Parasites of two coexisting invasive sailfin catfishes (Siluriformes: Loricariidae) in a tropical region of Mexico

Currently many species of Amazon sailfin catfishes (Loricariidae) have been introduced to wild environments outside their native range. There is, however, little knowledge about their role as vectors of parasites that can infect native fish or even humans through its consumption. The aim of the present study was to determine the parasitic fauna of the invasive sailfin catfish species Pterygoplichthys pardalis (leopard pleco) and P. disjunctivus (vermiculated pleco) from freshwater systems in the southeast of Mexico. Four ectoparasite species were found in P. pardalis (1 protozoan: Ichthyophthirius multifiliis; 2 monogeneans: Urocleidoides vaginoclastrum and Heteropriapulus heterotylus; 1 digenean: Clinostomum sp.), and only one in Heteropriapulus disjunctivus (H. heterotylus). No endoparasites were found. Ichthyophthirius multifiliis, U. vaginoclaustrum and Clinostomum sp., were considered as rare species (prevalence <5%) since they were found in a single individual of P. pardalis. H. heterotylus was the only species shared among both host species and it occurs throughout the year. This monogenean species represents 96% of total parasites recorded in P. pardalis and 100% in P. disjunctivus. Monthly values of prevalence, intensity and abundance of H. heterotylus in both host species showed important intra-annual variations, but not differ significantly between both hosts.


Introduction
Introduction of non-native fish species threatens native biodiversity and the ecosystem function of aquatic communities globally (Kelly et al., 2009;Cucherousset & Olden, 2011).Many non-native fish species may become predators, competitors (Hill & Lodge, 1999), or even spread parasites or pathogens to native species (Torchin et al., 2003;Torchin & Mitchell, 2004).The latter has been the subject of concern for many decades, since it is well known that parasites transmitted by non-native species can cause severe impacts on native species (Bakke et al., 1990;Poulin et al., 2011).
The first report of a wild sailfin catfish in Mexico was made by Guzmán & Barragán (1997) in the Mezcala River, Guerrero.More recently, Wakida-Kusunoki et al. (2007) reported the presence of Pterygoplichthys disjunctivus (Weber, 1991) and P. pardalis (Castelnau, 1855) in freshwater ecosystems of southeastern Mexico.Currently, these species are listed as one of the most invasive species in Mexico (Mendoza-Alfaro et al., 2011) and continue spreading and proliferating massively throughout rivers, streams, reservoirs and lakes of several parts of the country.The main problem caused by these species is the competition for space and food with native fish species, including those commercially important (Martínez-Palacios et al., 2009;Sandoval-Huerta et al., 2012).However, nowadays there is little information available about their parasitic fauna (Mendoza-Franco et al., 2012) and the role of these parasite species as potential pathogens to native fish species or humans (Pound et al., 2011).
Therefore, the aims of this study were (1) to determine the parasitic fauna of the invasive sailfin catfish species: P. pardalis and P. disjunctivus from riverine systems of southeastern Mexico, (2) to compare the population parameters of infection (prevalence, intensity and abundance) among these coexisting fish hosts and (3) to determine whether the intra-annual variability in the prevalence, mean intensity and mean abundance of parasites in each fish species is related to seasonal changes in water temperature and/or to the host size and weight.
Fish were caught using a casting net (mesh = 76 mm) at depths between 0.7 m to 2.0 m depth (from January to December 2012) and transported alive (individually) to the laboratory in plastic containers.At each sampling site, the bottom water temperature was recorded by using a multiparameter sensor (YSI-30 Salinity, Conductivity Temperature, Ohio 45387 USA).During the study period (January-December 2012) the average water temperature ranged from 24ºC (in February) to 30.8ºC (in August) (Fig. 1b).Fish were sacrificed by decapitation to ensure a fast dead, which is according to Mexican laws (NOM-033-ZOO-1995).The total length (cm) and weight (g) were recorded for each fish.The parasitological survey included the revision of external (eyes, fins, skin, and gill cavities) and internal organs (intestinal caeca, coelomic cavity, stomach, intestine, spleen, gallbladder, liver, kidneys, and heart) and muscle.Internal organs were individually separated and subsequently examined by compression between two 10 x 10 x 1.0 cm flat glasses (Guzmán-Cornejo et al., 2012).Parasites were isolated and preserved in labelled vials with 96% alcohol.
