New ChemDiv Library
“Beyond Flatland”
March 14, 2012
ChemDiv, Inc.
Prepared by
Sergey Tkachenko, PhD & Volodymyr Kysil, PhD
Union of Efficiency and Innovation
Yin & Yang Algorithm
New Literature & Screening Data
Knowledge Database
Discovery
and Development
of New
Reactions
Chemotype &
Compounds
Evaluation
New
Scaffolds
Generation
Available
Compound
libraries
High Throughput
Synthesis of
New
Compounds
New Focused Libraries
Diversity
Oriented
Synthesis;
Effective
Synthetic
Tools
“Escape from Flatland” – New Approach for
Scaffold & Library Design
Inspired by: Frank Lovering, et.al. Escape from Flatland: Increasing Saturation as an
Approach to Improving Clinical Success. J. Med. Chem. 2009, 52, 6752–6756
Fsp3 = number of sp3-hybridized carbons/ total carbon count
The increase of scaffold/molecule saturation leads to:
More diverse set of compounds
More highly complex molecules
Natural product-likeness
Access to greater chemical space
Better complement to the spatial subtleties of target proteins
3D-dimensionality may result in greater selectivity
Higher water solubility
Better phys-chemical parameters (logP and PSA)
Very low increase of MW
New stereo-centers
As result: Faster transition of compound from discovery to drugs
One difficulty: more complex scaffold/molecules require new perfect synthetic approaches.
Diversity oriented synthesis!!!
3D-Shape
Why do we need to escape the Flatland?
Nature “sees” and “recognizes” molecules as 3D surfaces of chemical
information. Therefore, the biological activity of any given molecule
strongly depends on its 3D shape.
In contrast to planar molecules, 3D-molecules with developed 3D shape
have more chances to be recognized by nature
The molecular shape diversity of a small molecule library is fingerprints
of overall functional diversity
Fsp3 parameter as a measure of saturation has become one of the
most important criterion of scaffold/molecule value
Diversity
The term “diversity” is somewhat subjective and needs
to be specified.
We recognize six key components of structural diversity that have
been consistently identified in the literature
 Scaffold diversity - presence of a range of distinct molecular
scaffolds;
 Functional groups diversity - variation in the functional groups
present;
 Appendage diversity (substituent or building-block diversity) variation in structural moieties around a common scaffold;
 Stereochemical diversity - variation in the orientation of potential
macromolecule-interacting elements;
 Conformational diversity - variation of possible conformers of
molecules;
 Chain diversity – presence of different distinct chains (especially if
scaffold is not determined uniquely)
Fsp3 of Small Molecules in Clinical Trials
Phase
#compounds
Fsp3%
Launched
1719
45.4
Phase 2&3
2315
42.7
Discontinued&
Withdrawn
Phase 1
2146
42.4
1223
41.1
Preclinical
21204
37.7
(compounds with MW>650 were excluded)
Fsp3 is important drug-like parameter
Comparison of Drug-like vs Natural Libraries 1
Content of sp3 carbons (frequency of occurrence)
Type of sp3 carbons
C
C
Content in
Kinase Targeted
library
Content in
Natural Product
library
23.2%
68.1%
59.8%
92.3%
87.5%
95.6%
69.3%
87.8%
C
H
C
C
H
C
H
H
H
H
Drug-like library – Kinase database (Integrity) ~25K compounds
Natural Product library – combined sources ~ 25K compounds
(compounds with MW>650 were excluded)
Comparison of Drug-like vs Natural Libraries 2
Content of flat fragments (frequency of occurrence)
Type of flat fragment
Content in
Kinase Targeted
library
Content in
Natural Product
library
91.9%
59.7%
34.8%
7.80%
2.03%
2.53%
4.48%
2.51%
5.09%
1.04%
N
N
S
Drug-like library – Kinase database (Integrity) ~25K compounds
