Supplemental material to this article can be found at: http://molpharm.aspetjournals.org/content/suppl/2017/03/09/mol.116.106419.DC1

1521-0111/91/5/533–544$25.00 https://doi.org/10.1124/mol.116.106419 MOLECULAR PHARMACOLOGY Mol Pharmacol 91:533–544, May 2017 Copyright ª 2017 by The American Society for Pharmacology and Experimental Therapeutics

Purinergic Receptor Transactivation by the b2-Adrenergic Receptor Increases Intracellular Ca21 in Nonexcitable Cells s

Wayne Stallaert,1,2 Emma T. van der Westhuizen,1,3 Anne-Marie Schönegge, Bianca Plouffe, Mireille Hogue, Viktoria Lukashova, Asuka Inoue, Satoru Ishida, Junken Aoki, Christian Le Gouill, and Michel Bouvier Department of Biochemistry (W.S., E.T.v.d.W., A.-M.S., B.P., M.B.) and Institute for Research in Immunology and Cancer (W.S., E.T.v.d.W., A.-M.S., B.P., M.H., V.L., C.L.G., M.B.), Université de Montréal, Montréal, QC, Canada; Graduate School

of Pharmaceutical Sciences, Tohoku University, Sendai, Miyagi, Japan (A.I., S.I., J.A.); Japan Science and Technology Downloaded from Agency, Precursory Research for Embryonic Science and Technology, Kawaguchi, Saitama, Japan (A.I.); and Japan Agency for Medical Research and Development, Core Research for Evolutional Science and Technology, Chiyoda-ku, Tokyo, Japan (J.A.) Received August 14, 2016; accepted March 6, 2017 molpharm.aspetjournals.org ABSTRACT 21 The b2 adrenergic receptor (b2AR) increases intracellular Ca in receptor antagonist, suggesting a role for this Gq-coupled a variety of cell types. By combining pharmacological and receptor family downstream of the b2AR activation. Consistent genetic manipulations, we reveal a novel mechanism through with this mechanism, b2AR stimulation promoted the extracel- 21 which the b2AR promotes Ca mobilization (pEC50 5 7.32 6 lular release of ATP, and pretreatment with apyrase inhibited the 21 0.10) in nonexcitable human embryonic kidney (HEK)293S cells. b2AR-promoted Ca mobilization. Together, these data sup- Downregulation of Gs with sustained cholera toxin pretreatment port a model whereby the b2AR stimulates a Gs-dependent and the use of Gs-null HEK293 (ΔGs-HEK293) cells generated release of ATP, which transactivates Gq-coupled P2Y receptors using the clustered regularly interspaced short palindromic through an inside-out mechanism, leading to a Gq- and IP3- 21 repeat-associated -9 nuclease (CRISPR/Cas9) system, dependent Ca mobilization from intracellular stores. Given that at ASPET Journals on September 25, 2021 combined with pharmacological modulation of cAMP formation, b2AR and P2Y receptors are coexpressed in various tissues, this revealed a Gs-dependent but cAMP-independent increase in novel signaling paradigm could be physiologically important and 21 intracellular Ca following b2AR stimulation. The increase in have therapeutic implications. In addition, this study reports the cytoplasmic Ca21 was inhibited by P2Y purinergic receptor generation and validation of HEK293 cells deleted of Gs using antagonists as well as a dominant-negative mutant form of Gq, a the CRISPR/Cas9 genome editing technology that will undoubt- Gq-selective inhibitor, and an inositol 1,4,5-trisphosphate (IP3) edly be powerful tools to study Gs-dependent signaling.

This work was supported, in part, by Canadian Institutes for Health Introduction Research (CIHR) [Grant MOP 11215 to M.B.] and grants from Precursory Research for Embryonic Science and Technology, Japan Science and The b2 adrenergic receptor (b2AR) has been shown to Technology Agency [to A.I.], and Japan Agency for Medical Research and regulate a vast signaling network, leading to the activation of Development, Core Research for Evolutional Science and Technology [to J.A.]. W.S. was supported by the Vanier Canada Graduate Scholarship from CIHR. key cellular effectors such as (AC), extracellu- E.T.v.d.W. was supported by postdoctoral research fellowships from CIHR, lar signal-regulated kinase 1/2, and Akt (De Blasi, 1990; Daaka Canadian Hypertension Society, Fonds de la Recherche en Santé du Quebec (FRSQ), and National Health and Medical Research Council Australia [Grant et al., 1998; Bommakanti et al., 2000) to control a variety of GNT-1013819]. A.-M.S. was supported by postdoctoral research fellowships physiologic processes, including the regulation of cardiac func- from FRSQ, and B.P. was supported by postdoctoral research fellowships from CIHR. M.B. holds the Canada Research Chair in and tion, smooth muscle tone, immunologic responses, fat metabo- Molecular Pharmacology. lism, as well as both central and peripheral nervous system 1W.S. and E.T.v.d.W. contributed equally to this work and should be considered co-first authors. activity (Guimarães and Moura, 2001; Collins et al., 2004; 2Current affiliation: Department of Systemic Cell Biology, Max Planck Sitkauskiene and Sakalauskas, 2005; Pérez-Schindler et al., Institute of Molecular Physiology, Dortmund, Germany. 3Current affiliation: Monash Institute for Pharmaceutical Sciences, Monash 2013). Although much is known about the signaling repertoire University, Parkville, Victoria, Australia. of the b2AR, new insights into its full signaling capabilities and https://doi.org/10.1124/mol.116.106419. s This article has supplemental material available at molpharm. its role in cellular biology continue to be discovered (Stallaert aspetjournals.org. et al., 2012; van der Westhuizen et al., 2014).

ABBREVIATIONS: AC, adenylyl cylase; b2AR, b2 adrenergic receptor; BRET, bioluminescent resonance energy transfer; Cch, carbamylcholine chloride; CRISPR/Cas9, clustered regularly interspaced short palindromic repeat-associated protein-9 nuclease; CTX, cholera toxin; FBS, fetal bovine serum; FRET, fluorescent resonance energy transfer; GPCR, -coupled receptor; HA, hemagglutinin; HEK, human embryonic kidney; IP3, inositol 1,4,5-trisphosphate; ISO, isoproterenol; PCR, polymerase chain reaction; PEI, polyethylenimine.

533 534 Stallaert et al.

In addition to the abovementioned signaling pathways, the [imino-3,1-phenylenecarbonylimino(4-fluoro-3,1-phenylene)carbonylimino]] 21 9 b2AR can also stimulate an increase in intracellular Ca . bis-1,3,5-naphthalenetrisulfonic acid hexasodium salt], NF340 [4,4 - This signaling response is well established in excitable cells, (Carbonylbis(imino-3,1-(4-methyl-phenylene)carbonylimino))bis 9 such as cardiomyocytes, via a mechanism involving cAMP- (naphthalene-2,6-disulfonic acid) tetrasodium salt], NF279 [8,8 - mediated regulation of plasma membrane L-type Ca21 chan- [Carbonylbis(imino-4,1-phenylenecarbonylimino-4,1-phenylenecarbonylimino)] bis-1,3,5-naphthalenetrisulfonic acid hexasodium salt], A-804598 nels (Zhang et al., 2001; Christ et al., 2009; Benitah et al., 99 9 21 [N-Cyano-N -[(1S)-1-phenylethyl]-N -5-quinolinyl-guanidine], 5-BDBD 2010). However, increases in intracellular Ca levels have [5-(3-Bromophenyl)-1,3-dihydro-2H-benzofuro[3,2-e]-1,4-diazepin- also been observed in nonexcitable cells, such as alveolar 2-one], CGS 15943 [9-Chloro-2-(2-furanyl)-[1,2,4]triazolo[1,5-c] epithelial cells (Keller et al., 2014), and in human embryonic quinazolin-5-amine] and NF449 [4,49,499,4999-[Carbonylbis(imino- kidney (HEK)293 cells endogenously or stably overexpressing 5,1,3-benzenetriyl-bis(carbonylimino))]tetrakis-1,3-benzenedisulfonic the b2AR (Schmidt et al., 2001; Stallaert et al., 2012; van der acid octasodium salt] were obtained from R&D Systems (Bio-Techne, Westhuizenetal.,2014).Previous observations from our labora- Minneapolis, MN). Coelenterazine 400a and coelenterazine h were purchased from Nanolight Technologies (Pinetop, AZ), and coelenter- tory indicated that such b2AR-mediated increases in intra- cellular Ca21 in HEK293 cells are inhibited by an inositol azine cp was purchased from Biotium (Hayward, CA). Arginine-vasopressin and A23187 [5-(Methylamino)-2-[[2R,3R,6S,8S,9R,11R)-3,9,11-trimethyl-8- 1,4,5-trisphosphate (IP3) receptor antagonist, 2-APB (Stallaert 1 [(1S)-1-methyl-2-oxo-2-(1H-pyrrol-2-yl)-ethyl]-1,7-dioxaspiro[5.5]undec-2- et al., 2012), suggesting the release of Ca2 from intracellular

yl]methyl]-4-benzoxazolecarboxylic acid] were from Tocris (Bio-Techne). Downloaded from stores, which points to a mechanism distinct from the plasma 21 Cell culture reagents were from Wisent (Montreal, QC, Canada). The Gq membrane Ca channel activation described in excitable cells. selective inhibitor, FR900359 (Schrage et al., 2015), was obtained from E. Several studies have reported that activation of the b2AR Kostenis and S. Kehraus from the University of Bonn (Bonn, Germany). can also stimulate the release of extracellular mediators as part The Gq and Gq(Q209L/D277N) constructs were purchased from Missouri of its signaling repertoire. Activation of the b2AR in mouse S&T cDNA Resource Center (Rolla, MO). HapII, PvuII, and HaeII skeletal muscle, vascular smooth muscle cells, or cardiac restriction enzymes were from Takara Bio (Kusatsu, Shiga, Japan). All fibroblasts promotes the extracellular release of cAMP, which other reagents were obtained from Sigma-Aldrich, unless otherwise stated. molpharm.aspetjournals.org is converted to AMP by ecto-phosphodiesterases and then to Cell Culture and Transfections. Parental HEK293S cells or adenosine by ecto-59-nucleotidase on the extracellular surface HEK293S cells stably expressing an amino-terminal tagged human b AR [3.17 6 0.32 pmol/mg protein; hemagglutinin (HA)-b AR- of the cells (Dubey et al., 1996, 2001; Duarte et al., 2012). In 2 2 HEK293S cells] (Galandrin and Bouvier, 2006) were grown at 37°C addition, extracellular adenosine can be produced by the with 5% CO2 in Dulbecco’s modified Eagle’s medium supplemented with release of intracellular adenine nucleotides to the extracellular 5% fetal bovine serum (FBS). HEK293S and HA-b2AR-HEK293S cells space, which are converted to adenosine by ecto-ATPase, were transiently transfected (600,000 cells/well) in six-well plates with ecto-ADPase, and ecto-59-nucleotidase (Jackson et al., 1996). biosensors (950 ng/1 Â 106 cells) for Ca21 measurements (Obelin lumines- Stimulation of human erythrocytes with isoproterenol (ISO) cence), bioluminescence resonance energy transfer (BRET), and fluores- increases extracellular ATP levels (Montalbetti et al., 2011), cence resonance energy transfer (FRET) assays using linear at ASPET Journals on September 25, 2021 suggesting that bARs can also be coupled to ATP release. In polyethylenimine (PEI; 1 mg/ml; Polysciences, Warrington, PA) diluted in some cases, this release of mediators into the extracellular NaCl (150 mM, pH 7.0) (PEI:DNA ratio 3:1), as previously described (Reed milieu can lead to the subsequent transactivation of other G et al., 2006). Twenty-four hours post-transfection, cells were replated protein-coupled receptors (GPCRs) or receptor tyrosine kinases. (50,000 cells/well) onto white 96-well CulturePlates (Perkin-Elmer, Wood- bridge, ON, Canada). Six hours after replating, growth medium was Previously, an inside-out signaling mechanism was demon- replaced with fresh starvation media consisting of Dulbecco’s modified strated for the b2AR, leading to the transactivation of adenosine Eagle’s medium supplemented with 0.5% FBS for the next 18 hours. receptors through the production of extracellular adenosine, as Parental HEK293 and ΔGs-HEK293 cells were cultured and trans- described above (Sumi et al., 2010), as well as the epithelial fected, as described for the HEK293S cells, with the following modifi- growth factor receptor through the membrane shedding of cation: medium was supplemented with 10% FBS, and PEI was diluted heparin-binding epithelial growth factor (Kim et al., 2002). in Dulbecco’s phosphate-buffered saline and added to cells in suspension In this study, we describe a new inside-out pathway activated (300,000–350,000 cells/ml). Cells were then transferred into white 96-well CulturePlates (30,000–35,000 cells/well). Cells were cultured by the b2AR in nonexcitable cells, involving the Gs-dependent, but cAMP-independent increase in intracellular Ca21 through with 10% FBS for the following 48 hours prior to experimentation. Δ the release of extracellular ATP and the subsequent trans- Generation of HEK293 Cells Devoid of Gs ( Gs-HEK293). The two members (Gas and Gaolf, encoded by the GNAS and the activation of Gq-coupled P2Y purinergic receptors, providing a GNAL , respectively; both were expressed in HEK293 cells) of previously unrecognized link between adrenergic and puriner- the Gas family were simultaneously targeted by a CRISPR/Cas9 gic receptors. This study also provides the first description and system to introduce frame shift into the coding sequence, as described use of clustered regularly interspaced short palindromic repeat- previously ( et al., 2013), with some modifications. Briefly, GNAS- associated protein-9 nuclease (CRISPR/Cas9) genomic editing targeting sgRNA sequence (59-CTACAACATGGTCATCCGGG-39) to knockout Gs in human cells, providing a powerful new tool to was inserted into the BbsI site of the pSpCas9(BB)-2A-GFP (PX458) explore the signaling events downstream of this Ga subunit. vector (a gift from Feng Zhang, Broad Institute, Cambridge, MA; plasmid # 48138, Addgene, Cambridge, MA) using two synthesized oligonucleotides (59-CACCGCTACAACATGGTCATCCGGG-39 and 9 9 Materials and Methods 5 -AAACCCCGGATGACCATGTTGTAGC-3 ; FASMAC, Atsugi- shi, Kanagawa, Japan). Similarly, GNAL-targeting sgRNA sequences 1 Reagents. ISO, carbamoylcholine chloride (Cch), thapsigargin, (59-TGTTTGATGTTGGTGGCCAG-39), 2 (59-GTAATGTTTGCCGTCA- 8-bromo-cAMP,forskolin,choleratoxin(CTX),suramin,andapyrase CCGG-39), and 3 (59-ATTGTGCACAGTCAATCAGC-39) were inserted using were purchased from Sigma-Aldrich (St. Louis, MO). ICI118,551, BAPTA- three sets of oligonucleotides (for 1, 59-CACCGTGTTTGATGTTGGTGGC- AM [1,2-Bis(2-aminophenoxy)ethane-N,N,N9,N9-tetraacetic acid tetrakis CAG-39 and 59-AAACCTGGCCACCAACATCAAACAC-39;for2, (acetoxymethyl ester)], 2-APB [2-aminoethoxydiphenyl borate], SQ22536 59-CACCGTAATGTTTGCCGTCACCGG-39 and 59-AAACCCGGTGACGG- [9-(Tetrahydro-2-furanyl)-9H-purin-6-amine], NF157 [8,89-[Carbonylbis CAAACATTAC-39; for 3, 59-CACCGATTGTGCACAGTCAATCAGC-39 b2AR Transactivation of Purinergic Receptors 535 and 59-AAACGCTGATTGACTGTGCACAATC-39). Correct insertion blots were performed according to standard procedures using an anti- of the sgRNA sequences was verified by sequencing using the Sanger Gas/olf mouse monoclonal antibody (clone E-7, cat. no. sc55546; Santa method (FASMAC). HEK293 (200,000 cells/well) were seeded into Cruz Biotechnology, Santa Cruz, CA) or anti–a-tubulin mouse mono- 12-well plates and transfected with a mixture of the GNAS-targeting clonal antibody (clone DM1A, cat. no. sc-32293; Santa Cruz Bio- PX458 vectors and either of the GNAL-targeting PX458 vectors (0.25 mg technology) (both at 1 mg/mL in Tris-buffered saline, 1% bovine serum each/well) with Lipofectamine 2000 (Life Technologies, Carlsbad, albumin, and 0.05% Tween 20), followed by horseradish peroxidase– CA). Twenty-four hours later, cells were detached and GFP-positive linked anti-mouse IgG secondary antibody (cat. no. NA9310, GE cells (approximately 30% of cells) were isolated using a cell sorter Healthcare (Buckinghamshire, England); 1:2000 diluted with Tris- (SH800; Sony, Minato-ku, Tokyo, Japan). After growing clonal cell buffered saline, 5% skim milk, 0.05% Tween 20). Chemiluminescence colonies, clones were analyzed for mutations in the GNAS and the signals were detected using a LAS-4000 (FujiFilm, GE Healthcare) and GNAL genes by polymerase chain reaction (PCR) and restriction visualized with Multi Gauge version 3.0 (FujiFilm). enzyme digestion, using the primer and restriction enzyme combina- Ca21 Measurements. Ca21 measurements were carried out, as tions shown in Supplemental Table 1. PCR was performed with an described previously (van der Westhuizen et al., 2014). Briefly, initial denaturation cycle of 95°C for 2 minutes, followed by 35 cycles of HEK293S, HA-b2AR-HEK293S, or ΔGs-HEK293 cells were tran- 95°C for 15 seconds, 64°C for 30 seconds, and 72°C for 30 seconds. The siently transfected with the mCherry-obelin– or GFP2-obelin–based resulting PCR product was digested with the corresponding restric- Ca21-sensitive biosensors. The day of experimentation, cells were tion enzyme and analyzed by agarose gel electrophoresis. Candidate washed twice and preincubated in stimulation buffer [modified Hanks’

