ARTICLE IN PRESS

Ecotoxicology and Environmental Safety 61 (2005) 247255

www.elsevier.com/locate/ecoenv

Health variables and gill morphology in the tropical sh Astyanax

fasciatus from a sewage-contaminated river
Andre Albertoa, Antonio F.M. Camargob, Jose R. Veranic, Oscar F.T. Costad,
Marisa N. Fernandesa,
a
Department of Physiological Sciences, Federal University of Sao Carlos-UFSCar, C. Postal 676, 13565-905 Sao Carlos, SP, Brazil
b
Department of Ecology, Paulista State University-UNESP, Av. 24/A, 1515, 13506-900 Rio Claro, SP, Brazil
c
Department of Hydrobiology, Federal University of Sao Carlos-UFSCar, C. Postal 676, 13565-905 Sao Carlos, SP, Brazil
d
Department of Morphology, Federal University of Amazonas-UFA, Av. Gen. Rodrigo O. J. Ramos, 3000, 69077-00 Manaus, AM, Brazil
Received 26 September 2003
Available online 5 November 2004

Abstract

The relative condition factor (Kn), gonadosomatic index (GSI), selected hematological variables and gill morphology of the sh
Astyanax fasciatus were analyzed in two sites (site 1 was unpolluted and site 2 was polluted with untreated domestic sewage) of a
tropical river (Camanducaia river, Sao Paulo State, Brazil). The relationship between the body mass (MB) and the standard length
(LS) of A. fasciatus from both sites was MB=0.00799 L3.51843
S . The Kn values from both males and females and the GSI of females
were signicantly higher in site 2. The mean corpuscular hemoglobin and mean corpuscular hemoglobin concentration were higher
in females from site 2. Gill tissue anomalies and gill parasites were rare in sh from both sites; however, the number of chloride cells
was signicantly higher in sh from site 1. A. fasciatus presents high capacity to live in ion-poor and soft water and is able to
compensate for environmental changes caused by untreated domestic sewage discharges.
r 2004 Elsevier Inc. All rights reserved.

Keywords: Fish; Untreated domestic sewage; Condition factor; Gonadosomatic index; Blood variables; Chloride cells; Characidae

1. Introduction conditions and a high capability to compensate for

Untreated sewage efuents may consist of complex
mixtures of toxicants that may impair the health of sh
and have been considered as an important factor
contributing to the reduction of species diversity in the
world (Schmidt et al., 1999; Amisah and Cowx, 2000).
The tolerance limits to toxicants found in domestic
efuents vary among species and their integrative effects
may lead to reproductive failure and reduction of
survival and growth of the more sensitive sh species
(Waring et al., 1996; Wendelaar Bonga, 1997). Some
species, however, are abundant in degraded waters,
displaying a high tolerance to changes in environmental
Corresponding author. Fax: +55 16 260 8327.
E-mail address: dmnf@power.ufscar.br (M.N. Fernandes).
homeostatic disturbances (Wootton, 1990).
Laboratory tests exposing sh to a given chemical or
physical factor do not reproduce real environmental
conditions and may under- or overestimate effects. Field
studies comparing impacted and nonimpacted areas
permit an evaluation of the health condition of sh in
their own environment, although it is not always
possible to determine with precision the causal agent
of any given organosomatic change. Representative
indices of the whole sh, hematological variables, and
some morphological tracts of individual sh, such as the
gills which are the main organs for gas exchange in most
species and important organs for osmo- and ionoregula-
tion, can be used to provide an indication of how well
sh are coping with their environment (Banerjee and
Homechaudhuri, 1990; Goede and Barton, 1990;
Martinez and Souza, 2002).

