Helminth fauna parasitizing Pimelodus pohli (Actinopterygii: Pimelodidae) from the upper São Francisco River, Brazil

The parasite fauna of catfish, Pimelodus pohli , from the São Francisco River Basin is presented. A total of 45 catfish from the upper São Francisco River (45°15’44”W 18°13’25”S), were examined from July 2009 to September 2011. Forty-three catfish (95.5%) were infected by at least one parasite species, with 885 parasite specimens being found, distributed across 17 species: Monogenea ( Demidospermus uncusvalidus , Pavanelliella pavanellii , and Scleroductus sp.); Eucestoda (plerocercoids of Proteocephalidea); Digenea (metacercariae of Austrodiplostomum compactum , adults of Auriculostoma platense and Kalipharynx sp., and juvenile of Prosthenhystera obesa ); Nematoda (larvae of Contracaecum sp., Hysterothylacium sp., Procamallanus pimelodus , Procamallanus sp., and unidentified of Cucullanidae, and adults of Cucullanus caballeroi, Philometra sp., and Procamallanus freitasi ); and Acanthocephala (adults of Neoechinorhynchus pimelodi ). Procamallanus freitasi and Scleroductus sp. were the taxa with the highest prevalence. Demidospermus uncusvalidus , P . freitasi , and Scleroductus sp . were the dominant species. The host’s sex did not influence parasitic indexes; however, the total length of the catfish did appear to have some influence. The parasites, with except for P . obesa , were registered for the first time in P . pohli, as well as the occurrence of Kalipharynx sp. and C . caballeroi among pimelodid hosts from São Francisco River and South America.


Introduction
Pimelodus pohli Ribeiro & Lucena, 2006, a fish species popularly known as white catfish, or "mandi-branco", is found at several localities along the São Francisco River Basin (RIBEIRO;LUCENA, 2006).According to these authors, P. pohli is a species identified by Britski et al. (1988) as Pimelodus sp..As was recently described, studies referring to its biological, behavioral, and parasitological aspects are unknown, and a comparison of its fauna with that of Pimelodus maculatus Lacepède, 1803 (congeneric of the same hydric system) and other pimelodids, for which its parasite faunas are already known, is urgent.Observations of the gastric content of P. pohli in the present study have revealed the use of aquatic invertebrates (insects, small arachnids, and crustaceans) and sediments as elements of its diet.Taking into account the study on the feeding habits of fish in the upper São Francisco River, as conducted by Alvim (1999), P. pohli may be considered a generalistic invertivorous and iliophagous.Thus, the endohelmintic fauna recorded in that study must reflect these feeding habits.
In this study, the helminth fauna of P. pohli is presented for the first time in the São Francisco River Basin, contributing to the knowledge on the parasite species of fish in the São Francisco River, and on the parasite species of pimelodids in the hydrographic basins in South America.

Materials and Methods
A total of 45 specimens of P. pohli were collected from July 2009 to September 2011 in the upper São Francisco River, downstream from the Três Marias Reservoir (45°15'44"W, 18°13'25"S) in the municipality of Três Marias, State of Minas Gerais, Brazil.The fish were captured by local fishers and were sent to the Centro Integrado de Recursos Pesqueiros e Aquicultura (CIRPA) of the Companhia de Desenvolvimento dos Vales do São Francisco e do Parnaíba (CODEVASF), where they were weighted (average body weight: 26.6 ± 11 g), measured (average total length: 14.5 ± 1.6 cm), and necropsied to investigate their parasite fauna.Voucher specimens of P. pohli were deposited in the Coleção de Peixes of the Museu de Zoologia of the Universidade de São Paulo (MZUSP: 105895), State of São Paulo, Brazil.Parasite voucher specimens were deposited in the Coleção Helmintológica of the Instituto Oswaldo Cruz (CHIOC), Rio de Janeiro, Brazil, under the numbers indicated in Table 1 of the Results section.
Ecological descriptors (prevalence, mean abundance, and mean intensity) were calculated according to Bush et al. (1997).The dispersion index was calculated for each species in order to determine the distribution standard of the species; it was tested by calculating the d statistic (LUDWIG; REYNOLDS, 1988).The interaction between couples of species co-occurring within the hosts was evaluated via the chi-square test (χ²) with Yates's correction and a 2 × 2 contingency table, and by Spearman's rank correlation coefficient (rs).Student's t-test was used to evaluate the total length among the male and female hosts.Spearman's rank correlation coefficient (rs) was used to verify possible correlations between the total length of hosts and parasite abundance and intensity.Pearson's correlation coefficient (r) was used to determine possible correlations between the total length of the hosts and parasite prevalence.The influence of the host's sex on parasite abundance and prevalence was tested using the normal approximation of Z via the Mann-Whitney U test and the chi-square test (χ²) with Yates's correction and a 2 × 2 contingency table, respectively.The statistical analysis was performed according to Zar (1996), and was applied to the study of parasite infracommunities among P. pohli, but only for species with a prevalence rate higher than 10% (BUSH et al., 1990).The adopted significance level was p < 0.05.

