Tumor suppressor genes may be viewed as the antithesis of oncogenes. These genes normally function not only to suppress transformation but also govern the regulatory proteins to suppress cellular proliferation. Mutation in these genes (and the resultant proteins) arising from environmental exposure, genetic susceptibility, other infectious agents etc. serves as the main cause of various types of cancers. Classical examples of these tumor suppressor genes are proteins like p53, Wilm's tumor and BAPI, which are associated with a wider range of neoplasm including breast tumor.
We have selected three tumor suppressor proteins, bovine. API, human WTI (C-terminal region) and human Prohibitin and predict the 3D structures based either on sequence homologies with the proteins of known structures (Homology Modeling) or by the fold recognition techniques (Threading).
The 3D structure of bovine .:API (AP lyase) has been predicted based on the crystal coordinates of ldew. BAPI repairs the apurinic/apyrimidinic sites produced as a result of damage to DNA. The sequence of BAPI is significantly homologous to human AP endonuclease. The model mainly consists of a 2 domain globular protein. Both domains display similar topology comprising of a six stranded beta sheet surrounded by alpha helices which pack together to form 4 layered alphalbeta sandwich. The active site is located in a pocket at the top of alphalbeta sandwich and is surrounded by loop regions. It has been proposed that His309, Asp283 and Thr265 play an important role in repairing AP sites.
The recognition and binding of AP endonucleases to DNA primarily involves 3 alpha helical loop regions. The helical loop, alpha8 (residues 222-228) is positioned within the DNA major groove where Asn222 and Asn226 make water mediated contacts with the phosphate of G9. The second helical loop, alpha5 (residues 175-181) is also positioned within the major groove with Arg 177 making two hydrogen bonds with phosphate (OIP) of nucleotide C8. The helical loop alpha11 (residues 266-277) is oriented near the DNA minor groove and could make additional phosphate backbone interactions between amino acid residues Phe266 and Met271 and nucleotide base 3DR7 and G25.
The structural characteristics of BAP1 were also studied by mutating amino acids involved in DNA binding or catalysis i.e. Arg177â†’7Ala, Asp283â†’7 Ala and His309â†’7 Asn. Thus in the Arg 177â†’7 Ala mutant, insertion of Ala at position 177 causes loss of DNA binding possibly arising from the uncapping of DNA bound active site which results in the destabilization of the extra-helical AP site conformation. Similarly mutation of Asp283â†’7 Ala and His309â†’7 Asn resulted in an elimination of enzymatic activity.
Wilm's tumor protein (WT1) is a nuclear transcription factor that regulates the activity of insulin growth factor (IGF) and transforming growth factor (TGF) system, both of which are implicated in breast tumorigenesis. The complete sequence of WT1 protein does not show significantly homology with other proteins of known structure. The C-terminal region of human WT1 (residues 310-449) was modeled using a chimeric template constructed by superposing the two proteins, GLI and Zif268. The 3D model shows that the C-terminal region of WT1 is folded into 4 helices and 8 beta pleated sheets. One cobalt and three zinc ions are coordinated by conserved residues, with 2 cysteines contributed by each Î²3 sheet and 2 histidines by each Î± helix. These are held together by a hydrophobic core and by Co/Zn ion
The protein-DNA complex in the predicted WTl model involves four zinc finger containing Cys(2)-His(2) located in the C-terminal region of WTl. These residues coordinate with cobalt and three zinc metal ions. The hydrogen bonding was observed only between His373, His401 and DNA base G7 and G4.
A 17 amino acid domain contain Cys-His-Zn motif is responsible for DNA and RNA binding which coordinate with Zn203 and DNA bases AI, G2, C3, G4, T5, C56, C57, A58 and G60.
DNA recognition is mediated through base contacts with the side chains of amino acids located at 4 positions of the recognition helix (Le. amino acid at positions 1,2, 3 and 6). Asp396, Asp429, present at 2nd position of helix 3 and 4 as well as Arg366 and Arg427 present at 1st position of helix 2 and 4 make hydrogen bonds with DNA bases C3, C55 and, A52, Tll, G4 and T5 respectively.
