Local adaptation to environmental stress at different levels of genetic polymorphism in various plants and animals has been documented through evolution of heavy metal tolerance. We used samples of Drosophila subobscura populations from two differently polluted environments to analyze the change of chromosomal inversion polymorphism as genetic marker during laboratory exposure to lead. Exposure to environmental contamination can affect the genetic content within a particular inversion and produce targets for selection in populations from different environments. The aims were to discover whether the inversion polymorphism is shaped by the local natural environments, and if lead as a selection pressure would cause adaptive divergence of two populations during the multigenerational laboratory experiment. The results showed that populations retain signatures from past contamination events, and that heavy metal pollution can cause adaptive changes in population. Differences in inversion polymorphism between the two populations increased over generations under lead contamination in the laboratory. The inversion polymorphism of population originating from the more polluted natural environment was more stable during the experiment, both under conditions with and without lead. Therefore, results showed that inversion polymorphism as a genetic marker reflects a strong signature of adaptation to the local environment, and that historical demographic events and selection are important for both prediction of evolutionary potential and long-term viability of natural populations.
Citation: Kenig B, Kurbalija Novičić Z, Patenković A, Stamenković-Radak M, Anđelković M (2015) Adaptive Role of Inversion Polymorphism of Drosophila subobscura in Lead Stressed Environment. PLoS ONE 10(6): e0131270. https://doi.org/10.1371/journal.pone.0131270
Editor: Owain Rhys Edwards, CSIRO, AUSTRALIA
Received: February 19, 2015; Accepted: May 31, 2015; Published: June 23, 2015
Copyright: © 2015 Kenig et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data have been deposited to figshare (http://dx.doi.org/10.6084/m9.figshare.1275137).
Funding: This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (http://www.mpn.gov.rs), Contract No. 173012. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Anthropogenic pollutants in the environment, including heavy metals, affect natural populations of organisms in many ways. One of the most significant outcomes with potentially large long-term effects is genetic change. Exposure to pollutants can cause selection pressure which may lead to genetic adaptation [1, 2]. Evidence for local adaptation to environmental stress on various levels of genetic polymorphism of different species has been provided in numerous empirical works, with documented evolution of heavy metal tolerance [3–5].
Chromosomal inversion polymorphism is known to be associated with environmental changes and local adaptation [6–12]. Several hypotheses have been suggested to explain the maintenance of inversion polymorphism in natural populations (reviewed in Hoffmann and Rieseberg ). The traditional explanation for the spread and evolution of the inversions by Dobzhansky  is that alleles within inversions have epistatic effects on fitness, and that combinations of the alleles are “coadapted” by having higher fitness than predicted from the sum of their independent effects. An alternative hypothesis states that inversions are favored because they bring together locally adapted alleles even without epistasis . Both of these hypotheses explain that the reduced recombination between inversion heterokaryotypes avoids gene exchange with other genetic backgrounds. The suppression of recombination in heterokaryotypes within or close to inverted regions may facilitate the capture of locally adaptive alleles across multiple linked loci, thus allowing inversions to spread through a local population [6, 11, 13].
Drosophila subobscura Collin (Diptera: Drosophilidae) possesses one of the richest inversion polymorphism in the genus, on all five acrocentric chromosomes with more than 60 different inversions, forming more than 90 different chromosomal gene arrangements described. Some of the inversions are rare and/or restricted to limited areas . Inversion polymorphism in D. subobscura is to a certain degree temporally and spatially associated with the variation and dynamics of ecological factors, and it shows clear-cut geographic, habitat-associated, and seasonal variations [9, 14–17]. Therefore, chromosomal inversions of D. subobscura are adaptive in response to environmental changes and represent a suitable model for studying adaptation to a changing environment. Moreover, Drosophila studies allow for detailed multigenerational genetic analyses in precisely controlled conditions , which facilitates better assessment of the effects under long-term chronic changes [19, 20]. The effect of history of heavy metal pollution on population fitness response to laboratory lead contamination in D. subobscura showed that a higher resistance was present in a population originating from the more polluted locality . It was also found that the extent of inversion polymorphism was higher in urban Drosophila populations, which was a result of the large number of ecological niches in this type of environment [22, 23]. However, only few studies have analyzed how historical differences in inversion polymorphisms affect the ability of populations to respond to new environmental conditions, although the prediction of population-specific genetic response to heavy metal pollution is one of the most important tasks of ecotoxicology [24, 25].
The present study analyzes the dynamics and variation of inversion polymorphisms in D. subobscura flies originating from two natural populations, sampled from ecologically distinct habitats. The two populations were maintained in laboratory conditions with different levels of lead concentrations during six generations. The aims were:
- to analyze initial population differentiation regarding the inversion frequencies, and monitor whether such differentiation would be retained over generations in the experimental conditions or whether the selection pressure imposed by lead would cause a convergence of populations to a similar level of inversion polymorphism.
