Transcript Lecture Contents -

Lecture Contents -- Unit 3
• Drug Discovery
–
–
–
–
–
–
Basic objectives and problems
Screening approach vs. rational design
Phytopharmacology
Databases, QSAR, and CoMFA
“Pharmacogenomics” and “proteomics”
Case study: GV 150526A
Basic Facts About Drug Discovery
• Almost any metabolic pathway with all it’s
adjuncts (receptors, enzymes, genes therefor, and
regulatory elements) is a potential drug target
• During the past century, pharmacology has
identified some 400 such targets; the human
genome project confirms that thousands must exist
• Independent of this, the present rate of drug
discovery is insufficient; new strategies are
required
Some Companies
Specialize in Drug Discovery
Drug Discovery Strategies
• Screening-based:
–
–
–
–
Traditional medicine
Bioprospecting
Mass screening of microbial strains
Combinatorial chemistry
• Rational Drug Design
– Target interaction analysis
and molecular modeling
Natural Product-Based
Drug Discovery
Natural Product Success Stories
• Microorganisms: Antibiotics
• Plants:
– Taxoids for cancer
– Artemisinin for malaria
– Huperzine A and galanthamine for Alzheimer
• Animals: Conotoxins as ultra-high potency
analgetics
Phytopharmacology: Decision Tree
„Microbial Pharmacology:“
Penicillin And Other ß-Lactames
• Fleming (1928): Growth of bacterial cultures
inhibited by co-infection with Penicillium notatum
 “penicillin” postulated as a secreted molecule
• 1938: Penicillin isolated and characterized as part
of British war preparations
• Beta-lactames became most important lead
structure ever since then
Benzylpenicillin (Penicillin V)
Phytopharmacology: Taxoids
• Diterpene from Taxus brevifolia
• Most significant anticancer agent developed
in the past two decades (“mitotic poison”)
Phytopharmacology: Artemisinin
• Unusual sesquiterpene endoperoxide from
Artemisia annua (Quinghaosu in Chinese
traditional medicine)
• Lead compound for new generation of
malaria therapeutics (including chloroquineresistant and cerebral malaria)
C15H22O5
MW = 282.3
Marine Pharmacology: Conotoxins
• Peptide neurotoxins (receptor channel
blockers) from molluscs (snails and shells)
-conotoxin PnIa:
nicotinic receptor blocker
-conotoxin MVIIc:
P-type Ca-channel blocker
The Ideal Combinatorial Library
Made by forming all possible combinations
of a series of sets of precursor molecules,
and applying the same sequence of reactions
to each combination
Combinatorial Chemistry:
Basic Theoretical Approach
R1
R2 TEMPLATE
R3
Combinatorial Chemistry:
Detection of Hits
Obstacles to Combinatorial Chemistry
• Restricted and specialized chemistry, needs
training
• Not yet suitable for large molecules
• Automated synthesis needs to be installed
and integrated with the laboratory workflow
• Equipment AND organization must be
tightly integrated with a tailored data
management infrastructure
A Well-Designed Library
Can Mean BIG Money...
• 1995: Schering-Plough pays $3 million for
access to certain parts of the Neurogen
compound library
• Payment estimates for unrestricted access to
targeted libraries run up to $15 million
• Construction of large (diverse or targeted)
combinatorial libraries) has become a
significant outsourcing business
Combinatorial Chemistry:
SAR By NMR
New Frontiers in Receptor
Ligand Screening
Databases In Drug Discovery
• Employ advanced search algorithms including
artificial intelligence (AI) systems
• “Data Mining” -- knowledge discovery in
databases:
–
–
–
–
Fuzzy logic -- “soft” search criteria
Structural similarity searches
Retrieve implicit information
Link structural information with bio-informatics
Tools for Rational Drug Design
• (Q)SAR: (Quantitative) Structure-Activity
Relationships
• SAFIR: Structure-Affinity Relationships
• SPAS: Structure-Property/Affinity Studies
• CoMFA: Comparative Molecular Field
Analysis
SARs, Easy and Obvious?
