Medicinal chemistry and biophysics
Lecture 1 Introduction
21/09/2020 Matthew Groves
Medicinal chemistry allows us to develop new medicines for existing and new diseases. The majority
of these medicines are small molecules. There is a recent trend towards biologics. These normally are
antibodies which are relatively easy to create, because you simply need the diseased target and then
raise competent antibodies, so all the work is done in the cell. However, biologics are very expensive
to create, store and ship. Whereas the small molecules are chemical entities which are synthesisable
in a lab and purifiable quite simply. Still, 80% of the new medicines are small molecules.
Medicines are limited in the number and types of atoms. The 3D structure of the medicines are very
important for how they interact with e.g. the body, the target.
Biophysics is the use of light, sound or particle emission (waves) to study a biological sample. A wave
has a dynamic path with crests and troughs. The distance between two crests or two troughs is called
a wavelength. The type of wavelength defines what we can look at. In case of life cell images
(microscopy), the information content has to be the same as the wavelength. So you can get away
with visible light which is compatible for cellular objects (membranes, organelles). When we want to
look in more detail, the wavelength needs to decrease so e.g. X-ray.
For all clinically used drugs that are covered in the lectures, the indication must be learned. This is
important for pharmaceutical knowledge (e.g. aspirin – inflammation).
Know the chemical structures of amino acids and the first two rows in the periodic table!!
When looking at aspirin and paracetamol, you see that they are small. This because it is easier for
small molecules to cross membranes and thereby reach the target. If they are too big they will not
cross the membrane. If they are too small, they will not be able to make sufficient interactions with
the protein target to be selective between different protein targets. C, O, H and N are the core
elements of most drugs since they are the biologically relevant atoms. The way that these atoms are
connected together will give a 3D structure that determines the biological activity of the drug. So
although aspirin and paracetamol are used for the same indication, due to their different structure
they will interact with fundamentally different targets.
Ethanol is the smallest available drug and it has its biological effect because it interacts with many
different proteins.
The less specific your drug is to the target site, the more side effects will occur. To get to the target
site, you first have to pass a series of membranes. So orally taken drugs must be soluble enough to
get into the gut. However, if they are too soluble, they cannot pass the GI tract membrane anymore.
Once the membrane is passed (right balance between solubility and lipophilicity), the drug is in the
blood stream. Then the drug has to survive metabolism, degradation and tissue deposition before it
reaches the membrane of its actual target site.
1
,The environment outside the membrane and within the membrane is water driven. Water molecules
carry a charge. So drugs are soluble because they can possess charges that can compensate with the
water molecule charges. So polar drugs (highly charged) are quite happy outside a membrane. If they
are too charged the drugs will simply not get into the membrane. The exterior of the lipid bilayer is
composed of polar headgroups. So the polar drugs can bind to the membrane or stay in solution.
Inside the membrane, it is highly uncharged. Amphiphilic drugs have a polar and a hydrophobic
nature. So they will insert themselves in the membrane, keeping one site to the solvent and one site
in the lipid membrane. They may flip across the membrane, but also these drugs are quite happy in
this position. Non-polar drugs are unhappy in the watery solution, so they will embed themselves in
the membrane and they will stay there. So these types of drugs are not quite good. There are ways
through this. There are protein pores that allow selective uptake of molecules across the membrane.
These transport proteins are natural proteins which make active transport possible. However, the
majority of drugs still passes the membrane via passive transport. Passive diffusion can be estimated
by the log P value which is defined by the physical/chemical properties of the drug. log P says
something about lipophilicity. So a balance between charged and non-polar sections is necessary so
that the drug is happy to get into the membrane but also is happy to leave again.
Log P is a measure of lipophilicity. It is the log of the maximum concentration of a drug in octanol
divided by the maximum concentration of a drug in un-ionized, uncharged water. Successful drugs
have a log P of about 5 or slightly lower. This indicates that successful drugs have large contributions
of hydrophobic/non-charged groups.
Lipinski’s rule of five:
1. Molecular mass less than 500 → because otherwise the membrane cannot be passed.
2. Log P less than 5
3. Less than ten hydrogen bond acceptors (-O-, -N-, etc.)
4. Less than five hydrogen bond donors (NH, OH, etc.) → atoms in an unionized state have a
protonated oxygen or nitrogen.
3 and 4 together mean that if you would have more acceptors and donors, the charge would go up
leading to increased solubility and a lower log P which is unwanted. However some charges are
needed to find selectivity. So a competent drug requires good solubility in both water and
membranes in order to have a good absorption. This is driven by the Lipinski’s rule of five.
Lipid membranes contain a polar headgroup and a lipid tail. These are phospholipids. There is a
connecting moiety (glycerol) that links the charged group to the long apolar tails which can consist of
unsaturated and saturated fatty acids. These tails exclude water from interactions with each other to
allow membranes to form.
