Introduction

The Amazon basin is a center of diversity for most groups of Neotropical fish, that is to say, it is an area of high species richness, due to its large extension of floodplains, which are important habitats for native fish, providing feeding and nursery zones. Conservative estimates suggest there are about 3,000 fish species in this basin (^{ALBERT & REIS, 2011}; ^{JUNK, 2013}; ^{FROESE & PAULY, 2016}). In the Amazonian estuary, there are 243 fish species, of which 23 are endemic species. Many of these fish are important for trade and economy in the Amazon, and constitute the main source of food for human populations in the region. Moreover, this basin has diverse tributaries draining its water levels, which vary enormously during the year (^{ALBERT & REIS, 2011}), including the Igarapé Fortaleza hydrographic basin.

In the Amazon River system, in the region of the state of Amapá (Northern Brazil), there is the Igarapé Fortaleza hydrographic basin, an important tributary of this river. The Igarapé Fortaleza basin, located at the estuarine coastal sector, is characterized for having a river system with extensive floodplains, which is drained by fresh water and connected to a main watercourse, influenced by high rainfalls and tides (every 12 hours) from the Amazonas River (^{TAVARES-DIAS et al., 2013}). This tributary of the Amazon River system harbors more than 80 species of freshwater fish (^{GAMA & HALBOTH, 2004}), including Characiformes species of the genus Triportheus Cope, 1872, which are popularly known as freshwater sardines and represents an important resource for artisanal fishing and subsistence of human population in the region.

Triportheus spp. are Characidae with a geographic wide distribution in Bolivia, Colombia, Peru, Argentina, Ecuador, Venezuela and Brazil (^{SANTOS et al., 2006}; ^{FROESE & PAULY, 2016}). Currently, 18 species of Triportheus Cope, 1872 are known, including Triportheus rotundatus Jardine, 1841 (^{MALABARBA, 2004}; ^{FROESE & PAULY, 2016}), the fish species that is the focus of the present study. The species of Triportheus inhabit most of the major river drainages of South America, and constitutes an important element in both commercial and subsistence fisheries in the Amazon basin (^{MALABARBA, 2004}), its reproduction occurs during the rainy season in the Amazon River system (^{FROESE & PAULY, 2016}). Triportheus rotundatus is a benthopelagic fish with an omnivorous diet, fed on fruits, seeds and insects that float on the water surface, besides microcrustaceans (^{PEREIRA et al., 2011}; ^{FROESE & PAULY, 2016}; ^{SUÇUARANA et al., 2016}). However, studies on the biology of T. rotundatus are reduced, mostly those regarding its parasitic fauna.

Studies on parasitic fauna aspects should be directed to T. rotundatus, due to the importance of knowledge on several factors that could influence the diversity and structure of the parasite infracommunities in fish populations (^{MOREIRA et al., 2009}; ^{LAGRUE et al., 2011}; ^{TAVARES-DIAS et al., 2013}; ^{COSTA-PEREIRA et al., 2014}; ^{OLIVEIRA et al., 2016}). Moreover, the knowledge about parasites community infracommunities and their relationship with the host fish is of great importance, because the parasites also play a key role in ecosystems by regulating the abundance or density of natural fish populations, thus stabilizing food web and host community structures (^{MOREIRA et al., 2009}; ^{MORLEY, 2012}; ^{TAVARES-DIAS et al., 2013}; ^{OLIVEIRA et al., 2016}). However, the ecological knowledge of parasites in Neotropical environments is yet very limited.

Although there is no information on the parasites of T. rotundatus, for other Triportheus spp. diverse Anacanthorus species, Ancistrohaptor sp., Ichthyophthirius multifiliis, Piscinoodinium pilullare, Procamallanus (Spirocamallanus) sp., Procamallanus (Procamallanus) peraccuratus, Procamallanus hilarii, Contracaecum sp, Goezia sp., Echinorhynchus paranensis, Ergasilus sp. and Dolops sp. have been reported (^{MACHADO, 1959}; ^{KRITSKY et al., 1992}; ^{COHEN et al., 2013}; ^{COSTA-PEREIRA et al., 2014}; ^{OLIVEIRA et al., 2016}). However, there are no studies on the parasites of T. rotundatus. In this way, the present study was the first investigation on several aspects of parasite communities in T. rotundatus from a tributary of the Amazon River, in the state of Amapá (Northern Brazil).