Stimulants/Anorectics in Medicine
SARs, Easy and Obvious?
Stimulant Drugs of Addiction
Can „Drug-Like“ Structures
Be Predicted?
• Only 32 basic templates describe half of all known
drugs (Bemis et al. 1996)
• Medicinal chemists essentially use their intuition
(“expert rules”) to gauge drug structures 
emulation by trainable (and self-entraining)
neuronal networks working from relatively few
molecular descriptors
• If “drug-likeness” can be quantified  targeted
design of combinatorial libraries
Comparative Molecular Field Analysis
• CoMFA: Method to analyze and predict
structure-activity relationships (Cramer 1988)
• Based on superimposition techniques:
–
–
–
–
–
Steric overlap (“distance geometry”)
Crystallographic data
Pharmacophore theory
Steric and electrostatic alignment algorithms
„Automated field fit“
Further reading:
http://www.netsci.org/Science/Compchem/feature11.html ; http://cmcind.far.ruu.nl/webcmc/camd/3dqsar.html
The Essence of CoMFA
• Superpose active and inactive analogues; calculate the
“receptor excluded volume,” the occupancy of which would
result in loss of activity
• Use ligand binding points and conformational restraints to
decompose the distance matrix into differences and
similarities
© Tripos Software
Somatostatin Receptor Ligand
Modeling
Science 282, 737-9 (23 Oct 98)
New Buzzwords in Drug Discovery
A Case Study In Drug Discovery
GV-150526A
(CAS: 153436-38-5)
3-[2-phenylaminocarbonyl)ethenyl]-4,6-dichloroindole-2-carboxylate,
a glycine antagonist currently completing Phase III studies for stroke
Glutamate, Receptors, And Stroke
The NMDA Receptor Complex
Starting Point: Known Antagonists of
Glycine Site at the NMDA Receptor
Kynureic acid (R1 and R1 can be H or Cl)
Nanomolar in vitro affinity but poor in vivo
activity due to insufficient CNS penetration
Improved CNS penetration
but lack of receptor selectivity
!
4,6-dichloroindole-2-carboxylate:
Good receptor selectivity and CNS penetration,
but in vitro affinity for glycine site (pKi=5.7)
needs to be improved; however:
A NEW LEAD STRUCTURE IS IDENTIFIED!
Input From Theory
Comparison with receptor model
predicts that a hydrogen bond accepting
group in the “northeast” of the template
is required for optimal binding
  C-3 unsaturated side chains
should be able to considerably enhance
the affinity to the glycine binding site
Template Derivatization At C-3
PRIMARY SCREENING SYSTEM:
In vitro binding inhibition of [3H]-glycine
to crude synaptic membrane preparations
from adult rat cerebral cortex
SARs From Primary Screening
R
H
CH2-CH2-COOH
CH2-CH2-CONH-Ph
CH=CH-COOH
CH=CH-COO-tBu
CH=CH-CONH-Ph
CH=CH-CONH-C10H7
CH=CH-CONH-CH2-Ph
CH=CH-SO2NH-Ph
pKi
5.7
7.4
7.6
7.7
6.3
8.5
7.4
6.9
6.1
pKi = inverse logarithm of binding constant to the glycine site of the NMDA receptor
Can The in vitro Characteristics of the
Refined Lead Be Improved Further?
Ro
Rm
Rp
pKi
H
H
H
NH2
H
NO2
H
CH3
NO2
H
H
H
H
H
H
H
NH2
H
H
H
OCH3
H
H
H
H
H
NO2
H
H
NH2
H
H
OH
H
OCH3
OCH3
F
COOH
N(CH3)2
O-CH2-CH3
Cl
CF3
8.5
8.9
8.3
8.5
8.7
7.6
8.1
7.7
7.5
7.2
7.9
8.3
6.9
6.8
The Glycine Site
of the NMDA Receptor