Investigating Diadromy in Fishes and Its Loss in an -Omics Era

Summary

Diadromy, the predictable movements of individuals between marine and freshwater environments, is biogeographically and phylogenetically widespread across fishes. Thus, despite the high energetic and potential fitness costs involved in moving between distinct environments, diadromy appears to be an effective life history strategy. Yet, the origin and molecular mechanisms that underpin this migratory behavior are not fully understood. In this review, we aim first to summarize what is known about diadromy in fishes; this includes the phylogenetic relationship among diadromous species, a description of the main hypotheses regarding its origin, and a discussion of the presence of non-migratory populations within diadromous species. Second, we discuss how recent research based on -omics approaches (chiefly genomics, transcriptomics, and epigenomics) is beginning to provide answers to questions on the genetic bases and origin(s) of diadromy. Finally, we suggest future directions for -omics research that can help tackle questions on the evolution of diadromy.

Graphical Abstract

Introduction

Diadromy refers to the predictable migration between marine and freshwater environments that certain species undertake during specific periods in their life (McDowall, 2008a). Although diadromy in fishes is rare (present in less than 1% of all fish species), it is widely distributed both phylogenetically and biogeographically, with many diadromous species known for their evolutionary, historical, cultural, or economic value (McDowall, 1999).

The fact that diadromy involves movement between such distinct environments suggests that it requires major physiological and behavioral adaptations. In turn, such movements have important ecological and evolutionary consequences. For instance, diadromy has played a role in the genetic structure of populations (e.g., Chubb et al., 1998; Delgado et al., 2019; Taillebois et al., 2013) and in postglacial colonization (e.g., Reusch et al., 2001; Mateus et al., 2016). Despite this relevant influence on species biology, little is known about the potential selective pressures leading to its origin, the molecular mechanisms underlying the capacity for diadromy, and the effects on species evolution. Why has diadromy evolved? Which genes give diadromous individuals the ability to migrate? And why have some diadromous populations stopped migrating? These are questions not yet adequately answered. Important efforts have, however, been made to improve our understanding of diadromy including the formulation of hypotheses about its origins (e.g., Gross 1987; Tsukamoto et al., 2009) and the search for genes that differentiate diadromous and non-migratory populations (e.g., Perrier et al., 2013; Taugbøl et al., 2014).

The development of otolith and bone tissue microchemistry during the last decade has facilitated the description and classification of diadromous fishes as this method traces the presence of individuals to marine or freshwater environments (e.g., Hale and Swearer 2008; Feutry et al., 2012; Hughes et al., 2014; Warburton et al., 2018; Górski et al., 2018). Additionally, technological advances in sequencing technologies, specifically in -omics (high-throughput sequencing to study large-scale genomes, transcriptomes, epigenomes, etc.), are facilitating significant advances in our understanding of the roles of genetics, the environment, and their interaction in the evolution of life history traits (see reviews on genomics [Orsini et al., 2013], transcriptomics [Alvarez et al., 2015], and epigenomics [Metzger and Schulte, 2016]). Studies based on -omics approaches have been conducted on many diadromous species (see section “-Omics studies in diadromous fishes”), although -omics studies analyzing diadromy as a common trait in taxa across the phylogeny of fishes are lacking.

Although migratory behavior is present in all major animal taxa from invertebrates to mammals (Merlin and Liedvogel, 2019), the genetic bases and the evolutionary consequences of migratory behavior are not fully understood. The combination of new sequencing technologies and -omics approaches is key for the study of the evolution of life histories including the study of diadromy. Here, we aim to review our knowledge on diadromy across the phylogeny of fishes and how -omics techniques are helping answer questions about the ecology and evolution of diadromous species. We organized this essay in three sections: (1) What is known about diadromy in fishes? Here, we describe the classification, the distribution of diadromy from a phylogenetic and biogeographic perspective, the main hypotheses proposed to explain its origin, and the presence of non-migratory populations in diadromous species. (2) The contribution of -omics research to our understanding of diadromy. In this section, we discuss how research in genomics, transcriptomics, and epigenomics is providing information about the life history of diadromous species, the facultative nature of this migratory behavior, the molecular bases underpinning this trait, and the origin of diadromy. (3) What questions could future research focus on? In this section, we elaborate on broad questions that future -omics research can help address regarding the evolution, genetic mechanisms, and maintenance of diadromy in fishes.

Diadromy in Fishes—What Is Known?

The term “diadromy” was first introduced by Myers to describe “truly” migratory fishes (Myers, 1949), with “truly” referring to the movement between marine and freshwater environments (McDowall, 1993). McDowall (1997) expanded the definition and proposed specific characteristics that all diadromous species must fulfill. These include: migration must be mediated through physiological changes, it must occur at predicted times, and it should involve reciprocal migrations (McDowall, 1997). Although most diadromous species are known to be euryhaline, some are amphihaline, meaning that they can only adapt to a different salinity at a particular life stage (McDowall, 2009).

Diadromous species include fishes, gastropod mollusks (family Neritidae, see Abdou et al., 2015), and crustaceans (families Atyidae and Palaemonidae) (McDowall, 1997); however, in this review, we focus on fishes. More than 440 fish species have been reported to be diadromous (Table S1). These species are distributed among 58 of the 482 recognized families of fishes (Nelson, 1994); however, almost 62% of all diadromous fishes are concentrated in only seven families (Table 1).

Table 1.