- to detect the level of inversion polymorphism within and between populations maintained on different lead concentration across generations.
- to check if heterozygosity (percent of heterokaryotypes) would remain the same over generations or it would change (increase or decrease) under the lead pollution, and furthermore, whether this change would differ between populations.
Materials and Methods
Two populations of Drosophila subobscura were sampled, one at the locality of Deliblato Sands (DS) (Orno-Quercetum cerris-virgiliane), about 60 km northeast from Belgrade, Serbia (44° 49' N; 21° 07' E), and the other in the Botanical Garden (BG) (Arboretum), in an urban part of Belgrade, capital of Serbia (44° 49' N; 20° 28' E). Verbal permissions for collecting the flies were obtained for both localities. For the Botanical garden, permission was obtained from the management of the Botanical Garden "Jevremovac", The University of Belgrade, and at the locality Deliblato Sands from the Forest Office “Deliblato”. The species studied is not classified as endangered or protected, the localities are not on protected land, and therefore no additional specific permissions were required for collection.
The literature data for lead concentration in soil sampled in DS (5–35 mg/kg) characterize this particular locality as unpolluted . The BG locality has chronically been exposed to anthropogenic activity, mostly related to traffic pollution. The soil in this area is heavily polluted, with the average concentration of lead in the soil of 298.6 mg/kg [27, 28].
The flies were collected by using fermented fruit traps. Approximately 100 isofemale lines (IF) per population were made, each from a single gravid wild-caught female. All lines were maintained and all experiments performed under constant laboratory conditions: temperature of 19°C, approx. 60% relative humidity, light of 300 lux and 12/12 h light/dark cycles. After two generations, three pairs of males and females from each IF line were used to establish two synthetic mass populations in order to preserve the original genetic variability. The F1 progeny of the mass populations was used to establish three experimental groups with three different media. The control group was reared on the standard Drosophila medium (water/cornmeal/yeast/sugar/agar/nipagine), without lead. In order to determine the toxic potential of lead acetate, eight different concentrations of lead acetate were prepared in triplicate for treating D. subobscura eggs. Exactly 100 eggs per concentration were used and the number of emerged adults was recorded. Based on the results obtained, a survival curve was made, and two concentrations of lead acetate [Pb (CH3COO)2·3H2O] were chosen. The first experimental group (low lead concentration group–LLC) had 10 μg/mL of lead acetate added to the medium, which is a sublethal concentration. The second experimental group (high lead concentration group–HLC) had 100 μg/mL of lead acetate added to the medium, which represented 20% of the lethal concentration (LC 20). Flies in all three groups (control, LLC, HLC) were kept en masse, each group consisting of 10 bottles (replicates) containing 25 mL of the medium. The flies from the F1 generation were mixed randomly within each group, and the F2 generation was initiated with 50 adult flies in an equal sex ratio for each bottle (500 adults in total per experimental group). Adult females were allowed to lay eggs for the first seven days after they were placed into the bottles. After that period, the adults were removed, and newly emerged adult flies were used to initiate the next generation. Each subsequent generation was obtained exactly in the same manner.
It was shown by Kenig et al  that these two populations initially showed different resistance to lead in laboratory conditions. Egg to adult viability was higher in the Botanical Garden population compared to Deliblato Sands population, and the difference was more expressed on media with both low and high concentrations of lead than on the control medium. Furthermore, the viability increased across generations in the HLC group from DS, while in the corresponding group from BG there were no significant changes across generations (viability remained high in all generations). Therefore, the initial higher resistance to contamination, meaning higher fecundity, higher viability, and faster egg to adult development in all the generations studied, was found in D. subobscura originating from the locality of Botanical Garden.
An analysis of inversion polymorphism was carried out with males from DS and BG populations in F1, F3, and F6 generations, from each experimental group (control, LLC and HLC). The males were individually crossed with virgin females from the Küsnacht laboratory line, which is homokaryotypic for all acrocentric chromosomes of the set (AST, JST, UST, EST, and OST). Salivary glands from third-instar larvae were squashed and their chromosomes stained with aceto-orcein solution. In order to minimize the possibility of error in determining the karyotype and to reduce the probability of wrong determination to (1/2)8 for each chromosome, we analyzed eight larvae from the progeny of each cross. For the cytological analysis of chromosome arrangements, the chromosome map of Kunze-Mühl and Müller  was used, as well as nomenclature according to Kunze-Mühl and Sperlich .
About 30 males (240 autosomes and 30 sex chromosomes) from each experimental group (C, LLC, HLC) in three generations (F1, F3 and F6), in both populations (DS and BG) were analyzed, making a total of about 420 males.
