Application of enzyme assays for toxicological water testing

Application of Enzyme Assays for Toxicological Water Testing U. OBST,* Stadtwerke Mainz AG, Rheinallee41,D-6500 Mainz, Federal Republic of Germany,A. HOLZAPFEL-PSCHORN, and M . WIEGAND- ROSINUS, ES WE -Znstitut fur Wasserforschung und Wassertechnologie, Sohnleinstrasse 158,D-6200 Wiesbaden-Schierstein, Federal Republic of Germany Abstract Two different criteria for biochemical toxicity testing are proposed the effect of hazardous substances on the metabolic activity of the natural microflora and the determination of such substances by a standardized enzyme testing system. To measure effects on the microbial metabolic activity, there are various simple in uiuo enzyme assays using chro- mogenic or fluorogenic substrates. To detect unknown substances with harmful effects, there exist some in uitro enzyme tests with commercial preparations of defined purity and sensitivity. The performance, sensitivity, and some data of the practical use of such tests are presented. The applicability of the two testing protocols is discussed. INTRODUCTION For critical examination of harmful water compounds there is a number of microbial tests that are partly documented in legislation (Anony- mous, 1967; Hamburger, 1983). There are principally two different criteria for testing toxicological effect (Fig. 1): 1. The effect of hazardous substances on the natural biocenosis in water (ecological toxicity) and 2. the determination of unknown substances and their possible effects in water by a standardized testing system (model-simu- lated toxicity). Damage to the turnover activity of a natural biocenosis can involve dramatic consequences for water quality because the microbial self- purification is central to ecologically intact waters and biological treat- *To whom correspondence should be addressed. Toxicity Assessment: An International Journal Vol. 3, 81-91 (1988) 0 1988 John Wiley & Sons, Inc. CCC 0884-8181/88/03081-11$0.400 82/OBST ET AL. 2a;dized \System Test \ Enzyme Activity Test of the Natural Model- Simulated -TcDcicity Determination of Inhibi - tor Effects of Water Sugstances in a Defined Enzyme System Fig. 1. Ecological and model-simulated toxicology in water analyses. ment processes. Poisoning will change the physiological behavior of microorganisms before they eventually die. Rapid and easy determina- tions of the catabolic activity can be performed with biochemical tests, and allow a direct measurement of different enzyme activities in uiuo. The effects of pollutants detected with these tests depend on the exam- ined microflora, which may already be adapted to a certain level of pollution. Therefore, enzyme activities in uiuo are often only inhibited at relatively high noxious levels. The detection of toxic substances in water can be performed with a defined test system. Under optimal conditions this system is related to the ecosystem; otherwise, it can be chosen by its sensitivity, reproduc- ibility, etc. These model systems afford a calibration of their sensitivity with known pure toxic substances in order to judge the effects of samples of unknown composition. Instead of distinct species of microorganisms such as luminous bacteria (Krebs, 1983), defined enzyme preparations in uitro can be used for tests of acute toxicity (Weil, 1974; Obst et al., 1985). Such measure- ments can be done with commercial products of standardized purity and sensitivity. The advantages of enzyme-testing systems in uitro are easy and mostly inexpensive supply by the distributor, standardized composition and activity, ENZYME ASSAYS APPLICATIONI83 easy storage, easy operation with common analytical equipment, rapid performance, and group- or substance-specific sensitivity. Although the data obtained with these tests can only be transferred on a limited basis to the ecosystem, we get a very effective early warning system that can be analyzed before more extensive tests with other organisms or chemical analyses are performed. METHODS Test Principles As the methods are already published in several papers (Obst et al., 1985; Obst, 1985; Holzapfel-Pschorn et al., 19871, we provide a short overview of the test principles and the available assays in the following. In order to have a simple test process for the determination of enzyme activities in uiuo, chromogenic or fluorogenic substrates are added directly to the water samples. The corresponding enzymes cleave these compounds and release the originally bound dye, which can be detected in a normal spectrophotometer or spectrofluorometer (Obst, 1985; Holzapfel-Pschorn et al., 1987). After calibration and under de- fined reaction