Unique haplotypes of co-segregating major histocompatibility class II <Emphasis Type="Italic">A</Emphasis> and class II <Emphasis Type="Italic">B</Emphasis> alleles in Atlantic salmon (<Emphasis Type="Italic">Salmo salar</Emphasis>) give rise to diverse class II genotypes

Sequence-based typing of a breeding population (G1) consisting of 84 Atlantic salmon individuals revealed the presence of 7 Sasa-DAA and 7 Sasa-DAB expressed alleles. Subsequent typing of 1,182 individuals belonging to 33 families showed that Sasa-DAA and Sasa-DAB segregate as haplotypes. In total seven unique haplotypes were established, with frequencies in the population studied ranging from 0.01 to 0.49. Each haplotype is characterized by a unique minisatellite marker size embedded in the 3' untranslated region of the Sasa-DAA gene. These data corroborate the fact that Atlantic salmon express a single class II locus, consisting of tightly linked class II A and class B genes. The seven haplotypes give rise to 15 genotypes with frequencies varying between 0.01 and 0.23; 21 class II homozygous individuals were present in the G1 population. We also studied the frequency distribution in another breeding population (G4, n=374) using the minisatellite marker. Only one new marker size was present, suggesting the presence of one new class II haplotype. The marker frequency distribution in the G4 population differed markedly from the G1 population. The genomic organizations of two Sasa-DAA and Sasa-DAB alleles were determined, and supported the notion that these alleles belong to the same locus. In contrast to other studies of salmonid class II sequences, phylogenetic analyses of brown trout and Atlantic class II A and class II B sequences provided support for trans-species polymorphism.     

Natural selection acts on Atlantic salmon major histocompatibility (MH) variability in the wild

Pathogen-driven balancing selection is thought to maintain polymorphism in major histocompatibility (MH) genes. However, there have been few empirical demonstrations of selection acting on MH loci in natural populations. To determine whether natural selection on MH genes has fitness consequences for wild Atlantic salmon in natural conditions, we compared observed genotype frequencies of Atlantic salmon (Salmo salar) surviving in a river six months after their introduction as eggs with frequencies expected from parental crosses. We found significant differences between expected and observed genotype frequencies at the MH class II alpha locus, but not at a MH class I-linked microsatellite or at seven non-MH-linked microsatellite loci. We therefore conclude that selection at the MH class II alpha locus was a result of disease-mediated natural selection, rather than any demographic event. We also show that survival was associated with additive allelic effects at the MH class II alpha locus. Our results have implications for both the conservation of wild salmon stocks and the management of disease in hatchery fish. We conclude that natural or hatchery populations have the best chance of dealing with episodic and variable disease challenges if MH genetic variation is preserved both within and among populations.     

Major histocompatibility polymorphism associated with resistance towards amoebic gill disease in Atlantic salmon (Salmo salar L.)

The association between major histocompatibility (MH) polymorphism and the severity of infection by amoebic gill disease (AGD) was investigated across 30 full sibling families of Atlantic salmon. Individuals were challenged with AGD for 19days and then their severity of infection scored by histopathological examination of the gills. Fish were then genotyped for the MH class I (Sasa-UBA) and MH class II alpha (Sasa-DAA) genes using polymorphic repeats embedded within the 3' untranslated regions of the Sasa-UBA and Sasa-DAA genes. High variation in the severity of infection was observed across the sample material, ranging from 0% to 85% gill filaments infected. In total, seven Sasa-DAA-3UTR and ten Sasa-UBA-3UTR marker alleles were identified across the 30 families. A significant association between the marker allele Sasa-DAA-3UTR 239 and a reduction in AGD severity was detected. There was also a significant association found between AGD severity and the presence of two Sasa-DAA-3UTR genotypes. While the associations between MH allele/genotypes and AGD severity reported herein may be statistically significant, the small sample sizes observed for some alleles and genotypes means these associations should be considered as suggestive and future research is required to verify their biological significance.     