A G-test was done to determine the homogeneity of chromosome arrangement frequencies within all five chromosomes, in pairwise comparisons among experimental groups (C, LLC and HLC), within generations (F1, F3 and F6), for each population (DS and BG). Z-statistics was used to assess the differences in frequencies of individual chromosome arrangements. Both tests were performed in pairwise comparisons among experimental groups and among generations, for each population (intra-population differences), and between the populations (inter-population differences). The inversion polymorphism parameters—the degree of heterozygosity (HZ) and Index of free recombination (IFR) were derived from arrangement frequencies according to the description of Krimbas . The IFR takes in consideration the length of the chromosome region within inversions, but there is considerable evidence that inversions affect recombination beyond their breakpoints . Therefore, HZ may be a preferable metric for the degree of inversion polymorphism. A sequential Bonferroni test  was used to adjust for the multiple pairwise comparisons within each experimental group.
Intra-population differences in parameters of chromosomal polymorphism
The chromosome arrangement frequencies, heterozygosity, and the index of free recombination (IFR) were calculated for D. subobscura from Deliblato Sands (Table 1) and Botanical Garden (Table 2) populations, in three generations (F1, F3 and F6) on standard medium (control group) and media with two lead concentrations (LLC and HLC). The inversion polymorphism parameters (heterozygosity and IFR) were similar in all experimental groups and generations, except in the controls from F3 and F6, which showed a lower value of IFR and an increase in heterozygosity compared to the F1 generation. Trends of variation of these parameters were found to be similar in both populations (Tables 1 and 2).
The statistical significance for differences in chromosomal arrangement frequencies between generations (F1, F3 and F6) within experimental groups (C, LLC and HLC) for both populations (DS and BG) is given in Table 3. Significant differences between generations were obtained for the DS population in all experimental groups. In the control group, differences were found between F3 and F6 for complex gene arrangements E1+2+9 (p < 0.05), O3+4 (p < 0.001) and O3+4+1 (p < 0.001), and between F1 and F6 generations for E8 inversion (p < 0.05). In the LLC group, significant differences between F3 and F6 generations were found only in frequencies of two E chromosome arrangements: ESt (p < 0.001) and E8 (p < 0.001). The highest intergenerational difference between F3 and F6 was found for HLC groups, in frequencies of arrangements on chromosomes U (USt: p < 0.05; U1+2: p < 0.01), E (E8: p < 0.001; E1+2+9: p < 0.001) and the O3+4 arrangement (p < 0.01) (Table 3).
The population from BG showed significant differences only between generations F3 and F6: within the LLC group for the O3+4+2 gene arrangement (p < 0.05), and in the HLC group for two E chromosome arrangements (ESt: p < 0.01; E8: p < 0.05) (Table 3).
Differences in the frequencies of chromosome arrangements among experimental groups (C, LLC, HLC) within generations (F1, F3, F6) from DS and BG are shown in S1 Table. The population from DS showed a significantly lower frequency of the complex chromosome arrangement O3+4+1 in the LLC group within generation F3 when compared to the control group (p < 0.05). Significant differences were found in F3 between the control and the HLC group, for frequencies of chromosome arrangements on chromosome A (ASt: p < 0.05; A1: p < 0.05) and chromosome J (JSt: p < 0.01; J1: p < 0.01).
A slightly different result was obtained within the F6 generation. Significant differences were found in frequencies of arrangements on chromosomes U, O, and the most significant on E, among experimental groups. The increase of the E8 frequency is positively associated with the lead concentration in the medium, revealed by a significant difference in the frequency of this inversion between the control and both groups with lead (p < 0.01 for LLC; p < 0.001 for HLC). The HLC group had a significantly lower frequency of E1+2+9 compared to both the control and the LLC group (p < 0.001). A significant difference in frequency of the U1+2+6 was detected between the control and the HLC group (p < 0.01) (S1 Table). Another significant difference in frequencies of the standard gene arrangement on chromosome O was registered between the control and the HLC group (p < 0.01).
When the experimental groups from the BG population within F3 and F6 generations were compared, a few statistically significant differences were found. Within the F3 generation, a significantly lower frequency of A1 chromosome inversion in HLC group was detected compared to the control (p < 0.05) and LLC group (p < 0.01). Within the F6 generation the only significant increase was obtained in the frequency of the EST arrangement in the HLC group, compared to the control (p < 0.01) (S1 Table).
In both populations (S2 and S3 Tables) significant differences in chromosome arrangement frequencies (per chromosome) were found between the experimental groups (C, LLC and HLC) within generations (F3 and F6). The results of the G test for both populations are in accordance with the results of the Z-statistics.
Inter-population differences in parameters of chromosomal polymorphism
Chromosomal polymorphism analysis of the initial generation (F1) showed that the Deliblato Sands population had 14 structural types, while there were 16 in the Botanical Garden population. This evidence pointed to a higher level of chromosomal variability in the BG population, which was indirectly confirmed by differences in IFR values, where BG population had lower IFR (81.581) compared to DS (85.074).
