Ligand-based Design - Leibniz Institute for Age Research

-2010-
3D Structures of Biological Macromolecules
Part 4: Drug Research and Design
Jürgen Sühnel
jsuehnel@fli
jsuehnel@fli-leibniz.de
jsuehnel@fli leibniz.de
Leibniz Institute for Age Research, Fritz Lipmann Institute,
Jena Centre for Bioinformatics
Jena / Germany
Supplementary Material: www.fli-leibniz.de/www_bioc/3D/

Example of Drug Discovery

Example of Drug Discovery

Example of Drug Discovery
Pacific yew tree
(Eibe)

Drug Research is
the Search for a Needle in a Haystack.
www.kubinyi.de

Development of Drug Research
www.kubinyi.de

Drug Timeline
www.kubinyi.de

Drug Timeline
www.kubinyi.de

Costs in Drug Research
4 Cost for discovering and developing a new drug:
several € 100 million up to € 1000 million (average € 802 M)
4 Time to market:
10 – 15 years

Global Company Sales 2006

Top Ethical Drugs by Sales in 2006
(Lowering blood cholesterol)
(Asthma treatment)
(Inhibits blood clots)
(Proton pump inhibitor; treatment of dyspepsia, peptic ulcer disease, …)
(Calcium channel blocker; anti-hypertensive agent)
http://www.p-d-r.com/ranking/Top_100_Ethical_Drugs_by_Sales.pdf

New Products Marketed for the First Time
http://www.p-d-r.com/ranking/Prous_TYND_2005.pdf

Disciplines Involved in Drug Development
Molecular Conceptor

The Role of Molecular Structure
Molecular Conceptor

The Pharmacophore Concept
Molecular Conceptor

Mechanisms of Drug Action – Definitions I
www.kubinyi.de

Mechanisms of Drug Action – Definitions II
www.kubinyi.de

Serendipity - Penicillin
Molecular Conceptor

Serendipity - Penicillin

Serendipity - Aspirin
Serendipity - Aspirin
Molecular Conceptor

Strategies in Drug Design
www.kubinyi.de

Computational Approaches to Drug Research
TTarget t id identification tifi ti
Lead discovery
Lead optimization
Ligand-based design
Receptor-based design (Docking)
D t b i (Vi t l i )
Database screening (Virtual screening)
Supporting combinatorial chemistry

3D Structures in Drug Design
www.kubinyi.de

Lead Structure Identification
www.kubinyi.de

Lead Structure Search Pipeline
www.kubinyi.de

Lead Structures:
Endogenous Neurotransmitters
www.kubinyi.de

Lead Structures:
Endogenous Neurotransmitters
Neurotransmitters are chemicals that are used to relay, amplify and
modulate electrical signals between a neuron and another cell.
Acetylcholine: voluntary movement of the muscles
Noradrenaline: wakefulness or arousal
Dopamine: voluntary movement and emotional arousal
Serotonin: sleep and temperature regulation
GABA: (gamma (g aminobutryic y acid) ) - motor behaviour
www.kubinyi.de

Lead Optimization
www.kubinyi.de

Ligand Ligand-based based Design: What is QSAR ?

Ligand Ligand-based based Design: Basic Requirements for QSAR Studies

Ligand Ligand-based based Design: QSAR
HHansch h analysis l i iis th the iinvestigation ti ti of f the th quantitative tit ti relationship l ti hi bbetween t th the
biological activity of a series of compounds and their physicochemical substituent
or global parameters representing hydrophobic, electronic, steric and other effects
using multiple regression correlation methodology
methodology.
www.kubinyi.de

Ligand Ligand-based based Design: QSAR
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters

Ligand Ligand-based based Design: QSAR Parameters - Lipophilicity

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
‐ reaction constant
‐ substituent constant
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: QSAR Parameters
www.kubinyi.de

Ligand Ligand-based based Design: A QSAR Success Story
www.kubinyi.de

Ligand Ligand-based based Design: A QSAR Success Story
pI50 – concentration of test compound required to reduce the protein content of cell by 50%
pI 50
www.kubinyi.de

Ligand Ligand-based based Design: 3D 3D-QSAR QSAR CoMFA
www.kubinyi.de

Molecular Superposition of D Receptor Ligands
www.kubinyi.de

Ligand Ligand-based based Design: 3D 3D-QSAR QSAR CoMFA
www.kubinyi.de

Ligand Ligand-based based Design: 3D 3D-QSAR QSAR CoMFA
www.kubinyi.de

Ligand Ligand-based based Design: 3D 3D-QSAR QSAR CoMFA
Partial least squares regression (PLS regression) is a statistical method that finds a linear regression model by
projecting the predicted variables and the observable variables to a new space. Because both the X and Y data are
projected to new spaces, spaces the PLS family of methods are known as bilinear factor models. models
PLS is used to find the fundamental relations between two matrices (X and Y), i.e. a latent variable approach to
modeling the covariance structures in these two spaces. A PLS model will try to find the multidimensional direction
in the X space that explains the maximum multidimensional variance direction in the Y space space. PLS‐regression is
particularly suited when the matrix of predictors has more variables than observations, and when there
is multicollinearity among X values. By contrast, standard regression will fail in these cases.
PLS regression is an important step in PLS path analysis analysis, a multivariate data analysis technique that employs latent
variables. This technique is often referred to as a form of variance‐based or component‐based structural equation
modeling.
Partial least squares was introduced by the Swedish statistician Herman Wold Wold, who then developed it with his son son,
Svante Wold, a professor of chemometrics at Umeå University. An alternative term for PLS (and more correct
according to Svante Wold [3] ) is projection to latent structures, but the term partial least squares is still dominant in
many areas. It is widely applied in the field of chemometrics, in sensory evaluation, and more recently, in the analysis
of functional brain imaging data [4]
of functional brain imaging data.

Electrostatic and Van Van-der der-Waals Waals Interactions

Ligand Ligand-based based Design: 3D 3D-QSAR QSAR CoMFA
Comparative
Molecular
Field
Analysis

Receptor Receptor-based based Design ( (Structure Structure-based based Design)
Molecular Conceptor

Receptor Receptor-based based Design ( (Structure Structure-based based Design)
Molecular Conceptor

Receptor Receptor-based based Design: Docking
Molecular Conceptor

Receptor Receptor-based based Design: Docking
Molecular Conceptor

Receptor Receptor-based based Design: Docking
Molecular Conceptor

Hydrophobic Amino Acids
Molecular Conceptor

Receptor Receptor-based based Design: Docking
Molecular Conceptor

H-Bond Bond Properties of Amino Acids
Molecular Conceptor

Receptor Receptor-based based Design: HH-bond
bond Effect
IC50 -
Drug concentration
required for 50% inhibition of a
biological effect
Molecular Conceptor

Receptor Receptor-based based Design: HH-bond
bond Effect
www.kubinyi.de

Charge Properties of Amino Acids
Molecular Conceptor

Receptor Receptor-based based Design: Salt Bridge
116.
Molecular Conceptor

Receptor Receptor-based based Design: Docking
Molecular Conceptor

Receptor Receptor-based based Design: SAR (Pharmacophore
Pharmacophore Features)
Molecular Conceptor

Receptor Receptor-based based Design: DNA Receptor
Molecular Conceptor

Receptor Receptor-based based Design: DNA Intercalating Agents
Molecular Conceptor

Receptor Receptor-based based Design: DNA Intercalating Agents
Molecular Conceptor

Receptor Receptor-based based Design: AIDS Drugs

Receptor Receptor-based based Design: AIDS Drugs

Combinatorial Diversity in Nature
www.kubinyi.de

Classical vs. Combinatorial Chemistry
ww.kubinyi.de

Combinatorial Library
ww.kubinyi.de

Combinatorial Library
ww.kubinyi.de

Types and Features of Combinatorial Libraries
ww.kubinyi.de

Receptor Receptor-based based Design: Virtual Screening
Virtual Screening:
Select subsets of compounds for assay that are more likely to contain
active hits than a sample chosen at random
Time Scales:
Docking of 1 compound 30 s
(SGI R10000 processor)
Docking of the 1.1 million data set 6 days
(64-processor SGI ORIGIN)
ACD-SC: Database from Molecular Design Ltd.
Agonists: Known active compounds
Docking of ligands to the estrogen receptor
(nuclear hormone receptor)

Receptor Receptor-based based Design: Virtual Screening

Lipinski‘s „Rule Rule of Five“ Five
Compounds are likely to have a good absorption and permeation
in biological systems and are thus more likely to be successful drug candidates
if they meet the following criteria:
•5 or fewer H-bond donors
•10 or fewer H-bond acceptors
•Molecular weight less than or equal to 500 daltons
•Calculated log P less than or equal to 5
•„Compound classes that are substrates for biological transporters are exceptions to the rule“.
Druggable gg compounds p

ADME
ADME

The Future: Pharmacogenomics and Personalized Medicine
www.kubinyi.de

Prediction Issues
www.kubinyi.de