proteins STRUCTURE O FUNCTION O BIOINFORMATICS
Computational identification of posttranslational modification-based nuclear import regulations by characterizing nuclear localization signal-import receptor interaction Jhih-Rong Lin, Zhonghao Liu, and Jianjun Hu* Department of Computer Science and Engineering, University of South Carolina, Columbia, South Carolina 29208
ABSTRACT The binding affinity between a nuclear localization signal (NLS) and its import receptor is closely related to corresponding nuclear import activity. PTM-based modulation of the NLS binding affinity to the import receptor is one of the most understood mechanisms to regulate nuclear import of proteins. However, identification of such regulation mechanisms is challenging due to the difficulty of assessing the impact of PTM on corresponding nuclear import activities. In this study we proposed NIpredict, an effective algorithm to predict nuclear import activity given its NLS, in which molecular interaction energy components (MIECs) were used to characterize the NLS-import receptor interaction, and the support vector regression machine (SVR) was used to learn the relationship between the characterized NLS-import receptor interaction and the corresponding nuclear import activity. Our experiments showed that nuclear import activity change due to NLS change could be accurately predicted by the NIpredict algorithm. Based on NIpredict, we developed a systematic framework to identify potential PTMbased nuclear import regulations for human and yeast nuclear proteins. Application of this approach has identified the potential nuclear import regulation mechanisms by phosphorylation of two nuclear proteins including SF1 and ORC6. Proteins 2014; 82:2783–2796. C 2014 Wiley Periodicals, Inc. V
Key words: post-translational modification; nuclear import regulation; protein sorting regulation; protein targeting; nuclear localization signals.
INTRODUCTION As the control center of the cell, the nucleus is separated from the cytoplasm by the nuclear envelope. While small nuclear proteins (5 0.5. This indicates that motif matches with disorder scores lower than 0.5 are less likely to be NLS. In the Yeast nuclear dataset, 117 nuclear proteins contain at least one matches, while 34 nuclear proteins contain a strong (NIpredict score >58) match in disorder region (disorder score >5 0.5). In the Human nuclear dataset, 339 nuclear proteins have at least one match, while 115 nuclear proteins has a strong (NIpredict score >58) match in disorder region (disorder score >5 0.5). In Supporting Information Tables S3 and S4, cNLS Mapper scores for matches were calculated using the activity-based profile in Supporting Information Table S2. Flanking residues adjacent to motif matches that are included in the position window of the activity-based profile (Supporting Information Table S2) were also included to calculate cNLS Mapper scores. The number of motif matches overlapped with annotated NLSs with scores >5 4 can be used to benchmark NIpredict and
2792
PROTEINS
cNLS Mapper, considering that a cutoff 4 on the NLS activity score was used to infer the basic level of NLS functionality in the work of Kosugi et al.10 For motif matches on the yeast nuclear proteins, all 8 motif matches that are overlapped with annotated NLSs have both NIpredict scores and cNLS Mapper scores greater or equal than 4. And 4 out of these 8 motif matches have score differences of less than or equal to 1 between NIpredict and cNLS Mapper. For motif matches on the human nuclear proteins, all 37 motif matches have NIpredict scores greater or equal than 4, while only 26 out of 37 motif matches have cNLS Mapper scores greater or equal than 4. And 10 out of those 26 motif matches have score differences of less or equal than 1 between NIpredict and cNLS Mapper. This result shows that for certain NLSs the predictions of NIpredict and cNLS Mapper are in close agreement while for certain NLSs their predictions are more different. The result also exposed the bias of cNLS Mapper as it failed to detect many annotated NLSs. Another approach to benchmark NIpredict and cNLS Mapper is comparing their scores of motif matches on non-nuclear proteins, which are shown in Supporting Information Table S5 (for Yeast non-nuclear dataset) and S6 (for Human non-nuclear dataset). To exclude the disorder factor, only motif matches located in disorder region (region with disorder score >5 0.5) were considered. In Supporting Information Table S5, 22 motif matches located in the disorder region of yeast nonnuclear proteins were found. Out of those 22 motif matches, 16 matches have NIpredict scores >54, while 10 matches have cNLS Mapper scores >54. In Supporting Information Table S6, 11 motif matches located in the disorder region of human non-nuclear proteins were found. Out of those 11 motif matches, 6 matches have NIpredict scores >54, while 2 matches have cNLS Mapper scores >54. The results in Supporting Information Table S5 and S6 show that cNLS Mapper gave better predictions on those motif matches than NIpredict. The reason that NIpredict over-estimated the NLS activities of motif matches that are unlikely to be NLS could be due to the bias introduced by the used structure template. The above benchmark results indicate that both NIpredict and cNLS Mapper have different biases in predicting NLS activities. NIpredict has higher recall performance but with lower precision while cNLS Mapper is more conservative in terms of its activity score prediction. For cNLS Mapper, the missed detection of many annotated NLSs may indicate that the additive rules incorporated in the activity-based profiles of cNLS Mapper may be insufficient to model the complicated relationship between NLSs and their corresponding import activity. For NIpredict, biased predictions may be introduced due to biased structure modeling. To identify potential PTM-based nuclear import regulations, we collected the experimentally verified PTM sites from DbPTM 3.072 that overlap the wildcard
Prediction of PTM-Based Nuclear Import Regulations
positions of the motif matches. The NLS activity scores of motif matches before and after PTM were predicted using NIpredict and are shown in Supporting Information Table S7. The predicted NLS activity scores of the motif matches increase, decrease, or remain roughly equal after phosphorylation or acetylation on the underlined residue(s). All possible combinations of the PTM sites within the predicted NLS were listed in Supporting Information Table S7, while some of the combinations may never happen during the life cycle of the protein. The motif matches of which the predicted NLS activity scores subject to apparent changes are likely to be the candidates of PTM-based nuclear import regulations. Assuming that the changes of real NLS activity scores for most of the NLSs whose functions are regulated by PTM are greater or equal than a threshold (e.g., 3), we defined confidence scores for nuclear import regulation to estimate the probability of nuclear import regulation given predicted NLS activity scores before and after PTM. The derivation was provided in the Supporting Information file. Case studies on potential PTM-based nuclear import regulations identified by NIpredict
Localization of nuclear proteins to the nucleus is the prerequisite for their participation in nuclear activities. Regulation of nuclear localization is a known mechanism to control and regulate other biological activities such as gene transcription and cell cycle progression.13 It is thus interesting to check what biological activities could be regulated by our identified potential PTM-based nuclear import regulations. By screening the candidates of potential PTM-based nuclear import regulations in Supporting Information Table S7, we identified two potential regulation mechanisms of biological activities based on NIpredict predictions and associated biological evidences. PKG inhibits spliceosome assembly by phosphorylation based regulation of SF1 nuclear import
A spliceosome is a large complex in the nucleus whose function is to remove introns from pre-mRNA (RNA splicing). Spliceosome assembly is thus a necessary event for RNA-splicing. SF1 and U2AF65 are both important components in spliceosome and their interaction is critical for spliceosome assembly. It was found that the Ser-20 phosphorylation mediated by cGMP-dependent protein kinase (PKG) on human SF1 (Q15637) inhibits its interaction with U2AF65, which leads to a block of spliceosome assembly.73 Wang et al.73 raised another possibility that the Ser20 phosphorylation may regulate localization of SF1 since Ser-20 is adjacent to a putative NLS, which is among our NIpredict predictions (Supporting Information Table S4). The motif match 13-PSKKRKRSR-21 of SF1 has a predicted NLS activity score and a disorder score of 8 and 0.65. The confidence score for NLS 0.99 and disorder score
