Nota de Estudos
Study Reminders
Text Version

Fundamental Principles of Drug Development Process

Set your study reminders

We will email you at these times to remind you to study.
  • Monday




























Welcome back. In the last session, we have given a brief introduction to the drug
discovery processes and we have seen that there are several stages in a drug discovery
process. Now, even a drug can fail and there is a high failure rate. And that is because
whatever you see in the test tube that is not reflected in the in-vivo studies. And whatever
you see in the preclinical studies may not be reflected in the human subjects, so that is
how the drugs can fail.
Now, suppose an intended drug molecule crosses the initial stages which includes target

identification, target validation followed by drug in-silico studies. And then from in-
silico studies, you get some molecules and then you synthesize that.

And then you study the actual interaction by a bioassay technique and then you try to
find out what are called hits; that means some preliminary effect you need and then you
tinker around the structure. And finally you get some lead and the lead has to be
optimized through what is called structure activity relationship. First you make a large
library of compounds, and then you try to find out a structure activity relationship.
That means, suppose there is a phenyl group at some point and then you replace the
phenyl group with electron donating aromatic ring and an electron withdrawing aromatic
ring and you see what is happening to the biological effect. Sometimes the electron
donor may be helpful, sometimes the electron withdrawing may be helpful, that is what
will give you the structure activity relationship. That means how the activity of the
molecule depends on the structure at different segments in the molecule, so that will help
you to identify a lead.
And then the lead has to be taken into the preclinical trials. And preclinical trials are
basically done on animals. And in animals, you study the toxicological effect. Some
people say that it is better to first study the toxicological effect. Because even if you
observe good ADME properties good PD properties (pharmacodynamics), by then you

have already invested money on these and then you see the toxicological effect and you
find that it is very toxic, then the whole money that you have put is gone.
So, some medicinal chemists say that better study the toxicological property first and
then you proceed to study and invest money on other type of effects.
(Refer Slide Time: 03:48)

I already said that all drugs are ultimately toxic, but that they are toxic at a certain
concentration. At lower concentration, as it is a drug, we can assure that it is having
some good activity. But as you slowly increase the concentration, you might see that
toxic effect developing. And in many occasions, it may also cross the lethal dose that
means it can kill the subject.
Now, here in this graph, I have 3-sigmoidal curves. The first is basically when you are
giving a certain lower concentration of a drug and you are measuring the response.
Suppose if it is in the animal stage, so you have different animals, 50 animals or 100
animals and you are giving this drug to these animals at different concentration. And so
you will see the effect, this is the average effect on the animals. This effect is what we
call response.
Now, so basically at lower concentration, there will be very less effect. So, it actually
picks up in this region. And the middle point of this curve corresponds to the ED50, ED50

is what is called effective dose, effective dose. The effective dose is somewhere in
between this point and that point, that means, here it is the 50 percent.
That means, if you have say 100 animals and you are slowly increasing the concentration
of the drug, if you see the desired effect in in 50 animals then that means, the
concentration needed to have the effect shown by 50 percent of the subjects that is what
is called effective dose.
If it is animal, then it is effective dose in animals. If it is in the clinical level, then it is
effective dose on human. So, there is the term called effective dose or this is also called
the therapeutic dose of a drug. Now, as you increase the concentration, you start seeing
the toxicity. The toxicity will be definitely shown at a higher concentration, but it also
follows a sigmoidal curve.
And there is also another terminology or another parameter which is called TD that is
called the toxic dose. The TD is called toxic dose, ED is called the effective dose and LD
is a lethal dose that means if you cross the toxic dose then you have a lethal does. So, the
subject will be killed. Now, we have ED50, as I told you this effective dose, that means,
the dose which gives you the actual effect that you wanted is between here and there.
And here it is the 50 percent of the population is showing the effective response. So,
according to that, we have ED50, we have TD50 and we have LD50. ED50 is the dose at
which 50 percent of the population therapeutically responds because we know that the
aspirin causes acidity to a person, but may not cause acidity to another person, may have
very well an analgesic effect.
So, you have 100 persons and then I give this aspirin to several of these people and then
if 50 percent of the people are saying that yes I am having this good effect (analgesic
effect), so that dose what is called the effective dose. Similarly, you have TD50 TD50 is a
dose that at which 50 percent of the population experiences toxicity and then you have
lethal dose.
Dose at which 50 percent of the population dies is the LD50. If you cross this then all the
subjects will die; if you cross this point, that means, this concentration, but in between
these two concentrations, in the middle of that sigmoidal curve that gives where 50
percent of the population dies. So, these three things are measured usually animals are,