Natural Product library – combined sources ~ 25K compounds
(compounds with MW>650 were excluded)
Increased 3D-Diversity of Flexible
vs Flat Structures
Example: Proline-like compounds
16 Isomers of Methyl-proline
O
OH
O
N
H
OH
O
N
H
OH
4 Isomers of flat
analogs
O
N
H
OH
N
H
O
N
OH
O
O
OH
O
N
H
OH
N
H
O
OH
O
N
H
N
H
OH
OH
N
H
O
O
O
OH
N
H
OH
O
O
N
N
OH
N
H
OH
O
OH
O
OH
N
H
N
H
O
OH
OH
N
H
O
OH
O
OH
N
H
N
H
N
H
Proline-like compounds 2
Structure
Flat compounds
O
OH
O
N
H
N
OH
O
CH3
OH
N
NH2
O
HN
CH3
N
Structure
Flexible
Fsp3=0.0000
logP=0.980
logSW=-3.44
PSA=53.09
O
OH
Fsp3=0.167
logP=1.180
logSW=-2.25
PSA=42.23
Fsp3=0.083
logP=2.81
logSW=-4.17
PSA=42.23
N
O
Phys-chemical
properties
Fsp3=0.167
logP=0.680
logSW=-2.15
PSA=48.02
Fsp3=0.286
logP=1.13
logSW=-3.71
PSA=34.03
O
N
H
N
OH
O
CH3
OH
N
NH2 CH3
O
HN
Fsp3=0.8000
logP=-2.330
logSW=-0.890
PSA=49.33
Fsp3=0.833
logP=-0.840
logSW=-0.826
PSA=40.54
Fsp3=0.417
logP=1.540
logSW=-1.445
PSA=40.54
N
O
Phys-chemical
properties
N
Fsp3=0.833
logP=-1.00
logSW=-0.784
PSA=46.33
Fsp3=0.857
logP=0.100
logSW=-0.454
PSA=32.34
Flexibility improves phys-chemical properties (logSW, ClogP, PSA)
of scaffolds or building blocks
Fsp3 Difference of Different GPCR Ligands and Libraries
Fsp3 for GPCR ligands
35
#compounds%
30
Virtual database from Integrity
& MedChem sources:
Adenosine
Serotonin
25
Chemokine
Chemokine - 4.2K ligands
20
Serotonin - 8.6K ligands
15
10
Adenosine - 2.4K ligands
5
0
0
10
20
30
40
50
60
70
80
90
100
Natural set - 25K compounds
Fsp3%
Fsp3 in different databases
Available collection of compounds:
35
Peptidomimetic Library ~
20K compounds
30
Chemokines
Peptidomimetics
# compounds %
25
Diverse Base1
Natural set
20
Diverse set - 250K compounds
15
10
Chemokines Set and Peptidomimetic
Library have a similar distribution of
the Fsp3 Chemokines are very different
from diverse and natural sets
5
0
0
10
20
30
40
50
Fsp3%
60
70
80
90
100
Criteria to be Considered for Novel Scaffold/Molecule Design
Parameter
Scaffold
Molecule
No undesirable functionalities (MedChem filters)
No undesirable chemotypes (MedChem filters)
Amide bonds
No more 2 amide bonds
(cyclic or linear)
No more 2 amide bonds
(cyclic or linear)
Aromatic rings
No more 2 aromatic
rings
No more 3 aromatic
rings
MW
100<MW<350
150<MW<450
Fsp3
>0.30
>0.30
ClogP
-1.0 <ClogP< 3.5
0 <ClogP<5.0
PSA
10<PSA<60
40<PSA<90
HBA/HBD
<6/2
<8/3
Rotatable bonds
<6
<8
Comparison of Newest Chemdiv’s Library (~25K),
Available Diversity Set (Internet, 250K) and
Natural Products (~25K)
New Old NP Fsp3
45
40
#compounds%
35
New chemistry 2011
30
250K diversity (Internet)
25
Natural Products (~25K)
20
15
10
5
0
0
10
20
30
40
50
60
70
80
90
100
Fsp3%
Our chemistry is becoming more similar to natural products
Available Library “Beyond Flatland”
Library (Edition March 2012) contains 26267 compounds
Rot_Bond & Acceptor Distribution
ClogP vs MW
30
6
5
Rot bonds
20
4
Acceptors
ClogP
# compounds %
25
15
10
3
2
1
5
0
0
0
1
2
3
4
5
6
7
8
9
10
-1
150
11
200
250
300
# atom or bond
sp3 distribution
400
450
non-H atom distribution
40
14
35
12
# compounds %
30
25
20
15
10
10
8
6
4
2
5
0
0
20
30
40
50
60
70
80
90
sp3 %
100
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
# non-H atoms
Ring distribution
60
50
# compounds %
# compounds %
350
MW
40
#Aromatic Rings %
# All Rings %
30
20
10
0
0
1
2
3
4
# Rings
5
6
7
New drug-like compounds
Natural-like compounds
Unique scaffolds
3D diversity
Enrichment of unique structures
(spiro ~7%)
33
Scaffold Selection
From Prototypes to Scaffolds
Prototypes: more 1,500 antibacterial small molecules (from new patents,
preclinical, phase I-III, under active development, natural, natural-like, etc.).
