Oxygen Tolerance of Sulfate-Reducing Bacteria in Activated Sludge

Mar 3, 2004 - Environmental Science & Technology 2014 48 (20), 11777-11786 .... Emilie Lefèvre , Luciana P. Pereyra , Sage R. Hiibel , Elizabeth M...
0 downloads 0 Views 142KB Size
Environ. Sci. Technol. 2004, 38, 2038-2043

Oxygen Tolerance of Sulfate-Reducing Bacteria in Activated Sludge KASPER U. KJELDSEN, CATHERINE JOULIAN,† AND KJELD INGVORSEN* Department of Microbiology, Aarhus University, Building 540, Ny Munkegade, 8000 Aarhus C., Denmark

The oxygen tolerance of sulfate-reducing bacteria (SRB) present in activated sludge was studied in batch incubations using radiolabeled [35S]sulfate and a most probable number (MPN) technique employing activated sludge medium. Sulfate reduction (SR) could not be detected in activated sludge during oxic incubation or in the presence of nitrate. However, upon anoxic incubation of both freshly sampled activated sludge and activated sludge preaerated for 40 min, SR resumed immediately at an initial rate of 2 µM h-1. During long-term aeration of activated sludge, the number of viable and culturable SRB remained constant at around 106 SRB mL-1 throughout a 121 h aeration period. During the first 9 h of the 121 h aeration period, the anaerobic SR activity was unaffected, as compared to that of an unaerated control sample, and recommenced instantaneously upon anoxic incubation. Even after 121 h of continuous aeration, SR took place within 1.5 h after anoxic incubation albeit at a rate less than 20% that of the unaerated control. As suggested by MPN estimates and the observed kinetics of SR, oxygen exposure resulted in temporary metabolic inactivation of SRB but did not cause cell death. Consequently, SRB have the potential for quick proliferation during anoxic storage of activated sludge.

Introduction The presence of sulfate-reducing bacteria (SRB) in periodically oxic environments has been demonstrated in several studies during the last two decades (19, 36, 41, 44). In some cases, sulfate reduction (SR) has actually been detected in supposedly oxic zones of microbial mats (7, 46) and marine sediments (22). However, to date no pure culture is known capable of carrying out dissimilatory SR in the presence of significant oxygen concentrations (>1 µM) (8). Despite traditionally being considered obligate anaerobes, oxygen exposure studies (4, 16, 19, 43) suggest that some SRB possess a relatively high degree of oxygen tolerance. In accordance with this, SRB hold various enzymatic protection mechanisms against oxidative stresssa trait likely to facilitate survival during periods of oxygen exposure. Some SRB produce the disproportionating enzymes catalase (EC 1.11.1.6) and superoxide dismutase (EC 1.15.1.1) for scavenging of the toxic reduced dioxygen species hydrogen peroxide and superoxide (12, 19). Recently two new classes of reductive enzymes, * Corresponding author phone: (+45) 89423245; fax: (+45) 86 12 71 91; e-mail: [email protected]. † Current address: Environnement et Proce ´ de´s, BRGM, 3 avenue Claude Guillemin, BP 6009, 45060 Orle´ans Cedex 2, France. 2038

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 7, 2004

peroxidases (EC 1.11.1.7) and superoxide reductases (EC 1.15.1.2), catalyzing the elimination of hydrogen peroxide and superoxide, respectively, were discovered in phylogenetically diverse SRB (1, 30, 31). Many SRB are furthermore capable of reducing oxygen at high rates either respiratorily (9) or nonrespiratorily (42). Oxygen reduction may result in ATP formation (14, 42) but apparently cannot support growth (8). Hence, a number of studies indicate that some taxons of SRB are well adapted to survive in periodically oxic environments. Activated sludge in wastewater treatment plants (WWTPs) is an example of a periodically oxic environment where SRB have been detected in relatively high numbers, both by cultivation (21, 28) and by molecular methods (32, 45). In activated sludge, SRB are subjected to regular and repeated cycles of oxygenation, and as a result this environment could select for highly oxygen tolerant SRB. The presence of SRB in activated sludge is of great technical interest since SR has several potential negative effects on the operation of WWTPs. Sulfide, the main end product of dissimilatory SR, may cause deflocculation of sludge flocs by reducing and precipitating ferric iron, which is an important flocculating agent in activated sludge (34). Furthermore, sulfide production can stimulate growth of filamentous bacteria and consequently result in concomitant sludge bulking (50, 51). Deflocculation and filamentous bulking are frequent problems in wastewater treatment systems, impeding the settling properties of the activated sludge, thereby reducing effluent water quality. If SRB survive the oxic conditions in the activated sludge systems, they may thrive during the following anoxic storage of settled activated sludge, where sulfide production can lead to deteriorated dewaterability (38). It should be noted that in other technical systems the activity of SRB may be highly advantageous. SRB are gaining increasing importance in wastewater technology, e.g., for treatment of high-strength sulfate waste streams and heavy metal removal (37) and the desulfurization of stack gases (29). A better understanding of the effects of oxygen on the activity and survival of SRB will be important for optimizing these applications. SRB apparently are ubiquitous in activated sludge, yet despite the potential negative effects of sulfide production on the operation of WWTPs, the activities of SRB in activated sludge systems have not been studied extensively. This especially pertains to the influence of oxygenation on SR and survival of SRB. The repeated cycles of aeration generally are thought to suppress the activity and growth of SRB in activated sludge systems primarily due to the conventionally assumed toxicity of oxygen. In addition, the nitrification potential of activated sludge is generally high, and oxygenation can result in transient accumulation of nitrate, which has been shown to suppress SR, for example, in the oil industry (10). The present study was conducted to investigate the oxygen tolerance of SRB present in activated sludge by addressing the effect of aeration on the SR activity and the survival of SRB in the sludge.

