当前位置:首页 >> 环境科学/食品科学 >>

SBR as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms


Appl Microbiol Biotechnol (1998) 50: 589±596

? Springer-Verlag 1998

ORIGINAL PAPER

M. Strous á J. J. Heijnen á J. G. Kuenen M. S. M. Jetten

The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms

Received: 14 May 1998 / Received last revision: 30 July 1998 / Accepted: 31 July 1998

Abstract Currently available microbiological techniques are not designed to deal with very slowly growing microorganisms. The enrichment and study of such organisms demands a novel experimental approach. In the present investigation, the sequencing batch reactor (SBR) was applied and optimized for the enrichment and quantitative study of a very slowly growing microbial community which oxidizes ammonium anaerobically. The SBR was shown to be a powerful experimental set-up with the following strong points: (1) e?cient biomass retention, (2) a homogeneous distribution of substrates, products and biomass aggregates over the reactor, (3) reliable operation for more than 1 year, and (4) stable conditions under substrate-limiting conditions. Together, these points made possible for the ?rst time the determination of several important physiological parameters such as the biomass yield (0.066 ? 0.01 C-mol/mol ammonium), the maximum speci?c ammonium consumption rate (45 ? 5 nmol/mg protein/min) and the maximum speci?c growth rate (0.0027 á hA1, doubling time 11 days). In addition, the persisting stable and strongly selective conditions of the SBR led to a high degree of enrichment (74% of the desired microorganism). This study has demonstrated that the SBR is a powerful tool compared to other techniques used in the past. We suggest that the SBR could be used for the enrichment and quantitative study of a large number of slowly growing microorganisms that are currently out of reach for microbiological research.

Introduction
In the past years molecular ecology has indicated that classical microbiological studies have been devoted to only a small part of the natural diversity (Pace 1997; Raskin et al. 1995, Ward et al.1990). Some of the as-yetuncultivated microorganisms may have escaped discovery because they grow too slowly to be easily cultivated by classical methods. One of these very slowly growing microbial communities is the one responsible for the anaerobic ammonium oxidation (Anammox) process studied in our institute. The doubling time of the Anammox culture has been reported to be 30 days (Van de Graaf et al. 1996). Although this culture grows very slowly, the Anammox process was shown to be a promising new way of removing nitrogen from wastewater (Jetten et al. 1997; Strous et al. 1997a, b). The Anammox culture oxidizes ammonium directly to dinitrogen gas using nitrite as the electron acceptor. The process is autotrophic, using CO2 as the only carbon source. A ?rst successful step in the partial enrichment of the organisms that catalyze this reaction was the use of ?uidized bed cultures fed with mineral media containing ammonium and nitrite only (Van de Graaf et al. 1996). In the ?uidized bed reactor the microbial community grew as bio?lms on sand particles. However, cultivation using a ?uidized bed reactor was not satisfactory, because the operation of laboratory-scale ?uidized beds was di?cult. Sometimes the biomass retention in the ?uidized bed reactor was not su?cient to maintain the Anammox culture. Due to the desired long duration of the cultures (more than 1 year), several cultivation attempts were unsuccessful due to pH control and recycle pump failures. In general, the bio?lm structure (thickness, population composition, and wall growth) was not constant over the reactor and over di?erent reactor runs. This last problem could be explained by a lack of complete bulk mixing in the ?uidized bed reactor. For example, some parts of the reactor

M. Strous á J. J. Heijnen á J. G. Kuenen á M. S. M. Jetten (8) Kluyver Laboratory for Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands Tel.: +31-15-2781193 Fax: +31-15-2782355 e-mail: M. Jetten@STM.TUDelft.NL

590

did not receive substrate continuously and the biomass in these areas was exposed to starvation, which led to a decreased Anammox activity. This made quantitative interpretation of the results very di?cult. Strati?cation (lack of incomplete bulk mixing) might be common in ?uidized bed reactors (Heijnen 1994). Several techniques have been developed to study attached microorganisms (Caldwell 1995; Characklis 1990; Chesbro et al. 1979; Gjaltema et al. 1994; Strous et al. 1997a). These techniques were based on two strategies or combinations thereof: (1) the use of bio?lms to improve the culturability of organisms that naturally occur in bio?lms (Caldwell 1995), and (2) the use of biomass retention to study slowly growing microorganisms under substrate limitation (Chesbro et al. 1979). The sequencing batch reactor (SBR) used in the study described here combines these two strategies. This paper presents evidence that the SBR is a technique that is more speci?cally suited for the long-term enrichment, cultivation and quantitative analysis of a very slowly growing microbial community.

