Parasite Community of Cichla kelberi ( Perciformes , Cichlidae ) in the Três Marias Reservoir , Minas Gerais , Brazil

A total of 112 specimens of yellow peacock bass, Cichla kelberi, were collected in the Três Marias Reservoir, upper São Francisco River, State of Minas Gerais, between 2004 and 2005. Twelve species of parasites were collected: two species of ectoparasites (the monogenetic Gussevia sp. and the copepod Ergasilus sp.); and ten species of endoparasites, including four species of adult endoparasites: the cestodes Proteocephalus macrophallus and P. microscopicus, and the nematodes: Capillostrongyloides sentinosa and Rhabdochona acuminata, and six species of larval endoparasites: the metacercariae of Austrodiplostomum compactum and the nematodes: Hysterothylacium sp., Contracaecum sp. Type 1, Procamallanus (Spirocamallanus) inopinatus, Spiroxys sp., and Goezia sp.. A significant positive correlation in the abundance of the two adult endoparasites species, P. macrophallus and P. microscopicus, may indicate that such associated species had the same ecological requirements. The positive correlations between the prevalence of the larval endoparasites, Hysterothylacium sp. and A. compactum, may indicate that there was opportunism for the two generalists species. Ergasilus sp., C. sentinosa, R. acuminata and Hysterothylacium sp. were recorded for the first time in C. kelberi, and this increased their list of known hosts. In addition, both proteocephalid species and C. sentinosa can extend their known biogeographical distribution to the Três Marias Reservoir.


Introduction
Species of the genus Cichla Schneider, 1801, which is also known as the peacock bass, originate from the Amazon Basin and represent the major group of piscivorous fish in the Cichlidae in South America (LOWE-MCCONNEL, 1991).
The introduction of species is a type of ecological change that can modify the biotic community in which the species has been introduced (LI; MOYLE, 1981).The presence of the exotic species Cichla kelberi Kullander & Ferreira, 2006, in the Três Marias Reservoir, upper São Francisco River, has been confirmed by experimental fishing since 1982, and there are indications that the species may have found favorable reproductive conditions in this reservoir (SATO; GODINHO, 1999).
There are scarce studies on the impacts of the introduction of exotic species into reservoirs in Brazil.According to Sunaga and Verani (1997), there is the possibility that exotic fishes, such as the peacock bass Cichla ocellaris Schneider, 1801 and piranha Pygocentrus nattereri Kner, 1858, have caused damage in the structure of fish communities in the lakes in the middle Doce River.Godinho et al. (1994) reported a lower richness and abundance of species; smaller sized individuals in the little ponds of the middle Doce River were subject to the introduction of C. ocellaris and P. nattereri.Moreover, losses in the fish community may have occurred in the Redonda Lagoon, State of Rio Grande do Norte, by the introduction of C. ocellaris (MOLINA et al., 1996).On the other hand, the introduction of peacock bass, Cichla monoculus Agassiz, 1831, and croaker Plagioscion squamosissimus (Heckel, 1840) in the Paraná River basin have had great economic importance in the region (AGOSTINHO; JÚLIO JR, 1999).Impacts on the structure of the native fish community of the Três Marias Reservoir caused by the introduction of C. kelberi are still unknown.
Cichla kelberi lives in lentic environments where it reproduces mainly during the rainy season (ZARET, 1980) and shows sexual dimorphism during the breeding season, when a post-occipital protuberance appears in male individuals (FONTENELE, 1948).This species provides parental care to its offspring and has a high fertility rate; there is the presence of adhesive organs in the larvae that prevent them from being carried by the currents (FONTENELE, 1950).The species also builds nests (BRAGA, 1952), and it presents fractional spawning (MAGALHÃES et al., 1996), an aeration capacity of its eggs independent of the rate of oxygen dissolved in water (ZARET, 1980), and piscivorous feeding behavior (GOMES, 2002).
The yellow peacock bass, as it is popularly known, has light spots on its pelvic, anal, and caudal fins -a characteristic that distinguishes it from other species of peacock bass.It is naturally distributed in the Araguaia River basin, in the states of Mato Grosso and Goiás, and the lower  (Travassos, Artigas & Pereira, 1928), in C. kelberi from the Ingazeira Weir, Paulistana, State of Piauí.Yamada (2008) recorded the monogeneans Sciadicleithrum ergensi Kritsky, Thatcher, and Boeger, 1989, Sciadicleithrum uncinatum Kritsky, Thatcher, and Boeger, 1989, Gussevia longihaptor (Mizelle & Kritsky, 1969), Gussevia undulata Kritsky, Thatcher, and Boeger, 1986, and Ascocotyle sp. (metacercariae) in peacock basses collected in the Itaipú Reservoir, State of Paraná.Yamada et al. (2011) inventoried the ectoparasites S. ergensi, Gussevia tucunarense Kritsky, Thatcher, and Boeger, 1986, G. undulata  The goal of the present study was to identify the component species of the parasite community of C. kelberi in the Três Marias Reservoir, so as to contribute to the known parasite biodiversity of cichlid fish from the São Francisco River, Brazil.Specimens of C. kelberi were identified according to Kullander and Ferreira (2006).During necropsy of the fish, their total length, weight, and sex were recorded on forms, and the parasite specimens were collected, fixed, and processed for identification in accordance with methodologies used for the study of fish parasites (AMATO et al., 1991).

