Oncoprotein CIP 2A is stabilized via interaction with tumor suppressor PP 2A/B56
2017; Springer Nature; Volume: 18; Issue: 3 Linguagem: Inglês
10.15252/embr.201642788
ISSN1469-3178
AutoresJiao Wang, Juha Okkeri, Karolina Pavic, Zhizhi Wang, Otto Kauko, Tuuli Halonen, Grzegorz Sarek, Päivi M. Ojala, Zihe Rao, Wenqing Xu, Jukka Westermarck,
Tópico(s)RNA modifications and cancer
ResumoArticle7 February 2017free access Source DataTransparent process Oncoprotein CIP2A is stabilized via interaction with tumor suppressor PP2A/B56 Jiao Wang Jiao Wang Department of Biological Structure, University of Washington, Seattle, WA, USA College of Life Sciences, Nankai University, Tianjin, China Search for more papers by this author Juha Okkeri Juha Okkeri Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Karolina Pavic Karolina Pavic Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Zhizhi Wang Zhizhi Wang Department of Biological Structure, University of Washington, Seattle, WA, USA Search for more papers by this author Otto Kauko Otto Kauko Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Department of Pathology, University of Turku, Turku, Finland Search for more papers by this author Tuuli Halonen Tuuli Halonen Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Grzegorz Sarek Grzegorz Sarek Research Programs Unit, Translational Cancer Biology, University of Helsinki, Helsinki, Finland Search for more papers by this author Päivi M Ojala Päivi M Ojala Research Programs Unit, Translational Cancer Biology, University of Helsinki, Helsinki, Finland Search for more papers by this author Zihe Rao Zihe Rao College of Life Sciences, Nankai University, Tianjin, China Search for more papers by this author Wenqing Xu Corresponding Author Wenqing Xu [email protected] Department of Biological Structure, University of Washington, Seattle, WA, USA Search for more papers by this author Jukka Westermarck Corresponding Author Jukka Westermarck [email protected] orcid.org/0000-0001-7478-3018 Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Department of Pathology, University of Turku, Turku, Finland Search for more papers by this author Jiao Wang Jiao Wang Department of Biological Structure, University of Washington, Seattle, WA, USA College of Life Sciences, Nankai University, Tianjin, China Search for more papers by this author Juha Okkeri Juha Okkeri Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Karolina Pavic Karolina Pavic Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Zhizhi Wang Zhizhi Wang Department of Biological Structure, University of Washington, Seattle, WA, USA Search for more papers by this author Otto Kauko Otto Kauko Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Department of Pathology, University of Turku, Turku, Finland Search for more papers by this author Tuuli Halonen Tuuli Halonen Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Search for more papers by this author Grzegorz Sarek Grzegorz Sarek Research Programs Unit, Translational Cancer Biology, University of Helsinki, Helsinki, Finland Search for more papers by this author Päivi M Ojala Päivi M Ojala Research Programs Unit, Translational Cancer Biology, University of Helsinki, Helsinki, Finland Search for more papers by this author Zihe Rao Zihe Rao College of Life Sciences, Nankai University, Tianjin, China Search for more papers by this author Wenqing Xu Corresponding Author Wenqing Xu [email protected] Department of Biological Structure, University of Washington, Seattle, WA, USA Search for more papers by this author Jukka Westermarck Corresponding Author Jukka Westermarck [email protected] orcid.org/0000-0001-7478-3018 Turku Centre for Biotechnology, University of Turku, Turku, Finland Åbo Akademi University, Turku, Finland Department of Pathology, University of Turku, Turku, Finland Search for more papers by this author Author Information Jiao Wang1,2,‡, Juha Okkeri3,4,‡, Karolina Pavic3,4,‡, Zhizhi Wang1,‡, Otto Kauko3,4,5, Tuuli Halonen3,4, Grzegorz Sarek6, Päivi M Ojala6, Zihe Rao2, Wenqing Xu *,1,‡ and Jukka Westermarck *,3,4,5,‡ 1Department of Biological Structure, University of Washington, Seattle, WA, USA 