| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biochemistry and Molecular Biology, University of Maryland School of Medicine, Baltimore, Maryland 21201, USA
1Correspondence: Department of Biochemistry and Molecular Biology, University of Maryland School of Medicine, 108 N. Greene St., Baltimore, MD 21201, USA. E-mail: rthompso{at}umaryland.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGlossaryREFERENCES |
|---|
Key Words: cholesterol • research funding • statins • HMG CoA reductase • mevastatin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGlossaryREFERENCES |
|---|
Millions of Americans now worry about their cholesterol, especially how
much they consume in their diet and how much is present in their blood.
However, it is only relatively recently that the importance of
cholesterol to the development of heart disease was identified. Even
now the molecular mechanisms of most cardiovascular disease remain to
be fully described. In fact, cholesterol had been isolated in the 1800s
and was known to be an important constituent of animal cell membranes;
by the 1930s, it was established as the precursor to hormones like
testosterone as well as the bile salts. Cholesterol is a lipid, a fatty
molecule that dissolves in liquids like olive oil but not in water.
Cholesterol was suspected to be important in cardiovascular disease
almost 90 years ago, when it became known that the arteriosclerotic
‘plaques’, which actually clog the arteries, contained substantial
cholesterol (1)
and that feeding some animals a
high-cholesterol diet accelerated the appearance of cardiovascular
disease (2)
. Despite these early indications, there was no
proof that cholesterol was more than a bystander in the process of
arteriosclerosis.
The Framingham heart study established that cholesterol levels in the
blood were important in heart disease. This was a very large
prospective study beginning in 1948 that examined 5209 residents of
Framingham, Massachusetts, looking for risk factors for atherosclerosis
leading to heart attack and stroke. The study was funded primarily by
the National Institutes of Health (NIH). The study consisted of
frequent, thorough physical examinations and extensive blood testing of
the volunteers over periods extending to decades. A key risk factor
identified in the study was a high level of serum cholesterol
(3)
. High levels of other serum lipids (fatty compounds)
such as triglycerides generally were not correlated with a higher risk
for heart attack. Naturally, this suggested that reducing serum
cholesterol might reduce atherosclerosis and thus the risk of death
from heart attack, which, in fact, is the case (4)
.
Large, long-term studies such as the Framingham study are expensive:
the Framingham study has cost more than $27 million just since 1983
(5)
. The cost is perhaps understandable when one considers
that in the early 1970s, more than 10,000 people participating in the
study were receiving medical exams that were much more frequent and
thorough than ordinary medical exams. The cost of those exams and
laboratory tests, together with the expense of collecting and analyzing
the data, can run into thousands of dollars per patient. Studies like
Framingham are also a gamble because there is no guarantee the study
will find an answer. It was far from certain that a risk factor for
heart disease would be found or that, if found, it would be useful for
treatment or prevention. Ultimately, nearly a dozen risk factors were
identified during the course of the study: high serum cholesterol is
merely the best known and most therapeutically important. The high
cost, multiyear duration, and uncertain payoff necessarily make such
studies seem unattractive and unrealistic for the private sector. Only
the federal government has been willing to provide adequate funding for
the duration of the Framingham study, as well as hundreds of other
research studies and clinical trials aimed at the reduction of
cardiovascular disease. The federal investment has paid off handsomely;
it identified a therapy that has proved effective for millions of
people—saving lives as well as billions of dollars in health care
costs.
One potential way of lowering serum cholesterol is by limiting intake in the diet. Most meat and dairy foods provide amounts of cholesterol adequate for dietary needs. The body needs about 1 gram of cholesterol every day to replace that lost primarily through the digestive tract. Controlling dietary intake has limited success in lowering cholesterol not only because of the difficulty some patients have in staying on a diet, but also because the body can biosynthesize all the cholesterol it needs. Vegetarians who do not consume animal tissue obtain essentially no cholesterol from their diet, yet they do not lack for the cholesterol they require because their own bodies can make it.
