Journal of Applied Microbiology 2001, 90, 337±342
Metabolic response of bio®lm to shear stress in ®xed-®lm
culture
Y. Liu and J.-H. Tay
Environmental Engineering Research Center, School of Civil and Structural Engineering, Nanyang Technological University, Singapore
539/9/00: received 6 September 2000 and accepted 31 October 2000
Y . L I U A N D J . - H . T A Y . 2001.
Aims: In a bio®lm reactor, detachment force resulting from hydraulic shear is a major factor
that determines the formation and structure of steady state bio®lm. The metabolic response of
bio®lm to change in shear stress was therefore investigated.
Methods and Results: A conventional annular reactor made of PVC was used, in which
shearing over the rotating disc surface was strictly de®ned. Results from the steady state aerobic
bio®lm reactor showed that the bio®lm structure (density and thickness) and metabolic
behaviour (growth yield and dehydrogenase activity) were closely related to the shear stress
exerted on the bio®lm. Smooth, dense and stable bio®lm formed at relatively high shear stress.
Higher dehydrogenase activity and lower growth yield were obtained when the shear stress was
raised. Growth yield was inversely correlated with the catabolic activity of bio®lm. The reduced
growth yield, together with the enhanced catabolic activity, suggests that a dissociation of
catabolism from anabolism would occur at high shear stress.
Conclusions: Bio®lms may respond to shear stress by regulating metabolic pathways
associated with the substrate ¯ux ¯owing between catabolism and anabolism. A biological
phenomenon, besides a simple physical effect, is underlying the observed relation between the
shear stress and resulting bio®lm structure.
Signi®cance and Impact of the Study: A hypothesis is proposed that the shear-induced
energy spilling would be associated with a stimulated proton translocation across the cell
membrane, which favours formation of a stronger bio®lm. This research may provide a basis
for experimental data on bio®lm obtained at different shear stresses to be interpreted in relation
to energy.
INTRODUCTION
Bio®lm is a promising biotechnology for wastewater treatment. Intensive research has shown that the morphological
characteristics of bio®lm are important for the stability and
performance of a bio®lm reactor. Detachment force is
considered a dominant factor that strongly in¯uences the
bio®lm structure and performance. In a bio®lm reactor,
detachment force may result from hydraulic shear. Much
research has been dedicated to the effects of the detachment
force on the morphological characteristics and structures of
Correspondence to: Dr Yu Liu, Environmental Engineering Research Center,
School of Civil and Structural Engineering, Nanyang Technological University,
50 Nanyang Avenue, Singapore 639798 (e-mail: [email protected]).
ã 2001 The Society for Applied Microbiology
bio®lm (Rittmann 1982; Chang et al. 1991; Chen et al. 1998;
Kwok et al. 1998). In the environmental engineering
literature, it is generally considered that a higher detachment
force makes the bio®lm denser and thinner. It has been
hypothesized that a physical rather than a biological
phenomenon is responsible for the observed relation
between detachment force and the resulting bio®lm structure (Vieira et al. 1993; Chen et al. 1998).
So far, little attention has been paid to the interrelationship
of the metabolism of ®xed bacteria to detachment force.
There is evidence that suspended cells can respond to
hydraulic shear by altering their growth rate, morphology,
cell size/density and metabolism (Nollert et al. 1991; Meijer
et al. 1993; Moreira et al. 1995; Chen and Huang 2000).
However, a clear correlation between the detachment force
338
Y. LIU AND J.-H. TAY
and the resulting bio®lm metabolism is still lacking for ®xed®lm culture. The performance of a bio®lm reactor is directly
associated with the metabolic activity of ®xed bacteria under
given operational conditions. Similar to suspended culture,
the metabolic network of ®xed bacteria basically includes
interrelated catabolic and anabolic reactions. As one of the
most important aspects, the relation between the detachment
force and the energy metabolism of bio®lm has been little
studied so far. Therefore, an attempt to show the effects of
detachment force on energy metabolism and the structure of
bio®lm is reported, and related mechanisms are discussed.
M A T E R I A LS A N D M E T H O D S
Experimental apparatus
A conventional annular reactor made of PVC was used. It
consists of two coaxial cylinders, a stationary outer cylinder
and a rotating inner cylinder made up of a stack of recti®ed
removable PVC discs. The liquid volume of the reactor is
4 l. This type of reactor has the advantage of providing
constant shearing over the rotating disc surface. The
diameter of the rotating disc is 23 cm. Polystyrene ®lms
(20 ´ 3 cm) were attached to the side surfaces of the discs to
serve as support material for bio®lm development.
