Parasites of juvenile Salminus brasiliensis and Brycon orbignyanus (Characiformes, Bryconidae) in the middle Paraná River, Argentina

. Piscivorous Salminus brasiliensis and herbivorous/omnivorous Brycon orbignyanus often form mixed schools in the same habitats. The aims of this study were: (a) to analyze and compare ecto-and endoparasite communities in juvenile S. brasiliensis and B. orbignyanus in the Middle Paraná River, Argentina; and (b) to evaluate the possible relationships between degree of parasitic infestation and body condition (Kn) of the fish. Fish necropsy was performed in search of endo-and ectoparasites, which were mounted in permanent or semi-permanent preparations for taxonomic study using helminthological techniques. A total 24 macroparasite taxa were recorded from the two host species. The parasitic infracommunities of both hosts were competitively structured. Ectoparasite communities were more similar than endoparasite communities. The abundance of parasitic species was not correlated with the standard length, weight and Kn of host individuals, except for the abundance of the copepod E. lacusauratus in the host S. brasiliensis , which was positively related to fish condition. New records of parasites were observed for native hosts, extending the geographic range of several species already known in the Upper Paraná River. The general structure of parasite communities can be determined by several factors, such as parasite-host relationship, host trophic level, and environment. In concordance with the similarity in habitat preferences, and the strong differences in trophic preferences, ectoparasite communities were found to be more similar than endoparasite communities between the two host species.


Introduction
The Neotropical region harbors the world's highest fish species richness (Reis, Albert, Mincarone, Petry, & Rocha, 2016).These fish serve as hosts for a great diversity of ichthyoparasites (Carlson, Dallas, Alexander, Phelan, & Phillips, 2020).Parasite community analysis is useful for understanding fish health status and the links among host, pathogen, and environment (Beldomenico & Begon, 2010).However, although many studies have endeavored to link the ecological characteristics of fish to the diversity of their parasites, there is no general consensus (Luque & Poulin, 2008).
Habitat is one of the most important factors in the parasite-stress-host relationship, as it is a source of parasites (Beldomenico et al., 2008).Fish species with similar evolutionary history, geographic range, ecological niche, and feeding behavior tend to have similar parasite communities (Marcogliese, 2016;Oliveira et al., 2018;Deflem et al., 2022).Additionally, top predators are more likely to have a higher parasite load since they may ingest prey that is already parasitized (Pardo, Zumaque, Hernando Noble, & Suarez, 2008).
Since juvenile S. brasiliensis and B. orbignyanus generally share the same habitat but differ markedly in feeding preferences, we hypothesize that their ectoparasite communities (available in the same habitat) will be more similar while their endoparasite communities (generally originating from their prey) will be more different.The aim of this study was to compare parasitism in mixed schools of juvenile S. brasiliensis and B. orbignyanus and evaluate the possible relationships between level of parasite infestation and body condition of fish caught in the Curtiembre River, Rio Paraná, Entre Ríos, Argentina.

Study area and capture of individuals
The Curtiembre Stream is a small, semi-permanent tributary of the Middle Paraná River, located near the town of General San Martín, Entre Ríos, Argentina (31° 27' 44.3'' S, 60° 10' 41.3" W) (Benzaquén, 2013; Dirección de Hidráulica de Entre Ríos, 2018) (Figure 1).The final stretch of the Curtiembre Stream is seasonally flooded during high-water periods of the Paraná River.Juvenile S. brasiliensis and B. orbignyanus were captured in January 2016 (summer) using gillnets (30and 35-mm mesh opening), arranged in different stretches of the stream.This study was approved by the Research Ethics and Safety Committee of Universidad Nacional del Litoral (Santa Fe, Argentina), and fish collection was authorized by Resolution 1721/14 of the Dirección General de Recursos Naturales, Ministerio de Producción, Turismo y Desarrollo Económico de la Provincia de Entre Ríos.

Fish necropsy, fixation and parasite identification
Fish were euthanized by immersion in a solution with a high dose of benzocaine until opercular movement ceased (Neiffer & Stamper, 2009).Standard length (SL, cm) and body weight (W, g) were measured, and specimens were stored at -20°C for subsequent necropsy in the laboratory.The integument, fins, mouth, nostrils and gill apparatus were examined under a stereoscopic microscope (20X, 40X magnification) for ectoparasites (Eiras, Takemoto, & Pavanelli, 2006).The digestive tract was extracted, and the stomach and intestine were examined.The extracted helminths were preserved in 70% alcohol and mounted in permanent or semi-permanent preparations, according to the usual techniques in helminthology.Reference taxonomic keys and original publications were used for taxonomic identification of the parasites.Voucher specimens were deposited in the Invertebrate Collection of the Museo Provincial de Ciencias Naturales Florentino Ameghino (MFA-ZI) (Santa Fe, Argentina).

