An ecological and comparative analysis of parasites in juvenile Mugil liza (Pisces, Mugilidae) from two sites in Samborombón bay, Argentina

Mugil liza Valenciennes, 1836 is an economically important food fish and has been recommended for aquaculture in South America. A total of 278 fishes were collected in the spring and summer of 2009 and 2010. These fish were sorted into sample groups according to their size class. We used Bayesian statistics and 95% credible intervals for each parameter tested were calculated. Fish studied harbored a total of 15 different species of parasites. Diversity of parasite species found on Mugil liza was greatest at the S.R.C. collection site, but evidenced a lower species richness than at A.R. site. The 1st size fishes of both sites evidenced greater parasite diversity than either 2nd or 3rd size fish. Differences observed could be explained by the different use of habitat types at the two sites or differential susceptibility to infection by parasites. The dominance of D. fastigatainfluenced observed results of lower community diversity indexes. New works elucidating different parasite life cycles within juvenile and adults ofM. liza in Argentina, promise to be important for determining the risk of the parasitism by zoonotic metacercariae A. (P.) longa and use of this fish as food and an economic resource, and the possible use of mullet parasites in other promising fields as indicators of biodiversity, and/ or water contamination.

The Estuary of the Rio de la Plata is the second largest hydrologic basin of South America (Urien, 1967;Framiñan & Brown, 1996) and the most important within Argentina. Samborombón Bay, found within the estuary of the Rio de la Plata, was declared a RAMSAR site since 1997 (Ramsar Convention Bureau, 2015). Within the Bay of Samborombón it is evident that a north-south gradient in the concentrations of the nutrients, phosphorus, nitrogen exists (Schenone et al., 2008) due to the northern lotic systems receiving nutrients from the high basin where extensive agricultural and livestock activity has developed (Fernandez Cirelli et al., 2006). In the south of the Bay the oceanic influence is higher than in the north for two reasons, first, the proximity of the Ajó River to the sea and the great distance from the discharge of fresh water from the Parana and Uruguay Rivers into La Plata River (Schenone et al., 2007(Schenone et al., , 2008. Mugil liza Valenciennes, 1836, an economically important food fish has been recommended for aquaculture in South America (Godinho et al., 1988;OIA, 2007), is the only mullet that breeds in Argentina (Gonzalez Castro et al., 2011) Many works have been published about lebranche mullet parasites (Chieffi, 1990;Chieffi et al., 1992;Knoff & Amato, 1992;Knoff & Boeger, 1994;Amado & Rocha,1995;Martinez Okumura et al., 1999;Scholz, 1999;Suriano et al., 2000;Fernandes & Cohen, 2006;Aparecida de Oliveira et al., 2007;Abdallah et al., 2009;Failla Siquier & Ostrowski de Nuñez, 2009;Simões et al., 2010;Montes et al., 2013), but ecological papers have focused mainly on adult fish (Knoff et al., 1997;Ranzani-Paiva & Silva Souza, 2004;Alarcos & Etchegoin, 2010), only Carnevia & Speranza (2003) working with juvenile from Uruguay reported some population indices.
The main objective of this work is to compare the metazoan parasite populations and parasite communities, at the component and infracommunity levels, of juvenile lebranche mullet in two sites within Samborombón Bay and analyze the distribution of the zoonotic metacercariae of A. (P.) longa, recently reported by Martorelli et al. (2012), in juvenile mullet of different sizes.

MATERIAL AND METHODS
The samples were collected in two sites of the Samborombón Bay, one in the north of the bay (Salado River relief Channel -S.R.C., 35°50'10"S, 57°50'20"W) and other in the south (Ajó River -A.R., 36°20'12"S, 56°54'17"W). Both localities were clearly defined and described by Schenone et al. (2007Schenone et al. ( , 2008. Juvenile lebranche mullet were sampled in the spring and summer of 2009 and 2010 and collected from the coastal region using cast-nets, and fixed nets. The mullet sampled were divided into study groups (Tab. I) following the sizes established by Gonzalez Castro et al. (2011) and, as such, size corresponds to the age in the following manner; size 1 fish were (age 0-1 year), size 2 (age 2 yr) and size 3 (age 3 yr). The mean and 95% credibility interval for the weight (in grams), total length (in cm) and standard length, (in cm), were calculated using Bayesian statistics (Tab. II).
