The Nature of Scientific Discovery
The curious history of reaction and resistance to the homocysteine theory of arteriosclerosis demands explanation. Why should a quarter-century of medical science elapse and a whole new generation of scientists mature before the significance of an important new theory of disease is recognized? Is there some flaw or internal inconsistency in the homocysteine theory that has prevented its timely acceptance and implementation? No disease has been investigated more thoroughly and painstakingly in the 20th century than arteriosclerosis. Although research on cancer has attracted the attention of generations of medical scientists, the sheer volume of effort devoted to understand arteriosclerosis exceeds even that devoted to understanding cancer. In view of tremendous interest by medical scientists, why has a significant new theory of the underlying cause of arteriosclerosis been widely ignored and criticized by others for so long?
An outstanding scientist who at tempted to answer questions about the nature of scientific discovery in another context was James Bryant Conant, chemist, educator, scientific administrator, public servant and president of Harvard University from 1933 until 1953. In his capacity as chemist and historian of science, Conant investigated and wrote an influential analysis of the reaction of early 19th-century chemists to the discovery of oxygen, the nature of combustion and the overthrow of the phlogiston theory of combustion. Phlogiston was a supposed property of matter that was consumed during combustion. Conant showed that adherents of the phlogiston theory ridiculed the discovery of oxygen and its combination with combustible materials to produce heat and fire. He analyzed their thinking, human nature and motives in an effort to understand their resistance to the new theory.
One of Conant's proteges, Thomas Kuhn, became interested in the nature of scientific discovery and the reaction of the scientific community to the development of a significant new scientific theory. Kuhn began his investigation of this topic as a junior fellow at Harvard. The result was his influential book The Structure of Scientific Revolutions, published as a monograph in 1962. 1 Although I did not know Kuhn when I was a student at Harvard College in the early 1950s, I was familiar with Conant's case studies of the history of science. In fact, one of the reasons I attended Harvard was the preeminence of Conant in chemistry, education and scientific leadership. I met Conant briefly several times as a student at Harvard and decided to concentrate in chemistry and biochemistry as an undergraduate.
The main concept Kuhn developed in his monograph is the insight that science often progresses by episodic revolutions in thinking rather than by incremental analysis of scientific problems. Kuhn pointed out that the history and development of science are intertwined with the personalities, motivations, human nature and thinking habits of practicing scientists. The ascendance of a particular scientific theory depends upon how well it explains a recognized set of observations and facts established by scientists over a period of years. When a new discovery is considered incompatible with current scientific theory, a crisis occurs in the field of investigation because of the anomalous new discovery. Adherents of the established scientific theory marshal the facts at their command and energetically gather new information that may support the existing theory. Only when sufficient data and facts are gathered by proponents of the new theory to demonstrate the incompatibility of the prevailing theory does the crisis in scientific understanding lead to the new theory's acceptance and the previous theory's overthrow. Kuhn termed this resolution of conflict between incompatible theories a "paradigm shift." In the resolution period following a paradigm shift, Kuhn found that scientists gather further information to support the new theory and reexamine and reinterpret the previous body of knowledge as to its compatibility with the new theory.
Albert Szent-Gyorgi, the brilliant 20th-century biochemist, described scientific discovery as a process that begins with analyzing the same facts other scientists examine but concludes with a new concept based on fresh observation. In this process, the discoverer realizes that the significant new observation will force reexamination and reinterpre-tation of an existing body of information about a scientific challenge. The more significant the challenge and the more extensive the backlog of previous knowledge in the field, the more significant is the new discovery and the more likely it is that the new theoretical interpretation will precipitate a "paradigm shift" in the sense described by Kuhn.
It is rare in the history of science that a previously widely held theory is totally rejected and discarded by proponents of a significant new scientific theory. An example of this rare occurrence in chemistry is the overthrow of the phlogiston theory of combustion, which was totally forgotten in the wake of Lavoisier's discovery of the reaction of oxygen with combustible materials. Another example, in biology and medicine, is the theory of spontaneous generation, which was overthrown by Pasteur's famous demonstrations and arguments concerning microorganisms before the French Academy of Science. Closely allied with the overthrow of the theory of spontaneous generation was the introduction of the germ theory of disease by Pasteur, leading to the science of microbiology, the early efforts to control contagious diseases in the 19th century and the discovery of antimicrobial therapy of infectious disease in the 20th century.
More commonly, new scientific theories that are based on new discoveries force scientists to reinterpret a large body of existing information according to the principles of the new theory. An example from physics is Albert Einstein's theory of relativity, prompted by the discovery of X-rays, radioactive decay and the constancy of the speed of light. One consequence was the overthrow of the concept of transmission of light by a hypothetical, but un-proven, substance called the "ether of space." Another consequence was that the known dual nature of matter, with both wave-like and particle-like properties, was reinterpreted and reincorporated by Einstein into the theories of relativity, quantum mechanics and structure of matter that underlie the development of modern physics.
