Abstract

ORIGINAL ARTICLE / ARTÍCULO ORIGINAL

Resumen

BIODIVERSITY OF FISH PARASITES FROM GUANDU RIVER, SOUTHEASTERN

BRAZIL: AN ECOLOGICAL APPROACH

BIODIVERSIDAD DE LOS PARÁSITOS DE PECES DE RÍO GUANDU, SURESTE DE

BRASIL: UNA APROXIMACIÓN ECOLÓGICA

1 1 2*

Rodney K. de Azevedo ; Vanessa D. Abdallah & José L. Luque

1 Curso de Pós-Graduação em Ciências Veterinárias, Universidade Federal Rural do Rio de Janeiro, Seropédica, Brasil.

2 Departamento de Parasitologia Animal, Universidade Federal Rural do Rio de Janeiro, Caixa Postal: 74.508, Seropédica,

Brazil, CEP: 23851-970. E-mail: jlluque@ufrrj.br

*Correspondence to author/ Autor para correspondencia: José L. Luque

Suggested citation: Azevedo, R.K., Abdallah, V.D. & Luque, J.L. 2011. Biodiversity of fish parasites from Guandu river,

Southeastern Brazil: an ecological approach. Neotropical Helminthology, vol 5, nº 2, pp.185-199.

Key words: Biodiversity- Brazil- ecology parasite- fishes- Guandu River- parasites.

En el presente trabajo, se realiza un análisis comparativo de las comunidades de parásitos de 21

especies de peces en el río Guandu, Brasil y se evaluó el efecto de algunas de las características del

huésped (tamaño corporal, comportamiento social, hábitat de los peces, categoría trófica y capacidad

de migrar) en la diversidad de sus comunidades de parásitos metazoarios. Como una medida de

diversidad, hemos utilizado la riqueza de especies de parásitos y la diversidad taxonómica. La riqueza

de especies de parásitos y la diversidad taxonómica no se vieron afectados por el número de los

espécimenes examinados por especie. El tamaño del cuerpo de los peces mostró una correlación

significativa con la riqueza de especies de parásitos, aunque no hay correlación con la diversidad

taxonómica de los parásitos. Las principales características asociadas a la diversidad taxonómica de los

parásitos son la formación de cardúmenes y la categoría trófica omnívora. Las comunidades de

parásitos de peces en el río Guandu presentaron comunidades parasitarias aisladas.

Palabras clave: Biodiversidad- Brasil- ecología parasitaria- parásitos- peces - río Guandu.

Neotrop. Helminthol., 5(2), 2011

2011 Asociación Peruana de Helmintología e Invertebrados Afines (APHIA)

Versión Impresa: ISSN 2218-6425 / Versión Electrónica: ISSN 1995-1043

185

Here, we performed a quantitative analysis of the parasite communities in 21 species of fish from

Guandu River, Brazil; we evaluated the effects of some host traits (body size, social behavior, fish's

habitat, trophic category and ability to migration) on the diversity of their communities of metazoan

parasites. To measure quantitative diversity, we used parasite species richness, as well as the average

taxonomic distinctness of the assemblage and its variance. The parasite species richness, the taxonomic

distinctness and the variance were unaffected by the number of host individuals examined per species.

Fish body length proved to be the main predictor of parasite species richness, although it did not

correlate with parasite taxonomic distinctiveness. The mains host features associated with the

taxonomic diversity of parasites were schooling behavior and omnivores trophic category. Parasite

communities found in fish from Guandu River isolationist communities.

INTRODUCTION

Parasite communities are playing an increasingly

important role as models for the study of

biodiversity and biogeography (Poulin & Morand,

2000). Given the integral roles played by parasites

in natural ecosystems, identifying hotspots of high

parasite diversity, as well as areas of relatively low

parasite diversity, is crucial for a complete

understanding of the functioning of the biosphere.

Currently, the biodiversity of marine and

freshwater ecosystems of Latin America is

threatened, mainly by environmental problems

resulting from the destruction and degradation of

the ecosystems. In this context, parasite

biodiversity can be very important because

parasitism plays key roles in ecosystems,

regulating the abundance or density of host

populations, stabilize food webs and structuring

animal communities (Luque & Poulin, 2007).

Takemoto et al. (2005) performed the first study

relating the different features of host species and

the parasite species richness in freshwater fishes

from Neotropical Region. In fish, for instance,

some previous studies have found that host body

size is a good predictor of parasite species richness

whereas others have found no effect of host size.

The same is true for a range of other host features. It

is therefore difficult to assess the relative

importance of different host traits for the evolution

of parasite diversity in general (Luque et al., 2004;

Luque & Poulin, 2008).