In the control groups of F3 generation, significant differences were found in the distribution of chromosome O gene arrangements between the two populations (Table 4). The arrangement O3+4 had a higher frequency in the control group from BG (p < 0.05), O3+4+1 had higher frequency in DS (p < 0.001), while O3+4+2 was completely absent in the control group from DS, but present in BG in low frequency (p < 0.05). Significant differences between arrangements of chromosome O were also confirmed by the G-test (p < 0.001) as shown in S4 Table. In the F6 generation the G-test revealed differences between two populations (p < 0.05) in the distribution of gene arrangements on the U chromosome in the control groups. Standard arrangements of chromosomes U and E had lower frequency in population DS, and the Z-test confirmed significance of these interpopulation differences: USt (p < 0.01) and ESt (p < 0.05) (Table 4). The heterozygosity on the control in the F3 generation was higher (0.675) in the BG population than in DS (0.570). The IFR value in the DS control group was 78.889, which is higher than the one obtained for BG (76.806). The heterozygosity was higher by 12.5% in the BG control group in the F6 generation (0.703) than in the corresponding group from DS (0.578). Furthermore, the IFR values were very different, and the value of this parameter in the DS control was 80.355, while in the corresponding group from BG it was 75.373 (Tables 1 and 2).
Between the LLC groups of DS and BG populations, in the F3 generation, there were significant differences only for gene arrangement O3+4+2 (p<0.05) due to the absence of this arrangement in the DS group (Table 4). The G-test did not reveal any significant difference between these groups from two populations in F3 for any of the analyzed chromosomes (S4 Table). In the F6 generation there were significant differences in frequencies of gene arrangements of chromosome E between two populations on LLC (Table 4). Arrangement ESt had a significantly lower frequency in the LLC group from DS than the one from BG (p < 0.001), while E8 inversion showed the opposite; the lower frequency was found in the LLC group from BG (p < 0.001). Significant interpopulation differences between the LLC groups for chromosome E (p < 0.001) and for all chromosomes (p < 0.05) were also revealed by the G-test (S4 Table).
The heterozygosity was similar in both populations (DS—0.508 and BG—0.500), as well as the IFR values (DS—82.434 and BG—82.339). In generation F6 the heterozygosity was higher in the LLC from population BG (0.555) compared to the corresponding group originating from DS (0.425), which represents 13% of the difference. The IFR value in the group reared on lower lead concentration from DS was 85.153 and from BG was 81 (Tables 1 and 2).
In the F3 generation, in the HLC group, the standard gene arrangement of chromosome A showed a significantly higher frequency in BG than the corresponding group from DS (z = 5.193, p < 0.001), while A1 inversion showed the opposite (z = -5.992, p < 0.001) (Table 4). Significant difference between these groups was also found for the O3+4+1 complex gene arrangement, with a higher frequency in DS than in BG (z = -3.058, p < 0.05). The G-test confirmed significant interpopulation differences in the F3 generation among HLC groups, on chromosome A, and among all chromosomes (G = 20.71, p < 0.001; G = 32.00, p < 0.01) (S4 Table). In the F6 generation between HLC groups significant differences were found in the frequencies of gene arrangements on chromosome E. Arrangement ESt had significantly lower frequency in the HLC group from DS than in the corresponding group from BG (z = 6.503, p < 0.001), while for the E8 inversion there was a lower frequency in the HLC group from BG (z = -8.375, p < 0.001). Arrangement E1+2+9 showed a higher frequency in the BG group reared on higher lead concentration through six generations (z = 3.422, p < 0.01) (Table 4). The G-test confirmed significant inter-population differences in the distribution of gene arrangements between HLC groups in the F6 generation, on chromosome E (G = 38.86, p < 0.001) and among all chromosomes (G = 50.14, p < 0.001) (S4 Table). Heterozygosity was higher in the BG population in both generations: in the F3 generation by 9.4%, and in the F6 generation by 5.5%. The value of IFR in the HLC group from DS was 86.403, while the corresponding value from BG was 83.998 (Tables 1 and 2).
The general goal of the present study was to determine the importance of D. subobscura inversion polymorphism in a lead polluted environment, as well as the importance of this type of polymorphism in the ability to adapt to environmental changes. By transferring the individuals originating from the populations with different background of environmental pollution into constant laboratory conditions with and without lead contamination, this study aimed to reveal the role of population history and selection in the evolution of inversion polymorphism.