positive clones were further analyzed by genomic DNA sequencing balanced salt solution: 137 mM NaCl, 5.4 mM KCl, 0.25 mM Downloaded from using a TA cloning method. The lack of functional Gs was also Na2HPO4, 0.44 mM KH2PO4, 1.8 mM CaCl2, 0.8 mM MgSO4, confirmed by assessing GPCR-stimulated cAMP production, as de- 4.2 mM NaHCO3, 0.2% (w/v) D-glucose, pH 7.4] containing the obelin scribed below. For the TA cloning, PCR-amplified genomic DNA substrate, coelenterazine cp (1 mM), at 25°C for 2 hours in the dark. fragments using an ExTaq polymerase (Takara Bio) were gel- Compounds diluted in stimulation buffer were injected into the wells, purified (Promega, Madison, WI) and cloned into a pMD20 T-vector and luminescence readings were taken every 0.3 seconds using the (Takara Bio). Ligated products were introduced into SCS1-competent SpectraMax L (Molecular Devices, Sunnyvale, CA). Luminescence cells (Stratagene, Agilent Technologies, Santa Clara, CA), and trans- was monitored for a total of 60 seconds poststimulation and expressed molpharm.aspetjournals.org formed cells were selected on an ampicillin-containing LB plate. At least as relative luminescence units. The area under the resulting curves 12 colonies were picked, and inserted fragments were PCR-amplified using and the peak response to stimulation were calculated. Where in- the ExTaq polymerase and primers [59-CAGGAAACAGCTATGAC-39 (M13 dicated, the area under the curves was transformed as percentage of primer RV) and 59-GTTTTCCCAGTCACGAC-39 (M13 primer M4)] peak ISO or A23187 response. designed to anneal the pME20 T-vectors. PCR products of the trans- cAMP Production. cAMP production measurements were car- formed pMD20 T-vector were sequenced using the Sanger method ried out, as described previously (van der Westhuizen et al., 2014). (FASMAC) and the M13 primer RV. Briefly, cells were transfected as indicated with either the FRET-

Whole-transcriptome expression profiling was performed using the based GFP10-EPAC1-vYFP or the BRET-based GFP10-EPAC1-RlucII Clariom S assay (human, analysis using the Transcriptome Analysis cAMP-sensitive biosensors in which binding of cAMP induces a

Console from Affymetrix, a part of Thermo Fisher Scientific, Waltham, conformational change that leads to a decrease in FRET or BRET. at ASPET Journals on September 25, 2021 MA) to assess differences in expression between the parental Compounds were diluted in stimulation buffer and added to cells at HEK293 and ΔGs-HEK293 cell lines (Supplemental Fig. 5; Supple- 37°C for 30 minutes prior to measurement. The light emitted from mental Tables 2 and 3). FRET donor and acceptor was measured with the FlexStation 6 mRNA Expression. Total RNA was isolated from 3 Â 10 II (Molecular Devices), following GFP10 excitation with a laser at 400 nm, HA-b2AR-HEK293S cells using the RNAeasy mini kit (Qiagen, using the 510 nm (GFP10) and 533 nm (vYFP) emission filters. BRET Hilden, Germany), according to the manufacturer’s protocol for animal was measured with the Mithras LB 940 (Berthold Technologies, Bad cells. Microarray analysis of mRNA samples was performed using Wildbad, Germany) using 410 6 70 nm (RlucII) and 515 6 20 nm

Illumina’s HumanRef-8 v3.0 expression bead chips at McGill Univer- (GFP10) emission filters or the SynergyNeo microplate reader from sity and Génome Québec Innovation Centre (Montreal, QC, Canada). BioTek using 410 6 80 nm and 515 6 30 nm emission filters. FRET The gene expression microarray data were analyzed and normalized or BRET ratios were calculated (acceptor/donor), and ΔFRET and by the bioinformatics platform at the Institute for Research in ΔBRET were determined by subtracting the FRET or BRET signal Immunology and Cancer at the University of Montreal. The raw generated in vehicle treatment control conditions. fluorescent values were then normalized using a quantile normaliza- In one instance, cAMP measurements were performed using the tion method (Bolstad et al., 2003; Lemieux, 2006), which re-ranks the homogeneous time resolved fluorescence (HTRF)-cAMP dynamic kit sampled based on the raw fluorescent values obtained for each probe. (Cisbio, Bedford, MA), as described previously (Stallaert et al., 2012).

Using this method of normalization, the distribution of the samples Gs Activation Assay. HA-b2AR-HEK293S cells were cotrans- becomes the reference point; therefore, no internal control is required. fected with a three-component G protein activation biosensor com-

For the microarray dataset analyzed, the distribution of the data is posed of Gas-67-RlucII (50 ng/well), Gb1 (100 ng/well), and GFP10-Gg1 defined by the median value 7.162, the minimum value 5.851, the (100 ng/well) in which Gs activation results in the separation between maximum value 15.760, the first quartile of 6.578, and the third the Ga and Gbg subunits, thus leading to a decrease in BRET (Galés quartile of 9.215. The normalized values were plotted alongside known et al., 2005). The day of the experiment, cells were washed twice with housekeeping genes, b-actin and a-tubulin, which were used for stimulation buffer and pretreated with various inhibitors diluted in comparison purposes to represent genes that are highly expressed in stimulation buffer for the indicated times at 37°C. Coelenterazine the cells. The inhibitory G protein, Gao, was also plotted because it is 400a, diluted in stimulation buffer (5 mM), was added to the wells for not expressed in HEK293 cells (Law et al., 1993) and therefore 5 minutes, and then increasing concentrations of ISO were added for represents an appropriate baseline background value. 2 minutes. Plates were read on the Mithras LB 940 with 410 6 70 nm

Western Blot Analysis. The parental HEK293 cells and the ΔGs- (RlucII) and 515 6 20 nm (GFP10) emission filters, and BRET ratios HEK293 cells were harvested, and approximately 1 Â 106 cells were were calculated as above. lysed in 500 mL SDS-PAGE sample buffer [62.5 mM Tris-HCl (pH 6.8), Quantification of Extracellular ATP. Extracellular ATP was 50 mM dithiothreitol, 2% SDS, 10% glycerol, and 4 M urea] containing measured using the luciferin-luciferase–based ENLITEN ATP Assay 1 mM EDTA and 1 mM phenylmethylsulfonyl fluoride. Cell lysates System (Promega), according to the manufacturer’s instructions. were homogenized with a handy ultrasonic homogenizer (Microtec, Briefly, cells were grown to 90–100% confluence in a 96-well plate. Funabashi, Chiba, Japan) and boiled at 95°C for 5 minutes. Western Cells were washed with stimulation buffer and then pretreated with 536 Stallaert et al.

10 mM ICI 118,551 or stimulation buffer (vehicle) for 30 minutes at Stable overexpression of the b2AR in HEK293S cells (HA- 37°C. Cells were stimulated with 10 mM ISO, and the cell supernatant b2AR-HEK293S cells) led to acceleration in the onset (2 (conditioned buffer) was collected at 15–120 seconds poststimulation. seconds) and increased magnitude of the response (Fig. 1B). – 1 Luciferin luciferase reagent (100 ml) was injected into 10 ml condi- This increase in intracellular Ca2 was inhibited in both cell tioned buffer, and luminescence was measured on the Mithras LB 940. lines following pretreatment with the b AR-selective antago- ATP concentration of each sample was determined by comparing the 2 nist, ICI 118,551, confirming the b AR specificity of the luminescence of samples with those of standards in the concentration 2 range of 1026 to 10210 M. response (Fig. 1C). The response was concentration- Data Analysis. Data analysis was performed using Microsoft dependent, yielding similar pEC50 values for ISO in both cell Excel (Microsoft, Redmond, WA) and GraphPad Prism (versions lines (7.32 6 0.10 and 7.05 6 0.16 in HEK293S and HA-b2AR- 5 and 6; GraphPad Software, La Jolla, CA). Statistical analyses were HEK293S cells, respectively) (Fig. 1D). performed as indicated in figure legends. b2AR Promotes a Gs-Dependent but cAMP- Independent Ca21 Response. To probe the mechanism 21 Results underlying the b2AR-mediated Ca response, we first exam- ined the Gs/cAMP pathway in the HA-b2AR-HEK293S cell 21 b2AR Activation Leads to an Increase in Intracellu- line, which produces a larger Ca response following stimu- lar Ca21 in HEK293S Cells. ISO stimulation of the endog- lation, thus facilitating its mechanistic dissection. To confirm Downloaded from enously expressed b2ARs in HEK293S cells induces a rapid the involvement of Gs, the primary G protein–coupling 21 and transient increase in intracellular Ca (Fig. 1A). The partner of the b2AR, we tested the effect of overnight pre- increase in intracellular Ca21 was detected as early as treatment with CTX to downregulate Gs signaling (Levis and 5 seconds following the addition of ISO, reaching a maximum Bourne, 1992), as well as the effect of acute pretreatment with at approximately 12 seconds, followed by a slow decrease a small molecule inhibitor of Gs, NF449 (which prevents thereafter and returning to basal levels 60 seconds post- GDP/GTP exchange) (Hohenegger et al., 1998). We found that stimulation, demonstrating a functional Ca21 response in the ISO-stimulated Ca21 response was significantly inhibited molpharm.aspetjournals.org nonexcitable cells that endogenously express the receptor. by both CTX and NF449 pretreatment (Fig. 2A), suggesting at ASPET Journals on September 25, 2021

2+ Fig. 1. ISO increases intracellular Ca via the b2AR. HEK293S (endogenously expressing b2AR at 18.74 6 3.14 fmol/mg protein) or HA-b2AR-HEK293S (stably expressing b2AR at 3174 6 320 fmol/mg protein) cells were transiently transfected with mCherry-obelin. (A and B) Cells were treated with vehicle or ISO (10 mM), and increases in intracellular Ca2+ [increases in obelin relative luminescence unit (RLU)] were measured. (C) Cells were pretreated with vehicle or ICI 118,551 (ICI, 100 nM) for 1 hour and then treated with ISO (10 mM). Data represent ISO-promoted peak Ca2+ responses. (D) The Ca2+ response was measured with increasing concentrations of ISO. EC50 values: 7.32 6 0.10 for HEK293S and 7.05 6 0.16 for HA-b2AR-HEK293S. Data are expressed as the mean 6 S.E.M. of three to four independent experiments, each performed in triplicate. Data in column graphs were analyzed by two- tailed paired Student’s t test, where P , 0.05 (*) was considered significant. b2AR Transactivation of Purinergic Receptors 537 Downloaded from molpharm.aspetjournals.org at ASPET Journals on September 25, 2021

2+ Fig. 2. The ISO-promoted increase in intracellular Ca is Gs-dependent. (A) HA-b2AR-HEK293S cells transiently transfected with mCherry-obelin were pretreated with CTX (200 ng/ml, 18 hours) or with NF449 (10 mM, 1 hour). Cells were treated with ISO (10 mM), and increases in intracellular Ca2+ were measured. Inset: The area under the curve (AUC) was calculated for each treatment. (B) HA-b2AR-HEK293S cells were transiently transfected with Gas-67-RlucII and GFP10-Gg1. Cells were pretreated with vehicle, CTX, or NF449, as indicated above, followed by ISO treatment, and Gs activation was measured as a decrease in the BRET response. (C) Protein extracts from parental HEK293 and three clones (CL1, CL2 and CL3) of ΔGs-HEK293 cells were separated on SDS-PAGE and immunoblotted with anti-Gas/olf antibody (top panel) or anti–a-tubulin antibody (bottom panel). Note that the parental HEK293 cells express the long (GasL) and short (GasS) Gas isoforms, whereas they were both undetectable in the mutant clones. (D) The parental HEK293 or DGs-HEK293 (clone 1) cells were transiently transfected with GFP10-EPAC-RlucII. Cells were then treated with either ISO (10 mM) or forskolin (Fsk; 10 mM) for 20 minutes. ISO increased intracellular cAMP levels in HEK293 cells at a similar level to Fsk, whereas the response to ISO was significantly blocked in DGs-HEK293 cells. (E and F) The parental HEK293 or DGs-HEK293 (clone 1) cells were transiently transfected with b2AR and Obelin-GFP2, and treated with ISO (10 mM) (E) or the Ca2+ ionophore A23187 (10 mM) (F). ISO increased intracellular Ca2+ in the parental HEK293 cells, whereas the response was blocked in ΔGs-HEK293 cells (E). The response to the Ca2+ ionophore A23187 was comparable in both cell lines (F). cAMP and Ca2+ data for the two other clones are shown in Supplemental Fig. 4. Data are expressed as the mean 6 S.E.M. of three to six independent experiments, each performed in duplicate or triplicate. Data in (A) were analyzed by one-way analysis of variance, with a Dunnett’s multiple comparison post hoc test, where P , 0.05 (*) was considered significant. Data in (D) were analyzed by two-way analysis of variance, followed by a Sidak’s multiple comparison post hoc test, where P , 0.05 (*) and P , 0.001 (***) were considered significant.