Influent pump

Stirrer

Effluent pump

PC

Gas buffer Fresh medium

{
41 % Ar 1 % CO2 56 % N2 2 % H2O

SBR

Effluent

95 % Ar 5 % CO2

Materials and methods
Inoculum Granules from a 27-l denitrifying ± and anaerobic ammoniumoxidizing ± ?uidized bed reactor (Mulder et al. 1995) were used as the source of the biomass. Mineral medium The composition of the mineral medium was (g á lA1): KHCO3 1.25, NaH2PO4 0.05, CaCl2.2H2O 0.3, MgSO4.7H2O 0.2, FeSO4 0.00625, ethylenediamine tetraacetic acid 0.00625, trace elements solution 1.25 ml á lA1 (Van de Graaf et al. 1996). NaNO2 and (NH4)2SO4 were added as speci?ed in the results section. Operation of the sequencing batch reactors The Anammox SBR (Fig. 1) was maintained in a 15-l (height 0.41 m, diameter 0.22 m) vessel without ba?es, equipped with a water jacket. The vessel was stirred at 80 ? 10 rpm (six-bladed turbine stirrer, diameter one-third of the vessel diameter). The temperature was 32±33 °C. Anaerobiosis was maintained by ?ushing (3±5.5 ml/min) with an Ar/CO2 (95/5%) gas mixture. The oxidation/reduction potential was monitored using a combined redox electrode (Ag/AgCl, Ingold) and Applicon 1020 biocontroller. The CO2 present in the gas was su?cient to bu?er the solution and to keep the pH between 7.0 and 8.0. The SBR was ?lled continuously with fresh medium over an 11.5-h period. After the ?lling period, the stirrer and in?uent supply were stopped and the aggregates were allowed to settle for 15 min. In the remaining 15 min of the total cycle, part of the liquid ± that containing the ?nely dispersed and suspended material ± was purged by an e?uent pump. To prevent entry of air and loss of anaerobiosis in the purgeperiod, a gas bu?er, ?lled with Ar/CO2 gas and water, was present. The minimum liquid volume, after the liquid was purged, was 5 l. The maximum volume, at the end of the ?lling period, was between 5.2 l and 13 l, depending on the in?uent ?ow rate. Figure 2 shows a schematic representation of the operation of the SBR. This ?gure ? ? includes NH? 4 , NO2 and NO3 concentration pro?les and the pH and oxidation-reduction potential of the SBR during one cycle.

Fig. 1 Experimental set-up of the sequencing batch reactor (SBR)

NH4+, NO2-, NO3-, pH, hydr. vol.

16 14 12 10 8 6 4 2 1 2 3 4 5 6 7 8 9

200 190 180 170 160 150 10 11 12 ORP (mV)

0 Time (h) Stirrer

Influent pump Effluent pump

Fig. 2 Concentration pro?les of soluble nitrogen compounds in a SBR cycle (mmol/l): ammonium (d), nitrite (n) and nitrate (h), pH (m), oxidation/reduction potential (ORP) versus standard H+/H2 potential (lower solid line) and hydraulic volume (hydr vol.; l) (upper solid line). Error margins fall within the symbols. The shaded parts of the time bars indicate the period of operation of the stirrer and pumps Analytical procedures Nitrogen compounds Nitrate was measured colorimetrically after a reaction with salicylic acid (detection range 0.2±10 mM; Cataldo et al. 1975). Nitrite was measured colorimetrically after a reaction with sulfanilic acid and N-naftyl-ethylenediamine (detection range 2±500 lM; GriessRomijn van Eck 1996). Ammonium was measured colorimetrically after a reaction with phenol (detection range 0.01±0.2 mM) (Fawcett and Scott 1960). Hydroxylamine was measured colorimetrically after a reaction with quinolinol and trichloroacetic acid (detection range 0.03±0.1 mM; Frear and Burrel 1955). Hydrazine was determined colorimetrically after a reaction with p-dimethylaminobenzaldehyde (detection range 2±90 lM; Watt and Chrisp 1952). To

591 correct for the in?uence of other medium components all of these nitrogen compounds were measured using a standard addition protocol. Dinitrogen gas and nitrous oxide were analyzed using gas chromatography (Otte et al. 1996). Nitric oxide and nitric dioxide were analyzed using chemiluminescence (Kester et al. 1994). Carbon compounds Protein was measured with the Biuret method (Stickland 1951). Dry weight was determined by drying the sample at 65 °C for at least 40 h. The ash content was measured after ashing at 700 °C for 1 h. Organic carbon in liquid samples was measured using a Dohrmann DC 190 IC/OC analyzer. Inorganic carbon in liquid samples was determined using gas chromatography after conversion of (bi)carbonate to CO2 by the addition of 4 M sulfuric acid. Carbon dioxide and methane in the gas samples were analyzed using gas chromatography (Otte et al. 1996). Sulfur compounds Reduced sulfur compounds such as thiosulfate and tetrathionate were analyzed colorimetrically after cyanolysis (detection range 0.02±0.5 mM; So ? rbo 1967). Elemental composition of the biomass The carbon, hydrogen, nitrogen and sulfur content of the enriched sludge was measured using a Hewlett Packard 185 B CHNS-analyzer after washing the sludge with a physiological salt solution and drying under vacuum. Oxygen content was calculated as the remainder after subtracting the ash and CHNS content of the samples. Mass ?ow To measure the gas ?ow rate, a glass bowl (diameter 22 cm) ?lled with water was placed on a precision balance (0.0±3.2 kg). A vertical glass cylinder (diameter 5 cm) was placed above the bowl, with the lower inlet 1 cm under the water surface. At the start of the measurement, water from the bowl was drawn into the cylinder and the top of the cylinder was closed. Gas from the SBRs was led into the cylinder via a sintered glass nozzle placed under the cylinder in the bowl, and the gas ?ow rate was calculated from the amount of water that was displaced by the gas ?ow in a ?xed amount of time. The liquid ?ow rate was also measured using a precision balance. Electron microscopy Samples were ?xed in glutaraldehyde, embedded in Spurr resin and stained with osmium tetroxide and ruthenium red (Van de Graaf et al. 1996). Size distribution of SBR aggregates The size distribution of the aggregates from three SBR runs was determined with the aid of image analysis (Tijhuis et al. 1994)

Results
Enrichment The Anammox population was enriched in two SBRs. A 2 l SBR (SBR 2 l) was inoculated at t ? 0 with 0.45 g dry weight of Anammox granules (Mulder et al. 1995).

