantifying Functional Selectivity and Agonist Bias. ACS Chem

Neurosci 3:193–203, American Chemical Society.

Liu JJ, Horst R, Katritch V, Stevens RC, and Wüthrich K (2012) Biased Signaling

Pathways in β2-Adrenergic Receptor Characterized by 19F-NMR. Science 335:

1106-10.

Luttrell LM, Maudsley S, and Bohn LM (2015) Fulfilling the Promise of Biased G

Protein–Coupled Receptor Agonism. Mol Pharmacol 88: 579-88. Downloaded from

Mailman RB (2007) GPCR functional selectivity has therapeutic impact. Trends

Pharmacol Sci 28:390–396. molpharm.aspetjournals.org

Neve K (2009) Functional selectivity of G protein-coupled receptor ligands: new

opportunities for drug discovery.

Nygaard R, Zou Y, Dror RO, Mildorf TJ, Arlow DH, Manglik A, Pan AC, Liu CW, Fung

JJ, Bokoch MP, Thian FS, Kobilka TS, Shaw DE, Mueller L, Prosser RS, and at ASPET Journals on October 2, 2021

Kobilka BK (2013) The Dynamic Process of β2-Adrenergic Receptor Activation.

Cell 152:532–542.

Rajagopal S, Ahn S, Rominger DH, Gowen-macdonald W, Lam CM, Dewire SM, Violin

JD, and Lefkowitz RJ (2011) Quantifying Ligand Bias at Seven-Transmembrane

Receptors. Mol Pharmacol 80:367-77.

Rajagopal S, Rajagopal K, and Lefkowitz RJ (2010) Teaching old receptors new tricks:

biasing seven-transmembrane receptors. Nat Rev Drug Discov 9:373–86, NIH

Public Access.

Rasmussen SGF, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, Thian FS,

Chae PS, Pardon E, Calinski D, Mathiesen JM, Shah STA, Lyons JA, Caffrey M,

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Gellman SH, Steyaert J, Skiniotis G, Weis WI, Sunahara RK, and Kobilka BK

(2011) Crystal structure of the β2 adrenergic receptor-Gs protein complex. Nature

477:549–55, NIH Public Access.

Shenoy SK, Drake MT, Nelson CD, Houtz DA, Xiao K, Madabushi S, Reiter E, Premont

RT, Lichtarge O, and Lefkowitz RJ (2006) beta-arrestin-dependent, G protein-

independent ERK1/2 activation by the beta2 adrenergic receptor. J Biol Chem

281:1261–73, American Society for Biochemistry and Molecular Biology. Downloaded from

Shukla AK, Singh G, and Ghosh E (2014) Emerging structural insights into biased GPCR

signaling. Trends Biochem Sci 39:594–602. molpharm.aspetjournals.org

Soergel DG, Subach RA, Burnham N, Lark MW, James IE, Sadler BM, Skobieranda F,

Violin JD, and Webster LR (2014) Biased agonism of the μ-opioid receptor by

TRV130 increases analgesia and reduces on-target adverse effects versus morphine:

A randomized, double-blind, placebo-controlled, crossover study in healthy at ASPET Journals on October 2, 2021

volunteers. Pain 155:1829–1835.

Stahl EL, Zhou L, Ehlert FJ, and Bohn LM (2015) A Novel Method for Analyzing

Extremely Biased Agonism at G Protein–Coupled Receptors. Mol Pharmacol 87:

866-77.

Tschammer N, Bollinger S, Kenakin T, Gmeiner P (2011) Histidine 6.55 is a major

determinant of ligand-biased signaling in dopamine D2L receptor. Mol Pharmacol

79:575-85.

Urban JD, Clarke WP, von Zastrow M, Nichols DE, Kobilka B, Weinstein H, Javitch JA,

Roth BL, Christopoulos A, Sexton PM, Miller KJ, Spedding M, and Mailman RB

(2006) Functional Selectivity and Classical Concepts of Quantitative Pharmacology.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

J Pharmacol Exp Ther 320: 1-13.

Vaidehi N, and Bhattacharya S (2016) Allosteric communication pipelines in G-protein-

coupled receptors. Curr Opin Pharmacol 30:76–83.

Vardy E, Mosier PD, Frankowski KJ, Wu H, Katritch V, Westkaemper RB, Aubé J,

Stevens RC, and Roth BL (2013) Chemotype-selective modes of action of Κ-opioid

receptor agonists. J Biol Chem 288:34470–34483. Downloaded from Violin JD, Crombie AL, Soergel DG, and Lark MW (2014) Biased ligands at G-protein-

coupled receptors: promise and progress. Trends Pharmacol Sci 35:308–316.