higher than 0.5 implicate that this motif match is very likely to be a NLS. As shown in Supporting Information Table S7, the predicted NLS activity score of this putative NLS decreased from 8 to 4 after the Ser-20 phosphorylation. The significant reduction on the predicted NLS activity score of SF1 indicates the repressing effect of the Ser-20 phosphorylation on the nuclear import of SF1. This hypothesis is supported by the 1-tailed Student’s t-test that the significance of the score variation caused by phosphorylation at Ser-20 achieved a p-value of 6.3E-04, and the confidence score for nuclear import regulation is 0.64. This result is consistent with the biological role of PKG with regard to SF1, which is known to prevent spliceosome assembly but through repressing the nuclear activity of SF1 by the Ser-20 phosphorylation. Therefore, the PTM-based nuclear import regulation of SF1 mediated by PKG could be another mechanism to block spliceosome assembly. CDK regulates the nuclear import of ORC6 via phosphorylation
DNA replication occurs only in dividing eukaryotic cells, which must be tightly controlled to ensure that the genome is only replicated once. Previous study74 found that DNA re-replication is prevented by Cyclindependent kinases (CDK) in Saccharomyces cerevisiae: pre-replicative complex (pre-RC) assembly takes place in G1 phase when Cdc28 kinase activity is low and is blocked in the other phases through phosphorylation mediated by Cdc28 kinase on different proteins through multiple levels of mechanisms.74 In particular, it is known that CDK mediated phosphorylation on ORC6 (P38826) has an effect to prevent helicase from loading with unknown mechanisms.74–76 One possible mechanism is that CDK mediated phosphorylation on ORC6 blocks Cdt1 recruitment through inhibiting Cdt1 binding.77 We found that ORC6 contains a motif match 115PSPKKNKRS-123 which is covered by our NIpredict predictions (Supporting Information Table S3). The predicted NLS activity score and the disorder score of this motif match is 5 and 0.67, respectively. The confidence score for NLS 0.81 and disorder score higher than 0.5 implicate that this motif match is likely to be a NLS, in which Ser-116 is the phosphorylation site mediated by CDK.78 This result is consistent with the previous study.79 As shown in Supporting Information Table S7, the predicted NLS activity score of this motif match drops from 5 to 1 after the Ser-116 phosphorylation, which indicates that the Ser-116 phosphorylation inhibits the nuclear import of ORC6. This hypothesis is supported by the 1-tailed Student’s t-test that the significance of the score variation caused by phosphorylation at Ser-116 achieved a P-value of 1.71E-02, and the confidence score for nuclear import regulation is 0.62. The repressing effect of nuclear import is consistent with the PROTEINS
2793
J.R. Lin et al.
biological role of CDK with regard to ORC6, which is known to prevent helicase from loading through CDKmediated phosphorylation. Therefore, the PTM-based nuclear import regulation of ORC6 mediated by CDK could be another mechanism to prevent DNA re-replication. The above two hypotheses on PTM-based nuclear import regulation mechanisms were based on the prediction results of NIpredict and the known biological roles of PTM-mediating enzymes (PKG and CDK). We found that a common characteristic of these two proteins is that their participations in the nuclear activities are also controlled by other mechanisms in addition to the PTMregulated nuclear import process. Such kinds of multiple regulation mechanisms are common in biological systems to make the regulated activity tightly controlled, which is also reported in experimental studies.13 The first hypothesized regulation mechanism regulates RNA splicing while the second hypothesized regulation mechanism regulates DNA replication. This result corresponds to the previous study that the PTM-based nuclear import regulation was found to regulate other biological events in gene expression and cell proliferation.16 CONCLUSIONS In this study, we proposed a computational method, NIpredict, for predicting nuclear import activity and discovery of PTM-based nuclear import regulations. This approach is based on characterizing the interaction between NLS and the import receptor in terms of MIECs and learning the relationship between the characterized interaction energies and the corresponding nuclear import activity by Support Vector regression. The accuracy of NIpredict is demonstrated by its high performance in leave-one-out cross-validation on the major-site dataset and accurate prediction in the real cases. NIpredict was then used to systematically scan the Yeast and Human genome for identifying potential PTM-based nuclear import regulations. On the basis of NIpredict predictions and the known biological roles of the PTMs (or PTM-mediating enzymes), we identified the potential regulation mechanisms of two biological activities through the identified PTM-based nuclear import regulation. It should be noted that the scope of analysis in this study was limited by the NLS mutation templates due to limited experimental dataset. This approach can be applied to identify the more comprehensive list of PTM-based regulations of protein sub-cellular localization given more experimental datasets. A web server for predicting nuclear import activity given the NLS sequence is available at http://mleg.cse.sc.edu/NIpredict. ACKNOWLEDGEMENT The authors thank Professor Paul R. Housley of the Department of Pharmacology, Physiology, and Neuro-
2794
PROTEINS
science at the University of South Carolina for comments and suggestions. REFERENCES 1. Davis LI. The nuclear pore complex. Annu Rev Biochem 1995;64: 865–896. 2. Sorokin AV, Kim ER, Ovchinnikov LP. Nucleocytoplasmic transport of proteins. Biochemistry (Mosc) 2007;72:1439–1457. 3. Chahine MN, Pierce GN. Therapeutic targeting of nuclear protein import in pathological cell conditions. Pharmacol Rev 2009;61:358– 372. 4. McLane LM, Corbett AH. Nuclear localization signals and human disease. IUBMB Life 2009;61:697–706. 5. Hung MC, Link W. Protein localization in disease and therapy. J Cell Sci 2011;124(Part 20):3381–3392. 6. Kau TR, Way JC, Silver PA. Nuclear transport and cancer: from mechanism to intervention. Nat Rev Cancer 2004;4:106–117. 7. Nakai K, Horton P. PSORT: a program for detecting sorting signals in proteins and predicting their subcellular localization. Trends Biochem Sci 1999;24:34–36. 8. Cokol M, Nair R, Rost B. Finding nuclear localization signals. EMBO Rep 2000;1:411–415. 9. Nguyen Ba AN, Pogoutse A, Provart N, Moses AM. NLStradamus: a simple Hidden Markov Model for nuclear localization signal prediction. Bmc Bioinform 2009;10:202. 10. Kosugi S, Hasebe M, Tomita M, Yanagawa H. Systematic identification of cell cycle-dependent yeast nucleocytoplasmic shuttling proteins by prediction of composite motifs. Proc Natl Acad Sci USA 2009;106:10171–10176. 11. Mehdi AM, Sehgal MS, Kobe B, Bailey TL, Boden M. A probabilistic model of nuclear import of proteins. Bioinformatics 2011;27: 1239–1246. 12. Lin JR, Hu J. SeqNLS: nuclear localization signal prediction based on frequent pattern mining and linear motif scoring. PloS One 2013;8:e76864. 13. Kaffman A, O’Shea EK. Regulation of nuclear localization: a key to a door. Annu Rev Cell Dev Biol 1999;15:291–339. 14. Poon IK, Jans DA. Regulation of nuclear transport: central role in development and transformation? Traffic 2005;6:173–186. 15. Harreman MT, Kline TM, Milford HG, Harben MB, Hodel AE, Corbett AH. Regulation of nuclear import by phosphorylation adjacent to nuclear localization signals. J Biol Chem 2004;279:20613– 20621. 16. Nardozzi JD, Lott K, Cingolani G. Phosphorylation meets nuclear import: a review. Cell Commun Signal 2010;8:32. 17. Jans DA, Xiao CY, Lam MH. Nuclear targeting signal recognition: a key control point in nuclear transport? Bioessays 2000;22:532–544. 18. Liu Y, Peng L, Seto E, Huang S, Qiu Y. Modulation of histone deacetylase 6 (HDAC6) nuclear import and tubulin deacetylase activity through acetylation. J Biol Chem 2012;287:29168–29174. 19. Madison DL, Yaciuk P, Kwok RP, Lundblad JR. Acetylation of the adenovirus-transforming protein E1A determines nuclear localization by disrupting association with importin-alpha. J Biol Chem 2002;277:38755–38763. 20. di Bari MG, Ciuffini L, Mingardi M, Testi R, Soddu S, Barila D. cAbl acetylation by histone acetyltransferases regulates its nuclearcytoplasmic localization. EMBO Rep 2006;7:727–733. 21. Dietschy T, Shevelev I, Pena-Diaz J, Huhn D, Kuenzle S, Mak R, Miah MF, Hess D, Fey M, Hottiger MO, Janscak P, Stagljar I. p300mediated acetylation of the Rothmund-Thomson-syndrome gene product RECQL4 regulates its subcellular localization. J Cell Sci 2009;122(Part 8):1258–1267. 22. Shimazu T, Horinouchi S, Yoshida M. Multiple histone deacetylases and the CREB-binding protein regulate pre-mRNA 3’-end processing. J Biol Chem 2007;282:4470–4478.