In the case of the homocysteine theory of arteriosclerosis, the discovery of the atherogenic effect of homocysteine resulting from the study of children with homocystinuria and the reproduction and demonstration of vascular disease in hyperhomocysteinemic animals did not immediately precipitate a scientific crisis in thinking about arteriosclerosis. There was at first insufficient knowledge about how homocysteine damages arteries, insufficient knowledge about homocysteine levels in subjects with arteriosclerosis in its different manifestations and insufficient knowledge about how homocysteine relates to fats, cholesterol and lipoproteins in their effects on vascular disease. Only when the accumulation of knowledge about homocysteine reached a critical level was it possible to explain and reinterpret the long history of research on arteriosclerosis. Only within the past few years, in the 1990s, has it become possible to consider that the homocysteine theory may constitute a "paradigm shift" in understanding about the cause of arteriosclerosis in susceptible populations in the sense described by Kuhn.
Controversy over the Homocysteine Theory
Why was the homocysteine theory of arteriosclerosis controversial and widely ignored when it was first developed? Why was the introduction of a new concept of arteriosclerosis based on a medical discovery about the damaging effect of homocysteine on arteries a threat to the conventional wisdom about the disease? Why was this new concept of arteriosclerosis as a disease of protein intoxication from dietary imbalance involving vitamins so difficult for medical scientists in the cholesterol, fat and lipoprotein fields to accept? Why was a new approach that had the potential to explain many otherwise imponderable observations about arteriosclerosis so widely ignored?
As explained in Chapter 1, the initial reaction to the medical discovery about homocysteine and vascular disease was one of interest and fascination on the part of medical scientists who were looking for a new approach to otherwise inexplicable observations about arteriosclerosis. Within the next five years, however, influential scientists in the homocystinu-ria field and in the cholesterol-lipoprotein field began to question the significance of the new approach. Some scientists perceived that it threatened to revise thinking radically about the underlying cause of the most pervasive degenerative disease in developed countries.
Perhaps the most controversial aspect of the new homocysteine theory, when it was first developed, was that cholesterol, fats and lipoproteins were relegated to a secondary role in understanding the underlying cause of arteriosclerosis. The discovery of the atherogenic effect of homocysteine in children with different forms of homo-cystinuria was considered an anomaly since arteriosclerosis was then observed to be caused without the evident participation of cholesterol, fats, or lipoproteins.
In the mid-20th century, elements of a crisis gradually developed in the conventional cholesterol/fat concept of the underlying cause of arteriosclerosis. A number of significant observations were found to be incompatible with the traditional approach, including the failure to relate blood cholesterol to dietary cholesterol and fat, the occurrence of arteriosclerosis in many individuals without abnormalities of blood cholesterol and lipoproteins, the failure to correlate major declines in risk of heart disease and stroke with changes in blood cholesterol or dietary cholesterol and the failure of several major trials of cholesterol lowering by diet or drug therapy to show decreased disease or increased longevity. The recent modest success of statin drug therapy in lowering blood cholesterol and reducing disease risk is compromised by the evidence of liver and muscle toxicity and carcinogenicity of these drugs in animals 2 and by the evidence that these drugs inhibit formation of ubiquinol, a key component of energy production in heart and other cells. 3 The success of vitamin E in prevention of arteriosclerosis, originally proposed by the Shute Clinic in Canada in the early 1950s, is contrasted with the recent failure to show a protective effeci of beta-carotene in disease prevention.
As pointed out by Kuhn, a crisis in scientific theory, such as the accumulation of the significant anomalies in the cholesterol/fat approach, never leads to rejection of the previous theory unless a plausible alternative theory is available to take its place. The cholesterol/fat approach persisted as the leading theory to explain the underlying cause of arteriosclerosis for many years because no other theory was available to explain the vast accumulation of observations and facts concerning the disease. The homocysteine theory of arteriosclerosis, when first introduced in 1975 4 and when first developed comprehensively in 1983, 5 did not attempt to explain the many anomalous observations in the cholesterol/fat field. This is because there was no comprehensive biochemical theory that related the observations on homocysteine and vascular disease to observations in the cholesterol/fat field. As Chapter 6 explains, such a comprehensive theory was not introduced until 1993.
Another reason the homocysteine theory of arteriosclerosis was not accepted is that the anomalous observation of vascular disease in children with homocystinuria was based on genetic diseases which are rare in the general population. If these diseases affected only 1 child in 100,000, how could the observation be applied to the population as a whole? Furthermore, the heterozygous state of homocystinuria caused by cystathionine beta synthase deficiency occurs at most in 1 to 2 percent of the population, and early studies failed to show increased risk of arteriosclerosis in these heterozygotes. Only with the recent observation of the high incidence—38 percent—of the heterozygous state of the enzyme methylenetetrahydrofo-late reductase in a susceptible population 7 does it become plausible that the homocysteine theory applies to a population at risk. This mutation could affect a major segment of the population by increasing the quantity of dietary folic acid that is necessary to prevent buildup of blood homocysteine.
Chapter 2 noted that scientists at the Harvard School of Public Health failed to confirm Rinehart's discovery of arteriosclerosis in monkeys with chronic vitamin B6 deficiency because of the experiment's flawed design and interpretation. Although other groups of investigators also observed arteriosclerosis in vitamin B6-deficient monkeys and pigs, the influence of the Harvard experiments brought into question the whole theory of vitamin deficiency as a cause of arteriosclerosis in the general population. Similarly, when medical scientists reported that they could not confirm our finding of early arteriosclerotic plaques in rabbits injected with homocysteine, the effect of their report was to question the entire validity of the homocysteine theory. Only with subsequent experimentation in animals over the next decade did it become apparent that elevated blood homocysteine is associated with arteriosclerosis in animals in the overwhelming majority of studies.
