Humans and mice may look very different but, being mammals, they share many common features. Examine the physiology, anatomy or metabolism of a mouse, and you can find many insights into what makes a human tick. Hence the mouse has become a favourite of researchers investigating almost any aspect of mammalian biology - whether embryonic development or disease, behaviour or cancer.
These similarities to humans are reflected in the mouse genome, the sequence of which was published in 2002. Almost every gene in the human genome has a counterpart in the mouse; researchers have allied this to powerful genetic tools and have developed thousands of mouse strains with mutations that mirror those seen in human genetic disease.
"As a model organism for understanding human biology, the mouse is the best," argues Ian Jackson of the MRC Human Genetics Unit, University of Edinburgh. "Flies, worms and frogs are all important for understanding how genes work in cells or for understanding the fundamental principles of development. But if you want to investigate the role of genes in the whole mammal, you have to use mice."
Humans and mice have an ancient relationship, which probably began when humans stopped being hunter-gatherers and took up farming - grain stores being an ideal food source for mice. While this may not have kept the relationship friendly, the unusually coloured mice that occur quite frequently in the wild have long been of interest - being mentioned in ancient Chinese references and depicted on early Egyptian artifacts.
During the 1700s, the collection and breeding of 'fancy' mice with different coloured coats became a popular hobby in Japan. ('Fancy' is a 19th-century English word for hobby, particularly a livestock hobby.) In the 1800s, these coloured mice began to find their way into Europe, where they increased in number and popularity, particularly in Victorian England. A National Mouse Club, which set standards for the different varieties of mice, was founded in 1895, and mouse-fancying clubs can be found worldwide to this day.
In mid-19th-century Austria, mice had also appealed to Gregor Mendel as he embarked on his studies on inheritance. He began breeding mice in his quarters, aiming to decipher the inheritance of their coat colour. Outraged by the thought of a monk living with animals having sex, Mendel's conservative bishop banned the mice, so Mendel had to turn his attention to a far more monkish pursuit - gardening.
Mendel's laws of inheritance, discovered through his pea research, languished for more than 30 years. Their rediscovery in 1900 launched a new era of genetics, and researchers immediately wanted to know whether Mendel's discoveries in peas applied to mammals. "If you go back to the original papers in early genetics, people were using lots of different organisms, such as guinea pigs, rats, chickens, rabbits and mice," says Dr Jackson. "Over time, the mouse became the preferred organism for mammalian genetics, as it is small and has a rapid generation time [of about nine weeks]. Over the years, sophisticated genetic tools have been developed, but the basic way of doing genetic studies - then and now - is the same. If you put two mice together, you get baby mice."
The mouse in the lab
In France, Lucien Cuénot was the first - in 1902 - to demonstrate Mendelian ratios for the inheritance of coat colour characters in mice. In Harvard, William Castle began his research in the same year, buying mice from a local mouse fancier who had quickly turned her hobby into a business. Together with his student Clarence Little, Castle produced a series of seminal papers on coat-colour genetics.
Little is probably best known for his development of 'lab mice' - inbred mouse strains that are still used today. "Inbred strains have been very important for mouse genetics," says Dr Jackson. "You need a uniform genetic background against which you can compare new variations. There are several hundred inbred strains, each with a different background, although only perhaps a dozen are used commonly by the research community." Little's first inbred mouse, DBA (dilute brown non-agouti), was developed in 1909; his most famous strain, C57BL/6, in 1921. C57BL/6 was the strain whose genome was sequenced and published in 2002.
Clarence Little's contribution to the field was not finished. In 1929, backed by two car barons, Edsel Ford (Henry's son) and Roscoe Jackson (head of the Hudson Motorcar Company), he set up the Jackson Laboratory in Maine, USA. The lab is a world-renowned centre for mouse genetics, and has been a major influence on keeping mice at the forefront of mammalian biology.
For several decades, researchers focused on finding mutants and variants, and mapping the genes involved. As well as coat colour, other easily identifiable traits were examined - such as ear shape, and tail length and shape. In the 1960s and 1970s, biochemical markers were developed, allowing researchers to look at protein variants. This led to a large increase in the number of known mutations and genes, but the true power of mouse genetics was not unleashed until 1977, when the first mouse gene was isolated.
