Skip to main content


Amino acids - the path of Lake Champlain Atlantic salmon restoration


Salmon complete some of the most amazing and longest distant (up to 6,000 km) migrations of any animals – traveling from their birth rivers to oceans or lakes as juveniles and back again as full-grown adults (Figure 1). Salmon not only complete these long distance migrations, but also do so with great accuracy, finding the same river they were born in nearly 95% of the time. Izaac Walton (1653) first recorded this behavior on the Thames River in London, England in his famous book The Compleat Angler. For over 200 years, scientist have studied this amazing behavior to try and unlock the mysteries of how, why, when, and where these animals make these immense migrations. This summer I traveled to Northern Japan on a NSF Office of International Science and Engineering (OISE) East Asia and Pacific Summer Institute (EAPSI) fellowship to understand the spatial and temporal variability of amino acids, the hypothesized natural odorant responsible for these amazing migrations.

Field and lab experiments by Hasler and Wisby (1951) generated the first hypothesis describing how salmon find their home-streams using odor cues. They proposed (1) each stream has a unique and stable chemical composition and, thus, a distinct odor, because of local differences in soil and vegetation of the drainage basin; (2) juvenile salmon imprint to the distinct odor of their home-stream before migrating to the sea or lake; and (3) adult salmon use this information as a cue for homing to their home-stream. Over the last 60 years, this hypothesis has been validated for many species in the salmon family, including both Pacific and Atlantic salmon (Hasler and Scholz 1983) , yet many of the mechanisms remain unexplained.

One of the greatest questions initially was, and remains today, what are the naturally occurring odorants responsible for the ability of salmon to find their home-streams and how do they function as such? Not all chemical constituents of stream waters are created equal when it comes to natural odorants – Sutterlin and Sutterlin (1971) found that amino acids “as a class of compounds were highly stimulatory” where as “little or no response” was observed with “a variety of simple sugars, n-aliphatic acids, alcohols, or amino-substituted alcohols in electrophysiological studies. Furthermore, over the last 20 years, Dr. Hiroshi Ueda of Hokkaido University in Sapporo, Japan, has shown positive results that amino acids function as olfactory odorants in 4 species of Pacific salmon (Oncorhynchus spp.) (Ueda 2011) .

This summer (2012) I went to Japan to work with Dr. Ueda to begin the path to understand if Atlantic salmon (Salmo salar) also use amino acids as their primary home-stream homing odorant. I had two main objectives while in Japan (1) observe and learn Dr. Ueda’s behavioral two-choice Y-maze techniques and design my own and (2) analyze stream, lake, and hatchery waters from around the Lake Champlain region for amino acids. Both objectives were completed this summer. The results of this summers work will improve the water management, rearing, and outplanting strategies of Lake Champlain’s fish hatcheries.

To determine if adult salmon prefer one water type to another, a simple two-choice “Y-maze” is used – the maze has a base or “holding tank” and two branches, like a Y. Waters of different origin (e.g. home vs. non-home stream) or amino acid compositions can then be run down the different branches of the maze and the behavior of the salmon monitored and recorded using Passive Integrated Transponder (PIT) systems – a small tag is inserted into the fish and the movement recorded by antennas in the maze. Working with my US Fish and Wildlife Service (USFWS) and Vermont Fish and Wildlife Department (VTFW) collaborators we designed and built a Y-maze according to the schematic below (Figure 2). We will use the maze to study the homing behavior of two groups of Atlantic salmon, (1) those that inappropriately return to one of Lake Champlain’s hatchery’s outlet ditch and (2) fish that I artificially imprinted to a 5 amino acid mixture at Berkshire National Fish Hatchery during the critical period of learning, the parr-smolt transformation, and am currently rearing to adulthood to test their preference for the same 5 amino acid mixture.

The analysis of water samples collected in the fall of 2011 and spring of 2012, the two key periods of salmon migration, homing and imprinting, from around the Lake Champlain watershed and shipped to Japan to be analyzed was multifaceted; (1) establish if the amino acids of the Winooski and Boquet Rivers, the two rivers currently prioritized by the USFWS for restoration, are unique and stable – Hasler and Wisby’s (1951) 1st postulate of their famous hypothesis, (2) determine the spatiotemporal variability within and between water types (groundwater, hatchery, river, and lake), and (3) verify the effectiveness of the artificial amino acid imprinting experiment.

While I experienced the trials and tribulations of operating a complex machine, a High Performance Liquid Chromatograph (HPLC), I successfully analyzed 106 of 166 samples. Of those 60 remaining samples, 13 are of high priority and potentially add interesting information to the analyses, while the remaining 47 are additional samples from the experiment at Berkshire Hatchery. Initial results (Figure 3) from this summers work on the Berkshire samples suggests both that the stripping of amino acids from the Hatchery water with activated carbon filers to create a neutral background of amino acids was effective (control) and that the addition of the 5 desired amino acids (glutamic acid, arginine, valine, threonine, and leucine) to the treatment tanks was effective.

So what’s next? This fall, with the help of the USFWS and VTFW, we will conduct pilot studies in the Y-maze – we will run 100 fish from Hatchery Brook, the outlet ditch to Ed Weed Fish Culture Station, in the Y-maze under two treatments (1) control – the same unfiltered Hatchery Brook water in both branches of the maze, and (2) filtered – one branch will contain activated carbon filtered Hatchery Brook water and the other will contain unfiltered water. Next fall (2013), we will use lessons learned from this years maze trials to analyze the behavior of Atlantic salmon artificially imprinted at Berkshire Hatchery – the true test of if Atlantic salmon also use amino acids as their home-stream homing odorant. Throughout both of these experiments, I will continue to explore the spatial and temporal variability of amino acids in hatchery, stream, and lake waters, conduct analyses of the physiology during imprinting and homing, and investigate the genetic expression of amino acid olfactory receptors.

1) Izaak, W. (1653) The compleat angler. New York, NY: Caxton Press.

2) Hasler, A.D., Wisby, W.J. (1951) Discrimination of stream odors by fishes and its relation to parent stream behavior. The American Naturalist, 85, 823, pp. 223-238.

3) Hasler, A.D., Scholz (1983) Olfactory imprinting and homing in salmon. Investigations into the mechanisms of the imprinting process. Zoophysiology, 14, 154 pp. New York, NY: Springer-Verlag.

4) Sutterlin, AM, Sutterlin, N (1971) Electrical responses of the olfactory epithelium of Atlantic Salmon (Salmo salar). J. of Fisheries Research Board of Canada, 28, 4, 565-572.

5) Ueda, H. (2011) Physiological mechanisms of homing migration in Pacific salmon from behavioral to molecular biological approaches. General and Comparative Endocrinology, 170, pp. 222-232.