The original experiments of Rinehart and Greenberg in vitamin B6-deficient monkeys suggested that the blood levels of B6 in the human population were in the deficient range that caused arteriosclerosis in monkeys. Despite published reports in the earlier literature that suggested that a widespread deficiency of vitamin B6 in the population could account for arteriosclerosis, 8 this finding was generally dismissed by nutritionists as lacking adequate foundation. Only with recent studies showing widespread deficiencies of vitamin B6, folic acid and vitamin B12 in the elderly 9 and in cardiac patients 10 is there increasing acceptance of Rinehart's finding that vitamin deficiency may be the underlying cause of the disease. Since fat and cholesterol were found only in small quantities in the arteriosclerotic plaques of a few of Rinehart's monkeys and the predominant plaques were fibrous and fibrocalcific, most scientists who adhered to the cholesterol/fat theory regarded the experimental results as atypical of the human disease. The logical conclusion of this line of thinking was that since the experimental plaques contained little fat and cholesterol, vitamin B6 deficiency was not relevant to arteriosclerosis.
Many medical scientists who investigate patients with arteriosclerosis are only generally familiar with the detailed pathological findings in arteriosclerosis. Except for scientists with experience in human pathology, many investigators have an incomplete concept of the cellular and tissue components of arteriosclerotic plaques. Many pathologists and medical scientists who investigate the cellular and tissue aspects of the disease, on the other hand, may have an incomplete understanding of the biochemical and physiological details of how cholesterol and lipoproteins are formed in the body. The controversy over acceptance of the homocysteine theory of arteriosclerosis is partly related to misinterpretation of the significance of fat and cholesterol deposits in plaques and their role in the development of plaques. The idea that plaques are filled with or obstructed only by greasy fat deposits is incorrect for the vast majority of plaques. As discussed in Chapter 2, many arteriosclerotic plaques, even in advanced disease, are of the fibrous and fibrocalcific type that are found in subjects with homocystinuria and in monkeys with vitamin B6 deficiency. Typically arteriosclerotic plaques are tough, inelastic, thickened and heavily encrusted with calcium deposits, making them difficult to dissect with scalpel or scissors. In advanced plaques, their complex structure also includes cholesterol crystals, fatty deposits, areas of degeneration or death of tissue, blood clots, protein deposits, the growth of small blood vessels into the artery wall and areas of bleeding that predispose to complete blockage by formation of blood clots.
If a medical investigator is convinced that cholesterol and fats are the underlying cause of arteriosclerotic plaques, then any observations or experimental results in which cholesterol and fatty deposits are inconspicuous are dismissed as irrelevant. On the other hand, experts in the development of plaques agree that lipoproteins do participate in the early stages of plaque formation by forming foam cells. If the exact nature of the interaction between tissue changes in the arteries produced by homocysteine and the deposits of cholesterol and lipoproteins is incompletely understood, the participation of homocysteine in the process is perceived as questionable, leading to controversy about the significance of homocysteine in the development of plaques.
Adherents of the cholesterol/fat hypothesis correctly point out that elevated cholesterol levels are associated with increased risk of arteriosclerosis, particularly at levels greater than 240 milligrams per deciliter. Practicing physicians know, however, that the majority of their patients with arteriosclerosis have normal or desirable cholesterol levels in the 180-220 range. In my autopsy study of almost 200 veterans, the group with the most severe disease had a mean cholesterol of 186, and two-thirds had no evidence of diabetes, high blood pressure or elevated cholesterol levels. 11 The response to these facts by cholesterol and fat experts is that even though blood cholesterol level is low in some cases, cholesterol must cause the disease because of the correlation of risk with elevated levels. One of the attractive features of the homocysteine theory is that a dietary, genetic, toxic or age-related factor (homocysteine) is able to explain many of the cases of severe arteriosclerosis in patients with lifelong normal or desirable cholesterol levels. The failure to acknowledge this disparity between the cholesterol and homocysteine approaches leads to another source of controversy about the underlying cause of the disease.
Another reason for controversy over the homocysteine theory is the inherent complexity and inscrutable nature of the degenerative diseases of aging. Although cholesterol levels increase after puberty and in the adult years, the reasons for this increase are only partially understood. Furthermore, in the elderly population there is little correlation between cholesterol levels and risk of arteriosclerosis, as shown by the Framingham Heart Study. As explained more fully in Chapter 6, the gradual rise in homocysteine levels with aging is the result of a gradual shift in the way the body processes homocysteine. This shift to higher levels with aging suggests a new explanation of the biochemical basis of the aging process. The detailed processing of methionine and homocysteine in aging is at the heart of how aging cells and tissues are impaired in their ability to use food and oxygen for the production of chemical energy in the body.
Resistance to the Homocysteine Theory
In the 28 years since its first discovery, the principles of the homocysteine theory have been explained repeatedly and have been supported by scientific studies in different disciplines from laboratories all over the world. 4,5,6 Although medical experts have had extensive opportunity to read about and understand these principles, the medical community in America has been unwilling, until the past half-decade, to consider the theory seriously. One reason, as explained in the previous section, is that the homocysteine theory relegates cholesterol and fats to a secondary role in causation and incriminates deficiencies of vitamins B6, folic acid and B12 and an imbalance between dietary protein and these vitamins as the underlying cause of arteriosclerosis.