conditions, such as a linear turnover rate during incuba- tion, the substrate turnover of distinct enzymes or enzyme groups can be calculated per sample, volume, and time. The tests are performed under laboratory conditions and with synthetic substrates in optimal concen- trations. Thus we get data that correspond to the tested biocenosis, but a direct transfer of the determined substrate turnover to the biotope is normally not possible. The effects of unknown pollutants in a water sample on a standard- ized testing system can be examined with in uitro enzyme assays. En- zyme reactions that are well examined, and for which the response and effect of certain inhibitors are known, should be preferred. Working with a commercial test kit seems an obvious choice. For the required results this procedure is disadvantageous, however, because the concen- trations of the reactants and the reaction conditions in these kits are optimized for the determination of enzymes or substrates. In order to get a sensitive detection of pollution effects, it is better to use “stressed systems, because the sensitivity for inhibitory effects increases. There- fore, it is advantageous to standardize the components and conditions of the used enzyme reaction on a level that allows, on the one hand, a 84/OBST ET AL. reproducible reaction under undisturbed conditions and that is, on the other hand, sensitive to low concentrations of inhibitors. For the stan- dardization of such a test an extensive screening is necessary in which the most suitable compromise between reproducibility of the test proce- dure and sensitivity of the test system has to be found. To avoid results that are too sensitive and unpredictive, it is more opportune to choose an inhibitor-free water sample with a composition similar to that investigated as a relative base and to control for undis- turbed activity instead of distilled water. For example, samples taken before introduction of waste water can be compared to samples taken afterward, etc. Available Tests The in uiuo and in uitro enzyme tests presented in Table I were presently applied to investigate turnover activities and toxic effects in water (Obst et al., 1985; Obst, 1985; Holzapfel-Pschorn et al., 1987). Until now, mainly central metabolic activities were selected for microbial enzyme tests in uiuo; chromogenic and fluorogenic substrates for these tests are both commercially available in the most cases (Table I). Analogous to the microbial degradation of water substances, we selected single enzymes or groups degrading polymeric substrates (es- terases, amylases, proteases), others degrading oligomeric compounds TABLE I Enzymatic in vivo and in vitro tests In vivo tests: Biochemical assay of the In uitro tests: Test principle microbial degradation of organic substrates Toxicity assay Investigated 1. enzyme activity 2. 3. Catalyse of polymeric substrates urease Esterases" (ester-bond cleavage) Proteases (protein cleavage) ATP-dependent Amylases (plysaccharide cleavage) luciferase Catalyse of oligomeric substrates Aminopeptidase" (peptide cleavage), a-respectively p-glucosidase" (disaccharide cleavage) Phosphatase" (phosphomonoester cleavage) Energy metabolism Dehydrogenasedelectron transport system 'Tests can be performed with photometric as well as with more sensitive fluorometric detection. ENZYME ASSAYS APPLICATION/85 (phosphatases, glucosidases, aminopeptidases), and enzyme groups that catalyze terminal metabolic reactions (electron transport system). The choice between chromogenic and fluorogenic substrates allows the se- lection of the sensitivity range according to the density of microbial colonization in a sample. The presented spectrum of tests cannot be regarded as complete, especially in relation to the complexicity of de- composition processes in natural biocenoses. Therefore, further tests concerning other metabolic pathways are presently underway. The number of enzyme assays in vitro is still relatively low. We recently developed two enzyme tests that are similar to the long-estab- lished cholinesterase test (Weil, 1974). The first test system is based on the enzymatic cleavage of urea to ammonia and carbon dioxide by urease (Obst etal., 1985). This enzyme reaction was selected because the microbial degradation of urea is a typical reaction in waste water. In addition, urease is very sensitive to heavy metals, and the whole proce- dure can easily be performed. The influences on the enzymatic cleavage of urea are determined either electrochemically or titrimetrically for the pH value increases due to the release of ammonia. The second test system is based on the ATP-dependent luciferase reaction (Wiegand-Rosinus, 1987, personal communication), which has been used for many years to determine ATP concentrations in biological materials. Although this enzyme reaction has no ecotoxicological signif- icance, it was selected as a “pretest” because of its good and simple performance. The development of the test instruction has yet not been completed, but some data concerning the test’s sensitivity can already be presented. More tests with commercially available enzyme prepara- tions are underway or planned, in which the focus will be the specific detection of inhibitory effects of pesticides. RESULTS The application of enzyme tests in uiuo is recommended if the effects of pollutants on the biocenosis are to be determined at specific sites as, for example, during embankment filtration. Showing the natural fluctua- tions of the metabolic activity of the investigated biocenosis baselines of the enzyme activities within a longer characteristic period are absolutely necessary. Only if such data are available can the consequences of accidents or changes in the biotope be clearly stated. The field of operation for model systems in vitro is, for example, monitoring the extent of inhibitors at treatment plants or surface waters. As with the enzyme tests in uiuo, it is useful to determine 86/OBST ET AL. a base level with the in viuo tests for samples that will be investigated periodically. Although the sensitivity and reactivity of the enzyme activities in uiuo are often lower than those of the standardized model systems, dis- tinct inhibitory effects were observed by adding pollutants to samples of the Rhine River. Fig. 2 shows the inhibition of several metabolic activi- ties in the Rhine water by mercury and cadmium. It is obvious that different microbial metabolic pathways react differently to the poisons. Therefore, an influence on the microflora is better seen with a spectrum of enzyme tests than with only a single one. Despite the relatively low sensitivity of adapted microorganisms, Fig. 3 shows that, compared to the standardized sensitive in vitro tests, the metabolism of the Rhine microflora is vulnerable to different haz- ardous heavy metals. Enzyme groups like the esterases represent a summarizing parameter of heterotrophic activity; they thus are less sensitive than specific enzymes such as P-glucosidase. 10000 - 5000 - 1000 - 500 - 0.05 0.01 Fig. 2. Comparison of ECSo levels of mercury and cadmium for various microbial enzymes of the Rhine River microflora. ENZYME ASSAYS APPLICATION/87 Fig. 3. EC50 levels of mercury, cadmium, and lead for the in v i t ~ o activity of enzymes and microbial enzyme activity in uiuo. The sensitivity of the in uitro enzyme tests with urease and ATP- dependent luciferase can be compared with well-known methods such as the test with luminous bacteria or the fish test (Table 11). With in uitro enzyme tests, substance-specific effects can be observed. For example, the urease reacts very sensitively to various heavy metals due to the SH group in their catalytic center. Almost no effect is recorded with organic substances (Table 11). For the ATP-dependent luciferase, no significant trend can be seen so far; but test development has still not been com- pleted. The values presented in Table I1 give only a tentative view of the results. In several research projects we could record baselines of some en- zyme activities in uiuo so that, e.g., a correlation with the water volume flow rate of the Rhine can be shown. Thus, we already had some princi- pal data sets when, in the autumn 1986, several chemical industry accidents occurred within a short period and toxic substances were intro- duced into the river. Despite the rising problems in our treatment plants, we had the opportunity to perform enzyme tests in uiuo as well as in uitro in order to characterize these accidents. In Fig. 4a the activity of urease in uitro is graphed together with the concentration of rhodamine, 88lOBST ET AL. TABLE I1 Comparison of urease EC50, luminous bacteria EC50, and fish LC50 data using pure compounds Investigated Urease assay Luciferase assay Bacterial luminescence Fish assay substance (EC50 in mg@ (EC50 in mg/L) (EC50 in mg/L)' (LC50 in mg/L,' Mercury (11) 0.007 0.007 0.046 0.37 Copper (II) 0.005 0.26 0.4 0.32 Cadmium (11) 5.0 0.2 8.0 4.2 Chromium (VI) 52 - 70 29- 133 Lead (11) 0.6 1.0 0.4 236 Zinc (11) 40 - 2.0 10 Formaldehyde 10 100 - 310 100 Phenol No inhibition 100 25 14 m-Cresol No inhibition - 8.2 17 Nitrite > 100 - 420 240 "EC, effective concentration; LC, lethal concentration. "From Obst et al. (1985). =From Krebs (1983). which accompanied some pesticides and other pollutants into the Rhine. Therefore, rhodamine