Intrapopulation variability (Deliblato Sands)
Laboratory conditions affected the change of gene arrangements frequencies, both between generations and treatments in the Deliblato Sands population. Even on the control medium, significant differences between F3 and F6 generations and between F1 and F6 generations in the frequencies of E and O chromosome arrangements were found, which possibly resulted from the population’s adaptation to laboratory conditions. The transfer of individuals from the natural environment leads to a process of a population’s biological and genetic refinement in response to such an environmental change [33, 34]. It has been shown that the response to this type of environmental disruption is population-specific for D. subobscura [35, 36]. Adaptation to the laboratory conditions probably depends on some selective advantages of the chromosomal arrangements carried by individuals, or to their linkage with a gene arrangement already participating in a major heterotic association, as predicted by Carson . If natural populations of Drosophila are set up with balanced systems of co-adapted genes in inversions, as proposed by the Dobzhansky’s hypothesis , transfer to the laboratory may break down such a genetic balance and enable a new co-adapted system to evolve. This could be supported by the hypothesis of Lewontin  stating that genetic polymorphism should be lost in a uniform environment, while it could also be in accordance with the idea of Carson  stating that genetic polymorphism may be lost in a uniform environment only if each heterozygote was particularly adapted to some slightly different environmental variables in nature, not present in laboratory conditions.
Significant differences in gene arrangement frequencies between the F3 and F6 generations were found for arrangements on the E chromosome (Est and E8) in the LLC group, and on chromosomes U (Ust, U1+2), E (E8, E1+2+9), and O (O3+4) in the HLC group. Changes in inversion polymorphism parameters which were positively or negatively associated with lead were, among other causes, the result of epistatic or additive effect of the gene content of the inversions. Additionally, when we looked at the frequency of chromosome arrangements within the generations, we observed that the initial response to the presence of lead was associated with chromosomes A and J, while the prolonged period of such conditions included changes in the arrangements frequency of the E chromosome. The changes in the E and O chromosome arrangement frequencies were in accordance with previous studies of D. subobscura inversion polymorphism dynamics in population cages . It was found that under such conditions, those particular chromosomes acted as flexible systems under predominant diversifying selection. Furthermore, a great divergence of sex chromosome arrangement frequencies was detected on the higher lead concentration. It was shown that increased tolerance to cadmium in D. melanogaster was linked to the presence of sex chromosome that originated from partially cadmium-resistant line of the species , as well as that one or more genes responsible for metallothionein structural gene expression regulation were located on sex chromosome . The A1 inversion in the DS population showed a significant change in frequency after three generations of exposure to lead. Although the inversion A1 is of a simple type, it covers a considerable region of the chromosome, with a great number of loci carrying different combinations of alleles.
Intrapopulation variability (Botanical Garden)
The results obtained on inversion polymorphism variability for experimental groups originating from the Botanical Garden showed significant differences in gene arrangement frequencies both between generations and treatments, and no significant difference between generations within the control group. Although arrangement O3+4+2 appeared already at a low frequency, it decreased over generations, especially on the LLC. Also, a significant change in gene arrangement frequencies on the E chromosome (Est and E8) for the HLC group between generations was obtained, suggesting that genes of the E and O chromosomes have an important role in the response to lead pollution. Studies of heavy metal resistance acquisition in various D. melanogaster lines have shown that if one regulatory gene is included in the resistance mechanism, a strong pleiotropic effect of that gene leads to changes in fitness components. It is unlikely that more regulatory genes have a strong pleiotropic effect . Genes that are involved in the resistance mechanism exhibit strong epistatic effects on the loci of allelogenic combinations in chromosomal inversion. Epistasis between genes involved in the resistance mechanism and the rest of the genome occurs during acquisition of resistance to heavy metals [10, 42].
Initial inter-population differences in chromosome arrangement frequencies were expected if we take into account the previous analysis of chromosomal polymorphism of the populations from the same locations . The differences are due to the diverse evolutionary history of each population originating from ecologically distinct habitats. Historical population differentiation remained high and even increased during the experiment, particularly in the lead treated groups. Initially, the BG population showed a significantly higher level of chromosomal variability compared to the DS population. It has already been shown that there is a positive correlation between the complexity of urban environment and the degree of inversion polymorphism [22, 43]. The genetic variability in the BG population, affected by a high anthropogenic activity, is most probably shaped by disruptive selection. Intraspecies and intrapopulation competition promotes diversification of the number and variability of ecological niches. In such conditions, different phenotypes occupy different niches, which promote less competition for resource usage. Disruptive selection is seen in high density populations rather than in low density populations, due to the fact that higher density populations often compete more intensely for resources, which either drives polymorphisms or creates changes in niches in order to relax competition .
The transfer from nature to standard laboratory conditions caused changes across generations in the frequencies of E and O chromosome arrangements in the DS control group, which indicates a recreation of genetic content inside inversions of these chromosomes. The absence of corresponding changes of the chromosomal arrangement frequency in the BG population control group indicates that sets of alleles within inversions of this population enabled a more efficient response to laboratory conditions [11, 22, 43]. Such a different response is in accordance with the hypothesis that populations of DS and BG have different sets of alleles inside the chromosomal arrangements and different existing adaptive potentials, shaped by the local environmental conditions of their native habitats. These differences remain high even after six generations, which is in accordance with the results of Fragata et al. . They found that a strong initial historical signature is maintained in the pattern of inversion frequencies in different D. subobscura populations, even after 40 generations of evolution in a common laboratory environment, regardless of the clear pattern of the convergent evolution on the level of live history traits .