Wacker D, Wang C, Katritch V, Han GW, Huang X-P, Vardy E, McCorvy JD, Jiang Y, molpharm.aspetjournals.org

Chu M, Siu FY, Liu W, Xu HE, Cherezov V, Roth BL, and Stevens RC (2013)

Structural Features for Functional Selectivity at Serotonin Receptors. Science 340:

615-9.

Warne T, Edwards PC, Leslie AGW, and Tate CG (2012) Crystal structures of a at ASPET Journals on October 2, 2021

stabilized β1-adrenoceptor bound to the biased agonists bucindolol and carvedilol.

Structure 20:841–849, Cell Press.

Weiss DR, Ahn S, Sassano MF, Kleist A, Zhu X, Strachan R, Roth BL, Lefkowitz RJ,

and Shoichet BK (2013) Conformation Guides Molecular E ffi cacy in Docking

Screens of Activated β2 Adrenergic G Protein Coupled Receptor. ACS Chem Biol,

17:1018-26.

Wisler JW, DeWire SM, Whalen EJ, Violin JD, Drake MT, Ahn S, Shenoy SK, and

Lefkowitz RJ (2007) A unique mechanism of beta-blocker action: carvedilol

stimulates beta-arrestin signaling. Proc Natl Acad Sci U S A 104:16657–62.

Wisler JW, Xiao K, Thomsen ARB, and Lefkowitz RJ (2014) Recent developments in

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

biased agonism. Curr Opin Cell Biol 27:18–24, NIH Public Access.

Yao X, Parnot C, Deupi X, P Ratnala VR, Swaminath G, Farrens D, and Kobilka B

(2006) Coupling ligand structure to specific conformational switches in the β2 -

adrenoceptor. Nat Chem Biol 2: 417-22.

Zhou L, and Bohn LM (2014) Functional selectivity of GPCR signaling in animals. Curr

Opin Cell Biol 27:102–108.

Zhou L, Lovell KM, Frankowski KJ, Slauson SR, Phillips AM, Streicher JM, Stahl E, Downloaded from

Schmid CL, Hodde P, Madoux F, Cameron MD, Prisinzano TE, Aubé J, and Bohn

LM (2013) Development of functionally selective, small molecule agonists at kappa molpharm.aspetjournals.org

opioid receptors. J Biol Chem 288:36703–36716.

Zhou X, Gao X, Barty A, Kang Y, He Y, and Liu W (2016) X-ray laser diffraction for

structure determination of the rhodopsin-arrestin complex. Sci data 3: 160021.

at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Footnotes

This work was funded by Boehringer-Ingelheim and by National Institutes of Health

[R01-GM097296].

Downloaded from molpharm.aspetjournals.org at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Figure Legends

Figure 1: The model used for calculating the ligand bias in GPCRs. The allosteric communication pipelines starting from the residues in the extracellular loop region and passing through the residues in the agonist binding site (agonist shown in blue spheres) and terminating in the residues that couple to the G-protein (GPI- shown in pink surface) are shown as pink sticks. The allosteric communication pipelines that connect to the β-arrestin Downloaded from interface (BAI shown in green surface) are shown in green sticks. The ratio of the strength of these two pipelines for a test ligand with respect to a reference ligand is defined as the molpharm.aspetjournals.org ligand bias.

Figure 2. Comparison of computational ligand bias (y axis) to the experimental ligand bias factors (x axis) for A. the β-arrestin biased agonists in β2AR and B. for G-protein at ASPET Journals on October 2, 2021 biased agonists in κOR. In figure A, the bias values obtained using epinephrine as the reference agonist are shown as green triangles, while those obtained using isoproterenol as the reference agonist are shown in green circles. The respective reference agonists from both studies are marked at zero. The error bars for the bias factors of all β2AR agonists represent standard errors with 95% confidence limits except for BI-167107 which is 99% confidence limit. The number of experiments for these cases were n3. In figure B, the bias values obtained for the G-protein biased system κOR is shown using

U69593 as the reference agonist. For the OR agonists, the experimental bias factor was calculated from at least three independent experiments. All computational ligand bias

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395 values are reported with standard error and were obtained from five independent MD simulations.