Identification of monogeneans was based on the characteristics of the sclerotized hard parts of haptor and male copulatory organ.Haptors were subjected to a partial digestion using a proteinase K-base method following Harris et al. (1999) in order to study haptoral sclerites.Trematodes were stained with Gomori's trichrome or Mayer's carmine to determine the internal features.Skin and gills smears were made in search of protozoans, which were impregnated with silver nitrate (Martorelli et al., 2008).Voucher specimens of helminths were deposited at the Colección Nacional de Helmintos, Instituto de Biología, Universidad Nacional Autónoma de México (Mexico).
Data analyses.Prevalence, intensity and abundance were determined as recommended by Bush et al. (1997).Assumptions of normality and homocedasticity in each variable were previously tested by Kolmogorov-Smirnov's and Bartlett's tests, respectively (Sokal & Rohlf, 1995).To assess significant variations in the overall abundance of parasites over the year, a non-parametric Analysis of Variance of Kruskal-Wallis was used (Steel & Torrie, 1986).Chi-square tests were used to compare the prevalence and intensity between both host species (Rózsa et al., 2000).Moreover, a Mann-Whitney U-test was used to assess significant differences in the abundance of parasites between hosts (Zar, 1999).Spearman rank correlation analyses were used to assess significant relationships between the population descriptors of parasites (prevalence, intensity and abundance) and fish host size and weight, and water temperature.For these analyzes, the average monthly values were used.P-values less than 0.05 were considered significant.
Temporal changes in prevalence, intensity and abundance were analyzed only in the monogenean H. heterotylus since this was the only species present during the whole year in both fish species (Fig. 1).In P. pardalis, the highest proportion of hosts infected with this parasite was recorded in June (93%) and the lowest in March (24%), where as in P. disjunctivus the highest values were in April and November (100%) and did not show any parasites in December (Fig. 1a).Moreover, in P. pardalis, the intensity of infection by this monogenean was relatively higher between June and September (6.5 ± 1.0 and 8.9 ± 1.5, respectively) and lower in January (1.0 ± 0.0).
In P. disjunctivus, the intensity was also higher between June and September (4.5 ± 0.7 and 4.6 ± 0.4, respectively) (Fig. 1b).Similarly, the abundance of H. heterotylus is higher between June and September (3.5 ± 0.9 ind.host -1 and 4.7 ± 1.0 ind.host -1 ) and the lowest in January (0.2 ± 0.1 ind.host -1 ) (Fig. 1c).Likewise, it was found that the abundance of this parasite in P. pardalis and P. disjunctivus showed significant variations (Kruskal-Wallis ANOVA, p< 0.01) over the year and was positively correlated with water temperature (r s = 0.89, p< 0.01 and r s = 0.79, p< 0.05, respectively), which ranged from 24ºC (in February) to 30.8ºC (in August).A significant correlation was also found between the intensity of H. heterotylus (in both fish hosts) and water temperature (P.pardalis, r s = 0.91, p< 0.01; P. disjunctivus, r s = 0.89, p< 0.01).Moreover, no significant correlations were found between the host size, weight, and infection parameters (abundance, prevalence and intensity).
Out of the four species found in this study for P. pardalis, three of them (I.multifiliis, U. vaginoclastrum and Clinostomum sp.) had not been previously recorded, although their presence appears to be occasional, as they were found on a single individual.These parasites were likely acquired from native host species (Jogunoori et al., 2004).
Another interesting result of this study was the absence of gastrointestinal parasites in both invasive fish, which is likely because these introduced species have not yet acquired a parasitic fauna that could contribute to regulating their population growth in the invaded environment (Anderson, 1978).In this regard, other studies have also suggested that the low diversity of parasite species recorded in sailfin catfish species could be related to their distinctive, very thick shell that is filled with tiny thorns, which could be a barrier for some parasite species (Whittington et al., 2000).Apparently, the small number of years that these fish have been present in the new geographical range along with its feeding habits and behaviour (bentic algae/ weeds) seems to be a more probable explanation.
