Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Beyond PKA: Evolutionary and structural insights that define a docking and dimerization domain superfamily

2021; Elsevier BV; Volume: 297; Issue: 2 Linguagem: Inglês

10.1016/j.jbc.2021.100927

ISSN

1083-351X

Autores

Heather R. Dahlin, Ning Zheng, John D. Scott,

Tópico(s)

RNA Research and Splicing

Resumo

Protein-interaction domains can create unique macromolecular complexes that drive evolutionary innovation. By combining bioinformatic and phylogenetic analyses with structural approaches, we have discovered that the docking and dimerization (D/D) domain of the PKA regulatory subunit is an ancient and conserved protein fold. An archetypal function of this module is to interact with A-kinase-anchoring proteins (AKAPs) that facilitate compartmentalization of this key cell-signaling enzyme. Homology searching reveals that D/D domain proteins comprise a superfamily with 18 members that function in a variety of molecular and cellular contexts. Further in silico analyses indicate that D/D domains segregate into subgroups on the basis of their similarity to type I or type II PKA regulatory subunits. The sperm autoantigenic protein 17 (SPA17) is a prototype of the type II or R2D2 subgroup that is conserved across metazoan phyla. We determined the crystal structure of an extended D/D domain from SPA17 (amino acids 1–75) at 1.72 Å resolution. This revealed a four-helix bundle-like configuration featuring terminal β-strands that can mediate higher order oligomerization. In solution, SPA17 forms both homodimers and tetramers and displays a weak affinity for AKAP18. Quantitative approaches reveal that AKAP18 binding occurs at nanomolar affinity when SPA17 heterodimerizes with the ropporin-1-like D/D protein. These findings expand the role of the D/D fold as a versatile protein-interaction element that maintains the integrity of macromolecular architectures within organelles such as motile cilia. Protein-interaction domains can create unique macromolecular complexes that drive evolutionary innovation. By combining bioinformatic and phylogenetic analyses with structural approaches, we have discovered that the docking and dimerization (D/D) domain of the PKA regulatory subunit is an ancient and conserved protein fold. An archetypal function of this module is to interact with A-kinase-anchoring proteins (AKAPs) that facilitate compartmentalization of this key cell-signaling enzyme. Homology searching reveals that D/D domain proteins comprise a superfamily with 18 members that function in a variety of molecular and cellular contexts. Further in silico analyses indicate that D/D domains segregate into subgroups on the basis of their similarity to type I or type II PKA regulatory subunits. The sperm autoantigenic protein 17 (SPA17) is a prototype of the type II or R2D2 subgroup that is conserved across metazoan phyla. We determined the crystal structure of an extended D/D domain from SPA17 (amino acids 1–75) at 1.72 Å resolution. This revealed a four-helix bundle-like configuration featuring terminal β-strands that can mediate higher order oligomerization. In solution, SPA17 forms both homodimers and tetramers and displays a weak affinity for AKAP18. Quantitative approaches reveal that AKAP18 binding occurs at nanomolar affinity when SPA17 heterodimerizes with the ropporin-1-like D/D protein. These findings expand the role of the D/D fold as a versatile protein-interaction element that maintains the integrity of macromolecular architectures within organelles such as motile cilia. Protein–protein interactions constrain macromolecules to form molecular machines (1Scott J.D. Pawson T. Cell signaling in space and time: Where proteins come together and when they're apart.Science. 