Common Genetic Basis of Eye and Pigment Loss in Two Distinct Cave Populations of the Isopod Crustacean Asellus aquaticus

Common Genetic Basis of Eye and Pigment Loss in Two Distinct Cave Populations of the Isopod... Abstract Repeated evolution of similar phenotypes is a widespread phenomenon found throughout the living world and it can proceed through the same or different genetic mechanisms. Cave animals with their convergent traits such as eye and pigment loss, as well as elongated appendages, are a striking example of the evolution of similar phenotypes. Yet, few cave species are amenable to genetic crossing and mapping techniques making it challenging to determine the genetic mechanisms causing their similar phenotypes. To address this limitation, we have been developing Asellus aquaticus, a freshwater isopod crustacean, as a genetic model. Many of its cave populations originate from separate colonization events and thus independently evolved their similar cave-related phenotypes which differ from the still existent ancestral-like surface populations. In our prior work, we identified genomic regions responsible for eye and pigment loss in a single cave population from Slovenia. In this study we examined another, independently evolved cave population, also from Slovenia, and asked whether the same or different genomic regions are responsible for eye and pigment loss in the two cave populations. We generated F2 and backcross hybrids with a surface population, genotyped them for the previously identified genomic regions, and performed a complementation test by crossing individuals from the two cave populations. We found out that the same genomic regions are responsible for eye and pigment loss and that at least one of the genes causing pigment loss is the same in both cave populations. Future studies will identify the actual genes and mutations, as well as examine additional cave populations to see if the same genes are commonly associated with eye and pigment loss in this species. Introduction A frequently asked question in evolutionary biology is whether populations or species with similar phenotypes have evolved their convergent traits using the same or different genetic mechanisms. Multiple studies have shown that both scenarios occur in nature (reviewed in Manceau et al. 2010; Kronforst et al. 2012; Domyan and Shapiro 2017) and there is evidence that repeated use of the same gene is more common in closely related than in distantly related species (Conte et al. 2012). Additional model organisms and focal traits are needed to better understand and predict when and why each scenario occurs. Cave animals, famous for their eyeless and depigmented appearance, are an excellent group in which to study the evolution of similar phenotypes. As these have evolved numerous times in very different cave organisms (Culver 1982), it is possible to compare the evolution of similar characters in unrelated species, closely related species, or populations of the same species (Protas and Jeffery 2012). Among cave animals, genetic mechanisms behind the evolution of similar phenotypes have been most intensively studied in the cavefish Astyanax mexicanus. Its many independently evolved cave populations and the ability to perform genetic crosses and mapping experiments make it a superb model to study this phenomenon at the population level. So far, studies have mainly focused on pigmentation, eye size, appetite regulation, feeding behavior, and sleep. These studies demonstrated that many times the same genes or loci are responsible for similar phenotypes of independent populations (e.g., Protas et al. 2006; Gross et al. 2009; Duboué et al. 2011; Elipot et al. 2014; Aspiras et al. 2015), though there are examples where different genes or loci were found to be the cause (e.g., Kowalko et al. 2013). Such bias toward gene or genetic pathway reuse raises the question of why a particular gene is repeatedly targeted. Several possible explanations have been suggested including advantageous secondary consequences of mutations, ease of mutation of that particular gene, or lack of negative pleiotropic effects (Stern 2013). An illustrative example of gene reuse is the repeated mutation of the Oca2 gene that causes the albino phenotype in multiple cave populations of A. mexicanus (Protas et al. 2006; Gross and Wilkens 2013). Consequential enhancement of the catecholamine pathway that could control behaviors adaptive in the cave environment is suggested to be the reason for the repeated use of this gene (Bilandžija et al. 2013). Few studies have examined whether the same or different genetic mechanisms are responsible for similar phenotypes of cave animals other than A. mexicanus. One example is the two independently evolved cave-adapted species of planthoppers that live in different continents but show a blockage in the same step of the melanin synthesis pathway which is also the same step that A. mexicanus is blocked at (Protas et al. 2006; Bilandžija et al. 2012). Furthermore, studies of the amphipod crustacean, Gammarus minus, revealed that the gene hedgehog is differently expressed in cave and surface populations of this species—as was shown in A. mexicanus (Yamamoto et al. 2004; Aspiras et al. 2012). Contrary to what is known in A. mexicanus, research based on morphology and gene expression in another cavefish species, Sinocyclocheilus anophthalmus, suggests that lens independent mechanisms might be responsible for eye degeneration (Yamamoto and Jeffery 2000; Meng et al. 2013). Although much progress has been made toward understanding whether the same or different genetic mechanisms are responsible for similar phenotypes of different cave animals, what has been lacking are species similar to A. mexicanus where this question can be asked on the population level, with the potential of utilizing genetic crosses and mapping techniques. The isopod crustacean, Asellus aquaticus, is an exciting emerging model organism in which genetic mechanisms behind the evolution of similar phenotypes can be studied on the population level. Multiple cave populations of this isopod originated from separate colonization events in Italy, Slovenia, Hungary, and Romania and thus have independently evolved their similar, cave-related phenotypes with characteristic traits such as loss of eyes, depigmentation, and elongated appendages (Konec et al. 2015; Pérez-Moreno et al. 2017). Closely related ancestral-like surface populations are still existent and are composed of individuals with markedly different phenotypes. The best-studied are the cave populations from Slovenia, where in the Planina Cave two distinct populations live in two separate cave channels, the Pivka Chanel and the Rak Channel (Verovnik et al. 2003, 2004). Several lines of evidence suggest that these two populations invaded caves and evolved independently from each other. They formed mutually exclusive genetic clusters in a randomly amplified polymorphic DNA analysis (Verovnik et al. 2003), and they do not share any common mtDNA haplotypes (Verovnik et al. 2004). Although both populations are eyeless and depigmented, they differ consistently in a number of other morphometric and morphological characters (Prevorčnik et al. 2004). Recent studies using microsatellites revealed that there is no gene flow between populations (Konec 2015), and that both populations differ also in their behavioral response to light, thigmotactic preference, and ability to withstand water current (Fišer 2017). The likely explanation, why the two populations are so different even though they live in proximity, is that they are distributed north (Rak Channel population) and south (Pivka Channel population) of a major geological discontinuity, the Idrija Fault, that significantly shaped the regional geomorphology (Placer et al. 2010). The Idrija Fault is older than the age of the cave populations, which does probably not exceed 1.3 Mya (Konec et al. 2015), making a vicariant split of a formerly contiguous subterranean population unlikely. As cave and surface forms can interbreed in captivity (Baldwin and Beatty 1941; Protas et al. 2011), genetic mapping of traits commonly associated with cave life can be applied (Protas et al. 2011). In the cave population from the Pivka Channel of Planina Cave genomic regions responsible for eye and pigment variation have already been mapped (Protas et al. 2011). Surprisingly, multiple mechanisms of both eye and pigment loss were found within this single population. Complete eye loss mapped to a different locus than eye reduction, while albinism appears to be achieved either by a recessive genotype at one locus or doubly recessive genotypes at two other loci. To investigate whether the same or different genetic mechanisms are responsible for the repeated evolution of eye and pigment loss in different cave populations of A. aquaticus, we mapped these traits for the first time in another cave population, i.e., Rak Channel of Planina Cave. Cave individuals from the Rak Channel and surface individuals from a closely related surface population were used to generate F2 and backcross individuals. These were genotyped for genes that were previously found to be associated with eye and pigment loss in the cave population from the Pivka Channel of Planina Cave. Additionally, we performed a complementation test between cave individuals from Rak and Pivka Channels. Geographic proximity and a common timeframe of cave habitat colonization of both cave populations imply that their surface ancestors had probably shared the same gene pool (Verovnik et al. 2003, 2004). Thus, we hypothesized that alleles responsible for eye and pigment loss in these two distinct cave populations originated from the same standing genetic variation. We predicted that genomic regions responsible for eye and pigment loss in both cave populations overlap. Methods Animals We collected cave individuals from the Rak Channel of Planina Cave and surface individuals from Rakov Škocjan (Fig. 1). The surface population from Rakov Škocjan inhabits the upstream surface section of the same river that flows underground in the Rak Channel and is closely related to the Rak Channel cave population (Verovnik et al. 2003, 2004). Animals were maintained in the laboratory as described by Protas et al. (2011). Cave males were mated to surface females and cave females were mated to surface males to generate F1 hybrids. These were mated either to siblings to generate F2 hybrids, or F1 females were mated back to cave males to generate backcross hybrids. F2 and backcross individuals had a high mortality, and often times only one or two individuals per brood survived embryonic development. We pooled together F2 and backcross individuals from many crosses and raised them until they were at least 4 mm long. A total of 82 individuals were phenotyped and genotyped. Fig. 1 View largeDownload slide Map of Planina Cave area showing the Rak and Pivka Channels. Upper left: the black rectangle marks the region in Europe where the Planina Cave is situated. Bottom: Planina Cave area magnified. Red dots indicate sampling sites of cave and surface Asellus aquaticus populations mentioned in the text. The omega-like black symbols mark cave openings. Dashed lines are used where the precise watercourse is unknown. Fig. 1 View largeDownload slide Map of Planina Cave area showing the Rak and Pivka Channels. Upper left: the black rectangle marks the region in Europe where the Planina Cave is situated. Bottom: Planina Cave area magnified. Red dots indicate sampling sites of cave and surface Asellus