Chemosphere 185 (2017) 746e753

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Organotropism of methylmercury in fish of the southeastern of Brazil

L.S. Azevedo a, *, M.G. Almeida a, W.R. Bastos b, M.S. Suzuki a, M.C.N.N. Recktenvald b,
M.T.S. Bastos b, C.S. Vergílio c, C.M.M. de Souza a
a
rio de Ci^
Laborato encias Ambientais, Centro de Bioci^
encias e Biotecnologia, Universidade Estadual do Norte Fluminense Darcy Ribeiro, Campos dos
Goytacazes, Rio de Janeiro, RJ, CEP: 28013-602, Brazil
b
rio de Biogeoquímica Ambiental, Universidade Federal de Rondo
Laborato ^nia, Porto Velho, Rondo^nia, RO, CEP: 76815-800, Brazil
c
Departamento de Biologia, Centro de Ci^
rias, Universidade Federal do Espírito Santo, Alegre, Espírito Santo, ES, CEP: 29500-000, Brazil
encias Agra

h i g h l i g h t s

CH3Hgþconcentration were correlated with biometric variables.

Concentration of CH3Hgþ did not vary among feed habits.

Sediment as a representative source of mercury to fish.

a r t i c l e i n f o a b s t r a c t

Article history: This is one of the first studies to evaluate the effect of biometric variables (total length and weight), diet,
Received 2 December 2016 and abiotic matrices (sediment and water column) on the bioaccumulation of methylmercury in tissues
Received in revised form (muscle, liver, and gills) of four fish (two carnivore-invertivores, Pimelodus fur and Pachyurus adspersus;
11 July 2017
one carnivore-piscivore, Oligosarcus hepsetus; and one omnivore, Pimelodella lateristriga) in the lower
Accepted 15 July 2017
section of a river in southeastern Brazil. Samples of fish (n ¼ 120), water (n ¼ 5) and sediment (n ¼ 5)
Available online 17 July 2017
were collected at five sites characterized by pollution with mercury due to the use of organomercury
Handling Editor: Martine Leermakers fungicides and stream bed gold mining, commonly carried out in that section of the river in the 1980s.
The results show that biometric variables are strongly correlated with methylmercury levels in muscle
Keywords: (r ¼ 0.61, p < 0.0005) of P. fur. As a rule, concentrations of total mercury and methylmercury did not vary
Bioaccumulation considerably between the organs of the species of different food habits, because of the environmental
Fish conditions in the study area. Despite the low concentrations of mercury in sediments (<0.05 mg kg1
Mercury wet. wt), this compartment is a representative source of this pollutant for the organisms investigated,
Methylmercury
due to the close contact these animals keep with it in view of the low water columns in that section of
Paraíba do Sul river
the river.
Sediment
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
The various uses of soils in southeastern Brazil, a region char-
acterized by large urban and industrial centers, place intense an-
thropic pressure on its main drainage basin, where the Paraíba do
Sul river (PSR) is one of the main water courses. These waters are
exposed to the discharge of municipal sewage and industrial ef-
fluents as well as to the effects of mining and construction of dams.
Several authors have taken biogeochemical approaches to investi-
gate the effects of pollution in various sites in the region (Araújo
* Corresponding author.
E-mail address: lucasazevs@gmail.com (L.S. Azevedo).
et al., 2010; Rocha et al., 2015), recording the circulation of mer-
cury (Hg) discharged by activities such as the use of organomercury
fungicides in sugar cane plantations and gold mining (aluvionar
sediments) in the lower section of the river (Ca ^mara, 1990; Lima,
1990). However, these studies did not take into account the po-
tential transportation of Hg and the association of the methyl-
mercury (CH3Hgþ) with the region's fish species.
Binding covalently with the sulfhydryl group of proteins,
CH3Hgþ is one of the most toxic forms of Hg. The compound is
characterized by long residence times in tissues of living organisms,
which promote its bioaccumulation and biomagnification (Trudel
and Rasmussen, 1997).
Diet is the main pathway of Hg and CH3Hgþ into fish (Hall et al.,

http://dx.doi.org/10.1016/j.chemosphere.2017.07.081
0045-6535/© 2017 Elsevier Ltd. All rights reserved.
 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753
1997), while absorption through gills comes second (Jaeger et al.,
2009). Importantly, the species at the top of the trophic chain
normally present the highest levels of the organic form of Hg
(Clarkson et al., 2003). In this regard, retention of Hg in different
fish tissues may induce the emergence of adverse toxic effects,
causing histological damage like hepatic lesions and reducing the
exchange area of gill lamellae (Mela et al., 2007; Fernandes et al.,
2013). Furthermore, ecological factors like balance, posture, as
well as performance in swimming (Garcia et al., 2000; Berntssen
et al., 2003) and in both predation and the ability to escape from
predators may also be affected.
In this scenario, the present study evaluated bioaccumulation of
total Hg (THg) and CH3Hgþ in tissues (muscle, liver, and gills) of
four fish species of different food habits living in the lower section
of PSR, southeastern Brazil, a region with a record of exposure to Hg
due to the use of organomercury fungicides and gold mining ac-
tivities. To find the factors that influence the bioaccumulation of
THg and CH3Hgþ, we looked into the relationship between bio-
accumulation of these Hg chemical species and biometric variables,
food habit, and abiotic matrices (sediment and water column). We
put forward the hypothesis that biometric characteristics, the
ecology of fish, and the environmental conditions in the study area
influence the accumulation of contaminants in muscle, liver, and
gills.
2. Materials and methods
2.1. Study area
Covering 62,074 km2, the drainage basin of PSR is used as water
source in three highly urbanized Brazilian states, namely Rio de
Janeiro, Sa~o Paulo, and Minas Gerais. In total, 6.7 million people live
in the area (IBGE, 2010). Discharge reaches 4384 m3 s1 in the rainy
season, from December to February, and drops to a minimum in the
dry season (between June and August), with 181 m3 s1 (ANA,
2006).
More specifically, the study area is located in the northwest
portion of the PSR drainage basin. Samples were collected in five
sites along approximately 120 km: (1) S~ ao Sebastia ~o do Paraíba
~o Fidelis (SFI), (4) Baltazar (RPO), and
(SSP), (2) Itaocara (ITA), (3) Sa
(5) Guarani (RDR). Sites RPO and RDR are located in the sub-basins
of rivers Pomba and Dois Rios, both of which are tributaries to PSR