could be used as a simple and rapidly detectable tracer for the pollutant wave after this accident. As expected, the activ- ity inhibition is distinct at the highest rhodamine concentration, which means in the maximum of the pollutant wave. Later we see an increase of the activity, but not to the previous level, then a decrease again. Comparing the urease inhibition in uitro to the in uiuo enzyme activity of the microflora, an opposite tendency between inhibition test and microbial turnover activity in viuo can be seen in the maximum of the pollutant wave (Fig. 4b). The same phenomenon, but more distinct, however, was observed, when 2,4-dichlorophenoxyacetic acid (2,4-D) was introduced by the BASF corporation (Fig. 5) . Again an opposite tendency between turn- over activity in uiuo (esterases) and enzyme inhibition in uitro (urease) was observed. We assume that the introduced pollutants have been partly metabolized and that the metabolites were even more toxic than the original pollutants. Other working groups observed a higher de- crease of the microbial turnover activities downstream, which means several days later. This was possibly once again due to the higher ENZYME ASSAYS APPLICATION189 actMty(Y$octlvity 200 150 - lOo--O.l 50- B 0 1,5--300 -250 l,O--ma -150 0.5 -1M) (% STI I h) - ureose (nvitro test) (in vim test) --- Esterases --0.2 ‘\ \ \ I 7 / I L-4 , - I I + A A b 5 Nw6 Date 9o/OBST ET AL. - Urease Esterases November 21 -22.1986 2 0 2 1 2 2 2 3 2 L 1 2 3 L 5 6 7 8 Local Time lhr l Fig. 5. Comparison of the inhibition of urease activity in uitro and of the activity of esterases in viuo in the Rhine River during the pollution with 2,4-D. concentration of toxic metabolites, which would underline our hypothe- ses. Unfortunately, no additional parameters were available to evalu- ate this assumption. These first experiences call for more detailed investigations of such phenomena. In our field of concern it is necessary to record, for a longer time period, the metabolic activity in the Rhine River together with water volume flow rate, chemical analyses, biomass, and inhibition tests with enzymes in uitro. In addition, we shall try to clarify the production of metabolites and their influences in a future research project. SUMMARY Determinations of microbial enzyme activities in uiuo as well as inhibi- tion tests with enzyme preparations in uitro have been shown by first experiences to be suitable for toxicological water analyses. The behavior of microbial enzyme activities in uiuo during accidents can allow an ecotoxicological characterization of the microbial self-purification. The ENZYME ASSAYS APPLICATIONI91 enzyme tests in vitro are recommended to be performed as simple detec- tion systems before more complicated analyses are started. It is important to determine a database for the investigated biotope for a characteristic time in order to predict the consequences of accidents or other influences. For the practical interpretation of the in vitro inhibition tests it is useful t o take a sample of water as control for undisturbed activity that is as unpolluted as possible, but similar to the sample to be tested. Investigations with a range of methods are advantageous, because the results obtained by only one single test system are often not suffl- cient in the case of an unexpected pollution. A combination of in vivo and in vitro tests is recommended, because ecotoxicological and model-simulated toxicological aspects can be de- termined with the same sample and prospectively important informa- tion about the possible formation and influence of metabolites can result. We thank the Bundesministerium fur Forschung und Technologie and the Umweltbundesamt for financial support. References Anonymous. 1967. Gesetz uber Abgaben fur das Einleiten von Abwasser in Gewasser Hamburger, B. 1983. Stand der Normung von Biotesten. Vom Wasser 61:227-235. Holzapfel-Pschorn, A,, U. Obst, and K. Haberer. 1987. Sensitive methods for the determi- nation of microbial activities in water samples using fluorigenic substrates. Fresenius Z. Anal. Chem. 327521-523. Krebs, F. 1983. Toxizitatstest mit gefriergetrockneten Leuchtbakterien. Gewasserschutz, Wasser, Abwasser 63:173-230. Obst, U., K. Resch, and Th. Feuerstein. 1985. Einfacher Toxizitatstest fur Wasser und Abwasser auf biochemischer Basis. Vom Wasser 65199-202. Obst., U. 1985. Test instructions for measuring the microbial metabolic activity in water samples. Fesenius Z. Anal. Chem. 321:166-168. Weil, L. 1974. Enzymatischer Test zur Bestimmung von Organophosphorpestiziden. Hydrochem. Hydrogeol. Mitt. 1:111-119. (Abwasserabgabengesetz). Bundesgesetzblatt Z2721-2726.