The samples for ecological studies were fixed in 10% buffered formalin. In the laboratory fish were examined for parasites following the protocol of Marcogliese (2007). All parasites were studied according to methods of Pritchard & Kruse (1982). The parasites studied were deposited in the helminthological (MLP) and in the crustacean collection of Museo de La Plata, Argentina (MLP-Cr).
Ecological terminology follows Bush et al. (1997). According with Magurran (1998) were calculated at the infracommunity level the species richness (S), number of parasites, Shannon-Wiener diversity index (H), Pielou evenness index (E), the Berger Parker dominance index (D), the dominant specie, the percent of uninfected fishes, and the percent of parasited fishes, and to the community component the prevalence of infection, the number of parasites (Ncc) , the species richness (Scc), the Shannon-Wiener diversity index (Hcc), Pielou evenness index (E) , and complement of Simpson dominance index (Simpson).
The Shannon-Wiener index of diversity was calculated using base 10 logarithms. All parasites species were considered for the calculation of the community index.
The Shannon-Wiener diversity index, Pielou evenness index and complement of Simpson dominance index were calculated with the codes proposed by Golicher et al. (2006). It was calculated the DIC (deviance information criterion) for both codes (LogNorm and Gamma) being the latter smaller (233,32 vs 263,77 in S1, 228,61 vs 246,69 in S2) and used for the component community calculations.
WinBUGS was used to generate 100,000 samples from the posterior distributions for each of the analyses after discarding the initial 10,000 samples as a 'burn in'. The mean and the 2,5th and 97,5th percentiles of the distribution of each parameters was calculated with the statistic programs Epidat 4.0 and WinBUGS software (http://www.sergas.es) and (http://www2.mrc-bsu.cam. ac.uk/bugs/winbugs).This interval was used to represent a 95% Bayesian credible interval. The first year sampled was used as "prior" of the second year. A significance level (a) of 5 % or less was considered significant (P <= 0.05).

Population parameters
Prevalence. When analyzed, the size 1 (S1) mullets from both sites, evidenced a significant higher prevalence of D. fastigata, A (P.) longa and Parabrachiella sp. 1 in A.R. For size 2 (S2) mullets, the digenean H. manteri evidenced a significant higher prevalence in the S.R.C. samples. In A.R. samples the prevalence of D. fastigata, and Ligophorus sp. were significant bigger in the size 1 (S1) compared with S2 fishes, but within size 3 (S3) fishes prevalence was not statistically significant different to both sizes (S1 and S2). The copepod Parabrachiella sp. 1 was more prevalent in S1 fish, M. macracantha was more prevalent in S2 fish and F. mugilis was more prevalent in S2 and S3 than in S1 fish at the A. R. sample site. In the Salado River relief Channel (S.R.C.) the prevalence of H. manteri, A. (P.) longa and F. mugilis were significant higher in S2 than in S1 fish.
Mean intensity. When analyzed the S1 from both sites D. fastigata evidenced significant higher mean intensity of parasite species observed in the A.R. samples. In S2 class juvenile mullets from S.R.C. the mean intensity of both H. manteri and Ligophorus sp. were significantly greater than observed in A.R. samples. In A.R. the mean intensity of Ligophorus sp. was significant higher in S3 fishes compared with S1 and S2 size classes, while intensity within S1 and S2 classes were not statistically significant different. The mean intensity of Ascocotyle (P.) longa was not statistically significant different in S1 and S2 size classes, but significant lower for S3 than S1 size class. In the S.R.C. there is no difference observed between sizes.