Screening of Mhc variation in Atlantic salmon (Salmo salar): a comparison of restriction fragment length polymorphism (RFLP), denaturing gradient gel electrophoresis (DGGE) and sequencing

We compared three different molecular methods currently used for screening of Mhc variation in population studies of Atlantic salmon. Restriction fragment length polymorphism (RFLP) of the entire class II gene detected 22 haplotypes. Seventeen exon 2 sequences were obtained from individuals carrying the 22 haplotypes, two of which had not been detected by RFLP. The six alleles (27%) detected by RFLP and not by exon 2 sequencing probably resulted from sequence variation outside exon 2. Within exon 2, RFLP differentiated 88% of the sequences. Alternatively, denaturing gradient gel electrophoresis (DGGE) performed under two run conditions detected 94% of the sequence variation. Both RFLP using different probes, and the two PCR-based methods using three different primer pairs, suggest that there is only a single Mhc class II B gene in the Baltic populations of Atlantic salmon.     

High levels of MHC class II allelic diversity in lake trout from Lake Superior

Sequence variation in a 216 bp portion of the major histocompatibility complex (MHC) II B1 domain was examined in 74 individual lake trout (Salvelinus namaycush) from different locations in Lake Superior. Forty-three alleles were obtained which encoded 71-72 amino acids of the mature protein. These sequences were compared with previous data obtained from five Pacific salmon species and Atlantic salmon using the same primers. Although all of the lake trout alleles clustered together in the neighbor-joining analysis of amino acid sequences, one amino acid allelic lineage was shared with Atlantic salmon (Salmo salar), a species in another genus which probably diverged from Salvelinus more than 10-20 million years ago. As shown previously in other salmonids, the level of nonsynonymous nucleotide substitution (dN) exceeded the level of synonymous substitution (dS). The level of nucleotide diversity at the MHC class II B1 locus was considerably higher in lake trout than in the Pacific salmon (genus Oncorhynchus). These results are consistent with the hypothesis that lake trout colonized Lake Superior from more than one refuge following the Wisconsin glaciation. Recent population bottlenecks may have reduced nucleotide diversity in Pacific salmon populations.     

Contrasting responses to selection in class I and class IIα major histocompatibility-linked markers in salmon

Comparison of levels and patterns of genetic variation in natural populations either across loci or against neutral expectation can yield insight into locus-specific differences in the strength and direction of evolutionary forces. We used both approaches to test the hypotheses on patterns of selection on major histocompatibility (MH)-linked markers. We performed temporal analyses of class I and class IIα MH-linked markers and eight microsatellite loci in two Atlantic salmon populations in Ireland on two temporal scales: over six decades and 9 years in the rivers Burrishoole and Delphi, respectively. We also compared contemporary Burrishoole and Delphi samples with nearby populations for the same loci. On comparing patterns of temporal and spatial differentiation among classes of loci, the class IIα MH-linked marker was consistently identified as an outlier compared with patterns at the other microsatellite loci or neutral expectation. We found higher levels of temporal and spatial heterogeneity in heterozygosity (but not in allelic richness) for the class IIα MH-linked marker compared with microsatellites. Tests on both within- and among-population differentiation are consistent with directional selection acting on the class IIα-linked marker in both temporal and spatial comparisons, but only in temporal comparisons for the class I-linked marker. Our results indicate a complex pattern of selection on MH-linked markers in natural populations of Atlantic salmon. These findings highlight the importance of considering selection on MH-linked markers when using these markers for management and conservation purposes.     

