Nav view search

Navigation

Search

Current and future GO FAM experiments

 

(% completion statistics from July 10, 2013)

In general, most scientific researchers closely guard the details of their current and future experiments.  It's all kept secret, until the research has been completed and published.  Revealing the finer details of future experiments is never done, because researchers don't want other scientists to "scoop them" and publish the research before they do.  But we will ignore that rule--advancing the fight against multi-drug-resistant mutant superbugs of malaria is far too important; keeping everything secret would slow down our progress.  In addition, by disclosing the details of our current and future GO FAM experiments, we are hoping to motivate and inspire other malaria researchers to plan their projects in a way that can complement the research we are performing on World Community Grid, instead of trying to compete against it. 

To the other computational malaria researchers:  please use different libraries of compounds, different docking software, and/or different target structures in your research against malaria, so that we can explore a much larger percentage of "chemical space" in as many ways as possible.  All of our GO Fight Against Malaria data are in the public domain--let's help each other out, instead of "stepping on each others' toes."  In addition, we do take requests:  if you want us to screen ~ 5.6 million compounds against a particular malaria target that is not listed below, then please contact ALP.  If the structural data are available, we will do our best to include the requested target.

To the experts in performing biological assays against the malaria parasite and/or against specific enzymes from Plasmodium falciparum:  please help us test our predictions.  If you have experience performing assays against one of the targets listed below, please let us know, and we can try to start collaborating together.  If you happen to know of an important drug target from malaria that is not included in the list below, then please share that information with us, and we'll try to include it in our future plans.  If you are aware of strong evidence which indicates that one of the targets below is not a useful target for fighting malaria infections, then please let us know that information, as well.

Detailed description of the experiments involved in the Global Online Fight Against Malaria

The numbers correspond to the "batch identification numbers." If you are contributing computer time to the GO FAM project, then click on the "advanced view" option, then click on the "tasks" tab, and you can see which particular experiment you are helping us advance.  Click on the "Details of the GO FAM project" link in the menu in the upper-left corner to learn more about these "docking" calculations, why we are performing them, and what naming convention we use for the jobs performed on World Community Grid.

 

We are organizing this information according to the type of target it involves, instead of just organizing it into a series of sequential "experiments."  Each "experiment" involves docking one entire library of compounds against one particular class of drug target (that is, against one particular type of enzyme).

 

Some of the information below is in an abbreviated, short-hand form.  I will add more details and additional positive control data for these experiments in the future.  This site is still a work-in-progress.  Please be patient.

 

See the following review article for more information about these targets:  Jana, S. & Paliwal, J. Novel molecular targets for antimalarial chemotherapy. Int. J Antimicrobial Agents. 30, 4-10 (2007).  Many of the quotes below are from this review article (or they are from the article that describes the specific PDB entry for the structure of that target).

 

Target #1 = DHFR

Dihydrofolate reductase, or "DHFR", is a very well-validated drug target for malaria.  DHFR inhibitors (such as pyrimethamine) have been used to cure malaria infections for decades, but multi-drug-resistant mutant superbugs that are not killed by DHFR inhibitors keep arising and spreading throughout the world.  The quadruple mutant superbug version of Pf DHFR is the most multi-drug-resistant mutant of this enzyme found in clinical settings--we're targeting the crystallographic structure of this superbug from 1j3k.pdb (which was crystallized with the potent inhibitor WR-99210---see the bottom of the GO FAM homepage for more information about this inhibitor, and see the images below for a positive control experiment that successfully predicted/reproduced the way in which this compound binds to this drug target).  The version of DHFR from extremely-drug-resistant tuberculosis also happens to bind well to the compound WR-99210, which is why we are including TB DHFR in this GO FAM experiment (that is, compounds that score well against both Pf DHFR and TB DHFR could show a lot of promise against both diseases).


Batch numbers against DHFR:

(each "batch" involves docking 10,000 different compounds against a single model of one target)

Experiment 1 = 100% completed

0 = multi-drug-resistant quadruple mutant of Pf DHFR from 1j3k.pdb (with 2 crystallographic water molecules included in the target) versus the compound library called "NCI Diversity Set II"

1 = Wild type Pf DHFR from 1j3i.pdb (with 2 crystallographic water molecules included in the target) versus the compound library called "NCI Diversity Set II"

 

Experiment 2 = 100% completed

[full NCI library = 316,179 models of compounds]
2 - 33  = Quadruple mutant of Pf DHFR from 1j3k (with 2 x-tal waters) versus full NCI library
34 - 65 = Wt Pf DHFR from 1j3i (with 2 x-tal waters) versus full NCI library
66 - 97   = Quadruple mutant of Pf DHFR from 1j3k (dry) versus full NCI library
98 - 129  = Wt Pf DHFR from 1j3i (dry) versus full NCI library
130 - 161 = P vivax DHFR from 2BL9 (dry) versus full NCI library
162 - 193 = Human DHFR from 1KMV (dry) versus full NCI library
194 - 225 = Human DHFR from 3DFR (dry+WHAT_IF Hs) vs full NCI library
                 **note:  all other targets have MolProbity's H's,
                   but 3DFR had a residue with missing atoms near the active site,
                   which WHAT_IF was able to fix (but MolProbity didn't)
226 - 257 = TB DHFR from 1DF7 (dry) versus full NCI library
258 - 289 = TB DHFR from 1DG7 (dry) versus full NCI library

 

Experiment 3 = 100% completed

[Enamine library= 2,345,014 models of compounds]
290 - 524 = Quadruple mutant of Pf DHFR from 1j3k (with 2 x-tal waters) versus full Enamine library
525 - 759 = Wt Pf DHFR from 1j3i (with 2 x-tal waters) versus full Enamine library
760 - 994     = Quadruple mutant of Pf DHFR from 1j3k (dry) versus full Enamine library
995 - 1229   = Wt Pf DHFR from 1j3i (dry) versus full Enamine library
1230 - 1464 = P vivax DHFR from 2BL9 (dry) versus full Enamine library
1465 - 1699 = Human DHFR from 1KMV (dry) versus full Enamine library
1700 - 1934 = Human DHFR from 3DFR (dry+WHAT_IF Hs) vs full Enamine library
1935 - 2169 = TB DHFR from 1DF7 (dry) versus full Enamine library
2170 - 2404 = TB DHFR from 1DG7 (dry) versus full Enamine library

 

Experiment 4 = 100% completed

[full ChemBridge library = 1,013,483 models of compounds]
2405 - 2506 = Quadruple mutant of Pf DHFR from 1j3k (with 2 x-tal waters) versus full ChemBridge library
2507 - 2608 = Wt Pf DHFR from 1j3i (with 2 x-tal waters) versus full ChemBridge library
2609 - 2710 = Quadruple mutant of Pf DHFR from 1j3k (dry) versus full ChemBridge library
2711 - 2812 = Wt Pf DHFR from 1j3i (dry) versus full ChemBridge library
2813 - 2914 = P vivax DHFR from 2BL9 (dry) versus full ChemBridge library
2915 - 3106 = Human DHFR from 1KMV (dry) versus full ChemBridge library
3017 - 3118 = Human DHFR from 3DFR (dry+WHAT_IF Hs) vs full ChemBridge library
3119 - 3220 = TB DHFR from 1DF7 (dry) versus full ChemBridge library
3221 - 3322 = TB DHFR from 1DG7 (dry) versus full ChemBridge library


++++skipped to ENR targets for several experiments, now jumping back to DHFR++++++++

 

Experiment 8 = 100% completed

[Asinex library = 507,000 models of compounds]
7013 - 7063 = Quadruple mutant of Pf DHFR from 1j3k (with 2 x-tal waters) versus Asinex library
7064 - 7114 = Wt Pf DHFR from 1j3i (with 2 x-tal waters) versus Asinex library
7115 - 7165 = Quadruple mutant of Pf DHFR from 1j3k (dry) versus Asinex library
7166 - 7216 = Wt Pf DHFR from 1j3i (dry) versus Asinex library
7217 - 7267 = P vivax DHFR from 2BL9 (dry) versus Asinex library
7268 - 7318 = Human DHFR from 1KMV (dry) versus Asinex library
7319 - 7369 = Human DHFR from 3DFR (dry+WHAT_IF Hs) vs Asinex library
7370 - 7420 = TB DHFR from 1DF7 (dry) versus Asinex library
7421 - 7471  = TB DHFR from 1DG7 (dry) versus Asinex library


++++skipped to ENR targets for one experiment, now jumping back to DHFR++++++++


Experiment 10 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
7982 - 8132 = Quadruple mutant of Pf DHFR from 1j3k (with 2 x-tal waters) versus VitasMLabs library
8133 - 8283 = Wt Pf DHFR from 1j3i (with 2 x-tal waters) versus VitasMLabs library
8284 - 8434 = Quadruple mutant of Pf DHFR from 1j3k (dry) versus VitasMLabs library
8435 - 8585 = Wt Pf DHFR from 1j3i (dry) versus VitasMLabs library
8586 - 8736 = P vivax DHFR from 2BL9 (dry) versus VitasMLabs library
8737 - 8887 = Human DHFR from 1KMV (dry) versus VitasMLabs library
8888 - 9038 = Human DHFR from 3DFR (dry+WHAT_IF Hs) vs VitasMLabs library
9039 - 9189 = TB DHFR from 1DF7 (dry) versus VitasMLabs library
9190 - 9340 = TB DHFR from 1DG7 (dry) versus VitasMLabs library


**note: since WR-99210 is known to inhibit both Pf DHFR and TB DHFR, we are screening compounds against the tuberculosis target and the malaria target.  Compounds that score well against both Pf DHFR and TB DHFR should be promising hits.

**However, we don't want these hits to score well against human DHFR; thus, the human DHFR targets are being included as a "negative design" element (but the top ligands versus human DHFR might help advance cancer research and/or drugs to prevent rejection of transplanted organs)

**note:  all other targets have MolProbity's H's, but 3DFR had a residue with missing atoms near the active site, which WHAT_IF (wif) was able to fix (but MolProbity didn't).

 

These first two images show a "positive control" experiment.  This experiment demonstrates that we can accurately predict/reproduce the specific way that a known inhibitor binds to this important drug target, Pf DHFR (for Plasmodium falciparum's dihydrofolate reductase).  The known, experimentally-determined, X-ray crystallographic binding mode for this inhibitor, WR-99210, is shown in magenta, while the binding mode that we predicted with Vina calculations is shown in cyan.  The left image shows a critical residue in the active site as magenta ball-and-sticks (while the inhibitor is shown in thicker stick mode, without the balls), while the right image shows this critical residue of the target as gray ball-and-sticks.

 


 

 

Target #2 = ENR

Enoyl-acyl-carrier-protein reductase  (or "ENR," which is also called Fab I) is part of a unique metabolic pathway in the apicoplast (that is, it's an enzyme that is part of a pathway that is not present in humans).  ENR is part of FAS (a Fatty Acid Synthesis pathway); FAS inhibitors are used to treat many different types of pathogens.  For example, isoniazid (or "Inh") is a front-line treatment for tuberculosis that inhibits the ENR enzyme from Mycobacterium tuberculosis.  The ENR enzyme from TB is called "InhA".  Pf ENR is a validated drug target for the development of anti-malarial compounds:  triclosan has been shown to kill malaria parasites via its inhibition of Pf ENR.  However, triclosan is not "orally bioavailable" (that is, similar to WR-99210, triclosan's chemical properties mean that it can never be turned into a pill, which is why we need to find newer, better inhibitors of this target that can actually be turned into pills to treat malaria patients).

 

Batch numbers against ENR:

Experiment 5 = 100% completed

[full NCI library = 316,179 models of compounds]
3323 - 3354 = Pf ENR (enoyl-acyl-carrier-protein reductase, also known as Fab I) from 1uh5.pdb vs full NCI
3355 - 3386 = Pf ENR from 1vrw.pdb versus full NCI library
3387 - 3418 = TB ENR (InhA) from 2aq8.pdb versus full NCI library
3419 - 3450 = TB ENR (InhA) I21V drug-resistant mutant from 2aqh.pdb versus full NCI library
3451 - 3482 = TB ENR (InhA) I47T drug-resistant mutant from 2aqi.pdb versus full NCI library
3483 - 3514 = TB ENR (InhA) S94A drug-resistant mutant from 2aqk.pdb versus full NCI library
3515 - 3546 = Pf ENR from 2oos.pdb versus full NCI library
3547 - 3578 = TB ENR (InhA) from 2x23.pdb versus full NCI library
3579 - 3610 = Pf ENR from 3lt0.pdb versus full NCI library
3611 - 3642 = Pf ENR from 3lt4.pdb versus full NCI library

 

Experiment 6 = 100% completed

[Enamine library = 2,345,014 models of compounds]
3643 - 3877 = Pf ENR (also known as Fab I) from 1uh5.pdb vs full Enamine library
3878 - 4112 = Pf ENR from 1vrw.pdb versus full Enamine library
4113 - 4347 = TB ENR (InhA) from 2aq8.pdb versus full Enamine library
4348 - 4582 = TB ENR (InhA) I21V drug-resistant mutant from 2aqh.pdb versus full Enamine library
4583 - 4817 = TB ENR (InhA) I47T drug-resistant mutant from 2aqi.pdb versus full Enamine library
4818 - 5052 = TB ENR (InhA) S94A drug-resistant mutant from 2aqk.pdb versus full Enamine library
5053 - 5287 = Pf ENR from 2oos.pdb versus full Enamine library
5288 - 5522 = TB ENR (InhA) from 2x23.pdb versus full Enamine library
5523 - 5757 = Pf ENR from 3lt0.pdb versus full Enamine library
5758 - 5992 = Pf ENR from 3lt4.pdb versus full Enamine library

Experiment 7 = 100% completed

[full ChemBridge library = 1,013,483 models of compounds]
5993 - 6094 = Pf ENR (also known as Fab I) from 1uh5.pdb vs full ChemBridge library
6095 - 6196 = Pf ENR from 1vrw.pdb versus full ChemBridge library
6197 - 6298 = TB ENR (InhA) from 2aq8.pdb versus full ChemBridge library
6299 - 6400 = TB ENR (InhA) I21V drug-resistant mutant from 2aqh.pdb versus full ChemBridge library
6401 - 6502 = TB ENR (InhA) I47T drug-resistant mutant from 2aqi.pdb versus full ChemBridge library
6503 - 6604 = TB ENR (InhA) S94A drug-resistant mutant from 2aqk.pdb versus full ChemBridge library
6605 - 6706 = Pf ENR from 2oos.pdb versus full ChemBridge library
6707 - 6808 = TB ENR (InhA) from 2x23.pdb versus full ChemBridge library
6809 - 6910 = Pf ENR from 3lt0.pdb versus full ChemBridge library
6911 - 7012 = Pf ENR from 3lt4.pdb versus full ChemBridge library



++++skipped back to DHFR targets for 1 experiment, now jumping back to ENRs++++++++

Experiment 9 = 100% completed

[Asinex library = 507,000 models of compounds]
7472 - 7522 = Pf ENR (also known as Fab I) from 1uh5.pdb vs Asinex library
7523 - 7573 = Pf ENR from 1vrw.pdb versus Asinex library
7574 - 7624= TB ENR (InhA) from 2aq8.pdb versus Asinex library
7625 - 7675 = TB ENR (InhA) I21V drug-resistant mutant from 2aqh.pdb versus Asinex library
7676 - 7726 = TB ENR (InhA) I47T drug-resistant mutant from 2aqi.pdb versus Asinex library
7727 - 7777 = TB ENR (InhA) S94A drug-resistant mutant from 2aqk.pdb versus Asinex library
7778 - 7828 = Pf ENR from 2oos.pdb versus Asinex library
7829 - 7879 = TB ENR (InhA) from 2x23.pdb versus Asinex library
7880 - 7930 = Pf ENR from 3lt0.pdb versus Asinex library
7931 - 7981 = Pf ENR from 3lt4.pdb versus Asinex library



++++skipped back to DHFR targets for 1 experiment, now jumping back to ENRs++++++++


Experiment 11 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
9341 - 9491 = Pf ENR (also known as Fab I) from 1uh5.pdb vs VitasMLabs library
9492 - 9642 = Pf ENR from 1vrw.pdb versus VitasMLabs library
9643 - 9793 = TB ENR (InhA) from 2aq8.pdb versus VitasMLabs library
9794 - 9944 = TB ENR (InhA) I21V drug-resistant mutant from 2aqh.pdb versus VitasMLabs library
9945 - 10095 = TB ENR (InhA) I47T drug-resistant mutant from 2aqi.pdb versus VitasMLabs library
10096 - 10246 = TB ENR (InhA) S94A drug-resistant mutant from 2aqk.pdb versus VitasMLabs library
10247 - 10397 = Pf ENR from 2oos.pdb versus VitasMLabs library
10398 - 10548 = TB ENR (InhA) from 2x23.pdb versus VitasMLabs library
10549 - 10699 = Pf ENR from 3lt0.pdb versus VitasMLabs library
10700 - 10850 = Pf ENR from 3lt4.pdb versus VitasMLabs library



           

**note: since triclosan is known to inhibit both Pf ENR and TB ENR (InhA), we are screening compounds against the tuberculosis target and the malaria target.  Compounds that score well against both Pf ENR and TB ENR could be very promising hits against both TB and malaria.

*since ENR is part of a FAS path not present in humans, ENR inhibitors should be less toxic
         *this is the reason why no human ENR targets were included (i.e., no human ENR's exist)

*note: all ENR targets = dry (no crystallographic waters were included in the targets)

 

These two images show that we can also use "AutoDock Vina" to correctly predict/reproduce the detailed "binding mode" that a known inhibitor uses to interact with the drug target "Pf ENR" (for Plasmodium falciparum's enoyl-acyl-carrier-protein reductase).  The know, experimentally-determined, X-ray crystallographic binding mode of this inhibitor is shown in magenta, while the binding mode predicted by Vina is shown in cyan.  The solvent-excluded surface of ENR is shown in green.  In the image on the right, part of this surface was clipped away ("undisplayed") to give us a better view of this inhibitor's binding mode.  In this image, the backbone of ENR is displayed as green "ribbons".

 

 


 


Target #3 =  HGPRTase

Hypoxanthine-guanine-xanthine phosphoribosyltransferase (or "HGPRTase" or "HGXPRTase") is a potential drug target for the treatment of multi-drug-resistant malaria infections.  The Plasmodium parasites lack the "de novo purine biosynthetic pathway."  That is, unlike humans, the malaria parasites are not able to make their own purine nucleotides (adenine, A, and guanine, G) from scratch--they must ingest them.  Starving the malaria parasites of purines kills malaria in cultured cells.  To obtain the purines (the A and G that are key components of DNA) that the parasites need to survive and replicate, they use HGPRTase to salvage/harvest these nucleotides from their food.  Although humans can synthesize these nucleotides from scratch, people who have a complete lack of HGPRTase activity get the disease called Lesch-Nyhan syndrome, which causes hyperuricemia and neural disorders.  Similarly, people who only have partially active HGPRTase enzymes tend to get gouty arthritis.  Since the lack of sufficient human HGPRTase activity causes health problems, we want to discover candidate compounds that only inhibit the HGPRTase enzyme from the parasite, without significantly impeding the activity of the human version of the enzyme.  To advance this goal, we will screen several million compounds against both the Plasmodium falciparum HGPRTase as well as the human HGPRTase, and we will harvest the compounds that score well against the malaria enzyme while also scoring poorly against the human enzyme.

 

   

The above image demonstrates that we can successfully reproduce/predict the known binding mode of an HGPRTase inhibitor.  The backbone of the target HGPRTase is shown with purple "ribbons," and the "CPK" spheres are displayed for the two magnesium ions (in green), the diphosphate (in red and orange), and a key aspartate in the active site (colored by atom type, with purple carbon atoms).  The known, experimentally-determined, X-ray crystallographic binding mode of this inhibitor "IMU" is shown in stick mode with purple carbon atoms, while the binding mode predicted by Vina is shown as sticks with cyan carbon atoms.  This "positive control" experiment successfully reproduced the known binding mode in 1BZY.pdb, which is the structure of the human version of HGPRTase.  On GO FAM we want to discover compounds that will score well against the versions of HGPRTase from malaria and that simultaneously score poorly against the human version.

 

 

Batch numbers against HGPRTase:

Experiment 12 = 100% completed

[full NCI library = 316,179 models of compounds]
10851 - 10882 = human HGPRTase from 1bzy.pdb (chain A) with POP cofactor deleted versus full NCI library
10883 - 10914 = human HGPRTase from 1bzy.pdb (chain A) versus full NCI library
10915 - 10946 = Pf HGPRTase from 1cjb.pdb (chain A) with POP cofactor deleted versus full NCI library
10947 - 10978 = Pf HGPRTase from 1cjb.pdb (chain A) versus full NCI library
10979 - 11010 = Pf HGPRTase from 3ozg.pdb (chain C) with POP cofactor deleted versus full NCI library
11011 - 11042 = Pf HGPRTase from 3ozg.pdb (chain C) versus full NCI library

 
Experiment 13 = 100% completed

[Enamine library = 2,345,014 models of compounds]
11043 - 11277 = human HGPRTase from 1bzy.pdb (chain A) with POP cofactor deleted versus Enamine
11278 - 11512 = human HGPRTase from 1bzy.pdb (chain A) versus Enamine library
11513 - 11747 = Pf HGPRTase from 1cjb.pdb (chain A) with POP cofactor deleted versus Enamine library
11748 - 11982 = Pf HGPRTase from 1cjb.pdb (chain A) versus Enamine library
11983 - 12217 = Pf HGPRTase from 3ozg.pdb (chain C) with POP cofactor deleted versus Enamine library
12218 - 12452 = Pf HGPRTase from 3ozg.pdb (chain C) versus Enamine library

 

Experiment 14 = 100% completed

[full ChemBridge library = 1,013,483 models of compounds]
12453 - 12554 = human HGPRTase from 1bzy.pdb (chain A) with POP cofactor deleted versus ChemBridge
12555 - 12656 = human HGPRTase from 1bzy.pdb (chain A) versus ChemBridge library
12657 - 12758 = Pf HGPRTase from 1cjb.pdb (chain A) with POP cofactor deleted versus ChemBridge library
12759 - 12860 = Pf HGPRTase from 1cjb.pdb (chain A) versus ChemBridge library
12861 - 12962 = Pf HGPRTase from 3ozg.pdb (chain C) with POP cofactor deleted versus ChemBridge library
12963 - 13064 = Pf HGPRTase from 3ozg.pdb (chain C) versus ChemBridge library

 
Experiment 15 = 100% completed

[Asinex library = 507,000 models of compounds]
13065 - 13115 = human HGPRTase from 1bzy.pdb (chain A) with POP cofactor deleted versus Asinex
13116 - 13166 = human HGPRTase from 1bzy.pdb (chain A) versus Asinex library
13167 - 13217 = Pf HGPRTase from 1cjb.pdb (chain A) with POP cofactor deleted versus Asinex library
13218 - 13268 = Pf HGPRTase from 1cjb.pdb (chain A) versus Asinex library
13269 - 13319 = Pf HGPRTase from 3ozg.pdb (chain C) with POP cofactor deleted versus Asinex library
13320 - 13370 = Pf HGPRTase from 3ozg.pdb (chain C) versus Asinex library


Experiment 16 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
13371 - 13521 = human HGPRTase from 1bzy.pdb (chain A) with POP cofactor deleted versus Vitas-M Labs
13522 - 13672 = human HGPRTase from 1bzy.pdb (chain A) versus Vitas-M Labs library
13673 - 13823 = Pf HGPRTase from 1cjb.pdb (chain A) with POP cofactor deleted versus Vitas-M Labs library
13824 - 13974 = Pf HGPRTase from 1cjb.pdb (chain A) versus Vitas-M Labs library
13975 - 14125 = Pf HGPRTase from 3ozg.pdb (chain C) with POP cofactor deleted versus Vitas-M Labs library
14126 - 14276 = Pf HGPRTase from 3ozg.pdb (chain C) versus Vitas-M Labs library

 


3OZG.pdb = Pf HGXPRTase with 2nd gen inhibitor = S-SerMe-ImmH phosphonate
1CJB.pdb = Pf HGPRTase with transition-state analog inhibitor
1BZY.pdb = human HGPRTase with transition-state analog inhibitor (same authors & citation as 1CJB)

 

The above images also display the results of "positive control" docking experiments against HGPRTase, but these two images depict the version of HGPRTase from Plasmodium falciparum (with white "ribbons" to show its backbone).  On the left, the known, experimentally-determined, X-ray crystallographic binding mode of the inhibitor "IRP" from 1CJB.pdb is shown as sticks with gray carbon atoms, while the binding mode predicted by Vina has magenta carbon atoms.  The "CPK" spheres in both images show the magnesium ions (in green), the diphosphate (in orange and red), and a key aspartate (in gray and red).  On the right, the X-ray crystallographic binding mode of the inhibitor "SSI" is depicted as sticks with light magenta carbon atoms, while the binding mode predicted by Vina has green carbon atoms.