Mean abundance. When analyzed the S1 fishes from both sites D. fastigata and Parabrachiella sp. 1 were significant greater mean abundance in the A.R. samples. In the S.R.C. samples, H. manteri and Ligophorus sp. evidenced significant higher mean abundance. The comparison of the S2 size classes of the two sample sites, indicated a significant higher mean abundance of H. manteri was observed in S.R.C. than in the A. R. In A.R. samples, only Parabrachiella sp. 1 was significant more abundant in the S1 size fish than o S2. In S.R.C. samples F. mugilis was significant more abundant in S2 size fish than in S1 fish.
Infracommunity level. The Table IV reflects the community indexes for both sites, S.R.C and A.R. The most dominant parasite species observed in all the sizes of juvenile mullet samples at both sites was D. fastigata. The S and the Shannon-Wiener diversity index were similar in both sites in all the sizes. The number of parasites sowed no meaning differences inside the S1 of mullets from both sizes. In A.R. the S1 have more parasites than the others sizes.
Evenness was bigger and significant in S1 and S2 from S.R.C than the A.R. In this latter site, the evenness of S1 was bigger and significant than S2, and this last two were not significant different than S3. The S1 from S.R.C. have a bigger and significant percent's of uninfected fishes than same sizes in A.R.
In A.R. (Tab. V) the fishes of S1 and S2 were infected with three and the S3 only with two parasite species. In S.R.C, in contrast with A.R., some fishes had 6-7 parasite species. In this site the fishes were parasitized more frequently with two parasite species.
The correlation (Tab. VI) between LT of mullets and S was negative in the S1 from A.R. and positive in S.R.C. The correlation with the N was positive in S1 and S2 from A.R. and S1 from the S.R.C. The correlation with the Shannon-Wiener diversity index was negative in S1 and S2 from A.R. but both sizes in the S.R.C were positive. Component community. The Berger Parker dominance index is plotted in Figs 1-3. In the S1 from R.A. the species with significant higher dominance after D. fastigata were Ligophorus sp., A. (P.) longa and Parabrachiella sp. In the S1 class samples of S.R.C. the The parasite samples of S1 size fish of S.R.C. were compared with S2 size there and found to evidence significantly great species richness (Scc), Shannon-Wiener diversity index (Shannon) and complementary of Simpson equitability index (Simpson). In the A.R. the S1 size class was compared with S2 and S3 in the same site and found the S1 fish had the highest values of species richness, number of parasites, and Shannon and Simpson indices. The S2 had a significant great species richness of parasites than S3 fish sampled in the same site.
When comparing parasite loads between the same fish sizes at the two sites, found the S1 class samples of S.R.C. had a significant higher prevalence (Pcc), Shannon-Wiener and Simpson index, but evidenced significant lower counts of parasites and significant lower species richness than the S1 from A.R. The S2 sample class of S.R.C. indicated greater Shannon and Simpson indices than S2 of A.R. The prevalence of parasites in both sites was not significant different than in S2 samples, but the species richness was greater for parasites in S2 samples of A.R.

DISCUSSION
The high number of parasite species found in comparison with past studies may be explained by the fact that here were analyzed juvenile mullets that could be more susceptible to infection by parasites, or alternately, there may have been other factors such as behavior differences, lower immunological competence, or environmental changes (Khan, 2012 There was evidence that reported differences between the environments of the two sites (Schenone et al., 2007(Schenone et al., , 2008 studied may affect and/or alter the parasite fauna present in each. Schenone et al. (2007Schenone et al. ( , 2008 clustered the sites in the north end of the Bay based in similar pH values, dissolved oxygen, high concentrations of nutrients, phosphorus and nitrogen and also containing the same kinds of contaminants, such as lead, zinc, copper and arsenic. In the south of the bay, the group of streams and channels present share similar values of pH, concentration of oxygen and conductivity due to being near the ocean and lack many of the contaminants and agricultural sourced nutrients as a factor of being far away from the discharge of Parana and Uruguay Rivers into the La Plata River. In A. R. the saline waters allow the development of parasites normally found in adults mullets (P.   Tab. IV. Community parameters of each lebranche mullet size in both environments, Samborombón bay, Argentina (A.R., Ajó River; Evenness, Pielou evenness index; Ncc, number of parasite; Pcc, prevalence of parasitic infection; S, species richness; S1, size 1; S2, size 2; S3, size 3; Scc, species richness of the community component; Shannon, Shannon-Wiener diversity index; Simpson, complement of Simpson dominance index; S.R.C., Salado River relief Channel). S1
In the S. R. C. sample site found H. manteri and Ligophorus sp. were more abundant than observed in A.R. samples, perhaps as result of fresh water or physicalchemical environmental conditions. The six individuals of S3 size class from A.R. in our sample were in poorly preserved condition. More fishes of this size must be examined to be positive of observed patterns being valid, but even with these preservation problems was observed a pattern of an increase of F. mugilis present as mullets parasitized got to be older/bigger. That pattern was also observed across the two sizes of mullets in S.R.C. samples.