Data analysis
Relative condition factor values were calculated for all individuals as described by Le Cren (1951).The logarithms of standard length (Ls) and total weight (Wt) of each individual host were used to adjust the curve for Wt Ls -1 (Wt = a•Lt b ), and the regression coefficients 'a' and 'b' were estimated.The values of 'a' and 'b' were used to estimate the theoretically predicted body weight (We) using the equation: We = a•Lt b .Then the relative condition factor (Kn) was calculated, which corresponds to the ratio between the observed weight and the theoretically expected weight for a given length (Kn = Wt We -1 ).
Prevalence of parasitic infestation (P) was calculated with confidence intervals obtained using Blaker's method, mean infestation abundance (MA), and mean infestation intensity (MI) with their confidence intervals calculated by bootstrap BCa 95% with 2000 replications (Reiczigel, Marozzi, Fábian & Rózsa, 2019;Klaschka & Reiczigel, 2021).To analyze the level of parasite diversity coverage, a species accumulation curve was generated using sampling rarefaction, estimating species richness as a function of the number of samples (Colwell, Mao, & Chang, 2004).The ecological diversity of the parasite infracommunities of both hosts was assessed by calculating the indices of true diversity of Order 0, 1 and 2 (0D, 1D and 2D, respectively) (Moreno, Barragán, Pineda, & Pavón, 2011) and Simpson's dominance index DS (Moreno, 2001).Ecological similarity between endoparasite and ectoparasite infracommunities of each host was calculated using the Jaccard Similarity and Bray-Curtis dissimilarity indices (Moreno, 2001).To analyze whether parasite communities were structured competitively or randomly, the C-Score index was calculated by evaluating its significance using null models (Gotelli & Entsminger, 2011).To compare the prevalence by parasitic taxon between host species, Fisher's exact test was applied.For the mean abundances and mean intensities, a bootstrap t-test with 1000 replications was performed (Reiczigel et al., 2019).To detect the distribution pattern of the parasite populations, the dispersion indices (ID) and Poulin discrepancy (D) were calculated with their confidence limits estimated by bootstrap BCa with 1000 replications (Reiczigel et al., 2019), only for species with prevalence >10%.To analyze the differences in the degree of parasite infestation, the mean values of abundance of ectoparasites, endoparasites and total parasite abundance were compared using Mann-Whitney U tests.To determine the relationship between host morphometric parameters and parasite infestation, Spearman's correlations between SL, body weight, Kn, and parasite abundance were used.The normality of the data was analyzed by Shapiro-Wilk tests.Paleontological Statistics PAST 4.03 (Hammer, Harper, & Ryan, 2001), EcoSim Version 7.0 (Gotelli & Entsminger, 2011), Quantitative Parasitology web QPweb 3.0 Version 1.0.15(6 December 2020) (Reiczigel et al., 2019) and R Studio software (RStudio Team, 2020), were used.Significance level was set at 5% (p < 0.05).
A total 224 macroparasites belonging to 24 taxonomic entities were recorded (Table 1).Ichthyoparasite prevalence was 100% in S. brasiliensis (all individuals were infested by at least one ichthyoparasite), and 68.4% in B. orbignyanus.The most prevalent taxa in S. brasiliensis were the ectoparasites Monogenea gen.sp.(81.82%) followed by Lernaea cyprinacea (63.64%) and Anacanthorus daulometrus (63.64%).The most prevalent taxa in B. orbignyanus were Amplexibranchius bryconis (26.32%) and Ergasilus sp.(21.05%).The accumulation curves of parasite species generated by sample rarefaction approached a plateau, but never became entirely flat (Figure 2).The total parasite community of S. brasiliensis consisted of 181 individuals belonging to 20 taxa, with dominance of 0.26 (± 0.128), and Monogenea being the most abundant taxon.The B. orbignyanus community was made up of 43 parasites from 14 taxa, with higher dominance of 0.40 (± 0.37), and A. bryconis outstanding for its abundance (Table 1).Although fewer S. brasiliensis than B. orbignyanus individuals were analyzed, 0D, 1D and 2 D were higher in S. brasiliensis (Table 2).The estimated Chao 2 diversity surpassed the 0D in both hosts (Table 2).The C-Score index showed that the parasitic infracommunities of both hosts were structured competitively (Table 2).Considering the total parasite abundance per host, similarity was greater in terms of species composition (Jaccard index) compared to considering the relative abundances of parasite species (Bray Curtis similarity) (Table 2).Specifically, the Jaccard index showed higher qualitative similarity in the ectoparasite communities than in the endoparasite communities of S. brasiliensis and B. orbignyanus (Table 2).However, the Bray Curtis similarity was nearly the same for both ectoparasite and endoparasite communities of both host species (Table 2).Comparison of the parasitological descriptors found for the two host populations showed significant differences only between the parasitic prevalence of the ectoparasites Ergasilus thatcheri and Lernaea cyprinacea, in the latter of which the mean infestation abundances also differed (Table 1).Most of the parasite populations with prevalence greater than 10% had an aggregated distribution (Table 3).The only parasites that presented random and aggregated distribution were L. cyprinacea, Anacanthorus daulometrus and Monogenea gen.sp.Total parasite abundance (Z = -4.03;p = 5.56E-05), endoparasite abundance (Z = -2.5;p = 0.01) and ectoparasite abundance (Z = -3.71;p = 1.9E-05) differed significantly between the two host species.In most cases, the abundance of parasitic species was not correlated with standard length, weight or Kn of individuals (Figures 3 and 4).Only the abundance of E. lacusauratus in S. brasiliensis juveniles was positively related to the relative body condition factor (Figure 3).

