Structure-functional study of two signaling protein complexes tensin2-PTB/DLC1 and Smad4-MH1/Hoxc9
by Chen Lihong
Ph.D. Life Science
xvii, 184 p. : ill. (some col.) ; 30 cm
The protein Deleted in Liver Cancer 1 (DLC1) has been demonstrated as a bona fide tumor suppressor in diverse human malignancies. The tumor suppressive function of DLC1 by the catalytic activity of its RhoGAP domain is tightly controlled by a wide variety of regulatory mechanisms temporally and spatially. The tensin family proteins are found to be DLC1’s spatial regulators. Tensin is a family of proteins localized at integrin-mediated focal adhesions bridging actin cytoskeletion and integrins. Identification of tensin2 as the first and novel interacting partner of DLC1 has provided the first evidence about the focal adhesion localization of DLC1. Subsequent studies have further demonstrated that the interaction of DLC1 with tensins plays unequivocal role in modulating the tumor suppress...[ Read more ]
The protein Deleted in Liver Cancer 1 (DLC1) has been demonstrated as a bona fide tumor suppressor in diverse human malignancies. The tumor suppressive function of DLC1 by the catalytic activity of its RhoGAP domain is tightly controlled by a wide variety of regulatory mechanisms temporally and spatially. The tensin family proteins are found to be DLC1’s spatial regulators. Tensin is a family of proteins localized at integrin-mediated focal adhesions bridging actin cytoskeletion and integrins. Identification of tensin2 as the first and novel interacting partner of DLC1 has provided the first evidence about the focal adhesion localization of DLC1. Subsequent studies have further demonstrated that the interaction of DLC1 with tensins plays unequivocal role in modulating the tumor suppressive activity of DLC1. To date, association with tensin proteins is the best characterized regulatory event in the subcellular localization of DLC1 at the protein level. Different interaction modes have been revealed in the interaction of DLC members with the C-terminal SH2 and/or PTB domains of tensin family proteins. DLC1 utilizes a SH2 binding motif to interact with the SH2 domain of tensin2 and cten, and a novel binding site to interact with tensin2 PTB domain. In order to gain insight into the molecular mechanism of DLC1-tensin2 mediated tumor suppression, it is important to understand the binding details of the interaction between DLC1 and tensin2.
One of the best methods to reveal the molecular mechanism underlying the cellular functions and interactions of proteins is to elucidate their three dimensional structures. In order to elucidate the biochemical and structural mechanisms governing the regulation of DLC1’s tumor suppressive function through the mediation of tensin family proteins, we employed multidimensional NMR spectroscopy technique and various other biochemical and functional experiments.
The first part of this dissertation describes the biochemical and structural study of the interaction between human tensin2 and DLC1 with its implications in tumor suppression. For the tensin2 SH2 domain (Chapter 4), we determined the NMR solution structure of tensin2 SH2 domain, studied the phosphor-Tyr independent interaction of the SH2 domain with DLC1 peptide, and identified the residues involved in the binding. This is the first structure of the SH2 domain from the tensin family, showing the characteristic SH2 fold, recognizing the SIYDNV motif of DLC1 peptide in a pTyr-independent manner similar to the ‘three-pronged’ binding of SAP SH2. For the tensin2 PTB domain (Chapter 5), we solved and described the NMR solution structure of tensin2 PTB in complex with the DLC1 peptide, which adopts a novel PTB-peptide binding mode un-discovered before. We characterized and verified the binding surface and key residues on the tensin2 PTB domain and DLC1 peptide. Based on the complex structure, we further demonstrated the functional significance of this novel binding mode in regulating the tumor suppressive activity of DLC1-tensin2 complex. Meanwhile, we studied the interaction between the integrin β3 cytoplasmic tail and tensin2 PTB domain, which is shown to compete with the DLC1 peptide. The PTB and SH2 domains were not detected to bind to liposomes in the sedimentation-based liposome binding assay compared to PH domain. The tensin2 PTB-DLC1 peptide complex structure, with a novel binding mode, shed new light on the versatile recognition modes and extends the binding repertoire of the PTB domains in mediating diverse cellular signaling pathways, as well as provides a molecular and structural basis for better understanding the tumor suppressive activity of DLC1 and tensin2.
In the future, we’ll try to study the undefined regulatory mechanisms for DLC family proteins from structural angle. Meanwhile, it is worth investigating that if the novel peptide binding mode found in tensin2 PTB domain exists in other PTB domains. As critical scaffold and adapter proteins for signaling networks, it is likely that understanding the protein complexes bound through PTB domains will provide targets for therapeutic intervention for specific diseases.
In the second part of this dissertation, the biochemical and structural study of the interaction between the smad4 MH1 domain and Hoxc9 homeodomain is described. Smad family proteins mediate signaling initiated by bone morphogenetic proteins (BMPs). Upon TGF-β/BMP stimulation the common Smad, Smad4, can interact directly with Hox proteins and suppresses their DNA-binding activity. Mapping the Hox-binding domain in Smad4 revealed that the N-terminal MH1 domain and MH1-linker boundary were involved. However, the molecular mechanism of the binding is not well characterized and direct contact residues remain to be elucidated.
In order to elucidate the molecular basis of the MH1-homeodomain interaction which was found to regulate the transcription activity of Hox proteins, we carried out the multidimensional NMR and other biochemical experiments. The direct interaction between the two domains was confirmed by GST pull-down assay. NMR titration experiments showed that a strong and specific binding occurred between the two domains. NMR triple-resonance experiments and backbone assignments revealed that the N-terminal arm of Hoxc9 was involved in the interaction with Smad4 MH1. In addition, SPR measurement verified a strong interaction, bearing the dissociation constant of 400 nM, between the two domains. Our results yield the first detailed insight into the interaction between the homeodomain of Hox proteins and the conserved MH1 domain of Smad family proteins, and provide possible mechanism for Hox-Smad interactions that could interrupt Hox-DNA binding through the identification of a number of Smad-contact residues, including a positive-charged DNA-binding segment located at the N-terminal arm of the homeodomain.
In the future, the Hox-interacting residues on the MH1 domain of smad4 would be studied by NMR titration and mutations. The detailed structure characterization of the complex between Hoxc9 and Smad4 MH1 domains by NMR and crystallography approaches is in progress. The results of these efforts will help to further explain how the conformational change of the flexible N-terminus of Hox influences its binding to the regulators and benefit the elucidation of the regulation of Hox transcription factors.