Total Number of Known Diadromous Species Reported by Family and Category

Order	Family	Diadromous	Anadromous	Catadromous	Amphidromous
Acipenseriformes	Acipenseridae (R)	18	18
	Ambassidae	4		1a	3a
Anguilliformes	Anguillidae (R)	16		16
	Muraenidae	1		1
	Ophichthidae	1	1a
Atheriniformes	Atherinidae	1			1a
	Atherinopsidae	2	2a
Characiformes	Citharinidae	2	2a
Clupeiformes	Clupeidae (R)	31	26	2	3
	Engraulidae (R)	11	5	1	5a
	Pristigasteridae	7	4a		3a
Cyprinoformes	Cyprinidae (R)	6	6
Elopiformes	Elopidae	1	1a
	Megalopidae	1		1a
Gadiformes	Gadidae	1	1
	Lotidae (R)	1	1a
Galaxiiformes	Galaxiidae (R)	11	1		10
Gobiesoformes	Gobiesocidae (R)	1			1
Gobiiformes	Eleotridae (R)	37		5a	32
	Gobiidae (R)	103	2		101
	Lutjanidae	2		2a
	Moronidae (R)	2	2
Mugiliformes	Mugilidae (R)	34	1a	27	6a
Osmeriformes	Osmeridae (R)	10	10
	Plecoglossidae	1			1
	Retropinnidae	5	1a		4
	Salangidae (R)	6	6
Perciformes	Carangidae	2			2a
	Centropomidae	9		2	7a
	Cheimarrichthyidae	1			1
	Cottidae	8		2	6
	Gasterosteidae (R)	2	2
	Gerreidae	7			7a
	Haemulidae	1			1a
	Kuhliidae	10		5	5
	Lateolabracidae	1		1
	Latidae	1		1
	Percichthyidae	1		1
	Percidae (R)	1	1
	Pseudaphritidae	1		1
	Rhyacichthyidae	2			2
	Scianidae	3			3a
	Terapontidae	1		1a
	Tetrarogidae	1		1
	Toxotidae	3			3a
Petromyzontiformes	Geotriidae	1	1
	Mordaciidae	2	2
	Petromyzontidae (R)	8	8
Pleuronectiformes	Pleuronectidae	2		2
Salmoniformes	Salmonidae (R)	35	35
Siluriformes	Ariidae (R)	13	3		10a
	Bagridae	1	1a
	Claroteidae	1	1a
	Pangasiidae	1	1
	Plotosidae	1			1a
	Schilbeidae	1			1
Syngnathiformes	Syngnathidae	5			5
Tetraodontiformes	Tetraodontidae (R)	2	2
Total		444	147	73	224

The complete list of diadromous species is in Table S1.

(R) report of resident populations.

^aLittle information available (e.g., no microchemistry analysis).

Classification

There are three categories of diadromy (Gross, 1987; McDowall, 1997); fishes can be anadromous, catadromous, or amphidromous (Box 1). Categories differ in the direction of the first migration, from rivers to the sea (i.e., anadromy and freshwater amphidromy) or vice versa (i.e., catadromy, marine amphidromy); the time of migration, particularly the life cycle stage when individuals return to their natal environment (i.e., juveniles versus adults); and the purpose of the return migration, i.e., if the return to their natal environment is for growing and/or spawning purposes.

Box 1. Diadromous Species Classification.

Of the 444 species reported here as diadromous, 147 are described as anadromous, 73 as catadromous, and 224 as amphidromous (Table 1), with almost all amphidromous species reported as freshwater amphidromous. Table S1 lists species that followed McDowall's definition of diadromy and that are referenced in a scientific paper or book; however, for many species particularly amphidromous species, little information is available (Tables 1 and S1). Although Riede (2004) reported a higher number of amphidromous species than in this paper, this was due to the use of a broader definition of amphidromy (Chalant et al., 2019). We can, however, expect the total number of diadromous (mainly amphidromous) species to increase, with the increased focus on understudied species, for instance, from the tropics.

Anadromy is the most phylogenetically widespread category, present in 29 families, but only two families (Salmonidae and Clupeidae) comprise 41% of all anadromous species. Catadromy is present in 19 families. However, two families (Mugilidae and Anguillidae) represent almost 59% of all catadromous species. Amphidromy holds the highest number of diadromous species and is present in 26 families, two of which (Elotridae and Gobiidae) include 62% of all amphidromous species. Although most species from the same order belong to the same category of diadromy, different categories can be present within an order or even within a family. Four orders (Clupeiformes, Gobiiformes, Mugiliformes, and Perciformes) have species of all three categories (Table 1 and Figure 1).

Figure 1.

Actinopterygians Phylogeny Adapted from Hughes et al. (2018)

The colors in each terminal branch have been added to indicate the most likely origin of the ancestor for diadromous taxa (see Table S3 for references). Taxa exhibiting diadromy have colored background labels reflecting the category of diadromy (anadromy, catadromy, or amphidromy). Some taxa exhibit more than one form of diadromy, and their proportion is indicated by the different colors in the label backgrounds. The dotted line represents the beginning of the Jurassic Period. The R to the right states the presence of resident populations and the numbers the number of known diadromous families and species for each taxon.

Phylogeny and Biogeography

Diadromy is present from agnathans to the most recent bony fishes, indicating that it is an evolutionarily successful strategy (McDowall, 1993). Despite the high cost of migration, which includes genetic, morphological, physiological, and behavioral requirements, diadromy is likely to have evolved multiple times (Corush, 2019; McDowall, 1997). The most recent Actinopterygii phylogeny developed from genomic and transcriptomic data (Hughes et al., 2018) confirms that diadromy is widespread across the fish phylogeny (Figure 1).

Anadromy is more phylogenetically widespread than catadromy or amphidromy. Although anadromy is found from lampreys, a lineage that appeared before the Actinopterygii to the most recent order, catadromy is present from the Anguilliformes order, a lineage that appeared during the Jurassic Period when marine species reappeared (Fyhn et al., 1999), to more recent families (i.e., Lutjanidae). Amphidromous species, on the other hand, are present from the order Clupeiformes to the order Perciformes (Figure 1).

From a biogeographic point of view, diadromy is widely distributed across the globe. Gross (1987) described a latitudinal shift where anadromous species are prevalent at relatively high latitudes, whereas catadromous species have a relatively high occurrence in the tropics. This pattern led to the productivity hypothesis (see “Hypotheses on the origin” section). Amphidromy also appears to be found predominantly in the tropics (McDowall, 2010). However, although categories have higher incidence at certain latitudes, diadromous species of all three categories can be found at both high and low latitudes, providing evidence against the productivity hypothesis (Figure 2). The widespread extent of diadromous species supports its important role in species dispersal, including transoceanic dispersal (Chubb et al., 1998; McDowall, 1998).

Figure 2.

World Map Depicting the Distribution of Five to Six Diadromous Species Present at Different Latitudes

(A) Anadromous species.

(B) Catadromous species.

(C) Amphidromous species.

Hypotheses on the Origin

Although phylogenetic data support the hypothesis that diadromy has evolved multiple times (Figure 1), the origin of diadromy and its evolutionary bases are still under debate (Bloom and Lovejoy, 2014). Below, we summarize the main hypotheses for the origin of diadromy.

• Productivity or Resource Availability Hypothesis

This is the most accepted hypothesis first proposed by Gross (1987). It is based on the findings of latitudinal trends for anadromous and catadromous species by Baker, 1978 and Northcote, 1978 and states that anadromous species are more prevalent at high latitudes because productivity in the sea at those latitudes is higher than in freshwater environments. Catadromy instead, is relatively common at tropical latitudes, given that the productivity of fresh water at low latitudes is higher than in the sea.