rea and the collection sites are shown in the map
(Fig. 1). The study a
published by Rocha et al. (2015).
The fish examined and water samples were collected in May and
July 2014. A prolonged drought affected the lower section of PSR in
2014 and 2015 (Agevap, 2016), when no significant statistical dif-
ferences were observed (Tukey test, a ¼ 0.05; p > 0.05) in total
rainfall between the dry season (April to September) and the rainy
season (October to March) (INMET, 2017). Therefore, it was
assumed that the influence of seasonality on the distribution of THg
and CH3Hgþ in fish, particulate matter, and sediment was
negligible.
2.2. Sampling
2.2.1. Fish
In total, 120 adult individuals of four fish species (N ¼ 30 for
each species) from three groups of food habits (two carnivore-
invertivores, Pimelodus fur and Pachyurus adspersus; one
carnivore-piscivore, Oligosarcus hepsetus; and one omnivore,
Pimelodella lateristriga) were captured with the assistance of Projeto
Piabanha, an initiative set up by local stakeholders interested in
preserving the PSR (with permission given by Brazil's Ministry for
the Environment, authorization 36,260-0). Net mesh sizes used
747
were 12, 15, 20, 25, 30, 35, 40, 45, 50, 60, and 70 mm, which
afforded to sort the adult individuals of each species.
The fish captured were identified according to site and bio-
metric parameters (total length, TL e total weight, TW). TL and TW
of species were obtained measuring with an ichthyometer and
weighing in analytical balance. Gender was determined by gross
inspection of gonads (Vazzoler, 1996). Muscle, liver, and gill tissues
were excised, lyophilized (dry wt.) and used in the chemical anal-
ysis to quantify THg and CH3Hgþ.
2.2.2. Fish ecology
Pimelodus fur is a catfish of the Siluriformes order of bentho-
pelagic habits at trophic level 3.3 ± 0.5 (Fishbase, 2017). The species
feeds mainly on small invertebrates (insects, mollusks, and others).
Pimellodela lateristriga is another siluriform catfish, with a trophic
level of 3.6 ± 0.5 (Fishbase, 2017). This species’ diet is more
generalist, thus differing form that of P. fur. Neither of the two
species has considerable commercial value, although the demersal
habit of these animals means that they touch the sediment directly,
which is the main THg and CH3Hgþ in aquatic ecosystems. There-
fore, both are appropriate models to study the influence of abiotic
matrices on Hg bioaccumulation. Oligosarcus hepsetus is a fish of the
Charachiformes order, with pelagic habits and at trophic level
4.2 ± 0.73. This piscivorous species is at the highest trophic level,
compared with the others included in this study.
Pachyurus adspersus (“brazilian croaker”) belongs to the Perci-
formes order, Sicaenidae family, of benthopelagic habits that may
be considered similar to those of P. fur (Fishbase, 2017) and it is at
trophic level 3.6 ± 0.4. This is the most commercially interesting
species examined in this study, and plays an important role in the
region's fishing industry (Vieira, 2010).
2.2.3. Suspended particulate matter and bottom sediment
Five liters of the river water were collected in polyethylene
flasks at each site. Suspended particulate matter (SPM) was
collected by saturation of 0.7-mm of porosity, f 47 mm GF/F filters
and lyophilized (L10, Liotop). Limnologic variables like temperature
(Thermo Scientific Orion 3 STAR), pH, electric conductivity, O2
saturation (Thermo Scientific Orion STAR A221), and transparency
(Secchi disk) were measured at every site during field excursions.
Bottom sediment samples were collected from the top 10-cm
layer in the five sites in October 2013. Samples were sieved and
the passing fraction (<63 mm) was lyophilized and macerated in a
ball mill.
2.3. Chemical analyses
2.3.1. Total mercury in fish tissues
The chemical digestion of muscle (0.1 g dry wt.), liver (0.01 g dry
wt.), and gill (0.1 g dry wt.) to determine THg was carried out in five
steps (adapted from Bastos et al., 1998): (i) extraction with 1 mL
H2O2 30% and 3 mL H2SO4 97%:HNO3 65% (1:1); (ii) digestion in
block digester system at 60  C upon total solubilization of the
sample; (iii) addition of 5 mL KMnO4 5%; (iv) titration with
NH2OH$HCl 12%; (v) filtration through Whatman 40 paper and
completion to a 20-mL volume with ultra-pure water (milli-Q,
Millipore, Cambridge, MA, USA). THg was detected using a mercury
analyzer (Quick trace M-7500, CETAC) with a detection limit of
0.001 mg kg1. The accuracy of the method was evaluated using the
dogfish muscle reference material (DORM-3, NRC, concentration
0.382 ± 0.060 mg kg1 dry wt) with recovery of >90.0%. Repro-
ducibility was assessed using triplicates for every 80 samples (co-
efficient of variation < 10.0%). The results were expressed as wet wt
using the equation: concentration in wet wt ¼ concentration dry wt
x (100 - water percent level).
748 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753
Fig. 1. Lower section of the drainage basin of river Paraíba do Sul and sampling sites.
2.3.2. Total mercury in suspended particulate matter and sediment
Digestion of the sediment matrix (0.5 g dry wt.) and of filters
with SPM (the samples comprised two filters per site) was carried
out in five steps: (i) extraction with 1e2 mL Milli-Q H2O and 8 mL 3
HCl 37%: 1HNO3 65%; (ii) digestion in a block digester system at
60  C for 2 h; (iii) addition of 5 mL Milli-Q H2O and 10 mL KMnO4 5%
after cooling; (iv) digestion in a block digester system at 60  C for
30 min; and (v) filtration through Whatman paper 40 and
completion to a 25-mL volume with ultra-pure water (milli-Q,
Millipore, Cambridge, MA, USA) (Bastos et al., 1998). The results of
THg in sediment were kindly provided by Rocha et al. (2015).
THg in SPM was quantified using a mercury analyzer (Quick
trace M-7500, CETAC) with detection limit of 6  108 mg L1. In
sediment, THg was analyzed using a mercury analyzer (Mercury
Analyses System, FIMS-400 e Perkin-Elmer) with detection limit of
0.0005 mg kg1. The accuracy of the method for sediment was
evaluated using the estuarine sediment sample reference material
(IAEA-405, Mel, concentration 0.81 mg kg1 dry wt) with recovery
of 93.8%. Reproducibility was assessed using triplicates in all sam-
ples (coefficient of variation < 8%). The accuracy of the method for
SPM was evaluated using the marine sediment reference material
(MESS-3, NRC, concentration 0.091 ± 0.009 mg kg1 dry wt) with
recovery of >94.0% and reproducibility was <10.0%.
2.3.3. Methylmercury in fish tissues
Quantification of CH3Hgþ was carried out as described in EPA
Method 1630 (2001) and Liang et al. (1994), in four steps: (i)
addition of 50 mL of KOH in methanol 25% w/v; (ii) heating in stove
(NI 1512, Nova Instruments) to 70  C for 6 h; (iii) addition of 300 mL
sodium acetate buffer 272 g L1 (pH 4.5) followed by 3 mL of sample
and 50 mL of a solution of sodium tetraethylborate 1% w/v (Taylor
et al., 2011); (iv) completion to 40 mL with ultra-pure water
(milli-Q, Millipore, Cambridge, MA, USA). The analysis was carried
out in gas chromatography-cold vapor-atomic fluorescence spec-
troscopy (GC-CV-AFS, MERX-TM automated system from Brooks
Rand Labs Seattle, USA). Accuracy was assessed using the fish
 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753 749

