Direct evidence that the GPCR CysLTR2 mutant causative of uveal melanoma is constitutively active with highly biased signaling

Abstract

Uveal melanoma is the most common eye cancer in adults and is clinically and genetically distinct from skin cutaneous melanoma. In a subset of cases, the oncogenic driver is an activating mutation in CYSLTR2, the gene encoding the G protein–coupled receptor cysteinyl-leukotriene receptor 2 (CysLTR2). The mutant CYSLTR2 encodes for the CysLTR2–L129Q receptor, with the substitution of Leu to Gln at position 129 (3.43). The ability of CysLTR2–L129Q to cause malignant transformation has been hypothesized to result from constitutive activity, but how the receptor could escape desensitization is unknown. Here, we characterize the functional properties of CysLTR2–L129Q. We show that CysLTR2–L129Q is a constitutively active mutant that strongly drives Gq/11 signaling pathways. However, CysLTR2–L129Q only poorly recruits β-arrestin. Using a modified Slack–Hall operational model, we quantified the constitutive activity for both pathways and conclude that CysLTR2–L129Q displays profound signaling bias for Gq/11 signaling pathways while escaping β-arrestin–mediated downregulation. CYSLTR2 is the first known example of a G protein–coupled receptor driver oncogene that encodes a highly biased constitutively active mutant receptor. These results provide new insights into the mechanism of CysLTR2–L129Q oncoprotein signaling and suggest CYSLTR2 as a promising potential therapeutic target in uveal melanoma. Uveal melanoma is the most common eye cancer in adults and is clinically and genetically distinct from skin cutaneous melanoma. In a subset of cases, the oncogenic driver is an activating mutation in CYSLTR2, the gene encoding the G protein–coupled receptor cysteinyl-leukotriene receptor 2 (CysLTR2). The mutant CYSLTR2 encodes for the CysLTR2–L129Q receptor, with the substitution of Leu to Gln at position 129 (3.43). The ability of CysLTR2–L129Q to cause malignant transformation has been hypothesized to result from constitutive activity, but how the receptor could escape desensitization is unknown. Here, we characterize the functional properties of CysLTR2–L129Q. We show that CysLTR2–L129Q is a constitutively active mutant that strongly drives Gq/11 signaling pathways. However, CysLTR2–L129Q only poorly recruits β-arrestin. Using a modified Slack–Hall operational model, we quantified the constitutive activity for both pathways and conclude that CysLTR2–L129Q displays profound signaling bias for Gq/11 signaling pathways while escaping β-arrestin–mediated downregulation. CYSLTR2 is the first known example of a G protein–coupled receptor driver oncogene that encodes a highly biased constitutively active mutant receptor. These results provide new insights into the mechanism of CysLTR2–L129Q oncoprotein signaling and suggest CYSLTR2 as a promising potential therapeutic target in uveal melanoma. The superfamily of G protein–coupled receptors (GPCRs) is the largest gene family encoding cell-signaling transmembrane proteins, and approximately one-quarter of ∼400 nonolfactory GPCRs are therapeutic drug targets. Large-scale genomic analysis has revealed that one in five individuals carries a missense variant (MV) in a clinically relevant GPCR gene. The rate of de novo germline MVs in a GPCR gene is one in every 300 newborns, and one in 7 MVs is observed at a potential functionally relevant site (1Hauser A.S. Chavali S. Masuho I. Jahn L.J. Martemyanov K.A. Gloriam D.E. Babu M.M. Pharmacogenomics of GPCR drug targets.Cell. 2018; 172: 41-54.e19Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar). In addition, GPCR genes are commonly mutated in cancer. Somatic mutations are found in 20% of tumor samples, but the lack of specific “hotspot” variants makes it difficult to identify and validate individual receptors as driver oncogenes (2Moore A.R. Ceraudo E. Sher J.J. Guan Y. Shoushtari A.N. Chang M.T. Zhang J.Q. Walczak E.G. Kazmi M.A. Taylor B.S. Huber T. Chi P. Sakmar T.P. Chen Y. Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma.Nat. Genet. 2016; 48: 675-680Crossref PubMed Scopus (137) Google Scholar). We recently reported the discovery of a recurrent “hotspot” somatic missense mutation of the GPCR gene CYSLTR2. The mutant CYSLTR2 encodes cysteinyl-leukotriene receptor 2 (CysLTR2)–L129Q that carries a single amino acid substitution at a highly conserved residue in helix 3 (Ballesteros–Weinstein generic position 3.43) (3Pandy-Szekeres G. Munk C. Tsonkov T.M. Mordalski S. Harpsoe K. Hauser A.S. Bojarski A.J. Gloriam D.E. GPCRdb in 2018: adding GPCR structure models and ligands.Nucleic Acids Res. 2018; 46: D440-D446Crossref PubMed Scopus (231) Google Scholar) and serves as a driver oncogene in patients with uveal melanoma (UVM) (2Moore A.R. Ceraudo E. Sher J.J. Guan Y. Shoushtari A.N. Chang M.T. Zhang J.Q. Walczak E.G. Kazmi M.A. Taylor B.S. Huber T. Chi P. Sakmar T.P. Chen Y. Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma.Nat. Genet. 2016; 48: 675-680Crossref PubMed Scopus (137) Google Scholar). UVM is the most common intraocular malignancy and is associated with a high rate of metastasis with short survival time for patients (4Chua V. Mattei J. Han A. Johnston L. LiPira K. Selig S.M. Carvajal R.D. Aplin A.E. Patel S.P. The Latest on Uveal Melanoma Research and Clinical Trials: Updates from the Cure Ocular Melanoma (CURE OM) Science Meeting (2019).Clin. Cancer Res. 2020; (In press)Crossref Scopus (6) Google Scholar). UVM shows a characteristic pattern of mutually exclusive activating mutations in the CysLTR2–Gq/11–PLCβ4 (phospholipase C-β4) pathway in almost all tumors (2Moore A.R. Ceraudo E. Sher J.J. Guan Y. Shoushtari A.N. Chang M.T. Zhang J.Q. Walczak E.G. Kazmi M.A. Taylor B.S. Huber T. Chi P. Sakmar T.P. Chen Y. Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma.Nat. Genet. 2016; 48: 675-680Crossref PubMed Scopus (137) Google Scholar, 5Moore A.R. Ran L. Guan Y. Sher J.J. Hitchman T.D. Zhang J.Q. Hwang C. Walzak E.G. Shoushtari A.N. Monette S. Murali R. Wiesner T. Griewank K.G. Chi P. Chen Y. GNA11 Q209L Mouse model reveals RasGRP3 as an essential signaling Node in uveal melanoma.Cell Rep. 2018; 22: 2455-2468Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar, 6Hitchman T.D. Bayshtok G. Ceraudo E. Moore A.R. Lee C. Jia R. Wang N. Pachai M.R. Shoushtari A.N. Francis J.H. Guan Y. Chen J. Chang M.T. Taylor B.S. Sakmar T.P. et al.Combined Inhibition of Gαq and MEK Enhances Therapeutic Efficacy in Uveal Melanoma.Clin. Cancer Res. 2020; (In press)Crossref PubMed Scopus (3) Google Scholar). The same CysLTR2–L129Q mutation has also been identified as an oncogenic driver mutation in several other melanocytic tumors (7Kusters-Vandevelde H.V.N. Germans M.R. Rabbie R. Rashid M. Ten Broek R. Blokx W.A.M. Prinsen C.F.M. Adams D.J. Ter Laan M. Whole-exome sequencing of a meningeal melanocytic tumour reveals activating CYSLTR2 and EIF1AX hotspot mutations and similarities to uveal melanoma.Brain Tumor Pathol. 2018; 35: 127-130Crossref PubMed Scopus (7) Google Scholar). CysLTR2 is a significantly mutated GPCR not only in UVM but also in gastrointestinal adenocarcinoma (8Wu V. Yeerna H. Nohata N. Chiou J. Harismendy O. Raimondi F. Inoue A. Russell R.B. Tamayo P. Gutkind J.S. Illuminating the Onco-GPCRome: novel G protein-coupled receptor-driven oncocrine networks and targets for cancer immunotherapy.J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). Here, we that the UVM oncogene CysLTR2–L129Q has a and signaling We show that CysLTR2–L129Q is a constitutively active mutant receptor that strongly to Gq/11 signaling pathways. However, the receptor only recruits and We quantified the signaling bias of the mutant receptor the constitutive activity for the and mutant from the modified Slack–Hall operational we that the receptor bias of the and from is to the of the receptor. characterize the functional of we first the signaling for CysLTR2–L129Q and CysLTR2 CysLTR2 to Gq/11 with the D.J. N. T. R. M. R. L. et of the 2 Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). is the of Gq/11 and results in that is as an of in the of S. A. is a of 2018; PubMed Scopus Google Scholar). We first a time of and in with for CysLTR2 and CysLTR2 the first a the CysLTR2 from with and The with the same of encoding for the CysLTR2–L129Q mutant of as CysLTR2 the of CysLTR2–L129Q the signaling of the receptor a We of CysLTR2 and with a of at the These of and Gq/11 signaling and activity The a in for CysLTR2 that as with the receptor the gene as a modified Slack–Hall operational model In the CysLTR2–L129Q mutant shows to with but the with gene the mutation is the essential for a the of receptors is signaling that with receptor The results also show that the the signaling pathway is not that the high receptor of for CysLTR2–L129Q is not to a high of the signaling the of constitutively active The models and to the of the constitutively active receptor are the model, the operational model P. models of R. Biol. PubMed Scopus Google and the Slack–Hall operational model J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google Scholar, of operational models of receptor constitutive receptor activity and to the of the at the receptor J. PubMed Scopus Google Scholar). cysteinyl-leukotriene receptor from active GPCRs are and downregulation. We the how is CysLTR2–L129Q of signaling at a with the CysLTR2 has been to in to several R. N. M. signaling of