aquaticus populations mentioned in the text. The omega-like black symbols mark cave openings. Dashed lines are used where the precise watercourse is unknown. Phenotyping F2 and backcross individuals from the Rak Channel crosses were anesthetized in a 0.4% clove oil solution and phenotyped for eye and pigment related traits using a Leica S8 Apo stereomicroscope and LAS Core software. Each individual was phenotyped for presence and color of eye pigment. Presence and color of head and body pigmentation generally followed suit with eye pigment and was not included in the analyses. As head and body pigmentation is expressed in at least two different patterns (Protas et al. 2011), individuals were phenotyped for the pattern of this pigmentation. Additionally, each individual was phenotyped for eye (i.e., ommatidia or fragmented ommatidia) presence. Genetic markers Genetic markers were chosen based on the previously constructed genetic map for A. aquaticus and results of mapping eye and pigment related traits in the cave population from the Pivka Channel of the Planina Cave (Protas et al. 2011). In this population, four genomic regions located at four separate linkage groups were identified to be responsible for eye and pigment traits. Within each of these four regions, we searched for genetic markers by which we could reliably distinguish between the Rak Channel and Rakov Škocjan origin. We tried multiple markers within each region and had to reject several as they were either monomorphic or too polymorphic between both populations, or the Rak Channel individuals were not uniform for the marker genotype. Ultimately, we selected one suitable genetic marker per region (Supplementary Table S1). Three of these markers were in genes from the existing genetic map, i.e., disconnected, nckx30, and pax2, though the polymorphism used for genotyping was not necessarily the same for the Rak Channel and Pivka Channel populations. In the Pivka Channel crosses, the genotype of disconnected was associated with presence versus absence of eye pigment, the genotype of nckx30 was associated with light versus dark eye pigment, and the genotype of pax2 was associated with red versus orange or brown eye pigment. To mark the fourth region, which was related to stellate versus diffuse head and body pigmentation pattern and eye presence versus absence in the Pivka Channel crosses, we could not use markers in genes from the existing genetic map as these were polymorphic within the Rak Channel population. Fortuitously, we had been adding additional genes onto the genetic map and found that the A. aquaticus ortholog of the gene son of bowl (sob) mapped to this fourth region and was associated with the phenotype of stellate versus diffuse head and body pigmentation pattern and eye presence versus absence in the Pivka Channel crosses. As the marker in sob was found suitable also for the Rak Channel and Rakov Škocjan individuals, we used it to mark this last genomic region. For details regarding the identification of the genetic marker in sob see Supplementary Information S1. Genotyping To genotype F2 and backcross individuals from the Rak Channel crosses, DNA was extracted from two legs of each animal using QiaAmp Micro Kit (Qiagen). Genetic markers in genes disconnected, nckx30, pax2, and sob were amplified using primers specified in Supplementary Table S1. PCR was performed using GoTaq Green Master Mix (Promega); a 53°C annealing temperature and a 1 min extension time for 30 cycles. Products were purified using ExoSAP-IT (Thermo Fisher Scientific) and sent for sequencing to ELIM Biopharm or MCLAB. Sequences were analyzed using FinchTV 1.4.0 (Geospiza, Inc.) software. For markers in disconnected, nckx30, and pax2 sequence chromatograms were inspected to determine genotype. F2 and backcross individuals were either homozygous for the cave allele, homozygous for the surface allele, or heterozygous. The marker in sob was genotyped differently as primers used to amplify it only amplified the surface allele but not the cave allele. Therefore, successful amplification indicated either a homozygous surface or a heterozygous genotype, but we could not discriminate between the two. To confirm the homozygous cave genotype in individuals for which amplification was not successful, we used the forward primer and a different reverse primer to obtain a product, sequenced the product, and inspected the sequence chromatogram to confirm the cave genotype. Additionally, backcross individuals from the Pivka Channel crosses generated in our previous study (Protas et al. 2011) were genotyped for the marker in sob using stored DNA isolates. This was needed because the gene sob was not included in the genetic map constructed in that study (see also above). The marker in sob was amplified and genotyped as described above. However, as these were backcross individuals (F1 females were bred to cave males), successful amplification indicated a heterozygous genotype only. Genotypes of markers in disconnected, nckx30, and pax2, as well as phenotypic traits for these hybrids, were previously determined in Protas et al. (2011). Statistical analyses F2 and backcross individuals from the Rak Channel crosses were categorized according to their phenotypic trait values in groups that matched the groups from Protas et al. (2011); only traits related to markers in disconnected and nckx30 were grouped slightly differently. This was needed because we observed that some cave individuals from the Rak Channel population exhibited light red pigment in the eye region (as in Fig. 2A′). Thus, for disconnected, we pooled the light red and the unpigmented phenotype into the same group, i.e., unpigmented, and treated individuals with red, orange, or brown eye pigmentation as pigmented. Consequently, for nckx30 orange versus red or brown eye pigmentation was investigated instead of light (i.e., light red or orange) versus dark (i.e., red or brown) eye pigmentation as in Protas et al. (2011). Fig. 2 View largeDownload slide Phenotypes of F2 and backcross hybrids between the Rak Channel and surface individuals. A, B, C, D, E, and F show animals’ heads, while A′, B′, C′, D′, E′, and F′ focus on the eyes of the same animals. Several phenotypes were observed: no pigment (not shown), light red (A, A′), red (B, B′, C, C′), orange (D, D′), and brown (E, E′, F, F′) eye pigment. The head and body pigmentation pattern was either stellate (E) or diffuse (D). In addition, ommatidia were either present (A′, C′, D′, F′) or absent (B′, E′), but fragmented ommatidia (not shown) were observed only in a few individuals. Fig. 2 View largeDownload slide Phenotypes of F2 and backcross hybrids between the Rak Channel and surface individuals. A, B, C, D, E, and F show animals’ heads, while A′, B′, C′, D′, E′, and F′ focus on the eyes of the same animals. Several phenotypes were observed: no pigment (not shown), light red (A, A′), red (B, B′, C, C′), orange (D, D′), and brown (E, E′, F, F′) eye pigment. The head and body pigmentation pattern was either stellate (E) or diffuse (D). In addition, ommatidia were either present (A′, C′, D′, F′) or absent (B′, E′), but fragmented ommatidia (not shown) were observed only in a few individuals. Fisher’s exact tests were performed for each genetic marker and its associated trait to explore whether a significant association between genotype and phenotype exists. The strength of association was assessed using Cramér’s V, which varies from 0 (no association) to 1 (complete association). All calculations were performed in PAST 3 (Hammer et al. 2001). Complementation test Complementation crosses were carried out between males from the Rak Channel and females from the Pivka Channel populations. Males were collected in cave Škratovka, which is located very close to the Planina Cave and harbors the same cave population as is present in the Rak Channel (Konec 2015; Fig. 1). In the Pivka Channel we collected females engaged in precopula (i.e., precopulatory mate guarding behavior during which a male firmly holds a female) to ensure that they were ready to mate. In the laboratory, females were gently separated from males of the same population, and each female was put in a separate Petri dish together with one male from the Rak Channel population. Pairs were kept in constant darkness at 10°C and left to feed ad libitum on decaying leaves of the black alder (Alnus glutinosa). Eventually, two out of seven crosses were successful and produced offspring. As there is evidence that sperm storage does not occur in A. aquaticus (Ridley and Thompson 1979; Marchetti and Montalenti 1990), we were confident that these were hybrids. Most of these died soon after hatching, but nine offspring of the same cross reached 18–43 days of age and 1.5–2 mm in length, and were fixed in 96% ethanol. As ommatidia phenotype can alter upon preservation in ethanol (M. Protas, personal observation), only pigment related traits could be reliably determined. For phenotyping, we used an Olympus SZX2-ILLT stereomicroscope with the Olympus ColorView III camera and Cell^B 2.8. (Olympus) software. Results Interestingly, F2 and backcross hybrids between cave individuals from the Rak Channel and surface individuals showed almost exactly the same phenotypes as were previously observed in backcross hybrids from the Pivka Channel crosses (Fig. 2). F2 and backcross hybrids had no pigment, light red, red, orange, or brown pigment in the eye region. In addition, the pattern of head and body pigmentation was either stellate or diffuse. Finally, F2 and backcross hybrids were either eyeless (i.e., had no ommatidia) or eyed (i.e., had ommatidia or fragmented ommatidia). In contrast to the Pivka Channel crosses, the majority of eyed hybrids had ommatidia and fragmented ommatidia were observed in few individuals only. All investigated eye and pigment related phenotypic traits of cave individuals from the Rak Channel were found to be associated with the same genes as in cave individuals from the Pivka Channel (Table 1). Fisher’s exact tests showed a statistically significant association in all cases. The genotype of disconnected was associated with the phenotype of presence versus absence of eye pigmentation, and the cave allele related to the unpigmented phenotype. The genotype of nckx30 was associated with the phenotype of orange versus red or brown eye pigmentation, and the cave allele related to the orange phenotype. The genotype of pax2 was associated with the phenotype of red versus orange or brown eye pigmentation, and the cave allele related to the red phenotype. Finally, the genotype of sob was associated with the phenotype of stellate versus diffuse head and body pigmentation pattern and with the phenotype of eye presence versus absence, and the cave allele related to the stellate and the eyeless phenotype, respectively. The association was strongest for phenotypes related to disconnected, nckx30, and pax2 with Cramér’s V reaching from 0.75 to 0.89, and a bit less strong for phenotypes related to sob with Cramér’s V reaching from 0.53 to 0.6. According to Cohen (1988), all associations can be considered as having a large effect size. The strength of association for disconnected and pax2 related phenotypes was similar between the Rak Channel and Pivka Channel crosses, but somewhat less strong in the Rak Channel crosses for nckx30 and sob related phenotypes. Raw data on phenotypes and genotypes of all Rak Channel hybrids used in this study are available in Supplementary Table S2. Table 1 Association between genotypes and phenotypes in the Pivka Channel hybrids and