Effect of pH, EDTA, and Anions on Heavy Metal Toxicity Toward a Bioluminescent Cyanobacterial Bioreporter

  • Ismael Rodea-Palomares
  • Coral González-García
  • Francisco Leganés
  • Francisca Fernández-Piñas
Article

Abstract

The bioavailability and therefore toxicity of a metal depends on the chemical species present in a particular environment. We evaluated the effect of a series of factors that could potentially modify metal speciation on the toxicity of Hg, Cu, Zn, and Cd toward a recombinant strain of the freshwater cyanobacterium Anabaena sp. PCC 7120 with cloned lux operon of luminescent terrestrial bacterium Photorhabdus luminescens. The strain, denoted as Anabaena CPB4337, showed a high constitutive luminescence with no need to add exogenous aldehyde. The tested factors were pH, EDTA (as organic ligand), and anions PO43–, CO32–, and Cl. Chemical modeling and correlation analyses were used to predict metal speciation and link it with toxicity. In general, metal toxicity significantly correlated to the predicted metal free-ion concentration, although Zn–EDTA complexes and certain Hg chloro-complexes could also exhibit some toxicity to cyanobacteria. An interesting feature of metal toxicity to strain Anabaena CPB4337 was that low amounts of PO43– and CO32– increased metal toxicity; this effect could not be related to significant changes in metal speciation and could be attributed to a modulating effect of these anions on metal/uptake toxicity. The combination of toxicity studies that take into account a range of factors that might modulate metal toxicity with chemical modeling to predict changes in metal speciation might be useful for interpreting complex toxicity data. Finally, this cyanobacterial bioreporter, due to its ecological relevance as a primary producer, could be used as a tool for toxicity assessment in freshwater environments.

Available freshwater resources are polluted by industrial effluents, domestic and commercial sewage, as well as mine drainage, agricultural runoff and litter. Among water pollutants, heavy metals are priority toxicants that pose potential risks to human health and the environment. The evaluation of heavy metal contamination traditionally relies on highly sensitive and specific physical and chemical techniques such as atomic absorption spectroscopy or mass spectrometry; however, such methods are not able to distinguish between available (potentially hazardous to biological systems) and nonavailable (potentially nonhazardous) fractions of metals that exist in the environment in inert or complexed forms. Toxicity bioassays and biosensors might complement physical and chemical methods by detecting the toxicity related with bioavailable metals in environmental samples, effectively integrating the complexity of environmental factors (pH, redox potential, exchangeable cations, biological activity, etc.) that contribute to bioavailability (Köhler et al. 2000). Bioavailability is strongly affected by the speciation of a metal in a particular environment. Water chemistry parameters such as alkalinity, pH, salinity, hardness, phosphates, or ionic strength (Cook et al. 2000; Ho et al. 1999) might influence metal ion toxicity either directly by lowering free metal ion concentration or indirectly through synergistic or antagonistic effects. Furthermore, bioavailable concentrations of metals in the environment can be altered by chelating substances from natural (humic and fulvic acids) or anthropogenic sources (EDTA or polyphosphates). Most studies have found that toxicity is usually a function of the free metal ion because this species is generally the most bioavailable one (Campbell 1995); however, there are many other reports that showed that the toxic response does not always conform to the free-ion model and that organic complexing agents and/or inorganic hydroxyl or carbonate complexes might also exhibit some toxicity to target organisms (Allen and Hansen 1996; Campbell et al. 2000; Deheyn et al. 2004; Fernandez-Piñas et al. 1991; Parent et al. 1996).

Cyanobacteria are the only prokaryotic organisms carrying out an oxygen-evolving photosynthesis. They originated during the Precambrian era (2.8 × 109 years ago), and as a group they are known to survive a wide spectrum of environmental stresses. As primary producers with a key role in the N and C cycles, they are a dominant component of marine and freshwater phytoplankton and any detrimental effect on this group might have a negative impact in nutrient availability to organisms of higher trophic level.

In this study, we report the effect of potential modifying factors on metal speciation such as pH, complexing agent EDTA, and anions PO43–, CO32–, and Cl on the toxicity of Cu, Zn, Hg, and Cd toward a self-luminescent recombinant strain of the freshwater cyanobacterium Anabaena sp. PCC 7120. As luminescence is directly proportional to the metabolic status of the cell and any inhibition of cellular activity is reflected in a decrease of bioluminescence; toxicity was measured as luminescence inhibition caused by biologically available metal species. We used chemical modeling (Visual MINTEQ and PHREEQC programs) and correlation analyses in an attempt to link toxicity with metal speciation.

Methods and Materials

Strain and Culture Conditions

Anabaena sp. PCC 7120 strain CPB4337 (hereinafter Anabaena CPB4337), which bears in the chromosome a Tn5 derivative with luxCDABE from the luminescent terrestrial bacterium Photorhabdus luminescens (formerly Xenorhabdus luminescens), was used in this study as a bioreporter of metal toxicity. This strain shows a high constitutive self-luminescence with no need to add exogenous aldehyde; also, cell viability is not significantly affected by the Tn5 insertion and the endogenous generation of aldehyde (Fernandez-Piñas and Wolk 1994). Luminescence was shown to be high in this strain in a range of temperatures between 20°C and 30°C, in accord with Photorhabdus luminescens luciferase having the greatest thermal stability (Fernandez-Piñas et al. 2000; Szittner and Meighen 1990). Anabaena CPB4337 was routinely grown at 28°C in the light, Ca.65 μmol photons m2/s on a rotary shaker in 50 mL AA/8 medium (Allen and Arnon 1955) supplemented with nitrate (5 mM) in 125-mL Erlenmeyer flasks. The strain was grown in liquid cultures with 10 μg of neomycin sulfate (Nm) per mL.