Initially, the medium (4.5 mM 4.5 mM NO? 2) was supplied with a hydraulic retention time of 2 days (based on a 2-l reactor volume). Between t ? 100 days and t ? 200 days the nitrogen load of SBR 2 l could be increased gradually by increasing the in?uent concen? tration to 30 mM NH? 4 and 30 mM NO2 , and by gradually decreasing the hydraulic retention time to 0.26 days. The anaerobic ammonium-oxidizing population accumulated in the reactor and was enriched, as determined by electron microscopy and the occurrence of a typical peak at 468 nm in the reduced cytochrome spectra (Van de Graaf et al. 1996). Biomass accumulated in the reactor as wall growth and biomass aggregates. Due to the wall growth, the distribution of biomass over the reactor was not homogeneous and representative biomass sampling was impossible. At t ? 200 days a second, larger (15 l) SBR (SBR 15 l) was inoculated with 0.5 g dry weight of aggregates from SBR 2 l. Initially, the medium (5 mM NH? 4 , 5 mM NO? 2 ) was supplied to SBR 15 l with a hydraulic retention time of 27 days (based on a 15-l reactor volume). Between t ? 200 days and t ? 246 days the nitrogen load was increased linearly to a hydraulic retention time ? of 10 days (30 mM NH? 4 and 30 mM NO2 ) with complete consumption of nitrite. Between t ? 246 days and t ? 336 days the nitrogen load to SBR 15 l could be increased exponentially in discrete steps. This exponential increase (under nitrite limitation) was realized in the following way: every week the nitrogen load was doubled. When the reactor did not accumulate nitrite the doubled nitrogen load was maintained. When nitrite had accumulated in the reactor the load was reduced to its original value. After this period the medium (30 mM ? NH? 4 , 30 mM NO2 ) was supplied at a hydraulic retention time of 0.93 days. This nitrogen load (1 kg N/m3/ day) was maintained until t ? 531 days. In the 15 l SBR, wall growth was almost absent and the biomass was present as small aggregates. These were distributed homogeneously over the reactor, enabling representative biomass sampling. The size distribution of the aggregates from SBR 2 l and SBR 15 l (two runs) was estimated using image analysis (Fig. 3). The degree of enrichment of the Anammox community in the aggregates was estimated and compared to the inoculum and to granules from a ?uidized bed reactor such as the one previously used to enrich an Anammox community (Van de Graaf et al. 1996). Two independent parameters were selected to determine this degree of enrichment: (1) the speci?c maximum Anammox activity of the biomass in the reactor, and (2) the speci?c absorption at 468 nm of a cell-free extract prepared from the culture. The speci?c absorption at 468 nm in di?erence (reduced minus oxidized) cytochrome spectra was shown previously to be a measure for the degree of Anammox enrichment (Van de Graaf et al. 1996). Samples were collected from the SBR 15 l between t ? 350 days and t ? 400 days and from a ?uidized bed reactor fed with the same mineral medium, 300±350 days after inoculation. NH? 4,

592
0.25 1.2 20 18 Fraction of total aggregates 0.20 1.0 16 0.15 N-load (kg N/m3/day) 0.8 12 0.6 10 8 0.4 6 0 0.8 1.1 1.4 1.7 1.9 2.2 2.5 2.8 Aggregate diameter (mm) 3.0 3.7 0.2 2 0 200 0 250 300 Time (days) 350 4 14 Biomass present (g protein)

0.10

0.05

Fig. 3 Aggregate size distribution in di?erent SBR reactors as determined by image analysis. Black bars SBR 2 l (average aggregate diameter 1.7 mm). Grey and white bars two runs of SBR 15 l (average aggregate diameter 1.5 and 1.2 mm)

The results are presented in Table 1. Both the speci?c activity and the speci?c absorption at 468 nm were higher for the aggregates from the SBR, indicating that the degree of enrichment achieved in the SBR was higher. Study of the stoichiometry and kinetics Figure 4 shows the exponential increase in nitrogen load and the resulting accumulation of Anammox aggregates between t ? 246 days and t ? 531 days. It appears that the biomass speci?c activity in the reactor was constant during these 90 days (20 ? 6 nmol NH? 4 /mg protein/min). This was shown by a good correlation between the nitrogen load and the biomass growth. Because the biomass accumulation was proportial to the amount of biomass present in the reactor, the biomass accumulation could only be described by exponential growth. The biomass present in the e?uent of the SBR was collected over this period and it appeared that only approximately 10% of the growing biomass was washed out and 90% remained in the reactor. Figure 4 also
Table 1 Assessment of the degree of enrichment of anaerobic ammonium-oxidizing microorganisms in Anammox granules and aggregates from di?erent sources Sludge source