and a novel receptor 2 in and PubMed Scopus Google Scholar). However, is known the and of CysLTR2 and CysLTR2–L129Q. We a 2 to the and of to CysLTR2 The are on in with to an variant of Sakmar T.P. receptor 3 variants signaling 2016; PubMed Scopus Google Scholar). We a to characterize the The shows an for approximately the of the to The in the with of the The of the time and but the for almost that of a time is for GPCRs with that with These receptors S.M. S. and PubMed Scopus Google Scholar). The of the for with and the from the that the of is for as with the the of the of the for both we to with the only the characterize the of the mutation on we a of CysLTR2–L129Q in the The results from the show a of with In the for the receptor is that to at We quantified the for both Gq/11 signaling and the modified Slack–Hall operational model to the of receptor bias Gq/11 and pathways J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google Scholar) for to The receptor bias to the pathway of the signaling activity of a receptor J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google in to the bias that pathway of a receptor T. C. V. A. S. for functional and Chem. PubMed Scopus Google Scholar). 2 operational the and Slack–Hall The of et is that the Slack–Hall model to the pathway bias of the constitutive signaling of a receptor to as receptor bias J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google Scholar, of operational models of receptor constitutive receptor activity and to the of the at the receptor J. PubMed Scopus Google Scholar). The Slack–Hall model is an of the operational model, to functional bias T. C. V. A. S. for functional and Chem. PubMed Scopus Google Scholar). In the Slack–Hall model, both the receptor and receptor a The the of an to a The Slack–Hall model the of the model into a of and The is and is as the of the receptor and the receptor in the of an In the in the model is the of and a is as the receptor in the of a The the of the We modified the of to for for the in the is the from the model, we from to The the of the The is that is to the receptor for the of receptor with potential on receptor In we the receptor the gene and the from the to the The and to a model, as is the as a of We the of the to for in the and the and from the to the modified Slack–Hall model the is to the receptor and the and to a with a of of We to a of These for an of and of of for receptor at a We that in the of a the Slack–Hall model to the of a M.A. at and J. PubMed Scopus Google Scholar). We the for CysLTR2 variant and the to a of and are not at only the the from the The are at of receptors and the for high receptor to The of a as a of the receptor is The of the Slack–Hall model as a of the receptor is The of and of the of for the and mutant receptors that in the and the that the receptor the and and on of the and Using the from the in and we are to for the receptor at a The are to the receptor. The of CysLTR2–L129Q to of the CysLTR2 at a receptor The of the receptor is the mutation results in a of the in the The in signaling of CysLTR2–L129Q is as the of to the results suggest that CysLTR2–L129Q displays a for signaling and a for signaling in the Gq/11 signaling with the of the receptor to the of the receptor is and that of the mutant is the of the mutation on the in the pathway is a is the in the the mutation a bias of the constitutive signaling and from is to receptor bias biased has been for with the of receptor bias is a is the first time a has been for the analysis of receptor The is and to mutations that cause receptor bias in We that the receptor bias from to a lack of in the of the receptor. The CysLTR2 has been to a characteristic of a Y. de Y. N. Y. K. A. et of for G protein-coupled Full Text Full Text PDF PubMed Scopus Google with and with with of a time We hypothesized that adding a of the receptor, carries a the of and the receptor to a to and of to the for a time S.M. S. and PubMed Scopus Google Scholar). We a variant as a in we and with of with G protein-coupled receptors the of receptor Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). the bias of CysLTR2–L129Q we the and to we the to CysLTR2 the of a and observed the on and the to is it to a in the The results from and of the and variants for CysLTR2 and are in The the signaling at gene for as with the variant the signaling of the mutant is for as with to the for the receptor at a The of the are with for the and for the the of the signaling of as an of from The both and of for CysLTR2 the of to CysLTR2–L129Q for the variant with the only to that of CysLTR2–L129Q the We also the time for CysLTR2 that a time for the and a for the is with a and with The significantly from for to for in the the variants