Heavy Metal Toxicity Assays

Toxicity response of the cyanobacterium was estimated as EC50 values, the median effective concentration of the metal that causes a 50% bioluminescence inhibition with respect to a nontreated control. Standard metal solutions of Hg, Cu, Zn, and Cd were serially diluted (five to eight serial dilutions) for the EC50 determinations and were buffered with 2 mM MES [2-(N-morpholino) ethanesulfonic acid] and adjusted to pH 5.8. The use of most buffers (like Tris or Tricine) might not be appropriate, as precipitation and complexation of metals might occur (Fernandez-Piñas et al. 1991) and the alkylsulfonate derivatives of morpholine, like MES, are reported to be noncomplexing for metals (Kandegedara and Rorabacher 1999). Previous experiments in the current study showed the following: luminescence of strain Anabaena CPB4337 was high in a pH range from 5.8 to 8 (not shown); pH 5.8 was finally chosen for the toxicity bioassays because most metals are biologically available; according to Visual MINTEQ and PHREEQC calculations, the free-ion species of Cu, Zn, and Cd ranged between 95% and 99.99% of the total metal species present at pH 5.8 (Table 1); however, in the case of Hg, chemical modeling predicted that for pH values between 1 and 9, in the range of Hg concentrations used in the study, less than 0.001% was present as free ion Hg2+.
Table 1

Predicted percentages of the total concentration of metal present as free-ion and dominant forms of metal complexes in the aqueous phase at increasing pH values and increasing phosphate and carbonate concentrations as calculated by Visual MINTEQ

 

Hg

Cu

Zn

Cd

pH

    5.8

Hg(OH)2

93.674

Cu2+

98.117

Zn2+

98.436

Cd2+

99.806

HgOH+

0.148

CuSO4

0.011

ZnOH+

1.558

CdCl+

0.187

HgCl2

1.395

CuOH+

1.227

    

HgClOH

4.719

Cu(OH)2

0.643

    

    7

Hg(OH)2

99.871

Cu2+

35.092

Zn2+

79.826

Cd2+

99.705

HgClOH

0.123

CuOH+

6.969

ZnOH+

20.046

CdCl+

0.187

  

Cu(OH)2

57.880

Zn(OH)2

0.123

CdOH+

0.097

  

Cu2(OH)22+

0.054

    

    7.5

Hg(OH)2

99.959

Cu2+

5.514

Zn2+

55.228

Cd2+

99.474

HgClOH

0.039

CuOH+

3.465

ZnOH+

43.912

CdCl+

0.187

  

Cu(OH)2

91.007

Zn(OH)2

0.856

CdOH+

0.305

  

Cu2(OH)22+

0.013

    

    8

Hg(OH)2

99.987

Cu2+

0.595

Zn2+

27.222

Cd2+

98.749

HgClOH

0.012

CuOH+

1.183

ZnOH+

68.548

CdCl+

0.185

  

Cu(OH)2

98.219

Zn(OH)2

4.226

CdOH+

0.959

PO43− (mg/L1)

    0.1

Hg(OH)2

96.180

Cu2+

94.943

Zn2+

97.530

Cd2+

99.792

HgCl+

0.013

CuHPO4

0.012

ZnOH+

2.445

CdCl+

0.187

HgCl2

0.606

CuOH +

2.332

ZnHPO4

0.013

CdHPO4

0.010

HgClOH

3.152

Cu(OH)2

2.423

    

    1

Hg(OH)2

96.188

Cu2+

96.193

Zn2+

97.403

Cd2+

99.701

HgOH+

0.048

CuH2PO4+

0.039

ZnOH+

2.432

CdCl+

0.186

HgCl+

0.013

CuHPO4

0.098

ZnH2PO4+

0.038

CdHPO4

0.102

HgCl2

0.604

CuOH+

1.869

ZnHPO4

0.120

  

HgClOH

3.146

Cu(OH)2

1.541

    

    10

Hg(OH)2

96.226

Cu2+

95.173

Zn2+

96.083

Cd2+

98.840

HgCl+

0.013

CuH2PO4+

0.369

ZnOH +

2.340

CdCl+

0.178

HgCl2

0.593

CuHPO4

0.933

ZnH2PO4+

0.382

CdHPO4

0.971

HgClOH

3.119

CuOH+

1.806

ZnHPO4

1.188

  
  

Cu(OH)2

1.477

    

    100

Hg(OH)2

93.950

Cu2+

92.786

Zn2+

89.383

Cd2+

94.679

HgCl+

0.032

CuH2PO4+

3.333

ZnOH +

1.257

CdCl+

0.152

HgCl2

1.328

CuHPO4

2.971

ZnH2PO4+

3.220

CdHPO4

5.163

HgClOH

4.611

CuOH+

0.586

ZnHPO4

6.135

  
  

Cu(OH)2

0.169

    

CO32− (mg/L)

    0.1

Hg(OH)2

93.677

Cu2+

97.921

Zn2+

98.432

Cd2+

99.801

HgCl2

1.433

CuOH+

1.205

ZnOH+

1.557

CdCl+

0.187

HgClOH

4.783

Cu(OH)2

0.628

    

    1

Hg(OH)2

93.692

Cu2+

97.838

Zn2+

98.396

Cd2+

99.759

HgOH+

0.074

CuOH+

0.059

ZnOH+

1.549

CdCl+

0.186

HgCl+

0.032

Cu(OH)2

1.199

ZnHCO3+

0.046

CdHCO3+

0.048

HgCl2

1.428

      

HgClOH

4.774

      

    10

Hg(OH)2

93.759

Cu2+

97.005

Zn2+

98.028

Cd2+

99.759

HgOH+

0.075

CuCO3

0.566

ZnOH+

1.502

CdCl+

0.186

HgCl+

0.032

CuOH+

1.159

ZnHCO3+

0.442

CdHCO3+

0.048

HgCl2

1.402

Cu(OH)2

0.597

    

HgClOH

4.732

Cu(OH)22+

0.020

    

    100

Hg(OH)2

99.897

Cu2+

4.015

Zn2+

44.349

Cd2+

86.488

HgClOH

0.101

CuCO3

39.639

ZnOH+

30.360

CdCl+

0.134

  

Cu(CO3)22-

0.147

ZnHCO3+

7.016

CdOH+

0.073

  

CuOH+

2.144

ZnCO3

16.662

CdHCO3+

12.734

Note: Total metal concentration for calculations of each metal is 10 μM

Toxicity bioassays were as follows: 160 μL from the serial dilutions of each heavy metal plus a control [double distilled water (ddH2O) buffered with 2 mM MES at pH 5.8] were disposed in an opaque white 96-well microtiter plates. Cells, grown as described, were washed twice and resuspended in ddH2O buffered with 2 mM MES at pH 5.8 and were added to the microtiter plate wells to reach a final cell density of 0.5 at optical density (OD) 750 nm. Luminescence of each sample was recorded every 5 min in a Centro LB 960 luminometer up to 30 min. Three independent experiments with quadruplicate samples were conducted.

