BACE-1 Inhibitors Alzheimer’s Disease Application of Fragment-Based NMR Screening, X-ray Crystallography, Structure-Based Design, and Focused Chemical.

BACE-1 InhibitorsAlzheimer’s Disease

Application of Fragment-Based NMR Screening, X-ray Crystallography, Structure-Based Design, and

Focused Chemical Library Design to Identify Novel μM Leads for the Development of nM BACE-1( β-Site APP Cleaving Enzyme 1) Inhibitors: JMC 2010 (Schering-Plough)

Fragment-based discovery and optimization of BACE1 inhibitors: BMCL 2010 (Evotec)

Discovery of Cyclic Acylguanidines as Highly Potent and Selective β-Site Amyloid Cleaving Enzyme (BACE) Inhibitors: Part I;Inhibitor Design and Validation: JMC 2010 (Schering-Plough)

Design and Synthesis of 5,5’-Disubstituted Aminohydantoins as Potent and Selective Humanβ-Secretase (BACE1) Inhibitors: JMC 2010 (Wyeth)

4/13/11 1CHEM E-120

Amyloid Plaque Formation

2

Insoluble, neurotoxic, protein clusters formed from Amyloid Precursor Protein (APP)

770 AA AA669 Asp

BACE β-secretase

AA669AA770

Β-CFT

γ-secretase

Aβ – β-amyloid1-40(42)

AChEfibril

peripheral siteof AChE

N terminusextracellular

C terminusintracellular

4/13/11 CHEM E-120

BACE β-secretase

Transmembrane protein : NH2-terminal protease domain, connecting region, transmembrane region, cytosol domain

Aspartic acid protease with a catalytic diad of 2 aspartic acid amino acids

BACE-1 cleaves APP in the interior of the cell at a pH of ~ 5

BACE patents filed 1997 – 2010 (C&EN 2010, April 5)

Amgen 5BMS 25Elan 65Eli Lilly 3GSK 15Merck 45Schering-Plough 25Pfizer 50Wyeth 18

4/13/11 3CHEM E-120

BACE β-secretase

Science 2000, 290, 150

OM-99-2

4/13/11 4CHEM E-120

BACE β-secretase

Science 2000, 290, 150

OM-99-2

4/13/11 5CHEM E-120

BACE β-secretase

Application of Fragment-Based NMR Screening, X-ray Crystallography, Structure-Based Design, and Focused Chemical Library Design to Identify Novel μM Leads for the Development of nM BACE-1 ( β-Site APP Cleaving Enzyme 1) Inhibitors

JMC 2010, 53, 942-950 from Schering-Plough

Strategy:

Discover low molecular weight compounds by Nuclear Magnetic Resonance – binding affinity

Validate hits by X-ray cystallography

Determine IC50 of enzyme inhibition

Optimize hits by small focused libraries

4/13/11 6CHEM E-120

7

Fragment-Based Screening

Discover of a “lead” compound – a compound that binds to the target, has functional activity, and possesses “drug-like” properties.

Modern methods:

1. High-throughput screening of large numbers of compounds: 10,000’s to 100,00’s2. Fragment-Based lead discovery: screen “components” of drugs, <5000 compounds

Screen small compounds of molecular weights 120-300Rule of 3 mw < 300 HBA < 3 rotatable bonds < 3

HBD < 3 clogP < 3

Thought is these compounds are more likely to fit into a binding site, the binding affinity

is usually quite low – KD of 100 μM to 10 mM“hit-rate” of 2-10%Target structure known either by X-ray crystallography or NMRStart to combine this fragment with others to fit better into the binding site with

greater affinity.Structure-guided modification of the lead fragment.

4/13/11 CHEM E-120

8

Fragment-Based Screening

Chemical Fragments that Hydrogen Bond to Asp, Glu, Arg, and His Side Chains in ProteinBinding Sites: J Med. Chem 2010, 533086

4/13/11 CHEM E-120

9

Fragment-Based Screening

4/13/11 CHEM E-120

10

Fragment-Based NMR Screening2D 15N-HSQC NMRMeasure the changes in chemical shifts of 15N-labled BACE-1 catalytic domain in the presence of a fragment.