0147-6513/$ - see front matter r 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ecoenv.2004.08.009
 ARTICLE IN PRESS
248 A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255
In this context, the present study quantied variations
in the condition factor (Kn), the gonadosomatic index
(GSI), hematological variables, and the gill morphology
of the characid sh Astyanax fasciatus Cuvier of two
sites: one unpolluted area of the Camanducaia River and
a downstream site polluted with domestic sewage. The
later site is representative of other rivers in Brazil that
run throughout small cities (around 50,000 inhabitants)
and receive daily untreated domestic sewage. A. fasciatus
is an abundant (Nomura, 1975; Araujo, 1996) and widely
distributed species across Central and South America
(Sterba, 1973). This species is territorial and is able to
live and reproduce in waters exhibiting a high degree of
variation in physico-chemical characters (Menni et al.,
1996; Garutti and Figueiredo Garutti, 1992).
2. Materials and methods
2.1. Study area
The Camanducaia River is located in the state of Sao
Paulo, Brazil and is approximately 150 km long; 14 km
of the river lie inside the urban area of Amparo
(221420 4300 S and 461450 5200 W), a city of 60,000
Fig. 1. Map of the Camanducaia River, showing the urban area of the Amparo city (inside of dashed line), untreated domestic sewage discharges
(small arrows) into the river and tributaries, sites 1 (unpolluted) and 2 (untreated-domestic-sewage-polluted) of water and sh collection (K), and the
approximate location in the state of Sao Paulo and Brazil.
inhabitants (Fig. 1). The river receives over
12,000,000 L of untreated domestic sewage daily. The
mean water ow is 10.8 m3 s1 during the rainy season
(from December to March) and 4.0 m3 s1 during the
dry season (from August to October). High levels of
sewage efuent concentration characterize the end of the
dry season due to the low level of water (Alberto, 2002).
A. fasciatus specimens and water were collected from
two sites at the end of the dry season (September and
October 20002001): site 1, the reference or unpolluted
area, was located 5 km upstream from the rst sewage
efuent discharge, above the city of Amparo, while site
2, the untreated-sewage-polluted area, was located
downstream from the city, below the sewage discharge
(Fig. 1). No industrial efuents are discharged into the
river upstream of sites 1 and 2. The distance between
sites 1 and 2 was 17 km, interspersed by several
waterfalls, which constitute physical barriers to hinder
the movement of sh from site 2 to site 1. Samples (four)
of sh (50 specimens) and water were collected from the
two sites on September and October 2000 and 2001 (one
sample in each period), always during the afternoon.
Each sampling period, three-gill nets (15, 20, and 25 mm
mesh size) were set stretched at three stations located
50 m apart from one another.
 ARTICLE IN PRESS
A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255
2.2. Water analyses
Dissolved oxygen (DO), conductivity, temperature,
and pH were measured in the eld using a Water Quality
Checker (Horiba). Suspended materials were quantied
by water ltration (Watman GFC lter). Total nitrogen
and total phosphorus concentrations were determined
using unltered water; dissolved and ammonical nitro-
gen, nitrite, dissolved phosphorus, and orthophosphate
were determined after the water was ltered. The
method described by Mackereth et al. (1978) was used
to determine different forms of nitrogen species, while
the method proposed by Golterman et al. (1978) was
used to determine the forms of phosphorus species. The
biochemical oxygen demand (BOD5,20), total phenol
concentration (C6H5OH), surfactants (lauryl alkylben-
zene sulfonate), chloride, manganese, sulde, metals,
total hardness, and fecal coliform concentrations were
determined following the APHA (1989) methodologies.
2.3. Fish analyses
Adult A. fasciatus (wet body mass, MB=21.670.9 g;
standard length, LS=9.470.1 cm; n 72 males and 115
females) from both sites were collected and kept in 250-
L aquaria with their respective river water until they
were sacriced (up to 12 h). A blood sample from each
sh was taken from the caudal vein into heparinized
plastic tubes; sh were then sacriced by medullar
transection and their gonads and gills removed. Blood
subsamples were used for hematological and plasma ion
analyses, and the gills were xed with 2.5% glutaralde-
hyde in 0.1 M phosphate buffer at pH 7.4 for light
microscopy processing.
The relative condition factor (Kn) was calculated
according to the method described by Le Cren (1951) as
Kn=MB  M1 e . MB is the sh wet body mass and Me is
the expected wet mass of the sh based on the
relationship between body mass (MB) and standard
length (LS) for sh collected from both sites and
described as MB=aLbS. The gonads were weighed
immediately after removal to estimate the gonadoso-
matic index [GSI (%)=100  wet gonad mass  total
M1B ] according to Godinho et al. (1974).
Blood analyses were made immediately after blood
sampling. Hematocrit (Hct) was determined by spinning
the blood sample contained in heparinized capillary
tubes in a microhematocrit centrifuge, hemoglobin (Hb)
concentration was determined by the cyanomethemo-
globin method (Drabkin reagent), and red blood count
(RBC) was estimated using a modied Neubauer
chamber after the blood had been diluted in a saline
solution. Mean corpuscular volume (MCV), mean
corpuscular hemoglobin (MCH), and mean corpuscular
hemoglobin concentration (MCHC) were calculated
249
with the blood measurement values following Heath
(1987) and Houston (1990).
Plasma samples were obtained by blood centrifuga-
tion. Plasma osmolarity was determined immediately
using a micrometer mOsmett (Precision System), and
then the remaining plasma was cooled (20 1C) until ion
analyses. The plasma sodium (Na+) concentration was
determined with a Zeiss M4Q2 amephotometer, and
the plasma chloride concentration (Cl) was determined
by the thiocyanate method, using a commercial kit
(Labtest).
2.4. Gill morphometry
Fixed gills from six sh (three males and three females
of similar size) from each site were taken at random for
morphological and morphometric analyses. The four gill
arches from the right side of the head were dehydrated
in graded ethanol solutions up to pure ethanol and
embedded in Historesin (Leica). Sagittal sections (3 mm
thick) from each gill arch, taken at a 50-mm distance
between sections, were stained with toluidine blue,
which allows for colored localization of mucous (pink),
chloride cells (CCs; light blue), pavement cells, and
nuclei (dark blue). The gill tissue and cell morphology
were analyzed under an Olympus-Micronal CBA-K
microscope. Briey, two sections from each gill arch
were randomly sampled and three random laments
from each section were analyzed. The CCs in the
epithelia of laments and lamellae were counted
separately, and the lament and lamellar thickness,
epithelium thickness, and interlamellar distance were
measured in ve adjacent randomly selected interlamel-
lar spaces, according to Jagoe and Haines (1997). The
interlamellar distance was measured at the base of
lamellae. The thickness of the lamellar and lament
epithelium was measured from the surface in contact
with external water to the respective basement mem-
brane. The counts and measurements were done using a
video camera connected to an Olympus microscope
CBA-K and an IBM microcomputer, in which a system
of test points (Merz, 1967) was superimposed over the
video monitor to ensure random orientation of the
measurements.
2.5. Statistical analysis
Data are presented as mean7standard error of the
mean (SE). All the tests used compared males from sites
1 and 2 and females from sites 1 and 2 and were
performed using the GraphPad InStat software (San
Diego, CA). The KolmogorovSmirnov test was applied
to verify the normal distribution of the data, the F test
was applied to verify the homogeneity of the two
variances, and the t-test was used to detect differences in
means between populations. When the assumptions of
 ARTICLE IN PRESS
250 A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255