Materials and Methods

Fish and parasite sampling

Thirty-two T. rotundatus (14.7 ±1.3 cm and 39.4 ± 10.8 g) were collected in the Igarapé Fortaleza basin, in the state of Amapá, eastern Amazon region, northern Brazil (Figure 1), in the period from December 2012 to August 2013 for parasitological analysis. The fish were caught with different nets and transported in box with ice to the Health Laboratory of Aquatic Organisms of Embrapa Amapá, Macapá, Amapá state (Brazil). The fish collected were weighed (g) and measured for total length (cm). The present work was developed according to the principles adopted by the Brazilian College of Animal Experiments (COBEA), with the authorization from Ethics Committee in the Use of Animal of the Embrapa Amapá (#004 - CEUA/CPAFAP) and ICMBio (# 23276-1).

Figure 1 Collection locality of Triportheus rotundatus in tributary from Amazon River system, Northern Brazil. 

Results

All examined fish (100%) were parasitized by one or more metazoan species, and 1,316 parasites were collected. Anacanthorus pithophallus Kritsky, Boeger & Van Every, 1992; Anacanthorus furculus Kritsky, Boeger & Van Every, 1992; Ancistrohaptor sp. (Dactylogyridae); Genarchella genarchella Travassos, Artigas & Pereira, 1928 (Derogenidae); Posthodiplostomum sp. (Diplostomidae); Procamallanus (Spirocamallanus) inopinatus Travassos, Artigas & Pereira, 1928 (Camallanidae); Echinorhynchus paranensis Machado, 1959 (Echinorhynchidae) and Ergasilus sp. (Ergasilidae) were found (Table 1). Monogenoideans were the dominant parasite species, and although these parasites were not possible to count by species, the predominance was de A. furculus and A. pithophallus. None protozoan parasite was found. Parasites had an aggregate dispersion pattern, except for P. (S.) inopinatus, which presented random dispersion pattern (Table 2).

The species richness of parasites, the Brillouin diversity (HB) and evenness (E) were low (Table 3). Species richness of parasites (rs = 0.243, p= 0.180) and HB (rs = 0.273, p= 0.130) did not show any significant correlation with total host length. Hosts parasitized by four to five parasite species were predominant (Figure 2).

Figure 2 Species richness of parasites in Triportheus rotundatus from Amazon River system (Brazil). 

The abundance of monogenoideans presented negative correlation with the length (rs = - 0.363, p= 0.041) and weight (rs = -0.511, p= 0.003) of the hosts. The abundance of G. genarchella showed no correlation with the length (rs = -0.013, p = 0.941) and weight (rs = -0.069, p = 0.705), as well as the abundance of Posthodiplostomum sp. with the length (rs = -0.132, p = 0.471) and weight (rs = -0.218, p = 0.228), and the abundance of P. (S.) inopinatus with the length (rs = 0.075, p = 0.680) and weight (rs = 0.038, p = 0.832) of the hosts.

The equation of weight (W)-length (L) relationship for this host was Wt = 0.1493Lt^2.0653, r² = 0.710), with negative allometric, indicating greater increase in body weight than in size. The condition factor (Kn = 0.999, t = -0.0041, p = 0.997) of the parasitized fish did not differ from the standard value (Kn = 1.000), thus indicating that the parasitism did not impair host body condition.

Discussion

In T. rotundatus, the parasitic community was composed of 3 species of Monogenoidea, 2 Digenea, 1 Nematoda, 1 Acanthocephala and 1 Copepoda, with a dominance of monogenoideans (A. pithophallus, A. furculus and Ancistrohaptor sp.). Moreover, low species richness, low Brillouin diversity, low evenness, a high diversity of endoparasites species, and overdispersion of parasites characterized this parasitic community. Many parasite species commonly show overdispersion in different host fish (^{COSTA-PEREIRA et al., 2014}; ^{TAVARES-DIAS et al., 2015}; ^{OLIVEIRA et al., 2016}), due to processes that produce variability in host exposure, host acceptability by the parasite and host immune response. For Triportheus nematurus from Pantanal Matogrossense (Brazil) 1 species of Monogenoidea, 3 Nematoda and 1 Copepoda were reported, being Anacanthorus sp. the main component and predominant in helminth species with overdispersion (^{COSTA-PEREIRA et al., 2014}). For Triportheus curtus Garman, 1890 and Triportheus elongatus Spix & Agassiz, 1829 from same region of this study 3 species of Protozoa, 1 Monogenoidea, 1 Argulidae, 2 Digenea and 2 Nematoda were reported, being I. multifiliis the predominant parasite species (^{OLIVEIRA et al., 2016}). However, we found a higher parasitic prevalence and abundance in T. rotundatus when compared to those observed by ^{Costa-Pereira et al. (2014)} in T. nematurus. Such differences for this fish species with a similar mode of life may be attributed to the environment and the opportunities of these hosts found infective stages of parasites with complex life cycle.