The biosynthesis of cholesterol was elucidated over a period of more
than 20 years beginning in the late 1940s principally by groups led by
K. Bloch of Harvard University, F. Lynen of the Max Planck Institute in
Munich, J. W. Cornforth of the National Institute for Medical
Research in London, and G. Popják of UCLA, with much of the work
in the U.S. supported by the National Heart Institute, the progenitor
of the current National Heart, Lung, and Blood Institute of the NIH
(6)
. Bloch, Lynen, and Cornforth later won Nobel prizes
for their work. Again, this extensive series of experiments represented
a large investment in essentially pure science, since it was not known
at the time that reducing cholesterol would be therapeutically
important. Indeed, even with the more powerful biochemical tools now
available, elucidating such a biosynthetic pathway remains a
substantial scientific effort.
The ‘biosynthetic pathway’ for cholesterol is a series of chemical
reactions in the cell that convert and assemble the precursor molecules
into the final product, cholesterol. The assembly is carried out by a
series of enzymes, each of which catalyzes one step. For instance, the
first step (of more than 20) is the coupling of two precursor
molecules, which each contain two carbon atoms to make one containing
four carbon atoms; cholesterol has 27 carbon atoms in all. The
biosynthesis of cholesterol is complex compared with that of other
small molecules and requires a lot of energy. In part because of the
energetic cost, the biosynthesis in the cell (most cholesterol
biosynthesis in the body occurs in the liver) is under tight control
exerted early in the synthetic process, before much energy is consumed.
The tight control is analogous to a thermostat: an automatic system
that turns off a furnace when the proper temperature is reached,
maintaining the temperature at a more or less constant level. The step
in cholesterol biosynthesis where control is exerted is an enzyme
called hydroxymethyl glutaryl coenzyme A reductase (HMG CoA reductase,
for short), which catalyzes the reduction of HMG-CoA (the
‘substrate’) to mevalonic acid (the ‘product’). Figure 1
. The step catalyzed by HMG CoA reductase is essentially irreversible in
that it is difficult for the body to convert mevalonic acid back to HMG
CoA. However, HMG-CoA can be readily recycled to a precursor, which is
used in synthesizing other fatty molecules. This step in the
biosynthesis is referred to as the ‘committed step’. Thus, the search
began for potential drugs that might stop the biosynthesis process,
preferably at the HMG CoA reductase step. The therapeutic importance is
twofold: 1) the reversibility of the prior steps means that
no potentially toxic precursors would build up if the biosynthesis were
halted at that stage; 2) the body itself controls
biosynthesis at this step, so any potential intervention would mimic
the body’s own function and thus be less risky.
|
An effective way to halt a pathway like cholesterol biosynthesis is to
add a chemical blocker, called an inhibitor, that will interfere with
an enzyme that catalyzes a step in the pathway and prevent its
function. Many drugs, including aspirin, act as enzyme inhibitors. The
way to tell whether a particular chemical is an inhibitor is to see how
it affects the activity of the enzyme: more precisely, one measures how
much product the enzyme makes in the presence and absence of the
inhibitor. The potency of an inhibitor is determined by how little is
necessary to achieve the same degree of inhibition. Thus, finding
inhibitors of HMG CoA reductase (or any enzyme) requires a ready source
of the enzyme to test against and a simple assay for measuring how fast
the enzyme converts HMG CoA into mevalonic acid in the presence of the
inhibitor. Again, the substantial investment NIH and the National
Science Foundation (NSF) made in fundamental studies of enzyme
purification, assay development, kinetic analysis, and enzyme function
beginning as early as the 1950s has provided the basis of technology
and understanding that made it feasible to search for such inhibitors.
Thus, in 1960 Durr and Rudney devised a straightforward method of
isolating the HMG CoA reductase from baker’s yeast as well as a simple
assay for the enzyme activity, which made it possible for anyone to
test a particular chemical to see whether it was an inhibitor
(7)
. Often, inhibitors are molecules very similar to the
natural substrate(s) of an enzyme, but ones the enzyme cannot convert
into a product; the penicillins are examples of such inhibitors.
Beginning in the early 1970s, a number of inhibitors were found, but
few were potent or specific or could be administered to animals and
induce a reduction in cholesterol levels (8
, 9)
.