in the Rak Channel hybrids Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel hybrids comprise backcross individuals examined previously in Protas et al. (2011), but data for sob were obtained in this study. Rak Channel hybrids comprise F2 and backcross individuals examined in this study. Hybrids were either homozygous for the cave allele (CC), heterozygous (CS), or homozygous for the surface allele (SS). Pivka Channel backcross individuals lack the SS genotype as backcrossing was done only with the cave parent individuals. For the Rak Channel hybrids the CS and SS genotypes of sob could not be distinguished due to the genotyping method used. In the Pivka Channel hybrids unpigmented individuals had no pigment in the eye region, whereas in the Rak Channel hybrids unpigmented individuals either had no or light red pigment in the eye region. For the genetic marker in nckx30, phenotypes were grouped into light (light red or orange) versus dark (red or brown) eye pigment for the Pivka Channel hybrids, whereas phenotypes were grouped into orange versus red or brown eye pigment for the Rak Channel hybrids. See main text for a more detailed explanation. Table 1 Association between genotypes and phenotypes in the Pivka Channel hybrids and in the Rak Channel hybrids Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel hybrids comprise backcross individuals examined previously in Protas et al. (2011), but data for sob were obtained in this study. Rak Channel hybrids comprise F2 and backcross individuals examined in this study. Hybrids were either homozygous for the cave allele (CC), heterozygous (CS), or homozygous for the surface allele (SS). Pivka Channel backcross individuals lack the SS genotype as backcrossing was done only with the cave parent individuals. For the Rak Channel hybrids the CS and SS genotypes of sob could not be distinguished due to the genotyping method used. In the Pivka Channel hybrids unpigmented individuals had no pigment in the eye region, whereas in the Rak Channel hybrids unpigmented individuals either had no or light red pigment in the eye region. For the genetic marker in nckx30, phenotypes were grouped into light (light red or orange) versus dark (red or brown) eye pigment for the Pivka Channel hybrids, whereas phenotypes were grouped into orange versus red or brown eye pigment for the Rak Channel hybrids. See main text for a more detailed explanation. Complementation tests between cave individuals from the Rak Channel and Pivka Channel populations showed lack of complementation for pigment related traits (Fig. 3). All nine F1 hybrid offspring from the same complementation cross had no pigment in the eye region, head, or body, just like their parents. The presence or absence of ommatidia could not be reliably determined (see the “Methods” section). Fig. 3 View largeDownload slide Complementation test between cave individuals from the Rak Channel and the Pivka Channel populations showed lack of complementation. Male from the Rak Channel population (father of the complementation cross) (A), female from the Pivka Channel population (a representative individual as mother of the complementation cross died before preservation) (B), as well as F1 hybrid offspring (C) had no pigment in the eye region, head, or body. Fig. 3 View largeDownload slide Complementation test between cave individuals from the Rak Channel and the Pivka Channel populations showed lack of complementation. Male from the Rak Channel population (father of the complementation cross) (A), female from the Pivka Channel population (a representative individual as mother of the complementation cross died before preservation) (B), as well as F1 hybrid offspring (C) had no pigment in the eye region, head, or body. Discussion Repeated evolution of similar phenotypes is a widespread phenomenon found throughout the living world and the onset of modern molecular tools made it possible to investigate the genetic mechanisms behind it (Peichel and Marques 2017). Here we examined this issue in two independently evolved cave populations of the freshwater isopod A. aquaticus. In line with our prediction, we have shown that in both populations the same genomic regions are responsible for loss of pigment and eyes (i.e., ommatidia). These convergent traits might be caused by the same genes via the same or different mutations or by different but tightly linked genes. Results of the complementation test allow us to narrow down these possibilities for the pigment related traits. Crossing individuals from the Pivka Channel population to individuals from the Rak Channel population yielded unpigmented individuals. As no complementation occurred, it is highly likely that the same and not different genes are responsible for pigment loss in both cave populations, though it is currently impossible to say whether this is due to the same or different mutations. One complication of the complementation test is that in the Pivka Channel population pigment loss can be achieved in two different ways: (i) by two copies of the cave allele for the gene that encodes absence of pigmentation and is marked by disconnected, or (ii) by two copies of the cave allele for the gene that encodes orange pigment and is marked by nckx30, and two copies of the cave allele for the gene that encodes red pigment and is marked by pax2 (Protas et al. 2011). Thus, convergence in pigment loss in the two cave populations from the Rak and Pivka Channels is either due to the same gene marked by disconnected, the same two genes marked by nckx30 and pax2, respectively, or the same three genes marked by nck30, pax2, and disconnected. Currently, we are not able to distinguish between these possibilities though this can be dissected in the future by crossing orange, red, and unpigmented F2 hybrids generated from the Rak Channel population to F2 hybrids of the same colors generated from the Pivka Channel population. Nevertheless, at least one of the genes responsible for pigment loss is apparently in common between the two cave populations. Unfortunately, we were unable to test the complementation of the eyeless phenotype and thus cannot exclude any of the plausible genetic mechanisms (i.e., different but linked genes, same or different mutations of the same gene) for the observed convergent phenotype of this trait. Future carefully planned complementation crosses will hopefully provide additional insight into the genetic mechanisms responsible for eye loss. Initially, our interpretation of the two distinct ways in which pigment loss is achieved in the Pivka Channel population was that perhaps pigment pathways were accumulating mutations as pigmentation is useless in the dark cave environment and selection for it is relaxed (Wilkens 2010; Protas et al. 2011). However, this evolutionary mechanism is much less likely to produce the degeneration of pigment pathways in the same way in two independently evolved populations as seems to be the case in the populations from the Rak and Pivka Channels. If all of the same genes are responsible for pigment loss in these two cave populations, repeated utilization of the same genetic pathways would be better explained by adaptive pleiotropic effect(s) of the mutations that cause pigment loss. Bilandžija et al. (2013) suggested that this is a possible reason for the repeated use of the Oca2 gene causing pigment loss in the cavefish A. mexicanus. In populations adapting to a novel environment or selective pressures, beneficial alleles originate either from standing genetic variation or new mutations (Barrett and Schluter 2008). The two cave populations examined in this study derive from surface ancestors that had probably shared the same gene pool (Verovnik et al. 2003, 2004), and we have shown that the same genomic regions and at least one of the same genes is responsible for their convergent unpigmented and eyeless phenotypes. Considering this, our hypothesis that standing genetic variation present in the founding populations was the source of “cave” alleles remains the most parsimonious explanation for the evolution of similar phenotypes in these two cave populations. Fixation of alleles from standing genetic variation has been previously shown to be the mechanism of evolutionary change in multiple examples (reviewed in Seehausen 2015; Casane and Rétaux 2016; Peichel and Marques 2017). Yet, based on currently available data, new mutations cannot be excluded as the source of alleles responsible for the convergent cave phenotypes. Repeated, de novo mutation of the same gene in independently evolved populations of the same species has also been documented in many examples including the Oca2 gene causing pigment loss in the cavefish A. mexicanus (Protas et al. 2006; Gross and Wilkens 2013). Although our data showed a striking match between the genomic regions responsible for pigment and eye loss in cave populations from the Pivka and Rak Channels, we also observed some minor differences between the two populations. First, in the Pivka Channel crosses, individuals homozygous for the cave alleles for the marker in disconnected generally had no eye pigment, whereas in the Rak Channel crosses such individuals generally either had no pigment or light red pigment in the eye region. This difference could be due to the interaction of additional genes in either of these populations. Second, in the Pivka Channel individuals we were able to map eye presence versus absence but also eye size because many crosses had fragmented ommatidia (Protas et al. 2011). Differently, in the Rak Channel crosses most of the F2 and backcross individuals were either eyeless or had ommatidia and individuals with fragmented ommatidia were too few to enable us to map eye size. Thus, it is possible that the alleles responsible for fragmented ommatidia are present in the Rak Channel population at a lower frequency than in the Pivka Channel population. Or, perhaps individuals with those alleles had a higher mortality during the breeding and crossing period in the lab and therefore did not survive to be phenotyped and genotyped. Another difference between cave populations from the Rak and Pivka Channels was that the associations between nckx30 and orange eye pigment, and also sob and eye loss as well as stellate pigmentation pattern were stronger in the Pivka Channel population. Although this could imply that different genes in the same genomic regions are responsible for these same traits in the two populations, due to the results of complementation tests, this seems unlikely and other explanations are more probable. The weaker association could be due to the smaller sample size available for the Rak Channel crosses in this study compared with the Pivka Channel crosses from Protas et al. (2011). Another possible explanation for the weaker association between sob and eye loss is that due to high mortality of the Rak Channel crosses we were unable to separate them into families (like we did for the Pivka Channel crosses) but had to pool them from many different parents. Therefore, if the allele responsible for eye loss is not fixed in the Rak Channel population (as is apparently the case in the Pivka Channel population), eyed individuals homozygous for the cave allele for the marker in sob would be expected and more common. The presence of such individuals would make the association between sob and eye loss appear weaker compared with the Pivka Channel crosses in which only families of cave parents with complete eye loss were examined for this association. Currently, we have only examined the genetic background of eye and