A side-effect free method for identifying cancer drug targets

Abstract

Identifying effective drug targets, with little or no side effects, remains an ever challenging task. A potential pitfall of failing to uncover the correct drug targets, due to side effect of pleiotropic genes, might lead the potential drugs to be illicit and withdrawn. Simplifying disease complexity, for the investigation of the mechanistic aspects and identification of effective drug targets, have been done through several approaches of protein interactome analysis. Of these, centrality measures have always gained importance in identifying candidate drug targets. Here, we put forward an integrated method of analysing a complex network of cancer and depict the importance of k-core, functional connectivity and centrality (KFC) for identifying effective drug targets. Essentially, we have extracted the proteins involved in the pathways leading to cancer from the pathway databases which enlist real experimental datasets. The interactions between these proteins were mapped to build an interactome. Integrative analyses of the interactome enabled us to unearth plausible reasons for drugs being rendered withdrawn, thereby giving future scope to pharmaceutical industries to potentially avoid them (e.g. ESR1, HDAC2, F2, PLG, PPARA, RXRA, etc). Based upon our KFC criteria, we have shortlisted ten proteins (GRB2, FYN, PIK3R1, CBL, JAK2, LCK, LYN, SYK, JAK1 and SOCS3) as effective candidates for drug development.

Figure 1

The approach for CaI construction and analysis. (a) The three databases BIOCARTA, PID, REACTOME and KEGG utilised for the extraction of pathways followed by disintegration into protein constituents and identification of any other pathways they are involved with. (b) The meta database STRING for finding the interactions of all the proteins pooled above. (c) The large component of the CaI constructed from pooled interactions above, coloured by 335 modules by Rosvall Algorithm with node size plotted as per degree (d) The analyses for power law, K-core, inter- and intra-modular connectivities for CaI constructed followed by the drug statuses against centrality measures, in clockwise manner.

Figure 2

The technical analysis of the constructed CaI. (a) Complementary Cumulative Degree Distribution (CCDF) of CaI showing Power-Law behaviour. (b) K-core analysis of CaI representing the size of each k-shell (number of proteins appearing in k-core but not in k + 1th core) from periphery (k = 1) to inner core (k-max). (c) Classification of CaI proteins (R) based on its role and region in network space, the P-Z space classified into 7 categories of hub and non-hub nodes. The latter has been assigned as ultra-peripheral (R1), peripheral (R2), non-hub connector (R3) and non-hub kinless nodes (R4) and the former has been assigned as provincial (R5), connector (R6) and kinless hubs (R7) as described by Guimera et al.. Kinless hubs nodes are supposed to be important in term of functionality, which has high connection within module as well as between modules.

Figure 3

The drug statuses of the CaI proteins. (a) The hierarchical statuses of the drugs based on the information from Drugbank database and FDA regulations. Nutraceuticals (shaded in yellow) need not undergo clinical trial (WCT) and projects out from the experimental phase (WOC). (b) Segregation of classified R proteins having WCT (blue bars) and WOC phases (in red). R1: n = 2422; R2: n = 3953; R3: n = 1272; R4: n = 400; R5: n = 33; R6: n = 63; R7: n = 34. (c) Classification of R proteins (Y-axis) into the drug status of investigational (INV, light blue) or higher viz. APP (green), ILL (orange) and WD (red). R1: n = 173; R2: n = 598; R3: n = 250; R4: n = 113; R5: n = 5; R6: n = 28; R7: n = 15.

Figure 4

Network centrality impact of the classified CaI proteins upon drug statuses. (a) The distribution of the centrality measures of betweenness (BC), degree (DC) and Eigenvector (EC) for the classified R proteins in the x-axis: R1, n = 2422; R2, n = 3953; R3, n = 1272; R4, n = 400; R5, n = 33; R6, n = 63; and R7, n = 34. The mean is shown through the box and whisker plot. (b) The mean distribution of the BC, DC and EC measures for the classified R proteins having WOC and WCT statuses for the drugs developed against them in the x-axis. For WOC: R1, n = 2249; R2, n = 3355; R3, n = 1022; R4, n = 287; R5, n = 28; R6, n = 35; and R7, n = 19 and for WCT: R1, n = 173; R2, n = 598; R3, n = 250; R4, n = 113; R5, n = 5; R6, n = 28; and R7, n = 15. (c) Impact of centrality measures on the individual drug statuses of the classified R group proteins. X-axis represents different drug statuses: Not Available (N/A; n = 6193), Experimental (Exp; n = 287), Nutraceutical (NUT; n = 515), Investigational (Inv; n = 144), Approved (APP; n = 837), Illicit (ILL; n = 20), and Withdrawn (WD; n = 181). Only the statistical comparison between APP and WD are shown in the graph (refer to Supplementary Table 2 for more in-depth statistical results). Y-axis of the graphs represents centrality measures BC, DC and EC for respective panels in rows. Error bars = standard error, *p < 0.05, **p < 0.01, ***p < 0.001.