In the present study, we examine the relationship

between different features of host species and the

diversity of metazoan parasite communities across

species of fish hosts from Guandu River (Fig. 1) in

function of the strategic importance of this river,

which is the main source of potable water in Rio de

Janeiro, Brazil. The Guandu River supply water to

90% of population of City of Rio de Janeiro and

although be a very impacted environment (Bizerril

& Primo, 2001), it maintains an important level of

biodiversity of fishes, and consequently, fish

parasites (Azevedo et al., 2010). In addition, data

about population and community and quantitative

descriptors from the fish hosts are given herein.

Between April 2003 to September 2009 were

analyzed 786 specimens of fish, belonging to 21

species (Table 1) from the Guandu river, near the

dam of water treatment station (WTS) (22°48'32"S,

43°37'35"W). The taxonomy of the fishes follows

that of Reis et al. (2003). Parasites were collected

from the body surface, gills, and body cavities, and

viscera after examination under a stereoscopic

microscope. Washings from gills and gut lumen

were strained using a sieve (53 and 75 μm mesh

size) to retain even the smallest parasites.

Following Bush et al. (1997), prevalence, intensity

and abundance mean were calculated for parasites

of all fish species.

Pearson's correlation coefficient r was used to

analyze the possible correlation between the host's

total body length and the abundance of parasites,

with previous logarithmic transformation Log

(x+1) (Zar, 1999). The analysis included only

parasite species with prevalence greater than 10%

(Bush et al., 1990).

The following descriptors were calculated at the

parasites infracommunity level: total prevalence,

total intensity, total abundance, total species

richness, endoparasite and ectoparasite species

richness, Margalef's richness index (d), Brillouin's

diversity index (H) (log 10 based), Pielou's

evenness index (J`) and Berger-Parker dominance

index. In addition the Bray-Curtis similarity index

was calculated among infracommunities within

host fish species (Magurran, 2004). These

descriptors were used for all parasites with

exception of myxozoans. Statistical significance

level was established at P<0.05. All results were

presented in tabular form (Tables 2 and 3). The

relative abundance of metazoan parasites was

calculated at the level of infracommunities for all

species of fish that had more than three species of

parasites. The results were presented in graphical

form (Figs. 2 and 3).

For each fish species, the average taxonomic

distinctness (Δ+) and variance in taxonomic

distinctness (Λ+) of the parasite component

community were computed, following the

procedures and taxonomies used by Luque &

Poulin (2008). The effect of host length on

taxonomic distinctness, on variance and on total

richness were evaluated using the Pearson

correlation coefficient on logarithmic transformed

data. The Student`s t test or ANOVA test,

depending of the number of variables with previous

transformation, was used to verify the influence of

MATERIAL AND METHODS

186

Biodiversity of fish parasites from Guandu river

Azevedoet al.

the followings variables in the taxonomic

distinctness and variance in taxonomic

distinctness: (1) whether the fish species forms

schools or not, with species adopting schooling

only in some parts of the year (e.g. during the

reproductive period) classified as schooling; (2)

whether the fish's habitat is benthic, benthopelagic

or demersal; (3) the trophic category, where the fish

species were distributed into four categories:

detritivores, herbivores, omnivores or carnivores

and (4) whether species accomplish or not

migration, being classificated as potamodromous

or diadromous. Data were obtained from Fishbase

(Froese & Pauly, 2010). Table 4 showed the entire

data set of host species included in the analyses.

The present study follows the classification and

systematic arrangements used by Azevedo et al.

(2010). Parasite species names follow those

provided in the most recent taxonomic literature.

Species of fishes are arranged in alphabetical

sequence and valid names are adopted from Froese

& Pauly (2010).

Figure 1. Map of the Guandu River and the area of collection

(circle) near to the dam of water treatment station (WTS)

(22°48'32” S, 43°37'35” W).

Figure 2. Species richness of fish parasites according to

zoological group reported in the Guandu River, State of Rio de

Janeiro, between 2003 to 2009.

Figure 3. Distribution of parasite genera in fish's species

collected from Guandu River, State of Rio de Janeiro, between

2003 to 2009.

RESULTS

A total of 786 specimens of fish was analyzed and

were found 15.630 specimens of parasite

belonging to 81 species. Nine groups of metazoan

parasites were found: Acanthocephala, Cestoda,

Crustacea, Digenea, Hirudinea, Mollusca,

Monogenea Myxozoa and Nematoda. In these fish,

187

Neotrop. Helminthol., 5(2), 2011

Table 1. Prevalence, mean abundance, mean intensity, site of infection / infestation of metazoan parasites from fish collected in the Guandu river, State of Rio de

Janeiro, Brazil. (s = standard deviation, * parasites that had a positive correlation between abundance and its host's total length through Pearson's correlation

coefficient r).