animals are sacrificed. So, you can go to the LD50, that means, you can see what is the
dose required in which the 50 percent of the animals die.
So, for animals you can measure ED50, and you can measure LD50. Now, this ratio is very
important this LD50 by ED50, if the ratio is very high, that means, you have a very safe
margin, you have a safe margin. That means, suppose are talking about this type of
graph; suppose this is your effective dose curve and if the if your lethal dose curve is also
very similar, so there is not much therapeutic safety level that you have.
If this goes here, then you have a better window that means, there is a less chance of
crossing the lethal dose in that. So, this ratio that means, LD50 by ED50 is what is the
therapeutic index, initially I told when I begin this. I told you about this.
(Refer Slide Time: 10:37)

So, therapeutic index is, there are two ways of doing this for human you cannot kill the
human. So, for human you have you just measure the toxic effect that when you slowly
increase the dose level. So, some point you will see toxic level appearing. So, you see
that what is the dose required to have 50 percent of the toxic level; in 50 percent of the
population that you have taken and you are studying. So, TD50 by ED50 is called
therapeutic index.
So, also LD50 by ED50, is usually done for animals because where you can scarifies the
animals. But for human it is usually the TD50 by ED50, but any way this is what is called

the therapeutic index, and the larger this therapeutic index the safer is the drug that is
most important.
(Refer Slide Time: 11:44)

So, suppose this is the scale and this is the amount at which drug is effective and this is
the amount the drug becomes toxic, that means, you have a safety margin of this. This is
what is basically the therapeutic index but it is expressed as a ratio of TD50, that means,
the toxic dose TD50 divided by ED50 or LD50 divided by ED50.
(Refer Slide Time: 12:19)

Today, also there is another, the margin of safety you can even be very sure that a person
does not cross or he is not given a dose which is the toxic dose. So, to do that, there is
something called margin of safety (MOS) which is calculated by TD01, that means, 1
percent of the population is having a toxic dose and ED99, that means, 99 percent of the
population is having the effective showing that responds effective response.
So, that gives you a better even a safer margin, because you do not want even 1 percent
of the population to be affected by the toxic dose. So, rather than going up to 50 percent
of the population, basically you increase the effective dose for the 99 of the population
and you reduce the toxic dose for only 1 percent of the population, so that gives what is
called the margin of safety.
(Refer Slide Time: 13:38)

In clinical trials, basically there are phases, phase 1 is safety, tolerability, and there are
different targets for this phase studies rather than efficacy. Firstly, as I said, they try
tolerability; that means, what is the tolerance limit and then the bioavailability, that
means, the ADME properties that whether the drug is absorbed and then shows effect.
Usually healthy volunteers are used for the trial period, because you are only checking
the safety level, you are only checking the tolerability, and you are only checking the
bioavailability. You are not focusing on the disease yet. So, healthy volunteers are

In Johns Hopkins University, I know people are really paid for acting as healthy
volunteers to participate as trial participants. But this is safe, because they are given very
minimum amount of the intended drug.
(Refer Slide Time: 15:01)

So, once these are through, that it has got a very good safety margin, then you go to
phase two clinical trials. In phase 2 clinical trials, you actually take the patients now,
several hundred suffering from the condition that the drug is intended to target. And then
you try to you get information about the efficacy. And also you estimate the safety in a
larger population rather than the healthy subjects earlier in the phase 1 clinical trial.
So, basically phase 1 is on healthy volunteers, so it measures the safety, it actually
concentrates on determining the toxicological effects and the ADME effects that means,
the absorption distribution metabolism and excretion those type of properties, but no PD
that means, we are not targeting the efficacy of the molecule. Efficacy of the molecule is
done on patients, but in a limited way not thousands of patients. And then try to find out
the minimum amount required to get the efficacy, the dose that is required to cure the