Prediction of PTM-Based Nuclear Import Regulations
23. Li T, Diner BA, Chen J, Cristea IM. Acetylation modulates cellular distribution and DNA sensing ability of interferon-inducible protein IFI16. Proc Natl Acad Sci USA 2012;109:10558–10563. 24. Thevenet L, Mejean C, Moniot B, Bonneaud N, Galeotti N, Aldrian-Herrada G, Poulat F, Berta P, Benkirane M, BoizetBonhoure B. Regulation of human SRY subcellular distribution by its acetylation/deacetylation. EMBO J 2004;23:3336–3345. 25. Spilianakis C, Papamatheakis J, Kretsovali A. Acetylation by PCAF enhances CIITA nuclear accumulation and transactivation of major histocompatibility complex class II genes. Mol Cell Biol 2000;20: 8489–8498. 26. Timney BL, Tetenbaum-Novatt J, Agate DS, Williams R, Zhang W, Chait BT, Rout MP. Simple kinetic relationships and nonspecific competition govern nuclear import rates in vivo. J Cell Biol 2006; 175:579–593. 27. Hodel AE, Harreman MT, Pulliam KF, Harben ME, Holmes JS, Hodel MR, Berland KM, Corbett AH. Nuclear localization signal receptor affinity correlates with in vivo localization in Saccharomyces cerevisiae. J Biol Chem 2006;281:23545–23556. 28. Yang SN, Takeda AA, Fontes MR, Harris JM, Jans DA, Kobe B. Probing the specificity of binding to the major nuclear localization sequence-binding site of importin-alpha using oriented peptide library screening. J Biol Chem 2010;285:19935–19946. 29. Ghosh S, Vassilev AP, Zhang J, Zhao Y, DePamphilis ML. Assembly of the human origin recognition complex occurs through independent nuclear localization of its components. J Biol Chem 2011;286: 23831–23841. 30. Asada A, Saito T, Hisanaga S. Phosphorylation of p35 and p39 by Cdk5 determines the subcellular location of the holokinase in a phosphorylation-site-specific manner. J Cell Sci 2012;125(Part 14): 3421–3429. 31. Chung J, Khadka P, Chung IK. Nuclear import of hTERT requires a bipartite nuclear localization signal and Akt-mediated phosphorylation. J Cell Sci 2012;125(Part 11):2684–2697. 32. Teng AC, Al-Montashiri NA, Cheng BL, Lou P, Ozmizrak P, Chen HH, Stewart AF. Identification of a phosphorylation-dependent nuclear localization motif in interferon regulatory factor 2 binding protein 2. PLoS One 2011;6:e24100. 33. Fernandez-Garcia P, Pelaez R, Herrero P, Moreno F. Phosphorylation of yeast hexokinase 2 regulates its nucleocytoplasmic shuttling. J Biol Chem 2012;287:42151–42164. 34. Fulcher AJ, Roth DM, Fatima S, Alvisi G, Jans DA. The BRCA-1 binding protein BRAP2 is a novel, negative regulator of nuclear import of viral proteins, dependent on phosphorylation flanking the nuclear localization signal. FASEB J 2010;24:1454–1466. 35. Yu G, Yan T, Feng Y, Liu X, Xia Y, Luo H, Wang JZ, Wang X. Ser9 phosphorylation causes cytoplasmic detention of I2PP2A/SET in Alzheimer disease. Neurobiol Aging 2013;34:1748–1758. 36. Barrientes S, Cooke C, Goodrich DW. Glutamic acid mutagenesis of retinoblastoma protein phosphorylation sites has diverse effects on function. Oncogene 2000;19:562–570. 37. Davy CE, Jackson DJ, Wang Q, Raj K, Masterson PJ, Fenner NF, Southern S, Cuthill S, Millar JB, Doorbar J. Identification of a G(2) arrest domain in the E1 wedge E4 protein of human papillomavirus type 16. J Virol 2002;76:9806–9818. 38. Ku NO, Liao J, Omary MB. Phosphorylation of human keratin 18 serine 33 regulates binding to 14-3-3 proteins. EMBO J 1998;17: 1892–1906. 39. Moore JD, Yazgan O, Ataian Y, Krebs JE. Diverse roles for histone H2A modifications in DNA damage response pathways in yeast. Genetics 2007;176:15–25. 40. Yan J, Olsen JV, Park KS, Li W, Bildl W, Schulte U, Aldrich RW, Fakler B, Trimmer JS. Profiling the phospho-status of the BKCa channel alpha subunit in rat brain reveals unexpected patterns and complexity. Mol Cell Proteomics MCP 2008;7:2188–2198. 41. Caillaud A, Hovanessian AG, Levy DE, Marie IJ. Regulatory serine residues mediate phosphorylation-dependent and phosphorylation-
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
independent activation of interferon regulatory factor 7. J Biol Chem 2005;280:17671–17677. Eldar-Finkelman H, Argast GM, Foord O, Fischer EH, Krebs EG. Expression and characterization of glycogen synthase kinase-3 mutants and their effect on glycogen synthase activity in intact cells. Proc Natl Acad Sci USA 1996;93:10228–10233. Shao J, Prince T, Hartson SD, Matts RL. Phosphorylation of serine 13 is required for the proper function of the Hsp90 co-chaperone, Cdc37. J Biol Chem 2003;278:38117–38120. Hou T, Zhang W, Case DA, Wang W. Characterization of domainpeptide interaction interface: a case study on the amphiphysin-1 SH3 domain. J Mol Biol 2008;376:1201–1214. Hou T, Xu Z, Zhang W, McLaughlin WA, Case DA, Xu Y, Wang W. Characterization of domain-peptide interaction interface: a generic structure-based model to decipher the binding specificity of SH3 domains. Mol Cell Proteomics MCP 2009;8:639–649. Li N, Hou T, Ding B, Wang W. Characterization of PDZ domainpeptide interaction interface based on energetic patterns. Proteins 2011;79:3208–3220. Hou T, Li N, Li Y, Wang W. Characterization of domain-peptide interaction interface: prediction of SH3 domain-mediated proteinprotein interaction network in yeast by generic structure-based models. J Proteome Res 2012;11:2982–2995. Marfori M, Mynott A, Ellis JJ, Mehdi AM, Saunders NF, Curmi PM, Forwood JK, Boden M, Kobe B. Molecular basis for specificity of nuclear import and prediction of nuclear localization. Biochim Biophys Acta 2011;1813:1562–1577. Marfori M, Lonhienne TG, Forwood JK, Kobe B. Structural basis of high-affinity nuclear localization signal interactions with importinalpha. Traffic 2012;13:532–548. Hahn S, Maurer P, Caesar S, Schlenstedt G. Classical NLS proteins from Saccharomyces cerevisiae. J Mol Biol 2008;379:678– 694. Case DA, Cheatham TE, III, Simmerling CL, Wang J, Duke RE, Luo R, Walker RC, Zhang W, Merz KM, Roberts B, Hayik S, Roitberg A, Seabra G, Swails J, Goetz AW, Kolossvary I, Wong KF, Paesani F, Vanicek J, Wolf RM, Liu J, Wu X, Brozell SR, Steinbrecher T, Gohlke H, Cai Q, Ye X, Wang J, Hsieh M-J, Cui G, Roe DR, Mathews DH, Seetin MG, Salomon-Ferrer R, Sagui C, Babin V, Luchko T, Gusarov S, Kovalenko A, Kollman PA. AMBER 12: University of California, San Francisco; 2012. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE. The Protein Data Bank. Nucleic Acids Res 2000;28:235–242. Yang Y, Zhan J, Zhao H, Zhou Y. A new size-independent score for pairwise protein structure alignment and its application to structure classification and nucleic-acid binding prediction. Proteins 2012;80: 2080–2088. Krivov GG, Shapovalov MV, Dunbrack RL, Jr. Improved prediction of protein side-chain conformations with SCWRL4. Proteins 2009; 77:778–795. Duan Y, Wu C, Chowdhury S, Lee MC, Xiong G, Zhang W, Yang R, Cieplak P, Luo R, Lee T, Caldwell J, Wang J, Kollman P. A pointcharge force field for molecular mechanics simulations of proteins based on condensed-phase quantum mechanical calculations. J Comp Chem 2003;24:1999–2012. Onufriev A, Bashford D, Case DA. Modification of the generalized Born model suitable for macromolecules. J Phys Chem B 2000;104: 3712–3720. Onufriev A, Bashford D, Case DA. Exploring protein native states and large-scale conformational changes with a modified generalized born model. Proteins 2004;55:383–394. Fan RE, Chen PH, Lin CJ. Working set selection using second order information for training support vector machines. J Mach Learn Res 2005;6:1889–1918. .Activities at the Universal Protein Resource (UniProt). Nucleic acids research 2014;42(Database issue):D191-D198.
PROTEINS
2795
J.R. Lin et al.
60. Kosugi S, Hasebe M, Matsumura N, Takashima H, Miyamoto-Sato E, Tomita M, Yanagawa H. Six classes of nuclear localization signals specific to different binding grooves of importin alpha. J Biol Chem 2009;284:478–485. 61. Craft JW, Legge GB. An AMBER/DYANA/MOLMOL phosphorylated amino acid library set and incorporation into NMR structure calculations. J Biomol Nmr 2005;33:15–24. 62. Papamokos GV, Tziatzos G, Papageorgiou DG, Georgatos SD, Politou AS, Kaxiras E. Structural role of RKS motifs in chromatin interactions: A molecular dynamics study of HP1 bound to a variably modified histone tail. Biophys J 2012;102:1926–1933. 63. Huh WK, Falvo JV, Gerke LC, Carroll AS, Howson RW, Weissman JS, O’Shea EK. Global analysis of protein localization in budding yeast. Nature 2003;425:686–691. 64. Garg A, Bhasin M, Raghava GP. Support vector machine-based method for subcellular localization of human proteins using amino acid compositions, their order, and similarity search. J Biol Chem 2005;280:14427–14432. 65. Hodel MR, Corbett AH, Hodel AE. Dissection of a nuclear localization signal. J Biol Chem 2001;276:1317–1325. 66. Hou T, Zhang W, Wang J, Wang W. Predicting drug resistance of the HIV-1 protease using molecular interaction energy components. Proteins 2009;74:837–846. 67. Srinivasan J, Miller J, Kollman PA, Case DA. Continuum solvent studies of the stability of RNA hairpin loops and helices. J Biomol Struct Dyn 1998;16:671–682. 68. Kollman PA, Massova I, Reyes C, Kuhn B, Huo S, Chong L, Lee M, Lee T, Duan Y, Wang W, Donini O, Cieplak P, Srinivasan J, Case DA, Cheatham TE, III. Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models. Acc Chem Res 2000;33:889–897. 69. Alvisi G, Jans DA, Guo J, Pinna LA, Ripalti A. A protein kinase CK2 site flanking the nuclear targeting signal enhances nuclear transport of human cytomegalovirus ppUL44. Traffic 2005;6:1002– 1013.
2796
PROTEINS
70. Alvisi G, Marin O, Pari G, Mancini M, Avanzi S, Loregian A, Jans DA, Ripalti A. Multiple phosphorylation sites at the C-terminus regulate nuclear import of HCMV DNA polymerase processivity factor ppUL44. Virology 2011;417:259–267. 71. Dosztanyi Z, Csizmok V, Tompa P, Simon I. IUPred: web server for the prediction of intrinsically unstructured regions of proteins based on estimated energy content. Bioinformatics 2005;21:3433–3434. 72. Lu CT, Huang KY, Su MG, Lee TY, Bretana NA, Chang WC, Chen YJ, Huang HD. DbPTM 3.0: an informative resource for investigating substrate site specificity and functional association of protein post-translational modifications. Nucleic Acids Res 2013;41(Database issue):D295–D305. 73. Wang X, Bruderer S, Rafi Z, Xue J, Milburn PJ, Kramer A, Robinson PJ. Phosphorylation of splicing factor SF1 on Ser20 by cGMP-dependent protein kinase regulates spliceosome assembly. EMBO J 1999;18:4549–4559. 74. Nguyen VQ, Co C, Li JJ. Cyclin-dependent kinases prevent DNA rereplication through multiple mechanisms. Nature 2001;411:1068– 1073. 75. Green BM, Morreale RJ, Ozaydin B, Derisi JL, Li JJ. Genome-wide mapping of DNA synthesis in Saccharomyces cerevisiae reveals that mechanisms preventing reinitiation of DNA replication are not redundant. Mol Biol Cell 2006;17:2401–2414. 76. Tanny RE, MacAlpine DM, Blitzblau HG, Bell SP. Genome-wide analysis of re-replication reveals inhibitory controls that target multiple stages of replication initiation. Mol Biol Cell 2006;17:2415– 2423. 77. Chen S, Bell SP. CDK prevents Mcm2–7 helicase loading by inhibiting Cdt1 interaction with Orc6. Genes Dev 2011;25:363–372. 78. Holt LJ, Tuch BB, Villen J, Johnson AD, Gygi SP, Morgan DO. Global analysis of Cdk1 substrate phosphorylation sites provides insights into evolution. Science 2009;325:1682–1686. 79. Li JJ, Herskowitz I. Isolation of ORC6, a component of the yeast origin recognition complex by a one-hybrid system. Science 1993; 262:1870–1874.