188

Biodiversity of fish parasites from Guandu river

Azevedoet al.

Site of infection

(metacerraride)

(Table 1)

outer surface of

the bladder Eyes

189

Neotrop. Helminthol., 5(2), 2011

Site of infection

Digenea

(Table 1)

190

Biodiversity of fish parasites from Guandu river

Azevedoet al.

Digenea

Site of infection

(Table 1)

191

Neotrop. Helminthol., 5(2), 2011

Site of infection

(Table 1)

192

Biodiversity of fish parasites from Guandu river

Azevedoet al.

Site of infection

(Table 1)

193

Neotrop. Helminthol., 5(2), 2011

Site of infection

Nematoda

(Table 1)

194

Biodiversity of fish parasites from Guandu river

Azevedoet al.

Site of infection

a total of 69% were parasitized by at least one

species of metazoan parasite. The percentage of

parasitism was 55% for endoparasite and 45% for

ectoparasites.

The monogeneans had higher species richness at

component community level. Of all the parasites

found, the digenean had lower specificity, since the

metacercaria Austrodiplostomum compactum

(Lutz, 1928) was found parasitizing nine different

species of fish. Of all the species of parasites

collected only four showed positive correlation

between the hosts' total length and abundance: in

Geophagus brasiliensis (Quoy and Gaimard,

1824) the digenean Posthodiplostomum

macrocotyle Dubois, 1937 (r=-0,289; p=0,041), in

Hypostomus affinis (Steindachner, 1877) the

monogenean Hyperopletes malmbergi Boeger,

Kritsky and Belmont- Jégu, 1994 (r=0,394;

p=0,028), in Mylossoma aureum (Spix and

Agassiz, 1829) the monogenean Anacanthorus

paraspathulatus Kritsky, Boeger and van Every,

1992 (r=0,484; p=0,049) and in Pimelodus

maculatus Lacépède, 1803 the nematoda

Cucullanus (Cucullanus) pinnai pinnai Travassos,

Artigas and Pereira, 1928 (r=0,513; p=0,001).

Centropomus undecimalis (Bloch, 1792) showed

the highest mean parasite diversity (H)=0.57±0.42

and Margalef's richness index (d)=0.64±0.48. The

values found for the index of interactivity CC in

parasite communities in fishes from Gaundu river

were high, indicative of isolationist communities.

The host's total length was positively correlated

with richness (r=0.999, p=0.00), but not with either

Δ+ (r=-0.169; p=0.563) or Λ+ (r=0.03, p=0.917).

The richness was not significantly correlated with

Δ+ (r=-0.017, p =0.558) nor with Λ+ (r=0.032,

p=0.912), showing that species with high parasite

richness does not necessarily have high taxonomic

distinctness and variance. The number of fish

examined was not correlated with richness

(r=0.223, p =0.602), with taxonomic distinctness

(r=-0.119, p =0.454), or with the variance

(r=0.031, p =0.908), showing that these indexes are

independent of the sampling effort.

Of the categorical variables considered, only two

significant results were obtained: Δ+ varied

significantly between schooling and non-schooling

fish species (t=2.527, p=0.026) and among fish

species with different trophic category (detritivores

and omnivores) (t=2.905, p=0.033). Omnivores

and schooling fish species had higher values of Δ+

than other groups.

DISCUSSION

The results of this study indicate that the parasite

communities of fish in the Guandu River were

characterized by low parasite species richness and

evenness, by isolationist communities and by

greater values of taxonomic diversity in omnivores

and schooling fish species.

According to Kennedy (2009) all parasite species

have a niche selection to a greater or lesser degree,

but isolationist communities have species poor and

the species are independent of each other. By

contrast, interactive communities have highest

species richness. Communities could be located

anywhere along this continuum, and those of

freshwater fish tend to be found towards the

isolationist end.

(Table 1)

195

Neotrop. Helminthol., 5(2), 2011

Our results suggesting that host feeding habits and

the formation of schools may influence the

taxonomic diversity, since omnivores and

schooling fish species had higher parasite diversity

than other groups. According Luque et al. (2004)

many researches have found that schooling fish

species are used by more species of parasites than

solitary species, for both external parasites and all

parasites combined. The fact of omnivores fish

species present greater diversity can also be

explained, because the greater the variety of food,

the greater the intake of various intermediate hosts

and easier to contamination by parasites is acquired

via the food web.

