Behavioural stress feedback loops in benthic invertebrates caused by pH drop-induced metabolites

Author's response

Lauric Feugere; Katharina C. Wollenberg Valero shared

• It is very intriguing that the time-to-success and avoidance mechanisms was different depending on the metabolite donors. As the heterospecific metabolite donor caused a difference in the measured parameters, how different was the metabolite composition from S. aurata compared to the conspecific metabolite composition?

Yes, as it is often the case, our answer in this study was – “it depends”. The reaction of recipients depended on the donor of the metabolites, but also on who was the recipient. For example, the ragworms reacted uniformly to Sparus aurata stress metabolites as also to their own pH-induced stress metabolites, but the same did not apply to crabs. Since this study was limited to behavioral (avoidance and time-to-success) and physiological (respiration rates) measurements, the honest answer is – we don’t know yet which metabolites were present here, and to which extent they were different among species. However, what we do know is that in both cases (conspecific and heterospecific donors), the metabolites released through pH stress are either different, or more than 10 times more concentrated, than the ones regularly released in the absence of pH stress (with our control being “control metabolites”, see Supplementary Information, Figure S6).  But stay tuned, we are currently working on a metabolomic characterization of similar stress metabolites, which were induced by thermal stress in zebrafish embryos (read our matching preprint here: https://doi.org/10.1101/2021.04.07.438623), which will hopefully shed light on the nature of the chemicals that can constitute stress metabolites.

Nevertheless, there are several possible explanations that could account for the observed variations in response to conspecific versus Sparus aurata stress metabolites, ranging from experimental sources (e.g., relative concentrations of cues, number of donors, conditioning protocol) to physiological reasons inherent to the donors’ differential response to acute pH drop between S. aurata and benthic invertebrates. It may be that the conspecific and S. aurata donors produce different types of metabolites, in different concentrations and relative ratios, which would differentially affect the behavior of the recipients. In addition, variation in the behavioral response to S. aurata cues may be explained by predator-prey relationships.

• What metabolites were the dominant ones and how do these metabolites create an awareness (in which potential pathways) to the studied organisms?

Pending chemical characterization of these metabolites, our study nevertheless allows us to shed light on the existence and potential role of the stress metabolites: they are specifically released upon pH stress, but not, or only to a lower extent, in an optimal pH environment. It is possible that stress metabolites may in fact be a cocktail of different compounds including chemicals resulting from the metabolic stress response, the purpose of which is to restore homeostasis. Further work will be necessary to unveil whether these stress metabolites are true signaling molecules (i.e., purposefully released to alert others of an upcoming threat), or if they are simply byproducts of the stress response (in this case, response to pH alteration and CO2) which are passively released into the environment and upon which bypassers eavesdrop and react accordingly.

While the novelty in our preprint is that such metabolites can be released by abiotic stressors such as acidified pH, there is a large body of literature on cues that are released by biotic stressors. From this we can see that disturbance cues released upon predation stress induce alertness in recipients in order to induce a response to deal with the imminent threat. Stress metabolites could be detected at sensory epithelia, either actively via receptors or passively via membrane diffusion depending on molecular properties, and then activate behavioral responses. In this preprint, this is a behavioral stress response (comparing the behaviors we observed to other papers where crabs or polychaetes showed stress behaviors, and to responses of these animals to the pH shock itself). However, we can add here evidence from our zebrafish preprint (linked above) that shows that this can in part also be a physiological stress response involving ROS (reactive oxygen species) and the immune system. We can therefore assume that such “positive feedback loops” provide the recipient with an anticipatory stress response to the risk of acute abiotic stress. The molecular pathways along which such information could travel are as little understood as the metabolites themselves, but there are studies that link acidified pH response to altered olfaction mediated by GABAA receptors (e.g., Chivers et al., 2014), and social alarm cues increasing whole-body cortisol which indicates activation of the HPI axis (e.g., Barkhymer et al., 2018). To shed some more light on these pathways, we are currently looking at transcriptomic data of the harbor ragworms exposed to pH drop-induced stress metabolites from this same experiment. This will hopefully help to elucidate the molecular pathways activated by stress metabolites in the recipients.

References:

Barkhymer, A.J., Garrett, S.G. and Wisenden, B.D., 2019. Olfactorily‐mediated cortisol response to chemical alarm cues in zebrafish Danio rerio. Journal of Fish Biology, 95(1), pp.287-292.

Chivers, D.P., McCormick, M.I., Nilsson, G.E., Munday, P.L., Watson, S.A., Meekan, M.G., Mitchell, M.D., Corkill, K.C. and Ferrari, M.C., 2014. Impaired learning of predators and lower prey survival under elevated CO2: a consequence of neurotransmitter interference. Global Change Biology, 20(2), pp.515-522.

• The effects of pH drop on small hermit crab and ragworm are more pronounce compared to green shore crab. Is it possible that this is due to the mobility of the chosen sample species; ragworm and hermit crab having less mobility than shore crab?

This is indeed a possible mechanism by which pH drop could induce different responses in the three tested species. The two crab species used alternative strategies to respond to the acute pH stress. As shown in Figure S1 (Supplementary Information), green shore crabs tended to attempt to escape the testing arena more often than small hermit crabs. Conversely, small hermit crabs used a dual strategy as they, in addition to escaping, were more prone to freezing whereby they retreated into their shell – which shore crabs weren’t able to do. Intriguingly, we noticed that this dual response in small hermit crabs often was characterized as an initial freezing period followed by an escape attempt shortly afterwards. It is also worth noting that the behavioral assays were different – for crabs we looked at how successful they were at finding a food cue, whereas for the ragworms we looked at sediment burrowing speed – so there were different trade-offs to make for these animals, either between actively seeking food vs freezing/escaping, or to alter a behavior the worms naturally do once removed from their sediment. The reason for this was that for this experiment, we couldn’t really convince the ragworms to come out of their burrow to feed in exposure to cues, something that we normally regularly achieve in our labs. Interestingly, our harbor ragworms were delayed in their burrowing response by the stress metabolites by displaying a behavior pattern which is normally associated with extreme physiological stress, such as ‘playing dead’, freezing, everting the proboscis, and sideways-undulating movement.