Transgenes and knockout mice
Armed with molecular biology techniques, researchers can isolate, examine and modify DNA, teasing apart its role in the body. In the early 1980s, 'transgenic mice' became all the rage, after it was shown that DNA injected into mouse eggs could be incorporated into the genome. The DNA could, for example, be a 'reporter gene' under the control of a promoter from a normal mouse gene. The protein made by the reporter gene is therefore produced in the same place, and at the same stage of development, as the normal gene.
The next innovation in mouse genetics came with the development of knockout mice - mice lacking a specific gene - in the late 1980s. "The crucial things here were embryonic stem cells," says Dr Jackson. "Matt Kaufman and [Sir] Martin Evans (Cambridge), and Gail Martin (San Francisco) grew cells from an early embryo. Evans, Liz Robertson, Alan Clark and Allan Bradley [now Director of the Wellcome Trust Sanger Institute] went on to show that those cells could contribute to a new embryo, grow into an adult mouse, and be part of the germline of the adult. So the idea was, if you can modify a gene in the embryonic stem cells, the modification could be passed onto future generations of mice."
The first knockout mice lacked the HPRT gene (hypoxanthine guanine phosphoribosyl transferase gene), mutations in which, in humans, cause a mental retardation disorder called Lesch-Nyhan syndrome. These knockout mice were produced by identifying a random mutation, but Oliver Smithers in the USA had been developing a way of swapping new, modified DNA directly into the genome - a technique called homologous recombination. When his 'gene targeting' technique was ported to embryonic stem cells, the production of knockout mice became far more efficient.
"There must be 3500-4000 genes that have been knocked out so far, and tens of thousands of papers have been written about them," says Dr Jackson. "People are now doing more sophisticated knockouts, where the gene is removed only in certain types of cells types, or the gene is turned on or off when a drug is added. Another approach is to 'knock-in' a gene - you can replace the existing gene with another gene, such as a reporter gene.
"This is the 'gene-up' approach to mouse genetics," he adds. "You say 'this is the gene I'm interested in; let's see what it does'. So you knock it out, see what happens. The other approach is 'phenotype down', where you find a mutation that is interesting and find the gene responsible."
Mouse and man
The sequence of the mouse genome was published in 2002. "The mouse and human genomes are very similar," says Dr Jackson. "There are a relatively small number of rearrangements, and the gene content is pretty much the same. Some gene families have expanded in the mouse, such as those involved in scent recognition, innate immunity to pathogens [humans and mice being exposed to different pathogens], and in reproduction. People have tended to comment on the differences between mice and humans, but there are so many similarities. If you find a mutation in a mouse gene, you almost always find a human disease with similar effects."
Some mouse models of human disease, such as for obesity, cancer or immune system defects, have arisen spontaneously. Many other models - notably for cystic fibrosis - have been generated as knockouts. Even so, there are still thousands of diseases, and thousands of genes, that researchers have not tackled as yet. With the genome sequenced, allied to high-throughput technology, the concept of testing what every gene does, and of fully untangling how genes contribute to disease, has become a realistic goal.
"Over the last five years, high throughput DNA sequencing has sped up research remarkably," says Dr Jackson. "To find a mutation, you don't have a do lots of crosses using lots of mice, you can look at the DNA directly. So you can find lots of gene mutations in a couple of months - experiments that would take years in mice alone. It has changed that way you think about genetics.
"Now that the genome has been sequenced, far fewer mice are required per discovery. It is a really powerful tool."
Mouse Genome Sequencing Consortium (2002) Initial Sequencing and comparative analysis of the mouse genome. Nature 420: 520-562
Nolan P M et al (2000) A systematic, genome-wide, phenotype-driven mutagenesis programme for gene function studies in the mouse. Nature Genetics 25: 440-443
Kile B T et al (2003) Functional genetic analysis of mouse chromosome 11. Nature 425: 81-86.
Nadeau J H et al (2001) Sequence interpretation. Functional annotation of mouse genome sequences. Science 291: 1251-1255
Healy E, Jordan S A, Budd P S, Suffolk R, Rees J L, Jackson I J (2001) Functional variation of MC1R alleles from red-haired individuals. Human Molecular Genetics 10: 2397-2402