Limits...
Genetic connectivity across marginal habitats: the elephants of the Namib Desert

View Article: PubMed Central - PubMed

ABSTRACT

Locally isolated populations in marginal habitats may be genetically distinctive and of heightened conservation concern. Elephants inhabiting the Namib Desert have been reported to show distinctive behavioral and phenotypic adaptations in that severely arid environment. The genetic distinctiveness of Namibian desert elephants relative to other African savanna elephant (Loxodonta africana) populations has not been established. To investigate the genetic structure of elephants in Namibia, we determined the mitochondrial (mt) DNA control region sequences and genotyped 17 microsatellite loci in desert elephants (n = 8) from the Hoanib River catchment and the Hoarusib River catchment. We compared these to the genotypes of elephants (n = 77) from other localities in Namibia. The mtDNA haplotype sequences and frequencies among desert elephants were similar to those of elephants in Etosha National Park, the Huab River catchment, the Ugab River catchment, and central Kunene, although the geographically distant Caprivi Strip had different mtDNA haplotypes. Likewise, analysis of the microsatellite genotypes of desert‐dwelling elephants revealed that they were not genetically distinctive from Etosha elephants, and there was no evidence for isolation by distance across the Etosha region. These results, and a review of the historical record, suggest that a high learning capacity and long‐distance migrations allowed Namibian elephants to regularly shift their ranges to survive in the face of high variability in climate and in hunting pressure.

No MeSH data available.


Related in: MedlinePlus

Genetic analyses of Namibian elephants using genotypes at 17 microsatellite loci. (A) Spatial genetic autocorrelograms of 55 Namibian elephants, implemented using the software GenAlEx 6.5 (Peakall and Smouse 2012). The genetic similarity between pairs of individuals (y‐axis) is shown relative to their geographic separation (x‐axis) (Peakall and Smouse 2012). The geographic distances between all possible pairs of individual were divided into quintiles (five groups each of the same size). r: spatial autocorrelation coefficient. U: upper 95% randomization limits of r. L: lower 95% randomization limits of r. In Namibian elephants, spatial distance and genetic distance were not correlated and isolation by distance was not observed at any distance class. (B) Principal coordinate analysis showing the genetic relationship of Namibian desert elephants (n = 4) to Etosha elephants (n = 51) performed on the genetic distance matrix. Only the first and second coordinates are shown here; neither differentiated between desert elephants and Etosha elephants, nor did the third coordinate (not shown).
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016642&req=5

ece32352-fig-0003: Genetic analyses of Namibian elephants using genotypes at 17 microsatellite loci. (A) Spatial genetic autocorrelograms of 55 Namibian elephants, implemented using the software GenAlEx 6.5 (Peakall and Smouse 2012). The genetic similarity between pairs of individuals (y‐axis) is shown relative to their geographic separation (x‐axis) (Peakall and Smouse 2012). The geographic distances between all possible pairs of individual were divided into quintiles (five groups each of the same size). r: spatial autocorrelation coefficient. U: upper 95% randomization limits of r. L: lower 95% randomization limits of r. In Namibian elephants, spatial distance and genetic distance were not correlated and isolation by distance was not observed at any distance class. (B) Principal coordinate analysis showing the genetic relationship of Namibian desert elephants (n = 4) to Etosha elephants (n = 51) performed on the genetic distance matrix. Only the first and second coordinates are shown here; neither differentiated between desert elephants and Etosha elephants, nor did the third coordinate (not shown).

Mentions: We conducted a spatial autocorrelation analysis to examine the association between the genetic differences between pairs of individuals and their geographic separation. For the 51 “nondesert” elephants, geographic distances based on the geographic coordinates were computed between each pair of elephants, with the distances placed into even quintiles (x‐axis in Fig. 3A). Genetic distances between pairs of elephants were also determined (y‐axis in Fig. 3A). Both permutation and bootstrap tests did not detect significant spatial genetic autocorrelations in Namibian elephants at any distance class (Fig. 3A). This indicated that gene flow across Namibian elephants was recent and sufficient enough to prevent the development of patterns that would have been suggestive of isolation by distance.