protein reference samples (DORM-2, NRC, concentration
4.47 ± 0.032 mg kg1 dry wt) with a recovery rate of 93% and
detection limit of 0.0014 mg kg1. Reproducibility was assessed
using duplicates for every samples (coefficient of variation < 10%).
The results were expressed as wet wt using the equation: con-
centration in wet wt ¼ concentration dry wt x (100 - water percent
level).
2.4. Total lipid extraction
The extraction of lipids from the muscle of the fish species (five
lyophilized samples per species) was carried out according to Folch
et al. (1957).
2.5. Statistical analysis
Normality of data was tested using the Shapiro-Wilk test. In
order to meet ANOVA requirements, data were transformed using a
maximum likelihood function (Box Cox). The differences in con-
centrations of THg, CH3Hgþ, and the percent CH3Hgþ levels in each
tissue were analyzed using ANOVA followed by the Tukey test
(p < 0.05). The Pearson correlation (a ¼ 0.05) was used to correlate
the concentrations of CH3Hgþ with biometric data. All statistical
tests were conducted using the software Rstudio 3.2.2.
3. Results and discussion
3.1. Bioaccumulation of THg and CH3Hgþ
Since there are no significant differences among males and fe-
males in terms of bioaccumulation of mercury (p > 0.05) we used
the whole data set without discriminate the number of female and
males individuals. Being liposoluble, Hg tends to accumulate in
liver tissue and muscle (McIntyre and Beauchamp, 2007). In the
present study, the concentrations of THg were highest in liver,
followed by muscle and gills (Table 1, p < 0.05), as observed in
previous research (Halvelkova
et al., 2008; Vieira et al., 2011;
Berge
s-Tiznado et al., 2015). The exception was observed for
O. hepsetus, when THg levels were high in liver and muscle samples
(with no statistically significant difference), followed by gills
(p < 0.05). A different pattern was observed for CH3Hgþ. In
O. hepsetus, P. lateristriga and P. fur, CH3Hgþ concentrations were
highest in muscle (p < 0.05), followed by liver and gill samples
(with no statistically significant difference between the last two
tissues), while in P. adspersus CH3Hgþconcentration was highest in
muscle, followed by liver and then gills (p < 0.05) (Table 1). High
levels of CH3Hgþ in fish muscle are expected due to the high affinity
of the pollutant for the thiol groups constituting amino acids like
cysteine, for instance (Ruelas-Inzunza et al., 2003; Leaner and
Mason, 2004). This relationship was confirmed by the high
percent values of CH3Hgþ in muscle of O. Hepsetus (73.63± 20.53%),
P. adspersus (78.74± 22.89%), P. lateristriga (85.00± 29.67%), and
P. fur (56.99± 29.44%) (Fig. 2).
The CH3Hgþ concentrations in the liver of the fish species
analyzed were statistically lower than those observed in muscle
(p < 0.05), and the percent values in the organ were the lowest,
differing statistically from muscle and gills (p < 0.05). The liver
plays a key role in the detoxification of CH3Hgþ, converting the
compound into less bioaccumulative inorganic forms of the metal,
thus reducing its toxicity (Suda and Hirayama, 1992; Wiener and
Spry, 1995; Watras et al., 1998; Gonzalez et al., 2005; Branco
et al., 2007, 2011).
The calculated levels of lipids (%) in the muscle of species
(P. lateristriga: 26.85 ± 21.34; P. adspersus: 31.53 ± 18.02; P. fur:
30.61 ± 18.45; O. Hepsetus: 5.60 ± 2.66) did not correlate signifi-
cantly with the levels of THg and CH3Hgþ. Non-significant corre-
lations between these variables have been observed by other
authors (Pethybridge et al., 2007).
The mean lipid level in O. hepsetus was low, compared with the
other species. Classified as a pelagic piscivore, its diet may be
presumed to contain fattier items, considering the prey consumed
by P. fur, P. adspersus, and P. lateristriga. Nevertheless, the prolonged
drought in 2014 and 2015 (Agevap, 2016) may have reduced the
availability of the prey consumed by O. hepsetus, forcing it to
explore other trophic resources in a restricted food scenario. In a
study about the trophic ecology of fish species in the lower section
of the PSR, Rocha et al. (2015) observed that the isotopic niche of
O. hepsetus overlapped significantly with that of P. fur, P. adspersus,
P. lateristriga, and other demersal fish, and suggested that these
species compete for trophic resources.
A correlation between CH3Hgþ concentration in muscle and
weight was also expected, since this chemical species is slowly
excreted from the fish muscle (Peterson and Sickle, 2007; Storelli
et al., 2007; Carrasco et al., 2011; Polak-Juszczak, 2012; Burger