Detected population differentiation remained high, and even increased, during the experiment in the lead treatment groups. In both DS and BG populations, we found inter-generational changes in gene arrangement frequencies of the E and O chromosomes, related to the presence of lead at both concentrations. Therefore, it is clear that individuals from this population are more sensitive to lead pollution in laboratory conditions, since they were not adapted to this kind of environmental stress, compared to the BG population originating from more polluted environmental background. The genetic structure of each population is shaped by current natural selection, but is also the outcome of the historical processes in a particular population. Demographic factors can affect the genetic content of a given inversion, producing different targets for selection between populations [6, 11, 13]. The structure of gene arrangements, as well as the specific combinations of those arrangements in different populations, determines the way that environmental stress can affect fitness. Differences in structure and integrity of the genetic systems of DS and BG led to a population-specific response. According to the Dobzhansky’s co-adaptation hypothesis , it is expected that inversions of geographically distant populations will differ in genetic content. Nevertheless, most studies did not find genetic differentiation within inversions, nor did they suggest low genetic differentiation within inversions across the European cline of D. subobscura, [47, 48]. In contrast with these results, Santos  detected recombination load in crosses of homokaryotypic D. subobscura lines for several inversions of the O chromosome, which shows that the genetic content of a given inversion can change even inside the same population. Our results do not exclude a local adaptation scenario , in which the chromosomal arrangements of a particular population carry combinations of alleles adapted to factors in a given local environment. Positive fitness effect of allelic combination involved in a local adaptation does not necessarily involve epistatic interactions, but the result from a cumulative effect of these combinations .
Although heterozygosity was different in initial generations between the populations DS and BG, we recorded similar changes in both populations, namely a negative correlation with the lead concentration. This trend was more expressed in the BG population, and enhanced in the F6 generation. However, under laboratory conditions without lead, an increase of heterozygosity in both populations was recorded, especially in BG. We have already mentioned that in D. subobscura, chromosomal inversions suppress recombination in heterokaryotypes, which may help to maintain positive epistatic interactions among groups of alleles at loci contained in the inversion . Three mutually not exclusive balancing selection mechanisms have been suggested for the maintenance of inversion polymorphism: co-adaptation , supergene selection , and local adaptation . Genetic exchange among chromosomes in heterokaryotypes from different populations would disrupt locally adapted allele complexes and produce unfit offspring. Consequently, homokaryotypes could suffer an extra disadvantage due to disruption of these favorable combinations, since their chromosomes can freely recombine. Our results suggest that lead contamination affected the composition of the genetic content in both populations through a decrease of heterozygosity and an increase of the portion of chromosomal segments which are ready to be subjected to recombination, a process that could lead to an increase of the recombination load. Although recombination load leads to fitness decrease because the recombination breaks up the associations between beneficial combinations of interacting alleles [11, 13, 52], new allele combinations could potentially have an advantage in novel environmental conditions such as lead contamination.
Based on the analysis of an adaptive genetic marker, such as chromosomal inversion polymorphism, we showed that populations retain signatures from past contamination events, and that heavy metal pollution can provoke adaptive changes in population. Our results show not only that the initial differences in inversion polymorphisms between populations remained throughout the laboratory experiment, but they even increased in conditions of lead contamination. We also observed that a population originating from a more polluted natural habitat had higher initial level of inversion polymorphism, which was also more stable during the experiment, both in conditions with and without lead. Past processes of environmental contamination can result in the ability for evolutionary change, and thus can lead to adaptive divergence of populations residing in differently polluted habitats. Therefore, we showed that both historical events, as well as selection processes, are important for prediction of evolutionary potential and long-term viability of natural populations, especially for further deliberation in parts of evolutionary ecology and conservation biology.
S1 Table. Z-test for inversion frequencies between groups.
S2 Table. G test for inversion frequencies between groups of DS.
S3 Table. G test for inversion frequencies between groups of BG.
Conceived and designed the experiments: BK MA MSR. Performed the experiments: BK AP. Analyzed the data: BK AP ZKN MSR. Contributed reagents/materials/analysis tools: MSR MA. Wrote the paper: BK ZKN AP MSR MA.
- 1. Reznick DN, Ghalambor CK. The population ecology of contemporary adaptations: what empirical studies reveal about the conditions that promote adaptive evolution. Genetica. 2001; 112–113: 183–198.
- 2. Loxdale HD. Rapid genetic changes in natural insect populations. Ecol Entomol. 2010; 35: 155–164.
- 3. Nevo E. Genetic diversity in nature: Patterns and theory. Evol Biol. 1988; 23: 217–246.
- 4. Koehn RK, Bayne BL. Towards a physiological and genetical understanding of the energetics of the stress response. Biol J Linn Soc. 1989; 1/2: 157–171.