Effect of Modifying Factors on Metal Toxicity

To investigate the effect of modifying factors on Hg, Cu, Zn, and Cd toxicity, a metal concentration of 10 μM was chosen to elicit a strong toxic response; the EC50 values for each of the metals after 30 min of exposure were below this concentration (see the Results section and Fig. 1).
Fig. 1

Bioluminescence inhibition curves for Hg, Cu, Zn, and Cd. The curves present the percent of bioluminescence inhibition versus logarithm of metal concentration (expressed in μM). EC50 values after 30 min of exposure to the metals (EC50–30 min) and standard deviation were calculated from three independent experiments with triplicate samples

To relate bioavailability and toxicity, the effect of pH, organic ligand (EDTA), phosphate (as NaH2PO4), carbonate (as Na2CO3), and chloride (as NaCl) on the toxicity of Hg, Zn, Cu, and Cd to strain Anabaena CPB4337 was checked. These potential modifying factors were varied within ranges that might be found in freshwater environments (Perona et al. 1999; Van Dijk et al. 1994): pH from 5.8 to 8; chelate/metal ratios between 0 and 2, phosphate, carbonate, and NaCl concentrations from 0.1 to 10 mg/L; a 10-fold higher phosphate/carbonate/chloride concentration (100 mg L−1) was also used. It was considered interesting to check the effect of a much higher salt concentration, 2% NaCl (w/v), on cyanobacterial luminescence and metal toxicity because this high salt content is present in the widely used bioassays based on the marine bioreporter Vibrio fischeri and previous studies have reported that it affected metal bioavailability and toxicity (Deheyn et al. 2004; Newman and McCloskey, 1996; Riba et al. 2003).

To check the effect of pH on metal toxicity, 160 μL of the appropriate metal solution buffered with 2 mM MES and adjusted at pHs 5.8, 7, 7.5, and 8 were disposed on the microtiter plates. Cells grown as described were centrifuged, washed, resuspended in ddH2O buffered with 2 mM MES and adjusted to pH 5.8, 7, 7.5, and 8, and were added to reach a final cell density (OD750 nm) of 0.5; the final pH of the bioassay was checked for each metal concentration.

To investigate the effect of the complexing agent EDTA on metal toxicity, EDTA/metal solutions were prepared to get final chelate/metal molar ratios of 0.5, 1, 1.5, and 2; three controls were included: 0 M2+ (untreated control), 10 μM M2+ (metal treated control for toxic response), and 20 μM EDTA (EDTA control to monitor any effect of the chelator on cell self-luminescence). The EDTA/metal solutions and controls were buffered with 2 mM MES and adjusted to pH 5.8 and were incubated for at least 48 h at room temperature to ensure metal complexation (Fernandez-Piñas et al. 1991; Riether et al. 2001; Tauriainien et al. 2000).

To investigate the effect of phosphate, carbonate, and chloride on metal toxicity, phosphate/metal, carbonate/metal, and chloride/metal solutions were prepared to get a final metal concentration of 10 μM and final concentrations of 0.1, 1, 10, and 100 mg/L phosphate (as NaH2PO4), carbonate (as Na2CO3), or chloride (as NaCl). Then 0 M2+ (untreated control), 10 μM M2+ (metal treated control for toxic response), 100 mg/L phosphate/carbonate/chloride (controls to monitor any effect of phosphate, carbonate, or choride on cell self-luminescence) were included. Two percent NaCl (w/v) was also used to study its effect on cyanobacterial luminescence and metal toxicity. The metal/phosphate, metal/carbonate, and metal/chloride mixtures were allowed to soak for at least 48 h at room temperature to allow complexes to be formed (Fernandez-Piñas et al. 1991). All solutions were buffered with 2 mM MES and adjusted to pH 5.8, except those containing 100 mg/L carbonate whose pH was adjusted to 7 to avoid equilibrium shifts to bicarbonate and CO2.

The bioassays with the modifying factors EDTA, phosphate, carbonate, and chloride were essentially carried out as for the standard metal assay; luminiscence messurements was recorded every 5 min in the Centro LB 960 luminometer up to 30 min. Three independent experiments with triplicate samples were carried out for each case.

Modelling of Metal Speciation

Two programs were used to predict metal speciation: Visual MINTEQ and PHREEQC. The chemical equilibrium model Visual MINTEQ (http://www.lwr.kth.se/English/OurSoftware/vminteq/index.htm) is based on the program PC MINTEQA2 version 4.0 (Allison et al. 1991). Assumptions of a fixed pH, fixed potential redox (Eh), closed system, and no precipitation of solid phases were made during computations. The geochemical model PHREEQC version 2 (Parkhurst and Appelo 1999) was used with the aid of the graphical user interface PHREEQCI (version 2) http://wwwbrr.cr.usgs.gov/projects/GWC_coupled/phreeqc/index.html). These chemical models have proved very useful for linking speciation to metal toxicity and biosorption processes in a number of organisms (Campbell et al. 2000; Deheyn et al. 2004; Herrero et al. 2005; Newman and McCloskey 1996). Both model calculations were very similar and, to simplify, only dominant metal species for each tested condition as calculated by Visual MINTEQ are shown in Tables 1 and 3.

