Aerobic Biological Treatment of

Synthetic Municipal Wastewater in

Membrane-Coupled Bioreactors
Christian G. Klatt, Timothy M. LaPara
University of Minnesota, Department of Civil Engineering, 500 Pillsbury
Drive SE, Minneapolis, Minnesota 55455; telephone: 612-624-6028; fax:
612-626-7750; e-mail: lapar001@umn.edu
Received 27 July 2002; accepted 1 October 2002
DOI: 10.1002/bit.10572

Abstract: Membrane-coupled bioreactors (MBRs) offer

many benefits compared to conventional biological
wastewater treatment systems; however, their performance characteristics are poorly understood. Laboratory-scale MBRs were used to study bacterial adaptations in physiology and community structure. MBRs
were fed a mixture of starch, gelatin, and polyoxyethylene-sorbitan monooleate to simulate the polysaccharide,
protein, and lipid components of municipal wastewater.
Physiological adaptations were detected by measuring
ectoenzyme activity while structural dynamics were
studied by denaturing gradient gel electrophoresis of
PCR-amplified 16S rRNA gene fragments. As cell biomass accumulated in the MBRs, pollutant removal efficiency initially improved and then stabilized with respect
to effluent concentrations of chemical oxygen demand,
protein, and carbohydrate. Comparison of the MBR effluent to filtered reactor fluid indicated that a portion of the
observed pollutant removal was due to filtration by the
membrane rather than microbial activity. The rates of
ectoenzyme-mediated polysaccharide (-glucosidase)
and protein (leucine aminopeptidase) hydrolysis became
relatively constant once pollutant removal efficiency stabilized. However, the maximum rate of lipid hydrolysis
(heptanoate esterase) concomitantly increased more
than 10-fold. Similarly, -glucosidase and leucine aminopeptidase ectoenzyme affinities were relatively constant,
while the heptanoate esterase affinity increased more
than 30-fold. Community analysis revealed that a substantial community shift occurred within the first 7 days
of operation. A Flavobacterium-like bacterial population
dominated the community (>50% of total band intensity)
and continued to do so for the remainder of the experiment. 2003 Wiley Periodicals, Inc. Biotechnol Bioeng 82: 313
320, 2003.

Keywords: DGGE; ectoenzymes; enzyme affinity; PCR

INTRODUCTION
The biodegradation of organic compounds found in industrial and municipal wastewaters is necessary to protect surface water quality. Aerobic biological techniques, such as
the activated sludge process, have proven effective for the
Correspondence to: T. M. LaPara

2003 Wiley Periodicals, Inc.

removal of these pollutants. Activated sludge treatment consists of a well-mixed aeration basin coupled to a quiescent
chamber that separates the biomass from the treated effluent
so that the biomass can either be recycled back to the aeration chamber or be removed from the system altogether as
a waste byproduct. Treatment efficiency by the activated
sludge process depends on the development of microbial
aggregates that are susceptible to gravitational separation.
Problems with poor settling biomass, however, are common
(Jenkins et al., 1993), such that biomass clarification efficiency often limits activated sludge treatment efficiency
(Dick, 1970).
Membrane-coupled bioreactors (MBRs) are an alternative biotechnology for wastewater treatment. MBRs operate
in a similar fashion as the activated sludge process, except
that an ultra- or microfiltration membrane replaces the
gravitational clarifier to separate the biomass from the
treated effluent. MBRs have the potential to achieve complete biosolids separation and cell recycle, thereby eliminating the need for the formation of flocs or aggregates,
while simultaneously increasing the treatment capacity of
the system (iek et al., 1998; Konopka et al., 1996; LaPara
et al., 2001a; Muller et al., 1995). MBRs can therefore treat
considerably greater amounts of wastewater per reactor volume compared to conventional batch, fed-batch, or continuous-flow bioreactors.
Despite these potentials advantages, the implementation
of MBRs for wastewater treatment has been infrequent due
to high membrane costs and large energy inputs for membrane operation (Visvanathan et al., 2000). A considerable
amount of research to date, therefore, has focused on membrane operationspecifically, on understanding membrane
fouling to reduce MBR operational costs (Bouhabila et al.,
2001; Defrance et al., 2000; Kang et al., 2002; Yoon et al.,
1999). In contrast, relatively little research on MBRs has
focused on the microbiological aspects of MBR operation,
with results thus far indicating that while pollutant removal
efficiency is very high, the biomass is less reactive to