Genetic connectivity across marginal habitats: the elephants of the Namib Desert
Genetic analyses of Namibian elephants using genotypes at 17 microsatellite loci. (A) Spatial genetic autocorrelograms of 55 Namibian elephants, implemented using the software GenAlEx 6.5 (Peakall and Smouse 2012). The genetic similarity between pairs of individuals (y‐axis) is shown relative to their geographic separation (x‐axis) (Peakall and Smouse 2012). The geographic distances between all possible pairs of individual were divided into quintiles (five groups each of the same size). r: spatial autocorrelation coefficient. U: upper 95% randomization limits of r. L: lower 95% randomization limits of r. In Namibian elephants, spatial distance and genetic distance were not correlated and isolation by distance was not observed at any distance class. (B) Principal coordinate analysis showing the genetic relationship of Namibian desert elephants (n = 4) to Etosha elephants (n = 51) performed on the genetic distance matrix. Only the first and second coordinates are shown here; neither differentiated between desert elephants and Etosha elephants, nor did the third coordinate (not shown).
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016642&req=5

ece32352-fig-0003: Genetic analyses of Namibian elephants using genotypes at 17 microsatellite loci. (A) Spatial genetic autocorrelograms of 55 Namibian elephants, implemented using the software GenAlEx 6.5 (Peakall and Smouse 2012). The genetic similarity between pairs of individuals (y‐axis) is shown relative to their geographic separation (x‐axis) (Peakall and Smouse 2012). The geographic distances between all possible pairs of individual were divided into quintiles (five groups each of the same size). r: spatial autocorrelation coefficient. U: upper 95% randomization limits of r. L: lower 95% randomization limits of r. In Namibian elephants, spatial distance and genetic distance were not correlated and isolation by distance was not observed at any distance class. (B) Principal coordinate analysis showing the genetic relationship of Namibian desert elephants (n = 4) to Etosha elephants (n = 51) performed on the genetic distance matrix. Only the first and second coordinates are shown here; neither differentiated between desert elephants and Etosha elephants, nor did the third coordinate (not shown).
Mentions: We conducted a spatial autocorrelation analysis to examine the association between the genetic differences between pairs of individuals and their geographic separation. For the 51 “nondesert” elephants, geographic distances based on the geographic coordinates were computed between each pair of elephants, with the distances placed into even quintiles (x‐axis in Fig. 3A). Genetic distances between pairs of elephants were also determined (y‐axis in Fig. 3A). Both permutation and bootstrap tests did not detect significant spatial genetic autocorrelations in Namibian elephants at any distance class (Fig. 3A). This indicated that gene flow across Namibian elephants was recent and sufficient enough to prevent the development of patterns that would have been suggestive of isolation by distance.

View Article: PubMed Central - PubMed

ABSTRACT

Locally isolated populations in marginal habitats may be genetically distinctive and of heightened conservation concern. Elephants inhabiting the Namib Desert have been reported to show distinctive behavioral and phenotypic adaptations in that severely arid environment. The genetic distinctiveness of Namibian desert elephants relative to other African savanna elephant (Loxodonta africana) populations has not been established. To investigate the genetic structure of elephants in Namibia, we determined the mitochondrial (mt) DNA control region sequences and genotyped 17 microsatellite loci in desert elephants (n = 8) from the Hoanib River catchment and the Hoarusib River catchment. We compared these to the genotypes of elephants (n = 77) from other localities in Namibia. The mtDNA haplotype sequences and frequencies among desert elephants were similar to those of elephants in Etosha National Park, the Huab River catchment, the Ugab River catchment, and central Kunene, although the geographically distant Caprivi Strip had different mtDNA haplotypes. Likewise, analysis of the microsatellite genotypes of desert‐dwelling elephants revealed that they were not genetically distinctive from Etosha elephants, and there was no evidence for isolation by distance across the Etosha region. These results, and a review of the historical record, suggest that a high learning capacity and long‐distance migrations allowed Namibian elephants to regularly shift their ranges to survive in the face of high variability in climate and in hunting pressure.

No MeSH data available.


Related in: MedlinePlus