21 the involvement of Gs in the b2AR-mediated Ca response. and Gg1 as a reporter for G protein activation (Galés et al., However, although chronic pretreatment with CTX led to the 2005), acute pretreatment with NF449 had no impact on Gs expected inhibition of Gs activation measured using a BRET- activation (Fig. 2B), indicating that its effect on the b2AR- based biosensor that monitors the separation between Gas promoted Ca21 response resulted from another mechanism of 538 Stallaert et al. action. This is consistent with the fact that, although a potent in all three DGs-HEK293 selected clones (Supplemental Fig. 4B). and selective inhibitor of Gas, NF449 contains multiple Furthermore, the ISO-promoted increase in intracellular Ca21 negative charges (El-Ajouz et al., 2012) that impede its cell was also found to be inhibited in DGs-HEK293 cells (Fig. 2E; permeability and most likely restrict its effect to extracellular Supplemental Fig. 4C), whereas the response to the Ca21 target(s) in cell-based assays. Observations that NF449 is also ionophore, A23187, was similar to that observed in parental a potent antagonist of purinergic P2X and P2Y receptors HEK293 cells (Fig. 2F; Supplemental Fig. 4C), thus confirming 21 (Braun et al., 2001; Kassack et al., 2004) provided valuable theessentialroleofGsintheb2AR-promoted Ca response. 21 insights into additional mechanistic components underlying The loss of Ca responses to b2AR stimulation was observed in 21 the b2AR-stimulated Ca response, as discussed below. all three DGs-HEK293 clones (Supplemental Fig. 4C), excluding However, the inhibitory effect of sustained CTX treatment a possible clonal effect. Furthermore, a gene array analysis supported a role for Gs in the Ca21 response. (Supplemental Fig. 5) of the parental and one of the DGs- To confirm the role of Gs in the cytoplasmic Ca21 increase, HEK293 cells (clone 1) revealed that none of the other G protein we genetically deleted functional Gs from HEK293 cells using subunits (a, b,org) were among the 489 genes (of 21,448 genes the CRISPR/Cas9 system (DGs-HEK293). The genetic char- detected) found to be significantly up- or downregulated with a acterization of three ΔGs cell clones is presented in Supple- minimum threshold of twofold in the ΔGs cells (Supplemental mental Figs. 1–4. Loss of both the long (GasL) and the short Table 2), excluding the possible implication of other G protein (GasS) forms of Gas was confirmed in three DGs-HEK293 expression changes in the observed responses. The expression Downloaded from clones (Fig. 2C). levels of the G protein subunits detected in the microarray for ΔGs cells were then tested for a loss of function using the parental and ΔGs cells are presented in Supplemental Table 3. BRET-based EPAC biosensor to monitor the effect of Gs We next assessed the role of the Gs- and AC-dependent 21 knockout on the b2AR-mediated cAMP response. In DGs- second-messenger cAMP in the ISO-promoted Ca response. HEK293 cells, the cAMP response to ISO stimulation was Pretreatment with SQ22536, a pharmacological inhibitor of significantly reduced compared with parental HEK293 cells AC, significantly inhibited ISO-promoted cAMP production molpharm.aspetjournals.org (Fig. 2D; Supplemental Fig. 4A). Similarly, stimulation of (by 43%; Fig. 3A), yet it failed to block the Ca21 response (Fig. another Gs-coupled receptor, the vasopressin type 2 receptor, 3B), producing a response higher than in the absence of the with arginine-vasopressin failed to promote cAMP accumulation inhibitor. To further confirm this surprising result, we directly at ASPET Journals on September 25, 2021

2+ Fig. 3. The ISO-promoted increase in Ca is independent of cAMP production. (A) HA-b2AR-HEK293S cells were pretreated with vehicle or the adenylyl cyclase inhibitor, SQ22536 (100 mM, 1 hour), and then treated with ISO (10 mM) and cAMP measured using the Cisbio kit. (B–E) HA-b2AR-HEK293S 2+ cells were transiently transfected with either GFP10-EPAC-vYFP (C) or mCherry-obelin (B, D, and E). (B) Cells were treated as in (A), and Ca response was measured. Inset: Ca2+ response was plotted using the area under the curve (AUC). SQ22536 significantly inhibited ISO-promoted cAMP production, but did not attenuate the ISO-induced Ca2+ response. (C and D) Cells were treated with either forskolin (Fsk; 100 mM, 30 minutes) or ISO and cAMP (C) and Ca2+ (D) responses were measured. Fsk activates cAMP production to a similar extent as ISO (C), but stimulates only a marginal Ca2+ response compared with ISO (D) and over vehicle (D, inset). (E) Cells were stimulated or not (veh) with ISO or the nonhydrolyzable cAMP analog, 8-bromo-cAMP (8-br, 100 mM), and Ca2+ response was measured. Inset: Ca2+ response was plotted using AUC. Data are expressed as the mean 6 S.E.M. of three to six independent experiments, each performed in duplicate or triplicate. Column graphs were analyzed by two-tailed paired Student’s t tests, where P , 0.05 (*) was considered significant. b2AR Transactivation of Purinergic Receptors 539 assessed whether increases in intracellular cAMP alone were acting in a paracrine manner. As a control experiment, we capable of generating a Ca21 response. Stimulation with forskolin, examined the Ca21 response to Cch stimulation in both coculture which increases cAMP independently of b2AR activation through conditions and found no significant difference (Fig. 5B), further the direct activation of AC (Seamon et al., 1981), increased the supporting the b2AR specificity of this paracrine signaling intracellular cAMP to a similar level as obtained for ISO (Figs. 2D response. and 3C), yet elicited only a marginal Ca21 response (Fig. 3D). Because both ATP and ADP adenine nucleotides act as Further supporting these results, the cell-permeable cAMP agonists for the P2Y11 receptor (van der Weyden et al., 2000), analog, 8-bromo-cAMP, elicited a similarly weak Ca21 response we assessed the role of extracellular adenine nucleotides in 21 (Fig. 3E). Taken together, our data indicate the b2AR-promoted the b2AR-promoted Ca response, by pretreating cells with increase in intracellular Ca21occurs predominantly independently apyrase, a cell-impermeable enzyme that hydrolyzes both of cAMP production. ATP and ADP to AMP. Apyrase treatment decreased both the Involvement of Purinergic Receptors in the b2AR- efficacy (34% decrease in Emax) and potency (increase in 21 28 26 Promoted Ca Response. To further investigate the EC50 from 9.2 Â 10 Mto1.7Â 10 M) of the ISO- above observation that NF449, a small molecule inhibitor of promoted response (Fig. 5C), consistent with the depletion Gs and a purinergic receptor antagonist (Braun et al., 2001; of an extracellular mediator released upon b2AR activation Kassack et al., 2004), blocked the ISO-promoted Ca21 re- and acting in trans on the P2Y11 receptor. The Cch- sponse (see Fig. 2A), we assessed additional purinergic re- mediated Ca21 response, in contrast, was not inhibited Downloaded from ceptor antagonists to test the potential role of this receptor and instead was increased following apyrase treatment family. Pretreatment with a pan-purinergic receptor antago- (Fig. 5D). Consistent with adenine nucleotides being the nist, suramin, significantly inhibited the ISO-promoted paracrine mediator responsible for the transactivation, ISO increase in intracellular Ca21 (Fig. 4A), whereas the pan- treatment promoted the release of ATP from HA-b2AR- adenosine receptor antagonist CGS 15943 had no effect (Fig. HEK293S cells (Fig. 6A), with kinetics similar to those of

21 molpharm.aspetjournals.org 4B). Gene expression microarray analysis of HA-b2AR- the ISO-promoted Ca response (reaching a maximum at HEK293S cells indicated the expression of several P2X 15 seconds and returning to basal levels by 60 seconds). This purinergic ion channels and P2Y purinergic GPCRs in this cell ATP release was blocked by the b2AR-selective antagonist, line (Supplemental Fig. 6). To determine the identity of the ICI 118,551, supporting a role for b2AR in the response (Fig. purinergic receptor(s) involved in the Ca21 response, we selec- 6B). These results are consistent with the hypothesis that tively inhibited the most highly expressed subtypes. NF279 b2AR activation leads to the release of adenine nucleotides, (P2X1-selective), 5-BDBD (P2X4-selective), and A804598 (P2X7- which can act in trans on P2Y11 receptors to activate a Ca21 selective) were without effect, whereas NF157 (P2Y11/P2X1- response. selective) and NF340 (P2Y11-selective) significantly inhibited P2Y11 Transactivation Elicits a Gq-Dependent In- 21 21 the ISO-promoted Ca response (Fig. 4C), suggesting a role crease in Ca from Intracellular Stores. To investigate at ASPET Journals on September 25, 2021 for the P2Y11 purinergic receptor. The effect of NF449 the potential role of Gq activation following P2Y11 trans- (Fig. 2A), NF340, and NF157 (Fig. 4C) were specific to the activation (Qi et al., 2001), we tested the effects of over- b2AR-stimulated response because these compounds did not expressing either wild-type Gaq or a dominant-negative significantly inhibit the Cch-stimulated increase in Ca21 via mutant form of Gq (Gq-Q209L/D277N). Overexpression of endogenously expressed Gq-coupled muscarinic receptors in wild-type Gq in HA-b2AR-HEK293S cells increased the 21 HA-b2AR-HEK293S cells (Fig. 4D). The effect of the P2Y11- magnitude of the ISO-promoted Ca response, whereas selective antagonist, NF340, was concentration-dependent and overexpression of Gq-Q209L/D277N or pretreatment with 21 resulted in a decrease in the Emax of the ISO-promoted Ca the Gq-selective inhibitor FR900359 (Schrage et al., 2015) response, with no significant effect on the EC50 (Fig.4E),confirming completely abolished the response (Fig. 7). These results that its effect does not occur directly through a competitive blockade further support the involvement of a Gq-coupled receptor 21 of the b2AR, but instead suggesting a noncompetitive antagonism of downstream of the b2AR in the Ca response. the response. Furthermore, both NF340 and NF157 were found to We next assessed whether the increase in intracellular Ca21 progressively decrease the ISO-promoted Ca21 response in a dose- occurs via release from intracellular stores, as is typical dependent manner, with potencies compatible with their affinity for for Gq-mediated responses. For this purpose, cells were the purinergic receptors (Fig. 4F). pretreated with the IP3 receptor antagonist, 2-APB; the cell- 21 A previous study demonstrated that b2AR activation could permeable Ca chelator, BAPTA-AM; or the sarco/endoplasmic induce the release of adenine nucleotides from HEK293 cells reticulum Ca21-ATPase inhibitor, thapsigargin. Upon stimula- 21 (Sumi et al., 2010). To determine whether the b2AR promotes a tion with ISO, the Ca response was completely blocked by Ca21 response through the release of an extracellular mediator, all three inhibitors (Fig. 8A). These inhibitors had a similar leading to the subsequent transactivation of a purinergic receptor, effect on the Cch-mediated Ca21 response, which activates a we performed coculture experiments in which parental HEK293S well-characterized IP3-dependent pathway via activation of cells transfected with the obelin Ca21 biosensor (obelin-HEK293S Gq-coupled muscarinic receptors (Caulfield, 1993) (Fig. 8B). responder cells) were cocultured at a 1:1 ratio with either parental Together these data are consistent with the proposed trans- HEK293S cells or HA-b2AR-HEK293S cells not expressing the activation of the Gq-coupled P2Y11 receptor upon activation biosensor (releaser cells). When obelin-HEK293S responders were of the b2AR. cocultured with releaser cells overexpressing the b2AR (HA- 21 b2AR-HEK293S), a significant potentiation in the Ca response Discussion was observed following ISO treatment compared with coculture with parental HEK293S releaser cells (Fig. 5A), suggesting a role In this study, we describe a new node in the b2AR signaling for the b2AR-dependent release of an extracellular mediator network involving a Gs-dependent but cAMP-independent 540 Stallaert et al. Downloaded from molpharm.aspetjournals.org at ASPET Journals on September 25, 2021

2+ Fig. 4. The ISO-promoted Ca response is inhibited upon purinergic receptor blockade. HA-b2AR-HEK293S cells were transiently transfected with mCherry-obelin. (A–D) Cells were pretreated with the pan-adenosine receptor antagonist CGS 15943 (CGS, 1 mM, 30 minutes) (B) or the purinergic receptor antagonists: suramin (500 mM, 1 hour) (A and C), NF279 (1 mM, 30 minutes) (C), 5-BDBD (10 mM, 30 minutes) (C), A804598 (1 mM, 30 minutes) (C), NF449 (10 mM, 30 minutes) (D), NF157 (10 mM, 30 minutes) (C and D), or NF340 (10 mM, 30 minutes) (C and D). Cells were then stimulated with either ISO (10 mM) (A–C) or Cch (100 mM) (D), and the Ca2+ response was measured. Inset in (A) and (B) shows the Ca2+ response plotted using the area under the curve (AUC). (E) Cells were pretreated with vehicle or the indicated concentration of NF340, followed by stimulation with increasing concentrations of ISO, and the Ca2+ response was measured. (F) Cells were pretreated with increasing concentration of either NF157 or NF340, followed by stimulation with ISO at an EC80 2+ concentration (150 nM). Ca response was then measured. IC50 values: 4.36 6 1.91 mM for NF157 and 50.4 6 27.5 mM for NF340. Data are expressed as the mean 6 S.E.M. of three to four independent experiments, each performed in triplicate. AUC data presented in column graphs were analyzed by two-tailed unpaired Student’s t tests, where P , 0.05 (*) was considered significant. release of ATP, which transactivates P2Y11 purinergic recep- production and/or release of an extracellular mediator acting tors and subsequently initiates a Gq-dependent mobilization on a distal receptor, represents a bona fide signaling event for of intracellular Ca21. This discovery is consistent with the many GPCRs (Ostrom et al., 2000; Gschwind et al., 2001; Lee accumulating reports that receptor transactivation, via the et al., 2002; Sumi et al., 2010). Given that the b2AR and the b2AR Transactivation of Purinergic Receptors 541 Downloaded from molpharm.aspetjournals.org

Fig. 5. ISO stimulates the releases of an extracellular mediator involved in the Ca2+ response. Parental HEK293S cells were transiently transfected with mCherry-obelin (responder cells) and cocultured with HEK293S (WT) or HA-b2AR-HEK293S (b2AR) cells not expressing the biosensor (releaser cells). (A and B) Cells were treated with either ISO (10 mM) (A) or Cch (100 mM) (B), and the Ca2+ response was measured. (C and D) Cells were pretreated with apyrase (1 UI/ml, 1 hour), followed by stimulation with increasing concentration of ISO (C) or Cch (100 mM) (D), and the Ca2+ response was measured. Inset for A, B, and D: Ca2+ response plotted using the area under the curve (AUC). Data are expressed as the mean 6 S.E.M. of three independent experiments, each at ASPET Journals on September 25, 2021 performed in triplicate. AUC data presented in column graphs were analyzed by two-tailed paired t tests, where P , 0.05 (*) was considered significant.

