Phosphorylated protein chip combined with artificial intelligence tools for precise drug screening

Abstract

We have developed a protein array system, named "Phospho-Totum", which reproduces the phosphorylation state of a sample on the array. The protein array contains 1471 proteins from 273 known signaling pathways. According to the activation degrees of tyrosine kinases in the sample, the corresponding groups of substrate proteins on the array are phosphorylated under the same conditions. In addition to measuring the phosphorylation levels of the 1471 substrates, we have developed and performed the artificial intelligence-assisted tools to further characterize the phosphorylation state and estimate pathway activation, tyrosine kinase activation, and a list of kinase inhibitors that produce phosphorylation states similar to that of the sample. The Phospho-Totum system, which seamlessly links and interrogates the measurements and analyses, has the potential to not only elucidate pathophysiological mechanisms in diseases by reproducing the phosphorylation state of samples, but also be useful for drug discovery, particularly for screening targeted kinases for potential drug kinase inhibitors.

Keywords

Phospho-Totum / protein array / signal transduction pathways / artificial intelligence tools / drug screening

Fundings

This study was supported by the State Key Program of National Natural Science Foundation of China (Grant No. 82230114 to F.H.) and the National Key Research and Development Program of China (Grant No. 2022YFE0104800 to F.H.).

Acknowledgments

We thank Mr. Takahiro Ohshima (Infocom Corporation) for the development of analysis software and Mr. Ryuki Kudo and Mr. Atsushi Kawasaki for the development of visualization software.

References

[1] Levitzki A, Gazit A. Tyrosine kinase inhibition: an approach to drug development[J]. Science, 1995, 267(5205): 1782-1788. doi: 10.1126/science.7892601

[2] Zhang J, Yang PL, Gray NS. Targeting cancer with small molecule kinase inhibitors[J]. Nat Rev Cancer, 2009, 9(1): 28-39. doi: 10.1038/nrc2559

[3] Steen H, Jebanathirajah JA, Rush J, et al. Phosphorylation analysis by mass spectrometry: myths, facts, and the consequences for qualitative and quantitative measurements[J]. Mol Cell Proteomics, 2006, 5(1): 172-181. doi: 10.1074/mcp.M500135-MCP200

[4] Pawlak M, Schick E, Bopp MA, et al. Zeptosens' protein microarrays: a novel high performance microarray platform for low abundance protein analysis[J]. Proteomics, 2002, 2(4): 383-393. doi: 10.1002/1615-9861(200204)2:4<383::AID-PROT383>3.0.CO;2-E

[5] Knezevic V, Leethanakul C, Bichsel VE, et al. Proteomic profiling of the cancer microenvironment by antibody arrays[J]. Proteomics, 2001, 1(10): 1271-1278. doi: 10.1002/1615-9861(200110)1:10<1271::AID-PROT1271>3.0.CO;2-6

[6] Zhang H, Pelech S. Using protein microarrays to study phosphorylation-mediated signal transduction[J]. Semin Cell Dev Biol, 2012, 23(8): 872-882. doi: 10.1016/j.semcdb.2012.05.009

[7] Edgar R, Domrachev M, Lash AE. Gene expression omnibus: NCBI gene expression and hybridization array data repository[J]. Nucleic Acids Res, 2002, 30(1): 207-210. doi: 10.1093/nar/30.1.207

[8] Rodchenkov I, Babur O, Luna A, et al. Pathway commons 2019 update: integration, analysis and exploration of pathway data[J]. Nucleic Acids Res, 2020, 48(D1): D489-D497. doi: 10.1093/nar/gkz946

[9] Tomonari T, Sato Y, Tanaka H, et al. Potential use of lenvatinib for patients with unresectable hepatocellular carcinoma including after treatment with sorafenib: Real-world evidence and in vitro assessment via protein phosphorylation array[J]. Oncotarget, 2020, 11(26): 2531-2542. doi: 10.18632/oncotarget.27640

[10] Manabe T, Yasuda H, Terai H, et al. IGF2 autocrine-mediated IGF1R activation is a clinically relevant mechanism of osimertinib resistance in lung cancer[J]. Mol Cancer Res, 2020, 18(4): 549-559. doi: 10.1158/1541-7786.MCR-19-0956

[11] Kagiwada H, Kiboku T, Matsuo H, et al. Assessing the activation/inhibition of tyrosine kinase-related pathways with a newly developed platform[J]. Proteomics, 2021, 21(16): 2000251. doi: 10.1002/pmic.202000251

[12] Kagiwada H, Motono C, Horimoto K, et al. Phosprof: pathway analysis database of drug response based on phosphorylation activity measurements[J]. Database, 2022, 2022: baac072. doi: 10.1093/database/baac072

[13] Kanehisa M, Furumichi M, Tanabe M, et al. KEGG: new perspectives on genomes, pathways, diseases and drugs[J]. Nucleic Acids Res, 2017, 45(D1): D353-D361. doi: 10.1093/nar/gkw1092

[14] Fabregat A, Jupe S, Matthews L, et al. The reactome pathway knowledgebase[J]. Nucleic Acids Res, 2018, 46(D1): D649-D655. doi: 10.1093/nar/gkx1132

[15] Hornbeck PV, Zhang B, Murray B, et al. PhosphoSitePlus, 2014: mutations, PTMs and recalibrations[J]. Nucleic Acids Res, 2015, 43(D1): D512-D520. doi: 10.1093/nar/gku1267

[16] Kagamu H, Kitano S, Yamaguchi O, et al. CD4⁺ T-cell immunity in the peripheral blood correlates with response to anti-PD-1 therapy[J]. Cancer Immunol Res, 2020, 8(3): 334-344. doi: 10.1158/2326-6066.CIR-19-0574

[17] Eisinga R, Breitling R, Heskes T. The exact probability distribution of the rank product statistics for replicated experiments[J]. FEBS Lett, 2013, 587(6): 677-682. doi: 10.1016/j.febslet.2013.01.037

[18] Fisher RA. Statistical methods for research workers[M]. 11th ed. Edinburgh: Oliver and Boyd, 1925.

[19] Saito S, Aburatani S, Horimoto K. Network evaluation from the consistency of the graph structure with the measured data[J]. BMC Syst Biol, 2008, 2: 84. doi: 10.1186/1752-0509-2-84

[20] Ben-Israel A, Greville TNE. Generalized inverses: theory and applications[M]. 2nd ed. New York: Springer, 2003.

[21] Whittaker J. Graphical models in applied multivariate statistics[M]. Chichester: John Wiley and Sons, 1990.

[22] Lamb J, Crawford ED, Peck D, et al. The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease[J]. Science, 2006, 313(5795): 1929-1935. doi: 10.1126/science.1132939

[23] Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles[J]. Proc Natl Acad Sci U S A, 2005, 102(43): 15545-15550. doi: 10.1073/pnas.0506580102

[24] Tokarski JS, Newitt JA, Chang CYJ, et al. The structure of Dasatinib (BMS-354825) bound to activated ABL kinase domain elucidates its inhibitory activity against imatinib-resistant ABL mutants[J]. Cancer Res, 2006, 66(11): 5790-5797. doi: 10.1158/0008-5472.CAN-05-4187

[25] Saito S, Hirokawa T, Horimoto K. Discovery of chemical compound groups with common structures by a network analysis approach (affinity prediction method)[J]. J Chem Inf Model, 2011, 51(1): 61-68. doi: 10.1021/ci100262s

[26] Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction[M]. 2nd ed. New York: Springer, 2009.