 

 


 

 
Target #4 =  PMT

Phosphoethanolamine methyltransferase is a potential drug target that was suggested by a member of World Community Grid on the GO FAM Forum.  On Jan. 7, 2012, l_mckeon posted a press release on the GO FAM Forum that led me to the paper from Soon Goo Lee, et al.  J. Biol. Chem. 287:1426-1434 (2012), which was a "paper of the week" for JBC.  This paper presented the first atomic-resolution "x-ray crystallographic" structure of this promising drug target.

Phosphoethanolamine methyltransferase is a key enzyme that the malaria parasite needs to build its membranes.  The phosphobase methylation path in the Plasmodium parasites is similar to the path that plants use, but this pathway is not found in mammals.  Since this pathway is not found in humans, inhibiting this malarial enzyme could potentially cause fewer toxic side effects than those that might result from inhibiting other proteins involved in malaria infections. 

This PMT enzyme catalyzes 3 sequential steps:

CH3 + pEA --> pMME + CH3 --> pDME + CH3 --> pCho

That is, PMT modifies phosphoethanolamine in 3 separate reactions that occur in sequence in order to produce phosphocholine.  This phosphocholine product is the precursor for the generation of phosphatidyl choline--a key phospholipid that is necessary to form the membranes in this parasite (i.e., it is needed for the parasite's "membrane biogenesis" process).  In plants, a bi-functional, two-domain enzyme catalyzes these 3 steps, but in Plasmodium, a single domain PMT performs all three reactions.

When the Pf PMT gene is "knocked out" (removed from the parasite's genome), it completely abolishes phosphatidyl choline synthesis via the primary metabolic route (the phosphobase path), and it shows that the secondary/back-up route (the Bremer-Greenberg path) can not compensate for the loss of PMT.  The loss of Pf PMT activity leads to significant defects in growth, reproduction, and viability of the malaria parasites, which is why we are trying to discover new, more potent inhibitors of this drug target.

These two images of "positive control" docking experiments demonstrate that we can use Vina to correctly predict/reproduce the known binding mode of compounds with phosphoethanolamine methyltransferase (PMT, from 3UJ7.pdb).  The image on the left shows the backbone of PMT as light blue "ribbons," and both images display the known, X-ray crystallographic binding mode of the cofactor "SAM" as sticks with light blue carbon atoms.  The sticks with yellow carbon atoms show the binding mode of SAM that was predicted by Vina calculations.  The image on the right shows part of the solvent-excluded surface of PMT in purple, but much of this surface had to be clipped away ("undisplayed") to enable us to view how the cofactor binds to this drug target.



Batch numbers against PMT:

Experiment 17 = 100% completed

[full NCI library = 316,179 models of compounds]
14277 - 14308 = Pf PMT from 3uj7 (SAM_and_PO4-induced; chain B, a conf) versus full NCI library
14309 - 14340 = Pf PMT from 3uj8 (SFG_and_PO4-induced; a conf) versus full NCI library
14341 - 14372 = Pf PMT from 3uj8 (SFG_and_PO4-induced; b conf) versus full NCI library
14373 - 14404 = Pf PMT from 3uj9 (PC_product-induced; b conf) versus full NCI library
14405 - 14436 = Pf PMT from 3uja (OPE_substrate-induced; a conf) versus full NCI library
14437 - 14468 = Pf PMT from 3ujb (OPE_substrate and SAH-induced; chain B, a conf) versus full NCI library
14469 - 14500 = Pf PMT from 3ujb with OPE substrate present (chain B, a conf) versus full NCI library
14501 - 14532 = Pf PMT from 3ujb with SAH_cofactor_End-state present (chain B, a conf) versus full NCI library

 

SAM = S-adenosyl-methionine, which is also called AdoMet.  SAM is a cofactor that the PMT enzyme (and other types of enzymes) needs in order to function properly.  After this cofactor has been used, its final/end state is called SAH (S-adenosyl-homocysteine).

PC = phosphocholine = pCho, which is the final product produced by PMT.

OPE = phosphoethanolamine = pEA, which is the substrate that PMT modifies in order to produce PC.

SFG = sinefungin = an inhibitor.  Sinefungin's structure is very similar to SAM, but SFG has a C-NH3+ instead of an +S-CH3.


Experiment 18 = 100% completed

[Enamine library = 2,345,014 models of compounds]
14533 - 14767 = Pf PMT from 3uj7 (SAM_and_PO4-induced; chain B, a conf) versus Enamine library
14768 - 15002 = Pf PMT from 3uj8 (SFG_and_PO4-induced; a conf) versus Enamine library
15003 - 15237 = Pf PMT from 3uj8 (SFG_and_PO4-induced; b conf) versus Enamine library
15238 - 15472 = Pf PMT from 3uj9 (PC_product-induced; b conf) versus Enamine library
15473 - 15707 = Pf PMT from 3uja (OPE_substrate-induced; a conf) versus Enamine library
15708 - 15942 = Pf PMT from 3ujb (OPE_substrate and SAH-induced; chain B, a conf) versus Enamine library
15943 - 16177 = Pf PMT from 3ujb with OPE substrate present (chain B, a conf) versus Enamine library
16178 - 16412 = Pf PMT from 3ujb with SAH_cofactor_End-state present (chain B, a conf) versus Enamine library


Experiment 19 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
16413 - 16514 = Pf PMT from 3uj7 (SAM_and_PO4-induced; chain B, a conf) versus ChemBridge library
16515 - 16616 = Pf PMT from 3uj8 (SFG_and_PO4-induced; a conf) versus ChemBridge library
16617 - 16718 = Pf PMT from 3uj8 (SFG_and_PO4-induced; b conf) versus ChemBridge library
16719 - 16820 = Pf PMT from 3uj9 (PC_product-induced; b conf) versus ChemBridge library
16821 - 16922 = Pf PMT from 3uja (OPE_substrate-induced; a conf) versus ChemBridge library
16923 - 17024 = Pf PMT from 3ujb (OPE_substrate and SAH-induced; chain B, a conf) versus ChemBridge library
17025 - 17126 = Pf PMT from 3ujb with OPE substrate present (chain B, a conf) versus ChemBridge library
17127 - 17228 = Pf PMT from 3ujb with SAH_cofactor_End-state present (chain B, a conf) versus ChemBridge library


Experiment 20 = 100% completed

[Asinex library = 507,000 models of compounds]
17229 - 17279 = Pf PMT from 3uj7 (SAM_and_PO4-induced; chain B, a conf) versus Asinex library
17280 - 17330 = Pf PMT from 3uj8 (SFG_and_PO4-induced; a conf) versus Asinex library
17331 - 17381 = Pf PMT from 3uj8 (SFG_and_PO4-induced; b conf) versus Asinex library
17382 - 17432 = Pf PMT from 3uj9 (PC_product-induced; b conf) versus Asinex library
17433 - 17483 = Pf PMT from 3uja (OPE_substrate-induced; a conf) versus Asinex library
17484 - 17534 = Pf PMT from 3ujb (OPE_substrate and SAH-induced; chain B, a conf) versus Asinex library
17535 - 17585 = Pf PMT from 3ujb with OPE substrate present (chain B, a conf) versus Asinex library
17586 - 17636 = Pf PMT from 3ujb with SAH_cofactor_End-state present (chain B, a conf) versus Asinex library


Experiment 21 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
17637 - 17787 = Pf PMT from 3uj7 (SAM_and_PO4-induced; chain B, a conf) versus Vitas-M Labs library
17788 - 17938 = Pf PMT from 3uj8 (SFG_and_PO4-induced; a conf) versus Vitas-M Labs library
17939 - 18089 = Pf PMT from 3uj8 (SFG_and_PO4-induced; b conf) versus Vitas-M Labs library
18090 - 18240 = Pf PMT from 3uj9 (PC_product-induced; b conf) versus Vitas-M Labs library
18241 - 18391 = Pf PMT from 3uja (OPE_substrate-induced; a conf) versus Vitas-M Labs library
18392 - 18542 = Pf PMT from 3ujb (OPE_substrate and SAH-induced; chain B, a conf) versus Vitas-M Labs library
18543 - 18693 = Pf PMT from 3ujb with OPE substrate present (chain B, a conf) versus Vitas-M Labs library
18694 - 18844 = Pf PMT from 3ujb with SAH_cofactor_End-state present (chain B, a conf) versus Vitas-M Labs library



The above two images are also from "positive control" docking experiments against PMT, but this time the X-ray crystallographic binding mode was taken from 3UJ8.pdb.  The backbone of PMT is shown as dark purple "ribbons" (on the left), while the known, experimentally-determined binding mode of the inhibitor sinefungin is depicted as sticks with dark green carbon atoms.  The detailed way in which this inhibitor binds to the target (or the "binding mode") was reproduced by Vina fairly well and is shown as sticks with orange carbon atoms.  The image on the right shows the surface of PMT in dark purple, but much of the surface was clipped away ("undisplayed") so that we could actually see the inhibitors contained within this drug target.

 

 


 

 
Target #5 =  PNP

Purine nucleotide phosphorylase (PNP) is a validated drug target to treat malaria infections.  PNP is a critical enzyme for the "purine salvage mechanisms" that Plasmodium falciparum must use to survive.  To learn more about why the malaria parasites must ingest/salvage purines, see the above description for target #3 = HGPRTase. 

The inhibitor immucillin-H (ImmH) is a transition state analog that has been shown to kill malaria parasites by inhibiting this PNP enzyme.  A derivative of ImmH, 5-methylthio-immucillin-H, kills Plasmodium falciparum strain 3D7 cultures at very low concentrations (that is, concentrations that selectively inhibit the parasite's PNP enzyme but that do not inhibit the human PNP significantly).

 

 

The three images above show the results of positive control experiments which demonstrate that we can accurately dock PNP inhibitors against their respective crystal structures.  These three images show the full view of the PNP targets, while the images below are zoomed in to highlight the agreement between the binding modes predicted by Vina and the experimentally-determined, X-ray crystallographic structure of their actual binding modes.  On the left is the Pf PNP target from 1nw4.pdb; the Vina-docked mode for the inhibitor immucillin-H (Imm-H) has cyan carbon atoms.  In the center is a mutant Pf PNP from 3fow.pdb that contains the Val66Ile_Val73Ile_Tyr160Phe mutations (shown as light purple spheres); the Vina-docked mode for Imm-H has tan carbon atoms.  On the right is the human version of PNP from 1rr6.pdb; the Vina-docked mode of Imm-H has green carbon atoms.  We want to find new inhibitors that bind well to the Pf PNP structures but that do NOT bind well to the human PNP (as a way to try to minimize potential toxic side effects).  However, compounds that do bind well to and strongly inhibit human PNP might be useful in treating T-cell cancer and autoimmune diseases.  Thus, scientists who are working on these human diseases might want a copy of the results of these virtual screens versus human PNP to help them advance their own research.  If you are a scientist who is interested in the results versus human PNP, please contact Dr. Alex Perryman.

 

Kudos to Stefano Forli, Ph.D.!!  The models of the targets that we are using in these GO FAM experiments against PNP, and the positive control re-docking experiments which showed that we can accurately dock compounds against these models, were prepared and performed by Dr. Stefano Forli, a postdoctoral fellow in Prof. Art Olson's lab at TSRI.  Alex L. Perryman, Ph.D., used these data from Stefano on the PNP system to create the molecular images of these positive control experiments for PNP.  Dr. Alex Perryman prepared all of the other GO FAM experiments described above and below, and he created all of the molecular images on this page. 

 

These two images display a close-up of the positive control re-docking experiments against Pf PNP from 1nw4.pdb.  The known, crystallographic binding mode for the inhibitor Imm-H has magenta carbon atoms, while the binding mode predicted by Vina has cyan carbon atoms.  The image on the left shows the solvent-excluded surface of PNP, while the image on the right displays the ribbon/cartoon mode of PNP's backbone.

 

Batch numbers against PNP:

Experiment 22 = 100% completed

[full NCI library = 316,179 models of compounds]
18845 - 18876 = Pf PNP from 1nw4 (ImmH_and_sulfate-induced) versus NCI library
18877 - 18908 = Pf PNP from 1q1g (5'-methylthio-ImmH-induced) versus NCI library
18909 - 18940 = Human PNP from 1rr6 (ImmH_and_phosphate-induced) versus NCI library
18941 - 18972 = Pvivax PNP from 3emv (sulphate-induced) versus NCI library
18973 - 19004 = Pf PNP from 3enz (hypoxanthine_ribose_and_arsenate-induced) versus NCI library
19005 - 19036 = Pf PNP from 3fow = V66I_V73I_Y160F mutant (ImmH_and_phosphate-induced) versus NCI library
19037 - 19068 = Human PNP from 3phb (DADMe-ImmG_and_phosphate-induced) versus NCI library
19069 - 19100 = Pf PNP from 3phc (DADMe-ImmG_and_K_and_phosphate-induced) versus NCI library


Experiment 23 = 100% completed

[Enamine library = 2,345,014 models of compounds]
19101 - 19335 = Pf PNP from 1nw4 (ImmH_and_sulfate-induced) versus Enamine library
19336 - 19570 = Pf PNP from 1q1g (5'-methylthio-ImmH-induced) versus Enamine library
19571 - 19805 = Human PNP from 1rr6 (ImmH_and_phosphate-induced) versus Enamine library
19806 - 20040 = Pvivax PNP from 3emv (sulphate-induced) versus Enamine library
20041 - 20275 = Pf PNP from 3enz (hypoxanthine_ribose_and_arsenate-induced) versus Enamine library
20276 - 20510 = Pf PNP from 3fow = V66I_V73I_Y160F mutant (ImmH_and_phosphate-induced) versus Enamine
20511 - 20745 = Human PNP from 3phb (DADMe-ImmG_and_phosphate-induced) versus Enamine library
20746 - 20980 = Pf PNP from 3phc (DADMe-ImmG_and_K_and_phosphate-induced) versus Enamine library



Experiment 24 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
20981 - 21082 = Pf PNP from 1nw4 (ImmH_and_sulfate-induced) versus ChemBridge library
21083 - 21184 = Pf PNP from 1q1g (5'-methylthio-ImmH-induced) versus ChemBridge library
21185 - 21286 = Human PNP from 1rr6 (ImmH_and_phosphate-induced) versus ChemBridge library
21287 - 21388 = Pvivax PNP from 3emv (sulphate-induced) versus ChemBridge library
21389 - 21490 = Pf PNP from 3enz (hypoxanthine_ribose_and_arsenate-induced) versus ChemBridge library
21491 - 21592 = Pf PNP from 3fow = V66I_V73I_Y160F mutant (ImmH_and_phosphate-induced) versus CB
21593 - 21694 = Human PNP from 3phb (DADMe-ImmG_and_phosphate-induced) versus ChemBridge library
21695 - 21796 = Pf PNP from 3phc (DADMe-ImmG_and_K_and_phosphate-induced) versus ChemBridge library


Experiment 25 = 100% completed

[Asinex library = 507,000 models of compounds]
21797 - 21847 = Pf PNP from 1nw4 (ImmH_and_sulfate-induced) versus Asinex library
21848 - 21898 = Pf PNP from 1q1g (5'-methylthio-ImmH-induced) versus Asinex library
21899 - 21949 = Human PNP from 1rr6 (ImmH_and_phosphate-induced) versus Asinex library
21950 - 22000 = Pvivax PNP from 3emv (sulphate-induced) versus Asinex library
22001 - 22051 = Pf PNP from 3enz (hypoxanthine_ribose_and_arsenate-induced) versus Asinex library
22052 - 22102 = Pf PNP from 3fow = V66I_V73I_Y160F mutant (ImmH_and_phosphate-induced) versus Asinex
22103 - 22153 = Human PNP from 3phb (DADMe-ImmG_and_phosphate-induced) versus Asinex library
22154 - 22204 = Pf PNP from 3phc (DADMe-ImmG_and_K_and_phosphate-induced) versus Asinex library


Experiment 26 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
22205 - 22355 = Pf PNP from 1nw4 (ImmH_and_sulfate-induced) versus Vitas-M Labs library
22356 - 22506 = Pf PNP from 1q1g (5'-methylthio-ImmH-induced) versus Vitas-M Labs library
22507 - 22657 = Human PNP from 1rr6 (ImmH_and_phosphate-induced) versus Vitas-M Labs library
22658 - 22808 = Pvivax PNP from 3emv (sulphate-induced) versus Vitas-M Labs library
22809 - 22959 = Pf PNP from 3enz (hypoxanthine_ribose_and_arsenate-induced) versus Vitas-M Labs library
22960 - 23110 = Pf PNP from 3fow = V66I_V73I_Y160F mutant (ImmH_and_phosphate-induced) versus Vitas-M
23111 - 23261 = Human PNP from 3phb (DADMe-ImmG_and_phosphate-induced) versus Vitas-M Labs library
23262 - 23412 = Pf PNP from 3phc (DADMe-ImmG_and_K_and_phosphate-induced) versus Vitas-M Labs library

 

ImmH = Immucillin-H = an inhibitor
DADMe-ImmG = IM5 = an inhibitor

 

The two images above display a close-up of the results of positive control re-docking experiments against the mutant Pf PNP from 3fow.pdb, which contains the Val66Ile_Val73Ile_Tyr160Phe mutations.  These 3 mutations are shown as light purple spheres on the left and as the speckled part of the surface on the right.  The known, crystallographic binding mode of the inhibitor Imm-H has purple carbon atoms, while the Vina-docked mode has tan carbon atoms.  The left image shows the ribbon/cartoon mode of PNP's backbone, while the right image displays the solvent-excluded surface.

 

 

The two images above display a close-up of the results of positive control re-docking experiments against the human version of PNP from 1rr6.pdb.  The known, crystallographic binding mode of the inhibitor Imm-H has light blue carbon atoms, while the binding mode predicted by Vina has green carbon atoms.  The left image shows the solvent-excluded surface of human PNP, while the right image displays the ribbon/cartoon mode of PNP's backbone.

 

 


 

 

Target #6 =  PfSUB1

When malaria parasites replicate themselves inside a red blood cell, the "daugher merozoites" eventually rupture the infected host cell, which allows the new parasites to invade and infect other red blood cells.  The subtilisin-like serine proteases from Plasmodium falciparum (subtilases = PfSUB1) are involved in this ability of malaria parasites to escape (or "egress") an infected red blood cell, which allows the parasites to spread the infection in that host.  Specifically, the PfSUB1 enzyme proteolytically modifies the SERA family of papain-like proteins (that is, the subtilases chop regulatory sections off of the papain-like enzymes in order to active them).  The Blackman group at the UK Medical Research Council's (MRC) National Institute for Medical Research (NIMR) has shown that the PfSUB1 enzyme has an additional role in "priming" the merozoite stage of the parasite prior to its invasion of red blood cells.  Right before the daughter parasites escape from an infected red blood cell, three critical merozoite surface proteins (MSP1, MSP6, and MSP7) are cleaved by PfSUB1 in order to activate them.  Inhibiting the PfSUB1 target causes the unprocessed, inactive MSP's to accumulate on the merozoite surface, which significantly reduces the ability of the parasites to invade new red blood cells (erythrocytes).  Thus, PfSUB1 has essential roles in both the escape of the daughter parasites from an infected red blood cell and in the remodelling and activation of its surface proteins to enable a new round of erythrocyte invasion.

Since there are not any crystal structures of PfSUB1 available to the public, we will be using "homology models" of this enzyme as targets on GO FAM.  The homology models of PfSUB1 that we will be targeting were shared with the GO FAM project by the Blackman group and by InhibOx.  For more details, see the Research Team page. Although InhibOx is a company, all data from these experiments against PfSUB1 are legally in the "public domain" (as is all data from all projects performed on World Community Grid).  Thus, we are not "shilling for some company"---InhibOx is sharing their data and insight with us for free to help us all advance the fight against superbugs of malaria.

 

A homology model from the Blackman group of the complex of PfSUB1 with a peptide substrate is shown above.  On the left, the ribbon/cartoon mode of the backbone of PfSUB1 is displayed in brown, while the image on the right shows the solvent-excluded surface of this target.  The peptide is depicted as green spheres.  To learn more about this homology model, see "Note2" below.

 

Kudos to Dr. Garrett M. Morris for proposing the idea to target PfSUB1 on GO FAM and for connecting the Olson lab with the Blackman group and with InhibOx!!  Garrett Morris worked in the Olson lab for ~ 17 years, he was one of the main AutoDock developers, and now he is a Research Manager at InhibOx.

 

 

Batch numbers against PfSUB1:

Experiment 27 = 100% completed

[full NCI library = 316,179 models of compounds]
23413 - 23444 = PfSUB1 homology model1 from Jean-Paul Ebejer et al. at InhibOx vs. NCI library
23445 - 23476 = PfSUB1 homology model from C. Withers-Martinez et al., JBC 2002, MRC's NIMR vs. NCI library
23477 - 23508 = PfSUB1 new homol. model (w/o flipped side-chains), Mike Blackman's group, NIMR, UK, vs. NCI
23509 - 23540 = PfSUB1 new homol. model (w/ flips) from Mike Blackman's group, NIMR, UK, vs. NCI library
23541 - 23572 = PfSUB1 minimized homol. model (w/o flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. NCI
23573 - 23604 = PfSUB1 minimized homol. model (w/ flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. NCI
23605 - 23636 = PfSUB1 homology model2 from Simone Fulle at InhibOx, UK, vs. NCI library


Experiment 28 = 100% completed

[Enamine library = 2,345,014 models of compounds]
23637 - 23871 = PfSUB1 homology model1 from Jean-Paul Ebejer et al. at InhibOx vs. Enamine library
23872 - 24106 = PfSUB1 homology model from C. Withers-Martinez et al., JBC 2002, MRC's NIMR vs. Enamine
24107 - 24341 = PfSUB1 new homol. model (w/o flips), Mike Blackman's group, NIMR, UK, vs. Enamine library
24342 - 24576 = PfSUB1 new homol. model (w/ flips) from Mike Blackman's group, NIMR, UK, vs. Enamine library
24577 - 24811 = PfSUB1 minimized model (w/o flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. Enamine
24812 - 25046 = PfSUB1 minimized model (w/ flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. Enamine
25047 - 25281 = PfSUB1 homology model2 from Simone Fulle et al. at InhibOx, UK, vs. Enamine library


Experiment 29 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
25282 - 25383 = PfSUB1 homology model1 from Jean-Paul Ebejer et al. at InhibOx vs. ChemBridge library
25384 - 25485 = PfSUB1 model from C. Withers-Martinez et al., JBC 2002, MRC's NIMR vs. ChemBridge
25486 - 25587 = PfSUB1 new homol. model (w/o flips), Mike Blackman's group, NIMR, UK, vs. ChemBridge
25588 - 25689 = PfSUB1 new homol. model (w/ flips) from Mike Blackman's group, NIMR, UK, vs. ChemBridge
25690 - 25791 = PfSUB1 minimized model (w/o flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. CB
25792 - 25893 = PfSUB1 minimized model (w/ flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. ChemBridge
25894 - 25995 = PfSUB1 homology model2 from Simone Fulle et al. at InhibOx, UK, vs. ChemBridge library


Experiment 30 = 100% completed

[Asinex library = 507,000 models of compounds]
25996 - 26046 = PfSUB1 homology model1 from Jean-Paul Ebejer et al. at InhibOx vs. Asinex library
26047 - 26097 = PfSUB1 homology model from C. Withers-Martinez et al., JBC 2002, MRC's NIMR vs. Asinex
26098 - 26148 = PfSUB1 new homol. model (w/o flips), Mike Blackman's group, NIMR, UK, vs. Asinex
26149 - 26199 = PfSUB1 new homol. model (w/ flips) from Mike Blackman's group, NIMR, UK, vs. Asinex library
26200 - 26250 = PfSUB1 minimized model (w/o flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. Asinex
26251 - 26301 = PfSUB1 minimized model (w/ flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. Asinex
26302 - 26352 = PfSUB1 homology model2 from Simone Fulle et al. at InhibOx, UK, vs. Asinex library


Experiment 31 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
26353 - 26503 = PfSUB1 homology model1 from Jean-Paul Ebejer et al. at InhibOx vs. VitasMLabs library
26504 - 26654 = PfSUB1 homology model from C. Withers-Martinez et al., JBC 2002, MRC's NIMR vs. VitasMLabs
26655 - 26805 = PfSUB1 new homol. model (w/o flips), Mike Blackman's group, NIMR, UK, vs. VitasMLabs
26806 - 26956 = PfSUB1 new homol. model (w/ flips) from Mike Blackman's group, NIMR, UK, vs. VitasMLabs
26957 - 27107 = PfSUB1 minimized model (w/o flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. VitasMLabs
27108 - 27258 = PfSUB1 minimized model (w/ flips) C. Withers-Martinez, S. Fulle & M. Blackman vs. VitasMLabs
27259 - 27409 = PfSUB1 homology model2 from Simone Fulle et al. at InhibOx, UK, vs. VitasMLabs library

 

Note: the "with flips" versus "without flips" comments refer to the fact that the MolProbity server predicted that some residues in the active site should have a lower energy when their side-chain is flipped.  We included both the original conformation of these residues (in targets labeled "w/o flips") as well as the flipped conformations.

