CHAPTER ONE

Reflections on Medicinal
Chemistry at Merck, West Point
Paul S. Anderson
Lansdale, PA, USA

Much has changed since I arrived at Merck’s West Point laboratories in 1964
as an organic chemist interested in learning how to do drug discovery.
It quickly became apparent that this endeavor would require a prolonged
period of study and learning. Later, I realized that it would, in fact, require
life-long learning. The learning experience began immediately as others tau-
ght me the process by which therapeutic targets were selected and pursued
at that time. Medicinal chemistry was make a one compound at a time en-
deavor that mainly relied on data from experiments in whole animals for
SAR information. Safety pharmacology and drug metabolism studies were
reserved mostly for development candidates. Despite these limitations, the
pharmaceutical industry had been remarkably successful in the discovery
of new medicines. One such example was the thiazide diuretics.
The West Point laboratories had successfully discovered and developed
the thiazide diuretic chlorothiazide 1.1 At the time, whole animal pharma-
cology was central to this major accomplishment. The biological assays in
place tended to measure pharmacological activity without necessarily defin-
ing the mechanism by which the activity was achieved. It was clear, how-
ever, that physiological understanding of kidney function had played a
critical role in the course of events that led to the discovery of the thiazide
diuretics. The discovery road began with an earlier observation that the
antibacterial agent sulfanilamide produced alkaline diuresis in patients.
This effect was tracked to inhibition of carbonic anhydrase in the kidney.
Carbonic anhydrase inhibitors induced excretion of sodium bicarbonate
by the kidney. Inhibition of sodium reabsorption by exchange of sodium
for hydrogen ion in the distal part of the kidney therefore was thought to
account for the alkaline diuresis induced by these enzyme inhibitors. Karl
Beyer set forth the idea that a drug that worked in the proximal portion
of the kidney would be better tolerated by excreting sodium chloride instead
of sodium bicarbonate.2 His efforts to find such a molecule in collaboration
with medicinal chemists James Sprague and Frederick Novello led to the

#
Annual Reports in Medicinal Chemistry, Volume 47 2012 Elsevier Inc. 3
ISSN 0065-7743 All rights reserved.
http://dx.doi.org/10.1016/B978-0-12-396492-2.00001-1
4 Paul S. Anderson

discovery of dichlorfenamide 2, a potent carbonic anhydrase inhibitor with

increased chloride secretion relative to acetazolamide. Further SAR
work revealed that addition of an amino group to the aromatic ring of
bis-sulfonamides decreased carbonic anhydrase inhibition without loss of
chloride secretion. A subsequent molecule in which the amino group
and a sulphonamido group were incorporated into a ring became known
as chlorothiazide 1, a potent diuretic that did not inhibit carbonic anhydrase.
It was the first of many thiazide diuretics that found broad clinical use in
treating edema and hypertension.
O2
Cl H3C S S
Cl N Cl SO2NH2

NH
H2NO2S S H2NO2S SO2NH2
O2 NHC2H5

1 2 3

Interestingly, carbonic anhydrase reappeared as a target of interest at

Merck in the 1980s as Robert Smith and his colleagues sought to design
an inhibitor for topical use in the eye to treat glaucoma. Although many
carbonic anhydrase inhibitors were known, those previously approved for
clinical use as oral medications failed to lower intraocular pressure on topical
administration. We believed that a topically administered drug would avoid
the side effects associated with the use of the oral medications. Eventually,
we were successful in finding such an inhibitor. Success required gaining
knowledge about ocular transport issues from in vitro models and finding
the right animal model for testing these insights in vivo. It also required
the identification of a potent carbonic anhydrase inhibitor that was soluble
in water and had good transport properties as well as a long residence time in
the ciliary process tissue where the enzyme resides in the eye. At West Point,
Jack Baldwin accomplished this difficult task with the discovery of dorz-
olamide 3, the first topically effective carbonic anhydrase inhibitor for the
treatment of glaucoma.3 In the course of this work, we had learned that
the challenges of getting acceptable physical, metabolic, and transport prop-
erties in a drug candidate could be even greater than those associated with
optimization for potency and selectivity. One of the ancillary benefits of this
program was an early opportunity to explore the use of X-ray crystal struc-
tures of an inhibitor bound to an enzyme in drug discovery.4,5 This early
experience in the manipulation of physical properties and the use of
protein–ligand crystal structures in drug design would pay a dividend in
later work on the design of inhibitors for HIV protease.
Reflections on Medicinal Chemistry 5