Known targets or mechanism of action.
HO
HC
3
H2N
Cl
O
N
N
O
Cl
H2N
H
N
O
O
O
N
N N
N
CH3
O
O
Cl
N
O
Cl
O
O
CH3
N
O
N
H3C
O
Cl
O
CH3
CH3
~ 550 new scaffolds proposed
~ 250 new IP-clean scaffolds
selected for synthesis
Cl
O
N
B
Cl
CH3
O
N
H
N
O
N
N
N
O
CH3
NH2
NH2
O
OH
Cl
O
O
H3C
S
N
F
CH3
O
F
N
H
O
S
O
N
H
O
N
N
H
NH2
N
O
N
S
S
N
O
O
S
F
Cl
O
N
NH
Cl
Cl
N
H
N
H
NH
N
H
CH3
O
N
O
O
N
N
O
H
N
N
N
H
N
H
O
O
O
“Beyond the Flatland”
Template Design Criteria:
at least one of the following:




High degree of (hetero-)cycle saturation (Preferable Fsp3= 0.3 - 1)
One or more stereogenic centers
Bridged or spiro-fused ring systems
Structural fragments of naturally occurring compounds
possessing interesting biological activity
 7 (or more) membered (hetero-)cyclic rings
 Rare or unique combinations of privileged heterocycles
 widely distributed in nature, e.g. indole, isoquinoline, etc.
 preferably with at least one stereogenic center
&
(if possible) biological annotation
New Anti-bacterial Scaffolds (Examples)
Scaffold
Possible Targets
Scaffold
Possible Targets
R2
N
N
N
N
N
Fsp3=0.875
R1
R2
N
R1a
N
Bacterial Efflux Pump
Toll-Like Receptors
Ligase dltA
N
R2
N
N
R3
Fsp3=0.714
MurF Synthetase
N
Fsp3=0.555
O
R1b
O
Cell Wall Biosynthesis
R1
Bacterial
transcriptional
regulatory repressor
protein EthR
Rb
N 1
R1a
N
R2
O
Undecaprenyl
pyrophosphate
synthase (UPPS)
Fsp3=0.400
R2c
R1
N
N
R2
Fsp3=0.375
R1a
N
R1b
N
R3 N
N
R2
O
Fsp3=0.571
H
Enoyl-(acyl-carrier
protein) Reductase
Fabl
N
N
R1
FtsZ Polymerization
Phosphopantetheine
Adenylyltransferase
(PPAT)
R3
R1 a
Fsp3=0.538
H
R1 b
N
H
N N
N
R2
N
Histidine protein
kinase
R3
Fsp3=0.750
Scaffolds with high Fsp3 and good phys-chemical properties
New Metabolic Scaffolds (Examples)
Scaffold
Possible Targets
Scaffold
R2
R2
N
O
Histamine H3
Receptor Antagonists
O
N
R1
O
R1a
N
N
N
R1
Fsp3=1.00
N
R1
O
R2
Bile Acid Responsive
TGR5 Receptors
(AXOR 109, GPCR19)
Agonists
N
N
R2
N
R1
O
N
R1
N
R2b
Fsp3=0.833
Orexin OX-1
Antagonists
N
Fsp3=0.750
R3b
O
R2a
Fsp3=0.571
O
Fsp3=0.444
R3a
5-HT2C Receptor
Ligands
N
R1
H
N
R2a
Somatostatin SRIF1B
(sst5) Antagonists
N
Melanocortin MC4
Agonists
R1b
Fsp3=0.875
R2b
N
N
N
H
Fsp3=0.875
O
R2
Possible Targets
Triacylglycerol Lipase
(Hormone-Sensitive
Lipase) Inhibitors
N
N
N
R1
R2
Acetyl-CoA
Carboxylase (ACC)
Inhibitors
Fsp3=1.00
Scaffolds with high Fsp3 and good phys-chemical properties
Chemistry
Synthetic Tools for Framework Assembly
 Reaction Types:
 multicomponent reactions (MCR), intramolecular and
post-MCR modifications;
 new MCRs;
 azomethine dipolar cycloaddition;
 Halocyclizatons;
 “classic” (hetero-)cyclization, but in new sequences (e.g.