Fig. 4 Accumulation of Anammox sludge in a SBR: Biomass present (g protein s), nitrate produced (kg N/m3/day h), nitrogen load (kg N/m3/day solid line). The dashed line represents the amount of biomass that was predicted to be present with the stoichiometry that followed from Table 2, and a biomass retention e?ciency of 90%

shows the nitrate produced in the reactor. Nitrate is probably produced from nitrite to generate reducing equivalents for CO2 ?xation (Van de Graaf et al. 1996) and is therefore a measure for biomass growth. Between t ? 336 days and t ? 531 days more than ten mass balances over the SBR were compiled. In?uent and e?uent liquids and gases were monitored for all important nitrogen compounds (NH? 4 , N2H4, NH2OH, ? N2, N2O, NO, NO? 2 , NO2, NO3 ) and carbon compounds [KHCO? 3 , CO2, biomass (as dry weight and protein)] total soluble organic carbon and CH4. The results are presented in Table 2. Carbon and nitrogen were completely recovered and the degree of reduction balance was closed within 5%. With these mass-balances, the conversion of nitrogen and carbon compounds could be calculated (Table 3). The protein content of the biomass was 0.6 ? 0.1 g protein/g biomass. The elemental composition of the
Absorption of di?erence spectra at 468 nm (absorption units/mg protein) 0 0.075 ? 0.05a 0.086 ? 0.05a

Maximum activity of the biomass in the reactor (nmol NH? 4 /mg protein/min) 6b 26c 45 ? 5

Inoculum Fluidized bed reactor Sequencing batch reactor
a

The error was mainly systematic, the speci?c absorption of the bio?lms from the SBR was always 10± 20% higher Mulder et al. (1995) c Van de Graaf et al. (1996)
b

593 Table 2 Carbon, nitrogen and the degree of reduction balances over the SBR between t = 336 days and t = 531 days. Values are given as the (mean ? SD) (IC Inorganic carbon, TOC total soluble organic carbon, DW dry weight as volatile solids) C (C-mmol/h) In IC (9.2 ? 0.2) TOC (0.20 ? 0.05) DW (0.10 ? 0.05) IC (8.0 ? 0.2) TOC (0.20 ? 0.05) DW (1.0 ? 0.1) 1.01 ? 0.03 N (N-mmol/h) c (e-mmol/h)a

NH? 4 ?23X1 ? 0X7? NO? 2 ?22X4 ? 0X7? DW (0.015 ? 0.008) NH? 4 ?4X6 ? 0X6? NO? 3 ?4X7 ? 0X3? N2 (36.2 ? 1.4) DW (0.15 ? 0.015) 1.00 ? 0.04

NO? 2 ??134 ? 4? DW (0.46 ? 0.04) NO? 3 ??37X6 ? 2X4? N2 (A109 ? 4) DW (4.6 ? 0.4) 0.94 ? 0.06

Out

Accumulation In Out + accumulation
a

c = degree of reduction (oxidation state). In the degree of reduction balance, H2O, H+, NH? 4 and CO2 were used as the reference (zero)

biomass was determined to be CH2O0.5N0.15S0.05 (molecular weight ? 25.7 g/C-mol). From Table 3 and the elemental composition of the biomass the stoichiometry of the Anammox process (at nitrite limitation, in the presence of 5 mM surplus ammonium and nitrate) was calculated (using data reconciliation; Van der Heijden et al. 1994) to be:
? ? ? 1 NH? 4 ? 1X32 NO2 ? 0X066 HCO3 ? 0X13 H

3 1X02 N2 ? 0X26 NO? 3 ? 0X066 CH2 O0X5 N0X15 ? 2X03 H2 O Using this stoichiometry (obtained from mass balancing between t ? 336 days and t ? 531 days and the experimentally found 90% retention of growing biomass) the biomass growth was modeled (in retrospect) for the exponential growth period shown in Fig. 4. The calculated biomass growth and the experimentally observed biomass growth correlated very well (Fig. 4). Apart from dinitrogen gas, only traces of other gaseous nitrogen compounds (N2O, NO, NO2) were produced. Of the nitrogen load only 0.03±0.06% was converted to nitrous oxide; 0.00025±0.0005% was con-

verted to nitric oxide and less than 0.00005% was converted to nitric dioxide. The free concentrations of N2O and NO in the reactor liquid were 60 ? 6 lM and 7 ? 1 nM, respectively. The speci?c ammonium conversion that was observed during the 90 days of exponential growth was 20 ? 6 nmol NH? 4 /mg protein/min. The maximum ammonium conversion (qmax, in the presence of excess nitrite) was 45 ? 5 nmol NH? 4 /mg protein/min. The di?erence between the maximum and average activities could have two explanations: (1) the nitrogen load was increased in a stepwise manner and, after each increase, all biomass was fully active, but since the amount of biomass increased continuously, less substrate was available per amount of biomass at the end of each step, and (2) the aggregates were nitrite-limited (the bulk nitrite concentration was between 0.0 and 0.1 mM) and nitrite gradients might have been common in the aggregates. Thus, although the bulk reactor conditions were constant and stable, the conditions inside the aggregates might not have been. By combining the maximum activity and the biomass yield, the maximum growth rate was calculated to be 0.0027 ? 0.0005 hA1 (doubling time 11 days).