and the the receptor bias of from We conclude that the receptor bias of the and from is to the of the receptor. the not strongly with the The of with an receptor one that with only with the of the receptor and a with with the and the of the receptor G signaling Y. Y. A. A. O. Han de J. Zhang C. et structure of to PubMed Scopus Google Scholar). The structure of a how one receptor with a G to the transmembrane and with to the G signaling R. T. S. A. F. C. et of an signaling Biol. PubMed Scopus Google Scholar). In to the G the of the with the receptor are only the with has been Y. Y. A. A. O. Han de J. Zhang C. et structure of to PubMed Scopus Google Scholar, T. Hauser A.S. N. Gloriam D.E. S. Babu M.M. of PubMed Scopus Google Scholar). the and we conclude that with the is to strongly with the G for the adding the to the the is to strongly with the of the receptor, the is to with the to the is a in activity is not is that with the G is to a receptor that is with but and for G and We that the receptor bias of CysLTR2–L129Q signaling is to a of an that is the In the G is the at the of the in the G only a active receptor the of with the receptor a active with We to the the receptor bias in CysLTR2–L129Q. In we a CYSLTR2 oncogene in UVM and that CysLTR2–L129Q drives Gq/11 signaling activity in malignant UVM and serves as a driver We that the CysLTR2–L129Q is highly biased Gq/11 signaling pathways and to significantly The lack of a in the to the signaling bias of the high of the mutant receptor. The biased constitutive signaling pattern of CysLTR2–L129Q it and We a that of and Gq/11 signaling and activity, as as receptor on with the encoding the The Gq/11 activity quantified the The activity quantified a from from from The from from and from from the modified and and from and from from and and from and from from and from are in at the from and from The encodes CysLTR2 in to an and a (2Moore A.R. Ceraudo E. Sher J.J. Guan Y. Shoushtari A.N. Chang M.T. Zhang J.Q. Walczak E.G. Kazmi M.A. Taylor B.S. Huber T. Chi P. Sakmar T.P. Chen Y. Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma.Nat. Genet. 2016; 48: 675-680Crossref PubMed Scopus (137) Google Scholar). The the to the to that to the are in and from sequencing The to the These from from E. T. E. Ceraudo E. R.D. M. Huber T. Sakmar T.P. analysis of the PubMed Scopus Google and from and from Sakmar T.P. receptor 3 variants signaling 2016; PubMed Scopus Google Scholar). The the on the with a specific to to the gene of for as as an to in The and from the in of the of the and of The as at for of from and 3 and with a 2 The on the for of and the at for to and a and The of all a The the to to of a the to the of and the to in the The in with of the of and of the at for and 2 of the to on and and the the site that is been of the in at the and the We the in to for The the of to the of Sakmar T.P. receptor 3 variants signaling 2016; PubMed Scopus Google Scholar). The encoding the amino from the of the receptor, to the of the CysLTR2 receptor in the a we also a variant of with G protein-coupled receptors the of receptor Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar) of the amino with the and with to the CysLTR2 receptor. both of amino acid we the in to a for and The to from the of the receptor and the the receptor and the The the on the and are in The of the and the of as that the the of a to a from the and sequencing in with at in to with The at in as in the of with In a the in and for The of with the and for in with 20% and with the and at a of in 7 the with the in and in the from in a and specific the from the in The a of the to the of GPCRs E. M. H. F. F. E. H. C. A. T. F. C. M. et as a of to G protein-coupled receptor PubMed Scopus Google Scholar). The is a the as the and the as the The is a that is the of the to of the has been that also from the of of the of of S. A. is a of 2018; PubMed Scopus Google Scholar). novel pathway not on is found at the the could a to signaling from the with a of and of and of as the of CysLTR2 and in individual with as the of in individual with of with and as the on an at and with 7 of of from to in a the with and to The at for 2 of 3 of and 3 of in the and The in the at on the in and in the in and in in the and The is first to at to at and the is at at and We a of and in with for CysLTR2 and with of and the on an as and the every with of in a at a of at the of in at a of of in and for 2 The adding 3 of the and the in a We a of CysLTR2 with that in in with to an variant of Sakmar T.P. receptor 3 variants signaling 2016; PubMed Scopus Google Scholar). with and the mutant in as with to for the The at an a of the in and for and the to In a the in and for The of with the and for in 20% and and with