Figure 3: The shape of the agonist binding cavity in β2AR of A) a neutral agonist epinephrine and, B) a biased agonist, formoterol. The allosteric communication pipelines to the G-protein and the β-arrestin coupling interfaces are also depicted. This figure shows

β2AR surfaces sliced to reveal the shape of the ligand binding cavity. Downloaded from

Figure 4: A. The residues within 5Å of formoterol in 2AR. B. the binding site residues molpharm.aspetjournals.org within 5Å of epinephrine in 2AR. The size of the circles is proportional to the percentage of snapshots from the MD simulations that show the contacts. Large circles are for agonist- residue contacts present in more than 75% of the snapshots from MD simulations; medium size circles - between 60-75% of the snapshots and small circles between 40-60% of the at ASPET Journals on October 2, 2021 snapshots. Residues shown in red text with yellow highlight are the functional hotspot residues.

Figure 5: A. The residues within 5Å of the G-protein biased agonist 1.1 in κOR. B. The binding site residues within 5Å of the balanced agonist U69593 in in κOR. The size of the circles is proportionate to the percentage of snapshots from the MD simulations that show the contacts. Large circles are for agonist-residue contacts present in more than 75% of the snapshots from MD simulations; medium size circles - between 60-75% of the snapshots and small circles between 40-60% of the snapshots. Residues shown in red text with yellow highlight are the functional hotspot residues.

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

MOL #110395

Figure 6: Heat map depicting the extent of bias in β2AR and in κOR. The thickness of the cartoon is directly proportional to the contribution of that residue to arrestin or G-protein bias. A. β2AR in complex with formoterol (β-arrestin biased) regions of the receptor which are strongly β-arrestin biased are represented by shades of blue to red. B. κOR in complex with probe 1.1 (G-protein biased) regions of the receptor which are strongly G-protein biased are represented by shades of blue to red. Downloaded from molpharm.aspetjournals.org at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 1

Extracellular Region

Binding site TM2

TM1

TM5 TM7 Downloaded from TM6

GPI molpharm.aspetjournals.org BAI

at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 2

A β-arrestin bias B G-protein bias

0.25 -25 -20 -15 -10 -5 0

s a

i 0.02 s

B 0.20

form a

d B

n carm

a 0.15 d

g u69 -0.02 i n 2.1

L iso

a L

form

n 0.10

o a

i -0.06 t

BI n

o t

0.05 i

t u

epi a p

t -0.10

u m Downloaded from

0.00 p 1.1

o iso

C m

sal o

-0.05 C -0.14 -1 0 1 2 3 4 -0.18 Experimental Bias Factor Experimental Bias Factor

molpharm.aspetjournals.org

at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 3

Downloaded from molpharm.aspetjournals.org

β-arrestin β-arrestin pipeline pipeline G-protein pipeline G-protein at ASPET Journals on October 2, 2021 pipeline

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 4

Downloaded from

molpharm.aspetjournals.org

at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 5

Downloaded from

molpharm.aspetjournals.org at ASPET Journals on October 2, 2021

Molecular Pharmacology Fast Forward. Published on January 24, 2018 as DOI: 10.1124/mol.117.110395 This article has not been copyedited and formatted. The final version may differ from this version.

Figure 6

Downloaded from molpharm.aspetjournals.org

at ASPET Journals on October 2, 2021

Molecular Pharmacology

Identifying Functional Hotspot Residues for Biased Ligand Design in G-protein- coupled Receptors Anita K. Nivedha, Christofer S. Tautermann, Supriyo Bhattacharya, Sangbae Lee, Paola Casarosa, Ines Kollak, Tobias Kiechle, and Nagarajan Vaidehi

Department of Molecular Immunology, Beckman Research Institute of the City of Hope, 1500, E. Duarte Road, Duarte, CA – 91010. (A.N., S.B., S.L., N.V.) Dept. Medicinal Chemistry, Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Straße 65, 88397 Biberach, Germany (C.T.) Dept. Imm. and Resp. Diseases Research, Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Straße 65, 88397 Biberach, Germany (I.K., T.K.) Corp. Dept. Bus. Development & Licensing, C.H. Boehringer Sohn, Binger Str. 173, 55216 Ingelheim, Germany (P.C.)

Supplementary Information

Molecular Pharmacology

Materials and Methods

Computational Methods

Receptor Structure Preparation β2 Adrenergic Receptor (β2AR): The β2AR structure for our study was taken from the nanobody-stabilized active state crystal structure of the receptor in complex with an agonist (PDB ID: 3P0G) (Rasmussen et al., 2011). The starting structure for epinephrine, carvedilol, isoproterenol and carmeterol were obtained from PDB’s 4LDO (Ring et al., 2013), 4AMJ (Warne et al., 2012), 2Y03 and 2Y02 (Warne et al., 2011). The initial positions of formeterol and BI-167107 were obtained by making modifications to the structure of carmeterol from PDB 2Y02, and aligning the two receptors (from 2Y02 and 3P0G), while the starting structure for salbutamol was obtained by making modifications to the structure of isoproterenol from PDB 2Y03 and aligning the two receptors (from 2Y03 and 3P0G).