P2Y receptors are coexpressed in various nonexcitable tissues, engineered HEK293 cell line lacking both members of Gs including the brain, gastrointestinal tract, lung, adipocytes, (Gs and Golf), offering a powerful tool to directly assess the kidney, skeletal muscle, heart, and liver (André et al., 1996; contribution of Gs to cellular events. Current methods to Nicholas et al., 1996; Moore et al., 2001), the results obtained investigate the role of Gs involve either overnight treatment in the current study could provide additional insight into how with CTX, which initially causes chronic activation of Gs to b2AR and P2Y receptors might collaborate in normal condi- stimulate its downregulation; the use of the few currently tions and pathophysiology. available small molecule inhibitors, which provide poor target In addition to characterizing a new signaling modality for specificity; and/or inhibition of the downstream effectors AC or the b2AR, the present study introduces a new genetically A. Genetic knockout of Gs provides a useful tool

Fig. 6. ISO promotes ATP release in HA-b2AR-HEK293S cells. HA-b2AR-HEK293S cells were stimulated with ISO (10 mM) for different time points, and ATP was measured in the bulk media. (A) ISO promoted a rapid and transient release of ATP over a period of 60 seconds, with peak release measured at 30 seconds. (B) The response to ISO was blocked in both HEK293S and HA-b2AR-HEK293S cells pretreated with the b2AR-selective antagonist ICI 118551. Data are the mean 6 S.E.M. of two to four independent experiments performed in duplicate. 542 Stallaert et al.

increase in intracellular Ca21, through a mechanism involv- ing exchange protein directly activated by cAMP (EPAC) and « or calcium release-activated calcium chan- nels (Schmidt et al., 2001; Keller et al., 2014). Although we cannot exclude a contribution of these mechanisms to the Ca21 response studied in this work, our data clearly indicate that 21 b2AR activation in HEK293 cells can elicit a Ca response independently of cAMP because blocking approximately 50% of the cAMP production did not affect the Ca21 response (Fig. 3, A and B). Moreover, direct elevation of cAMP following treatment with forskolin or the cAMP analog 8-bromo-cAMP 21 elicited only marginal Ca responses compared with b2AR stimulation (Fig. 3, D and E). Whether calcium release- activated calcium channels could subsequently contribute to the response following the initial Ca21 mobilization from the 2+ endoplasmic reticulum pool described in the current study has

Fig. 7. The ISO-promoted Ca response is modulated by wild-type or Downloaded from dominant-negative Gq overexpression. HA-b2AR-HEK293S were tran- not been investigated. Interestingly, these previous studies siently transfected or not (Mock) with either wild-type Gq or a dominant- used the Fura-2 indicator dye to measure intracellular Ca21, negative mutant of Gq (Q209L/D277N) (Gq-DN), or pretreated with the which is dissolved in the nonionic detergent pluronic F127. Gq-selective inhibitor FR900359, and the Ca2+ response was measured. Data are expressed as the mean 6 S.E.M. of four to five independent This detergent inhibits the multidrug-resistant proteins experiments, each performed in triplicate. (ABC-B transporters), which have been implicated in ATP transport to the extracellular space (Guan et al., 2011). The use of the genetically encoded obelin Ca21 biosensor in the molpharm.aspetjournals.org to specifically investigate the role of Gs and, as was the case in current study should not interfere with this ATP release the present study, allow the discovery of Gs-dependent mechanism and may have aided in the elucidation of the responses that do not require cAMP production. These cells mechanism described in the current work. should prove useful in future studies to explore additional This collaboration between the b2AR and purinergic recep- noncanonical Gs signaling pathways. tors may be of particular relevance in the context of the airway Although the b2AR-mediated release of adenine nucleotides epithelium. Indeed, the b2AR plays an important role in the has been described previously (Baker et al., 2004; Sumi et al., cAMP-dependent secretion of airway surfactants, ciliary 2010), the functional consequences of this signaling event are beating, and regulation of mucosal clearance (Wright and just beginning to be explored. P2Y11 mRNA was the most Dobbs, 1991; Salathe, 2002), properties that contribute to the at ASPET Journals on September 25, 2021 abundant in HEK293S cells, and it has the highest affinity for therapeutic success of b2AR agonists for the treatment of ATP compared with other purinergic receptors expressed in asthma (Giembycz and Newton, 2006). Interestingly, activa- these cells (von Kügelgen, 2006). Yet, the capacity of the b2AR tion of Gq-coupled receptors and increases in intracellular to stimulate the release of ATP may in fact lead to the Ca21 promote undesirable inflammatory responses in pulmo- activation of other purinergic receptor subtypes in cells nary tissues (Barnes, 1998; Rider et al., 2011). Furthermore, expressing a different repertoire or spatial organization of an increase in purine nucleotide release from the airway receptors. epithelium is a hallmark of inflammatory diseases in the The mechanism described in this study differs from pre- lungs (Burnstock et al., 2012), and purinergic signaling has 21 vious reports of b2AR-mediated Ca responses in nonexcit- been proposed to participate in asthmatic airway inflamma- able HEK293 cells, which suggested a cAMP-dependent tion (Basoglu et al., 2005; Idzko et al., 2007), as well as in the

2+ Fig. 8. Ca originating from the intracellular stores is involved in the ISO and Cch-mediated responses. HA-b2AR-HEK293S cells were pretreated or not with the sarco/endoplasmic reticulum Ca2+-ATPase inhibitor thapsigargin (Tg; 5 mM, 1 hour), the cell-permeable Ca2+ chelator BAPTA-AM (20 mM, 2+ 1 hour), or the IP3 receptor antagonist 2-APB (200 mM, 1 hour), followed by stimulation with ISO (10 mM) (A) or Cch (100 mM) (B), and the Ca response was measured. Inset for (A) and (B): Ca2+ response plotted using the area under the curve (AUC). Data are expressed as the mean 6 S.E.M. of three to six independent experiments, each performed in triplicate. AUC data in column graphs were analyzed by one-way analysis of variance, with a Dunnett’s multiple comparison post hoc test, where P , 0.05 (*) was considered significant. b2AR Transactivation of Purinergic Receptors 543 pathogenesis of chronic obstructive pulmonary disorder Bolstad BM, Irizarry RA, Astrand M, and Speed TP (2003) A comparison of nor- malization methods for high density oligonucleotide array data based on variance (Adriaensen and Timmermans, 2004; Mortaz et al., 2010). and bias. Bioinformatics 19:185–193. Given the beneficial effects of cAMP production and deleteri- Bommakanti RK, Vinayak S, and Simonds WF (2000) Dual regulation of Akt/protein 21 kinase B by subunits. J Biol Chem 275:38870–38876. ous consequences of increases in intracellular Ca in airway Braun K, Rettinger J, Ganso M, Kassack M, Hildebrandt C, Ullmann H, Nickel P, epithelia, the identification of functionally selective b2AR Schmalzing G, and Lambrecht G (2001) NF449: a subnanomolar potency antago- nist at recombinant rat P2X1 receptors. Naunyn Schmiedebergs Arch Pharmacol ligands that stimulate cAMP production without promoting 364:285–290. 21 Ca mobilization could represent a desirable class of b2AR Burnstock G, Brouns I, Adriaensen D, and Timmermans JP (2012) Purinergic sig- naling in the airways. Pharmacol Rev 64:834–868. agonists for the treatment of pulmonary disorders. Salme- Caulfield MP (1993) Muscarinic receptors: characterization, coupling and function. terol, a clinically effective b2AR agonist used in the treatment Pharmacol Ther 58:319–379. of asthma, exhibits less side effects than what is typically Cazzola M, Page CP, Rogliani P, and Matera MG (2013) b2-agonist therapy in lung disease. Am J Respir Crit Care Med 187:690–696. associated with chronic use of other b2AR agonists (Cazzola Christ T, Galindo-Tovar A, Thoms M, Ravens U, and Kaumann AJ (2009) Inotropy et al., 2013). Biased signaling studies have also shown that and L-type Ca21 current, activated by beta1- and beta2-adrenoceptors, are dif- ferently controlled by phosphodiesterases 3 and 4 in rat heart. Br J Pharmacol salmeterol is highly efficacious toward cAMP production but 156:62–83. does not increase intracellular Ca21 (van der Westhuizen Collins S, Cao W, and Robidoux J (2004) Learning new tricks from old dogs: beta- adrenergic receptors teach new lessons on firing up adipose tissue metabolism. Mol et al., 2014). Given the results obtained in the current work, Endocrinol 18:2123–2131. future studies examining the influence of purinergic receptor Daaka Y, Luttrell LM, Ahn S, Della Rocca GJ, Ferguson SS, Caron MG, and Lefkowitz RJ (1998) Essential role for G protein-coupled receptor endocytosis Downloaded from transactivation in the context of b2AR-targeted treatment of in the activation of mitogen-activated protein kinase. J Biol Chem 273:685–688. pulmonary disorders could be of great interest. Indeed, it is De Blasi A (1990) Beta-adrenergic receptors: structure, function and regulation. Drugs Exp Clin Res 16:107–112. tempting to speculate that optimal therapy for such pulmonary Duarte T, Menezes-Rodrigues FS, and Godinho RO (2012) Contribution of the ex- disorders might involve harnessing the therapeutic benefit of tracellular cAMP-adenosine pathway to dual coupling of b2-adrenoceptors to Gs – and Gi proteins in mouse skeletal muscle. J Pharmacol Exp Ther 341:820–828. b-adrenergic mediated cAMP/ activity while Dubey RK, Gillespie DG, Mi Z, and Jackson EK (2001) Endogenous cyclic AMP- avoiding the proinflammatory effect of purinergic transactiva- adenosine pathway regulates cardiac fibroblast growth. Hypertension 37:

21 1095–1100. molpharm.aspetjournals.org tion and subsequent Ca -dependent signaling. Dubey RK, Mi Z, Gillespie DG, and Jackson EK (1996) Cyclic AMP-adenosine Our study demonstrates a novel inside-out pathway for the pathway inhibits vascular smooth muscle cell growth. Hypertension 28:765–771. El-Ajouz S, Ray D, Allsopp RC, and Evans RJ (2012) Molecular basis of selective b2AR, adding to its signaling repertoire the transactivation of 21 antagonism of the P2X1 receptor for ATP by NF449 and suramin: contribution of purinergic receptors and stimulation of Ca mobilization. basic amino acids in the cysteine-rich loop. Br J Pharmacol 165:390–400. Given the preponderance of data supporting such transacti- Galandrin S and Bouvier M (2006) Distinct signaling profiles of beta1 and beta2 adrenergic receptor ligands toward adenylyl cyclase and mitogen-activated protein vation mechanisms for both the b2AR and other GPCR family kinase reveals the pluridimensionality of efficacy. Mol Pharmacol 70:1575–1584. members, we propose that such inside-out signaling events Galés C, Rebois RV, Hogue M, Trieu P, Breit A, Hébert TE, and Bouvier M (2005) Real-time monitoring of receptor and G-protein interactions in living cells. Nat are not rare occurrences and instead represent a common Methods 2:177–184. signaling motif for GPCRs that may provide tissue-specific Giembycz MA and Newton R (2006) Beyond the dogma: novel beta2-adrenoceptor signalling in the airways. Eur Respir J 27:1286–1306. signaling in cells expressing a given repertoire of receptors. Gschwind A, Zwick E, Prenzel N, Leserer M, and Ullrich A (2001) Cell communica- at ASPET Journals on September 25, 2021 tion networks: epidermal growth factor receptor transactivation as the paradigm for interreceptor signal transmission. Oncogene 20:1594–1600. Acknowledgments Guan Y, Huang J, Zuo L, Xu J, Si L, Qiu J, and Li G (2011) Effect of pluronic P123 The authors thank Dr. Monique Lagacé for critical reading of the and F127 block copolymer on P-glycoprotein transport and CYP3A metabolism. Arch Pharm Res 34:1719–1728. manuscript. The authors also thank Dr. Kumiko Makide and Yuji Guimarães S and Moura D (2001) Vascular adrenoceptors: an update. Pharmacol Rev Shinjo for technical assistance of flow cytometry cell isolation and 53:319–356. Kouki Kawakami for Western blot analysis. Hohenegger M, Waldhoer M, Beindl W, Böing B, Kreimeyer A, Nickel P, Nanoff C, and Freissmuth M (1998) Gsalpha-selective G protein antagonists. Proc Natl Acad Sci USA 95:346–351. Authorship Contributions Idzko M, Hammad H, van Nimwegen M, Kool M, Willart MA, Muskens F, Hoog- steden HC, Luttmann W, Ferrari D, Di Virgilio F, et al. (2007) Extracellular ATP Participated in research design: Stallaert, van der Westhuizen, Le triggers and maintains asthmatic airway inflammation by activating dendritic Gouill, Bouvier. cells. Nat Med 13:913–919. Conducted experiments: Stallaert, van der Westhuizen, Schönegge, JacksonEK,KoehlerM,MiZ,DubeyRK,TofovicSP,CarcilloJA,andJonesGS(1996) Possible role of adenosine deaminase in vaso-occlusive diseases. J Hypertens 14:19–29. Plouffe, Hogue, Lukashova. Kassack MU, Braun K, Ganso M, Ullmann H, Nickel P, Böing B, Müller G, Contributed new reagents or analytic tools: Inoue, Ishida, Aoki, Le and Lambrecht G (2004) Structure-activity relationships of analogues of NF449 Gouill. confirm NF449 as the most potent and selective known P2X1 receptor antagonist. – Performed data analysis: Stallaert, van der Westhuizen, Schönegge, Eur J Med Chem 39:345 357. Keller MJ, Lecuona E, Prakriya M, Cheng Y, Soberanes S, Budinger GR, Plouffe, Hogue, Lukashova, Inoue, Bouvier. and Sznajder JI (2014) Calcium release-activated calcium (CRAC) channels me- Wrote or contributed to the writing of the manuscript: Stallaert, van diate the b(2)-adrenergic regulation of Na,K-ATPase. FEBS Lett 588:4686–4693. der Westhuizen, Inoue, Bouvier. Kim J, Eckhart AD, Eguchi S, and Koch WJ (2002) Beta-adrenergic receptor- mediated DNA synthesis in cardiac fibroblasts is dependent on transactivation of the epidermal growth factor receptor and subsequent activation of extracellular References signal-regulated kinases. J Biol Chem 277:32116–32123. Adriaensen D and Timmermans JP (2004) Purinergic signalling in the lung: impor- Law SF, Yasuda K, Bell GI, and Reisine T (1993) Gi alpha 3 and G(o) alpha selec- tant in asthma and COPD? Curr Opin Pharmacol 4:207–214. tively associate with the cloned somatostatin receptor subtype SSTR2. J Biol Chem – André C, Erraji L, Gaston J, Grimber G, Briand P, and Guillet JG (1996) Transgenic 268:10721 10727. mice carrying the human beta 2-adrenergic receptor gene with its own promoter Lee FS, Rajagopal R, and Chao MV (2002) Distinctive features of Trk neurotrophin overexpress beta 2-adrenergic receptors in liver. Eur J Biochem 241:417–424. receptor transactivation by G protein-coupled receptors. Cytokine Growth Factor Baker JG, Hall IP, and Hill SJ (2004) Temporal characteristics of cAMP response Rev 13:11–17. element-mediated gene transcription: requirement for sustained cAMP production. Lemieux S (2006) Probe-level linear model fitting and mixture modeling results in Mol Pharmacol 65:986–998. high accuracy detection of differential gene expression. BMC Bioinformatics 7:391. Barnes PJ (1998) Pharmacology of airway smooth muscle. Am J Respir Crit Care Med Levis MJ and Bourne HR (1992) Activation of the alpha subunit of Gs in intact cells 158:S123–S132. alters its abundance, rate of degradation, and membrane avidity. J Cell Biol 119: Basoglu OK, Pelleg A, Essilfie-Quaye S, Brindicci C, Barnes PJ, and Kharitonov SA 1297–1307. (2005) Effects of aerosolized adenosine 59-triphosphate vs adenosine 59-mono- Montalbetti N, Leal Denis MF, Pignataro OP, Kobatake E, Lazarowski ER, phosphate on dyspnea and airway caliber in healthy nonsmokers and patients with and Schwarzbaum PJ (2011) Homeostasis of extracellular ATP in human eryth- asthma. Chest 128:1905–1909. rocytes. J Biol Chem 286:38397–38407. Benitah JP, Alvarez JL, and Gómez AM (2010) L-type Ca(21) current in ventricular Moore DJ, Chambers JK, Wahlin JP, Tan KB, Moore GB, Jenkins O, Emson PC, cardiomyocytes. J Mol Cell Cardiol 48:26–36. and Murdock PR (2001) Expression pattern of human P2Y receptor subtypes: a 544 Stallaert et al.

quantitative reverse transcription-polymerase chain reaction study. Biochim Bio- Seamon KB, Padgett W, and Daly JW (1981) Forskolin: unique diterpene activator of phys Acta 1521:107–119. adenylate cyclase in membranes and in intact cells. Proc Natl Acad Sci USA 78: Mortaz E, Folkerts G, Nijkamp FP, and Henricks PA (2010) ATP and the patho- 3363–3367. genesis of COPD. Eur J Pharmacol 638:1–4. Sitkauskiene B and Sakalauskas R (2005) The role of beta(2)-adrenergic receptors in Nicholas AP, Hökfelt T, and Pieribone VA (1996) The distribution and significance of CNS inflammation and allergy. Curr Drug Targets Inflamm Allergy 4:157–162. adrenoceptorsexaminedwithinsituhybridization.Trends Pharmacol Sci 17:245–255. Stallaert W, Dorn JF, van der Westhuizen E, Audet M, and Bouvier M (2012) Ostrom RS, Gregorian C, and Insel PA (2000) Cellular release of and response to ATP Impedance responses reveal b2-adrenergic receptor signaling pluridimension- as key determinants of the set-point of signal transduction pathways. J Biol Chem ality and allow classification of ligands with distinct signaling profiles. PLoS One 275:11735–11739. 7:e29420. Pérez-Schindler J, Philp A, and Hernandez-Cascales J (2013) Pathophysiological Sumi Y, Woehrle T, Chen Y, Yao Y, Li A, and Junger WG (2010) Adrenergic receptor relevance of the cardiac b2-adrenergic receptor and its potential as a therapeutic activation involves ATP release and feedback through purinergic receptors. Am J target to improve cardiac function. Eur J Pharmacol 698:39–47. Physiol Cell Physiol 299:C1118–C1126. Qi AD, Kennedy C, Harden TK, and Nicholas RA (2001) Differential coupling of the van der Westhuizen ET, Breton B, Christopoulos A, and Bouvier M (2014) Quanti- human P2Y(11) receptor to phospholipase C and adenylyl cyclase. Br J Pharmacol fication of ligand bias for clinically relevant b2-adrenergic receptor ligands: im- 132:318–326. plications for drug taxonomy. Mol Pharmacol 85:492–509. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, and Zhang F (2013) Genome van der Weyden L, Adams DJ, Luttrell BM, Conigrave AD, and Morris MB (2000) engineering using the CRISPR-Cas9 system. Nat Protoc 8:2281–2308. Pharmacological characterisation of the P2Y11 receptor in stably transfected Reed SE, Staley EM, Mayginnes JP, Pintel DJ, and Tullis GE (2006) Transfection of haematological cell lines. Mol Cell Biochem 213:75–81. mammalian cells using linear polyethylenimine is a simple and effective means of von Kügelgen I (2006) Pharmacological profiles of cloned mammalian P2Y-receptor producing recombinant adeno-associated virus vectors. J Virol Methods 138:85–98. subtypes. Pharmacol Ther 110:415–432. Rider CF, King EM, Holden NS, Giembycz MA, and Newton R (2011) Inflammatory Wright JR and Dobbs LG (1991) Regulation of pulmonary surfactant secretion and stimuli inhibit glucocorticoid-dependent transactivation in human pulmonary ep- clearance. Annu Rev Physiol 53:395–414. ithelial cells: rescue by long-acting beta2-adrenoceptor agonists. J Pharmacol Exp Zhang ZS, Cheng HJ, Ukai T, Tachibana H, and Cheng CP (2001) Enhanced cardiac

Ther 338:860–869. L-type calcium current response to beta2-adrenergic stimulation in heart failure. J Downloaded from Salathe M (2002) Effects of beta-agonists on airway epithelial cells. J Allergy Clin Pharmacol Exp Ther 298:188–196. Immunol 110:S275–S281. Schmidt M, Evellin S, Weernink PA, von Dorp F, Rehmann H, Lomasney JW, and Jakobs KH (2001) A new phospholipase-C-calcium signalling pathway medi- Address correspondence to: Dr. Michel Bouvier, Institute for Research in ated by cyclic AMP and a Rap GTPase. Nat Cell Biol 3:1020–1024. Immunology and Cancer, Université de Montréal, P.O. Box 6128, Succursale Schrage R, Schmitz AL, Gaffal E, Annala S, Kehraus S, Wenzel D, Büllesbach KM, Centre-Ville, Montréal, QC, Canada, H3C 3J7. E-mail: michel.bouvier@ Bald T, Inoue A, Shinjo Y, et al. (2015) The experimental power of FR900359 to umontreal.ca study Gq-regulated biological processes. Nat Commun 6:10156. molpharm.aspetjournals.org at ASPET Journals on September 25, 2021 Supplemental Data

Article's title: Purinergic receptor transactivation by the β2-adrenergic receptor increases intracellular Ca2+ in non-excitable cells

Author's names: Wayne Stallaert, Emma T van der Westhuizen, Anne-Marie Schönegge, Bianca Plouffe, Mireille Hogue, Viktoria Lukashova, Asuka Inoue, Satoru Ishida, Junken Aoki, Christian Le Gouill & Michel Bouvier

Journal title: Molecular Pharmacology

1 Supplemental Table 1 – Primer sequences and restriction enzyme combinations to screen for mutants.

Target Primer 1 Primer 2 Enzyme

GNAS 5’-TCAACGGTAGGATGCTGTGG-3’ 5’-CTACAAGAAGGGAGGCCGTG-3’ Hap II

GNAL #1 5’-AAGCACGTTTGCCATTGTCC-3’ 5’-CTTCAGGTTATCCGCCCTCC-3’ Hae III

GNAL #2 5’-ACTGTCACCAAAGCCTCCAG-3’ 5’-TACTGCTTGAGGTGCATCCG-3’ Hap II

GNAL #3 5’-ACACTAAACATAGAGTGGGTGC-3’ 5’-TGAACAAAACTTTCTGGTTGTCAG-3’ Pvu II

2 Supplemental Table 2: List of genes found to be significantly up or down regulated in the ΔGs cells (clone 1) with a minimum of 2 fold change (<-2 or >2) vs the parental cells. Significance was established using unpaired one-way ANOVA between-samples, using the transcriptome analysis console from Affymetrix (p < 0.05 was considered statistically significant). The false discovery rate (FDR) adjustment has been calculated for each gene and is indicated.

Fold Change FDR Gene Description (linear) p-value Symbol (parental vs. ∆Gs) (parental vs ∆Gs) SHISA2 shisa family member 2 11.41 0.110671 PADI3 peptidyl arginine deiminase, type III -2.03 0.110671 ANKRD1 ankyrin repeat domain 1 (cardiac muscle) -16.70 0.110671 CNTFR ciliary neurotrophic factor receptor 2.82 0.110671 CD44 CD44 molecule (Indian blood group) -3.47 0.112203 MPPED2 metallophosphoesterase domain containing 2 2.11 0.112203 TCF24 transcription factor 24 5.14 0.112203 KCTD12 potassium channel tetramerization domain containing 12 36.44 0.112203 EFNB2 ephrin-B2 6.82 0.112203 RPS6KA3 ribosomal protein S6 kinase, 90kDa, polypeptide 3 -2.14 0.112203 GDF11 growth differentiation factor 11 2.09 0.112203 ERVMER34-1 endogenous retrovirus group MER34, member 1 4.41 0.112203 GJB2 gap junction protein beta 2 28.92 0.112203 MRPL50 mitochondrial ribosomal protein L50 2.46 0.112203 OSMR oncostatin M receptor -2.81 0.112203 EPHA7 EPH receptor A7 12.08 0.112203 GPC5 glypican 5 2.29 0.112203 ODF2 outer dense fiber of sperm tails 2 3.02 0.112203 LYST lysosomal trafficking regulator -2.05 0.112203 GREM1 gremlin 1, DAN family BMP antagonist -8.63 0.112203 SRPX2 sushi-repeat containing protein, X-linked 2 -7.93 0.112203 MMP2 matrix metallopeptidase 2 -4.43 0.112203 EMP3 epithelial membrane protein 3 -2.41 0.112203 SERPINF1 serpin peptidase inhibitor, clade F (alpha-2 antiplasmin, pigment epithelium derived factor), member 1 4.50 0.112203 SLC6A9 solute carrier family 6 (neurotransmitter transporter, glycine), member 9 2.01 0.112203 SP100 SP100 nuclear antigen -2.58 0.112203 FAXDC2 fatty acid hydroxylase domain containing 2 -2.01 0.112203 CPS1 carbamoyl-phosphate synthase 1 -3.74 0.112203 NR6A1 nuclear receptor subfamily 6, group A, member 1 2.92 0.112203 NME5 NME/NM23 family member 5 5.70 0.112203 NALCN sodium leak channel, non selective -4.25 0.112203 IL2RG interleukin 2 receptor, gamma -6.77 0.112203 ISOC1 isochorismatase domain containing 1 2.04 0.112438 MLLT11 myeloid/lymphoid or mixed-lineage leukemia; translocated to, 11 -3.38 0.112438 TMEM154 transmembrane protein 154 -2.47 0.113586 GGT1 gamma-glutamyltransferase 1 -4.26 0.113586 STX3 syntaxin 3 -2.01 0.113586 MXRA8 matrix-remodelling associated 8 3.07 0.113586 ALPPL2 alkaline phosphatase, placental like 2 2.41 0.114411 TLR6 toll-like receptor 6 -2.53 0.114411 TRIB2 tribbles pseudokinase 2 -2.50 0.114411 DNAJC3 DnaJ (Hsp40) homolog, subfamily C, member 3 -2.04 0.114411 PAX7 paired box 7 -2.05 0.114411 HHIP hedgehog interacting protein 2.08 0.114411 BMP2 bone morphogenetic protein 2 -5.67 0.114411 ROR2 receptor tyrosine kinase-like orphan receptor 2 5.05 0.114411 F10 coagulation factor X -2.02 0.114411 PARP14 poly(ADP-ribose) polymerase family member 14 -8.88 0.114411 PDGFD platelet derived growth factor D -2.12 0.114411 NFKBIA nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha -3.53 0.114411 KNOP1 lysine-rich nucleolar protein 1 2.26 0.114411 H1F0 H1 histone family, member 0 -2.67 0.114771 TINAGL1 tubulointerstitial nephritis antigen-like 1 -2.86 0.115278 RIT1 Ras-like without CAAX 1 -2.19 0.115278 RIBC2 RIB43A domain with coiled-coils 2 2.08 0.115927 GBP2 guanylate binding protein 2, interferon-inducible -2.20 0.115927 TOR2A torsin family 2, member A 3.30 0.115927 CA2 carbonic anhydrase II 8.75 0.115927 IQGAP2 IQ motif containing GTPase activating protein 2 2.20 0.115992 WDR34 WD repeat domain 34 3.83 0.115992 SULT1C4 sulfotransferase family 1C member 4 10.70 0.115992 FHL2 four and a half LIM domains 2 -2.29 0.115992 GABRA3 gamma-aminobutyric acid (GABA) A receptor, alpha 3 -5.18 0.115992 ABCC2 ATP binding cassette subfamily C member 2 -2.25 0.115992 HOXB-AS3 HOXB cluster antisense RNA 3 20.44 0.119418 CITED1 Cbp/p300-interacting transactivator, with Glu/Asp rich carboxy-terminal domain, 1 4.71 0.121838 TNFRSF9 tumor necrosis factor receptor superfamily, member 9 -16.22 0.122868 FILIP1L filamin A interacting protein 1-like -2.09 0.122868 AMBN ameloblastin 5.54 0.122868 GGT2 gamma-glutamyltransferase 2 -2.08 0.125061 OLFM1 olfactomedin 1 -15.15 0.125061 VRK1 vaccinia related kinase 1 2.43 0.125061 YPEL3 yippee like 3 -2.03 0.125061 CNKSR1 connector enhancer of kinase suppressor of Ras 1 3.26 0.125061 TBC1D13 TBC1 domain family, member 13 2.95 0.125061 GPRC5C G protein-coupled receptor, class C, group 5, member C -2.09 0.125061 ZBTB7B zinc finger and BTB domain containing 7B -2.15 0.125061 GREM1 gremlin 1, DAN family BMP antagonist [Source:HGNC Symbol;Acc:HGNC:2001] -5.89 0.125061 EDIL3 EGF-like repeats and discoidin I-like domains 3 -2.76 0.125061 Supplemental Table 2: continue