The presence of all categories of diadromy at different latitudes questions the generality of this hypothesis (McDowall 2008b; Figure 2), for example, temperate eels migrate to freshwater environments that have lower productivity than marine environments (Edeline, 2007). Bloom and Lovejoy (2014) tested this hypothesis using the phylogeny of the order Clupeiformes. Their results did not support the productivity hypothesis as the ancestry of diadromy could not be predicted based on latitude (Bloom and Lovejoy, 2014).

Historical processes including the expansion and invasion of newly available environments following post-Pleistocene deglaciation could explain the prevalence of anadromous species in northern temperate latitudes (McDowall, 2008b). Temperature and particularly temperature fluctuations could also explain the incidence of diadromy. For example, at high latitudes temperature fluctuates more in fresh water than in the sea, a factor that has led to the hypothesis that the invasion of fresh water (i.e., catadromy) is more frequent at low latitudes (Lee and Bell, 1999).

• Random Escapement Hypothesis

This hypothesis, proposed by Tsukamoto et al. (2009), is based on behavioral models from observations of the amphidromous Ayu (Plecoglossus altivelis) and argues that diadromy originated as an escapement behavior of fishes to leave unfavorable environments, instead of as migration to a more nutrient-rich habitat (i.e., productivity hypothesis). A three-step model is used to explain the start of migration. First, an individual needs to reach a threshold age or size; second, it needs to be physiologically prepared; and third, it needs to receive an endogenous or exogenous cue to initiate migration (Tsukamoto et al., 2009).

Similarly, a “safe-site” hypothesis was proposed, where migration to fresh water or a “safe haven” is a consequence of the need to protect early life history stages from marine predators. An example is the early larval migration of osmeroids, which has been hypothesized to maximize their survival, and therefore their fitness, due to the presence of safe sites (Dodson et al., 2009).

• Ecological Opportunity Hypothesis

Proposed by Feutry et al. (2013), this hypothesis states that diadromy appears as a response to ecological opportunities (Feutry et al., 2013). Using the case of the Kuhlia family, within which catadromous species migrate to nutrient-poor environments, the authors proposed that diadromy originated owing to the opportunity to colonize insular ecosystems. These isolated habitats would be characterized, for example, by an absence of predators, making them ideal for colonization. This idea to move to an ecological advantageous site is similar to the “safe-site” hypothesis.

• Intermediate State Hypothesis

This hypothesis states that diadromous fishes have appeared as an intermediate state between fully freshwater and fully marine fishes (Gross, 1987). Gross (1987) also proposed that amphidromy is the ancestral state of both anadromy and catadromy. The hypothesis suggested that anadromous species evolved from amphidromous species, which evolved from euryhaline wanderers that evolved from freshwater species whereas catadromy evolved from amphidromous species that originated from euryhaline wanderers that evolved from marine species (Gross, 1987). The improvement in our understanding regarding the biology of amphidromy, specifically, the short time (i.e., days or weeks) amphidromous species spend in the secondary environment, led to the rejection of the idea that amphidromy was an intermediate step between fully freshwater and marines fishes (Gross, 1997).

Recently, Corush (2019) tested this hypothesis by simulating the rate of transitions in and out of diadromy and comparing it between freshwater, marine, and diadromous fishes. Transition rates out of diadromy into strictly marine or strictly freshwater life histories were higher than transitions in the opposite directions (from marine or freshwater life histories into diadromy), leading to the conclusion that diadromy may sometimes be an intermediate state between freshwater and marine fishes, but not always (Corush, 2019).

• Conditional Evolutionary Stable Strategy Model

This model proposed that diadromy is a phenotypically plastic trait in which an individual expresses a migratory phenotype depending on environmental variables and will migrate if this migration leads to higher fitness (Edeline, 2007). This model is supported by the fact that the migration in diadromous species reduces inter- and intra-specific competition (Edeline, 2007). Although the Conditional Evolutionary Stable Strategy Model (CESSM) may apply for some species, the presence of sympatric migratory and non-migratory populations that show high levels of genetic differentiation (e.g., Salisbury et al., 2018; Delgado et al., 2019) suggests that in such cases the decision to migrate does not just depend on environmental variables.

Loss of Diadromy

Major reductions in dispersal ability have evolved many times across numerous taxa (Waters et al., 2020). Non-migratory (hereafter called resident) populations exist among all three categories of diadromous fish species (Table 1). Many resident populations within a species have evolved multiple independent times and derived from a common diadromous ancestor, as the case of the anadromous Alosa pseudoharengus in Connecticut, USA (Palkovacs et al., 2008), or the amphidromous Galaxias maculatus in Chile (Delgado et al., 2019). Resident populations need not be landlocked; they can inhabit environments with access to the sea despite which they do not migrate but remain in their natal habitat. Examples of resident populations have been described for all categories of diadromy as follows.

• Anadromous Species

Atlantic salmon (S. salar), brown trout (S. trutta), brook trout (S. fontinalis), Arctic charr (S. alpinus), white-spotted charr (S. leucomaensis), Dolly Varden (S. malma), rainbow trout (O. mykiss), masu salmon (O. masau), coastal cutthroat Trout (O. clarki), sockeye salmon (O. nerka), Chinook salmon (O. tshawytscha) (e.g., Dodson et al., 2013), three-spined stickleback (G. aculeatus) (e.g., Bell and Foster 1994), and Japanese smelt (H. nipponensis) (e.g., Arai et al., 2006b).

• Catadromous Species

European eel (A. Anguilla) (e.g., Arai et al., 2006a), Japanese eel (A. japonica) (e.g., Tsukamoto and Arai 2001), American eel (A. rostrata) (e.g., Lamson et al., 2009), and tupong (P. urvillii) (e.g., Crook et al., 2010).

• Amphidromous Species

Common galaxias (G. maculatus) (e.g., Delgado et al., 2019), spotted galaxias (G. truttaceus) (e.g., Waters et al. 2001), big-scaled redfin (T. hakonensis) (e.g., Sakai et al., 2002), New Zealand eleotrid (G. cotidianus) (e.g., Michel et al., 2008), and Rhinogobius sp. (Tsunagawa et al., 2010).