Materials and Methods Origin of Sludge. Activated sludge was obtained from a municipal WWTP in Odder (Jutland, Denmark), an OCOtype plant (Degre´mont Puritek A/S) treating domestic wastewater. The OCO activated sludge tank consists of two concentric compartments, the outer of which is repeatedly operated in cycles of oxic (approximately 55 min, e25% air saturation) and anoxic (approximately 30 min) conditions. 10.1021/es034777e CCC: $27.50

 2004 American Chemical Society Published on Web 03/03/2004

The inner compartment of the OCO tank is permanently anoxic. The plant receives a daily input of 4000 m3 of wastewater with a mean chemical oxygen demand (COD) of 550 mg of O2 L-1. The average sludge age in the aeration zone of the OCO tank is 15 days. Samples of activated sludge were collected from the outer compartment of the OCO tank during the anoxic period. The samples were immediately stored at 4 °C under anoxic conditions and used for experiments within e24 h. Short-Term Aeration Experiments. Effect of Preincubatory Aeration on SR. Initial experiments were conducted to investigate the effect of short-term aeration on the ensuing SR activity upon anoxic incubation. A 75 mL sample of activated sludge ([volatile suspended solids (VSS)] ) 3500 mg L-1; [SO42-] ) 800 µM) was transferred into a 120 mL serum bottle and quickly heated to 20 °C in a water bath. The sludge was subsequently aerated with sterile (0.2 µM) prehumidified atmospheric air at 100% atmospheric oxygen saturation for 40 min with vigorous magnetic stirring to ensure proper suspension of the activated sludge. Following aeration, the serum bottle was immediately sealed with a butyl rubber stopper, the headspace gas replaced with sterile (0.2 µm) oxygen-free N2, and the serum bottle finally injected with 350 kBq of carrier-free 35SO42- (Isotope Laboratory, Risø, Denmark). A control bottle containing 75 mL of fresh unaerated activated sludge was sealed under an atmosphere of N2, the temperature adjusted, and the control bottle immediately injected with a similar amount of 35SO42-. Upon tracer addition, the bottles were incubated in the dark at constant agitation, and triplicate sludge samples (0.5 mL) were removed anaerobically at suitable time intervals for determination of SR activity. The withdrawn samples were preserved in a mixture of 2 mL of 20% zinc acetate and 300 µL of 0.45 M zinc sulfide and stored at -20 °C until analysis. Total reduced inorganic sulfur (TRIS) was recovered from the zinc acetate preserved frozen samples using the singlestep chromium reduction method described by Fossing and Jørgensen (15). In the following, TRIS will be referred to as sulfide. SR rates were calculated from the initial linear increase in sulfide concentration. Effect of Oxygen and Nitrate on SR. A custom-made incubation system was used to study whether SR occurred in the presence of nitrate or oxygen in the activated sludge. The incubation system consisted of a 250 mL Bluecap bottle (Schott) with a sampling outlet at the bottom. The Bluecap bottle was sealed with a rubber stopper fitted with a Clarktype oxygen electrode (40) and a macroscale nitrate biosensor (26). The latter had a 90% response time of 60 s, and a lower detection limit of 0.5 µM NO3-. Both the oxygen electrode and the nitrate sensor were connected to strip chart recorders. Activated sludge (250 mL; [VSS] ) 3400 mg L-1; [SO42-] ) 615 µM) was heated to 20 °C in a water bath, transferred to the incubation bottle, and aerated with sterile (0.2 µm) prehumidified atmospheric air through a long hypodermic needle penetrating the rubber stopper. The activated sludge was constantly kept in suspension by magnetic stirring. Two different experiments were performed in the above-described incubation system: (Experiment 1) Activated sludge was initially aerated at 100% atmospheric oxygen saturation for a total of 30 min. After 15 min of aeration, 1200 kBq of carrierfree 35SO42- was added to the sludge. When the aeration was stopped 15 min later, 300 µM nitrate (final concentration) was added. Subsequently the nitrate and tracer amended sludge was incubated for an additional period of 100 min while the headspace of the incubation bottle was continuously flushed with sterile (0.2 µm) N2. (Experiment 2) The oxygen concentration of activated sludge containing 1200 kBq of carrier-free 35SO42- was maintained at 15-25% atmospheric saturation for a period of 1 h.

