Transcript Protein Data Bank Advisory Committee

New Targets for Old Drugs: Ideas
from in silico Analysis
Philip E. Bourne
University of California San Diego
[email protected]
WPS-AMEFAR Meeting February 10, 2010
Agenda
• Motivation
• Computational Methodology
• Repositioning an Existing Drug - The TB Story
• The Future? - The Human vs Pathogen
Drugome
Big Questions in the Lab
1.
2.
3.
4.
August 14, 2009
Can we improve how
science is disseminated
and comprehended?
What is the ancestry of the
protein structure universe
and what can we learn
from it?
Are there alternative ways
to represent proteins from
which we can learn
something new?
What really happens when
we take a drug?
Motivation
• The truth is we know very little about how the
major drugs we take work
• We know even less about what side effects
they might have
• Drug discovery seems to be approached in a
very consistent and conventional way
• The cost of bringing a drug to market is huge
>$800M
• The cost of failure is even higher e.g. Vioxx >$4.85Bn
Motivation
Motivation
• The truth is we know very little about how the
major drugs we take work – receptors are
unknown
• We know even less about what side effects
they might have - receptors are unknown
• Drug discovery seems to be approached in a
very consistent and conventional way
• The cost of bringing a drug to market is huge
~$800M – drug reuse is a big business
• The cost of failure is even higher e.g. Vioxx $4.85Bn - fail early and cheaply
Motivation
Why Don’t we Do Better?
A Couple of Observations
• Gene knockouts only effect phenotype in 1020% of cases , why?
– redundant functions
– alternative network routes
– robustness of interaction networks
A.L. Hopkins Nat. Chem. Biol. 2008 4:682-690
• 35% of biologically active compounds bind to
more than one target
Paolini et al. Nat. Biotechnol. 2006 24:805–815
Motivation
Implications
• Ehrlich’s philosophy of magic bullets
targeting individual chemoreceptors has
not been realized
• Stated another way – The notion of one
drug, one target, one disease is a little
naïve in a complex biological system
Motivation
How Can we Begin to Address
the Problem?
• Systematic screening for multiple
targets
• Integration of knowledge from multiple
sources
• Analyze the impact of multiple targets
on the complete biological network
Motivation
2. What is the ancestry of the
protein structure universe?
4. What really happens when
we take a drug?
Valas, Yang & Bourne 2009
Current Opinions in Structural Biology 19:1-6
What if only the binding
pocket was conserved and the
global structure of the protein
has changed?
A drug could potentially bind to
distinctly different gene families
Put More Simply:
Can We Find Off-targets and What Do
They Tell Us?
•
They tell us one of four things:
1. Nothing
2. A possible explanation for a side-effect of a drug
3. A possible repositioning of a drug to treat a
completely different condition
4. A multi-target strategy to attack a pathogen
Today I will give you examples of 3 and 4 while
illustrating the complexity of the problem
Agenda
• Motivation
• Computational Methodology
• Repositioning an Existing Drug - The TB Story
• The Future? - The Human vs Pathogen
Drugome
Need to Start with a 3D Drug-Receptor
Complex - The PDB Contains Many
Examples
Generic Name
Other Name
Treatment
PDBid
Lipitor
Atorvastatin
High cholesterol
1HWK, 1HW8…
Testosterone
Testosterone
Osteoporosis
1AFS, 1I9J ..
Taxol
Paclitaxel
Cancer
1JFF, 2HXF, 2HXH
Viagra
Sildenafil citrate
ED, pulmonary
arterial
hypertension
1TBF, 1UDT,
1XOS..
Digoxin
Lanoxin
Congestive heart
failure
1IGJ
Computational Methodology
A Reverse Engineering Approach to
Drug Discovery Across Gene Families
Characterize ligand binding
site of primary target
(Geometric Potential)
Identify off-targets by ligand
binding site similarity
(Sequence order independent
profile-profile alignment)
Extract known drugs
or inhibitors of the
primary and/or off-targets
Search for similar
small molecules
…
Dock molecules to both
primary and off-targets
Statistics analysis
of docking score
correlations
Computational Methodology
Characterization of the Ligand Binding
Site - The Geometric Potential
 Conceptually similar to hydrophobicity
or electrostatic potential that is
dependant on both global and local
environments
• Initially assign Ca atom with
a value that is the distance
to the environmental
boundary
• Update the value with those
of surrounding Ca atoms
dependent on distances and
orientation – atoms within a
10A radius define i
GP  P 
Pi
cos(ai)  1.0