Note2: the above-referenced JBC paper =  Withers-Martinez C, Saldanha JW, Ely B, Hackett F, O'Connor T, Blackman MJ.  (2002)  Expression of recombinant Plasmodium falciparum subtilisin-like protease-1 in insect cells. Characterization, comparison with the parasite protease, and homology modeling. 
Journal of Biological Chemistry.  277(33):29698-709.

 

 

Above are images of the homology model provided by InhibOx of the complex of PfSUB1 with a peptide substrate.  The two images on the top display the ribbon/cartoon mode of PfSUB1's backbone, while the image on the bottom shows the solvent-excluded surface of this target.  The peptide substrate is depicted as dark blue spheres in the top-left image, while the other two images display the peptide as dark blue sticks.  The top-left image shows the side view, while the top-right image and the bottom image display the view from the top.

 

 

Batch numbers for the extension to the experiments versus PfSUB1:

Experiment 62 = 100% completed

[full NCI library = 316,179 models of compounds]
43969 - 44000 = PfSUB1 new x-tal str (C. Withers-Martinez at MRC's NIMR, UK) vs NCI library
44001 - 44032 = PfSUB1 new x-tal str w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs NCI library

 

Experiment 63 = 100% completed

[Enamine library = 2,345,014 models of compounds]
44033 - 44267 = PfSUB1 new x-tal str (C. Withers-Martinez at MRC's NIMR, UK) vs Enamine library
44268 - 44502 = PfSUB1 new x-tal str w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Enamine

Experiment 64 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
44503 - 44604 = PfSUB1 new x-tal str (C. Withers-Martinez at MRC's NIMR, UK) vs ChemBridge library
44605 - 44706 = PfSUB1 new x-tal str w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs ChemBridge
 

Experiment 65 = 100% completed

[Asinex library = 507,000 models of compounds]
44707 - 44757 = PfSUB1 new x-tal str (C. Withers-Martinez at MRC's NIMR, UK) vs Asinex library
44758 - 44808 = PfSUB1 new x-tal str w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Asinex
 

Experiment 66 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
44809 - 44959 = PfSUB1 new x-tal str (C. Withers-Martinez at MRC's NIMR, UK) vs Vitas-M Labs library
44960 - 45110 = PfSUB1 new x-tal str w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Vitas-M Labs

 

Note:  this new 2.25 Angstrom crystal structure of PfSUB1 from Chrislaine Withers-Martinez and Mike Blackman's group, et al., at MRC's NIMR, UK, has not yet been published.

Note2:  w/ WATs = all the crystallographic water molecules were included as part of the target.

 

 

 

Batch numbers for the second extension to the experiments versus PfSUB1:

Experiment 82 = 100% completed

[full NCI library = 316,179 models of compounds]
61099 - 61130 = PfSUB1 newest x-tal str2 (C. Withers-Martinez at MRC's NIMR, UK) vs NCI library
61131 - 61162 = PfSUB1 newest x-tal str2 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs NCI library

 

Experiment 83 = 100% completed

[Enamine library = 2,345,014 models of compounds]
61163 - 61397 = PfSUB1 newest x-tal str2 (C. Withers-Martinez at MRC's NIMR, UK) vs Enamine library
61398 - 61632 = PfSUB1 newest x-tal str2 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Enamine library

Experiment 84 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
61633 - 61734 = PfSUB1 newest x-tal str2 (C. Withers-Martinez at MRC's NIMR, UK) vs ChemBridge library
61735 - 61836 = PfSUB1 newest x-tal str2 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs ChemBridge
 

Experiment 85 = 100% completed

[Asinex library = 507,000 models of compounds]
61837 - 61887 = PfSUB1 newest x-tal str2 (C. Withers-Martinez at MRC's NIMR, UK) vs Asinex library
61888 - 61938 = PfSUB1 newest x-tal str2 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Asinex library
 

Experiment 86 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
61939 - 62089 = PfSUB1 newest x-tal str2 (C. Withers-Martinez at MRC's NIMR, UK) vs Vitas-M Labs library
62090 - 62240 = PfSUB1 newest x-tal str2 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Vitas-M Labs

 

***now jumping back to target class #15 = OAR = FabG***

 

 

Batch numbers for the third extension to the experiments versus PfSUB1:

Experiment 97 = 100% completed

[full NCI library = 316,179 models of compounds]
78229 - 78260 = PfSUB1 newest x-tal str3 (C. Withers-Martinez at MRC's NIMR, UK) vs NCI library
78261 - 78292 = PfSUB1 newest x-tal str3 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs NCI

 

Experiment 98 = 100% completed

[Enamine library = 2,345,014 models of compounds]
78293 - 78527 = PfSUB1 newest x-tal str3 (C. Withers-Martinez at MRC's NIMR, UK) vs Enamine library
78528 - 78762 = PfSUB1 newest x-tal str3 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Enamine

Experiment 99 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
78763 - 78864 = PfSUB1 newest x-tal str3 (C. Withers-Martinez at MRC's NIMR, UK) vs ChemBridge library
78865 - 78966 = PfSUB1 newest x-tal str3 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs ChemBridge
 

Experiment 100 = 100% completed

[Asinex library = 507,000 models of compounds]
78967 - 79017 = PfSUB1 newest x-tal str3 (C. Withers-Martinez at MRC's NIMR, UK) vs Asinex library
79018 - 79068 = PfSUB1 newest x-tal str3 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Asinex
 

Experiment 101 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
79069 - 79219 = PfSUB1 newest x-tal str3 (C. Withers-Martinez at MRC's NIMR, UK) vs Vitas-M Labs library
79220 - 79370 = PfSUB1 newest x-tal str3 w/ WATs (C. Withers-Martinez of M. Blackman's lab) vs Vitas-M Labs

 

***now jumping back to target class #17 = actin depolymerization factor = ADF1 and 2***

 

 

 


 

 

Target #7 =  AM1

M1 neutral aminopeptidase, or AM1, is an enzyme that the parasite needs to properly digest the hemoglobin in our red blood cells.  AM1 plays a critical role in providing the amino acids (the individual building blocks that are joined in specific ways to create proteins) that the parasite needs to grow and mature within the erythrocyte (within the infected red blood cell).  AM1 is a validated drug target for the treatment of malaria.  An inhibitor of Pf AM1 has been shown to impede the growth of chloroquine-resistant malaria parasites in cell culture assays, and it reduced malaria infections in mice (with Plasmodium chabaudi) by 92%, as compared to control studies.

 

 

 

The three above images display the results of positive control re-docking experiments against Pf AM1.  The experimentally-determined, known X-ray crystallographic binding mode of the inhibitor BEY is displayed as sticks with purple carbon atoms, while the binding mode predicted by docking calculations with Vina is displayed as sticks with green carbon atoms.  For this particular re-docking experiment, the 4th-ranked binding mode displayed the best agreement with the known, crystallographic binding mode; consequently, the 4th-ranked mode is displayed.  For the other target classes, the top-ranked binding mode from Vina generally displayed the best agreement with the known binding mode.  The slight decrease in accuracy for these positive control re-docking experiments against Pf AM1 might be caused by the presence of two metal ions in the active site (zinc and magnesium).  Docking calculations against binding sites that contain metals are often less accurate (regarless of the type of docking program being utilized), due to the large charges present on the metals and the unique geometric requirements that metals display.  The Pf AM1 target molecule from 3EBI.pdb is displayed as either light purple ribbons or as a clipped solvent-excluded surface with a light purple color.  The two images on the top display close-up views to highlight the binding modes, while the image on the bottom displays the full view of the target, to indicate the general location of the active site we are trying to block.

 

 

Batch numbers against Pf AM1:

Experiment 32 = 100% completed

[full NCI library = 316,179 models of compounds]
27410 - 27441 = Pf AM1 from 3EBH.pdb (chain B; bestatin-induced) versus NCI library
27442 - 27473 = Pf AM1 from 3EBI.pdb (chain A; BEY-induced) versus NCI library
27474 - 27505 = Pf AM1 from 3EBI.pdb (chain B with 2 x-tal waters; BEY-induced) versus NCI library


Experiment 33 = 100% completed

[Enamine library = 2,345,014 models of compounds]
27506 - 27740 = Pf AM1 from 3EBH.pdb (chain B; bestatin-induced) versus Enamine library
27741 - 27975 = Pf AM1 from 3EBI.pdb (chain A; BEY-induced) versus Enamine library
27976 - 28210 = Pf AM1 from 3EBI.pdb (chain B with 2 x-tal waters; BEY-induced) versus Enamine library

 

Experiment 34 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
28211 - 28312 = Pf AM1 from 3EBH.pdb (chain B; bestatin-induced) versus ChemBridge library
28313 - 28414 = Pf AM1 from 3EBI.pdb (chain A; BEY-induced) versus ChemBridge library
28415 - 28516 = Pf AM1 from 3EBI.pdb (chain B with 2 x-tal waters; BEY-induced) versus ChemBridge library

 

Experiment 35 = 100% completed

[Asinex library = 507,000 models of compounds]
28517 - 28567 = Pf AM1 from 3EBH.pdb (chain B; bestatin-induced) versus Asinex library
28568 - 28618 = Pf AM1 from 3EBI.pdb (chain A; BEY-induced) versus Asinex library
28619 - 28669 = Pf AM1 from 3EBI.pdb (chain B with 2 x-tal waters; BEY-induced) versus Asinex library

 

Experiment 36 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
28670 - 28820 = Pf AM1 from 3EBH.pdb (chain B; bestatin-induced) versus Vitas-M Labs library
28821 - 28971 = Pf AM1 from 3EBI.pdb (chain A; BEY-induced) versus Vitas-M Labs library
28972 - 29122 = Pf AM1 from 3EBI.pdb (chain B with 2 x-tal waters; BEY-induced) versus Vitas-M Labs library

 

BES = bestatin = dipeptide analog inhibitor
BEY = phosphinate dipeptide analog inhibitor

x-tal waters = water molecules that were present in the X-ray crystallographic structure of the target.  For 3EBI.pdb chain B, keeping two particular crystallographic waters present as part of the target improved the accuracy of the positive control re-docking experiments with Vina.

 

 

 

The three above images also display the results of positive control re-docking experiments against Pf AM1.  The experimentally-determined, known X-ray crystallographic binding mode of the inhibitor BES is displayed as sticks with cyan carbon atoms, while the binding mode predicted by Vina is displayed as sticks with magenta carbon atoms.  For this particular re-docking experiment, the 2nd-ranked binding mode displayed the best agreement with the known, crystallographic binding mode; consequently, the 2nd-ranked mode is displayed.  For the other target classes, the top-ranked binding mode from Vina generally displayed the best agreement with the known binding mode.  The Pf AM1 target molecule from 3EBH.pdb is displayed as either cyan ribbons or as a clipped solvent-excluded surface with a cyan color.  The two images on the top display close-up views to highlight the binding modes, while the image on the bottom displays the full view of the target, to indicate the general location of the active site we are trying to block.

 

 

 

 


 

 

Target #8 =  FP2 and FP3

Falcipains (FP2 and FP3) are cysteine proteases (that is, they use a cysteine residue to chop up certain other proteins at specific places), and they are validated drug targets for the treatment of malaria.  The falcipains are another class of enzymes that are critically involved in the parasite's ability to digest hemoglobin (in our red blood cells).  Blocking the ability of the falcipains to function can kill the malaria parasite.  When the gene for falcipain 2 is "knocked out" (deleted from the parasite's genome), undegraded hemoglobin accumulates in the parasite.  In other words, removing the parasite's ability to produce FP2 causes the parasite to starve, since it can no longer digest its food.  In addition, compounds that can inhibit the function of FP2 and FP3 have been shown to stop the growth of Plasmodium falciparum parasites in cell culture studies and, more importantly, these inhibitors were able to cure malaria infections in animal models.

The known inhibitors of FP2 and FP3 are "covalent inhibitors," which means that after they bind to the active site of this enzyme, the inhibitors undergo a chemical reaction that causes them to permanently attach themselves to the target.  When a covalent inhibitor fuses itself to the active site of a drug target, that particular copy of the enzyme can never function again.  Because the known inhibitors of FP2 and FP3 are covalent inhibitors, special types of docking protocols are generally needed to model how these covalent inhibitors bind to and block the function of the drug target.  However, those specialized "covalent docking" approaches are only suitable for docking a very small number of compounds--they are not meant to be used in virtual screens of millions of (non-reactive) compounds.  Consequently, our positive control studies against this system used only the traditional docking approach--covalent docking protocols were not utilized.  This made the positive control docking results against FP2 and FP3 less accurate than our positive control studies against other systems (that is, the known and docked binding poses did not superimpose as well as they did against the other types of drug targets that are part of GO FAM).  Fortunately (and surprisingly), the positive control results from Vina docking calculations with the known inhibitors were still accurate enough to justify screening millions of compounds against this system, using the standard, non-covalent docking approaches.  Thus, the experiments against FP2 and FP3 will be searching for new types of non-covalent inhibitors of this drug target, but the chemical features of the best inhibitors we discover could also assist in the discovery and design of new types of covalent inhibitors, as well.

 

 

 

The top two images compare the known crystallographic binding mode of the covalent inhibitor "K11017" (from 3BWK.pdb) in dark purple to the Vina-docked mode (using conventional, non-covalent docking protocols) in magenta.  Most of the images on this page display the top-ranked Vina mode, but these two images show the 8th-ranked mode, because it had the best agreement with the experimentally-determined binding mode.  The image on the top left shows the ribbon/cartoon mode of the backbone of Pf FP3, while the image on the top right displays the surface mode.  The image on the bottom compares the known crystallographic binding mode of the covalent inhbitor "leupeptin" (from 3BPM.pdb) in cyan to the Vina-predicted mode in light purple.  This image displays the 6th-ranked mode from the docking studies versus Pf FP3.  These positive control experiments versus FP2 and FP3 were the least accurate out of all of the different classes of targets we have used on GO Fight Against Malaria, but as I stated above, these compounds are covalent inhibitors, which generally require specialized docking protocols.  Since the covalent docking protocols are not feasible for large virtual screening experiments, we had to use the normal, non-covalent docking protocols for these compounds, which is why they do not superimpose as well with the known binding mode (as compared to the other types of targets we are attacking).  These images display a close-up view of the active site, while the image at the end of the description of these experiments shows the full view of the FP3 target.

 

 

 

Batch numbers against Pf FP2 and FP3:

Experiment 37 = 100% completed

[full NCI library = 316,179 models of compounds]
29123 - 29154 = Pf FP2 from 1YVB.pdb (chain A; cystatin-induced) versus NCI library
29155 - 29186 = Pf FP3 from 3BPM.pdb (chain B; leupeptin-induced) versus NCI library
29187 - 29218 = Pf FP3 from 3BWK.pdb (chain B; K11017-induced) versus NCI library

 

Experiment 38 = 100% completed

[Enamine library = 2,345,014 models of compounds]
29219 - 29453 = Pf FP2 from 1YVB.pdb (chain A; cystatin-induced) versus Enamine library
29454 - 29688 = Pf FP3 from 3BPM.pdb (chain B; leupeptin-induced) versus Enamine library
29689 - 29923 = Pf FP3 from 3BWK.pdb (chain B; K11017-induced) versus Enamine library

Experiment 39 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
29924 - 30025 = Pf FP2 from 1YVB.pdb (chain A; cystatin-induced) versus ChemBridge library
30026 - 30127 = Pf FP3 from 3BPM.pdb (chain B; leupeptin-induced) versus ChemBridge library
30128 - 30229 = Pf FP3 from 3BWK.pdb (chain B; K11017-induced) versus ChemBridge library

 

Experiment 40 = 100% completed

[Asinex library = 507,000 models of compounds]
30230 - 30280 = Pf FP2 from 1YVB.pdb (chain A; cystatin-induced) versus Asinex library
30281 - 30331 = Pf FP3 from 3BPM.pdb (chain B; leupeptin-induced) versus Asinex library
30332 - 30382 = Pf FP3 from 3BWK.pdb (chain B; K11017-induced) versus Asinex library

 

Experiment 41 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
30383 - 30533 = Pf FP2 from 1YVB.pdb (chain A; cystatin-induced) versus Vitas-M Labs library
30534 - 30684 = Pf FP3 from 3BPM.pdb (chain B; leupeptin-induced) versus Vitas-M Labs library
30685 - 30835 = Pf FP3 from 3BWK.pdb (chain B; K11017-induced) versus Vitas-M Labs library

 

 

 


 

 

Target #9 =  GR

Glutathione reductase from Plasmodium falciparum, or Pf GR, is currently a "potential drug target" that we are trying to help validate.  Oxidative stress is an important mechanism for defeating intracellular parasites, including malaria.  The "redox metabolism" pathways for the malaria parasite involve the key enzymes glutathione reductase, glutathione peroxidase, thioredoxin reductase, and thioredoxin peroxidase.  It is known that two inhibitors of thioredoxin reductase are able to kill malaria parasites, which suggests that GR could become a valid drug target for curing malaria.  If we can discover potent inhibitors of the Pf GR enzyme, then those inhibitors will be tested against the malaria parasites.  If potent Pf GR inhibitors are able to enter and then kill the malaria parasites, then we will know that GR is a "valid drug target" for curing malaria, and we will have advanced the discovery of Pf GR inhibitors that could potentially be developed into new malaria cures.  If potent Pf GR inhibitors do not kill malaria parasites, then we will know that Pf GR is not a good target for further research and development.

 

 

The above three images represent positive control re-docking experiments against the human version of the GR enzyme.  These images show that we can accurately dock the inhibitor "pyocyanin" against this target.  Since we only have an "apo" (or unbound) crystal structure of the malarial GR enzyme available, there is not yet any experimental proof regarding how pyocyanin binds to Pf GR.  Consequently, we could only perform these positive control experiments against the pyocyanin-induced conformation of human GR.  The X-ray crystallographic conformation of pyocyanin (that is, the experimentally-determined structure) is displayed in stick mode with purple carbon atoms, while the docked mode predicted by Vina is shown as sticks with green carbon atoms.  The upper-left image shows a close-up view of the active site, while the upper-right image displays the full structure of the GR target.  The bottom image is also a close-up, but it includes a small subset of the surface of GR (while the other two images just display the cartoon mode of the backbone of GR).

 

Batch numbers against Pf GR:

Experiment 42 = 100% completed

[full NCI library = 316,179 models of compounds]
30836 - 30867 = Pf GR from 1ONF.pdb (apo dimer) versus NCI library
30868 - 30899 = human GR from 3SQP.pdb (chain B; pyocyanin-induced) versus NCI library

 

Experiment 43 = 100% completed

[Enamine library = 2,345,014 models of compounds]
30900 - 31134 = Pf GR from 1ONF.pdb (apo dimer) versus Enamine library
31135 - 31369 = human GR from 3SQP.pdb (chain B; pyocyanin-induced) versus Enamine library

Experiment 44 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
31370 - 31471 = Pf GR from 1ONF.pdb (apo dimer) versus ChemBridge library
31472 - 31573 = human GR from 3SQP.pdb (chain B; pyocyanin-induced) versus ChemBridge library

 

Experiment 45 = 100% completed

[Asinex library = 507,000 models of compounds]
31574 - 31624 = Pf GR from 1ONF.pdb (apo dimer) versus Asinex library
31625 - 31675 = human GR from 3SQP.pdb (chain B; pyocyanin-induced) versus Asinex library

 

Experiment 46 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
31676 - 31826 = Pf GR from 1ONF.pdb (apo dimer) versus Vitas-M Labs library
31827 - 31977 = human GR from 3SQP.pdb (chain B; pyocyanin-induced) versus Vitas-M Labs library

 

 

pycocyanin = an inhibitor of human GR that is also known to have some antimalarial activity

apo = unbound conformation.  Since the structure of the Pf GR target was not bound to a known inhibitor (i.e., it's conformation was not induced into an inactive state by a small molecule compound), it means that the experiments against Pf GR probably have a lower chance of success than most of our other GO FAM experiments.  We have to take some risks now and then to try to make significant, innovative breakthroughs.


 

 

 


 

 

Target #10 = GST

Glutathione S-transferase from Plasmodium falciparum, or Pf GST, catalyzes the conjugation (covalent attachment) of glutathione to a wide range of hydrophobic compounds (such as the parasitotoxic hemin molecule), which renders them non-toxic and enables their elimination from the parasite.  Without this conjugation process, then the hemin that is produced when the malaria parasite eats our red blood cells would start to accumulate within the parasite, which would eventually kill it.  Unlike most other organisms, the malaria parasite only has one GST isoenzyme (it only makes one version of this type of protein).  The Pf GST enzyme is highly abundant within the malaria parasite, and Pf GST has a very different structure than human GST.  Consequently, Pf GST is a very promising target for the development of new antimalarial drugs.

 

These images depict positive control re-docking experiments against the malarial version of GST from 1Q4J.pdb.  The X-ray crystallographic conformation (that is, the experimentally-determined structure) of the inhibitor GTX is displayed as sticks with dark green carbon atoms, while the docked mode predicted by Vina has orange carbon atoms.  The image on the left shows a close-up view of the binding site, while the image on the right displays the full view of the GST target.