At about this time, Akira Ando and colleagues at Sankyo Pharmaceuticals
in Japan began testing the fermentation broth of molds and other
microorganisms for HMG CoA reductase inhibitors. Many simple organisms
and plants produce chemicals that are toxic or noxious to discourage
animals from eating them; examples of this strategy include poisonous
mushrooms and hot peppers. A sufficiently potent HMG CoA reductase
inhibitor might be quite toxic to noncarnivorous animals that obtain no
cholesterol in their diet. Fermentation technology is well developed in
Japan and is a potentially rich source of pharmaceuticals. Ando’s
group isolated about a teaspoonful of a compound they called ML-236b
(see Fig. 1
) from more than 800 gallons of fermentation broth of a mold
species related to the one that produces penicillin. ML-236b was a
potent inhibitor of HMG CoA reductase, able to shut down the enzyme
activity when present at microgram per liter levels (10)
.
Moreover, when administered to rats it induced a significant reduction
in their serum cholesterol (11)
. Brown and colleagues in
Britain isolated and characterized the same compound at about the same
time as an antifungal compound and named it compactin; they did not
suspect that it was an HMG CoA reductase inhibitor, but the name stuck
(12)
. The similarity of the portion of the inhibitors
highlighted in red to HMG CoA and mevalonic acid can easily be seen
(Fig. 1)
; many enzyme inhibitors are molecules that are structurally
similar to the substrate or product of the reaction.
Unlike the earlier compounds, compactin showed real potential as a drug
for reducing cholesterol levels due to its potency and lack of overt
toxicity in mammals. As a result, several groups began to isolate and
synthesize related compounds, including mevinolin, lovastatin, and
pravastatin (see below). Several of these compounds ultimately entered
wide use as anticholesterol drugs. In the case of a molecule like
compactin, the amount isolated from the fermentation broth was enough
to permit compactin’s structure to be determined (12)
but
not to test it as extensively as is necessary. Knowing the structure,
however, it is usually feasible to synthesize the molecule in the test
tube in amounts large enough to test. Much of the chemical technology
developed for synthesizing these compounds in the laboratory and
determining their structure has also been developed with the support of
NIH and NSF. As a result, a large ‘chemical toolbox’ is available
that permits essentially any molecule known to be made synthetically.
Similarly, determination of the structure of small molecules like
compactin has become straightforward due to technology enhancements in
X-ray crystallography and infrared and NMR spectroscopy, developed
largely with federal grant support. Although these techniques have
important practical applications, their development was aimed primarily
at answering research questions. Perhaps most important to the whole
enterprise are the skilled biomedical scientists and biochemists who do
the work, a large fraction of whom are supported by NIH during their
predoctoral and postdoctoral training in this country. Before support
from NIH became available, Ph.D.’s in the biomedical sciences were a
rarity. Even though it may be argued whether too few or too many
biomedical scientists are currently being trained in the U.S., there is
no question that the American pharmaceutical and biotechnology
industries (which lead the world) could not exist, let alone thrive,
without those thousands of trained people.
To be therapeutically useful, a drug must do more than simply inhibit
an enzyme or exert some other therapeutic effect. It must be otherwise
nontoxic, or at least there should be a large difference between levels
in the bloodstream that are therapeutically useful and those that are
toxic. Beyond this, there should be a minimum of side effects
associated with the drug. Ideally, the drug should be absorbed rapidly
enough and eliminated from the body slowly enough to make only one or a
few doses a day necessary. All these issues come into play after a
potentially useful compound is identified. Many times it is necessary
to modify the structure of the drug to achieve the desired
pharmacological attributes. Typically, medicinal chemists will make or
isolate many compounds similar to the original, in search of a better
molecule. In the case of compactin, scientists at Merck isolated a
compound that differs only slightly from compactin (Fig. 1)
, but is
more effective (13)
. This compound was mevinolin, sold as
Mevacor. Subsequently, Merck also introduced lovastatin (Zocor).