Results
Among the 45 examined specimens of P. pohli, 18 (40%) were male and 27 (60%) were female (representing a ratio of 1:1.5).Male and female catfish presented with a mean total length of 14.3 ± 1.4 cm and 14.7 ± 1.7 cm, respectively, and a mean body weight of 25.6 ± 9.8 g and 28.2 ± 13.2 g, respectively.There was no significant difference in the total length (t = 0.84; p = 0.20) and weight (t = 0.75; p = 0.45) between male and female hosts.
Nematoda and Monogenea were the taxa whose prevalence values were higher, representing 82% and 75%, respectively, of the total fauna.
In Nematoda, the adult specimens of Procamallanus freitasi (Moreira, Oliveira & Costa, 1991) presented with a higher prevalence rate, mean abundance, and mean intensity, followed by Procamallanus (Spirocamallanus) sp. and Contracaecum sp., respectively, which consisted of specimens at the larval stage.In Monogenea, Scleroductus sp.presented with the highest prevalence rate, and Demidospermus uncusvalidus Gutiérrez & Suriano, 1992, was the most abundant species with the highest mean intensity (Table 1).
The parasite community of P. pohli demonstrated the standard aggregated distribution pattern, except for the uniform distribution of Neoechinorhynchus pimelodi Brasil-Sato & Pavanelli, 1998.The dominance frequency and mean relative dominance of the parasite community were most marked in D. uncusvalidus, followed by P. freitasi and Scleroductus sp..The frequency of shared dominance was highest in P. freitasi and Scleroductus sp., respectively (Table 2).
The total length of the hosts negatively influenced the parasitic indexes of the parasite fauna of P. pohli -that is, the intensity and abundance of D. uncusvalidus, the abundance of Contracaecum sp., and the prevalence of P. freitasi, were highest in fish with a smaller total length (Table 3).

Discussion
Nematodes, both at their larval stage and as adults, are common parasites of freshwater fish, and they can attack all organs in the body; many of them are known to be disease agents in fish and humans (MORAVEC, 1998).Nematoda was the prevailing taxon in several studies on the helminth fauna of P. maculatus.In Brazil, Kohn and Fernandes (1987) 2003).
In addition to the ecological aspects that involve the use of alimentary items that predispose congeneric hosts to the same types of parasites, and which lead to the parasite-host specificity in several hydrographic basins listed in this study, it is equally significant to add that P. freitasi indexes were significantly higher in smaller specimens of P. pohli -an endemic species to the São Francisco River.
Procamallanus spp.perforate the intestinal wall of hosts with their buccal capsule and feed on blood; these nematodes are commonly found upon necropsy, and are red or dark brown in color due to such hematophagous activity (BRASIL-SATO, 2003).They can cause inflammatory reactions at the place where they adhere to, and they may cause anemia in the host due to blood loss.They can occupy the intestinal lumen in small or juvenile fish and, if they do not cause death under such conditions, they can affect growth (THATCHER, 2006).Besides the availability of the intermediate host (crustaceans and insects), which justifies the constant record of C. pinnai in Pimelodus spp., Albuquerque et al. (2008) have also suggested the possibility of a histotrophic phase in C. pinnai cycle.As most Brazilian rivers are related to reservoirs with hydroelectric potential, the abundance of nematodes may be enhanced, i.e. the parasitic transmission can be faster and/or more efficient among the hosts kept in smaller areas (e.g., due to fragmentation of the rivers and also due changing the flow of the rivers).Host abundance is favored in lentic environments, such as in lakes and dams (MORAVEC, 1998), and thus high indexes of infection by nematodes are found in the fish species that dwell in areas close to the reservoirs of hydroelectric power stations (FELTRAN et al., 2004).Therefore, all of these aspects must be considered in proposals that suggest raising Pimelodus spp.under confined conditions, and a lot of attention must be paid to parasitism in endemic species, such as P. pohli (endemic to the São Francisco River), when dealing with biodiversity preservation in aquatic systems.
The development stage and the parasitic indexes of P. freitasi and C. caballeroi have suggested that P. pohli serves as the definitive host for those nematodes, and as an intermediate host for the larvae of Contracaecum sp. and Hysterothylacium sp., which will become adult specimens in birds, according to Moravec (1998).Santos et al. (2007) and Gutiérrez and Martorelli (1999) recorded high prevalence values for D. uncusvalidus and low prevalence rates for Scleroductus sp. in P. maculatus in the Guandu River, Brazil and in the la Plata River, Argentina, respectively.In the present study, the prevalence of Scleroductus sp. and the prevalence and mean intensity of D. uncusvalidus were high.In the upper São Francisco River, these two direct-transmission ectoparasite species (THATCHER, 2006) are the ones that characterized the ectoparasite community of P. pohli (mainly in fish with smaller body sizes), and the high values of their parasitic indexes optimized the observed co-occurrence.
The biometric data of siluriform hosts have been analyzed and were related to the parasitic indexes in several works (MACHADO et al., 1994;BRASIL-SATO;PAVANELLI, 2004;BRASIL-SATO;SANTOS, 2005;MOREIRA et al., 2005;SANTOS;BRASIL-SATO, 2006;BACHMANN et al., 2007).In the São Francisco River, parasitic indexes have been minimally influenced by fish size, and the possible cumulative effect of parasites, as proposed by Rohde (1993), has not been observed in parasite communities of fish in that hydric system.In this study, smaller fish presented with a higher abundance of ectoparasites (D. uncusvalidus) and, except for A. compactum, endohelminths also had their abundance favored, with the number of larval specimens of Contracaecum sp. and adults of P. freitasi being significantly higher in smaller P. pohli (Table 4).For camallanid nematodes, it is possible that the smaller fish presented with predilection for some alimentary items such as copepods, which were indicated as intermediate hosts for these parasites by Moravec (1998).
The sex of siluriform hosts has not been a determining factor in the structure of fish in the São Francisco River.In the present study, and in the studies conducted by Brasil-Sato and Santos (2005) and Santos andBrasil-Sato (2004, 2006), in the São Francisco River, there was no influence of host sex on the parasitic indexes, and this fact was supported by the similarities in feeding strategies and habits observed between male and female hosts.
Sympatric or related hosts must be similar in composition to their parasite faunas due to the level of similarity in their biological and behavioral features.In the São Francisco River, Pinto et al. (1974), Pinto andNoronha (1976), andVicente et al. (1985) reported on the larvae of Procamallanus (Spirocamallanus) sp. in Conorhynchos conirostris, and Moreira et al. (1991) and Vicente and Pinto (1999) 1928, Sanguinicola coelomica (Szidat, 1951) Gibson and Bray, 2002 (cited as the original combination: Plehniella coelomica Szidat, The significance of the analysis of parasitic indexes among host species with a wide distribution in the South American hydrographic basins, and among congeneric and sympatric species in this region, is highlighted to detect the main groups of parasite species that are able to affect fish.This type of predictive analysis indicates the relative specificity of parasite-host relationships in the hydrographic basins.The parasitism analysis of P. pohli highlighted that the smaller specimens were significantly more infested by Monogenea and infected by Nematoda; P. pohli is an endemic pimelodid host, whose niche and immunological plasticity is related to the available habitats that is restricted to the São Francisco River in relation to congeneric species or other pimelodids that present with a wide geographic distribution in the South and Central Americas.It was reinforced that, among the endoparasites, cucullanid and camallanid nematodes are the parasite representatives that prevail in the parasite fauna of the Pimelodus species.For P. pohli, the presence and parasitic indexes of those nematodes (as well as the endoparasite community) prove its invertivorous alimentary habits, which were not previously known.
In this study, the parasite species of P. pohli were presented, as well as those that are common in the fauna of P. maculatus in the São Francisco River and other basins.With the exception of P. obesa, which was recorded in P. pohli by Martins et al. (2012), this study revealed the parasite community of P. pohli and highlighted the first record of Kalipharynx sp. and C. caballeroi in the parasite community of fish in the upper São Francisco River, and in the pimelodids from South America.

