Limits to genetic rescue by outcross in pedigree dogs.

Outcrossing should reduce inbreeding levels and associated negative effects in highly inbred populations. In this study, we investigated the effectiveness of different outcrossing schemes using computer simulations. The inbreeding rate estimated for a 25-year period of 2.1% per generation in a highly inbred dog breed reduced to 1.8% when a single litter was produced by an outcross without backcrosses. To reduce the inbreeding rate below 1%, more than eight of the 14 litters born yearly in the recipient breed had to be outcrossed. However, outcrossing in pedigree dogs is usually followed by backcrossing and generally involves one or a few litters. Backcrossing reduced the effect of outcrossing considerably. When two litters were produced by an outcross followed by one generation of backcross, the inbreeding rate was 2.0% per generation. Continuously outcrossing was more effective than a single or a few outcrosses. When each newborn litter during 25 years had a 5% chance of being produced by an outcross, the inbreeding rate reduced to -0.2%. To investigate the possibility that new alleles were introduced from the donor population into the recipient population, the fate of different type of alleles (varying from completely lethal to beneficial) before and after an outcross was investigated by first simulating 80 years of natural selection prior to the outcross and then different types of outcross. Because natural selection reduced the frequency of lethal alleles before outcrossing, the introduction of a lethal allele that was segregating in the donor breed but not in the recipient breed occurred rarely. Introduction of slightly detrimental alleles or neutral alleles occurred more frequently. In conclusion, outcrossing only had a limited short-term effect unless repeated continuously. Nevertheless, it may help to buy time in which the population structure can be changed so that the effective population size increases.