2College of Life Sciences, Nankai University, Tianjin, China 3Turku Centre for Biotechnology, University of Turku, Turku, Finland 4Åbo Akademi University, Turku, Finland 5Department of Pathology, University of Turku, Turku, Finland 6Research Programs Unit, Translational Cancer Biology, University of Helsinki, Helsinki, Finland ‡These authors contributed equally as first authors ‡These authors contributed equally as second authors ‡These authors contributed equally as senior authors *Corresponding author. Tel: +1 206 221 5609; E-mail: [email protected] *Corresponding author. Tel: +358 2 333 8621; E-mail: [email protected] EMBO Reports (2017)18:437-450https://doi.org/10.15252/embr.201642788 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Protein phosphatase 2A (PP2A) is a critical human tumor suppressor. Cancerous inhibitor of PP2A (CIP2A) supports the activity of several critical cancer drivers (Akt, MYC, E2F1) and promotes malignancy in most cancer types via PP2A inhibition. However, the 3D structure of CIP2A has not been solved, and it remains enigmatic how it interacts with PP2A. Here, we show by yeast two-hybrid assays, and subsequent validation experiments, that CIP2A forms homodimers. The homodimerization of CIP2A is confirmed by solving the crystal structure of an N-terminal CIP2A fragment (amino acids 1–560) at 3.0 Å resolution, and by subsequent structure-based mutational analyses of the dimerization interface. We further describe that the CIP2A dimer interacts with the PP2A subunits B56α and B56γ. CIP2A binds to the B56 proteins via a conserved N-terminal region, and dimerization promotes B56 binding. Intriguingly, inhibition of either CIP2A dimerization or B56α/γ expression destabilizes CIP2A, indicating opportunities for controlled degradation. These results provide the first structure–function analysis of the interaction of CIP2A with PP2A/B56 and have direct implications for its targeting in cancer therapy. Synopsis CIP2A promotes malignancy by inhibiting the tumor suppressor activity of the protein phosphatase PP2A. This study presents the first crystal structure of CIP2A and characterizes the structural requirements for the binding of CIP2A to the B56 subunits of PP2A. Cancerous inhibitor of PP2A (CIP2A) is an obligate dimer that directly interacts with the regulatory PP2A subunits B56α and B56γ. CIP2A binds to B56 proteins via a conserved N-terminal region, and CIP2A homodimerization promotes B56 binding. Inhibition of either CIP2A dimerization or B56α/γ expression destabilizes CIP2A. Introduction Protein phosphatase 2A (PP2A) is a critical tumor suppressor that normally acts by preventing cellular transformation, whereas its inhibition promotes the various malignant characteristics of human cancer cells 12. PP2A also regulates various physiological processes. Therefore, further understanding of structural mechanisms of PP2A regulation is highly relevant for various disciplines. In cancer cells, PP2A inhibition results in hyperphosphorylation of a large number of oncogenic drivers and synergizes with other oncogenic events, such as constitutive RAS activity 1345. Importantly, PP2A complex components are mutated at a relatively low frequency in most types of human cancer. This establishes the reactivation of PP2A as an attractive novel approach in cancer therapy 26. Furthermore, the recent discovery of small molecules and peptides that are capable of restoring PP2A activity in human cancer cell lines provides convincing support to this strategy by demonstrating robust in vivo efficacy in preclinical studies 27. PP2A is inhibited in cancer by a group of otherwise unrelated PP2A inhibitor proteins 28. Among them, cancerous inhibitor of PP2A (CIP2A) is the most prevalent oncoprotein. CIP2A is a long-lived protein in cancer cells 9, and its depletion results in inactivation of many oncogenic PP2A targets (e.g., MYC, E2F1, Akt) 6. Importantly, these effects have been shown to be reversible upon PP2A co-inhibition 6101112. Regarding functional synergism between PP2A inhibition and RAS signaling upon cell transformation, and cell cycle progression 1345, CIP2A overexpression is