Serotonin Receptors: The three-dimensional structure for the two serotonin receptor serotypes, 5HT1B and 5HT2B, in complex with ergotamine, were obtained from PDBs 4IAR and 4IB4 (Wang et al., 2013), respectively. For both 5HT1B and 5HT2B, missing loops were added using the MMTSB Toolbox (Feig et al., 2004), followed by annealing the added loops keeping the rest of the protein structures fixed.

Κ-Opioid Receptor (κOR): The active state structure of κOR is not available but the structure of the inactive state of κOR with antagonist bound (PDB ID: 4DJH) (Wu et al., 2012) and the nanobody bound active state of μOR (PDB ID: 5C1M) (Huang et al., 2015) are available. In order to generate a homology model of the active-state κOR we first generated a chimeric template PDB by using the structures of the inactive κOR and the active μOR. TMs1-4, ECL2 and helix 8 were obtained from the inactive κOR and TMs5-7 were obtained from the active μOR by aligning the two receptors using PyMOL (“The PyMOL Molecular Graphics System, Version 1.6 Schrödinger, LLC.,”), and connecting the interface residues using the Draw mode in Maestro. A local minimization of the regions which were joined to create the chimeric template model (choosing ± 2 residues 151-156, 173-177, 289-293 at the three points of chimerization) was performed, followed by unrestrained minimization using MacroModel. Missing residues were added using Prime. Homology modeling was performed using MODELLER 9.0 (Šali and Blundell, 1993) and the top 20 models were analyzed. The best homology model was chosen based on the lowest RMSD (0.91 Å) to the chimeric template model.

The homology modeled active state structure of the κOR was passed through the Protein Preparation Wizard in Maestro ver. 9.1.207, followed by Receptor Grid Generation. A van der Waals scaling factor of 0.5 was used, while the partial charge cut-off was set to 1.0. The grid box was centered at the centroid of residues determined as important for binding based on mutation studies of the receptor (Zhou et al., 2013). The inner box dimension was set as 15x15x15Å.

Molecular Pharmacology

Ligand Preparation & Docking: The ligands for β2AR and κOR were built using Maestro ver. 9.1.207. They were minimized in MacroModel using the OPLS 2005 forcefield in MacroModel, with a distance-dependent dielectric. A single-point quantum calculation was performed using the Hartree-Fock theory with 6-31G** basis set and a PBF water solvation model using Jaguar (Bochevarov et al., 2013). Prior to docking, a conformational ensemble for each ligand was generated using the MacroModel Conformational Search. This ensemble of structures was then docked to the receptor using Standard Precision docking in Glide. A van der Waals scaling factor of 0.5 and a partial charge cut-off of 1.0 was used, and the output format requested was to write out at most 10 poses per ligand. The docked poses were evaluated based on the formation of critical interactions within the binding pocket, determined using mutation studies and based on similarity to conformations of comparable structures obtained from the Cambridge Structural Database. The interactions deemed as important for the binding of U69593 were with residues D1383.32, Y1192.64 and Y3207.43. For probe 1.1, it was D1383.32, Y1192.64, Y3137.36, I2627.39 and Y3207.43, and finally for probe 2.1 the residues selected were D1383.32, Y1192.64, Y3137.36 and Y3207.43 (Mansour et al., 1997; Kane et al., 2006, 2008, Frankowski et al., 2012; Martinez-Mayorga et al., 2013; Vardy et al., 2013).