Molecular evolution at Mhc genes in two populations of chinook salmon Oncorhynchus tshawytscha

The DNA sequences of four exons of the MHC (major histocompatibilty complex) were examined in chinook salmon ( Oncorhynchus tshawytscha ) from an interior (Nechako River) and a coastal (Harrison River) population in the Fraser River drainage of British Columbia. Mhc class I A1, A2 and A3 sequences and a class II B1 sequence were obtained by PCR from each of 16–20 salmon from each population. The class I A1 and a pair of linked A2–A3 exons were derived from two different classical salmonid class I genes, Sasa-A and Onmy-UA , respectively. Allelic variation for B1, A1 and A2 was characterized by the high levels of nonsynonymous substitution indicative of the effects of natural selection on Mhc domains that contain peptide binding regions. The number of alleles detected at each of the four exons ranged from three ( B1 ) to 22 ( A1 ), but levels of nucleotide sequence divergence at all four exons were low relative to classical mammalian Mhc genes. The nucleotide similarity among alleles ranged between 89 and 99% over all exons, and all four domains possessed only two major sequence motifs. Allelic distributions at B1, A1 and A3 confirmed the genetic distinctiveness of the Harrison and Nechako chinook salmon populations revealed in previous studies. The two major allelic motifs of B1 and A1 segregated strongly between the populations. In spite of evidence that allelic diversity at these chinook salmon Mhc exons has been generated by selection, the level and distribution of diversity in the two salmon populations strongly reflected the demographic history of the species, which has been characterized by repeated bottlenecks and isolation-by-distance in glacial refugia.     

Patterns of variability at the major histocompatibility class II alpha locus in Atlantic salmon contrast with those at the class I locus

In order to investigate the mechanisms creating and maintaining variability at the major histocompatibility (MH) class II alpha (DAA) locus we examined patterns of polymorphism in two isolated Atlantic salmon populations which share a common post-glacial origin. As expected from their common origin, but contrary to the observation at the MH class I locus, these populations shared the majority of DAA alleles: out of 17 sequences observed, 11 were common to both populations. Recombination seems to play a more important role in the origin of new alleles at the class II alpha locus than at the class I locus. A greater than expected proportion of sites inferred to be positively selected (potentially peptide binding residues, PBRs) were found to be involved in recombination events, suggesting a mechanism for increasing MH variability through an interaction between recombination and natural selection. Thus it appears that although selection and recombination are important mechanisms for the evolution of both class II alpha and class I loci in the Atlantic salmon, the pattern of variability differs markedly between these classes of MH loci.     

Genetic variation within and among domesticated Atlantic salmon broodstocks in British Columbia, Canada

Atlantic salmon have been reared in the British Columbia, Canada aquaculture industry since the early 1980s. No breeding programmes spanned the entire production period and pedigree records were not kept for broodstocks prior to or since importation. Of the three recognized industry strains, two are of European ancestry ('Mowi' from Norway and 'McConnell' from Scotland) and one is of North American heritage ('Cascade' from Gaspe, Quebec). We evaluated the amount and distribution of genetic variation within industry broodstocks by surveying microsatellite variation at 11 loci in 20 broodstock groups sampled from major production facilities. Allelic richness averaged 10.9 (range 5.8-13.8), compared with a value of 20.3 obtained for a North American wild population. Pairwise genetic distances (D(S)) between samples within strains were generally less than those between strains, with samples attributed to the same strain clustering together in a neighbour-joining dendrogram. Nevertheless, average distances between samples within the European strains were high (0.41 for Mowi; 0.71 for McConnell) but lower (0.06) for the Cascade strain. The reduced intra-sample and increased intra-strain genetic variation observed for the BC domesticated samples compared with wild populations was similar to observations for European domesticated Atlantic salmon. Evidence of introgression of the Cascade strain into European broodstocks was provided by the presence of large Ssa202 alleles (confined to North America in wild populations) in some Mowi and McConnell samples. Introgression likely also contributed to the decreased intercontinental genetic distance for the domesticated samples of this study compared with that observed for wild populations.     

MHC polymorphism and disease resistance in Atlantic salmon (<Emphasis Type="Italic">Salmo salar</Emphasis>); facing pathogens with single expressed major histocompatibility class I and class II loci

Few studies have yet addressed the functional aspects of MHC molecules in fish. To lay the foundation for this, we evaluated the association between disease resistance and MHC class I and class II polymorphism in Atlantic salmon. Standardized disease challenge trials were performed on a semi-wild Atlantic salmon population with subsequent MHC typing and statistical analysis. The pathogens employed were infectious salmon anaemia virus (ISAV) causing infectious salmon anaemia and the Aeromonas salmonicida bacteria causing furunculosis. The material consisted of 1,182 Atlantic salmon from 33 families challenged with A. salmonicida and 1,031 Atlantic salmon from 25 families challenged with ISAV. We found highly significant associations between resistance towards infectious diseases caused by both pathogens and MH class I and class II polymorphism in Atlantic salmon. The observed associations were detected due to independently segregating MH class I and class II single loci, and inclusion of a large number of fish allowing an extensive statistical analysis.     