Computational identification of posttranslational modification-based nuclear import regulations by characterizing nuclear localization signal-import receptor interaction Jhih-Rong Lin, Zhonghao Liu, and Jianjun Hu* Department of Computer Science and Engineering, University of South Carolina, Columbia, South Carolina 29208
ABSTRACT The binding affinity between a nuclear localization signal (NLS) and its import receptor is closely related to corresponding nuclear import activity. PTM-based modulation of the NLS binding affinity to the import receptor is one of the most understood mechanisms to regulate nuclear import of proteins. However, identification of such regulation mechanisms is challenging due to the difficulty of assessing the impact of PTM on corresponding nuclear import activities. In this study we proposed NIpredict, an effective algorithm to predict nuclear import activity given its NLS, in which molecular interaction energy components (MIECs) were used to characterize the NLS-import receptor interaction, and the support vector regression machine (SVR) was used to learn the relationship between the characterized NLS-import receptor interaction and the corresponding nuclear import activity. Our experiments showed that nuclear import activity change due to NLS change could be accurately predicted by the NIpredict algorithm. Based on NIpredict, we developed a systematic framework to identify potential PTMbased nuclear import regulations for human and yeast nuclear proteins. Application of this approach has identified the potential nuclear import regulation mechanisms by phosphorylation of two nuclear proteins including SF1 and ORC6. Proteins 2014; 82:2783–2796. C 2014 Wiley Periodicals, Inc. V
Key words: post-translational modification; nuclear import regulation; protein sorting regulation; protein targeting; nuclear localization signals.
INTRODUCTION As the control center of the cell, the nucleus is separated from the cytoplasm by the nuclear envelope. While small nuclear proteins (5 0.5. This indicates that motif matches with disorder scores lower than 0.5 are less likely to be NLS. In the Yeast nuclear dataset, 117 nuclear proteins contain at least one matches, while 34 nuclear proteins contain a strong (NIpredict score >58) match in disorder region (disorder score >5 0.5). In the Human nuclear dataset, 339 nuclear proteins have at least one match, while 115 nuclear proteins has a strong (NIpredict score >58) match in disorder region (disorder score >5 0.5). In Supporting Information Tables S3 and S4, cNLS Mapper scores for matches were calculated using the activity-based profile in Supporting Information Table S2. Flanking residues adjacent to motif matches that are included in the position window of the activity-based profile (Supporting Information Table S2) were also included to calculate cNLS Mapper scores. The number of motif matches overlapped with annotated NLSs with scores >5 4 can be used to benchmark NIpredict and
2792
PROTEINS
cNLS Mapper, considering that a cutoff 4 on the NLS activity score was used to infer the basic level of NLS functionality in the work of Kosugi et al.10 For motif matches on the yeast nuclear proteins, all 8 motif matches that are overlapped with annotated NLSs have both NIpredict scores and cNLS Mapper scores greater or equal than 4. And 4 out of these 8 motif matches have score differences of less than or equal to 1 between NIpredict and cNLS Mapper. For motif matches on the human nuclear proteins, all 37 motif matches have NIpredict scores greater or equal than 4, while only 26 out of 37 motif matches have cNLS Mapper scores greater or equal than 4. And 10 out of those 26 motif matches have score differences of less or equal than 1 between NIpredict and cNLS Mapper. This result shows that for certain NLSs the predictions of NIpredict and cNLS Mapper are in close agreement while for certain NLSs their predictions are more different. The result also exposed the bias of cNLS Mapper as it failed to detect many annotated NLSs. Another approach to benchmark NIpredict and cNLS Mapper is comparing their scores of motif matches on non-nuclear proteins, which are shown in Supporting Information Table S5 (for Yeast non-nuclear dataset) and S6 (for Human non-nuclear dataset). To exclude the disorder factor, only motif matches located in disorder region (region with disorder score >5 0.5) were considered. In Supporting Information Table S5, 22 motif matches located in the disorder region of yeast nonnuclear proteins were found. Out of those 22 motif matches, 16 matches have NIpredict scores >54, while 10 matches have cNLS Mapper scores >54. In Supporting Information Table S6, 11 motif matches located in the disorder region of human non-nuclear proteins were found. Out of those 11 motif matches, 6 matches have NIpredict scores >54, while 2 matches have cNLS Mapper scores >54. The results in Supporting Information Table S5 and S6 show that cNLS Mapper gave better predictions on those motif matches than NIpredict. The reason that NIpredict over-estimated the NLS activities of motif matches that are unlikely to be NLS could be due to the bias introduced by the used structure template. The above benchmark results indicate that both NIpredict and cNLS Mapper have different biases in predicting NLS activities. NIpredict has higher recall performance but with lower precision while cNLS Mapper is more conservative in terms of its activity score prediction. For cNLS Mapper, the missed detection of many annotated NLSs may indicate that the additive rules incorporated in the activity-based profiles of cNLS Mapper may be insufficient to model the complicated relationship between NLSs and their corresponding import activity. For NIpredict, biased predictions may be introduced due to biased structure modeling. To identify potential PTM-based nuclear import regulations, we collected the experimentally verified PTM sites from DbPTM 3.072 that overlap the wildcard
Prediction of PTM-Based Nuclear Import Regulations
positions of the motif matches. The NLS activity scores of motif matches before and after PTM were predicted using NIpredict and are shown in Supporting Information Table S7. The predicted NLS activity scores of the motif matches increase, decrease, or remain roughly equal after phosphorylation or acetylation on the underlined residue(s). All possible combinations of the PTM sites within the predicted NLS were listed in Supporting Information Table S7, while some of the combinations may never happen during the life cycle of the protein. The motif matches of which the predicted NLS activity scores subject to apparent changes are likely to be the candidates of PTM-based nuclear import regulations. Assuming that the changes of real NLS activity scores for most of the NLSs whose functions are regulated by PTM are greater or equal than a threshold (e.g., 3), we defined confidence scores for nuclear import regulation to estimate the probability of nuclear import regulation given predicted NLS activity scores before and after PTM. The derivation was provided in the Supporting Information file. Case studies on potential PTM-based nuclear import regulations identified by NIpredict
Localization of nuclear proteins to the nucleus is the prerequisite for their participation in nuclear activities. Regulation of nuclear localization is a known mechanism to control and regulate other biological activities such as gene transcription and cell cycle progression.13 It is thus interesting to check what biological activities could be regulated by our identified potential PTM-based nuclear import regulations. By screening the candidates of potential PTM-based nuclear import regulations in Supporting Information Table S7, we identified two potential regulation mechanisms of biological activities based on NIpredict predictions and associated biological evidences. PKG inhibits spliceosome assembly by phosphorylation based regulation of SF1 nuclear import
A spliceosome is a large complex in the nucleus whose function is to remove introns from pre-mRNA (RNA splicing). Spliceosome assembly is thus a necessary event for RNA-splicing. SF1 and U2AF65 are both important components in spliceosome and their interaction is critical for spliceosome assembly. It was found that the Ser-20 phosphorylation mediated by cGMP-dependent protein kinase (PKG) on human SF1 (Q15637) inhibits its interaction with U2AF65, which leads to a block of spliceosome assembly.73 Wang et al.73 raised another possibility that the Ser20 phosphorylation may regulate localization of SF1 since Ser-20 is adjacent to a putative NLS, which is among our NIpredict predictions (Supporting Information Table S4). The motif match 13-PSKKRKRSR-21 of SF1 has a predicted NLS activity score and a disorder score of 8 and 0.65. The confidence score for NLS 0.99 and disorder score