The A.R. samples evidenced differences between juvenile fish size classes in the prevalence and abundances of some parasites (D. fastigata, Ligophorus sp. and Parabrachiella sp.) perhaps reflecting different use of the habitat, different habitats, or different susceptibility to the parasites. In the S.R.C. study area the habitat frequented by the young mullets appeared to be more homogenous, because there was not a great difference in the parasite species present across mullet age classes, and, only a moderate increase in the presence of H. manteri and F. mugilis observed in the S2 age class mullets.
A special analysis must be done to identify parasites of Ligophorus species. They are very small monogenans and determine them is complex because species only differ in characteristics of the vagina or penis. Thus, differences of Ligophorus species found within and between both sample sites (and fish sizes in each) may obscure different species identification, including some which could be new to science.
The increase in the prevalence and mean abundance of monogenean in the higher sizes could be explain by a grown of the fish or a bigger encounter of mullets facilitating the transmission of this parasites.
In Argentina the metacercariae of A. (P.) longa was not reported in adult mullets, but have been mentioned parasitizing juvenile mullets by Carnevia & Speranza (2003) and Lado et al. (2013) from Uruguayan coast of La Plata River and ), Martorelli et al. (2012 and Montes et al. (2013), in Argentinean waters. In this work we increased the distribution between different sizes of mullets analyzed by Martorelli et al. (2012). In both sites studied, the prevalence of A (P) longa increased as the mullets parasitized got bigger, and the mean abundance and mean intensity remained similar within each size for the same site. The mullets from A.R. and the S2 from the S.R.C. had a greater prevalence of parasitism by A. (P.) longa because they seem to live in waters with a higher salt concentration (or marine influence). That environment is suitable for the development of A. (P.) longa or the intermediate host (Helobia australis d'Orbigny, 1835). The absence of this parasite in the mullets analyzed by Alarcos & Etchegoin (2010) could be explained by the type of life cycle of the parasite, more vulnerability of the juveniles, sampling errors or environmental condition in Mar Chiquita lagoon.
The great numerical dominance played by D. fastigata reduced values of community indexes at the S. R. C. study site and more in A.R. sample even when species richness was greater than in S.R.C. samples. The digenean may be filling the habitat for other intestinal parasites being a strong competitor and perhaps displacing other species. Dicrogaster fastigata may be used as sentinel for controlling ambient changes. The high presence in A.R of this parasite agrees with Lado et al. (2013) who found a high prevalence of this parasite in mullets from saline waters of Uruguay. In neither site of this study did we find the snail species that serves as host of the immature stages of this parasite.
New works about juvenile (and mainly adults) of M. liza in the Samborombón bay as in other localities of Argentina, help provide elucidation of how the different life histories of the various parasites are be important in understanding the risk of finding zoonotic metacercariae of A. (P.) longa in local mullet populations. They also may help find ways of using mullet parasites in other promising fields, such as functioning as indicators of biodiversity, environmental contamination, and expanded use of this fish as food and as an economic resource for local export.