Zocor is now among the most widely prescribed drug in
existence, accounting for sales in 1998 of 4.7 billion dollars
a year worldwide, or 
40% of the market (14)
. Other
important ‘statins’ include Lipitor from the Parke-Davis division of
Warner-Lambert (now Pfizer), Pravachol from Sankyo/Bristol-Myers
Squibb, and Lescol from Novartis. The production, packaging,
quality assurance, and marketing of these drugs employ thousands of
people here and abroad; Merck alone employs 57,000 people. Export of
these drugs also improves the U.S. balance of trade.
The HMG-CoA reductase inhibitors shown in Fig. 1
are all related to
compactin; as a result, they share a couple of drawbacks. First,
compactin and its relatives are natural products, meaning they are
isolated from an organism in nature. Although significant improvements
can be made to the rate of production from organisms like molds and
fungi, the small amounts of compactin isolated by Ando indicate the
enormous amount of effort necessary to scale up fermentation to
commercially useful quantities. Like many natural products, compactin
can exist in several subtly different configurations, called
stereoisomers. Often only a single stereoisomer (of the 256 possible
for compactin) is made in nature and is active, with the other possible
stereoisomers being inactive. As a result, chemical synthesis of
molecules like compactin can be extremely difficult and prohibitively
expensive for production purposes.
Recognizing these problems, scientists began to consider whether other,
simpler molecules might not also be HMG CoA inhibitors. It is apparent
from the red highlighting in Fig. 1
that only a modest portion of
compactin actually has a structure similar to mevalonic acid, and
perhaps the complex two-ring structure of compactin is not really
required. As structural information on enzymes such as HMG CoA
reductase became available in the 1980s, scientists working with the
support of NIH and industry began to use these structures to understand
how current inhibitors worked and help design new inhibitors. In
particular, scientists developed something called ‘quantitative
structure activity relationships’ (mercifully abbreviated as QSARs) to
judge the potential efficacy of inhibitors. At about this time,
powerful computer graphics programs were developed at the University of
California at San Diego and elsewhere that permitted the 3-dimensional
structures of enzymes and inhibitors to be displayed and manipulated in
real time. These developments were important because it was no longer
necessary to synthesize a series of related molecules to find the best
inhibitor; one could essentially create models of the proposed
inhibitor molecules in the computer and bring them together
with the structure of the enzyme to predict whether a particular
molecule would inhibit the enzyme. An image of a portion of human HMG
CoA reductase is depicted in Fig. 2
(15)
. These software programs were an outgrowth not only
of biomedical research, but also of computer science research funded by
several agencies, including NSF and the Department of Defense. This
approach was much faster and cheaper than actually finding or making
and testing many new inhibitors. In the mid-1980s Sandoz developed
fluvastatin and the Parke-Davis division of Warner Lambert developed
atorvastatin. Both these compounds are HMG CoA reductase inhibitors
that differ much more from compactin than Mevacor (see Fig. 3
). The advantage of these compounds (from a production and therefore
profitability standpoint) is they are much simpler: each has only four
possible stereoisomers and, consequently, is dramatically cheaper to
produce.
|
|
More recently, the entire process of creating potential medicines (called leads) and testing them for efficacy has become highly automated, which has led to revolutionary increases in efficiency and cost effectiveness. For a century, drug companies had identified potential drug candidates by collecting thousands of field specimens of plants/animals and testing them (or their constituents) for efficacy in treating disease. The National Cancer Institute of the NIH also invested heavily in a search for antitumor drugs in the same way, identifying many of the anticancer drugs currently in use. After they had identified an active compound (as in the case of compactin), chemists synthesized piecemeal a series of similar chemicals whose structures differ slightly from the parent and could be tested for improved efficacy, reduced side effects, improved uptake and persistence, and other desirable properties. This piecemeal synthesis and testing are still very expensive and inefficient, even when guided by known enzyme structures and QSARs.
