There has recently been a rapid progress in computational methods for determining protein targets of small molecule drugs, which will be termed as compound protein interaction (CPI). In this review, we comprehensively review topics related to computational prediction of CPI. Data for CPI has been accumulated and curated significantly both in quantity and quality. Computational methods have become powerful ever to analyze such complex the data. Thus, recent successes in the improved quality of CPI prediction are due to use of both sophisticated computational techniques and higher quality information in the databases. The goal of this article is to provide reviews of topics related to CPI, such as data, format, representation, to computational models, so that researchers can take full advantages of these resources to develop novel prediction methods. Chemical compounds and protein data from various resources were discussed in terms of data formats and encoding schemes. For the CPI methods, we grouped prediction methods into five categories from traditional machine learning techniques to state-of-the-art deep learning techniques. In closing, we discussed emerging machine learning topics to help both experimental and computational scientists leverage the current knowledge and strategies to develop more powerful and accurate CPI prediction methods.

Membrane proteins sense extracellular cues and transduce intracellular signaling to coordinate directionality and speed during cellular migration. They are often localized to specific regions, as with lipid rafts or tetraspanin-enriched microdomains; however, the dynamic interactions of tetraspanins with diverse receptors within tetraspanin-enriched microdomains on cellular surfaces remain largely unexplored. Here, we investigated effects of tetraspan(in) TM4SF5 (transmembrane 4 L6 family member 5)-enriched microdomains (TERMs) on the directionality of cell migration. Physical association of TM4SF5 with epidermal growth factor receptor (EGFR) and integrin α5 was visualized by live fluorescence cross-correlation spectroscopy and higher-resolution microscopy at the leading edge of migratory cells, presumably forming TM4SF5-enriched microdomains. Whereas TM4SF5 and EGFR colocalized at the migrating leading region more than at the rear, TM4SF5 and integrin α5 colocalized evenly throughout cells. Cholesterol depletion and disruption in TM4SF5 post-translational modifications, including -glycosylation and palmitoylation, altered TM4SF5 interactions and cellular localization, which led to less cellular migration speed and directionality in 2- or 3-dimensional conditions. TM4SF5 controlled directional cell migration and invasion, and importantly, these TM4SF5 functions were dependent on cholesterol, TM4SF5 post-translational modifications, and EGFR and integrin α5 activity. Altogether, we showed that TM4SF5 dynamically interacted with EGFR and integrin α5 in migratory cells to control directionality and invasion.-Kim, H.-J., Kwon, S., Nam, S. H., Jung, J. W., Kang, M., Ryu, J., Kim, J. E., Cheong, J.-G., Cho, C. Y., Kim, S., Song, D.-G., Kim, Y.-N., Kim, T. Y., Jung, M.-K., Lee, K.-M., Pack, C.-G., Lee, J. W. Dynamic and coordinated single-molecular interactions at TM4SF5-enriched microdomains guide invasive behaviors in 2- and 3-dimensional environments.

No abstract

Active c-Src non-receptor tyrosine kinase localizes to the plasma membrane via N-terminal lipid modification. Membranous c-Src causes cancer initiation and progression. Even though transmembrane 4 L six family member 5 (TM4SF5), a tetraspan(in), can be involved in this mechanism, the molecular and structural influence of TM4SF5 on c-Src remains unknown. Methods: Here, we investigated molecular and structural details by which TM4SF5 regulated c-Src devoid of its N-terminus and how cell-penetrating peptides were able to interrupt c-Src activation via interference of c-Src-TM4SF5 interaction in hepatocellular carcinoma models. Results: The TM4SF5 C-terminus efficiently bound the c-Src SH1 kinase domain, efficiently to the inactively-closed form. The complex involved protein tyrosine phosphatase 1B able to dephosphorylate Tyr530. The c-Src SH1 domain alone, even in a closed form, bound TM4SF5 to cause c-Src Tyr419 and FAK Y861 phosphorylation. Homology modeling and molecular dynamics simulation studies predicted the directly interfacing residues, which were further validated by mutational studies. Cell penetration of TM4SF5 C-terminal peptides blocked the interaction of TM4SF5 with c-Src and prevented c-Src-dependent tumor initiation and progression in vivo. Conclusions: Collectively, these data demonstrate that binding of the TM4SF5 C-terminus to the kinase domain of inactive c-Src leads to its activation. Because this binding can be abolished by cell-penetrating peptides containing the TM4SF5 C-terminus, targeting this direct interaction may be an effective strategy for developing therapeutics that block the development and progression of hepatocellular carcinoma.

The transmembrane 4 L six family proteins TM4SF1, TM4SF4, and TM4SF5 share 40-50% overall sequence identity, but their C-terminus identity is limited. It may be likely that the C-termini of the members are important and unique for own regulatory functions. We thus examined how the TM4SF5 C-terminus affected cellular functions differentially from other family members. Using colon cancer cells expressing wildtype (WT), C-terminus-deleted, or chimeric mutants, diverse cellular functions were explored in 2-dimensional (2D) and 3-dimensional (3D) condition. The C-termini of the proteins were relatively comparable with respect to 2D cell proliferation, although each C-terminal-deletion mutant exhibited increased proliferation relative to the WT. Using chimeric constructs, we found that the TM4SF5 C-terminus was critical for regulating the diverse metastatic functions of TM4SF5, and could positively replace the C-termini of other family members. Replacement of the TM4SF1 or TM4SF4 C-terminus with that of TM4SF5 increased spheroids growth, transwell migration, and invasive dissemination from spheroids in 3D collagen gels. TM4SF5-mediated effects required its extracellular loop 2 linked to the C-terminus via the transmembrane domain 4, with causing c-Src activation. Altogether, the C-terminus of TM4SF5 appears to mediate pro-migratory roles, depending on a structural relay from the second extracellular loop to the C-terminus.

Transmembrane 4 L six family member 5 (TM4SF5) is highly expressed in hepatocellular carcinoma tissues and enhances migration in two-dimensional environments. Here, we investigated how TM4SF5 is involved in diverse pro-metastatic phenotypes in in vivo-like three-dimensional (3D) extracellular matrix gels. TM4SF5-positive cells aggressively formed invasive foci in 3D Matrigel, depending on TM4SF5-mediated signaling activity, cytoskeletal organization, and matrix metallopeptidase (MMP) 2-mediated extracellular remodeling, whereas TM4SF5-null cells did not. The TM4SF5-null cells did, however, form invasive foci in 3D Matrigel following inhibition of Rho-associated protein kinase or addition of collagen I, suggesting that collagen I compensated for TM4SF5 expression. Similarly, TM4SF5-positive cells expressing vascular endothelial-cadherin formed network-like vasculogenic mimicry in 3D Matrigel and collagen I mixture gels, whereas TM4SF5-negative cells in the mixture gels displayed the network structures only upon further treatment with epidermal growth factor. The foci formation also required MMP2-mediated remodeling of the extracellular matrix. Co-cultures exhibited TM4SF5-positive or cancer-associated fibroblasts at the outward edges of TM4SF5-null cell clusters. Compared with TM4SF5-null cells, TM4SF5-positive cells in 3D collagen gels showed a more invasive outgrowth with dramatic invadopodia. These observations suggest that TM4SF5 plays roles in the promotion of diverse metastatic properties with fewer environmental requirements than TM4SF5-negative cells.