The toxicity of cadmium to three aquatic organisms (Photobacterium
phosphoreum, Daphnia magna and Carassius auratus) under
different pH levels

R.-J. Qu, X.-H. Wang, M.-B. Feng, Y. Li, H.-X. Liu, L.-S. Wang, Z.-Y. Wang

State Key Laboratory of Pollution Control and Resources Reuse, School of Environment, Xianlin Campus, Nanjing University, Nanjing 210023,
Jiangsu, PR China

a r t i c l e i n f o

Article history:
Received 27 February 2013
Received in revised form
15 May 2013
Accepted 16 May 2013

Available online 12 June 2013

Keywords:
Cadmium
Toxicity
pH
Photobacterium phosphoreum
Daphnia magna
Carassius auratus

a b s t r a c t

This study investigated the effect of pH on cadmium toxicity to three aquatic organisms: Photobacterium
phosphoreum, Daphnia magna and Carassius auratus. The acute toxicity of Cd

to P. phosphoreum and

D. magna at

ﬁve pH values (5.0, 6.0, 7.0, 8.0, and 9.0) was assessed by calculating EC

values.

We determined that Cd

was least toxic under acidic conditions, and D. magna was more sensitive to

the toxicity of Cd than P. phosphoreum. To evaluate Cd

-induced hepatic oxidative stress in C. auratus at

three pH levels (5.0, 7.25, 9.0), the activity of antioxidant enzymes (superoxide dismutase, catalase and
glutathione peroxidase), the level of glutathione and the malondialdehyde content in the liver were
measured. Oxidative damage was observed after 7 d Cd exposure at pH 9.0. An important

ﬁnding of the

current research was that Cd

was generally more toxic to the three test organisms in alkaline

environments than in acidic environments.

1. Introduction

The global increase in freshwater contamination by numerous

natural and industrial chemical compounds is a major environ-
mental problem in the world (

Schwarzenbach et al., 2006

). Heavy

metals are important contaminants of aquatic environments
worldwide. Among these metals, cadmium has received consider-
able attention in recent years because its concentration in water
body has been markedly increased by human activities such as
sewage treatment, production of pulp and paper, and processing
of metals (

Hare, 1992

). As a nonessential element, cadmium

may endanger the growth and development of aquatic life. For
example, cadmium may inhibit the bioluminescence of bacteria
(

Ishaque et al., 2006

), cause limited activity even death in

daphnias (

Canizares-Villanueva et al., 2000

), and induce oxidative

stress in

ﬁsh (

Livingstone, 2001

). The toxicity of cadmium in

contaminated ecosystems depends not only on the concentration
of this metal but also on the water chemistry. An important
environmental stressor that affects most chemical and biological
processes in water is pH. The pH of aquatic systems can be
decreased or increased by a variety of anthropogenic sources,
including agriculture, urbanization, industry, and mining (USEPA,

http://www.epa.gov/caddis/ssr_ph4s.html

). Fluctuations in pH may

lead to changes in cadmium speciation, thereby in

ﬂuencing its

bioavailability and toxicity to exposed organisms. Thus, it is of
signi

ﬁcance to study cadmium toxicity to different aquatic species

as a function of pH.

In aquatic toxicological studies, bacteria, daphnids and

ﬁsh are

the most frequently used test species (

Farre and Barcelo, 2003

These organisms represent different trophic levels in the aquatic
food chain and are capable of re

ﬂecting the water quality.

A bioluminescence inhibition assay using the marine bacterium
Photobacterium phosphoreum is often chosen as the

ﬁrst toxicity

screening method in a test battery because it is fast and cost
effective (

;

). In the assay, light production is directly proportional to the

metabolic activity of the bacterial population, and inhibition of
enzymatic activity causes a corresponding decrease in lumines-
cence intensity. Recently, this simple and sensitive biotest has
been widely used to investigate the toxicity of various inorganic
and organic compounds in water samples (

Ren and Frymier, 2005

;

Katritzky et al., 2010

Acute toxicity testing using freshwater daphnids, particularly

Daphnia magna, is a popular bioassay used internationally for
toxicity screening of chemicals and for monitoring of ef

ﬂuents and

contaminated waters (

Persoone et al., 2009

). D. magna has been

recommended as a standard test organism by many international
organizations (e.g., ISO and OECD). The use of D. magna has many

Contents lists available at

SciVerse ScienceDirect

journal homepage:

www.elsevier.com/locate/ecoenv

Ecotoxicology and Environmental Safety

0147-6513/$ - see front matter

http://dx.doi.org/10.1016/j.ecoenv.2013.05.020

Corresponding author. Fax:

+86 25 89680358.

E-mail address:

wangzun315cn@163.com (Z.-Y. Wang)

Ecotoxicology and Environmental Safety 95 (2013) 83

–90

advantages, such as a high sensitivity and a short reproductive
cycle. Since its initial application in 1928, D. magna has been used
routinely in toxicological studies (

Biesinger and Christensen, 1972

;

Hermens et al., 1984

;

De Schamphelaere et al., 2004

Fish are an indispensable component of integrated toxicity

testing of aquatic environments. The prominence of

ﬁsh in

environmental risk assessment is demonstrated by several toxicity
tests in the OECD guidelines. In particular, the

ﬁsh acute toxicity

test is a mandatory component of the base set of data require-
ments for ecotoxicity testing (

Lammer et al., 2009

). However, in

routine acute tests with mortality as the endpoint, it is dif

ﬁcult to

evaluate the physiological changes that occur in experimental

ﬁsh.

