High diet overlap is assumed to cause competition between the three dominant pelagic planktivorous mesopredators in the Baltic Sea, sprat, herring, and stickleback11,24,25. Despite this assumption, stickleback populations have increased dramatically over the past decades, which raises the question of whether and how resources are partitioned26. While previous studies of fish diet overlap have mainly relied on microscopic identification of gut content, we implemented a DNA metabarcoding approach targeting two different gene regions, the 18S rRNA gene (18S) and the mitochondrial cytochrome c oxidase I gene (COI) to reveal the taxonomic diversity of prey, and a qPCR step to quantify rotifers that are at times abundant in the Baltic Sea. Our study highlights consistency between methods, with DNA metabarcoding resolving the plankton-fish link at the highest taxonomic resolution. Our results suggest a unique niche of stickleback that may enable high population growth in the open water, despite high competition between mesopredators, although this finding needs to be confirmed at larger scale. More than half of the DNA found in herring and sprat stomach contents was assigned to Pseudocalanus, supporting previous observations of high diet overlap between the two clupeids11,12. On the other hand, the diet of stickleback differed substantially from the two clupeids, with rotifers appearing as main prey DNA in spring. The high rotifer biomass in the environment and lack of competition from other predators indicate that this novel niche utilization may support the drastic increase of pelagic stickleback in the Baltic Sea.
We find that copepods dominated the gut content of the two clupeids sprat and herring. Pseudocalanus and Temora occupied most of the sequence reads of the clupeid metabarcoding, two species that are often reported as preferred prey in previous studies11,12. Despite high contributions of these two copepods, Pseudocalanus was more than four times as abundant as Temora in clupeid gut contents. A strong preference for this copepod with marine origin can further confirm the increased competition between the clupeids, as Pseudocalanus has decreased due to decreased salinity12 and shares a similar vertical distribution as clupeid during daytime27. Our study using metabarcoding further reveals a large relative quantity (11%) of the ctenophore Mertensia in the gut samples of both clupeids. Similar, Clarke et al.28 reported an important contribution of gelatinous zooplankton to upper trophic levels in the Southern Ocean. Despite high abundances of ctenophores in the Baltic Sea and their assumed importance in marine food webs19, they are not reported as food for planktivorous fish. A possible explanation is the difficulty observing them microscopically, as their digestion rate is faster than crustaceans29, and no hard parts remain in the digestive system. Further, COI detected the presence of cladocerans, which was confirmed by the microscopic survey, but underrepresented with 18S that strongly amplify copepods20. Interestingly, more than twice annelid COI reads, including the benthic macroinvertebrates Bylgides and Marenzellaria, were associated to stickleback (15%) and herring (8%) than to sprat (4%), highlighting their ability to migrate vertically. These interactions suggest that together stickleback and herring contribute to benthic-pelagic coupling when oxygen is not restricting vertical migration in the southern Baltic Sea30.
Sprat and herring share a similar feeding niche, which may explain previously observed declines in body mass and stomach fullness, and supports the theory of competition between the two species31. In contrast, stickleback revealed little diet overlap with the other mesopredators. The low relative abundances of Pseudocalanus (1–8%) in metabarcoding analyses indicates that the density-dependent competition may not limit the population growth of stickleback. The copepods that were shared in the diet of stickleback, sprat, and herring were Temora, Acartia, and Centropages have increased over the last decades, as opposed to Pseudocalanus32. Our results show that stickleback are able to feed on a broader spectrum of prey and highlight that stickleback utilizes the rotifer Synchaeta baltica as prey, which is an important component of the plankton community composition in the Baltic Sea18,20. Due to the difference of prey size, we can expect an overrepresentation of copepod to rotifer sequences compared with microscopic count data. High predation rate on S. baltica is supported by both the qPCR assay as well as microscopic counts, although only the eggshells were visible but not the soft-bodied rotifer. Despite the considerably lower carbon content per S. baltica (ca. 6 µg C ind−1) compared to copepods (ca. 20 µg C ind−1)33, the high number of rotifers likely act as a major food source for stickleback. These results propose that stickleback, due to their opportunistic feeding behaviour34 and smaller size35, have a distinct feeding niche from sprat and herring in the open water, as they feed on a smaller size class of zooplankton compared to the clupeids. Thus, we cannot assume the same process of competition between clupeids and stickleback.
Rotifers can at times be very abundant in the Baltic Sea, reaching densities up to 25,000 ind m−3, but their natural predators are poorly studied. An increasing trend in biomass of the two main rotifer genera (Synchaeta and Keratella) was observed since the 1990s36. In a recent study, we showed that rotifers might occupy a unique feeding niche, as direct grazers of dinoflagellate spring bloom, as well as in the recycling of organic matter in summer20. The low level of predation on rotifers by clupeid adults (< 1% of the reads) observed here indicate that this trophic niche may not be fully utilized. Further, qPCR did not identify predation on rotifers by other zooplankton species, including several species of copepods and cladocerans, which is supported by previous observations showing that limited DNA reads were associated to dominant zooplankton species in spring20,22. Thus, stickleback appears to have little or no competition for rotifers as a food resource in the southern Baltic Sea in spring. This abundant and increasing resource36 may therefore sustain the expanding Baltic stickleback population during its pelagic phase.
Similar feeding patterns but different taxonomic resolutions were found between approaches for adult fish. DNA metabarcoding has the highest taxonomy resolution, targeting the full prey spectrum, which allow for an exploration of the whole prey community for fish and other organisms37,38. As previously reported by Clarke et al.39, both barcodes showed consistent outcomes despite their different taxonomy resolutions. 18S identified the highest prey diversity by having an increased resolution for copepods genera revealing the consumption of Pseudocalanus, Acartia, Centropages, and Temora, while COI was limited to the identification of the genera Pseudocalanus and Acartia. Moreover, 18S identified gelatinous zooplankton prey, while COI was more efficient in identifying cladocerans, annelids and rotifers at lower taxonomy levels. qPCR, that is more sensitive than DNA metabarcoding for identifying a single species40, allowed us to confirm and quantify the interaction between the rotifer S. baltica and stickleback. However, this approach does not allow for a diet assessment, but rather for a post-hoc quantification, due to the specific primers used. Microscopy is the traditional method to assess fish diet, but is limited to prey with hard remaining parts, such as exoskeleton23,41. Our visual observations are consistent with the fish prey species identified at the same sampling location in 2020 (J. Hentati-Sundberg, pers. comm.) and both show the above-mentioned limitation. Our study confirms that combining molecular tools, including DNA metabarcoding and qPCR, with traditional microscopy observations is a robust approach to explore the full prey spectrum of planktivorous fish, and reveals a diverse prey spectrum for mesopredators in the Baltic Sea.
In this study, we highlight the importance of soft-bodied zooplankton organisms as potential prey for planktivorous fish, suggesting that these neglected prey taxa might be important players in the ecosystem. While several prey species are difficult to observe using traditional gut content analysis, molecular techniques such as metabarcoding and qPCR can be used as a suitable complement. In this study, we show that rotifers, which are often not identified with microscopic analysis of gut samples, may contribute to niche partitioning between sprat, herring, and stickleback. Although this result must be confirmed by more intensive sampling, the opportunistic feeding in the pelagic Baltic Sea may reduce competition with clupeids and support the increasing stickleback population.
Source: Ecology - nature.com