Express BACE-1 with 15NH4Cl as sole nitrogen sourcePurifyAssign resonances - Identify the amino acids in the catalytic domain in the NMR spectrum

Crystal structure helpsAdd a mixture of 10 compounds and measure the spectrum, compare with spectrum of the active site without compoundsHit – measure each compound separately to identify hitQuantify by Kd

1/τ = koff + kon[S]y = b + mx

kobs (s-1)

[S]

koff (Kd)

Δδ (ppm) vs [drug]

4/13/11 CHEM E-120

11

Heteronuclear Two-Dimensional Coupling

HMQC (heteronuclear multiple quantum correlation)HSQC (heteronuclear single quantum correlation)

Correlation between the carbon atoms and the hydrogens attached DIRECTLY to the carbon, 1H vs 13C

4/13/11 CHEM E-120

12

Fragment-Based NMR Screening

4/13/11 CHEM E-120

13

Fragment-Based NMR ScreeningScreened 10,000 compound fragment library, ~ 50% obeyed Rule-of-3

4/13/11 CHEM E-120

14

Fragment-Based NMR Screening

Selected 204 isothioureas from fragment library and IC50 for BACE-1 inhibition measured15 compounds active at 50 μg/mL which where characterized by NMR to generate 336 fold increase binding affinity

got X-Ray structure

Ligand Efficiency (LE) free binding energy per non-hydrogen atom (kcal/mole/heavy atom)LE > 0.3

4/13/11 CHEM E-120

Fragment-Based NMR Screening

4/13/11 15CHEM E-120

16

Fragment-Based NMR ScreeningIsothioureas will have hydrolytic stability problems - bioisosters

4/13/11 CHEM E-120

17

Fragment-Based NMR ScreeningAnalyzed 32 2-aminopyridines by NMR giving compound 4

Kd = 32 μMLE = 0.38similar to 3but no inhibition of BACE-1 up to 1 mMbut pKa ~ 7.2good for BBB penetration

3

4/13/11 CHEM E-120

Hit-to-Lead Optimization

Small-focused libraries by parallel synthesisstep e most likely

4/13/11 18CHEM E-120

Fragment-based discovery and optimization of BACE1 inhibitors: BMCL 2010, 20, 5329 (Evotec)

Screened a 20,000 compound fragment library, average mw = 250screened at 1 mM in typical enzyme assayhits confirmed using surface plasmon resonance (SPR) and X-ray crystallography

4/13/11 19CHEM E-120

Fragment-based discovery and optimization of BACE1 inhibitors: BMCL 2010, 20, 5329 (Evotec)

Able to cocrystallize 3 with BACE1Asp32&228 catalytic diad

4/13/11 20CHEM E-120

Fragment-based discovery and optimization of BACE1 inhibitors: BMCL 2010, 20, 5329 (Evotec)

4/13/11 21CHEM E-120

Fragment-based discovery and optimization of BACE1 inhibitors: BMCL 2010, 20, 5329 (Evotec)

4/13/11 22CHEM E-120

23

Discovery of Cyclic Acylguanidines as Highly Potent and Selective β-Site Amyloid Cleaving Enzyme (BACE) Inhibitors: Part I;Inhibitor Design and Validation: JMC 2010, 53, 951-965

Problem – need to cross BBB (pKa’s of ~ 7 are best) but the enzyme exists at a pH ~ 5

They focused on the second heterocycle – bioisotere of thioisourea with pKa > 6-10

4 HBD in protonated state4/13/11 CHEM E-120

Test the idea

Computational ligand binding studies suggested the chlorophenyl group will fit into S1

Parallel Synthesis

CH3NH2 PhCH2NH2 XPhCH2NH2 RCH2NH2

P1A

P1B

P1C

P1D P4D

ROW A

4/13/11 24CHEM E-120

Type 2 did not bind by NMR and did not inhibit BACE-1

Type 1 hits

Discovered two modes of binding

4/13/11 25CHEM E-120

Problem! Have two binding modes that are not obvious to control. Know want to extend into S1, S3 and S1’. With what??

Prepare a library 500 compounds using high throughput organic synthesis. 3 sites or points of diversity.

Fix R1 and R2 = CH3 and isobutylsteroechemistry?? Use racemic R3 = 500 amines

4/13/11 26CHEM E-120

A-siteneeds 2 HBD

4/13/11 27CHEM E-120

IC50 = 27 nM

3 with IC50 = 0.2 mMto39 with IC50 = 27 nM

7407 fold enhancment

clog p = 7.5mw = 545LE = 0.25modest bioavailability

4/13/11 28CHEM E-120

Can extend from S1 into S3?

40IC50 = 605 nMBUT brain penetrationIC50 = 27 nM

4/13/11 29CHEM E-120

4/13/11 30CHEM E-120

Design and Synthesis of 5,5’-Disubstituted Aminohydantoins as Potent and Selective Human β-Secretase (BACE1) Inhibitors: JMC 2010, 53, 1146 (Wyeth)

Discovered through FRET assay

4/13/11 31CHEM E-120

Types of Compounds Tested to Develop SAR

4/13/11 32CHEM E-120

N NN O

H2N CH3

CH3

37

IC50 (nM)BACE1 40BACE2 780cathepsin D 12,500

N NN

O

H2NCH3

H3C O

F

F

(S)-55

IC50 (nM)BACE1 10BACE2 810cathepsin D 820

100 mg/kg po 69% reduction of plasma Aβ40

limited brain exposure!!!4/13/11 33CHEM E-120