Toxic Substances Program, University of California, Davis, California 95616-8723
If you ever wondered why the well-known rat poison is called Warfarin, it is from the initials WARF for Wisconsin Alumni Research Foundation. WARF funded the initial studies on coumarins in Karl Paul Link's laboratory, which gave rise to this compound and to its close relative, dicoumarol. Professor Link was a pioneer in what would now be called basic research in toxicology. Perhaps because his research was so oriented to biochemistry and so interdisciplinary, his name seems to be lost when pioneers in toxicology are discussed. His seminal work began with a problem among dairy cattle, the major industry in Wisconsin, which needed a solution.
Sweet clover (Melilotus alba and M. officinalis) is grown as a green manure and hay crop in the northern U.S. and in Canada. Its coumarin content gives it a distinctive sweet odor similar to vanilla. Its use as hay was widespread in the 1920s, when a series of wet summers led to an epidemic of "bleeding disease" in cattle. Epidemiological detective work by veterinarians traced the cause to sweet clover hay that had been improperly cured and had become infected with molds. When sweet clover is put up as hay, it is easy for it to be contaminated with molds. This is particularly likely when conditions are wet at cutting or curing, or when foliage is lush. Molds such as Penicillium nigricans, P. jensi, and the Aspergilli metabolize the coumarin into dicoumarol. Cattle or sheep consuming the spoiled hay showed less clotting power in their blood, a condition leading to fatal internal hemorrhage.
Link came to this problem from a project to develop a strain of sweet clover that would be low in the coumarin that gave it a sweet smell but a bitter taste. A farmer in Wisconsin came to him in 1933 with spoiled hay and dying cattle. In the depths of the great depression, this represented a very serious loss for the farmer, who could not afford to replace the hay. Link set out to isolate the active principal from the spoiled hay. An improved in vitro clotting assay, using rabbit plasma, was employed to guide chemical fractionation of the spoiled hay. Large-scale isolation of dicoumarol was accomplished eventually by a graduate student, Mark Stahmann, who went on to become a professor in the biochemistry department at Wisconsin for all of his own long and productive career.
After six years of work, Link's laboratory had crystallized dicoumarol (3,3`-methylene-bis[4-hydroxycoumarin]); a year later the actual structure was determined and active dicoumarol was synthesized. Dicoumarol was released into clinical medicine in 1941, where it has enjoyed widespread use ever since as an anticoagulant, a drug property much in demand at the time, since heparin was the only available alternative. Dicoumarol is similar in structure to vitamin K. When consumed by livestock, it inhibits vitamin K production. Vitamin K is necessary in the body to activate prothrombin. When tissue is damaged, thromboplastin is released and converts prothrombin to thrombin. Thrombin alters the solubility of fibrinogen in blood and causes it to clot and seal the tissue damage. Dicoumarol prevents this process. The relationship between blood clotting rate, vitamin K, and dicoumarol remained to be understood clinically until later, despite early work by Link that clearly demonstrated the reversal of dicoumarol effects in rabbits by giving them vitamin K.
During a six-month "sabbatical" at a local sanatorium in 1945, while recovering from "wet pleurisy," Link conceived of the use of a coumarin derivative as a rat poison. During this period of enforced idleness, he reviewed all of the chemical and bioassay data from his laboratory and selected candidate relatives of dicoumarol that had been synthesized between 1940 and 1944, as potential "better mousetraps (rat traps)." A more potent analogue of dicoumarol named warfarin was first promoted in 1948 as a rodenticide. It was potent enough to kill rats ingesting multiple doses in poisoned bait or water, despite their simultaneous ingestion of vitamin K (which was not the case for dicoumarol). This was a radical departure from previous practice; the concept of using a slow-acting poison rather than vastly more toxic poisons that killed rats acutely, such as cyanide and strychnine, was novel and required a good deal of salesmanship by Link and WARF. Warfarin is used in rat, gopher, and ground squirrel poisons, and also acts as a vitamin K inhibitor to block the blood clotting process and provoke hemorrhaging. After achieving success as a rat poison, warfarin successfully made the transition to a clinically useful anticoagulant in the 1950s (under the name "Coumadin") because of its high water solubility and its bioavailability after oral dosing. It is more potent than dicoumarol, but retains the property of reversibility of its anticoagulant effects with administration of high doses of vitamin K. The first of several WARF patents for Link's blood anticoagulant discoveries was granted in 1947 for dicoumarin, the agent in spoiled sweet clover that causes cattle to hemorrhage.
In another version of this story, Link was extremely concerned by a 1948 report from England that dicoumarol might be useful as a rat poison. He felt that such a development would doom the use of dicoumarol as an anticoagulant, as clinicians would be reluctant to prescribe "rat poison" to patients. He set out to find an analogue from the lab's collection of previously synthesized relatives of dicoumarol, and selected Warfarin (compound #42, a coumarin derivative) for this purpose. It is ironic that Warfarin (Coumadin) rapidly also became the drug of choice for clinical use, largely replacing dicoumarol, and achieved perhaps its greatest fame (notoriety) when used to treat then President Eisenhower after his heart attack in 1955.
According to Link, dicoumarol is not a good rat poison. It acts too slowly on rodents in field conditions, where their diet is high in vitamin K due to consumption of natural grains as food. Higher doses in pellets or baits might kill rats, but would also pose a threat to domestic animals and humans. Warfarin, on the other hand, is a much more potent and efficient rodenticide. It kills rats by acting as an anticoagulant, so that they bleed to death internally. It also increases capillary fragility, which enhances internal hemorrhaging. It is safe to use; literally millions of tons have been used worldwide for over 50 years, and it is still in wide use with very little risk to humans and nonrodent animals. Resistance to Warfarin was first reported in rats in 1960, and further studies of this phenomenon opened up a whole new field of research on the mechanism of activation of prothrombin and the discovery of gamma-carboxyglutamic acid as an essential component in this process.