Table 3 The conversions of carbon and nitrogen compounds by Anammox biomass, calculated from Table 2. The conversions were balanced for element and charge conversion (Van der Heijden et al. 1994) and the resulting stoichiometry was calculated. (n.d. not determined) Compound Conversion (mmol/h) A18.5 A22.4 4.7 18.1 A1.2 1.1 n.d. n.d. ? ? ? ? ? ? 1 0.7 0.3 0.7 0.3 0.15 Balanced conversion (mmol/h) A17.5 A23.1 4.5 17.9 A1.16 1.16 35.5 A2.2 ? 0.5 ? 0.5 ? 0.3 ? 0.5 ? 0.13 ? 0.13 ?1 ? 0.2 Stoichiometry

Discussion
This study has shown that the SBR was a very suitable experimental set-up for the cultivation, enrichment and study of a very slowly growing microbial community. The SBR had the following six strong points: 1. Reliable operation over periods of more than 1 year. 2. In the settling period, the aggregates settled rapidly and e?ciently, leading to a 90% retention of the growing biomass. This means that of every 10 g protein generated in the reactor, only 1 g was washed out. The settling properties of the aggregates may have been improved during the enrichment by selection for well-settling aggregates.

NH? 4 NO? 2 ? NO3 N2 HCO? 3 CH2O0.5N0.15S0.05 H2O H+

A1 A1.32 +0.26 +1.02 A0.066 +0.066 +2.03 A0.13

594

Due to the reliable operation and the e?cient biomass retention, large amounts of enriched Anammox biomass could be produced. The degree of enrichment reported previously in a ?uidized bed reactor was 64% (based on microscopic counts; Van de Graaf et al. 1996). When the speci?c absorption at 468 nm in the cytochrome spectra of SBR aggregates and ?uidized bed granules were compared, the degree of enrichment in the SBR aggregates was 15% higher (74%). This indicated that the selective pressure in the SBR was slightly higher than in the ?uidized bed reactor since the duration of the experiment was the same in both cases (400 days). Apparently, contaminating organisms persisted longer in the ?uidized bed enrichment culture. This might have been caused by the occurrence of a local high shear stress generated by the SBR turbine stirrer blades, leading to a continuous disruption of part of the aggregates and e?cient exchange of organisms between aggregates (retained in the reactor) and suspended organisms (washed out). In a ?uidized bed culture, the shear stress was lower and granules might have been more stable once they were formed. 3. A homogeneous distribution of substrates and aggregates made representative sampling and the performance of experiments under de?ned bulk conditions possible for the ?rst time for all chemical and biological assays. As stated in the results section, the conditions inside the aggregates were not de?ned. The penetration depth of nitrite inside the aggregates was calculated to be 0±0.1 mm [at bulk nitrite concentrations of 0±100 lM, assuming that mass transfer only occurred by di?usion, that the a?nity of the microorganisms for the substrate was high, and that the protein density in the granules was 5% (w/v)]. This means that with a 1-mm e?ective aggregate diameter, 50% of the biomass was active in the SBR, which was con?rmed by the lower Anammox activity observed in the reactor (20 ? 6 nmol NH? 4 /mg protein/ min compared to qmax ? 45 ? 5 nmol NH? 4 /mg protein/min). As explained in the results section, the stepwise increase in the nitrogen load also contributed to the lower reactor activity. Because nothing is known about the substrate a?nity of the microorganisms and mass transfer by convection inside the aggregates, the exact nitrite concentration inside the aggregates remains unknown. 4. Stable conditions (comparable to a steady state in a chemostat) were achieved, enabling for the ?rst time mass-balancing under de?ned conditions at low substrate concentrations. 5. For the SBR experimental set-up no special equipment is required, apart from the usual equipment used for continuous cultivation, making this cultivation technique accessible to many microbiology laboratories. 6. The results obtained with the SBR can be translated directly to a practical application of the Anammox process, since reliable biomass retention would be one of the key factors when the Anammox process is applied on a full-scale level to remove ammonium from the wastewater. Furthermore, a scale-up of SBR would be relatively easy.