Analysis of Results

The toxic response of Anabaena CPB4337 as EC50 values was estimated by fitting the experimental luminescence inhibition data to a three-parameter logarithmic function:
$$ f = a/\left\{{1+\text{exp}\left[ {-\left({x-x_{0} } \right)/b} \right]} \right\}, $$
where f is the percentage of bioluminescence inhibition, x is the logarithm of metal concentration, a, b, and x0 are the parameters of the equation estimated by the model.

One-way analyses of variance (ANOVA) and linear regression analyses were computed using MINITAB Release 14 for Windows (Minitab Inc., USA).

Results

Toxicity Assays of Heavy Metals to Strain Anabaena CPB4337

For each metal treatment, a concentration–response curve could be established from which the EC50 values were derived (Fig. 1). The bioluminescent cyanobacterium responded sensitively to the four metals tested; the 30-min EC50 values and 95% confidence intervals calculated for each of the metals were as follows: Hg, 1.99 ± 0.29 μM (1.41–2.57); Cu, 1.52 ± 0.10 μM (1.32–1.72); Zn; 3.94 ± 0.19 μM (3.56–4.32); Cd, 6.76 ± 1.21 μM (4.34–9.18). Based on these values, the order of sensitivity of Anabaena CPB4337 toward the tested metals was Cu ≥ Hg > Zn > Cd.

Effect of Modifying Factors on Metal Toxicity

Figure 2 illustrates the effect of increasing pH values (5.8–8) on Hg, Zn, Cu, and Cd toxicity to strain Anabaena CPB4337 after 30 min of exposure. A pH increase significantly ameliorated Hg, Zn, and Cu toxicity (ANOVA, p < 0.05). Cd toxicity was not significantly remediated at higher pH values (ANOVA, p < 0.05). As shown in Table 1, the amelioration of Zn and particularly Cu toxicity at increasing pH values could be explained by the formation of nonavailable or less available metal hydroxides (particularly neutral hydroxyl complexes) and the concomitant decrease of the free-ion form; in fact, significant and negative correlations were found between percent maximum luminescence and percent free Cu2+ (= −0.92, p < 0.1) and between percent maximum luminescence and percent free Zn2+ (= −0.82, < 0.05) (Table 2). In the case of Cd, up to pH 8 the free-ion form predominated (Table 1) and no significant correlation was found between percent maximum luminescence and percent free Cd2+ (Table 2), which might explain why Cd toxicity was not reduced with increasing pH values. Whereas for Zn, Cu, and Cd, the free-ion form clearly governed toxicity in the tested pH range, Hg essentially appeared as the neutral species, mainly HgCl2, Hg(OH)Cl, and Hg(OH)2 between pH 5.8 and 8 (Table 1).Within this pH range, the predominant Hg-containing ion pair was the dihydroxy species; the neutral dichloro-complex was present below pH 7 and the free-ion Hg2+ species was present at extremely low concentrations (from 2.48 × 10−11 M at pH 5.8 to 1.05 × 10−15 M at pH 8 as calculated by Visual MINTEQ and PHREEQC). The most available and probably toxic Hg species could be the free-ion form even at these very low concentrations and perhaps the neutral dichloro-complex that was not present at the higher pH values (Table 1). A significant negative correlation was found between percent maximum luminescence and the logarithm of free Hg2+ concentration (= −0.96, p < 0.05), which could explain most of the observed Hg toxicity variation.
Fig. 2

Effect of increasing pH on Hg, Zn, Cu, and Cd toxicity to Anabaena CPB4337. All measurements were conducted in quadruplicate and were repeated at least twice. Values with the same superscript letter were not significantly different (p < 0.05) as determined by ANOVA between the different pH treatments for each metal concentration tested

Table 2

Correlation analyses between percent maximum luminescence (y) and percent free ions for Cu, Zn, and Cd or the logarithm of Hg2+ concentration (μM) for Hg (x)

Factor

Free ion

Regression parameters

x0

m

r

R2

p

pH

Cu2+

89.44

−0.78

−0.92

0.84

<0.1

Zn2+

92.30

−0.67

−0.82

0.68

<0.05

Cd2+

Hg2+

105.24

−3.68

−0.96

0.93

<0.05

EDTA

Cu2+

132.81

−1.32

−0.97

0.94

<0.05

Zn2+

Cd2+

184.15

−1.82

−0.98

0.97

<0.01

Hg2+

PO42−

Cu2+

Zn2+

207.07

−1.91

−0.85

0.72

<0.1

Cd2+

Hg2+

−194.47

−19.26

−0.81

0.65

<0.1

CO32−

Cu2+

114.80

−1.07

−0.98

0.95

<0.05

Zn2+

163.06

−1.56

−0.99

0.99

<0.01

Cd2+

801.61

−7.85

−0.99

0.98

<0.01

Hg2+

−184.29

−18.19

−0.99

0.99

<0.01

Note: Parameters of linear regression equations: x0 (value of y when x = 0); m (slope) and r (correlation coefficient) as well as R2 (goodness-of-fit coefficient) and p-values are given. Analyses were computed using MINITAB Release 14 for Windows. –: p-values > 0.1