higher than 0.5 implicate that this motif match is very likely to be a NLS. As shown in Supporting Information Table S7, the predicted NLS activity score of this putative NLS decreased from 8 to 4 after the Ser-20 phosphorylation. The significant reduction on the predicted NLS activity score of SF1 indicates the repressing effect of the Ser-20 phosphorylation on the nuclear import of SF1. This hypothesis is supported by the 1-tailed Student’s t-test that the significance of the score variation caused by phosphorylation at Ser-20 achieved a p-value of 6.3E-04, and the confidence score for nuclear import regulation is 0.64. This result is consistent with the biological role of PKG with regard to SF1, which is known to prevent spliceosome assembly but through repressing the nuclear activity of SF1 by the Ser-20 phosphorylation. Therefore, the PTM-based nuclear import regulation of SF1 mediated by PKG could be another mechanism to block spliceosome assembly. CDK regulates the nuclear import of ORC6 via phosphorylation
DNA replication occurs only in dividing eukaryotic cells, which must be tightly controlled to ensure that the genome is only replicated once. Previous study74 found that DNA re-replication is prevented by Cyclindependent kinases (CDK) in Saccharomyces cerevisiae: pre-replicative complex (pre-RC) assembly takes place in G1 phase when Cdc28 kinase activity is low and is blocked in the other phases through phosphorylation mediated by Cdc28 kinase on different proteins through multiple levels of mechanisms.74 In particular, it is known that CDK mediated phosphorylation on ORC6 (P38826) has an effect to prevent helicase from loading with unknown mechanisms.74–76 One possible mechanism is that CDK mediated phosphorylation on ORC6 blocks Cdt1 recruitment through inhibiting Cdt1 binding.77 We found that ORC6 contains a motif match 115PSPKKNKRS-123 which is covered by our NIpredict predictions (Supporting Information Table S3). The predicted NLS activity score and the disorder score of this motif match is 5 and 0.67, respectively. The confidence score for NLS 0.81 and disorder score higher than 0.5 implicate that this motif match is likely to be a NLS, in which Ser-116 is the phosphorylation site mediated by CDK.78 This result is consistent with the previous study.79 As shown in Supporting Information Table S7, the predicted NLS activity score of this motif match drops from 5 to 1 after the Ser-116 phosphorylation, which indicates that the Ser-116 phosphorylation inhibits the nuclear import of ORC6. This hypothesis is supported by the 1-tailed Student’s t-test that the significance of the score variation caused by phosphorylation at Ser-116 achieved a P-value of 1.71E-02, and the confidence score for nuclear import regulation is 0.62. The repressing effect of nuclear import is consistent with the PROTEINS
2793
J.R. Lin et al.
biological role of CDK with regard to ORC6, which is known to prevent helicase from loading through CDKmediated phosphorylation. Therefore, the PTM-based nuclear import regulation of ORC6 mediated by CDK could be another mechanism to prevent DNA re-replication. The above two hypotheses on PTM-based nuclear import regulation mechanisms were based on the prediction results of NIpredict and the known biological roles of PTM-mediating enzymes (PKG and CDK). We found that a common characteristic of these two proteins is that their participations in the nuclear activities are also controlled by other mechanisms in addition to the PTMregulated nuclear import process. Such kinds of multiple regulation mechanisms are common in biological systems to make the regulated activity tightly controlled, which is also reported in experimental studies.13 The first hypothesized regulation mechanism regulates RNA splicing while the second hypothesized regulation mechanism regulates DNA replication. This result corresponds to the previous study that the PTM-based nuclear import regulation was found to regulate other biological events in gene expression and cell proliferation.16 CONCLUSIONS In this study, we proposed a computational method, NIpredict, for predicting nuclear import activity and discovery of PTM-based nuclear import regulations. This approach is based on characterizing the interaction between NLS and the import receptor in terms of MIECs and learning the relationship between the characterized interaction energies and the corresponding nuclear import activity by Support Vector regression. The accuracy of NIpredict is demonstrated by its high performance in leave-one-out cross-validation on the major-site dataset and accurate prediction in the real cases. NIpredict was then used to systematically scan the Yeast and Human genome for identifying potential PTM-based nuclear import regulations. On the basis of NIpredict predictions and the known biological roles of the PTMs (or PTM-mediating enzymes), we identified the potential regulation mechanisms of two biological activities through the identified PTM-based nuclear import regulation. It should be noted that the scope of analysis in this study was limited by the NLS mutation templates due to limited experimental dataset. This approach can be applied to identify the more comprehensive list of PTM-based regulations of protein sub-cellular localization given more experimental datasets. A web server for predicting nuclear import activity given the NLS sequence is available at http://mleg.cse.sc.edu/NIpredict. ACKNOWLEDGEMENT The authors thank Professor Paul R. Housley of the Department of Pharmacology, Physiology, and Neuro-
2794
PROTEINS
science at the University of South Carolina for comments and suggestions. REFERENCES 1. Davis LI. The nuclear pore complex. Annu Rev Biochem 1995;64: 865–896. 2. Sorokin AV, Kim ER, Ovchinnikov LP. Nucleocytoplasmic transport of proteins. Biochemistry (Mosc) 2007;72:1439–1457. 3. Chahine MN, Pierce GN. Therapeutic targeting of nuclear protein import in pathological cell conditions. Pharmacol Rev 2009;61:358– 372. 4. McLane LM, Corbett AH. Nuclear localization signals and human disease. IUBMB Life 2009;61:697–706. 5. Hung MC, Link W. Protein localization in disease and therapy. J Cell Sci 2011;124(Part 20):3381–3392. 6. Kau TR, Way JC, Silver PA. Nuclear transport and cancer: from mechanism to intervention. Nat Rev Cancer 2004;4:106–117. 7. Nakai K, Horton P. PSORT: a program for detecting sorting signals in proteins and predicting their subcellular localization. Trends Biochem Sci 1999;24:34–36. 8. Cokol M, Nair R, Rost B. Finding nuclear localization signals. EMBO Rep 2000;1:411–415. 9. Nguyen Ba AN, Pogoutse A, Provart N, Moses AM. NLStradamus: a simple Hidden Markov Model for nuclear localization signal prediction. Bmc Bioinform 2009;10:202. 10. Kosugi S, Hasebe M, Tomita M, Yanagawa H. Systematic identification of cell cycle-dependent yeast nucleocytoplasmic shuttling proteins by prediction of composite motifs. Proc Natl Acad Sci USA 2009;106:10171–10176. 11. Mehdi AM, Sehgal MS, Kobe B, Bailey TL, Boden M. A probabilistic model of nuclear import of proteins. Bioinformatics 2011;27: 1239–1246. 12. Lin JR, Hu J. SeqNLS: nuclear localization signal prediction based on frequent pattern mining and linear motif scoring. PloS One 2013;8:e76864. 13. Kaffman A, O’Shea EK. Regulation of nuclear localization: a key to a door. Annu Rev Cell Dev Biol 1999;15:291–339. 14. Poon IK, Jans DA. Regulation of nuclear transport: central role in development and transformation? Traffic 2005;6:173–186. 15. Harreman MT, Kline TM, Milford HG, Harben MB, Hodel AE, Corbett AH. Regulation of nuclear import by phosphorylation adjacent to nuclear localization signals. J Biol Chem 2004;279:20613– 20621. 16. Nardozzi JD, Lott K, Cingolani G. Phosphorylation meets nuclear import: a review. Cell Commun Signal 2010;8:32. 17. Jans DA, Xiao CY, Lam MH. Nuclear targeting signal recognition: a key control point in nuclear transport? Bioessays 2000;22:532–544. 18. Liu Y, Peng L, Seto E, Huang S, Qiu Y. Modulation of histone deacetylase 6 (HDAC6) nuclear import and tubulin deacetylase activity through acetylation. J Biol Chem 2012;287:29168–29174. 19. Madison DL, Yaciuk P, Kwok RP, Lundblad JR. Acetylation of the adenovirus-transforming protein E1A determines nuclear localization by disrupting association with importin-alpha. J Biol Chem 2002;277:38755–38763. 20. di Bari MG, Ciuffini L, Mingardi M, Testi R, Soddu S, Barila D. cAbl acetylation by histone acetyltransferases regulates its nuclearcytoplasmic localization. EMBO Rep 2006;7:727–733. 21. Dietschy T, Shevelev I, Pena-Diaz J, Huhn D, Kuenzle S, Mak R, Miah MF, Hess D, Fey M, Hottiger MO, Janscak P, Stagljar I. p300mediated acetylation of the Rothmund-Thomson-syndrome gene product RECQL4 regulates its subcellular localization. J Cell Sci 2009;122(Part 8):1258–1267. 22. Shimazu T, Horinouchi S, Yoshida M. Multiple histone deacetylases and the CREB-binding protein regulate pre-mRNA 3’-end processing. J Biol Chem 2007;282:4470–4478.
