How ecology constrains grey wolf gene flow

Not new, but this 2004 paper by Eli Geffen and colleagues concerning population structure in grey wolves has a good discussion of gene flow may be limited even in cosmopolitan species with long-distance dispersal.

Considering the known cline in pelt color variation (white wolves further north), the appearance of clines for other genes should not have been unexpected. But the force of selection on color is perhaps more obvious than the force attributable to other kinds of ecological features, such as vegetation variants. So finding the genetic variation serves as a cue to look for the ecology, rather than vice-versa:

We suggest two possible mechanisms underlying the observed correlation between an east-west axis and genetic variation in North American wolf populations. First, North America has several dominant topographic divisions that extend several thousand kilometres on a north-south axis (e.g. Pacific Coast range, Rocky Mountains, Great Plains; see Fig. 1). Each of these regions creates a broad longitudinal band of unique vegetation zones. The reluctance of dispersing individuals to move between these major zones would generate an isolation-by-distance pattern on an east-west axis. Second, wolf packs usually do not follow migratory caribou but maintain year-round resident territories. However, during years when prey densities are low, up to 17% of all wolf packs follow migratory caribou and then return to their original territory for denning (Ballard et al. 1997). The caribou migration is a well-documented phenomenon that occurs across Alaska and Canada (Kelsall 1968). The pursuit of migratory caribou by wolves across large distances would diminish the possibility of having a north-south genetic cline in wolf populations along the migration route because wolf populations along this gradient could mix, whereas the dominance of north-south movements could reduce dispersal east and west (Carmichael et al. 2001). Specific data on long-range dispersal in relation to landscape features for wolves and caribou are needed to test these hypotheses. In conclusion, our results indicate the need for future genetic studies to assess the potential importance of climate and habitat as underlying causes of genetic patterns of differentiation (Geffen et al. 2004:2488).

This last is an interesting idea -- that long-distance dispersal routes across some subset of the range should impede differentiation along them, but might still allow differentiation along a perpendicular axis. Geometrically obvious, but it takes a fairly unique set of circumstances to set it up -- and it might be more important to population structure in a low-density predator than in the relatively higher-density prey species.

The authors also suggest a way that clines might be maintained in a high-dispersal species without selection:

Additionally, we suggest that developing grey wolves may become imprinted with regard to climate and habitat. The tenure of wolves in their natal pack can be long, as young wolves are often recruited as helpers (Mech 1988; Mech 1999), and could drive the development of hunting skills for prey in local habitats. When young wolves disperse, often in their second or third year (Gese & Mech 1991), they may direct their movements toward familiar landscapes. Dispersing wolves that select familiar ground have a better chance of survival (Gese & Mech 1991). Such behaviour may account for why wolves hunting migrating caribou appear to be differentiated from nearby resident wolves that hunt nonmigratory game (Carmichael et al. 2001). Ecotypes of this nature have been described in killer whales (Orcinus orca; Saulitis et al. 2000) but not in large terrestrial carnivores (but see Rueness et al. 2003).

That sounds similar in principle to the ecological differentiation of black and grizzly bears, although there the differences are enforced by direct competition and maintained by a species boundary. Here, the differences are proposed to be more fluid, but still capable of creating clines in neutral alleles.

But that's in addition to selection:

The importance of climate as an explanation for population genetic structure has been more widely invoked for plants than animals (Sork et al. 1999). The low dispersal ability of plants allows for adaptation to local environmental conditions (Sork et al. 1999). Climatic and phytogeographical factors may also restrict movements of small terrestrial species (e.g. King 1987; Arter 1990) or species that are associated with specific host plants (e.g. Keyghobadi et al. 1999). The impact of such factors on highly mobile animals is much harder to envision. An environmental cline can provide a mechanism for differentiation (e.g. Endler 1980; Lande 1982; Turelli et al. 2001) and a recent model of speciation suggests that evolutionary branching can occur along clines of moderate steepness given the presence of intraspecific competition (Doebeli & Dieckmann 2003). Competition, both inter- and intraspecific, may be intense in larger carnivores (Van Valkenburgh & Wayne 1994), and in wolves, intraspecific strife between wolf packs is one of the largest sources of natural mortality (Mech 1994). Consequently, a model of differentiation along a cline could conceivably apply to North American wolves.

The article begins by pointing out that the grey wolf's range is the largest of any extant terrestrial mammal. Of course, that is except for humans -- which continues to make wolves an interesting model for Pleistocene human population structure.

The important point here is this: widespread distribution and rapid long-distance dispersal in wolves have not inhibited the formation of significant clines of genetic variation. These clines are maintained by selection and/or structured migration (here proposed to correlate with learned ecological roles). These factors operate at the level of local populations, and there is no need to invoke distinct origins for different wolf regional populations to account for them.

And it is of great interest that the major cline here is an east-west differentiation, correlated with climate zones. The differentiation among Pleistocene Eurasian hominids on an east-west (China-Europe) axis remains to be explained in similar terms, and it will be worth reviewing ecological information to find relevant barriers to gene flow.

References:

Geffen E, Anderson MJ, Wayne RK. 2004. Climate and habitat barriers to dispersal in the highly mobile grey wolf. Mol Ecol 13:2481-2490. DOI link