Other reasons for the reluctance to accept the homocysteine theory are (1) incomplete understanding by some investigators of the pathological changes and pathogensis of the disease, (2) the inherent complexity of degenerative diseases associated with aging, (3) inability or unwillingness to acknowledge that a completely different complex area of biochemistry other than cholesterol and fats is involved in the cause of the disease, (4) misinterpretation of aspects of experimental arteriosclerosis induced by cholesterol in animals, (5) failure to appreciate the significance of cholesterol oxides in plaque formation, (6) the observed correlation between damage to arteries and deposition of cholesterol and fats in developing plaques, (7) the inherent difficulty in attributing disease in populations to specific nutritional factors and (8) the failure of previous science to advance a theory explaining the significance of the protein intoxication approach to understanding the origin of arteriosclerosis.
When treating manifestations of arteriosclerosis, heart attack, stroke, kidney failure or gangrene of feet and legs, the medical profession frequently encounters the disease in its late stages. Usually decades of gradual narrowing of coronary, carotid, renal or iliac arteries by arteriosclerosis have occurred silently before symptoms occur. In attempting to treat a serious disease late in its course, difficult problems are encountered by physicians and surgeons. Drug therapy may be ineffective in reversing decades of damage and gradual narrowing of the arteries. Surgical therapy by angioplasty or by bypass grafts may be temporarily effective in relieving symptoms in the late stages of the disease, but the process may recur and cause narrowing of the treated arteries or the grafted segments. Treatment of associated prediposing conditions such as diabetes or high blood pressure may help to delay the onset of further complications of arteriosclerosis, but therapy for these conditions is frequently only partially successful.
Efforts by medical practitioners or nutritional experts to arrest the progress of arteriosclerosis by controlling blood cholesterol and lipoproteins through dietary modification and drug therapy are fraught with difficulty. Changing lifelong habits of poor diet and nutritional abuse are frequently ineffective in elderly patients. Drug therapy to lower blood cholesterol is complicated by toxic side effects in some cases and by failure of the cholesterol and lipoprotein levels to respond satisfactorily in other cases. Because medical practitioners encounter the disease after major complications have already occurred and because the disease is so far advanced, efforts at prevention of further progression, therefore, are frequently ineffective.
Because of their training and orientation, some medical and surgical practitioners in the arteriosclerosis field have traditionally neglected nutritional and preventive measures when treating patients with advanced disease. Because of the difficulties in treating advanced disease, the use of proper nutrition is limited. Because of some unsubstantiated claims by nutritionists and because of reluctance to accept the causative role of nutritional imbalance between proteins and vitamins, many medical practitioners have been reluctant to accept a new approach. Furthermore, nutritional scientists concentrated for many years on treating those diseases caused by extreme and serious deficiencies of vitamins; for example, they used niacin therapy for pellagra, vitamin C therapy for scurvy and vitamin B12 therapy for pernicious anemia. Only in recent years have they begun to concentrate on the role of partial vitamin deficiencies and subtle nutritional imbalances in degenerative diseases.
Because the cholesterol approach to the treatment and prevention of arteriosclerosis has prevailed for 80 years, the pharmaceutical industry has concentrated on developing better drugs to lower blood cholesterol levels. The current generation of the "statin" drugs used for this purpose impairs the function of the liver and other organs in the formation of cholesterol in the body. These drugs have had modest success in preventing the complications of arteriosclerosis in recent trials, but the price in terms of safety is potential toxicity to the liver, muscles and optic lenses and the threat of cancer, indicated by their carcinogenic effects in animals. 23 Another price is the high cost of those drugs which are now recommended for a large segment of the population, including children who may be at risk for developing elevated cholesterol levels as adults. The pharmaceutical industry has concentrated on developing these lucrative drugs because the prevention of arteriosclerosis by improving diet or adding vitamin supplements to enriched foods offers little economic incentive. The pharmaceutical industry is reluctant to support the homocysteine theory because little profit can be made from marketing vitamins and because the theory obviates the need for expensive cholesterol-lowering drugs.
Furthermore, the nutritional community and the food industry have been resistant to the homocysteine theory for several reasons. First, the theory implicates the modern food supply in the cause of arteriosclerosis because highly processed foods are deficient in vitamins and nutritionally imbalanced. The marketing of these foods, deficient in B vitamins, is profitable because of their long shelf life and ease of distribution. Second, the homocysteine theory obviates the need for marketing low-cholesterol foods, polyunsaturated oils and "light" foods which are popular and profitable. To reduce the toxic buildup of homocysteine, the food industry would instead need to concentrate on marketing and distributing fresh, minimally processed foods which are more perishable and fragile than packaged, boxed, frozen, canned or highly preserved foods. A future threat to the food supply is the proposal to market irradiated foods which are seriously deficient in vitamins B6, folic acid and vitamin B12 because of the exquisite sensitivity of these vitamins to damage by the oxidizing effects of radiation. On the other hand, there has been a major improvement in the marketing and transporting of fresh vegetables and fruits in the past 30 years, possibly contributing to the decline in heart attacks and strokes because of the increased availability of these foods all year long.
