Evaluation of biocide efficacy by microcalorimetric determination of microbial activity in biofilms

Journal of Microbiological Methods 33 (1998) 227–235

Journal of Microbiological Methods

Evaluation of biocide efficacy by microcalorimetric determination of microbial activity in biofilms ` Henry von Rege, Wolfgang Sand* ¨ Hamburg, Institut f ur ¨ Allgemeine Botanik und Botanischer Garten, Abteilung Mikrobiologie Ohnhorststraße 18, Universitat D-22609 Hamburg, Germany Received 10 December 1997; received in revised form 26 May 1998; accepted 27 May 1998

Abstract The microcalorimetric measurement of microbial activity of biofilm samples allows easy testing of the efficacy of biocides. This has been demonstrated in experiments with biofilm samples consisting of sulfate reducing (SRB) and chemoorganotrophic (COT) bacteria formed in batch culture on mild steel coupons. Additionally, biofilms were produced in continuous culture on the surface of a flow-through gold tubing in a measuring cylinder of the calorimeter. The biofilm samples were treated with the biocides formaldehyde, tetramethylammoniumhydroxide, 1,8-dihydroxyanthraquinone, and a commercial biocide with glutaraldehyde as one of the active compounds at varying concentrations and incubation times. For evaluation of biocide efficacy, microbial activity was monitored and cell counts were determined. All biocides were able to reduce microbial activity, but cell numbers did not decrease correspondingly. Formaldehyde exhibited the best effect. Only 3% of the original microbial activity remained, and a reduction in cell numbers of five orders of magnitude in the case of SRB was measured. In contrast, tetramethylammoniumhydroxide had only a slight effect. Microbial activity was reduced only to 20%, and the cell numbers did not decrease at all. The other biocides exhibited intermediate effects. In general, cell numbers of chemoorganotrophic bacteria in these biofilm samples decreased more than did the numbers of SRB. If the biocide containing medium was substituted by a biocide-free one, regrowth and reactivation of biofilm cells resulted. However, the activity did not reach initial values within the experimental time.  1998 Elsevier Science B.V. All rights reserved. Keywords: Microbial activity; Biocide; Biofilm; Biofouling; Microcalorimetry

1. Introduction Biofilms are ubiquitous and, thus, are found on different types of surfaces like metal, concrete, or plastic (Flemming, 1995; Donlan et al., 1994; Ridgway et al., 1983). Microbial cells are embedded in a matrix of extracellular polymeric substances (EPS) and form a slimy layer on surfaces of materials *Corresponding author. Tel.: 149 40 8228243; fax: 149 40 82282423.

known as biofouling. The undesired effects of biofilms are corrosion of constructional materials, heat loss in heat exchangers, or the decrease of product quality as in the paper industry (Flemming, 1995; Donlan et al., 1994; Ridgway et al., 1983). The most common microorganisms found in biofilms are sulfate reducing bacteria, chemoorganotrophic bacteria like Pseudomonas, and bacteria involved in iron–manganese cycles (in the case of metal surfaces; Geesey, 1991; Little et al., 1992). In a biofilm matrix bacteria are largely protected against toxic

0167-7012 / 98 / $ – see front matter  1998 Elsevier Science B.V. All rights reserved. PII: S0167-7012( 98 )00055-4

