Parasites and diet of Serrasalmus maculatus in a hydroelectric reservoir in Brazil

Abstract Serrasalmus maculatus is a species of piranha which, despite being abundant in a reservoir environment, has few studies related to its parasitological and diet aspects. Thus, we aimed to document the parasitic fauna and diet of the S. maculatus in a hydroelectric reservoir in Brazil. In addition, we perform two literature reviews for the Neotropical region, recording the parasitic fauna already associated with S. maculatus and the occurrence of parasite genera identified in this study parasitizing Characiformes from other aquatic systems. Thirty-one hosts were collected with gillnets, from August 2014 to September 2016. Serrasalmus maculatus had a piscivorous feeding habit and a low richness parasitic component community, including two taxa of monogeneans, Anacanthorus lepyrophallus and Mymarothecium sp.; no endohelminths were observed. Data from the literature review, together with the findings of the study, showed that S. maculatus in the Neotropical region harbors 25 helminth taxa, with the monogenean being the most prevalent parasitic group and Brazil is the country with the most reports of the parasitic genera. These findings provide information on the relationships between diet, social behavior, and parasitic fauna of S. maculatus and on the patterns of distribution and infection of the observed parasite rates.


Introduction
Parasites can influence local communities by affecting host physiology, morphology, reproduction, and behaviour, thereby affecting population, community, and ecosystem structures, and host behaviours (e.g., feeding habits and predator-prey relationships) in turn, can affect the structures of parasite communities (Timi & Poulin, 2020). However, even though the ecological relevance of parasitism is widely recognised, many studies have neglected the effects of these organisms on their hosts (Timi & Poulin, 2020). For example, even though Brazil harbours a megadiverse freshwater ichthyofauna (~3500 species) (Froese & Pauly, 2020a), the parasitology of only 13% of the region's species has been evaluated, of which the majority are economically important species. Nevertheles little is known about the parasitology of fish species with low commercial importance (Eiras et al., 2010(Eiras et al., , 2011. The piranha, or pirambeba, Serrasalmus maculatus (Kner, 1858) is a medium-sized freshwater fish belonging to Characiformes, that is widely distributed in South America, throughout both the Amazon and Paraguay-Paraná River basins (Froese & Pauly, 2020b). The species is piscivorous, preferentially consuming fish musculature, fins, and scales. Eventually, invertebrates are the speciesmost common prey (Agostinho & Marques, 2001;Agostinho et al., 2003;Villares et al., 2008). It is also generally gregarious and, although has low economic importance, is one of the most abundant species in hydroelectric reservoirs, because readily adapts to artificial lentic environments (Sazima & Machado, 1990;Hoffmann et al., 2005;Behr & Signor, 2008). Despite the abundance of S. maculatus in hydroelectric reservoirs, there are few studies on its parasitological aspects.
Most studies of the parasitology of S. maculatus have focused on populations in the Upper Paraná River floodplain region (Pavanelli et al., 1997(Pavanelli et al., , 2004Takemoto et al., 2009;Casali & Takemoto, 2016;Moreira et al., 2019), and few studies have examined this species ecology or parasitology in artificial environments. In addition, considering the diet is an important factor in host-parasite interactions and hosts with more diverse diets tend to be more susceptible to endoparasite infections (Lima et al., 2016), we aimed (i) document the parasitic fauna and (ii) characterize the diet of S. maculatus in a hydroelectric reservoir in Brazil. We targeted also (iii) to verify the parasite fauna already associated with S. maculatus in the Neotropical region; and (iv) the occurrence of parasite genera -identified in the present study -in characiform fishes from other aquatic systems (natural or artificial) in the Neotropical region.

Study area
The Ilha Solteira hydroelectric reservoir is an accumulation basin that was formed in 1978 and is situated along the Upper Paraná River, between the states of São Paulo, Minas Gerais, and Mato Grosso do Sul, Brazil (Figure 1). With a mean depth of 17.6 m, maximum volume of21.06 × 10 9 m 3 , hydrographic basin area of 1195 km 2 , and residence time of 46.7 days, it is one of the largest artificial reservoirs in the neotropics (Garcia et al., 2014). For the present study, host sampling was conducted in the Can-Can arm in municipality of Santa Clara D'Oeste, São Paulo state, Brazil (50° 55ʹ 59.65″ W and 20° 02ʹ 30.54″ S).

Host sampling
Serrasalmus maculatus specimens were collected using gill nets (3,4,5,6,7,8,10,12 and 14 cm between nonadjacent nodes) between August 2014 to September 2016 (authorization SISBio nº 42229-1). The collected specimens were euthanized (Authorization CEUA/FEIS nº 001/2014 and Certified SisGen A9038DB) and identified as described by Ota et al. (2018). The total weight (g, with viscera) and standard length (cm, from snout to last vertebra) of each specimen were recorded, and the fish were subsequently individually stored in plastic bags, frozen and sent to the laboratory for additional analyses. All measurements are expressed as the mean ± standard deviation followed by the range.