Molecular Dynamics (MD) simulation protocol: All MD simulations were performed using the GROMACS ver. 5.1 package and GROMOS forcefield (Oostenbrink et al., 2004) at 300K. The receptor-agonist complex was placed in a POPC lipid bilayer box with 128 POPC and surrounded by water molecules and neutralizing Cl- ions. The topology file (.itp) and GROMACS file (.gro) for the ligand was obtained using the webserver PRODRG (Schüttelkopf and van Aalten, 2004). The solvent and solute were coupled to a temperature bath with a relaxation time of 0.1 ps. The pressure was calculated using a molecular virial and was held constant by weak coupling to a pressure bath with a relaxation time of 0.5 ps. For all the equilibration simulations, positional restraints were applied, and the simulations were performed at constant volume (NVT), during which, no pressure coupling was applied. The bond lengths and geometry of water molecules were constrained using the SHAKE algorithm. The equations of motion were integrated at a timestep of 2 fs and using the leapfrog algorithm. Center of mass motion was removed every 20 fs. The short-range van der Waals and electrostatic interactions were evaluated at every time step by using a charge-group pair list with a cutoff radius of 8 Å between the centers of geometry of the charge groups. Longer-range van der Waals and electrostatic interactions, between pairs at a distance greater than 8 Å and shorter than a long-range cutoff (14 Å), were evaluated every fifth time step, during which the pair list was also updated, and kept unchanged between updates. The receptor-agonist complexes were equilibrated under NVT conditions, for 10ns in the case of β2AR, β2ARTYY and serotonin receptors and for 25ns for the homology model of κOR. During the equilibration of the receptor−agonist complexes, the atoms of the complex were restrained in their positions using a harmonic restraining force with a force constant of 1000 kJ/(mol·nm) during the 200 ps of equilibration. In this step, the water molecules and lipid bilayer were allowed to move to optimize their packing around the receptor. The system was further equilibrated using NPT ensemble, while the force constant of the restraining force was set to 5 kcal/mol and reduced to zero stepwise at each 5 ns, and the Molecular Pharmacology

pressure coupling was switched on. This results in a total equilibration time of 40 ns. We performed an additional 10 ns of simulations without restraints before the production runs. Five independent simulations each to 200 ns were performed with different starting velocities amounting to a total of 1μs of total simulation time per system. The average representative structures from MD simulations were calculated by first performing conformation clustering on the combined MD simulation trajectories from the five simulations for any given system. The most representative structure of the most populated cluster was calculated as the frame that has the smallest RSMD to the center of this cluster of conformations. All MD simulation runs were analyzed for convergence by plotting the RMSDs of the Cα atoms of the residues in the transmembrane regions of the receptor and that of the ligand.

Analysis of the standard error in the ligand bias calculations The error in the calculation of the strength of the allosteric pipelines is dependent on two variables: (i) The number of snapshots used from the MD trajectories and (ii) the simulation time length of each MD simulation. We calculated the variation in the standard error in the pipeline strength as a function of the number of snapshots stored from MD simulations. We stored snapshots ranging from 5000 to 40,000 within the 200ns of MD simulation length. The standard error in the calculation of the strength of the allosteric communication pipeline using different number of snapshots is shown in Figure S2. Upon analysis of the standard error values, we concluded that storing snapshots from MD simulations every 20ps should be adequate for a 200ns length simulation time. We collect statistics from MD simulations by running multiple simulations rather than one long simulation. This provides a wider conformational sampling given the limitations of reaching ergodicity. To analyze how long of a simulation time should each MD simulation run be, we analyzed the standard error in the ratio of the strength of the allosteric pipelines to the beta-arrestin to that of the G-protein PRlig in equation (1) in the main text, as a function of simulation time as shown in Figure S3. The standard error is calculated as Std. error = Std. Dev./sqrt(N) where N = 5. Running individual simulations for 200ns shows lower standard errors in PRlig values calculated for several ligand-GPCR systems. To assess the sensitivity of the ligand bias calculations to the docked structure of the agonist we performed the following test. We performed MD simulations following the same protocol starting from a docked structure of carmoterol that is 0.9Å away from the crystal structure of carmoterol used in this study. Additionally, we intentionally chose a carmoterol docked structure that did not have contacts between the catechol group and all the three serines, S203, S204 and S207, of the receptor as seen in the crystal structure of carmoterol bound β-1-adrenergic receptor. The calculated ligand bias is 0.1395 with mean error of 0.016. The ligand bias value calculated previously using MDs starting from the crystal structure was 0.14. Thus, the docked structure leads to ligand bias that is similar to that of the crystal structure.

Calculation of the residue-based allosteric hub score We calculated the extent of contribution to the allosteric pipelines to both the - arrestin and G-protein coupling interfaces by each residue in the GPCR. We define this as the allosteric hub score of each residue. The allosteric hub score is the simply the number of allosteric communication pathways passing through each residue in the GPCR either to Molecular Pharmacology

the -arrestin or G-protein coupling interface. For more details see reference Bhattacharya 2014, and Bhattacharya 2016.

Molecular Pharmacology

Table S1: List of all systems used in the current study, including the Computational Ligand Bias values and details about the studies from which the Experimental Bias Factors were obtained.