this parametric test were violated, the nonparametric
MannWhitney U test was used to detect differences in
the groups of sh from different sites.
3. Results
Water temperature, pH, and the concentrations of
sulde and aluminium did not differ between sites
(Table 1). DO at site 2 was approximately half that
measured at site 1 (Table 1). All other variables were
higher at site 2. Most notably, the fecal coliform count
was 10 times higher at site 2 than at site 1. Ammoniacal
nitrogen was 13 times higher at site 2 than at the
unpolluted site 1; total phosphorus was 5 times higher
and the BOD5,20 value was almost 2 times higher than at
the unpolluted site. Conversely, the DO measured at site
2 was half (po0:05) of that measured at site 1 (Table 1).
The relationship between the wet mass and the
standard length of male and female A. fasciatus
combined across sites was described by the equation
MB=0.008 L3.518
S (Fig. 2a). Fig. 2b shows the individual
Kn values of sh from sites 1 and 2 in relation to LS of
the sh. The mean Kn values of males (Kn=0.9070.01)
and females (Kn=0.9670.01) from site 1 were signi-
cantly lower (po0.05) than those of males
(Kn=1.0370.03) and females (Kn=1.1270.02) from
site 2. The mean GSI of males did not differ between
sites (site 1 GSI=2.570.2% and site 2 GSI=2.47
0.3%; t-test, pX0.05); however, the mean GSI of
females from site 1 (GSI=7.570.9%) was lower
(t-test, po0.05) than that of females from site 2
(GSI=10.870.8%).
The mean values of the measured red blood cell
variables and plasma osmolarity and the Na+ and Cl
concentrations of male and female A. fasciatus are
shown in Table 2. No signicant differences between sh
from sites 1 and 2 were found in these variables except
for the blood index values, MCH and MCHC, which
were signicantly higher (po0.05) in females from site 2.
However, no correlations between these indices and the
GSI of females were found (Pearson coefcients were
r 0:17 and r 0:01 for the correlations between GSI
and MCH and between GSI and MCHC, respectively).
The epithelium of the gill lament of male and female
A. fasciatus had four to ve cell layers. The outermost
layer consisted of pavement cells, with a scattering of
mucous and chloride cells among them. Lamellae (100
lamella/mm of lament length) were regularly spaced on
both sides of the lament. The lamellar epithelium
consisted of two cell layers: the innermost cell layers
were undifferentiated cells and the outermost cell layers
consisted of pavement and chloride cells. Rare parasites
were found in the gill tissue of sh from both sites and,

Table 1
Mean values7SE of water variables from unpolluted (site 1) and untreated-domestic-sewage-polluted (site 2) sites on the Camanducaia River,
Amparo, SP, Brazil

Water variables Site 1 Site 2 Statistical signicance

1
Dissolved oxygen (mg L ) 7.270.1 3.670.3** po0.0001
BOD5,20 (mg L1) 5.570.3 10.570.3* po0.001
Fecal coliforms (MPN 100 mL1) 4.3  104733 4.3  1057333* po0.0001
Total phosphorus (mg L1) 5573 252715* po0.0001
Dissolved phosphorus (mg L1) 3470.5 15675.5* po0.0001
Orthophosphate (mg L1) 20.570.6 7574.9* po0.001
Total nitrogen (mg L1) 0.870.07 370.04* po0.0001
Dissolved nitrogen (mg L1) 0.470.05 2.370.04* po0.0001
Ammoniacal nitrogen (mg L1) 7.870.4 10471.2* po0.0001
Nitrite (mg L1) 15.270.5 29.970.4* po0.0001
Suspended materials (mg L1) 4.470.5 10.270.3* po0.001
Conductivity (mS cm1) 5370.01 85.370.3* po0.0001
Total hardness (mg L1) 1671.2 2771.8* po0.001
Chloride (mg L1) 10.270.7 14.970.4* po0.0001
Manganese (mg L1) 4472.6 70710.8** po0.05
Iron (mg L1) 954722 1430744* po0.0001
Aluminium (mg L1) 1974.4 1971.4NS pX0.05
Surfactants (mg L1) 0.0470.0003 0.19570.002* po0.0001
Total phenols C6H5OH (mg L1) 0.00370.00006 0.00870.00006* po0.0001
Sulde (mg L1) o0.5 o0.5
PH 7.170.1 6.970.1NS pX0.05
Temperature (1C) 2172 2372NS pX0.05