Invasion success and genetic diversity of introduced populations of guppies Poecilia reticulata in Australia

High genetic diversity is thought to characterize successful invasive species, as the potential to adapt to new environments is enhanced and inbreeding is reduced. In the last century, guppies, Poecilia reticulata, repeatedly invaded streams in Australia and elsewhere. Quantitative genetic studies of one Australian guppy population have demonstrated high additive genetic variation for autosomal and Y-linked morphological traits. The combination of colonization success, high heritability of morphological traits, and the possibility of multiple introductions to Australia raised the prediction that neutral genetic diversity is high in introduced populations of guppies. In this study we examine genetic diversity at nine microsatellite and one mitochondrial locus for seven Australian populations. We used mtDNA haplotypes from the natural range of guppies and from domesticated varieties to identify source populations. There were a minimum of two introductions, but there was no haplotype diversity within Australian populations, suggesting a founder effect. This was supported by microsatellite markers, as allelic diversity and heterozygosity were severely reduced compared to one wild source population, and evidence of recent bottlenecks was found. Between Australian populations little differentiation of microsatellite allele frequencies was detected, suggesting that population admixture has occurred historically, perhaps due to male-biased gene flow followed by bottlenecks. Thus success of invasion of Australia and high additive genetic variance in Australian guppies are not associated with high levels of diversity at molecular loci. This finding is consistent with the release of additive genetic variation by dominance and epistasis following inbreeding, and with disruptive and negative frequency-dependent selection on fitness traits.     

Adaptive and Neutral Genetic Variation and Colonization History of Atlantic Salmon, Salmo salar

Synopsis I combined neutral microsatellite markers with the major histocompatibility complex (MHC) class IIB to study genetic differentiation and colonization history in Atlantic salmon, Salmo salar, in the Baltic Sea and in the north-eastern Atlantic. Baltic salmon populations have lower levels of microsatellite genetic variation, in terms of heterozygosity and allelic richness than Atlantic populations, confirming earlier findings with other genetic markers, suggesting that the Baltic Sea populations have been exposed to genetic bottlenecks, most likely at a founding event. On the other hand, the level of MHC variation was similar in the Baltic and in the north-eastern Atlantic, indicating that positive balancing selection has increased the level of MHC-variation. Both microsatellite and MHC class IIB genetic variation give strong support to the hypothesis that the Baltic salmon are of a biphyletic origin, the southern population in this study is strongly differentiated from both the northern Baltic salmon populations and from the north-eastern Atlantic populations. Salmon may have colonized the northern Baltic Sea either from the south, via the so called “N?rke strait” or from the north, via a proposed historical connection between the White Sea and the northern Baltic. At microsatellites, no significant isolation-by distance was found at either colonization route. At the MHC, populations were significantly isolated by distance when assuming that colonization occurred via the “N?rke strait”.     

Zygosity at the major histocompatibility class IIB locus predicts susceptibility to Renibacterium salmoninarum in Atlantic salmon (Salmo salar L.)

Major histocompatibility (MH) class II genes play an important role in the vertebrate immune response. Here, we investigate the relationship between Atlantic salmon (Salmo salar) MH class IIB zygosity and susceptibility to Renibacterium salmoninarum, the causal agent of bacterial kidney disease. By combining DNA sequences from the salmon MH class IIB gene with quantitative ELISA data on R. salmoninarum antigen levels, we found that MH class IIB homozygotes were significantly more susceptible to R. salmoninarum than heterozygotes. These findings are discussed in the context of current evolutionary theory.     