Several techniques have been developed for the culture and study of aggregated or slowly growing organisms, such as the microstat (Caldwell 1995), the RotoTorque (Characklis 1990), the retentostat (Chesbro et al. 1979), the ?xed bed reactor (Strous et al. 1997a) and the ?uidized bed reactor (Van de Graaf et al. 1996). In the present study, the SBR was demonstrated to be an interesting new method for the enrichment and study of slowly growing microorganisms. In the microstat the conditions are more de?ned than in the SBR, but this set-up was not intended for enrichment and contaminants are not washed out. In the RotoTorque, the ?xed bed reactor and the ?uidized bed reactor, biomass, substrates and products are not distributed homogeneously and representative experiments are not possible (Gjaltema et al. 1994; Strous et al. 1997a; Van de Graaf et al. 1996). Even if the ?uidized bed reactor could be optimized to improve mixing, biomass retention and reliability would still be less optimal compared to the SBR (as explained in the introduction). In both the microstat and the RotoTorque only very small amounts of biomass can be accumulated, compared to the SBR. The conditions in the retentostat are more de?ned than in the SBR, but operation of the retentostat is not reliable over the required long periods of time because the biomass is retained by a membrane that is easily clogged. Furthermore, the retentostat is not suitable for the study of aggregated micro-organisms. The development of SBR cultivation is essential for the microbiological and applied research of the Anammox process because it generates a steady supply of large amounts of highly enriched Anammox biomass of constant composition. For the ?rst time, the determination of important parameters, such as the a?nity and inhibition constants, and other characteristics of this unusual microbial community has become possible. In the study described here, the stoichiometry of the process was determined via mass balances, and better estimates for the maximum ammonium conversion rate and growth rate became available. The maximum speci?c ammonium consumption (qmax ? 45 ? 5 nmol NH? 4 /mg protein/min) and the maximum growth rate (lmax ? 0.0027 hA1, doubling time 11 days) were much higher than reported previously (qmax ? 26 nmol A1 NH? 4 /mg protein/min, lmax ? 0.001 h ). This could be explained partly by the higher degree of enrichment that occurred in the former. At the same time, in the ?uidized bed culture part of the population might have been inactive due to the occurrence of strati?cation (incomplete bulk mixing of substrates and bio?lm granules; Heijnen 1994) and this may have caused starvation in parts of the ?uidized bed culture. Part of the Anammox population might have been present in a dormant phase. The maximum growth rate of the anaerobic ammonium-oxidizing population in the absence of masstransfer limitations was 0.0027 hA1. Although this is as yet the highest growth rate found for Anammox organisms, it must be stressed that this is equivalent to a

595

doubling time of as much as 11 days for these anaerobic chemolithoautotrophs. Anaerobic ammonium oxidation yields only slightly more energy per mol of ammonium oxidized (DG ? A358 kJ/mol; Van de Graaf et al. 1996) than aerobic chemolithoautotrophic ammonium oxidation (DG ? A315 kJ/mol), but these aerobic organisms are often able to grow ten times faster (Wiesmann 1994). The slow growth of Anammox organisms cannot be explained by an ine?cient energy metabolism since the biomass yield of aerobic and anaerobic ammonium oxidation di?er by only 30% [0.1 C-mol/mol NH? 4 (Wiesmann 1994) compared to 0.07 C-mol/mol NH? 4 for Anammox]. This 30% di?erence can be attributed to the low growth rate of Anammox organisms. The slower an organism grows, the more energy is required for its maintenance. A thermodynamic model that takes maintenance energy into account (Tijhuis et al. 1993) predicts a yield of 0.1 C-mol/mol NH? 4 for aerobic nitri?cation and 0.06 C-mol/mol NH? 4 for the Anammox process. Thus, the low Anammox growth rate can only be related to a low speci?c ammonium consumption rate. The speci?c activity of Anammox is more than seven times lower than the speci?c activity reported for aerobic ammonium oxidation (0.3 lmol NH? 4 /mg protein/min compared to 0.04 lmol NH? 4 /mg protein/min). However, when aerobic nitri?ers are subjected to oxygen limitation, which is common in many natural habitats, the speci?c nitri?cation activity is much lower and is comparable to Anammox (Bock et al. 1995). The low activity of anaerobic ammonium-oxidizing organisms results not necessarily from a K-strategy (a high substrate a?nity as a microbial specialism that is usually associated with low maximum growth rates) as opposed to an r-strategy (high maximum growth rate and low substrate a?nity; Schlegel and Jannasch 1989). Anaerobic ammonium oxidation may simply be a kinetically di?cult metabolic strategy. As pointed out in the Introduction, molecular (mainly 16S-rRNA) microbial ecology has shown that interesting microbial resources are being left unexplored. Furthermore, the low levels of ribosomal RNA frequently encountered in natural samples (Amann et al. 1995) indicate that many of these interesting organisms can be expected to have a low growth rate, such as the anaerobic ammonium-oxidizing population studied in our laboratory. Interestingly, this Anammox community has so far resisted characterization using molecular ecological methods. The sequence of the 16S-ribosomal RNA gene of the dominant microorganism in this culture is still unknown, despite the investment of considerable research e?orts. This study has shown that reliable, convenient techniques such as the SBR can lead to the enrichment and study of organisms that may be out of reach for classical microbiological techniques.
Acknowledgements This research was supported ?nancially by the Foundation for Applied Sciences (STW), the Royal Netherlands

Academy of Arts and Sciences (KNAW), Gist-brocades and DSM companies. We gratefully acknowledge M. Veenhuis, I. Keizer, K. Sjollema (RUG) for electron microscopy, K. Schoonen and J. de Bruijn for careful maintenance of the ?uidized bed reactors, A. Van Uijen for help with mass-balancing, S. Van Hateren and S. Peeters for image analysis, and M. van Loosdrecht and D. Sorokin for critical reading of the manuscript. All experiments conducted comply with the current laws of the Kingdom of The Netherlands.