In the 1970s, Merck had initiated a program to expand access to new

compounds being synthesized in academic chemistry departments by
offering to screen these materials for biological activity under a collabo-
rative agreement. One compound that proved to be interesting was 5,6,7,
8-tetrafluoro-1,4-dihydronaphthalene-1,4-imine, which unexpectedly
was found to have benzodiazepine-like behavioral activity as well as po-
tent antiseizure activity in rodents without the usual sedative side effects
of this well-known class of compounds. Pursuit of this interesting obser-
vation led to a more drug-like molecule dizocilpine 4, better known as
MK-801.6 Studies focused on the mechanism of action of MK-801
used (3H)MK-801 to identify specific high-affinity binding sites in rat
brain membranes that proved to be associated with NMDA receptors.
Neurophysiological studies were then used to establish that MK-801
was a noncompetitive antagonist of the NMDA receptor.7 While
MK-801 was not developed as a drug, it has served as a valuable, com-
monly used tool in neuropharmacology. The MK-801 work taught us a
great deal about the importance of mechanism of drug action studies in
drug discovery work.
While the MK-801 work was in progress, Dan Veber’s group at West
Point was pursuing several other opportunities for drug discovery based
on hypotheses that antagonists of certain peptide hormones might be useful
therapeutic agents. The peptide hormone CCK was selected as the initial
target based on the current knowledge of its physiology and pharmacology.
CCK-A receptors were believed to mediate pancreatic and biliary secretion
as well as gut motility, while CCK-B receptors were implicated in gastric
acid secretion and panic–anxiety attacks. The plan was to screen fermenta-
tion broths for molecules with affinity for CCK receptors using a radio
receptor assay in the hope of finding nonpeptidal compounds that would
prove to be antagonists. The screening effort identified asperlicin as a mod-
estly potent nonpeptidal CCK antagonist. Subsequent fragmentation and
reassembly analysis of the natural product sought to identify and optimize
the pharmacophore responsible for this activity. As the project proceeded,
Ben Evans and Mark Bock were successful in designing 5 and 6 that were
orally bioavailable, potent, and selective antagonists for CCK-A and -B
receptors, respectively.8,9 While these molecules demonstrated the
expected pharmacological consequences of selective receptor blockage,
neither became a drug because of side effect and efficacy issues. A similar
approach to the study of the peptide hormone oxytocin also gave rise to
new nonpeptidal antagonists that were not subsequently developed as
drugs. It became apparent that there are many targets for which a
6 Paul S. Anderson

preclinical hypothesis for a mechanism of drug action can be developed, but

few of these hypotheses actually translate to new medicines for a variety of
reasons. One that did achieve the desired objective was the West Point effort
to discover and develop an antagonist for the aIIbb3 integrin receptor,
which recognizes the RGD sequence present in many integrins. The
Merck effort led by Ruth Nutt and George Hartman produced both
peptide and nonpeptide antagonists. The nonpeptide antagonist was
developed as tirofiban 7,10 which was approved by the FDA in 1998 for
treatment of unstable angina. These experiences with MK-801 and the
search for peptide antagonists as new medicines taught us that target
selection is a critical component of the drug discovery process that
requires careful analysis before committing to a course of action. A clear
understanding of what will be required to translate the preclinical
hypothesis into a successful clinical outcome needs to be a key part of this
analysis.

H3C
O
H3C N H
N
N
H H
HN N O

4 5

H3C HO2C O O
O H
N S
H H N
N N H
H
N O

CH3
O

NH
6 7

Discovery of drugs to manage HIV infection was a major goal of the

pharmaceutical industry during the closing years of the twentieth century.
Once human immunodeficiency virus was identified as the etiological agent
of AIDS, understanding of molecular events in viral replication revealed the
virus-specific enzymes HIV protease, reverse transcriptase, and integrase to
Reflections on Medicinal Chemistry 7