The traditional approach to addressing the nutritional cause of arteriosclerosis has been to incriminate dietary cholesterol and fats. This approach is based on the facts that feeding cholesterol to animals induces arteriosclerosis and elevated blood lipoprotein levels and diets high in the fats and cholesterol of animal origin are correlated with susceptibility to the disease. This ingrained pattern of thinking about the cause of arteriosclerosis, sometimes termed the "cholesterol myth," has a certain direct appeal because of its apparent cause-and-effect relationship. Unfortunately, nutrition and its relation to induction of disease are highly complex and difficult to interpret.
The homocysteine theory of arteriosclerosis has less direct appeal because the production of the disease is related to a subtle imbalance of dietary proteins and vitamins, relegating the increase of lipoprotein and cholesterol levels in some cases to secondary effects. In addition, the simple concept that foods with high sugar or high fat content are highly processed with little vitamin content has been difficult for adherents of the cholesterol/fat approach to accept.
My Personal Story
New scientific theories are related to the social, cultural and personal contexts of their discoverers. It has often been claimed that the time is right for a new scientific discovery because of development of a particular field; dissatisfaction with previous theories; appearance of new methods, concepts or technology; widespread awareness among scientists of opportunities for advancing understanding in a given field; and a chance observation that precipitates the development of a new theory. These influences affect the personal situation of the discoverer of a new theory, and the discovery of the new theory in turn affects the discoverer.
In the case of the homocysteine theory of arteriosclerosis, the observation of the relation between homocysteine and vascular disease, the development of the theory and its publication had some remarkable effects on my career and personal situation. As a person with a solid educational background in chemistry, biochemistry, biology and medicine; an interest in understanding the relation between biochemistry and disease; and the experience of training in several superb scientific laboratories under the guidance of outstanding scientists, my situation at the time I discovered the homocysteine theory was promising. I had recently completed my residency training in pathology, a discipline of medicine that had only begun to respond to the revolutionary developments in molecular biology, cell biology and biochemistry that occurred in the 1950s and 1960s.
My position was one of high visibility in pathology and medicine: Associate Pathologist at the Massachusetts General Hospital and Assistant Professor of Pathology at Harvard Medical School. I truly believed that my discovery of the homocysteine theory could be developed and applied to patients with arteriosclerosis effectively and expeditiously. The facilities for medical research and the atmosphere where I worked were stimulating and conducive to basically oriented scientific research on a significant problem in contemporary medicine. However, for a variety of reasons, my promising career became sidetracked and shunted in another direction by forces beyond my control. In reviewing what happened to my career, the events that occurred, the apparent reasons for the changes in my opportunities and the final outcome from a vantage point of 20 years, several conclusions are of interest in regard to the development and fate of the homocysteine theory.
Beginning in 1968 with the observation of arteriosclerosis in children with different genetic diseases causing ho-mocystinuria and continuing with the publication of experiments with animals, cell cultures, physiological and biochemical preparations from 1969 to 1976, my research associates, students and colleagues were active in determining the elements of the homocysteine theory of arteriosclerosis. Because of my background and interests, this research was concentrated on the basic biochemical, pathophysiological and cellular effects of homocysteine on arterial cells and tissues. Only one unsuccessful attempt was made to measure blood homocysteine levels in patients with arteriosclerosis. Furthermore, the abnormal processing of homocysteine thiolactone by cancer cells and the discovery of a series of new homocysteine compounds that affect cancer growth in animals were under active study in 1976, occupying the laboratory efforts of my associates.
In 1976 I was contacted by two neurophysiologists from Massachusetts Institute of Technology, Stephen A. Raymond and Edward R. Gruberg. They had heard a seminar by Dr. Moses Suzman, who described the development of the homocysteine theory, and had read all of my published articles on the subject. They became fascinated with the subject and decided to write a book for the general reader to explain the theory and its ramifications. The result of their efforts was a fascinating book Beyond Cholesterol-Vitamin B6, Arteriosclerosis and Your Heart, published in 1981. 12 Besides reviewing my discovery of the homocysteine theory, they interviewed prominent medical scientists in the homocystinuria and arteriosclerosis fields. In the book they described the nature of arteriosclerosis, the cholesterol hypothesis, evidence for the homocysteine theory, risk factors for arteriosclerosis, the recommended dietarv allowance for vitamin B6, dietary composition of methionine and vitamin B6 and other implications of the theory. Prior to publication of the book, an article based on the book was published in Atlantic Monthly. 13
These publications were remarkable in several ways. The authors were not experts in the field of arteriosclerosis. They were not physicians, but they were highly educated and experienced medical scientists in another field, neurophysiology. Yet they were able to capture the essence of the disease, to describe the development and implications of the homocysteine theory, and to contrast the new theory with the traditional cholesterol hypothesis. They admirably reviewed the evidence for the theory that was available in the late 1970s, described the background of the discovery and development of the theory, and made a cogent, well-reasoned presentation of its implications. The article and book were minor successes, generating comment in the media, generally favorable reviews, and realizing several printings.
Partly as the result of Beyond Cholesterol, the media paid more attention to the homocysteine theory. An interview was published in Prevention magazine in which I explained the use of the homocysteine theory in preventing arteriosclerosis. 14 An article was published in Time magazine introducing the homocysteine theory and contrasting it with the recent observation of arteriosclerosis induced in chickens by the herpes virus, as alternatives to the cholesterol approach. 15 In the article in Time, an expert in the cholesterol field was quoted as saying that "taking vitamin B6 . . . [for arteriosclerosis is] . . . crazy."