Thus, in the present study, we evaluated contaminant-induced
oxidative stress in

ﬁsh, which may ultimately lead to cell death.

Several oxidation-related biomarkers of the liver, including the
activity of antioxidant enzymes such as superoxide dismutase
(SOD), catalase (CAT) and glutathione peroxidase (GPx), the level
of nonenzymatic antioxidant glutathione (GSH), and the concen-
tration of malondialdehyde (MDA), were measured in the gold

ﬁsh

Carassius auratus. This freshwater

ﬁsh was chosen as the test

animal due to its extensive distribution in China and the wide-
spread use in ecotoxicological researches (

;

To the best of our knowledge, previous toxicity testing con-

cerning the in

ﬂuence of pH on cadmium toxicity was either

performed within a narrow pH range or was limited to only a
single species. These shortcomings prompted us to conduct the
current study. By investigating the toxicity of cadmium to three
aquatic organisms (P. phosphoreum, D. magna and C. auratus)
across a relatively wide pH range (5.0

–9.0), we were able to

characterize the contribution of pH to cadmium toxicity in
different organisms. The results may provide useful information
for evaluating the toxicological effects of Cd in various environ-
ments with different pH.

2. Materials and methods

2.1. Chemicals and instruments

Chemicals: Cadmium sulfate, hydrochloric acid and sodium hydroxide, bought

from Sinopharm Chemical Reagent Co., Ltd., are of analytical grade. 3-(N-morpho-
lino) propanesulfonic acid (MOPS) with a purity of 99% was supplied by Aladdin

Reagent. The kits for the analysis of oxidative stress biomarkers were purchased
from Nanjing Jiancheng Bioengineering Institute.

Instruments: METTLER-TOLEDO S20 SevenEasy pH Meter (METTLER-TOLEDO,

China), Tecan In

ﬁnite 200

PRO multimode microplate reader (Tecan, Switzerland),

PRX-250B Intelligent Arti

ﬁcial Climate Chamber (Safe，China), Eppendorf 5417R

centrifuge (Eppendorf, Germany), IKA T10 homogenizer (IKA, Germany), TU-1800
UV

–vis spectrophotometer (Persee, China), and Atomic absorption spectrophoto-

metry (SOLLAR M6, Thermo, USA).

2.2. Test species and corresponding treatments

Freeze-dried powder of P. phosphoreum (T3 mutation) was obtained from the

Institute of Soil Science, Chinese Academy of Sciences (Nanjing, China). After
injection of 0.5 mL of cold sterilized 2.0% NaCl solution into a vial containing 0.5 g
of freeze-dried powder, the solution was mixed thoroughly by shaking.
After 2 min, P. phosphoreum was revived, and 10

μL of the bacterial liquid was

diluted with 2 mL of 3.0% NaCl solution to serve as the working

ﬂuid for

subsequent tests.

The D. magna strain was supplied by the Research Center for Eco-

Environmental Sciences, Chinese Academy of Sciences (Beijing, China). Tap water
that had been adsorbed by activated carbon and aerated for more than 48 h was
used as culture water. Daphnids were kept in the culture water (pH 7.25

70.25) in a

14 h: 10 h light: dark cycle at 20

1C, and were fed daily with green algae,

Scenedesmus obliquus. Juvenile

ﬂeas that had undergone three generations of

parthenogenesis (6

–24 h old) were used in the experiment.

C. auratus (weight: 30.15

74.35 g; length: 13.870.9 cm) were purchased from

a local aquatic breeding center. Before the experiments, the gold

ﬁsh were

acclimatized in tanks containing 150 L dechlorinated and aerated water at

71 1C for 10 days. The water used for acclimation and subsequent experiments

had a pH of 7.25

70.25, conductivity of 340.6716.4 μs/cm, total hardness of

135.5

79.3 mg CaCO

/L, and alkalinity of 40.7

75.2 mg CaCO

/L. The following

ion levels were measured: Na

, 11.2

70.2 mg/L; K

, 2.34

70.07 mg/L; Mg

7.74

70.02 mg/mL; Ca

, 41.07

70.82 mg/L and Cl

−

, 28.3

71.2 mg/L. The aquaria

were aerated with air stones attached to an air compressor to saturate with oxygen
(6.76

70.84 mg O

/L). The

ﬁsh were fed twice a day with commercial pellets. The

experiments were initiated when the total mortality fell to below 1%.

2.3. Experimental design

2.3.1. P. phosphoreum

The test was carried out according to the National Standard Method of China

(Water quality

– Determination of the acute toxicity – Luminescent bacteria test.

GB/T15441-1995). The metal stock was prepared from 3CdSO

O by dissolving

6.016 g in 3.0% NaCl solution, resulting in 2.637 g Cd/L. Based on the preliminary
test, six gradient concentrations at each pH value (i.e., 1.319, 2.637, 13.185, 26.370,
131.85, and 263.70 mg Cd/L for pH 5.0 and 6.0; 0.264, 1.319, 2.637, 13.185, 26.370,
and 131.85 mg Cd/L for pH 7.0; and 0.132, 0.264, 1.319, 2.637, 13.185, and 26.370 mg
Cd/L for pH 8.0 and 9.0) were used to determine the EC

values. The concentration

series in octuplicate and eight controls were arranged in a 96-well (8 rows

columns) black

ﬂat-bottom microplate (GRE, USA.). First, each well in the ﬁrst

column of the microplate was

ﬁlled with 180 μL of 3.0% NaCl solution to serve as

the control group. Second, the same volume of HgCl

standard solution was added

to each of the eight wells in the second column to serve as a reference to verify the
reliability of the experimental results. The third column which was injected with
180

μL of the pH-adjusted 3.0% NaCl solution was set as the pH-control group. Next,

180

μL of pH-adjusted metal solutions was added to the wells from the fourth

column to the ninth column in order of increasing concentration. Then, 20

μL of

bacterial suspension was added into each test well to get a total volume of 200

μL.