required for RAS-driven human cell transformation 10. Moreover, we recently demonstrated significant overlap between CIP2A and RAS-regulated phosphoproteomes 13. Clinically, CIP2A overexpression is an equally strong predictor of poor survival in TCGA pan-cancer data as KRAS mutation, and corroborating the functional synergism between CIP2A and RAS, the patients with both of these alterations constituted the patient population with clearly the worst outcome 13. In addition to robust effects of CIP2A depletion by siRNA on malignant cell growth in vitro 6, several studies have demonstrated that CIP2A inhibition very potently inhibits xenograft tumor growth of different types of cancer cells 101415161718. CIP2A also mediates resistance to many cancer therapeutics 6911192021. Importantly, even though CIP2A deficiency inhibits MYC activity, and Her2-driven mammary tumorigenesis in vivo 1122, it does not compromise normal mouse development or growth, except for a defect in spermatogenesis 112223. Notably, high CIP2A protein expression predicts poor patient survival in over a dozen different cancer types 20, and thus, its prognostic and functional relevance equals, or exceeds, that of most oncoproteins that have been traditionally considered important oncogenic drivers. Based on all these data, inhibition of CIP2A protein expression and/or activity could constitute a very efficient cancer therapy strategy without detrimental side effects. However, a lack of structural information for the CIP2A protein has thus far hampered the advancement of this potential cancer therapy target in drug development. PP2A functions as a protein complex consisting of either a core dimer between the scaffolding A subunit (PR65) and the catalytic subunit PP2Ac, or a trimer in which one of the regulatory B subunits interacts with the AC core dimer 24. Our current understanding supports the view that different B subunits mediate the substrate specificity of the PP2A trimer 24 and that only a subset of the numerous B subunits are relevant for the tumor suppressor activity of PP2A 2526. For example, B56α mediates PP2A complex recruitment and the PP2A-mediated dephosphorylation of MYC serine 62 2728. Another B56 family protein, B56γ, functions as a human tumor suppressor 2526 and negatively regulates Akt kinase phosphorylation 2529. CIP2A has been shown to promote phosphorylation and activity of both of these critical PP2A targets 69101216. However, thus far there has not been any evidence of whether CIP2A would directly bind to any of the numerous PP2A complex components. Here, we present the first crystal structure of CIP2A and reveal that CIP2A binds to PP2A B56α and B56γ tumor suppressor subunits directly. Both the CIP2A N-terminal region and CIP2A dimerization contribute to maximal B56 binding. We further show that B56 binding determines CIP2A protein stability in human cell lines. Together, these results provide important insights into poorly understood oncogenic protein CIP2A and may help designing approaches for inhibiting CIP2A protein expression for cancer therapy. Results CIP2A homodimerization Our understanding of proteins that interact with CIP2A is limited 30. Therefore, we conducted a yeast two-hybrid (Y2H) analysis with full-length CIP2A as bait, and using commercial Hybrigenics platform with over 80 × 106 prey clones (Fig EV1A). As CIP2A is expressed at a very low level in most normal tissues but is overexpressed in breast cancer 91112, we used a mixed cDNA library from several breast cancer cell lines (T47D, MDA-MB-468, MCF7, BT20). Using Hybrigenics Global Predicted Biological Score (Global PBS®) computational platform that scores probability of an interaction to be specific, we surprisingly found CIP2A itself as a very high-confidence interaction partner for full-length CIP2A bait (Fig EV1B). The various CIP2A prey clones that interacted with full-length CIP2A bait are depicted as green bars in Fig 1A. Number of independent interacting CIP2A prey clones allowed selected interaction domain (SID) analysis that delineates the shortest fragment that is shared with all interacting clones, and thus represents a