(Refer Slide Time: 16:25)

If satisfactory results from phase 2 trials are obtained, the drug will enter phase 3 clinical
trials. These are larger versions of the previous trials that means, now it is a broad
approach. You take the patients and you give the drugs which cross the phase 2 trials,
and then you actually go to large number of patients and give these drugs, and then see
the risk and the benefit analysis.
Risk and benefit that means, the toxic effect versus the benefit as I said every drug has a
toxic effect. But the beneficial effect is that you have to compare the cost of this risk and
benefit; which is better; which is more important to you. Like some people are taking
anti-hypertensive drugs and many of these anti-hypertensive drugs in the long run is
going to affect the kidney.
But question is that whether you want to live 20 years from now on by taking drugs
which will reduce the blood pressure or you take the risk that you can die anytime if you
do not take the drug. So, this benefit and risk you have to calculate and then finally, the
regulatory authority approves the drug.
(Refer Slide Time: 18:00)

There is something which is called phase 4 clinical trials which is done after this trial
phase 3. Once the drug is available in the market, still many pharmaceutical companies
study a wider variety of people what they had their patients, and see that in the next 5
years, 10 years, what is the patient’s report, how are they feeling, whether there are
reporting something else which was over looked at that time; that is called the phase 4
clinical trials.
In many cases, some drugs have been approved, it entered the market and later it has to
be withdrawn because they are having some other effects which may be detrimental.
(Refer Slide Time: 18:49)

So, this is the drug developmental process: validation, assay development, high
throughput screening, hit to lead and then lead optimization, pre-clinical and then clinical
drug development.
(Refer Slide Time: 19:04)

There are some empirical rules when you design certain molecules for in-silico
screening; it is not that you write randomly whatever would like to do. You have to
follow certain things like time and again I am saying the drugs are low molecular weight
compounds. But what is the optimum molecular weight? That you should not cross or
what is the threshold like lethal dose which you should not cross to have a lethal dose.
Just like molecular weight, there is some restriction on the lipophilicity, which means,
how soluble it is in a lipophilic solvent like organic solvents.
And then for interacting with the binding site, these interactions are electrostatics or
could be hydrogen bonding; hydrogen bonding plays a very dominant role in this
interactions. So, how many hydrogen bonds you need in the molecule to be there.
Lipinski made some rule which are called Lipinski rule of five. Rule of five does not
mean that there are five rules. Rule of five means that whatever numbers are shown here,
they are either five or multiples of five.
So, Lipinski’s rule of five says that whenever you select a molecule, you are thinking
that it may be a possible drug, first check whether the molecular weight is greater than
500? If so, then discard that; it should be preferably below 500; though it is not a

sacrosanct that it has to be less than 500 always, there may be some molecules which are
having 550 or 600, but this is a general guideline that a molecule is drug like when it has
molecular weight of less than 500.
It has a logP value of less than 5. What is logP value? It is the logarithm of the partition
coefficient between octanol and water, why octanol? Because octanol mimics the
membrane through which the drugs crosses. Membrane is actually lipophilic this is large
chains, this fatty acyl glycerol chains with a polar head group, but a majority portion in a
membrane is lipophilic.
So, what happens? The drug has to cross this lipophilic membrane and then go inside, so
that means, it should have some balanced lipophilicity or hydrophobicity. So, here this
logP is the lipophilicity that means, the logarithm of partition coefficient between octanol
and water. And octanol was found to resemble the biological membranes.
And then it should have less than or equal to 5 hydrogen bond donors, that means, the
sum of hydrogen bonds donors which are usually OH and NH. So, sum of NH and OH in
the molecule should not cross 5, ok, it could be 4, but the threshold value is 5.
And then the 10 hydrogen bond acceptors; that means, what are hydrogen bond
acceptors? Nitrogen lone pair in free nitrogen, not NH and oxygen lone pair. So, oxygen
is the hydrogen bond acceptor, OH is the donor, NH is the donor and N, a tertiary
nitrogen, is an acceptor. So, this is what is called Lipinski rule.
So, now, all pharmaceutical companies initially screen the molecule based on the
Lipinski rule and reject those molecules which do not satisfy these type of parameters.
Remember this is only true for oral drugs, because we are talking about logP value; we
are talking about this molecular weight, we are talking about this hydrogen bond, all
these things. This is for oral drugs only.
And majority of the drugs we have are orally taken, only many of the life-saving drugs
are usually given intravenously because of their poor bioavailability and absorption.

