The authors have declared that no competing interests exist.
Conceived and designed the experiments: MIMP MDLL FP JPMC. Performed the experiments: MIMP. Analyzed the data: MIMP MDLL RG FP JPMC. Contributed reagents/materials/analysis tools: MIMP MDLL RG FP JPMC. Wrote the paper: MIMP MDLL RG FP JPMC.
The grasshopper
Supernumerary chromosomes, also called accessory or B chromosomes, are additional chromosomes found in some individuals at some populations of about 15% of eukaryotic species
Gene flow may be directly analyzed by demographic studies, e.g. capture-mark-recapture methods, but it can also be indirectly estimated through population genetic analysis by molecular markers. In this latter case, the calculation of Wright's indices to estimate inbreeding at several hierarchical levels (
ISSR markers have been used for a variety of genetic and population-genetic purposes, such as: the characterization of different varieties in maize
One of the main caveats of ISSR markers, and other markers based on PCR amplification, is allelic dropout, i.e. a stochastic failure of PCR amplification in one of the alleles carried by an individual at a given locus, leading to the determination of a wrong genotype
In this paper, we analyze several parameters of population genetics and structure in 10 natural populations of the grasshopper
The six primers employed yielded a total of 97 ISSR markers ranging in size from 180 to >2000 bp in the 10 populations as a whole, with 99% of the loci being polymorphic. Per region, however, the percentage of polymorphic loci fell to 77.9% (SD = 3.13) in the southern populations and 76.7% (SD = 1.78) in the eastern ones (
Polymorphic loci | |||
Region | Population | Number | % |
South | Algarrobo | 76 | 78.4 |
Torrox | 79 | 81.4 | |
Nerja-0 | 78 | 80.4 | |
Nerja-2 | 73 | 75.3 | |
Salobreña | 72 | 74.2 | |
Average | 75.6 (3.05) | 77.9 (3.13) | |
East | Mundo | 76 | 78 |
Claras | 76 | 78.4 | |
Socovos | 74 | 76 | |
Calasparra | 72 | 74 | |
Caravaca | 74 | 76 | |
Average (SD) | 74.4 (1.67) | 76.5 (1.78) |
Data analysis with Hickory v1.1
Parameter | Mean | SD | 2.50% | 97.50% |
theta-I, θ(I) | 0.299 | 0.021 | 0.26 | 0.344 |
theta-II, θ(II) | 0.129 | 0.008 | 0.114 | 0.145 |
theta-III, θ(III) | 0.098 | 0.004 | 0.091 | 0.105 |
hs[Algarrobo] | 0.229 | 0.005 | 0.219 | 0.239 |
hs[Torrox] | 0.23 | 0.005 | 0.219 | 0.24 |
hs[Nerja-0] | 0.218 | 0.005 | 0.208 | 0.228 |
hs[Nerja-2] | 0.228 | 0.005 | 0.219 | 0.238 |
hs[Salobreña] | 0.229 | 0.005 | 0.218 | 0.24 |
hs[Mundo] | 0.235 | 0.007 | 0.221 | 0.249 |
hs[Claras] | 0.22 | 0.007 | 0.207 | 0.233 |
hs[Socovos] | 0.205 | 0.006 | 0.193 | 0.216 |
hs[Calasparra] | 0.21 | 0.005 | 0.199 | 0.22 |
hs[Caravaca] | 0.216 | 0.007 | 0.203 | 0.23 |
Hs | 0.222 | 0.002 | 0.218 | 0.226 |
Ht | 0.253 | 0.002 | 0.249 | 0.257 |
Gst-B | 0.123 | 0.005 | 0.114 | 0.133 |
2.5% and 97.5% show confidence interval
Note that hs values are estimates of genetic diversity (panmictic or expected heterozygosity) for each population sample, whereas Hs is the average of hs values for all population samples. Ht is the heterozygosity that would be observed if all population samples would come from a single population.