Figure 3 shows the effect of complexing agent EDTA on metal toxicity in strain Anabaena CPB4337 after 30 min of exposure. The results showed that EDTA addition had a significant (ANOVA, p < 0.05) effect on remediation of all the tested metal toxicities. As calculated by Visual MINTEQ, the chelator markedly decreased the free-ion proportion of Cu, Cd, and Zn: At ratio EDTA/metal = 0.5, only 46% of Cu and Zn and 49% of Cd remained as the free-ion form, at ratios EDTA/metal = 1 and higher, metals were present as EDTA–metal complexes (not shown); for Cu and Cd, the toxic response most probably related to the amount of noncomplexed metal present, suggesting that the metal–EDTA complex was nontoxic to the cyanobacterial strain and that toxicity was governed by the free-ion concentration. This is confirmed by the significant and negative correlations found between percent maximum luminescence and percent free Cu2+ (= −0.97, p < 0.05) and between percent maximum luminescence and percent free Cd2+ (= −0.98, p < 0.01) in the EDTA experiment (Table 2). In the case of Hg, Visual MINTEQ and PHREEQC did not predict the formation of Hg–EDTA complexes and the concentration of free-ion Hg2+ (2.5 × 10−11 M) was constant through the increasing EDTA/Hg ratios. Furthermore, no correlation existed between percent maximum luminescence and the logarithm of the predicted Hg2+ concentration (Table 2). However, Hg toxicity amelioration by EDTA followed the pattern of Cu and Cd toxicity; so, most probably Hg–EDTA complexes were likely to be formed. For Zn, although there was also a decrease in toxicity with increasing EDTA/metal ratios, the results were somewhat different in that for an EDTA/metal ratio of 1:1, bioluminescence was already below 100% maximum luminescence. Additionally, no significant correlation was found between percent maximum luminescence and percent free Zn2+ (Table 2); possibly, the Zn–EDTA complexes might be available and toxic to cyanobacteria. It should also be noticed that EDTA itself was not toxic to the cells; in fact, an enhancement of bioluminescence could be seen in the EDTA controls treated with the highest EDTA concentration, 20 μM (Fig. 3).
Fig. 3

Effect of the addition of chelating agent EDTA on Hg, Cu, Zn, and Cd toxicity to Anabaena CPB4337. * Statistically significant differences (p < 0.05) as determined by ANOVA with respect to the metal-treated control. Error bars represent standard deviation of the means of at least two independent experiments with triplicate samples

Figure 4a shows the effect of increasing phosphate concentrations (0.1, 1, 10, and 100 mg/L) on Hg, Zn, Cu, and Cd toxicity to strain Anabaena CPB4337. Only Zn toxicity was significantly remediated by 100 mg/L of phosphate in the medium (ANOVA, p < 0.05). Hg and Cu toxicity significantly increased with the addition of the lowest phosphate concentration, 0.1 mg/mL, and did not recover with increasing phosphate concentration (ANOVA, p < 0.05); this could not be explained by significant changes in metal speciation (Table 1). Cd toxicity did not change with the addition of phosphate at the tested concentrations. The Visual MINTEQ program predicted that the free-ion form predominated for Cu, Zn, and Cd at the four tested phosphate concentrations (Table 1); however, for 100 mg/L of added phosphate, the Zn2+ free-ion species was the lowest of the free-ion species of the four metals. Additionally, the program predicted the formation of neutral monohydrogen phosphate and charged dihydrogen phosphate with Cu, Zn, and Cd, but the amount of ZnHPO4 formed accounted for a higher percentage than the monohydrogen phosphates formed with the other three metals; in fact, only for Zn, a significant correlation between percent maximum luminescence and percent free ion (= −0.85, p < 0.1) was found (Table 2), which would explain, at least partially, the observed remediation of Zn toxicity by a decrease of free Zn2+ ions at 100 mg/L of added phosphate. In the case of Hg, phosphate addition did not substantially changed Hg speciation (Table 1): The neutral dihydroxy species was the predominant form in the tested phosphate range, the free-ion form slightly increased, as calculated by Visual MINTEQ and PHREEQC, from 2.84 × 10−11 M at 0 mg/mL phosphate to 7.07 × 10−11 M at 100 mg/mL added phosphate; this slight increase along with the increase of the neutral dichloro-complex from 0.6% to 1.32% (Table 1) might partially explain the observed Hg toxicity increase (Fig. 4a). In fact, a negative and significant correlation was found between percent maximum luminescence and the logarithm of predicted Hg2+ concentration, (= −0.81, p < 0.1) (Table 2). The observed increase of Cu toxicity could not be explained by changes in speciation (Table 1); additionally, no significant correlation was found between percent maximum luminescence and percent free Cu2+ (Table 2). Finally, 100 mg/L phosphate, on its own, increased luminescence.
Fig. 4

Effect of the addition of phosphate and carbonate on Hg, Cu, Zn, and Cd toxicity to Anabaena CPB4337. * Statistically significant differences (p < 0.05) as determined by ANOVA with respect to the metal-treated control. Error bars represent standard deviation of the means of at least two independent experiments with triplicate samples

Figure 4b shows the effect of increasing carbonate (0.1, 1, 10, and 100 mg/L) on Hg, Cu, Zn, and Cd toxicity to strain Anabaena CPB4337. The addition of 100 mg/L carbonate significantly ameliorated metal toxicity (ANOVA, p < 0.05). Similar to the results with phosphate, there was increased toxicity of the four metals with lower carbonate concentrations (up to 10 mg/L); this could not be explained by significant changes in metal speciation in this range of carbonate concentrations (Table 1). In the case of Cu, Zn, and Cd, the chemical modeling program predicted that the free-ion form predominated up to 10 mg/L of added carbonate, whereas at 100 mg/L, the free-ion form significantly decreased due to the formation of bicarbonate and carbonate complexes (Table 1) that might be nonavailable or less available than the free-ion form and thus less toxic; this might explain the observed significant (ANOVA, p < 0.05) remediation of Cu, Zn, and Cd toxicity with 100 mg/L of added carbonate. Regarding Hg, the neutral dihydroxy complex predominated and its proportion increased with increasing carbonate concentrations. The neutral dichloro-complex significantly decreased at 10 mg/L added carbonate and was no longer present at 100 mg/L (Table 1); the free-ion form concentration, as calculated by Visual MINTEQ and PHREEQC, significantly decreased with increasing carbonate concentrations (from 2.84 × 10−11 M at 0 mg/L to 1.29 × 10−15 M at 100 mg/L added carbonate). The decrease of free-ion Hg2+ and HgCl2 at 100 mg/L added carbonate might account for the observed amelioration of Hg toxicity. In agreement with the toxicity results at higher carbonate concentrations, a significant and negative correlation existed between percent maximum luminescence and percent Cu2+ (= −0.98, p < 0.05), percent Zn2+ (= −0.99, p < 0.01), percent Cd2+ (= −0.99, p < 0.01) and the logarithm of predicted Hg2+ concentration (= −0.99, p < 0.01) (Table 2). Finally, the addition of 100 mg/L carbonate, on its own, increased luminescence significantly (ANOVA, p < 0.05).