In the life cycle of trematodes of genus Clinostomum, their eggs are dispersed in the feces of fish-eating birds that usually are the definitive host.The egg gives rise to a miracidial covered with cilia that swim freely and infects a snail, resulting in a sporocyst that migrates to the intestine or liver of the host snail.Subsequently, through the food web this sporocyst is transmitted to a fish where it becomes metacercaria.The cycle is completed when a fish-eaten bird eats the fish (Olsen, 1986).In monogeneans, most species are oviparous.The ciliated larva that arises from the egg can swim and infect the only host of their life cycle.Only a few monogeneans are viviparous, among which highlight those belonging to the genus Gyrodactylus, in which until three generations can occur in a single host individual (Bakke et al., 2007).It is also known that this type of ectoparasites can be transmitted, even without direct contact whit an infected host, and some are so specific that when the host dies they are able to find a new host of the same species (Cable & Harris, 2002).Regarding I. multifiliis, free-swimming theront stage parasites invade the skin and gills of host fish and rapidly transform into feeding trophonts within the epidermis (Cassidy-Hanley et al., 2011).Mature trophonts exit the host tissue, giving rise to the tomont stage.Then, tomonts attach to an inert support, encyst within a gelatinous capsule, and rapidly divide to give rise to 100-1000 new infective theronts within 18-24 h at 22ºC (Cassidy-Hanley et al., 2011).
In this study, H. heterotylus was the most important parasite (in terms of prevalence and abundance) and the only one shared between both fish hosts.This parasite seems to be specific of loricarids (e.g.Nagasawa & Uyeno, 2012;Nitta & Nagasawa, 2013).The fact of not having found significant differences between them in the prevalence, intensity and abundance of this parasite between both hosts could be because these fish species share the same habitat, have similar behavior and food habits (Torchin et al., 2001).
Moreover, our data of prevalence, intensity and abundance of this monogenean (in both hosts) were consistent to those recorded in the species of Heteropriapulus recorded in P. pardalis by Porto et al. (2012).In the case that these species of Heteropriapulus were the same, our results could indicate that the population characteristics of this parasite have been maintained, even outside its native range.Although to demonstrate it molecular studies are required.
On the other hand, monthly data of prevalence, intensity and abundance of this monogenean showed a similar temporal variability in both invasive fish.Also, a significant correlation was found (in both hosts) between the intensity and abundance of H. heterotylus and the water temperature, suggesting that intra-annual variability of this parasite is likely influenced by this environmental parameter, as has been documented in other parasite species (Chubb, 1977;Scott & Nokes, 1984;Khidr, 2012).For example, a study conducted on Puntius spp.(Cyprinidae) from India revealed that abundance of monogeneans showed an increasing or decreasing trend with the rise and fall in temperature in general, although some monogenean species were more abundant during the monsoon and winter seasons (Shrivastava et al., 2012).Similarly, a previous study conducted within our study area revealed that the abundance of these loricariid species was higher from June to September and was positively correlated with the gonadosomatic index of these fish (Wakida-Kusunoki & Amador-del Ángel, 2011).Therefore, it could be possible that the season of higher abundance of parasites is also related to the reproductive stage of the host, since it is well known that some fish species are more susceptible to parasite infection in periods of higher reproductive investment (Šimková et al., 2005).
Most parasite species recorded in this study have not been documented to be pathogenic to humans, although it is known that a species of Clinostomum (C.complanatum) has been found to infect humans (two in Korea and one in Japan) (Witenberg, 1944).This parasite was transmitted by eating raw or undercooked fish, which is linked to cultural factors of certain human populations (Park et al., 2009), and was found attached to the human mucous layer of larynx and pharynx.However, although the prevalence of these trematodes was very low in these hosts (P.pardalis and P. disjunctivus), it was not recommended to consume undercooked meat fish.
In summary, this study documents the parasite composition of the Amazon sailfin catfish species P. pardalis and P. disjunctivus, which are known to be highly invasive in many regions worldwide.In general, the parasite species richness was relatively poor (three species in P. pardalis and one in P. disjunctivus).Through this study, P. pardalis is assigned as new host for Clinostomum sp., I. multifiliis, U. vaginoclastrum and H. heterotylus and the study area as a new geographic location for the four species of parasites.
Finally, this research highlights the importance of conducting samplings in different seasons of the year in order to make a better characterization of the parasite fauna and to determine the environmental factors influencing its population dynamics.

Fig. 1 .
Fig.1.Intra-annual variations in: a) prevalence, b) intensity of infection and c) abundance of parasites in the invasive loricarids Pterygoplichthys disjunctivus and P. pardalis from a freshwater ecosystem of southeastern Mexico.

Table 1 .
Infection parameters of the parasite from Pterygoplichthys pardalis and P. disjunctivus.