Materials and Methods
Voucher specimens of C. kelberi were deposited in the Museum of Zoology, University of São Paulo (MZUSP), São Paulo, No.
95148.Voucher specimens of parasites of C. kelberi were deposited in the Coleção Helmintologica of the Instituto Oswaldo Cruz (CHIOC), Rio de Janeiro, according to the numbering presented in the Results section.The identification of parasites follows specific scientific literature, and the larval nematodes were classified according to Moravec (1998).
Statistical tests were only applied to the parasite species that showed a parasitic prevalence higher than 10% (BUSH et al., 1990).The ecological descriptors used were in accordance with Bush et al. (1997).Species dominance in communities, as based on the proportion of infected/infested fish, was tested using Simpson's index (C) (STONE; PENCE, 1978); dominance was considered when values were greater than or equal to 0.25.The frequency of dominance, the shared frequency of dominance, and the mean relative dominance of each parasite species were calculated according to Rohde et al. (1995).The ratio between the mean parasite variance and abundance (dispersion index [DI]) was calculated for each parasite species in order to determine their distribution pattern.Their significance was tested using the statistical d-test (d>1.96)(LUDWIG; REYNOLDS, 1988).
Student's t-test was used to check for possible differences between the weights and sizes of hosts in relation to their sex.The chi-square test (Χ 2 ), using the Yates correction, and a 2 × 2 contingency table, were used to determine the influence of sex and the period of collection on the prevalence of parasites.Spearman's correlation coefficient by ranks (r s ) was used to evaluate possible correlations between the hosts' total length and intensity, abundance, richness, and parasite diversity.Pearson's correlation coefficient (r) was used to assess the correlation between parasite prevalence (using prior angular transformation of the data) in relation to the hosts' size classes, as estimated by Sturges's formula (STURGES, 1926).The Mann-Whitney U-test was used to assess the possible differences between intensity, abundance, richness, and parasite diversity in relation to the sex and period of collection of hosts.The statistical analyses applied to the parasite infrapopulations and parasite community were performed in accordance with the methods of Zar (2006).
The community descriptors of parasitism that were calculated included the following: mean parasite abundance; mean parasite richness; Shannon-Wiener's parasite diversity (H'); and Hill's equivalence modified (E).The DivEs software was used to calculate these indexes (RODRIGUES, 2007).Berger-Parker's numerical dominance (Dbp), as based on parasite infracommunities, was calculated from the respective abundances (MAGURRAN, 1988).The level of significance for all tests was p<0.05.

Component parasite community
Not all fish of the sample analyzed were parasitized (n=9).Twelve species of metazoan parasites were collected.Two species were ectoparasites and ten were endoparasites; there were four species of adult endoparasites and six species of larval endoparasites, totaling 3205 specimens and an average of 28.6 parasites per fish.
Simpson's dominance index (C) indicated that there was no dominance in the parasite community of C. kelberi (C=0.21), even though P. macrophallus and Hysterothylacium sp.presented with higher frequency of dominance values (Table 2).
The metazoan parasites of C. kelberi showed the typical pattern of aggregated or superdispersed distribution, which is a common ictioparasitology pattern, as shown in Table 3.The parasite community of C. kelberi presented with H'=0.55,E=2.40, D bp =0.55, a mean parasite richness of 2.12 ± 1.13, and a mean parasite abundance of 28.6 ± 66.6.