be excellent targets for therapeutic intervention. Both established and early

stage pharmaceutical companies joined the effort to find drugs that worked
by these mechanisms. At Merck, our entry into the fray was spearheaded by
the synthesis of HIV protease as an initial source of the enzyme for inhibitor
assay development and protein crystallographic studies.11 Recognition of
HIV protease as an aspartic acid protease provided insight for inhibitor
design gleaned from earlier study on other aspartic acid proteases such as
renin. The Merck sample collection contained a number of peptides and
peptide-like molecules that had been designed to be renin inhibitors. These
molecules all had a core secondary alcohol component believed to be a
transition-state mimic for the cleavage site in substrates. Several of these
renin inhibitors were found to also be HIV protease inhibitors. While they
provided guidance for the SAR optimization work, they lacked potency in
cells and were not orally bioavailable. X-ray crystal structures of HIV
protease with and without early inhibitors bound to it facilitated the design
process, as did molecular modeling. These structures confirmed the impor-
tance of proper positioning of the secondary alcohol in the enzyme active
site and revealed other interesting interactions between inhibitors and the
flap region of the active site. With these insights in hand, the march toward
molecules with high potency in cell culture was fairly rapid, however, find-
ing molecules with acceptable ADMET properties proved to be challenging.
Very few molecules satisfied the ADMET criteria for development as an
orally active drug. Eventually, the West Point team led by Joel Huff and
Joe Vacca overcame these challenges with the discovery of indinavir 8.12
Nonnucleoside reverse transcriptase inhibitors also have become
important in the treatment of HIV infection. Screening of small molecule
sample collections for leads to this class of compounds was an important
step toward their discovery. Several companies including Johnson &
Johnson, Boehringer-Ingelheim, Merck, and Pharmacia Upjohn obtained
leads from their sample collections that evolved into clinical candidates.
The Merck lead evolved into the widely used drug efavirenz 9, which
was discovered by Steve Young.13 The Boehringer-Ingelheim lead prog-
ressed to nevirapine and the Pharmacia Upjohn compound became
delavirdine. These inhibitors bind to a flexible, allosteric site in the enzyme,
which may account for the structural diversity permitted by this target.
Crystal structures of inhibitors bound to this site were subsequently
obtained. Although this information did not contribute to the early work
on nonnucleoside inhibitors, it has enhanced our understanding of this
flexible binding site.
8 Paul S. Anderson

N OH OH
H F3C
N N
N Cl
O
O
O NH
N O
C(CH3)3 H
8 9
One of the most important contributions to medicine in the second half
of the twentieth century was the development of the statins. The conversion
of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) to mevalonic
acid by the enzyme HMG-CoA reductase is the rate-limiting step in cho-
lesterol biosynthesis. Thus this enzyme was an obvious target for the discov-
ery of drugs with utility in controlling cholesterol in the blood. Subsequent
understanding of the regulatory role that HMG-CoA reductase plays in LDL
receptor expression further enhanced the desire to find inhibitors with drug-
like properties. An enzyme assay suitable for use in screening fermentation
broths for inhibitory activity was developed and used to discover natural
product inhibitors such as compactin and lovastatin 10. Merck developed
and launched lovastatin as a product for controlling blood cholesterol level.
This successful product served as a template for further work to improve its
drug-like properties. It was noted that metabolic hydrolysis of the ester in
lovastatin gave rise to a metabolite with much less enzyme inhibitory activ-
ity. Increasing bulk by adding a methyl group alpha to the ester carbonyl in
lovastatin enhanced metabolic stability and more than doubled in vivo
potency. The new molecule called simvastatin 11 became a very successful
treatment for hypercholesterolemia.14 Both of these inhibitors are prodrugs
because the active species is the beta hydroxy acid formed on metabolic
hydrolysis of the lactone. It was assumed that this part of the inhibitor resem-
bled an intermediate in the reaction pathway of HMG-CoA reductase and
therefore was key to how the enzyme recognizes and binds to this class of
inhibitor. This assumption suggested that it might be possible to replace the
stereochemically complex decalin part of simvastatin with a less complex
hydrocarbon structure. Exploration of this hypothesis at Merck led to the
discovery of a family of biphenyl replacements for the decalin substructure
such as 12.15 A common feature of these and latter HMG-CoA reductase
inhibitors was the p-fluorophenyl entity thought at the time to be a replace-
ment for the hydrophobic ester present in simvastatin. Subsequent X-ray
crystal structures of enzyme-bound inhibitors validated this concept.16
Reflections on Medicinal Chemistry 9

The new biphenyl inhibitors were very potent but did not appear to have
additional preclinical advantages that merited clinical development. How-
ever, other companies pursued these structures to arrive at fluvastatin and
atorvastatin using a strategy that often is referred to as variation of a known
active compound. Over time, this has been one of the most productive
approaches to the discovery of new medicines, as one can readily see by
reviewing the discovery of beta blockers, histamine antagonists, ACE inhib-
itors, and many other majors classes of drugs.