In addition to the book and subsequent articles, a number of newspapers, some reputable and some of the tabloid variety, reported on the publicity surrounding the homocysteine theory. A leading proponent of the cholesterol/fat approach denounced the theory in a newspaper interview as "errant nonsense" and suggested that failure to prescribe cholesterol-lowering drugs for arteriosclerosis amounted to "malpractice" or worse. Even Canadian television interviewed me regarding the controversy.
The chairman of the department at the hospital where I worked retired in 1975, and the new chairman informed me that I would need to find a position elsewhere. In explanation I was told by the hospital director that I had "failed to prove my theory." Harvard had decided against renewing my appointment or promoting me to a tenured position. I had held an appointment as assistant professor of pathology at Harvard Medical School for eight years. My 28-year association with Harvard as an undergraduate, medical student, intern, resident, research fellow, instructor, assistant professor and associate pathologist came to an end in December of 1978.
Although I had declined several inquiries about positions in other medical centers during the period when I was developing the homocysteine theory, I was now ready to accept a new position that would enable me to continue my career as a medical scientist. I was surprised to find, however, that my attempts to secure such a position were repeatedly frustrated during a two year period by unexplained obstacles. After this substantial threat to my survival as a medical scientist, assistance from a colleague and indirect pressure on my former employer finally resulted in my appointment as pathologist at the Veterans Affairs Medical Center in Providence, Rhode Island in 1981. This position, which I still occupy today, has a primary responsibility for the practice of anatomic and clinical pathology in the care of U.S. veterans. In the past 16 years, I have continued to investigate the importance of homocysteine in arteriosclerosis, cancer and aging as best as I can with the limited facilities that are available to me. My position in Providence has enabled me to develop my own individual approach to understanding the homocysteine theory, which I will describe more fully in Chapter 6.
Watson's Rules for Success in Science
In an article published by one of my former mentors, James D. Watson, on the occasion of the 40th anniversary of the discovery of the DNA double helix, he outlined the strategies for his success in science. 16 Watson's rules for success certainly applied to his career as a revolutionary molecular biologist. His rules also apply to my own experience in advancing the homocysteine theory of arteriosclerosis.
Watson's first rule is to learn from superb scientists who are winners in the intense competition in the world of science. He was able to learn about the new field of molecular biology from Salvatore Luria, an outstanding scientist in the genetics of microbiology and bacteriophage. He decided to explore his interest in DNA structure during a postdoctoral fellowship at Cambridge University in England, where he made his famous discovery with Francis Crick.
Watson's second rule is to take risks by exploring a new unrecognized approach to a significant scientific problem. In taking this risk, Watson said, a scientist has to be prepared to "get into deep trouble" because colleagues and rivals will tell you that you are "very likely to be unqualified to succeed." He added that this risk taking often leads to criticism because "your very willingness to take on a very big goal will offend some people who will think that you are . . . crazy."
Watson's third rule is to have scientific allies "who will save you when you find yourself in deep [trouble]." Watson's mentor Luria saved him when he offended one of his thesis advisors, and Max Perutz and John Kendrew saved him when his American fellowship stipend was cut off for tackling the DNA structure problem. These allies at Cambridge University supported Watson and Crick's work on DNA in its early stages, allowing successful resolution of the problem.
Watson's fourth rule is to "never do anything that bores you." You must persist with a scientific problem that is exciting and appealing to you. As Louis Pasteur said, the scientist's most precious asset is his enthusiasm for investigating a problem that interests him. Because the scientist who takes risks is likely to encounter criticism and ridicule, it is necessary to turn to experts who have knowledge in other fields for advice and to "constantly expose your ideas to informed criticism."
My own risk in advancing a new theory of the cause of arteriosclerosis was that I would offend the generations of scientists who had devoted their careers to propounding the cholesterol hypothesis. Although I had experience in superb research laboratories concerned with cholesterol biosynthesis, steroid hormones, molecular biology of protein synthesis and transfer RNA, molecular and microbial genetics, and methionine and homocysteine metabolism, I had never worked directly on the problem of arteriosclerosis. I was able to take a fresh look at the fundamental causes of this disease because I had not participated in this area of research during my years of fellowship training. I was able to apply my knowledge and background in protein synthesis, cholesterol and hormone biochemistry, molecular genetics and experimental pathology in new ways that had not been considered by previous investigators in the arteriosclerosis field.
After the first eight years of research on the homocysteine theory, I found myself in 1976 in deep trouble. My career development grant from the National Institutes of Health had expired, and the new department chairman, the hospital director and Harvard Medical School were not convinced that my approach was promising. After my involuntary termination from a promising career two years later, I found myself unable to continue my research into the homocysteine theory, and unable to investigate my new discoveries about abnormal homocysteine metabolism in cancer cells. Thereby I found myself fulfilling Watson's rule about getting into trouble over ambitious scientific ideas.
