Pollution of the environment with plastic waste has long been an ignored issue, but is now considered a major global threat to aquatic systems and their inhabitants. Microplastics, comprising plastic fragments, beads, and fibers smaller than 5 mm, are detected in rivers, lakes and oceans all over the world. Due to their small size, they can be ingested by a wide range of aquatic organisms, including teleost fish. To date, little is known about how severely native freshwater fish species are affected by microplastics. There is also limited knowledge about how the differing gastrointestinal morphologies and foraging strategies of fish affect the uptake mechanisms and the retention time of microplastics. The aim of the thesis was to tackle some of these knowledge gaps in order to better understand the interaction of fish with microplastics in freshwater systems.
First, a new method for the detection of microplastics in fish was developed, which allowed efficient and rapid (<1 h) digestion of the entire fish gastrointestinal tract, and included an optional density separation step to reduce mineral components. (Manuscript I). This novel method made it possible to reliably and rapidly examine a large number of samples, allowing a large-scale analysis of microplastic burden in fish. This method was then used to investigate the microplastic burden of native fish species across the German state of Baden-Württemberg (Manuscript II). The overall burden of microplastics was found to be low, with an average prevalence of ~19 % and an intensity of between one and four particles per individual. Several relevant biotic and abiotic factors, such as sampling site and trophic state, were shown to have only a minor influence on microplastic burden. The results also revealed a major limitation with currently available microplastic detection methods: particles <40 μm could not be reliably detected in the gastrointestinal tract of the examined fish. However, by using the dataset acquired in this thesis it was possible to calculate the theoretical total microplastic burden in local fish with a size distribution analysis. It was found that as particle size decreases, particle concentration increases – with a power law growth fit likely indicating that over 95 % of all microplastic particles in fish are currently being excluded from collected data. This means that only a fraction of the potential size spectrum of microplastics can currently be considered in research data.
It is still not fully understood how microplastics are taken up by fish. To gain a more holistic understanding of microplastic uptake pathways, pre-existing and recently developed theories were explored through a number of practical and theoretical approaches (Manuscript III). Four fish species (rainbow trout(Oncorhynchus mykiss), grayling (Thymallus thymallus), common carp (Cyprinus carpio), crucian carp (Carassius carassius)), representing different foraging styles and domestic status, were exposed to a range of particles (varying by type and colour) with or without the provision of food; the abundance of microplastics was subsequently determined in their gastrointestinal tract. These experiments revealed that visually-orientated fish ingest microplastics actively and/or accidentally with their food much more frequently than fish that are chemosensory-orientated. In addition to the microplastic concentration in the water and fish size, the colour of the plastic particles played an important role in uptake: particles were taken up significantly more often if they resembled the colour of the food. By contrast, chemosensory foraging fish were able to discriminate larger plastic particles, and only ingested microplastics on occasion, by chance. At smaller particle sizes, uptake pathways other than feeding become more relevant; statistical models showed that in large marine fish species, notable amounts of microplastics were ingested simply through drinking. Finally, these experiments showed for the first time that domestication plays an important role in the uptake of microplastics. Relative to wild fish, farmed fish discriminated less between differently coloured plastic particles, and were more likely to actively ingest microplastics when no food source was available.
The next step was to investigate the duration that microplastic particles remained in the gastrointestinal tract of fish (Manuscript IV). A special diet was developed that contained differently sized microplastic particles. The number of retained particles in the gastrointestinal tract was determined up to 72 h after administration in two fish species (rainbow trout (Oncorhynchus mykiss), common carp (Cyprinus carpio)) that have distinct gastrointestinal morphologies. The laboratory experiments showed size-dependent differences in the T50 value (time at which 50 % of the particles are excreted) of plastic particles in fish with a true stomach; particles with a size of ~1000 μm were excreted approximately three times faster than particles with a size of ~40 μm. In fish without a stomach, the differences were substantially smaller, suggesting purely passive excretion with the chyme. It was thus concluded that the morphology of the gastrointestinal tract plays a vital role in the retention of microplastics, and that large plastic particles must be actively excreted in fish with a true stomach.
Finally, controlled laboratory experiments were conducted to investigate whether realistic microplastic concentrations have detrimental short- and long-term effects on fish (Manuscript V). In addition to an analysis of established performance and health parameters, the entire rainbow trout (Oncorhynchus mykiss) liver proteome was examined and the results confirmed with the help of gene expression analysis. Two groups of fish were exposed to a realistic current environmental concentration of microplastics, and a slightly elevated microplastic concentration that reflects expected microplastic exposure levels in the near future. These two groups were then compared with a control group (no exposure to microplastics) after 120 days of continuous exposure. Microplastic exposure was shown to have a significant dose-dependent effect on growth and other performance parameters (i.e. specific growth rate, feed conversion rate). There were no significant differences in blood glucose, hematocrit levels and oxidative stress levels between the groups. The proteomic analysis identified over 6000 proteins, but no clear difference in their regulation or correlation with gene expression was found between treatments. However, a number of single proteins and their respective transcripts were identified as potential biomarkers for future studies. The results therefore conclusively showed that even low microplastic concentrations have a notable impact on fish with long-term exposure. Importantly, they provide the basis for future investigations of microplastic effects on health, and demonstrates the potential of novel state-of-the-art methods that are now emerging in the field.