pigment loss in the Pivka Channel and Rak Channel populations that both live in Planina Cave, although in its two distinct channels. Still to examine are other independently arisen cave populations that are geographically and phylogenetically more distant from the two populations examined so far. We imagine that those cave populations might be less likely to have the same four genomic regions responsible for eye and pigment loss. Yet if the same surface lineage founded these cave populations, it could still be possible that standing genetic variation in the ancestral surface population is responsible for the phenotypic variation in some of these populations. One population that would be especially interesting to examine in future studies is the A. aquaticus infernus population from Romania which lives in a chemoautotrophic cave environment that is ecologically very different from the environment in Planina Cave (Sarbu et al. 1996; Turk-Prevorčnik et al. 1998; Konec et al. 2015). If different genetic mechanisms do exist in some cave populations, we would expect to find them in this cave population. To lead back to the original question, the same genomic regions seem to be responsible for similar phenotypes in A. aquaticus cave populations from the two channels of Planina Cave. Complementation tests provide further evidence that at least one of the same genes is responsible for pigment loss in these two populations. Further work identifying the actual genes and mutations responsible for eye and pigment traits will help us to more specifically address the evolutionary history of these traits and why the same regions are responsible. Finally, examining additional, especially geographically and phylogenetically more distant populations will allow us to examine whether for eye and pigment related traits a bias toward gene reuse exists in this species. Acknowledgments We thank Gregor Bračko and Teo Delić for help with collecting animals, and Valerija Zakšek for providing a template of the cave map. We thank Franco Fernandez, Hafasa Mojaddidi, Isaac Villalpando, Kaitlyn Vitangcol, and John Wallace for assistance with the animals. This symposium was generously sponsored by the Company of Biologists (http://www.biologists.com), the Paleontological Association (PA-GA201707), the American Microscopical Society, the Crustacean Society, and the SICB divisions DEDB, DEE, DIZ, DNB, and DPCB. Funding This work was supported by the Cave Conservancy Foundation, National Speleological Foundation, National Speleological Society, Old Timer’s Reunion Cave Society, and by the Slovenian Research Agency through the Research Core Funding P1-0184, research project N1-0069, and a Ph.D. grant (Contract No. 1000-12-0510) to Ž.F. Supplementary data Supplementary data available at ICB online. References Aspiras AC , Prasad R , Fong DW , Carlini DB , Angelini DR. 2012 . Parallel reduction in expression of the eye development gene hedgehog in separately derived cave populations of the amphipod Gammarus minus . J Evol Biol 25 : 995 – 1001 . Google Scholar CrossRef Search ADS PubMed Aspiras AC , Rohner N , Martineau B , Borowsky RL , Tabin CJ. 2015 . Melanocortin 4 receptor mutations contribute to the adaptation of cavefish to nutrient-poor conditions . Proc Natl Acad Sci U S A 112 : 9668 – 73 . Google Scholar CrossRef Search ADS PubMed Baldwin E , Beatty RA. 1941 . The pigmentation of cavernicolous animals . J Exp Biol 18 : 136 – 43 . Barrett RDH , Schluter D. 2008 . Adaptation from standing genetic variation . Trends Ecol Evol 23 : 38 – 44 . Google Scholar CrossRef Search ADS PubMed Bilandžija H , Ćetković H , Jeffery WR. 2012 . Evolution of albinism in cave planthoppers by a convergent defect in the first step of melanin biosynthesis . Evol Dev 14 : 196 – 203 . Google Scholar CrossRef Search ADS PubMed Bilandžija H , Ma L , Parkhurst A , Jeffery WR. 2013 . A potential benefit of albinism in Astyanax cavefish: downregulation of the oca2 gene increases tyrosine and catecholamine levels as an alternative to melanin synthesis . PLoS One 8 : e80823. Google Scholar CrossRef Search ADS PubMed Casane D , Rétaux S. 2016 . Evolutionary genetics of the cavefish Astyanax mexicanus . Adv Genet 95 : 117 – 59 . Google Scholar PubMed Cohen J. 1988 . Statistical power analysis for the behavioral sciences . 2nd ed. Hillsdale (NJ ): Lawrence Erlbaum Associates . Conte GL , Arnegard ME , Peichel CL , Schluter D. 2012 . The probability of genetic parallelism and convergence in natural populations . Proc R Soc B Biol Sci 279 : 5039 – 47 . Google Scholar CrossRef Search ADS Culver DC. 1982 . Cave life: evolution and ecology . Cambridge : Harvard University Press . Google Scholar CrossRef Search ADS Domyan ET , Shapiro MD. 2017 . Pigeonetics takes flight: evolution, development, and genetics of intraspecific variation . Dev Biol 427 : 241 – 50 . Google Scholar CrossRef Search ADS PubMed Duboué ER , Keene AC , Borowsky RL. 2011 . Evolutionary convergence on sleep loss in cavefish populations . Curr Biol 21 : 671 – 6 . Google Scholar CrossRef Search ADS PubMed Elipot Y , Hinaux H , Callebert J , Launay JM , Blin M , Rétaux S. 2014 . A mutation in the enzyme monoamine oxidase explains part of the Astyanax cavefish behavioural syndrome . Nat Commun 5 : 3647. Google Scholar CrossRef Search ADS PubMed Fišer Ž. 2017 . Evolution of reproductive isolation through adaptation to the subterranean environment [dissertation]. University of Ljubljana. 93 pp. Gross JB , Borowsky R , Tabin CJ. 2009 . A novel role for Mc1r in the parallel evolution of depigmentation in independent populations of the cavefish Astyanax mexicanus . PLoS Genet 5 : e1000326. Google Scholar CrossRef Search ADS PubMed Gross JB , Wilkens H. 2013 . Albinism in phylogenetically and geographically distinct populations of Astyanax cavefish arises through the same loss-of-function Oca2 allele . Heredity 111 : 122 – 30 . Google Scholar CrossRef Search ADS PubMed Hammer Ø , Harper DAT , Ryan PD. 2001 . PAST: paleontological statistics software package for education and data analysis . Palaeontol Electron 4 : 1 – 9 . Konec M. 2015 . Genetic differentiation and speciation in subterranean and surface populations of Asellus aquaticus (Crustacea: Isopoda) [dissertation]. University of Ljubljana. 87 pp. Konec M , Prevorčnik S , Sarbu SM , Verovnik R , Trontelj P. 2015 . Parallels between two geographically and ecologically disparate cave invasions by the same species, Asellus aquaticus (Isopoda, Crustacea) . J Evol Biol 28 : 864 – 75 . Google Scholar CrossRef Search ADS PubMed Kowalko JE , Rohner N , Linden TA , Rompani SB , Warren WC , Borowsky R , Tabin CJ , Jeffery WR , Yoshizawa M. 2013 . Convergence in feeding posture occurs through different genetic loci in independently evolved cave populations of Astyanax mexicanus . Proc Natl Acad Sci U S A 110 : 16933 – 8 . Google Scholar CrossRef Search ADS PubMed Kronforst MR , Barsh GS , Kopp A , Mallet J , Monteiro A , Mullen SP , Protas M , Rosenblum EB , Schneider CJ , Hoekstra HE. 2012 . Unraveling the thread of nature’s tapestry: the genetics of diversity and convergence in animal pigmentation . Pigment Cell Melanoma Res 25 : 411 – 33 . Google Scholar CrossRef Search ADS PubMed Manceau M , Domingues VS , Linnen CR , Rosenblum EB , Hoekstra HE. 2010 . Convergence in pigmentation at multiple levels: mutations, genes and function . Philos Trans R Soc B Biol Sci 365 : 2439 – 50 . Google Scholar CrossRef Search ADS Marchetti E , Montalenti SG. 1990 . The imperfect fitness of the male Asellus aquaticus (L.) for the identification of the receptive female . Rend Lincei Sci Fis Nat 1 : 327 – 33 . Google Scholar CrossRef Search ADS Meng F , Braasch I , Phillips JB , Lin X , Titus T , Zhang C , Postlethwait JH. 2013 . Evolution of the eye transcriptome under constant darkness in Sinocyclocheilus cavefish . Mol Biol Evol 30 : 1527 – 43 . Google Scholar CrossRef Search ADS PubMed Peichel CL , Marques DA. 2017 . The genetic and molecular architecture of phenotypic diversity in sticklebacks . Philos Trans R Soc B Biol Sci 372 : 20150486. Google Scholar CrossRef Search ADS Pérez-Moreno JL , Balázs G , Wilkins B , Herczeg G , Bracken-Grissom HD. 2017 . The role of isolation on contrasting phylogeographic patterns in two cave crustaceans . BMC Evol Biol 17 : 247. Google Scholar CrossRef Search ADS PubMed Placer L , Vrabec M , Celarc B. 2010 . The bases for understanding of the NW Dinarides and Istria Peninsula tectonics . Geologija 53 : 55 – 86 . Google Scholar CrossRef Search ADS Prevorčnik S , Blejec A , Sket B. 2004 . Racial differentiation in Asellus aquaticus (L.) (Crustacea: Isopoda: Asellidae) . Arch Hydrobiol 160 : 193 – 214 . Google Scholar CrossRef Search ADS Protas ME , Hersey C , Kochanek D , Zhou Y , Wilkens H , Jeffery WR , Zon LI , Borowsky R , Tabin CJ. 2006 . Genetic analysis of cavefish reveals molecular convergence in the evolution of albinism . Nat Genet 38 : 107 – 11 . Google Scholar CrossRef Search ADS PubMed Protas ME , Jeffery WR. 2012 . Evolution and development in cave animals: from fish to crustaceans . Wiley Interdiscip Rev Dev Biol 1 : 823 – 45 . Google Scholar CrossRef Search ADS PubMed Protas ME , Trontelj P , Patel NH. 2011 . Genetic basis of eye and pigment loss in the cave crustacean, Asellus aquaticus . Proc Natl Acad Sci U S A 108 : 5702 – 7 . Google Scholar CrossRef Search ADS PubMed Ridley M , Thompson DJ. 1979 . Size and mating in Asellus aquaticus (Crustacea: Isopoda) . Z Tierpsychol 51 : 380 – 97 . Google Scholar CrossRef Search ADS Sarbu SM , Kane TC , Kinkle BK. 1996 . A chemoautotrophically based cave ecosystem . Science 272 : 1953 – 5 . Google Scholar CrossRef Search ADS PubMed Seehausen O. 2015 . Process and pattern in cichlid radiations—inferences for understanding unusually high rates of evolutionary diversification . New Phytol 207 : 304 – 12 . Google Scholar CrossRef Search ADS PubMed Stern DL. 2013 . The genetic causes of convergent evolution . Nat Rev Genet 14 : 751 – 64 . Google Scholar CrossRef Search ADS PubMed Turk-Prevorčnik S , Blejec A . 1998 . Asellus aquaticus infernus, new subspecies (Isopoda: Asellota: Asellidae), from Romanian hypogean waters . J Crustacean Biol 18 : 763 – 73 . Google Scholar CrossRef Search ADS Verovnik R , Sket B , Prevorčnik S , Trontelj P. 2003 . Random amplified polymorphic DNA diversity among surface and subterranean populations of Asellus aquaticus (Crustacea: Isopoda) . Genetica 119 : 155 – 65 . Google Scholar CrossRef Search ADS PubMed Verovnik R , Sket B , Trontelj P. 2004 . Phylogeography of subterranean and surface populations of water lice Asellus aquaticus (Crustacea: Isopoda) . Mol Ecol 13 : 1519 – 32 . Google Scholar CrossRef Search ADS PubMed Wilkens H. 2010 . Genes, modules and the evolution of cave fish . Heredity 105 : 413 – 22 . Google Scholar CrossRef Search ADS PubMed Yamamoto Y , Jeffery WR. 2000 . Central role for the lens in cavefish eye degeneration . Science 289 : 631 – 3 . Google Scholar CrossRef Search ADS PubMed Yamamoto Y , Stock DW , Jeffery WR. 2004 . Hedgehog signalling controls eye degeneration in blind cavefish . Nature 431 : 844 – 7 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Integrative and Comparative Biology Oxford University Press

Common Genetic Basis of Eye and Pigment Loss in Two Distinct Cave Populations of the Isopod Crustacean Asellus aquaticus

Loading next page...
 
/lp/ou_press/common-genetic-basis-of-eye-and-pigment-loss-in-two-distinct-cave-7tBtCgkFvM
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: journals.permissions@oup.com
ISSN
1540-7063
eISSN
1557-7023
D.O.I.