(Refer Slide Time: 24:00)

Now, let us go into some of this. Before that I think I should tell you about some of the
references. There is a medicinal chemistry book that real name is Medicinal Chemistry:
An Introduction, which is by Gareth Thomas. Then Introduction to Medicinal Chemistry,
just the introduction to medicinal chemistry by G. L. Patrick that is a good book and
finally, R. Silverman’s book on Organic Chemistry of Drug Design, organic chemistry of
drug design. These are very good books on the topic that I will be covering.
(Refer Slide Time: 24:42)

Once the target is known and validated, then you have to ultimately molecules, you have
to synthesize molecules that is where the organic chemists are required. So, they have to
make the molecules. And one thing is very clear, greater the number of molecules one
makes, more is the chance of striking a hit. Suppose you have made five molecules only.
These five molecules may not produce anything because you have a very small pool of
molecules. So, larger the number of molecules, larger the molecular library, greater are
the chances of finding a hit. If you are looking for a book and if the library has only 100
books, the chance that you will find that book is very less.
If the library has 1 million books, then the chances that the book you are talking about is
there, is much more. Similarly, in drug discovery process, the more number of molecules
that you make, the greater is the chance of success, there is no doubt about that. Because
as I said whatever you do, computational studies can assist you in making the library a
little bit smaller, but ultimately you have to make large number of molecules.
Now, initially what happened, organic chemistry including synthetic chemistry
developed in a rapid stride in the 40s, 50s, and 60s. Starting from Robert Robinson, and
then it went to Woodward and then many of the stalwarts in organic chemistry; they have
now established the art of synthesizing molecules. So, organic molecules can be made.
But how many molecules you can make per day? Usually you do one reaction or two
reactions per day. So, one person can make one or two molecules per day.
What happened after the 1980s, there has been an enormous development in biology.
And earlier what happened was that suppose some biologist has isolated an enzyme. And
the amount of enzyme he or she has is very little. And then he or she once wants to check
what is the interaction of different molecules with this new enzyme.
Now, the chemists supply one compound and the actual system was such that it takes
about few days to ultimately come up to a conclusion whether the molecule has really
interacted, how much effective was the interactions. So, it is taking long time. But after
the 1980s something came which is called high-throughput screening, that means, you
can screen several hundred molecules within a very short time, and that means, now
gone are those days where the organic chemist synthesizes one molecule and gives it to
the biologist and 1 month later get the result that nothing has happened.

Then you make another molecule. So, things were very slow, but then after the high
throughput screening that every day you can screen large number of molecules, so that
immediately you can put pressure on the organic chemist to come up with methods
which can produce large molecules per day, so that can be supplied to the biologists.
Now these issues have given rise what is called combinatorial chemistry.
Combinatorial chemistry was developed to produce large number of compounds
required for a throughput screening. It allows the simultaneous synthesis of a large
number of possible compounds that could be formed from a number of building blocks.
The products of such process are known as combinatorial library. Libraries may be
collection of individual compounds or mixture of compounds.
This is very interesting. Libraries may be collection of individual compounds that means,
suppose you have 10 test tubes, each test tube is numbered as 1, 2, 3, 4, 5, 6, 7, 8, 9, and
10. Now, suppose I take CH3COCl in all these test tubes suppose (the same acid
chloride) and then I add different amines, suppose this is methylamine I add.
And here I add ethylamine CH3CH2NH2, then propyl amine and so on. Since I am doing
in parallel fashion, so I will get the amide. In this 1st case, it will be amide or
methylamine and acetyl chloride, in this 2nd case ethyl amide, in this 3rd case, propyl
amide or propionyl amide. So, you will get all these different amides, if you do the
reaction in this fashion.
So, here basically each test tube has only collection of individual compounds, which
means each test tube contains only one pure compound. And the other is mixture of
compounds; that means, basically what you are doing, you are adding both methylamine
and ethylamine (something like that) into each test tube. So, you get two compounds;
one is the amide of the methyl another is an amide of the ethyl.
Now, you do the testing. Now, chances of getting more success if you have more number
of molecules in a particular test tube, earlier case I said that 10 molecules in 10 test
tubes, but there are techniques by which you can make 10 molecules in a particular test
tube, so that you have hundred 10 multiplied with 10, that means, the 100 molecules in
these 10 test tubes and then you test through high throughput screening.