The existence of resident populations that can migrate (i.e., inhabiting environments with access to the sea), but do not, suggests that migration may not always be beneficial and that ecological factors likely play an important role in the decision to migrate. Facultative diadromy demonstrates that there is a balance between the benefits and costs of migration and residency tactics (Hogan et al., 2014). Ferguson et al. (2019) introduced the threshold-trait model to explain which factors determine or affect the decision of whether or not to migrate. This model involves two components: a genetic and environmental threshold. Individuals remain resident if the energy status is high and exceeds a given threshold. Alternatively, individuals migrate if their energy status is low (i.e., nutritionally deficient).

Gross (1987) proposed a model to explain diadromy, which is simplified in Figure 3. In basic terms, the fitness of migrating adding the cost of migration must be higher than the fitness acquired by remaining in the natal environment. Following Gross hypothesis and Ferguson's threshold model, we propose to explain the loss of diadromy from an ecological and evolutionary perspective following one or the combination of four scenarios in nature (Figure 3).

Figure 3.

A simplified model to explain the “decision” to migrate (Diadromy) or remain in the natal environment (Loss of diadromy) from an Eco-Evo perspective.

• No Alternative - Figure 3A

There are two possible explanations for the loss of diadromy (Figure 3A). Either a population becomes physically landlocked unable to leave its natal environment or the presence of mutations makes individuals physiologically unable to migrate (i.e., the genetic component of the threshold-trait model). Mutations that would affect the ability to migrate in diadromous fishes may be linked to osmoregulatory genes, nutrient assimilation genes, and signaling (see “Molecular bases” section).

• Increase in the Cost of Migration - Figure 3B

This could be due to changes in the river systems; for example, a change in the landscape could lead to a longer distance to reach the estuary increasing the cost of migration (Figure 3B). Natural or anthropogenic barriers such as dams would also increase the cost of migration.

• Decrease in the Fitness of Migration - Figure 3C

This would be the product of changes in the environmental conditions of the secondary habitat, for example, reduction of food supply, an increase in the number of predators, or an increase in parasite threats (Figure 3C). These changes would lead to a reduction in fitness and even the survival of the migrating phenotype.

• Increase in the Fitness of Staying - Figure 3D

Changes in the environmental conditions in the natal habitat that would lead to a higher fitness payoff for a decision to stay (i.e., the environmental component of the threshold-trait model) (Figure 3D). This could result from a decrease in the fitness value of the secondary environment (Figure 3C) or positive changes in the natal habitat. Examples of the latter are the increment in the quality and quantity of resources and a decrease in competition.

Omics Studies in Diadromous Fishes—What Current Research Tells Us?

Since the introduction of next-generation sequencing (NGS) and the drop in sequencing cost, the number of studies using DNA, RNA, and methylation to address ecological and evolutionary dynamics questions has increased. Here, we present a list of studies on diadromous species that used -omics techniques, chiefly genomics, transcriptomics, and epigenomics (Table 2). This list is based on published articles and excludes books, theses, articles in bioRxiv, and conference abstracts.

Table 2.

Representation of Research in Diadromous Fishes that Have Used an -Omics Approach