10.1093/icb/icy028
Publisher site
See Article on Publisher Site

Abstract

Abstract Repeated evolution of similar phenotypes is a widespread phenomenon found throughout the living world and it can proceed through the same or different genetic mechanisms. Cave animals with their convergent traits such as eye and pigment loss, as well as elongated appendages, are a striking example of the evolution of similar phenotypes. Yet, few cave species are amenable to genetic crossing and mapping techniques making it challenging to determine the genetic mechanisms causing their similar phenotypes. To address this limitation, we have been developing Asellus aquaticus, a freshwater isopod crustacean, as a genetic model. Many of its cave populations originate from separate colonization events and thus independently evolved their similar cave-related phenotypes which differ from the still existent ancestral-like surface populations. In our prior work, we identified genomic regions responsible for eye and pigment loss in a single cave population from Slovenia. In this study we examined another, independently evolved cave population, also from Slovenia, and asked whether the same or different genomic regions are responsible for eye and pigment loss in the two cave populations. We generated F2 and backcross hybrids with a surface population, genotyped them for the previously identified genomic regions, and performed a complementation test by crossing individuals from the two cave populations. We found out that the same genomic regions are responsible for eye and pigment loss and that at least one of the genes causing pigment loss is the same in both cave populations. Future studies will identify the actual genes and mutations, as well as examine additional cave populations to see if the same genes are commonly associated with eye and pigment loss in this species. Introduction A frequently asked question in evolutionary biology is whether populations or species with similar phenotypes have evolved their convergent traits using the same or different genetic mechanisms. Multiple studies have shown that both scenarios occur in nature (reviewed in Manceau et al. 2010; Kronforst et al. 2012; Domyan and Shapiro 2017) and there is evidence that repeated use of the same gene is more common in closely related than in distantly related species (Conte et al. 2012). Additional model organisms and focal traits are needed to better understand and predict when and why each scenario occurs. Cave animals, famous for their eyeless and depigmented appearance, are an excellent group in which to study the evolution of similar phenotypes. As these have evolved numerous times in very different cave organisms (Culver 1982), it is possible to compare the evolution of similar characters in unrelated species, closely related species, or populations of the same species (Protas and Jeffery 2012). Among cave animals, genetic mechanisms behind the evolution of similar phenotypes have been most intensively studied in the cavefish Astyanax mexicanus. Its many independently evolved cave populations and the ability to perform genetic crosses and mapping experiments make it a superb model to study this phenomenon at the population level. So far, studies have mainly focused on pigmentation, eye size, appetite regulation, feeding behavior, and sleep. These studies demonstrated that many times the same genes or loci are responsible for similar phenotypes of independent populations (e.g., Protas et al. 2006; Gross et al. 2009; Duboué et al. 2011; Elipot et al. 2014; Aspiras et al. 2015), though there are examples where different genes or loci were found to be the cause (e.g., Kowalko et al. 2013). Such bias toward gene or genetic pathway reuse raises the question of why a particular gene is repeatedly targeted. Several possible explanations have been suggested including advantageous secondary consequences of mutations, ease of mutation of that particular gene, or lack of negative pleiotropic effects (Stern 2013). An illustrative example of gene reuse is the repeated mutation of the Oca2 gene that causes the albino phenotype in multiple cave populations of A. mexicanus (Protas et al. 2006; Gross and Wilkens 2013). Consequential enhancement of the catecholamine pathway that could control behaviors adaptive in the cave environment is suggested to be the reason for the repeated use of this gene (Bilandžija et al. 2013). Few studies have examined whether the same or different genetic mechanisms are responsible for similar phenotypes of cave animals other than A. mexicanus. One example is the two independently evolved cave-adapted species of planthoppers that live in different continents but show a blockage in the same step of the melanin synthesis pathway which is also the same step that A. mexicanus is blocked at (Protas et al. 2006; Bilandžija et al. 2012). Furthermore, studies of the amphipod crustacean, Gammarus minus, revealed that the gene hedgehog is differently expressed in cave and surface populations of this species—as was shown in A. mexicanus (Yamamoto et al. 2004; Aspiras et al. 2012). Contrary to what is known in A. mexicanus, research based on morphology and gene expression in another cavefish species, Sinocyclocheilus anophthalmus, suggests that lens independent mechanisms might be responsible for eye degeneration (Yamamoto and Jeffery 2000; Meng et al. 2013). Although much progress has been made toward understanding whether the same or different genetic mechanisms are responsible for similar phenotypes of different cave animals, what has been lacking are species similar to A. mexicanus where this question can be asked on the population level, with the potential of utilizing genetic crosses and mapping techniques. The isopod crustacean, Asellus aquaticus, is an exciting emerging model organism in which genetic mechanisms behind the evolution of similar phenotypes can be studied on the population level. Multiple cave populations of this isopod originated from separate colonization events in Italy, Slovenia, Hungary, and Romania and thus have independently evolved their similar, cave-related phenotypes with characteristic traits such as loss of eyes, depigmentation, and elongated appendages (Konec et al. 2015; Pérez-Moreno et al. 2017). Closely related ancestral-like surface populations are still existent and are composed of individuals with markedly different phenotypes. The best-studied are the cave populations from Slovenia, where in the Planina Cave two distinct populations live in two separate cave channels, the Pivka Chanel and the Rak Channel (Verovnik et al. 2003, 2004). Several lines of evidence suggest that these two populations invaded caves and evolved independently from each other. They formed mutually exclusive genetic clusters in a randomly amplified polymorphic DNA analysis (Verovnik et al. 2003), and they do not share any common mtDNA haplotypes (Verovnik et al. 2004). Although both populations are eyeless and depigmented, they differ consistently in a number of other morphometric and morphological characters (Prevorčnik et al. 2004). Recent studies using microsatellites revealed that there is no gene flow between populations (Konec 2015), and that both populations differ also in their behavioral response to light, thigmotactic preference, and ability to withstand water current (Fišer 2017). The likely explanation, why the two populations are so different even though they live in proximity, is that they are distributed north (Rak Channel population) and south (Pivka Channel population) of a major geological discontinuity, the Idrija Fault, that significantly shaped the regional geomorphology (Placer et al. 2010). The Idrija Fault is older than the age of the cave populations, which does probably not exceed 1.3 Mya (Konec et al. 2015), making a vicariant split of a formerly contiguous subterranean population unlikely. As cave and surface forms can interbreed in captivity (Baldwin and Beatty 1941; Protas et al. 2011), genetic mapping of traits commonly associated with cave life can be applied (Protas et al. 2011). In the cave population from the Pivka Channel of Planina Cave genomic regions responsible for eye and pigment variation have already been mapped (Protas et al. 2011). Surprisingly, multiple mechanisms of both eye and pigment loss were found within this single population. Complete eye loss mapped to a different locus than eye reduction, while albinism appears to be achieved either by a recessive genotype at one locus or doubly recessive genotypes at two other loci. To investigate whether the same or different genetic mechanisms are responsible for the repeated evolution of eye and pigment loss in different cave populations of A. aquaticus, we mapped these traits for the first time in another cave population, i.e., Rak Channel of Planina Cave. Cave individuals from the Rak Channel and surface individuals from a closely related surface population were used to generate F2 and backcross individuals. These were genotyped for genes that were previously found to be associated with eye and pigment loss in the cave population from the Pivka Channel of Planina Cave. Additionally, we performed a complementation test between cave individuals from Rak and Pivka Channels. Geographic proximity and a common timeframe of cave habitat colonization of both cave populations imply that their surface ancestors had probably shared the same gene pool (Verovnik et al. 2003, 2004). Thus, we hypothesized that alleles responsible for eye and pigment loss in these two distinct cave populations originated from the same standing genetic variation. We predicted that genomic regions responsible for eye and pigment loss in both cave populations overlap. Methods Animals We collected cave individuals from the Rak Channel of Planina Cave and surface individuals from Rakov Škocjan (Fig. 1). The surface population from Rakov Škocjan inhabits the upstream surface section of the same river that flows underground in the Rak Channel and is closely related to the Rak Channel cave population (Verovnik et al. 2003, 2004). Animals were maintained in the laboratory as described by Protas et al. (2011). Cave males were mated to surface females and cave females were mated to surface males to generate F1 hybrids. These were mated either to siblings to generate F2 hybrids, or F1 females were mated back to cave males to generate backcross hybrids. F2 and backcross individuals had a high mortality, and often times only one or two individuals per brood survived embryonic development. We pooled together F2 and backcross individuals from many crosses and raised them until they were at least 4 mm long. A total of 82 individuals were phenotyped and genotyped. Fig. 1 View largeDownload slide Map of Planina Cave area showing the Rak and Pivka Channels. Upper left: the black rectangle marks the region in Europe where the Planina Cave is situated. Bottom: Planina Cave area magnified. Red dots indicate sampling sites of cave and surface Asellus aquaticus populations mentioned in the text. The omega-like black symbols mark cave openings. Dashed lines are used where the precise watercourse is unknown. Fig. 1 View largeDownload slide Map of Planina Cave area showing the Rak and Pivka Channels. Upper left: the black rectangle marks the region in Europe where the Planina Cave is situated. Bottom: Planina Cave area magnified. Red dots indicate sampling sites of cave and surface Asellus aquaticus populations mentioned in the text. The omega-like black symbols mark cave openings. Dashed lines are used where the precise watercourse is unknown. Phenotyping F2 and backcross individuals from the Rak Channel crosses