256
M.L. Cancela et al.
have been confounded by the difficulty of disentangling environmental from genetic
effects (Endler 1986). Accordingly, there has been a great interest in identifying the
footprints of selection at the molecular level. Three issues come to mind, namely
genomic inferences on (1) the genetic basis and architecture of local adaptations; i.e.
the genomic basis of why resident individuals commonly have higher fitness than
individuals from other environments (Kawecki and Ebert 2004), (2) demonstrating
and understanding the genomic basis of evolution in response to the on-going global
change (Reusch and Wood 2007, Gienapp et al. 2008), but also studies of (3) the
evolutionary impact of selective harvesting (Coltman 2008, Allendorf et al. 2008).
All three issues are highly relevant in relation to marine fish evolution as well as for
the management of genetic resources in marine fish. First of all, if marine fish are
locally adapted then conservation of the local populations is essential, not only to
maintain adaptive genetic diversity within the species, but also to assure the survival
of the species in particular geographical areas/habitats and to maximise productivity.
Immigrant non-adapted fish would have lower survival and/or growth; reproduction
would be strongly impeded. Secondly, global change is expected to alter the distribution and abundance of many marine species and related fisheries (Roessig et al.
2004, Dulvy et al. 2008). Little is known whether the fish populations will be able
to adapt genetically over this – in an evolutionary context – rather short period with
rapidly changing environmental variables such as temperature (Davis et al. 2005).
Finally, fisheries induced evolution has been inferred based on changes in age and
size at maturation for fish (Jørgensen et al. 2008), albeit without direct genetic evidence of change at the molecular level. If fisheries are altering important life-history
traits, fitness under natural conditions as well as population productivity may be
strongly reduced.
A number of studies have demonstrated environmental selection on single protein markers (e.g. Sick 1965, Christiansen and Frydenberg 1974) or DNA markers
(e.g. Case et al. 2005, Hemmer-Hansen et al. 2007) suggesting genetic adaptations
to the local environment. Although such studies have provided valuable insights
into the evolutionary processes in marine fish, they, nevertheless, only infer selection of a particular gene, i.e. not at the genomic level. Instead general information
is warranted about the proportion and types of genes subject to selection, and
the genetic architecture (number, function and interplay among genes) of phenotypic traits. Therefore, genomic approaches allowing simultaneous assessment of
many genes potentially subject to selection would provide a major leap forward
in relation to understanding adaptation to natural and anthropogenic drivers of
evolution.
Selection in space and time can subsequently be identified by a number of different approaches. The “candidate gene approach” exclusively investigates variation in
genes thought to play a major role for particular adaptive traits of interest. Common
relatively long genomic sequences are generated from individuals from different
populations and subjected to statistical methods which enable the identification of
molecular footprints of selection (Guinand et al. 2004). Such a directed approach has
many advantages when investigating particular traits with very good “candidates”. A
historical example is the characterisation of the natural variation and physiological
M.L. Cancela et al.
have been confounded by the difficulty of disentangling environmental from genetic
effects (Endler 1986). Accordingly, there has been a great interest in identifying the
footprints of selection at the molecular level. Three issues come to mind, namely
genomic inferences on (1) the genetic basis and architecture of local adaptations; i.e.
the genomic basis of why resident individuals commonly have higher fitness than
individuals from other environments (Kawecki and Ebert 2004), (2) demonstrating
and understanding the genomic basis of evolution in response to the on-going global
change (Reusch and Wood 2007, Gienapp et al. 2008), but also studies of (3) the
evolutionary impact of selective harvesting (Coltman 2008, Allendorf et al. 2008).
All three issues are highly relevant in relation to marine fish evolution as well as for
the management of genetic resources in marine fish. First of all, if marine fish are
locally adapted then conservation of the local populations is essential, not only to
maintain adaptive genetic diversity within the species, but also to assure the survival
of the species in particular geographical areas/habitats and to maximise productivity.
Immigrant non-adapted fish would have lower survival and/or growth; reproduction
would be strongly impeded. Secondly, global change is expected to alter the distribution and abundance of many marine species and related fisheries (Roessig et al.
2004, Dulvy et al. 2008). Little is known whether the fish populations will be able
to adapt genetically over this – in an evolutionary context – rather short period with
rapidly changing environmental variables such as temperature (Davis et al. 2005).
Finally, fisheries induced evolution has been inferred based on changes in age and
size at maturation for fish (Jørgensen et al. 2008), albeit without direct genetic evidence of change at the molecular level. If fisheries are altering important life-history
traits, fitness under natural conditions as well as population productivity may be
strongly reduced.
A number of studies have demonstrated environmental selection on single protein markers (e.g. Sick 1965, Christiansen and Frydenberg 1974) or DNA markers
(e.g. Case et al. 2005, Hemmer-Hansen et al. 2007) suggesting genetic adaptations
to the local environment. Although such studies have provided valuable insights
into the evolutionary processes in marine fish, they, nevertheless, only infer selection of a particular gene, i.e. not at the genomic level. Instead general information
is warranted about the proportion and types of genes subject to selection, and
the genetic architecture (number, function and interplay among genes) of phenotypic traits. Therefore, genomic approaches allowing simultaneous assessment of
many genes potentially subject to selection would provide a major leap forward
in relation to understanding adaptation to natural and anthropogenic drivers of
evolution.
Selection in space and time can subsequently be identified by a number of different approaches. The “candidate gene approach” exclusively investigates variation in
genes thought to play a major role for particular adaptive traits of interest. Common
relatively long genomic sequences are generated from individuals from different
populations and subjected to statistical methods which enable the identification of
molecular footprints of selection (Guinand et al. 2004). Such a directed approach has
many advantages when investigating particular traits with very good “candidates”. A
historical example is the characterisation of the natural variation and physiological