The metal solutions and controls were adjusted with HCl (0.12 mol/L and
0.012 mol/L) and NaOH (0.05 mol/L and 0.005 mol/L) to obtain

ﬁnal pH values of

approximately 5.0, 6.0, 7.0, 8.0, and 9.0. These

ﬁve pH values were selected because

the previous research has shown that within this range, the pH has no effect on
light emission by luminescent bacteria (

Fulladosa et al., 2004

). The non-complexing

buffer MOPS was used at a concentration of 2 mM to stabilize the pH.

To accurately determine Cd toxicity to P. phosphoreum at different pH levels, it

was necessary to adjust the pH without signi

ﬁcantly altering Cd concentrations in

solution. The pH of the bacterial suspension was approximately 7.58. Repeated
trials indicated that the pH of the metal solutions should be adjusted to 4.90, 5.92,
6.95, 8.10, and 9.16 to achieve the desired pH of 5.0, 6.0, 7.0, 8.0, and 9.0,
respectively, after the addition of the bacterial liquid. All the

ﬁnal pH values were

adjusted to be within

70.1 of the desired value.

For example, to prepare 263.70 mg Cd/L solution (pH 4.90), the following steps

were performed. First, the pH of the 3.0% NaCl solution was adjusted to 4.90 with
HCl (0.12 mol/L and 0.012 mol/L). Next, 2.5 mL of the metal stock, 10 mL of 3.0%
NaCl solution (pH 4.90), and 2.5 mL of the buffering agent (2 mM) were added to a
50 mL glass beaker. After readjustment to pH 4.90, the mixture was transferred to a
25 mL volumetric

ﬂask. The beaker was washed three times with the 3.0% NaCl

solution (pH 4.90), which was also transferred to the volumetric

ﬂask. Additional

3.0% NaCl solution was added until the liquid level reached the scale line on the
volumetric

ﬂask. Before use, the solution was poured into the pipetting reservoir,

and the pH was remeasured and adjusted to 4.90 if needed. This pH adjustment did
not signi

ﬁcantly alter the Cd concentration in solution because the amount of HCl

or NaOH added was negligible (less than 0.03 mL). After mixing in a 9:1 volume
ratio with the bacterial suspension, the pH and the metal concentration were
5.0 and 237.33 mg Cd/L, respectively. Due to dilution by the bacterial liquid, all the
concentrations were scaled by a factor of 0.9 for EC

calculation. The exposure

solution was used immediately after preparation. The bioluminescence of various
treatments and controls was determined using a Tecan In

ﬁnite 200

PRO multi-

mode microplate reader after exposure of 15 min at 25

1C. The toxicity of each

treatment was expressed as the relative light rate (RLR, %), which is calculated as
follows:

RLR

ð%Þ ¼ L=L

100%;

where L

and L are the average light units of the controls and the treatments,

respectively.

ﬁtting a straight line between the RLR values falling within the 10–90%

range and the corresponding concentrations with the linear regression method, the
regression equations were obtained and used to calculate EC

values (i.e., the

concentration at which RLR is 50%).

2.3.2. D. magna

The acute toxicity of the metal-spiked samples to D. magna was determined in

accordance with the National Standard Method of China (Water quality

—Determi-

nation of the acute toxicity of substance to Daphnia (D. magna straus) GB/T 13266-
1991). Preliminary experiments were performed to investigate the effect of pH on
D. magna. The experiments revealed that the activity of the organism was not
reduced by 24 h exposure to culture medium of pH 5.0, 6.0, 7.0, 8.0 or 9.0,

R.-J. Qu et al. / Ecotoxicology and Environmental Safety 95 (2013) 83

–90

indicating that the effect of pH in this range is negligible. Consequently, toxicity test
was performed at these pH values. The pH was adjusted using the test procedure
for P. phosphoreum, except that the 3.0% NaCl solution was replaced with standard
dilution water. The most suitable ranges of Cd

concentrations for toxicity testing

were 0.44

–2.19 mg Cd/L for pH 5.0 and 6.0, 0.04–2.19 mg Cd/L for pH 7.0, and 0.04–

0.44 mg Cd/L for pH 8.0 and 9.0. Five or six concentrations within each range were
used to determine EC

values. Glass culture dishes (120 mm) were used as test

vessels. Ten daphnids were placed in each vessel containing 100 mL of test solution,
and the experiment was replicated four times. Toxicity tests were conducted at
20

72 1C with a 14 h: 10 h light: dark photoperiod in an illuminating incubator. No

food was provided during the experimental period. The number of immobilized D.
magna was recorded after 24 h exposure. Each test was accompanied by a control
test with standard dilution water.

The EC

values and their 95% con

ﬁdence intervals were calculated using

Spearman

–Karber software (the 1978 version of the Trimmed Spearman–Karber

method) developed by Montana State University.