MYO5B myosin VB -2.24 0.125061 TCIRG1 T-cell, immune regulator 1, ATPase, H+ transporting, lysosomal V0 subunit A3 -2.03 0.125061 LINC00649 long intergenic non-protein coding RNA 649 7.92 0.125061 MXD1 MAX dimerization protein 1 -2.65 0.125061 SMAD7 SMAD family member 7 -2.05 0.125061 FAM111A family with sequence similarity 111, member A -3.67 0.125061 SPINT1 serine peptidase inhibitor, Kunitz type 1 -3.30 0.125061 MYL9 myosin light chain 9 7.77 0.125061 LGALS1 lectin, galactoside-binding, soluble, 1 -4.66 0.125061 TAPBP TAP binding protein (tapasin) -2.07 0.129017 LRRN1 leucine rich repeat neuronal 1 -3.64 0.129017 MOB3B MOB kinase activator 3B 3.46 0.129017 TGM2 transglutaminase 2 -2.11 0.129017 SLC27A4 solute carrier family 27 (fatty acid transporter), member 4 2.63 0.129017 INA internexin neuronal intermediate filament protein, alpha 6.31 0.129017 CXCL10 chemokine (C-X-C motif) ligand 10 -2.64 0.129017 C1QTNF1 C1q and tumor necrosis factor related protein 1 -3.53 0.129017 DOLK dolichol kinase 2.54 0.129017 TRUB2 TruB pseudouridine (psi) synthase family member 2 2.66 0.129017 DDR1; discoidin domain receptor tyrosine kinase 1; -2.07 0.129017 MIR4640 microRNA 4640 FAM174B family with sequence similarity 174, member B -2.20 0.129179 RGS9 regulator of G-protein signaling 9 -3.45 0.129920 KDM5B lysine (K)-specific demethylase 5B -2.04 0.130577 ZNF711 zinc finger protein 711 2.40 0.131949 HSPB8 heat shock 22kDa protein 8 -2.52 0.133236 MFAP2 microfibrillar associated protein 2 4.33 0.133506 NFKB2 nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100) -3.33 0.133506 NDUFAF2 NADH dehydrogenase (ubiquinone) complex I, assembly factor 2 2.09 0.133506 TFAP2A transcription factor AP-2 alpha (activating enhancer binding protein 2 alpha) -2.02 0.133506 ADAM28 ADAM metallopeptidase domain 28 -2.55 0.133506 FAM27E3 family with sequence similarity 27, member E3 5.35 0.133506 CAV1 caveolin 1 -2.69 0.133506 COQ4 coenzyme Q4 3.77 0.133506 ACSM3 acyl-CoA synthetase medium-chain family member 3 2.08 0.135266 EFNA1 ephrin-A1 -2.24 0.135266 PCYT1B phosphate cytidylyltransferase 1, choline, beta -2.08 0.136233 KIAA1217 KIAA1217 -2.36 0.136233 DTX4 deltex 4, E3 ubiquitin ligase 2.29 0.136366 HBP1 HMG-box transcription factor 1 -2.06 0.137726 POU6F2 POU class 6 homeobox 2 3.42 0.138586 ACTBL2 actin, beta-like 2 -6.53 0.138788 NEK6 NIMA-related kinase 6 2.27 0.139162 RPP25L ribonuclease P/MRP 25kDa subunit-like 2.72 0.140728 FEZF1 FEZ family zinc finger 1 -3.16 0.140857 SPTLC3 serine palmitoyltransferase, long chain base subunit 3 -8.29 0.140857 PHF14 PHD finger protein 14 2.33 0.141430 SV2A synaptic vesicle glycoprotein 2A -2.33 0.141430 PRDM6 PR domain containing 6 3.28 0.141430 SDC4 syndecan 4 -3.34 0.141482 RUNX3 runt-related transcription factor 3 2.50 0.141482 NEFL neurofilament, light polypeptide 2.83 0.142040 DDT D-dopachrome tautomerase 2.10 0.142188 ACADVL acyl-CoA dehydrogenase, very long chain -2.01 0.143238 TNFRSF12A tumor necrosis factor receptor superfamily, member 12A -2.43 0.143238 KRT17 keratin 17, type I -2.56 0.143625 SERPINB8 serpin peptidase inhibitor, clade B (ovalbumin), member 8 -4.11 0.146230 TNFAIP3 tumor necrosis factor, alpha-induced protein 3 -3.62 0.146230 METTL7A methyltransferase like 7A 2.09 0.146230 IER3 immediate early response 3 -7.50 0.149465 URM1 ubiquitin related modifier 1 3.77 0.149465 CASP4 caspase 4 -10.88 0.149833 IL13RA2 interleukin 13 receptor, alpha 2 -2.55 0.149863 FLT1 fms-related tyrosine kinase 1 2.48 0.150351 ETV5 ets variant 5 -5.05 0.151176 PAMR1 peptidase domain containing associated with muscle regeneration 1 -3.90 0.153221 PLTP phospholipid transfer protein 7.87 0.153221 NUAK2 NUAK family, SNF1-like kinase, 2 -2.95 0.153221 LAMP1 lysosomal-associated membrane protein 1 -2.23 0.153221 CCL2 chemokine (C-C motif) ligand 2 -12.25 0.153262 TMCO3 transmembrane and coiled-coil domains 3 -2.13 0.154608 NCKAP1L NCK-associated protein 1-like -2.74 0.154608 SLAIN1 SLAIN motif family member 1 3.82 0.155068 GLE1 GLE1 RNA export mediator 4.51 0.155068 EYA1 EYA transcriptional coactivator and phosphatase 1 -2.62 0.155068 MAL2 mal, T-cell differentiation protein 2 (gene/pseudogene) -3.04 0.155068 SPON1 spondin 1, extracellular matrix protein 2.27 0.155068 MAPKAP1 mitogen-activated protein kinase associated protein 1 2.43 0.155068 YDJC YdjC homolog (bacterial) 2.01 0.155068 TP53INP1 tumor protein p53 inducible nuclear protein 1 -2.46 0.155068 GPR160 G protein-coupled receptor 160 -2.01 0.155068 RINL Ras and interactor like -3.72 0.155068 EID2B EP300 interacting inhibitor of differentiation 2B 2.02 0.155068 SP140L SP140 nuclear body protein-like -4.04 0.155068 SVIL supervillin -2.24 0.155068 GCOM1; MYZAP; POLR2M GRINL1A complex locus 1; myocardial zonula adherens protein; polymerase (RNA) II (DNA directed) polypeptide M -2.01 0.155124 FABP5 fatty acid binding protein 5 (psoriasis-associated) 2.22 0.155124 DUSP9 dual specificity phosphatase 9 2.03 0.155124 TENM1 teneurin transmembrane protein 1 4.35 0.155124 PPP1R15A 1, regulatory subunit 15A -2.55 0.155124 ZER1 zyg-11 related, cell cycle regulator 2.64 0.155124 Supplemental Table 2: continue

GUCY1A3 1, soluble, alpha 3 7.58 0.155124 RALGPS1 Ral GEF with PH domain and SH3 binding motif 1 2.01 0.155144 ATF3 activating transcription factor 3 -3.08 0.155596 SYT11 synaptotagmin XI -4.12 0.155596 NOV nephroblastoma overexpressed -5.79 0.155773 GALNT3 polypeptide N-acetylgalactosaminyltransferase 3 -2.73 0.155773 PRKAR2A protein kinase, cAMP-dependent, regulatory, type II, alpha 2.03 0.155773 RELB v-rel avian reticuloendotheliosis viral oncogene homolog B -3.11 0.155773 CD83 CD83 molecule -2.24 0.155773 GPNMB glycoprotein (transmembrane) nmb -4.50 0.155773 RAC2 ras-related C3 botulinum toxin substrate 2 (rho family, small GTP binding protein Rac2) -3.67 0.155824 MYD88 myeloid differentiation primary response 88 -7.29 0.155824 SPATA16 spermatogenesis associated 16 16.93 0.155959 IQCK IQ motif containing K 2.09 0.156641 MYOF myoferlin -3.62 0.156641 FAM129A family with sequence similarity 129, member A -3.39 0.156641 REXO1L4P REX1, RNA exonuclease 1 homolog-like 4, pseudogene -2.33 0.156641 RNF122 ring finger protein 122 2.78 0.157160 TMEM129 transmembrane protein 129, E3 ubiquitin protein ligase 2.71 0.157433 OPTN optineurin -3.04 0.157570 GABRE; MIR224; MIR452 gamma-aminobutyric acid (GABA) A receptor, epsilon; microRNA 224; microRNA 452 -2.90 0.157570 GREM1 gremlin 1, DAN family BMP antagonist [Source:HGNC Symbol;Acc:HGNC:2001] -10.02 0.157570 CD109 CD109 molecule -3.07 0.157570 LOXL4 lysyl oxidase-like 4 -2.23 0.157570 CYBRD1 cytochrome b reductase 1 -2.28 0.157570 ZBED8 zinc finger, BED-type containing 8 2.65 0.157570 DUSP5 dual specificity phosphatase 5 -2.15 0.157570 FLVCR2 feline leukemia virus subgroup C cellular receptor family, member 2 2.03 0.157570 CPA4 carboxypeptidase A4 -4.72 0.157570 CAPN6 calpain 6 -4.49 0.157845 TFPI2 tissue factor pathway inhibitor 2 -3.61 0.157845 QPCT glutaminyl-peptide cyclotransferase -2.51 0.158768 TRPC6 transient receptor potential cation channel, subfamily C, member 6 -2.10 0.158782 AIFM2 -inducing factor, mitochondrion-associated, 2 -2.07 0.158984 DERL3 derlin 3 3.09 0.160890 PSAT1 phosphoserine aminotransferase 1 2.38 0.160890 CNN1 calponin 1, basic, smooth muscle -2.60 0.161490 MALSU1 mitochondrial assembly of ribosomal large subunit 1 2.24 0.162899 LPAR5 lysophosphatidic acid receptor 5 -2.32 0.162957 GPX3 glutathione peroxidase 3 -2.67 0.162957 RGS7BP regulator of G-protein signaling 7 binding protein 2.59 0.162957 IRX4 iroquois homeobox 4 -2.35 0.162957 DACH1 dachshund family transcription factor 1 3.85 0.162957 JUNB jun B proto-oncogene -3.22 0.162957 DHRS2 dehydrogenase/reductase (SDR family) member 2 -3.84 0.162999 S100A3 S100 calcium binding protein A3 -4.88 0.162999 LIN7A lin-7 homolog A (C. elegans) 2.01 0.164273 DCBLD2 discoidin, CUB and LCCL domain containing 2 -2.12 0.164354 PTDSS2 phosphatidylserine synthase 2 2.11 0.165015 PARD6G par-6 family cell polarity regulator gamma 2.65 0.165351 TXLNG taxilin gamma 2.14 0.165737 CDC42EP3 CDC42 effector protein (Rho GTPase binding) 3 -2.11 0.167513 SPRY4 sprouty RTK signaling antagonist 4 -3.05 0.167513 IRS2 insulin receptor substrate 2 2.49 0.167513 SYTL5 synaptotagmin-like 5 -6.40 0.167769 PLCXD1 phosphatidylinositol-specific phospholipase C, X domain c ontaining 1 2.77 0.168301 GRHPR glyoxylate reductase/hydroxypyruvate reductase 2.07 0.168301 MAP2 microtubule associated protein 2 -2.66 0.168423 ABHD13 abhydrolase domain containing 13 -2.21 0.168423 TSPAN18 tetraspanin 18 2.19 0.168423 MVP; PAGR1 major vault protein; PAXIP1 associated glutamate-rich protein 1 -4.83 0.168423 KCNH2 potassium channel, voltage gated eag related subfamily H, member 2 -3.66 0.168423 C12orf56 12 open reading frame 56 5.71 0.168423 RCAN2 regulator of 2 10.42 0.168423 PCDH9 protocadherin 9 3.51 0.168696 SFN stratifin -2.68 0.168775 RNASE1 ribonuclease, RNase A family, 1 (pancreatic) -2.02 0.169077 HLA-DMA major histocompatibility complex, class II, DM alpha -2.95 0.169077 NR2F1 nuclear receptor subfamily 2, group F, member 1 2.14 0.169077 TMEM80 transmembrane protein 80 2.27 0.169077 selenoprotein M; selenoprotein M [Source:EntrezGene; Acc:140606]; Salzman2013 ANNOTATED, CDS, coding, OVCODE, OVERLAPTX, OVEXON, UTR3 best transcript NM_080430; Salzman2013 ANNOTATED, CDS, coding, OVCODE, SELM -2.36 0.169077 OVERLAPTX, OVEXON, UTR3, UTR5 best transcript NM_080430; Transcript Identified by AceView, Gene ID(s) 140606 HLA-DRB1 major histocompatibility complex, class II, DR beta 1 -2.47 0.169077 S100A16 S100 calcium binding protein A16 -6.34 0.169077 SEL1L3 sel-1 suppressor of lin-12-like 3 (C. elegans) -2.79 0.169077 SLC16A4 solute carrier family 16, member 4 -2.14 0.170338 CD68 CD68 molecule -2.74 0.170338 RAB26 RAB26, member RAS oncogene family 2.56 0.170338 HOXA4 homeobox A4 2.50 0.170400 HLA-DPA1 major histocompatibility complex, class II, DP alpha 1 -4.10 0.170400 PRKCB protein kinase C, beta 2.86 0.170400 HOXC8 homeobox C8 2.49 0.170611 TCEAL3 transcription elongation factor A (SII)-like 3 -2.89 0.170793 GPC3 glypican 3 2.22 0.171328 COL4A1 collagen, type IV, alpha 1 -2.11 0.172224 CD163L1 CD163 molecule-like 1 -8.28 0.172224 GBP3 guanylate binding protein 3 -2.11 0.172739 PNMAL1 paraneoplastic Ma antigen family-like 1 -2.13 0.172779 DSEL dermatan sulfate epimerase-like -5.03 0.172930 Supplemental Table 2: continue

