Substrains
How substrains arise
Substrains may arise for any of the following reasons:
- residual heterozygosity or incomplete inbreeding at the time of separation from progenitors (Bailey 1977; Bailey DW. 1982. Immunology Today 3:210-14),
- undetected spontaneous mutations that become fixed in a colony (genetic drift) (Radulovic et al. 1998; Sluyter et al. 1999; Stiedl et al. 1999; Specht et al. 2001; Roth et al. 2002; Wotjak 2003),
- undetected genetic contamination (Naggert et al. 1995),
- deliberate (and either unrecorded or forgotten) outcrossing of strains for specific experimental purposes (Bailey 1977; Bailey DW. Immunology Today 1982; 3:210-14; Simpson et al. 1997; Threadgill et al. 1997; Wotjak 2003),
- separation of a subcolony from its parent colony for a combined total of 20 or more generations (for example, when, since being separated, it and its parent colony have each been bred for 10 generations - 20 generations between the two), and
- a new health status assignment to a subcolony (for example a pathogen free status).
Substrain phenotypes can vary
Numerous studies report physiological and behavioral differences among substrains. As examples:
- behaviors differ among substrains (Crawley et al. 1997; Stiedl et al. 1999);
- tissue rejection differs among related 129 strains (Simpson et al. 1997);
- tumor susceptibility differs among C3H substrains (Glant et al. 2001);
- fear responses differ between C57BL/6J and C57BL/6N mice ((Radulovic et al. 1998; Stiedl et al. 1999);
- effects of anesthetics on cardiac function differ between C57BL/6J and C57BL/6N mice (Roth et al. 2002); and
- electroconvulsive thresholds differ between C57BL substrains (Yang et al. 2003).
Once a subcolony is determined to be a substrain, it should be given a laboratory code. Laboratory codes are assigned by the Institute of Laboratory Animal Research (ILAR).
Immunologists uncover many substrain differences
Substrain differences may be particularly important to immunologists, whose studies depend on well-defined, homogenous backgrounds. In fact, immunologists seem to uncover more genotypic variations in inbred strains than do other scientists, perhaps because the molecular traits they often investigate are more sensitive than are other traits to subtle changes (Bailey DW. 1982. Immunology Today 3:210-14).
Differences among 129 substrains
Reports of potential and actual confounding results because of 129 substrain differences have served as a wake-up call to the research community. The 129 strain originated in 1928 and has since differenciated into numerous substrains. Because embryonic stem (ES) cells derived from 129 mice colonize germlines so efficiently, the 129 strain is one of the most widely used strains in genetic studies. Yet, according to Threadgill et al. (1997), ES cell lines from numerous 129 substrains are being used with little attention to their differences. Realizing that 1) the origin and the reported physiological differences between 129 substrains were unknown (Hogan B. Beddington R, Costantini F, Lacy E. 1994. Manipulating the mouse embryo: a laboratory manual, 2nd ed. Cold Spring Harbor (NY)), 2) many loci in the R1 ES cell line appeared to be heterozygous, and 3) efficient gene targeting depended on isogenic DNA (te Riele et al. 1992; van Deursen and Wieringa 1992). Threadgill and his colleagues decided to conduct a thorough molecular analysis of the relatedness of the various 129 substrains. They found that strain 129/SvJ is significantly different from other 129 substrains and should be more accurately classified as a recombinant congenic strain (129X/Sv) derived from 129/Sv and an unknown strain, "X."
Based on this finding, they suggest that:
- confusing results of their experiments involving targeted Egfr alleles are due to genetic differences in the 129 substrains (129X1/SvJ, an F1 hybrid between 129X1/SvJ and 129S1/Sv-+Tyr+Oca2, and 129S1/Sv-+Tyr+Oca2) used to derive the ES cell lines carrying the targeted mutations.*
genetic differences between the 129 substrains explains why 129X1/SvJ is a high ovulator in response to exogenous gonadotropins, whereas 129P3/J and 129P1/ReJ are low ovulators (Hogan B. Beddington R, Costantini F, Lacy E. 1994. Manipulating the mouse embryo: a laboratory manual, 2nd ed. Cold Spring Harbor (NY)). - because efficient homologous recombination in ES cells depends on isogenic DNA, the most common gene-targeting experiments, which use constructs derived from one of the 129X1/SvJ-derived libraries in a 129S1/Sv-+Tyr+Oca2 -derived ES cell line, are either suboptimal or impossible (they mention numerous reported and unreported examples where this problem may have already occurred).
* Petkov and his colleagues (Petkov et al. 2004), using the panel of SNPs mentioned in the Genetic Contamination section, determined that 129X1/SvJ has genetic contributions from C57BL/6J on Chromosomes 5, 7, 14, 18, and 19, and from BALB/cJ on Chromosomes 7, 8, 10, 18, 19, and X, suggesting that the "X" in 129X1/SvJ is an F1 hybrid between C57BL/6J and BALB/cJ.
A bibliography is available for sources referenced on this page.