2009; 326: 1220-1224Crossref PubMed Scopus (438) Google Scholar). A-kinase-anchoring proteins (AKAPs) confine PKA within 'signaling islands' to create highly organized signaling compartments (2Langeberg L.K. Scott J.D. Signalling scaffolds and local organization of cellular behaviour.Nat. Rev. Mol. Cell Biol. 2015; 16: 232-244Crossref PubMed Scopus (169) Google Scholar, 3Smith F.D. Reichow S.L. Esseltine J.L. Shi D. Langeberg L.K. Scott J.D. Gonen T. Intrinsic disorder within an AKAP-protein kinase A complex guides local substrate phosphorylation.Elife. 2013; 2e01319Crossref PubMed Google Scholar, 4Smith F.D. Esseltine J.L. Nygren P.J. Veesler D. Byrne D.P. Vonderach M. Strashnov I. Eyers C.E. Eyers P.A. Langeberg L.K. Scott J.D. Local protein kinase A action proceeds through intact holoenzymes.Science. 2017; 356: 1288-1293Crossref PubMed Scopus (89) Google Scholar). A defining attribute of AKAPs is an amphipathic α-helix that binds with high affinity to the docking and dimerization (D/D) domain of PKA regulatory (PKA-R) subunits (5Carr D.W. Stofko-Hahn R.E. Fraser I.D. Bishop S.M. Acott T.S. Brennan R.G. Scott J.D. Interaction of the regulatory subunit (RII) of cAMP-dependent protein kinase with RII-anchoring proteins occurs through an amphipathic helix binding motif.J. Biol. Chem. 1991; 266: 14188-14192Abstract Full Text PDF PubMed Google Scholar, 6Carr D.W. Hausken Z.E. Fraser I.D. Stofko-Hahn R.E. Scott J.D. Association of the type II cAMP-dependent protein kinase with a human thyroid RII-anchoring protein. Cloning and characterization of the RII-binding domain.J. Biol. Chem. 1992; 267: 13376-13382Abstract Full Text PDF PubMed Google Scholar). The PKA holoenzyme is composed of two catalytic (C) subunits constrained by an R subunit dimer (7Turnham R.E. Scott J.D. Protein kinase A catalytic subunit isoform PRKACA; history, function and physiology.Gene. 2016; 577: 101-108Crossref PubMed Scopus (80) Google Scholar, 8Taylor S.S. Ilouz R. Zhang P. Kornev A.P. Assembly of allosteric macromolecular switches: Lessons from PKA.Nat. Rev. Mol. Cell Biol. 2012; 13: 646-658Crossref PubMed Scopus (276) Google Scholar). This broad-spectrum kinase and the signaling events that it influences have been a focus of research spanning over 65 years (9Sutherland Jr., E.W. Wosilait W.D. Inactivation and activation of liver phosphorylase.Nature. 1955; 175: 169-170Crossref PubMed Scopus (93) Google Scholar, 10Fischer E.H. Krebs E.G. Conversion of phosphorylase b to phosphorylase a in muscle extracts.J. Biol. Chem. 1955; 216: 121-132Abstract Full Text PDF PubMed Google Scholar, 11Krebs E.G. The phosphorylation of proteins: A major mechanism for biological regulation. Fourteenth Sir Frederick Gowland Hopkins memorial lecture.Biochem. Soc. Trans. 1985; 13: 813-820Crossref PubMed Scopus (120) Google Scholar). The R-subunit is composed of an amino-terminal D/D domain connected via an autoinhibitory or pseudosubstrate linker to tandem cAMP-binding pockets. There are four known isoforms of PKA-R subunits: RIα, RIβ, RIIα, and RIIβ (8Taylor S.S. Ilouz R. Zhang P. Kornev A.P. Assembly of allosteric macromolecular switches: Lessons from PKA.Nat. Rev. Mol. Cell Biol. 2012; 13: 646-658Crossref PubMed Scopus (276) Google Scholar). Each gene arose from duplication and expansion from an ancestral R-subunit. These PKA–R-subunit isoforms display distinct physiochemical properties and exhibit differential binding affinities for AKAPs (2Langeberg L.K. Scott J.D. Signalling scaffolds and local organization of cellular behaviour.Nat. Rev. Mol. Cell Biol. 2015; 16: 232-244Crossref PubMed Scopus (169) Google Scholar, 4Smith F.D. Esseltine J.L. Nygren P.J. Veesler D. Byrne D.P. Vonderach M. Strashnov I. Eyers C.E. Eyers P.A. Langeberg L.K. Scott J.D. Local protein kinase A action proceeds through intact holoenzymes.Science. 