C11orf70 chromosome 11 open reading frame 70 2.09 0.173044 CDK9 -dependent kinase 9 2.07 0.173832 GLIPR1 GLI pathogenesis-related 1 -4.11 0.173832 KRT222 keratin 222, type II -2.40 0.174048 PUDP pseudouridine 5-phosphatase 4.97 0.174048 SLC16A7 solute carrier family 16 (monocarboxylate transporter), member 7 -3.40 0.174229 SPANXB1 SPANX family, member B1 -2.60 0.174229 ALS2CR11 amyotrophic lateral sclerosis 2 chromosome region candidate 11 2.60 0.175651 CCDC102B coiled-coil domain containing 102B -2.32 0.175738 GUCY1B3 guanylate cyclase 1, soluble, beta 3 2.96 0.175976 TNFRSF11B tumor necrosis factor receptor superfamily, member 11b -2.20 0.175976 PLCXD1 phosphatidylinositol-specific phospholipase C, X domain containing 1 2.72 0.175976 TRPV3 transient receptor potential cation channel, subfamily V, member 3 -4.30 0.175976 AIMP2 aminoacyl tRNA synthetase complex-interacting multifunctional protein 2 2.02 0.175976 NPTX2 neuronal pentraxin II 2.66 0.175976 SARM1 sterile alpha and TIR motif containing 1 2.48 0.176147 LOXL1 lysyl oxidase-like 1 -2.22 0.178033 DPM2 dolichyl-phosphate mannosyltransferase polypeptide 2, regulatory subunit 2.68 0.179237 SYNGR4 synaptogyrin 4 -2.64 0.179237 RPA3 replication protein A3 2.07 0.179296 STXBP1 syntaxin binding protein 1 2.21 0.179296 LIPG lipase, endothelial -2.72 0.179296 KLRG1 killer cell lectin-like receptor subfamily G, member 1 2.40 0.179296 NUDCD2 NudC domain containing 2 2.08 0.179296 FAM84B family with sequence similarity 84, member B 3.16 0.179296 IRF1 interferon regulatory factor 1 -2.36 0.179296 ZNF334 zinc finger protein 334 4.81 0.179883 SLC25A1 solute carrier family 25 (mitochondrial carrier; citrate transporter), member 1 2.20 0.180060 MAN2A2 mannosidase, alpha, class 2A, member 2 -2.01 0.180101 UNC5C unc-5 netrin receptor C 2.78 0.181190 KCNT2 potassium channel, sodium activated subfamily T, member 2 -3.69 0.181454 CCBL1 cysteine conjugate-beta lyase, cytoplasmic 2.93 0.181636 MARCH3 membrane associated ring finger 3 2.64 0.182556 TMEM40 transmembrane protein 40 -4.92 0.182630 KCNK1 potassium channel, two pore domain subfamily K, member 1 -2.23 0.183459 ITGA3 integrin alpha 3 -2.05 0.183675 RABEPK Rab9 effector protein with kelch motifs 2.75 0.183675 PTX3 pentraxin 3, long -2.63 0.183693 CLDN1 claudin 1 -3.24 0.184224 PCDHB9 protocadherin beta 9 -2.90 0.185098 ASRGL1 asparaginase like 1 2.17 0.185098 NOVA1 neuro-oncological ventral antigen 1 -3.47 0.185098 SLC1A4 solute carrier family 1 (glutamate/neutral amino acid transporter), member 4 -2.15 0.185098 PLAU plasminogen activator, urokinase -2.16 0.185098 SLC43A1 solute carrier family 43 (amino acid system L transporter), member 1 2.51 0.185361 Homo sapiens DiGeorge syndrome critical region gene 6 (DGCR6), mRNA.; protein DGCR6; Homo sapiens DiGeorge DGCR6; LOC102724770 syndrome critical region gene 6, mRNA (cDNA clone MGC:54086 IMAGE:5229172), complete cds.; Protein DGCR6 2.03 0.185639 [Source:UniProtKB/Swiss-Prot;Acc:Q14129] SLC13A4 solute carrier family 13 (sodium/sulfate symporter), member 4 -2.94 0.185696 SPANXN3 SPANX family, member N3 -3.62 0.186944 SLC35F1 solute carrier family 35, member F1 4.41 0.186944 MCTP2 multiple C2 domains, transmembrane 2 -3.56 0.187292 OSTN osteocrin 3.26 0.187429 GALNT14 polypeptide N-acetylgalactosaminyltransferase 14 4.35 0.187521 SWI5 SWI5 homologous recombination repair protein 2.84 0.187521 AFF2 AF4/FMR2 family, member 2 3.80 0.187557 FAIM2 Fas apoptotic inhibitory molecule 2 -2.35 0.187846 HLA-DRB5 major histocompatibility complex, class II, DR beta 5 -3.91 0.188078 SFXN3 sideroflexin 3 -2.12 0.188313 DDR2 discoidin domain receptor tyrosine kinase 2 2.02 0.189055 POLR3K polymerase (RNA) III (DNA directed) polypeptide K, 12.3 kDa 2.28 0.189702 ZMYM5 zinc finger, MYM-type 5 -2.12 0.189702 GABRQ gamma-aminobutyric acid (GABA) A receptor, theta -3.47 0.189935 TTC39B tetratricopeptide repeat domain 39B 2.17 0.190122 IFI44L interferon-induced protein 44-like -4.29 0.190453 COLEC12 collectin sub-family member 12 2.19 0.190889 FLII flightless I actin binding protein -2.04 0.191209 PCDHB6 protocadherin beta 6 -4.55 0.191353 ARPC5L actin related protein 2/3 complex subunit 5-like 2.04 0.192061 LAMB3; MIR4260 laminin, beta 3; microRNA 4260 -7.53 0.192103 SMAD3 SMAD family member 3 -2.46 0.192233 TFPI tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) -3.55 0.192353 GSTO2 glutathione S-transferase omega 2 -2.69 0.192431 DNAJB9 DnaJ (Hsp40) homolog, subfamily B, member 9 -2.37 0.193172 HAND1 heart and neural crest derivatives expressed 1 3.77 0.194813 ADAP1 ArfGAP with dual PH domains 1 2.24 0.194813 ADCYAP1R1 adenylate cyclase activating polypeptide 1 (pituitary) receptor type I 4.75 0.194813 ADGRA2 adhesion G protein-coupled receptor A2 2.06 0.194813 GALC galactosylceramidase -3.94 0.194839 BTN3A2 butyrophilin, subfamily 3, member A2 -2.05 0.195830 C16orf91 chromosome 16 open reading frame 91 2.05 0.196597 APOBEC3B apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like 3B -2.45 0.196597 PDIA4 protein disulfide isomerase family A, member 4 -2.03 0.196618 USP17L7 ubiquitin specific peptidase 17-like family member 7 -3.34 0.197758 SETSIP SET-like protein 2.96 0.197988 GYPC glycophorin C (Gerbich blood group) 2.96 0.197988 SCIN scinderin 2.05 0.199238 PCDHB16 protocadherin beta 16 -2.71 0.199555 CATSPER1 cation channel, sperm associated 1 -2.12 0.199555 ANKFN1 ankyrin-repeat and fibronectin type III domain containing 1 -2.35 0.200357 DLX5 distal-less homeobox 5 2.99 0.200747 Supplemental Table 2: continue

SEMA3A sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3A -2.50 0.201399 TUFT1 tuftelin 1 -2.03 0.202677 ANK1 ankyrin 1, erythrocytic -2.41 0.203372 FAM214A family with sequence similarity 214, member A -2.21 0.203372 IL32 interleukin 32 -6.65 0.203372 SEMA4B sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4B -2.02 0.203372 UGGT2 UDP-glucose glycoprotein glucosyltransferase 2 -2.29 0.203372 CCNA1 cyclin A1 2.29 0.203372 MAP3K14 mitogen-activated protein kinase kinase kinase 14 -2.12 0.203380 BIRC3 baculoviral IAP repeat containing 3 -3.25 0.203380 AKR1B10 aldo-keto reductase family 1, member B10 (aldose reductase) -2.06 0.203380 RASGRF2 Ras protein-specific guanine nucleotide-releasing factor 2 2.36 0.203530 KHDRBS3 KH domain containing, RNA binding, signal transduction associated 3 4.20 0.203732 HIST1H3I histone cluster 1, H3i 2.83 0.204472 CDR1 cerebellar degeneration related protein 1 -2.02 0.204472 RUNX1 runt-related transcription factor 1 -2.34 0.204472 MR1 major histocompatibility complex, class I-related -2.51 0.204472 MYRF myelin regulatory factor -2.10 0.205442 SLC35D2 solute carrier family 35 (UDP-GlcNAc/UDP-glucose transporter), member D2 -6.37 0.205971 IL7R interleukin 7 receptor -3.62 0.206261 MARC1 mitochondrial amidoxime reducing component 1 2.15 0.206410 LRRC49 leucine rich repeat containing 49 -2.18 0.206552 TCFL5 transcription factor-like 5 (basic helix-loop-helix) 2.05 0.206700 TRIM10 tripartite motif containing 10 -2.06 0.207171 TGFB1 transforming growth factor beta 1 -2.57 0.207319 FURIN furin (paired basic amino acid cleaving enzyme) -2.33 0.208054 LIPH lipase, member H -2.11 0.208054 ANKRD18A; FAM95C ankyrin repeat domain 18A; family with sequence similarity 95, member C 2.07 0.208054 PAQR4 progestin and adipoQ receptor family member IV 2.23 0.208054 DPYD dihydropyrimidine dehydrogenase -3.99 0.209838 STS steroid sulfatase (microsomal), isozyme S 2.91 0.211094 HIST1H1T histone cluster 1, H1t -3.12 0.211155 CARD6 caspase recruitment domain family, member 6 -2.23 0.211783 LUM lumican -2.78 0.212696 SYT1 synaptotagmin I -2.54 0.212696 LBH limb bud and heart development -2.60 0.212696 DMRTA1 DMRT-like family A1 -2.69 0.214015 SERINC2 serine incorporator 2 -2.12 0.214149 ZCCHC10 zinc finger, CCHC domain containing 10 2.10 0.214175 PRKAR1B protein kinase, cAMP-dependent, regulatory, type I, beta 2.32 0.214304 VGLL3 vestigial-like family member 3 -2.92 0.216748 CYP1B1 cytochrome P450, family 1, subfamily B, polypeptide 1 -2.08 0.217849 LRP2 LDL receptor related protein 2 3.22 0.218946 TSHZ2 teashirt zinc finger homeobox 2 3.35 0.220753 EMILIN2 elastin microfibril interfacer 2 -2.58 0.220772 IL6R interleukin 6 receptor -2.85 0.220898 BICC1 BicC family RNA binding protein 1 -2.21 0.221128 RTN4RL1 reticulon 4 receptor-like 1 3.65 0.221128 LIG4 ligase IV, DNA, ATP-dependent -2.66 0.221500 CDH23 cadherin-related 23 2.50 0.222620 GPM6A glycoprotein M6A -2.63 0.223775 ADGRE5 adhesion G protein-coupled receptor E5 -2.35 0.223775 LMX1B LIM homeobox transcription factor 1, beta 2.45 0.224040 TMEM130 transmembrane protein 130 -2.50 0.224040 TNF tumor necrosis factor -5.37 0.224040 KIAA1549L KIAA1549-like 2.39 0.225116 EPHA4 EPH receptor A4 2.27 0.225245 CCSER1 coiled-coil serine rich protein 1 2.31 0.225443 ZNF503 zinc finger protein 503 2.23 0.225601 STMN3 stathmin-like 3 2.76 0.226887 LRRC8C leucine rich repeat containing 8 family, member C -2.04 0.227782 PGR progesterone receptor -4.12 0.227782 MMP13 matrix metallopeptidase 13 -8.58 0.227782 ZNF32 zinc finger protein 32 2.02 0.227782 GABBR2 gamma-aminobutyric acid (GABA) B receptor, 2 2.43 0.227782 TRIM55 tripartite motif containing 55 -2.06 0.227782 ANKRD18B ankyrin repeat domain 18B 2.03 0.227782 FRK fyn-related Src family tyrosine kinase -2.25 0.227782 ZDHHC22 zinc finger, DHHC-type containing 22 2.22 0.227782 RASA4 RAS p21 protein activator 4 -2.13 0.227974 COLCA2 colorectal cancer associated 2 2.06 0.228844 ARRDC3 arrestin domain containing 3 2.13 0.228864 PKIG protein kinase (cAMP-dependent, catalytic) inhibitor gamma -2.03 0.228864 ZFP36 ZFP36 ring finger protein -2.04 0.229162 KMT5C lysine (K)-specific methyltransferase 5C 2.04 0.229550 VWA5A von Willebrand factor A domain containing 5A -2.67 0.229568 SCAI; GOLGA1 suppressor of cancer cell invasion; golgin A1 2.28 0.229568 XPNPEP3 X-prolyl aminopeptidase 3, mitochondrial 2.54 0.229568 SYT14 synaptotagmin XIV -3.78 0.230916 FAM216A family with sequence similarity 216, member A 2.37 0.231097 CEACAM1 carcinoembryonic antigen-related cell adhesion molecule 1 (biliary glycoprotein) -2.04 0.231097 AR androgen receptor 2.21 0.231097 PCBP3 poly(rC) binding protein 3 -2.21 0.232701 SPRY2 sprouty RTK signaling antagonist 2 2.38 0.232874 RUNX2 runt-related transcription factor 2 -2.50 0.233182 MIR99AHG mir-99a-let-7c cluster host gene -2.26 0.233279 GJB7 gap junction protein beta 7 2.45 0.233571 FPGS folylpolyglutamate synthase 2.10 0.233936 SERF1B; SERF1A small EDRK-rich factor 1B (centromeric); small EDRK-rich factor 1A (telomeric) 2.03 0.234194 SERF1A small EDRK-rich factor 1A (telomeric) 2.03 0.234194 IL31RA interleukin 31 receptor A -5.13 0.235161 Supplemental Table 2: continue

