(2007) 99, 1–2 & 2007 Nature Publishing Group All rights reserved 0018-067X/07 $30.00 www.nature.com/hdy NEWS AND COMMENTARY

Selection and the rate of loss of ...... strong selection! Perhaps such strong and genetic selection should not be surprising, given the harsh climate and the limited grass cover of the island. It is becoming diversity very clear that inbred individuals are DH Reed more sensitive to environmental stress ...... (for example, Armbruster and Reed, Heredity (2007) 99, 1–2; doi:10.1038/sj.hdy.6800990; published online 16 May 2007 2005; Reed et al., 2006) and that this is an important consideration in conserva- tion . A combination of high fecundity, a large variance in volutionary and conser- effects’). The latter occurs because in- coefficients among the offspring born in vation biologists are very much dividuals within the population vary in a given year and strong selection E interested in how various evolu- their inbreeding levels. Kaeuffer et al. against inbred individuals creates ideal tionary forces interact to determine the (2006) suggest that their results are more circumstances for selection to maintain amount and type of genetic variation likely to be due to local than general heterozygosity levels within a popula- found in natural populations and how, effects, but admit that there is no tion, even in the face of the potential for in turn, that genetic variation affects the concrete way to test this. The inability significant . fitness and evolutionary potential of to discriminate between these two hy- Like most good studies, this one populations. In populations smaller potheses is not, in my opinion, of great raises more questions than it answers. than a few thousand individuals, it is concern as ultimately all -wide It has generally been felt that selection generally believed that genetic variation effects must have a local cause. The real strong enough to maintain or increase is lost via random genetic drift faster question is whether inbreeding depres- genetic variation, would also be strong than can replace it. This is of sion is due to increased homozygosity enough to imperil small populations particular concern with regard to the at a few loci with large effect or a large with rapid . However, Kaeuf- conservation of endangered of number of loci with small effect. fer et al. (2006) state that selection had a plants and animals, as the loss of genetic Several studies of inbred populations negligible effect on the demography of diversity through random genetic drift performed in the laboratory had already the population in their models. This can compromise the future ability of suggested that the decrease in hetero- might reflect reality if there is consider- populations to evolve in the face of at neutral loci is slower than able reproductive excess and mortality changing environmental conditions. that expected by theory (for example, is determined primarily by intraspecific Thus, a recent paper published in Rumball et al., 1994; Latter et al., 1995; competition. If mortality is strictly den- the Proceedings of the Royal Society B Gilligan et al., 2005). However, there are sity-dependent and due primarily to (Kaeuffer et al., 2006) demonstrating two reasons why the paper by Kaeuffer competition for limited food resources, that an isolated population of mouflon et al. (2006) is so intriguing. First, they then truncation selection such as that (Ovis aries) introduced to a remote did not just see a slower decline in modeled is possible and would not island as a single founding male–female heterozygosity than what is expected push the population to extinction. The pair, has apparently increased in hetero- under common models of neutral ex- fact that the goats are not native to the zygosity (a measure of genetic varia- pectations. They actually saw an in- island and there are no predators, tion) over the course of approximately crease in heterozygosity, as estimated makes this type of scenario not so far- 23 generations is of great interest. from 25 loci, in a popula- fetched. However, the population re- Kaeuffer et al. (2006) suggest that this tion with a harmonic mean population mained very small for at least five increase in heterozygosity is due to size of less than 20 individuals. This generations after introduction, as natural selection acting against inbred suggests that the strength of selection though struggling with inbreeding de- individuals. against inbred (more homozygous) in- pression. Then after a short climb to The idea that natural selection can act dividuals may be much stronger than carrying capacity, the population began as a force to help maintain genetic was previously suspected and therefore to cycle in a way that suggests strong diversity is not new. Associative