Problem Set #7: Catalysis and Inhibitors: HIV Protease

Due November 4 at the beginning of class



1.HIV protease is member of the aspartic protease family, which means that the active site contains a pair of aspartates.  Propose a mechanism for the protease activity and draw out the steps below. (hint: there is no acyl intermediate)





















2. HIV protease cleaves the gag and pol “polyproteins,” releasing matrix, capsid, protease, reverse transcriptase, integrase, and other proteins.   Interstingly, the substrate cleavage sites have little apparent sequence similarity.  Below, draw out the natural substrate SQNY*PIVQ (the * indicates the bond that is cleaved) and show what kinds of interactions might be present in the enzyme active site to recognize, stabilize, read out, and cleave the substrate peptide.























3. Fortunately for us, many structures have been solved of HIV protease.  Some have been solved without substrate, while others have been solved with inhibitors bound, and still others have been solved for mutant versions of HIV protease with bound substrate.  Why are there no structures of wild-type HIV protease with substrate bound?


4. One of the first HIV protease inhibitor structures was that published in Science in December 1989 using a small, modified-peptide inhibitor.  Find this file in the PDB (4HVP.pdb), download it onto your computer, and look at it with SwissPDB Viewer. Convince yourself that HIV protease is a homodimer formed almost entirely of beta strands.  The substrate peptide binds at the interface between the two subunits, covered by a “flap.”   As a result, the binding site is fairly symmetrical, with catalysis being performed by Asp 25 and 25’, hydrogen bonding occuring with Gly 27 and 27’ etc. (the primes indicate identical residues on two different subunits).   


Use SwissPDB Viewer to highlight and explore the interactions between the inhibitor and the substrate. You may want to use some of the following functions to enhance your visualization of these interactions:

  1. To reduce confusion, you may want to use the control panel to “show” only the inhibitor, or at the very least show the protein backbone as only a ribbon.
  2. Next, draw a van der Waals surface around the inhibitor.  In the control panel, find the four dots with the arrow underneath—click on the arrow and select “VDW”, then put a check in that column for whatever groups you want a VDW surface around.
  3. In the control panel, select the inhibitor.
  4. in the Select menu, pick “neighbors of a selected a.a.”  and try different distances.  Neighboring side chains should appear AND should be checked off in the control panel window.  If you check the “labl” column in the control panel for the selected neighbors, their names will appear.
  5. In the Tools menu, select “compute H-bonds”.
  6. In the Display menu, select “show H-bonds” and “show H-bond distances”
  7. Use the Tools panel to zoom in and rotate.


Then, on the next page, fill in the picture of the hydrogen bond interactions between the substrate peptide and the protease with the names and numbers of each of the side chains, and the lengths in angstroms of the hydrogen bonds.  Some of them have been done for you already!


Also, identify with a big arrow the location on the substrate at which catalysis should occur.  What is unusual about the substrate at the central “peptide bond,” and how does the inhibitor resemble the transition state? Why is this related to its function as an inhibitor?









Viral resistance to HIV protease inhibitors has been associated with mutations at postitions 82 and 84 on the protein.  Where are these residues located relative to the inhibitor or substrate, and what effect would these mutations have on the enzyme’s activity?










5.   The structures of a newer generation of HIV protease inhibitors all complexed to the protein were published together in 1999 in Biochemistry.  Based on the figure on the next page, which of these “macrocyclic peptidomimetic inhibitors” binds to HIV protease most strongly? 






Looking at their structures, what IS a “macrocyclic peptidomimetic inhibitor”? What advantages might a cyclic inhibitor have over a linear one?








All of these inhibitors have a hydroxyl group in place of a carbonyl at the peptide bond cleavage site.  In what ways might this substitution interfere with normal catalysis in the active site?








PDB files 1b6j,1b6k, 1b6l, 1b6m, 1b6n, 1b6o, and 1b6p.pdb