Parasitological procedures
The organs (skin, fins, nasal cavities, gills, eyes, heart, liver, gonads, intestines, swim bladder, spleen, gallbladder, and mesentery) were analysed for parasitological procedures, using a stereomicroscope, and parasites preserved in 70% ethanol or mounted on semipermanent slides using Gray and Wess medium. The parasite specimens were then subject to morphological analysis, using a computerised image analysis system with differential interference contrast (DIC) -LAS V3 (Leica Application Suite V3; Leica Microsystems, Wetzlar, Germany) and identified according to Kritsky et al. (1992) and Kritsky et al. (1996). Parasite prevalence (P, in percentage), mean intensity of infestation (MII), and mean abundance (MA) were then calculated according to Bush et al. (1997). Mean intensity of infestation and mean abundance are expressed as the mean ± standard error followed by the range.

Literature review
Two literature reviews were conducted to verify the parasite fauna already associated with S. maculatus in the Neotropical region; and to verify the occurrence of parasite genera -identified in the present study -in characiform fishes from other aquatic systems (natural or artificial) in the Neotropical region. In the first review, we collected data on the helminth fauna previously reported for S. maculatus and its synonymy (= Serrasalmus spilopleura Kner, 1860) from the Neotropical region, from the first report in 1997 to 2021. In the second review, we collected data regarding the occurrence of monogenean species belonging to Anacanthorus and Mymarothecium genera in S. maculatus, as well as in other characiforms from the Neotropical region, from the first report of each genus (1965 to 2021 for Anacanthorus, and 1996 to 2021 for Mymarothecium).

Diet analysis
The stomachs of the host specimens were removed, fixed in 4% formaldehyde, and preserved in 70% alcohol, and stomach contents were analysed using an optical stereomicroscope. Recovered food items were quantified using the volumetric method (displacement of each measured food item from stomach contents using a gridded Petri dish) (Hyslop, 1980). Glass slides were used to compress food items to 1.0 mm in height, and the number of quadrants occupied by each food item was multiplied by 0.001 to calculate the volume in ml (Hellawell & Abel, 1971). All food items were identified to lowest possible taxonomic (Bicudo & Bicudo, 1970;Mugnai et al., 2010;Ota et al., 2018).

Results
The weight and standard length of the 31 S. maculatus specimens ranged from 32.24 to 650.40 g (139.95 ± 24.42 g) and from 9.5 to 24.0 cm (14.44 ± 0.53), respectively. Data from the literature review jointly with data from the specimens evaluated here demonstrated that S. maculatus in the Neotropical region harbour 25 helminth taxa ( Table 1). Of these 25 taxa, 10 are monogeneans, nine nematodes, three digeneans, two acanthocephalans, and one cestode ( Figure 2). Monogeneans most commonly infect host gills, followed by the nasal cavities and body surface (mucus), whereas the endohelminth groups with higher richness, nematodes and acanthocephalans, most commonly infect host intestines (Table 1 and Figure 3).    (Jégu & dos Santos, 2001;Rossin et al., 2019). However, for the Northern Brazil basin, the identification of S. spilopleura is still valid, so records of S. spilopleura in the northern basins were not included in the review. Furthermore, it is noteworthy that the occurrence of S. maculatus is recorded for the Amazon and Paraguay-Paraná River basins (Froese & Pauly, 2020b), while S. spilopleura is restricted to the basins of the Northern region of Brazil (Jégu & dos Santos, 2001).  (Jégu & dos Santos, 2001;Rossin et al., 2019). However, for the Northern Brazil basin, the identification of S. spilopleura is still valid, so records of S. spilopleura in the northern basins were not included in the review. Furthermore, it is noteworthy that the occurrence of S. maculatus is recorded for the Amazon and Paraguay-Paraná River basins (Froese & Pauly, 2020b), while S. spilopleura is restricted to the basins of the Northern region of Brazil (Jégu & dos Santos, 2001).