Table 1
Biometric data and THg and CH3Hg þ concentration in four fish species captured in Paraíba do Sul river (mg kg1 wet wt.).

Species Biometric data (mean ± SD) THg (mean ± SD) CH3Hgþ (mean ± SD)

Min-Max Min-Max Min-Max

TL (cm) TW (g) Muscle (N) Liver (N) Gills (N) Muscle (N) Liver (N) Gill (N)

Pimelodella 16.48 ± 1.74 30.09 ± 11.93 0.085 ± 0.054aA 0.189 ± 0.106bA(30) 0.032 ± 0.021cA 0.079 ± 0.053dC 0.014 ± 0.008eC 0.016 ± 0.008e(3)
lateristriga (30) (30) (20) (18)
(Omnivorous) 12.4e20.0 14.8e71.7 0.015e0.207 0.032e0.4683 0.003e0.099 0.019e0.236 0.001e0.040 0.007e0.021
Pachyurus 19.51 ± 2.81 57.87 ± 29.0 0.061 ± 0.042aA 0.211 ± 0.13bA(30) 0.025 ± 0.015cA 0.065 ± 0.039dC 0.044 ± 0.020eD 0.020 ± 0.011f(19)
adspersus (30) (30) (16) (20)
(Invertivore) 15.3e25.7 25.62 0.007e0.164 0.020e0.458 0.001e0.06 0.011e0.163 0.007e0.086 0.004e0.044
e161.38
Pimelodus fur 19.76 ± 3.89 75.65 ± 45.63 0.144 ± 0.076aB 0.249 ± 0.128bA(30) 0.030 ± 0.016cA 0.074 ± 0.039dC 0.017 ± 0.010eE 0.013 ± 0.008e(24)
(30) (30) (28) (21)
(Invertivore) 13.4e27.5 21.14e206.2 0.062e0.351 0.065e0.637 0.009e0.076 0.023e0.227 0.003e0.046 0.003e0.033
Oligosarcus 16.64 ± 2.52 37.55 ± 22.71 0.077 ± 0.041aA 0.106 ± 0.074aB (30) 0.029 ± 0.014bA 0.059 ± 0.031dC 0.018 ± 0.013eE 0.019 ± 0.013e (7)
hepsetus (30) (30) (30) (16)
(Piscivore) 11.3e22.7 15.5e112.5 0.019e0.203 0.012e0.341 0.007e0.062 0.018e0.146 0.003e0.049 0.005e0.041

Different lowercase letters indicate significant statistical difference (p < 0.05) in THg and CH3Hg þ between tissues of the same specie.
Different uppercase letters indicate significant statistical difference (p < 0.05) in THg and CH3Hg þ between the same tissues of different species.
N ¼ number of samples.
TL ¼ total length.
TW ¼ total weight.
750 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753

Fig. 2. Percent CH3Hgþ values in three tissues of four fish species captured in Paraíba do Sul river. Different letters indicate statistically significant difference (p < 0.05).

et al., 2013; Storelli et al., 2005; Quiu et al., 2011; Maceda-Veiga

et al., 2012). Nevertheless, P. fur was the only species in which the
levels of CH3Hgþ in muscle correlated positively and significantly
with TW (p < 0.0005) (Table 2 and Fig. 3). The consumption of preys
from higher trophic levels by an individual of a given fish species as
it grows in size may explains the positive correlation between these
variables (Trudel and Rasmussen, 2006).

Table 2
Pearson correlation (r) between CH3Hgþ and total weigth of four fish species
captured in Paraiba do Sul river.