The longtime friend and colleague in medical science who helped me to salvage my research career was William Sunderman Jr. of the University of Connecticut. He recommended me for my current postion at the V.A. Medical Center in Providence. Within the first two years in that position, I was able to complete and publish my first monograph on the homocysteine theory. 5 With help and collaboration from a new colleague in Providence, Michael Vezeridis, I was able to resume laboratory research on my synthetic homocysteine compounds that antagonize carcinogenesis and inhibit the growth of malignant cells in culture and in animals, as I describe in Chapter 6. By learning from outstanding scientists in different fields, taking a risk to explore a new approach to a very large scientific and medical problem, turning to scientific allies who could help me out of career troubles and persisting in my enthusiasm for investigating an exciting scientific observation, I had fulfilled Watson's rules for success in science.
A revolution occurred in molecular biology following the discovery of the DNA double helix. Likewise, there promises to be a revolution in the understanding of degenerative diseases of aging as a result of observations on how homocysteine controls the growth, aging, and death of body cells. The observation of changes in homocysteine with aging, are an extraordinary sequel to an event in my career that occurred before I began my studies on homocysteine. In the course of an interview, I was asked about how a biomedical scientist could investigate the fundamental nature of aging at a cellular and molecular level. My response to the question was inadequate and uninformed because I was at an early stage in my career and had not considered aging as a promising field of study. Although I was later granted an American Cancer Society Faculty Research Award, I was intrigued by the question and disappointed with my lack of knowledge in the field of aging at the time.
If there is to be a revolution in understanding the role of homocysteine in the aging process, two basic conditions must be fulfilled. First, there needs to be a plausible theoretical framework for understanding how changes in homocysteine processing could explain the basic phenomena observed in aging. Second, there must be experimental validation of the theoretical formulation of changes in homocysteine with aging and with the degenerative diseases of aging, including arteriosclerosis and cancer. If these conditions are fulfilled, a fifth rule could be added to Watson's rules for success. The rule is that one's scientific success can be judged by the creative advances a new scientific theory inspires in the younger generation of scientists who follow in one's field of study.
The New Status of the Homocysteine Theory
Although the homocysteine theory is no longer new, it has yet to be widely accepted among medical investigators and practitioners. A paradigm shift has not yet occurred in the arteriosclerosis field. The homocysteine theory is currently undergoing reexamination by increasing numbers of medical investigators worldwide. Some consider the homocysteine approach increasingly interesting, promising and productive because of its power to explain otherwise inexplicable facts.
The principle reason for the new status of the homocysteine theory as a hot new area for medical research on arteriosclerosis is that a reliable test for blood homocysteine levels has begun to be applied to human studies. While studies with experimental animals, cell cultures, and biochemical pathways are of theoretical interest to medical experts and investigators, only with successful human studies 910 do the results begin to convince the skeptics. Surveys using the test for blood homocysteine on populations at risk and groups of patients with arteriosclerotic disease have now fulfilled the major predictions of the homocysteine theory. 17 As explained in Chapter 3, these human studies have shown that elevated blood homocysteine, hyperhomocysteinemia, is a major independent risk factor for the development of arteriosclerosis.
The status of the homocysteine theory has now reached a critical stage. Large-scale testing of the theory in populations at high risk of arteriosclerosis is urgently needed to demonstrate its effectiveness in preventing and treating the disease. Such a successful demonstration may well fulfill the prediction made in an article published in Science this year, "heart attack: gone with the century." 18 By showing that nutritional measures, supplemental vitamins, and other nontoxic strategies to control elevated blood homocysteine levels reduce the risk of vascular disease, a dramatically lower risk of heart attack and stroke and increased longevity will be demonstrated in a successful study of this type will constitute final proof of the validity of the approach. It is time that some of the funding which has been lavished for decades on the cholesterol/fat hypothesis with equivocal, disappointing or inconclusive results should now be directed to planning and concluding long-term prospective trials of the homocysteine theory.
Over the past half-decade there has commenced a worldwide interest in planning a prospective trial of the homocysteine theory. As yet, these efforts have been only desultory and incipient. The past record of disappointing results in the large-scale intervention trials aimed at controlling arteriosclerosis by lowering of cholesterol levels is a warning that poorly conceived or inadequately executed trials may fail to accomplish ambitious objectives.
In the development of a prospective trial to test the homocysteine theory of arteriosclerosis, several principles must be followed. Since arteriosclerosis develops slowly over a period of decades, the trial must be of sufficient duration to demonstrate a difference between treated and control groups in slowing progression of the disease. Prevention by dietary modification should emphasize limitation of methionine consumption and augmentation of natural sources of vitamins B6, folic acid and B12. Prevention and treatment with supplemental vitamins must be at sufficient doses to insure control of blood homocysteine levels. Other factors affecting homocysteine levels, such as age, gender, family history, thyroid function, kidney function, blood pressure, diabetes, drugs, hormones and toxins must be controlled and compared between those in treated and control groups. Batteries of clinical tests are needed to document changes in blood homocysteine, levels of vitamins B6, folic acid and B12, LDL, HDL and tests of major organ function. Finally, documentation of the clinical complications of arteriosclerosis, such as heart attack, stroke, gangrene and embolism needs to be established by reliable criteria.
The interpretation of results of a prospective trial to test the homocysteine theory needs to focus on changes in disease incidence, the relation to nutritional status and dietary consumption, the results of laboratory testing and clinical studies, and the importance of genetic and toxic factors. Modern molecular methods to detect hidden genetic defects affecting both homocysteine and lipoproteins will add to the decisiveness of the results. Successful conclusion of such prospective trials in the future will demonstrate beyond doubt that an optimal diet with or without supplemental vitamins over a sufficient period of time will substantially reduce heart attack, stroke, kidney failure and arteriosclerotic gangrene as killer diseases.