Discussion
Previous studies on ichthyoparasites of the Middle Paraná River have mostly focused on endoparasites (Chemes & Takemoto, 2011, 2020;Ostrowski de Núñez, Arredondo, & Gil de Pertierra, 2017;Chemes, Gervasoni, Rossi, & Lizama, 2022;Rossi & Chemes, 2022).The current paper provides information on both endo-and ectoparasites of two Characiform hosts that are ecologically and economically important in the region.It is the first report of parasitic fauna associated with B. orbignyanus, a native species valued by sport fishers in the Middle Paraná River.The study reveals the presence of several ectoparasites that were previously recorded in Brazil, but have been found for the first time in Argentina, extending their known geographic range.These ectoparasites include the copepods Amplexibranchius bryconis, Ergasilus lacusauratus and E. thatcheri, and the monogeneans Anacanthorus contortus, A. daulometrus, A. douradensis and A. parakruidenieri.Additionally, the fish studied are reported as new hosts for some parasite species.Specifically, S. brasiliensis is recorded as a new host for the crustaceans E. thatcheri and Lernaea cyprinacea, as well as the acanthocephalan Quadrigyrus machadoi.Brycon orbignyanus is associated for the first time with the branchiurans Argulus ichesi and Dolops geayi, and the copepods Amplexibranchius bryconis, E. lacusauratus, E. thatcheri and L. cyprinacea.This study provides novel information regarding new parasite records and expands the distribution range of several species.However, the species accumulation curve and Chao 2 index indicate the presence of undetected parasite species for these hosts.
Parasite species richness and abundance was higher in S. brasiliensis than in B. orbignyanus.Both species had more diversity in ectoparasites than endoparasites.In S. brasiliensis, monogeneans had higher values of prevalence, mean intensity, and mean abundance of infestation, and were found mainly in the gills.They are of sanitary interest because a high abundance of parasites can cause asphyxia in fish and decrease sanitary conditions, leading to a disease known as 'dactylogyridosis' (Eiras et al., 2006;Pavanelli, Eiras, & Takemoto, 2008).In the Upper Paraná River, Cohen et al. (2012) observed monogeneans in S. brasiliensis, though with a lower prevalence than in the current study.
In B. orbignyanus, the highest prevalence, abundance and mean intensity of infestation was found for the family Ergasilidae and the species Amplexibranchius bryconis, which has also been reported in B. cephalus in the Amazon basin (Thatcher & Paredes, 1985).The genus Brycon is widely farmed in Brazilian rivers, including the Paraná River basin (Zaniboni Filho, Reynalte-Tataje, & Weingartner, 2006).Fish farming involves moving fish among facilities in different basins, and may account for the dispersal of the parasite A. bryconis in the Middle Paraná River.A similar situation occurred with the dispersal throughout Latin America of the copepod Lernaea cyprinacea, a parasite originally associated with Cyprinus carpio, which has recently been recorded in the region parasitizing another native species (Gervasoni, Chemes, Scaglione, & Cerutti, 2018).In the Upper Paraná River, Lehun et al. (2020) observed monogeneans in S. brasiliensis, though with lower parasitic prevalence than in the present study.In Argentina, there is an aquaculture market for S. brasiliensis.Proper management in the different fish farm systems in the Paraná River could help prevent the spread of its parasites.
Differences between parasitological descriptors and diversity indices in hosts that use the same h abitats could be linked to different interactions among life histories, population dynamics of host fish and parasites, and niche availability (non-infested fish), which do not only involve the environment as a main factor (Luque & Poulin, 2007;Takemoto et al., 2009;Marcogliese & Pietrock, 2011;Alcântara & Tavares Dias, 2015;Soto et al., 2016).Juvenile S. brasiliensis had a smaller body size, so they are not strictly piscivorous (> 21 cm) (Bechara et al., 2005).However, S. brasiliensis showed greater abundance and diversity of parasites, and less similarity in the species composition of its endoparasite community than that of B. orbygnianus ectoparasites.According to Bechara et al. (2005), S. brasiliensis juveniles feed mainly on fish, insects and crustaceans, while B. orbignyanus juveniles feed primarily on zooplankton, small aquatic insects, and filamentous algae (Tonella, Dias, Vitorino, Fugi, & Agostinho, 2019).The piscivorous feeding of dorado juveniles could foster intake of fish or invertebrates already infested with parasites, which is known as a process of parasitic bioaccumulation through food chains.In B. orbignyanus, the present study found a lower diversity of parasites than that reported by Fernandes, Casali, and Takemoto (2019) in a study carried out in the Alto Paraná River, where the authors analyzed 104 fish This suggests that differences in sample size affect the parasite richness observed.Our results are in concordance with studies that indicate a higher abundance of endoparasites in carnivorous vertebrates than in herbivores or omnivores (Marcogliese, 2002;Lafferty et al., 2008;Pardo et al., 2008).However, these differences are more pronounced in adult fish than in juveniles since, in some hosts, trophic specificity b ecomes more accentuated with ontogenetic development (Bellay et al., 2015).In agreement with other studies that have analyzed fish body condition, the juveniles of both characiform species had good body condition in the natural environment studied (Lizama et al., 2006;Satake, Mayumi Ishikawa, Hisano, Benites de Pádua, & Tavares-Diaz, 2009).Unlike studies that surveyed more than 150 hosts, finding a directly proportional relationship (Lizama et al., 2006;Österlinga & Larsenb, 2013), the current study found no relationship between the parasite abundance and body condition in either host.One possible explanation is the limited variation in parasite abundance among the collected individuals.It is important to consider that the diversity and abundance of ichthyoparasites are influenced by both host characteristics (such as body size, diet, and biogeographical distribution) and environmental factors (Takemoto, Pavanelli, Lizama, Luque, & Poulin, 2005;Timi et al., 2011).
The only parasite in which abundance was positively associated with the body size of S. brasiliensis juveniles was E. lacusauratus, with larger fish harboring a greater abundance of this parasite species.In the fish Leporinus fasciatus, gill area (gill size) was positively related to the abundance of Ergasilus sp.copepods (Oliveira, Prestes, Adriano, & Tavares-Dias, 2022), suggesting that larger fish with larger gill arches could host higher abundances of parasites.This observation should be considered in future research, because the family Ergasilidae includes species with high pathogenic potential, some of which are responsible for high mortality among farmed fish in both freshwater and brackish environments (Piasecki, Goodwin, Eiras, & Nowak, 2004).
The current study provides original information on the parasitic communities of S. brasiliensis and B. orbignyanus in the Middle Paraná River Argentina.New parasites were recorded for the native hosts, extending the geographic range of several parasite species already known in the Upper Paraná River.The general structure of the parasitic communities was determined by several factors, such as the parasite-host relationship, host trophic level, and the similarity of the environment they share.