The addition of NaCl up to 100 mg/L did not have any significant effect on metal toxicity or speciation (not shown); however, the addition of 2% NaCl, the salt concentration needed for Vibrio fischeri-based bioassays, had a significant effect on cyanobacterial cell luminescence, metal toxicity, and speciation of the four metals. As shown in Fig. 5, 2% NaCl, on its own, significantly (ANOVA, p < 0.05) inhibited cell luminescence (50% inhibition) of the strain Anabaena CPB4337; furthermore, high salt induced substantial metal speciation changes (Table 3) and also affected metal toxicity. In the case of Hg, the addition of 2% NaCl significantly (ANOVA, p < 0.05) increased toxicity (Fig. 5) and substantially changed speciation (Table 3), the hydroxyl complexes were no longer present, and the charged chloro-complex HgCl42– clearly predominated (53.37%); the free-ion form, Hg2+, was present at extremely low amounts (5.33 × 10−18 M), as calculated by Visual MINTEQ and PHREEQC. As the free-ion form concentration was negligible, the toxic metal species could be HgCl42– and, to some extent, the neutral dichloro-complex HgCl2. On the other hand, 2% NaCl significantly (ANOVA, p < 0.05) ameliorated Cu, Zn, and Cd toxicity to the strain Anabaena CPB4337 (Fig. 5). This might be explained by changes in metal speciation (Table 3); the addition of 2% NaCl decreased the available and toxic free-ion forms Cu2+ and Zn2+ by 20% and Cd2+ by almost 95%, with the concomitant increase of chloro-complexes that might be nonavailable and nontoxic.
Fig. 5

Effect of the addition of 2% NaCl on Hg, Cu, Zn, and Cd toxicity to Anabaena CPB4337. * Statistically significant differences (p < 0.05) as determined by ANOVA with respect to the metal-treated control. The 2% NaCl control was significantly (ANOVA, p < 0.05) different with respect to the cell luminescence control. Error bars represent standard deviation of the means of at least two independent experiments with triplicate samples

Table 3

Predicted percentages of the total concentration of metal present as free-ion and dominant forms of metal complexes in the aqueous phase at 2% NaCl as calculated by Visual MINTEQ

NaCl

Hg

Cu

Zn

Cd

0%

Hg(OH)2

93.674

Cu2+

98.117

Zn2+

98.436

Cd2+

99.806

HgOH+

0.148

CuSO4

0.011

ZnOH+

1.558

CdCl+

0.187

HgCl2

1.395

CuOH+

1.227

    

HgClOH

4.719

Cu(OH)2

0.643

    

2%

HgCl2

12.353

Cu2+

77.080

Zn2+

75.323

Cd2+

5.498

HgCl3

34.272

CuCl+

20.585

ZnCl+

21.245

CdCl+

48.748

HgCl42−

53.373

CuCl2

1.026

ZnCl2

1.277

CdCl2

32.845

  

CuOH+

0.747

ZnCl3

1.428

CdCl3

7.530

  

Cu(OH)2

0.525

ZnCl42−

0.443

CdCl42−

5.364

Note: Total metal concentration for calculations of each metal is 10 μM

Discussion

One important characteristic of whole-cell bioreporters is that they reflect the real physiological impact of toxic compounds, as they report on the bioavailable fraction of toxicants. Bioavailability and toxicity of metals depends on their speciation in aqueous environments (Allen and Hansen 1996). Of the possible chemical forms of a metal, the free ion is usually the most toxic one (Campbell 1995); however, that does not necessarily mean that the free metal ion is the only toxic species (Allen and Hansen 1996; Campbell et al. 2000; Fernandez-Piñas et al. 1991).

The study of the effect of modifying factors indicated that, in general, cyanobacterial toxicity correlated with free metal ion concentration (Table 2); chemical modeling predicted that any decrease in the free-ion concentration generally correlated with the corresponding formation of neutral and charged metal hydroxides, metal phosphate, carbonate, and chloride species, and EDTA complexes. The formation of all of these metal complexes, with some exceptions, correlated with toxicity amelioration, implying that they were not toxic probably due to low or no bioavailability to the cyanobacterial cells. One exception was the observed toxicity of the Zn–EDTA complexes at a ratio EDTA/Zn 1:1 (Fig. 3). These complexes could be toxic to the cyanobacteria either by direct uptake or by the ability of the cyanobacterial strains to actively release Zn from the complex, as already pointed out by Paton et al. (1997) in their study with a bioluminescent strain of Pseudomonas fluorescens. Campbell et al. (2000) found a similar behavior in the toxicity of Zn and Cd to a bioluminescent construct of Escherichia coli in the presence of EDTA and fulvic acid. These authors also reported an stimulatory effect of EDTA on cell luminescence; they suggested that this stimulatory effect could be due to a surface permeability effect or uptake/metabolism of EDTA by E. coli cells.

