W26 Choose a target and find hit/lead compounds I and II Flashcards
Choosing a disease/ drug target
What are the things to consider? (6)
- Biomacromolecules involved (proteins, DNA/RNA, etc…)
- Target specificity/selectivity between species: penicillin, antiviral drugs (viral
DNA polymerase) - Target specificity/selectivity within the body (i.e. selectivity among various
isoenzymes, etc…) - Targeting specific organs/tissues (i.e. ᵯ 1 in the heart and ᵯ 2 in the lungs, etc..)
- Multi-target drugs: combination therapy (i.e. cancer, HIV) and multi-target
direct ligand (promiscuous ligand or dirty drugs- one drug hits more than one target) - Target Validation: disease association, cell-based models, protein interactions,
etc
Identifying a Bioassay:
What are the In vitro and In vivo methods?
(What happens?)
In vitro
* Specific tissues, cells, enzymes
* Use bacteria and yeast to produce enzymes (i.e.
HIV protease). IC50, competitive/non-
competitive* Isolated tissues or cells expressing a
receptor
* Intracellular and extracellular events
* PK properties (i.e. Caco-2 cells, microsomes,
etc…)
In vivo-
* Induce a clinical condition in the animal
* Transgenic animals
* Slow and expensive
* Sometimes results are invalid
* Variability according to the species
In vitro tests:
What is High-throughput screening (HTS)?
Automated test of large number of
compounds (several thousands) against a large number of targets (30-50); efficient to hit identification; false-positive hits (promiscuous inhibitors and PAINS-Pan-assay interference compounds)
In vitro tests:
What is Screening by NMR?
Detect whether a compound binds to a protein target; screen a mixture of compounds; 1000 small molecules a day; detect weak binding; no false-positive
1) NMR spectrum of the drug is taken
2)Protein is added and the spectrum is re-run (protein signals are not detected)
3)Drug not binding: its NMR spectrum will still be detected
4)Drug binding: no NMR spectrum will be detected
In vitro tests:
What is Isothermal titration calorimetry (ITC)
Determine the thermodynamic proper-
ties of binding between a drug and its protein target— the binding affinity and enthalpy change;
What is the difference between hit and lead compounds?
Hit= Compound with desired activity on a given screening assay, with low cytotoxicity
Lead= Compound highly active on the target and selective, effective in the disease model, amenable to synthetic modifications, drug-like properties (ADMET)
Finding a Hit compound
I. Screening of natural products
- Plants Source: morphine, cocaine, taxol,
etc. - Microorganisms: bacteria, fungi (antimicrobial agents-cephalosporins)
- Marine sources: coral, sponges, fishes, marine microorganisms
- Animal sources, Venoms and toxins
Complicated structures (chiral centres, strange bonds): extraction
Fleming’s original plate
studying staphylococci colonies, left for
holidays.
A mould (fungus) had destroyed the surrounding colonies…
Finding a Hit compound
II. Screening synthetic compound libraries
- Compounds or synthetic intermediate which have been previously synthesised
Finding a Hit compound
III. Existing Drugs
- ‘Me too and “me better’ drugs: use established drugs as hit compounds in
order to design a drug that gives them a foothold in the same market area.
Modify the structure sufficiently such that it avoids patent restrictions,
retains activity, and, ideally, has improved therapeutic properties - Selective optimisation of side activities (SOSA): enhance the desired side
effect and to eliminate the major biological activity of existing drugs - Repurposing: screening existing drugs, compounds that are either in clinical
use or have reached late-stage clinical trials against new disease/targets
Advantages= Cheaper and Faster
Finding a Hit compound
IV. Starting from the natural ligand or modulator
- Natural ligands for receptors: used as hit to design agonists (i.e. adrenaline
and noradrenaline); design of antagonists (i.e. histamine); - Natural substrate for enzymes: used as hit to design inhibitors (i.e. natural
substrate HIV protease -development of the first HIV protease inhibitor) - Enzyme products as hit compounds: used as hit to design inhibitors (i.e.
carboxypeptidase inhibitors) - Natural modulators as hit compounds: receptors and enzymes are under
allosteric control. The natural or endogenous chemicals that exert this
control (modulators) could also serve as hit compounds
Finding a Hit compound
V. Serendipity
- Hit compounds found by chance
- Research to improve a drug can have unexpected and beneficial spin offs (i.e.