changes in reactor conditions compared conventional treatment bioreactors (Konopka et al., 1996, 1998; Konopka,
2000; Tappe et al., 1999).
The purpose of this study was to elucidate the relationships between microbial community structure and function
in an aerobic MBR treating a synthetic municipal wastewater. This synthetic feed media contained starch, gelatin, and
polyoxyethylene-sorbitan monooleate to simulate the polysaccharide, protein, and lipid components of municipal
wastewater (Raunkjr et al., 1994). Reactor performance
was investigated by measuring the removal of COD, protein, and carbohydrate as biomass accumulated in the MBR.
Ectoenzyme activity (the combination of extracellular and
surface attached enzyme activity) was quantified using fluorogenic substrate analogs. Bacterial community dynamics
were investigated by denaturing gradient gel electrophoresis
of PCR-amplified 16S rRNA gene fragments (PCRDGGE).
MATERIALS AND METHODS
Membrane-Coupled Bioreactors
Laboratory-scale MBRs were inoculated with 1 ml of biomass from a 100-ml batch culture that had been fed synthetic wastewater medium and incubated overnight at 25C.
This batch culture had been inoculated with 1 ml of cryopreserved activated sludge collected from the aeration tanks
of the Metropolitan Wastewater Treatment Plant (St. Paul,
MN). MBRs consisted of a 580-ml fermentor (CYTOLIFT
glass airlift bioreactor; Kontes, Vineland, NJ) coupled to
0.2-m pore size polysulfone microfilter membrane cartridge (surface area 0.011 m 2 ; A/G Technology,
Needham, MA) (Fig. 1). Culture fluid was rapidly (residence time <30 sec) withdrawn from the bioreactor,
pumped through the membrane cartridge, and returned to
the bioreactor. Membrane permeate was pumped from the
filter cartridge at a specific rate to maintain a liquid volume
(700 50 ml) in the bioreactor. Sterile feed medium was
added using a peristaltic pump (Masterflex variable-speed
console drive pump; Cole-Parmer, Vernon Hills, IL). The
reactor hydraulic residence time was approximately 8.5 h.
Aeration rates were delivered in the range of 0.51.0 L
min1 to keep the culture sufficiently mixed and aerated.
Dissolved oxygen concentrations were continuously maintained >80% of saturation (data not shown).

Figure 1.

314

Schematic of the MBR system used in this study.

The feed medium was designed to represent low-strength

municipal wastewater and contained the following (per liter
of deionized water): 150 mg gelatin, 70 mg starch, 120 mg
polyoxyethylene-sorbitan monooleate 10 mg yeast extract,
10 mg casamino acids, 150 mg ammonium chloride, 100 mg
sodium bicarbonate, 25 mg sodium phosphate, 30 mg potassium phosphate, 40 mg magnesium chloride, 60 mg calcium chloride, and 0.1 mL SL7 trace mineral solution (Biebl
and Pfennig, 1981). This medium had a chemical oxygen
demand (COD) of 400 mg l1.
Analytical Methods
Biomass was measured as particulate protein using the
Lowry et al. (1951) method and bovine serum albumin
(BSA) as a protein standard. COD analysis was determined
colorimetrically using low-range accu-Test vials (Bioscience, Bethlehem, PA) and potassium hydrogen phthalate as
a standard. Quantification of soluble protein was determined using the Hartree (1971) modification of the Lowry
method using BSA as a protein standard. Soluble carbohydrate was determined using the anthrone method and glucose as a carbohydrate standard (Herbert et al., 1971). All
assays were performed in triplicate. Data are presented as
the arithmetic means; the standard deviations for all samples
were <5% of the mean (data not shown).
Enzymatic activity for the hydrolysis of organic biopolymers was quantified using fluorogenic model substrates
involving 4-methylumbelliferone (MUF) or 7-amino-4methylcoumarin (AMC) as substrate analogs (Hoppe,
1993). The model substrates used were MUF-glucopyranoside (-glucosidase), MUF-heptanoate (heptanoate esterase), and L-leucine-AMC (leucine aminopeptidase). Fluorescence output was measured on a TD-700 fluorometer (Turner Designs, Sunnyvale, CA) using a long UV
filter. Aliquots of culture media were diluted with 10 mM
Tris-HCl (pH 7.4) and mixed with a fluorogenic substrate
analog. The amount of substrate analog was controlled to
measure either enzyme activity under saturation conditions
(i.e., S >> Km) or to produce a reaction velocity vs. substrate
profile. Enzymatic activity of each sample was determined
from a linear regression of fluorescence units over time and
correlated to a MUF or AMC standard curve. One unit of
enzyme activity corresponded to the production of 1 mol
of product per minute.
Community Analysis
Biomass samples were collected from the MBR and centrifuged; the cell pellet was resuspended in 1 ml of lysis buffer
(120 mM sodium phosphate buffer, 5% sodium dodecyl
sulfate, pH 8.0). Cells were lysed during a 90-min incubation at 70C, followed by three consecutive freeze-thaw
cycles. Genomic DNA was purified from these samples
using the Fast DNA Spin Kit (Qbiogene, Vista, CA) per the
manufacturers instructions.
Partial 16S rRNA genes were amplified from the ex-

