Drug discovery lec 2 Flashcards
1
Q
What does pre-clinical pharmaceutical R&D do
A
- Optimisation of properties through changes in structure- turn weak agonist into potent agonists
- Improve activity and pharmacokinetics
- Pharma R&D chemists and biologists discover and improve weakly acting molecules turning them into optimised drugs with high activity and good pharmacokinetic properties
2
Q
Time-honoured approaches
A
- Natural products and synthetic chemistry
- Hard work + cleverness
- But mostly luck
- Modern pharma R&D is about efficiency and removing reliance on cleverness and luck
3
Q
Taxol
A
- Taxol or Paclitaxel was discovered by the U.S National cancer institute in 1967 isolating it from the Pacific yew tree, Taxus brevifolia
- It was developed commercially by Bristol-Myers Squibb
- Taxol is used to treat patients with lung, ovarian, breast, head and neck cancer and Kaposi sarcoma
- Paclitaxel stabilizes microtubules and inhibits cell division
4
Q
Natural products
A
- Structurally complex
- Often rigid
- Lots of chirality
- More 3-dimensional
- More functionality: HB donors; HB acceptors
- Natural products still very importanty part of drug discovery
- A quater of small molecule drugs are natural products or derived synthetically from natural products
5
Q
Synthetic approaches
A
6
Q
Screening
A
- High throughput screening (HTS)
- Simple robotic assay (single-point inhibition)
- 10,000+ compounds
- Poor S: N
- Random error
- High content screening (HCS)
- Complex manual assays (IC50s, Kids)
- Labour intensive
- Time-consuming
- Few compounds
- Slow but reliable
- Take the output from HTS and put it through HCS, so we can identify which hits are actual hits and are potentially viable for use as a drug
7
Q
Computer-Aided compound selection (CACS)
A
- All screening is time consuming, expensive, labour intensive and logistically problematic
- How can we reduce the search space, yet not miss anything, while making the process more efficient and reliable
- CACS- impacts compounds selected from internal and external compound collections; selection of fragments for fragment-based design; and the design of libraries
8
Q
Compound selection: the good
A
- Imbue our selected compounds with features common to known drugs
- E.g. few if any drugs are acyclic
- Imbue our selected compounds with features common to those drugs which bind a particular receptor class
- E.g. Privileged fragments
9
Q
Compound selection: The bad
A
- Heavy metal ions, such Hg or U
- Inorganic structures
- Organometallics
- Reactive groups
- Groups (long alkyl chains) not consistent with the properties of good drugs
- Promiscuous inhibitors
- I.e. molecules that regularly come up as hits
- Can be determined by substructure (e.g. chromone) or physical properties
10
Q
Reactive groups
A
11
Q
Other problematic chemical classes
A
12
Q
Promiscuous inhibitors
A
- Non-aggregate formers
- Fluconazole
- Ketoconazole
- Aggregate formers
- Clotrimazole
- Econazole
- Miconazole
- Sulconazole
13
Q
Compound selection: The Ugly
A
- Synthetically interactable molecules
- Too big
- Too complex
- Too many H bond donors and acceptors
- Too difficult to make
14
Q
A
- Possession of certain structural features can increase or decrease the chances that a molecule can be developed into an active and effective drug
- Others can preclude molecules from being reversible inhibitors, agonist or antagonist
15
Q
Lipinski’s Rule of 5
A
- Max of 5 H bond donors (Nitrogen or Oxygen with 1+ H atom)
- Max of 10 H bond acceptors (Nitrogen or Oxygen)
- MW <500
- Octonal-water partition co-efficient logP <5
16
Q
Selecting compounds
A
- Drug-like
- Remove reactive molecules
- Filter on properties
- Lipsinki’s rule of 5, ASA
- Predict drugness
- Lead-like
- More restrictive filtering
- Ligand like
- Similarity to known ligands
- Ligands of related receptors
- Privileged fragments
17
Q
Virtual Screening
A
- Replace expensive and time consuming
- Blind searches with computer-guided alternatives
- 2D virtual screening } knowledge of known ligands
- Pharmacophore screening} knowledge of known ligands
- Structure based automated protein docking} knowledge of target
18
Q
Chemical similarity
A
19
Q
Automated protein docking
A
- Finding, filtering, screening, scoring and ranking compound for drug discovery
