How the house mouse became biology's default model organism

Almost every mammalian genetics paper published in the last century rests, one way or another, on the house mouse. Mus musculus is the animal behind inbred strains, knockout technology, and a genome sequence that arrived only a year after the human one. None of that was planned from the start. The mouse's rise began not in a university laboratory but in Victorian pet shows and a small commercial breeding farm in Massachusetts, run by a woman with no formal scientific training who happened to keep meticulous records of which mice got sick.

The fancy mouse trade

Long before geneticists took any interest, people bred mice purely for how they looked. "Fancy mice", selectively bred for coat colour, coat pattern, and temperament rather than for any scientific purpose, had been kept as pets in Japan and China for centuries, and the hobby had a considerable following in Britain and the United States by the late 1800s. The National Mouse Club was founded in England in 1895 to standardise and judge fancy mouse varieties at shows, in much the same way dog and cage-bird fanciers organised their own hobbies. Breeders traded animals selected for traits like waltzing behaviour (a circling, head-bobbing movement caused by an inner-ear mutation), particular coat colours, and unusual markings.

It was this hobbyist trade, not any university, that produced the first large, well-documented stocks of genetically distinctive mice available in bulk. The most important of these commercial breeders was Abbie E. C. Lathrop.

Lathrop was a retired schoolteacher who moved to a farm in Granby, Massachusetts, in 1900. When an attempt at poultry farming failed, she began breeding mice and rats instead, starting from a small stock of waltzing mice and expanding the operation until, at its peak, her farm housed more than 11,000 animals in straw-filled boxes. She sold them as pets and show animals, but her careful attention to coat colour and breeding lines soon caught the eye of scientists. In 1902 the Harvard geneticist William Ernest Castle, working at the university's Bussey Institute, placed an order for mice from her farm, and Lathrop went on to supply breeding stock to laboratories for the rest of her working life.

Lathrop's contribution to mouse research went beyond simply supplying animals. From around 1908 she noticed unusual skin growths appearing in some of her mice and began sending samples to researchers, eventually establishing a working correspondence with the pathologist Leo Loeb. Between 1913 and 1919 the two co-authored ten papers in journals including the Journal of Experimental Medicine, showing that susceptibility to particular tumours varied by breeding line and appeared to be heritable, among the earliest evidence that cancer risk could have a genetic basis rather than being purely a matter of chance or infection. Lathrop died of pernicious anaemia in 1918, having spent nearly two decades supplying the animals that, in the hands of others, would go on to define mammalian genetics.

Turning fancy mice into a genetic tool

William Castle's Bussey Institute laboratory at Harvard trained a generation of mammalian geneticists, and it was there, from around 1909, that an undergraduate named Clarence Cook Little began investigating whether coat colour in mice followed Mendelian inheritance patterns, using breeding stock obtained from Lathrop. By carefully mating siblings over successive generations, a technique that concentrates existing variation into predictable, uniform lines, Little produced what is usually credited as the first deliberately inbred strain of mice, DBA, named for the dilute, brown, and non-agouti coat-colour genes it carried.

Little went on, in 1921, to establish another line by mating a female (numbered 57) and a male (numbered 52) from Lathrop's stock, which became known as C57BL. In 1937 that line was split into two separately maintained sublines, C57BL/6 and C57BL/10, of which C57BL/6 would go on to become the single most widely used inbred mouse strain in the world. The idea underlying all of this work was straightforward but powerful: if every animal in a strain is genetically near-identical to every other, differences in disease susceptibility, tumour incidence, or response to treatment between strains can be attributed to genotype rather than to chance individual variation. That principle, established with Lathrop's mice at Harvard in the 1900s and 1910s, is still the basis of inbred-strain research today.

Bar Harbor and the founding of the Jackson Laboratory

Little's career moved from Harvard to the presidency of the University of Maine and then the University of Michigan, but his central interest remained mammalian genetics and cancer. In 1929, with backing from Detroit industrialists he had cultivated as university patrons, including the Ford, Webber, and Jackson families, Little founded a dedicated research institute in Bar Harbor, Maine, on land donated by the conservationist George B. Dorr. The laboratory was named, after his death, for Roscoe B. Jackson, president of the Hudson Motor Car Company and one of its funders. It opened with a small staff and modest funding, and the coastal site was chosen partly because its climate suited animal breeding.

The timing was almost immediately disastrous: construction was barely finished when the stock market crashed later that year. The laboratory survived the Depression years in part by selling standardised, well-documented inbred mice, including strains descended directly from Little's Harvard work, to researchers elsewhere, a commercial sideline that from 1933 onward became a core part of its business model and remains so today. A further catastrophe struck in October 1947, when a wildfire that burned across Mount Desert Island destroyed most of the laboratory's buildings and killed the great majority of its mouse colonies, along with its library and records. Staff, led by the geneticist Elizabeth "Tibby" Russell, rebuilt almost every strain from breeding animals that had been sent out to collaborating laboratories before the fire, and the institution recovered with support from bodies including the National Institutes of Health and the American Cancer Society.

