Population genetic analysis of the recently rediscovered Hula painted frog (Latonia nigriventer) reveals high genetic diversity and low inbreeding

Abstract

After its recent rediscovery, the Hula painted frog (Latonia nigriventer) has remained one of the world's rarest and least understood amphibian species. Together with its apparently low dispersal capability and highly disturbed niche, the low abundance of this living fossil calls for urgent conservation measures. We used 18 newly developed microsatellite loci and four different models to calculate the effective population size (N_e) of a total of 125 Hula painted frog individuals sampled at a single location. We compare the N_e estimates to the estimates of potentially reproducing adults in this population (N_ad) determined through a capture-recapture study on 118 adult Hula painted frogs captured at the same site. Surprisingly, our data suggests that, despite N_ad estimates of only ~234-244 and N_e estimates of ~16.6-35.8, the species appears to maintain a very high genetic diversity (H_O = 0.771) and low inbreeding coefficient (F_IS = -0.018). This puzzling outcome could perhaps be explained by the hypotheses of either genetic rescue from one or more unknown Hula painted frog populations nearby or by recent admixture of genetically divergent subpopulations. Independent of which scenario is correct, the original locations of these populations still remain to be determined.

Figure 1

Estimated genetic clustering (K = 2) of 134 Hula painted frog individuals captured at two close-by locations in northern Israel as obtained using Bayesian analysis implemented in STRUCTURE. Each bar represents an individual; * = individuals captured within the Hula Nature Reserve (individuals captured outside the reserve are unmarked); ♦ = individuals assigned to the other cluster using the Likelihood Relatedness analysis as implemented in ML-RELATE.

Figure 2

Dendrogram showing the relationship among 134 Hula painted frog individuals captured at two close-by locations in northern Israel. Hierarchical clustering was performed on the likelihood relatedness output obtained by ML-RELATE. Approximately unbiased (au) P values (%), as computed by the R package pvclust, are given for each node in the dendrogram (distance: euclidean, cluster method: ward.D2). Colours of branches were chosen to match the clusters obtained by STRUCTURE analysis (see Fig. 1); * = individuals captured within the Hula Nature Reserve (individuals captured outside the reserve are unmarked).

Figure 3

Digital photographs of the ventral spot pattern of three adult Hula painted frog individuals. (a) Example images of two adult female individuals with snout-vent lengths of 98.5 mm (left) and 88.0 mm (middle), and an adult male individual with a snout-vent length of 106.0 mm (right) as used by simple eye matching. The yellow arrows indicate examples of spots with a distinct shape and the blue lines indicate examples of characteristic strings of spots (to the right of the blue lines) that, together with the general appearance of the spot pattern, were used for the identification of recaptured individuals. (b) Cropped photographs of the same individuals (same order as above) as used for the automatic identification with Wild-ID. Photos by RGBP.