 

Batch numbers against Pf GR:

Experiment 47 = 100% completed

[full NCI library = 316,179 models of compounds]
31978 - 32009 = Pf GST from 1OKT (apo tetramer) versus NCI library
32010 - 32041 = Pf GST from 1Q4J (GTX-induced dimer) versus NCI library
32042 - 32073 = Pf GST from 1Q4J (GTX-induced tetramer) versus NCI library
32074 - 32105 = Pf GST from 2AAW (GTX-induced dimer w/ P33) versus NCI library
32106 - 32137 = Pf GST from 3FR3 (GDS-induced dimer w/ x-tal waters) versus NCI library
32138 - 32169 = Pf GST from 3FR3 (GDS-induced tetramer w/ x-tal waters) versus NCI library

 

Experiment 48 = 100% completed

[Enamine library = 2,345,014 models of compounds]
32170 - 32404 = Pf GST from 1OKT (apo tetramer) versus Enamine library
32405 - 32639 = Pf GST from 1Q4J (GTX-induced dimer) versus Enamine library
32640 - 32874 = Pf GST from 1Q4J (GTX-induced tetramer) versus Enamine library
32875 - 33109 = Pf GST from 2AAW (GTX-induced dimer w/ P33) versus Enamine library
33110 - 33344 = Pf GST from 3FR3 (GDS-induced dimer w/ x-tal waters) versus Enamine library
33345 - 33579 = Pf GST from 3FR3 (GDS-induced tetramer w/ x-tal waters) versus Enamine library

Experiment 49 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
33580 - 33681 = Pf GST from 1OKT (apo tetramer) versus ChemBridge library
33682 - 33783 = Pf GST from 1Q4J (GTX-induced dimer) versus ChemBridge library
33784 - 33885 = Pf GST from 1Q4J (GTX-induced tetramer) versus ChemBridge library
33886 - 33987 = Pf GST from 2AAW (GTX-induced dimer w/ P33) versus ChemBridge library
33988 - 34089 = Pf GST from 3FR3 (GDS-induced dimer w/ x-tal waters) versus ChemBridge library
34090 - 34191 = Pf GST from 3FR3 (GDS-induced tetramer w/ x-tal waters) versus ChemBridge library

 

Experiment 50 = 100% completed

[Asinex library = 507,000 models of compounds]
34192 - 34242 = Pf GST from 1OKT (apo tetramer) versus Asinex library
34243 - 34293 = Pf GST from 1Q4J (GTX-induced dimer) versus Asinex library
34294 - 34344 = Pf GST from 1Q4J (GTX-induced tetramer) versus Asinex library
34345 - 34395 = Pf GST from 2AAW (GTX-induced dimer w/ P33) versus Asinex library
34396 - 34446 = Pf GST from 3FR3 (GDS-induced dimer w/ x-tal waters) versus Asinex library
34447 - 34497 = Pf GST from 3FR3 (GDS-induced tetramer w/ x-tal waters) versus Asinex library

 

Experiment 51 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
34498 - 34648 = Pf GST from 1OKT (apo tetramer) versus Vitas-M Labs library
34649 - 34799 = Pf GST from 1Q4J (GTX-induced dimer) versus Vitas-M Labs library
34800 - 34950 = Pf GST from 1Q4J (GTX-induced tetramer) versus Vitas-M Labs library
34951 - 35101 = Pf GST from 2AAW (GTX-induced dimer w/ P33) versus Vitas-M Labs library
35102 - 35252 = Pf GST from 3FR3 (GDS-induced dimer w/ x-tal waters) versus Vitas-M Labs library
35253 - 35403 = Pf GST from 3FR3 (GDS-induced tetramer w/ x-tal waters) versus Vitas-M Labs

 

GTX = inhibitor s-hexyl-GSH = S-HEXYL-GLUTATHIONE

GDS = glutathione disulfide = 2 glutathione molecules joined by a disulfide bond.  GDS gets reduced and thus split into 2 separate glutathione molecules (the x-S-S-x disulfide bond gets split into x-SH + HS-x) in order to help oxidize the critical cofactor NADPH.

P33 = PEG 330 = heptaethylene glycol = a surfactant

w/ x-tal waters = some of the crystallographic water molecules were included as part of the target during the docking calculation.  The positive control re-docking calculations were more accurate when these crystallographic water molecules were included as part of the target.

Note:  The original crystal structures of these targets were dimers (they are made of two identical pieces that come together to make the full, intact version of the protein), but for some of the targets, the symmetry information in the crystal structure was used in PyMOL to generate a model of the tetrameric structure.  The "tetramer" is a dimer of dimers (it has four identical pieces that come together).  In positive control re-docking experiments, the results were sometimes more accurate when the inhibitor was docked to the full tetrameric structure, since the inhibitor made a few contacts (interactions at a close distance) with both dimers.

 

 

 

 


 

 

Target #11 = DHODH

DHODH, or dihydroorotate dehydrogenase, is an enzyme that is part of the malaria parasite's "de novo" pathway for synthesizing pyrimidine nucleotides (such as cytosine and thymine, which are key building blocks in DNA).  The cells in mammals (including us humans) are able to use both the de novo synthesis path (they can build the pyrimidine nucleotides from scratch) and the salvage paths (they can obtain the pyrimidines by breaking down components in food and extracting them).  However, the malaria parasites can only utilize the de novo path.  If we can shut down this de novo pyrimidine synthesis pathway in the parasite, then the parasite will not be able to make DNA, which should prevent it from replicating effectively.  DHODH is one of the many malarial enzymes in this pathway, and it is one of the few enzymes in this path for which an atomically-detailed "crystal structure" is available.  In some research papers, this Pf DHODH is considered to be a "potential drug target" that needs further exploration, but in other recent papers, inhibitors of Pf DHODH (such as DSM265) have been shown to kill both drug-sensitive and drug-resistant strains of malaria (in mouse models of malaria infections).

On GO Fight Against Malaria, we will be searching for compounds that can selectively and strongly inhibit the malarial version of this enzyme (Pf DHODH), without significantly inhibiting the human version of DHODH.  Consequently, we will identify compounds that are both predicted to bind well to Pf DHODH and are also predicted to bind poorly to human DHODH.  However, the compounds that are predicted to bind strongly to human DHODH might be useful to other types of scientists in the world.  Potent inhibitors of human DHODH have the potential to help treat autoimmune diseases, such as rheumatoid arthritis and psoriatic arthritis.  Scientists who are trying to discover new drugs to treat arthritis should contact us and request a copy of the results we produce against human DHODH.

 

 

These images display the results of positive control re-docking experiments against the malarial version of DHODH from 3I68.pdb.  The ribbon / cartoon mode of the protein backbone is shown as thin tubes in each image, and the spheres represent orotic acid and the flavin mononucleotide (which were included as part of the model of the target for these experiments).  A water molecule that was also part of the model of the target is depicted as that little V-shaped molecule (with white ends and a red vertex).  The left image is a close-up view, while the right image displays the full target.  The crystallographic binding mode of the inhibitor J4Z is shown as sticks with magenta carbon atoms, while the binding mode predicted by Vina has cyan carbon atoms.

 

Batch numbers against Pf DHODH:

Experiment 52 = 100% completed

[full NCI library = 316,179 models of compounds]
35404 - 35435 = human DHODH from 1D3G (BRE-induced) versus NCI library
35436 - 35467 = human DHODH from 1D3G (BRE-induced; w/ DDQ & 1 x-tal water) versus NCI library
35468 - 35499 = Pf DHODH from 1TV5 (A26-induced) versus NCI library
35500 - 35531 = human DHODH from 2FPV (ILC-induced; w/ 1 x-tal water) versus NCI library
35532 - 35563 = human DHODH from 2FPY (ILF-induced; w/ 1 x-tal water) versus NCI library
35564 - 35595 = human DHODH from 2FQI (ILH-induced) versus NCI library
35596 - 35627 = human DHODH from 3G0X (MD7-induced; w/ 1 x-tal water) versus NCI library
35628 - 35659 = Pf DHODH from 3I65 (JZ8-induced) versus NCI library
35660 - 35691 = Pf DHODH from 3I65 (JZ8-induced; w/ 1 x-tal water) versus NCI library
35692 - 35723 = Pf DHODH from 3I68 (J4Z-induced) versus NCI library
35724 - 35755 = Pf DHODH from 3I68 (J4Z-induced; w/ 1 x-tal water) versus NCI library
35756 - 35787 = Pf DHODH from 3O8A (O8A-induced) versus NCI library

 

Experiment 53 = 100% completed

[Enamine library = 2,345,014 models of compounds]
35788 - 36022 = human DHODH from 1D3G (BRE-induced) versus Enamine library
36023 - 36257 = human DHODH from 1D3G (BRE-induced; w/ DDQ & 1 x-tal water) versus Enamine library
36258 - 36492 = Pf DHODH from 1TV5 (A26-induced) versus Enamine library
36493 - 36727 = human DHODH from 2FPV (ILC-induced; w/ 1 x-tal water) versus Enamine library
36728 - 36962 = human DHODH from 2FPY (ILF-induced; w/ 1 x-tal water) versus Enamine library
36963 - 37197 = human DHODH from 2FQI (ILH-induced) versus Enamine library
37198 - 37432 = human DHODH from 3G0X (MD7-induced; w/ 1 x-tal water) versus Enamine library
37433 - 37667 = Pf DHODH from 3I65 (JZ8-induced) versus Enamine library
37668 - 37902 = Pf DHODH from 3I65 (JZ8-induced; w/ 1 x-tal water) versus Enamine library
37903 - 38137 = Pf DHODH from 3I68 (J4Z-induced) versus Enamine library
38138 - 38372 = Pf DHODH from 3I68 (J4Z-induced; w/ 1 x-tal water) versus Enamine library
38373 - 38607 = Pf DHODH from 3O8A (O8A-induced) versus Enamine library

Experiment 54 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
38608 - 38709 = human DHODH from 1D3G (BRE-induced) versus ChemBridge library
38710 - 38811 = human DHODH from 1D3G (BRE-induced; w/ DDQ & 1 x-tal water) versus ChemBridge library
38812 - 38913 = Pf DHODH from 1TV5 (A26-induced) versus ChemBridge library
38914 - 39015 = human DHODH from 2FPV (ILC-induced; w/ 1 x-tal water) versus ChemBridge library
39016 - 39117 = human DHODH from 2FPY (ILF-induced; w/ 1 x-tal water) versus ChemBridge library
39118 - 39219 = human DHODH from 2FQI (ILH-induced) versus ChemBridge library
39220 - 39321 = human DHODH from 3G0X (MD7-induced; w/ 1 x-tal water) versus ChemBridge library
39322 - 39423 = Pf DHODH from 3I65 (JZ8-induced) versus ChemBridge library
39424 - 39525 = Pf DHODH from 3I65 (JZ8-induced; w/ 1 x-tal water) versus ChemBridge library
39526 - 39627 = Pf DHODH from 3I68 (J4Z-induced) versus ChemBridge library
39628 - 39729 = Pf DHODH from 3I68 (J4Z-induced; w/ 1 x-tal water) versus ChemBridge library
39730 - 39831 = Pf DHODH from 3O8A (O8A-induced) versus ChemBridge library

 

Experiment 55 = 100% completed

[Asinex library = 507,000 models of compounds]
39832 - 39882 = human DHODH from 1D3G (BRE-induced) versus Asinex library
39883 - 39933 = human DHODH from 1D3G (BRE-induced; w/ DDQ & 1 x-tal water) versus Asinex library
39934 - 39984 = Pf DHODH from 1TV5 (A26-induced) versus Asinex library
39985 - 40035 = human DHODH from 2FPV (ILC-induced; w/ 1 x-tal water) versus Asinex library
40036 - 40086 = human DHODH from 2FPY (ILF-induced; w/ 1 x-tal water) versus Asinex library
40087 - 40137 = human DHODH from 2FQI (ILH-induced) versus Asinex library
40138 - 40188 = human DHODH from 3G0X (MD7-induced; w/ 1 x-tal water) versus Asinex library
40189 - 40239 = Pf DHODH from 3I65 (JZ8-induced) versus Asinex library
40240 - 40290 = Pf DHODH from 3I65 (JZ8-induced; w/ 1 x-tal water) versus Asinex library
40291 - 40341 = Pf DHODH from 3I68 (J4Z-induced) versus Asinex library
40342 - 40392 = Pf DHODH from 3I68 (J4Z-induced; w/ 1 x-tal water) versus Asinex library
40393 - 40443 = Pf DHODH from 3O8A (O8A-induced) versus Asinex library

 

Experiment 56 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
40444 - 40594 = human DHODH from 1D3G (BRE-induced) versus Vitas-M Labs library
40595 - 40745 = human DHODH from 1D3G (BRE-induced; w/ DDQ & 1 x-tal water) versus Vitas-M Labs library
40746 - 40896 = Pf DHODH from 1TV5 (A26-induced) versus Vitas-M Labs library
40897 - 41047 = human DHODH from 2FPV (ILC-induced; w/ 1 x-tal water) versus Vitas-M Labs library
41048 - 41198 = human DHODH from 2FPY (ILF-induced; w/ 1 x-tal water) versus Vitas-M Labs library
41199 - 41349 = human DHODH from 2FQI (ILH-induced) versus Vitas-M Labs library
41350 - 41500 = human DHODH from 3G0X (MD7-induced; w/ 1 x-tal water) versus Vitas-M Labs library
41501 - 41651 = Pf DHODH from 3I65 (JZ8-induced) versus Vitas-M Labs library
41652 - 41802 = Pf DHODH from 3I65 (JZ8-induced; w/ 1 x-tal water) versus Vitas-M Labs library
41803 - 41953 = Pf DHODH from 3I68 (J4Z-induced) versus Vitas-M Labs library
41954 - 42104 = Pf DHODH from 3I68 (J4Z-induced; w/ 1 x-tal water) versus Vitas-M Labs library
42105 - 42255 = Pf DHODH from 3O8A (O8A-induced) versus Vitas-M Labs library

 

Notes:
BRE = human DHODH inhibitor; IC50 = 7 nanoM
A26 = weak Plasmodium falciparum DHODH inhibitor; IC50 = 190 microM
ILC = human DHODH inhibitor; IC50 = 44 nanoM
ILF = human DHODH inhibitor; IC50 = 2 nanoM
ILH = human DHODH inhibitor; IC50 = 7 nanoM
MD7 = human DHODH inhibitor; IC50 = 53 to 1580 nanoM
JZ8 = simple, naphthalene-based Plasmodium falciparum DHODH inhibitor; IC50 = 47 nanoM (target with 92% sequence identity)
J4Z = simple, anthracene-based Plasmodium falciparum DHODH inhibitor; IC50 = 56 nanoM (target with 92% sequence identity)
O8A = Plasmodium falciparum DHODH inhibitor Genz667348; IC50 = 22 nanoM

w/ x-tal waters = some of the crystallographic water molecules were included as part of the target during the docking calculation.  The positive control re-docking calculations were more accurate when these crystallographic water molecules were included as part of the target.

IC50 = half maximal inhibitory concentration, which measures how effective a compound is at inhibiting a particular enzyme.  It is the dose of a compound that is required to inhibit an enzyme's activity by 50% in vitro (in a test tube).  The smaller the IC50 value, the more potent the compound is.

 

 

 


 

 

Target #12 = hCD81

hCD81, or human CD81, is a receptor protein that belongs to the tetraspanin family.  Tetraspanins are part of a widely distributed superfamily of proteins that each have four transmembrane regions (parts of the protein that are threaded through the lipid membrane, which forms the barrier that surrounds our cells) and two extracellular domains (extracellular domains are independent structural units of a receptor protein that are located on the outside of the membrane barrier and help the cell receive signals from and/or interact with the environment outside the cell).  Some of the different types of human tetraspanins are only found in particular tissues, but human CD81 is nearly ubiquitous (it is found throughout the body, in many different types of tissues).  CD81 possess a "pleiotropic function" (when one gene is responsible for or affects more than one phenotypic characteristic or function), which includes being involved in cell adhesion, cell migration, cell fusion, co-stimulation, signal transduction, and differentiation. In addition, it was identified as a key player that is involved in the infection process of several different viruses, including HIV and hepatitis C.

These two images display the structure of the human CD81 protein and the location of the 3 different binding sites that are being targeted in these experiments.  Each box shows the location of 1 particular binding site.  The image on the left shows the surface representation of CD81, colored with the "David Goodsell" convention, while the image on the right displays the cartoon mode of the backbone of CD81.  The six images that are displayed below the description for Experiment 61 contain the exact same type of information and the same target molecule, but each particular binding site is shown by itself.

 

 

Surprisingly, CD81 is currently the only human surface protein known to play a role in the process used by sporozoites of several Plasmodium species to invade/infect the human hepatocytes (liver cells). In 2008, Rubinstein et al. were able to identify approximately 35 residues in the large extracellular loop of CD81 that play a significant role in the progression of malaria infections and in sporozoite invasion.  We are defining three different "grid boxes" on CD81 (three different regions of the target that the compounds will be allowed to explore during the docking calculations) in order to discover new compounds that can block the interactions between these 35 key residues in CD81 and the Plasmodium proteins that the parasites use to invade our liver cells.  Thus, these virtual screens against CD81, in combination with experimental assays performed by our collaborators, could help us discover promising new types of drugs that might help prevent the spread of malaria infections.

As was stated at the top of this page, we do take requests.  These experiments against human CD81 are an example of us following through on that promise.  Reem Al Olaby, a scientist who visited and worked in the Olson lab a couple years ago on a hepatitis C project, recently learned about the GO Fight Against Malaria project, and she asked us if we could help expand the scope of her research.  We welcomed this opportunity to tackle a new target and a new approach to fighting malaria infections.

Reem Al Olaby is a Ph.D. student (Yousif Jameel fellow) at the American University in Cairo, Egypt, under the supervision of Dr. Hassan Azzazy, Dr. Rod Balhorn, Dr. Brett Chromy, and Dr. Shoshana Levy.  She has been advancing the fight against hepatitis C for several years.  Reem is currently a research intern in Dr. Ali Sultan's lab at Weill Cornell Medical College in Doha, Qatar (WCMC-Q), where she is working on another aspect of her Ph.D. project by performing research to develop "smart drugs" against malaria.  This research is being performed in collaboration with Dr. Ali Sultan (Prof. of Immunology and Microbiology at WCMC-Q), Prof. Art Olson's lab (TSRI), Dr. Jost Vielmitter (California Institute of Technology), and Dr. Brett Chromy (University of California, Davis).

 

Batch numbers against human CD81:

Experiment 57 = 100% completed

[full NCI library = 316,179 models of compounds]
42256 - 42287 = human CD81 (from 1G8Q.pdb) target site 1 vs full NCI library
42288 - 42319 = human CD81 (from 1G8Q.pdb) target site 2 vs full NCI library
42320 - 42351 = human CD81 (from 1G8Q.pdb) target site 3 vs full NCI library

 

Experiment 58 = 100% completed

[Enamine library = 2,345,014 models of compounds]
42352 - 42586 = human CD81 (from 1G8Q.pdb) target site 1 vs Enamine library
42587 - 42821 = human CD81 (from 1G8Q.pdb) target site 2 vs Enamine library
42822 - 43056 = human CD81 (from 1G8Q.pdb) target site 3 vs Enamine library

Experiment 59 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
43057 - 43158 = human CD81 (from 1G8Q.pdb) target site 1 vs ChemBridge library
43159 - 43260 = human CD81 (from 1G8Q.pdb) target site 2 vs ChemBridge library
43261 - 43362 = human CD81 (from 1G8Q.pdb) target site 3 vs ChemBridge library

 

Experiment 60 = 100% completed

[Asinex library = 507,000 models of compounds]
43363 - 43413 = human CD81 (from 1G8Q.pdb) target site 1 vs Asinex library
43414 - 43464 = human CD81 (from 1G8Q.pdb) target site 2 vs Asinex library
43465 - 43515 = human CD81 (from 1G8Q.pdb) target site 3 vs Asinex library

 

Experiment 61 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
43516 - 43666 = human CD81 (from 1G8Q.pdb) target site 1 vs Vitas-M Labs library
43667 - 43817 = human CD81 (from 1G8Q.pdb) target site 2 vs Vitas-M Labs library
43818 - 43968 = human CD81 (from 1G8Q.pdb) target site 3 vs Vitas-M Labs library

 

1G8Q.pdb = crystal structure of extracellular domain of human CD81 = a receptor for hepatitis C virus and for Plasmodium parasites

 

 

 

 

 

 

 


*******************jumped back to target class #6 = PfSUB1, to submit new experiments versus

           a brand new, not-yet-published crystal structure from Chrislaine Withers-Martinez

                   of Mike Blackman's group at MRC's NIMR, UK.  *************************


 

 

 

 


 

 

Target #13 = ODCase

ODCase, or orotidine 5'-phosphate decarboxylase, is a potential drug target that is also part of the critical de novo pyrimidine synthesis path.  See target #11, DHODH, to learn more about what this pathway is and why we want to discover new, very potent inhibitors that can block several key steps within it.

 

These images represent the results of positive control re-docking experiments against the version of ODCase from Plasmodium falciparum (in 3N34.pdb).  The seven V-shaped, red and white "ball and stick" molecules show different crystallographic water molecules that were included as part of the model of the target in some of these experiments, while the white cartoon shows the backbone of Pf ODCase.  The image on the left is a close-up view of the binding site, while the image on the right shows the full target.  The X-ray crystallographic conformation of FNU is displayed as sticks with pink-ish carbon atoms, while the binding mode predicted by Vina has magenta carbon atoms.