HO O HO
CO2H
O OH
O
F
O
H
H3C R CH3 CH3

H3C
CH3
10 R = H
11 R = CH3 12

The above advances in health care offered useful teaching about the nature
of drug discovery. Studying the physiology of organ function can put one on a
productive track for drug discovery, as was the case with the thiazide
diuretics. Understanding mechanism of drug action also has great value, as
was the case with the integrin antagonists, statins, and HIV infection treat-
ments. It was clear that the hurdles for drug discovery are high. Discovering
a molecule that is selective for the desired mechanism of drug action and has
an acceptable ADMET profile for the intended route of administration is a
formidable challenge even with the tools that are available today. In most pro-
jects, very few molecules satisfy all of the necessary criteria.
New tools such as genomics, metabolomics, bioinformatics, high-
throughput screening and chemistry, fragment-based drug design, and
structure-based drug design have been incorporated into the drug discovery
process without fundamentally changing the paradigm or increasing produc-
tivity. This can be put into perspective by realizing that technology supports
drug discovery, but people discover drugs. Process makes you functional,
but process does not make you creative. Creativity is still a key factor in
the transformation of a drug discovery concept into a new medicine. To this
end, target selection is key, because most new target opportunities begin
10 Paul S. Anderson

with a preclinical hypothesis that lacks formal clinical validation. History tells
us that most of these hypotheses fail to achieve clinical validation. However,
it is likely that some of the new tools may eventually reduce failure rate by
helping drug discovery scientists to pick the best targets for the discovery of
new medicines. To this end, Sir David Jack17 has reminded us that successful
organizations engage all of their people in the selection of the best ideas for
drug discovery with recognition that these are likely to be the ones that are
“simple, practicable with available resources and novel enough to yield
medicines that are likely to be better than probable competitors in ways that
will be obvious to both doctors and their patients.” In a related statement,
George Merck addressed the challenge of balancing business interests and
the interest of patients by saying in a 1950 speech at the Medical College
of Virginia, “We try to remember that medicine is for the patient. . .it is
not for the profits. The profits follow, and if we have remembered that, they
have never failed to appear. The better we have remembered it, the larger
they have been.” It would be good for us to remember this as the drug
discovery enterprise moves forward in the twenty-first century. The phar-
maceutical industry has undergone changes in recent years such that building
shareholder value has come to dominate creation of value for patients and
doctors. Better balance is needed between business interests and the interests
of patients and their physicians in order to have a productive industry that
can better serve the healthcare needs of society in the twenty-first century.
A return to better balance will enable creative medicinal chemists to
continue to discover important new medicines.

REFERENCES
(1) Novello, F.C.; Sprague, J.M. J. Am. Chem. Soc. 1957, 79, 2028.
(2) Beyer, K.H. Trends Pharmacol. Sci. 1980, 1, 114.
(3) Sugrue, M.F.; Harris, A.; Adamsons, I. Drugs Today 1997, 33, 283.
(4) Baldwin, J.J.; Ponticello, G.S.; Anderson, P.S.; Christy, M.E.; Murcko, M.A.;
Randall, W.C.; Schwam, H.; Sugrue, M.F.; Springer, J.P.; Gautheron, P.; Grove, J.;
Mallorga, P.; Viader, M.P.; McKeever, B.M.; Navia, M.A. J. Med. Chem. 1989,
32, 2510.
(5) Smith, G.M.; Alexander, R.S.; Christianson, D.V.; McKeever, B.M.; Ponticello, G.S.;
Springer, J.P.; Randall, W.C.; Baldwin, J.J.; Habecker, C.N. Protein Sci. 1994, 3, 118.
(6) Thompson, W.J.; Anderson, P.S.; Britcher, S.F.; Lyle, T.A.; Thies, J.E.; Magill, C.A.;
Varga, S.L.; Schwering, J.E.Z.; Lyle, P.A.; Christy, M.E.; Evans, B.E.Z.; Colton, C.D.;
Holloway, M.K.; Springer, J.P.; Hirshfield, J.M.; Ball, R.G.; Amato, J.S.; Larsen, R.D.;
Wong, E.H.F.; Kemp, J.A.; Tricklebank, M.D.; Singh, L.; Oles, R.; Priestly, T.;
Marshall, G.M.; Knight, A.R.; Middlemiss, D.N.; Woodruff, G.N.; Iversen, L.L.
J. Med. Chem. 1990, 33, 789.
Reflections on Medicinal Chemistry 11

