Threespine stickleback (Gasterosteus aculeatus) infected with the tapeworm Schistocephalus solidus display impairments in their anti-predator responses. They also have increased expression of the gene encoding the IMPase 1 enzyme in their brains, which is part of a key step in myo-inositol synthesis. IMPase 1 and myo-inositol levels are the targets of lithium treatment in patients with bipolar disorder. Although promising candidates, we do not know if IMPase 1 and myo-inositol are directly implicated in the changes in risky behaviors measured in Schistocephalus-infected fish. Understanding the molecular mechanisms directly or indirectly involved in these behavioral alterations is crucial to understand the evolution of host-parasite interactions. Here, we increased myo-inositol levels of uninfected fish and inhibited IMPase 1 activity in infected fish to test the prediction that it would decrease and increase their anti-predator behaviour, respectively. We found that uninfected fish with increased myo-inositol levels (by injecting exogenous myo-inositol or by inducing endogenous production using an osmotic challenge) did not decrease their anti-predator responses. However, infected fish treated with lithium chloride had some of their anti-predator behaviors restored, but not all. They spent less time swimming close to the surface, swam a shorter distance, had a higher latency to feed, and spent more time frozen after a predator attack. Our results suggest that the target of lithium treatment is implicated in the risky behaviors of infected fish and supports the idea that the parasite-associated alteration in behavior has a multifactorial nature.

SAMPLING: Using a beach seine, we caught threespine sticklebacks from the wild population of Lac-Témiscouata (47°48'37.1"N 68°51'56.6"W, Québec, Canada) in June 2018 (n = 250). We brought all fish to the “Laboratoire Aquatique de Recherche en Sciences Environnementales et Médicales” (LARSA) at Université Laval. During transportation, we kept fish in coolers filled with the lake water and kept it aerated with air stones. In the animal facility, we held all fish in three 1000 L water tanks under a Light: Dark photoperiod of 13 h: 11 h and a temperature of 12 °C, similar to their natural conditions. We fed fish daily, twice a day, with a mix of artemia (Hikari Bio-Pur) and flakes (Nutrafin-Basix).

EXPERIMENTAL GROUPS: For the uninfected fish experiment, we created four groups. Uninfected control for the stress of the injection (UNI CTRL, n = 20); uninfected control for the injection of phosphate-buffered saline (PBS), which was used as the vehicle (UNI PBS, n = 20); uninfected fish exposed to exogenous myo-inositol (UNI INO, n = 20); and uninfected exposed to an osmotic challenge (UNI OSC, n = 20). For the infected fish experiment, we had two groups. The infected control group (INF CTRL, n = 20) and infected treated with lithium group (INF LIT, n = 20).

TREATMENTS: We exposed uninfected sticklebacks to exogenous myo-inositol (CAS #87-89-8, Sigma-Aldrich, ≥ 99% purity) at a dose of 90 mM dissolved in buffered saline (PBS, pH 7.4; Life Technologies, Carlsbad, CA, USA) by intraperitoneal injection (with a total volume of 100 µL) during three days, once per day. In another treatment group, we performed an intraperitoneal injection of a saline solution at 25 ppt in uninfected sticklebacks, by dissolving Instant Ocean salt mix (Spectrum Brands, Blacksburg, VA, USA) in PBS in a final volume of 100 µL. We exposed fish infected with Schistocephalus solidus for three days to lithium chloride (CAS #7447-41-8, Sigma-Aldrich) using static exposures, by dissolving lithium in the aquarium water at a concentration of 12.5 mM (in a total volume of 2.7 L).

BEHAVIOUR QUANTIFICATION: During the control week, fish were submitted to the handling of the experiment for three days but without receiving any treatment. We changed the water of the aquarium completely 48h after the beginning of the experiment and from this moment they did not receive any food until the behavioral tests. On the fourth day, each fish was transferred to the behavioral tanks for testing, which lasted one hour in total. After the behavioral tests, each fish returned to its home tank where they could rest for three days. After this period, the second week (treatment) started and we treated all fish, except those from the control group, and then we submitted fish to the behavior tests, as in the first week. We measured time spent in the upper part of the aquarium, time spent in the center of aquarium, total distance travelled, latency to feed, latency to freeze and time spent frozen after a predator attack. We recorded all behavioral experiments using a digital camera (Super Circuits PC212XS) placed in the front or at the top of the behavioral tanks. Videos were analyzed using Ethovision software (Ethovision XT 11.5, Noldus Information Technology, Wageningen, The Netherlands, Noldus et al., 2001). Water depth preference, distance traveled and total time spent in the center were measured with an automatic tracking module of Ethovision and latency to feed, latency to freeze, and time spent frozen were measured using a manual behavior setting module of the software. All tests and analyses were done by V.A.A., while blind for the fish experimental group.

FINAL SAMPLING: At the end of the behavioral tests, we anesthetized each fish with an overdose of MS-222 (75 mg/L, pH 7.5). We quickly measured the standard length of the fish, their mass and rapidly dissected their whole brains. Brains were weighed and placed individually in 0.5 mL sample tubes and snap-frozen in liquid nitrogen. Brains were then kept at -80 °C until analysis. We verified every fish infection status (parasitized or not by S. solidus) upon dissection. We calculated the parasite index as the proportion of infected fish mass that is contributed by parasite tissue.