Herbicide bioconcentration in algae: studies on lipophilicity-sorption-activity relationships (LSAR) with Chlorella fusca

Herbicide bioconcentration in algae: studies on lipophilicity-sorption-activity relationships (LSAR) with Chlorella fusca

The Science of the Total En~ronment, Supplement 1993 Elsevier Science Publishers B.V., Amsterdam 453 Herbicide bioconcentration in algae: studies on...

324KB Sizes 0 Downloads 6 Views

The Science of the Total En~ronment, Supplement 1993 Elsevier Science Publishers B.V., Amsterdam

453

Herbicide bioconcentration in algae: studies on lipophilicity-sorption-activity relationships (LSAR) with Chlorellafusca Martina Manthey, Michael Faust, Susanne Smolka and L. H o r s t Grimme Institut fi~r Zellbiologie, Biochemie und Biotechnologie, FB 2, Unioersiti~tBremen, D-2800 Bremen 33, Germany

ABSTRACT Lipophilicity-sorption-activity relationships (LSAR) were studied for a congeneric series of 15 herbicidal phenylureas with the unicellular green alga Chlorella fusca as a model organism. Data for substance lipophilicity (log kw), algal bioconcentration (log BCF) and algal toxicity (pls0 CV/CN) were experimentally determined and subsequently subjected to a series of regression analyses. These studies revealed only moderate correlations between log k w and log BCF and between log k w and pls0 values. In contrast, a highly significant relationship was observed between bioconcentration and toxicity. The following conclusions were drawn on basis of the results. (i) The predictive value of lipophilicity for estimating herbicide bioconcentration and toxicity in microalgae is limited. (ii) Different levels of phenylurea toxicity for Chlorella seem to be attributable to different degrees of bioconcentration. (iii) An integrated consideration of lipophilicity, bioconcentration and toxicity should be advantageous in aquatic ecotoxicological hazard assessment. Key words: Aquatic toxicology; Herbicides; Micro-algae (Chlorella); Bioconcentration;

Lipophilicity; Phenylureas INTRODUCTION Bioconcentration and toxicity of pesticides, determined by the use of selected organisms from different trophic levels and taxonomic groups, are important criteria for ecotoxicological hazard assessment for aquatic environments. Alternatively to an experimental determination, predictive modelling of these parameters from the physicochemical properties of a substance is gaining increasing acceptance. In correspondence to a simple partition model for the bioconcentration of xenobiotics by aquatic 1993 Elsevier Science Publishers B.V.

454

u. MaNa'HEVEr AL.

organisms, lipophilicity has been established as a suitable indicator of both the bioaccumulative and the toxicological potentials of organic chemicals with unspecific modes of action [1]. Concerning the prediction of the effects of pesticides on the level of aquatic primary production three major gaps in knowledge can be identified: -

-

--

Investigations on aquatic bioconcentration have been focused mainly on fish, while data on algae are rare. The predictive value of lipophilicity for the estimation of bioconcentration and toxicity of substances with specific modes of action is not clear. The existence of specific binding sites might result in a more complex, multiphasic process of adsorption and absorption of toxicant molecules [2]. The interrelationships between bioconcentration and algal toxicity have not been examined.

The purpose of this work, therefore, was the integrative characterization of a lipophilicity-sorption-activity relationship (LSAR) for a congeneric series of herbicidal phenylureas using the unicellular green alga Chlorella fusca as a model organism. METHODS

Common names of the 15 phenylureas used are listed in Table 1 by order of increasing lipophilicity. They are all known to be specific inhibitors of the electron flow in photosystem II of autotrophic plants. In order to characterize phenylurea lipophilicity the reversed-phase liquid chromatographic retention parameter log k w [3,4] was determined with an rp-HPLC system for each compound. For these agents log k w is correlated linearly to the octanol/water partition coefficient log P [4], thus permitting a qualitative comparison of our results with those of similar studies. Herbicide sorption by synchronous populations of algae was determined as the disappearance of the toxicants from the nutrient medium [5]. Bioconcentration factors BCF, defined as the ratio between the toxicant concentration in the algae and in the surrounding medium [6], were calculated from steady-state concentrations achieved within 30 min [5]. Control experiments confirmed that equilibrium conditions (constant concentrations in the aqueous phase) were reached for each test compound. Blanks without algae were run within each experiment to correct results for the loss of substances due to glass wall sorption and the sample preparation procedures. Figure 1 illustrates the experimental procedure.