A study by Takemoto et al. (2005) in the floodplain

of the upper Paraná River found that parasite

species richness in freshwater fish species was not

associated with several host characteristics, with

the exception of host population density. However,

in the study by Luque et al. (2004), the fish size

proved to be the main predictor of total parasite

species richness in marine fish. This result was also

found in this study. According to Luque et al.

(2004) following from island biogeographical

theory larger-bodied hosts should be able to

accommodate more parasite species than small

ones; they may also incur higher exposure to

internal parasites because of the quantities of food

they ingest, and to external parasites because their

larger surface area facilitates contact with infective

stages.

Considering both the approach and the results, the

present study includes both key improvements on

earlier studies of this kind and novel findings,

making its contribution particularly relevant. The

majority of earlier studies on the richness of

freshwater fish parasites have used data from fish

species that do not occur in the same geographical

areas, with exception to work of Takemoto et al.

(2005). The present study focused on a set of fish

species from the same general area (Guandu river),

thus minimizing any differences in parasite

availability. In addition, this study was the first to

incorporate the average taxonomic distinctness of

the assemblage and its variance as a measure of

taxonomic diversity in freshwater fish. For data, all

study in freshwater fish has used species richness

Rodney K. de Azevedo was supported by a student

fellowship from FAPERJ (Fundação Carlos

Chagas Filho de Amparo à Pesquisa do Estado do

Rio de Janeiro); Vanessa D. Abdallah was

supported by a student fellowship from CNPq

( C o n s e l h o N a c i o n a l d e P e s q u i s a e

Desenvolvimento Tecnológico, Brazil). José L.

Luque was supported by a Research fellowship

from CNPq and a grant from FAPERJ.

ACKNOWLEDGMENTS

study in freshwater fish has used species richness

as their sole measure of the diversity of parasite

assemblages. According to Luque et al. (2004) the

richness is a convenient measure, but it does not

capture all facets of diversity. It ignores the

evolutionary relationships among species

coexisting in an assemblage. Applied to parasite

assemblages, measures of diversity that

incorporate information on the relationships

among parasite species can shed light on how the

assemblage has been structured.

196

Biodiversity of fish parasites from Guandu river

Azevedoet al.

Hosts (N) P(%) MA MI ML

(cm)

MR Parasite

richness §

Astronotus ocellatus

35 74.00 13.71±0.38 18.46±0.51 20.44 1.77±1.23 2 (5) 0.43±0.41

Astyanax bimaculatus

40 30.00 0.95±0.05 3.16±0.15 9.92 0.37±0.58 2 (2) 0.03±0.12

Astyanax parahybae

40 35.00 0.65±0.03 1.86±0.08 10.39 0.35±0.48 2 (2) 0

Centropomus undecimalis

31 90.00 18.73±0.57 20.81±0.27 28.75 2.76±1.63 7 (5) 0.64±0.48

Cichla ocellaris

26 100.00 32.50±2.10 32.50±2.10 26.36 2.30±1.31 3 (4) 0.51±0.46

Cyphocharax gilbert 60 58.00 5.50±0.002 9.42±0.004 16.32 0.86±0.89 6 (1) 0.17±0.37

Geophagus brasiliensis

50 90.00 11.68±0.25 12.97±0.28 15.49 1.26±0.69 7 (3) 0.19±0.30

Gymnotus carapo

30 67.00 2.33±0.10 3.50±0.15 36.46 1.00±0.83 9 (1) 0.33±0.55

Hoplosternum littorale

10 60.00 2.91±0.05 4.85±0.08 19.65 1.03±1.13 5 (2) 0.25±0.52

Hypostomus affinis

31 87.00 26.35±1.09 30.25±1.26 27.75 2.09±1.30 2 (5) 0.46±0.38

Leporinus conirostris

18 61.00 27.33±3.80 44.72±6.23 36.60 0.72±0.57 1 (1) __

Leporinus copelandii 30 27.00 1.86±0.25 7.00±0.96 34.75 0.70±0.75 2 (3) 0.01±0.05

Loricariichthys castaneus 32 75.00 21.06±0.76 28.08±1.01 27.90 1.25±0.95 2 (3) 0.21±0.34

