Experimental tests for heritable morphological color plasticity in non-native brown trout (Salmo trutta) populations

Redundancy analysis, genome-wide association studies and the pigmentation of brown trout (Salmo trutta L.)

The association of molecular variants with phenotypic variation is a main issue in biology, often tackled with genome-wide association studies (GWAS). GWAS are challenging, with increasing, but still limited, use in evolutionary biology. We used redundancy analysis (RDA) as a complimentary ordination approach to single- and multitrait GWAS to explore the molecular basis of pigmentation variation in brown trout (Salmo trutta) belonging to wild populations impacted by hatchery fish. Based on 75,684 single nucleotide polymorphic (SNP) markers, RDA, single- and multitrait GWAS allowed the extraction of 337 independent colour patterning loci (CPLs) associated with trout pigmentation traits, such as the number of red and black spots on flanks. Collectively, these CPLs (i) mapped onto 35 out of 40 brown trout linkage groups indicating a polygenic genomic architecture of pigmentation, (ii) were found to be associated with 218 candidate genes, including 197 genes formerly mentioned in the literature associated to skin pigmentation, skin patterning, differentiation or structure notably in a close relative, the rainbow trout (Onchorhynchus mykiss), and (iii) related to functions relevant to pigmentation variation (e.g., calcium- and ion-binding, cell adhesion). Annotated CPLs include genes with well-known pigmentation effects (e.g., PMEL, SLC45A2, SOX10), but also markers associated with genes formerly found expressed in rainbow or brown trout skins. RDA was also shown to be useful to investigate management issues, especially the dynamics of trout pigmentation submitted to several generations of hatchery introgression.

Heritable morphological differentiation in salmonids from two distinct stream types

We tested for phenotype-to-habitat associations in brown trout Salmo trutta populations from two ecologically different habitat types; i.e., groundwater and surface-water-fed streams. Additionally, we raised captive offspring from two such populations under standardised conditions to test whether potential phenotypic differentiation would be passed on to offspring. We found analogous differentiation by habitat in multiple wild populations. Some of these morphological differences were at least partially inherited by offspring. We suggest that this could have implications for both scientists and fisheries authorities studying or managing trout populations.

Melanin-Based Skin Coloration Predicts Antioxidant Capacity in the Brown Trout (Salmo trutta)

In many vertebrate species, individuals exhibit large variation in the degree of melanin-based coloration on their body. Dark and pale individuals differ in diverse physiological and behavioral traits, suggesting that melanic coloration may reveal individual quality. However, research into the relationships between physiological and skin traits, in terms of melanin-based skin coloration, in wild fish is scant. Our correlative study aimed at investigating the relationships between physiology and melanin-based coloration of the skin of free-living brown trout (Salmo trutta Linnaeus, 1758). We scrutinized the relationships between body condition (body mass and Fulton's K condition factor), oxidative status (plasma total antioxidant capacity and amount of pro-oxidant molecules), and the degree of melanin-based skin coloration assessed by digital photography and image analysis. We found heavier fish to be, on average, darker colored than paler conspecifics. Moreover, a significant covariation between plasma total antioxidant capacity and melanic coloration was noted. Our findings suggest that the melanin-based coloration of brown trout serves as a signal to communicate a better antioxidant defense to conspecifics.

Genetic and molecular evidence that brown trout Salmo trutta belonging to the Danubian lineage are a single biological species

We tested for reproductive isolation between Salmo trutta abanticus, S.t. labrax, S.t. caspius and S.t. fario by conducting crosses to produce F1 and F2 offspring. We also estimated the extent of genetic divergence between all three entities by examining sequence variation across the coI, d-loop and cytb mitochondrial genes. All of the F1 cross-types were successfully produced. After 2 years of culturing, F2 generation were produced as well. Fertilization, hatching and survival rates and hatching performance of F1 and F2 generations were evaluated. F2 generation had similar performance to their parent. Fertilization, hatching, larval survival rate and hatchery performance of F1 and F2 generation were similar except pure bred F2 S.t. abanticus. Purebred F1 individuals shared similar coloration patterns and spots with their parents but direction of the hybridization appeared to be decisive on morphology of hybrids. Some of the hybrids exhibit different morphological characters than their parents. Based on partial alignments of the three genes, phylogenetic analysis showed that these S. trutta are gathering within the same clade and appeared as monophyletic group. We found that there were some morphologic and genetic variation among S. trutta subspecies but the degree of variation does not warrant species level recognition. These findings indicate that the four subspecies constitute a single biological entity, corresponding to different morphs of the Danubian lineage. We therefore recommend that S. trutta belonging to Danubian lineage in Turkey be referred to as Salmo trutta and that strains be named according to location, such as Abant, Caspian, Black Sea and Anatolian.

Maternal allocation of carotenoids increases tolerance to bacterial infection in brown trout

Life-history theory predicts that iteroparous females allocate their resources differently among different breeding seasons depending on their residual reproductive value. In iteroparous salmonids there is typically much variation in egg size, egg number, and in the compounds that females allocate to their clutch. These compounds include various carotenoids whose functions are not sufficiently understood yet. We sampled 37 female and 35 male brown trout from natural streams, collected their gametes for in vitro fertilizations, experimentally produced 185 families in 7 full-factorial breeding blocks, raised the developing embryos singly (n = 2960), and either sham-treated or infected them with Pseudomonas fluorescens. We used female redness (as a measure of carotenoids stored in the skin) and their allocation of carotenoids to clutches to infer maternal strategies. Astaxanthin contents largely determined egg colour. Neither egg weight nor female size was correlated with the content of this carotenoid. However, astaxanthin content was positively correlated with larval growth and with tolerance against P. fluorescens. There was a negative correlation between female skin redness and the carotenoid content of their eggs. Although higher astaxanthin contents in the eggs were associated with an improvement of early fitness-related traits, some females appeared not to maximally support their current offspring as revealed by the negative correlation between female red skin colouration and egg carotenoid content. This correlation was not explained by female size and supports the prediction of a maternal trade-off between current and future reproduction.

Maladaptation and phenotypic mismatch in hatchery-reared Atlantic salmon Salmo salar released in the wild

Changes in body shape, fluctuating asymmetry (FA) and crypsis were compared among Atlantic salmon Salmo salar fry kept as controls in captivity and those released and subsequently recaptured in the wild according to a before-after-control-impact (BACI) design. Hatchery fish that survived in the wild became more cryptic and displayed a much lower incidence of fin erosion and of asymmetric individuals than control fish kept in captivity. Significant differences in body shape were also apparent, and survivors had longer heads, thicker caudal peduncles and a more streamlined body shape than hatchery controls as early as 20 days following stocking, most likely as a result of phenotypic plasticity and non-random, selective mortality of maladapted phenotypes. Hatchery-reared fish typically perform poorly in the wild and the results of this study indicate that this may be due to phenotypic mismatch, i.e. because hatcheries generate fish that are phenotypically mismatched to the natural environment.