 

Batch numbers against ODCase:

Experiment 67 = 100% completed

[full NCI library = 316,179 models of compounds]
45111 - 45142 = Pb ODCase (from 2FDS.pdb; apo = not ligand-bound) versus NCI library
45143 - 45174 = Pv ODCase (from 2FFC_a.pdb, monomer w/ 6 xtal WATs) versus NCI library
45175 - 45206 = Pv ODCase (from 2FFC_b.pdb; dimer w/ 6 xtal WATs) versus NCI library
45207 - 45238 = Pv ODCase (from 2FFC_b.pdb; dry dimer) versus NCI library
45239 - 45270 = Pf ODCase (from 2Q8Z_b.pdb; w/ 6 xtal WATs) versus NCI library
45271 - 45302 = Pf ODCase (from 2Q8Z_b.pdb; dry) versus NCI library
45303 - 45334 = Pf ODCase (from 3BPW_a.pdb; w/ 9 xtal WATs) versus NCI library
45335 - 45366 = Pf ODCase (from 3BPW_a.pdb; dry) versus NCI library
45367 - 45398 = Pf ODCase (from 3N34_b.pdb; w/ 7 xtal WATs) versus NCI library
45399 - 45430 = Pf ODCase (from 3N34_b.pdb; dry) versus NCI library
45431 - 45462 = Pf ODCase (from 3N3M_a.pdb; w/ 8 xtal WATs) versus NCI library
45463 - 45494 = Pf ODCase (from 3N3M_a.pdb; dry) versus NCI library
45495 - 45526 = Pf ODCase (from 3S9Y_a.pdb; w/ 7 xtal WATs) versus NCI library
45527 - 45558 = Pf ODCase (from 3S9Y_a.pdb; dry) versus NCI library

 

Experiment 68 = 100% completed

[Enamine library = 2,345,014 models of compounds]
45559 - 45793 = Pb ODCase (from 2FDS.pdb; apo = not ligand-bound) versus Enamine library
45794 - 46028 = Pv ODCase (from 2FFC_a.pdb, monomer w/ 6 xtal WATs) versus Enamine
46029 - 46263 = Pv ODCase (from 2FFC_b.pdb; dimer w/ 6 xtal WATs) versus Enamine library
46264 - 46498 = Pv ODCase (from 2FFC_b.pdb; dry dimer) versus Enamine library
46499 - 46733 = Pf ODCase (from 2Q8Z_b.pdb; w/ 6 xtal WATs) versus Enamine library
46734 - 46968 = Pf ODCase (from 2Q8Z_b.pdb; dry) versus Enamine library
46969 - 47203 = Pf ODCase (from 3BPW_a.pdb; w/ 9 xtal WATs) versus Enamine library
47204 - 47438 = Pf ODCase (from 3BPW_a.pdb; dry) versus Enamine library
47439 - 47673 = Pf ODCase (from 3N34_b.pdb; w/ 7 xtal WATs) versus Enamine library
47674 - 47908 = Pf ODCase (from 3N34_b.pdb; dry) versus Enamine library
47909 - 48143 = Pf ODCase (from 3N3M_a.pdb; w/ 8 xtal WATs) versus Enamine library
48144 - 48378 = Pf ODCase (from 3N3M_a.pdb; dry) versus Enamine library
48379 - 48613 = Pf ODCase (from 3S9Y_a.pdb; w/ 7 xtal WATs) versus Enamine library
48614 - 48848 = Pf ODCase (from 3S9Y_a.pdb; dry) versus Enamine library

 

Experiment 69 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
48849 - 48950 = Pb ODCase (from 2FDS.pdb; apo = not ligand-bound) versus ChemBridge library
48951 - 49052 = Pv ODCase (from 2FFC_a.pdb, monomer w/ 6 xtal WATs) versus ChemBridge library
49053 - 49154 = Pv ODCase (from 2FFC_b.pdb; dimer w/ 6 xtal WATs) versus ChemBridge library
49155 - 49256 = Pv ODCase (from 2FFC_b.pdb; dry dimer) versus ChemBridge library
49257 - 49358 = Pf ODCase (from 2Q8Z_b.pdb; w/ 6 xtal WATs) versus ChemBridge library
49359 - 49460 = Pf ODCase (from 2Q8Z_b.pdb; dry) versus ChemBridge library
49461 - 49562 = Pf ODCase (from 3BPW_a.pdb; w/ 9 xtal WATs) versus ChemBridge library
49563 - 49664 = Pf ODCase (from 3BPW_a.pdb; dry) versus ChemBridge library
49665 - 49766 = Pf ODCase (from 3N34_b.pdb; w/ 7 xtal WATs) versus ChemBridge library
49767 - 49868 = Pf ODCase (from 3N34_b.pdb; dry) versus ChemBridge library
49869 - 49970 = Pf ODCase (from 3N3M_a.pdb; w/ 8 xtal WATs) versus ChemBridge library
49971 - 50072 = Pf ODCase (from 3N3M_a.pdb; dry) versus ChemBridge library
50073 - 50174 = Pf ODCase (from 3S9Y_a.pdb; w/ 7 xtal WATs) versus ChemBridge library
50175 - 50276 = Pf ODCase (from 3S9Y_a.pdb; dry) versus ChemBridge library

 

Experiment 70 = 100% completed

[Asinex library = 507,000 models of compounds]
50277 - 50327 = Pb ODCase (from 2FDS.pdb; apo = not ligand-bound) versus Asinex library
50328 - 50378 = Pv ODCase (from 2FFC_a.pdb, monomer w/ 6 xtal WATs) versus Asinex library
50379 - 50429 = Pv ODCase (from 2FFC_b.pdb; dimer w/ 6 xtal WATs) versus Asinex library
50430 - 50480 = Pv ODCase (from 2FFC_b.pdb; dry dimer) versus Asinex library
50481 - 50531 = Pf ODCase (from 2Q8Z_b.pdb; w/ 6 xtal WATs) versus Asinex library
50532 - 50582 = Pf ODCase (from 2Q8Z_b.pdb; dry) versus Asinex library
50583 - 50633 = Pf ODCase (from 3BPW_a.pdb; w/ 9 xtal WATs) versus Asinex library
50634 - 50684 = Pf ODCase (from 3BPW_a.pdb; dry) versus Asinex library
50685 - 50735 = Pf ODCase (from 3N34_b.pdb; w/ 7 xtal WATs) versus Asinex library
50736 - 50786 = Pf ODCase (from 3N34_b.pdb; dry) versus Asinex library
50787 - 50837 = Pf ODCase (from 3N3M_a.pdb; w/ 8 xtal WATs) versus Asinex library
50838 - 50888 = Pf ODCase (from 3N3M_a.pdb; dry) versus Asinex library
50889 - 50939 = Pf ODCase (from 3S9Y_a.pdb; w/ 7 xtal WATs) versus Asinex library
50940 - 50990 = Pf ODCase (from 3S9Y_a.pdb; dry) versus Asinex library

 

Experiment 71 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
50991 - 51141 = Pb ODCase (from 2FDS.pdb; apo = not ligand-bound) versus Vitas-M Labs library
51142 - 51292 = Pv ODCase (from 2FFC_a.pdb, monomer w/ 6 xtal WATs) versus Vitas-M Labs library
51293 - 51443 = Pv ODCase (from 2FFC_b.pdb; dimer w/ 6 xtal WATs) versus Vitas-M Labs library
51444 - 51594 = Pv ODCase (from 2FFC_b.pdb; dry dimer) versus Vitas-M Labs library
51595 - 51745 = Pf ODCase (from 2Q8Z_b.pdb; w/ 6 xtal WATs) versus Vitas-M Labs library
51746 - 51896 = Pf ODCase (from 2Q8Z_b.pdb; dry) versus Vitas-M Labs library
51897 - 52047 = Pf ODCase (from 3BPW_a.pdb; w/ 9 xtal WATs) versus Vitas-M Labs library
52048 - 52198 = Pf ODCase (from 3BPW_a.pdb; dry) versus Vitas-M Labs library
52199 - 52349 = Pf ODCase (from 3N34_b.pdb; w/ 7 xtal WATs) versus Vitas-M Labs library
52350 - 52500 = Pf ODCase (from 3N34_b.pdb; dry) versus Vitas-M Labs library
52501 - 52651 = Pf ODCase (from 3N3M_a.pdb; w/ 8 xtal WATs) versus Vitas-M Labs library
52652 - 52802 = Pf ODCase (from 3N3M_a.pdb; dry) versus Vitas-M Labs library
52803 - 52953 = Pf ODCase (from 3S9Y_a.pdb; w/ 7 xtal WATs) versus Vitas-M Labs library
52954 - 53104 = Pf ODCase (from 3S9Y_a.pdb; dry) versus Vitas-M Labs library




Notes:
Pb = Plasmodium berghei (the strain of malaria that infects mice; mice are used as a model system to study malaria infections)
Pv = Plasmodium vivax (a common strain of malaria, but it is not as deadly)
Pf = Plasmodium falciparum (the deadliest strain of malaria)


XXXX_a.pdb  or XXXX_b.pdb = either the alternate conformation "A" or alternate conformation "B" from the crystal structure was used for the model of the target, respectively (since it performed better in positive control experiments).

For 2FFC.pdb, the original crystal structure was a monomer, but the biologically active form is a dimer.  PyMOL was used (with the symmetry information in the pdb file) to create a model of the full dimeric structure.  Both the monomer and the dimer performed well in positive control experiments, so for this and only this structure, we included both as targets.  The other targets are all dimers.

w/ N xtal WATs = N water molecules from the original crystallographic structure were included as part of the target, since that increased the accuracy of the positive control experiments.

dry = the crystallographic water molecules were not included as part of the model of the target, since we also want to search for compounds that can displace/replace some of these waters.

 

 

These images represent the results of positive control re-docking experiments against the version of ODCase from Plasmodium vivax (in 2FFC.pdb).  The dry version of the target (i.e., the model that did not contain any crystallographic water molecules) performed very well in these positive control experiments.  The image on the left is a close-up view of the binding site, while the image on the right shows the full target.  The X-ray crystallographic conformation of U5P is displayed as sticks with light purple carbon atoms, while the binding mode predicted by Vina has green carbon atoms.

 

 


 

 

Target #14 = Profilin

Profilin is a potential drug target for curing malaria infections.  Profilins have a critical role--they are one of the main regulators of actin dynamics.  Actin is a protein that polymerizes to form "microfilaments," which are one of the thee main components of the cytoskeleton.  Profilins sequester the actin monomers (that is, they bind to the monomers and gather them together to form a pool).  By forming a pool of actin monomers, it allows the parasite to polymerize these monomers together to form actin microfilaments more easily and more rapidly.  Plasmodium parasites use a highly specialized system of microfilaments for both moving around inside a mosquito or inside a human and for invading the cells within the infected host.  The malaria parasite's genome contains only a small number of proteins that can regulate actin.  Of these important actin regulators, profilin seems to be the only one for which atomically-detailed structures are available.  The critical importance of profilin for the malaria parasite is supported by "reverse genetics" studies in rodents (using Plasmodium berghei, the strain of malaria that infects mice).  These experiments in rodents show that profilin is absolutely essential for the ability of the parasite to invade blood cells, which makes profilin a very attractive target for therapeutic intervention.

We are targeting two different sites on profilin:  the actin binding site and the poly-proline peptide binding site.  The poly-proline peptide binding site is involved in proflin's interactions with different scaffolding proteins (that is, with proteins that control where profilin is located within the parasite and when it can try to interact with other proteins).  The actin binding site is used by profilin to sequester the actin monomers to form a pool (that can then polymerize into microfilaments).  Blocking either of these binding sites on profilin might be useful, and blocking both of them should be even better.  However, we want to discover compounds that preferentially bind to the malarial version of profilin but that do not bind strongly to human profilin, since we do not want to disrupt the cytoskeleton within our human cells.  Consequently, we are including several different structures of human profilin in these experiments, to enable us to harvest compounds that are predicted to both bind strongly to Plasmodium profilin and to bind weakly (or not at all) to human profilin.

 

Batch numbers against the poly-proline peptide binding site of profilin:

Experiment 72 = 100% completed

[full NCI library = 316,179 models of compounds]
53105 - 53136 = human platelet profilin (from 1AWI.pdb) versus NCI library
53137 - 53168 = human platelet profilin (from 1CF0.pdb) versus NCI library
53169 - 53200 = human profilin II (from 1D1J.pdb) versus NCI library
53201 - 53232 = human profilin I (from 1PFL.pdb; NMR structure) versus NCI library
53233 - 53264 = Pf profilin (from 2JKF.pdb) versus NCI library
53265 - 53296 = Pf profilin (from 2JKG.pdb) versus NCI library
53297 - 53328 = human profilin I (from 2PBD, A conformation) versus NCI library

 

Experiment 73 = 100% completed

[Enamine library = 2,345,014 models of compounds]
53329 - 53563 = human platelet profilin (from 1AWI.pdb) versus Enamine library
53564 - 53798 = human platelet profilin (from 1CF0.pdb) versus Enamine library
53799 - 54033 = human profilin II (from 1D1J.pdb) versus Enamine library
54034 - 54268 = human profilin I (from 1PFL.pdb; NMR structure) versus Enamine library
54269 - 54503 = Pf profilin (from 2JKF.pdb) versus Enamine library
54504 - 54738 = Pf profilin (from 2JKG.pdb) versus Enamine library
54739 - 54973 = human profilin I (from 2PBD, A conformation) versus Enamine library

 

Experiment 74 =100% completed

[ChemBridge library = 1,013,483 models of compounds]
54974 - 55075 = human platelet profilin (from 1AWI.pdb) versus ChemBridge library
55076 - 55177 = human platelet profilin (from 1CF0.pdb) versus ChemBridge library
55178 - 55279 = human profilin II (from 1D1J.pdb) versus ChemBridge library
55280 - 55381 = human profilin I (from 1PFL.pdb; NMR structure) versus ChemBridge library
55382 - 55483 = Pf profilin (from 2JKF.pdb) versus ChemBridge library
55484 - 55585 = Pf profilin (from 2JKG.pdb) versus ChemBridge library
55586 - 55687 = human profilin I (from 2PBD, A conformation) versus ChemBridge library

 

Experiment 75 = 100% completed

[Asinex library = 507,000 models of compounds]
55688 - 55738 = human platelet profilin (from 1AWI.pdb) versus Asinex library
55739 - 55789 = human platelet profilin (from 1CF0.pdb) versus Asinex library
55790 - 55840 = human profilin II (from 1D1J.pdb) versus Asinex library
55841 - 55891 = human profilin I (from 1PFL.pdb; NMR structure) versus Asinex library
55892 - 55942 = Pf profilin (from 2JKF.pdb) versus Asinex library
55943 - 55993 = Pf profilin (from 2JKG.pdb) versus Asinex library
55994 - 56044 = human profilin I (from 2PBD, A conformation) versus Asinex library

 

Experiment 76 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
56045 - 56195 = human platelet profilin (from 1AWI.pdb) versus Vitas-M Labs library
56196 - 56346 = human platelet profilin (from 1CF0.pdb) versus Vitas-M Labs library
56347 - 56497 = human profilin II (from 1D1J.pdb) versus Vitas-M Labs library
56498 - 56648 = human profilin I (from 1PFL.pdb; NMR structure) versus Vitas-M Labs library
56649 - 56799 = Pf profilin (from 2JKF.pdb) versus Vitas-M Labs library
56800 - 56950 = Pf profilin (from 2JKG.pdb) versus Vitas-M Labs library
56951 - 57101 = human profilin I (from 2PBD, A conformation) versus Vitas-M Labs library

 

 

 

Batch numbers against the actin binding site of profilin:

Experiment 77 = 100% completed

[full NCI library = 316,179 models of compounds]
57102 - 57133 = human platelet profilin (from 1AWI.pdb), actin binding site, versus NCI library
57134 - 57165 = human platelet profilin (from 1CF0.pdb), actin binding site, versus NCI library
57166 - 57197 = human profilin II (from 1D1J.pdb), actin binding site, versus NCI library
57198 - 57229 = human profilin I (from 1PFL.pdb; NMR structure), actin binding site, versus NCI library
57230 - 57261 = Pf profilin (from 2JKF.pdb), actin binding site, versus NCI library
57262 - 57293 = Pf profilin (from 2JKG.pdb), actin binding site, versus NCI library
57294 - 57325 = human profilin I (from 2PBD, A conformation), actin binding site, versus NCI library

 

Experiment 78 = 100% completed

[Enamine library = 2,345,014 models of compounds]
57326 - 57560 = human platelet profilin (from 1AWI.pdb), actin binding site, versus Enamine library
57561 - 57795 = human platelet profilin (from 1CF0.pdb), actin binding site, versus Enamine library
57796 - 58030 = human profilin II (from 1D1J.pdb), actin binding site, versus Enamine library
58031 - 58265 = human profilin I (from 1PFL.pdb; NMR structure), actin binding site, versus Enamine library
58266 - 58500 = Pf profilin (from 2JKF.pdb), actin binding site, versus Enamine library
58501 - 58735 = Pf profilin (from 2JKG.pdb), actin binding site, versus Enamine library
58736 - 58970 = human profilin I (from 2PBD, A conformation), actin binding site, versus Enamine library

 

Experiment 79 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
58971 - 59072 = human platelet profilin (from 1AWI.pdb), actin binding site, versus ChemBridge library
59073 - 59174 = human platelet profilin (from 1CF0.pdb), actin binding site, versus ChemBridge library
59175 - 59276 = human profilin II (from 1D1J.pdb), actin binding site, versus ChemBridge library
59277 - 59378 = human profilin I (from 1PFL.pdb; NMR structure), actin binding site, versus ChemBridge library
59379 - 59480 = Pf profilin (from 2JKF.pdb), actin binding site, versus ChemBridge library
59481 - 59582 = Pf profilin (from 2JKG.pdb), actin binding site, versus ChemBridge library
59583 - 59684 = human profilin I (from 2PBD, A conformation), actin binding site, versus ChemBridge library

 

Experiment 80 = 100% completed

[Asinex library = 507,000 models of compounds]
59685 - 59735 = human platelet profilin (from 1AWI.pdb), actin binding site, versus Asinex library
59736 - 59786 = human platelet profilin (from 1CF0.pdb), actin binding site, versus Asinex library
59787 - 59837 = human profilin II (from 1D1J.pdb), actin binding site, versus Asinex library
59838 - 59888 = human profilin I (from 1PFL.pdb; NMR structure), actin binding site, versus Asinex library
59889 - 59939 = Pf profilin (from 2JKF.pdb), actin binding site, versus Asinex library
59940 - 59990 = Pf profilin (from 2JKG.pdb), actin binding site, versus Asinex library
59991 - 60041 = human profilin I (from 2PBD, A conformation), actin binding site, versus Asinex library

 

Experiment 81 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
60042 - 60192 = human platelet profilin (from 1AWI.pdb), actin binding site, versus Vitas-M Lab library
60193 - 60343 = human platelet profilin (from 1CF0.pdb), actin binding site, versus Vitas-M Lab library
60344 - 60494 = human profilin II (from 1D1J.pdb), actin binding site, versus Vitas-M Lab library
60495 - 60645 = human profilin I (from 1PFL.pdb; NMR structure), actin binding site, versus Vitas-M Lab library
60646 - 60796 = Pf profilin (from 2JKF.pdb), actin binding site, versus Vitas-M Lab library
60797 - 60947 = Pf profilin (from 2JKG.pdb), actin binding site, versus Vitas-M Lab library
60948 - 61098 = human profilin I (from 2PBD, A conformation), actin binding site, versus Vitas-M Lab library

 

 

Notes:
1AWI = human platelet profilin complexed with L-Pro10 peptide, 2.2 A
1CF0 = human platelet profilin complexed with L-Pro10-IodoTyr peptide, 2.2 A
1D1J = human profilin II, apo, 2.2 A
1PFL = NMR structure of human profilin I, apo
2JKF = Plasmodium falciparum profilin, apo, 2.31 A
2JKG = Plasmodium falciparum profilin bound to OctaProline peptide, 1.89 A
2PBDa = human profilin I bound to actin and to poly-Pro-GAB domain of VASP (Vasodilator-stimulated phosphoprotein), alternate conformation "A" of side-chains, 1.5 A

 

 

 

 


*******************jumped back to target class #6 = PfSUB1, to submit new experiments versus

           a second brand new, not-yet-published crystal structure from Chrislaine Withers-Martinez

                   of Mike Blackman's group at MRC's NIMR, UK.  *************************


 

 

 


 

 

Target #15 =  3-oxoacyl acyl-carrier-protein reductase

3-oxoacyl acyl-carrier-protein reductase (OAR, which is also called FabG), is a potential drug target for curing malaria.  OAR is also called beta-ketoacyl-acyl-carrier-protein reductase.  This enzyme performs the second reductive step in fatty acid synthesis (FAS).  Importantly, this type II FAS pathway used by the malaria parasite is a path that is not present in human cells, which reduces the probability that Pf OAR inhibitors would be toxic to us.  This type II FAS pathway occurs in the "apicoplast," which is a specialized organelle within the parasite.  Human cells do not have apicoplasts.  See target #2 = ENR for additional details on why targeting this FAS pathway should be useful.

The small molecule compound "hexachlorophene" and analogs of it (such as bithionol) have been shown to possess antimalarial activity in vitro (that is, in test tubes), which highlights the therapeutic potential of disabling the activity of this enzyme.  Hexachlorophene inhibits Pf OAR with an EC50 of 6.2 micro-Molar.  Since hexachlorophene is a fairly weak inhibitor of Pf OAR, we are searching for more potent compounds in these experiments.  If we do not find something more potent than hexachlorophene, it will still be useful to discover new types of weak inhibitors that are structurally different than hexachlorophene, to provide new scaffolds that can hopefully be optimized into more potent compounds by medicinal chemists.  Having many different types of scaffolds against a particular enzyme enhances our understanding of the structure-function relationships that govern the activity of a target, and it increases the chances that at least one of them could be developed into an actual drug.

 

These images depict the results of positive control re-docking experiments against the version of OAR from Staphylococcus aureus (from 3SJ7.pdb).  The image on the left is a close-up view of the binding site, while the image on the right is the full view.  The crystallographic conformation of NADP has green carbon atoms, while the binding mode predicted by Vina calculations has purple carbon atoms.  The two images at the bottom of the description for OAR represent the same information, but those images display the surface of OAR (while the two above images show the cartoon mode of the backbone of OAR).

 

 

Batch numbers for the experiments versus OAR (also known as FabG):

Experiment 87 = 100% completed

[full NCI library = 316,179 models of compounds]
62241 - 62272 = E.coli OAR (from 1Q7B.pdb) vs NCI library
62273 - 62304 = TB OAR (from 1UZM.pdb; conf. A; w/ 5 WATs) vs NCI library
62305 - 62336 = TB OAR (from 1UZM.pdb; conf. A) vs NCI library
62337 - 62368 = TB OAR (from 1UZM.pdb; conf. B; w/ 5 WATs) vs NCI library
62369 - 62400 = TB OAR (from 1UZM.pdb; conf. B) vs NCI library
62401 - 62432 = TB OAR (from 1UZN.pdb; w/ 6 WATs) vs NCI library
62433 - 62464 = TB OAR (from 1UZN.pdb) vs NCI library
62465 - 62496 = Pf OAR (from 2C07.pdb; conf. A; loop_amber; w/ 9 WATs) vs NCI library
62497 - 62528 = Pf OAR (from 2C07.pdb; conf. A; loop_amber) vs NCI library
62529 - 62560 = Pf OAR (from 2C07.pdb; conf. A; loop_charm; w/ 9 WATs) vs NCI library
62561 - 62592 = Pf OAR (from 2C07.pdb; conf. A; loop_charm) vs NCI library
62593 - 62624 = Pf OAR (from 2C07.pdb; conf. B; loop_amber; w/ 9 WATs) vs NCI library
62625 - 62656 = Pf OAR (from 2C07.pdb; conf. B; loop_amber) vs NCI library
62657 - 62688 = Staph a. OAR (from 3OSU.pdb; w/ 5 WATs) vs NCI library
62689 - 62720 = Staph a. OAR (from 3OSU.pdb; w/ 9 WATs) vs NCI library
62721 - 62752 = Staph a. OAR (from 3OSU.pdb) vs NCI library
62753 - 62784 = Staph a. OAR (from 3SJ7) vs NCI library
62785 - 62816 = TB OAR (from 1UZN.pdb; w/ NADP) vs NCI library
62817 - 62848 = Staph a. OAR (from 3SJ7; w/ NADP) vs NCI library

 

Experiment 88 = 100% completed

[Enamine library = 2,345,014 models of compounds]
62849 - 63083 = E.coli OAR (from 1Q7B.pdb) vs Enamine library
63084 - 63318 = TB OAR (from 1UZM.pdb; conf. A; w/ 5 WATs) vs Enamine library
63319 - 63553 = TB OAR (from 1UZM.pdb; conf. A) vs Enamine library
63554 - 63788 = TB OAR (from 1UZM.pdb; conf. B; w/ 5 WATs) vs Enamine library
63789 - 64023 = TB OAR (from 1UZM.pdb; conf. B) vs Enamine library
64024 - 64258 = TB OAR (from 1UZN.pdb; w/ 6 WATs) vs Enamine library
64259 - 64493 = TB OAR (from 1UZN.pdb) vs Enamine library
64494 - 64728 = Pf OAR (from 2C07.pdb; conf. A; loop_amber; w/ 9 WATs) vs Enamine library
64729 - 64963 = Pf OAR (from 2C07.pdb; conf. A; loop_amber) vs Enamine library
64964 - 65198 = Pf OAR (from 2C07.pdb; conf. A; loop_charm; w/ 9 WATs) vs Enamine library
65199 - 65433 = Pf OAR (from 2C07.pdb; conf. A; loop_charm) vs Enamine library
65434 - 65668 = Pf OAR (from 2C07.pdb; conf. B; loop_amber; w/ 9 WATs) vs Enamine library
65669 - 65903 = Pf OAR (from 2C07.pdb; conf. B; loop_amber) vs Enamine library
65904 - 66138 = Staph a. OAR (from 3OSU.pdb; w/ 5 WATs) vs Enamine library
66139 - 66373 = Staph a. OAR (from 3OSU.pdb; w/ 9 WATs) vs Enamine library
66374 - 66608 = Staph a. OAR (from 3OSU.pdb) vs Enamine library
66609 - 66843 = Staph a. OAR (from 3SJ7) vs Enamine library
66844 - 67078 = TB OAR (from 1UZN.pdb; w/ NADP) vs Enamine library
67079 - 67313 = Staph a. OAR (from 3SJ7; w/ NADP) vs Enamine library