But the question is suppose this test tube shows the activity; that means, there are 10
molecules out of which may be one is showing activity. Now, you can go back and then
individually synthesize those 10 molecules and then see which one is active. But this is
much better option because out of 100, there is more chance of getting a hit; and then
you know which test tube is giving some activity and you know what are the ingredients
that you have added and you try to separate them and then separately study those 10. So,
it is much better option.
(Refer Slide Time: 32:28)

(Refer Slide Time: 32:35)

Now, this combinatorial chemistry was developed by using solid phase synthesis,
because ultimately it was realized that this solution phase chemistry you have to do
chromatography to purify the molecules.
Whereas, if the reaction is done on a solid phase, as we have seen in solid phase peptide
synthesis, your purification is very easy; you just take the solid beads and then wash
them after each reaction and then you detach it from the solid phase which ultimately
gives the compounds. So, all these combinatorial chemistry are usually done on solid
surfaces, like utilizing Merrifield’s resin.
Now, I will give you an example of this combinatorial synthesis. Suppose you take
RCOCl. Now, different types of R which are exemplified by or denoted by A1, A2, A3,
that means, suppose A1 is methyl, that means, acetyl chloride, this is propionyl chloride
and this is say butyryl chloride. And then you are adding an amino acid NH2 with R’ and
then amino acid ester.
Now, for this ester, this R can be varied; you can take different amino acids, you can
have 20 amino acids here depending on this R. These are your variables R R’ and R’’.
So, you can vary this R’ and you can vary these R’’ also. And so suppose I say that B1.
So, basically if R1 is something say methyl; that means alanine and this is COOR’

this R’
is suppose methyl, so that is exemplified by B1.
So, similarly varying this R’

and R’’ I can have another entity called B2, and then I can
have a third entity called B3. So, when I take A1 in one test tube I add this B1 that means,
this is methyl and that is methyl. So, I will get A1B1. And in another test tube, I will get
A1B2 and then A1B3 and if A2 is the starting point, you will get A2B1 you are adding the
same components B1, B2, B3, A2B2, A2B3.
And then with A3, you will get A3B1, A3B2 and A3B3. I am not actually proceeding any
further, but you see that the same reaction is operated on three different acid chlorides
and you get 9 compounds.

(Refer Slide Time: 35:56)

We know what the advantage of the solid support is.
(Refer Slide Time: 36:02)

These are some of the resin beads in the Merrifield resin; we know that it is a polystyrene
which is chloromethylated. Then there is polyethylene glycol chain, which means, you
have this type of moiety. The beauty of this type of moiety is that it swells in water. So,
you have the resin bead, then polyethylene glycol chain and then a reactive X.
As you put it in aqua solution, this swelling is important because that allows percolation
of the other entity that is reacting with it. X is NH2, OH, SH all these reactive

functionalities. There are other resins also; this is called one resin for carboxylic acid.
This is the THP, THP ether for alcohols and then you have this chloroformate type of
resin for amines.
These resins which are attached to a chloroformate resin, which are attached to a
tetrahydropyranyl moiety. This is the tetrahydropyranyl linker. And this is the one resin
which is the benzylic alcohol. These are the resin that you will be using depending on
you are starting entities.
(Refer Slide Time: 37:39)

Now, this is how this combinatorial chemistry is done. Basically you take a plate. And
there are wells in the plate; wells in the plate that means grooves. Now, suppose you are
doing this type of reaction that this resin bead is attached to the where the ester linkage to
NBoc amino acid with the R1
and R2
as shown here.