Species M  TW L  TW G  TW

Oligosarcus hepsetus 0.08 0.28 0.06

Pimelodus fur 0.61* 0.012 0.36
Pimellodela lateristriga 0.11 0.02 IS
Pachyurus adspersus 0.17 0.15 0.37

*p < 0.0005. Fig. 3. Pearson correlation between concentrations of CH3Hgþ (wet wt.) and total
M ¼ muscle; L ¼ liver; G ¼ gills; TW ¼ total weight; IS ¼ Insufficient samples. weight of P. fur captured in Paraíba do Sul river.
 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753
Table 3
Concentration of THg in the SPM (volumetric basis mg L1) and sediment (mg kg1 wet wt.) in five sites in Paraíba do Sul river.
Geographical Coordinates
SSP S 21º440 54.700
W 42º200 32.700
ITA S 21º380 0400
W 42º020 01.900
RPO S 21º340 51.600
W 42º080 25.100
RDR S 21º440 14.600
W 41º0550 55.100
SFI S 21º380 04.900
W 41º450 41.200
SSP: S~ ~o do Paraíba, ITA: Itaocara, SFI: S~
ao Sebastia ao Fidelis, POR: Baltazar, RDR: Guarani, SPM: Suspended Particulate Matter.
The non-significant correlations between the levels of CH3Hgþ
in the muscle of the other three species may be explained by the
poor distribution of individuals along the weight intervals consid-
ered (<20 g, 20e40 g, 41e60 g, 61e80 g, 81e100 g, 101e120 g,
121e140 g, 141e160 g, and >160 g). A large number of the in-
dividuals of O. hepsetus (63%), P. adserpsus (36%), and P. lateristriga
(70%) were in the 20e40-g range. For P. fur, the highest percent
number of individuals in the same weight range (which was also
20e40 g) was 23%. Except for the <20% and 141e160-g weight in-
tervals, all other weight classes represented between 10% and 16%
of individuals. This indicates a more even distribution of P. fur
samples between the weight intervals.
Also, the non-significant correlations between TW and the
CH3Hgþ levels in liver and gills suggest that these tissues are only
lightly influenced by weight gain (Table 2). Even though the liver is
a tissue that can accumulate Hg levels that exceed those stored in
muscle (Table 1), the organ plays an important role in the detoxi-
fication and excretion of CH3Hgþ through demethylation reactions
that convert the organic form of Hg into less toxic and bio-
accumulative species. Such metabolic function prevents the accu-
mulation of CH3Hgþ in the liver in view of weight gain. Though the
correlation coefficients for L and TW in the four species (Table 2)
were not significant, they are negative, which agrees with the
liver's demethylation and excretion ability.
3.2. Influence of diet
The invertivore species (P. fur) presented the highest THg levels
in muscle, while the piscivorous one (O. hepsetus) had the lowest
THg in liver (p < 0.05). The levels of THg in the gills of the four
species did not differ significantly (p > 0.05). Concerning CH3Hgþ,
the concentrations in muscle did not differ across species (p > 0.05),
and the levels in the liver of P. fur, O. Hepsetus, and P. lateristriga
were lower than those recorded in the liver of P. adspersus
(p < 0.05). The comparisons between the levels of contaminant in
gills were not carried out, because of the small number of samples
from P. lateristriga and O. herpsetus.
These results do not support the expected hierarchy of food
habits in terms of the bioaccumulation of THg and CH3Hgþ
(piscivores > invertivores > omnivores). However, the distribution
of these contaminants in the tissues of the species studied may
indicate that these organisms are exploring other, similar food re-
sources. The main reasons backing this statement are the envi-
ronmental conditions in the study area during sample collection
and the literature data about isotopic analyses (C and N) in the
muscle of the four species examined.
During droughts, the diversity of the prey consumed by fish may
decline (Abelha et al., 2001). In this scenario, fish may have to
explore alternative trophic resources as a survival strategy (Begon
et al., 2005).
751
SPM (Mean ± SD) Sediment (Mean ± SD)
1.36  107 ± 2.03  108 0.041 ± 0.004
1.97  106 ± 1.3  106 0.047 ± 0.001
6 6
1.12  10 ± 1.38  10 0.045 ± 0.005
6 7
1.57  10 ± 7.67  10 0.035 ± 0.002
8.69  107 ± 6.02  108 0.062 ± 0.001
The determination of the isotopic signatures of d13C and d15N
conducted by Rocha et al. (2015) for the four species included in the
present study indicate that the niche of the pelagic carnivore spe-
cies living in the lower section of PSR (represented mainly by
O. hepsetus) overlaps significantly with the niches taken by P. fur,
P. lateristriga, and P. adspersus, and that these animals share trophic
resources. Therefore, the intake of THg and CH3Hgþ from nutrition
may not have varied considerably between the fish examined,