Table 1 .
Ribeiro & Lucena, 2006(prevalence, mean intensity, intensity amplitude, mean abundance), and site of infestation/infection of helminth parasites (the developmental stages of the parasite specimens, except for adults, were indicated in parentheses following the nominal taxa) of Pimelodus pohliRibeiro & Lucena, 2006from the upper São Francisco River, State of Minas Gerais, Brazil.

Table 4 .
VenancFERNANDES, 1987)08)pecies of helminth parasites of Pimelodus pohliRibeiro & Lucena, 2006from the upper São Francisco River, State of Minas Gerais, BrazilAlbuquerque et al. (2008)in the Guandu River, State of Rio de Janeiro;Venancio et al. (2010)found that C. pinnai was the sole helminth in the fauna, presenting with a high prevalence rate in this host in the Paraíba do Sul River, State of Rio de Janeiro.In addition to the occurrence of these nematodes in high indexes in several hydrographic basins, Cucullanus spp.[especially C. pinnai (cucullanids) and Procamallanus spp.(camallanids)] have been constantly presented in the parasite fauna of P. maculatus.In this study, P. freitasi was the species that presented the highest prevalence among the endoparasites of P. pohli, with remarkable record of larvae of Procamallanus sp. and P. pimelodus, as well as adults of Cucullanus caballeroiPetter, 1977;this pattern was relevant, as it showed the same structural composition of the representative cucullanid and camallanid species in the helminth fauna of congeneric pimelodid hosts.This parasite-host relationship had been cited for pimelodid hosts in the Mogi Guaçu(KOHN;FERNANDES, 1987), Paraná (PAVANELLI et al., 1997), and  São Francisco Rivers (BRASIL-SATO, 2003)when, in addition to the Eucestoda larvae, Nematoda was highlighted as a common parasite that was represented by the larvae of the species allotted to Anisakidae, the species of Procamallanus Baylis, 1923, Cucullanus Müller, 1777 and, among the latter, C. pinnai (BRASIL-SATO, (Chi-square test (χ²) with Yates's correction; Spearman's rank (rs) correlation coefficient.*significantvalues: p < 0.05).