Crystallographic structure solution of Brazilian subtypes of HIV-1 proteases and structural studies of E. coli L-asparaginase

Author(s)

Mario Sanches Matilde Junior

• Document type: Doctoral Thesis
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Instituto de Física de São Carlos (IFSC/BT)
• Defense date: 2004-12-17

Examining board members

Igor Polikarpov; Maria Cristina Nonato Costa; Debora Foguel; Marcos Roberto de Mattos Fontes; Richard Charles Garratt

• Advisor: Igor Polikarpov

Abstract

One of the major problems faced in antiviral therapy against AIDS is the emergence of viral variants that exhibit drug resistance, as well as viral subtypes naturally more liable to development of therapeutic failure. In this work we solved the crystal structure of four HIV-1 proteases complexed with the universal C2 symmetry-based inhibitor TL-3. These proteases where obtained from patients with AIDS: one of the subtype B wild type (Bwt), one of the subtype F wild type (Fwt), and a mutant of each subtype (Bmut and Fmut), these last two out of patients showing therapeutic failure. The proteins were produced by Escherichia coli bacteria heterologous expression, purified out of the inclusion bodies, and refolded. The collected diffraction data were processed to 2.1 A resolution for the Bwt:TL-3 complex, 1.75 A for Bmut:TL-3, 2.1 A for Fwt:TL-3 and 2.8 A for Fmut:TL-3. These data were initially processed in the P6122 space group and latter reprocessed in P6i. The four structures were solved by molecular replacement using a published HIV-1 protease structure as a model. The structural analysis shown that the TL-3 inhibitor binds in a similar fashion in the active site of all four structures. On the proteases Bmut and Fmut the mutation V82A causes a repacking of the SI\' pocket which rerranges the inhibitor\'s side chain at P1\' subsite. Our analysis further indicate that some polymorphic substitutions between subtypes B and F could lead to a stabilization of naturally flexible regions on subtype F proteases, creating a intrinsically less active resistant enzyme. On the proteases Fwt and Fmut the polymorphic substitution M36I leads to the displacement of the loop between residues 35-41, which would cause a flexibility loss of the flaps and of the loop 76-83 in the active site. Our comparisons further indicate that the polymorphic substitution L89M on non-B subtypes could be equivalent to the L90M resistance mutation on subtype B proteases. Lastly, based on these structural data we were able to suggest a few structural modifications on the TL-3 inhibitor that could furnish a more potent inhibitor. On this thesis we also present the data of structural solution and analysis of the Escherichia coli L-asparaginase solved at 1.95 A resolution in C2 space group. Based on structural and kinetical data we proposed a general reaction mechanism for amidohidrolases, which include L-asparaginases, involving the formation of two catalytic triads Thr-Tyr-Glu and Thr-Lys-Asp, which acounts for the two threonines in the active site. Our cavity volume analysis of three amidohidrolases also indicate that the increasing in the L-glutaminase activity in some enzimes is directly proportional to the increase in the cavity volume. (AU)