228

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

substances like antibiotics or biocides (Foley and Gilbert, 1996). Strategies to limit the growth of undesired microoganisms are often the addition of biocides or other antimicrobials like chlorine (LeChevallier et al., 1988), mechanical cleaning, or nutrient depletion (Griebe et al., 1996). Of these techniques the most common practice is the application of biocides. However, often suppliers recommend dosages which neither seriously inhibit microbial activity nor reduce the cell numbers of biofilm microorganisms. Since the costs for countermeasures are high and environmental restrictions are increasingly enforced, there is obviously a need to use appropriate methods for evaluation of biocide efficacy. The classical determination of cell numbers, which for specialized microorganisms often needs several weeks, is a costly and time consuming process. In addition, no information becomes available about the activity of microorganisms and their physiological status. To overcome this problem, several methods for the determination of activity in biofilms have been developed, such as the use of fluorescent dyes (Kalmbach et al., 1997; Schaule and Flemming, 1996), determination of RNA-turnover rate (Yu and McFeters, 1994), sulfate reduction rate, or hydrogenase activity in the case of SRB (Maxwell and Hamilton, 1986; Beech et al., 1994). Nevertheless, the results of these techniques are often difficult to interpret (Hamilton et al., 1988) and can be attributed only to the activity of one single group of bacteria in the biofilm community. Another difficulty results from the fact that for most techniques the biofilm must be removed from the sample surface resulting in a total alteration of the bacterial environment and, hence, bacterial metabolism (e.g. diffusion, uptake of nutrients). Yu et al. (1993) demonstrated different results for determinations of cell numbers of living biofilm cells. They were dependent on the use of intact or dispersed biofilm samples. Furthermore, the samples can only be used once. These problems can be overcome by the use of microcalorimetry. Microbial activity may be quantified by the detection of heat output accompanying all biochemical redox reactions. Modern instruments allow heat quantities as small as 10 26 W, e.g. evolved by bacteria, to be recorded. Thus, microcalorimetry can be used for measurements of metabolism of aerobic or anaerobic bacteria (Traore

et al., 1981; von Stockar and Marison, 1989), since both of them produce heat in the course of metabolism. The determination of microbial activity of attached cells in a biofilm can easily be carried out with this technique (Humphrey and Marshall, 1984; Lock and Ford, 1983; Wentzien et al., 1994). For this reason we developed a microcalorimetric test to directly determine microbial activity of biofilm samples with unaltered, intact biofilm samples. A further advantage of this technique is the rapid sample handling because further processing for ` sample preparation is not needed (von Rege and Sand, 1996a). The advantages of this technique for use in biocide evaluation in the case of biofilm problems are demonstrated in these experiments. Different groups of biocides were tested for their efficacy against biofilm samples containing SRB and chemoorganotrophic bacteria on mild steel and gold surfaces in batch or continuous culture.

2. Materials and methods

2.1. Microcalorimetry 2.1.1. Batch experiments A Thermal Activity Monitor (TAM 2277, Thermometric AB, Bromma, Sweden or C3-Analysentechnik, Baldham, Germany) equipped with an ampoule-cylinder (TAM, Nr. 2277-205) was used. The ampoules were made of stainless steel and have a volume of 25 ml. A basic description of the ¨ 1974). microcalorimeter is given elsewhere (Wadso, The calorimetric system is able to quantify exothermic and endothermic processes from biological and chemical reactions. Temperature differences of less than 10 26 8C are detectable. The heat flows from a heat producing sample, e.g. in an ampoule, via Peltier elements to a 25 l thermostatted water bath. The latter surrounds the measuring cylinder and serves as heat sink. The Peltier elements convert this heat flow into a voltage signal. The signal is expressed as mW. Calibration is achieved by calibration heater resistors in contact with the measuring position. A known current is passed through the channel heater resistor to dissipate a specific thermal power. The calibration range for all experiments was 0–1000 mW. The limit of signal detectability is

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

1 mW, the base line stability is 62 mW. The thermostat was set to 308C (62310 24 8C) as operating temperature.

2.1.2. Continuous culture experiments A flow-through measuring cylinder (TAM, Nr. 2277-202) was used. The cylinder consisted of a gold tubing mounted around the measuring cup. The tubings are made of 24 carat gold with an internal diameter of 1 mm. An external solution, with or without bacteria, is pumped via a peristaltic pump (flow-rate: 20 ml / h) through the tube. In the measuring cylinder the solution passes firstly through a heat-exchange coil for temperature equilibration before reaching the measuring position. The heat is again registered by Peltier elements. Calibration range for these experiments was 0–300 mW. The limit of detectability is 0.5 mW, the base line stability is 60.3 mW. The experiments were run at 308C (62310 24 8C).