The Jackson Laboratory, still based in Bar Harbor, is today one of the world's principal repositories of characterised mouse strains, and its early decision to treat inbred mice as a distributable, standardised resource rather than a private laboratory tool is arguably as important to the mouse's dominance as any single genetic strain it produced.

From bred strains to engineered genomes

Inbred strains gave researchers control over genetic background, but for most of the twentieth century they still could not alter a chosen gene directly and predictably. That changed through work carried out independently on two sides of the Atlantic during the 1980s.

In Cambridge, the developmental biologist Martin Evans, working with Matthew Kaufman, showed in a 1981 Nature paper that pluripotent cells could be isolated directly from mouse blastocysts and grown in culture while retaining the ability to differentiate into other cell types or form tumours when reintroduced into a mouse. These were the first mouse embryonic stem (ES) cell lines, and they solved a critical practical problem: they gave researchers a cell population that could be genetically modified outside the animal, checked, and then used to generate a whole living mouse carrying that modification.

Meanwhile, at the University of Utah, Mario Capecchi and, working separately, Oliver Smithies at the University of Wisconsin were pursuing a related idea: that a technique already used to introduce specific mutations into bacterial and yeast genomes through homologous recombination might also work in mammalian cells, despite the vastly larger and more repetitive mammalian genome making a matching event seem, to many contemporaries, close to statistically impossible. Capecchi's early NIH grant application proposing to test this in mammalian cells was turned down by reviewers who judged the odds of success too low to justify funding; he pursued the work regardless, using discretionary funds from an already-funded project. The approach was validated in mammalian cells during the mid-1980s, and in 1987 Capecchi, together with Kirk Thomas, published a paper in Cell demonstrating site-directed mutagenesis of a chosen gene in mouse ES cells by gene targeting. Combined with Evans's ES cell technology, that meant a specific, deliberately altered gene could now be carried from a cultured cell all the way into a living, breeding mouse: the "knockout mouse".

"for their discoveries of principles for introducing specific gene modifications in mice by use of embryonic stem cells"

That is the exact wording the Nobel committee used in 2007 when it awarded the Nobel Prize in Physiology or Medicine jointly to Mario Capecchi, Martin Evans, and Oliver Smithies. By the time of the award, more than ten thousand mouse genes had been individually knocked out across the research community, using the combined ES cell and gene targeting method the three had developed. Gene targeting turned the mouse from an organism whose genetic differences could only be observed and bred for, as Little and Lathrop had done decades earlier, into one whose genome could be deliberately rewritten to test a specific hypothesis about what a single gene does.

Reading the whole genome

Knockout technology let researchers ask precise questions about individual genes, but by the late 1990s a further resource was needed: a complete reference sequence against which any single gene, mutant, or comparison with human biology could be checked. The Mouse Genome Sequencing Consortium, an international collaboration built around centres including the Whitehead Institute/MIT Center for Genome Research, the Wellcome Trust Sanger Institute, and Washington University in St Louis, published a high-quality draft sequence of the C57BL/6J mouse genome in Nature in December 2002, in a paper titled "Initial sequencing and comparative analysis of the mouse genome".

The scale of what the sequence revealed reinforced, in genomic terms, what breeders and geneticists had been assuming by inference for a century: the mouse genome runs to roughly 2.5 billion base pairs, modestly smaller than the human genome, with an initial estimate of around 30,000 genes (a figure, in line with the equivalent human estimate, that was later revised down as annotation improved). Around 90% of the mouse genome could be aligned to matching segments of the human genome despite roughly 75 million years of separate evolution, and the substantial majority of genes had a recognisable human counterpart. The mouse genome project arrived only a year after the initial human genome sequence, a deliberate sequencing choice that let researchers immediately compare the two, gene for gene and region for region, in a way that was not previously possible at this resolution.

Why the mouse, and not something else

None of the individual pieces of this story, hobbyist breeding, inbred strains, a research institute, embryonic stem cells, or a sequenced genome, would alone have made the mouse the standard laboratory mammal. What made the difference was how they accumulated on top of each other over more than a century. Lathrop's farm produced well-documented genetic variation. Little turned that variation into standardised, reproducible inbred strains and built an institution designed to distribute them widely rather than hoard them. Evans, Capecchi, and Smithies then gave researchers a way to make precise, deliberate changes to the mouse genome rather than relying on naturally occurring or randomly induced mutations. The 2002 genome sequence gave every one of those tools a common reference to work against, and let mouse findings be checked directly against human biology.

Each step depended on the one before it. Inbred strains would have been far less useful without a way to alter genes directly; gene targeting would have been far less interpretable without a reference genome; and none of it would have started at all without a Massachusetts farm that, by chance rather than design, kept careful pedigree records of pet mice. It is a reminder that dominance in model organism research rarely comes from a single breakthrough. It comes from a long chain of people, most of them not trying to reshape biology, who happened to build tools that the next generation could not do without.