Biochemistry at the University of Wisconsin, as with most of the land grant universities in the heartland, grew out of a tradition of agricultural chemistry, and the department has had no formal affiliation with the medical school on campus. As a result, biomedical research has generally been performed with little clinical collaboration, and such was the case for Link's early work. Dicoumarol's narrow therapeutic window can expose any patient to the risk of abnormal bleeding, so accurate assessment of their response to therapy is crucial to maintaining safe, therapeutic dosages. Link was deeply suspicious that clinicians would not use dicoumarol properly, as the correct dosage must be titrated for each patient, and is a function of their vitamin K status as well as other complex factors including their liver and kidney function. In addition, there is a 12- to 24-hour lag phase before the effect of dicoumarol as an anticoagulant is manifested, and there is a cumulative effect of repeated administration. Link's concerns were apparently justified; dicoumarol gained the reputation of being a dangerous drug to use because of episodes of uncontrolled bleeding. It took about a decade for the drug to be routinely used with careful determination of prothrombin clotting times and vitamin K ready as an antidote. Dicoumarol, and generic copies made mainly by European laboratories, remained the anticoagulant of choice until the mid-1950s, when the better qualities of warfarin (Coumadin) for clinical use became clearly apparent. Warfarin, currently the most commonly used anticoagulant, antagonizes vitamin K, a key player in the clotting cascade. Without vitamin K, the liver can't activate proteins necessary to form and maintain a clot. Warfarin and other oral anticoagulants are used to prevent clots or the extension of clots associated with various conditions, including venous thrombosis, pulmonary embolism, atrial fibrillation, and myocardial infarction. It is also used prophylactically to prevent stroke in certain high-risk populations. It has been estimated that sodium Warfarin and dicoumarol have been administered to more than one million patients to prevent life-threatening conditions associated with heart disease and stroke.
Link had entered graduate school in Agricultural Chemistry at the University of Wisconsin in Madison in 1922, and became its first Professor of Biochemistry in 1930. His thesis research with Professor W. E. Tottingham involved the analysis of various constituents of plant tissues. From about 1934 on, he focused much of his laboratory's research on dicoumarol-related questions. He also maintained an active research interest in carbohydrate chemistry for the entirety of his career. He trained many of the future leaders in this field among his 55 Ph.D. students and 43 M.S. degree recipients. He retired from the University of Wisconsin in 1971 after a successful career there that spanned almost 50 years. He died in 1978.
Professor Link also left a significant legacy to science in the many students he trained as Ph.D.s, including Stanford Moore (who shared the 1972 Nobel Prize with W. H. Stein and Christian Anfinson for their studies on the structure of ribonuclease). Moore's thesis research required that he learn the new microanalytical methods of Pregl for the determination of C, H, and N. Link had recently returned from Europe where he had studied in the laboratory of Fritz Pregl in Graz, so had intellectual and familial ties to Germany. He was a friend of Max Bergmann, who had recently arrived from Germany to lead a laboratory at the Rockefeller Institute for Medical Research in New York. Through that friendship, in 1939 Moore was encouraged to join the Bergmann Laboratory, which was an internationally renowned center of research on the chemistry of proteins and enzymes. Moore stayed at the Rockefeller Institute for the remainder of his career.
I had the pleasure of encountering Professor Link while I was a graduate student at the University of Wisconsin. He was one of three members of my M.S. examination committee in 1961. I had been warned that his questions might cover a wide range of topics, but that the history of biochemistry was a favorite area, so I was prepared when we got into a series of questions on who discovered what, when. He was especially gratified that the history of dicoumarol and Warfarin was an area still being studied by graduate students more than 20 years after the seminal discoveries had been made. I was reminded of this when I read a book entitled One Hundred Years of Agricultural Chemistry and Biochemistry at Wisconsin, the proceedings of a symposium held in Madison in 1983 (Nelson, D.L., and Soltvedt, B.C., Eds., 1989, Science Tech Publishers, Madison, WI). This book contains chapters by Professor Link, from a talk he gave in 1958 to the New York Academy of Medicine, and by one of his former students, Clinton Ballou, that were source material for much of this biography.
It is noteworthy that Link continued a Wisconsin tradition of reinvesting royalty income from commercially valuable research discoveries made with WARF support into increased funds to support research by University of Wisconsin faculty. This nonprofit Wisconsin Alumni Research Foundation (WARF) was organized in 1925 to commercialize the discovery by Professor Harry G. Steenbock that vitamin D could be produced in milk by ultraviolet radiation. This invention led to the virtual worldwide elimination of rickets by the 1940s. It should be noted for the post-Sputnik generations of scientists that the importance of WARF as a vehicle for supporting scientific research in its first 30 years cannot be overestimated. The massive federal support for research via NIH and NSF that we take for granted today came only after the United States made this a priority in the late 1950s. Steenbock's, and later, Link's altruism in turning over potential profits from their discoveries for reinvestment into research at the University of Wisconsin played a pivotal role in making WARF the source of almost five hundred million dollars of research support for many scientists, past and present.
NOTES
* For correspondence via fax: (530) 752-5593. E-mail: jalast{at}ucdavis.edu.