2017; 356: 1288-1293Crossref PubMed Scopus (89) Google Scholar, 12Omar M.H. Scott J.D. AKAP signaling islands: Venues for precision pharmacology.Trends Pharmacol. Sci. 2020; 41: 933-946Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar). The D/D domain was discovered as a unique region of PKA–RIIα subunit that mediates dimerization and AKAP binding (13Scott J.D. Stofko R.E. McDonald J.R. Comer J.D. Vitalis E.A. Mangili J.A. Type II regulatory subunit dimerization determines the subcellular localization of the cAMP-dependent protein kinase.J. Biol. Chem. 1990; 265: 21561-21566Abstract Full Text PDF PubMed Google Scholar). Subsequent NMR and crystallographic studies have characterized the structure of the RI and RII D/D domains in complex with AKAPs (14Newlon M.G. Roy M. Morikis D. Carr D.W. Westphal R. Scott J.D. Jennings P.A. A novel mechanism of PKA anchoring revealed by solution structures of anchoring complexes.EMBO J. 2001; 20: 1651-1662Crossref PubMed Scopus (172) Google Scholar, 15Sarma G.N. Kinderman F.S. Kim C. von Daake S. Chen L. Wang B.C. Taylor S.S. Structure of D-AKAP2:PKA RI complex: Insights into AKAP specificity and selectivity.Structure. 2010; 18: 155-166Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar). R-subunit protomers dimerize to form an X-type helix bundle in an antiparallel arrangement (Fig. 1A). A hydrophobic groove formed at the top of this substructure docks with an amphipathic α-helix on the surface of the AKAP (16Newlon M.G. Roy M. Morikis D. Hausken Z.E. Coghlan V. Scott J.D. Jennings P.A. The molecular basis for protein kinase A anchoring revealed by solution NMR.Nat. Struct. Biol. 1999; 6: 222-227Crossref PubMed Scopus (178) Google Scholar, 17Gold M.G. Lygren B. Dokurno P. Hoshi N. McConnachie G. Tasken K. Carlson C.R. Scott J.D. Barford D. Molecular basis of AKAP specificity for PKA regulatory subunits.Mol. Cell. 2006; 24: 383-395Abstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar, 18Gold M.G. Fowler D.M. Means C.K. Pawson C.T. Stephany J.J. Langeberg L.K. Fields S. Scott J.D. Engineering A-kinase anchoring protein (AKAP)-selective regulatory subunits of protein kinase A (PKA) through structure-based phage selection.J. Biol. Chem. 2013; 288: 17111-17121Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). A helical segment at the amino terminus of RI subunits orients key determinants for AKAP binding (Fig. 1A). In RII subunits, the first five amino acids form β-strands that are essential for docking, with isoleucine's 3 and 5 serving as key PKA-anchoring determinants (19Hausken Z.E. Coghlan V.M. Hastings C.A. Reimann E.M. Scott J.D. Type II regulatory subunit (RII) of the cAMP-dependent protein kinase interaction with A-kinase anchor proteins requires isoleucines 3 and 5.J. Biol. Chem. 1994; 269: 24245-24251Abstract Full Text PDF PubMed Google Scholar) (Fig. 1A). About 60 AKAPs have been identified, each containing a PKA-anchoring helix that associates with D/D domains (Omar and Scott, 2020). These regions of secondary structure have degenerate sequences of 14 to 18 residues, but with a discernable pattern of hydrophobic amino acids critical for docking (5Carr D.W. Stofko-Hahn R.E. Fraser I.D. Bishop S.M. Acott T.S. Brennan R.G. Scott J.D. Interaction of the regulatory subunit (RII) of cAMP-dependent protein kinase with RII-anchoring proteins occurs through an amphipathic helix binding motif.J. Biol. Chem. 1991; 266: 14188-14192Abstract Full Text PDF PubMed Google Scholar, 12Omar M.H. Scott J.D. AKAP signaling islands: Venues for precision pharmacology.Trends Pharmacol. Sci. 2020; 41: 933-946Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar). Peptide studies