Mercury also represented an exception due to the fact that chemical modeling of Hg speciation predicted that, under almost all the tested conditions, the predominant species (more than 98%) were the neutral Hg(OH)2, Hg(OH)Cl, and HgCl2 while the amount of free Hg2+ ion was very low, almost negligible; so, it was not easy to determine which was the toxic mercury species to cyanobacteria. However, the very low concentration of free Hg2+ ion decreased even more with increasing pH, phosphate, and carbonate and this correlated with the observed remediation of toxicity (Table 2). If free-ion Hg2+ is the main toxic species, the EC50 calculated for Hg (Fig. 1) is actually much lower, with a value of 6.27 × 10−12 M, as calculated by Visual MINTEQ and PHREEQC; thus, in terms of free-ion concentration, Hg2+ appears to be more toxic than Cu2+, Zn2+, or Cd2+ to Anabaena CPB4337. However, we cannot discard that the neutral species HgCl2 might also show some toxicity under certain conditions, as an increase in pH or carbonate concentration clearly diminished toxicity and the concentration of this species; in this regard, Newman and McCloskey (1996) also considered HgCl2 to be bioavailable due to its lipophilicity (Simkiss 1983). Herrero et al. (2005) also found that the macroalga Cystoseira baccata accumulated Hg mainly as HgCl2. In addition, 2% NaCl had a significant effect on Hg speciation, with HgCl42– as the predominant form, and a further decrease in the concentration of the free-ion form, Hg2+, which reached 5.33 × 10−18 M—a value almost identical to that reported by Deheyn et al. (2004) in their Microtox assay medium— indicating that our speciation calculations by Visual MINTEQ and PHREEQC were also accurate for Hg. In spite of the negligible free Hg2+ concentration, this high salt concentration increased Hg toxicity to the strain Anabaena CPB4337, suggesting that the charged chloro-complex HgCl42– might also be toxic species to cyanobacteria.

An interesting feature of metal toxicity to the strain Anabaena CPB4337 is that low amounts of anions such as phosphate and carbonate increased metal toxicity; this could not be related to significant changes in metal speciation (Table 1) but could be due to a modulating effect of these anions, both nutrients for cyanobacteria (Rippka 1988), on metal uptake/toxicity. Heijerick et al. (2003) also found that low hardness levels (as CaCO3) increased Zn toxicity to Daphnia magna; Herrero et al. (2005) reported a slight increase in Hg uptake with background salt concentrations as nitrate but not as chloride salts. Deryabin and Aleshina (2008) recently reported that carbonates and hydrocarbonates had a pronounced inhibitory effect on the bioluminescence of Photobacterium phosphoreum and a recombinant luminescent E. coli strain. As these low anion concentrations can be found in natural waters and they could affect the performance of other bioreporters in a similar way, laboratory tests should be done before assaying environmental samples in order to fully understand ecotoxicity data.

Finally, the point of ecological relevance/significance is important when assessing the performance of bacterial bioreporters. The assays based on marine luminescent bacteria such as Vibrio fischeri might not be very appropriate for soil and freshwater ecotoxicity testing because sample filtration is required and they work only in saline solution (2% NaCl) (Villaescusa et al. 1996). Because of the salinity, the insolubility of some organic substances that might be present in the environmental sample is enhanced, producing turbid solution. As discussed earlier and similar to other authors (Deheyn et al. 2004; Newman and McCloskey 1996), we have found that 2% NaCl substantially changed metal speciation (Table 3); this high salt concentration significantly increased the proportion of chloro-complexes while significantly decreasing the proportion of free ions, which, as discussed earlier, are usually the most toxic forms of a metal. For this reason, the bioassays based on marine luminescent bacteria might underestimate metal toxicity in freshwater simples, as already suggested (Deheyn et al. 2004). In this report, we present an application of a recombinant self-luminescent cyanobacterial bioreporter to assess heavy metal toxicity; this organism is a derivative of a freshwater cyanobacterium, and due to its ecological relevance as a primary producer, it could be used as a potential tool for toxicity assessment in freshwater environments (rivers, effluents, lakes, groundwater, etc.).

Conclusion

We report an application of an ecologically relevant self-luminescent cyanobacterial bioreporter for the assessment of metal toxicity and its modulation in the presence of a range of potential modifying factors. Chemical modeling and correlation analyses proved very useful for linking toxicity and bioavailability. In general, there was a good correlation between the observed toxic effects and free-ion metal concentration. Low concentrations of phosphate and carbonate increased heavy metal toxicity toward the cyanobacterium. This approach of combining toxicity studies with chemical modeling to predict changes in metal speciation might help to interpret complex toxicity data when testing real environmental samples.

Notes

Acknowledgments

This work was funded by Comunidad de Madrid grants 07M/0052/2002, GR/AMB/0084/2004, and S-0505/AMB/0321. Ismael Rodea-Palomares is the recipient of a Ph.D. research contract from Comunidad de Madrid.