Allelic polymorphism in MHC class II B in four populations of Atlantic salmon (Salmo salar)

We sequenced exon 2 of the MHC class II B gene in Atlantic salmon from the Baltic Sea and identified 17 different exon 2 alleles among 22 different restriction fragment length polymorphism haplotypes. The sequences differed at between 1 and 34 bases. Two different tests were used to estimate the importance of recombination in the generation of new alleles. Recombination events appear to have occurred between three and nine times. Only two pairs of sequences differed by less than five nucleotides, minimizing the importance of point mutations for generating new alleles. Phylogenetic analysis showed that sequences did not cluster according to populations, and genetic distances between populations were small compared to those obtained by allele frequency data. These results, together with the similarity found between exon 2 sequences from Baltic salmon and Norwegian salmon, indicate that all of the identified alleles were present in the ancient salmon population colonizing the Baltic rivers after the last glaciation.     

Global major histocompatibility Class II β (mh-IIβ)-polymorphism in Arctic charr Salvelinus alpinus

This study explored the use of the gene encoding the β subunit of the major histocompatibility (MH) receptor as a population marker in Arctic charr Salvelinus alpinus. The use of this polymorphic marker allowed differentiation of the S. alpinus lineages previously defined using mitochondrial DNA (mtDNA) but also allowed differentiation between the populations studied within those lineages. The majority of the variation observed here occurred prior to the last glaciation event. Nevertheless, all S. alpinus populations were differentiated using both MH Class II β (mh-IIβ) sequences and allelic frequencies. The fact that all the populations studied presented high rates of non-synonymous: synonymous substitutions and high levels of interpopulation variation, suggested mh-IIβ as an ideal marker to assess differentiation among S. alpinus populations in ways that may represent divergence both by genetic drift and natural adaptation to the local environment.     

Comparative analysis of population structure across environments and geographical scales at major histocompatibility complex and microsatellite loci in Atlantic salmon (Salmo salar)

Evidence of selection acting on major histocompatibility complex (MHC) genes has been illustrated with the analysis of their nucleotide sequences and allele frequency distribution. Comparing the patterns of population differentiation at neutral markers and MHC genes in the wild may provide further insights about the relative role of selection and neutrality in shaping their diversity. In this study, we combine both methods to assess the role of selection on a MHC gene in Atlantic salmon. We compare variation at a MHC class II B locus and microsatellites among 14 samples from seven different rivers and seven subpopulations within a single river system covering a variety of habitats and different geographical scales. We show that diversifying selection is acting on the sites involved in antigen presentation and that balancing selection maintains a high level of polymorphism within populations. Despite important differences in habitat type, the comparison of the population structure at MHC and microsatellites on large geographical scales reveals a correlation between patterns of differentiation, indicating that drift and migration have been more important than selection in shaping population differentiation at the MHC locus. In contrast, strong discrepancies between patterns of population differentiation at the two types of markers provides support for the role of selection in shaping population structure within rivers. Together, these results confirm that natural selection is influencing MHC gene diversity in wild Atlantic salmon although neutral forces may also be important in their evolution.     

Genetic differentiation among natural European populations of Atlantic salmon, Salmo salar L., from drainages of the Atlantic Ocean

Previously published allelic frequencies at four polymorphic protein coding loci were used as a basis for examining genetic relationships among 19 European populations of Atlantic salmon, Salmo salar L.--exclusive of Baltic drainages--from the Barents Sea to Spain. The data did not support a model of distinct ancestral (e.g. Boreal and Celtic) origins, but were consistent with all populations descending from a single ancestral group within this region with genetically diverged populations drawn together through limited local migrations.     

MH class IIα polymorphism in local and global adaptation of Arctic charr (<Emphasis Type="Italic">Salvelinus alpinus</Emphasis> L.)