In experiments 1 and 2, samples (0.5 mL) were taken at 5-10 min intervals for measurement of SR. The withdrawn samples were preserved and analyzed as described above. In experiment 1, the nitrate reduction rate was calculated from the initial linear decrease in nitrate concentration measured using the biosensor. The obtained rate was confirmed in a parallel experiment devoid of 35SO42- where the nitrate concentration was measured by HPLC analysis. Long-Term Aeration Experiments. The effects of longterm aeration on the survival and SR activity of SRB in activated sludge were studied using another custom-made incubation system. Incubation System. The incubation system consisted of a 2 L (nominal capacity) glass bottle (Schott Bluecap) modified with a sampling outlet at the bottom and an oxygen electrode (same type as described above), mounted through the side of the bottle for continuous monitoring of the oxygen concentration. Sterile (0.2 µm) prehumidified atmospheric air was supplied via a diffuser positioned on the side of the bottle, opposite the oxygen electrode. The bottle was placed on a magnetic stirrer to ensure constant and complete suspension of the sludge particles. Incubation Conditions. Activated sludge (1.8 L) was transferred to the aeration bottle and quickly heated to 20 °C in a water bath, before aeration was initiated. The oxygen level in the sludge was maintained at 95-100% atmospheric oxygen saturation during an aeration period of 121 h. Samples were withdrawn after 0, 3, 9, 33, 73, and 121 h of aeration for most probable number (MPN) enumeration of SRB, determination of SR activity, and measurement of sulfate, nitrate, and VSS concentration. Tracer MPN Enumeration of SRB. Tracer MPN enumerations of SRB were conducted in a growth medium prepared from sterilized concentrated activated sludge from the Odder WWTP. The use of a natural medium for MPN enumerations of SRB has previously been described as being superior to that of conventional synthetic media for detection of SRB in several types of environmental samples (6, 21, 48). To minimize the risk of false positives (48), the sludge medium was amended with Na2SO4 to a final concentration of 2.9 mM (in situ sulfate concentration approximately 0.9 mM). Dilution series (10-fold) of activated sludge were made using 1 mL of inoculum which had been homogenized in advance under anoxic conditions by repeated passage through a 23 gauge needle as described by Thiele et al. (47). After inoculation, 220 kBq of carrier-free 35SO42- was aseptically added to each tube by syringe. The presence of SRB in a tube was scored by analyzing the percentage [35S]sulfide produced (i.e., the activity of the [35S]sulfides relative to the total 35S activity in a sample) using a threshold [35S]sulfide percentage of 0.2 for positive tubes (48). All MPN series were made in triplicate and incubated upside down for 2 months in the dark at 20 °C. Samples (0.5 mL) removed from the MPN tubes for analysis of [35S]sulfide percentage were handled and preserved as described above. MPN values and confidence limits were calculated using statistical MPN tables published by The American Public Health Association (3). Determination of SR Activity in Aerated Sludge Samples. To reduce possible biases from differences in substrate availability due to progressing aerobic carbon mineralization during the 121 h aeration period, all samples (2 mL) taken for determination of SR activity were injected into MPN tubes containing 8 mL of sterile unaerated sludge, 100 kBq of carrier-free 35SO42-, and 905 µM sulfate (unamended). Tubes were incubated in the dark at 20 °C, and 0.5 mL samples were withdrawn anaerobically at appropriate time intervals for analysis of SR activity. Some time points were sampled in triplicate to assess the standard deviation of the measurements. Samples were preserved and analyzed as described above. A sample taken after 121 h of aeration was pasteurized VOL. 38, NO. 7, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2039

FIGURE 1. Time course of sulfate reduction during anoxic incubation of untreated activated sludge (4) and activated sludge aerated for 40 min prior to incubation (O). Lines show linear curve fits. Error bars represent standard deviations (n ) 3). (80 °C for 15 min) prior to transfer to sterile sludge media containing 35SO42- and used as a negative control. SR rates were calculated as described above. Measurement of the Oxygen Consumption Rate. The oxygen consumption rate of the aerated activated sludge was determined from the initial linear decrease in oxygen concentration after the aeration was temporarily stopped (