Molecular Stress Responses In Estuarine Fish

Abstract

Estuaries cover only about 6% of the global coastal areas but are among the most biologically productive ecosystems. They have a massive influence on carbon cycle, serve as nurseries, migration routes, and sources for fish populations. However, they are also among the ecosystems most affected by direct human interventions and anthropogenically caused climate change. Despite their high ecological importance and the ecosystem services they provide, only one-tenth of a percent of all climate change research is conducted through observational studies in estuaries. Global models used for predicting the effects of climate change often fail in these understudied systems. Some of the strongest influencing factors include oxygen deficiency, particularly from anthropogenic nutrient input, rising temperatures, extreme events such as heatwaves, altered salinity patterns due to changes in precipitation and sea level rise, as well as increases in sediment load due to construction or altered flow patterns. Fish populations in estuaries have a significant impact on the ecosystem services provided, from the transport and storage of carbon to being a direct food source, yet they are increasingly exposed to environmental stressors. The multiple stressors that can simultaneously affect an organism in wild habitats can have unexpected interactive effects on the organism itself and its interaction with associated microbiota, the holobiont. This aspect has been little explored in laboratory experiments and even less in field studies, but it is important for understanding and predicting changes in ecosystems and providing a foundation for protection and restoration. Multicellular organisms have a highly conserved cellular stress response that directly reacts to stressors, protecting macromolecules, degrading damaged macromolecules, halting the cell cycle, or inducing controlled cell death to protect the entire organism in case of severe damage. Vertebrates like fish also have additional stress responses controlled by stress perception in the brain and hormone release, which allow responses on various levels. The organismic and cellular response to one stressor generally limits the capacity to respond to additional stressors. The universal mechanism underpinning physiological stress responses is the adjustments of abundance in messenger RNA that can be determined and described with high precision via transcriptome studies. As the costs of holistic omics methods continue to decrease, large-scale applications for identifying stressors, stress interactions, and previously unconsidered biotic interactions become feasible. In the first part of this work (Chapter 2), the impact of abiotic gradients along the estuarine stressor mosaic on tissue-specific stress responses in immunologically important gill tissues and metabolically important liver tissues of a top-predatory freshwater species, Sander lucioperca L., was investigated using differential expression and abundance analyses alongside network-based approaches. As an approach to assess the holobiont's reaction, bacterial composition patterns on the gills were determined in parallel. The results showed turbidity-induced starvation in fish and increased cellular stress and immune responses in reaction to hypoxia and eutrophication coincided with the increase in abundances of various opportunistic bacteria. The work also served for method validation and demonstrated that the different omics methods (mRNA and bacterial 16S gene sequencing) complement each other well, and the differential analyses support the network results. To track movement patterns, habitat use, and spatial connectivity, we conducted stable isotope analyses of δ13C and δ15N on the most dominant fish species in the study system (Chapter 3). For potamodromous ruffe, it was shown that they remain in the estuary throughout their life cycle, as there were hardly any marine or fluvial isotope signatures observed. This suggests a brackish water population of this species that is spatially distinct from upstream freshwater populations. Additionally, a high degree of site fidelity was observed, suggesting limited short-term movement along the salinity gradient. The stable isotope patterns of ruffe and smelt were then combined in the next chapter (Chapter 4) with spatially and seasonally extended gill mucus bacterial composition data. Gills are a multifunctional organ in constant and direct exchange with the surrounding water, making them a potential entry point for pathogens. The composition of the microbial community can therefore be an important indicator of fish health. In the site-faithful ruffe, the composition of the microbiota was heavily influenced by spatial conditions. In anadromous smelt particularly during spawning migration a gradual adaptation of bacterial composition was detected. The stable isotope data and bacterial patterns largely overlapped, providing mutual insights into migration patterns. In both fish species, a stable core microbiome comprising few but highly abundant bacterial taxa was identified, which diminished in both fish species under prolonged hypoxia and eutrophication conditions. This effect was accompanied by an increase in opportunistic taxa (Acinetobacter, Shewanella, Aeromonas). To capture the fish responses, we further integrated 340 tissue-specific transcriptome datasets from livers and gills of smelt and ruffe with the bacterial data, isotopes, fish abundances, physiological and abiotic data to unravel the interactive effects of dominant stressors on a large scale (Chapter 5). In both fish species, rising water temperatures in spring were accompanied by the occurrence of opportunistic Flavobacteriaceae and a strong increase in adaptive and innate immune responses across tissues. Persistent hypoxia in summer in the upper river, on the other hand, was associated with rising abundances of opportunistic Acinetobacter taxa alongside cellular stress responses, particularly regarding cell cycle control and cell death. In anadromous smelt especially, generalized additive models showed a parallel decline in immune responses, indicating energy reallocation-driven immunosuppression, potentially making the fish more susceptible to infections. Population genetic analyses of both species showed high connectivity among populations but functional diversity and signs of selection in genes important for adaptation to high temperatures and hypoxia, related to the HIF transcription factor and pathway genes, as well as the immune system. Finally, based on a 40-year time series, we examined changes in the abundances and composition of life cycle guilds of fish in the Elbe estuary (Chapter 6). While there was an improvement in habitat quality with increasing water quality throughout the 1990s, since 2010, catches have been declining again, and total fish density has decreased by 92% in the last decade. Drivers identified included recurring hypoxia situations, high concentrations of suspended particulate matter (SPM), and the reduction of nursery habitats. In summary, this Phd-project provides spatiotemporal insights into the molecular stress responses of dominant fish species in conjunction with host-associated bacteria as an estimate of the holobiont response in an estuarine system for the first time. Various methods for large-scale integration of omics techniques were evaluated and established. The use of different methods particularly enabled the validation of individual results. A minimally invasive, non-lethal monitoring of fish-associated bacteria, supported by point investigations of molecular stress responses within the host organism, could represent a cost-efficient extension of current long-term monitoring methods. The current results can already be used to guide ecosystem recovery efforts. The data show a need for action, as the decline in fish abundance is expected to be accompanied by a genetic bottleneck, reducing genetic diversity and stripping populations of adaptive capacity, especially considering increasing stressors with ongoing climate change (e.g., hypoxia, temperature). However, the longer the population collapse lasts, the stronger these effects can manifest themselves due to genetic drift.