Parasite infracommunities
The sample analyzed included 53 females and 59 males of C. kelberi.The females had a mean diversity 0.21 ± 0.14 (0-0.60) and a richness of 2.11 ± 1.28 (0-5) species.The males showed 0.20 ± 0.17 (0-0.62) and 2.10 ± 0.99 (0-5), respectively.There was no statistically significant difference between the parasite diversity and parasite richness values with respect to the hosts' sex (U=1221.5, p=0.58;U=1529.0,p=0.84, respectively).The parasite indexes of the five parasite species analyzed were not influenced by the sex of C. kelberi (Table 4) during the entire collection period.The mean total abundance was not correlated with the sex of the fish (U=1520.0,p=0.80).
The abundance of P. macrophallus was higher in fish with smaller total length, representing a negative correlation (Table 5).The diversity and richness of parasite infracommunities were not correlated with the total length of the hosts (r s =-0.15, p=0.12; r s =-0.015, p=0.88, respectively).The mean total abundance proved to be negatively correlated with the size of the fish (r s =-0.19, p=0.04).
A total of 62 specimens of C. kelberi were collected during the dry season and 50 specimens were collected during the rainy season.The mean prevalence, intensity, and abundance of the parasites found are shown in Table 6.The fish collected during the dry season showed a mean parasite diversity of 0.19 ± 0.15 (0-0.62) and a mean parasite richness of 2.69 ± 1.08 (0-5 species).
Some of the parasite parameters of Hysterothylacium sp. were significantly higher during the dry season, unlike the prevalence and abundance of P. macrophallus, which were significantly higher in the rainy season (Table 7).
There were two associations in the parasite community of C. kelberi: one among adult endoparasites with a correlation between their abundances; and one association between the prevalence of larval endoparasites (Table 8).

Discussion
Even though endoparasites predominated over the ectoparasites with respect to the number of species found, there was no dominance of any parasite species in the parasite community of C. kelberi.
Austrodiplostomum compacutm showed the highest prevalence value in the parasite community of C. kelberi, followed by Hysterothylacium sp..The proteocephalid cestodes had intermediate prevalence values, and ectoparasites -which were represented by only two species -had low parasitic indexes; this was also the case for the other nematode larvae.
Despite the piscivorous feeding behavior of peacock basses, Gomes and Verani (2003) pointed out that they vary their feeding behavior from microcrustaceans and rotifers to insects and shrimps, and they also feed on fish as they move between habitats.This may explain why there is a low prevalence of adult and larval nematode species, with the exception of the Hysterothylacium whose presence appeared to be accidental.These species should have low transmission rates and less potential to colonize C. kelberi, which may have occurred due to the lower availability of the intermediate hosts in the case of endoparasites, or it may be due to the higher pathogenicity of these parasites, which would cause a more efficient immune response in the hosts (GUIDELLI et al., 2003).It is likely that these nematode larvae use various species of fish to ensure successful parasite transmission.
According to Szidat and Nani (1951), cercariae of diplostomids penetrate actively into the skin of the fish, and they subsequently develop into metacercariae.When fish-eating birds (cormorants and gulls) prey on infected fish, they can acquire these metacercariae which, in this case, will develop into adult worms in the intestine.When the eyes of the hosts have a high number of metacercariae, the fish are immediately known by the fishermen because they present a "verminous cataract" due to the opacity in the lens of the eyes.In the present study, specimens of A. compactum, which are representative of the Diplostomidae found free (not encysted) in the eyes (lens and vitreous humor) of peacock bass, include these fish in this listing of intermediate hosts.
Larvae of Hysterothylacium sp.found in the coelomic cavity, stomach, and anterior, middle, and posterior intestines showed high prevalence, with more than 50% of the fish collected parasitized.It is probable that when specimens of C. kelberi feed on smaller prey (fish), they acquire these larvae that accumulate in these fish until they find their definitive hosts.In this case, C. kelberi can also be a paratenic host.According to Bell and Burt (1991), piscivorous fish host more species of endoparasites than non-piscivorous fish; this occurs because of the fact that when piscivorous fish feed on forage fish (smaller), they acquire possible helminth larvae that are present in the food.
In the present day, P. macrophallus and P. microscopicus are unique endohelminths of Cichla spp., having never been found in other fish species in Brazil (REGO, 1994;MACHADO et al., 2000;THATCHER, 2006;AZEVEDO et al., 2010;KOHN et al., 2011;FRANCESCHINI et al., 2013;LACERDA et al., 2013a;YAMADA;TAKEMOTO, 2013).According Guégan and Kennedy (1993), the fauna of the parasites of a species of fish is influenced from the time it is introduced into a new habitat.Fish introduced to a new location can usually carry a few species of parasites that parasitized them at their location of origin.Subsequently, these fish acquire new parasite species in their new habitat (POULIN; MORAND, 2004).Since C. kelberi is an exotic species, it is possible that the proteocephalid cestodes that were found had been introduced in conjunction with their hosts, because no other species of fish from the Três Marias Reservoir in the upper São Francisco River region, whose parasitic fauna was investigated, presented with such helminths (BRASIL-SATO, 2003;BRASIL-SATO;SANTOS, 2003SANTOS, , 2005;;SANTOS;BRASIL-SATO, 2006;SANTOS, 2008;MONTEIRO et al., 2009).Lacerda et al. (2013b) accepted this hypothesis when they verified that both species  of proteocephalids present in C. piquiti Kullander & Ferreira, 2006, from the original environment (Lajeado Reservoir and São Salvador, Tocantins River) continued to parasitize those hosts that were introduced in new places (Jupiá and Itaipu Reservoirs, Paraná River).All of the parasite species that were analyzed showed a typical pattern of aggregated or superdispersed distribution, which are common patterns in studies on parasites of freshwater fish (MACHADO et al., 2000;GUIDELLI et al., 2003;SANTOS;BRASIL-SATO, 2006).
Significant positive associations were observed in pairs of species of adult endoparasites (P.macrophallus and P. microscopicus), with a positive correlation noted between their abundances; this may indicate that such associated species have the same ecological requirements and that there are probably frequent and common intermediate hosts for the associated species (GUIDELLI et al., 2003).The relationship observed between both generalist species of the larval endoparasites, Hysterothylacium sp.(passive transmission) and A. compactum (active transmission), exhibited positive correlations between their prevalence rates; this may indicate that opportunism occurred between the two species, especially since their transmission routes are quite distinct from one another.Santos et al. (2009) reported the presence of Spiroxys sp. in C. kelberi of Três Marias Reservoir and considered such occurrence as accidental due to the low prevalence rate recorded.The authors also reported the occurrence of these larvae in Pygocentrus piraya (Cuvier, 1819) and Serrasalmus brandtii Lütken, 1875, which are characiforms fish from the Três Marias Reservoir.
Ergasilus sp., C. sentinosa, R. acuminata, and Hysterothylacium sp. have been recorded for the first time in C. kelberi, thus expanding the list of hosts.In addition, for both species of proteocephalids and C. sentinosa, the Três Marias Reservoir, in the upper São Francisco River, has now become a new location for these parasite species, thus expanding upon their known geographical distribution.

