The effects of historical fragmentation on major histocompatibility complex class II β and microsatellite variation in the Aegean island reptile, Podarcis erhardii

Abstract

The major histocompatibility complex (MHC) plays a key role in disease resistance and is the most polymorphic gene region in vertebrates. Although habitat fragmentation is predicted to lead to a loss in MHC variation through drift, the impact of other evolutionary forces may counter this effect. Here we assess the impact of selection, drift, migration, and recombination on MHC class II and microsatellite variability in 14 island populations of the Aegean wall lizard Podarcis erhardii. Lizards were sampled from islands within the Cyclades (Greece) formed by rising sea levels as the last glacial maximum approximately 20,000 before present. Bathymetric data were used to determine the area and age of each island, allowing us to infer the corresponding magnitude and timing of genetic bottlenecks associated with island formation. Both MHC and microsatellite variation were positively associated with island area, supporting the hypothesis that drift governs neutral and adaptive variation in this system. However, MHC but not microsatellite variability declined significantly with island age. This discrepancy is likely due to the fact that microsatellites attain mutation-drift equilibrium more rapidly than MHC. Although we detected signals of balancing selection, recombination and migration, the effects of these evolutionary processes appeared negligible relative to drift. This study demonstrates how land bridge islands can provide novel insights into the impact of historical fragmentation on genetic diversity as well as help disentangle the effects of different evolutionary forces on neutral and adaptive diversity.

Keywords: drift; historical fragmentation; immunity; major histocompatibility complex; selection.

Figure 1

Podarcis erhardii in typical habitat on Naxos, of the Cyclades in the Aegean Sea

Figure 2

Map of central Cyclades in the Aegean Sea. Pie charts are clustered by fragmentation group and represent the frequency of different MHC class II β alleles in each population as indicated by the relevant colors). Population codes are as follows: AK—Antikeros, AN—Andreas, DA—Daskalio, GL—Glaronissi, IR—Irakleia, KE—Keros, KO—Koufonissi, KP—Kopria, MA—Megalos Ambelas, MK—Makronissi, MP—Megali Plaka, NX—Naxos, OV—Ovriokastro, SK—Schoinoussa. Numbers refer to individual Poer‐DAB* alleles

Figure 3

Frequency distributions of alleles per individual (A _i) estimates for each island population of Podarcis erhardii

Figure 4

Phylogeny of major histocompatibility complex class II β (exon 2) in Podarcis erhardii and related taxa. The numbers at each node represent posterior probability support, and branch lengths are proportional to number of base pair substitutions

Figure 5

Regression of either island age or area against MHC and microsatellite variability. (a) Island age versus A _i (p = .0058, R ² = .44) and H _e (p = .003, R ² = .50). (b) The log of island area versus AR _msat and AR_MHC (p = .039, R ² = .25), and AR _msat (p = .0016, R ² = .54). (c) Island age versus A _i (p = .39, R ² = −.015) and H _e (p = .31, R ² = .01). (d) The log of island area versus AR_MHC. (p = .903, R ² = −.082) AR _msat (p = .62, R ² = −.06). Regression lines for MHC are shown with a solid line and for microsatellite variation with a dotted line. Thickened lines indicate a significant relationship

Figure 6

Scatter plots of Podarcis erhardii microsatellite and MHC variation in relation to island age and log Area. (a) A _i (A _i × Age R ² = −.483; A _i × Log area R ² = .001). (b) AR_MHC (AR_MHC × Age R ² = −.063; AR_MHC × Log area = 0.308). (c) H _e (H _e × Age R ² = −.143; H _e × Log area R ² = .481). (d) AR _msat × Log area = 0.624; AR _msat × Age R ² = −.2932)