2.2. Determination of microbial activity 2.2.1. Batch experiments The steel ampoule was sterilized by immersion in 70% ethanol for 15 min prior to any measurement of heat output. To avoid an autooxidation of reduced compounds in biofilm samples in case of anaerobic experiments (e.g. FeS), the following steps were performed under anaerobic conditions in an anaerobic glovebox (Du-Scientific, USA, containing (v / v) 2% CO 2 , 10% H 2 , and 88% N 2 ). For thermal analysis coupons were withdrawn from a sample holder, rinsed with degassed, deionized water, and inserted into the ampoule. The ampoule was filled with 15 ml of fresh, anaerobic, sterile medium. Afterwards, the closed ampoule was allowed to thermostat in the calorimeter at 308C for 75 min prior to measurement of the heat output. The signal for heat output was recorded via computer (software Digitam 4.01, Thermometric, Sweden). The heat output of a sample was determined after 2 h when a stable value had been established. Sterile control coupons produced 0.25 mW/ cm 2 heat output, indicating that only quantitatively negligible abiotic reactions occurred between coupons and the bulk

229

aqueous phase. Therefore, values for biofilm samples were not corrected.

2.2.2. Continuous culture experiments Biofilm formation on gold tubing has been reported for thiobacilli (Wentzien et al., 1994) and nitrifyers (Rudert, 1989). Thus, this system was used to test biocide efficacy against biofilms in continuous culture. 2.3. Biofilm generation 2.3.1. Batch experiments All batch experiments were run with coupons of mild steel (99.76% iron, size 20330 mm, 1 mm thickness), which had been polished with emery grit paper (grit 180, 600) and cleaned with acetone. Six coupons each were fixed on teflon holders which were attached to butyl rubber stoppers (Maagtechnik, Switzerland). These were hung into 1-l bottles, the latter were closed and sterilized together with the coupons for 40 min at 1118C. Afterwards, the bottles were aseptically filled with sterile, anaerobic Postgate C nutrient solution (Postgate, 1984). An enrichment culture containing non-characterized SRB and facultatively anaerobic, chemoorganotrophic bacteria (COT) was used for inoculation and for biofilm formation with the coupons. The culture was obtained from a water sample of the river Elbe (Assel, Germany) by consecutively incubating it under anaerobic conditions in Postgate E-medium. The enrichment produced a strong biofilm on metal ` surfaces (von Rege and Sand, 1996a). Each bottle was inoculated with 20 ml of a 3-day-old preculture and gassed for 1 h with nitrogen to remove oxygen. The coupons were transferred every week into fresh, sterile, anaerobic medium. The bottles were incubated at 288C for 10 weeks to obtain a mature biofilm. For comparison with the activity measurements, cell numbers were determined by the MPNtechnique. For this purpose, the biofilm was removed by scraping with a scalpel and vortexed in 0.9% NaCl solution. For quantification of the SRB, tubes containing Postgate E-medium were used (Postgate, 1984). COT were enumerated by Standard I-agar plate counts (Merck, 7818). Incubation was done in a glovebox (Du-Scientific, USA) with an atmosphere containing 2% CO 2 , 10% H 2 , and 88% N 2 (v / v).