Diadromy Category	-Omics Approach (Methods)	Research Objective	Species Studied	Include Residents	References Examples
Anadromy	Genomics (GBS, RADseq, ddRADseq, RADcap, SNP array, NextRAD, Pool-seq, whole-genome sequencing)	Development of molecular markers/SNP panels	Salmo salar; Alosa pseudoharengus; Alosa aestivalis	No	Houston et al. (2014); Yáñez et al. (2016); Baetscher et al. (2017)
		Applicability of SNP array from close-related species	Salmo trutta	No	Drywa et al. (2013)
		Assembly of a reference genome	Salmo salar; Oncorhynchus tshawytscha; Pungitius pungitius	No	Davidson et al. (2010); Christensen et al. (2018); Varadharajan et al. (2019)
			Salmo salar	Yes	Hauge et al. (2016)
		Assembly of mitochondrial genome	Takifugu obscurus; Lethenteron camtschaticum; Coilia nasus	No, Yes, No	Kim et al. (2014); Balakirev et al. (2016); Zhang et al. (2016)
		Development of linkage map/chromosome rearrangements	Salmonidae	No	Sutherland et al. (2016)
		Population diversity and structure	Leuciscus idus; Oncorhynchus mykiss; Salvelinus fontinalis; Salmo trutta	Yes	Skovrind et al. (2016); Leitwein et al. (2017); Elias et al. (2018); Lemopoulos et al., 2018b
			Salmo salar; Salvelinus alpinus; Brachymystax lenok; Tenualosa ilisha; Alosa pseudoharengus; Alosa aestivalis	No	Asaduzzaman et al. (2019); Aykanat et al. (2015); Madsen et al. (2020); Moore et al. (2017), 2014; Reid et al. (2018); Roman et al. (2018)
		Genomic divergence/local adaptation	Thaleichthys pacificus; Oncorhynchus tshawytscha	No	Candy et al. (2015); Narum et al. (2018)
			Coregonus clupeaformis; Salmo salar; Entosphenus tridentatus; Lampetra fluviatilis; Oncorhynchus nerka; Lampetra planeri; Salvelinus alpinus	Yes	Bourret et al. (2013); Hume et al. (2018); Mateus et al. (2013); Nichols et al. (2016); O'Malley et al. (2019); Parker et al. (2019); Renaut et al. (2011); Rougemont et al. (2017); Salisbury et al. (2020); Veale and Russello (2017)
			Gasterosteus aculeatus	Yes	Hohenlohe et al. (2010); Jones et al. (2012); Guo et al. (2015); Ferchaud and Hansen (2016); Currey et al., 2019; Dean et al. (2019); Marques et al. (2019); Rennison et al. (2019); Terekhanova et al. (2019)
			Oncorhynchus mykiss	Yes	Hale et al. (2013); Hecht et al. (2013); Pearse et al. (2014); Bowersox et al. (2016); Matala et al. (2017); Arostegui et al. (2019)
			Salmo salar	Yes	Bourret et al. (2011); Culling (2013); Perrier et al. (2013)
		Introgression/hybridization	Gasterosteus aculeatus; Leucopsarion petersii; Alosa pseudoharengus; Pungitius pungitius	Yes	Guo et al. (2019); Hirase et al. (2020); Reid et al. (2020); Yoshida et al. (2016)
		Genotype - migration associations	Salmo salar; Oncorhynchus tshawytscha; Oncorhynchus mykiss; Salmo trutta	No	Brieuc et al. (2015); Cauwelier et al. (2018); Johnston et al. (2014); Lemopoulos et al. (2018a); Micheletti et al. (2018a); Prince et al. (2017); Thompson et al. (2020)
		Genotype - sex - migration associations	Oncorhynchus mykiss	Yes	Kelson et al. (2019)
		Genotype - environment associations	Oncorhynchus mykiss; Gasterosteus aculeatus	Yes	Micheletti et al., 2018b; Stuart et al. (2017); Haenel et al. (2019)
			Salmo salar; Oncorhynchus mykiss	No	Jeffery et al. (2017); Willoughby et al. (2018)
		Genotype - microbiota associations	Gasterosteus aculeatus	Yes	Steury et al. (2019)
		Sex determination	Gasterosteus aculeatus	No	Bissegger et al. (2019)
	Transcriptomics (RNA-seq, cDNA arrays, microarrays)	Assembly of transcriptomic profiles	Oncorhynchus mykiss; Salmo salar; Salmo trutta, Salvelinus alpinus, Coregonus lavaretus; Oncorhynchus tshawytscha	Yes	Salem et al. (2015); Carruthers et al. (2018); Christensen et al. (2018)
		Detection of lncRNAs	Oncorhynchus mykiss	No	Al-Tobasei et al. (2016)
		Expression profiles of spermatogenesis	Coilia nasus	No	Zhou et al. (2015)
		Expression profiles before migration	Salvelinus fontinalis	Yes	Boulet et al. (2012)
		Expression profiles of infection response	Gasterosteus aculeatus; Oncorhynchus mykiss; Salmo salar & Salmo trutta	Yes	Lenz et al. (2013); Sutherland et al. (2014)
		Expression profiles hatchery versus wild	Salmo salar; Oncorhynchus mykiss	No	Bicskei et al. (2014); Fox et al. (2014)
		Expression profiles juvenile brains	Oncorhynchus mykiss	Yes	Hale et al. (2016)
		Expression profile hybrids	Salvelinus fontinalis	Yes	Mavarez et al. (2009)
		Genomic population/divergence	Coregonus clupeaformis; Alosa pseudoharengus; Oncorhynchus nerka; Salmo salar	Yes; No	Jeukens et al. (2010); Czesny et al. (2012); Lemay et al. (2013); Warren et al. (2014)
		Salinity adaptation	Alosa pseudoharengus; Gasterosteus aculeatus; Salmo salar	Yes	Lemmetyinen et al. (2013); Gibbons et al. (2017); Kusakabe et al. (2017); Rastorguev et al. (2017); Velotta et al. (2017)
		Temperature adaptation	Salmo trutta; Gasterosteus aculeatus	Yes	Meier et al. (2014); Morris et al. (2014)
		Migratory life history	Salmo trutta; Salmo Salar	Yes	Giger et al. (2008)
		Environmental stress associations	Takifugu obscurus	No	Xu et al. (2018)
		Freshwater colonization	Gasterosteus aculeatus & non-diadromous sister species	Yes	Kitano et al., 2019; Ishikawa et al. (2019)
	Proteomics	Salinity adaptation	Coregonus lavaretus	Yes	Papakostas et al. (2012)
	Epigenomics (RRBS, MSAP)	Genotype - environment associations/stressors	Gasterosteus aculeatus; Salmon trutta	No	Aniagu et al. (2008); Morán et al. (2013); Fellous and Shama (2019)
		Genotype-phenotype associations	Salmon trutta; Gasterosteus aculeatus	No; Yes	Covelo-Soto et al. (2015); Smith et al. (2015)
		Salinity adaptation	Gasterosteus aculeatus	Yes	Artemov et al. (2017)
		Migration effects	Oncorhynchus mykiss	Yes	Baerwald et al. (2016)
		Hatchery effects	Oncorhynchus kisutch; Oncorhynchus mykiss	No	Le Luyer et al. (2017); Gavery et al. (2018)
Catadromy	Genomics (genome sequencing, RADseq, Pool-seq)	Molecular markers development	Anguilla japonica	No	Sekino et al. (2016)
		Assembly of mitochondrial genome	Trachidermus fasciatus	No	Zhu et al. (2018)
		Population structure	Mugil cephalus; Anguilla japónica, Anguilla anguilla & Anguilla rostrata; Trachidermus fasciatus	No	Krück et al. (2013); Igarashi et al. (2018); Gong et al. (2019); Li et al. (2019)
		Adaptive divergence	Anguilla rostrata; Cottus asper	Yes	Pavey et al. (2015); Dennenmoser et al. (2017)
		Hybridization	Anguilla anguilla & Anguilla rostrata	No	Pujolar et al. (2014); Nikolic et al., 2020
	Transcriptomics (RNA-seq, cDNA array)	Transcriptomic profiles	Anguilla anguilla; Trachidermus fasciatus	No	Churcher et al. (2015); Ma et al. (2018)
		Genotype-phenotype associations	Anguilla rostrata	No	Côté et al. (2014)
	Proteomics	Salinity acclimation	Anguilla marmorata; Trachidermus fasciatus	No	Jia et al. (2016); Ma et al. (2018)
	Epigenomics (MSAP)	Methylation changes between life stages	Anguilla anguilla	No	Trautner et al. (2017)
Amphidromy	Genomics (genome sequencing, RADcap)	Development of molecular markers	Oncorhynchus clarki lewisi	No	Campbell et al. (2012)
		Assembly of mitochondrial genome	Sicyopterus lagocephalus	No	Chiang et al. (2015)
		Phylogeny (mitogenome)	Sicyopterus genus	No	Lord et al. (2019)
		Populations diversity and structure	Galaxias maculatus	Yes	Delgado et al. (2019)
		Salinity adaptation	Galaxias maculatus	Yes	Delgado et al. (2020)
	Transcriptomics (transcriptome sequencing)	Salinity adaptation	Plecoglossus altivelis	Yes	Lu et al. (2016)

Anadromy is by far the category with the highest number of studies (Table 2). From the >120 papers listed, 84% concern anadromous species. Within anadromy, the most intensely studied species are three-spined stickleback (Gasterosteus aculeatus), rainbow trout (O. mykiss), and Atlantic salmon (Salmo salar). Catadromy is the second category in terms of the number of studies; however, most research on catadromy has focused on species from the Anguilla family, particularly, the European eel (Anguilla anguilla). Most genetic research on amphidromous species has thus far been based on a few mitochondrial and nuclear markers (e.g., Crandall et al., 2010; Lord et al., 2015; Taillebois et al., 2013) with studies within this group using -omics approaches being scarce (Table 2). These findings are not surprising as anadromous species are more prevalent in the Northern Hemisphere, where most research takes place, and they are of economic importance.