Arctic charr, a highly plastic salmonid that inhabits the circumpolar region, colonized its current environment after the last glaciation. Recent colonization limits the capacity of many techniques to define and characterize constituent populations. As a novel approach, we used the major histocompatibility (MH) class IIalpha gene polymorphism as a marker that would characterize the genetic divergence of global Arctic charr populations caused by drift and by local adaptation to pathogens. We were able to detect significant isolation of all the lineages previously defined by mitochondrial DNA sequencing and also isolation of some populations within those groups. We found that most of the polymorphism of the class IIalpha gene was distributed globally, which indicates ancestral selection; however, in most cases, distinctive allele frequencies and specific haplotypes distinguished each population suggesting that recent selection has also occurred. Although all studied populations showed similar MH class IIalpha polymorphisms, we also found variation in which particular amino acid positions were polymorphic and which were constant in the different populations studied. This variation provides a greater adaptive capacity for the MH class IIalpha receptors in Arctic charr and is yet another illustration of the extraordinary plasticity of the species.     

Geographic heterogeneity in natural selection on an MHC locus in sockeye salmon

Balancing selection maintains high levels of polymorphism and heterozygosity in genes of the MHC (major histocompatibility complex) of vertebrate organisms, and promotes long evolutionary persistence of individual alleles and strongly differentiated allelic lineages. In this study, genetic variation at the MHC class II DAB-beta1 locus was examined in 31 populations of sockeye salmon (Oncorhynchus nerka) inhabiting the Fraser River drainage of British Columbia, Canada. Twenty-five percent of variation at the locus was partitioned among sockeye populations, as compared with 5% at neutral genetic markers. Geographic heterogeneity of balancing selection was detected among four regions in the Fraser River drainage and among lake systems within regions. High levels of beta1 allelic diversity and heterozygosity, as well as distributions of alleles and allelic lineages that were more even than expected for a neutral locus, indicated the presence of balancing selection in populations throughout much of the interior Fraser drainage. However, proximate populations in the upper Fraser region, and four of six populations from the lower Fraser drainage, exhibited much lower levels of genetic diversity and had beta1 allele frequency distributions in conformance with those expected for a neutral locus, or a locus under directional selection. Pair-wise FST values for beta1 averaged 0.19 and tended to exceed the corresponding values estimated for neutral loci at all levels of population structure, although they were lower among populations experiencing balancing selection than among other populations. The apparent heterogeneity in selection resulted in strong genetic differentiation between geographically proximate populations with and without detectable levels of balancing selection, in stark contrast to observations at neutral loci. The strong partitioning and complex structure of beta1 diversity within and among sockeye populations on a small geographic scale illustrates the value of incorporating adaptive variation into conservation planning for the species.     

Genetic Variation within and Between Domesticated Chinook Salmon, Oncorhynchus tshawytscha, Strains and their Progenitor Populations

Domesticated chinook salmon strains in British Columbia (BC), Canada are believed to have originated primarily from populations of the Big Qualicum (BQ) River and Robertson Creek (RC) on Vancouver Island in the early 1980s. The number of parental fish that gave rise to the domesticated strains and their subsequent breeding history during approximately five ensuing generations of domestication were not documented. Genetic variation at 13 microsatellite loci was examined in samples from two domesticated strains and the two progenitor populations to determine the genetic relationships among them. The domesticated strains had lower allelic diversity and tended to have lower levels of expected heterozygosity than did the BQ and RC progenitor populations. Only three alleles over all 13 loci were detected in the domesticated strains that were not present in the BQ and RC samples, whereas the progenitor strains possessed over 25 (BQ) and 43 (RC) private alleles. Genetic distance and F_ST values also indicated a closer relationship of the domesticated strains with the BQ than the RC population. One domesticated strain had a significant excess of heterozygosity compared with that expected under conditions of mutation-drift equilibrium, indicative of a recent genetic bottleneck. Genetic differentiation between the domesticated strains was as great as that distinguishing them from the progenitor populations, indicating that the genetic base of domesticated chinook salmon could be increased by hybridization. The existence of genetically distinct domesticated strains of chinook salmon in coastal BC generates the need for an evaluation of potential genetic interactions between domesticated escapees and natural spawning populations.