Paul Ehrlich introduced the idea that the biological effect of almost all compounds was due to its
binding to a target. So bodies do not act if they are not bound. So the drug will do nothing unless it
binds to the target. This has been refined significantly resulting in the Lock and Key principle. So
2
,there are molecular entities with a specific shape that allows them to choose the correct target. If
the Key (substrate) fits in the Lock (enzyme), biological action will be exerted. Nowadays, we moved
to the concept of induced fit. So this is not only that the key fits the lock, but also that the process of
fitting the lock to the key actually changes both things. So the process of binding is dynamic. Once
the substrate is bound to the active site of the enzyme, there is initial binding resulting in a complex.
Then the protein is allowed to go through a process of movement to make the fit even more precise.
This will provide better energetics. Then the enzyme will form a chemical reaction on the substrate,
providing two products which can then release from the enzyme so that the active site can be
occupied again.
Thermodynamics → things exist in different energy states. And nature drives things towards lower
energy states. It describes the final equilibrium state. In case of medicinal chemistry and drug design,
we hope that the equilibrium state is the bound state where the molecule is bound to the target
protein. The parameter for this is K. E.g. when you hold a cup of coffee high in the air, you have to
put energy in it or otherwise the cup of coffee will fall.
Kinetics → the rate of the process. The parameter for this is k. E.g. the rate at which the coffee cup
falls.
In this case, the complex AB will only happen when that is
the lower energy state. If A and B separate have a lower
energy state, the equilibrium will drive the system to this
state where the substrate is not bound to the protein.
The more product we have (more C and D), so the more
bound state, the higher K is going to be. The lower the K,
the more unbound substrate there is. So since we would
like to see a lot of bound state, we prefer a high K.
Within this, there are still different kinds of interactions:
- Covalent interactions → drug molecules that bind to the target and then form a
physical/chemical bond between the drug and the target. So in this case there is no
dissociation. So there is no equilibrium. If you give this enough time, all of the system will
become bound. So there is no K, there only is a kinetic rate k. These kind of drugs are difficult
to generate. They often bind to the wrong protein target and then they will not become
unbound. This results in significant side effects.
- Non-covalent interactions → this allows a potential target protein to bind a drug molecule. If
it is the wrong one, it can unbind again so that the drug can look for a new binding partner.
Here there is an association rate and a dissociation rate. So there is an equilibrium.
Keq is then an association (binding) constant that can be
referred to as Kass.
For Kdiss (= [A]+[B]/[AB]) you want K to be as small as possible.
3
, Each new interaction provides a change in the Gibb’s free energy. Gibb’s free energy (G) is the
energy required to build the system from nothing. So you have your reactants which have a certain
energy. Then there is an activation barrier. Then you reach the products which have a lower energy
than the reactants. This means that the equilibrium lies to the right. The rate at which this reaction
happens is driven by how much of the activation
barrier is there. It can also be that the energy state of
the reactants is lower than that of the products. This
means that in order to get to the products, you have to
put in energy. If you do not put in energy, the reaction
will not happen and the equilibrium will lie to the left.
ΔG is the change in energy between the reactants and
the products. If ΔG is negative, so you have moved to a
lower energy state, the reaction will happen naturally.
The change in Gibb’s free energy (ΔG) has two components: enthalpy and entropy. These are the
driving forces of a reaction.
Enthalpy (ΔH) is a way of saying heat. Ice soaks up heat from the environment. This means that
heat/energy is absorbed, meaning that it takes energy from the environment. So the overall energy
that is going into it is positive (ΔH > 0). The reaction is then endothermic. The opposite case is a fire,
where heat/energy is released. Then ΔH < 0 resulting in an exothermic reaction. Ice melts naturally
and fire burns naturally, so the overall ΔG must be negative.
Entropy (ΔS) is a statistical measure of the ordering of the
system. For example, you have a protein and a ligand which
are not bound to each other. Both are bound to a water
molecule. These water molecules are coordinated so are not
able to explore the rest of the universe. They are in a fixed
position. However, when the protein binds to the ligand, the
two water molecules will come free and be able to explore the
universe. This creates a significant increase in the disordering
of the universe around it → entropy has increased. This is also
the case in hydrophobic interactions between the protein and
the ligand. When you have a hydrophobic atom, the water
molecule cannot get close to it since the water carries a charge that does not want to get close to the
hydrophobic surface. When the protein and the ligand undergo a hydrophobic interaction, water is
allowed to explore a much larger region of the universe → entropy has increased.
There is a link between the dissociation/association constant and the Gibb’s free energy. The
dissociation constant is related to the exponential of the ΔG divided by the universal gas constant
and the absolute temperature.
4