The Structure software, which analyzes genetic structure through a Bayesian approach, showed that the number of genetic groups (K value) best fitting our data, inferred following Evanno
These results were found using all 97 ISSR markers analyzed. Note the highest peak for K = 2.
Each vertical bar represents one of the 255 individuals analyzed. Group 1 (southern region) is represented in red color, and Group 2 (eastern region) is shown in green color. Bar length is proportional to the inferred ancestry values into each group for each individual.
Population | N | Group 1 | Group 2 |
Algarrobo | 29 | 0.940 | 0.060 |
Torrox | 27 | 0.964 | 0.036 |
Nerja-0 | 30 | 0.968 | 0.032 |
Nerja-2 | 30 | 0.962 | 0.038 |
Salobreña | 23 | 0.949 | 0.051 |
Mundo | 15 | 0.170 | 0.830 |
Claras | 21 | 0.058 | 0.942 |
Socovos | 27 | 0.037 | 0.963 |
Calasparra | 30 | 0.091 | 0.909 |
Caravaca | 23 | 0.064 | 0.936 |
The highest assignment proportion to each group is marked in bold.
The Mantel test revealed significant isolation by distance (IBD) when we used pairwise-Fst values (r = 0.90, P = 0.0004) or Dice's dissimilarity indices (r = 0.68, P = 0.00003) for genetic distance (see genetic and geographical distances in
An unrooted neighbor-joining phylogenetic tree built with the
This was built with Pairwise-Fst genetic distances determined with the 97 ISSR markers analyzed. Note the two groups of populations corresponding to the southern (upper half) and eastern (lower half) regions. All populations harbored B chromosomes excepting Caravaca, Socovos and Claras.
More than half of the ISSR markers (51 out of 97) displayed consistent patterns under the two controls performed, but 3 markers failed to show repeatability under control 1 only, 40 failed for control 2 only, and 4 markers failed for both controls (
A total of 101 progeny analyses (PAs) were performed to test inheritance of 23 ISSR markers (
To analyze the possible incidence of allelic dropout on population genetics and structure results, we repeated calculations with Hickory and Structure after discarding the 10 markers showing high incidence of dropout detected by the PCR repeatability analysis (26-500, 26-650, 26-750, 26-1000, 39-250 and 43-540) or inheritance analysis (7-1000, 7-1100, 7-1400 and 14-750). The population parameters estimated with the 87 remaining markers, listed in
The analysis of population structure with the 87 ISSR markers showing slight dropout or none at all, with the Structure software, gave very similar results to those found when including all the markers, with K = 2 best fitting our data, and high proportions of individuals from the southern populations being included in one group and those from the eastern populations in the other group (see
Even with only 46 markers, Structure showed that K = 2 best fitted our data, and the genetic differentiation between the southern and eastern regions was clearly evident, with high proportions of individual assignment to the correct region (see
As in other organisms previously analyzed
Hs (i.e. the average for hs in all populations) was lower than Ht (expected heterozygosity in all populations as a whole), suggesting a possible heterozygous deficiency which could be explained by allelic dropout, inbreeding or population subdivision. As explained above, dropout effects are negligible in this case, and inbreeding is unlikely since
The significant values found for θ(II) and Gst-B, and the highest variance between regions, compared to within regions, revealed the existence of population subdivision for the analyzed areas. Population structure was also shown by the Structure software. A K of 2 best explained the data, with population groups being well defined in coincidence with the southern and eastern regions, so that more than 90% of individuals were properly assigned to them. Remarkably, our analysis of population genetics and structure parameters, after discarding those markers showing allelic dropout, showed almost identical results to those performed with the complete data set, suggesting that allelic dropout has a minor effect on these estimations. Therefore, although ISSR, and other markers being affected by allelic dropout, become less relevant with the advent of new sequencing technologies and genotyping methods, they still constitute an efficient and low-cost method of revealing population structure, especially in non-model organisms.
