On this page
Your body can make antibodies against molecules it has never met. NCERT tells you an antibody has two heavy and two light chains, H₂L₂. But if every antibody needed its own gene, there would not be enough DNA. Susumu Tonegawa solved that puzzle. This post explains his experiment, then works through the combinatorics that make the answer click.
Who was Susumu Tonegawa?
Tonegawa was born in Nagoya, Japan, on 5 September 1939, and the Nobel facts page records his death on 11 July 2026. He won the 1987 Nobel Prize in Physiology or Medicine alone, "for his discovery of the genetic principle for generation of antibody diversity".
His Nobel autobiography is a story of chance turns. He failed his first entrance exam to Kyoto University's chemistry department and got in a year later, in 1959. Reading François Jacob and Jacques Monod on the operon pulled him into molecular biology. He did his PhD at the University of California, San Diego, then worked with Renato Dulbecco at the Salk Institute. When his US visa was running out, Dulbecco suggested the new Basel Institute for Immunology. He arrived in 1971 and writes that he "had very little knowledge of immunology" when he accepted the post.
What was the puzzle?
The 1987 Nobel press release puts it simply. The body has hundreds of millions of different antibodies ready before it meets the antigens they recognise. Yet the human genome was then thought to contain about 100,000 genes. One gene per antibody "simply didn't add up".
The alternative idea was that DNA in developing B cells is changed so that a limited set of inherited pieces can make many different products. Tonegawa set out to test it with the then-new tools of molecular biology, restriction enzymes and recombinant DNA.
What did the 1976 experiment show?
In Hozumi and Tonegawa (1976), PNAS, they cut mouse DNA with the restriction enzyme BamHI and sorted the fragments by size. Then they used radioactive antibody messenger RNA to find which fragments carried the code for the variable (V) and constant (C) parts of a kappa light chain.
In embryo DNA, the V and C sequences were on two different fragments, of about 3.9 million and 6.0 million daltons.
In DNA from an antibody-producing tumour (a plasmacytoma), both were on a single, smaller fragment of about 2.4 million daltons.
The conclusion in the abstract: V and C genes "which are some distance away from each other in the embryo cells, are joined to form a contiguous polynucleotide stretch during differentiation of lymphocytes".
Worked example: how many antibodies from a few hundred pieces?
The press release gives figures for the human heavy chain: about 200 V pieces, about 20 D pieces and 4 J pieces. Each B cell picks one of each at random.
Step 1: heavy-chain variable regions.
200 × 20 × 4 = 16,000 combinations.
Why multiply? For each of the 200 V choices there are 20 D choices, so 200 × 20 = 4000 V-D pairs, and each pair can join any of 4 J pieces.
Step 2: add joining variation. The press release adds that V, D and J join imprecisely and that the V and D sets often differ between the copies inherited from each parent. Together this gives "something like five million" heavy-chain variable forms.
Step 3: pair with a light chain. The light chain adds "more than 10,000 variants". Since any heavy chain can pair with any light chain:
5,000,000 × 10,000 = 5 × 10¹⁰, or 50 billion.
That matches the press release's "many billions" of possibilities.
Common mistakes:
Adding instead of multiplying. 200 + 20 + 4 = 224 is the number of pieces, not combinations.
Thinking the genome stores billions of antibody genes. It stores a few hundred pieces; the variety comes from combining them.
Forgetting the light chain. The final count multiplies heavy and light options.
These are the 1987 figures quoted by the Nobel Committee for humans. Exact counts of gene segments vary with the source and species, so learn the method rather than the numbers.
Why does it matter for the immune system?
The press release notes that each B lymphocyte produces its own unique antibody, and an adult has about 10¹² B lymphocytes. Most will never be used, but the large, varied pool means some cell is ready for almost any antigen. Tonegawa's autobiography adds that his group went on to find the T cell receptor genes, the "other half" of the immune system.
For the chapter view of antibodies, read Human health and immunity. For the first use of antibodies as medicine, read Emil von Behring and the diphtheria antitoxin.
What can a student take from Tonegawa's way of working?
Bring a new tool to an old question. He was a molecular biologist, not an immunologist, and that was his advantage.
Compare two states of the same thing. Embryo versus antibody-making cell was the whole design.
Ask for help early. He credits an immunologist and a geneticist for guiding him into the field.
If combinatorics problems like this one feel slippery, the Fermi Problem Library has a step-by-step solver that explains the reasoning rather than printing an answer.
Next step: redo the calculation with 100 V, 25 D and 6 J pieces and check you get 15,000 heavy-chain combinations.