over- the ability of selection to maintain food limitation. This suggests that there dominance is the term used to describe genetic diversity greater. Second, this must have been strong selection against the fitness difference between hetero- is not a highly fecund species in a inbred individuals for survival to ma- zygotes and homozygotes at a neutral laboratory environment. The fact that turity in the absence of limited food locus. Since the locus itself is typically a these changes in heterozygosity oc- supplies early on, or there would have noncoding portion of DNA being used curred in a wild population of a been little genetic variation to preserve as a molecular marker in the study (for vertebrate makes it more compelling. in the later density-dependent phase. example, microsatellite loci), differences Kaeuffer et al. (2006) performed a The biggest question was raised by in do not directly contribute to number of computer simulations to test Kaeuffer et al. (2006) themselves: Can fitness. The correlation between hetero- how various founder scenarios and the results from this population be zygosity at neutral loci and the fitness of selection regimes are predicted to im- generalized to other populations? Was an individual can come about because pact genetic variation in the introduced there a special set of circumstances that the marker loci are directly linked to loci population of mouflon. They found that allowed genetic variation to increase in that do directly affect fitness (‘local truncation selection against individuals this small population following the effects’) or heterozygosity at these mar- with heterozygosity levels less than 0.40 bottleneck or are the conditions for this kers can correlate with fitness because resulted in an increase in heterozygosity type of phenomena fairly liberal? Or they accurately reflect genomic hetero- similar to the one they observed over is this type of scenario rare but will zygosity (‘general’ or ‘genome-wide time in this population. This is very show up often in surviving populations News and Commentary 2 because only the subset of bottlenecked cular genetic variation in finite populations of of genetic diversity in captive and wild populations that maintain genetic varia- Drosophila melanogaster. Genet Res 85: 47–55. populations of two nearly extinct species of tion survive long enough to be tested? Kaeuffer R, Coltman DW, Chapuis J-L, Pontier D, Goodeid fish reveals that one is inbred in Re´ale D (2006). Unexpected heterozygosity in the wild. Heredity (advance online publication, This paper should spur further research an island mouflon population founded by a doi:10.1038/sj.hdy.6800947). on natural and experimental popula- single pair of individuals. Proc R Soc B, Gle´min S, Vimond L, Ronfort J, Bataillon T, tions to test the boundary conditions doi:10.1098/rspb.2006.3743. Mignot A (2006). Marker-based investigation where genetic diversity can be main- Latter BD, Mulley JC, Reid D, Pascoe L (1995). of in the endangered Reduced genetic load revealed by slow in- species Brassica insularis. Heredity 97: 304–311. tained in small populations and what it breeding in Drosophila melanogaster. Hamill RM, Doyle D, Duke EJ (2006). Spatial means to their future and 139: 287–297. patterns of genetic diversity across European persistence. Reed DH, Nicholas AC, Stratton GE (2006). subspecies of the mountain hare, Lepus timidus Inbreeding levels and prey abundance interact L. Heredity 97: 355–365. Dr DH Reed is at the Department of Biology, to determine fecundity in natural populations Slate J, David P, Dodds KG, Veenvliet BA, Glass BC, University of Mississippi, PO Box 1848, Univer- of two species of wolf spider. Conserv Genet, Broad TE et al. (2004). Understanding the doi:10.1007/s10592-006-9260-4. relationship between the inbreeding coefficient sity, MS 38677, USA. Rumball W, Franklin IR, Frankham R, Sheldon BL and multilocus heterozygosity: theoretical expec- (1994). Decline in heterozygosity under full-sib tations and empirical data. Heredity 93: 255–265. and double first-cousin inbreeding in Droso- Zachos FE, Hartln GB, Suchentrunk F (2007). e-mail: [email protected] phila melanogaster. Genetics 136: 1039–1049. Fluctuating asymmetry and Armbruster P, Reed DH (2005). Inbreeding de- in the roe deer (Capreolus capreolus): a test of the pression in benign and stressful environments. Editor’s suggested reading developmental stability hypothesis in mam- Heredity 95: 235–242. mals using neutral molecular markers. Heredity Gilligan DM, Briscoe DA, Frankham R (2005). Bailey NW, Macias Garcia C, Ritchie MG (2007). (advance online publication, doi:10.1038/ Comparative losses of quantitative and mole- Beyond the point of no return? A comparison sj.hdy.6800954).

Heredity