explaining the statistically similar concentrations of CH3Hgþ in the
muscle of these species. It should be emphasized, however, that the
differences observed in the concentrations of THg and CH3Hgþ in
the liver of these species are not good indicators of the influence of
food habits, since this tissue is able to demethylate the organic form
of Hg, promoting the excretion of the metal.
3.3. Influence of abiotic matrices
In the present study, the levels of THg measured in sediment
were as many as 1000 times higher than in the water column (the
mean values of physicochemical parameters observed in the two
collection periods were: transparency (m) ¼ 2.77 ± 0.71; T
( C) ¼ 23.35 ± 1.61; pH ¼ 7.4 ± 0.1; O2 saturation (%) ¼ 99.62;
electric conductivity (mS cm1) ¼ 46.96 ± 6.04), considering the
overall extension of the sampling sites (approximately 120 km) and
the fact that all species were captured in all collection sites
(Table 3). The effect of exposure to sediment, concerning protection
of aquatic life, is represented by two parameters, the threshold
effect level (TEL ¼ 0.147 mg kg1) and the probable effect level
(PEL ¼ 0.486 mg kg1) (Macdonald et al., 1996), which indicate the
risk of deleterious effects to the aquatic biota (Table 3). The com-
parison of sediment variables observed in the present study and
TEL and PEL values indicate that the sediment in all collection sites
poses only a very low hazard of adverse effects to the local biota.
However, even under such conditions, we observed high
CH3Hgþ percent values in the gills of all fish species analyzed,
especially the demersal species P. lateristriga (Fig. 2). These results
may be associated with the closer contact the species has with
sediment, as discussed in previous studies about the relationship
between the extent of contamination in this compartment and THg
and CH3Hgþ bioaccumulation in fish (Kasper et al., 2009; Hosseini
et al., 2013; Rocha et al., 2015). Abdolvand et al. (2013) also
attributed the influence of sediment on the high percent levels
(>90%) and high concentrations of CH3Hgþ observed in gills of the
benthonic specie Epinephelus diacanthus (1.025 ± 0.05 mg kg1, dry
wt.) in river Arvand, Iran.
The effect of sediment on bioaccumulation in gills may be
potentiated in shallow aquatic ecosystems, as in the PSR, during
low rainfall periods (considering the transparency of the water
column, which was as low as 2.77 ± 0.71 m). These conditions
prompt the need for a foraging strategy characterized by incursions
752 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753
along the entire water column (Rocha et al., 2015), when the bot-
tom sediment is stirred, increasing the system's turbid zone and
therefore the amounts of available suspended particulate matter,
mainly at greater depths. Therefore, physical characteristics of the
system like water column height as well as ecologic factors such as
a species' habitat and foraging strategy may either increase or
decrease the influence of abiotic matrices on the bioaccumulation
of Hg and CH3Hgþ by fish.
4. Conclusion
The levels of CH3Hgþ in muscle were influenced by weight gain
of the fish analyzed. Nevertheless, the number of individuals should
distribute the most homogeneous way possible across the weight
intervals for each species. Also, TW did not affect THg and CH3Hgþ
levels in liver and gills considerably.
It was not possible to define the order of the concentrations of
THg and CH3Hgþ in muscle, liver, and gills of the four species
investigated based on the food habits of each, due to the prolonged
drought in the PSR during the collection period and the overlapping
of trophic niches of the species studied. The high percent values of
CH3Hgþ in the gills of these fish, especially in the demersal
P. lateristriga indicate the importance of sediment as a represen-
tative source of Hg and CH3Hgþ in PSR.
Acknowledgements
The authors are grateful to Projeto Piabanha (permission given
by Brazil's Ministry for the Environment, 36260-2) for capturing
fish, to Prof Carlos E. V. de Carvalho and Prof Marcos S. M. de B.
Saloma~o (Universidade Estadual do Norte Fluminense Darcy
Ribeiro) for providing physico-chemical data, to Laborato
rio de
Radioiso
topos Eduardo Penna Franca (LREPF) of Universidade
Federal do Rio de Janeiro for permission to use the mercury ana-
lyses system (FIMS-400). Cristina M Souza received a grant from
Fundaça~o de Amparo a  Pesquisa do RJ (FAPERJ - E-26/111.790/2013).
References
Abelha, M.C.F., Agostinho, A.A., Goulart, E., 2001. Plasticidade tro
fica em peixes de