Reference Biased Computational System Description Experimental Ref. Receptor Agonist Receptor Agonist Ligand Bias Bias β2AR WT isoproterenol 0.12 β2AR WT epinephrine β2AR WT formoterol β-arrestin 0.21 [ref] • Same WT receptor β2AR WT salbutamol -0.02 • Different β-arrestin biased ligands β2AR WT isoproterenol β2AR WT BI-167107 β-arrestin 0.04 [ref] β2AR WT formoterol 0.08 β2AR WT isoproterenol β-arrestin SI Table S3 β2AR WT carmeterol 0.14 • Same WT receptor κOR WT probe 1.1 -0.13 • Different G-protein κOR WT u69593 G-protein biased ligands κOR WT probe 2.1 -0.04 • Different receptor serotypes 5HT1B ergotamine 5HT2B ergotamine β-arrestin 0.32 [ref] • Same ligand • WT and Mutant receptors β2AR WT epinephrine β2AR TYY epinephrine β-arrestin 0.40 Unavailable • Same ligand

Molecular Pharmacology

Table S2: The list of residues in the GPCR β-arrestin interface and G-protein interface that we have used in this study for calculating the allosteric pipelines for the different GPCRs. G-protein Extracellul β-arrestin interface interface Binding Site -ar Region System residues (BAI) residues residues (BS) residues (GPI) (EC) 63, 64, 65, 66, 131, 134, 135, 136, 131, 134, 135, 136, 93, 109, 110, 113, 114, 117, 30-41, 88-110, β-2AR 137, 138, 139, 140, 142, 143, 146, 138, 139, 141, 142, 165, 169, 191, 192, 193, 195, 162-204, 291- and 222, 225, 271, 274, 278, 329, 331 143, 222, 225, 226, 200, 203, 204, 207, 208, 286, 313 TYY 229, 230, 232, 233, 289, 290, 293, 294, 297, 305, β-2AR 239, 271, 274, 275 308, 309, 312, 316 35, 38, 40, 41, 70, 73, 102, 105, 106, 102, 105, 106, 107, 8, 9, 12, 54, 55, 57, 58, 61, 62, 1-9, 60-79, 135- 107, 108, 109, 110, 112, 113, 114, 109, 110, 112, 113, 64, 65, 66, 68, 70, 80, 81, 82, 173, 238-259 195, 198, 199, 202, 203, 205, 209, 114, 193, 196, 197, 83, 84, 85, 87, 88, 156, 158, κ-OR 210, 211, 213, 214, 221, 222, 280, 200, 201, 203, 204, 172, 173, 176, 177, 232, 233, 281, 282 210, 217, 220, 221 235, 236, 237, 239, 240, 243, 255, 258, 259, 262, 265, 266, 268, 269 79, 80, 82, 84, 85, 147, 150, 151, 147, 150, 151, 152, 38, 39, 109, 110, 125, 126, 129, 38-49, 96-118, 152, 153, 154, 155, 156, 158, 159, 157, 158, 159, 231, 130, 133, 134, 199, 200, 201, 170-212, 340- 5HT1B 161, 162, 231, 234, 237, 238, 239, 234, 235, 238, 239, 212, 213, 217, 327, 330, 331, 362 306, 307, 310, 311, 312, 313, 314, 313, 314, 317 334, 351, 352, 355, 359 315, 316, 370, 375 84, 85, 87, 42, 43, 153, 156, 157, 153, 156, 157, 158, 114, 115, 121, 131, 132, 135, 48-60, 108-132, 158, 159, 160, 161, 162, 163, 164, 160, 161, 162, 163, 136, 139, 140, 207, 208, 209, 186-221, 342- 165, 166, 170, 240, 243, 244, 246, 240, 243, 244, 247, 217, 218, 221, 222, 225, 337, 362 5HT2B 247, 248, 314, 315, 316, 317, 318, 323, 326, 327 340, 341, 344, 347, 359, 362, 319, 320, 321, 322, 323, 324, 325, 363, 366, 370 326, 380, 383, 384, 386

Molecular Pharmacology

Table S3: Experimental bias factors for agonists formeterol and carmeterol using reference agonist isoproterenol. We used the Black-Leff model to calculate the bias factors as described in the Methods section of the main text. The error bars correspond to 95% confidence limits corresponding to 2 tailed t-test. The number of experiments is n=3.