Prediction of PTM-Based Nuclear Import Regulations
23. Li T, Diner BA, Chen J, Cristea IM. Acetylation modulates cellular distribution and DNA sensing ability of interferon-inducible protein IFI16. Proc Natl Acad Sci USA 2012;109:10558–10563. 24. Thevenet L, Mejean C, Moniot B, Bonneaud N, Galeotti N, Aldrian-Herrada G, Poulat F, Berta P, Benkirane M, BoizetBonhoure B. Regulation of human SRY subcellular distribution by its acetylation/deacetylation. EMBO J 2004;23:3336–3345. 25. Spilianakis C, Papamatheakis J, Kretsovali A. Acetylation by PCAF enhances CIITA nuclear accumulation and transactivation of major histocompatibility complex class II genes. Mol Cell Biol 2000;20: 8489–8498. 26. Timney BL, Tetenbaum-Novatt J, Agate DS, Williams R, Zhang W, Chait BT, Rout MP. Simple kinetic relationships and nonspecific competition govern nuclear import rates in vivo. J Cell Biol 2006; 175:579–593. 27. Hodel AE, Harreman MT, Pulliam KF, Harben ME, Holmes JS, Hodel MR, Berland KM, Corbett AH. Nuclear localization signal receptor affinity correlates with in vivo localization in Saccharomyces cerevisiae. J Biol Chem 2006;281:23545–23556. 28. Yang SN, Takeda AA, Fontes MR, Harris JM, Jans DA, Kobe B. Probing the specificity of binding to the major nuclear localization sequence-binding site of importin-alpha using oriented peptide library screening. J Biol Chem 2010;285:19935–19946. 29. Ghosh S, Vassilev AP, Zhang J, Zhao Y, DePamphilis ML. Assembly of the human origin recognition complex occurs through independent nuclear localization of its components. J Biol Chem 2011;286: 23831–23841. 30. Asada A, Saito T, Hisanaga S. Phosphorylation of p35 and p39 by Cdk5 determines the subcellular location of the holokinase in a phosphorylation-site-specific manner. J Cell Sci 2012;125(Part 14): 3421–3429. 31. Chung J, Khadka P, Chung IK. Nuclear import of hTERT requires a bipartite nuclear localization signal and Akt-mediated phosphorylation. J Cell Sci 2012;125(Part 11):2684–2697. 32. Teng AC, Al-Montashiri NA, Cheng BL, Lou P, Ozmizrak P, Chen HH, Stewart AF. Identification of a phosphorylation-dependent nuclear localization motif in interferon regulatory factor 2 binding protein 2. PLoS One 2011;6:e24100. 33. Fernandez-Garcia P, Pelaez R, Herrero P, Moreno F. Phosphorylation of yeast hexokinase 2 regulates its nucleocytoplasmic shuttling. J Biol Chem 2012;287:42151–42164. 34. Fulcher AJ, Roth DM, Fatima S, Alvisi G, Jans DA. The BRCA-1 binding protein BRAP2 is a novel, negative regulator of nuclear import of viral proteins, dependent on phosphorylation flanking the nuclear localization signal. FASEB J 2010;24:1454–1466. 35. Yu G, Yan T, Feng Y, Liu X, Xia Y, Luo H, Wang JZ, Wang X. Ser9 phosphorylation causes cytoplasmic detention of I2PP2A/SET in Alzheimer disease. Neurobiol Aging 2013;34:1748–1758. 36. Barrientes S, Cooke C, Goodrich DW. Glutamic acid mutagenesis of retinoblastoma protein phosphorylation sites has diverse effects on function. Oncogene 2000;19:562–570. 37. Davy CE, Jackson DJ, Wang Q, Raj K, Masterson PJ, Fenner NF, Southern S, Cuthill S, Millar JB, Doorbar J. Identification of a G(2) arrest domain in the E1 wedge E4 protein of human papillomavirus type 16. J Virol 2002;76:9806–9818. 38. Ku NO, Liao J, Omary MB. Phosphorylation of human keratin 18 serine 33 regulates binding to 14-3-3 proteins. EMBO J 1998;17: 1892–1906. 39. Moore JD, Yazgan O, Ataian Y, Krebs JE. Diverse roles for histone H2A modifications in DNA damage response pathways in yeast. Genetics 2007;176:15–25. 40. Yan J, Olsen JV, Park KS, Li W, Bildl W, Schulte U, Aldrich RW, Fakler B, Trimmer JS. Profiling the phospho-status of the BKCa channel alpha subunit in rat brain reveals unexpected patterns and complexity. Mol Cell Proteomics MCP 2008;7:2188–2198. 41. Caillaud A, Hovanessian AG, Levy DE, Marie IJ. Regulatory serine residues mediate phosphorylation-dependent and phosphorylation-
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
independent activation of interferon regulatory factor 7. J Biol Chem 2005;280:17671–17677. Eldar-Finkelman H, Argast GM, Foord O, Fischer EH, Krebs EG. Expression and characterization of glycogen synthase kinase-3 mutants and their effect on glycogen synthase activity in intact cells. Proc Natl Acad Sci USA 1996;93:10228–10233. Shao J, Prince T, Hartson SD, Matts RL. Phosphorylation of serine 13 is required for the proper function of the Hsp90 co-chaperone, Cdc37. J Biol Chem 2003;278:38117–38120. Hou T, Zhang W, Case DA, Wang W. Characterization of domainpeptide interaction interface: a case study on the amphiphysin-1 SH3 domain. J Mol Biol 2008;376:1201–1214. Hou T, Xu Z, Zhang W, McLaughlin WA, Case DA, Xu Y, Wang W. Characterization of domain-peptide interaction interface: a generic structure-based model to decipher the binding specificity of SH3 domains. Mol Cell Proteomics MCP 2009;8:639–649. Li N, Hou T, Ding B, Wang W. Characterization of PDZ domainpeptide interaction interface based on energetic patterns. Proteins 2011;79:3208–3220. Hou T, Li N, Li Y, Wang W. Characterization of domain-peptide interaction interface: prediction of SH3 domain-mediated proteinprotein interaction network in yeast by generic structure-based models. J Proteome Res 2012;11:2982–2995. Marfori M, Mynott A, Ellis JJ, Mehdi AM, Saunders NF, Curmi PM, Forwood JK, Boden M, Kobe B. Molecular basis for specificity of nuclear import and prediction of nuclear localization. Biochim Biophys Acta 2011;1813:1562–1577. Marfori M, Lonhienne TG, Forwood JK, Kobe B. Structural basis of high-affinity nuclear localization signal interactions with importinalpha. Traffic 2012;13:532–548. Hahn S, Maurer P, Caesar S, Schlenstedt G. Classical NLS proteins from Saccharomyces cerevisiae. J Mol Biol 2008;379:678– 694. Case DA, Cheatham TE, III, Simmerling CL, Wang J, Duke RE, Luo R, Walker RC, Zhang W, Merz KM, Roberts B, Hayik S, Roitberg A, Seabra G, Swails J, Goetz AW, Kolossvary I, Wong KF, Paesani F, Vanicek J, Wolf RM, Liu J, Wu X, Brozell SR, Steinbrecher T, Gohlke H, Cai Q, Ye X, Wang J, Hsieh M-J, Cui G, Roe DR, Mathews DH, Seetin MG, Salomon-Ferrer R, Sagui C, Babin V, Luchko T, Gusarov S, Kovalenko A, Kollman PA. AMBER 12: University of California, San Francisco; 2012. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE. The Protein Data Bank. Nucleic Acids Res 2000;28:235–242. Yang Y, Zhan J, Zhao H, Zhou Y. A new size-independent score for pairwise protein structure alignment and its application to structure classification and nucleic-acid binding prediction. Proteins 2012;80: 2080–2088. Krivov GG, Shapovalov MV, Dunbrack RL, Jr. Improved prediction of protein side-chain conformations with SCWRL4. Proteins 2009; 77:778–795. Duan Y, Wu C, Chowdhury S, Lee MC, Xiong G, Zhang W, Yang R, Cieplak P, Luo R, Lee T, Caldwell J, Wang J, Kollman P. A pointcharge force field for molecular mechanics simulations of proteins based on condensed-phase quantum mechanical calculations. J Comp Chem 2003;24:1999–2012. Onufriev A, Bashford D, Case DA. Modification of the generalized Born model suitable for macromolecules. J Phys Chem B 2000;104: 3712–3720. Onufriev A, Bashford D, Case DA. Exploring protein native states and large-scale conformational changes with a modified generalized born model. Proteins 2004;55:383–394. Fan RE, Chen PH, Lin CJ. Working set selection using second order information for training support vector machines. J Mach Learn Res 2005;6:1889–1918. .Activities at the Universal Protein Resource (UniProt). Nucleic acids research 2014;42(Database issue):D191-D198.
PROTEINS
2795
J.R. Lin et al.
60. Kosugi S, Hasebe M, Matsumura N, Takashima H, Miyamoto-Sato E, Tomita M, Yanagawa H. Six classes of nuclear localization signals specific to different binding grooves of importin alpha. J Biol Chem 2009;284:478–485. 61. Craft JW, Legge GB. An AMBER/DYANA/MOLMOL phosphorylated amino acid library set and incorporation into NMR structure calculations. J Biomol Nmr 2005;33:15–24. 62. Papamokos GV, Tziatzos G, Papageorgiou DG, Georgatos SD, Politou AS, Kaxiras E. Structural role of RKS motifs in chromatin interactions: A molecular dynamics study of HP1 bound to a variably modified histone tail. Biophys J 2012;102:1926–1933. 63. Huh WK, Falvo JV, Gerke LC, Carroll AS, Howson RW, Weissman JS, O’Shea EK. Global analysis of protein localization in budding yeast. Nature 2003;425:686–691. 64. Garg A, Bhasin M, Raghava GP. Support vector machine-based method for subcellular localization of human proteins using amino acid compositions, their order, and similarity search. J Biol Chem 2005;280:14427–14432. 65. Hodel MR, Corbett AH, Hodel AE. Dissection of a nuclear localization signal. J Biol Chem 2001;276:1317–1325. 66. Hou T, Zhang W, Wang J, Wang W. Predicting drug resistance of the HIV-1 protease using molecular interaction energy components. Proteins 2009;74:837–846. 67. Srinivasan J, Miller J, Kollman PA, Case DA. Continuum solvent studies of the stability of RNA hairpin loops and helices. J Biomol Struct Dyn 1998;16:671–682. 68. Kollman PA, Massova I, Reyes C, Kuhn B, Huo S, Chong L, Lee M, Lee T, Duan Y, Wang W, Donini O, Cieplak P, Srinivasan J, Case DA, Cheatham TE, III. Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models. Acc Chem Res 2000;33:889–897. 69. Alvisi G, Jans DA, Guo J, Pinna LA, Ripalti A. A protein kinase CK2 site flanking the nuclear targeting signal enhances nuclear transport of human cytomegalovirus ppUL44. Traffic 2005;6:1002– 1013.
2796
PROTEINS
70. Alvisi G, Marin O, Pari G, Mancini M, Avanzi S, Loregian A, Jans DA, Ripalti A. Multiple phosphorylation sites at the C-terminus regulate nuclear import of HCMV DNA polymerase processivity factor ppUL44. Virology 2011;417:259–267. 71. Dosztanyi Z, Csizmok V, Tompa P, Simon I. IUPred: web server for the prediction of intrinsically unstructured regions of proteins based on estimated energy content. Bioinformatics 2005;21:3433–3434. 72. Lu CT, Huang KY, Su MG, Lee TY, Bretana NA, Chang WC, Chen YJ, Huang HD. DbPTM 3.0: an informative resource for investigating substrate site specificity and functional association of protein post-translational modifications. Nucleic Acids Res 2013;41(Database issue):D295–D305. 73. Wang X, Bruderer S, Rafi Z, Xue J, Milburn PJ, Kramer A, Robinson PJ. Phosphorylation of splicing factor SF1 on Ser20 by cGMP-dependent protein kinase regulates spliceosome assembly. EMBO J 1999;18:4549–4559. 74. Nguyen VQ, Co C, Li JJ. Cyclin-dependent kinases prevent DNA rereplication through multiple mechanisms. Nature 2001;411:1068– 1073. 75. Green BM, Morreale RJ, Ozaydin B, Derisi JL, Li JJ. Genome-wide mapping of DNA synthesis in Saccharomyces cerevisiae reveals that mechanisms preventing reinitiation of DNA replication are not redundant. Mol Biol Cell 2006;17:2401–2414. 76. Tanny RE, MacAlpine DM, Blitzblau HG, Bell SP. Genome-wide analysis of re-replication reveals inhibitory controls that target multiple stages of replication initiation. Mol Biol Cell 2006;17:2415– 2423. 77. Chen S, Bell SP. CDK prevents Mcm2–7 helicase loading by inhibiting Cdt1 interaction with Orc6. Genes Dev 2011;25:363–372. 78. Holt LJ, Tuch BB, Villen J, Johnson AD, Gygi SP, Morgan DO. Global analysis of Cdk1 substrate phosphorylation sites provides insights into evolution. Science 2009;325:1682–1686. 79. Li JJ, Herskowitz I. Isolation of ORC6, a component of the yeast origin recognition complex by a one-hybrid system. Science 1993; 262:1870–1874.