Michael addition followed by cyclization);
 ring closure metathesis (RCM);
 Unique unions of known and recently developed
reactions (e.g., Ugi-MCR – Diels-Alder-skeletal
rearrangement; intramolecular Ugi-MCR – selective
reduction etc.);
 and more…
The chemistry should meet DOS (Diversity Oriented
Synthesis) criteria!
Multicomponent Reactions (MCR)
New MCR: unique tool for functionally enriched pyrazines and 1,4diazepines synthesis developed at ChemDiv
TL 2007,48, 6239-6244;
TL 2009, 50, 2854–2856;
Eur. JOC 2010, 1525–1543.
+
N
C
NH2
N
TMSCl
+
n
NH2
n = 1, 2
R3
N
N
NH
R2
N
H R2
R2
R3
R1
n
R1
O
R3
NH
R1
R3
N
N
R1
NH
R2
N
N
H
H
N
H
N
R2
N
N
R1
N R1
NH
R2
H
N
N
N
NH
R2
R1
Michael-type Azaanulation
Acylation of N-Bn-enamines based on activated ketones with acryloyl
chloride followed by Michael-type cyclization has been reported by
Paulvannan and Stille [JOC 1994, 59, 1613-1620].
bn
O
R1
z
NH
R1
z
Cl
O
O
O
bn
bn
N
N
R1
R1
z
Z = CO2Et, COR, CONHPh, SO2R, CN; R1 = Alk.
How can we use this chemistry for unique scaffolds
generation?
z
Michael-type Azaanulation
Study of the scope of the reaction and/or reactivity of functional
groups that could be incorporated into the core might provide a great
number of unique scaffolds
Potential point
of diversity
R
O
bn
NH
R
R
O
N
Cl
R1
z
O
R
z
R
R
z
R
Potential site for
scaffold modification
Potential sites for
scaffold transformations
Potential points
of diversity
Michael-type Azaanulation
Structural variety of activated ketones ensures cores variety
O
O
R
R
R
N
O
NH
R
O
R
+
O
Cl
R
z
OAlk
N
R
OAlk
O
O
O
R
O
N
O
O
O
R
N
R
N
R
N
R
N
O
O
OAlk
O
N
PG
O
N
OAlk
OAlk
N
PG
OAlk
O
O
PG
OAlk
Our Expertise in Novel Library/Compound
Design
Tsaloev A., Ilyin A., Tkachenko S., Ivachtchenko A., Kravchenko D., Krasavin M.
Cyclic products of the Ugi reaction of aldehydo and keto carboxylic acids: chemoselective
modification. Tetrahedron Letters. 2011, 52: 1800–1803.
Kysil V., Khvat A., Tsirulnikov S., Tkachenko S, Williams C., Churakova M., Ivachtchenko A.
General Multicomponent Strategy for the Synthesis of 2-Amino-1,4-diazaheterocycles: Scope,
Limitations, and Utility. European Journal of Organic Chemistry. 2010; 1525–1543.
Kysil V.M., Khvat A., Tsirulnikov S., Tkachenko S., Ivachtchenko A. Multicomponent approach to
unique 1,4-diazepine-2-amines. Tetrahedron Letters. 2009; 50(24): 2854-2856.
Balakin K.V., Ivanenkov Y.A., Tkachenko S.E., Kiselyov A.S., Ivachtchenko A.V. Regulators of
chemokine receptor activity as promising anticancer therapeutics. Current Cancer Drug Targets.
2008; 8(4): 299-34.
Kiselyov A.S., Tkachenko S.E., Balakin K.V., Ivachtchenko A.V. Small-molecule modulators of Hh
and Wnt signaling pathways. Expert Opinion on Therapeutic Targets. 2007; 11(8): 1087-1101.
 Savchuk N.P., Tkachenko S.E., Balakin K.V. Design of pGPCR-targeted Libraries. In Rognan D.,
ed. Ligand Design for G Protein-coupled Receptors. Methods and Principles in Medicinal
Chemistry (Volume 30). Weinheim: Wiley VCH. 2006, pp. 137-164.
 Kysil V., Tkachenko S., Khvat A., Williams C., Tsirulnikov S., Churakova M., Ivachtchenko A.
TMSCl-Promoted Isocyanide-Based MCR of Ethylenediamines: an Efficient Assembling of
2-Aminopyrazine Core. Tetrahedron Letters, 2007; 48(36): 6239-6244.
Thank You