Beginning in the early 1990s, several chemists with NIH support began
to devise means of systematically synthesizing scores or hundreds of
analogs essentially simultaneously (16
, 17)
; this ability
is termed ‘combinatorial chemistry’. At the same time, other workers
began to devise means of testing very large numbers (into the millions)
of compounds for efficacy (17)
. For instance, instead of
synthesizing one or a few analogs of compactin and testing them one
after the other for their ability to inhibit HMG CoA reductase, a
‘library’ of thousands of compounds could be synthesized and tested
within months, essentially automatically. An important advantage of the
combinatorial approach is that potential new drugs can be found without
knowing much about their target in the cell or what the structure of an
inhibitor should be like. This approach has revolutionized the
pharmaceutical development business, making it much easier and cheaper
to identify potential drug candidates. For instance, during the 60 years from 1934 to 1994, Merck’s scientists synthesized, purified, and
screened for therapeutic efficacy 250,000 different chemicals at
huge expense. Using combinatorial synthesis and high throughput
screening techniques, Merck scientists in the 4 ensuing years
synthesized and tested 4.5 million compounds (14)
. This
approach was developed principally in the U.S. with NIH support, and
the U.S. leads the world in the development and implementation of this
technology. Indeed, providing the instrumentation, chemical reagents,
and software to implement the combinatorial approach is now a
multibillion dollar business worldwide (18)
, one that is
dominated by American firms. It is unclear whether the combinatorial
approach is currently being used to find HMG CoA reductase inhibitors,
but it clearly is now the principal route to new medicines of all
kinds, particularly inhibitors of enzymes newly identified during the
sequencing of the human genome.
Thus, at the present time we have a group of medicines that reduce
serum cholesterol levels by up to 40%, which substantially reduces the
risk of heart attack and other consequences of atherosclerosis such as
stroke and kidney failure (4)
. These drugs are generally
well tolerated by those taking them and typically need to be taken only
once a day. As a result, they are used by millions all over the world,
providing substantial benefits in combating disease and enhancing the
quality of life. The production, distribution, and marketing of these
drugs comprise a multibillion dollar business internationally,
employing thousands of persons in this country and contributing
favorably to America’s foreign trade balance. It should be evident
that it would have been difficult or impossible to develop and produce
these drugs in the absence of federally sponsored research, especially
without the knowledge base and substantial number of trained scientists
it provides. This has made our research enterprise an important
national asset, one we should encourage and support for the
future.
|
Glossary |
|---|
|
TOPABSTRACTINTRODUCTIONGlossaryREFERENCES |
|---|
Cholesterol: A steroid molecule that plays three main roles in the body: it is an essential component of cell membranes; it is the precursor from which bile salts are synthesized in the body; and it is the precursor for steroids in the body, such as the hormones testosterone and progesterone.
Enzyme: A protein molecule that catalyzes the transformation of one molecule (the substrate) into another (the product). An enzyme may be one of a series that act in succession to synthesize a molecule like cholesterol from simple precursors. Enzymes are said to catalyze the conversion of molecules called substrate(s) into product(s). There are more than 30,000 different enzymes in the human body.
Inhibitor: A molecule that inhibits the functioning of an enzyme.
Lipid: A fatty molecule used in the body. Lipids are oily or ‘hydrophobic’ molecules that separate from water. Examples are cholesterol and other steroids, phospholipids, and triglycerides.
Steroid: A class of lipid with a characteristic fused ring structure; examples include cholesterol, testosterone, progesterone, and the bile salts.
Triglycerides: A class of lipid molecules that are the main constituent of ‘fat’ in fatty tissue, vegetable oil, and butter. Triglycerides consist of three fatty acids attached to a glycerol backbone; the structure of the fatty acids determines whether the triglyceride is liquid or solid at room temperature.
|
ACKNOWLEDGMENTS |
|---|
; Wolfgang
Mergner, Kim Collins, Mark Jenkins, and Richard Engel for critical
reading of the manuscript; and Howard Garrison and Tamara Zemlo of
FASEB for critical reading of the manuscript and many helpful
suggestions. The author is grateful for support from the National
Institutes of Health, Office of Naval Research, and National Science
Foundation. Received for publication January 2, 2001. Accepted for publication March 23, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGlossaryREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Hamerman Osteoporosis and atherosclerosis: biological linkages and the emergence of dual-purpose therapies QJM, July 1, 2005; 98(7): 467 - 484. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
