by Tasha Thompson, Michael Miller, Daniel Prince and Sean O’Rourke

Adult spring Chinook salmon, Salmon River, California (Photo credit: Peter Bohler)

Spring Chinook and summer steelhead (premature migrators) have been extirpated or are in decline across most of their range while fall Chinook and winter steelhead populations (mature migrators) remain relatively healthy. Because premature migrating fish are closely related to mature migrating fish within the same river, conservation policy typically lumps them into the same conservation unit. Thus, spring Chinook and summer steelhead, in most situations, don’t receive special conservation protections despite sharp declines.

In our recently published study, we show that incredibly important genetic adaptations can rely on rare evolutionary events in single genes, and that current conservation policies can fail to protect this type of adaptive variation. Most current policies protect genetic adaptations between distantly related population units, but they don’t necessarily protect adaptations within closely related population units, and the consequences of that can be substantial: in the case of Chinook and steelhead, the consequences could be the permanent loss of an economically, culturally, and ecologically important life history. To account for this type of adaptive variation, current conservation policies will likely need to be improved.

Chinook salmon ‘holding’ and circling before heading upstream to spawn (Photo credit: Jan Jaap Dekker)

Pacific salmon are born in freshwater streams, migrate to the ocean as juveniles, spend a few years there, then return to the stream they were born in to spawn. Some species of Pacific salmon, specifically steelhead (a legendary sport fish) and Chinook (a workhorse of the West Coast fishing industry), exhibit two strikingly distinct life history types within their species when it comes to spawning migration time. Mature migrators (a.k.a. fall Chinook and winter steelhead), return from the ocean in a sexually mature state. These fish migrate directly to their spawning grounds and spawn almost immediately. In contrast, premature migrators (a.k.a. spring Chinook and summer steelhead) return to freshwater months before sexual maturity. These fish migrate high into the watershed and hold in cold, deep pools over the summer while their gonads develop, then spawn at about the same time as mature migrators.

Premature migrators are special for a number of reasons: they play an important ecological role by carrying marine nutrients higher into watersheds than mature migrators, they are very significant to the cultures and traditions of the indigenous peoples of the Pacific Northwest and Northern California, they provide a larger window of fishing opportunity, and they have much higher fat content than mature migrators so taste much better.

Policies that lump together premature and mature populations have been justified by two  assumptions that our study shows are incorrect. The first assumption was that spring Chinook and summer steelhead had evolved from their mature migrating counterparts independently in each river; the second was that spawning migration time was controlled by many genes that each has a small effect. These led to the belief that premature migration had evolved many times and therefore could easily re-evolve in the future if lost.

Genomic data analysis (Photo credit: Gus Tolley)

To identify the genetic basis for migration timing, we used an inexpensive and efficient technology called RAD (restriction-site associated DNA) sequencing to test hundreds of thousands of points throughout the steelhead genome, and then compared the results of summer steelhead to the results of winter steelhead to see where their genomes differed. We did the same thing with spring and fall Chinook.

Strikingly, we found that the same genetic region differed between summer and winter steelhead as between spring and fall Chinook, and that variation in this single region (a gene called GREB1L) completely explains the difference between migration types in both species.

Next, we investigated if premature migration versions of GREB1L had arisen once or multiple times. We found that all summer steelhead versions had arisen from a single event and all spring Chinook versions had arisen from a single event. The evolutionary events were different between the species, so both evolutionary events occurred sometime in the past 15 million years since the two species diverged. Finding that the same gene is crucial for premature migration in two separate species and that all the premature migration versions of this gene we examined arose from a single evolutionary event within each species strongly suggests that the genetic mechanisms for evolving premature migration are limited and happen very rarely across evolutionary time.

For the Pacific Northwest and Northern California, our study indicates that we should be much more concerned about the decline of spring Chinook and summer steelhead than we previously were. The premature life history depends on a particular version of the GREB1L gene. However, the number of fish carrying that version has declined dramatically. If premature migrating fish are lost, that version will be lost and may take many thousands to millions of years to re-evolve.

Juvenile steelhead sampling on the Salmon River, California (Photo credit: Mikal Jakubal)

This study is also significant for many specific rivers and local communities, such as the Klamath Basin in Northern California, that have seen dramatic declines of spring Chinook and summer steelhead. In many of these locations, grass roots efforts are among the only things keeping these fish from totally disappearing. Premature migrators have been completely lost from many California rivers where they used to be abundant, and most populations that remain are severely depressed. For example, the Salmon River in Siskiyou County only had approximately 100 spring Chinook return this year, where it historically had tens of thousands. The same pattern is common throughout Oregon and Washington too.

Identifying the premature migration gene has also allowed us to develop genetic markers to easily test the migration type (premature or mature) of ambiguous samples such as juveniles or carcasses for which the migration type was not previously able to be determined. This will enable a better local scale understanding of the ecology of premature vs. mature migration, factors behind the decline of premature migrators, and steps that can be taken to bolster premature populations.

Now that genomic technologies allow us to determine the genetic basis and evolutionary history of important adaptations, we can use this information to improve conservation policies. More specifically, we can better protect adaptations that exist within closely related population units, are disproportionately impacted by human activities, and are unlikely to re-evolve in human timeframes.

Tasha Thompson and Daniel Prince are Ph.D. Candidates in the Integrative Genetics and Genomics Graduate Group at University of California, Davis. Michael Miller is an Assistant Professor of Population and Quantitative Genetics in the Department of Animal Science at University of California, Davis. Sean O’Rourke is an Assistant Project Scientist in the Department of Animal Science at University of California, Davis.

Further reading

D. J. Prince, S. M. O’Rourke, T. Q. Thompson, O. A. Ali, H. S. Lyman, I. K. Saglam,T. J. Hotaling, A. P. Spidle, M. R. Miller, The evolutionary basis of premature migration in Pacific salmon highlights the utility of genomics for informing conservation. Sci. Adv. 3, e1603198 (2017).