potential region mediating the CIP2A homodimerization. The SID analysis for interaction between two CIP2A molecules indicated that CIP2A homodimerization is mediated by a region encompassing amino acids 388–558 (Figs 1A and EV1C). Interestingly, structural foldability (flexibility) analysis indicates that this potential homodimerization domain of CIP2A comprises of a well-folded domain that is followed by a flexible linker and a predicted coiled-coil domain that is most likely disordered (Appendix Fig S1). This prediction was supported by notification that based on gel filtration analysis, the C-terminal fragment per se tends to aggregate. Also, consistent with an earlier publication 31, full-length CIP2A is not stable enough to be purified from either E. coli or insect cells. In contrast, the N-terminal 1–560 fragment of CIP2A spanning the SID could be produced in E. coli in large quantities and was relatively stable. Therefore, we focused on the human CIP2A(1–560) fragment to further confirm CIP2A homodimerization. Click here to expand this figure. Figure EV1. Investigating CIP2A-interacting proteins by yeast two-hybrid assay Summary of screen parameters and description of global PBS classification for confidence of interactions observed in the screen. List of CIP2A (KIAA1524) and B56γ (PPP2R5C) prey clones found to interact with full-length CIP2A bait in Y2H screen. Description of selected interaction domain (SID) determination and graphical representation of SID for both CIP2A (KIAA1524) and B56γ (PPP2R5C) interaction with full-length CIP2A bait. The SID is depicted in relation to predicted structural domains of both proteins (other color bars). Download figure Download PowerPoint Figure 1. Verification of homodimerization of the CIP2A(1–560) Summary of CIP2A prey fragments (green) interacting with full-length CIP2A(1-905) bait (orange) in yeast two-hybrid screen. Original Y2H data are shown in Fig EV1. Number of independent interacting CIP2A prey clones allowed selected interaction domain (SID) analysis that delineates the shortest fragment that is shared with all interacting clones, and thus represents a potential region mediating the CIP2A homodimerization. Gray dotted lines illustrate location of SID across all prey fragments and in full-length CIP2A. Dimerization of CIP2A(1–560) fragment analyzed by GST pulldown. Equal molar amounts of GST and GST-CIP2A(1–560) were incubated with CIP2A(1–560)-V5 fragment for 1 h at 37°C before pulldown. GST-tagged CIP2A(1–560) (90 kDa), but not GST, can pulldown untagged CIP2A(1–560) (60 kDa) in a stoichiometric manner. The SDS–PAGE was stained with Coomassie Blue. SEC-MALS analysis of untagged CIP2A(1–560) on a Superdex 200 Increase 10/300 GL column. The blue curve is the UV absorbance profile, whereas the black line shows the measured molar mass for the major peak. Untagged CIP2A(1–560) has a nominal MW of 62 kDa whereas SEC-MALS chromatogram show shape-independent MW reading at 117.2 kDa which corresponds to the molecular weight of a CIP2A(1–560) dimer. Thermophoresis analysis of interaction between labeled and non-labeled CIP2A(1–560) proteins. Proximity ligation assay (PLA) for interaction between two differently tagged full-length CIP2A proteins. HEK293T cells co-transfected with CIP2A-V5 and EGFP-CIP2A constructs were subjected to PLA with either V5 and GFP antibodies (left panel), or as control with only secondary PLA probes (middle panel). Red dots indicate the association between two CIP2A proteins. As another specificity control, mock-transfected cells were analyzed with PLA including both V5 and GFP primary antibodies. Shown is a representative image from two PLA experiments. Scale bars, 10 μm. Analysis of endogenous CIP2A dimerization by size-exclusion chromatography of HeLa total cell extract and cytoplasmic extracts. Estimated molecular weights are based on column calibration with standard proteins. Shown is a representative result of three independent experiments. Size difference between CIP2A in different fractions is indicative of post-translational regulation of CIP2A upon complex formation. Gray line in the cytoplasm blot is an