de Melo E. B., Bruni A. T., Ferreira M. M. C., "VALIDATION OF SEMI-EMPIRICAL METHODOLOGY FOR GEOMETRY OPTIMIZATION OF HIV-IN INHIBITORS". Ribeirão Preto, SP, Brazil, 25-28/09/2005: 5th International Congress of Pharmaceutical Sciences (CIFARP2005), Braz. J.  Pharm. Sci., 41(Suppl 1) (2005), 265. Poster MM001. Section: Molecular Modeling (MM).

MM001-VALIDATION  OF  SEMI-EMPIRICAL  METHODOLOGY  FOR  GEOMETRY  OPTIMIZATION  OF  HIV-IN
NHIBITORS.
 

EDUARDO B. DE MELO (PG)⁽¹⁾; ALINE T. BRUNI (PQ)⁽²⁾; MÁRCIA M. C. FERREIRA (PQ)⁽³⁾

⁽¹⁾UNIOESTE; ⁽²⁾IBILCE/UNESP; ⁽³⁾IQ/UNICAMP.
 

The integrase (IN) is indispensable for the replication of  HIV.  The only HIV-IN/inhibitor complex is the  PDB  1QS4  (5CITEP/IN).
When  not  much  crystallographic  information   is  available,   the  best  starter point  for   a  theoretical  study  are  the  most  stable
calculated  geometries.  In  this  trial  were  selected  the  best  semi-empirical  theory  (AM1 or PM3)  for  the optimization  of  these
compounds. The comparison criteria utilized were: the bond lenghts of 5CITEP PDB; the properties of the optimized geometries; and
literature data.  The  5CITEP,  a keto-enol,  was optimized for these two theories.  The bond lengths of the obtained geometries were
similar  to  5CITEP  PDB.  Overlap  between  AM1  and  PM3  geometries  showed that they were identical.  HOMO,  LUMO  and
dipole moment  (D) values are also similar.  However,  the heats of formation values  (DH_f)  presented great difference  (D = 43 kcal/
mol).  The  same  was  observed for the partial charges of the atoms.  In  the  literature, the  AM1  theory  is  described as capable to
compute most exact values for D,  HOMO  and partial charges than the PM3.  Furthemore, it is indicated for the study of keto-enols.
Therefore, AM1 is the semi-empirical theory indicated for a molecular modeling study of keto-enol inhibitors of HIV-IN.
 

Supervisor: Márcia M. C. Ferreira.