230

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

2.3.2. Continuous culture experiments Either the enrichment culture (containing SRB and COT) or a pure culture of Vibrio natriegens (DSM 759) were pumped through the gold tubing to establish a biofilm. The cultures were externally incubated at 308C. For anaerobic experiments 1.0 g / l ascorbic acid was added to the sterile nutrient solution (Postgate C-medium), which was additionally sparged with nitrogen gas. Vibrio natriegens was cultivated in an aerated, sterile medium containing 0.5 g / l peptone, 0.1 g / l meat extract, 0.2 g / l yeast extract, 1.21 g / l tris(hydroxymethyl)-aminoethane, and 25.0 g / l NaCl at pH 7.5. The respective culture solutions were continuously pumped (20 ml / h) through the gold tubing of the flow-through cylinder. Biofilm formation was traced by replacing the bacterial culture solution by a sterile one. This occurred at the end of the logarithmic growth-phase of the various test cultures (5 h for Vibrio natriegens, 48 h for the enrichment). If a heat output remained, it could be exclusively attributed to attached cells on the surface of the gold tubing in the measuring cylinder. 2.4. Biocide application 2.4.1. Batch experiments The biocides Dilurit 946 (BK Ladenburg, Germany), tetramethylammoniumhydroxide (Sigma), 1,8-dihydroxyanthraquinone (Aldrich), and a 37% formaldehyde-solution (Merck) were used. Dilurit (Dilu) and tetramethylammoniumhydroxide (TM) were diluted with deionized water and sterile filtered. Formaldehyde (FA) was diluted with deionized water, 1,8-dihydroxyanthraquinone (DHA) was dissolved in acetone according to Cooling et al. (1996). The latter two compounds were freshly prepared and not sterilized. Dilurit is a commercial biocide recommended for use against slime forming bacteria. One of the active compounds is pentandial (5glutaraldehyde). The biocide 1,8-dihydroxyanthraquinone shall be active specifically against SRB by uncoupling ATP-synthesis from electron transfer reactions and, thus, reduce sulfide production (Cooling et al., 1996). Up to now it has only been used against planktonic cells in a concentration range of 3–5 mg / l. Tetramethylammoniumhydroxide belongs to the group of quater-

nary ammonium compounds, which are claimed to be active against biofilm problems (Paulus, 1996). Formaldehyde is a well-known antimicrobial compound which disturbs cell activity by the formation of methyl bridges. The biocides were added to assays containing coupons with a 10-week-old biofilm. Before and after biocide addition three coupons each were withdrawn for evaluation of biocide efficacy by determination of cell numbers and microbial activity. Dilurit was tested in three concentrations (25, 100, 500 mg / l) and two contact times (24 h, 7 days). To test whether living cells remained in the biofilm after biocide addition, coupons were transferred to fresh, biocide-free medium after rinsing with deionized water and consecutively incubated. After 1 week of incubation they were measured again for microbial activity and cell numbers. Further experiments were run for 24 h with addition of the biocides Dilurit and TM at 500 mg / l, DHA at 4.8 mg / l, and FA at 50 000 mg / l concentration. The latter was used as a control because of the high concentration. In case of FA, the coupons were directly incubated in 20 ml of the solution under anaerobic conditions.

2.4.2. Continuous culture experiments Biocides were added to the continuous culture assays after the start of the experiment at about 46 h in case of Vibrio natriegens and at 120 h in case of the enrichment. At these times a constant heat output had indicated that a mature biofilm had developed. Dilurit was tested in three concentrations (25, 100, 500 mg / l). Control experiments aiming at the detection of surviving cells in the biofilm were run by replacing the biocide-containing by biocide-free nutrient solution followed by continuous heat-output recording.

2.5. Statistical analyses The values of three parallel coupons each are given as arithmetric mean values. A possible correlation between two parameters was calculated using Spearman’s rank correlation. The significance was tested using the student’s t-test.

