Recombinant Inbred RI Strains

Definition and Development

Sets of recombinant inbred (RI) strains are produced by crossing two standard inbred strains such as C57BL/6 and DBA/2 to produce an F1 hybrid, then brother x sister mating from these for at least 20 generations, but maintaining several separate lines. Ideally, at least 20 new lines are developed, though smaller sets can still be useful. In these strains, the genes for which the parental strains differ have been sorted into new combinations in each of the resulting RI strains. At any one genetic locus, such as the major histocompatibility complex, at which the parental strains differ, about half the strains should resemble each of the progenitor strains. There are many sets of RI strains available, developed from different pairs of progenitor strains.46

Nomenclature of RI Strains

RI strains are known by an abbreviation of the two parental strains, with the female parent given first, separated by an upper case X, followed by numbers for the individual strains. For example, AXB2 is a strain derived from a cross between a strain A female and a strain C57BL/6 male, identified as strain 2 of this set. If the male parental strain already ends in a number, a hyphen may be used to distinguish the individual strains.

Research Uses of RI Strains

Sets of RI strains are useful for investigating characters that differ between the two parental strains. If the parental strains differ for some phenotype, such as the activity of a particular enzyme or susceptibility to a pharmaceutical or toxic agent, it is often of interest to determine whether the difference is due to a single genetic locus, and if so, it would be desirable to genetically map that locus.

The procedure is to phenotype a full set of the RI strains developed from a cross between the strains for the character of interest, such as high or low activity of an enzyme. Obviously, this is only worthwhile if a suitable set of RI strains already exists, because it takes several years to develop and characterize each set. Thus, the first step would be to screen the parental strains of several sets of RI strains to see whether they differ. Assuming a suitable set of these strains can be found, small numbers of animals of each strain are phenotyped, and the distribution of strain means is studied. If the phenotypes fall into two distinct groups, each of which resembles one of the parental strains, then that would be evidence that the phenotype is controlled by a single genetic locus. In contrast, if there were a continuum of phenotypes covering many intermediate values, then this would suggest that the character has a more complex mode of inheritance. If there are sufficient numbers of individual RI strains in the set, it may be possible to resolve a character controlled by two loci, but it is rare to be able to distinguish more than about two loci.

If the phenotype indicates that the character is controlled by a single locus, then this can usually be genetically mapped to some chromosomal region. Suppose the study involves the BXD set of RI strains in which, in the original set, there are 26 individual strains, and the parental strains (C57BL/6 and DBA/2) differ in the activity of a particular enzyme. Suppose all 26 strains are typed, and in 16 of them, the enzyme activity resembles strain C57BL/6, and in the other ten, it resembles DBA/2, and all strains can unambiguously be classified into these two groups. The ratio of 16 to 10 is approximately a 50:50 ratio. The individual strains are designated BXD1, BXD2, BXD5, BXD6, BXD8, and so on (BXD3 and BXD4 were lost during inbreeding). The resemblance of each strain to one of the parental strains can be indicated by a B or a D, and a pattern of response such as DBBDD BBDBB, etc., can be built up. This is called the "strain distribution pattern" or SDP. Two genes tightly linked on the same chromosome will tend to have a similar or even identical SDP, because the loci will rarely have recombined during the inbreeding. The SDP of two unlinked genes will not be different from random. As an example, Modi (an enzyme locus) and D9Mit10 (a microsatellite locus) have identical SDPs of DBBDD BBDBB BDDBB DDBBB BBBDB B, and D9Mit10 has been mapped to a particular location on mouse chromosome 9. Thus, Mod1 can also be placed close to D9Mit10 on chromosome 9 (in fact, Mod1 was mapped before D9Mit10, but this example illustrates the method).

In contrast, the SDP of D1Mit5 on chromosome one is -BBDD BBBBB DDBDB BBBBD DDD (a "-" indicates no information). Of the 22 strains typed at both Mod1 and D1Mit5, only 12 loci are concordant, and this ratio of 12/22 is not different from 1/2, indicating no linkage between the two, as expected.

Most of the large sets of RI strains have been typed at many mapped loci, so if a single gene phenotype is indicated, then there is a high chance that it can be mapped simply by getting a computer to look for matches in the SDPs. No actual genotyping is required. Gene loci defined in terms of DNA can be mapped easily, because DNA samples from most of the RI strain sets are readily available.

Sets of RI strains can also sometimes be used to identify and map quantitative trait loci (QTLs) that are involved in the inheritance of polygenic characters. For example, there is a significant association between susceptibility to urethane-induced lung tumors and genotype at the Kras2 locus in the AXB and BXA set of RI strains.47 However, few sets of RI strains are sufficiently numerous to make it possible to resolve more than about two or three loci.

Maintenance of RI Strains

Each RI strain is an inbred strain in its own right, and should be maintained as an inbred strain using methods described above.

0 0

Post a comment