- 5. Hummel H, Patarnello T. Genetic effects of pollutants on marine and estuarine invertebrates. In: Beaumont AR, editor. Genetics and Evolution of Aquatic Organisms. Chapman and Hall; 1994. pp 425–434.
- 6. Dobzhansky T. Genetics of the Evolutionary Process. Columbia University Press, New York; 1970.
- 7. Krimbas CB. Drosophila subobscura: Biology, Genetics and Inversion Polymorphism. Verlag Dr Kovac; 1993.
- 8. Noor MA, Grams KL, Bertucci LA, Reiland J. Chromosomal inversions and the reproductive isolation of species. Proc Natl Acad Sci. 2001; 98: 12084–12088. pmid:11593019
- 9. Balanya J, Oller JM, Huey RB, Gilchrist GW, Serra L. Global genetic change tracks global climate warming in Drosophila subobscura. Science. 2006; 313: 1773–1775. pmid:16946033
- 10. Hoffmann AA, Rieseberg LH. Revisiting the impact of inversions in evolution: from population genetic markers to drivers of adaptive shifts and speciation? Annu Rev Ecol Syst. 2008; 39: 21–42. pmid:20419035
- 11. Kirkpatrick M. How and why chromosome inversions evolve. PLoS Biol. 2010; 8(9): e1000501. pmid:20927412
- 12. Lowry D, Willis J. A widespread chromosomal inversion polymorphism contributes to a major life-history transition, local adaptation, and reproductive isolation. PLoS Biol. 2010; 8: e1000500. pmid:20927411
- 13. Kirkpatrick M, Barton N. Chromosome Inversions, Local Adaptation and Speciation. Genetics. 2006; 173: 419–434. pmid:16204214
- 14. Anđelković M, Savković V, Kalajdžić P. Inversion polymorphism in Drosophila subsobscura from two different habitats from the mountain of Goč. Hereditas. 2003; 138: 241–243. pmid:14641490
- 15. Santos M, Céspedes W, Balanyà J, Trotta V, Calboli FCF, Fontdevila A, et al. Temperature-related genetic changes in laboratory populations of Drosophila subobscura: evidence against simple climatic-based explanations for latitudinal clines. Am Nat. 2005a; 165: 258–273. pmid:15729655
- 16. Stamenković-Radak M, Rašić G, Savić T, Kalajdžić P, Kurbalija Z, Kenig B, et al. Monitoring of the genetic structure of natural populations: change of the effective population size and inversion polymorphism in Drosophila subobscura. Genetica. 2008; 133: 57–63. pmid:17668277
- 17. Rodríguez-Trelles F, Tarrío R, Santos M. Genome-wide evolutionary response to a heat wave in Drosophila. Biol Lett. 2013; 9(4): 20130228 pmid:23740296
- 18. Hoffmann AA. Physiological climatic limits in Drosophila: patterns and implications. J Exp Biol. 2010; 213: 878–880.
- 19. Carroll SP, Hendry AP, Reznick DN, Fox CW. Evolution on ecological time-scales. Funct Ecol. 2007; 21: 387–393.
- 20. Lagisz M, Laskowski R. Evidence for between-generation effects in carabids exposed to heavy metals pollution. Ecotoxicology. 2008; 17: 59–66 pmid:17940866
- 21. Kenig B, Stamenkovic-Radak M, Andjelkovic M. Population specific fitness response of Drosophila subobscura to lead pollution. Insect Sci. 2013; 20(2):245–53. pmid:23955864
- 22. Singh BN. Chromosome inversions and linkage disequilibrium in Drosophila. Curr Sci. 2008; 94: 459–464.
- 23. Kenig B, Jelić M, Kurbalija Z, Stamenković-Radak M, Anđelković M. Inversion polymorphism in populations of Drosophila subobscura from urban and non-urban environments. Arch Biol Sci. 2010; 62(3): 565–574.
- 24. Grant A. Pollution Tolerant Species and Communities: Intriguing Toys or Invaluable Monitoring Tools? Hum Ecol Risk Assess. 2002; 8: 955–970.
- 25. Newman MC, Unger MA. Fundamentals of Ecotoxicology, 2nd edn. Lewis Publishers, CRC Press; 2003.
- 26. Kadović R, Knežević M. Teški metali u šumskim ekosistemima Srbije. Šumarski fakultet Univerziteta u Beogradu; 2002. pp. 114–149.
- 27. Grzetic I, Ghariani RHA. Potential health risk assessment for soil heavy metal contamination in the central zone of Belgrade (Serbia). J Serb Chem Soc. 2008; 73(8–9): 923–934.
- 28. Marjanović MD, Vukčević MM, Antonović DG, Dimitrijević SI, Jovanović ĐM, Matavulj MN, et al. Heavy metals concentration in soils from parks and green areas in Belgrade. J Serb Chem Soc. 2009; 74: 697–706.