2.0
neighbors Di  1.0

Computational Methodology
Xie and Bourne 2007 BMC Bioinformatics, 8(Suppl 4):S9
Discrimination Power of the Geometric
Potential
4
binding site
non-binding site
3.5
• Geometric
potential can
distinguish
binding and
non-binding
sites
3
2.5
2
1.5
1
0.5
100
99
88
77
66
55
44
33
22
11
0
0
Geometric Potential
Computational Methodology
0
Geometric Potential Scale
Xie and Bourne 2007 BMC Bioinformatics, 8(Suppl 4):S9
Local Sequence-order Independent Alignment
with Maximum-Weight Sub-Graph Algorithm
Structure A
Structure B
LER
VKDL
LER
VKDL
• Build an associated graph from the graph representations of two
structures being compared. Each of the nodes is assigned with a
weight from the similarity matrix
• The maximum-weight clique corresponds to the optimum alignment
of the two structures
Xie and Bourne 2008 PNAS, 105(14) 5441
Similarity Matrix of Alignment
Chemical Similarity
• Amino acid grouping: (LVIMC), (AGSTP), (FYW), and
(EDNQKRH)
• Amino acid chemical similarity matrix
Evolutionary Correlation
• Amino acid substitution matrix such as BLOSUM45
• Similarity score between two sequence profiles
d   f a Sb   f b S a
i
i
i
i
i
i
fa, fb are the 20 amino acid target frequencies of profile a
and b, respectively
Sa, Sb are the PSSM of profile a and b, respectively
Computational Methodology
Xie and Bourne 2008 PNAS, 105(14) 5441
Nothing in Biology {Including
Drug Discovery} Makes Sense
Except in the Light of
Evolution
Theodosius Dobzhansky
(1900-1975)
Agenda
• Motivation
• Computational Methodology
• Repositioning an Existing Drug - The TB Story
• The Future? - The Human vs Pathogen
Drugome
Found..
• Evolutionary linkage between:
– NAD-binding Rossmann fold
– S-adenosylmethionine (SAM)-binding domain of SAMdependent methyltransferases
• Catechol-O-methyl transferase (COMT) is SAMdependent methyltransferase
• Entacapone and tolcapone are used as COMT
inhibitors in Parkinson’s disease treatment
• Hypothesis:
– Further investigation of NAD-binding proteins may
uncover a potential new drug target for entacapone
and tolcapone
Repositioning an Existing Drug - The TB Story
Functional Site Similarity between
COMT and InhA
• Entacapone and tolcapone docked onto 215 NADbinding proteins from different species
• M.tuberculosis Enoyl-acyl carrier protein reductase ENR
(InhA) discovered as potential new drug target
• InhA is the primary target of many existing anti-TB drugs
but all are very toxic
• InhA catalyses the final, rate-determining step in the fatty
acid elongation cycle
• Alignment of the COMT and InhA binding sites revealed
similarities ...
Kinnings et al. 2009 PLoS Comp Biol 5(7) e1000423
Repositioning an Existing Drug - The TB Story
Binding Site Similarity between
COMT and InhA
COMT
SAM (cofactor)
BIE (inhibitor)
InhA
NAD (cofactor)
641 (inhibitor)
Repositioning an Existing Drug - The TB Story
Summary of the TB Story
• Entacapone and tolcapone shown to have potential
for repositioning
• Direct mechanism of action avoids M. tuberculosis
resistance mechanisms
• Possess excellent safety profiles with few side effects
– already on the market
• In vivo support
• Assay of direct binding of entacapone and tolcapone
to InhA reveals a possible lead with no chemical
relationship to existing drugs
Kinnings et al. 2009 PLoS Comp Biol 5(7) e1000423
Repositioning an Existing Drug - The TB Story
Summary from the TB Alliance
– Medicinal Chemistry
• The minimal inhibitory concentration (MIC)
of 260 uM is higher than usually
considered
• MIC is 65x the estimated plasma
concentration
• Have other InhA inhibitors in the pipeline
• The chemistry is novel and may be
revisited
• Interested in our approach
Repositioning an Existing Drug - The TB Story
Agenda
• Motivation
• Computational Methodology
• Repositioning an Existing Drug - The TB Story
• The Future? - The Human vs Pathogen
Drugome
Existing Drugs
3. Protein-ligand
Docking
TB Structural
Proteome
…
TB Protein-drug
Interactome
2. Binding site
Similarity
Drugome/TB
1. Structural
Determination
& Modeling
TB Genome
4.2 Network
Integration
4.1 Network
Reconstruction
TB Metabolome
Target identification
Drug repurposing
Side effect prediction
New therapeutics
for MDR and XDR-TB
Drug resistance
mechanism
The TB Drugome Bioinformatics 2009 25(12) 305-312
The Future? - The Human vs Pathogen Drugome
Predicted protein-ligand interaction network of M.tuberculosis. Proteins that are
predicted to have similar binding sites are connected. Squares represent the
top 18 most connected proteins.
The Future? - The Human vs Pathogen Drugome
Some Limitations
• Structural coverage of the given
proteome
• False hits / poor docking scores
• Literature searching
• It’s a hypothesis – need experimental
validation
• Money 
Summary
• We have established a protocol to look for offtargets for existing therapeutics and NCEs
• Understanding these off-targets in the context
of pathways and complete biological systems
would seem to be the next step towards a
new understanding – cheminfomatics meets
systems biology
Example On-going
Collaborations
• Metabolic Modeling of CETP inhibitor-induced
hypertension (Roger Chang / Bernhard Palsson)
• Drug target identification in P. aeruginosa using an
associated metabolic network (Josh Lerman /
Bernhard Palsson)
• Detecting off-targets of NSC45208 an inhibitor of T.
brucei RNA editing ligase I (Jacob Durant /
Rommie Amaro / J. Andrew McCammon)
• Organic Anion Transporters (OATs) towards
determining substrate specificity (Sanjay Nigam)
Acknowledgements
Lei Xie
Li Xie
Jian Wang
Sarah Kinnings
http://funsite.sdsc.edu