Experiment 89 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
67314 - 67415 = E.coli OAR (from 1Q7B.pdb) vs ChemBridge library
67416 - 67517 = TB OAR (from 1UZM.pdb; conf. A; w/ 5 WATs) vs ChemBridge library
67518 - 67619 = TB OAR (from 1UZM.pdb; conf. A) vs ChemBridge library
67620 - 67721 = TB OAR (from 1UZM.pdb; conf. B; w/ 5 WATs) vs ChemBridge library
67722 - 67823 = TB OAR (from 1UZM.pdb; conf. B) vs ChemBridge library
67824 - 67925 = TB OAR (from 1UZN.pdb; w/ 6 WATs) vs ChemBridge library
67926 - 68027 = TB OAR (from 1UZN.pdb) vs ChemBridge library
68028 - 68129 = Pf OAR (from 2C07.pdb; conf. A; loop_amber; w/ 9 WATs) vs ChemBridge library
68130 - 68231 = Pf OAR (from 2C07.pdb; conf. A; loop_amber) vs ChemBridge library
68232 - 68333 = Pf OAR (from 2C07.pdb; conf. A; loop_charm; w/ 9 WATs) vs ChemBridge library
68334 - 68435 = Pf OAR (from 2C07.pdb; conf. A; loop_charm) vs ChemBridge library
68436 - 68537 = Pf OAR (from 2C07.pdb; conf. B; loop_amber; w/ 9 WATs) vs ChemBridge library
68538 - 68639 = Pf OAR (from 2C07.pdb; conf. B; loop_amber) vs ChemBridge library
68640 - 68741 = Staph a. OAR (from 3OSU.pdb; w/ 5 WATs) vs ChemBridge library
68742 - 68843 = Staph a. OAR (from 3OSU.pdb; w/ 9 WATs) vs ChemBridge library
68844 - 68945 = Staph a. OAR (from 3OSU.pdb) vs ChemBridge library
68946 - 69047 = Staph a. OAR (from 3SJ7) vs ChemBridge library
69048 - 69149 = TB OAR (from 1UZN.pdb; w/ NADP) vs ChemBridge library
69150 - 69251 = Staph a. OAR (from 3SJ7; w/ NADP) vs ChemBridge library
 

Experiment 90 = 100% completed

[Asinex library = 507,000 models of compounds]
69252 - 69302 = E.coli OAR (from 1Q7B.pdb) vs Asinex library
69303 - 69353 = TB OAR (from 1UZM.pdb; conf. A; w/ 5 WATs) vs Asinex library
69354 - 69404 = TB OAR (from 1UZM.pdb; conf. A) vs Asinex library
69405 - 69455 = TB OAR (from 1UZM.pdb; conf. B; w/ 5 WATs) vs Asinex library
69456 - 69506 = TB OAR (from 1UZM.pdb; conf. B) vs Asinex library
69507 - 69557 = TB OAR (from 1UZN.pdb; w/ 6 WATs) vs Asinex library
69558 - 69608 = TB OAR (from 1UZN.pdb) vs Asinex library
69609 - 69659 = Pf OAR (from 2C07.pdb; conf. A; loop_amber; w/ 9 WATs) vs Asinex library
69660 - 69710 = Pf OAR (from 2C07.pdb; conf. A; loop_amber) vs Asinex library
69711 - 69761 = Pf OAR (from 2C07.pdb; conf. A; loop_charm; w/ 9 WATs) vs Asinex library
69762 - 69812 = Pf OAR (from 2C07.pdb; conf. A; loop_charm) vs Asinex library
69813 - 69863 = Pf OAR (from 2C07.pdb; conf. B; loop_amber; w/ 9 WATs) vs Asinex library
69864 - 69914 = Pf OAR (from 2C07.pdb; conf. B; loop_amber) vs Asinex library
69915 - 69965 = Staph a. OAR (from 3OSU.pdb; w/ 5 WATs) vs Asinex library
69966 - 70016 = Staph a. OAR (from 3OSU.pdb; w/ 9 WATs) vs Asinex library
70017 - 70067 = Staph a. OAR (from 3OSU.pdb) vs Asinex library
70068 - 70118 = Staph a. OAR (from 3SJ7) vs Asinex library
70119 - 70169 = TB OAR (from 1UZN.pdb; w/ NADP) vs Asinex library
70170 - 70220 = Staph a. OAR (from 3SJ7; w/ NADP) vs Asinex library
 

Experiment 91 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
70221 - 70371 = E.coli OAR (from 1Q7B.pdb) vs Vitas-M Labs library
70372 - 70522 = TB OAR (from 1UZM.pdb; conf. A; w/ 5 WATs) vs Vitas-M Labs library
70523 - 70673 = TB OAR (from 1UZM.pdb; conf. A) vs Vitas-M Labs library
70674 - 70824 = TB OAR (from 1UZM.pdb; conf. B; w/ 5 WATs) vs Vitas-M Labs library
70825 - 70975 = TB OAR (from 1UZM.pdb; conf. B) vs Vitas-M Labs library
70976 - 71126 = TB OAR (from 1UZN.pdb; w/ 6 WATs) vs Vitas-M Labs library
71127 - 71277 = TB OAR (from 1UZN.pdb) vs Vitas-M Labs library
71278 - 71428 = Pf OAR (from 2C07.pdb; conf. A; loop_amber; w/ 9 WATs) vs Vitas-M Labs library
71429 - 71579 = Pf OAR (from 2C07.pdb; conf. A; loop_amber) vs Vitas-M Labs library
71580 - 71730 = Pf OAR (from 2C07.pdb; conf. A; loop_charm; w/ 9 WATs) vs Vitas-M Labs library
71731 - 71881 = Pf OAR (from 2C07.pdb; conf. A; loop_charm) vs Vitas-M Labs library
71882 - 72032 = Pf OAR (from 2C07.pdb; conf. B; loop_amber; w/ 9 WATs) vs Vitas-M Labs library
72033 - 72183 = Pf OAR (from 2C07.pdb; conf. B; loop_amber) vs Vitas-M Labs library
72184 - 72334 = Staph a. OAR (from 3OSU.pdb; w/ 5 WATs) vs Vitas-M Labs library
72335 - 72485 = Staph a. OAR (from 3OSU.pdb; w/ 9 WATs) vs Vitas-M Labs library
72486 - 72636 = Staph a. OAR (from 3OSU.pdb) vs Vitas-M Labs library
72637 - 72787 = Staph a. OAR (from 3SJ7) vs Vitas-M Labs library
72788 - 72938 = TB OAR (from 1UZN.pdb; w/ NADP) vs Vitas-M Labs library
72939 - 73089 = Staph a. OAR (from 3SJ7; w/ NADP) vs Vitas-M Labs library

 

Notes:  since they are structurally similar, the versions of OAR from other pathogens were included in these experiments.  Discovering compounds that can inhibit OAR from several different types of bacteria could be very useful to the medical community, especially since the discovery and development of new types of antibiotics has generally been neglected by most pharmaceutical companies for many decades.  Finding new types of inhibitors that shut down the activity of OAR from any one of these pathogens would also be very useful.

E.coli OAR is the version of this enzyme from Escherichia coli, which are very common bacteria that occasionally cause severe illnesses.

TB OAR is the version of this enzyme found in Mycobacterium tuberculosis, which are the deadliest bacteria on the planet.

Staph a. OAR is the version of this enzyme from Staphylococcus aureus, which are bacteria that cause severe and sometimes deadly infections.  The most well-known drug-resistant form of Staphylococcus aureus is called "MRSA" (for methicillin-resistant Staphylococcus aureus).

"w/ WATS" indicates that this model of the target contains some of the crystallographic water molecules bound in the active site (the region we are trying to block and disable).

Since the crystal structures of Pf OAR were missing the coordinates of a flexible loop that is adjacent to the active site, the program called "loopy" (from Barry Honig's lab) was used to create different models of possible conformations of this flexible loop.  Loopy can be obtained at http://trantor.bioc.columbia.edu or at http://wiki.c2b2.columbia.edu/honiglab_public/index.php/Software .  Different "force fields" were used within loopy to obtain different models of this flexible loop, such as AMBER and CHARMM.

Conf. A versus conf. B indicates which flavor of coordinates were used in the models for some residues of these targets.  In some crystallographic structures of proteins, the electron density data show that two or more different conformations of particular residues are being sampled by the target.  Since both of these conformations are being displayed at different times (and/or being displayed by different copies of the protein at the same time), each conformation was used to create a model of the target.

 

 

 

 


 

 

Target #16 =  beta-hydroxyacyl-acyl-carrier-protein dehydrase = FabZ

FabA/Z  (FabZ), or the beta-hydroxyacyl-acyl-carrier-protein dehydrase, is another enzyme that is part of the type II Fatty Acid Synthesis (FAS) pathway, a path (or group of enzymes that function together in sequence) that is not found in human cells.  The structural and functional differences between these malarial enzymes in the type II pathway and the human enzymes in the type I pathway make these malarial enzymes attractive targets for the discovery of new drugs that selectively kill the parasites.  FabZ is still only a "potential drug target," meaning that it has not yet been validated as a real drug target.  But FabZ has a lot of potential for treating multi-drug-resistant malaria infections.  To support this hypothesis, other enzymes in this type II FAS pathway have been validated as real drug targets:  FabH is inhibited by thiolactomycin, which inhibits Plasmodium falciparum growth in cell cultures, and FabI (also known as ENR) is inhibited by triclosan, which is antiparasitic (but tricolosan is not orally bioavailable; that is, it cannot be turned into a pill).  For additional information on the type II FAS pathway, see target classes #2 (ENR) and #15 (OAR).

FabZ is an enzyme that catalyzes the dehydration of beta-hydroxy fatty acids that are linked to the acyl carrier protein, and it has broad substrate specificity.  This enzyme functions as a dimer, which means that two identical copies of it come together to form one functional unit.  The backbone of each dimer of FabZ displays what is known as the "hot dog" fold, which is very rare and has been observed in only a few other protein structures.

 

Batch numbers for the experiments versus FabZ:

Experiment 92 = 100% completed

[full NCI library = 316,179 models of compounds]
73090 - 73121 = Pf FabZ (from 1Z6B; loop_amber; apo) vs NCI library
73122 - 73153 = Pf FabZ (from 1Z6B; loop_charmm; apo) vs NCI library
73154 - 73185 = Pf FabZ (from 1Z6B; missing loops not added; apo) vs NCI library
73186 - 73217 = Pf FabZ (from 1ZHG; loop_charmm; apo) vs NCI library
73218 - 73249 = Pf FabZ (from 3AZ8; loop_amber; NAS21-bound) vs NCI library
73250 - 73281 = Pf FabZ (from 3AZ9; loop_amber; NAS91-bound) vs NCI library
73282 - 73313 = Pf FabZ (from 3AZ9; loop_charmm; NAS91-bound) vs NCI library
73314 - 73345 = Pf FabZ (from 3AZB; loop_amber; NAS91-11-bound) vs NCI library
73346 - 73377 = Yersinia pestis FabZ (from 3Q62; loop_charmm; apo) vs NCI library

 

Experiment 93 = 100% completed

[Enamine library = 2,345,014 models of compounds]
73378 - 73612 = Pf FabZ (from 1Z6B; loop_amber; apo) vs Enamine library
73613 - 73847 = Pf FabZ (from 1Z6B; loop_charmm; apo) vs Enamine library
73848 - 74082 = Pf FabZ (from 1Z6B; missing loops not added; apo) vs Enamine library
74083 - 74317 = Pf FabZ (from 1ZHG; loop_charmm; apo) vs Enamine library
74318 - 74552 = Pf FabZ (from 3AZ8; loop_amber; NAS21-bound) vs Enamine library
74553 - 74787 = Pf FabZ (from 3AZ9; loop_amber; NAS91-bound) vs Enamine library
74788 - 75022 = Pf FabZ (from 3AZ9; loop_charmm; NAS91-bound) vs Enamine library
75023 - 75257 = Pf FabZ (from 3AZB; loop_amber; NAS91-11-bound) vs Enamine library
75258 - 75492 = Yersinia pestis FabZ (from 3Q62; loop_charmm; apo) vs Enamine library

Experiment 94 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
75493 - 75594 = Pf FabZ (from 1Z6B; loop_amber; apo) vs ChemBridge library
75595 - 75696 = Pf FabZ (from 1Z6B; loop_charmm; apo) vs ChemBridge library
75697 - 75798 = Pf FabZ (from 1Z6B; missing loops not added; apo) vs ChemBridge library
75799 - 75900 = Pf FabZ (from 1ZHG; loop_charmm; apo) vs ChemBridge library
75901 - 76002 = Pf FabZ (from 3AZ8; loop_amber; NAS21-bound) vs ChemBridge library
76003 - 76104 = Pf FabZ (from 3AZ9; loop_amber; NAS91-bound) vs ChemBridge library
76105 - 76206 = Pf FabZ (from 3AZ9; loop_charmm; NAS91-bound) vs ChemBridge library
76207 - 76308 = Pf FabZ (from 3AZB; loop_amber; NAS91-11-bound) vs ChemBridge library
76309 - 76410 = Yersinia pestis FabZ (from 3Q62; loop_charmm; apo) vs ChemBridge library
 

Experiment 95 = 100% completed

[Asinex library = 507,000 models of compounds]
76411 - 76461 = Pf FabZ (from 1Z6B; loop_amber; apo) vs ChemBridge library
76462 - 76512 = Pf FabZ (from 1Z6B; loop_charmm; apo) vs ChemBridge library
76513 - 76563 = Pf FabZ (from 1Z6B; missing loops not added; apo) vs ChemBridge library
76564 - 76614 = Pf FabZ (from 1ZHG; loop_charmm; apo) vs ChemBridge library
76615 - 76665 = Pf FabZ (from 3AZ8; loop_amber; NAS21-bound) vs ChemBridge library
76666 - 76716 = Pf FabZ (from 3AZ9; loop_amber; NAS91-bound) vs ChemBridge library
76717 - 76767 = Pf FabZ (from 3AZ9; loop_charmm; NAS91-bound) vs ChemBridge library
76768 - 76818 = Pf FabZ (from 3AZB; loop_amber; NAS91-11-bound) vs ChemBridge library
76819 - 76869 = Yersinia pestis FabZ (from 3Q62; loop_charmm; apo) vs ChemBridge library
 

Experiment 96 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
76870 - 77020 = Pf FabZ (from 1Z6B; loop_amber; apo) vs ChemBridge library
77021 - 77171 = Pf FabZ (from 1Z6B; loop_charmm; apo) vs ChemBridge library
77172 - 77322 = Pf FabZ (from 1Z6B; missing loops not added; apo) vs ChemBridge library
77323 - 77473 = Pf FabZ (from 1ZHG; loop_charmm; apo) vs ChemBridge library
77474 - 77624 = Pf FabZ (from 3AZ8; loop_amber; NAS21-bound) vs ChemBridge library
77625 - 77775 = Pf FabZ (from 3AZ9; loop_amber; NAS91-bound) vs ChemBridge library
77776 - 77926 = Pf FabZ (from 3AZ9; loop_charmm; NAS91-bound) vs ChemBridge library
77927 - 78077 = Pf FabZ (from 3AZB; loop_amber; NAS91-11-bound) vs ChemBridge library
78078 - 78228 = Yersinia pestis FabZ (from 3Q62; loop_charmm; apo) vs ChemBridge library

 

Note:  the crystal structures for FabZ had a few missing loops (that is, those loops were likely too flexible or polymorphic to difract well; thus, the crystal structures contained no coordinates for those loops).  The "loopy" tool from the Honig lab was used to create models for these missing loops.  If the AMBER force-field was used to make the model within loopy, then "loop_amber" is listed above.  Conversely, if the CHARMM force-field was used to make the model within loopy, then "loop_charmm" is listed above).

Note:  Although most of these FabZ targets are from malaria (that is, from Plasmodium falciparum), one of the targets is from Yersinia pestisYersinia pestis causes bubonic plague, which is also known as the "black death" or the "Justinian plague".  Although bubonic plague is not very prevalent these days, it is a potential bio-terror weapon, which is why we included it as a target in these experiments.

 

 

 


*******************jumped back to target class #6 = PfSUB1, to submit new experiments versus

           a third brand new, not-yet-published crystal structure from Chrislaine Withers-Martinez

                   of Mike Blackman's group at MRC's NIMR, UK.  *************************


 

 

 

 

 

 


 

 

Target #17 =  actin depolymerization factor 1 and 2 = ADF1 and ADF2

Plasmodium parasites use an actin-based motor that allows them to rapidly move around within the infected human or mosquito hosts.   The parasite also uses this actin-based motor in the process of invading the cells of the host.  The actin-based motor is called "actomyosin," and it interacts with short, dynamic actin filaments and generates the force that allows the parasite to swim around in our bloodstream and to invade our cells.  In order for the parasite to be able to move rapidly, its unusually short actin filaments must be able to turnover very quickly (that is, they have to be able to rapidly polymerize and depolymerize).  In human cells, approximately 50% of the actin is present in the polymerized form (called "actin filaments"), while the other half is depolymerized (that is, it exists in monomeric form and is bound to different sequestering proteins).  Unlike human cells, in the parasite most of the actin is present in the monomeric form, and it forms only transient (short-lived), short filaments.  The polymerization and depolymerization process in the parasite is controlled by a small set of regulatory proteins.  These regulatory proteins in the parasite have different sequences, structures, and biochemical properties than the types of actin regulators that exist in human cells.  The most critically important actin regulators in the Plasmodium parasites include the "actin depolymerization factors," ADF1 and ADF2.

In these experiments versus ADF1 and ADF2, two different regions of this protein are being targeted:  the first set of experiments target the site that interacts with actin monomers, while the second set of experiments target a different site that is involved in severing the actin filaments.  ADF2 plays a less important role for the parasite:  parasites that are engineered to lack ADF2 display relatively mild defects at later stages of the parasites lifecycle.  But ADF1 is the major, essential regulator of actin filaments.  If we can discover compounds that impede the ability of the parasite to rapidly control the length and stability of its actin filaments, we should be able to prevent the parasite from being able to move around within our bodies and prevent it from being able to invade our blood and liver cells.  Consequently, the ADFs are potential drug targets for curing malaria infections.

 

Batch numbers for the experiments versus ADF1 and ADF2, box 1 (targets actin monomer binding site):

Experiment 102 = 100% completed

[full NCI library = 316,179 models of compounds]
79371 - 79402 = human ADF2, N-terminal domain (from 2VAC; conf A) vs NCI library
79403 - 79434 = human ADF2, C-terminal domain (from 2W0I.pdb; conf A) vs NCI library
79435 - 79466 = Pf ADF1 (from 2XF1.pdb; conf A) vs NCI library
79467 - 79498 = Pb ADF2 (from 2XFA.pdb) vs NCI library
79499 - 79530 = Pf ADF1 (from 3Q2B.pdb; conf A) vs NCI library
79531 - 79562 = Pf ADF1 (from 3Q2B.pdb; conf B) vs NCI library

 

Experiment 103 = 100% completed

[Enamine library = 2,345,014 models of compounds]
79563 - 79797 = human ADF2, N-terminal domain (from 2VAC; conf A) vs Enamine library
79798 - 80032 = human ADF2, C-terminal domain (from 2W0I.pdb; conf A) vs Enamine library
80033 - 80267 = Pf ADF1 (from 2XF1.pdb; conf A) vs Enamine library
80268 - 80502 = Pb ADF2 (from 2XFA.pdb) vs Enamine library
80503 - 80737 = Pf ADF1 (from 3Q2B.pdb; conf A) vs Enamine library
80738 - 80972 = Pf ADF1 (from 3Q2B.pdb; conf B) vs Enamine library

Experiment 104 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
80973 - 81074 = human ADF2, N-terminal domain (from 2VAC; conf A) vs ChemBridge library
81075 - 81176 = human ADF2, C-terminal domain (from 2W0I.pdb; conf A) vs ChemBridge library
81177 - 81278 = Pf ADF1 (from 2XF1.pdb; conf A) vs ChemBridge library
81279 - 81380 = Pb ADF2 (from 2XFA.pdb) vs ChemBridge library
81381 - 81482 = Pf ADF1 (from 3Q2B.pdb; conf A) vs ChemBridge library
81483 - 81584 = Pf ADF1 (from 3Q2B.pdb; conf B) vs ChemBridge library
 

Experiment 105 = 100% completed

[Asinex library = 507,000 models of compounds]
81585 - 81635 = human ADF2, N-terminal domain (from 2VAC; conf A) vs Asinex library
81636 - 81686 = human ADF2, C-terminal domain (from 2W0I.pdb; conf A) vs Asinex library
81687 - 81737 = Pf ADF1 (from 2XF1.pdb; conf A) vs Asinex library
81738 - 81788 = Pb ADF2 (from 2XFA.pdb) vs Asinex library
81789 - 81839 = Pf ADF1 (from 3Q2B.pdb; conf A) vs Asinex library
81840 - 81890 = Pf ADF1 (from 3Q2B.pdb; conf B) vs Asinex library
 

Experiment 106 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
81891 - 82041 = human ADF2, N-terminal domain (from 2VAC; conf A) vs Vitas-M Labs library
82042 - 82192 = human ADF2, C-terminal domain (from 2W0I.pdb; conf A) vs Vitas-M Labs library
82193 - 82343 = Pf ADF1 (from 2XF1.pdb; conf A) vs Vitas-M Labs library
82344 - 82494 = Pb ADF2 (from 2XFA.pdb) vs Vitas-M Labs library
82495 - 82645 = Pf ADF1 (from 3Q2B.pdb; conf A) vs Vitas-M Labs library
82646 - 82796 = Pf ADF1 (from 3Q2B.pdb; conf B) vs Vitas-M Labs library

 

+++++++++++++++++++++

 

Batch numbers for the experiments versus ADF1 and ADF2, box 2 (targets ability to sever actin filaments):

Experiment 107 = 100% completed

[full NCI library = 316,179 models of compounds]
82797 - 82828 = human ADF2 site 2, N-terminal domain (from 2VAC; conf A) vs NCI library
82829 - 82860 = human ADF2 site 2, C-terminal domain (from 2W0I.pdb; conf A) vs NCI library
82861 - 82892 = Pf ADF1 site 2 (from 2XF1.pdb; conf A) vs NCI library
82893 - 82924 = Pb ADF2 site 2 (from 2XFA.pdb) vs NCI library
82925 - 82956 = Pf ADF1 site 2 (from 3Q2B.pdb; conf A) vs NCI library
82957 - 82988 = Pf ADF1 site 2 (from 3Q2B.pdb; conf B) vs NCI library

 

Experiment 108 = 100% completed

[Enamine library = 2,345,014 models of compounds]
82989 - 83223 = human ADF2 site 2, N-terminal domain (from 2VAC; conf A) vs Enamine library
83224 - 83458 = human ADF2 site 2, C-terminal domain (from 2W0I.pdb; conf A) vs Enamine library
83459 - 83693 = Pf ADF1 site 2 (from 2XF1.pdb; conf A) vs Enamine library
83694 - 83928 = Pb ADF2 site 2 (from 2XFA.pdb) vs Enamine library
83929 - 84163 = Pf ADF1 site 2 (from 3Q2B.pdb; conf A) vs Enamine library
84164 - 84398 = Pf ADF1 site 2 (from 3Q2B.pdb; conf B) vs Enamine library

Experiment 109 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
84399 - 84500 = human ADF2 site 2, N-terminal domain (from 2VAC; conf A) vs ChemBridge library
84501 - 84602 = human ADF2 site 2, C-terminal domain (from 2W0I.pdb; conf A) vs ChemBridge library
84603 - 84704 = Pf ADF1 site 2 (from 2XF1.pdb; conf A) vs ChemBridge library
84705 - 84806 = Pb ADF2 site 2 (from 2XFA.pdb) vs ChemBridge library
84807 - 84908 = Pf ADF1 site 2 (from 3Q2B.pdb; conf A) vs ChemBridge library
84909 - 85010 = Pf ADF1 site 2 (from 3Q2B.pdb; conf B) vs ChemBridge library
 

Experiment 110 = 100% completed

[Asinex library = 507,000 models of compounds]
85011 - 85061 = human ADF2 site 2, N-terminal domain (from 2VAC; conf A) vs Asinex library
85062 - 85112 = human ADF2 site 2, C-terminal domain (from 2W0I.pdb; conf A) vs Asinex library
85113 - 85163 = Pf ADF1 site 2 (from 2XF1.pdb; conf A) vs Asinex library
85164 - 85214 = Pb ADF2 site 2 (from 2XFA.pdb) vs Asinex library
85215 - 85265 = Pf ADF1 site 2 (from 3Q2B.pdb; conf A) vs Asinex library
85266 - 85316 = Pf ADF1 site 2 (from 3Q2B.pdb; conf B) vs Asinex library
 

Experiment 111 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
85317 - 85467 = human ADF2 site 2, N-terminal domain (from 2VAC; conf A) vs Vitas-M Labs library
85468 - 85618 = human ADF2 site 2, C-terminal domain (from 2W0I.pdb; conf A) vs Vitas-M Labs library
85619 - 85769 = Pf ADF1 site 2 (from 2XF1.pdb; conf A) vs Vitas-M Labs library
85770 - 85920 = Pb ADF2 site 2 (from 2XFA.pdb) vs Vitas-M Labs library
85921 - 86071 = Pf ADF1 site 2 (from 3Q2B.pdb; conf A) vs Vitas-M Labs library
86072 - 86222 = Pf ADF1 site 2 (from 3Q2B.pdb; conf B) vs Vitas-M Labs library

 

 

 

 


 

 

Target #18 =  cyclophilin = Cyp

Cyclophilins, or Cyp's, are peptidyl-prolyl cis-trans isomerases, which means that these enzymes catalyze the conversion of the trans form of certain peptide bonds to the cis form.   This isomerization of the trans to the cis form of peptide bonds at proline residues helps proteins fold properly.  This family of proteins is called cyclophilins, because they bind to cyclosporine.  Cyclosporine is an immunosuppressant, and it is used to help prevent the rejection of transplanted organs.  When cyclosporine binds to cyclophilin, that complex then binds to and inhibits calcineurin (which is a calcium/calmodulin-dependent phosphatase; phosphatases are enzymes that remove phosphate groups from certain other proteins in order to regulate their activity or subcellular localization, while kinases are enzymes that add these phosphate groups to certain proteins in order to regulate their activity or preferred location).  Inhibiting calcineurin decreases the production of pro-inflammatory molecules such as TNF alpha and interleukin 2, which is thought to be the reason why cyclosporins are able to help prevent the rejection of transplanted organs. 

Cyclosporin A can also bind to the cyclophilin from Plasmodium falciparum.  In studies in mice and in test tubes or cell cultures, cyclosporin A has been shown to display potent anti-malarial properties (according to the reference for the 1QNG.pdb crystal structure).  In our experiments, we are searching for new, "small molecule" compounds that can inhibit the cyclophilin from malaria without causing the immunosuppressant effects that are caused by cyclosporine.  Although we want to find new ways to kill Plasmodium falciparum, we do not want the new drugs to suppress the immune system of patients who are infected with malaria parasites.  Similarly, the cyclophilin from Brugia malayi was included as a target, because we also want to find compounds that can inhibit its Cyp without causing any suppression of the human immune system.  Brugia malayi is a parasite that causes filariasis, which is a horribly painful and disfiguring disease that affects more than 120 million people in tropical and subtropical regions of Asia (especially in India), Africa, Central America, and South America.  Filariasis also affects some Pacific Island nations, such as Samoa, Tonga, the Cook Islands, Fiji, Papua New Guinea, and Vanuatu.  Elephantiasis is caused by filariasis.  In addition to trying to discover new compounds that can inhibit the activity of cyclophilin from the parasites that cause malaria or filariasis, we are also searching for new compounds that can inhibit the Cyp from Mycobacterium tuberculosis, which are the deadliest bacteria on Earth.  If we can find compounds that inhibit the Cyp's from all three of these pathogens simultaneously, it could be tremendously beneficial to the medical community.  Finding compounds that can inhibit any 1 of these 3 types of Cyp's should also be very helpful.

Human cyclophilin A also binds to the Gag polyprotein that is produced by HIV.  Catalyzing the conversion of the trans to the cis form of peptide bonds at proline residues in HIV's Gag polyprotein helps the HIV proteins fold properly.  Consequently, the interaction between human cyclophilin A and HIV's Gag is essential to the infectivity of HIV.  The human version of cyclophilin is included as a target in these experiments, because we want to find compounds that bind strongly to malaria's cyclophilin but that do not bind with high affinity to human cyclophilin (that is, we want to find inhibitors that are selective against malaria).  However, if we can find new inhibitors of human cyclophilin that do not also lead to the inhibition of calcineurin, then these compounds might help fight HIV infections (without also suppressing the immune system of the HIV-infected patients). 




Batch numbers for the experiments versus cyclophilin:

Experiment 112 = 100% completed

[full NCI library = 316,179 models of compounds]
86223 - 86254 = Brugia malayi Cyp (from 1A58.pdb; apo) vs NCI library
86255 - 86286 = human CypA (from 1AWQ.pdb; HAGPIA-induced) vs NCI library
86287 - 86318 = human CypA (from 1BCK.pdb; cyclosporin C-induced) vs NCI library
86319 - 86350 = Brugia malayi Cyp (from 1C5F.pdb; cyclosporin A-induced) vs NCI library
86351 - 86382 = human CypA (from 1CWJ.pdb; cyclosporin D-induced) vs NCI library
86383 - 86414 = human CypA (from 1CWK.pdb; cyclosporin D-induced) vs NCI library
86415 - 86446 = human CypA (from 1FGL.pdb; HIV-1 Gag fragment-induced) vs NCI library
86447 - 86478 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; No flips) vs NCI library
86479 - 86510 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; w/ flips) vs NCI library
86511 - 86542 = TB CypA (from 1W74.pdb; apo) vs NCI library
86543 - 86574 = human CypA (from 1YND; sanglifehrin A-induced) vs NCI library
86575 - 86606 = Pf Cyp (from 2FU0.pdb; apo) vs NCI library
86607 - 86638 = human CypA (from 3K0N; apo; conf. A; room-temperature x-tal) vs NCI library
86639 - 86670 = human CypA (from 3RDD; conf. A; EA4-induced) vs NCI library

 

Experiment 113 = 100% completed

[Enamine library = 2,345,014 models of compounds]
86671 - 86905 = Brugia malayi Cyp (from 1A58.pdb; apo) vs Enamine library
86906 - 87140 = human CypA (from 1AWQ.pdb; HAGPIA-induced) vs Enamine library
87141 - 87375 = human CypA (from 1BCK.pdb; cyclosporin C-induced) vs Enamine library
87376 - 87610 = Brugia malayi Cyp (from 1C5F.pdb; cyclosporin A-induced) vs Enamine library
87611 - 87845 = human CypA (from 1CWJ.pdb; cyclosporin D-induced) vs Enamine library
87846 - 88080 = human CypA (from 1CWK.pdb; cyclosporin D-induced) vs Enamine library
88081 - 88315 = human CypA (from 1FGL.pdb; HIV-1 Gag fragment-induced) vs Enamine library
88316 - 88550 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; No flips) vs Enamine library
88551 - 88785 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; w/ flips) vs Enamine library
88786 - 89020 = TB CypA (from 1W74.pdb; apo) vs Enamine library
89021 - 89255 = human CypA (from 1YND; sanglifehrin A-induced) vs Enamine library
89256 - 89490 = Pf Cyp (from 2FU0.pdb; apo) vs Enamine library
89491 - 89725 = human CypA (from 3K0N; apo; conf. A; room-temperature x-tal) vs Enamine library
89726 - 89960 = human CypA (from 3RDD; conf. A; EA4-induced) vs Enamine library

Experiment 114 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
89961 - 90062 = Brugia malayi Cyp (from 1A58.pdb; apo) vs ChemBridge library
90063 - 90164 = human CypA (from 1AWQ.pdb; HAGPIA-induced) vs ChemBridge library
90165 - 90266 = human CypA (from 1BCK.pdb; cyclosporin C-induced) vs ChemBridge library
90267 - 90368 = Brugia malayi Cyp (from 1C5F.pdb; cyclosporin A-induced) vs ChemBridge library
90369 - 90470 = human CypA (from 1CWJ.pdb; cyclosporin D-induced) vs ChemBridge library
90471 - 90572 = human CypA (from 1CWK.pdb; cyclosporin D-induced) vs ChemBridge library
90573 - 90674 = human CypA (from 1FGL.pdb; HIV-1 Gag fragment-induced) vs ChemBridge library
90675 - 90776 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; No flips) vs ChemBridge library
90777 - 90878 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; w/ flips) vs ChemBridge library
90879 - 90980 = TB CypA (from 1W74.pdb; apo) vs ChemBridge library
90981 - 91082 = human CypA (from 1YND; sanglifehrin A-induced) vs ChemBridge library
91083 - 91184 = Pf Cyp (from 2FU0.pdb; apo) vs ChemBridge library
91185 - 91286 = human CypA (from 3K0N; apo; conf. A; room-temperature x-tal) vs ChemBridge library
91287 - 91388 = human CypA (from 3RDD; conf. A; EA4-induced) vs ChemBridge library
 

Experiment 115 = 100% completed

[Asinex library = 507,000 models of compounds]
91389 - 91439 = Brugia malayi Cyp (from 1A58.pdb; apo) vs Asinex library
91440 - 91490 = human CypA (from 1AWQ.pdb; HAGPIA-induced) vs Asinex library
91491 - 91541 = human CypA (from 1BCK.pdb; cyclosporin C-induced) vs Asinex library
91542 - 91592 = Brugia malayi Cyp (from 1C5F.pdb; cyclosporin A-induced) vs Asinex library
91593 - 91643 = human CypA (from 1CWJ.pdb; cyclosporin D-induced) vs Asinex library
91644 - 91694 = human CypA (from 1CWK.pdb; cyclosporin D-induced) vs Asinex library
91695 - 91745 = human CypA (from 1FGL.pdb; HIV-1 Gag fragment-induced) vs Asinex library
91746 - 91796 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; No flips) vs Asinex library
91797 - 91847 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; w/ flips) vs Asinex library
91848 - 91898 = TB CypA (from 1W74.pdb; apo) vs Asinex library
91899 - 91949 = human CypA (from 1YND; sanglifehrin A-induced) vs Asinex library
91950 - 92000 = Pf Cyp (from 2FU0.pdb; apo) vs Asinex library
92001 - 92051 = human CypA (from 3K0N; apo; conf. A; room-temperature x-tal) vs Asinex library
92052 - 92102 = human CypA (from 3RDD; conf. A; EA4-induced) vs Asinex library
 

Experiment 116 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
92103 - 92253 = Brugia malayi Cyp (from 1A58.pdb; apo) vs Vitas-M Labs library
92254 - 92404 = human CypA (from 1AWQ.pdb; HAGPIA-induced) vs Vitas-M Labs library
92405 - 92555 = human CypA (from 1BCK.pdb; cyclosporin C-induced) vs Vitas-M Labs library
92556 - 92706 = Brugia malayi Cyp (from 1C5F.pdb; cyclosporin A-induced) vs Vitas-M Labs library
92707 - 92857 = human CypA (from 1CWJ.pdb; cyclosporin D-induced) vs Vitas-M Labs library
92858 - 93008 = human CypA (from 1CWK.pdb; cyclosporin D-induced) vs Vitas-M Labs library
93009 - 93159 = human CypA (from 1FGL.pdb; HIV-1 Gag fragment-induced) vs Vitas-M Labs library
93160 - 93310 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; No flips) vs Vitas-M Labs library
93311 - 93461 = Pf Cyp (from 1QNG.pdb; cyclosporin A-induced; w/ flips) vs Vitas-M Labs library
93462 - 93612 = TB CypA (from 1W74.pdb; apo) vs Vitas-M Labs library
93613 - 93763 = human CypA (from 1YND; sanglifehrin A-induced) vs Vitas-M Labs library
93764 - 93914 = Pf Cyp (from 2FU0.pdb; apo) vs Vitas-M Labs library
93915 - 94065 = human CypA (from 3K0N; apo; conf. A; room-temperature x-tal) vs Vitas-M Labs library
94066 - 94216 = human CypA (from 3RDD; conf. A; EA4-induced) vs Vitas-M Labs library

 

Notes:
HAGPIA = hexapeptide that is part of the HIV capsid protein

1BCK = crystallized with 2-Thr cyclosporin C bound to CypA

cyclosporins = large, flexible, macrocyclic inhibitors (that is, they are (modified) peptides that have their backbones connected together in head-to-tail fashion to form a quasi-circular ring); they are used clinically as potent immunosuppressants (to prevent the rejection of transplanted organs)

1CWJ = crystallized with 2-Val 3-S-methyl-sarcosine cyclosporin = cyclosporin D derivative

1CWK = crystallized with 1-(6,7-dihydro)MEBMT 2-Val 3-D-(2-S-methyl)sarcosine cyclosporin = cyclosporin D derivative

1FGL = crystallized with residues 81-105 from the HIV-1 Gag protein

1W74 = cycylophilin A from Mycobacterium tuberculosis, apo (that is, it was not crystallized with any inhibitors)

3K0N = X-ray crystallographic structure was determined at room temperature (most crystal structures are determined by collecting data at cryo-temperatures => very cold => less flexibility is displayed by the protein)

3RDD = crystallized with the small molecule inhibitor "EA4" bound = ethyl N-[(4-aminobenzyl)carbamoyl]glycinate

"No flips" = the original crystallographic conformations of Asn, Gln, and His were used in the model of the target
"w/ flips" = The other version of that target has the flipped conformations of some of these N, Q, and H residues, because the MolProbity server indicated that the flipped conformations were more stable

 

 

 

 


 

 

Target #19 =  apical membrane antigen 1 = AMA1

*see Science, 333:410 (July 22, 2011)

*several mAbs (monoclonal antibodies) that target Pf AMA1 significantly reduce parasite invasion of red blood cells


Although AMA1 mutates frequently, we are targeting a more highly-conserved region of AMA1 (that is, a region that does not mutate quite as much).  This conserved region is the location where the Pf RON2 peptide binds.  When Pf RON2 binds to Pf AMA1, it blocks the formation and/or function of the "moving junction".  During host cell invasion, the malaria parasite injects the RON complex into the human host cell's membrane.  RON2 is a transmembrane component of the RON complex, and RON2 interacts directly with Pf AMA1.  When apicocomplexan parasites, including Plasmodium falciparum, invade a cell, a "moving junction" (MJ) is formed and is critically involved in the process the parasite uses to infect that cell.  Adding free Pf RON2 peptide has been shown to strongly inhibit red blood cell invasion, by impeding the formation and/or function of a proper MJ.  Although AMA1 displays many polymorphisms, the strong inhibitory potency of the Pf RON2 peptide is not affected by these different sequence varations that AMA1 naturally displays.  In these GOFAM experiments against AMA1, we are targeting the specific region where Pf RON2 binds.  This Pf RON2 binding site is a hydrophobic groove and a region that becomes exposed when the flexible Domain II loop of AMA1 is displaced.  This site is also the same region where some of the antibodies (including "IgNAR") bind to AMA1.

 

 



These three images display the results of a positive control re-docking experiment against the AMA1 target from Plasmodium falciparum (from 3SRI.pdb).  The particular model of Pf AMA1 that was used in the re-docking calculation shown here included several of the water molecules that were present in the crystal structure.  Using the "wet version" of this target increased the accuracy of the positive control experiment.  The cyclic peptide called Pf RON2 was first pruned a bit (that is, a few of the very floppy residues at its ends were deleted from the models); consequently, only residues Ser2036 to Asn2051 were included in these re-docking calculations.  Even with this pruning, the model of Pf RON2 still had 37 different rotatable bonds.  Many docking programs start to lose their accuracy when the compound being docked has 12 or more rotatable bonds, but AutoDock Vina was surprisingly accurate in these positive control experiments.  The crystallographic conformation of Pf RON2 (that is, the "correct" answer) has gray carbon atoms, while the binding mode predicted by AD Vina calculations has scarlet carbon atoms.  The two images on the top display a close-up view, while the image on the bottom displays the full version of Pf AMA1 that was present in the crystal structure.  The image in the upper-right corner displays the surface representation of this target, while the other two images display the "cartoon mode" (or "ribbon mode") of the backbone of AMA1.

 

 

Batch numbers for the experiments versus apical membrane antigen1 :

Experiment 117 = 100% completed

[full NCI library = 316,179 models of compounds]
94217 - 94248 = Pf AMA1 (from 2Q8B.pdb; Ab-induced; loopAmb) vs NCI library
94249 - 94280 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced) vs NCI library
94281 - 94312 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced; w/ WATs) vs NCI library
94313 - 94344 = Pf AMA1 (from 2Z8W.pdb; IgNAR-induced) vs NCI library
94345 - 94376 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced) vs NCI library
94377 - 94408 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced; w/ WATs) vs NCI library
94409 - 94440 = Pf AMA1 (from 3SRI.pdb; conf. B; loopChrm2; PfRON2-induced) vs NCI library
94441 - 94472 = Pf AMA1 (from 3SRJ.pdb; loopChrm; R1-induced) vs NCI library
94473 - 94504 = Pf AMA1 (from 3ZWZ.pdb; PfRON2-induced) vs NCI library

 

Experiment 118 = 100% completed

[Enamine library = 2,345,014 models of compounds]
94505 - 94739 = Pf AMA1 (from 2Q8B.pdb; Ab-induced; loopAmb) vs Enamine library
94740 - 94974 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced) vs Enamine library
94975 - 95209 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced; w/ WATs) vs Enamine library
95210 - 95444 = Pf AMA1 (from 2Z8W.pdb; IgNAR-induced) vs Enamine library
95445 - 95679 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced) vs Enamine library
95680 - 95914 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced; w/ WATs) vs Enamine
95915 - 96149 = Pf AMA1 (from 3SRI.pdb; conf. B; loopChrm2; PfRON2-induced) vs Enamine library
96150 - 96384 = Pf AMA1 (from 3SRJ.pdb; loopChrm; R1-induced) vs Enamine library
96385 - 96619 = Pf AMA1 (from 3ZWZ.pdb; PfRON2-induced) vs Enamine library

 

Experiment 119 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
96620 - 96721 = Pf AMA1 (from 2Q8B.pdb; Ab-induced; loopAmb) vs ChemBridge library
96722 - 96823 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced) vs ChemBridge library
96824 - 96925 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced; w/ WATs) vs ChemBridge library
96926 - 97027 = Pf AMA1 (from 2Z8W.pdb; IgNAR-induced) vs ChemBridge library
97028 - 97129 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced) vs ChemBridge library
97130 - 97231 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced; w/ WATs) vs ChemBridge
97232 - 97333 = Pf AMA1 (from 3SRI.pdb; conf. B; loopChrm2; PfRON2-induced) vs ChemBridge library
97334 - 97435 = Pf AMA1 (from 3SRJ.pdb; loopChrm; R1-induced) vs ChemBridge library
97436 - 97537 = Pf AMA1 (from 3ZWZ.pdb; PfRON2-induced) vs ChemBridge library
 

Experiment 120 = 100% completed

[Asinex library = 507,000 models of compounds]
97538 - 97588 = Pf AMA1 (from 2Q8B.pdb; Ab-induced; loopAmb) vs Asinex library
97589 - 97639 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced) vs Asinex library
97640 - 97690 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced; w/ WATs) vs Asinex library
97691 - 97741 = Pf AMA1 (from 2Z8W.pdb; IgNAR-induced) vs Asinex library
97742 - 97792 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced) vs Asinex library
97793 - 97843 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced; w/ WATs) vs Asinex
97844 - 97894 = Pf AMA1 (from 3SRI.pdb; conf. B; loopChrm2; PfRON2-induced) vs Asinex library
97895 - 97945 = Pf AMA1 (from 3SRJ.pdb; loopChrm; R1-induced) vs Asinex library
97946 - 97996 = Pf AMA1 (from 3ZWZ.pdb; PfRON2-induced) vs Asinex library
 

Experiment 121 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
97997 - 98147 = Pf AMA1 (from 2Q8B.pdb; Ab-induced; loopAmb) vs Vitas-M Labs library
98148 - 98298 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced) vs Vitas-M Labs library
98299 - 98449 = Pf AMA1 (from 2Z8V.pdb; IgNAR-induced; w/ WATs) vs Vitas-M Labs library
98450 - 98600 = Pf AMA1 (from 2Z8W.pdb; IgNAR-induced) vs Vitas-M Labs library
98601 - 98751 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced) vs Vitas-M Labs library
98752 - 98902 = Pf AMA1 (from 3SRI.pdb; conf. A; loopChrm0; PfRON2-induced; w/ WATs) vs Vitas-M Labs
98903 - 99053 = Pf AMA1 (from 3SRI.pdb; conf. B; loopChrm2; PfRON2-induced) vs Vitas-M Labs library
99054 - 99204 = Pf AMA1 (from 3SRJ.pdb; loopChrm; R1-induced) vs Vitas-M Labs library
99205 - 99355 = Pf AMA1 (from 3ZWZ.pdb; PfRON2-induced) vs Vitas-M Labs library

 

Notes:
IgNAR = single-variable-domain of an Ab that has been shown to inhibit erythrocyte invasion

PfRON2 = Pf Rhoptry Neck Protein 2; RON = the receptor for AMA1.

R1 peptide = a peptide that has been shown to inhibit the parasite's invasion process.  It binds in
the same region as PfRON2.

 

 

 

 

 


 

 

Target #20 =  mitogen-activated protein kinase kinase = MAP2K

(also known as MEK6)

According to the research manuscript "Global kinomic and phospho-proteomic analyses of the human malaria parasite Plasmodium falciparum," by Lev Solyakov, Jean Halbert, Mahmood M. Alam, Christian Doerig, et al., in Nature Communications, vol. 2, pages 565 - 576, (2011), and the review article "Plasmodium kinases as targets for new-generation antimalarials," by Isabelle S. Lucet, Andrew Tobin, David Drewry, Andrew F. Wilks, and Christian Doerig, in Future Medicinal Chemistry, vol. 4, number 18, pages 2295-2310, (2012), there are several different kinase enzymes that have been proven to be essential to the life-cycle of the malaria parasite.  Of these essential kinases (that is, enzymes that add a phosphate group to other proteins, in order to regulate their activity and/or subcellular location), only a few have had their atomically-detailed, 3-D structures determined and deposited in the databases that are accessible to the public.  One of these kinds of kinases that are known to be essential to the life-cycle of Plasmodium falciparum is called "mitogen-activated protein kinase kinase," or "MAP2K".  This enzyme adds a phosphate group to the mitogen-activated protein kinase (hence, it is a kinase that regulates another kinase).  Pf MAP2K is thus a potential drug target for curing malaria infections.

Two other kinds of kinases that are known to be essential to the life-cycle of malaria parasites are:  Pf Lammer/CLK1 and Pf PK5.  These other two kinds of kinases will be the 21st and 22nd classes of targets for GO FightAgainstMalaria.

 

 

These two images display the results of positive control re-docking studies against the MAP2K enzyme from Plasmodium falciparum (3NIE.pdb).  The inhibitor, a non-hydrolyzable ATP analog (i.e., it is a structural mimic of ATP that can't be easily chopped apart), is called ANP and is shown as sticks.  The crystallographic conformation (i.e., the "correct" binding mode) has cyan carbon atoms, while the docked mode predicted by AutoDock Vina has magenta carbon atoms.  The image on the left is a close-up view, while the image on the right shows the full structure of the target.

Batch numbers for the experiments versus MAP2K :

Experiment 122 = 100% completed

[full NCI library = 316,179 models of compounds]
99356 - 99387 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Amber2) vs NCI library
99388 - 99419 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Charmm1) vs NCI library
99420 - 99451 = human MAP2K (from 3FME.pdb; STU-induced; loops_Amber1and2) vs NCI library
99452 - 99483 = human MAP2K (from 3FME.pdb; STU-induced; loops_Charmm2and1) vs NCI library
99484 - 99515 = human MAP2K (from 3FME.pdb; STU-induced; No missing loops were added) vs NCI
99516 - 99547 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1) vs NCI library
99548 - 99579 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1; w/ 2 x-tal WATs) vs NCI library
99580 - 99611 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2) vs NCI library
99612 - 99643 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2; w/ 2 x-tal WATs) vs NCI library
99644 - 99675 = Pf MAP2K (from 3NIE.pdb; ANP-induced; No missing loops were added) vs NCI library

 

Experiment 123 = 100% completed

[Enamine library = 2,345,014 models of compounds]
 99676  -  99910  = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Amber2) vs Enamine library
 99911 -  100145 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Charmm1) vs Enamine library
100146 - 100380 = human MAP2K (from 3FME.pdb; STU-induced; loops_Amber1and2) vs Enamine
100381 - 100615 = human MAP2K (from 3FME.pdb; STU-induced; loops_Charmm2and1) vs Enamine
100616 - 100850 = human MAP2K (from 3FME.pdb; STU-induced; No missing loops were added) vs Enamine
100851 - 101085 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1) vs Enamine library
101086 - 101320 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1; w/ 2 x-tal WATs) vs Enamine
101321 - 101555 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2) vs Enamine library
101556 - 101790 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2; w/ 2 x-tal WATs) vs Enamine
101791 - 102025 = Pf MAP2K (from 3NIE.pdb; ANP-induced; No missing loops were added) vs Enamine

 

Experiment 124 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
102026 - 102127 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Amber2) vs ChemBridge library
102128 - 102229 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Charmm1) vs ChemBridge
102230 - 102331 = human MAP2K (from 3FME.pdb; STU-induced; loops_Amber1and2) vs ChemBridge
102332 - 102433 = human MAP2K (from 3FME.pdb; STU-induced; loops_Charmm2and1) vs ChemBridge
102434 - 102535 = human MAP2K (from 3FME.pdb; STU-induced; No missing loops were added) vs ChemBridge
102536 - 102637 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1) vs ChemBridge library
102638 - 102739 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1; w/ 2 x-tal WATs) vs ChemBridge
102740 - 102841 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2) vs ChemBridge library
102842 - 102943 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2; w/ 2 x-tal WATs) vs ChemBridge
102944 - 103045 = Pf MAP2K (from 3NIE.pdb; ANP-induced; No missing loops were added) vs ChemBridge
 

Experiment 125 = 100% completed

[Asinex library = 507,000 models of compounds]
103046 - 103096 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Amber2) vs Asinex library
103097 - 103147 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Charmm1) vs Asinex library
103148 - 103198 = human MAP2K (from 3FME.pdb; STU-induced; loops_Amber1and2) vs Asinex library
103199 - 103249 = human MAP2K (from 3FME.pdb; STU-induced; loops_Charmm2and1) vs Asinex
103250 - 103300 = human MAP2K (from 3FME.pdb; STU-induced; No missing loops were added) vs Asinex
103301 - 103351 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1) vs Asinex library
103352 - 103402 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1; w/ 2 x-tal WATs) vs Asinex
103403 - 103453 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2) vs Asinex library
103454 - 103504 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2; w/ 2 x-tal WATs) vs Asinex
103505 - 103555 = Pf MAP2K (from 3NIE.pdb; ANP-induced; No missing loops were added) vs Asinex
 

Experiment 126 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
103556 - 103706 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Amber2) vs Vitas-M Labs library
103707 - 103857 = human MAP2K (from 3ENM.pdb; apo; chain A; loop_Charmm1) vs Vitas-M Labs
103858 - 104008 = human MAP2K (from 3FME.pdb; STU-induced; loops_Amber1and2) vs Vitas-M Labs
104009 - 104159 = human MAP2K (from 3FME.pdb; STU-induced; loops_Charmm2and1) vs Vitas-M Labs
104160 - 104310 = human MAP2K (from 3FME.pdb; STU-induced; No missing loops were added) vs Vitas-M Labs
104311 - 104461 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1) vs Vitas-M Labs library
104462 - 104612 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber1; w/ 2 x-tal WATs) vs Vitas-M Labs
104613 - 104763 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2) vs Vitas-M Labs library
104764 - 104914 = Pf MAP2K (from 3NIE.pdb; ANP-induced; loop_Amber2; w/ 2 x-tal WATs) vs Vitas-M Labs
104915 - 105065 = Pf MAP2K (from 3NIE.pdb; ANP-induced; No missing loops were added) vs Vitas-M Labs

 

Notes:
STU = staurosporine, a kinase inhibitor with 3.4 nM Kd against human MAP2K

ANP = phosphoaminophosphonic acid-adenylate ester = an ATP analog that is not easily hydrolyzed (it can thus bind to the kinase in a stable manner, without being split apart by the reactions that the kinase catalyzes).  Non-hydrolyzable ATP analogs are often used in X-ray crystallographic studies, in order to develop a reasonable model of the ATP-bound conformation.
 

 

 

 

 

 


 

 

Target #21 =  "protein kinase 5" = PK5

The enzyme "PK5" (for "protein kinase 5") from Plasmodium falciparum is the most well-characterized member of the parasite's "CDK" family of proteins.  "CDK" stands for "cyclin-dependent kinase".  For more information about what a kinase is, see the description for target class #18.  Similar to the MAP2K enzyme (target class #20), Pf PK5 is another kinase that has been proven to be essential to the life-cycle of the malaria parasite (according to the same manuscripts cited in the description for target class #20).  Pf PK5 is a promising "potential drug target" (that is, it has not yet been proven to be a "valid drug target"). 

 

The kinase from humans that has the highest % sequence identity and the closest structural similarity to Pf PK5 is called "CDK5".  We want to find compounds that are potent inhibitors of Pf PK5 but that do not strongly inhibit human CDK5 (that is, we want inhibitors that are specific for the malarial kinase).

 

 

These two images display the results of positive control re-docking experiments against PK5 from Plasmodium falciparum (from 1V0O.pdb).  The image on the left depicts a close-up view of the binding site, with the cartoon mode of the backbone of PK5 shown as light green tubes, while the image on the right shows the full view of the target.  The X-ray crystallographic conformation of the inhibitor INR is displayed as sticks with dark green carbon atoms, while the binding mode predicted by Vina has orange carbon atoms.

 

Batch numbers for the experiments versus Pf PK5 and human CDK5 :

Experiment 127 = 100% completed

[full NCI library = 316,179 models of compounds]
105066 - 105097 = Pf PK5 (from 1OB3.pdb; apo; w/ 4 x-tal WATs) vs NCI library
105098 - 105129 = Pf PK5 (from 1OB3.pdb; apo) vs NCI library
105130 - 105161 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced; w/ 4 x-tal WATs) vs NCI library
105162 - 105193 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced) vs NCI library
105194 - 105225 = Pf PK5 (from 1V0B.pdb; T198A mut; apo; w/ 4 x-tal WATs) vs NCI library
105226 - 105257 = Pf PK5 (from 1V0B.pdb; T198A mut; apo) vs NCI library
105258 - 105289 = Pf PK5 (from 1V0O.pdb; INR-induced; w/ 4 x-tal WATs) vs NCI library
105290 - 105321 = Pf PK5 (from 1V0O.pdb; INR-induced) vs NCI library
105322 - 105353 = Pf PK5 (from 1V0P.pdb; Purv B-induced; w/ 5 x-tal WATs) vs NCI library
105354 - 105385 = Pf PK5 (from 1V0P.pdb; Purv B-induced) vs NCI library
105386 - 105417 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced; w/ 6 x-tal WATs) vs NCI library
105418 - 105449 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced) vs NCI library

 

Experiment 128 = 100% completed

[Enamine library = 2,345,014 models of compounds]
105450 - 105684 = Pf PK5 (from 1OB3.pdb; apo; w/ 4 x-tal WATs) vs Enamine library
105685 - 105919 = Pf PK5 (from 1OB3.pdb; apo) vs Enamine library
105920 - 106154 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced; w/ 4 x-tal WATs) vs Enamine
106155 - 106389 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced) vs Enamine library
106390 - 106624 = Pf PK5 (from 1V0B.pdb; T198A mut; apo; w/ 4 x-tal WATs) vs Enamine library
106625 - 106859 = Pf PK5 (from 1V0B.pdb; T198A mut; apo) vs Enamine library
106860 - 107094 = Pf PK5 (from 1V0O.pdb; INR-induced; w/ 4 x-tal WATs) vs Enamine library
107095 - 107329 = Pf PK5 (from 1V0O.pdb; INR-induced) vs Enamine library
107330 - 107564 = Pf PK5 (from 1V0P.pdb; Purv B-induced; w/ 5 x-tal WATs) vs Enamine library
107565 - 107799 = Pf PK5 (from 1V0P.pdb; Purv B-induced) vs Enamine library
107800 - 108034 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced; w/ 6 x-tal WATs) vs Enamine
108035 - 108269 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced) vs Enamine library

 

Experiment 129 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
108270 - 108371 = Pf PK5 (from 1OB3.pdb; apo; w/ 4 x-tal WATs) vs ChemBridge library
108372 - 108473 = Pf PK5 (from 1OB3.pdb; apo) vs ChemBridge library
108474 - 108575 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced; w/ 4 x-tal WATs) vs ChemBridge
108576 - 108677 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced) vs ChemBridge library
108678 - 108779 = Pf PK5 (from 1V0B.pdb; T198A mut; apo; w/ 4 x-tal WATs) vs ChemBridge library
108780 - 108881 = Pf PK5 (from 1V0B.pdb; T198A mut; apo) vs ChemBridge library
108882 - 108983 = Pf PK5 (from 1V0O.pdb; INR-induced; w/ 4 x-tal WATs) vs ChemBridge library
108984 - 109085 = Pf PK5 (from 1V0O.pdb; INR-induced) vs ChemBridge library
109086 - 109187 = Pf PK5 (from 1V0P.pdb; Purv B-induced; w/ 5 x-tal WATs) vs ChemBridge library
109188 - 109289 = Pf PK5 (from 1V0P.pdb; Purv B-induced) vs ChemBridge library
109290 - 109391 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced; w/ 6 x-tal WATs) vs ChemBridge
109392 - 109493 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced) vs ChemBridge library
 

Experiment 130 = 100% completed

[Asinex library = 507,000 models of compounds]
109494 - 109544 = Pf PK5 (from 1OB3.pdb; apo; w/ 4 x-tal WATs) vs Asinex library
109545 - 109595 = Pf PK5 (from 1OB3.pdb; apo) vs Asinex library
109596 - 109646 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced; w/ 4 x-tal WATs) vs Asinex
109647 - 109697 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced) vs Asinex library
109698 - 109748 = Pf PK5 (from 1V0B.pdb; T198A mut; apo; w/ 4 x-tal WATs) vs Asinex library
109749 - 109799 = Pf PK5 (from 1V0B.pdb; T198A mut; apo) vs Asinex library
109800 - 109850 = Pf PK5 (from 1V0O.pdb; INR-induced; w/ 4 x-tal WATs) vs Asinex library
109851 - 109901 = Pf PK5 (from 1V0O.pdb; INR-induced) vs Asinex library
109902 - 109952 = Pf PK5 (from 1V0P.pdb; Purv B-induced; w/ 5 x-tal WATs) vs Asinex library
109953 - 110003 = Pf PK5 (from 1V0P.pdb; Purv B-induced) vs Asinex library
110004 - 110054 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced; w/ 6 x-tal WATs) vs Asinex
110055 - 110105 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced) vs Asinex library
 

Experiment 131 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
110106 - 110256 = Pf PK5 (from 1OB3.pdb; apo; w/ 4 x-tal WATs) vs Vitas-M Labs library
110257 - 110407 = Pf PK5 (from 1OB3.pdb; apo) vs Vitas-M Labs library
110408 - 110558 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced; w/ 4 x-tal WATs) vs Vitas-M Labs
110559 - 110709 = human CDK5 (from 1UNL.pdb; P25 and RRC-induced) vs Vitas-M Labs library
110710 - 110860 = Pf PK5 (from 1V0B.pdb; T198A mut; apo; w/ 4 x-tal WATs) vs Vitas-M Labs library
110861 - 111011 = Pf PK5 (from 1V0B.pdb; T198A mut; apo) vs Vitas-M Labs library
111012 - 111162 = Pf PK5 (from 1V0O.pdb; INR-induced; w/ 4 x-tal WATs) vs Vitas-M Labs library
111163 - 111313 = Pf PK5 (from 1V0O.pdb; INR-induced) vs Vitas-M Labs library
111314 - 111464 = Pf PK5 (from 1V0P.pdb; Purv B-induced; w/ 5 x-tal WATs) vs Vitas-M Labs library
111465 - 111615 = Pf PK5 (from 1V0P.pdb; Purv B-induced) vs Vitas-M Labs library
111616 - 111766 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced; w/ 6 x-tal WATs) vs Vitas-M Labs
111767 - 111917 = human CDK5 (from 4AU8.pdb; chain B; Z3R-induced) vs Vitas-M Labs library

 

 

Notes:  "w/ 4 x-tal WATs" = 4 crystallographic water molecules were included as part of the model of the target.  In some positive control re-docking calculations, the presence of these water molecules improved accuracy.

Purv B = purvalanol B = a "small molecule" inhibitor of Pf PK5 that has an IC50 of ~ 130 nM

Unless stated otherwise above, "chain A" of each target molecule's crystal structure was used.

 

These two images show the results of positive control re-docking experiments against human CDK5 (from 4AU8.pdb).  The image on the left is a close-up view, while the image on the right shows the full target.  The cartoon mode of the backbone of CDK5 is depicted in both images.  The crystallographic conformation of the inhibitor Z3R is shown as sticks with dark blue carbon atoms, while the binding mode predicted by Vina calculations has light purple carbon atoms.  In these experiments, we want to discover new compounds that can bind strongly to Pf PK5 and block its activity, but those candidate compounds should not bind strongly to human CDK5.

 


 

 

Target #22 =  "Pf Lammer"

Similar to the MAP2K enzyme (target class #20) and Pf PK5 (target class #21), Pf Lammer is another kinase that has been proven to be essential to the life-cycle of the malaria parasite (according to the same manuscripts cited in the description for target class #20).  Pf Lammer is a promising "potential drug target" (that is, it has not yet been proven to be a "valid drug target"). 

 

The kinase from humans that has the highest % sequence identity and the closest structural similarity to Pf Lammer is called "CLK1".  We want to find compounds that are potent inhibitors of Pf Lammer but that do not strongly inhibit human CLK1 (that is, we want inhibitors that are specific for the malarial kinase).

 

 

These images represent the results of positive control re-docking experiments against human CLK1 (from 1Z57.pdb).  The cartoon mode of the backbone of human CLK1 is shown in both images, with a close-up view on the left and the full view on the right.  The experimentally-determined, X-ray crystallographic conformation of the inhibitor DBQ is shown as sticks with black carbon atoms, while the binding mode predicted by Vina has yellow carbon atoms.  Like the other GO FAM experiments, we want to discover new compounds that can bind strongly to the malarial version of this enzyme and block its activity, but we do not want compounds that would also block the activity of the human version of this kinase (since doing so might cause bad side effects).

 

Batch numbers for the experiments versus Pf Lammer and human CLK1 :

Experiment 132 = 100% completed

[full NCI library = 316,179 models of compounds]
111918 - 111949 = human CLK1 (from 1Z57.pdb; DBQ-induced) vs NCI library
111950 - 111981 = human CLK1 (from 1Z57.pdb; DBQ-induced; w 7 x-tal WATs) vs NCI library
111982 - 112013 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced) vs NCI library
112014 - 112045 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced; w/ 7 x-tal WATs) vs NCI library
112046 - 112077 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced) vs NCI library
112078 - 112109 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced; w/ 7 x-tal WATs) vs NCI library
112110 - 112141 = Pf Lammer (from 3LLT.pdb; ANP-induced) vs NCI library
112142 - 112173 = Pf Lammer (from 3LLT.pdb; ANP-induced; w/ 5 x-tal WATs) vs NCI library

 

Experiment 133 = 100% completed

[Enamine library = 2,345,014 models of compounds]
112174 - 112408 = human CLK1 (from 1Z57.pdb; DBQ-induced) vs Enamine library
112409 - 112643 = human CLK1 (from 1Z57.pdb; DBQ-induced; w 7 x-tal WATs) vs Enamine library
112644 - 112878 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced) vs Enamine library
112879 - 113113 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced; w/ 7 x-tal WATs) vs Enamine
113114 - 113348 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced) vs Enamine library
113349 - 113583 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced; w/ 7 x-tal WATs) vs Enamine
113584 - 113818 = Pf Lammer (from 3LLT.pdb; ANP-induced) vs Enamine library
113819 - 114053 = Pf Lammer (from 3LLT.pdb; ANP-induced; w/ 5 x-tal WATs) vs Enamine library

 

Experiment 134 = 100% completed

[ChemBridge library = 1,013,483 models of compounds]
114054 - 114155 = human CLK1 (from 1Z57.pdb; DBQ-induced) vs ChemBridge library
114156 - 114257 = human CLK1 (from 1Z57.pdb; DBQ-induced; w 7 x-tal WATs) vs ChemBridge library
114258 - 114359 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced) vs ChemBridge library
114360 - 114461 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced; w/ 7 x-tal WATs) vs ChemBridge
114462 - 114563 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced) vs ChemBridge library
114564 - 114665 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced; w/ 7 x-tal WATs) vs ChemBridge
114666 - 114767 = Pf Lammer (from 3LLT.pdb; ANP-induced) vs ChemBridge library
114768 - 114869 = Pf Lammer (from 3LLT.pdb; ANP-induced; w/ 5 x-tal WATs) vs ChemBridge library

Experiment 135 = 100% completed

[Asinex library = 507,000 models of compounds]
114870 - 114920 = human CLK1 (from 1Z57.pdb; DBQ-induced) vs Asinex library
114921 - 114971 = human CLK1 (from 1Z57.pdb; DBQ-induced; w 7 x-tal WATs) vs Asinex library
114972 - 115022 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced) vs Asinex library
115023 - 115073 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced; w/ 7 x-tal WATs) vs Asinex
115074 - 115124 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced) vs Asinex library
115125 - 115175 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced; w/ 7 x-tal WATs) vs Asinex
115176 - 115226 = Pf Lammer (from 3LLT.pdb; ANP-induced) vs Asinex library
115227 - 115277 = Pf Lammer (from 3LLT.pdb; ANP-induced; w/ 5 x-tal WATs) vs Asinex library
 

Experiment 136 = 100% completed

[Vitas-M Labs library = 1,503,273 models of compounds]
115278 - 115428 = human CLK1 (from 1Z57.pdb; DBQ-induced) vs Vitas-M Labs library
115429 - 115579 = human CLK1 (from 1Z57.pdb; DBQ-induced; w 7 x-tal WATs) vs Vitas-M Labs library
115580 - 115730 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced) vs Vitas-M Labs library
115731 - 115881 = human CLK1 (from 2VAG.pdb; conf. A; V25-induced; w/ 7 x-tal WATs) vs Vitas-M Labs
115882 - 116032 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced) vs Vitas-M Labs library
116033 - 116183 = human CLK1 (from 2VAG.pdb; conf. B; V25-induced; w/ 7 x-tal WATs) vs Vitas-M Labs
116184 - 116334 = Pf Lammer (from 3LLT.pdb; ANP-induced) vs Vitas-M Labs library
116335 - 116485 = Pf Lammer (from 3LLT.pdb; ANP-induced; w/ 5 x-tal WATs) vs Vitas-M Labs library

 

 

 

Notes:  w/ N x-tal WATs = N crystallographic water molecules were included as part of the model of the target.  In some positive control re-docking calculations, the presence of these water molecules improved accuracy.

conf. A vs. conf. B = this crystal structure of the target had some residues that displayed two different conformations, and these alternate conformations were in the ligand's binding site that is being targeted in these virtual screening calculations.  One model of the target includes the "A" conformation, while the other model of the target contains the "B" conformation for those particular residues.


DBQ = the small molecule inhibitor "debromohymenialdisine," which has specificity for CLK1 over CLK3 (that is, it inhibits the human CLK1 version of the kinase enzyme much more strongly than it inhibits the closely-related CLK3 enzyme).  It has mid-nanoMolar activity vs CLK1 (that is, at 100 nanoMolar concentration, the CLK1 enzyme has 34.1% of the activity it normally displays without an inhibitor being present).

V25 = the potent small molecule inhibitor "ethyl 3-[(E)-2-amino-1-cyanoethenyl]-6,7- dichloro-1-methyl-1H-indole-2-carboxylate".  It has an IC50 value of 19.7 nanoMolar (that is, when that compound is present at ~ 20 nanoMolar concentration, it inhibits the activity of the human CLK1 enzyme by 50%).

ANP = non-hydrolyzable ATP analog called "phosphoaminophosphonic acid-adenylate ester".  Non-hydrolyzable means that the compound cannot easily be broken apart by the kinase enzyme.  Non-splittable ATP analogs like this are often used to try to obtain/trap the conformation of the target that would normally be induced by ATP.


human CLK1 = "cdc-2 like kinase" = cell-division control protein 2 = a kinase enzyme (that is, an enzyme that adds phosphate groups to itself and to other proteins, in order to regulate their sub-cellular localization and/or activity).  CLK1 is a kinase with dual specificity; it adds phosphate groups to serine or threonine residues of its substrates (and it can also add a phosphate group to one of its own tyrosine residues, in order to regulate its own activity).  Human CLK1 is the kinase enzyme with the highest structural similarity to the malarial kinase called "Pf Lammer".

Lammer = the "Lammer" family of enzymes contain a unique and conserved signature motif, which has the sequence "EHLAMMERILG" (using the 1-letter abbreviations for the amino acids).  For the malarial version of Lammer, it actually has LAMMES instead of LAMMER in that signature motif (but it's still called "Lammer").

 

 

These images show the results of positive control re-docking experiments against Pf Lammer (from 3LLT.pdb).  The image on the left shows a close-up view of the binding site, with the backbone of the kinase shown in cartoon mode, while the image on the right shows the full view of the target.  The small, V-shaped, red and white "ball and stick" molecules represent crystallographic waters that were included as part of the model of this target in some of these experiments.  The experimentally-determined, X-ray crystallographic conformation of ANP is shown as sticks with light purple carbon atoms, while the binding mode predicted by docking calculations with Vina is depicted with green carbon atoms.

 



 

 

The future of GO FAM

After the Vina calculations against target classes #21 and #22 are finished, the GFAM project will go on a long pause (for probably a couple years).  During the pause, we'll continue to process, measure, and analyze the data that we have already generated (especially the screens that involve the NCI library of compounds, since we can order those compounds for free from the NIH).  And we'll continue extending the collaborations that we already began against PfSUB1 (to try to validate it as a drug target and potentially help cure malaria infections), InhA from Mycobacterium tuberculosis (to advance the treatment of extremely-drug resistant tuberculosis), and human CD81 (to try to prevent malaria infections from spreading).  We will also try to initiate new collaborations against other classes of targets from GFAM.  Hopefully, this will also allow us to generate the results for a few great research papers. After we get a paper or two published on the GFAM data, then we will try to obtain grants to fund these projects. (We need to get some papers published in these areas to more firmly establish our credibility in the anti-malaria field and the anti-tuberculosis field, according to the NIH's general perspective.) When we get some grant funding for the GFAM project (to help pay for personnel costs, purchasing the commercially-available compounds, and performing the assays of the candidate compounds from GFAM), then we will end the long pause and resume phase 2 of GFAM (which will involve AutoDock calculations of compounds that performed well with Vina and perhaps some new Vina-based virtual screens of new libraries of compounds and/or new targets).

GFAM is still an unfunded project. We need to get some grant funding in order to extend these lines of research.  If you want to help increase the chances that we will be able to get some grant funding, then please contact your Senators and members of the House of Representatives (via e-mail, Twitter, FaceBook, and the telephone--using multiple forms of media is much more effective).  Tell them that you are strongly opposed to the sequester-related cuts to the NIH budget. Funding for the NIH needs to be increased, not cut in a drastic and arbitrary manner.  Let them know that this issue will affect how you vote--in both the primary elections and the general elections.

Thank you very much for your interest and your continued support,
Alex L. Perryman, Ph.D.

 

 

 

 

Note:  We shared our input files with the people who run the ZINC server at UCSF:  all of the "pdbqt" docking input files for the compounds in all of the libraries that we are using in these GO Fight Against Malaria experiments are available to the public, for free, at: 

http://zinc.docking.org/pdbqt

These "pdbqt" files for all of the libraries of "small molecule" compounds we screen on GO Fight Against Malaria and FightAIDS@Home were prepared by Stefano Forli, Ph.D.,  using the tool he created called "Raccoon."

 

% Completion Status, as of July 10, 2013

The images and text were updated on October 9, 2013