agua doce. Acta Sci. Biol. Sci. 23, 425e434.
Agevap, Age ^ncia, 2016. da Bacia do rio Paraíba do Sul. http://www.agevap.org.br
(Accessed July 2017).
ANA, 2006. Age ^ncia Nacional de Aguas.
http://www2.ana.gov.br/Paginas/default.
aspx (Accessed November 2015).
Abdolvand, S., Esfahani, S.K., Dmirchi, S., 2013. Mercury (Hg) and methyl mercury
(MMHg) bioaccumulation in three fish species (sea food) from Persian Gulf.
ToxEHS 6, 192e198.
Araújo, B.F., Carvalho, C.E.V., Andrade, D.R., Gomes, R.S., Souza, C.T., 2010. Distrib-
~o de Mercúrio Total em Tecido Muscular de Hypostomus affinis (Stein-
uiça
dachner, 1877) e Hypostomus c.f. luetkini (Barlenger, 1887) do rio Muriae
,
Itaperuna e RJ. J. Braz. Ecotoxicol. 5, 49e54.
Bastos, W.R., Malm, O., Pfeifer, W.C., Cleary, D., 1998. Establishment and analytical
quality control of laboratories for Hg determination in biological and geological
samples in the Amazon. Braz. Tech. Rev. Cie ^nc. Cult. 50, 255e260.
Begon, M., Townsend, C.R., Harper, J.L., 2005. Ecology: from Individuals to Ecosys-
tems, 4 ed. Blackwell Publishing, Incorporated.
Berge
s-Tiznado, M.E., M
arques-Farías, F., Lara-Mendoza, R.E., Torres-Rojas, Y.E.,
Galva
n-Magan ~ a, F., Bojo
rquez-Leyva, H., P
aez-Osuna, F., 2015. Mercury and
selenium in muscle and target organs of scalloped hammerhead sharks Sphyrna
lewini of the SE Gulf of California: dietary intake, molar ratios, loads, and hu-
man health risks. Arch. Environ. Contam. Toxicol. 69, 440e452.
Berntssen, M., Aatland, A., Handly, R., 2003. Chronic dietary mercury exposure
causes oxidative stress, brain lesions, and altered behaviour in Atlantic salmon
(Salmo salar) parr. Aquat. Toxicol. 65, 55e72.
Branco, V., Vale, C., Cana
rio, J., Santos, M.N., 2007. Mercury and selenium in blue
shark (Prionace glauca, L.1758) and swordfish (Xiphias gladius, L.1758) from two
areas of the Atlantic ocean. Environ. Pollut. 150, 373e380.
Branco, V., Cana
rio, J., Holmgren, A., Carvalho, C., 2011. Inhibition of the thioredoxin
system in the brain and liver of zebra-seabreams exposed to waterborne
methylmercury. Toxicol. Appl. Pharmacol. 251, 95e103.
Burger, J., Jeitner, C., Donio, M., Pittfield, T., Gochfeld, M., 2013. Mercury and sele-
nium levels, and selenium:mercury molar ratios of brain, muscle and other
tissues in bluefish (Pomatomus saltatrix) from New Jersey, USA. Sci. Total Env-
viron. 43, 278e286.
^mara, V.M., 1990. O caso de Campos, RJ: estudo do quadro de morbidade causado
Ca
pela exposiça ~o pregressa dos trabalhadores aps fungicidas mercuriais. In:
Hacon, S., Lacerda, L.D., Pfeiffer, W.C., Carvalho, D. (Eds.), Riscos e conseque ^ncias
do uso do Mercúrio. FINEP/CNPq/MS/IBAMA, Rio de Janeiro, pp. 229e246.
Carrasco, L., Barata, C., García-Berthou, E., Tobias, A., Bayona, J.M., Díez, S., 2011.
Patterns of mercury and methylmercury bioaccumulation in fish species
downstream of a long-term mercury- contaminated site in the lower Ebro River
(NE Spain). Chemosphere 84, 1642e1649.
Clarkson, T.W., Magos, L., Myers, G.J., 2003. Human exposure to mercury: the three
modern dilemmas. J. Trace Elem. Exp. Med. 16, 321e343.
EPA, 2001. Method 1630: Methyl Mercury in Water by Distillation, Aqueous Eth-
ylation, Purge and Trap, and CVAFS. U.S. Environmental Protection Agency: EPA-
821-R-01e020. EPA Office of Water, Washington DC, p. 49.
Fernandes, M.N., Paulino, M.G., Sakuragui, M.M., Ramos, C.A., Pereira, C.D.S.,
Sadauskas-Henrique, H., 2013. Organochlorines and metals induce changes in
the mitochondria-rich cells of fish gills: an integrative field study involving
chemical, biochemical and morphological analyses. Aquat. Toxicol. 126,
180e190.
Fishbase, 2017. Information. http://www.fishbase.org (Accessed July 2017).
Folch, J., Lee, M., Sloane-Stanley, G.H., 1957. A simple method for the isolation and
purification of total lipids from animal tissues. J. Biol.. Chem. 226, 497e509.
Garcia, L.M., Castro, B., Ribeiro, R., Guilhermino, L., 2000. Characterization of
cholinesterase from guppy (Poecilia reticulata) muscle and in vitro inhibition by
environmental contaminants. Biomark 5, 274e284.
Gonzalez, P., Dominique, P., Massabuau, J.C., Boudou, A., Bourdineaud, J.P., 2005.
Comparative effects of dietary methylmercury on Gene Expression in liver,
skeletal muscle and brain of the zebrafish (Danio rerio). Environ. Sci. Technol.
39, 3972e3980.
Hall, B.D., Bodaly, R.A., Fudge, R.J.P., Rudd, J.W.M., Rosenberg, D.M., 1997. Food as the
dominant pathway of methylmercury uptake by fish. Water Air Soil Pollut. 100,
13e24.
Halvelkova
, M., Dusek, L., Ne
methova
, D., Poleszczuk, G., Svobodova
, Z., 2008.
Comparison of mercury distribution between liver and muscle e a bio-
monitoring of fish from lightly and heavily contaminated localities. Sensor 8,
4095e4109.
Hosseini, M., Nabavi, S.M.B., Parsa, Y., 2013. Bioaccumulation of trace mercury in
trophic levels of benthic, benthopelagic, pelagic fish species, and sea birds from
arvand river. Iran. Biol. Trace. Elem. Res. 156, 175e180.
IBGE Instituto Brasileiro de Geografia e Estatística, 2010. Censo Demogr
afico
de 2010. http://www.ibge.gov.br/home/estatistica/populacao/censo2010/
(Accessed June 2015).
INMET, 2017. Instituto Nacional de Meteorologia. http://www.inmet.gov.br