Conclusion
During the course of our study, a total of 24 macroparasite taxa were found in S. brasiliensis and B. orbignyanus.These findings extend the geographic range of several species known in the Upper Paraná River.The parasitic infracommunities were competitively structured.In concordance with the similarity in habitat preferences, and the strong differences in trophic preferences, ectoparasite communities were found to be more similar than endoparasite communities between the two host species.Factors such as parasite-host relationship, host trophic level, and environment play significant roles in shaping parasite communities.

Figure 1 .
Figure 1.(a) Location of the Curtiembre stream in Entre Ríos Province in Argentina.(b) Satellite image showing the distribution of sampling points along the last reach of the Curtiembre stream and its mouth into the Middle Paraná River.

Figure 2 .
Figure 2. Accumulation curves of parasite species generated by sample rarefaction from juvenile Salminus brasiliensis and Brycon orbignyanus from the Curtiembre Stream, Middle Paraná River (Argentina).Grey lines represent ± standard error.

Figure 3 .
Figure 3. Spearman correlation among relative condition factor (Kn), standard length (SL) and weight (W) of Salminus brasiliensis (SB) and its parasites.The color gradient represents the intensity, and the asterisk represents the significance of the correlation.

Figure 4 .
Figure 4. Spearman correlation coefficients among relative condition factor (Kn), standard length (SL) and weight (W) of Brycon orbignyanus (BO) and its parasites.The color gradient represents the intensity of the correlation.

Table 1 .
Parasites of Salminus brasiliensis and Brycon orbignyanus from the floodplain of the Middle Paraná River, Entre Ríos (Argentina).

Table 2 .
Ecological attributes of the parasitic infracommunities of juvenile Salminus brasiliensis and Brycon orbignyanus from the Curtiembre Stream, Middle Paraná River (Argentina).