Research Articles Toxicity Testing of HVC Toxicity Testing of Highly Volatile - A New Assay Chemicals with Green Algae Werner Brack, Horst Rottler Chair of Ecological Chemistry and Geochemistry, University of Bayreuth, D-95447 Bayreuth, Germany Abstract A gas-tight system for toxicity testing of highly volatile chemicals with the green alga Chlamydomonas reinhardtii was developed. The procedure permits maintenance of constant and defined concentra- tions of the tested compounds in the vessels. To ensure sufficient CO2-supply , new bipartite test vessels were used. These vessels allowed spatial separation of a HCO3/CO3 2" buffer used for CO2 supply and the alga culture to avoid growth inhibition due to ionic strength. Severalvolatile chlorinated hydrocarbons have been tested. Their EC10 valueswere severalorders of magnitude lower than those obtained with open test systems. 1 Introduction Until now, there has been no standardised or widely used method for toxicity testing of volatile chemicals in a growth inhibition bioassay with green algae. A suggestion for such an assay is presented in this investigation. A cell multiplication assay with green algae for volatile com- pounds must meet two important requirements: 1. It must provide a sufficient supply of CO2 to avoid growth inhibition due to CO2 deficiency. 2. There must be a well-defined, constant concentration of the tested compounds during incubation time. The easiest way to meet the first demand is to use systems allowing free gas exchange with the atmosphere. Such tests are standardised in the OECD Guideline for Testing of Chemicals, 201 (1984), which does not specify a dinstict alga species and in DIN 38412 Part 33 (1989) with Scenedesmus subspicatus. With volatile chemicals (BRINGMANN& KOHN, 1978 and GEYER et al., 1985) the evaporation during the test makes it impossible to define the effective concentrations in the test flasks. If the calculated EC-values are based on the concentrations of stock solutions, the toxicity of the volatile compounds will be highly underestimated. BRINGMANN & KOHN (1978), for example, achieved toxici- ty thresholds for tetrachloromethane and trichlorethylene beyond the solubility in water. However, even in closed test systems, analytical control of the actual concentration of the test compound is very important because of inevitable losses during the preparation of the test solutions. Several attempts have been undertaken to maintain constant and well-defined concentrations of the tested chemicals. All of them used sealed systems, however with different ap- proaches to achieve a sufficient supply of CO 2. KOHN & PATTAgD (1990) shortened the incubation time from 72 hours, as recommended in DIN 38412 Part 33 (1989), to 48 hours to reduce the CO 2 demand. GALASSl & VIGHI (1981) used 2-L flasks, containing only 100 mL of medium, to provide a head space volume large enough to avoid CO2 deficiency. HERMANet al. (1990) added 0.4 % NaHCO 3 to the medium to ensure CO2 supply. GALASSl& VIGHI(1981) and HERMAN et al. (1990) controlled the acutal concentra- tions of the test compounds analytically, whereas KOHN & PATTARD (1990) did not. All of these investigations showed a drastically higher sen- sitivity of green algae to volatile compounds compared to those obtained by BglNGMANN & KOHN (1978) or GEaR et al. (1985). All of the methods are capable of reducing the problem of CO 2 deficiency in the cultures, but our own ex- periments showed that CO 2 remained the limiting factor for algal growth in such systems. According to NYHOLM & K~LQViST (1989), COz-limited test systems are less sen- sitive because toxic effects will not be expressed unless poten- tial growth is reduced so far that COz transport to the cells does no longer limit algal growth. Also CHEN'(1989) showed in a theoretical approach that nutrient limitation reduces sensitivity of growth inhibition tests. Further disadvantages are the high space demand for the method of GALASSI & VIGHI (1981) and growth inhibition due to the high ionic strength if NaHCO 3 is added to the medium in concentra- tions high enough to avoid CO 2 deficiency. Therefore, in this investigation a new, closed system is used, in which a KHCOJKzCO 3 buffer to supply the algae with CO2 is employed, but the buffer is separated from the test medium to avoid growth inhibition due to the ionic strength. This was done by using bipartite culture flasks as described later in detail. Similar vessels have been already used in plant physiology for the culturing of Chenopodium cells (HOSEMANN & BARZ, 1977). To estimate sensitivity of the test organism Chlamydomonas reinhardtii, toxicities of several nonvolatile and low volatile compounds, which have been already tested with Scenedesmus subspicatus by BgrNGMAhrN & KOHN (1978), have been determined. Aside from these non-volatile com- pounds, highly volatile chlorinated hydrocarbons have been tested with this system. These substances, primarily used as solvents, are produced in large amounts and released to the environment. They have been found in almost every com- partment of the environment. Until now, almost no realistic toxicity data achieved with closed test systems have been available for these chemicals. ESPR-Environ. Sci. & Pollut. Res. 1 (4) 223-228 (1994) 9 ecomedpublishers, D-86899 Landsberg,Germany 223