In helix 1, His339 present at the 3rd position, stacks against G4, T5 and G6 making van der Waalâ€™s interaction which may be significant for site specific recognition. Helix 2 and 4 have glutamine (Gln369) and glutamic acid (Glu430) at the 3rd position of helix .which helps in stabilizing the conformation of other residues of the helix. Thr400 is at the 6th position of helix 3 making a water mediated contact with 01P of G4. Arg372 and Arg433 are present at the 6th position of helix 2 and 4 respectively which make contacts with C9, G54 and AI, C3 respectively.
Linker sequences in WTl capping the helix recognizes the DNA and facilitates the specificity of DNA binding with zinc finger. The residues of the linker region are stabilized by H-bonds present between the linker residues as well as the extra linker amino acids. Only Lys351 make H-bond with DNA base C8.
Mutation of important site in the WTl molecule did not indicate any significant change in the case of Lys371â†’Ala and Ser415â†’Ala mutants. In the case of mutant Cys416â†’Ala, reduction in the van der Waalâ€™s contact between the amino acids was observed. Also a loss of coordination with the metal ion Zn appears possible. Two other mutants i.e. His434â†’Asp and His434â†’Arg showed a loss of coordination with the metal ion (Zn203). Also His434 does not interact directly with any DNA base, whereas the mutated Arg434. may interact directly with Al which may affect the DNA binding pattern.
PROSITE search shows that Ser313, Ser365 and Ser393 may undergo phosphorylation by casein kinase II whereas Tyr353 may be targeted by tyrosine kinase.
Prohibitin contributes to the control of the G1â†’S transition in the cell cycle in a complex manner which involves both transcriptional and post translational mechanisms. The target sequence of Prohibitin does not show any homology with other proteins of known 3D structure thus excluding the possibility of direct homology modeling. The method adopted in this case is. based on fold recognition technique. Three-dimensional model of human Prohibitin was constructed using the crystal coordinates of 1dkx and 1fpo. The C-terminal region of prohibitin shows a predominant helical topology whereas N-terminal region has a sheeted structure. The overall structure may be divided into three domains. The sequence patch 74-82 is homologous to the loop region in the Dnak, which helps in binding the substrate (Rb protein) while the sequence patch 107-116 is homologous to the J-domain of Hsc20, which shows ATPase activity. The region 185-214 acts as a transcriptional repressor of E2F activity. This region consists of 2 alpha helices (D and E) with structural homology to Hsc20 and may serve as a site of interaction with other proteins. The functional domain of prohibitin which interacts with Raf-l is the carboxyl terminal region Le. residues 243-272 (Raf-l can bind to prohibitin which in turn could bind to Rb). The region between 261-272 showed no homology with any known structure and as such this region was analyzed by secondary structure prediction. This region consists of helix-loop-helix motif and Asn256 (a glycosylated residue) which play an important role in recognition of Raf-l.
Among the five prohibitin mutants modeled no significant change was seen in the Glu52â†’Ala and Val88â†’Ala. In the Argl05â†’His mutant, both Argl05 as well as the Hisl05 mutant show 4 H-bonds although with a slightly different pattern. Besides, a salt bridge originally seen in the prohibitin model between Argl05 and Glu112 is also lost. This can account for the charge disruption which may in turn affect the antiproliferative activity of the protein. In Thrl08â†’I1e mutant, a significant increase in the accessibility of mutated He was observed. Also this mutation resulted in loss of a number of H-bonds. Mutation Thrâ†’I1e shows that presence of hydrophobic amino acid in helix A might account for more interactions. within the helix. In Glul12â†’Ala mutant, a significant decrease in the accessibility of Ala was observed. Glu112 forms ion pair with Arg 105 and Arg 117. However the mutant form is devoid of these interactions.
Thus these tumor suppressor proteins appear to play an important role not only in DNA repair but also in the control of transcription, cell proliferation as well as protection of the newly synthesized protein from misfolding.