(7) Wong, E.H.F.; Kemp, J.A.; Priestley, T.; Knight, A.R.; Woodruff, G.N.; Iversen, L.L.
Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 7104.
(8) Evans, B.E.; Bock, M.G.; Rittle, K.E.; DiPardo, R.M.; Whitter, W.L.; Veber, D.F.;
Anderson, P.S.; Freidinger, R.M. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 4918.
(9) Bock, M.G.; DiParto, R.M.; Evans, B.E.; Rittle, K.E.; Whitter, W.L.; Veber, D.F.;
Anderson, P.S.; Freidinger, R.M. J. Med. Chem. 1989, 32, 13.
(10) Lynch, J.F.; Cook, J.J.; Sitko, G.R.; Holahan, M.A.; Ramjit, D.R.; Mellott, M.J.;
Stranieri, M.T.; Stabilito, I.I.; Zhang, G.; lynch, R.J.; Manno, P.D.;
Cjang, C.T.-C.; Egbertson, M.S.; Halczenko, W.; Duggan, M.E.; Laswell, W.L.;
Vassallo, L.M.; Shafer, J.A.; Anderson, P.S.; Friedman, P.A.; Hartman, G.D.;
Gould, R.J. JPET 1995, 272, 20.
(11) Nutt, R.F.; Brady, S.F.; Darke, P.L.; Ciccarone, T.M.; Colton, C.D.; Nutt, E.M.;
Rodkey, J.A.; Bennett, C.D.; Waxman, L.H.; Sigal, I.S.; Anderson, P.S.;
Veber, D.F. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 7129.
(12) Vacca, J.P.; Dorsey, B.D.; Schleif, W.A.; Levin, R.B.; McDaniel, S.L.; Darke, P.L.;
Zugay, J.; Quintero, J.C.; Blahy, O.M.; Roth, E.; Sardana, V.V.; Schlabach, A.J.;
Graham, P.I.; Condra, J.H.; Gotlib, L.; Holloway, M.K.; Lin, J.; Chen, I.-W.;
Vastag, K.; Ostovic, D.; Anderson, P.S.; Emini, E.A.; Huff, J.R. Proc. Natl. Acad.
Sci. U.S.A. 1994, 91, 4096.
(13) Young, S.D.; Britcher, S.F.; Tran, L.O.; Payne, L.S.; Lumma, W.C.; Lyle, T.A.;
Huff, J.R.; Anderson, P.S.; Olsen, D.B.; Carrol, S.S.; Pettibone, D.J.; O’Brien, J.A.;
Ball, R.G.; Balani, S.K.; Lin, J.A.; Chen, I.-W.; Schleif, W.A.; Sardana, V.V.;
Long, W.J.; Barnes, V.W. Antimicrob. Agents Chemother. 1995, 39, 2602.
(14) Hoffman, W.F.; Alberts, A.W.; Anderson, P.S.; Chen, J.S.; Smith, R.L.; Willard, A.K.
J. Med. Chem. 1986, 29, 849.
(15) Stokker, G.E.; Alberts, A.W.; Anderson, P.S.; Cragoe, E.J.; Deana, A.A.; Gilfillan, J.L.;
Hirshfield, J.; Holtz, W.J.; Hoffman, W.F.; Huff, J.W.; Lee, T.J.; Novello, F.C.;
Prugh, J.D.; Rooney, C.S.; Smith, R.L.; Willard, A.K. J. Med. Chem. 1986, 29, 170.
(16) Istvan, E.S.; Deisenhofer, J. Science 2001, 292, 1160.
(17) Jack, D. Creating the Right Environment for Drug Discovery. Quay Publishing: Lancaster,
1991.