References

  1. Allen MB, Arnon DI (1955) Studies on nitrogen-fixing blue grenn algae. I Growth and nitrogen fixation byAnabaena cylindrica Lemm. Plant Physiol 30:366–372. doi:10.1104/pp.30.4.366 CrossRefGoogle Scholar
  2. Allen HE, Hansen DJ (1996) The importance of trace metal speciation to water quality criteria. Water Environ Res 68:42–54. doi:10.2175/106143096X127307 CrossRefGoogle Scholar
  3. Allison JD, Brown DS, Novo-Gradac KJ (1991) MINTEQA2/PRODEFA2, a geochemical assessment model for environmental systems: Version 3.0 User’ manual. EPA/600/3-91/021. US Environmental Protection Agency, Office of Research and Development, Washington, DCGoogle Scholar
  4. Campbell CD, Hird M, Lumsdon DG, Meeussen JCL (2000) The effect of EDTA and fulvic acid on Cd, Zn and Cu toxicity to a bioluminescent construct (pUCD607) of Escherichia coli. Chemosphere 40:319–325. doi:10.1016/S0045-6535(99)00302-1 CrossRefGoogle Scholar
  5. Campbell PG (1995) Interactions between trace metals and aquatic organisms: a critique of the free-ion activity model. In: Tessier A, Turner DR (eds) Metal speciation and bioavailability in aquatic systems. Wiley, New York, pp 45–103Google Scholar
  6. Cook SV, Chu A, Goodman RH (2000) Influence of salinity on Vibrio fischeri and lux-modified Pseudomonas fluorescens toxicity bioassays. Environ Toxicol Chem 19:2474–2477. doi:10.1897/1551-5028(2000)019<2474:IOSOVF>2.3.CO;2CrossRefGoogle Scholar
  7. Deheyn DD, Bencheikh-Latmani R, Latz MI (2004) Chemical speciation and toxicity of metals assessed by three bioluminescence-based assays using marine organisms. Environ Toxicol 19:161–178. doi:10.1002/tox.20009 CrossRefGoogle Scholar
  8. Deryabin DG, Aleshina ES (2008) Effect of salts on luminescence of natural and recombinant luminescent bacterial biosensors. Appl Biochem Microbiol 44:292–296. doi:10.1134/S0003683808030113 CrossRefGoogle Scholar
  9. Fernandez-Piñas F, Wolk CP (1994) Expression of luxCD-E in Anabaena sp. can replace the use of exogenous aldehyde for in vivo localization of transcription by luxAB. Gene 150:169–174CrossRefGoogle Scholar
  10. Fernandez-Piñas F, Mateo P, Bonilla I (1991) Binding of cadmium by cyanobacterial growth media: free ion concentration as a toxicity index to the cyanobacterium Nostoc UAM208. Arch Environ Contam Toxicol 21:425–431. doi:10.1007/BF01060366 CrossRefGoogle Scholar
  11. Fernandez-Piñas F, Leganes F, Wolk CP (2000) Bacterial lux gene as reporters in cyanobacteria. Methods Enzymol 305:513–527CrossRefGoogle Scholar
  12. Heijerick DG, Janssen CR, De Coen WM (2003) The combined effects of hardness, pH, and dissolved organic carbon on the chronic toxicity of Zn to D. magna: development of a surface response model. Arch Environ Contam Toxicol 44:210–217CrossRefGoogle Scholar
  13. Herrero R, Lodeiro P, Rey-Castro C, Vilariño T, Sastre de Vicente ME (2005) Removal of inorganic mercury from aqueous solutions by biomass of the marine macroalga Cystoseira baccata. Water Res 39:3199–3210CrossRefGoogle Scholar
  14. Ho KT, Kuhn A, Pelletier MC, Hendricks TL, Helmstetter A (1999) pH dependent toxicity of five metals to three marine organisms. Environ Toxicol 14:235–240CrossRefGoogle Scholar
  15. Kandegedara A, Rorabacher DB (1999) Noncomplexing tertiary amines as better buffers covering the range of pH 3–11. Temperature dependence of their dissociation contants. Anal Chem 71:3140–3144CrossRefGoogle Scholar
  16. Köhler S, Belkin S, Schmid RD (2000) Reporter gene bioassays in environmental analysis. Fresenius J Anal Chem 366:769–779CrossRefGoogle Scholar
  17. Newman MC, McCloskey JT (1996) Predicting relative toxicity and interaction of divalent metal ions: Microtox® bioluminescence assay. Environ Toxicol Chem 15:75–281CrossRefGoogle Scholar
  18. Parent L, Twiss MR, Campbell PG (1996) Influences of natural dissolved organic matter on the interaction of aluminium with the microalga Chlorella: a test of the free-ion model of trace-metal toxicity. Environ Sci Technol 30:1713–1720CrossRefGoogle Scholar
  19. Parkhurst DL, Appelo CAJ (1999) User’s guide to PHREEQC (Version 2): A computer program for speciation, batch-reaction. One-dimensional transport and inverse geochemical calculations: US Geological Survey Water-Resources Investigations Report 99–4259. US Geological SurveyGoogle Scholar
  20. Paton GI, Rattray EAS, Campbell CD, Cresser MS, Glover LA, Meeussen JCL, Killham K (1997) Use of genetically modified microbial biosensors for soil ecotoxicity testing. In: Pankhurst CF, Doube BM, Gupta VVSR et al (eds) Biological indicators of soil health. CAB International Press, Oxford, pp 394–418Google Scholar
  21. Perona E, Bonilla I, Mateo P (1999) Spatial and temporal changes in water quality in a Spanish river. Sci Total Environ 241:75–90CrossRefGoogle Scholar
  22. Riba F, Garcia-Luque E, Blasco J, Del Valls TA (2003) Bioavailability of heavy metals bound to estuarine sediments as a function of pH and salinity. Chem Spec Bioavail 15:101–114CrossRefGoogle Scholar
  23. Riether KB, Dollard MAD, Billard P (2001) Assesment of heavy metal bioavailability using Escherichia coli ZntAP::lux and CopAP::lux-based biosensors. Appl Microbiol Biotechnol 57:712–716CrossRefGoogle Scholar
  24. Rippka R (1988) Isolation and purification of cyanobacteria. Methods Enzymol 167:3–27CrossRefGoogle Scholar
  25. Simkiss K (1983) Lipid solubility of heavy metals in saline solutions. J Marine Biol Assoc UK 63:1–7CrossRefGoogle Scholar
  26. Szittner R, Meighen E (1990) Nucleotide sequence, expression and properties of luciferase coded by lux genes of a terrestrial bacterium. J Biol Chem 265:16581–16587Google Scholar
  27. Tauriainien SM, Virta MPJ, Karp MT (2000) Detecting bioavailable toxic metals and metalloids from natural water samples using luminescent sensor bacteria. Water Res 34:2661–2666CrossRefGoogle Scholar
  28. van Dijk GM, Van Liere L, Admiraal W, Bannik BA, Cappon JJ (1994) Present state of the water quality of European rivers and implications for management. Sci Total Environ 145:187–195CrossRefGoogle Scholar
  29. Villaescusa I, Martinez M, Pilar M, Murat JC, Hosta C (1996) Toxicity of cadmium species on luminescent bacteria. Fresenius J Anal Chem 354:566–570Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  • Ismael Rodea-Palomares
    • 1
  • Coral González-García
    • 1
    • 2
  • Francisco Leganés
    • 1
  • Francisca Fernández-Piñas
    • 1
  1. 1.Departamento de Biología, Facultad de CienciasUniversidad Autónoma de MadridMadridSpain
  2. 2.Instituto de Salud Carlos IIICentro Nacional de MicrobiologíaMadridSpain

Personalised recommendations