References
Amann RI, Ludwig W, Schleifer KH (1995) Phylogenetic identi?cation and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59: 143±169 Bock E, Schmidt I, Stuven R, Zart D (1995) Nitrogen loss caused by denitrifying Nitrosomonas cells using ammonium or hydrogen as electron donors and nitrite as electron acceptor. Arch Microbiol 163: 16±20 Caldwell DE (1995) Cultivation and study of bio?lm communities. In: Lappin-Scott HM, Costerton JW (eds) Microbial bio?lms. Cambridge University Press, Cambridge Cataldo DA, Haroon M, Schrader LE, Youngs VL (1975) Rapid colorimetric determination of nitrate in plant tissue by nitration of salicylic acid. Commun Soil Sci Plant Analysis 6: 71±80 Characklis WG (1990) Laboratory bio?lm reactors. In: Characklis WG, Marshall KC (eds) Bio?lms. Wiley, New York Chesbro W, Evans T, Eifert R (1979) Very slow growth of Escherichia coli. J Bac 139: 2625±2638 Fawcett JK, Scott JE (1960) A rapid and precise method for the determination of urea. J Clin Pathol 13: 156±159 Frear DS, Burrel RC (1955) Spectrophotometric method for determining hydroxylamine reductase activity in higher plants. Anal Chem 27:10 1664±1665 Gjaltema A, Arts PAM, Van Loosdrecht MCM, Kuenen JG, Heijnen JJ (1994) Heterogeneity of bio?lms in rotating annular reactors: occurrence, structure and consequence. Biotechnol Bioeng 44: 194±204 Griess-Romijn van Eck (1996) Physiological and chemical tests for drinking water. NEN 1056, IV-2. Nederlands Normalisatie Instituut, Rijswijk, The Netherlands Heijnen JJ (1994) Biological industrial waste-water treatment: minimizing biomass production and maximizing biomass concentration. PhD-thesis, TU Delft Jetten MSM, Horn SJ, Van Loosdrecht MCM (1997) Towards a more sustainable wastewater treatment system. Wat Sci Tech 35: 9 171±180 Kester RA, Wijlhuizen AG, Duyts H, Laanbroek HJ (1994) Chemiluminescence analysis of nitric oxide in small-volume samples by a modi?ed injection method. Biol Fertil Soils 18: 260±262 Mulder A, Van de Graaf AA, Robertson LA, Kuenen JG (1995) Anaerobic ammonium oxidation discovered in a denitrifying ?uidized bed reactor. FEMS Microbiol Ecol 16: 177±183 Otte S, Grobben NG, Robertson LA, Jetten MSM, Kuenen JG (1996) Nitrous oxide production by Alcaligenes faecalis under transient and dynamic aerobic and anaerobic conditions. Appl Environ Microbiol 62: 2421±2426 Pace NR (1997) A molecular view of microbial diversity and the biosphere. Science 276: 734±740 Raskin L, Amann RI, Poulsen LK, Rittmann BE, Stahl DA (1995) Use of ribosomal RNA-based molecular probes for characterization of complex microbial communities in anaerobic bio?lms. Water Sci Technol 31:1 261±272 Schlegel HG, Jannasch HW (1992) Prokaryotes and their habitats. In: Balows A, Tru ? per HG, Dworkin M, Harder W, Schleifer KH (eds) The prokaryotes I, 2nd edn. Springer, New York, pp 75±126 So ? rbo BO (1967) A colorimetric method for the determination of thiosulphate. Biochim Biophys Acta 23: 412±416

596 Stickland HL (1951) The determination of small quantities of bacteria by means of the biuret reaction. J Gen Microbiol 5: 698±703 Strous M, Van Gerven E, Kuenen JG, Jetten MSM (1997a) E?ects of aerobic and micro-aerobic conditions on anaerobic ammonium oxidizing (Anammox) sludge. Appl Environ Microbiol 63: 2446±2448 Strous M, Van Gerven E, Ping Z, Kuenen JG, Jetten MSM (1997b) Ammonium removal from concentrated waste streams with the Anaerobic Ammonium Oxidation (Anammox) process in di?erent reactor con?gurations. Wat Res 31:1 1955± 1962 Tijhuis L, Van Loosdrecht MCM, Heijnen JJ (1993) A thermodynamically based correlation for maintenance Gibbs energy requirements in aerobic and anaerobic chemotrophic growth. Biotechnol Bioeng 42: 509±519 Tijhuis L, Van Benthum WAJ, Van Loosdrecht MCM, Heijnen JJ (1994) Solid retention time in spherical bio?lms in a bio?lm airlift suspension reactor. Biotechnol Bioeng 44: 867±879 Van de Graaf AA, De Bruijn P, Robertson LA, Jetten MSM, Kuenen JG (1996) Autotrophic growth of anaerobic ammonium-oxidizing micro-organisms in a ?uidized bed reactor. Microbiology-UK 142: 2187±2196 Van der Heijden RTJM, Heijnen JJ, Hellinga C, Romein B, Luyben KChAM (1994) Linear constraint relations in biochemical reaction systems: I. Classi?cation of the calculability and balanceability of conversion rates. Biotechnol Bioeng 43: 3±10 Ward DM, Weller R, Bateson MM (1990) 16S-rRNA sequences reveal numerous uncultured micro-organisms. Nature 345: 63±65 Watt GW, Chrisp JD (1952) A spectrophotometric method for the determination of hydrazine. Anal Chem 24: 2006±2008 Wiesmann U (1994) Biological nitrogen removal from wastewater. Adv Biochem Eng Biotechnol 51: 113±153