A
total of 112 specimens of C. kelberi were collected in the Três Marias Reservoir (18°12'59" S; 45°17'34" W) located in the upper São Francisco River, near the Três Marias municipality, State of Minas Gerais, Brazil.The collection was carried out in August 2004, July and August 2005 (dry season); and in January 2004, December 2004, and January 2005 (rainy season).Fish collections were performed with the aid of gillnets that were placed the night before by fishermen of the Estação de Hidrobiologia e Piscicultura of the Companhia de Desenvolvimento dos Vales dos Rios São Francisco e Parnaíba (EPT/CODEVASF).
Tocantins River, State of Pará.It has been introduced into several reservoirs in the states of Rio Grande do Norte, Minas Gerais, and Ceará, and C. monoculus and C. ocellaris have been introduced into the Paraíba do Sul River, State of Rio de Janeiro, and the Rio Paraná basin, State of Paraná, extending into Paraguay (KULLANDER; FERREIRA, 2006).Few studies have been conducted on the parasite fauna of C. kelberi to date.Araujo et al. (2009) recorded the nematode,

Table 2 .
the Três Marias Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.Frequency of dominance, shared frequency of dominance, and mean relative dominance of the metazoan parasites of Cichla kelberiKullander & Ferreira, 2006, Três Marias Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.

Table 3 .
Values of the dispersion index and the statistical d-test of the metazoan parasites of Cichla kelberiKullander & Ferreira, 2006, Três  Marias Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.

Table 4 .
Analysis of parasite indexes under possible influence of the sex of Cichla kelberiKullander & Ferreira, 2006, Três Marias Reservoir,  upper São Francisco River, State of Minas Gerais, Brazil.

Table 5 .
Analysis of parasite indexes under possible influence of the total length of Cichla kelberiKullander & Ferreira, 2006, Três Marias  Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.

Table 6 .
Mean prevalence, intensity, and abundance of the metazoan parasites of Cichla kelberiKullander & Ferreira, 2006, in the respective collection period (dry and rainy seasons), Três Marias Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.

Table 7 .
Analysis of parasite indexes under the possible influence of the collection period (dry and rainy seasons) of Cichla kelberiKullander  & Ferreira, 2006, Três Marias Reservoir, upper São Francisco River, State of Minas Gerais, Brazil.