On this page
Hormones such as insulin circulate in blood at very low concentrations. In the 1950s doctors had no direct way to measure them; they estimated insulin from its effects on rats or on pieces of animal tissue. A physicist trained in measuring radioactivity changed that, working in a hospital with a doctor who taught her medicine. This post tells Rosalyn Yalow's story in her own words, explains how the radioimmunoassay works, and works through the ratio at the heart of it.
How did Rosalyn Yalow become a physicist?
Her Nobel autobiography records that she was born in New York City on 19 July 1921. Her parents had not been to high school but never doubted their children would go to college. A chemistry teacher first caught her interest, and at Hunter College, a women's college, physics professors drew her into physics. Hunter offered a physics major for the first time in September 1940, when she was a senior.
Her family thought she should become an elementary school teacher, and she doubted good graduate schools would support a woman in physics. She took a part-time secretarial job at Columbia, on condition that she learn stenography, hoping it would lead to graduate courses. Then, in February 1941, she was offered a teaching assistantship in physics at the University of Illinois. She tore up her stenography books.
At Illinois she was the only woman among the 400 members of the Faculty of Engineering, the first since 1917. When she earned an A- in an optics laboratory course, the department chairman said, "That A- confirms that women do not do well at laboratory work". She finished a PhD in nuclear physics in January 1945 and, she writes, became skilled in making and using apparatus for measuring radioactive substances.
How did she end up in a hospital?
Back in New York, her autobiography continues, she taught physics at Hunter and volunteered in a medical physics laboratory to learn about medical uses of radioisotopes. In December 1947 she joined the Bronx Veterans Administration Hospital as a part-time consultant, setting up its Radioisotope Service with, in her words, "nothing more than a janitor's closet and a small grant". She joined full time in January 1950. That July a young physician, Solomon Berson, joined the service. Their partnership lasted 22 years, until his death on 11 April 1972.
How did they discover the radioimmunoassay?
Yalow and Berson used radioactive tracers to study how substances move through the body. With insulin, they saw it disappeared more slowly from the blood of patients who had been treated with animal insulin. They reported in 1956 the demonstration of an insulin-binding globulin in the circulation of insulin-treated subjects: these patients had made antibodies to the animal insulin. Yalow writes that, studying how insulin reacts with these antibodies, they saw they had a tool to measure insulin itself, and that "the era of radioimmunoassay (RIA) can be said to have begun in 1959."
How does the radioimmunoassay work?
Their 1960 paper, Immunoassay of endogenous plasma insulin in man, explains the principle. Antibodies from guinea pigs immunised with beef insulin bind beef insulin labelled with radioactive iodine-131. Human insulin also binds those antibodies, and competes with the labelled insulin for them. The more human insulin in a sample, the less labelled insulin is bound.
The team separated bound from free labelled insulin and measured the ratio B/F (bound to free). Comparing a sample's B/F with a standard curve made from known amounts of human insulin gave its concentration. The paper describes setting up the antiserum so that B/F was between 2 and 4 with tracer alone, and notes that with sensitive antisera B/F fell by about 50 percent at 15 microunits per ml of human insulin. The method was sensitive to less than one microunit and needed only 10 to 20 microlitres of plasma. The earlier methods, the paper notes, depended on insulin's effects on rats or tissue and were of doubtful specificity.
Worked example: from B/F to fraction bound
Fraction bound. If B/F = 3, then for every 3 parts bound there is 1 part free. Fraction bound = 3 / (3 + 1) = 0.75, or 75 percent.
Add human insulin. If B/F halves to 1.5, fraction bound = 1.5 / (1.5 + 1) = 0.6, or 60 percent.
Read the curve. Using the 1960 paper's figure for a sensitive antiserum, a fall in B/F of about half corresponds to roughly 15 microunits per ml. A real lab reads the value off its own standard curve, made the same day.
Common mistakes:
Treating B/F as a percentage. A B/F of 3 means 75 percent bound, not 300 percent.
Assuming the curve is a straight line. You compare with standards rather than scale up from one point.
Reading a value outside the range of the standards.
For the physics of radioactive tracers, see our guide to radioactive decay and half-life. For why insulin levels matter, read our post on hormones and feedback loops.
How was the work recognised?
Yalow received half of the 1977 Nobel Prize in Physiology or Medicine "for the development of radioimmunoassays of peptide hormones". She wrote that Berson would have shared it had he lived, and asked for their laboratory to be named after him so his name would stay on her papers. By 1977, she wrote, RIA was used to measure hundreds of substances in thousands of laboratories. She died on 30 May 2011.
What can a student take from Yalow's way of working?
Carry your skills across subjects. A physicist's skill in measuring radioactivity became a tool in medicine.
Learn from a partner. She learned medicine from Berson, and he learned from her; she wrote that they were each other's severest critics.
Follow an odd result. Insulin clearing slowly from the blood led to the antibody, and the antibody led to the assay.
If physics ideas such as radioactivity are hard to picture, the Fermi Simulation Lab has interactive simulations for ideas that are easier to see than read.
Work out the fraction bound for B/F values of 4, 2 and 1, and sketch how it changes.