the and at a of in the from with of and of from all of the to of the with of in the to of the to all other for at the on the of and the the from the to the to at a of and the at the the with of and of and of the to and the to at of and of to in the The into the that as time the of the and as The to and that a every for with of and of and of to of from to to and for at the the and the The from the into a and from a the The to a the The and are the and from the and the and of in are as The from individual into the the and the The and are as and not In cases, to to the to and to the receptor to to specific models in 2 and the of with of CysLTR2 encoding the in the of These first to a The and the and The of the for the the for the that the for We an for with the and for the for that show we also with a as an We the model, the that the in the has the the the of the is to is a the for all 2 operational the and Slack–Hall The of et is that the Slack–Hall model to the pathway bias of the constitutive signaling of a receptor to as receptor bias J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google Scholar, of operational models of receptor constitutive receptor activity and to the of the at the receptor J. PubMed Scopus Google Scholar). The Slack–Hall model is an of the operational model, to functional bias T. C. V. A. S. for functional and Chem. PubMed Scopus Google Scholar). In the Slack–Hall model, both the receptor and receptor a The the of an to a The Slack–Hall model the of the model into a of and The is and is as the of the receptor and the receptor in the of an In the in the model is the of and a is as the receptor in the of a The the of the We modified the of to for for the in the is the from the model, we from The of modified Slack–Hall model is as shows the for the Slack–Hall operational model to the in In is the of the and is the to the for the first for and the from as a for all The is the of the that is the of the The and for all to an for The and for from the the time the in time in These are to a model the is the at time while is the at is the rate of the is the time of and is the the of the the the the The the only the all the time in to the time of the and These to a model the model is the of one and the other The the of the the the We and to the to of the rate and are rate for the The the and is the at time the of is on the of the active receptor and is as the in the the of the and is the of the and are for all and The is first the to a and as the for the of the and is for The for the are in the the of and The to a with an model as a as the We quantified the a modified Slack–Hall operational model to the of the receptor bias Gq/11 and pathways for to and to characterize the of the The receptor bias to the pathway of the signaling activity of a receptor J. adding constitutive receptor activity biased signaling Full Text Full Text PDF PubMed Scopus (6) Google in to the bias that pathway of a receptor T. C. V. A. S. for functional and Chem. PubMed Scopus Google Scholar). The the of the The is that is to the receptor for the of receptor with potential on receptor In we the receptor the gene and the from the to the The first the the and the of the Sakmar T.P. receptor 3 variants signaling 2016; PubMed Scopus Google Scholar) from the that as the to the The are for and the are for all These are to a model, is the as a of The is in the the is to the is to the of We to for in the and the and from the to the modified Slack–Hall model the is to the receptor and the and to a with a of of We to a of These for an of and of of for receptor at a We that in the of a the Slack–Hall model to the of a M.A. at and J. PubMed Scopus Google Scholar). We the of CysLTR2 variant These are to a the The and for and is the of the to is a for the of all is the of the The for The is to the for the receptor. of and are not at only the the from the The are at of receptors and the for high receptor to The of a as a of the receptor is The of the Slack–Hall model as of the receptor is The of and of the of for the and mutant receptors that in the and the that the receptor the and and on of the and Using and we are to the for the receptor at a We the to the receptor. These are reported the in and and to the in the are in the the to the The that of with the of We for the at the and for and with the E. M. J. M. T. A. R. P. Y. T. P. and T. H. the E. M. J. M. T. A. R. M. A. K. the E. M. J. M. T. A. R. P. Y. T. P. and T. H. and the E. M. J. M. T. P. and T. H. the E. C. the and the in M. H. and J. M. M. the in and M. H. We also from the the and the to T. P. S. and T. H. The Cancer and from the Cancer Y. Y. Y. P. P. P. of P. Cancer Cancer C. and P. Cancer Research C. and P. and for

Topics & Concepts