were anesthetized in a 0.4% clove oil solution and phenotyped for eye and pigment related traits using a Leica S8 Apo stereomicroscope and LAS Core software. Each individual was phenotyped for presence and color of eye pigment. Presence and color of head and body pigmentation generally followed suit with eye pigment and was not included in the analyses. As head and body pigmentation is expressed in at least two different patterns (Protas et al. 2011), individuals were phenotyped for the pattern of this pigmentation. Additionally, each individual was phenotyped for eye (i.e., ommatidia or fragmented ommatidia) presence. Genetic markers Genetic markers were chosen based on the previously constructed genetic map for A. aquaticus and results of mapping eye and pigment related traits in the cave population from the Pivka Channel of the Planina Cave (Protas et al. 2011). In this population, four genomic regions located at four separate linkage groups were identified to be responsible for eye and pigment traits. Within each of these four regions, we searched for genetic markers by which we could reliably distinguish between the Rak Channel and Rakov Škocjan origin. We tried multiple markers within each region and had to reject several as they were either monomorphic or too polymorphic between both populations, or the Rak Channel individuals were not uniform for the marker genotype. Ultimately, we selected one suitable genetic marker per region (Supplementary Table S1). Three of these markers were in genes from the existing genetic map, i.e., disconnected, nckx30, and pax2, though the polymorphism used for genotyping was not necessarily the same for the Rak Channel and Pivka Channel populations. In the Pivka Channel crosses, the genotype of disconnected was associated with presence versus absence of eye pigment, the genotype of nckx30 was associated with light versus dark eye pigment, and the genotype of pax2 was associated with red versus orange or brown eye pigment. To mark the fourth region, which was related to stellate versus diffuse head and body pigmentation pattern and eye presence versus absence in the Pivka Channel crosses, we could not use markers in genes from the existing genetic map as these were polymorphic within the Rak Channel population. Fortuitously, we had been adding additional genes onto the genetic map and found that the A. aquaticus ortholog of the gene son of bowl (sob) mapped to this fourth region and was associated with the phenotype of stellate versus diffuse head and body pigmentation pattern and eye presence versus absence in the Pivka Channel crosses. As the marker in sob was found suitable also for the Rak Channel and Rakov Škocjan individuals, we used it to mark this last genomic region. For details regarding the identification of the genetic marker in sob see Supplementary Information S1. Genotyping To genotype F2 and backcross individuals from the Rak Channel crosses, DNA was extracted from two legs of each animal using QiaAmp Micro Kit (Qiagen). Genetic markers in genes disconnected, nckx30, pax2, and sob were amplified using primers specified in Supplementary Table S1. PCR was performed using GoTaq Green Master Mix (Promega); a 53°C annealing temperature and a 1 min extension time for 30 cycles. Products were purified using ExoSAP-IT (Thermo Fisher Scientific) and sent for sequencing to ELIM Biopharm or MCLAB. Sequences were analyzed using FinchTV 1.4.0 (Geospiza, Inc.) software. For markers in disconnected, nckx30, and pax2 sequence chromatograms were inspected to determine genotype. F2 and backcross individuals were either homozygous for the cave allele, homozygous for the surface allele, or heterozygous. The marker in sob was genotyped differently as primers used to amplify it only amplified the surface allele but not the cave allele. Therefore, successful amplification indicated either a homozygous surface or a heterozygous genotype, but we could not discriminate between the two. To confirm the homozygous cave genotype in individuals for which amplification was not successful, we used the forward primer and a different reverse primer to obtain a product, sequenced the product, and inspected the sequence chromatogram to confirm the cave genotype. Additionally, backcross individuals from the Pivka Channel crosses generated in our previous study (Protas et al. 2011) were genotyped for the marker in sob using stored DNA isolates. This was needed because the gene sob was not included in the genetic map constructed in that study (see also above). The marker in sob was amplified and genotyped as described above. However, as these were backcross individuals (F1 females were bred to cave males), successful amplification indicated a heterozygous genotype only. Genotypes of markers in disconnected, nckx30, and pax2, as well as phenotypic traits for these hybrids, were previously determined in Protas et al. (2011). Statistical analyses F2 and backcross individuals from the Rak Channel crosses were categorized according to their phenotypic trait values in groups that matched the groups from Protas et al. (2011); only traits related to markers in disconnected and nckx30 were grouped slightly differently. This was needed because we observed that some cave individuals from the Rak Channel population exhibited light red pigment in the eye region (as in Fig. 2A′). Thus, for disconnected, we pooled the light red and the unpigmented phenotype into the same group, i.e., unpigmented, and treated individuals with red, orange, or brown eye pigmentation as pigmented. Consequently, for nckx30 orange versus red or brown eye pigmentation was investigated instead of light (i.e., light red or orange) versus dark (i.e., red or brown) eye pigmentation as in Protas et al. (2011). Fig. 2 View largeDownload slide Phenotypes of F2 and backcross hybrids between the Rak Channel and surface individuals. A, B, C, D, E, and F show animals’ heads, while A′, B′, C′, D′, E′, and F′ focus on the eyes of the same animals. Several phenotypes were observed: no pigment (not shown), light red (A, A′), red (B, B′, C, C′), orange (D, D′), and brown (E, E′, F, F′) eye pigment. The head and body pigmentation pattern was either stellate (E) or diffuse (D). In addition, ommatidia were either present (A′, C′, D′, F′) or absent (B′, E′), but fragmented ommatidia (not shown) were observed only in a few individuals. Fig. 2 View largeDownload slide Phenotypes of F2 and backcross hybrids between the Rak Channel and surface individuals. A, B, C, D, E, and F show animals’ heads, while A′, B′, C′, D′, E′, and F′ focus on the eyes of the same animals. Several phenotypes were observed: no pigment (not shown), light red (A, A′), red (B, B′, C, C′), orange (D, D′), and brown (E, E′, F, F′) eye pigment. The head and body pigmentation pattern was either stellate (E) or diffuse (D). In addition, ommatidia were either present (A′, C′, D′, F′) or absent (B′, E′), but fragmented ommatidia (not shown) were observed only in a few individuals. Fisher’s exact tests were performed for each genetic marker and its associated trait to explore whether a significant association between genotype and phenotype exists. The strength of association was assessed using Cramér’s V, which varies from 0 (no association) to 1 (complete association). All calculations were performed in PAST 3 (Hammer et al. 2001). Complementation test Complementation crosses were carried out between males from the Rak Channel and females from the Pivka Channel populations. Males were collected in cave Škratovka, which is located very close to the Planina Cave and harbors the same cave population as is present in the Rak Channel (Konec 2015; Fig. 1). In the Pivka Channel we collected females engaged in precopula (i.e., precopulatory mate guarding behavior during which a male firmly holds a female) to ensure that they were ready to mate. In the laboratory, females were gently separated from males of the same population, and each female was put in a separate Petri dish together with one male from the Rak Channel population. Pairs were kept in constant darkness at 10°C and left to feed ad libitum on decaying leaves of the black alder (Alnus glutinosa). Eventually, two out of seven crosses were successful and produced offspring. As there is evidence that sperm storage does not occur in A. aquaticus (Ridley and Thompson 1979; Marchetti and Montalenti 1990), we were confident that these were hybrids. Most of these died soon after hatching, but nine offspring of the same cross reached 18–43 days of age and 1.5–2 mm in length, and were fixed in 96% ethanol. As ommatidia phenotype can alter upon preservation in ethanol (M. Protas, personal observation), only pigment related traits could be reliably determined. For phenotyping, we used an Olympus SZX2-ILLT stereomicroscope with the Olympus ColorView III camera and Cell^B 2.8. (Olympus) software. Results Interestingly, F2 and backcross hybrids between cave individuals from the Rak Channel and surface individuals showed almost exactly the same phenotypes as were previously observed in backcross hybrids from the Pivka Channel crosses (Fig. 2). F2 and backcross hybrids had no pigment, light red, red, orange, or brown pigment in the eye region. In addition, the pattern of head and body pigmentation was either stellate or diffuse. Finally, F2 and backcross hybrids were either eyeless (i.e., had no ommatidia) or eyed (i.e., had ommatidia or fragmented ommatidia). In contrast to the Pivka Channel crosses, the majority of eyed hybrids had ommatidia and fragmented ommatidia were observed in few individuals only. All investigated eye and pigment related phenotypic traits of cave individuals from the Rak Channel were found to be associated with the same genes as in cave individuals from the Pivka Channel (Table 1). Fisher’s exact tests showed a statistically significant association in all cases. The genotype of disconnected was associated with the phenotype of presence versus absence of eye pigmentation, and the cave allele related to the unpigmented phenotype. The genotype of nckx30 was associated with the phenotype of orange versus red or brown eye pigmentation, and the cave allele related to the orange phenotype. The genotype of pax2 was associated with the phenotype of red versus orange or brown eye pigmentation, and the cave allele related to the red phenotype. Finally, the genotype of sob was associated with the phenotype of stellate versus diffuse head and body pigmentation pattern and with the phenotype of eye presence versus absence, and the cave allele related to the stellate and the eyeless phenotype, respectively. The association was strongest for phenotypes related to disconnected, nckx30, and pax2 with Cramér’s V reaching from 0.75 to 0.89, and a bit less strong for phenotypes related to sob with Cramér’s V reaching from 0.53 to 0.6. According to Cohen (1988), all associations can be considered as having a large effect size. The strength of association for disconnected and pax2 related phenotypes was similar between the Rak Channel and Pivka Channel crosses, but somewhat less strong in the Rak Channel crosses for nckx30 and sob related phenotypes. Raw data on phenotypes and genotypes of all Rak Channel hybrids used in this study are available in Supplementary Table S2. Table 1 Association between genotypes and phenotypes in the Pivka Channel hybrids and in the Rak Channel hybrids Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel hybrids comprise backcross individuals examined previously in Protas et al. (2011), but data for sob were obtained in this study. Rak Channel hybrids comprise F2 and backcross individuals examined in this study. Hybrids