455

LIPOPHILICITY-SORPTION-ACTIVITY RELATIONSHIPS (LSAR) IN CHLORELLA FUSCA

The methods used do not allow for a distinction between adsorption and absorption of toxicant molecules to microalgae; in accordance with other authors [2,5] we therefore use the term 'sorption' to comprise both processes. However, there is strong evidence that adsorption of phenylureas to the cell walls of microalgae is negligible [5]. Additionally, rp-HPLC analysis revealed no relevant potential for phenylurea (bio-) transformation under the experimental conditions. Thus, for the purpose of determining bioconcentration factors, we can calculate 'concentrations of phenylureas in the algae' by relating the total amount of toxicant sorbed to the volume of the cells. The inhibition of cell volume growth I(CV) and reproduction I(CN) during one generation of synchronized algal culturing were used as toxicity parameters and determined as described previously [7]. pls0 values ( = -log 150) were estimated statistically by probit analysis. RESULTS

The phenylureas examined accumulated by a factor of 35 to 2600 relative to the environmental concentration in C. fusca. Toxicity, expressed as 150 TABLE 1 Lipophilicity (logkw) , algal bioaccumulation (log BCF) and algal toxicity (pls0) of phenylurea herbicides Common name

logk w

log BCF

pls0(CV) a

pls0(CN) b

1. Fenuron 2. Metoxuron 3. Monuron 4. Monulinuron 5. Chlortoluron 6. Metobromuron 7. Fluometuron 8. Isoproturon 9. Diuron 10. Buturon 11. Difenoxuron 12. Linuron 13. Chlorbromuron 14. Chloroxuron 15. Neburon

1.47 2.08 2.28 2.65 2.76 2.80 2.88 2.95 3.01 3.02 3.16 3.32 3.43 3.95 4.33

1.54 2.39 2.33 2.28 2.46 2.16 1.96 2.60 3.41 2.25 2.66 2.88 3.03 3.02 3.10

5.41 6.49 6.25 5.92 6.90 6.06 5.88 6.70 7.42 6.19 7.04 7.14 7.45 7.52 7.45

5.49 6.45 6.39 5.98 6.95 6.13 5.88 6.86 7.48 6.21 6.89 7.15 7.52 7.48 7.38

a _ log concentration causing 50% inhibition of cell volume growth (14 h) b _ log concentration causing 50% inhibition of algal reproduction (24 h)

456

M.MANTHEYETAL.

harvesting of synchronized algae (autospores) sorption experiment -cell density: -toxicant concentration: -incubation: -parallel controls without

8 * 108.,cell/ml 7 * 10° ~ mol/l 30 min, in the dark, 22°C, pH 6.3

algae sample preparation -solid phase extraction -vacuum concentration rp-HPLC analysis of sorption determined as the disappearance of the toxicant from the nutrient medfum during incubation

calculation of BCF -from steady state concentrations -related to the algal volume Fig. 1. Determination of the bioconcentration factor B C F in Chlorella fusca.

values, also spanned two orders of magnitude from 3.10 -8 to 4.10 - 6 tool/1 (0.009-0.6 ppm). These data are compiled in Table 1 together with the values obtained for the lipophilicity parameter log kw. They were subjected to a series of regression analyses as illustrated in Figs. 2 to 4.

Lipophilicity (log k w) vs. algal bioconcentration (log BCF) Linear regression analysis with phenylureas revealed a moderate correlation between log k w and log BCF (Fig. 2). Limitations in the predictive value of lipophilicity for the estimation of phenylurea bioconcentration became particularly evident for the exceptionally high BCF value observed for diuron.

Lipophilicity (log k w) vs. algal toxicity (plso) Log k w values and the inhibition of algal reproduction (pls0 CN) (Fig. 3) and algal growth (pls0 CV) (not shown) were also only moderately correlated. The correlation coefficient was similar to that of log k w vs. log BCF. The results confirm that lipophilicity is not a precise instrument in modelling the bioconcentration and toxicity, e.g. long-term effects during one generation, of herbicides on algae. The correlations are also weaker than those reported for the relationship between log P and the short-term inhibition of photosynthetic electron flow in isolated chloroplast membranes (Hill reaction) within less than 1 rain [8].

L I P O P H I L I C I T Y - S O R P T I O N - A C T I V I T Y RELATIONSHIPS (LSAR) IN CHLORELLA FUSCA

457

4

09 O

g.

° 15

~

(J D0

J

07

y = 0.53x + 0.99

O,

r = 0.7496

I

i

(n=15) i

I

i

n

i

I

2

i

n

a

3

I

a

i

4

n

5

log kw Fig. 2. Correlation between the lipophilicity parameter log k w and the algal bioconcentration factor log BCF for 15 phenylurea herbicides (see Table 1).

