Predation is one of the strongest selective pressures that drive prey phenotype (Tollrian and Harvell, 1999).

Prey organisms have thus developed appropriate anti-predatory behaviors, specific morphological traits, and highly accurate sensory systems to detect and assess the potential predation risk, localize and identify predators, and therefore optimize the energy costs associated with predator avoidance strategies (Atkins et al., 2016, Lima and Dill, 1990).

The effectiveness of anti-predator defenses is considerably influenced by the environment in which an organism lives (Apfelbach et al., 2005). In aquatic ecosystems, prey species evolved highly sensitive receptors for detecting chemical cues which inform them about the presence, proximity, physiological state, and diet of potential predators (Dicke & Sabelis, 1988), and therefore adjust their phenotype accordingly (Atema, 1995).

Our model organism, Lymnaea stagnalis (Linnaeus 1758) – as its name suggests - lives in stagnant aquatic environments where visual information can easily be hindered by dense vegetation and/or turbidity (Rundle & Brönmark, 2001). Lymnaea uses chemicals released by predators (i.e., kairomones) to assess the type and degree of predation risk (Orr et al., 2007) and to then respond with the appropriate defensive vigilance behaviors (Rundle & Brönmark, 2001).

Previous studies demonstrated that snails which have been maintained in predator-free laboratory conditions since the early 1950s (i.e., over 250 generations) are capable of detecting the presence of a crayfish predator effluent (i.e., CE) and responding appropriately (Batabyal et al., 2021, Batabyal and Lukowiak, 2021, Orr et al., 2007, 2009), even if they have never experienced a natural predator such as crayfish in their laboratory lifetime. Thus, these inbred W strains exhibit innate recognition of crayfish as a threat and are thus called predator-experienced snails. In particular, exposure to CE increases the time snails take to re-emerge from their shell and elicits the shadow withdrawal response (Orr & Lukowiak, 2008).

Importantly, these defensive vigilance behaviors are not observed when the serotonergic system is disrupted by the serotonin-receptor antagonist mianserin (Il-Han et al., 2010). These data further confirm the high level of conservation of the role of the serotonergic system in modulating stress-induced arousal and vigilance behaviors associated with predation risk in Lymnaea as in vertebrates (Diamond et al., 1999, Edsinger and Dölen, 2018, Rillich and Stevenson, 2018). However, other studies have also shown that there are strain-specific differences in the ability to form long-term memories between populations of L. stagnalis (Orr et al., 2008).

Along with common anti-predatory behaviors L. stagnalis strains also show a higher-order predator-induced learning called configural learning (Swinton et al., 2019b). During configural learning animals simultaneously experience the aversive predator cue (CE) along with a food cue (carrot slurry) and following this exposure (configural learning training) the food now invokes fear and is avoided by the animal (Rivi, Batabyal, et al., 2021; Rivi, Benatti, et al., 2022; Teskey et al., 2012). The inbred W strain shows configural learning memory; while a strain of wild snails called Margo snails does not show configural learning memory. Crayfish are not found in the Margo Lake pond and as they do not innately respond to CE, we have labeled them predator-naïve (Batabyal and Lukowiak, 2022).

Thus, these strain-specific differences may provide a great experimental platform in which to mechanistically distinguish scent detection from aversive behavior and memory formation as well as to address how fear learning occurs and prospective future directions to understand the mechanism of innate fear recognition from a learned fear recognition.

Adaptation through natural selection has been shown to account for the divergence between aquatic populations experiencing different predation pressure (Cousyn et al., 2001, Dalesman et al., 2007, Meyer et al., 2006). Thus, in this study, we predicted that populations that co-existed with certain predators would have adapted to show higher innate responses to those predators than prey populations that did not experience such predation risk. In particular, we hypothesized that the innate predator recognition observed in lab-inbred snails even though they were not exposed to the predator for over 250 generations, can be due to the historical presence of a predator in the parental Dutch generation which may have led to the modification of genetic elements which have been fixed in the lab population (Batabyal & Lukowiak, 2021).

Here, we investigated whether there are also strain-specific differences in the transcription of several genes across these two strains (i.e., W and Margo Lake) of Lymnaea when they are exposed to the predator cue (i.e., CE).

We performed behavioral and molecular experiments to ask: 1) Are there strain-specific molecular differences in responses to CE? 2) At the behavioral level do only predator-experienced snails form configural learning even though neither strain has actually experienced CE before training?

To answer these questions, we first investigated the transcriptional effects induced by CE exposure on the expression levels of serotonin-related genes. As serotonin plays a key role in modulating behavioral arousal, including predator detection and vigilance behaviors (Rillich & Stevenson, 2018), we thought it was important to study the differences in the expression levels of these targets in both strains. Therefore, we focused our attention on the enzyme tryptophan hydroxylase (LymTPH) which catalyzes the synthesis of serotonin (Koert et al., 2001), the serotonin-specific receptors - LymHTR1 and LymHTR2 (Sugamori et al., 1993) and transporter LymSERT (Sadamoto et al., 2008). Further, we investigated the effects of CE exposure on the expression levels of the transcription factor cAMP response element binding protein 1 (i.e., LymCREB1). In both mammals and invertebrates, the activation of the serotonergic pathway is associated with the release of genes of the cAMP cascade, including CREB1 (Benatti et al., 2017, Kandel, 2012), which, in turn, plays a key role in learning and memory formation (Kandel 2012). As we recently demonstrated that long-term memory for configural learning is enhanced via CREB upregulation (Batabyal et al., 2021), we hypothesized that predator perception would upregulate the serotoninergic system, which in turn, would lead to the upregulation of CREB1, resulting in predator-induced learning.

The last target we investigated was the heat shock protein 70 (LymHSP70), whose upregulation has proven to be induced by many physicochemical stressors in Lymnaea, including temperature, hypoxia, and food deprivation (Foster et al., 2015; Rivi, Batabyal, et al., 2021, 2021). We hypothesized that predator detection, by imposing stress, would upregulate HSP70, which, in turn, may be involved in predator-induced learning.

Thus, we used the configural learning procedure to further investigate the differences in the ability of predator-experienced W-strain and predator-naive Margo strains to form configural learning (which is a reproduction of previous studies from our lab) and how those differences correlate with molecular differences elicited by CE exposure.