Reference ligand Biased Experimental Error Experimental Bias Factor Receptor Agonist Receptor Agonist Negative Positive

formeterol 1.06 isoproterenol β2AR 0.43 0.55 β2AR WT WT carmeterol 2.14 0.77 1.01

Molecular Pharmacology

Figure S1: Two-dimensional representation of the agonists studied here

Molecular Pharmacology

0.25

0.20

Frame Rate No. of

i l

R (200ns) Snapshots P

/ 0.15

r 5ps 40000

r r

e 10ps 20000

d 0.10 t

S 20ps 10000 40ps 5000 0.05

0.00 5000 10000 20000 40000 epinephrine carvedilol formoterol isoproterenol carmoterol salbutamol bi-167107 tyy

Figure S2: Variation of mean error in allosteric pipeline strength as a function of the number of snapshots from MD simulations. Molecular Pharmacology

0.14"

0.12"

r 0.1"

r E

0.08"

d t

S 0.06"

0.04"

0.02"

0" 50ns" 100ns" 150ns" 200ns" epinephrine" carvedilol" formoterol" isoproterenol" carmoterol" salbutamol" bi<167107" tyy"

Figure S3: Plot of the standard error in the ratio of the strength of the allosteric pipelines to the beta-arrestin to that of the G-protein PRlig in equation (1) in the main text, as a function of simulation time. Std. error = Std. Dev./sqrt(N); N = 5. Running individual simulations for 200ns shows lower standard errors in PRlig values calculated for several ligand-GPCR systems.

Molecular Pharmacology

A 0.50 B 250 0.40 225 200 0.40 0.30 175 0.30 150 125 0.20

100 Bias 0.20 Factors 75 0.10 50

0.10 Theoretical Bias Theoretical Experimental Bias Experimental 25

0 0.00 Ligand Computational 0.00 5HT1B 5HT2B

Figure S4: A. Calculated Ligand Bias for the β2ARTYY mutant in complex with epinephrine compared to that of the WT β2AR in complex with epinephrine. There is no experimental quantitative bias factor available for epinephrine in the β2ARTYY mutant. B. Comparison of computational ligand bias values to experimental bias factors for ergotamine in the two serotonin receptors. The bias factors were taken from reference (Wacker et al., 2013).

Molecular Pharmacology

Epi 3.32 A Asp-113 Form 50 Ser-2035.42 50 40 40

30 30

20 20

10 10

0 0 Populatio Distribution Populatio Population Distribution Population 0 60 120 180 240 300 360 0 60 120 180 240 300 360 -10 -10 Χ1 (deg) Χ1 (deg)

50 50 Ser-2045.43 Phe-193(ECL2) 40 40

30 30

20 20

10 10

0 0 Population Distribution Population 0 60 120 180 240 300 360 Distribution Population 0 60 120 180 240 300 360 -10 -10 Χ (deg) Χ (deg) 1 2

Figure S5: A. Side-chain rotameric angle distribution of residues altering the binding site shape for formoterol vs. epinephrine. The residues F193 and S204 are functional hotspot residues. These residues are common to epinephrine and formeterol that shape the binding site in β2AR. Molecular Pharmacology

Formoterol B D192ECL2 Y3087.35 N2936.55

S2035.42 F193ECL2

5.43 7.39 S204 I3097.36 N312

S2075.46 W1093.28

T1103.29 D1133.32 T1183.37 V1173.36

Figure S6: Three-dimensional orientation of the formeterol binding site in 2AR. The residues shown are present in the formoterol binding cavity in WT β2AR with contacts for over 75% of the MD simulation snapshots. The residues that show differences in rotamer populations between the unbiased epinephrine and biased formeterol are also shown (F193, S203, S204, D113). The agonist, formoterol, is shown in ball and stick representation. The connection to the functional hotspot residues are shown by dashed lines.

Molecular Pharmacology

Figure S7: Side-chain rotameric angle distribution of residues altering the binding site shape for U69593 vs. Probe 1.1. These residues are additionally functional hotspot residues in 1.1.

Molecular Pharmacology

Figure S8: The residues in the probe 1.1 binding cavity in KOR that contact the agonist for over 75% of the MD simulation are shown. The residues that show differences in rotamer populations between the unbiased U69593 and biased probe 1.1 are also shown (Y65, D84). The agonist probe 1.1, is shown in ball and stick representation. Y65, I240, Y258, K173 and D84 are binding site hotspot residues in probe 1.1, and dashed lines show the connections to these residues.