This is very similar to the peptide chemistry that we have done. And this amine is a free
amine which can react with an isocyanate and that will give you a substituted urea. And
it is known that if it is heated with HCl, then a reaction that takes place is this, attacks the
carbonyl and the resin is released, so that makes a five membered ring that is called a
Now, hydantoins are privileged skeleton for drug development because it satisfies The
Lipinski’s rule of five.

(Refer Slide Time: 39:30)

See these are my wells, all these grooves. So, I say that this is A, B, C, D, E, F, G, H on
this row and on the column I said 1, 2, 3, 4 up to 12. So, what I do, I add the first
component, the first component is this bead with the NHBoc. So, I add the bead with
different amino acids with 8 different amino acids.
So, X1, X2, X3, X4, X5, X6, X7, and X8. So, these are resin beads which are attached to
this X1. I added the beads which are having different amino acid attachments. On this
side, I add the same bead with same amino acid attachment.
So, different beads are attached to different amino acids. And in this direction, that
means, the column I have, the same bead attached to the same amino acid, but these are
different as in the rows we have different ones because these are X2s, this is X1 rows,
X3, X4 like that. On this side, it is X1, X2, X3, X4, and X5.
Now, you deprotect the amino acid Boc and then you add the next amino acid and
couple. You have 12 grooves on this side on the in the column you have to select 12
amino acids and then couple with. So, Y1, now earlier your X1 X2, that means, different
beads are having different amino acids here and on this side beads are having the same
amino acid.
Now, what you do, you add the second amino acid, the same amino acid on this side on
the row. So, you add the same amino on this row. So, what you will get X1Y1, X2Y1,

and X3Y1. So, Y1 will be attached to all the beads via the X. And in this you add, so
basically what you are doing in the second one you are adding only Y2. So, you get
X1Y2, X2Y2 these. Remember these are different.
So, then you add the Y3. So, in this case if you add all these, you get a library of
compounds. How many compounds? Usually there are commercial plates available,
usually having 96 plates. So, 96 compounds you can prepare in one go.
(Refer Slide Time: 42:56)

Another way you can do is this is very good technique that you have a glass plate.
Suppose, you have a glass plate and the glass plates are made of silica. So, using the
silica OH groups, you can hook up some organic species ending up with a NH2; glass
plates are available today that has lot of NH2s attached to the glass surface, not directly
attached, it is attached through the that OH and then an alkyl chain and that ends up with
the NH2.
This NH2 can be used for further reaction and then that means, the plate is the solid
phase, it has got this NH2, then you are adding a compound which reacts with the NH2.
So, the compound will now stick to the NH2. There is one medicinal chemist who
developed this technique. What he did was that on a glass plate with NH2, you react with
a molecule which is called N-veratryl oxycarbonyl.

So, these group is what is called the nitro veratryl oxycarbonyl. The beauty of this is that
first of all it is a chloroformate. So, this is going to react. So, you form the urethane or
the carbamate. And then as you shine UV, this is the protecting group which is labile
under UV. So, it cleaves. The NH2 is again liberated.
So, this is a group which is called nitro veratryl oxy carbonyl; there is another group
which is only nitro and a benzyl. The similar group: ortho nitro benzyl oxycarbonyl, also
cleavable under light, but this is a better one because the wave length of the light that is
used for removal of nitro veratryl oxy is around 365 nm. This is the photo labile group.
So, you can again re generate your NH2.
(Refer Slide Time: 46:10)

Now, let us see how you can utilize this technique. So, you take a glass plate like this,
and see ultimate objective is to make library of compounds by a simple technique and
then also a simple assay and you can check the bio activity. So, suppose this is a glass
plate. So, first you cover one side of the glass plate with a opaque material, so that light
cannot go through and fall on the glass surface.
You have made four quadrants and then two of these are covered by a by a opaque s
screen which covers this part. Now, this X is actually the NH2 attached to veratryl oxy,
this NH2 attached to veratryl group. So, all are attached to NVOC.

So, now, you have covered this part and shine light. So, what will happen? This will