have uncovered primary structure determinants that influence AKAP binding to RI and to RII (6Carr D.W. Hausken Z.E. Fraser I.D. Stofko-Hahn R.E. Scott J.D. Association of the type II cAMP-dependent protein kinase with a human thyroid RII-anchoring protein. Cloning and characterization of the RII-binding domain.J. Biol. Chem. 1992; 267: 13376-13382Abstract Full Text PDF PubMed Google Scholar, 20Alto N.M. Soderling S.H. Hoshi N. Langeberg L.K. Fayos R. Jennings P.A. Scott J.D. Bioinformatic design of A-kinase anchoring protein-in silico: A potent and selective peptide antagonist of type II protein kinase A anchoring.Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 4445-4450Crossref PubMed Scopus (140) Google Scholar, 21Huang L.J. Durick K. Weiner J.A. Chun J. Taylor S.S. Identification of a novel protein kinase A anchoring protein that binds both type I and type II regulatory subunits.J. Biol. Chem. 1997; 272: 8057-8064Abstract Full Text Full Text PDF PubMed Scopus (251) Google Scholar, 22Burns-Hamuro L.L. Ma Y. Kammerer S. Reineke U. Self C. Cook C. Olson G.L. Cantor C.R. Braun A. Taylor S.S. Designing isoform-specific peptide disruptors of protein kinase A localization.Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 4072-4077Crossref PubMed Scopus (110) Google Scholar, 23Carlson C.R. Lygren B. Berge T. Hoshi N. Wong W. Tasken K. Scott J.D. Delineation of type I protein kinase A-selective signaling events using an RI anchoring disruptor.J. Biol. Chem. 2006; 281: 21535-21545Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, 24Kennedy E.J. Scott J.D. Selective disruption of the AKAP signaling complexes.Methods Mol. Biol. 2015; 1294: 137-150Crossref PubMed Scopus (19) Google Scholar, 25Hundsrucker C. Krause G. Beyermann M. Prinz A. Zimmermann B. Diekmann O. Lorenz D. Stefan E. Nedvetsky P. Dathe M. Christian F. McSorley T. Krause E. McConnachie G. Herberg F.W. et al.High-affinity AKAP7delta-protein kinase A interaction yields novel protein kinase A-anchoring disruptor peptides.Biochem. J. 2006; 396: 297-306Crossref PubMed Scopus (46) Google Scholar). While docking grooves were initially believed to be less modular than peptide motifs, the D/D domain is now designated as a bona fide modular unit (26Bhattacharyya R.P. Remenyi A. Yeh B.J. Lim W.A. Domains, motifs, and scaffolds: The role of modular interactions in the evolution and wiring of cell signaling circuits.Annu. Rev. Biochem. 2006; 75: 655-680Crossref PubMed Scopus (352) Google Scholar, 27Fujita A. Nakamura K. Kato T. Watanabe N. Ishizaki T. Kimura K. Mizoguchi A. Narumiya S. Ropporin, a sperm-specific binding protein of rhophilin, that is localized in the fibrous sheath of sperm flagella.J. Cell Sci. 2000; 113: 103-112Crossref PubMed Google Scholar, 28Carr D.W. Fujita A. Stentz C.L. Liberty G.A. Olson G.E. Narumiya S. Identification of sperm-specific proteins that interact with A-kinase anchoring proteins in a manner similar to the type II regulatory subunit of PKA.J. Biol. Chem. 2001; 276: 17332-17338Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). Studies on spermatozoa revealed that anchoring disruptor peptides, such as Ht31, impair flagellar motility (29Vijayaraghavan S. Goueli S.A. Davey M.P. Carr D.W. Protein kinase A-anchoring inhibitor peptides arrest mammalian sperm motility.J. Biol. Chem. 1997; 272: 4747-4752Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar). Flagellar motility was unaffected by the kinase inhibitor PKI, or the drug H-89 (29Vijayaraghavan S. Goueli S.A. Davey M.P. Carr D.W. Protein kinase A-anchoring inhibitor peptides arrest mammalian sperm motility.J. Biol. Chem. 1997; 272: 4747-4752Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar). This allowed these authors to conclude that anchored PKA was not involved in this process. Since then, D/D proteins such as sperm autoantigenic protein 17 (SPA17), ROPN1, ropporin-1-like protein (ROPN1L), and CABYR have been recognized as nonkinase AKAP helix–binding partners (28Carr D.W. Fujita A. Stentz C.L. Liberty G.A. Olson G.E. Narumiya S. Identification of sperm-specific proteins that interact with A-kinase anchoring proteins in a manner similar to the type II regulatory subunit of PKA.J. Biol. Chem. 2001; 276: 17332-17338Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar, 30Wen Y. Richardson R.T. Widgren E.E. O'Rand M.G. Characterization of Sp17: A ubiquitous three domain protein that binds heparin.Biochem. J. 2001; 357: 25-31Crossref PubMed Scopus (59) Google Scholar, 31Newell A.E. Fiedler S.E. Ruan J.M. Pan J. Wang P.J. Deininger J. Corless C.L. Carr D.W. Protein kinase A RII-like (R2D2) proteins exhibit differential localization and AKAP interaction.Cell Motil. Cytoskeleton. 2008; 65: 539-552Crossref PubMed Scopus (43) Google Scholar). Functional studies report that genetic ablation of these D/D proteins or loss of association with AKAPs impairs motile ciliary action or flagellar motility (32Fiedler S.E. Dudiki T. Vijayaraghavan S. Carr D.W. Loss of R2D2 proteins ROPN1 and ROPN1L causes defects in murine sperm motility, phosphorylation, and fibrous sheath integrity.Biol. Reprod. 2013; 88: 41Crossref PubMed Scopus (54) Google Scholar, 33Young S.A. Miyata H. Satouh Y. Aitken R.J. Baker M.A. Ikawa M. CABYR is essential for fibrous sheath integrity and progressive motility in mouse spermatozoa.J. Cell Sci. 2016; 129: 4379-4387PubMed Google Scholar). The bioinformatic and phylogenetic studies reported herein classifies the D/D domain superfamily. Eighteen superfamily members are subdivided into type I (R1D2) and type II (R2D2) lineages. Many of these proteins are ancient and present in diverse eukaryotic kingdoms. Others result from a gene expansion that took place at the advent of metazoan multicellularity. To gain further insight into the R2D2 lineage, we determined the crystal structure of apo SPA17 1 to 75 from Danio rerio. SPA17 can form homotetramers and displays a low affinity for AKAPs. AKAP binding is considerably enhanced when SPA17 heterodimerizes with another R2D2 protein ROPN1L. Thus, cross-member heterodimerization expands the repertoire and functionality of D/D domains. A combined strategy for data mining was utilized to generate an improved inventory and annotation of D/D domain–containing proteins. Three databases were interrogated to define relationship hierarchies (Fig. 1B). First, the NCBI Conserved Domain Database was searched for proteins within the "Dimerization/docking domain of the regulatory subunit of cAMP-dependent kinase and similar domains". Second, the SuperFamily library was searched for proteins with the "dimerization-anchoring domain of cAMP-dependent PK regulatory subunit." Third, protein BLAST analyses against metazoan and excluding metazoan species generated a comprehensive list of the RIIα clan across all taxa. Screening of the Pfam database refined the RIIα clan as consisting of Dumpy-30 (DPY-30) and PKA-R subunit superfamilies. The output of our data mining strategy is diagrammatically presented in Figure 1B. We defined the RIIα clan as the group comprising the DPY-30 and PKA-R superfamilies. This has led to the identification of 18 PKA-R superfamily members based on sequence identity (Fig. 1C and Table 1). The group is further subdivided into type I and type II PKA-R subunit-like D/D proteins (Fig. 1, B and C).Table 1Human D/D domain of PKA-R superfamily membersPKA-R DD superfamilyChrAAMouse KO phenotypeRopporin-1 (ROPN1)373♂SubfertilityRopporin-1-Like (ROPN1L)573Ciliary dysmotility, ♂subfertilityRopporin-1B (ROPN1B)373Not applicableCa2+-binding Tyr-phosphorylation regulated (CABYR)1875Fibrous sheath dysplasia, ♂subfertilitySperm autoantigenic protein 17 (SPA17)1175Not availablePKA type II regulatory subunit α (PKA-RII)345Reduced interaction with AKAPsPKA type II regulatory subunit β (PKA-RII)745↑ Metabolic rate (RIIβ)↓ Body weight/fat (RIIβ)Ciliogenesis-associated TTC17-interacting protein (CATIP)260♂InfertilityRIIα domain containing protein 1 (RIIAD1)185Absent whiskers, abnormal body wall, neonatal lethalityTubulin polyglutamylase subunit 1 (TPGS1)1945♂Infertility, ↓ body fat, teratozoospermiaPKA type I regulatory subunit α (PKA-RI)1750Carney complex (RIα)PKA type I regulatory subunit β (PKA-RI)750↓ LTD and ↓ LTP (RIβ)Vestibule-1 (VEST1)860Not availableAdenylate kinase 8 (AK8)960HydrocephalyAdenylate kinase 5 (AK5)165Not availableF-Box and Leu-rich repeat protein 13 (FBXL13)770Abnormal eye interiorChamber depthTestis expressed 55 (TEX55)337Not availableEF-Hand Ca2+ bindingProtein 10 (EFCAB10)760Not availableThe gene name of each PKA-R superfamily member as listed in Figure 1C. The chromosomal location (Chr) and number of amino acids (AA) are indicated. Putative functions of each family member are inferred by listing the mouse KO phenotype obtained from the Mouse Genomics Data consortium. Not available denotes that a KO mouse has not been generated.Abbreviations: LTP, long-term potentiation; LTD, long-term depression. Open table in a new tab The gene name of each PKA-R superfamily member as listed in Figure 1C. The chromosomal location (Chr) and number of amino acids (AA) are indicated. Putative functions of each family member are inferred by listing the mouse KO phenotype obtained from the Mouse Genomics Data consortium. Not available denotes that a KO mouse has not been generated. Abbreviations: LTP, long-term potentiation; LTD, long-term depression. A total of 249 D/D domain–containing proteins across all taxa were selected for further analysis (Supplemental material). Metazoans have a full complement of PKA-R-like proteins (Fig. 2A). These include SPA17, ROPN1L, RIIAD1, CATIP, EFCAB10, TPGS1, AK5, AK8, VEST1, FBXL13, and TEX55. Taxa outside the metazoan kingdom contain the PKA-R-like proteins ROPN1L/RSP11, RSP7, TPGS1, EFCAB10, AK8, enolase, and RIIAD1 (Fig. 2A). Higher animals, including humans, additionally evolved the sperm fibrous sheath R2D2 proteins CABYR, ROPN1, and ROPN1B (Fig. 2A). The full gene name of each PKA-R superfamily member is listed in Table 1. An evolutionary tree illustrates how D/D domains evolved across major eukaryotic clades (Fig. 2B). Metazoans lost the D/D domain on enolase despite the expansion of the domain to other proteins (Fig. 2, A and B; Fig. S1). Dendrograms displaying the phylogenetic topology of the D/D superfamily were generated using the RAxML and IQ-tree platforms. Virtually identical branch alignments were obtained on both platforms (Fig. 2C; Fig. S1). Interestingly, organisms which do not rely on flagella for reproduction experienced an evolutionary loss of PKA-R-like D/D proteins. For example, gymnosperms and angiosperms use pollen to produce fertile seeds and do not have R2D2 proteins, but mosses and ferns which utilize sperm have R2D2 proteins (Fig. S2). As previously mentioned, the Pfam algorithm assigns the DPY-30 and related proteins to the RIIα clan. Our analyses designate the DPY-30 clade as an outgroup that is equally related to RI and RII (Figs. 1C and 2C). The crystal structure of DPY-30 reveals a D/D fold similar to PKA-R domains (16Newlon M.G. Roy M. Morikis D. Hausken Z.E. Coghlan V. Scott J.D. Jennings P.A. The molecular basis for protein kinase A anchoring revealed by solution NMR.Nat. Struct. Biol. 1999; 6: 222-227Crossref PubMed Scopus (178) Google Scholar, 34Wang X. Lou Z. Dong X. Yang W. Peng Y. Yin B. Gong Y. Yuan J. Zhou W. Bartlam M. Peng X. Rao Z. Crystal structure of the C-terminal domain of human DPY-30-like protein: A component of the histone methyltransferase complex.J. Mol. Biol. 2009; 390: 530-537Crossref PubMed Scopus (28) Google Scholar). Accordingly, the DPY-30 structure is superimposable over RII with an RMSD of 2.6 Å (Fig. S3A). Further delineation between DPY-30 and RII is evident from probabilistic hidden Markov modeling (Fig. S3B). This algorithm predicts that evolutionary changes have occurred at different positions within the D/D domains of both superfamilies. Sequence determinants that delineate the RI and RII families predominantly lie in the amino-terminal flanking region and in the loop between helix I and helix II of the D/D domain (Fig. 1A). Phylogenetic and topological analyses have used this information to subdivide the PKA-R superfamily D/D domains into two distinct but overlapping groups (Fig. 1, B and C). A hallmark of RI subunits is the presence of two prolines on each end of the loop between helix I and helix II. Hence, enolase, EFCAB10, AK5, AK8, VEST1, FBXL13, and TEX55 are prototypic of the R1D2 clade (Fig. 1C, gold underlined). In contrast, a defining feature of the R2D2 clade is replacement of the second proline with a hydrophobic side chain (Ile, Leu, or Val, Fig. 1C, blue underlined). Five proteins, ROPN1, ROPN1L, SPA17, RIIAD1, and CATIP, follow this convention (Figs. 1C and 2C; Fig. S1). Other determinants also contribute to the R1D2 or R2D2 designations. For example, TPGS1 is considered an R1D2 protein because it contains a predicted helical flanking amino-terminal motif and a second proline in the loop region. Yet, TPGS1 can also be considered an R2D2 protein because of features such as a glycine at the start of helix I and a conserved "YF" motif in helix II (Figs. 1C and 2C). Likewise, RIIAD1 and CATIP are intermediate to the R2D2 clade because they are predicted to have an amino-terminal helix rather than a β strand (Fig. 1C). Thus, our phylogenetic analyses have defined primary, secondary, and tertiary structure characteristics that are emblematic of the DPY-30, R1D2, and R2D2 subgroups of the RIIα clan. All data have been deposited in the Dryad server. We chose to focus our structural analyses on SPA17 because of its extended D/D domain. The zebrafish ortholog is 72% identical to the human ortholog and proved amenable to crystallization in multispecies trials (Fig. 3A). A construct spanning amino acids 1 to 75 of SPA17 from D. rerio was expressed in Escherichia coli, and the resultant protein was purified with a sequential three-step affinity, anion-exchange, and size-exclusion chromatography (SEC) protocol (Fig. S4). Crystals of SPA17 diffracted X-ray to 1.72 Å. The structure was determined by molecular replacement using the D/D domain of PKA-RIIα as a search model (PDB ID: 2IZX) and subsequently refined to an Rwork of 0.154 and Rfree of 0.165 (Fig. 3B and Table 2).Table 2Crystallographic data and refinement statisticsPropertyValueSpace groupP 32Cell constants a, b, c, α, β, γ60.96 Å 60.96 Å 89.02 Å90.00° 90.00° 120.00°Resolution (Å)34.03–1.7245.41–1.72% Data completeness (in resolution range)98.5 (45.41–1.72)Wavelength0.99996 ÅRmeas0.098Rpim0.035Rmerge (1.72–1.75 Å)0.91 (0.269)Data redundancy (1.72–1.75 Å)7.6 (5.6)CC1/20.993< I/σ(I) >1.02 (at 1.72 Å)Refinement programphenix.refine 1.18.2_3874, PHENIX 1.18.2_3874R, Rfree0.154, 0.1650.154, 0.166RMS (angles), RMS (bonds)0.92, 0.008Ramachandran favored100%Rfree test set1995 reflections (5.11%)Wilson B-factor (Å2)11.8Anisotropy0.632Bulk solvent ksol(e/Å3), Bsol(Å2)0.42, 34.0L-test for twinning< |L| > = 0.51, < L2 > = 0.34Estimated twinning fraction0.479 for -h,-k,l0.480 for h,-h-k,-l0.479 for -k,-h,-lFo,Fc correlation0.96Total number of atoms4799Average B, all atoms (Å2)17.0 Open table in a new tab Four copies of SPA17 are observed in the asymmetric unit of the crystal. The two central copies of SPA17 form the canonical four-helix bundle as previously observed in the homodimeric D/D domains of R1 and RII (Figs. 1A and 3B). The other two copies each form a similar homodimer with a symmetry related SPA17 chain. Interestingly, the two conserved sequence regions flanking the central helices of SPA17 both adopt a regular β-strand conformation. A four-stranded β-sheet is formed from the amino-terminal β-strand of two SPA17 molecules and the carboxyl-terminal β-strands of two other SPA17 chains (Fig. 3B). Because of the close involvement of these β-strands in crystal packing, the formation of the four-stranded β-sheet is likely a crystallization artifact. Nonetheless, these β-strands might mediate SPA17 oligomerization, as biochemical studies indicate that SPA17 can exist in higher order configurations (Fig. 3, C and D). SEC coupled to multiangle light scattering (SEC-MALS) verifies that SPA17 tetramers and dimers exist in solution (Fig. 3, C and D). The SPA17 1 to 75 fragment (8.8-kDa monomer) has molecular masses of 35 and 17 kDa (Fig. 3C). Although less evident, multimerization of full-length SPA17 (17.4-kDa monomer) was also observed. The purified protein ensemble elutes with molecular masses of 72 and 36 kDa (Fig. 3D). Collectively, the data in Figure 3 imply that, unlike RIIα, SPA17 can form higher order homo-oligomeric complexes. Although SPA17 exhibits distinctive structural features, it still retains many hallmarks of a canonical R2D2 protein. Alignment of the SPA17 structure to the D/D domain of apo RIIα dimers results in an associated RMSD of 0.482 Å. Similarly, the apo structure of SPA17 superimposes over RIIα in complex with AKAP-in silico with an RMSD of 0.514 Å (Fig. 4A, left panel). The AKAP-binding site is, therefore, retained in the SPA17 homodimer, although it appears to be occluded by the β-sheet formed among the terminal strands of SPA17 protomers in the crystal (Fig. 4A, right panel). As expected, key hydrophobic residues necessary for dimerization (magenta squares) are strictly conserved, but only five of the docking determinants (purple dots) are invariant in the sequence alignment (Fig. 4B). Together, these features suggest that the extended D/D domain of SPA17 contains most necessary determinants for binding to AKAPs. Thus, the mode of SPA17 interaction with AKAPs might be slightly different than how RII interfaces with its anchoring proteins. SPA17 coexists with ropporin-1-like proteins in the flagellum of mammalian sperm and motile cilia (31Newell A.E. Fiedler S.E. Ruan J.M. Pan J. Wang P.J. Deininger J. Corless C.L. Carr D.W. Protein kinase A RII-like (R2D2) proteins exhibit differential localization and AKAP interaction.Cell Motil. Cytoskeleton. 2008; 65: 539-552Crossref PubMed Scopus (43) Google Scholar). Sequence similarities between these members of the R2D2 clade raised the possibility that SPA17 and its close relative ROPN1L may form heterodimers (Fig. 1C). In keeping with this notion, full-length SPA17 and ROPN1L comigrate as assessed by SEC-MALS analysis (Fig. 5A). Likewise, SPA17 1 to 75 and ROPN1L 1 to 75 multimerize when analyzed by SEC-MALS (Fig. 5B). Gel filtration traces further indicate that SPA17–ROPN1L complexes migrate with an apparent molecular weight that is consistent with a heterodimer with a minor tetrameric species (Fig. 5, A and B). Protein pulldowns verified interaction between glutathione-S-transferase (GST)-ROPN1L and SPA17 (Fig. 5C). Reciprocal pull-down experiments confirmed

Referência(s)