The First International Conference on Homocysteine Metabolism from Basic Science to Clinical Medicine held in Ireland in 1995, gathered over 300 medical scientists working in the homocysteine field. 19 Over 120 separate scientific studies from different countries were presented to demonstrate increased acceptance among the medical community of the homocysteine approach to control vascular disease and other medical conditions, including birth defects, cancer and kidney failure.
Studies on the intake of folic acid to prevent spina bifida and other serious neurological birth defects have prompted the U.S. Food and Drug Administration to require the addition of this vitamin to all enriched foods, including flour, pasta, cereals and rice by 1998. The action of folic acid not only prevents neural tube birth defects but also prevents elevation of blood homocysteine in mothers during the first trimester of pregnancy. 20 Sufficient folic acid in enriched foods will help to insure that all pregnant women will consume 400 meg of the vitamin per day from the moment of conception and throughout their pregnancies. An additional anticipated benefit will be the prevention of up to 50,000 deaths from heart disease per year attributable to elevation of blood homocysteine.
REFERENCES
1. Thomas S. Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962).
2. Thomas B. Newman and Stephen B. Hulley, "Carcinogenicity of lipid-lowering drugs." Journal of the American Medical Association 275:55-60, 1996.
3. Richard A. Willis, Karl Folkers, J. Lan Tucker, Chun-Qu Ye, Li-Jun Xia and Hiroo Tamagawa, "Lovastatin decreases coenzyme Q levels in rats." Proceedings of the National Academy of Sciences USA 87:8928-8930, 1990.
4. Kilmer S. McCully and Robert B. Wilson, "Homocysteine theory of arteriosclerosis." Atherosclerosis 22:215-227, 1975.
5. Kilmer S. McCully, "The homocysteine theory of arteriosclerosis: Development and current status." In: Antonio M. Gotto, Jr. and Rodolfo Paoletti, editors, Atherosclerosis Reviews, volume 11 (New York: Raven Press, 1983), pp. 157-246.
6. Kilmer S. McCully, "Chemical pathology of homocysteine. I. Atherogenesis." Annals of Clinical and Laboratory Science 23:477-493, 1993.
7. Phyllis Frosst, Henk J. Blom, R. Milos, Philippe Goyette, O.A. Shephard, Rowena G. Matthews, Godfried J.H. Boers, Martin den Heijer, Leo AJ. Kluijtmans, Lambert P. W.J. van den Heuvel and Rima Rozen, "A candidate genetic risk factor for vascular disease: A common mutation in methylenetetrahy-drofolate reductase." Nature Genetics 10:111-113, 1995.
8. "Vitamin status of cardiac patients." Nutrition Reviews 25:116-118, 1967.
9. Jacob Selhub, Paul F. Jacques, Peter W.F. Wilson, David Rush and Irwin H. Rosenberg, "Vitamin status and intake as primary determinants of homocysteinemia in an elderly population." Journal of the American Medical Association 270:2693-2698, 1993.
10. Killian Robinson, Ellen L. Mayer, Dave P. Miller, Ralph Green, Frederick van Leute, Anjan Gupta, Kandice Kottke-Marchant, Susan R. Savon, Jacob Selhub, Steven E. Nis-sen, Michael Kutner, Eric J. Topol and Donald W. Jacob-sen, "Hyperhomocysteinemia and low pyridoxal phosphate. Common and independent reversible risk factors for coronary artery disease." Circulation 92:2825-2830. 1995.
11. Kilmer S. McCully, "Atherosclerosis, serum cholesterol and the homocysteine theory: A retrospective study of 194 consecutive autopsies." American Journal of the Medical Sciences 299:217-221, 1990.
12. Edward R. Gruberg and Stephen A. Raymond, Beyond Cholesterol. Vitamin B6, Arteriosclerosis and your Heart (New York: St. Martin's Press, 1981).
13. Edward R. Gruberg and Stephen A. Raymond, "Beyond cholesterol. A new theory on arteriosclerosis." Atlantic Monthly 243:59-65, 1979.
14. Jane Kinderlehrer, "B6—maybe the answer to heart disease." Prevention 31:138-145, 1979.
15. "Diet debate. Is cholesterol the culprit?" Time August 6, 1979, p. 61.
16. James D. Watson, "Succeeding in science: some rules of thumb." Science 261:1812-1813, 1993.
17. Kilmer S. McCully, "Homocysteine and vascular disease." Nature Medicine 2:386-398, 1996.
18. Michael S. Brown and Joseph L. Goldstein, "Heart attacks: Gone with the century?" Science 272:629, 1996.
19. International Conference on Homocysteine Metabolism from Basic Science to Clinical Medicine. Irish Journal of Medical Science 164 Suppl 15:1-36, 1995.
20. Johan B. Ubbink, "Is an elevated circulating maternal homocysteine concentration a risk factor for neural tube defects?" Nutrition Reviews 53:173-175, 1995.
21. Carol J. Boushey, Shirley A.A. Beresford, Gilbert S. Omenn and Arno G. Motulsky, "A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes." Journal of the American Medical Association 274:1049-1057, 1995.