Unformatted text preview:

Rational Drug Design Approach to Synthesizing HIV-1 Protease InhibitorsBy: Lingxian DingMarch 8, 1999IntroductionHuman Immunodeficiency Virus (HIV), the causitive agent of AIDS, now infects anestimated 20 million people worldwide, and while a cure has not yet been found for this fataldisease, rapid advances in molecular biology along with the 3-D elucidation of HIV proteinshave led to new drug targeting approaches for designing antiviral agents that specifically bind tokey regulatory proteins that are essential for HIV replication. The HIV-1 protease is one suchenzyme crucial for the maturation and assembly of infectious viral particles. Thus by inhibitingHIV-1 protease activity, a potential cure for AIDS is at hand. Armed with the 3D crystalstructure of the HIV-1 protease, rational drug design approaches have been widely employed inthe development of HIV protease inhibitors. While in clinical trials existent HIV-1 proteaseinhibitors show promise in causing an initial dramatic decline in viral plasma RNA in HIV-infected individuals, most patients taking protease inhibitors alone, however, by year’s end willshow an increase in plasma viral RNA to near baseline levels (Kaplan 1996). Failure of HIV-1protease inhibitor therapy is attributed to the development of resistant viral protease. Whiletechniques to combat resistance are currently being investigated, we have yet many obstacles toclear before discovering a cure for HIV.Structure and Function of HIV-1 Protease: Potential Drug Target IdentifiedThrough X-ray crystallography techniques, the 3D structure of HIV-1 protease has beenextensively studied and characterized. In its mature form, the viral protease exists as a dimer,whose subunits each consist of 99 amino acids. The folded subunits together interact to form acore hydrophobic, cylindrical catalytic cavity (24Å long by 6-8 Å in diameter) and two flexibleflaps (one per subunit) that can close around the substrate. Centered in the hydrophobic activesite are two symmetrically disposed catalytic aspartyl residues (Asp25 and Asp25’) that areinvolved in hydrolysis of the peptide bond (West 1995). Studies have shown that thehydrophobic cavern can hold six amino acids of the substrate in an extended conformation forcleavage (Molla 1998). Because structural and enzymatic proteins of the HIV virus aretranslated as part of two polyprotein precursors (Gag and Gag//Pol) cleavage of theseprecursors to generate gag matrix (p17), capsid (p24), nucleocapsid (p15) proteins, pol reversetranscriptase, integrase enzymes, and other viral proteins is vital for the production of mature,infectious viral particles (Molla 1998). Since this cleavage process is critical to viralpropagation, inhibition of viral protease proves to be an attractive drug target.Rational Drug Design: Structure and Receptor Based ApproachesInitial approaches to developing HIV-1 protease inhibitors were based on characterizingsubstrate-protease interactions that led to cleavage. Studies by Griffiths, J.T. et al. (1992) foundpreference for proteolysis of substrates with scissile hydrophobic-hydrophobic (usually involvingleucine, alanine, or valine) or aromatic-proline (i.e. phenylalanine-proline) peptide bonds.Hydrogen bonding and electrostatic interactions along the cylindrical groove were also taken intoaccount. But beyond optimal fitting of amino acid side chains into the binding groove,characterization of transition state structures has also guided the design of protease inhibitors. Ina functional protease a water molecule is held between the two active apartic acids. Duringhydrolysis the water molecule is added across the cleaved peptides. With this information athand, the initial inhibitors consisted of small polypeptides of seven amino acids in length thatwere designed to mimic normal cleavage sites but with the replacement of a nonhydrolyzableisostere at the cleavage site. Furthermore, the central carbonyl groups of these potent proteaseinhibitors also interact, via hydrogen bonding, with the water molecule held at the active site inorder to stabilize the close formation of the flaps on the enzyme (West 1995).While inhibitors such as Ro31-8959, LY289612, and other similar protease inhibitorsdesigned with the above described criteria prove to be potent antivirals in vitro, they suffer frompoor absorption, poor oral bioavailability, short serum half-life values, high susceptibility tohydrolysis by degradative enzymes, and other pharmacological problems (Reich 1995). Becauseof these limitations on peptidomimetic inhibitors, efforts have been made to minimizepharmacological problems by developing inhibitors that are structural mimics of peptides butwith little or no peptidic character.Reich et al. (1995) approached this problem by beginning with the cocrystal structure ofHIV-1 protease with LY289612 (a peptidomimetic inhibitor). As their base design they retainedonly the non-peptidic hydroxyethyl-t-butylbenzamide portion of the original inhibitor. Fromhere, the researchers synthesized and attached a variety of substituents to fit into the variousactive site pockets of the viral protease. By repeatedly solving the co-crystal structure of theprotease and a newly designed non-peptide inhibitor (seven in all), they were able to improveupon their design and maximize binding affinities with each repetition. Further characterizationin vivo showed improved bioavailability of up to 30% and maintenance in plasma levels, at orabove, the antiviral IC90 for five or more hours in rats, dogs, and monkeys.A different approach to designing non-peptide inhibitors involves the replacement of thewater molecule at the catalytic site with a component of the inhibitor itself. In designing thesecompounds computer modeling techniques have been employed to generate a class of cyclicinhibitors (i.e. XM323). In XM323 and analogous compounds, the inhibitor’s carbonyl oxygenis designed to displace the water molecule and hydrogen bond to Ile50. Cocrystalization ofprotease with the inhibitor confirmed computer generated predictions. While XM323 has 50%bioavailability and drug levels in plasma exceeds the IC90 for 4 to 16 hours, Phase I clinical trialswere abandoned in 1993 in favor of the derivative XM412 which was found to be more watersoluble and bioavailable (West 1995).As the above example demonstrates, in recent years computer simulation of molecularactivity has become a powerful tool in rational drug design. Because of


View Full Document

Stanford BIO 118 - Study Notes

Documents in this Course
Surrogacy

Surrogacy

14 pages

Load more
Download Study Notes
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Study Notes and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Study Notes 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?