Molecular Pharmacology

References: Bochevarov A, Harder E, Hughes TF, Greenwood JR, Braden DA, Philipp DM, Rinaldo D, Halls, MD, Zhang J, and Friesner RA (2013) Jaguar: A high-performance quantum chemistry software program with strengths in life and materials sciences. J Quantum Chem 113:2110-2142. Feig M, Karanicolas J, and Brooks C (2004) MMTSB Tool Set: enhanced sampling and multiscale modeling methods for applications in structural biology. J Mol Graph Model 22:377-395. Frankowski KJ, Hedrick MP, Gosalia P, Li K, Shi S, Whipple D, Ghosh P, Prisinzano TE, Schoenen FJ, Su Y, Vasile S, Sergienko E, Gray W, Hariharan S, Milan L, Heynen- Genel S, Mangravita-Novo A, Vicchiarelli M, Smith LH, Streicher JM, Caron MG, Barak LS, Bohn LM, Chung TDY, and Aubé J (2012) Discovery of small molecule kappa opioid receptor agonist and antagonist chemotypes through a HTS and Hit refinement strategy. ACS Chem Neurosci 3:221–236. Huang W, Manglik A, and Venkatakrishnan A (2015) Structural insights into μ-opioid receptor activation. Nature 524:315-321. Kane BE, McCurdy CR, and Ferguson DM (2008) Toward a structure-based model of salvinorin A recognition of the kappa-opioid receptor. J Med Chem 51:1824–1830. Kane BE, Nieto MJ, McCurdy CR, and Ferguson DM (2006) A unique binding epitope for salvinorin A, a non-nitrogenous kappa opioid receptor agonist. FEBS J 273:1966– 1974. Mansour a, Taylor LP, Fine JL, Thompson RC, Hoversten MT, Mosberg HI, Watson SJ, and Akil H (1997) Key residues defining the mu-opioid receptor binding pocket: a site-directed mutagenesis study. J Neurochem 68:344–353. Martinez-Mayorga K, Byler KG, Yongye AB, Giulianotti MA, Dooley CT, and Houghten RA (2013) Ligand/kappa-opioid receptor interactions: Insights from the X-ray crystal structure. Eur J Med Chem 66:114–121. Oostenbrink C, Villa A, Mark AE, and Van Gunsteren WF (2004) A biomolecular force field based on the free enthalpy of hydration and solvation: The GROMOS force-field parameter sets 53A5 and 53A6. J Comput Chem 25:1656–1676. Rasmussen SGF, Choi H-J, Fung JJ, Pardon E, Casarosa P, Chae PS, Devree BT, Rosenbaum DM, Thian FS, Kobilka TS, Schnapp A, Konetzki I, Sunahara RK, Gellman SH, Pautsch A, Steyaert J, Weis WI, and Kobilka BK (2011) Structure of a nanobody-stabilized active state of the β(2) adrenoceptor. Nature 469:175–80. Ring A, Manglik A, Kruse A, Enos M, and Weis W (2013) Adrenaline-activated structure of the β2-adrenoceptor stabilized by an engineered nanobody. Nature 502:575-579. Šali A, and Blundell T (1993) Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol 234:779-815. Schüttelkopf AW, and van Aalten DMF (2004) PRODRG : a tool for high-throughput crystallography of protein–ligand complexes. Acta Crystallogr Sect D Biol Crystallogr 60:1355–1363. The PyMOL Molecular Graphics System, Version 1.6 Schrödinger, LLC. Vardy E, Mosier PD, Frankowski KJ, Wu H, Katritch V, Westkaemper RB, Aubé J, Stevens RC, and Roth BL (2013) Chemotype-selective modes of action of kappa- opioid receptor agonists. J Biol Chem 288:34470–34483. Wacker D, Wang C, Katritch V, Han GW, Huang X-P, Vardy E, McCorvy JD, Jiang Y, Molecular Pharmacology

Chu M, Siu FY, Liu W, Xu HE, Cherezov V, Roth BL, and Stevens RC (2013) Structural Features for Functional Selectivity at Serotonin Receptors. Science 340:615-619. Wang C, Jiang Y, Ma J, Wu H, and Wacker D (2013) Structural basis for molecular recognition at serotonin receptors. Science 340:610-614. Warne T, Edwards P, Leslie A, and Tate C (2012) Crystal structures of a stabilized β1- adrenoceptor bound to the biased agonists bucindolol and carvedilol. Structure 20:81- 849. Warne T, Moukhametzianov R, Baker J, and Nehmé R (2011) The structural basis for agonist and partial agonist action on a β1-adrenergic receptor. Nature 469:241-244. Wu H, Wacker D, Katritch V, Mileni M, Han G, and Vardy E (2012) Structure of the human kappa opioid receptor in complex with JDTic. Nature 485:327-332. Zhou L, Lovell KM, Frankowski KJ, Slauson SR, Phillips AM, Streicher JM, Stahl E, Schmid CL, Hodde P, Madoux F, Cameron MD, Prisinzano TE, Aubé J, and Bohn LM (2013) Development of functionally selective, small molecule agonists at kappa opioid receptors. J Biol Chem 288:36703-36716.