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 3, MAY 5, 2003

tracted genomic DNA by PCR using a PTC 100 thermal

cycler (MJ Research, Watertown, MA) as described previously (LaPara et al., 2000a). The variable V3 region of the
16S rRNA gene from members of the domain Bacteria was
amplified using the PRBA338F (E. coli positions 338358)
(Lane, 1991) and PRUN518R (E. coli positions 534518)
(Muyzer et al., 1993) primers with a GC-clamp attached to
the reverse primer (Muyzer et al., 1993). The final 50-l
reaction mixture contained: 1 PCR buffer with MgCl2
(Promega, Madison, WI), 4 nmol deoxynucleoside triphosphates, 25 pmol of forward and reverse primers, and 1 ng
of template DNA. The PCR protocol included a 5-min initial denaturation at 94C, 30 cycles of 92C for 30 sec, 55C
for 30 sec, and 72C for 30 sec, followed by 7 min at 72C.
DGGE was performed on a D-Code apparatus (BioRad,
Hercules, CA). Samples containing approximately equal
amounts of PCR amplicons were loaded onto 8% (wt/vol)
polyacrylamide gels (37.5:1, acrylamide:bisacrylamide) in
0.5 TAE buffer (Sambrook et al., 1989) using a denaturing
gradient ranging from 3055% denaturant (100% denaturant contains 7 M urea, 40% (vol/vol) formamide in 0.5
TAE). Electrophoresis was performed at 60C, initially at
20 V (15 min) and then at 200 V (270 min). Following
electrophoresis, the gel was stained with SYBR Green I
(Molecular Probes, Eugene, OR; diluted 1:5,000 in 0.5
TAE). Gels were visualized on a UV transilluminator and
photographed with a digital CCD camera (BioChemi System; UVP, Upland, CA). Photographs were enhanced for
contrast and brightness using Adobe PhotoShop v 6.0. Band
intensities were quantified using LabWorks Image Acquisition software (UVP).
Specific PCR-DGGE bands were manually excised from
the gel, suspended in 30 l of sterile water, and incubated
overnight at room temperature. PCR-DGGE was repeated
on these samples until only a single band was detectable. A
final PCR step was performed without the GC-clamp attached to the reverse primer. PCR products were purified
using the Wizard PCR preps DNA purification system (Promega, Madison, WI) prior to nucleotide sequence determination. All nucleotide sequences were determined at the
Advanced Genetic Analysis Center at the University of
Minnesota using an ABI 3100 Genetic Analyzer (Applied
Biosystems, Foster City, CA). Nucleotide sequences were
determined fully in both directions for each PCR-DGGE
band using PRBA338F and PRUN518R as sequencing
primers. Reported nucleotide sequences do not include the
original PCR primer sequence. Reference nucleotide sequences were obtained from the GenBank database. The
nucleotide sequences obtained in this study have been deposited in the GenBank database under accession nos.
AF529281-AF529285.
Data Analysis
Michaelis-Menten constants were calculated from the enzyme activity assays by nonlinear regression using SigmaPlot regression software (SPSS, Chicago, IL) to the following equation:

V = Vm

S
Km + S

(1)

where V is the enzymatic reaction velocity, S is the substrate

analog concentration, Km is the half-saturation constant, and
Vm is the reaction velocity when S >> Km.
Enzyme affinity is used as a measure of the ability of an
enzyme to catalyze a reaction at low substrate concentration
and is calculated by the following equation:
Affinity =

Vm
Km

(2)

Nucleotide sequences were compared with sequences in the

GenBank database (Benson et al., 2002) with the BLASTn
program (Altschul et al., 1997). Nucleotide sequences
were checked for possible chimeric sequences with the
CHECK_CHIMERA program at the Ribosomal Database
Project website (Maidak et al., 2001). Putative chimeric
sequences were also manually split into subsections and
resubmitted to the GenBank database to determine if the
segments were from different phylogenetic groups.
RESULTS
Two different MBRs were operated with virtually 100% cell
recycle; some biomass, however, was lost from each MBR
due to sample collection and accidental cell loss caused by
foaming. In the first experiment (Fig. 2A), biomass levels
were approximately constant at 100200 mg l1 particulate
protein for the first 20 days. Biomass density then increased
linearly (27 mg l1 d1; r2 0.98) over the next 30 days.
COD levels in the membrane permeate decreased initially
and then remained reasonably constant at 130150 mg l1.
Soluble protein and carbohydrate levels in the membrane
permeate were less than 10 mg l1 and 2 mg l1 throughout
this experiment. In the second MBR experiment (Fig. 2B),
biomass levels increased more than 10-fold in a quasilinear
fashion (38 mg l1 d1; r2 0.89) throughout the 35-day
experiment. COD levels in the membrane permeate reached
a maximum of 235 mg l1 on the fourth day, but then
declined so that over the next 30 days it was approximately
130 mg l1. Soluble protein (1030 mg l1) and carbohydrate (<8 mg l1) concentrations were again low throughout
the duration of the experiment.
The gross pollutant removals reported above are the sum
of both biological (i.e., conversion of nutrients to new biomass, CO2, and other microbial products) and physical (i.e.,
containment by the membrane) removal. The relationship
between these two removal mechanisms was investigated in
the second MBR experiment by comparing the protein and
carbohydrate concentrations in the membrane permeate vs.
filtered (0.2 m pore size) reactor contents (Fig. 3). Filtered
carbohydrate concentrations were initially similar to that of
the membrane permeate, but then increased 4-fold and remained relatively constant throughout the duration of the
experiment. Filtered protein concentrations were 35-fold

KLATT AND LAPARA: TREATMENT OF SYNTHETIC MUNICIPAL WASTEWATER

315

Figure 2. The accumulation of biomass with simultaneous removal of

COD, protein, and carbohydrate during the first (A) and second (B) MBR
experiments. cell protein, effluent COD, effluent protein,
effluent carbohydrate.

higher than membrane permeate levels after the tenth day of

the experiment.
Metabolic activity of the MBR biomass was studied in
both experiments by measuring ectoenzyme activity at satu-

Figure 3. Comparison between carbohydrate (A) and protein (B) concentrations in the membrane permeate and in filtered reactor fluid samples
during the second MBR experiment. membrane permeate,
filtered reactor fluid.

316

rating substrate concentrations (i.e., V Vm) related to the

components of the synthetic wastewater media (Fig. 4).
-Glucosidase activity initially decreased more than 90% in
the first experiment, after which it remained reasonably constant at 12 U (g protein)1; in the second MBR experiment, -glucosidase activity was maintained at 12 U (g
protein)1 throughout the experiment. Leucine aminopeptidase activity was 6090 mU (g protein)1 during the first 13
days and then decreased to 2545 mU (g protein)1 for the
rest of the first MBR experiment. Leucine aminopeptidase
activity in the second MBR experiment was initially 17 mU
(g protein)1 and then slowly declined throughout the experiment to less than 2 mU (g protein)1. Heptanoate esterase activity, which was measured only in the second experiment, increased throughout the experiment in a complex
pattern. Activity increased more than 6-fold in the first 4
days, then declined by almost by 50% over the next 7 days,
followed by another increase in activity to more than 10fold higher than the heptanoate esterase activity at the beginning of the second experiment.
In the second MBR experiment, ectoenzyme affinity was
measured to help elucidate the physiological adaptations
made by the biomass (Fig. 5). -Glucosidase and leucine
aminopeptidase affinities were both reasonably constant
throughout the experiment (Fig. 6). In contrast, heptanoate
esterase affinity increased more than 15-fold in the first 4
days of the experiment. Heptanoate esterase affinity then
declined by almost 50% in the next 3 days, followed by a
gradual increase for the rest of the experiment. The final

Figure 4. Maximum ectoenzyme activity in replicate MBRs as a function

of time during the first (A) and second (B) MBR experiments.
-glucosidase, leucine aminopeptidase, heptanoate esterase.

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 3, MAY 5, 2003

Figure 7. Community profile of bacterial community structure as a function of time during the second MBR experiment as detected by denaturing
gradient gel electrophoresis of PCR-amplified 16S rRNA gene fragments.
Numbers shown in each lane represent the time of sample collection (days).
Letters identify prominent bacterial populations.

Figure 5. Heptanoate esterase activity as a function of time and of substrate analog concentration during the second MBR experiment. Symbols
represent measured enzyme activities; solid lines represent the best fit to
the Michaelis-Menten model; the slope of the broken line shows enzyme
affinity.

heptanoate esterase affinity was more than 30-fold higher

than the initial value.
Bacterial community structure during the second MBR
experiment was studied by denaturing gradient gel electrophoresis of PCR-amplified 16S rRNA gene fragments
(PCR-DGGE) (Fig. 7). A substantial community shift occurred within the first 4 days of the experiment, as all of the
bacterial populations disappeared with one exception (Band
A). Band A became the dominant bacterial population
(>55% of total band intensity) for the rest of the experiment
(Fig. 8). Two additional bands were detected (Bands C and
D) after 7 days, of which only one (Band C) remained
detectable for the rest of the experiment. An additional band
(Band B) also became detectable during the last 2 days of
the experiment.

Figure 6. Affinity of different ectoenzymes as a function of time during

the second MBR experiment. -glucosidase, leucine aminopeptidase, heptanoate esterase.

Several of the prominent PCR-DGGE bands were excised

and their nucleotide sequences were determined (Table I).
Two of the bands were associated with the CytophagaFlavobacteria-Bacteroides division of Bacteria (Bands A
and C), while the other three were from the Proteobacteria
(- and -subdivisions) (Bands E, F, and G). Band excision
and sequence analysis from day 7 and day 35 confirmed that
comigrating bands were the same bacterial population (data
not shown). Band B was identified as a chimeric sequence
and excluded from further analysis (data not shown).
DISCUSSION
This research was performed to better understand the relationships between bacterial activity and bacterial community dynamics in MBRs used for biological wastewater
treatment. MBRs impose increasingly stringent nutrient
limitation by providing a constant nutrient supply while
biomass accumulates. During the present research, a sub-

Figure 8. The fraction of total band intensity comprised by individual

bacterial populations as detected by PCR-DGGE during the second MBR
experiment. Band labels are shown in Figure 7. Band A, Band
B, Band C, Band D, Band E, Band F.

KLATT AND LAPARA: TREATMENT OF SYNTHETIC MUNICIPAL WASTEWATER

317

Table I.

Sequence length and closest phylogenetic affiliation of the PCR-DGGE bands analyzed in this study.
Phylogenetic relationship

PCR-DGGE
band

Sequence length
(bases)

Most closely related sequence

Bacterial
division

%
Similarity

A
C
D
E
F

155
155
163
160
160

Flavobacterium sp. IsoA1 (AJ319015)

Unidentified bacterium WP13 (AF303937)
Delftia acidovorans (AF149849)
Stenotrophomonas detusculanense (AF280434)
Comamonas testosteroni Q10 (AF519533)

CFB
CFB
-Proteobacteria
-Proteobacteria
-Proteobacteria

100
93.5
93.8
99.4
100

GenBank accession numbers are shown in parentheses. CFB Cytophaga-Flavobacteria-Bacteroides.

stantial shift in microbial activity and bacterial community

structure was detected within the first several days of MBR
operation. This physiological and community adaptation
was observed through improved pollutant removal efficiency, modulation of ectoenzyme activity and affinity, and
a substantial shift in the dominant bacterial populations.
These results are consistent with previous observations that
both physiological adaptations and community structure
shifts occur within the first few days of MBR operation
(Konopka et al., 1996, 1998; LaPara et al., 2001a).
In this study, however, further increases in heptanoate
esterase activity and affinity were detected, indicating that
although the performance of an MBR stabilized with respect
to pollutant removal efficiency, the bacterial community
continued to adapt its physiology to optimize bioreactor
performance. The detection of this continuing physiological
adaptation corresponded to a continuous structural shift,
suggesting that there is a connection between bacterial dynamics and bioreactor function. Unfortunately, bacterial
community dynamics in wastewater treatment bioreactors
are poorly understood (Amann et al., 1998).
The proliferation of a single bacterial population (Band
A) corresponded to the initial increase (<7 days) in heptanoate esterase activity and affinity. Although this connection is circumstantial, it is novel and may indicate a connection between bacterial community structure and biological wastewater treatment reactor function. A thorough
knowledge of these relationships between community structure and function that may be gained with further research
would be helpful to understand the performance of MBRs
and other full-scale wastewater treatment bioreactors in
which mixed and dynamic bacterial communities are responsible for the biodegradation of mixtures of substrates.
In previous work, community adaptations were observed in
laboratory bioreactors as functions of temperature (Konopka et al., 1999; LaPara et al., 2000b, 2001b), hydraulic
residence time (LaPara et al., 2000c), and stringency of
nutrient limitation in an MBR (LaPara et al., 2001a), but
were not correlated to any specific operational variable.
After this initial period, two other bacterial populations
(Bands C and D) were detected, suggesting that a combination of a structural shift to specialized bacteria and a
physiological adaptation by the Band A-population were
responsible for the continued increase in both heptanoate

318

esterase activity and affinity. Isolation and physiological

characterization of these bacterial populations would be enlightening; however, isolation of the relevant bacteria from
complex ecosystems is often difficult (Amann et al., 1995).
In the competition for a limited supply of nutrients, bacteria can gain an advantage by increasing the quantity of
enzyme per cell (i.e., increase Vm) (Matin, 1979) or by
producing an enzyme that is more efficient at low substrate
concentrations (i.e., improved enzyme affinity) (Button,
1994). In the present experiments, the increases in heptanoate esterase activity and affinity in the second MBR experiment were substantially different from previous work where
only moderate (1025%) increases in normalized Vm rates
were observed (LaPara et al., 2001a). While the modifications in heptanoate esterase activity and affinity reflect a
physiological adaptation to the increasingly stringent nutrient limitation imposed by the MBR, the question remains
why similar trends were not observed with the -glucosidase and leucine aminopeptidase measurements. Here
again, bacterial growth on individual compounds within a
mixture of substrates is poorly understood (KovrovaKovar and Egli, 1998). We are currently experimenting with
MBRs fed a single substrate to elucidate the relationships
between ectoenzyme activity, nutrient uptake, and bacterial
community dynamics.
The behavior of the MBR with respect to pollutant removal is consistent with previous results for MBR operation
(iek et al., 1998; Konopka et al., 1996; LaPara et al.,
2001a). Effluent COD levels typically improve initially and
then reach a pseudo-steady state. One of the goals of this
research was to simulate the treatment of municipal wastewater that contains protein, polysaccharide, and lipid as the
principal organic constituents (Raunkjr et al., 1994). Our
overall COD removal performance, however, was quite low
(70%) compared to full-scale municipal wastewater treatment bioreactors (>90%) (Tchobanoglous and Burton,
1991). Further reductions in COD concentrations were not
detected when the membrane permeate was collected, reinoculated with activated sludge, and incubated at 25C for 30
days (data not shown). This suggests that virtually all of the
COD measured in the membrane permeate was soluble microbial products (see Barker and Stuckey, 1999, for a review) instead of the relatively biodegradable components of
the feed media (starch, gelatin, and polyoxyethylene-sobitan

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 3, MAY 5, 2003

monooleate). We therefore conclude that the synthetic

wastewater used in the present experiments was helpful to
elucidate microbial activity during the biodegradation of
individual compounds in a mixture of substrates, but poorly
simulated municipal wastewater with respect to its susceptibility for COD removal.
From a practical perspective, the removal of COD by
MBRs can result from both biological (i.e., microbial attack) and physical (i.e., membrane filtration) mechanisms.
Because the goal of MBRs is to promote biological over
physical removal, we used a microfiltration membrane with
a 0.2-m pore size because it could achieve 100% cell
capture but allow large macromolecules to pass. Our results
showed that this occurred initially, but the membrane also
hindered both polysaccharide and protein from leaving the
MBR relatively soon after the initiation of the experiment.
This undoubtedly occurred as a result of membrane fouling
that reduced the effective membrane pore size. We are currently exploring the relationship between membrane pore
size and the retention of large macromolecules within
MBRs.
In conclusion, these experiments have further characterized the relationship between bacterial community structure
and function in membrane-coupled bioreactors. The bacterial community simultaneously adapts its community structure and activity, although the connection between these
two adaptations remains unclear. Further research is needed
to understand bacterial community dynamics in wastewater
treatment bioreactors and to correlate ectoenzyme activity
with individual substrate biodegradation for wastewaters
containing a mixture of compounds.
REFERENCES
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W,
Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation
of protein database search programs. Nucleic Acids Res 25:
33893402.
Amann RI, Ludwig W, Schleifer K-H. 1995. Phylogenetic identification
and in situ detection of individual microbial cells without cultivation.
Microbiol Rev 59:143169.
Amann R, Lemmer H, Wagner M. 1998. Monitoring the community structure of wastewater treatment plants: a comparison of old and new
techniques. FEMS Microbiol Ecol 25:205215.
Barker DJ, Stuckey DC. 1999. A review of soluble microbial products
(SMP) in wastewater treatment systems. Water Res 33:30633082.
Benson, DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Rapp BA, Wheeler
DL. 2002. GenBank. Nucleic Acids Res 30:1720.
Biebl H, Pfennig N. 1981. Isolation of members of the family Rhodospirillacaea. In: Starr MP, Stolp HG, Troper HG, Balows A, Schlegel
HG, editors. The prokaryotes. Berlin: Springer. p 267273.
Bouhabila E, Ben Aim R, Buisson H. 2001. Fouling characterisation in
membrane bioreactors. Separ Purif Technol 22:123132.
Button DK. 1994. The physical base of marine bacterial ecology. Microb
Ecol 28:273285.
iek N, Winnen H, Suidan MT, Wrenn BE, Urbain V, Manem J. 1998.
Effectiveness of the membrane bioreactor in the biodegradation of
high molecular weight compounds. Water Res 32:15531563.
Defrance L, Jaffrin MY, Gupta B, Paullier P, Geaugey V. 2000. Contribution of various constituents of activated sludge to membrane bioreactor fouling. Bioresource Technol 73:105112.

Dick RI. 1970. Role of activated sludge final settling tanks. J Sanit Eng Div
ASCE 96:423426.
Hartree EF. 1972. Determination of protein: a modification of the Lowry
method that gives a linear photometric response. Anal Biochem 48:
422427.
Herbert D, Phipps PJ, Strange RE. 1971. Chemical analysis of microbial
cells. Methods Microbiol 5B:209234.
Hoppe H-G. 1993. Use of fluorogenic model substrates for extracellular
enzyme activity (EEA) measurement of bacteria. In: Kemp PF, Sherr
BF, Sherr EB, Cole JJ, editors. Handbook of methods in aquatic microbial ecology. Ann Arbor, MI: Lewis. p 423432.
Jenkins D, Richards MG, Daigger GT. 1993. Manual on the causes and
control of activated sludge bulking and foaming, 2nd ed. Chelsea, MI:
Lewis.
Kang IJ, Yoon SH, Lee CH. 2002. Comparison of the filtration characteristics of organic and inorganic membranes in a membrane-coupled
anaerobic bioreactor. Water Res 36:18031813.
Konopka A. 2000. Microbial physiological state at low growth rate in
natural and engineered ecosystems. Curr Opin Microbiol 3:244247.
Konopka A, Zakharova T, Oliver L, Camp D, Turco RF. 1996. Biodegradation of organic wastes containing surfactants in a biomass recycle
reactor. Appl Environ Microbiol 62:32923297.
Konopka A, Zakharova T, Oliver L, Paseuth E, Turco RF. 1998. Physiological state of a microbial community in a biomass recycle reactor.
J Ind Microbiol Biotechnol 20:232237.
Konopka A, Zakharova T, LaPara TM. 1999. Bacterial function and community structure in reactors treating biopolymers and surfactants in a
continuous-flow reactor. J Ind Microbiol Biotechnol 18:235240.
Kovrov-Kovar K, Egli T. 1998. Growth kinetic of suspended microbial
cells: From single-substrate-controlled growth to mixed-substrate kinetics. Microbiol Mol Biol Rev 62:646666.
Lane DJ. 1991. 16S/23S rRNA sequencing. In: Stackebrandt E, Goodfellow M, editors. Nucleic acid techniques in bacterial systematics. New
York: John Wiley & Sons. p 115175.
LaPara TM, Nakatsu CH, Pantea L, Alleman JE. 2000a. Phylogenetic
diversity of mesophilic and thermophilic aerobic reactors treating
pharmaceutical wastewater. Appl Environ Microbiol 66:39513959.
LaPara TM, Konopka A, Nakastu CH, Alleman JE. 2000b. Effects of
elevated temperature on bacterial community structure and function in
bioreactors treating a synthetic wastewater. J Ind Microbiol Biotechnol
24:140145.
LaPara TM, Konopka A, Nakastu CH, Alleman JE. 2000c. Thermophilic
aerobic wastewater treatment in continuous-flow bioreactors. J Environ Eng 126:739644.
LaPara TM, Konopka A, Nakastu CH, Alleman JE. 2001a. Thermophilic
aerobic treatment of a synthetic wastewater in a membrane-coupled
bioreactor. J Ind Microbiol Biotechnol 26:203209.
LaPara TM, Nakatsu CH, Pantea L, Alleman JE. 2001b. Aerobic biological
treatment of a pharmaceutical wastewater: effect of temperature on
COD removal and bacterial community development. Water Res 35:
44174425.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193:265275.
Maidak BL, Cole JR, Lilburn TG, Parker CT, Saxman PR, Farris RJ,
Garrity GM, Olsen GJ, Schmidt TM, Tiedje JM. 2001. The RDP-II
(Ribosomal Database Project). Nucleic Acids Res 29:173174.
Matin A. 1979. Microbial regulatory mechanisms at low nutrient concentrations as studied in chemostat. In: Shilo M, editor. Strategics of
Microbial Life in Extreme Environments. Weinheim: Chemie. p
323339.
Muller EB, Stouthamer AH, van Verseveld HW, Eikelboom DH. 1995.
Aerobic domestic wastewater treatment in a pilot plant with complete
sludge retention by cross-flow filtration. Water Res 29:11791189.
Muyzer G, Dewaal EC, Uitterlinden AG. 1993. Profiling of complex mi-

KLATT AND LAPARA: TREATMENT OF SYNTHETIC MUNICIPAL WASTEWATER

319

crobial populations by denaturing gradient gel electrophoresis analysis

of polymerase chain reaction-amplified genes coding for 16S rRNA.
Appl Environ Microbiol 59:695700.
Raunkjr K, Hvitved-Jacobsen T, Nielsen PH. 1994. Measurement of
pools of protein, carbohydrate and lipid in domestic wastewater. Water
Res 28:251262.
Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory
manual, 2nd ed. Cold Spring, NY: Cold Spring Harbor Laboratory
Press.
Tappe W, Laverman A, Bohland M, Braster M, Rittershaus S, Groeneweg
J, van Verseveld HW. 1999. Maintenance energy demand and starva-

320

tion recovery dynamics of Nitrosomonas europaea and Nitrobacter

winogradskyi cultivated in a retentostat with complete biomass retention. Appl Environ Microbiol 65:24712477.
Tchobanoglous G, Burton FL. 1991. Wastewater engineering: treatment,
disposal, and reuse, 3rd ed. New York: McGraw-Hill.
Visvanathan C, Ben Aim R, Parameshwaran K. 2000. Membrane separation bioreactors for wastewater treatment. Crit Rev Environ Sci Technol 30:148.
Yoon SH, Kang IJ, Lee CH. 1999. Fouling of inorganic membrane and flux
enhancement in membrane-coupled anaerobic bioreactor. Separ Sci
Technol 34:709724.

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 3, MAY 5, 2003