Propanolol and Practolol) - Research projects carried out in a totally different field (i.e tolbutamide)
Finding a Hit compound
VI. Computer-aided design (CADD)
- Study the target 3D structure using a computer
- “Rise of PDB”: X-ray crystallography, NMR spectroscopy, cryo-EM
PDB= Protein Data Bank
Finding a Hit compound-using CADD
Examination of the target structure allows to rationally design compounds that can bind to it
Easier if there are co-crystallised ligands we know where the
active/binding site is (or we can use software for binding site prediction)
E.g. Rational design of HIV protease inhibitors:
Active as a homodimer,
essential for the viral
replication: produces the
active form of the viral non-
structural proteins
Ligand = peptide-base inhibitor
The residues forming the active/binding site can be analysed to
identify key interactions (more difficult if we do not have a
ligand bound) to design new ligands/inhibitors
Finding a Hit compound-using CADD
Virtual Screening
Instead of testing large number of compounds in the lab
(HTS), we can rely on computer simulations which can predict if a compound
is good or not, e.g. to bind a given target
Structure-based virtual screening: molecular docking
Given a target site, the software takes each virtual
compound and explore its possible conformations (docking
poses) within the target site it identifies those ones with
the best predicted binding
The software scores and ranks each generated
pose according to a mathematical evaluation
(scoring function) of the predicted free energy
change upon binding (docking score):
ligand-based virtual screening: shape similarity
Virtual compound libraries can be screened for their shape-similarity with known active molecules. All the possible conformations of the screened molecules are explored and compared for their shape and functional group matching with a query molecule
Ligand-based virtual screening are usually associated with higher (actual) hit rates
Finding a Hit compound
VII. Fragment-based hit discovery
Rather than screening large, high-affinity molecules, small molecular
fragments are screened against a given target. Due to their size, they will
bind weakly Once multiple active fragments are identified (mM-high μM IC50), the crystal
structures of their complexes with the target are resolved, and used to link
different fragments together= optimised high-affinity molecule
By linking the fragments together, we usually obtain bigger molecules with
very good affinity for the given target: good hit compounds for further
optimisation
Fragments used in fragment-based drug discovery usually obey “rule of 3”:
- Relative molecular mass <300 (and >150)
- Number of H-bond donating groups ≤ 3
- Number of H-bond accepting groups ≤ 3
- ClogP ≤ 3 (NB: calculated!)
- Number of rotatable bonds ≤ 3
Big advantage: it saves a lot of efforts related to the synthesis (and the
screening) of large molecules only those ones very likely to be active are
considered
NB: Software for in silico fragment-based drug design available
Hit to Lead process
Lead= compound highly active on the target, effective in the disease model, amenable to synthetic modifications, drug-like properties
- Improve Pharmacokinetic properties: range of factors which affect a
compound’s ability to reach its target in living systems. PK properties and
life-time - Improve Pharmacodynamic properties: ability of a compound to
selectively bind to a target and produce a desired pharmacological
effect. Low side effects, non-toxic. - Easily synthesised & chemically stable
Pharmacodynamics and pharmacokinetics should have equal priority in
influencing drug design strategies and determining which analogues are
synthesised
What is Lipinski’s Rule of five?
Rule of thumb, formulated by Chris Lipinski of Pfizer in 1997, which provides
predictors of drug-likeness for successful ORAL drugs:*
Relative molecular mass
<50
* Number of H-bond donating groups (HBD) ≤ 5 (i.e. OH, NH,
etc.)
* Number of H-bond accepting groups ≤ 10 (C=O, NH, COOR,
CONHR, etc.)
* LogP < 5 (log of partition coefficient between octanol and water: <0
favours water, >0 favours octanol; Drugs need to be lipophilic to pass
membranes and distribute to tissues, but not too lipophilic and not too
hydrophilic)
Works best for synthetic compounds, which occasionally break one of the points, but rarely two. Several
natural drugs break at least one of the points