The type of -omics approach used is a function of the research objectives (Table 2). Box 2 lists the most common methods with Reduced Representation Sequencing (RRS) being the most widely used. RRS is a cost-effective method that provides a large but limited number of markers across the genome (Wright et al., 2019). In some cases, RRS proved to be more powerful than previous genetic markers (e.g., few microsatellites, few mitochondrial genes) in differentiating and assigning individuals to populations (e.g., Moore et al., 2014; Candy et al., 2015; Yoshida et al., 2016), yet this is not necessarily always the case as shown by a study based on a relatively larger number of sequenced microsatellite markers (Layton et al., 2020). An alternative to RSS is whole-genome sequencing, and this can be done sequencing individuals or pools of individuals (Pool-seq) (Fuentes-Pardo and Ruzzante, 2017). However, given its cost, it has been used mainly in economically important species.

Box 2. Most Common Methods Used in -Omics Studies about Diadromous Species.

Genomic approaches allow the exploration of both neutral and adaptive markers, thus facilitating the examination of the genetic bases and mechanisms of adaptation (Orsini et al., 2013). Markers distributed across the genome have been used to estimate levels of genetic diversity (e.g., Bowersox et al., 2016; Gong et al., 2019), effective populations size (e.g., Li et al., 2019), bottlenecks (e.g., Ferchaud and Hansen 2016), and fine-scale populations structure or lack thereof (e.g., Mateus et al., 2013; Aykanat et al., 2015; Skovrind et al., 2016) in many diadromous species (Table 1). Outlier markers are used for the assessment of genetic divergence among populations (e.g., Box 3A). Estimating this divergence can provide information on the colonization history of diadromous species, the description of glacial lineages, and the effects of secondary contact (e.g., Bourret et al., 2013; Dean et al., 2019). Genome-wide studies have also revealed the molecular mechanisms (i.e., genes or islands of differentiation) supporting population divergence (e.g., Larson et al., 2016). Genetic markers differentiating phenotypes and populations can be used to assist in the assessment of relevant traits for fisheries/production (e.g., Yáñez et al., 2016), in the assignment of regional fisheries stocks (e.g., Baetscher et al., 2017), in the detection of introgression (e.g., Bourret et al., 2011), in the assessment of hybridization due to secondary contact (e.g., Reid et al., 2020), and in the detection of anthropogenic effects (e.g., Leitwein et al., 2017); this type of genotype-phenotype association studies has also given insights on which genes (i.e., loci under selection) underpin migratory behavior (e.g., Micheletti et al., 2018a).

Box 3. Examples of -Omics Research in Diadromous Species.

The facultative nature of diadromy in many species suggests that variation between migratory and resident behavior could happen at a transcriptional (i.e., transcriptome) or post-transcriptional (i.e., epigenome) level; thus, the molecular variation would be found in the expression of genes rather than genes themselves. Transcriptomic studies in diadromous species have been used to examine differences in gene expression between populations, revealing critical genes involved in, for example, spermatogenesis (Zhou et al., 2015), and also involved in processes related to migratory behavior including osmoregulation (Velotta et al., 2017), signaling or sensory perception (how fishes process light [Hale et al., 2016]), nutrient assimilation (Ishikawa et al., 2019), immune response (Lenz et al., 2013), and growth (Box 3B). These studies have also highlighted the importance of regulatory regions and their effects (e.g., Czesny et al., 2012). Furthermore, transcriptomics has been used to study salinity and temperature can affect the expression of genes differentially across populations (e.g., Meier et al., 2014 and Côté et al., 2014: Box 3B). This information is likely to be useful for predicting responses to changes in environmental conditions.

Phenotypic variance in migratory traits can be a product of genetics but may also be solely due to phenotypic plasticity as a response to environmental triggers, suggesting the important role of epigenetics (Merlin and Liedvogel, 2019). Epigenetics focuses on modifications of genetic material due to environmental factors (Merlin and Liedvogel, 2019). Research in this area is more recent and, therefore, less developed than genomics or transcriptomics. However, research has been conducted in some diadromous species to examine the effects of migration in reared individuals providing clues about the molecular mechanisms that distinguish migratory and resident populations. For instance, the primary location of methylation modifications that distinguish diadromous from resident populations varies in diadromous species, whereas these modifications are found predominantly within genes in three-spined sticklebacks (Smith et al., 2015) and in Rainbow Trout they are found largely in regulatory regions (Baerwald et al., 2016 -Box 3C).

Life History

Genomic analyses have been conducted to assess population structure and contribute to our understanding of migration in numerous diadromous species including the anadromous hilsa shad (Tenualosa ilisha) (Asaduzzaman et al., 2019), the catadromous Japanese eel (Anguilla japonica) (Igarashi et al., 2018), and the amphidromous common galaxias (Galaxias maculatus) (Box 3A). For example, site fidelity led to genetic differentiation in hilsa shad (Asaduzzaman et al., 2019) and Japanese eel (Igarashi et al., 2018) populations, whereas the presumed absence of site fidelity perhaps combined with a relatively large effective population size in common galaxias resulted in a panmictic or nearly panmictic migratory systems among diadromous collections (Delgado et al., 2019).

These genomic results regarding gene flow can also contribute to corroborate the classification of diadromous species. Common galaxias, for instance, has been classified as a marginal catadromous species as it was hypothesized that larvae only migrate to the estuaries (without reaching the ocean) and after a few weeks migrate back up the river streams (McDowall, 2009). The fact that populations across their Chilean distribution showed high levels of gene flow support the amphidromous nature of the common galaxias and is consistent with the hypothesis that larvae do indeed enter the ocean (Delgado et al., 2019).

The analyses of genomic markers have also provided information on species dispersal and reproductive behavior beyond that obtained through other methods such as telemetry. For example, a telemetry study in Arctic charr showed high dispersal levels, yet this high dispersal did not lead to high gene flow (Moore et al., 2017). This result is consistent with the notion that Arctic charr, an anadromous species, overwinters in non-natal freshwater environments in years when they do not reproduce (Jørgensen and Johnsen, 2014). Additionally, information on successful dispersal contributes to our understanding of the balance regarding the costs and benefits of this complex life history trait.