Bioconcentration (log BCF) vs. algal toxicity (plso) A highly significant correlation, however, was observed between the bioconcentration and the toxicity (Fig. 4). This suggests that different degrees of accumulation might account for the differences in the toxic potencies of the various phenylurea herbicides in Chlorella.

ols

¢¢ 0 4.,* 0 0 lb. @

m"

j

.,;

el

(n= 15) I

2

l

l

l

l

l

l

3

l

l

l

4

l

l

5

log kw Fig. 3. Correlation between the lipophilicity parameter log k w and the inhibition of algal reproduction pls0(CN) for 15 phenylurea herbicides (see Table 1).

458

M.u~VrHEVET~.

.o "o

7

"to o Lo

rr

y =

6

(n

i

I

2

i

i

i

I

i

3

+ 3.43

Z 0.9584

r

=

1.28x

i

151

i

I

i

i

i

4

log BCF Fig. 4. Correlation between the algal bioconcentration factor log BCF and the inhibition of algal reproduction pls0(CN) for 15 phenylurea herbicides (see Table 1).

The knowledge of the bioconcentration factor allows calculation of the internal amount of a toxicant taken up by a single algal cell of known size at a given environmental concentration. Thus, dose-response relationships, usually unknown in aquatic toxicology, can be estimated from the concentration-response relationships. Applying this procedure to our data, i.e. calculating theoretically- the internal amounts of penylureas in Chlorella, reduces the range of 150 values to less than one order of magnitude (0.8-5.5.10-18mol/cell; mean cell volume 23.6/xm3). DISCUSSION The results of this study support the idea that bioconcentration is a prerequisite for herbicide toxicity in microalgae but that it is not always adequately described by the overall lipophilicity of the molecule. Besides partitioning between the lipid phase of cell membranes and the medium, specific binding to photosystem II as well as unspecific binding to proteins might play a considerable role. As we demonstrated for the example of phenylureas, this more complex process of sorption and the resulting degree of bioconcentration can influence the toxicity. Thus, predicting bioconcentration and toxicity via lipophilicity may lead to underestimations in the case of substances with a specific mode of action. In an effort to expand the predictive capabilities in aquatic toxicology the integrated consideration of toxicity and bioconcentration should provide

LIPOPHILICITY-SORPTION-ACTIVITY RELATIONSHIPS (LSAR) IN C H L O R E L L A FUSCA

459

good first approximate estimates of toxic internal concentrations evoking a specific biological response. Here it is necessary to gain more experience in working with internal doses of chemicals so that processes and interrelationships involved with bioconcentration and specific toxicological endpoints may be understood better. Summing up, the following conclusions can be drawn from our results: (i)

The predictive value of lipophilicity for estimating herbicide bioconcentration and toxicity in microalgae is limited. (ii) Different levels of phenylurea toxicity for Chlorella seem to be attributable to different degrees of bioconcentration. (iii) An integrated consideration of lipophilicity, bioconcentration and toxicity should be advantageous in ecotoxicological risk assessment. REFERENCES 1 L.S. McCarty, The relationship between aquatic toxicity QSARs and bioconcentration for some organic chemicals. Environ. Toxicol. Chem., 5 (1986) 1071-1080. 2 DFG (Deutsche Forschungsgemeinschaft), Bioakkumulation in Nahrungsketten, VCH, Weinheim (1987). 3 T. Branmann, G. Weber and L.H. Grimme, Quantitative structure-activity relationships for herbicides - - Reversed-phase liquid chromatographic retention parameter (log k w) versus liquid-liquid partition coefficient as a model of the hydrophobicity of phenylureas, s-triazines, and phenoxycarbonic acid derivatives. J. Chromatogr. 261 (1983) 329-343. 4 T. Braumann, Determination of hydrophobic parameters by reversed-phase liquid chromatography: Theory, experimental techniques, and application in studies on quantitative structure-activity relationships. J. Chromatogr., 373 (1986) 191-225. 5 W. Neumann, H. Laasch and W. Urbach, Mechanisms of herbicide sorption in microalgae and the influence of environmental factors. Pestic. Biochem. Physiol., 27 (1987) 189-200. 6 C.H. Walker, Kinetic models for predicting bioconcentration factors. Environ. Pollut., 44 (1987) 227-239. 7 R. Altenburger, W. B6deker, M. Faust and L.H. Grimme, Evaluation of the isobologram method for the assessment of mixtures of chemicals. Combination effect studies with pesticides in algal biotests. Ecotoxicol. Environ. Saf., 20 (1990) 98-114. 8 E. Kakkis, V.C. Palmire, C.D. Strong, W. Bartsch, C. Hansch and U. Schirmer, Quantitative structure-activity relationships in the inhibition of photosytem II in chloroplasts by phenylureas. J. Agric. Food Chem., 32 (1984) 133-144.