COMPARISON OF INHIBITION ASSAYS USING NITROGEN REMOVING BACTERIA: APPLICATION TO INDUSTRIAL WASTEWATER CAMILLA GRUNDITZ* * M , LENA GUMAELIUS and GUNNEL DALHAMMAR * M Department of Biochemistry and Biotechnology, Royal Institute of Technology, KTH, SE-100 44 Stockholm, Sweden (First received March 1997; accepted in revised form January 1998) AbstractÐThree pure culture assays for measurements of inhibition of the nitrogen removal process were tested and compared. The methods were performed using pure cultures of bacteria responsible for three reactions; ammonia oxidation, nitrite oxidation and nitrite reduction. The inhibition caused by 48 samples of industrial wastewater was determined and the results for the three inhibition assays were compared. The correlation coecients were calculated to be between 0.63 and 0.73. The inhibition of the pure culture methods was also compared with results from another investigation, where the inhi- bition by the same wastewaters was determined by a screening method based on activated sludge. The best correlation was found with the ammonia oxidation assay. The heavy metal content was statistically analysed for relationships with inhibition using a linear model with stepwise linear regression. Although zinc was a signi®cant inhibitor for all three methods, these responded dierently in the second step, which indicate that the bacteria are not inhibited according to the same pattern. The relevance of the three pure culture assays as an inhibition test is discussed and the conclusion can be drawn that they are all needed to give a complete inhibition picture of wastewaters. # 1998 Elsevier Science Ltd. All rights reserved Key wordsÐInhibition assays, pure cultures, nitri®cation, denitri®cation, wastewater treatment, nitrogen removal, industrial wastewater INTRODUCTION In Sweden, requirements for nitrogen reduction in wastewater treatment plants were introduced in the early 1990's in order to prevent eutrophication of coastal waters. Treatment plants employing biologi- cal nitrogen removal encounter problems arising from shock or sustained loads of toxic compounds in the in¯uent. It is of great importance that the nitrogen removal processes are not disturbed. Industrial wastewaters that are fed to the municipal treatment plant contain potential inhibitors of these processes. It is therefore important to have methods for investigating the inhibitory eects of such waters. These methods should be sensitive, relevant and convenient. Assessment of toxicity of complex waters such as industrial wastewaters cannot be made simply from chemical component analysis; there are many constituents present and there is po- tential for a range of synergistic and antagonistic eects (Williamson and Johnson, 1981; Bitton and Dutka, 1986; Koopman and Bitton, 1986). A better approach would be to use methods based on living cells responsible for the speci®c reactions of interest. Pure culture organisms are preferred in order to yield reproducible results (King and Painter, 1986). In this work three dierent assays for measure- ments of inhibition were compared. These methods are based on three microbial reactions in the nitro- gen removal process; ammonia oxidation, nitrite oxidation and nitrite reduction. NH 4 4 Ammonia oxidation NO 2 4 Nitrite oxidation NO 3 4 Nitrate reduction NO 2 4 Nitrite reduction N 2 Pure cultures of nitrifying and denitrifying bac- teria are used, namely Nitrosomonas, Nitrobacter and a denitrifying strain named No. 110 (Gumaelius et al., 1996). These bacteria were iso- lated from activated sludge. The purpose of this work was to compare the assays by screening a large number of industrial wastewaters. The inhi- bition caused by 48 samples of industrial waste- water was determined. The tests were performed at two dierent times. In the ®rst phase of the study, fresh samples were tested with the Nitrobacter culture. In the second phase of the study, samples that had been frozen were tested with Nitrosomonas and No. 110. To investigate the eect of freezing, some tests were repeated with Nitrobacter. Most of the methods for esti- Wat. Res. Vol. 32, No. 10, pp. 2995±3000, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/98 $19.00 + 0.00 PII: S0043-1354(98)00050-5 *Author to whom all correspondence should be addressed. [Tel: +46-8-7907699; Fax: +46-8-245452; E-mail: camilla@biochem.kth.se]. 2995