Facultative Behavior

Although some diadromous species like Atlantic salmon (Salmo salar) and Arctic cisco (Coregonus autumnalis) are considered obligatory diadromous, others like the goby Awaous stamineus do not need to visit the marine environment to complete their life cycle (Hogan et al., 2014). The importance of migration for an individual's development thus appears to vary depending on the species or family. The presence of sympatric diadromous and resident populations with little genetic differentiation (e.g., Rainbow trout, Kendall et al., 2014) suggests that migrating is not a requisite for the development of individuals and that diadromy can in some groups be facultative. From an ecological perspective, facultative diadromy may be beneficial as the decision to migrate would depend on environmental pressures.

In some species, resident populations are clearly genetically divergent from their diadromous counterparts (e.g., common galaxias; Delgado et al., 2019). In Arctic charr, sympatric anadromous and resident populations were until recently considered genetically indistinguishable, yet recent genomic data revealed genetically differentiated sympatric anadromous and resident populations in Labrador (Salisbury et al, 2019, 2020). Sockeye Salmon (O. nerka) on the other hand, exhibits examples of both genetically differentiated and non-differentiated sympatric resident and anadromous populations (Nichols et al., 2016). These differences in genetic differentiation between diadromous and resident populations are likely the product of local adaptation and/or genetic drift, implying that the ability to migrate can be maintained or lost as a result of selection or random processes.

Common garden studies have been conducted to test individual fitness in different environmental conditions, and the use of reaction norms has helped determine that diadromous species evolved a plastic response to different environmental variables. Examples are the studies on the response to varying salinities in three-spined stickleback (McCairns and Bernatchez, 2010) and the response to different temperatures in brown trout (Meier et al., 2014). Both studies showed that diadromous individuals exhibit higher fitness when reared under a variety of environmental conditions than do resident individuals and are thus more plastic than resident individuals. The plastic nature of diadromy has also been demonstrated in steelhead, where diadromous or resident parents can express alternative offspring (e.g., Zimmerman et al., 2009).

Reciprocal transplant experiments with common galaxias under laboratory conditions also revealed that resident populations can differ in their response to salinity changes with some populations maintaining their osmoregulatory performance necessary for migration and others not being able to survive such changes (Delgado et al., 2020). Given that these populations were similarly genetically differentiated from their diadromous counterpart, we suspect genetic drift may be one of the factors playing an important role in determining whether the ability to migrate is maintained or lost. Yet, other forms of relaxed selection could also be playing a role in the differences in persistence and loss of salinity adaptation found between these two populations (Lahti et al., 2009). The loss of osmoregulatory capacity can also be partial as revealed by gene expression studies with alewives (Alosa pseudoharengus) resident populations (Velotta et al., 2014).

Molecular Bases

The physiological adaptations required to survive in both marine and freshwater habitats are extensive (Figure 4). Freshwater species rely on adaptations related to ions uptake (specific type mitochondria-rich cells, Bartels et al., 2017) and visual pigmentations (specific types of chromophores, Toyama et al., 2008). Although many genes described as relevant to marine-freshwater adaptations have been reported in diadromous species (Table S3), the genetic variation responsible for this migratory behavior is still unknown. This raises the question of whether there is one “diadromous” gene or genes (i.e., islands of differentiation).

Figure 4.

Examples of physiological adaptations necessary to survive in marine and freshwater environments.

Research on stickleback sister species provides a clear example of a gene essential for the ability to colonize fresh water: Fads2 (Ishikawa et al., 2019). This gene, involved in the assimilation of fatty acids, highlights the importance of food resources in the evolution of diadromous behavior, as well as the significance of one gene (or copy number variant) to promote or constrain the dispersal to a new environment. However, the fact that diadromous species should have evolved all these adaptations (Figure 4) required to survive in freshwater and marine environments at one point seems unlikely. A more plausible scenario is that adaptations such as wide osmoregulatory capacity may be inherited before a species becomes diadromous and that what makes a species to become diadromous must be a gene or genes related to signaling that would start the migration. In birds, for instance, it has been hypothesized that a gene related to behavior such as circadian behavior or photoreceptors may be responsible for migratory behavior (Lugo Ramos et al., 2017).

Migratory genes may be linked together in chromosomal rearrangements and genomic islands of differentiation (Wellenreuther and Bernatchez, 2018), as it has been suggested that these associations of genes in genomic regions facilitate the selection in favor of or against complex life history traits. Chromosomal rearrangements including inversions and duplications appear to maintain co-adapted alleles facilitating adaptation in many contexts (Sutherland et al., 2016; Varadharajan et al., 2019; Wellenreuther and Bernatchez, 2018). Genomic islands of differentiation also appear to play key roles in linking co-adaptive traits in diadromous species (e.g., Veale and Russello 2017). For example, the Omy5 linkage group, a large region located in chromosome 5 of steelhead shows a strong non-random association with life history differentiation between anadromous and non-migratory populations (Pearse et al., 2014). A small genomic region (∼30 kb) was recently shown to also be closely associated with spawning migration timing in Chinook salmon (Thompson et al., 2020).

Genomic and transcriptomic analyses have revealed many putative genes that differentiated migratory and resident populations; these genes are related to osmoregulation and muscle contraction, among other processes (Table S4). Research in salmonids has shown little parallelism in genes differentiating diadromous and resident populations across species (Schneider et al., 2019). Even within the same species, replicate resident populations show that local adaptation and genetic drift can lead to different genes being fixed or lost (Delgado et al., 2020; Salisbury et al., 2020). Thus, further research on a variety of species and tissues including brain is necessary to improve our understanding of the genes that may be common among diadromous species. Such efforts could help address the overarching question of whether or not there is a “diadromous gene” or gene complex.

Origin and Ancestry

The fact that diadromy is present only in a small percentage of species, yet it is widely present across the phylogeny of fishes (Figure 1), would suggest that diadromy could have originated in either of two scenarios (Figure 5). The first hypothesis posits that diadromy appeared multiple independent times across the phylogeny of fishes; the second hypothesis instead posits that diadromy is ancestral and has been lost on multiple occasions (Figure 5).

Figure 5.

Hypotheses Regarding the Origin of Diadromy

(A) Diadromy appeared in multiple events as a response to different scenarios (one or more of previously proposed hypotheses). (B) Diadromy is an ancestral trait that has been lost multiple times in fish phylogeny.

The second hypothesis, common to other migratory taxa including birds, assumes that species have an ancestral migratory predisposition; therefore, migration, in theory, could appear in many lineages (Zink, 2011). In fishes, the rarity of diadromy (<1%) and the lack of parallelism in genes associated with a migratory phenotype suggest that diadromy may not be predisposed in all lineages. Although the loss of diadromy does appear to be common in diadromous species, as seen by the existence of resident populations in many diadromous species across the phylogeny (Figure 1), many more instances of loss of diadromy are required under this second hypothesis than are instances of the appearance of diadromy under the first hypothesis. The hypothesis that diadromy appeared multiple independent times across the phylogeny of fishes thus seems more parsimonious.

Under the multiple appearance hypothesis, it can be also assumed that in some lineages/orders where this life history was advantageous, it rapidly expanded, resulting in a relatively large number of related species evolving from a common diadromous ancestor. Examples could include the Gobiiformes and Salmoniformes, where single families within each of these orders exhibit a high number of diadromous species (Figure 1). The first hypothesis also assumes the independence of each diadromy appearance events, suggesting that any of the proposed hypotheses on the origin of diadromy (see “Hypothesis on the origin” section), all of which have evidence for and against, may be true for a given taxon. This is also consistent with the suggestion that no single hypothesis explains the origin of diadromy for all diadromous species adding support to the first hypothesis. Recently, Alò and collaborators tested different hypotheses (e.g., productivity and genetic predisposition) to find one comprehensive migration model. Their results, however, showed that different migratory strategies including the different categories of diadromy cannot be explained by a single model but by different environmental, phylogenetic, and productivity variables (Alò et al., 2020).

There is also an ongoing debate about the salinity at which the ancestors of diadromous species lived. The ancestors have been hypothesized to be of freshwater, marine, or diadromous origin (McDowall, 1997). The ancestor species can be assumed to be of freshwater or marine environment as a function of the hypothesis of the origin of diadromy. For instance, under the intermediate state hypothesis, anadromy is derived from a freshwater ancestor, but under the safe-site hypothesis, anadromy derives from a marine ancestor. Depending on the taxa, both assumptions can be valid.

The phylogenetic tree suggests that both anadromous and amphidromous species appear to have both marine and freshwater ancestors (Figure 1). Catadromous species, on the other hand, appear to be present mainly in clades where the most recent ancestor was of marine origin (Figure 1). For most clades, however, there is no consensus on the habitat of the ancestral species. Indeed, we were unable to find information on the habitat of the ancestor for many taxa (Table S2). Although no single hypothesis explains diadromy for every diadromous species, the presence of a marine ancestor for catadromous species suggests that diadromous species may migrate for ecological reasons (i.e., increase fitness) but return to their natal habitat because they lack adaptations that would allow them to reproduce in the secondary environment. Thus, catadromous and marine amphidromous likely have a marine ancestor that passed on the ability to reproduce in marine environments, and similarly, anadromous and freshwater amphidromous most likely have a freshwater ancestor. However, there are notable exceptions like the order Clupeiformes, which held mostly anadromous species yet its most recent ancestor is marine.

The presence of marine ancestry for some anadromous or freshwater amphidromous could be explained by novel mutations that allow these species to reproduce in fresh water or by standing genetic variation of a slightly older freshwater ancestor. The appearance of novel mutations seems unlikely as these mutations would have had to appear multiple times in different lineages. The latter hypothesis (standing genetic variation) seems more plausible, as it is known that most actinopterygians are derived from a common freshwater ancestor (Vega and Wiens, 2012). The presence of pre-existing or cryptic genetic variation could explain why some anadromous and freshwater amphidromous species from multiple and independent lineages have a recent marine ancestor. This idea that most actinopterygians have standing genetic variation to reproduce in fresh water may also explain why anadromy and freshwater amphidromy are more prevalent than catadromy (<20% of diadromous species are catadromous, Table S2). However, the higher prevalence of anadromous versus catadromous species could also be explained by the higher speciation rate of anadromous species as the consequence of more opportunities to isolate and differentiate in freshwater environments than in marine environments.

Future Directions of Diadromy in an -Omics Era—What We Can Learn?

Research using -omics approaches regarding diadromy is still in its infancy. Further research on different species from all three categories of diadromy would help answer questions on the origin and molecular bases of diadromy. Questions such as whether diadromy evolved multiple independent times or is an ancestral trait could be addressed by having more genomic data of diadromous species within and among different orders, as thus far, most research has focused on Salmoniformes. Genomic data of species from different categories of diadromy from the same and different orders or even families (e.g., Clupeidae) would also contribute to resolving the question of whether the different categories of diadromy have the same genomic bases. Also, studies focusing on orders that show rapid diversification (i.e., Gobiiformes) could shed light on questions such as why freshwater amphidromy is more prevalent than other categories.

Transcriptomic and epigenomic research with diadromous and resident populations exhibiting little genetic differentiation, and which therefore have not undergone local adaptation, would help assess which differentially expressed genes are responsible for this migratory behavior and if indeed there is a “diadromous” gene(s) or gene complex. Research thus far has focused on osmoregulatory organs (i.e., gills); however, wide resistance to osmoregulatory changes is present too in non-diadromous species as well, suggesting that this adaptation is an ancestral trait. Thus, examining other organs including brains where genes related to signaling and photoreceptor are expressed may provide an answer to the question of what gene is responsible for starting the migration of larvae/juveniles.

Research on ecological factors that may influence the decision to migrate, including the presence of predators or the influence of parasites, may also lead to improvements in our understanding of the facultative nature of diadromy, which is present in many diadromous species. Finally, understanding the evolutionary consequences of diadromy and its loss as a source of genetic diversity and also considering the increased risk of extinction that this life history trait carries can be important to predict the evolution of these species.

Concluding Remarks

Diadromy is a life history trait that has an important role in the distribution and even diversification of species in aquatic systems, directly influencing the ability of populations to disperse and colonize new habitats and niches. This migratory behavior involved a series of adaptations that grants individuals the ability to survive in such different environments. And, although many questions regarding the origin, genetic bases, and evolutionary consequences of diadromy still need an answer, our understanding of diadromy has improved during the last decade thanks to advances in sequencing technologies. For example, research has revealed key genes and processes responsible for these adaptations. Also, studies suggest that diadromy may have appeared convergently in different taxa, proving to be an important source of genetic variation in fishes. Unfortunately, many diadromous species have experienced population declines throughout the last century (Duarte, 2018; McDowall, 2009; Righton et al., 2012), and many diadromous species are in peril owing to their migratory nature. Conservation strategies for a diadromous species require the preservation not only of their natal and secondary environment but also the connection between them (McDowall, 1999) as well as the consideration of future climate-driven changes in both freshwater and marine environments (Walter et al., 2012). Thus, given the conservation status of many diadromous species and the current climate change scenario, understanding the variability and potential to adapt to different environments of diadromous species is crucial for their preservation.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.