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

3. Results

3.1. Batch experiments for biocide evaluation The mild steel coupons exhibited after an incubation time of 10 weeks a stable, visible biofilm with cell numbers amounting to log 7 / cm 2 for SRB and COT each (Fig. 1). Microbial activity amounted to 58 mW/ cm 2 coupon. For sterile coupons the physicochemical heat output was determined to be 0.25 mW/ cm 2 and can thus be neglected. Incubating coupons with biofilm in a 5%-formaldehyde solution to thoroughly kill the biofilm cells, decreased the activity to 11% of the initial value. This result indicates, that 90% of the heat output originated from biological processes. These data are in agreement with those from earlier experiments with ` and Sand, 1996b). In anaerobic bacteria (von Rege

231

order to optimize dosage and incubation time for an evaluation of the efficacy, the biocide Dilurit 946 was chosen. It was added to the nutrient solution to produce a final concentration of 25, 100 or 500 mg / l. Biofilm coupons were incubated for 24 h or 7 days (Fig. 1). Dilurit in a concentration of 25 or 100 mg / l and a contact time of 24 h decreased microbial activity only by about 30%. The cell numbers remained stable in case of the COT or even slightly increased in case of the SRB. Thus, at these conditions the biocide remained without a serious effect on the biofilm population. Only an incubation with 500 mg / l biocide for 24 h or 7 d caused an intermediate and strong reduction of microbial activity (down to less than 8 mW/ cm 2 ). Interestingly, the contact time of 24 h did not cause a reduction of the cell numbers. Obviously, the contact time was too short. Only a contact time of 7 days caused a

Fig. 1. Influence of biocide (Dilurit) concentration and contact time on microbial activity and cell numbers of biofilm samples on mild steel coupons containing sulfate reducing bacteria (SRB) and chemoorganotrophic bacteria (COT). Microbial activity was measured as heat output in batch experiments. Error bars for each mean data point are included. N5cell count.

232

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

significant reduction of the SRB and COT cell numbers (by two orders of magnitude). Planktonic cells were considerably more sensitive than biofilm cells to biocide application. Generally, their cell numbers decreased more than those of their biofilm counterparts (not shown). If the biofilm coupons were transferred after biocide application into fresh, biocide-free medium and consecutively incubated for 1 week, it could be seen that the cell counts of both groups increased again to the initial level (log 7.6 / cm 2 for SRB and log 6.4 / cm 2 for COT). The activity did not fully reach the original value within one week. It increased only to 13 mW/ cm 2 , about 25% of the initial (not shown). Obviously, the combination of the two techniques allowed us to detect the insufficient effect of low dosages of (this) biocide and a too short incubation period, the survival of a part of the biofilm microorganisms, and the long-lasting inhibitory effect of the biocidal compounds (probably because of precipitation in the biofilm matrix) on microbial metabolism and, hence, activity. In a second set of experiments three other biocides were tested (Table 1). Again mild steel coupons with a 10-week-old biofilm (containing SRB and COT) were incubated with biocide. The test conditions were 24 h contact time each with Dilurit or TM at 500 mg / l, DHA at 4.8 mg / l, and FA at 50 000 mg / l concentration. In all assays microbial activity became considerably reduced. For FA only 3% of the initial activity remained measurable. For the first

three biocides, however, as noted in the previous experiments for the short contact time of 24 h, the cell numbers for SRB remained at the same level, while those for COT were slightly reduced. In the case of FA the cell numbers for both groups were strongly reduced. The results indicate that biofilm SRB seem to be less sensitive against biocides than biofilm COT. In contrast, planktonic cells of SRB and COT exhibited the same sensitivity. Although both groups were almost equally present in the biofilm samples, the SRB contributed only slightly to heat output. This may be deduced from the existence of a significant correlation (based on regression analysis) found between the cell numbers of COT and the amount of heat output, whereas no correlation existed for the cell numbers of SRB and the heat output (N536, a 51%).

3.2. Continuous culture experiments for biocide evaluation A stable biofilm was obtained on the inner surface of the gold tubing of the flow-through cylinder with Vibrio natriegens after 46 h and for the enrichment culture after 120 h. The heat output amounted to 100 mW with Vibrio (Fig. 2), in the case of the enrichment to 270 mW (Fig. 3). Experiments were performed with the Vibrio-biofilm and Dilurit additions at 25, 100, and 500 mg / l concentrations. The lowest biocide concentration resulted only in a slight and transient decrease in microbial activity (Fig. 2).