Discussion
This is the first study to report the parasitic fauna of S. maculatus from the northwest region of the Upper Paraná River basin, São Paulo, Brazil. In addition, represents the first report of monogeneans belonging to Mymarothecium in this host species and first report of Anacanthorus lepyrophallus in the Ilha Solteira Reservoir. For monogeneans that parasitise fish gills, the phylogenetic relationships and evolutionary history between host orders are important factors for host-parasite interaction and distribution (Braga et al., 2014).
Previous studies have demonstrated that most monogeneans prefer to parasitise specific host lineages Moreira et al., 2019) (e.g., Mymarothecium taxa parasitise members of the Serrasalmidae) (Braga et al., 2015). However, in some cases, members of other monogenean families have been reported to colonize phylogenetically distant hosts. In both cases, host-parasite relationships result from a combination of factors, including cospeciation, host-switching, and ecological fitting (Janzen, 1985;Brooks et al., 2006;Braga et al., 2014Braga et al., , 2015. Considering the monophyly of the Characiformes and the diversification of the group only in the continental neotropics, the phylogenetic contiguity between the order's families may indicate the sharing of a range of intrinsic resources (Braga et al., 2015). Anacanthorus spp. are widely distributed in hosts of the five families of the order Characiformes (Figure 4). The sharing of resources (e.g., phylogenetic conservatism and phenotypic flexibility) may have favoured its occurrence within individuals of the same order and family (see Braga et al., 2014Braga et al., , 2015 and cited references).
The predominance of monogeneans in S. maculatus in Neotropical region could be associated with both the parasites' monoxenous biology and host species' gregarious habit (Sazima & Machado, 1990;Strona, 2015). Indeed, the proximity of fish in shoals can facilitate monogenean transmission, which occurs through simple contact between hosts (Thatcher, 2006). Furthermore, gregarious behaviour also allows free-native larval forms (oncomiracidia) to locate hosts more easily (Thatcher, 2006), which would justify the results observed in the present study.
The low parasite richness and absence of endoparasites observed in the present study may be related to host behaviour and/or foraging. Several studies have reported that heteroxenous parasites are transmitted via food web interactions and that intermediate hosts are nearly always dietary components of the parasites' definitive hosts (Luque & Poulin, 2008;Lima et al., 2016). Therefore, host diet is considered an important factor in host-parasite interactions, and hosts with more diverse diets tend to be more susceptible to endoparasite infections and, thus, usually harbour greater parasite richness (Lima et al., 2016).
The dietary components of S. maculatus identified in the present study were like the findings of previous studies in the Upper Paraná floodplain, including the Ibicuí River, Rio Grande do Sul state, and a lower stretch of the Sorocaba River basin, São Paulo state, Brazil (Agostinho & Marques, 2001;Agostinho et al., 2003;Behr & Signor, 2008;Villares et al., 2008). Serrasalmus maculatus is piscivorous, preferentially ingesting fish fragments (instead of ingesting the host's entire body), and its feeding behaviour includes the mutilation of prey scales, fins, and muscle tissue, which we infer can hinder the ingestion of endoparasites (Sazima & Pombal-Jr, 1988;Sazima & Machado, 1990;Casali & Takemoto, 2016). In the present study, the dietary components of S. maculatus were fish fragments, terrestrial plants, and decapods (Macrobrachium sp.). However, even though Macrobrachium sp. is one of the most common of S. maculatus' prey items, this genus of shrimp is native from Amazon basin (Collart & Moreira, 1993), and was introduced in Paraná River basin (Bialetzki et al., 1997). When a species is introduced to a new area, it may lose part of its natural parasite fauna (i.e., Enemy Release Hypothesis - Keane & Crawley, 2002;Tourchin et al., 2002;Mitchell & Power, 2003;Torchin et al., 2003) and, thereby, break the natural network of complex interactions between intermediate and definitive hosts, which alters the infection dynamics and enables the loss of parasite taxa (Madi & Ueta, 2009).
Several authors have reported rich endoparasite fauna for S. maculatus in the Upper Paraná River floodplain, whereas endoparasites were completely absent in the present study, and the richness of ectoparasites was low (Pavanelli et al., 1997;Pavanelli et al., 2004;Takemoto et al., 2009;Casali & Takemoto, 2016 see Table 1). It is possible that the dynamics of parasitic infections are negatively affected by abiotic and biotic homogenisation in artificial habitats (Agostinho et al., 2007), such as hydroelectric reservoirs, especially for endoparasites with heteroxenous life cycles.
Floodplains are highly dynamic and complex systems because they include a wide variety of aquatic habitats (e.g., rivers, lakes, and canals) (Junk, 1980;Power et al., 1995), when compared to artificial reservoirs, since the hydrodynamics and biotic communities of such last environments are altered during the damming process. The conversion of lotic to lentic environments involves a series of negative biotic and abiotic impacts, including changes in flow and channel granulometry, increases in fish mortality, increased predation rates, simplification of trophic chains, interruption of fish migration, eutrophication, deterioration of water quality, reduction of benthic community stability, colonisation by macrophytes, invasion by non-native species, and simplification of habitats (Agostinho et al., 1992;. Furthermore, these changes can ultimately reduce the abundance and richness of local biota, disrupt the dynamics of host-parasite relationships, and, consequently, alter the structure of parasitic communities (Morley, 2007), and these seem to be the drivers involved here regarding the low parasite richness observed for S. maculatus.
In summary, the richness of the component parasite community of S. maculatus in the Ilha Solteira hydroelectric reservoir in Brazil was low, in contrast to what has been previously reported in other water environments (Pavanelli et al., 1997(Pavanelli et al., , 2004Takemoto et al., 2009;Casali & Takemoto, 2016). These findings provide insight into the relationships between S. maculatus diet, social behaviour, and parasite fauna and the distribution and infection patterns of the observed parasite taxa. The present study also illustrates the possible effects of habitat homogenisation on parasite infection dynamics in artificial reservoirs. However, additional multidisciplinary research is needed to elucidate the effects of biotic and abiotic factors on the structure and dynamics of component communities of fish parasites in natural and artificial habitats in the neotropics.