(Accessed July 2017).
Jaeger, I., Hop, H., Gabrielsen, G.W., 2009. Biomagnification of mercury in selected
species from an Arctic marine foodweb in Svalbard. Sci. Total Environ. 407,
4744e4751.
Kasper, D., Palermo, E.F.A., Dias, A.C.M.I., Ferreira, G.L., Leita ~o, R.P., Branco, C.W.C.,
Malm, O., 2009. Mercury distribution in different tissues and trophic levels of
fish from a tropical reservoir, Brazil. Neotrop. Ichthyol. 7, 751e759.
Leaner, J., Mason, R.P., 2004. Methylmercury uptake and distribution kinetics in
Sheepshead Minnows, Cyprinodon variegatus, after exposure to CH3Hg-spiked
food. Environ. Toxicol. Chem. 23, 2138e2146.
Liang, L., Bloom, N.S., Horvat, M., 1994. Simultaneous determination of mercury
speciation in biological materials by GC/CVAFS after ethylation and room-
temperature precollection. Clin. Chem. 40, 602e607.
Lima, E.C.R., 1990. Riscos e conseque ^ncias do uso do mercúrio: a situaça ~o do Rio de
Janeiro. In: Riscos e conseque ^ncias do uso do mercúrio. Finep/CNPQ/MS/IBAMA,
Rio de Janeiro, pp. 268e274.
Macdonald, D.D., Carr, R.S., Calder, F.D., Long, E.R., Ingersoll, C.G., 1996. Development
and evaluation sediment quality guidelines for from three rivers in Otago. New
Zeal. Sci. Total Environ. 262, 37e47.
Maceda-Veiga, A., Monroy, M., Sostoa, A., 2012. Metal bioaccumulation in the
Mediterranean barbel (Barbus meridionalis) in a Mediterranean river receiving
effluents from urban and industrial wastewater treatment plants. Ecotoxicol.
Environ. Saf. 76, 93e101.
McIntyre, J.K., Beauchamp, D.A., 2007. Age and trophic position dominate bio-
accumulation of mercury and organochlorides in the food web of Lake Wash-
ington. Sci. Total Environ. 372, 571e584.
Mela, M., Randi, M.A., Ventura, D.F., Carvalho, C.E., Pelletier, E., Oliveira Ribeiro, C.A.,
2007. Effects of dietary methylmercury on liver and kidney histology in the
neotropical fish Hoplias malabaricus. Ecotoxicol. Environ. Saf. 68, 426e435.
Peterson, S.A., Sickle, J.V., 2007. Mercury concentration in fish from streams and
rivers throughout the western United States. Environ. Sci. Technol. 41, 58e65.
Pethybridge, H., Daley, R., Virtue, P., Butler, E.C.V., Cossa, D., Nichols, P.D., 2007. Lipid
and mercury profiles of 61 mid-trophic species collected of south-eastern
Australia. Mar. Freshw. Res. 61, 1092e1108.
Polak-Juszczak, L., 2012. Bioaccumulation of mercury in the trophic chain of flatfish
from the Baltic Sea. Chemosphere 89, 585e591.
Quiu, Y.W., Lin, D., Liu, J.Q., Zeng, E.Y., 2011. Bioaccumulation of trace metals in
farmed fish from South China and potential risk assessment. Ecotoxicol. Envi-
ron. Saf. 74, 284e293.
Rocha, A.R.M., Di Beneditto, A.P.M., Pestana, I.A., Souza, C.M.M., 2015. Isotopic
profile and Mercury concentration in fish of the lower portion of the rio Paraíba
 L.S. Azevedo et al. / Chemosphere 185 (2017) 746e753
do Sul watershed, southeastern Brazil. Neotrop. Ichthyol. 13, 723e732.
Ruelas-Inzunza, J.R., Horvat, M., Pe
rez-Corte
s, H., P
aez-Osuna, F., 2003. Methyl-
mercury and total mercury distribution in tissues of gray whales (Eschrichtius
robustus) and spinner dolphins (Stenella longirostris) stranded along the lower
Gulf of California. Mexico. Cienc. Mar. 29, 1e8.
Storelli, M.M., Storelli, A., Giacominelli-Stuffer, R., Marcotrigiano, G.O., 2005. Mer-
cury speciation in the muscle of two commercially important fish, hake (Mer-
luccius merluccius) and striped mullet (Mullus barbatus) from Mediterranean
Sea: estimated weekly intake. Food Chem. 89, 295e300.
Storelli, M.M., Barone, G., Piscitelli, G., Marcotrigiano, G.O., 2007. Mercury in fish:
concentration vs. fish size and estimates of mercury intake. Food Addit. Contam.
24, 1353e1357.
Suda, I., Hirayama, K., 1992. Degradation ofmethyl and ethylmercury into inorga-
nicmercury by hidroxyl radical produced from rat liver microsomes. Arch.
Toxicol. 66, 398e402.
Taylor, V.F., Carter, A., Davies, C., Jackson, B.P., 2011. Trace-level automated mercury
speciation analysis. Anal. Methods 3, 1143e1148.
753
Trudel, M., Rasmussen, J.B., 1997. Modeling the elimination of mercury by fish.
Environ. Sci. Technol. 31, 1716ee1722e.
Trudel, M., Rasmussen, J.B., 2006. Bioenergetics and mercury dynamics in fish: a
modeling perspective. Can. J. Fish. Aquat. Sci. 63, 1890e1902.
Vazzoler, A.E.A.M., 1996. Biologia da reproduç~ ao de peixes teleo
steos: teoria e
pra
tica. EDUEM, Maringa
.
Vieira, F., 2010. Distribuiça~o, impactos ambientais e conservaç~ ao da fauna de peixes
da bacia do rio Doce. MGBiota 2, 5e48.
Vieira, J.L.F., Gomes, A.L.S., Santos, J.P.N., Lima, T.C.D., Freitas Jr., J.A., Pinheiro, M.C.N.,
2011. Mercury distribution in organs of two species of fish from Amazon region.
Bull. Environ. Contam. Toxicol. 87, 377e380.
Watras, C.J., Back, R.C., Halvorsen, S., Hudson, R.J.M., Morrison, K.A., Wente, S.P.,
1998. Bioaccumulation of mercury in pelagic freshwater food webs. Sci. Total
Environ. 219, 183e208.
Wiener, J.G., Spry, D.J., 1995. Toxicological significance of mercury in freshwater fish.
In: Heinz, G., Beyer, N. (Eds.), Interpreting Concentrations of Environmental
Contaminants in Wildlife Tissures. Lewis Publishers, Chelsea, Michigan.