were either homozygous for the cave allele (CC), heterozygous (CS), or homozygous for the surface allele (SS). Pivka Channel backcross individuals lack the SS genotype as backcrossing was done only with the cave parent individuals. For the Rak Channel hybrids the CS and SS genotypes of sob could not be distinguished due to the genotyping method used. In the Pivka Channel hybrids unpigmented individuals had no pigment in the eye region, whereas in the Rak Channel hybrids unpigmented individuals either had no or light red pigment in the eye region. For the genetic marker in nckx30, phenotypes were grouped into light (light red or orange) versus dark (red or brown) eye pigment for the Pivka Channel hybrids, whereas phenotypes were grouped into orange versus red or brown eye pigment for the Rak Channel hybrids. See main text for a more detailed explanation. Table 1 Association between genotypes and phenotypes in the Pivka Channel hybrids and in the Rak Channel hybrids Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel backcross individuals Fisher’s exact test Cramér’s V Rak Channel F2 and backcross individuals Fisher’s exact test Cramér’s V disconnected CC CS disconnected CC CS SS  unpigmented 82 12 P < 0.001 0.865  unpigmented 21 2 6 P < 0.001 0.830  pigmented 1 89  pigmented 0 41 11 nckx30 CC CS nckx30 CC CS SS  light 21 0 P < 0.001 0.912  orange 9 0 0 P < 0.001 0.750  dark 3 57  red/brown 5 30 5 pax2 CC CS pax2 CC CS SS  red 30 6 P < 0.001 0.818  red 10 2 0 P < 0.001 0.894  orange/brown 2 55  orange/brown 0 26 14 sob CC CS sob CC CS or SS  stellate 24 1 P < 0.001 0.899  stellate 8 2 P = 0.001 0.598  diffuse 2 34  diffuse 4 19 sob CC CS sob CC CS or SS  eyeless 53 2 P < 0.001 0.922  eyeless 14 5 P < 0.001 0.533  eyed 3 75  eyed 9 43 Pivka Channel hybrids comprise backcross individuals examined previously in Protas et al. (2011), but data for sob were obtained in this study. Rak Channel hybrids comprise F2 and backcross individuals examined in this study. Hybrids were either homozygous for the cave allele (CC), heterozygous (CS), or homozygous for the surface allele (SS). Pivka Channel backcross individuals lack the SS genotype as backcrossing was done only with the cave parent individuals. For the Rak Channel hybrids the CS and SS genotypes of sob could not be distinguished due to the genotyping method used. In the Pivka Channel hybrids unpigmented individuals had no pigment in the eye region, whereas in the Rak Channel hybrids unpigmented individuals either had no or light red pigment in the eye region. For the genetic marker in nckx30, phenotypes were grouped into light (light red or orange) versus dark (red or brown) eye pigment for the Pivka Channel hybrids, whereas phenotypes were grouped into orange versus red or brown eye pigment for the Rak Channel hybrids. See main text for a more detailed explanation. Complementation tests between cave individuals from the Rak Channel and Pivka Channel populations showed lack of complementation for pigment related traits (Fig. 3). All nine F1 hybrid offspring from the same complementation cross had no pigment in the eye region, head, or body, just like their parents. The presence or absence of ommatidia could not be reliably determined (see the “Methods” section). Fig. 3 View largeDownload slide Complementation test between cave individuals from the Rak Channel and the Pivka Channel populations showed lack of complementation. Male from the Rak Channel population (father of the complementation cross) (A), female from the Pivka Channel population (a representative individual as mother of the complementation cross died before preservation) (B), as well as F1 hybrid offspring (C) had no pigment in the eye region, head, or body. Fig. 3 View largeDownload slide Complementation test between cave individuals from the Rak Channel and the Pivka Channel populations showed lack of complementation. Male from the Rak Channel population (father of the complementation cross) (A), female from the Pivka Channel population (a representative individual as mother of the complementation cross died before preservation) (B), as well as F1 hybrid offspring (C) had no pigment in the eye region, head, or body. Discussion Repeated evolution of similar phenotypes is a widespread phenomenon found throughout the living world and the onset of modern molecular tools made it possible to investigate the genetic mechanisms behind it (Peichel and Marques 2017). Here we examined this issue in two independently evolved cave populations of the freshwater isopod A. aquaticus. In line with our prediction, we have shown that in both populations the same genomic regions are responsible for loss of pigment and eyes (i.e., ommatidia). These convergent traits might be caused by the same genes via the same or different mutations or by different but tightly linked genes. Results of the complementation test allow us to narrow down these possibilities for the pigment related traits. Crossing individuals from the Pivka Channel population to individuals from the Rak Channel population yielded unpigmented individuals. As no complementation occurred, it is highly likely that the same and not different genes are responsible for pigment loss in both cave populations, though it is currently impossible to say whether this is due to the same or different mutations. One complication of the complementation test is that in the Pivka Channel population pigment loss can be achieved in two different ways: (i) by two copies of the cave allele for the gene that encodes absence of pigmentation and is marked by disconnected, or (ii) by two copies of the cave allele for the gene that encodes orange pigment and is marked by nckx30, and two copies of the cave allele for the gene that encodes red pigment and is marked by pax2 (Protas et al. 2011). Thus, convergence in pigment loss in the two cave populations from the Rak and Pivka Channels is either due to the same gene marked by disconnected, the same two genes marked by nckx30 and pax2, respectively, or the same three genes marked by nck30, pax2, and disconnected. Currently, we are not able to distinguish between these possibilities though this can be dissected in the future by crossing orange, red, and unpigmented F2 hybrids generated from the Rak Channel population to F2 hybrids of the same colors generated from the Pivka Channel population. Nevertheless, at least one of the genes responsible for pigment loss is apparently in common between the two cave populations. Unfortunately, we were unable to test the complementation of the eyeless phenotype and thus cannot exclude any of the plausible genetic mechanisms (i.e., different but linked genes, same or different mutations of the same gene) for the observed convergent phenotype of this trait. Future carefully planned complementation crosses will hopefully provide additional insight into the genetic mechanisms responsible for eye loss. Initially, our interpretation of the two distinct ways in which pigment loss is achieved in the Pivka Channel population was that perhaps pigment pathways were accumulating mutations as pigmentation is useless in the dark cave environment and selection for it is relaxed (Wilkens 2010; Protas et al. 2011). However, this evolutionary mechanism is much less likely to produce the degeneration of pigment pathways in the same way in two independently evolved populations as seems to be the case in the populations from the Rak and Pivka Channels. If all of the same genes are responsible for pigment loss in these two cave populations, repeated utilization of the same genetic pathways would be better explained by adaptive pleiotropic effect(s) of the mutations that cause pigment loss. Bilandžija et al. (2013) suggested that this is a possible reason for the repeated use of the Oca2 gene causing pigment loss in the cavefish A. mexicanus. In populations adapting to a novel environment or selective pressures, beneficial alleles originate either from standing genetic variation or new mutations (Barrett and Schluter 2008). The two cave populations examined in this study derive from surface ancestors that had probably shared the same gene pool (Verovnik et al. 2003, 2004), and we have shown that the same genomic regions and at least one of the same genes is responsible for their convergent unpigmented and eyeless phenotypes. Considering this, our hypothesis that standing genetic variation present in the founding populations was the source of “cave” alleles remains the most parsimonious explanation for the evolution of similar phenotypes in these two cave populations. Fixation of alleles from standing genetic variation has been previously shown to be the mechanism of evolutionary change in multiple examples (reviewed in Seehausen 2015; Casane and Rétaux 2016; Peichel and Marques 2017). Yet, based on currently available data, new mutations cannot be excluded as the source of alleles responsible for the convergent cave phenotypes. Repeated, de novo mutation of the same gene in independently evolved populations of the same species has also been documented in many examples including the Oca2 gene causing pigment loss in the cavefish A. mexicanus (Protas et al. 2006; Gross and Wilkens 2013). Although our data showed a striking match between the genomic regions responsible for pigment and eye loss in cave populations from the Pivka and Rak Channels, we also observed some minor differences between the two populations. First, in the Pivka Channel crosses, individuals homozygous for the cave alleles for the marker in disconnected generally had no eye pigment, whereas in the Rak Channel crosses such individuals generally either had no pigment or light red pigment in the eye region. This difference could be due to the interaction of additional genes in either of these populations. Second, in the Pivka Channel individuals we were able to map eye presence versus absence but also eye size because many crosses had fragmented ommatidia (Protas et al. 2011). Differently, in the Rak Channel crosses most of the F2 and backcross individuals were either eyeless or had ommatidia and individuals with fragmented ommatidia were too few to enable us to map eye size. Thus, it is possible that the alleles responsible for fragmented ommatidia are present in the Rak Channel population at a lower frequency than in the Pivka Channel population. Or, perhaps individuals with those alleles had a higher mortality during the breeding and crossing period in the lab and therefore did not survive to be phenotyped and genotyped. Another difference between cave populations from the Rak and Pivka Channels was that the associations between nckx30 and orange eye pigment, and also sob and eye loss as well as stellate pigmentation pattern were stronger in the Pivka Channel population. Although this could imply that different genes in the same genomic regions are responsible for these same traits in the two populations, due to the results of complementation tests, this seems unlikely and other explanations are more probable. The weaker association could be due to the smaller sample size available for the Rak Channel crosses in this study compared with the Pivka Channel crosses from Protas et al. (2011). Another possible explanation for the weaker association between sob and eye loss is that due to high mortality of the Rak Channel crosses we were unable to separate them into families (like we did for the Pivka Channel crosses) but had to pool them from many different parents. Therefore, if the allele responsible for eye loss is not fixed in the Rak Channel population (as is apparently the case in the Pivka Channel population), eyed individuals homozygous for the cave