Computational modeling of Repeat1 region of INI1/hSNF5: An evolutionary link with ubiquitin

Abstract

The structure of a protein can be very informative of its function. However, determining protein structures experimentally can often be very challenging. Computational methods have been used successfully in modeling structures with sufficient accuracy. Here we have used computational tools to predict the structure of an evolutionarily conserved and functionally significant domain of Integrase interactor (INI)1/hSNF5 protein. INI1 is a component of the chromatin remodeling SWI/SNF complex, a tumor suppressor and is involved in many protein-protein interactions. It belongs to SNF5 family of proteins that contain two conserved repeat (Rpt) domains. Rpt1 domain of INI1 binds to HIV-1 Integrase, and acts as a dominant negative mutant to inhibit viral replication. Rpt1 domain also interacts with oncogene c-MYC and modulates its transcriptional activity. We carried out an ab initio modeling of a segment of INI1 protein containing the Rpt1 domain. The structural model suggested the presence of a compact and well defined ββαα topology as core structure in the Rpt1 domain of INI1. This topology in Rpt1 was similar to PFU domain of Phospholipase A2 Activating Protein, PLAA. Interestingly, PFU domain shares similarity with Ubiquitin and has ubiquitin binding activity. Because of the structural similarity between Rpt1 domain of INI1 and PFU domain of PLAA, we propose that Rpt1 domain of INI1 may participate in ubiquitin recognition or binding with ubiquitin or ubiquitin related proteins. This modeling study may shed light on the mode of interactions of Rpt1 domain of INI1 and is likely to facilitate future functional studies of INI1.

Keywords: SNF5; integrase interactor (INI)1/hSNF5; protein structure modeling; repeats; ubiquitin.

Figure 1

Cartoon illustrating the domain structure of INI1 and the Secondary Structure prediction within Rpt1(183–265) domain. A. Arrangement of two repeat domains of INI1 (Rpt = repeat; NES = nuclear export signal). B. Secondary structures within Rpt1 as predicted by different methods; (a) Jpred (b) PSIPRED (c) YASPIN (d) PHD. Numbers (0–9) represent the confidence/reliability of prediction.

Figure 2

Best models of S6(Rpt1) generated by different programs. (A) Robetta, (B) I‐TASSER, (C) QUARK. Coloring from N (blue) to C (brown) terminus. Both the front view and transverse view is depicted.

Figure 3

Intramolecular interactions, distribution of residues and conservation: (A) Hydrophobic core in the structure and interhelical H‐bonds (salt bridge) stabilizing the helices. Arrangement of hydrophobic residues (grey) projecting at the center of the protein. (B) Arrangement of charged residues (magenta) and Hydrophobic residues (cyan) on the helices. (C) Molecular surface of S6(Rpt1) colored by sequence conservation among SNF5 homologues, analyzed by the program ConSurf. The colors vary from dark green for highly conserved residues to magenta for residues with little conservation. Conserved surface patch is in the center of the structure. (D) Surface map of hydrophobic (grey)/polar (red: negatively charged, blue: positively charged) residues.

Figure 4

Comparison of (A) Sequence and secondary structures, (B) 3D Structures of PFU domain of PLAA (2k89A.pdb), S6(Rpt1) of INI1 and Ubiquitin. (C) Superposition of the structural homologous region of INI1 Rpt1 and PFU domain of PLAA. (D) Surface representation of charged and hydrophobic regions PFU and S6(Rpt1), blue: electropositive, red: electronegative. Selected region within the dotted oval shows the hydrophobic patch at the center. (E) Superposition of the structural homologous N‐terminal region of INI1 Rpt1 and ubiquitin.

Figure 5

Sequence and Structure comparison of Rpt1 and Rpt2. Modeled structure of (A) Rpt2 (B) Rpt1. Ionizable residues (red: negative, blue: positive). (C) Superposition of Rpt1 (green) and Rpt2 (red).