- 29. Kunze-Mühl E, Müller E. Weitere untersuchungen uber die chromosomale, Struktur und die naturlichen strukturtypen von Drosophila subobscura. Chromosoma (Berl.). 1958; 9: 559–570. pmid:13619011
- 30. Kunze-Mühl E, Sperlich D. Inversionen und chromosomale Strukturtypen bei Drosophila subobscura Coll. Z Indukt Abstam Vererbungs. 1955; 87: 65–84. pmid:13312460
- 31. Kulathinal RJ, Stevison LS, Noor MAF. The Genomics of Speciation in Drosophila: Diversity, Divergence, and Introgression Estimated Using Low-Coverage Genome Sequencing. PLoS Genet. 2009; 5(7): e1000550. pmid:19578407
- 32. Rice WR. Analysing tables of statistical tests. Evolution. 1989; 43: 223–225. pmid:2561774
- 33. de Frutos R. Changes of chromosomal polymorphism in experimental cage populations of Drosophila subobscura. Genetica. 1978; 49(2): 139–151.
- 34. Singh P, Sigh BN. Population Genetics of Drosophila ananassae: Variation in the degree of genetic divergence in populations transferred to laboratory conditions. Zool Stud. 2008; 47: 704–712.
- 35. Simoes P, Rose MR, Duarte A, Goncalves R, Matos M. Evolutionary domestication in Drosophila subobscura. J Evol Biol. 2007; 20: 758–766. pmid:17305841
- 36. Simoes P, Santos J, Fragata I, Mueller LD, Rose MR, Matos M. How repeatable is adaptive evolution? The role of geographical origin and founder effects in laboratory adaptation. Evolution. 2008; 62: 1817–1829. pmid:18489721
- 37. Carson HL. Relative fitness of genetically open and closed experimental populations of Drosophila robusta. Genetics. 1961; 46: 553–567. pmid:13691102
- 38. Lewontin RC. The adaptations of populations to varying environments. Cold Spring Harb Sym. 1957; 22: 395–408. pmid:13543723
- 39. Christie NT, Williams MW, Jacobson KB. Genetic and physiological parameters associated with cadmium toxicity in Drosophila melanogaster. Biochem Genet. 1985; 23: 571–583. pmid:3931625
- 40. Gill HJ, Nida DL, Dean DA England MW, Jacobson KB. Resistance of Drosophila to cadmium: biochemical factors in resistant and sensitive strains. Toxicology. 1989; 56: 315–321. pmid:2734808
- 41. Rose MR, Charlesworth B. Genetics of Life History in Drosophila melanogaster. II. Exploratory Selection Experiments. Genetics. 1981b; 97: 187–196. pmid:6790341
- 42. Kalajdžić P, Stamenković-Radak M, Anđelković M. The effect of different concentrations of lead on inversion polymorphism in Drosophila subobscura. Hereditas. 2006; 143: 41–46. pmid:17362332
- 43. Valiati VH, Valente VLS. Chromosomal polymorphism in urban populations of Drosophila paulistorum. Revta Bras Genet. 1997; 20(4): 567–581.
- 44. Bolnick DI. Can intraspecific competition drive disruptive selection? An experimental test in natural populations of sticklebacks. Evolution. 2004; 87: 608–618.
- 45. Fragata I, Lopes-Cunha M, Barbaro M, Kellen B, Lima M, Santos MA, et al. How much can history constrain adaptive evolution? A real-time evolutionary approach of inversion polymorphisms in Drosophila subobscura. J Evol Biol. 2014; 27: 2727–2738. pmid:25430492
- 46. Fragata I, Simoes P, Lopes-Cunha M, Lima M, Kellen B, Barbaro M, et al. Laboratory Selection Quickly Erases Historical Differentiation. PLoS ONE. 2014; 9(5): e96227. pmid:24788553
- 47. Simoes P, Calabria G, Picao-Osorio J, Balanya J, Pascual M. The Genetic Content of Chromosomal Inversions across a Wide Latitudinal Gradient. PLoS ONE. 2012; 7(12): e51625. pmid:23272126
- 48. Pegueroles C, Aquadro CF, Mestres F, Pascual M. Gene flow and gene flux shape evolutionary patterns of variation in Drosophila subobscura. Heredity. 2013; 110: 520–529. pmid:23321709
- 49. Santos M. Recombination load in a chromosomal inversion polymorphism of Drosophila subobscura. Genetics. 2009; 181: 803–809. pmid:19033153
- 50. Th Dobzhansky. Genetics of Natural Populations. XVIII. Experiments on Chromosomes of Drosophila pseudoobscura from Different Geographic Regions. Genetics. 1948; 33(6): 588–602. pmid:18100290
- 51. Wasserman M. Recombination-induced chromosomal heterosis. Genetics. 1968; 58: 125–139. pmid:5656342
- 52. Charlesworth B, Charlesworth D. An experiment on recombination load in Drosophila melanogaster. Genet Res. 1975; 25: 267–274. pmid:810390