In the 10 populations of the grasshopper
The only previous analysis of FST and IBD in a grasshopper species, performed in the South American grasshopper
The high gene flow inferred from the low GST values observed for these ten populations of
Whereas the five populations from the southern region (Algarrobo, Torrox, Nerja-0, Nerja-2 and Salobreña) all carried B chromosomes, only two (Mundo and Calasparra) from the eastern region carried them. In this latter region, the observed pairwise Fst values (see
Specimens of the grasshopper
Altitude | ISSR6 | ISSR7 | ISSR14 | ISSR26 | ISSR39 | ISSR43 | ||||||||||
Population | Province | Region | UTM (X–Y) | (m) | Nm | Nf | Nm | Nf | Nm | Nf | Nm | Nf | Nm | Nf | Nm | Nf |
Algarrobo | Malaga | South | 406506.77 – 4067270.76 | 4 | 28 | – | 29 | – | 29 | – | 28 | – | 26 | – | 28 | – |
Torrox | Malaga | South | 414886.32 – 4066101.15 | 37 | 26 | – | 23 | – | 27 | – | 25 | – | 18 | – | 25 | – |
Nerja-0 | Malaga | South | 419750.87 – 4066921.35 | 3 | 30 | – | 29 | – | 29 | – | 28 | – | 26 | – | 28 | – |
Nerja-2 | Malaga | South | 423550.39 – 4068087.00 | 79 | 29 | – | 28 | – | 29 | – | 29 | – | 26 | – | 30 | – |
Salobreña | Granada | South | 447096.84 – 4066013.37 | 1 | 23 | – | 21 | – | 23 | – | 23 | – | 21 | – | 22 | – |
Mundo | Albacete | East | 605562.49 – 4258313.37 | 450 | 15 | – | 15 | – | 15 | – | 15 | – | 10 | – | 15 | – |
Claras | Albacete | East | 568336.33 – 4242425.03 | 640 | 20 | – | 20 | – | 20 | – | 20 | – | 20 | – | 7 | – |
Socovos | Albacete | East | 590227.22 – 4242649.56 | 655 | 22 | 5 | 21 | 4 | 20 | 5 | 21 | 5 | 21 | 4 | 9 | 3 |
Calasparra | Murcia | East | 614230.74 – 4234235.38 | 260 | 28 | – | 29 | – | 28 | – | 27 | – | 27 | – | 30 | – |
Caravaca | Murcia | East | 602055.17 – 4218476.58 | 570 | 13 | 10 | 12 | 10 | 12 | 10 | 13 | 9 | 11 | 9 | – | – |
Total | 234 | 15 | 227 | 14 | 232 | 15 | 229 | 14 | 206 | 14 | 194 | 3 |
Nf = Number of females; Nm = Number of males.
ISSR markers were amplified by polymerase chain reaction (PCR) using the primers shown in
To test the incidence of PCR errors generating the absence of bands suggesting false recessive homozygotes (allelic dropout), we performed two types of replicates for the same individual, one where the two replicates shared the same master mix and were amplified at the same time (control 1), and another where the replicates were made from different master mixes and were amplified on different days (control 2). The frequency of allelic dropout was quantified as a percentage of individuals showing inconsistent genotypes in the two replicates.
ISSR data were analyzed by Bayesian methods, using the Hickory v1.1 software
The existence of isolation by distance (IBD) was analyzed by means of a Mantel Test comparing the matrix of pairwise-Fst values with the matrix of geographical distances between the sampled populations, with the Zt-win software
Also, we investigated IBD by using Dice's dissimilarity index
To elucidate the population structure in
With the genetic matrix distance (pairwise-Fst), estimated using the program AFLP-surv from the ISSR marker data, we inferred a phylogenetic tree under the Neighbor-Joining method
To identify genetically homogeneous groups within our samples, we used a Bayesian algorithm implemented in the Structure v2.3.1 software
To analyze inheritance of ISSR markers, we performed 10 controlled crosses analyzing the markers derived from one or more of the ISSR6, ISSR7, and ISSR14 primers (
Inheritance expectations for autosomal markers, shown in
(TIF)
(TIF)
(TIF)
(TIF)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)