Table 1 Influence of four different biocides on planktonic cells and on cell numbers and microbial activity of biofilm samples on mild steel a

biofilm SRB (log N / cm 2 ) biofilm COT (log N / cm 2 ) Planktonic SRB (log N / ml) Planktonic COT (log N / ml) Microbial activity (mW/ cm 2 ) a

Control without biocide

Dilu 500 mg / l

TM 500 mg / l

FA 50 000 mg / l

DHA 4.8 mg / l

6.42 60.50 7.85 60.71 7.17

6.85 60.74 6.85 60.68 5.28

8.15 60.51 7.29 60.15 7.97

1.28 61.14 4.77 60.36 0

5.97 60.20 6.57 60.33 7.68

7.60

4.93

6.77

0

6.63

65 615.8

6 62.3

12 60.3

2 62.2

8 61.2

Batch experiments lasted for 24 h at 308C N5cell count; SRB5sulfate reducing bacteria; COT5chemoorganotrophic bacteria; Dilu5Dilurit; DHA51,8-dihydroxyanthraquinone; 65standard deviation (not determined for planktonic cells); TM5tetramethylammoniumhydroxide; FA5formaldehyde.

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

233

Fig. 2. Influence of biocide concentration (Dilurit) on a pure culture biofilm with Vibrio natriegens in continuous culture (flow-rate 20 ml / h) at 308C. P5heat output.

With 100 mg / l Dilurit, microbial activity decreased somewhat but remained during the whole experiment at around 70 mW. At the highest concentration microbial activity rapidly decreased to 3 mW after 2 h and remained at that level until the end of the experiment. If in the latter case the biocide-containing medium was replaced by a biocide-free one, metabolic activity was restored to 85 mW after 10 h (not shown). This finding is in agreement with data from the batch experiments. Obviously, several

bacteria survive the biocide action (probably those which are deeply embedded in the biofilm and thus are protected) and start to regrow. The biofilm of the enrichment culture was tested in the same way. In these experiments the biocide concentrations 25 and 100 mg / l displayed no influence on microbial activity (not shown). Only at a concentration of 500 mg / l Dilurit microbial activity (after a short transient increase) was strongly reduced to 3 mW after 23 h (Fig. 3). After nutrient solution

Fig. 3. Influence of biocide concentration (Dilurit) on a mixed culture biofilm with SRB and COT in continuous culture (flow-rate 20 ml / h) at 308C. P5heat output.

234

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege

exchange, biofilm regrowth also occurred. A heat output of 110 mW was noted after 18 h of consecutively culturing.

4. Discussion Many techniques are in use for testing biocide efficacy. However, plate counts and MPN-techniques remain the most widely used ones because of their low cost and the unequivocal results concerning the killing effect of the biocidal agents. The results presented here indicate that microcalorimetry has considerable advantages over those techniques. Although the latter is primarily more expensive than the classical methods, it may replace the former, because it allows the rapid measurement, even online, of an inhibition of microbial activity by biocidal action. Furthermore, in combination with the classical techniques it becomes possible to differentiate between killing or inhibition as a result of a biocidal action. Even a residual effect, due to adsorbed biocide to the biofilm matrix, becomes detectable. Thus, the time span until a further biocide application becomes necessary, can be determined. This may be of considerable importance for industries like the paper industry, where biocide actions can only reduce but not totally remove microbial biofilmbound activity. Another consequence can also be deduced from the results: it concerns the development of biocides, biocidal compounds, or formulations of these. Because of the possibility of testing batch-wise or in continuous culture, almost all conditions, which influence biocidal efficacy, can rapidly be simulated. This includes pH, redox, oxygen content (up to anaerobiosis), pure and mixed cultures, different materials etc. Of special importance is the possibility of also detecting anaerobic metabolic processes. No other technique is known which unequivocally allows this. ATP-measurements are only related to the energy charge of the cells, which at best may be used as an indirect indicator of metabolic state, besides the fact that it gives only single values and does not allow on-line measurements. The use of CTC is disturbed by reduced compounds being produced under anaerobic conditions (Stewart et al., 1994). The application of an oxygen electrode is for obvious