Monogenoideans are parasites that are highly host specific when compared to other groups of parasites. Studies suggest that the distribution of monogenoideans on their fish hosts is strongly influenced by evolutionary history, both between and within fish orders (^{KRITSKY et al., 1992}; ^{BRAGA et al., 2014}). The gills of T. rotundatus were infected by A. pithophallus, A. furculus and Ancistrohaptor sp., parasites whose abundance presented a negative correlation with the size of the hosts. Such monogenoidean species has been also recorded in other Triportheus species from Brazil (^{KRITSKY et al., 1992}; ^{COHEN et al., 2013}; ^{OLIVEIRA et al., 2016}). However, while Ancistrohaptor spp. seems restricted to Triportheus spp. (^{COHEN et al., 2013}; ^{BRAGA et al., 2014}), Anacanthorus spp. presents a widespread distribution among freshwater fish species, once that they are also found in diverse Characiformes species of fish from several families (^{BRAGA et al., 2014}).

Characteristics of the habitat may facilitate the transmission and establishment of fish parasites (^{MOREIRA et al., 2009}; ^{COSTA-PEREIRA et al., 2014}; ^{BITTENCOURT et al., 2014}; ^{OLIVEIRA et al., 2016}). The abundance and diversity of the invertebrate fauna are also key components in the formation of a parasite community in fish populations (^{MOREIRA et al., 2009}; ^{MORLEY, 2012}; ^{TAVARES-DIAS et al., 2015}). Moreover, environments with abundance of aquatic vegetation (e.g. macrophytes), as the ecosystem of this study (^{THOMAZ et al., 2004}), may influence the abundance of parasites (^{MOREIRA et al., 2009}; ^{MORLEY, 2012}), because its floodplains areas are widely used for shelter and feeding by many fish species. Consequently, this increases the possibility of the fish finding different infective stages of the parasite species in the environment. Therefore, these factors favored the transmission of G. genarchella, Posthodiplostomum sp., P. (S.) inopinatus and E. paranensis in T. rotundatus. Most of these endoparasites present a wide distribution in freshwater fish from Brazil and have different macroinvertebrates as intermediate hosts. Procamallanus (S.) inopinatus has chironomid species as intermediate hosts (^{MOREIRA et al., 2009}), while digeneans G. genarchella and Posthodiplostomum sp. have mollusk species as intermediate hosts and aquatic fish-eating birds as definitive host. Concerning E. paranensis, it was described by the first time on Triportheus paranensis Günther, 1874 from Mato-Grosso, Brazil (^{MACHADO, 1959}). Recently, this acanthocephalan has also been reported infecting Pygocentrus nattereri Kner, 1860 (Serrasalmidae) from Negro River, at Pantanal of the Mato Grosso do Sul (Brazil), as well as Chaetobranchus flavescens Heckel, 1840 and Chaetobranchopsis orbicularis Steindachner, 1875 (Cichlidae) from the region of this study (^{BITTENCOURT et al., 2014}). However, as there are few reports on the occurrence of E. paranensis in fish, little is known about its biology and life cycle.

Concerning Ergasilus sp., these copepods occurred only on the gills of two specimens of T. rotundatus, because the water dynamic of the Igarapé Fortaleza basin seems to hinder the encounter of the parasitic crustacean species with host fish, once these ectoparasites need swimming to infect its host fish. Similarly, ^{Costa-Pereira et al. (2014)} also reported low prevalence and abundance of Ergasilus sp. in T. nematurus. Moreover, some parasitic crustacean species are host and site-specific, especially in relation to fish in particular habitats and life styles, while other parasites frequently have no preference. There is a dominance of Ergasilidae, mainly Ergasilus Nordmann, 1832, among the parasitic crustaceans in freshwater fish from Brazil. However, Ergasilus spp. have infested species of Characidae, Pimelodidae, Anostomidae and Cichlidae from the Amazon (^{TAVARES-DIAS et al., 2015}; ^{TABORDA et al., 2016}). Twenty-two freshwater species of Ergasilus are known from the gills of Neotropical fish (^{TABORDA et al., 2016}).

In summary, for T. rotundatus the ectoparasites community consisted of species with high prevalence and abundance, while the endoparasites community presented low prevalence and abundance, which did not affected the body condition of the hosts. The presence of endoparasites with a complex life cycle indicates that in this environment, the diet of T. rotundatus consists mostly of chironomids, mollusks and microcrustaceans. Thus, T. rotundatus is an intermediate or paratenic host for G. genarchella, Posthodiplostomum and E. paranensis, and a definitive host for P. (S.) inopinatus. The size of the host had little influence on parasite communities, once they had influence only on monogenoideans. Finally, the behavior and availability of infective stages, which intermediate hosts for endoparasites, were factors structuring the communities of parasites in this Amazonian host.