I had a WU from batch 10340

If my math is good we should be almost half of experiment 31

Yesterday the breaking news rolled up in scientific community:
"researchers had grown a crystal that enabled them to see the structure of an enzyme called integrase"

FAAH is going to test docking with the model of integrase:
"Grant number 3P01GM083658-0251 will enable us to perform virtual High-Throughput Screens against our
brand new models of the catalytic core domain of HIV integrase"

So I just wonder if are you going to update the model before running tests and then if we have to rerun some tests if old model was already used?
Keep up excelent work!

Dear Alex,
Always a pleasure to assist in your efforts. I just have received the preliminary PDB file of the discovered protease, what is your email to send it to you?

Keep up great work
Wojciech

To follow up on our previous discussion regarding the new Prototype Foamy Virus integrase crystal structures published in Nature: we do not need to update our models of HIV integrase before targeting them on FightAIDS@Home, but our new modeling protocols could help further refine the new PFV integrase crystal structures. Our new models were published recently; the paper can be found at:
http://www.ncbi.nlm.nih.gov/pubmed/20096702

A member of World Community Grid (wojciech_czyz) obtained early access to their pdb files of PFV integrase, and he sent me a copy. After studying those pdb files for a couple hours, I think that their new crystal structures of PFV integrase could be useful when searching for new scaffolds to bis chelate the two Mg ions in the active site (that is, after minimizing their new crystal structures with our new NMR-type restraint protocol applied to the magnesium--oxygen interactions). Finding these new scaffolds should help advance the development of inhibitors against the G140S/Q148H drug-resistant mutant. I most likely won't use their crystal structures for lead optimization purposes (i.e., for trying to improve a weakly binding hit into a much more potent, drug-like compound), but their new crystal structures could be helpful when using fragments to search for new hits against that drug-resistant mutant.

To advance the development of HIV integrase inhibitors, I advise using proper models of HIV integrase, followed by experimental assays against HIV integrase. PFV integrase has a different secondary structure and different residues composing an active site with a different overall shape. Some aspects of the PFV integrase crystal structures are relevant to the HIV integrase system (as discussed below), but the precise atomic details of the entire active sites are definitely not identical.

In that Nature paper on PFV integrase, they make an interesting statement on p. 3:
"Although the crystals could be grown in the presence of MgCl2, which considerably improved their diffraction limit, data resolution did not allow unambiguous visualization of Mg2+ cations in the active site."
-As a result, they still have some improper bidentate coordination geometries in their crystal structures (some of which contain Zn in the active site, instead of Mg..............Zn has very different preferences regarding coordination geometries). The precise locations and interactions displayed by the two catalytic magnesium ions are a critical issue.

On a positive note: "This result confirmed the expected two-metal binding mode of retroviral INs." Their data on bis chelation of the 2 active site metals agrees well with our AutoDock-based predictions.

Since the DNA-bound PFV integrase crystal structures have a closed conformation of the active site loop, this suggests that the closed conformations of the 140s loop in HIV integrase are likely the active ones. When searching for allosteric inhibitors against the E92Q/N155H mutant (see Experiment 33 on the Status page), I'll target the snapshots that have the most open conformations of the 140s loop, to try to develop "first-in-class" 3' processing inhibitors. A collaborator in Prof. John Elder's lab at TSRI has already developed a good 3' processing assay for HIV integrase. :)

Their data on the induced fit caused by drug binding supports the importance of our conclusions on the more rigid G140S/Q148H mutant of HIV integrase.

PFV integrase seems to have a catalytic Tyr in the active site, while in the HIV integrase system, His67 is more likely to be the catalytic residue. See our discussion in JMB on the infectivity of Y143 mutants versus mutants of H67.


Thus, I am still confident that we are targeting the best representations of HIV integrase in our new experiments on FightAIDS@Home.

Thank you all very much for your interest and your help,
Alex L. Perryman, Ph.D.