ALDH1L2 aldehyde dehydrogenase 1 family, member L2 2.36 0.235237 TLR1 toll-like receptor 1 -2.28 0.235469 ARMCX2 armadillo repeat containing, X-linked 2 -2.51 0.235892 C8orf4 chromosome 8 open reading frame 4 -2.25 0.235892 KRTAP21-2 keratin associated protein 21-2 -3.44 0.235919 ZBTB20; MIR568 zinc finger and BTB domain containing 20; microRNA 568 -3.82 0.236947 RFPL4A ret finger protein-like 4A -2.75 0.237185 SDR16C5 short chain dehydrogenase/reductase family 16C, member 5 -2.19 0.237315 FAM20C family with sequence similarity 20, member C 2.05 0.240017 DDIT3 DNA-damage-inducible transcript 3 -2.56 0.240083 PLIN2 perilipin 2 2.42 0.240226 ARL15 ADP-ribosylation factor like GTPase 15 2.17 0.240227 SYNPO synaptopodin -2.29 0.240227 ALPK3 alpha kinase 3 -2.23 0.240474 SOX11 SRY box 11 -2.41 0.241008 COL9A3 collagen, type IX, alpha 3 2.64 0.241459 SAMD9 sterile alpha motif domain containing 9 -3.80 0.241735 F2RL2 coagulation factor II (thrombin) receptor-like 2 -3.64 0.241735 ARRDC4 arrestin domain containing 4 -2.12 0.241735 CAMK1G calcium/calmodulin-dependent protein kinase IG -2.65 0.241735 LHPP phospholysine phosphohistidine inorganic pyrophosphate phosphatase -2.15 0.241735 DES desmin -2.05 0.241735 FOSL1 FOS-like antigen 1 -2.79 0.241735 RARB retinoic acid receptor, beta 3.94 0.242651 REXO1L10P REX1, RNA exonuclease 1 homolog-like 10, pseudogene -3.05 0.243730 LMNA lamin A/C -2.01 0.244792 SLITRK5 SLIT and NTRK-like family, member 5 -2.26 0.245036 THBS1 thrombospondin 1 -2.74 0.245486 DGKA diacylglycerol kinase alpha -3.09 0.246302 FZD4 frizzled class receptor 4 2.18 0.246357 ANKRD19P ankyrin repeat domain 19, pseudogene 2.32 0.246769 SLC45A3 solute carrier family 45, member 3 -2.06 0.246971 PRR5-ARHGAP8 PRR5-ARHGAP8 readthrough 2.32 0.247324 IFIH1 interferon induced, with helicase C domain 1 -2.47 0.247324 CYP3A5 cytochrome P450, family 3, subfamily A, polypeptide 5 -2.38 0.248155 LHX2 LIM homeobox 2 2.40 0.248155 HLA-DOB major histocompatibility complex, class II, DO beta -2.27 0.248159 CCL20 chemokine (C-C motif) ligand 20 -2.52 0.252810 WFDC2 WAP four-disulfide core domain 2 -2.25 0.252810 USP17L1 ubiquitin specific peptidase 17-like family member 1 -2.46 0.253684 BEGAIN brain-enriched guanylate kinase-associated -2.46 0.254223 DBF4 Transcript Identified by AceView, Entrez Gene ID(s) 10926 2.09 0.254235 MYOM3 myomesin 3 -2.33 0.254235 PCDH10 protocadherin 10 -2.18 0.254758 C1S complement component 1, s subcomponent -2.27 0.257500 HOXA5 homeobox A5 2.62 0.258495 CSF1 colony stimulating factor 1 (macrophage) -2.09 0.258596 CXCL8 chemokine (C-X-C motif) ligand 8 -3.45 0.258596 ENDOG endonuclease G 3.89 0.262148 PLA2G7 phospholipase A2, group VII (platelet-activating factor acetylhydrolase, plasma) 2.83 0.263764 RFPL4AL1 ret finger protein-like 4A-like 1 -2.74 0.264409 RENBP renin binding protein -2.80 0.265119 Supplemental Table 3: Expression levels of the different G protein subunits (α, β and γ) in the parental and ∆Gs cells. The statistical analysis (unpaired one-way ANOVA between-samples; p < 0.05 being considered statistically significant) was done using the “transcriptome analysis console” from Affymetrix. Using a threshold of a minimum of 2 fold change (<-2 or >2), none of the G-protein encoding genes was found to be differentially expressed. The false discovery rate (FDR) adjustment has been calculated for each gene and is indicated.

parental Bi-weight ∆Gs Bi-weight Fold Change ANOVA FDR Gene Description Avg Signal Avg Signal (linear) p-value p-value Symbol (log2) (log2) (parental vs ∆Gs) (parental vs ∆Gs) (parental vs ∆Gs) GNA11 guanine nucleotide binding protein (G protein), alpha 11 12.79 12.85 -1.04 0.135977 0.394487 GNA12 guanine nucleotide binding protein (G protein) alpha 12 11.64 11.22 1.34 0.126440 0.383426 GNA13 guanine nucleotide binding protein (G protein), alpha 13 11.84 12.22 -1.30 0.023307 0.211094 GNA14 guanine nucleotide binding protein (G protein), alpha 14 5.35 5.38 -1.03 0.490967 0.723730 GNA15 guanine nucleotide binding protein (G protein), alpha 15 7.75 7.64 1.08 0.630218 0.815803 GNAI1 guanine nucleotide binding protein (G protein), alpha i1 12.77 13.38 -1.52 0.034347 0.236947 GNAI2 guanine nucleotide binding protein (G protein), alpha i2 12.92 13.29 -1.30 0.000435 0.112203 GNAI3 guanine nucleotide binding protein (G protein), alpha i3 13.29 12.97 1.25 0.106507 0.355931 GNAO1 guanine nucleotide binding protein (G protein), alpha o 8.66 7.80 1.82 0.179868 0.447122 GNAQ guanine nucleotide binding protein (G protein), alpha q 10.67 9.73 1.92 0.014903 0.187846 GNAT1 guanine nucleotide binding protein (G protein), alpha t-rod 6.45 6.49 -1.03 0.823536 0.920852 GNAT2 guanine nucleotide binding protein (G protein), alpha t-cone 7.20 7.21 -1.01 0.920552 0.966923 GNAT3 guanine nucleotide binding protein (G protein), alpha t-gus 4.29 4.73 -1.35 0.312938 0.583998 GNAZ guanine nucleotide binding protein (G protein), alpha z 10.50 9.83 1.59 0.001963 0.129017 GNB1 guanine nucleotide binding protein (G protein), beta 1 15.07 15.24 -1.12 0.128356 0.386013 GNB2 guanine nucleotide binding protein (G protein), beta 2 12.73 13.33 -1.52 0.011808 0.179296 GNB3 guanine nucleotide binding protein (G protein), beta 3 7.91 7.67 1.19 0.198847 0.469394 GNB4 guanine nucleotide binding protein (G protein), beta 4 11.51 11.39 1.09 0.370855 0.631538 GNB5 guanine nucleotide binding protein (G protein), beta 5 7.53 7.98 -1.37 0.039106 0.244321 GNGT1 guanine nucleotide binding protein (G protein), gamma 1 4.98 5.45 -1.39 0.240187 0.512642 GNG2 guanine nucleotide binding protein (G protein), gamma 2 7.33 7.48 -1.11 0.511698 0.739180 GNG3 guanine nucleotide binding protein (G protein), gamma 3 4.16 4.28 -1.09 0.750803 0.884181 GNG4 guanine nucleotide binding protein (G protein), gamma 4 11.38 12.32 -1.91 0.002438 0.133236 GNG5 guanine nucleotide binding protein (G protein), gamma 5 13.84 13.82 1.02 0.503797 0.733463 GNG7 guanine nucleotide binding protein (G protein), gamma 7 9.57 9.31 1.20 0.074743 0.308821 GNG8 guanine nucleotide binding protein (G protein), gamma 8 5.77 5.83 -1.04 0.729230 0.872955 GNGT2 guanine nucleotide binding protein (G protein), gamma 9 6.14 6.69 -1.46 0.332215 0.600536 GNG10 guanine nucleotide binding protein (G protein), gamma 10 10.74 10.19 1.46 0.047841 0.262132 GNG11 guanine nucleotide binding protein (G protein), gamma 11 7.36 8.22 -1.81 0.006633 0.157845 GNG12 guanine nucleotide binding protein (G protein), gamma 12 11.50 12.22 -1.65 0.019686 0.202782 GNG13 guanine nucleotide binding protein (G protein), gamma 13 6.21 5.98 1.17 0.380751 0.638756 Supplemental Figure 1. Genomic sequences of GNAS- and GNAL-double mutant HEK293 cells. sgRNA-target sequences of the three mutant clones (A, CL1; B, CL2; C, CL3) were determined by a TA-cloning method. sgRNA target sequences are boxed and PAM sequences (NGG) are underlined. Arrows indicate a putative double-stranded break site. Restriction enzyme sites (Hap II (GNAS and GNAL #2), Pvu II (GNAL #3) and Hae III (GNAL #1)) are highlighted in red.

10 Supplemental Figure 2. Genotyping of Gs-mutant clones identified by a restriction enzyme- digested fragment method. sgRNA targets including one site (A) in the GNAS gene site and the three sites (sgRNA #1, #2 and #3, shown in B, C and D, respectively) in the GNAL gene were PCR- amplified and treated with the corresponding restriction enzymes (RE; Hap II (A and C), Hae III (B) or Pvu II (D)). The digested samples were subjected to a capillary electrograph analysis (MultiNA, Shimadzu) and pseudo-gel images of electropherogram were visualized by an accessory software with DNA markers of Hae III-digested phaiX174. Filled (grey) and open arrowheads indicate undigested and RE-digested, respectively, PCR fragments of the targeted sites in the parental HEK293 cells. An open arrowhead with an asterisk (CL2 in A) denotes a heteroduplex between the small and large single-stranded DNA. Note that there are four Hae III sites in a parental PCR fragment (B).

11 A An alignment of the Gs amino acids

GNAs-targeting sgRNA #2

B An alignment of the Golf amino acids

GNAL-targeting sgRNA #3

GNAL-targeting sgRNA #1

GNAL-targeting sgRNA #2

Supplemental Figure 3. An alignment of deduced amino acid sequences of the GNAS-(Gαs, A) and GNAL- (Gαolf, B) mutant alleles from the ∆Gs-HEK293 clones. Genomic sequences of the GNAS (A) and GNAL (B) gene from the three mutant clones (CL1, CL2 and CL3) were determined with a TA cloning method. A) Note that all alleles were introduced with a frame-shift mutation. Parent Gαs sequence refers to the isoform GNASL (long isoform). B) Note that except for one allele (CL3 allele #1, which has 28-amino acid insertion), all alleles were introduced with a frame-shift mutation. Parent Gαolf sequence refers to the isoform 2. Arrows indicate the GNAS-andGNAL-(#1, #2 and #3 constructs were used to generate the CL1, CL2 and CL3, respectively) targeting sgRNA site. Aligned regions with 100% identity are shaded in black and those above 50% identity are shaded grey.

12 epnefloigete aorsi AP 0n)o osoi Fk 10 (Fsk, Forskolin or 100nM) ∆ (AVP, vasopressin either following response ∆ epnefloigete spoeeo IO 10 (ISO, isoproterenol either following response the of of Characterization clones – 4 Figure Supplemental A esrdfloigete S rinpoeA38 1 M ciain h ubra h o fteISO the of top the Ca at number the The of activation. mM) % (10 the A23187 represent ionophore or columns ISO either following measured ...o needn xeiet,ec efre ntilct.Clm rpswr nlzdb two- by analyzed were graphs Column comparison triplicate. multiple in Bonferroni’s a performed by each followed ANOVA experiments, way independent 3 of S.E.M. C B sHK9Tclswr rninl rnfce with transfected transiently were cells Gs-HEK293T with transfected transiently cells Gs-HEK293T cAMP response (BRET2) cAMP response (BRET2) Ca2+ response (AUC x 105) 0 2 4 6 8 sHK9Tclstasetytasetdwt h PCBE isno eetse o cAMP for tested were biosensor EPAC-BRET the with transfected transiently cells ∆Gs-HEK293T HEK293 HEK293 HEK293 Overexpressed Overexpressed V2R L L CL3 CL2 CL1 Endogenous ∆Gs-HEK293 L L CL3 CL2 CL1 L L CL3 CL2 CL1 ∆Gs-HEK293 ∆Gs-HEK293 β 2AR 2+ β 2 epneotie ihA28.Dt r xrse stema ± mean the as expressed are Data A32187. with obtained response AR  )o osoi Fk 10 (Fsk, Forskolin or M) A23187 ISO 2 n PCBE isno eetse o cAMP for tested were biosensor EPAC-BRET and V2R 13 sHK9 cells. ∆Gs-HEK293 β A n blnboesr n Ca and biosensor, obelin and 2AR post-hoc test. A aetlHK9Tadthree and HEK293T Parental (A)   )atvto.()HK9Tor HEK293T (B) activation. M) )atvto.()HK9Tor HEK293T (C) activation. M) +2 epnewas response Supplemental Figure 5 – Scatter plot of the microarray comparing gene expression between parental and ∆Gs-HEK293 cells. Gene Level Differential Expression Analysis of the parental vs ∆Gs (clone 1) cells was performed on 21,448 genes using the Clarion S Human array type (Affymetrix). Among these, 491 genes were found to be either transcriptionally up-regulated (198, green) or down-regulated (293, red), including GNAL and GNAS that were knockdown using the CRISPR/Cas9 system. The list of 489 up- or down-regulated genes (with a minimum threshold of 2 fold) are presented in Supplemental Table 2. A Purinergic (P2X) Microarray signal (a.u.)

B Adenosine (P1Y) Purinergic (P2Y) Microarray signal (a.u.)

Supplemental Figure 6 – mRNA expression of P2X and P2Y purinergic receptor subtypes in HA-β2AR- HEK293S cells. Total mRNA expression in HA-β2AR-HEK293S cells was analyzed by gene expression array. Data were normalized using a quantile normalization procedure (see Methods). For comparison purposes, the expression levels of the housekeeping genes β-actin and α-tubulin are shown, since these genes are highly expressed in HA-β2AR-HEK293S cells. Since HEK293 cells have previously been found to not express Gαo to detectable amounts by Western blotting (Law et al., 1993), the normalized value for this probe is included on the graph to represent a baseline level of fluorescence of the microarray probes. The expression of the endogenous β2AR (not the overexpressed HA-β2AR, due to the location of the β2AR probe in the 3’- untralnslated region) is shown for comparison, as well as the expression of all of the purinergic receptor subtypes in HA-β2AR cells.

15