*Students t-test.
**MannWhitney U test.
NS
not statistically different.
MPN, most probable number.
 ARTICLE IN PRESS
A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255 251

when present, the hosts reaction consisted of prolifera-
tion of epithelial cells forming a local fusion of two to
three adjacent lamellae. Gill tissue anomalies such as
epithelial lifting of lament and lamellar epithelium,
epithelial rupture, necrosis, and blood vessel anomalies
such as lamellar telangiectasis and aneurysms were also
rare. Representative light micrographs of A. fasciatus
gills from sites 1 and 2 (Figs. 3A and B) illustrate the
general morphological appearance of the gill laments
and lamellae. Marked differences between sh of both
sites but not between males and females from the same
site were found in the CC number. Pale-blue-stained
CCs clearly protruded from the entire surface of the gill
epithelium in sh from the unpolluted reference site (site
1) (Fig. 3C), while the CCs of sh from the sewage-
polluted site (site 2) were smaller and less numerous
(Fig. 3D). Quantitative data revealed a signicantly
higher number of CCs (po0.05) in sh from site 1 in the
lamellar epithelium of all gill arches and in the lament
epithelium of the rst and second gill arches (Fig. 4).
The thickness of the gill lamellae of sh from the
untreated-domestic-sewage-polluted site (site 2) was
signicantly lower (po0.05) than of sh from the
unpolluted reference site (site 1) (Table 3) as a
consequence of the lamellar epithelium thickness that
was signicantly lower in the rst and second gill arches
(po0.05). In the third and fourth gill arches no
signicance difference between sh from both sites was
found in the lamellar epithelium thickness although this
variable was slightly lower in sh from site 2. The
distance between adjacent lamellae was signicantly
higher in the rst and second gill arches of sh from
site 2.

4. Discussion

A. fasciatus showed high tolerance to the sewage-

Fig. 2. Relationship between the body mass (MB) and the standard
length (LS) (a) and the relative condition factor (Kn) plotted against
standard length (b) of males (n 72) and females (n 115) of A.
fasciatus from sites 1 and 2 of the Camanducaia River, Amparo, SP,
Brazil.
polluted site. This site was characterized by a high
concentration of fecal coliforms, nitrogen, phosphorus,
surfactant, phenol, and suspended material in the water
and a low concentration of dissolved oxygen relative to
those characters of the unpolluted site (Table 1). The
high BOD5,20 values found in the untreated-domestic-
sewage-polluted site explains the reduction in DO, a
situation that is aggravated by the high temperatures of
water in the tropical region.

Table 2
Mean values7SE of selected blood variables of red blood cells and plasma of the red tail lambari, A. fasciatus, from unpolluted (site 1) and
untreated-domestic-sewage-polluted (site 2) sites of the Camanducaia River, Amparo, SP, Brazil

Males Females

Site 1 (n 18) Site 2 (n 13) Site 1 (n 20) Site 2 (n 26)

NS
Hct (%) 4071.5 3771.4 3971.3 4071.0NS
Hb (g 100 mL1) 9.570.5 9.470.6NS 8.970.5 9.970.3NS
RBC (105 mm3) 2671 2471NS 2571 2676NS
MVC (mm3) 16078 15677NS 15676 15374NS
MHC (pg cell1) 3671.4 38.370.6NS 3471.5 3871*
MCHC (%) 2371 2571.3NS 2371 2571*
Osmolarity (mOsm KgH2O1) 26079 236713NS 24578 23977NS
Na+ (mEq L1) 127713 12376NS 12275.4 11177.6NS
Cl (mEq L1) 9075 8976NS 9174 8175NS

Students t-test.
*po0.05.
NS
not statistically different.
 ARTICLE IN PRESS
252 A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255

Fig. 3. Representative gills laments of A. fasciatus from unpolluted site 1 (a) and untreated-domestic-sewage-polluted site 2 (b) of the Camanducaia
River, Amparo, SP, Brazil. Note the numerous chloride cells (arrows) in the lamellar epithelium in (a) and (c) and the few chloride cells in (b) and (d).
High magnication of lamella, showing the chloride cells in the lamellar epithelium of A. fasciatus from site 1 (c) and site 2 (d). L, lamella; F, lament;
PVC, pavement cell. Scale bar in mm.

human food fragments were found in the digestive tract

Fig. 4. Chloride cell number in the lamellar and lament epithelia of
ve adjacent interlamellar spaces of 16 gill lament of A. fasciatus
(n 6 animals) from unpolluted site 1 (black bars) and untreated-
domestic-sewage-polluted site 2 (open bars) of the Camanducaia
River, Amparo, SP, Brazil. Values are mean7SE; * indicates a
signicant difference (po0.05, MannWhitney U test) between the two
sites.
The high Kn values of sh from site 2, particularly of
females, indicate the better condition of sh from the
untreated-domestic-sewage-polluted location. High Kn
is usually interpreted as high reserves of energy in sh,
reecting abundant food availability. Large amounts of
of sh from site 2 (personal observation). In general,
organically polluted environments show high produc-
tivity levels (Colinvaux, 1993) and the few species, such
as A. fasciatus in the present study, which persist in such
environments may become abundant.
A positive relationship between Kn and GSI has been
reported for sh (Barbieri and Verani, 1987); however,
in our study the population trends did not support the
results of these authors because the GSI of females and
their relatives Kns were higher at the polluted site, but
the GSI of males did not differ between the polluted and
the control sites. Several eld studies have demonstrated
high Kn value and a disruption (generally downward) in
sh GSI in a number of contaminated areas (Greeley,
2002). The evaluation of Kn and the GSI of A. fasciatus
from two rivers heavily loaded with organic and
industrial sewage revealed a negative relationship
between Kn and GSI due to the smaller oocyte diameter
of females from these polluted rivers (Schulz and
Martins-Junior, 2001). In this study, the high levels of
heavy metals from leather processing and metallurgical
factories and the extremely low DO (0.2 mgO2 L1;
84 mg L1 BOD5,20) were considered the major reasons
for the decrease in GSI. In the present study, the
 ARTICLE IN PRESS
A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255
Table 3
Mean value7SE of gill morphometry of lambari, A. fasciatus (n 6) from unpolluted (site 1) and untreated-domestic-sewage-polluted (site 2) sites of
the Camanducaia River, Amparo, SP, Brazil
Gill measurements Site 1
Gill arches
I and II
Filament thickness (mm) 28.870.9
(107)
Lamellar thickness (mm) 8.370.1
(542)
Lamellar epithelium thickness (mm) 2.670.1
(596)
Interlamellar distance (mm) 4.870.2
(226)
The sample size is presented in parentheses. MannWhitney U test.
*po 0.05.
NS
not statistically different.
contamination of water with untreated domestic sewage
did not evidence any inuence on the GSI of A. fasciatus
as the GSI of females increased and no changes in males
from both sites were found.
A. fasciatus from the unpolluted and polluted sites
also differed with respect to characters related to
respiration and ionoregulation. Most Brazilian fresh-
waters are ion-poor and soft, which is the case of the
Camanducaia River water, implying the morphofunc-
tional adaptation of sh gills to optimize ion uptake
(Wood et al., 1998; Wood, 2001; Fernandes and Perna-
Martins, 2001,2002; Gonzalez et al., 1998, 2002; Moron
et al., 2003) that, conversely, may compromise the
effectiveness of respiration (Greco et al., 1995; Sakur-
agui et al., 2003).
Untreated domestic sewage discharge inuences the
DO concentration in water and may serve as an
important stress agent for aquatic animals. Reduced
DO should elicit physiological regulatory mechanisms
involved in the maintenance of the O2 gradient from
water to tissues, which is essential to maintain the
metabolic aerobic pathways (Hughes, 1973; Fernandes
and Rantin, 1994). The rst responses of sh to
environmental hypoxia (low DO levels) are changes in
the ventilation (tidal volume and respiratory frequency)
and cardiovascular (stroke volume, heart frequency, and
blood pressure) functions stimulated by the catechola-
mines released from both adrenergic nerves and
chromafn tissue, where they are synthesized and
stored. These responses result in the paradoxical
situation wherein the increase of lamellar perfusion
favors the uptake of O2, which concomitantly may
increase the passive diffusion of the ion into water,
leading to a hydrolytic imbalance (Heath, 1987)
mainly if low calcium is available such as in soft
water. Furthermore, other compensatory responses
253
Site 2
Gill arches
III and IV I and II III and IV
2671 3171.2NS (98) 29.071.4NS
(74) (98) (109)
8.970.3 7.770.1* 7.470.1*
(628) (548) (591)
2.070.1 1.970.1* 1.770.1NS
(648) (711) (534)
5.870.2 5.470.2* 5.970.2NS
(228) (298) (262)
must be elicited to maintain homeostasis, including
morphological changes of some organs and tissues.
The evaluation of hematological variables of A.
fasciatus from the untreated-domestic-sewage-polluted
site (site 2) did not provide evidence of physiological
changes to increase the O2 carrying capacity and
maintain the O2 ow from water to tissue in such an
hypoxic environment. Also, no changes in the blood
variables, including methemoglobin, were found in A.
altiparanae exposed to nitrite, a water variable that
usually may increase in domestic sewage (Martinez and
Souza, 2002). The signicantly higher blood indices,
MCH and MCHC, of females from site 2 may be related
to the higher O2 demand of the reproductive tissue in
these females, since the GSI was signicantly higher in
these females than in those from site 1, although no
correlation was found between MCH, MCHC, and GSI.
The mean Hb concentration values obtained in this
study for A. fasciatus were generally in the upper limit of
the range reported for other tropical sh (Table 4),
suggesting a high capacity for efcient oxygen uptake
that facilitates persistence in hypoxic waters. The
absence of hydrolytic imbalance given by plasma
osmolarity and ion concentration of A. fasciatus from
both sites also suggests high ion and osmoregulatory
efciency in this species.
The gills provide evidence that A. fasciatus may
alternately optimize the ionoregulatory and respiratory
functions in response to water characteristics of the sites.
High CC density in the gill epithelium of sh from ion-
poor and soft water favors ion uptake in very diluted
freshwater (Perry, 1997). Conversely, the CCs in the
lamellar epithelium increase the waterblood diffusion
distance (Bindon et al., 1994; Moron et al., 2003),
reducing the effectiveness of the respiratory gas exchange
(Greco et al., 1995; Perry, 1998; Sakuragui et al., 2003).
 ARTICLE IN PRESS
254 A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255

Table 4 untreated domestic sewage pollution and the physiolo-

Mean value7SE of hemoglobin concentration (Hb) of blood from
gical and morphological differences observed in this
some freshwater tropical sh
study will be important in generalizing these results.
Species Hb (g 100 mL1) References

Hypostomus regani 4.570.2 Fernandes et al.

Ihering (1999) Acknowledgments
Colossoma 5.670.3 Ranzani-Paiva et al.
macropomum Cuvier (1999) Financial support was provided by the Brazilian
Piaractus 8.170.2 Ranzani-Paiva et al.
research funding agency CAPES. The authors are
mesopotamicus (1999)
Holmberg grateful to Rosa Daolio, Emlio C. Nardini, and
Prochilodus scrofa 8.670.4 Ranzani-Paiva et al. Ricardo E. dos Santos, from the city of Amparos
Steindachner (1999) Autonomous Water and Sewage Service, and to Carlos
Prochilodus lineatus 9.770.2 Ranzani-Paiva et al. Sanches and Sandra Veloso, from the Laboratory of
Valenciennes (2000)
Schizodon borellii 7.570.1 Ranzani-Paiva et al.
Aquatic Ecology of UNESP/Rio Claro, for the water
Boulenger (2000) analyses. A. Alberto thanks CAPES for the award of a
Serrasalmus 7.670.4 Ranzani-Paiva et al. fellowship.
marginatus (2000)
Valenciennes
Pimelodus maculatus 7.370.8 Ranzani-Paiva et al. References
Lacepede (2000)
Pseudoplatystoma 4.970.5 Ranzani-Paiva et al.
Alberto, A., 2002. Morfologia branquial, hematologia e fator de
corruscans Agassiz (2000)
condic- ao relativo (Kn) do lambari Astyanax fasciatus Cuvier, 1819
Hoplias malabaricus 6.670.2 Ranzani-Paiva et al.
(Teleostei, Characidae) do rio Camanducaia, regiao de Amparo,
Bloch (2000)
SP. Aspectos sioecologicos associados a poluic- ao organica. MS
Parauchenipterus 7.070.2 Ranzani-Paiva et al.
Thesis, Federal University of Sao Carlos, Sao Carlos, SP, Brazil.
galeatus Linnaeu (2000)
Amisah, S., Cowx, I.G., 2000. Response of the sh populations of the
Astyanax fasciatus 9.470.5 Present study
River Don in South Yorkshire to water quality and habitat
Cuvier
improvements. Environ. Poll. 108, 191199.
APHA, 1989. Standard Methods for the Examination of Water and
Wastewater, 17 ed. American Public Health Association, Wa-
shington.
The higher CC density in sh from site 1 suggests a Araujo, F.G., 1996. Composic- ao e estrutura da comunidade de peixes
response to increase ion uptake as the ion concentration do medio e baixo rio Paraba do sul, RJ. Rev. Bras. Biol. 56,
in water was lower and the dissolved O2 was higher than 111126.
those in site 2. On the other hand, the gill epithelium Banerjee, S., Homechaudhuri, S., 1990. Hematological monitoring of a
bio-indicator sh, Heteropneustes fossilis, on exposure to copper
morphology of sh from site 2 suggests a response to toxicity. Israel J. Aquacult. Bamiggeh 42, 4651.
optimize the oxygen uptake as dissolved O2 is very low. Barbieri, G., Verani, J.R., 1987. O fator de condic- ao como indicador
The greater interlamellar distance and reduced epithe- do perodo de desova em Hypostomus aff. plecostomus (Linnaeus,
lium thickness in the gill lamellae of A. fasciatus from 1758) (Osteichthyes Loricariidae), na represa do Monjolinho (Sao
site 2 may reduce the resistance of water ow and favor Carlos, SP). Cienc. Cult. 39, 655658.
Bindon, S.D., Fenwick, J.C., Perry, S.F., 1994. Branchial chloride cell
the diffusion of O2 from water to blood, as pointed out proliferation in the rainbow trout, Oncorhynchus mykiss: implica-
by Hughes (1973). Onconrhynchus mykiss in hypoxic tions for gas transfer. Can. J. Zool. 72, 13951402.
environments increases its respiratory capability by Colinvaux, P., 1993. Ecology, vol. 2. Wiley, New York.
increasing the number of perfused lamellae and reducing Fernandes, M.N., Perna-Martins, S.A., 2001. Epithelial gill cells in the
armoured catsh, Hypostomus plecostomus (Loricariidae). Brazil J.
the waterblood diffusion barrier (Soivio and Tuurala,
Biol. 61, 6978.
1981). The substantially greater interlamellar distance Fernandes, M.N., Perna-Martins, S.A., 2002. Chloride cell responses
found in the two rst gill arches of A. fasciatus suggests to long-term exposure to distilled and hard water in the gill of the
that a larger volume of water passes through the armored catsh, Hypostomus tietensis (Loricariidae). Acta Zool.
interlamellar space in these gill arches, which may 83, 321328.
facilitate gas transfer from hypoxic water to blood. Fernandes, M.N., Rantin, F.T., 1994. Relationships between oxygen
availability and metabolic cost of breathing in Nile tilapia
In conclusion, because pollution constitutes a selector (Oreochromis niloticus): aquacultural consequences. Aquaculture
agent of the ichthyofauna in rivers and lakes our results 127, 339346.
demonstrate that the A. fasciatus has high capacity to Fernandes, M.N., Sanches, J.R., Matsuzaki, M., Panepucci, L.,
live in ion-poor and soft water and is able to compensate Rantin, F.T., 1999. Aquatic respiration in facultative air-breathing
for environmental changes caused by untreated domes- sh: effects of temperature and hypoxia. In: Val, A.L., Almeida-
Val, V.M.F. (Eds.), Biology of Tropical Fishes. INPA, Manaus,
tic sewage discharges. Future studies that have replica- pp. 341352.
tions of polluted and unpolluted sites and experimental Garutti, V., Figueiredo Garutti, M.L., 1992. Caracterizac- ao de
assays to directly test the relationship between the populac- oes do lambari, Astyanax bimaculatus (Pisces, Characidae),
 ARTICLE IN PRESS
A. Alberto et al. / Ecotoxicology and Environmental Safety 61 (2005) 247255
procedentes do Campus de Jaboticabalm UNESP, SP. Naturalia
17, 109119.
Godinho, H.M., Fenerich, N.A., Martins, M.A.B., 1974. Maturation
curve of ovary of Pimelodus maculates. Bol. Inst. Pesca 3, 120.
Goede, R.W., Barton, B.A., 1990. Organismic indices and an autopsy-
based assessment as indicators of health and condition of sh. In:
Adams, S.M. (Ed.), Biological Indicators of Stress in Fish. Amer.
Fish. Symp. 8, Bestheda, pp. 93108.
Golterman, H.L., Clymo, R.S., Ohstad, M.A.M., 1978. Methods for
Physical and Chemical Analysis of Fresh Waters. vol. 8, second ed.
IBP, Oxford.
Gonzales, R.J., Wood, C.M., Wilson, R.W., Patrick, M.L., Bergman,
H., Narahara, A., Val, A.L., 1998. Effects of water pH and Ca2+
concentration on ion balance in sh of the Rio Negro, Amazon.
Physiol. Zool. 71, 1522.
Gonzales, R.J., Wilson, R.W., Wood, C.M., Patrick, M.L., Val, A.L.,
2002. Diverse strategies for ion regulation in sh collected
from the ion-poor, acidic Rio Negro. Physiol. Biochem. Zool. 75,
3747.
Greco, A.M., Gilmour, K.M., Fenwick, J.C., Perry, S.F., 1995. The
effects of softwater acclimation on respiratory gas transfer in the
rainbow trout Oncorhynchus mykiss. J. Exp. Biol. 198, 25572567.
Greeley, M.S., 2002. Reproductive indicators of environmental stress
in sh. In: Adams, S.M. (Ed.), Biological Indicators of Aquatic
Ecosystem Stress. American Fish. Soc., Bethesda, pp. 321377.
Heath, A.G., 1987. Water Pollution and Fish Physiology. CRC Press,
Florida.
Houston, A.H., 1990. Blood and Circulation. In: Schreck, C.B.,
Moyle, P.B. (Eds.), Methods for Fish Biology. American Fisheries
Society, Bethesda, pp. 273334.
Hughes, G.M., 1973. Respiratory responses to hypoxia in sh. Am.
Zool. 13, 475489.
Jagoe, C.H., Haines, T.A., 1997. Changes in gill morphology of
Atlantic salmon (Salmo salar) smolts due to addition of acid and
aluminium to stream water. Environ. Poll. 97, 137146.
Le Cren, E.D., 1951. The length-weight relationship and seasonal cycle
in gonad weight and condition in the perch (Perca fluviatilis).
J. Animal Ecol. 20, 201209.
Mackereth, F.J.H., Heron, J., Talling, J.F., 1978. Water Analysis:
Some Revised Methods for Limnologists, vol. 36. Scientic
Publication, Cumbria.
Martinez, C.B.R., Souza, M.M., 2002. Acute effects of nitrite on ion
regulation in two neotropical sh species. Comp. Biochem. Physiol.
133A, 151160.
Menni, R.C., Gomes, S.E., Armengol, F.L., 1996. Subtle relationships:
freshwater sh and water chemistry in southern South America.
Hydrobiology 328, 173197.
Merz, W.A., 1967. Die Streckenmessungen an gerichteten Strukturen
im Mikroskop und ihr Anwendung zur Bestimmung von Ober-
achen-Volumen-Relationen im Knochengewebe. Mikroskopie 22,
132142.
255
Moron, S.E., Oba, E.T., Andrade, C.A., Fernandes, M.N., 2003.
Chloride cell responses to ion challenge in two tropical freshwater
sh, the erythrinids Hoplias malabaricus and Hoplerythrinus
unitaeniatus. J. Exp. Zool. 298A, 93104.
Nomura, H., 1975. Comparac- ao da idade e do crescimento de tres
especies de peixes do genero Astyanax Baird & Girard, 1854
(Osteichthyes, Characidae) do rio Mogi Guac- u, SP. Rev. Bras.
Biol. 35, 531547.
Perry, S.F., 1997. The chloride cell: structure and function in the gills
of freshwater shes. Annu. Rev. Physiol. 59, 325347.
Perry, S.F., 1998. Relationships between branchial chloride cells and
gas transfer in freshwater sh. Comp. Biochem. Physiol. 119A,
916.
Ranzani-Paiva, M.J.T., Salles, F.A., Eiras, J.C., Eiras, A.C., Ishikawa,
C.M., Alexandrino, A.C., 1999. Analises hematologicas de
curimbata (Prochilodus scrofa), pacu (Piaractus mesopotamicus) e
tambaqui (Colossoma macropomum) das estac- oes de piscicultura do
instituto de pesca, estado de Sao Paulo. Bol. Inst. Pesca 25, 7783.
Ranzani-Paiva, M.J.T., Souza, A.T.S., Pavanelli, G.C., Takemoto,
R.M., Eiras, A.C., 2000. Hematological evaluation in commercial
sh species from the oodplain of the upper Parana River, Brazil.
Acta Scient. 22, 507513.
Sakuragui, M.M., Sanches, J.R., Fernandes, M.N., 2003. Gill chloride
cells proliferation and respiratory responses to hypoxia of the
neotropical erythrinid sh Hoplias malabaricus. J. Comp. Physiol.
B 173, 309317.
Schmidt, H., Bernet, D., Wahli, T., Meier, W., Burkhardt-Holm, P.,
1999. Active biomonitoring with brown trout and rainbow trout in
diluted sewage plant efuents. J. Fish Biol. 54, 585596.
Schulz, U.H., Martins-Junior, H., 2001. Astyanax fasciatus as
bioindicator of water pollution of rio dos Sinos, RS, Brazil. Braz.
J. Biol. 61, 615622.
Soivio, A., Tuurala, H., 1981. Structural and circulatory responses to
hypoxia in the secondary lamellae of Salmo gairdneri gills at two
temperatures. J. Comp. Physiol. 145, 3743.
Sterba, G., 1973. Freshwater Fisher of the World. T.F.H. Publications,
New York.
Waring, C.P., Stagg, R.M., Fretwell, K., Mclay, H.A., Costello, M.J.,
1996. The impact of sewage sludge exposure on the reproduction of
the sand goby, Pomatoschistus minutus. Environ. Pollut. 93, 1725.
Wendelaar Bonga, S.E., 1997. The stress response in sh. Physiol. Rev.
77, 591625.
Wood, C.M., 2001. Toxic responses of the gill. In: Schlenk, D.S.,
Benson, W.H. (Eds.), Targer Organ Toxicity in Marine and
Freshwater Teleosts, vol 1. Taylor & Francis, London, pp. 189.
Wood, C.M., Wilson, R.W., Gonzalez, R.J., Patrick, M.L., Bergman,
H., Narahara, A., Val, A.L., 1998. Responses of an Amazonian
teleost, the tambaqui (Colossoma macropomum) to low pH in
extremely soft water. Physiol. Zool. 71, 658670.
Wootton, R.J., 1990. Ecology of Teleost Fishes. Chapman & Hall,
London.