allele for the marker in sob would be expected and more common. The presence of such individuals would make the association between sob and eye loss appear weaker compared with the Pivka Channel crosses in which only families of cave parents with complete eye loss were examined for this association. Currently, we have only examined the genetic background of eye and pigment loss in the Pivka Channel and Rak Channel populations that both live in Planina Cave, although in its two distinct channels. Still to examine are other independently arisen cave populations that are geographically and phylogenetically more distant from the two populations examined so far. We imagine that those cave populations might be less likely to have the same four genomic regions responsible for eye and pigment loss. Yet if the same surface lineage founded these cave populations, it could still be possible that standing genetic variation in the ancestral surface population is responsible for the phenotypic variation in some of these populations. One population that would be especially interesting to examine in future studies is the A. aquaticus infernus population from Romania which lives in a chemoautotrophic cave environment that is ecologically very different from the environment in Planina Cave (Sarbu et al. 1996; Turk-Prevorčnik et al. 1998; Konec et al. 2015). If different genetic mechanisms do exist in some cave populations, we would expect to find them in this cave population. To lead back to the original question, the same genomic regions seem to be responsible for similar phenotypes in A. aquaticus cave populations from the two channels of Planina Cave. Complementation tests provide further evidence that at least one of the same genes is responsible for pigment loss in these two populations. Further work identifying the actual genes and mutations responsible for eye and pigment traits will help us to more specifically address the evolutionary history of these traits and why the same regions are responsible. Finally, examining additional, especially geographically and phylogenetically more distant populations will allow us to examine whether for eye and pigment related traits a bias toward gene reuse exists in this species. Acknowledgments We thank Gregor Bračko and Teo Delić for help with collecting animals, and Valerija Zakšek for providing a template of the cave map. We thank Franco Fernandez, Hafasa Mojaddidi, Isaac Villalpando, Kaitlyn Vitangcol, and John Wallace for assistance with the animals. This symposium was generously sponsored by the Company of Biologists (http://www.biologists.com), the Paleontological Association (PA-GA201707), the American Microscopical Society, the Crustacean Society, and the SICB divisions DEDB, DEE, DIZ, DNB, and DPCB. Funding This work was supported by the Cave Conservancy Foundation, National Speleological Foundation, National Speleological Society, Old Timer’s Reunion Cave Society, and by the Slovenian Research Agency through the Research Core Funding P1-0184, research project N1-0069, and a Ph.D. grant (Contract No. 1000-12-0510) to Ž.F. Supplementary data Supplementary data available at ICB online. References Aspiras AC , Prasad R , Fong DW , Carlini DB , Angelini DR. 2012 . Parallel reduction in expression of the eye development gene hedgehog in separately derived cave populations of the amphipod Gammarus minus . J Evol Biol 25 : 995 – 1001 . Google Scholar CrossRef Search ADS PubMed Aspiras AC , Rohner N , Martineau B , Borowsky RL , Tabin CJ. 2015 . Melanocortin 4 receptor mutations contribute to the adaptation of cavefish to nutrient-poor conditions . Proc Natl Acad Sci U S A 112 : 9668 – 73 . Google Scholar CrossRef Search ADS PubMed Baldwin E , Beatty RA. 1941 . The pigmentation of cavernicolous animals . J Exp Biol 18 : 136 – 43 . Barrett RDH , Schluter D. 2008 . Adaptation from standing genetic variation . Trends Ecol Evol 23 : 38 – 44 . Google Scholar CrossRef Search ADS PubMed Bilandžija H , Ćetković H , Jeffery WR. 2012 . Evolution of albinism in cave planthoppers by a convergent defect in the first step of melanin biosynthesis . Evol Dev 14 : 196 – 203 . Google Scholar CrossRef Search ADS PubMed Bilandžija H , Ma L , Parkhurst A , Jeffery WR. 2013 . A potential benefit of albinism in Astyanax cavefish: downregulation of the oca2 gene increases tyrosine and catecholamine levels as an alternative to melanin synthesis . PLoS One 8 : e80823. Google Scholar CrossRef Search ADS PubMed Casane D , Rétaux S. 2016 . Evolutionary genetics of the cavefish Astyanax mexicanus . Adv Genet 95 : 117 – 59 . Google Scholar PubMed Cohen J. 1988 . Statistical power analysis for the behavioral sciences . 2nd ed. Hillsdale (NJ ): Lawrence Erlbaum Associates . Conte GL , Arnegard ME , Peichel CL , Schluter D. 2012 . The probability of genetic parallelism and convergence in natural populations . Proc R Soc B Biol Sci 279 : 5039 – 47 . Google Scholar CrossRef Search ADS Culver DC. 1982 . Cave life: evolution and ecology . Cambridge : Harvard University Press . Google Scholar CrossRef Search ADS Domyan ET , Shapiro MD. 2017 . Pigeonetics takes flight: evolution, development, and genetics of intraspecific variation . Dev Biol 427 : 241 – 50 . Google Scholar CrossRef Search ADS PubMed Duboué ER , Keene AC , Borowsky RL. 2011 . Evolutionary convergence on sleep loss in cavefish populations . Curr Biol 21 : 671 – 6 . Google Scholar CrossRef Search ADS PubMed Elipot Y , Hinaux H , Callebert J , Launay JM , Blin M , Rétaux S. 2014 . A mutation in the enzyme monoamine oxidase explains part of the Astyanax cavefish behavioural syndrome . Nat Commun 5 : 3647. Google Scholar CrossRef Search ADS PubMed Fišer Ž. 2017 . Evolution of reproductive isolation through adaptation to the subterranean environment [dissertation]. University of Ljubljana. 93 pp. Gross JB , Borowsky R , Tabin CJ. 2009 . A novel role for Mc1r in the parallel evolution of depigmentation in independent populations of the cavefish Astyanax mexicanus . PLoS Genet 5 : e1000326. Google Scholar CrossRef Search ADS PubMed Gross JB , Wilkens H. 2013 . Albinism in phylogenetically and geographically distinct populations of Astyanax cavefish arises through the same loss-of-function Oca2 allele . Heredity 111 : 122 – 30 . Google Scholar CrossRef Search ADS PubMed Hammer Ø , Harper DAT , Ryan PD. 2001 . PAST: paleontological statistics software package for education and data analysis . Palaeontol Electron 4 : 1 – 9 . Konec M. 2015 . Genetic differentiation and speciation in subterranean and surface populations of Asellus aquaticus (Crustacea: Isopoda) [dissertation]. University of Ljubljana. 87 pp. Konec M , Prevorčnik S , Sarbu SM , Verovnik R , Trontelj P. 2015 . Parallels between two geographically and ecologically disparate cave invasions by the same species, Asellus aquaticus (Isopoda, Crustacea) . J Evol Biol 28 : 864 – 75 . Google Scholar CrossRef Search ADS PubMed Kowalko JE , Rohner N , Linden TA , Rompani SB , Warren WC , Borowsky R , Tabin CJ , Jeffery WR , Yoshizawa M. 2013 . Convergence in feeding posture occurs through different genetic loci in independently evolved cave populations of Astyanax mexicanus . Proc Natl Acad Sci U S A 110 : 16933 – 8 . Google Scholar CrossRef Search ADS PubMed Kronforst MR , Barsh GS , Kopp A , Mallet J , Monteiro A , Mullen SP , Protas M , Rosenblum EB , Schneider CJ , Hoekstra HE. 2012 . Unraveling the thread of nature’s tapestry: the genetics of diversity and convergence in animal pigmentation . Pigment Cell Melanoma Res 25 : 411 – 33 . Google Scholar CrossRef Search ADS PubMed Manceau M , Domingues VS , Linnen CR , Rosenblum EB , Hoekstra HE. 2010 . Convergence in pigmentation at multiple levels: mutations, genes and function . Philos Trans R Soc B Biol Sci 365 : 2439 – 50 . Google Scholar CrossRef Search ADS Marchetti E , Montalenti SG. 1990 . The imperfect fitness of the male Asellus aquaticus (L.) for the identification of the receptive female . Rend Lincei Sci Fis Nat 1 : 327 – 33 . Google Scholar CrossRef Search ADS Meng F , Braasch I , Phillips JB , Lin X , Titus T , Zhang C , Postlethwait JH. 2013 . Evolution of the eye transcriptome under constant darkness in Sinocyclocheilus cavefish . Mol Biol Evol 30 : 1527 – 43 . Google Scholar CrossRef Search ADS PubMed Peichel CL , Marques DA. 2017 . The genetic and molecular architecture of phenotypic diversity in sticklebacks . Philos Trans R Soc B Biol Sci 372 : 20150486. Google Scholar CrossRef Search ADS Pérez-Moreno JL , Balázs G , Wilkins B , Herczeg G , Bracken-Grissom HD. 2017 . The role of isolation on contrasting phylogeographic patterns in two cave crustaceans . BMC Evol Biol 17 : 247. Google Scholar CrossRef Search ADS PubMed Placer L , Vrabec M , Celarc B. 2010 . The bases for understanding of the NW Dinarides and Istria Peninsula tectonics . Geologija 53 : 55 – 86 . Google Scholar CrossRef Search ADS Prevorčnik S , Blejec A , Sket B. 2004 . Racial differentiation in Asellus aquaticus (L.) (Crustacea: Isopoda: Asellidae) . Arch Hydrobiol 160 : 193 – 214 . Google Scholar CrossRef Search ADS Protas ME , Hersey C , Kochanek D , Zhou Y , Wilkens H , Jeffery WR , Zon LI , Borowsky R , Tabin CJ. 2006 . Genetic analysis of cavefish reveals molecular convergence in the evolution of albinism . Nat Genet 38 : 107 – 11 . Google Scholar CrossRef Search ADS PubMed Protas ME , Jeffery WR. 2012 . Evolution and development in cave animals: from fish to crustaceans . Wiley Interdiscip Rev Dev Biol 1 : 823 – 45 . Google Scholar CrossRef Search ADS PubMed Protas ME , Trontelj P , Patel NH. 2011 . Genetic basis of eye and pigment loss in the cave crustacean, Asellus aquaticus . Proc Natl Acad Sci U S A 108 : 5702 – 7 . Google Scholar CrossRef Search ADS PubMed Ridley M , Thompson DJ. 1979 . Size and mating in Asellus aquaticus (Crustacea: Isopoda) . Z Tierpsychol 51 : 380 – 97 . Google Scholar CrossRef Search ADS Sarbu SM , Kane TC , Kinkle BK. 1996 . A chemoautotrophically based cave ecosystem . Science 272 : 1953 – 5 . Google Scholar CrossRef Search ADS PubMed Seehausen O. 2015 . Process and pattern in cichlid radiations—inferences for understanding unusually high rates of evolutionary diversification . New Phytol 207 : 304 – 12 . Google Scholar CrossRef Search ADS PubMed Stern DL. 2013 . The genetic causes of convergent evolution . Nat Rev Genet 14 : 751 – 64 . Google Scholar CrossRef Search ADS PubMed Turk-Prevorčnik S , Blejec A . 1998 . Asellus aquaticus infernus, new subspecies (Isopoda: Asellota: Asellidae), from Romanian hypogean waters . J Crustacean Biol 18 : 763 – 73 . Google Scholar CrossRef Search ADS Verovnik R , Sket B , Prevorčnik S , Trontelj P. 2003 . Random amplified polymorphic DNA diversity among surface and subterranean populations of Asellus aquaticus (Crustacea: Isopoda) . Genetica 119 : 155 – 65 . Google Scholar CrossRef Search ADS PubMed Verovnik R , Sket B , Trontelj P. 2004 . Phylogeography of subterranean and surface populations of water lice Asellus aquaticus (Crustacea: Isopoda) . Mol Ecol 13 : 1519 – 32 . Google Scholar CrossRef Search ADS PubMed Wilkens H. 2010 . Genes, modules and the evolution of cave fish . Heredity 105 : 413 – 22 . Google Scholar CrossRef Search ADS PubMed Yamamoto Y , Jeffery WR. 2000 . Central role for the lens in cavefish eye degeneration . Science 289 : 631 – 3 . Google Scholar CrossRef Search ADS PubMed Yamamoto Y , Stock DW , Jeffery WR. 2004 . Hedgehog signalling controls eye degeneration in blind cavefish . Nature 431 : 844 – 7 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Integrative and Comparative BiologyOxford University Press

Published: Jun 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off