reasons not feasible. Besides, almost all techniques are destructive and do not allow a further use of the test samples. Thus, regrowth and similar phenomena may hardly be detected. The use of microscopic techniques is restricted by biofilm thickness and chemistry, as well as by the existence of non-penetrable microcolonies and precipitates (Stewart et al., 1994; Yu and McFeters, 1994), an argument which holds also for the classical cultivation techniques. Summarizing, microcalorimetry allows by rapid, if necessary on-line tests to screen biocides for their efficacy under a manifold of conditions. It also becomes possible to test various materials for an optimal resistance against biofouling and MIC. Because of the direct response of the instrument to any heat evolution, the experimental time becomes considerably reduced. In times where economical considerations together with environmental responsibility are of utmost importance, this novel technique will surely find its place in the biocide business.

Acknowledgements This study is based in part on the doctoral-thesis ` of Henry von Rege in the faculty of Biology, ¨¨ University of Hamburg. We thank Dr. Ute Hootmann for the Dilurit sample. This work was supported by grants of AIF / BMWi (No. 9524 and 10653 N / 1 to W.S.).

References Beech, I.B., Cheung, C.W.S., Chan, C.S.P., Hill, M.A., Franco, R., Lino, A.-R., 1994. Study of parameters implicated in the biodeterioration of mild steel in the presence of different species of sulphate-reducing bacteria. Int. Biodeter. Biodeg. 34, 289–303. Cooling III, F.B., Maloney, C.L., Nagel, E., Tabinowski, J., Odom, J.M., 1996. Inhibition of sulfate respiration by 1,8dihydroxyanthraquinone and other anthraquinone derivates. Appl. Environ. Microbiol. 62, 2999–3004. Donlan, R.M., Pipes, T.L., Yohe, T.L., 1994. Biofilm formation on cast iron substrata in water distribution systems. Wat. Res. 28, 1497–1503. Flemming, H.-C., 1995. Biofouling and biocorrosion-effects of undesired biofilms. Chemie Ingenieur Technik 67, 1425–1430. Foley, I., Gilbert, P., 1996. Antibiotic resistance of biofilms. Biofouling 10, 331–346.

` , W. Sand / Journal of Microbiological Methods 33 (1998) 227 – 235 H. von Rege Geesey, G., 1991. What is biocorrosion? In: Flemming, H.-C., Geesey, G. (Eds.), Biofouling and Biocorrosion in Industrial Water Systems. Springer-Verlag, Berlin, pp. 155–164. Griebe, T., Schaule, G., Flemming, H.-C., 1996. A biocide-free anti-fouling Strategy. In: Kreysa, G., Sands, W. (Eds.), Biodeterioration and Biodegradation. DECHEMA Monographs, vol. 133, VCH, Weinheim, Germany, pp. 435–458. Hamilton, W.A., Moosavi, A.N., Pirrie, R.N., 1988. Mechanism of anaerobic microbial corrosion in the marine environment. In: Sequeira, C.A.C., Tiller, A.K. (Eds.), Microbial Corrosion-1. Elsevier, New York, pp. 13–19. Humphrey, B.A., Marshall, K.C., 1984. The triggering effect of surfaces and surfactants on heat output, oxygen consumption and size reduction of a starving marine Vibrio. Arch. Microbiol. 140, 166–170. Kalmbach, S., Manz, W., Szewzyck, U., 1997. Dynamics of biofilm formation in drinking water: phylogenetic affiliation and metabolic potential of single cells assessed by formazan reduction and in situ hybridization. FEMS Microbiol. Ecol. 22, 265–279. LeChevallier, M.W., Cawthon, C.D., Lee, R.G., 1988. Inactivation of biofilm bacteria. Appl. Environ. Microbiol. 54, 2492–2499. Little, B.P., Wagner, P., Mansfield, F., 1992. An overview of microbiologically influenced corrosion. Electrochim. Acta 37, 2185–2194. Lock, M.A., Ford, T.E., 1983. Inexpensive flow microcalorimeter for measuring heat production of attached and sedimentary aquatic microorganisms. Appl. Environ. Microbiol. 46, 463– 467. Maxwell, S., Hamilton, W.A., 1986. Modified radiorespirometric assay for determining the sulfate reduction rate acitivity of biofilms on metal surfaces. J. Microbiol. Meth. 5, 83–91. Paulus, W., 1996. Biocides-modes of action. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Microbially Influenced Corrosion. Springer-Verlag, Berlin, pp. 105–120. Postgate, J.R., 1984. Cultivation and growth, Chapter 3. In: Postgate, J.R. (Ed.), The sulphate-reducing bacteria. Cambridge University Press, pp. 30–50. Ridgway, H.F., Kelly, A., Justice, C., Olson, B.H., 1983. Microbial fouling of reverseosmosis membranes used in advanced wastewater treatment technology: chemical, bacteriological, and

235

ultrastructural analyses. Appl. Environ. Microbiol. 45, 1066– 1084. Rudert, M., 1989. Der Einfluß des Sauerstoffs auf die Nitritoxidation von Nitrobacter. PhD-thesis, University of Hamburg. Schaule, G., Flemming, H.-C., 1996. Quantification of respiratory active bacteria in water and biofilms with a fluorescent redox dye. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Microbially Influenced Corrosion. Springer-Verlag, Berlin, pp. 159–165. Stewart, P.S., Griebe, T., Srinivasan, R., Chen, C.-I., Yu, F.P., deBeer, D., McFeters, G.A., 1994. Comparison of respiratory activity and culturability during monochloramine disinfection of binary population biofilms. Appl. Environ. Microbiol. 60, 1690–1692. Traore, A.S., Hatchikian, C.E., Le Gali, J., Belaich, J.-P., 1981. Microcalorimetric studies of the growth of sulfate-reducing bacteria: Energetics of Desulfovibrio vulgaris growth. J. Bacteriol. 145, 191–199. von Stockar, U., Marison, W., 1989. The use of calorimetry in biotechnology, In: Fichter, A. (Ed.), Advances in Biochemical Engineering / Biotechnology, vol. 40. Springer-Verlag, Berlin, pp. 94–136. ` von Rege, H., Sand, W., 1996a. Simulation of metal-MIC for evaluation of countermeasures. Mater. Corrosion 47, 486–494. ` von Rege, H., Sand, W., 1996b. Influence of oxygenation on the strength of MIC on metals. In: Kreysa, G., Sand, W. (Eds.), Biodeterioration and Biodegradation, DECHEMA Monographs, vol. 133. VCH Weinheim, Germany, pp. 325–331. ¨ I., 1974. A microcalorimeter for biological analysis. LKB Wadso, J. 21, 18–21. Wentzien, S., Sand, W., Albertsen, A., Steudel, R., 1994. Thiosulfate and tetrathionate degradation as well as biofilm generation by Thiobacillus intermedius and Thiobacillus versutus studied by microcalorimetry, HPLC and ion-pair chromatography. Arch. Microbiol. 161, 116–125. Yu, F.P., McFeters, G.A., 1994. Physiological responses of bacteria in biofilms to disinfection. Appl. Environ. Microbiol. 60, 2462–2466. Yu, F.P., Pyle, B.H., McFeters, G.A., 1993. A direct viable count method for the enumeration of attached bacteria and assessment of biofilm disinfection. J. Microbiol. Meth. 17, 167–180.