相关文章:
...a powerful tool for the study of slowly growing anaerobic ....pdf
SBR as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms_环境科学/食品科学_工程科技_专业资料。sharon-anammox,生物脱...
结合厌氧氨氧化与薄膜过滤程序之生物反应槽进行废水生物除氮之....ppt
? 11. Strous M., “The sequencing batch reactor as a powerful tool for the study ofslowly growing anaerobic ammonium-oxidizing microorganisms,” Applied ...
新型生物脱氮工艺研究进展_图文.pdf
张可方等[32] 在 SBR 内, 以模拟生活污水为处理对 C / N 取 3. 3 、 ...as a powerful tool for the study of slowly growing anaerobic ammonium - ...
ANAMMOX强化A_2_O系统耐受低C_N冲击的效能研究_林剑波.pdf
The sequencing batch anaerobic ammonium-oxidizing microorganisms [J].Mocrobiol ...reactor as a powerful tool for the study of slowly growing [2] Van ...
味精废水的治理技术及其发展趋势.pdf
reactor as a powerful tool to study very slowly growing micro organisms[ J...通过后续生物处理( 可采 用 SBR 或厌氧- SBR 工艺) , 可使废水 处理最终达...
氮肥工业废水处理技术探讨(3)_图文.pdf
as a powerful tool for the study of slowly growing anaerobic ammonium - ... Partial SBR nitrification and SHARON process[J] . Process Biochem, 2007 ,...
基于单片机的防酒后驾驶控制装置.pdf
ANAMM OX 菌的自养固定 CO2 过程会导致周边环境呈碱性 , 为保证反应的顺 利...as a powerful tool for the study of slowly growing anaerobic ammonium -...
超滤_反渗透工艺用于化工废水提标改造的处理效果_图文.pdf
厌氧污泥床反应器运行[J].华东理工大学学报:自然 科学版,2005,31(1):99102. G.Thesequencingbatchgrowing powerful toolforthestudyofslowly anaerobicammonium?...
ANAMMOX反应器的启动方法研究进展_图文.pdf
朱静平等。列采用sBR反应器,在142 d内于 较高负荷下成功启动了厌氧氨氧化反应...as a powerful tool for the study of slowly growing anaerobic ammonium-...
ANAMMOX滤池的启动及运行研究.pdf
as a powerful tool for the study of slowly growing anaerobic ammonium ...基于SBR反应器的ANAMMOX... 7页 2下载券 ANAMMOX与反硝化协同脱氮... ...
...a powerful tool for the study of slowly growing anaerobic ....unkown
as a powerful tool for the study of slowly growing anaerobic ammonium-...The SBR was shown to be a powerful experimental set-up with the following...
华南理工大学开题报告_图文.pdf
as a powerful tool for the study of slowly growing anaerobic ammonium ...不断加大 ANAMMOX-UASB 反应器进水中 SBR 出水渗滤液的比例,并最终全部 进渗滤...
华南理工大学开题报告.doc
as a powerful tool for the study of slowly growing anaerobic ammonium ... 通过调试 SBR 部分亚硝化反应器的运行参数, 使得 SBR 出水中基质比例 NO2--...
垃圾渗滤液颗粒污泥驯化试验及主要微生物种群变化.pdf
11d后,SBR出水中COD口去除率 达到80%,NH,一N去除....sequencinggrowing [6]StropsM.Thestudyofslowly ...powerfultoolfor the anaerobic ammonium-oxidizing...
《中国环境科学》获评“2012中国最具国际影响力学术期刊”_论文_....pdf
[12]Strous M.The sequencing batch reactor as a powerful tool f0r the study of slowly growing anaerobic ammonium-oxidizing microorgniasms[Z】.1998:589...
厌氧氨氧化技术及其在废水脱氮领域的应用.pdf
Sequencing barch reactor as a powerful tool for the study of slowly growing anaerobic Ammonium-oxidizing microorganisms[J]. Appl. Microbiol. Biotechnol. ...
The GRASS GIS has proved a powerful tool for the development ....pdf
a powerful tool for the development of complex..... such as land morphology, change slowly in time.... 3. Study area The higher part of Val di ...
推流固定化生物反应器培养ANAMMOX菌_杜兵.pdf
荷兰研究者认为 SBR 是适合 ANAM MOX 菌培养的反应器 , 并且在该反应器中...as a powerful tool for the study of slowly growing anaero bic ammomium- ...
基于单片机的防酒后驾驶控制装置_图文.pdf
powerful tool fortlle studyof slowly growing anaerobicammoniumoxidizing ...Theanaerobic ox ANAMMOX反应对温度的变化比较敏感.目前大部分研究中的 ammonium[...
...a powerful tool for the study of slowly growing anaerobic ....unkown
as a powerful tool for the study of slowly growing anaerobic ammonium-...The SBR was shown to be a powerful experimental set-up with the following...
更多相关标签: