Chemical Biology 1 – Pharmacology 10-17-14. Methods for studying protein function – Loss of Function 1. Gene knockouts 2. Conditional knockouts 3. RNAi.

Chemical Biology 1 –Pharmacology

10-17-14

Methods for studying protein function – Loss of Function

• 1. Gene knockouts

• 2. Conditional knockouts

• 3. RNAi

• 4. Pharmacology (use of small molecules to turn off protein function)

pre-translational

Pharmacology

• Disadvantage– Unlike genetic methods it is difficult to identify

ligands that are highly selective for a target.

• Advantages1. Fast time scale

2. Only perturbs targeted sub-domains

3. graded dose response - tunability

4. Most drugs are small molecules

Weiss WA, Taylor SS, Shokat KM. “Recognizing and exploiting differences between RNAi and small-molecule inhibitors.” Nat Chem Biol. 2007 Dec;3(12):739-44.

Time Scale and Specificity

Small molecules are subdomain specific

Example: PAK1 Kinase

Small molecules affect only one domain, while pre-translational methodsremove the entire protein from the cell.

Tunability

Allows the amount of inhibition/activity that is necessary

Reverse Chemical Genetics (Pharmacology)

1. Identify a protein target of interest– Develop an activity assay (enzymes) or a binding

assay (protein-ligand interactions) to screen compounds

3. Optimize your initial lead compound by making analogs (SAR) and by using any additional biochemical/structural information. In parallel, screen optimized analogs against other targets (selectivity)

2. Test biased or unbiased panels of compounds against protein target of interest

Major challenges

• Druggability– Many proteins do not appear to make

favorable interactions with drug-like small molecules

Molecular Weight <900 Da

Kd < 1 M (∆G < -8.4 kcal/mol)

No more than one or two fixed charges

– Estimated that only ~10% of all proteins are druggable

Hopkins and Groom, Nat Reviews Drug Disc, 2002

Major challenges

• Selectivity– Finding selective agonists and antagonists is

very challenging– Knowing which other proteins to

counterscreen is difficult (easier for mechanism-based or enzyme family-directed ligands)

In some cases, chemistry and genetics can be used to circumvent these problems.

Knight ZA, Shokat KM. “Chemical genetics: where genetics and pharmacology meet. Cell. 2007 Feb 9;128(3):425-30.”

Koh JT. “Engineering selectivity and discrimination into ligand-receptor interfaces.”Chem Biol. 2002 Jan;9(1):17-23. Review.

Identification of small molecule inhibitors

2 classes– 1. Enzyme Inhibitors

• Many effective strategies for identifying enzyme inhibitors.

– 2. Protein-Protein Interaction Inhibitors• Difficult to identify potent inhibitors of protein-

protein interactions.

Methods for discovering enzyme inhibitors

• High throughput screening (parallel synthesis and combinatorial chemistry)

• Mechanism-based (incorporate a functionality that is unique for an enzyme enzyme class (For example, proteases)

• Privileged scaffolds (kinases, phosphodiesterases)

• Transition state analogs

Turk B.Targeting proteases: successes, failures and future prospects.Nat Rev Drug Discov. 2006 Sep;5(9):785-99.

Aspartyl Protease Inhibitors

HIV Protease InhibitorsINHIBITORS OF HIV-1 PROTEASE: A “Major Success of Structure-Assisted Drug Design” Alexander Wlodawer, Jiri Vondrasek. Annual Review of Biophysics and Biomolecular Structure. Volume 27, Page 249-284, 1998

HIV Protease Inhibitors

HIV Protease Inhibitors Resistance

More protease inhibitors

• Ketones (serine and cysteine proteases)

• Phosphonic and hydroxamic acids (metalloproteases)

transition state analog

hydroxamic acidschelatethe active sitezinc

Protein Kinases

The human genome encodes538 Protein kinases (483 arecatalytically active)

Kinase Inhibitors

Almost all inhibitors that have been developed bind in the ATP pocket

Synthesis of kinase inhibitors

OlomucineCdc2/CyB: 1µMCdk2/CyA: 1µM

LibrarySynthesis

10,000

LibraryScreening

10,000

HitCdc2/CyB: 340 pMCdk2/CyA: 340 pM

Gray et. al. Science (1998) 281, 533-538.

Approved kinase inhibitors

28 small molecule kinase inhibitors are now in the clinicGleevec (Imatinib) was the first clinically approved kinase inhibitor (2003)

Protein-Protein Interaction (PPI) Inhibitors

• Identification of potent PPI inhibitors is very challenging. In general, standard screening strategies don’t work.

• Conversion of Peptides/Proteins to Small Molecules

• Innovative new strategies are needed– for example, SAR by NMR

“SAR” by NMR Abbott Laboratories (Stephen Fesik)

Fragment based approach- library of small compounds (several thousands)- build up larger ligands- n fragments may yield n2 compounds

NMR: 15N-HSQC of target protein (2D NMR)

Requirements3D structure of target protein (NMR or other)large quantities of 15N-labeled protein (> 100 mg)NMR assignments of backbone N and HN atomssize of protein <40 kDasolubility: protein and ligands

Principlestart with known protein structure and 15N assignments15N-HSQC of protein15N-HSQC of protein plus ligand: identify shifted peaksmap these on protein surface: binding site

Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.;Fesik, S. W. Science 1996, 274, 1531.

Conversion of Peptides/Proteins to Small Molecules

“SAR” by NMR

“SAR” by NMR

1. Screen 100-5000 low molecular weight (150 -300 MW) ligands to identify weak binders. HSQC perturbations identigy the site of binding

2. Screen for a second site of binding in the presence of the first ligand

3. Use structural information to design a linkage between the two identified ligands

∆G(linked ligand) = ∆G(fragment 1) + ∆G(fragment 2) + ∆G(linker) + ∆G(cooperativity)

∆G(linker) usually positive (entropic cost)

∆G(cooperativity) is a non-additive effect

Application: Bcl-xL/BH3 Proteins

First Site Ligands

First Site Ligands

Second Site Ligands

Linked Inhibitor

(Bcl-xL-ABT-737)

OHN

N

N

Cl

NO2

NH

S

N

SOO

Nature 2005 Jun 2;Vol. 435(7042):p. 677-81.J. Med Chem. 2006 Jan 26;Vol. 49(2):p. 656-63.J Med Chem. 2006 Feb 9;Vol. 49(3):p. 1165-81