artifact from film development and does not affect result interpretation. Download figure Download PowerPoint To biochemically verify CIP2A homodimerization, we used thrombin cleavage to remove the GST tag from GST-CIP2A(1–560)-V5 and used this as prey in a GST pulldown experiment with the parental GST-CIP2A(1–560) protein. Using V5 epitope antibody in Western blot analysis of GST pulldown samples, CIP2A(1–560)-V5 was found to not significantly associate with GST alone, whereas a robust interaction was observed between the two CIP2A fragments (Fig 1B). Cleavage of GST tag from GST-CIP2A(1–560)-V5 in the previous assay excluded the possibility that the observed CIP2A dimerization would be mediated by GST dimerization. However, to further exclude the possibility that interaction was mediated due to dimerization via the affinity tags, CIP2A dimerization was further demonstrated by Coomassie staining of gel after pulldown of CIP2A(1–560) without any tags (Fig 1C). Furthermore, our size-exclusion chromatography-coupled multi-angle light scattering (SEC-MALS) analysis clearly show that purified untagged CIP2A(1–560) has a shape-independent molecular mass of 117.2 kDa in solution, which is in good agreement with the calculated MW of 124.5 kDa for a CIP2A(1–560) dimer (Fig 1D). Interestingly, in addition to confirming the CIP2A homodimerization indicated by the Y2H assay, the ability to detect dimerization between two different CIP2A proteins in pulldown assays suggests that binding affinity of the dimerization interface may be relatively modest and that there is detectable exchange between interacting monomers. This conclusion is supported by results of microscale thermophoresis (MST) analysis revealing that CIP2A(1–560) homodimerizes with a modest affinity (Kd) of 290 nM (Fig 1E). Whether the C-terminal sequences lacking from CIP2A(1–560) might further stabilize the dimer remains to be studied. On the other hand, many Y2H prey clones of CIP2A that interacted with full-length CIP2A bait contained long stretches of amino acids C-terminally from amino acid 560 (Fig 1A). This clearly indicates that the homodimerization region included in CIP2A(1–560) is functional also in the presence of C-terminal regions of CIP2A. To further verify dimerization of full-length CIP2A containing those C-terminal sequences, we analyzed physical interaction between two differentially epitope-tagged full-length CIP2A proteins in cells by proximity ligation assay (PLA). PLA has been validated recently by numerous studies to detect protein–protein interaction in cultured cells and in vivo 2232. Here, by using V5 and GFP antibodies coupled with specific PLA probes, we could detect typical PLA dots clearly indicative of interaction between co-transfected CIP2A-V5 and EGFP-CIP2A fusion proteins (Fig 1F, left panel). On the other hand, only random background PLA signals were observed from non-transfected cells with primary V5 and GFP antibodies, or from CIP2A-V5 and EGFP-CIP2A co-transfected cells subjected to PLA without primary antibodies (Fig 1F). To further establish dimerization of endogenous CIP2A, we subjected HeLa cell extracts to size-exclusion chromatography. Consistently with all other results, this analysis clearly showed that both in whole-cell lysate, and in cytoplasmic soluble fraction, CIP2A monomer (~90 kDa) is a minor fraction of total cellular CIP2A pool, whereas majority of CIP2A is found in both dimer (~150–200 kDa) and higher molecular weight complex (> 440 kDa; Fig 1G). This result further supports our conclusions that CIP2A is an obligate dimer. These results reveal that CIP2A homodimerizes, and suggest that the dimerization is mediated by a region containing amino acids 338–558. Crystal structure of CIP2A(1–560) reveals the homodimerization interface To date, no structural information about CIP2A is available. In order to gain structural insights into CIP2A dimerization, the CIP2A(1–560) fragment was crystallized, and its crystal structure was determined at 3.0 Å resolution using the selenium-methionine single-wavelength anomalous scattering (SAD) method (Appendix Table S1 and Appendix Fig S2). In the crystal lattice, there are two CIP2A(1–560) molecules, related by a non-crystallographic twofold axis, in each asymmetric unit (Fig 2A). This finding is fully consistent with both Y2H and biochemical data, that CIP2A(1–560) forms a homodimer. Moreover, in the crystal structure, the dimer interface joining two CIP2A(1–560) molecules is located in the C-terminal end of CIP2A(1–560) which also is fully in line with Y2H SID prediction that postulated the dimerization domain to be located in the region 338–558 of CIP2A. Overall, the CIP2A(1–560) dimer structure resembles an oppositely twisted double hook (Fig 2A). Figure 2. Overall structure of the CIP2A(1–560) dimer Overall structure of the CIP2A(1–560) dimer. Two views of the crystal structure are related by a 90-degree rotation. Positions of N- and C-termini are labeled. Separated view of a CIP2A(1–560) monomer, in three orthogonal views. Positions of the three subdomains are boxed. Download figure Download PowerPoint The CIP2A(1–560) monomer is an all-helical protein, with most of the molecule formed by armadillo or armadillo-like repeats (Fig 2B), and can be roughly divided into "tip", "stem", and C-dimerization subdomains. The first 185 residues form a "tip" domain consisting of five shortened armadillo repeats. Following a twist-forming loop, residues 188–505 form the "stem" domain, consisting of atypical armadillo repeats 6–11; residues 507–559 form three helices that are responsible for CIP2A(1–560) dimerization (Fig 2B). Some of the armadillo repeats in the "stem" subdomain display the structural features of HEAT repeats, as revealed by protein folding similarity searches using the Dali server 33. In addition to the armadillo repeat domains of β-catenin and APC, the atypical HEAT-repeat domain of Wapl is among the closest structural neighbors of the stem subdomain of CIP2A(1–560) (Appendix Table S1 and Appendix Fig S3). Mutational analysis of CIP2A(1–560) dimerization interface The dimerization subdomain is formed by the last three helices of CIP2A(1–560) (Fig 3A and B). The last two helices and the loop linking to the previous helix form a relatively flat and highly hydrophobic surface, mediating the homodimerization of CIP2A(1–560) (Fig 3A and B). Formation of this homodimer interface buries an accessible surface area of 1,913 Å2, which is typical for specific protein–protein interactions. The two C-terminal ends of the CIP2A(1–560) homodimer are spatially very close to each other, and both point to the "top" side of the twisted double hook (Figs 2A and 3A). The key residues involved in the interaction between CIP2A monomers include V525, L529, L532, L533, L546, and I550 (Fig 3C), and all these residues, with the exception of L533, are evolutionarily conserved across different species (Appendix Fig S4). Figure 3. Mutations at the dimer interface of CIP2A negatively affect its dimerization efficiency Detailed interactions in the CIP2A dimer interface mediated by the three helices of C-dimerization domain. The two CIP2A molecules are shown in blue and green. Alternative image of dimer interface in which one CIP2A monomer is shown in space-filled model. The structural image was generated by Pymol. Positions of key hydrophobic residues in the CIP2A homodimerization interface are shown in a "peeled-apart" view. Dimerization of indicated GST-CIP2A(1–560) WT and mutant proteins analyzed by GST pulldown. Equal molar amounts of GST and GST-CIP2A(1–560) proteins were incubated with CIP2A(1–560)-V5 fragment for 1 h at 37°C before pulldown. Samples were analyzed by Western blot using V5 and GST antibodies. Representative image from six experiments is shown. All samples are from the same gel, and black vertical lines indicate where the blot has been cut to remove irrelevant lanes. Quantification of effects of dimerization interface point mutations on CIP2A dimerization. Western blot representative result is shown in (D). Shown is relative dimerization efficiency of indicated CIP2A mutants as compared to GST-CIP2A(1–560) WT, quantified as a ratio between CIP2A(1–560)-V5 and GST-CIP2A(1–560) in pulldown sample. Shown is mean + SEM from six independent experiments. Two-sided t-test between mutant and WT proteins for their relative CIP2A dimerization **P < 0.01. CIP2A–dimer interface with R522D and L533E mutations. The helices of two monomeric CIP2A units are shown in green and blue. Residues at the dimer interface are shown as sticks in magenta. R522 and L533 which are substituted for D and E, respectively, are shown as red sticks and indicated by red text. E523 which might contribute to disrupting the dimer interface by creating electrostatic repulsions with R522D mutant of CIP2A are shown as black sticks. The structure was generated in Pymol. Source data are available online for this figure. Source Data for Figure 3 [embr201642788-sup-0004-SDataFig3.pdf] Download figure Download PowerPoint To interfere with the CIP2A homodimerization interface, we introduced series of single-point mutations to residues that were directly involved in the interaction between CIP2A monomers, or were predicted to potentially interfere with dimerization, and examined their impact on CIP2A dimerization. All created mutations are depicted in Fig EV2. While some of these CIP2A mutants, especially the ones with multiple mutations, had low solubility that prohibited further in vitro test, two soluble single-point mutants, R522D and L533E, repeatedly demonstrated significantly impaired dimerization across six independent assays (Fig 3D and E). L533 is directly involved in the interaction surface between CIP2A monomers (Fig 3C and F). Its substitution by a bulky negatively charged amino acid is therefore likely to destabilize the dimerization interface. On the other hand, mutation of another conserved residue, arginine 522, to a negatively charged aspartate can be predicted to interfere with dimerization by steric and/or electrostatic clashes with the proximal residues such as E523 (Fig 3F), which also is a strictly conserved residue throughout evolution (Appendix Fig S4). Notably, the mode of interference in dimerization by these mutants was reflected with their potency on reducing pulled-down parental CIP2A(1–560)-V5 protein; L533E inhibited dimerization by up to 70%, whereas R552D being not directly involved in interaction surface caused ~50% inhibition (Fig 3D and E). Click here to expand this figure. Figure EV2. Screening for mutants of CIP2 deficient for dimerizationThe helices of two monomeric CIP2A units are shown in light gray and blue. Mutated residues are shown as red sticks. The following mutants were generated and tested: L529A, L532A, L529A L532A (2A), L529A L532A L533A (3A), R522D, Q526E, L529E, L533E, R522D Q526E, Q526E L529E L533E (3E). The structure was generated in RasWin. Download figure Download PowerPoint These results reveal that previously unappreciated homodimerization of CIP2A is mediated by an evolutionary conserved three-helix subdomain (residues 507–559), which form a planar interaction surface. CIP2A directly interacts with PP2A B56 tumor suppressor subunits Regardless of functional evidence that PP2A inhibition mediates CIP2A's oncogenic effects 6101112, no evidence for direct interaction between CIP2A and any of the PP2A complex components has been demonstrated as yet. Importantly, in addition to CIP2A homodimerization, we identified PP2A B subunit B56γ (PPP2R5C) as one of the direct interaction partners of full-length CIP2A by Y2H assay (Fig EV1). On the other hand, Y2H analysis did not reveal direct interaction between CIP2A and scaffolding A subunit, or catalytic C subunit. Direct binding of CIP2A to B56γ is a very exciting result, as together with B56α, B56γ has been shown to be one of the most important tumor suppressor B subunits 2526. To verify these results, the CIP2A(1–560) was demonstrated to interact directly with both B56γ and B56α in a GST pulldown experiment (Fig 4A). The interaction between CIP2A and B56γ and B56α was confirmed by MST analysis, allowing the determination of approximate Kd values for these interactions (Fig 4B). We further verified that full-length CIP2A interacts with B56α and B56γ by PLA in HEK293T cells either co-transfected with HA-tagged versions of B56 proteins and CIP2A-V5 (Fig 4C left pan
Referência(s)