2.3.3. C. auratus

Nine glass tanks (28 cm

60 cm 36 cm) were used for the experiment. Each

tank contained ten acclimated

ﬁsh that were randomly selected. The group of ﬁsh

was exposed to control (no cadmium addition in natural aerated water, pH 7.25),
low pH-acid medium (pH 5.0), high pH-alkaline medium (pH 9.0), 0.01 mg Cd/L in
acid medium (pH 5.0), 0.01 mg Cd/L in natural medium (pH 7.25), 0.01 mg Cd/L in
alkaline medium (pH 9.0), 0.1 mg Cd/L in acid medium (pH 5.0), 0.1 mg Cd/L in
natural medium (pH 7.25), or 0.1 mg Cd/L in alkaline medium (pH 9.0). A

ﬁnal pH of

5.0 or 9.0 was achieved through adjustment with 1.2 mol/L HCl or 2.0 mol/L NaOH
solution. Because the pH of the test medium is dif

ﬁcult to control, we only studied

three representative pH values for the biomarker evaluation. The two Cd doses we
evaluated were selected based on a series of toxicity tests, particularly those
involving oxidative stress responses of

ﬁsh to Cd exposure (

;

). During the experimental period, 0.001 mol/L HCl or

0.05 mol/L NaOH solution was added dropwise at an appropriate rate to set the pH
to within

70.1 of the desired value. The experimental ﬁsh were fed twice a day

with commercial pellets during the toxicity tests but were fasted 24 h prior to
biochemical analysis. Food residue was removed daily by an automatic water-
changing system to minimize contamination from metabolic waste. Half of the
dirty water was released and an equal volume of experimental water was
replenished. Afterward, metal solution was added and the pH was adjusted to
maintain the test conditions. Five

ﬁsh were randomly sampled at 1 d and 7 d for

analysis.

The

ﬁsh were killed by a sharp blow to the head at the end of the exposure

period. Liver tissues were carefully dissected, washed with cold physiological saline
(0.9 percent NaCl solution), weighed, and homogenized (1:10, w/v) in cold
physiological saline using an IKA T10 homogenizer (IKA, Germany). The homo-
genates were centrifuged (Eppendorf, Germany) at 4000 g for 15 min at 4

1C. The

supernatants were used as an enzyme source for biochemical analysis.

The enzyme (SOD, CAT, GPx) activity, GSH level, MDA content and protein

concentration of the supernatants were measured using the Diagnostic Reagent
Kits according to the manufacturer

’s instructions. SOD activity was measured

at 550 nm using the xanthine oxidase method (

McCord and Fridovich, 1969

CAT activity was determined by monitoring residual H

absorbance at 405 nm

(

Goth, 1991

). GSH level was determined at 420 nm through the reaction between

5,5-dithiobis-2-nitrobenzoic acid (DTNB) and thiol-containing compounds. GPx activity,
estimated by the rate of GSH oxidation, was measured at 412 nm (

Hafeman

et al., 1974

). The MDA content, a biomarker for lipid peroxidation, was determined

at 532 nm by the thiobarbituric acid reactive species (TBARS) assay, which
measures the amount of MDA that reacts with thiobarbituric acid (

Livingstone

et al., 1990

). The protein concentration was measured at 595 nm by the Coomassie

Brilliant Blue dye binding technique (

Bradford, 1976

), with bovine serum albumin

as a standard. The absorbances were recorded using a TU-1810 UV

–vis spectro-

photometer (Persee, China). The speci

ﬁc activity of enzymes is expressed as U/mg

protein, while GSH level and MDA content are denoted by

μmol/g protein and

nmol/mg protein, respectively.

2.4. Chemical measurements

For the toxicity tests with P. phosphoreum and D. magna, the stock solution was

acidi

ﬁed with 1‰ volume of HNO

then measured by a

ﬂame-atomic absorption

spectrophotometer (SOLLAR M6, Thermo, USA) to check the concentration of Cd
actually present. The measured concentration was 97.1% and 97.9% of the nominal
value, respectively. Therefore, the nominal concentrations were used in the toxicity
assessments.

For the toxicity test with C. auratus, the dissolved cadmium concentration in

the glass tanks was determined at the start and at the end of the test using a

ﬂame-

atomic absorption spectrophotometer (SOLLAR M6, Thermo, USA). Similarly, prior
to analysis, the collected water samples were acidi

ﬁed with 1‰ volume of HNO

The measurements were done in triplicate. In the mediums with Cd addition,
the initial Cd concentration was measured to be 0.0095

70.00029 and

0.094

70.0018 mg/L. The increase in the volume of exposure water caused by pH

adjustments will de

ﬁnitely alter Cd concentration in solution. Due to the difference

in the dropping speed of HCl and NaOH solution, changes in the concentration of
Cd

were different. In pH 5.0 water, Cd

was measured as 0.00946

70.00005 mg/L

and 0.0934

70.00053 mg/L, whereas in pH 9.0 water, the concentration was

0.0094

70.0001 mg/L and 0.09270.0021 mg/L, respectively. To facilitate the dis-

cussion, the nominal values of 0.01 and 0.1 mg/L were used in the following
paragraphs.

2.5. Modeling of metal speciation

In the acute toxicity tests for P. phosphoreum and D. magna, Cd speciation at the

several studied pH values was calculated using the chemical equilibrium model
Visual MINTEQ (

http://www.lwr.kth.se/English/OurSoftware/vminteq/index.htm

)

for the interpretation of the laboratory results. For each analysis, the EC

value

was used as the total Cd concentration. It was assumed that the pH was

ﬁxed

during computations. We used the corresponding water quality parameters for
each test as the input variables.

2.6. Statistical analysis

Experimental data were presented as the mean

7standard deviation (SD). The

values of oxidative stress biomarkers (SOD, CAT, GPx, GSH and MDA) were checked
for normality using the Shapiro

–Wilk test and for homogeneity of variance using

the Levene test. One-way analysis of variance (ANOVA) followed by Duncan

’s test

was used to determine the signi

ﬁcance of differences (Po0.05 or Po0.01) between

individual treatments and controls. All statistical analyses were performed using
the SPSS statistical package (ver. 17.0, SPSS Company, Chicago, USA).

2.7. Integrated biomarker response

The Integrated Biomarker Response (IBR) (

Beliaeff and Burgeot, 2002

), a

method for combining all the measured biomarker responses into one general
stress index, was applied to assess the potential toxicity of different exposure
protocols to

ﬁsh. The procedure of IBR calculation is described here brieﬂy. Data

were

ﬁrst standardized to allow direct visual comparison of the biomarker

responses under the test conditions. The standardized data (Y) were calculated as
Y

¼(X−m)/s, where X is the value of each biomarker response, m is the mean value

of the biomarker, and s is the standard deviation of the biomarker. Next, we
computed Z

¼Y in the case of activation or Z¼–Y in the case of inhibition. The

minimum value (Min) was obtained for each biomarker. Finally, the score (S) was
computed as S

¼Y+|Min|, where S≥0 and |Min| is the absolute value of Min.

Star plots were used to display the biomarker results. A star plot radius

coordinate represents the score of a given biomarker. When S

and S

are assigned

as two consecutive clockwise scores of a given star plot, n is assigned as the number
of radii corresponding to the biomarkers. Thus, the area A

obtained by connecting

the ith and the (i

+1)th radius coordinates can be calculated as

sin

βðS

cos

β þ S

þ1

sin

βÞ;

where

β ¼ Arc tan

þ1

sin

−S

þ1

cos

;

α is 2π/n radians, and S

is S

The total area corresponding to a given situation (IBR value) was obtained as

IBR

¼ ∑

¼ 1

;

where n is the number of biomarkers.

3. Results

3.1. Metal speciation

For P. phosphoreum, there was almost no change in cadmium

speciation when the pH increased from 5.0 to 8.0 (

Fig. 1

A). At these

pH values, the dominant species were the two chloro-complexes
CdCl

and CdCl

, which resulted from the complexation of cadmium

ion with chloride ions that are abundant in the medium (3.0% NaCl
solution). As pH increased to 9.0, the percentage of Cd(CO

)

−

increased appreciably to 12%, while CdCl

and CdCl

exhibited

approximately the same level of dominance. Throughout the whole
pH range, the free ion form of cadmium was least abundant.

R.-J. Qu et al. / Ecotoxicology and Environmental Safety 95 (2013) 83

–90

For D. magna, the species distribution of cadmium changed

markedly with pH (

Fig. 1

B). The four major species, Cd

, CdHCO

CdCO

, and Cd(CO

)

−

, accounted for more than 98% of the total

dissolved cadmium in solutions with pH in the 5.0

–9.0 range.

However, the relative abundance of each species varied at different
pH levels. At pH 5.0, Cd

was predominant, comprising 95% of the

total species. At pH 6.0, Cd

was still in dominance but its

proportion was reduced to 77%. In contrast, the percentage of
CdHCO

increased to 23%. Under neutral pH conditions (pH 7.0),

level was still higher than CdHCO

level, and CdCO

cannot

be negligible. As the water became alkaline, the dominant species
changed. At pH 8.0, CdCO

became more abundant than the other

three species. At pH 9.0, CdCO

and Cd(CO

)

−

were the two major

species. In the pH range we evaluated, the percentage of Cd

decreased dramatically with increasing pH.

3.2. Acute toxicity

3.2.1. P. phosphoreum

In general, the EC

values presented a gradient descent trend,

i.e., the toxicity to P. phosphoreum increased with increasing pH
(

Fig. 2

A). The largest EC

value was 19.3 mg Cd/L at pH 5.0, and

the smallest was only 1.03 mg Cd/L at pH 9.0. Moreover, when the
EC

was expressed on a Cd

basis as Cd

, a linear relation-

ship between Cd toxicity and pH was evident (

Fig. 3

A).

The regression equation describing this trend is Cd

1.805

−0.214 pH (R

¼0.961).

3.2.2. D. magna

A general downward trend in EC

values was observed

across the pH range we studied (

Fig. 2

B). When the pH increased

from 5.0 to 6.0, the EC

value decreased slightly from 1.21 mg Cd/L

to 1.16 mg Cd/L. Then, it dropped drastically to 0.42 mg Cd/L
at pH 7.0. As the pH further increased, the EC

value declined

slightly until a minimum value of 0.35 mg Cd/L was obtained
at pH 9.0. Additionally, when the Cd

was expressed as a

function of pH (

Fig. 3

B), a positive relationship was observed. The

correlation equation was Cd

¼3.075−0.376 pH (R

¼0.913).

3.2.3. C. auratus
3.2.3.1. Biochemical measurements. Control values for the biochemical
parameters SOD, CAT, GPx, GSH and MDA at day 1 were measured as
87.6

77.5 U/mg protein, 24.473.6 U/mg protein, 121712 U/mg

protein, 14.0

71.3 μmol/g protein, and 2.8170.42 nmol/mg protein,

respectively. At day 7, control values for these parameters were
90.3

75.5 U/mg protein, 28.572.8 U/mg protein, 119711 U/mg

protein, 13.4

71.4 μmol/g protein, and 2.6470.39 nmol/mg protein,

respectively (

Table 1

Antioxidant enzymes: Generally, no signi

ﬁcant changes in SOD,

CAT and GPx activity were observed in any of the test groups after
1 d exposure. However, a longer exposure duration resulted in
signi

ﬁcant decreases (Po0.05 or Po0.01) in the activity of

antioxidant enzymes in several groups compared to controls. The
greatest changes appeared in the two groups exposed to Cd at pH
9.0, which had signi

ﬁcant (Po0.05) and occasionally very sig-

ﬁcant (Po0.01) decreases in the activity of all three indicators.

Speci

ﬁcally, 0.78-fold and 0.77-fold differences in SOD activity,

0.71-fold and 0.53-fold differences in CAT activity, and 0.58-fold
and 0.63-fold differences in GPx activity were observed in the

Fig. 1. Species distribution diagram of Cd at various pH levels for P. phosphoreum
(A) and D. magna (B).

Fig. 2. Effects of pH on Cd toxicity to P. phosphoreum (A) and D. magna (B). Error
bars represent 95% con

ﬁdence interval.

R.-J. Qu et al. / Ecotoxicology and Environmental Safety 95 (2013) 83

–90

pH(9)-Cd(0.01) group and pH(9)-Cd(0.1) group, respectively.
The effects of Cd were less pronounced in other groups, such that
only the pH(5)-Cd(0.1) group had signi

ﬁcant SOD depletion (Po0.05).

Nonenzymatic antioxidant: There were no signi

ﬁcant changes in

GSH level in any test group relative to control after 1 d exposure,
but signi

ﬁcant GSH differences (Po0.01) were observed with

prolonged exposure. Compared with the control, GSH was sig-
ni

ﬁcantly reduced by 0.69-fold (Po0.05) in the pH(9)-Cd(0.01)

group and by 0.63-fold (P

o0.01) in the pH(9)-Cd(0.1) group.

However, a signi

ﬁcant 1.38-fold increase (Po0.01) occurred in

the pH(7.25)-Cd(0.1) group.

Lipid peroxidation: For MDA content, no test groups exhibited

signi

ﬁcant changes relative to control after 1 d exposure. As exposure

time increased, MDA content was signi

ﬁcantly increased by 1.90-fold

o0.05) in the pH(7.25)-Cd(0.01) group and by 3.05-fold (Po0.01)

in the pH(9.0)-Cd(0.1) group.

3.2.3.2. Integrated biomarker response. The IBR values were different
among the various exposure protocols after 7 days, ranging from 1.19
in the pH(9) group to 12.50 in the pH(9)-Cd(0.1) group (

Fig. 4

The change in IBR caused by the addition of Cd is smaller at pH
5.0 than at pH 9.0. Moreover, for any given waterborne Cd
concentration, the IBR at pH 9.0 is larger than the corresponding
value at pH 5.0. IBR calculations were not performed at day 1 due to
the short exposure duration.

4. Discussion

4.1. P. phosphoreum

In general, the toxicity of Cd to P. phosphoreum was inversely

related to pH. It has been reported that the effect of pH on metal
toxicity is twofold: the hydrogen ion may exert its effect either
directly by affecting metal uptake or indirectly by affecting the
chemical speciation and bioavailability of the dissolved metal pool
(

Peterson et al., 1984

). There was an approximately 13-fold

increase in the 15 min-EC

value when the pH was decreased

from 8.0 to 5.0, whereas the distribution of metal species was

Fig. 3. The Cd

value (with 95% con

ﬁdence interval) as a function of pH for P.

phosphoreum (A) and D. magna (B).

Table 1
Effects of short-term exposure to Cd

under different pH on enzyme (SOD, CAT and GPx) activity, GSH level and MDA content in liver of C. auratus.

Biomarkers

Duration (d)

No Cd addition

Cd (0.01 mg/L)

Cd (0.1 mg/L)

¼5.0

¼7.25

¼9.0

¼5.0

¼7.25

¼9.0

¼5.0

¼7.25

¼9.0

SOD

96.3

713.9

87.6

77.5

82.1

74.5

91.0

715.8

81.9

79.9

67.6

77.7

87.0

79.2

67.7

78.1

70.7

74.8

80.4

74.1

90.3

75.5

86.2

76.7

91.0

711.1

101.2

713.0

70.7

78.3

72.3

74.7

78.8

77.2

69.9

76.1

CAT

17.0

71.2

24.4

73.6

25.9

72.3

25.6

72.9

21.3

73.7

22.7

72.1

28.0

73.8

21.1

71.8

22.1

71.1

41.6

72.7

28.5

72.8

36.8

74.6

28.7

73.6

32.4

73.8

20.5

71.5

24.1

72.6

26.3

72.3

bcd

15.2

71.6

GPx

126.8

723.4

abc

121

712

abc

114.9

77.8

abc

140

97.7

710.4

135

711

107

714

106

714

115

713

abc

114.2

74.5

119

711

109.6

719.6

108

bcd

138

710

69.8

76.8

94.3

715.3

abc

94.1

720.8

abc

75.8

712.1

GSH

18.0

71.2

14.0

71.3

15.3

70.76

16.9

71.9

18.5

71.4

12.3

71.6

18.5

71.1

13.7

71.5

13.1

71.9

12.8

73.4

13.4

71.4

14.3

71.7

12.4

71.6

11.3

71.2

abc

9.26

71.03

9.81

71.58

abc

18.6

72.2

8.49

71.41

MDA

3.92

71.2

2.81

70.42

3.05

70.40

2.57

70.32

2.94

70.24

2.44

70.20

2.70

70.28

3.95

70.42

2.83

70.26

2.57

70.72

2.64

70.39

2.80

70.54

4.21

70.45

5.02

70.22

3.94

70.35

4.91

70.29

3.94

70.35

8.08

70.59

Data are presented as means

7standard deviation (SD), n is 5 for each data point.

–d

indicate signi

ﬁcant differences from control for different exposure conditions during the same exposure time (Po0.05).

Denotes very signi

ﬁcant differences from control for a test group in the same exposure period (Po0.01).

Fig. 4. Integrated biomarker response (IBR) values for different exposure protocols
after 7 days.

R.-J. Qu et al. / Ecotoxicology and Environmental Safety 95 (2013) 83

–90

almost unchanged in this pH region. This inconsistency suggests
that the protective effect of low pH against Cd toxicity is probably
the consequence of decreased uptake of Cd into the cell, which is
likely due to the competitive exclusion of Cd

binding to the cell

surface by H

. The partial inhibition of Cd bioaccumulation by H

was also found in the clam Corbicula

ﬂuminea (

Graney et al., 1984

)

and the mussel Unio pictorum (

Pynnonen, 1990

). At pH 9.0, the

decreased to a minimum value, and the cadmium species

distribution was changed considerably. Interactions of the three
major species, Cd(CO

)

−

, CdCl

, and CdCl

, might be responsible

for the high toxicity (

Villaescusa et al., 1996

). Moreover, at this pH,

the decline in proton concentration weakens the competition of
H

with Cd

, thereby enhancing Cd toxicity. Over the studied pH

range (5.0

–9.0), the linear correlation between Cd

and pH

also indicated that there was possibly a modest competition
between H

and Cd

4.2. D. magna

According to the free-ion activity model (FIAM) proposed by

Campbell (1995)

, metal toxicity is governed by the activity of the

free hydrated metal ion. Consequently, Cd should be more toxic to
D. magna in acidic environments than in neutral and alkaline ones
due to the predominance of Cd

(as con

ﬁrmed by our metal

speciation analysis results). However, our results show the oppo-
site trend, i.e., the acute toxicity of Cd

was decreased by a lower

pH. This

ﬁnding is consistent with that of

Clifford and McGeer

(2010)

, who reported a general trend of increasing EC

values for

Cd in Daphnia pulex when pH was decreased from 8.02 to 6.10.
Although H

has a dual effect on the toxicity of metals, changes in

Cd toxicity to D. magna largely re

ﬂect changes directly related to

concentration. Hence, the toxicity test results in the present

study may be better explained by the competition between
protons and the metal ions at the cell surface. The competition
of protons with free cadmium ions at the biotic ligand was
con

ﬁrmed by the linear relationship between Cd

and pH.

In a previous research work,

Playle (2004)

found a similar

competitive interaction, reducing the binding of Cd to the gills of
fathead minnows (Pimephales promelas) as pH decreased to 4.8.
Furthermore, low pH may trigger some physiological reactions
within organisms, which are involved in weakening the toxicity of
Cd.

Guan and Wang (2006)

proposed that the induction of

metallothionein by low pH is the major Cd detoxi

ﬁcation mechan-

ism in D. magna.

4.3. C. auratus

The fact that no test groups showed statistically signi

ﬁcant

changes in SOD, CAT and GPx activity, GSH level, or MDA content
after 1 d exposure seems to indicate that the exposure duration
was too short for the two concentrations of Cd to produce
oxidative stress in the liver. However, with prolonged exposure,
signi

ﬁcant changes in several biochemical indexes were observed

in certain experimental groups, suggesting that cadmium induced
oxidative stress. Particularly, the antioxidants were most strongly
inhibited in the groups kept in an alkaline environment (pH 9.0).
Also, reductions were observed in the groups at pH 5.0, but the
vast majority of these reductions were not statistically signi

ﬁcant.

In all, there was little induction of antioxidants in this study. One
reason for this could be the combined action of Cd and pH. Of the
antioxidant enzymes, SOD and associated CAT are considered as
the vital

ﬁrst-line defenses against oxygen toxicity (

Yu, 1994

). They

can protect organisms from oxidative damage by partially elim-
inating reactive oxygen species (ROS). Under physiological condi-
tions, SOD catalyzes the dismutation of superoxide radical (O

−

)

into hydrogen peroxide (H

), whereas CAT converts H

into

O and O

. The decreased SOD and CAT activity in

ﬁsh may be

caused by the interactions between Cd and essential trace ele-
ments. Cd has an adverse effect on the enzymatic systems of cells
due to its substitution for divalent metals (Zn

, Cu

, and Mn

)

in metalloenzymes and its very strong af

ﬁnity for sulfhydryl-

containing biological macromolecules such as proteins, enzymes
and nucleic acids (

Lionetto et al., 2000

). The competition between Cd and Zn, Cu or Mn in SOD

molecules could explain the decrease in SOD activity, which may
imply a reduction in H

production. Because the activity of CAT

is directly proportional to the substrate H

level which is

assumed to be produced by SOD (

Kono and Fridovich, 1982

;

Aitken and Roman, 2008

) the CAT activity is decreased. The Cd-

induced reduction in hepatic CAT activity may re

ﬂect a reduced

capacity of the liver to eliminate H

the

second-line

defenses

against

oxidative

damage

(

Cnubben et al., 2001

), GSH and GSH-related enzymes play a

major role in cellular metabolism and free radical scavenging
(

Pena-Llopis et al., 2003

;

Liu et al., 2008

). GSH is a major none-

nzymatic low-molecular-weight antioxidant and functions as a
direct free radical scavenger that quenches oxyradicals through its
sulfhydryl group (

Zhang et al., 2004

). GSH also serves as an

available co-substrate for GPx activity, and it is conjugated in
xenobiotic reactions (

). Our

experimental results are consistent with earlier studies (

Chatterjee

and Bhattacharya, 1984

;

Cirillo et al., 2012

) that showed decreased

levels of GSH in tissues exposed to Cd.

Pandey et al. (2008)

stated

that decreased intracellular GSH levels and simultaneous inhibi-
tion of GSH-related antioxidant enzymatic activity could result in
oxidative imbalance and subsequent cell death. There are two
possible explanations for lowered GPx activity observed in our
study. First, reduced GSH might cause the inactivation of GPx
because GPx is highly dependent on GSH concentration. Second,
the interaction between Cd and Se in the GPx molecule could lead
to the inhibition of GPx activity.

Moreover, MDA content was generally increased in

ﬁsh liver

after 7 d exposure, indicating an elevation of lipid peroxidation in
this organ. Lipid peroxidation is one of the main manifestations of
oxidative damage. Enhanced peroxidation of lipids in intra- and
extracellular membranes causes damages to the cells, tissues and
organs because the reactive carbonyls produced during lipid
peroxidation may spread damage far from the original site of
radical production (

Cheeseman, 1993

). The enhanced lipid perox-

idation in the Cd-treated

ﬁsh in our study might result from the

decrease in the liver activity of antioxidants and the generation of
radicals during normal metabolism.

In summary, suppressed activity of antioxidant enzymes,

decreased GSH level, and enhanced lipid peroxidation in the liver
of

ﬁsh after 7 d exposure to Cd

(particularly at pH 9.0) con

ﬁrmed

the occurrence of oxidative stress arising from insuf

ﬁcient neu-

tralization of ROS generated. It is noteworthy that at pH 5.0, the
two concentrations of Cd

we investigated almost triggered no

oxidative stress, although few indices exhibited statistically con-
ﬁrmed changes. Under alkaline conditions, especially in pH
9.0 water, signi

ﬁcant changes in the measured oxidative stress

biomarkers were commonly observed in both high-dose and low-
dose Cd

exposure groups after 7 days. This demonstrates that

the toxicity of Cd

to the aquatic organism is increased by an

alkaline environment, to a certain extent.

The IBR index, a simple tool for visualizing biological effects,

also con

ﬁrmed the ﬁndings presented above. Generally, the higher

the IBR value is, the more stressful the environment is. According
to the IBR index, the relative toxicity of the exposure conditions in
this study is as follows: pH(9)

opH(5)-Cd(0.01)opH(5)opH(5)-

Cd(0.1)

opH(7.25)-Cd(0.1)opH(9)-Cd(0.01)opH(7.25)-Cd(0.01)o

pH(9)-Cd(0.1). As waterborne Cd concentration is increased, the

R.-J. Qu et al. / Ecotoxicology and Environmental Safety 95 (2013) 83

–90

change in IBR index is larger at pH 9.0 than at pH 5.0. In addition,
the IBR value at pH 9.0 is larger than at pH 5.0 for a

ﬁxed Cd

exposure dosage. These results suggest that cadmium is more toxic
to

ﬁsh in an alkaline environment.

5. Conclusions

Overall, Cd

was more toxic to the three test organisms in

alkaline environments than in acidic environments. Nevertheless,
changes in Cd

toxicity with respect to pH were organism-

speci

ﬁc. The EC

values for D. magna were one order of magni-

tude smaller than those for P. phosphoreum, suggesting that
D. magna was more sensitive to the toxicity of Cd than P.
phosphoreum. The low sensitivity of P. phosphoreum to Cd may
question its applicability as the

ﬁrst screening method in assessing

environmental samples. For C. auratus, the signi

ﬁcant reduction in

antioxidants and enhanced lipid peroxidation in alkaline water
(pH 9.0) after 7 d Cd

treatment is indicative of the occurrence of

oxidative damage. Taking these results into account, changes in
toxicity with pH should be considered in laboratory assessments of
ﬁeld metal toxicity. The present study may provide useful infor-
mation to evaluate the toxicological effects of Cd on organisms in
various environments with different pH. However, the exact
mechanism involved in the combined toxicity effect of metal and
pH remains unclear and requires further investigation.

Acknowledgments

This research was

ﬁnancially supported by the National Natural

Science Foundation of China (nos. 41071319, 20977046), the
Fundamental Research Funds for the Central Universities of China
(no. 1112021101) and the Major Science and Technology Program
for Water Pollution Control and Treatment of China (no. 2012
ZX07506-001).

Appendix A. Supporting information

Supplementary data associated with this article can be found in

the online version at

http://dx.doi.org/10.1016/j.ecoenv.2013.05.020

References

Aitken, R.J., Roman, S.D., 2008. Antioxidant systems and oxidative stress in the

testes. Oxidative Med. Cell. Longevity 1, 15

–24

Atli, G., Canli, M., 2010. Response of antioxidant system of freshwater

ﬁsh

Oreochromis niloticus to acute and chronic metal (Cd, Cu, Cr, Zn, Fe) exposures.
Ecotoxicol. Environ. Saf. 73, 1884

–1889