Mugil liza 34 79.00 12.50±0.83 15.74±1.04 34.05 1.82±1.78 6 (7) 0.48±0.66

Mylossoma aureum

17 100.00 223.70±12.96 223.70±12.96 15.66 1.64±0.49 1 (1) 0.18±0.22

Oligosarcus hepsetus

40 25.00 0.57±0.04 2.30±0.17 16.61 0.25±0.44 2 (1) 0

Pimelodus maculatus

40 100.00 59.42±1.50 59.42±1.50 22.92 2.57±1.22 5 (5) 0.42±0.28

Rhamdia quelen

32 80.00 40.16±1.99 50.20±2.49 32.68 1.46±1.00 5 (3) 0.21±0.25

Tilapia rendalii 30 84.00 2.8±0.09 3.36±0.10 22.14 0.83±0.38 0 (1) __

Trachelyopterus striatulus

60 95.00 40.83±0.64 42.98±0.67 19.20 1.46±0.74 8 (3) 0.16±0.22

Table 2. Number of hosts examined (N), total prevalence P(%), mean total abundance (MA), mean total intensity

(MI), mean length of hosts (ML), mean total species richness (MR), parasites richness and Margalef's richness index

(d) of metazoan parasites from Guandu River, State of Rio de Janeiro, Brazil.

(§) Endoparasites and ectoparasites (in parentheses).

Hosts ( H) J’ Dominant

taxon

Bray-Curtis Berger-

Parker

Δ+

Λ+

Astronotus ocellatus

0.45±0.34 0.62±0.43 Monogenea 41.70 (33.90-48.40) 0.45±0.30 95.60 311.30

Astyanax bimaculatus

0.02±0.09 0.04±0.19 Digenea 8.10 (6.10-9.50) 0.31±0.46 __ __

Astyanax parahybae

0 0 Digenea 6.30 ( 4.40-7.70) 0.35±0.48 __ __

Centropomus undecimalis

0.57±0.42 0.60±0.39 Monogenea 33.04 (26.90-38.40) 0.57±0.29 87.50 176.30

Cichla ocellaris

0.31±0.32 0.45±0.41 Cestoda 43.10 (38.60-48.06) 0.80±0.20 83.80 589.60

Cyphocharax gilbert 0.08±0.16 0.16±0.32 Digenea 13.60 (11.70-15.80) 0.51±0.46 91.60 68.89

Geophagus brasiliensis

0.11±0.18 0.20±0.32 Digenea 52.10 (48.80-55.30) 0.85±0.28 92.60 341.70

Gymnotus carapo

0.13±0.19 0.31±0.46 Nematoda 9.00 (5.90-13.00) 0.52±0.42 94.90 115.70

Hoplosternum littorale

0.13±0.25 0.19±0.37 Digenea 20.40 (17.80-22.60) 0.50±0.45 90.10 355.50

Hypostomus affinis

0.31±0.26 0.43±0.34 Monogenea 42.90 (36.40-48.50) 0.72±0.30 90.90 278.30

Leporinus conirostris

__ __ Digene 27.70 ( 22.70-27.70) __ __ __

Leporinus copelandii 0 0.005±0.03 Myxozoa 2.20( 1.40-2.90) 0.26±0.45 89.60 176.30

Loricariichthys castaneus 0.13±0.20 0.23±0.33 Monogenea 32.90 (28.70-37.40) 0.67±0.41 97.94 24.80

Mugil liza 0.26±0.40 0.32±0.42 Monogenea 9.40 (7.06-12.20) 0.62±0.39 80.60 642.10

Mylossoma aureum

0.14±0.18 0.25±0.32 Nematoda 59.50 (49.60-66.90) 0.92±0.13 __ __

Oligosarcus hepsetus

0 0 Digenea 4.70 ( 3.10-5.40) 0.25±0.44 __ __

Pimelodus maculatus

0.46±0.31 0.51±0.34 Monogenea 43.70 (37.90-490) 0.78±0.17 87.50 556.60

Rhamdia quelen 0.12±0.20 0.19±0.27 Monogenea 9.60 (6.60-13.20) 0.73±0.39 96.10 104.70

Tilapia rendalii __ __ Crustacea 55.10 ( 55.10-55.10) __ __ __

Trachelyopterus striatulus 0.07±0.13 0.11±0.19 Monogenea 60.90 (58.10-63.50) 0.92±0.22 92.60 35.80

Table 3. Brillouin's diversity index (H), Pielou's evenness index (J'), dominant taxon, Bray-Curtis similarity index,

+ +

Berger-Parker dominance index, average taxonomic distinctness (Δ) and variance in taxonomic distinctness (Λ) of

metazoan parasites from Guandu river, State of Rio de Janeiro, Brazil.

197

Neotrop. Helminthol., 5(2), 2011

Hosts Family Mean

length of

hosts (cm)

Formation

schools *

Environment§ Trophic