4. The drug discovery process Flashcards
What are the two schools of thought for drug discovery and what do these methods entail?
Which strategy works best? Why?
Target-based discovery:
- Identify the target
- Identify hits
- Lead optimization
- Clinical trials
Phenotype-based discovery
- Identify the disease phenotype
- Identify hits
- Lead optimization
- Clinical trials
You either know what target you want to hit or you know what disease phenotype you want to treat
Few first-in-class drugs found through target-based screening as there is a poor understanding of disease complexity
However, higher innovation rate in follower drugs (slighly modified first in class drugs) in target-based approach. This is because optimization is possible once a molecular mechanism of action is established.
What makes good drug targets? What doesn’t? Why?
Enzymes, receptors, and channels make good targets
Protein:protein interactions don’t
MMOAs are not sufficient for biological activity
Two components are required for efficient drug action:
Mass-action binding (concentration dependent)
Biochemical/structural rearrangements that shift the target away from mass-action binding (locking the target in an inactive state)
Describe the 4 steps involved in structure-based drug design
Target selection:
A trivial first step, however finding the proper target is far from trivial
Can approach this biochemically (basic research that elucidate biochemical mechanisms of disease, typically analysis of cell lysate) or genetically (screen for mutations in patients suffering from disease, find commonality)
Target can be validated by CRISPR/Cas9 Knockout/knockin
Pharmacophore identification/screening
A hit is a compound that has the desired biological/chemical properties in an assay
Many ways to identify hits, two broad categories
High throughput screening (brute force), tests a large category of possibly unrelated drugs
Structure-based approaches, rational design/FBDD/Virtual screening, require strcuture of target to build something custom for that target (will see later)
Pharmacophore identification/optimization
The pharmacophore is the part of the drug that causes a biological response
If you discovered a hit from the previous step, you need to find what part of it actually exerts its effects
Test a bunch of molecules that have slight variations from this hit. Measure activity in each one
Use a training/testing set to build an algorithm that lets you predict activity molecules
AT THE SAME TIME start developing ADMET and DMPK (no use in creating a drug that affects a receptor perfectly if it can’t even reach it)
Lead optimization
Keep modifying and improving drug
You want a nanomolar Kd (residence time in seconds)
What mechanism of action would be most resilient when designing a viral protease inhibitor?
A compound that interacts with the protein backbone (rather than the labile amino acids) would be less susceptible to resistance via mutation
Describe the 3 basic steps of mass spectrometry
Why ionize your molecule?
An ionization source creates charged molecules, these molecules enter a mass analyzer that separates the ions by mass and charge, and the detector senses and quantifies the ions
The detector only detects ions, separates ions by “mass to charge ratio” (m/z)
How do you read a mass spectrum? How do you identify charge states and isotopomers?
What is the monoisotopic mass? is this often the highest intensity peak?
Different peaks represent different ions (difference varies by approximately 1-2 protons or 2 AMU)
Can multiply m/z by intensity to get mass
Zooming into one peak will show you a new spectrum called the isotopic envelope (contains isotopomers)
Each isotopomer within this zommed in envelope vary by 1 Da/x where x is the charge of the overall envelope (+15 in the slide)
To get the mass of the molecule, multiply the charge by the most intense peak, but subtract m/z * mass of one proton
Monoisotopic mass: the mass of the molecule with all carbons being C12, all hydrogens being H1, etc. Statistically very unlikely and is often not the case for large molecules.
Average mass arises from most frequent isotopomer
How is resolution calculated in mass spectrometry?
What is the difference between resolution and accuracy?
Full width at half maximum
Take a peak, find the maximum amplitude, and calculate the width at amplitude/2
Resolution: clustering
Accuracy: how right you are
Name the 4 methods of ionisation learned in class and describe the steps involved in the 2 that were discussed in detail
EI, FAB, MALDI*, ESI**
*Matrix assisted Laser desorption Ionization
Analyte/matrix blasted with a monochromatic laser
Matrix distributes the energy and dissipates it to analyte chich gets dispersed in gas phase
Matrices are usually organic, aromatic acids
*NOT quantitative, just tells you there is a protein
**Electrospray ionization
Droplet containing ions dries from heated gas causes repulsion of single ions
Gives a mass range, but sensitive to salt (can be coupled to LC)
What are the three analyzers for mass spectrometry discussed in class?
1: Time of flight (TOF)
Ions are accelerated with voltage difference
Move along a flight tube for a long distance, then detected
M/Z is determined from time of flight, which is a function of distance and acceleration potential
2: Quadrupole:
Ions move through 4 charged electrodes with opposite charge
Charges are inverted frequently, the ions oscillate between them
Certain frequencies cause a certain m/z ratio to be in resonance, while others will be filtered out (good for filtering m/z range)
3: Orbitrap
Ions trapped in ovoid, spin around and cause a current, Fourier transform allows m/z to be calculated from current
Describe the function of fragmentation and tandem MS
Fragmentation allows the specific filtration of ions
Different mass analyzers can be used in tandem (“MS/MS”), and using fragmentation between analyzers can allow specificity
On the fly (can be switched on and off) and are used to determine structure of the parent molecule