Operation of reactor
Bacteria from a steady state, laboratory-scale, aerobicsettling-anaerobic reactor, fed with peptone as sole carbon
source, were used as inoculum. The reactor was started
under identical seeding conditions for each rotation speed of
the discs. After the seeding period, the reactor was switched
to once-through continuous ¯ow at a constant hydraulic
retention time of 2 h. The synthetic substrate, containing
peptone as sole carbon source, KH2PO4, Na2HPO4, NH4Cl,
MgCl2, CaCl2 and FeCl3, was continuously fed into the
reactor by a peristaltic pump. The in¯uent chemical oxygen
demand (COD) concentration was kept at 200 mg l)1,
equivalent to a volumetric loading rate of 2á4 kg COD
m)3 day)1 for the experiments. The shear stress on bio®lm
was controlled by the rotation speed of the disc in the range
of 40±120 rev min)1 and was expressed as the tip velocity of
the rotating disc in terms of meters per second. The
temperature was kept constant at 25 ‹ 1°C, while dissolved
oxygen (DO) was supplied by pre-aeration of the substrate
solution. In order to maintain the DO concentration at above
4á0 mg l)1 in all experimental runs, a DO controller was
employed to function as follows. Once the DO concentration
of the reactor dropped below 4á0 mg l)1, the DO probe
located at the middle of the reactor transmits a signal to the
DO controller to activate the electro-valve controlling the
oxygen supply.
Analytical methods
Bio®lm samples were collected at steady state, indicated by
constant ef¯uent and ®xed biomass concentrations, from the
removable ¯exible PS slides (8 ´ 3 cm). The slides were
located on the middle disc of the reactor. The soluble COD
concentration was analysed using the standard method
(APHA 1995). The dry weight of ®xed biomass was also
analysed by the standard method (APHA 1995). Bio®lm
thickness was determined according to the method proposed
by Trulear and Characklis (1982). The sample slide was
placed on a microscope. The 10´ objective was lowered
micrometre by micrometre until the bio®lm surface was in
focus, and the micro-adjustment dial setting was recorded.
The objective was then lowered further until the inert
plastic surface was in focus. The difference between the two
focus lengths was then compared with a calibration curve,
and the bio®lm thickness was determined. The bio®lm
thickness was then expressed as a mean value evaluated from
at least 10 measurements along the slide for each sample.
Ohashi and Harada (1994) also used this method to
determine bio®lm thickness. The dehydrogenase activity of
the bio®lm is measured by following the reduction of
2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium
chloride (INT) to iodonitrotetrazolium (INT-formazan),
according to the procedure established by Lopez et al.
(1986) and then used by Lertpocasombut (1991). The INTformazan concentration is then determined spectrophotometrically at 490 nm. The speci®c INT-dehydrogenase
activity of the bio®lm is expressed in terms of the optical
density at 490 nm per milligram of dry weight of bio®lm. In
fact, the speci®c INT-dehydrogenase activity closely correlates with electron transport system activity (Trevors 1984).
According to Trulear and Characklis (1982), for steady state
®xed-®lm culture, the observed growth yield (Yobs) is
calculated as the increase in biomass concentration, including both ®xed and detached biomass, divided by the
corresponding decrease of COD concentration.
RESULTS
A series of experiments was conducted at different tip
velocities (Vt) of the rotating disc, ranging between 0á48 and
1á45 m s)1. In the annular reactor, the tip velocity of the
rotating disc acts as a major shear stress on bio®lm
(Rittmann 1982; Trulear and Characklis 1982). The tip
velocity vs steady state bio®lm thickness and density is
presented in Fig. 1. Results indicate that a higher tip
velocity leads to a thinner and denser bio®lm, i.e., a compact
bio®lm structure. Figure 2 shows the effects of Vt on the
observed growth yield (Yobs) and speci®c INT-dehydrogenase activity of the bio®lm. The reduced Yobs and increased
INT-dehydrogenase activity are observed with the increase
ã 2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 90, 337±342
RESPONSE OF BIOFILM TO SHEAR STRESS
Fig. 1 Effect of tip velocity on the thickness and density of bio®lm. (·)
Thickness; (s) density
Fig. 2 Effect of tip velocity on Yobs and speci®c INT-dehydrogenase
activity of bio®lm. (m)Yobs; (n) speci®c INT-dehydrogenase activity
in tip velocity. In a three-phase ¯uidized bed reactor,
Lertpocasombut (1991) also found that the INT-dehydrogenase activity of bio®lm was increased from 3á2 to 7á8
Absorbance (Abs) mg)1 dry weight as the speci®c velocity of
air was shifted from 5á7 m h)1 to 10á2 m h)1. The interrelationship of the growth yield to INT-dehydrogenase
activity is further shown in Fig. 3. The growth yield is
apparently inversely dependent on the INT-dehydrogenase
activity of the bio®lm.
During the aerobic oxidation process, the respiratory
activity of cells can be indirectly determined by measurement of the proton translocation activity, and the INT can
be used as arti®cial proton acceptor to quantify the catabolic
activity of the bio®lm. In order to demonstrate the possible
interrelationship of catabolic activity of the bio®lm to the
resulting bio®lm structure, bio®lm density vs speci®c INTdehydrogenase activity of the bio®lm is shown in Fig. 4. A
clear correlation between the catabolic activity of the bio®lm
339
Fig. 3 Relationship between Yobs and speci®c INT-dehydrogenase
activity of bio®lm
Fig. 4 Relationship between bio®lm density and speci®c INT-dehydrogenase activity of bio®lm
and the bio®lm density can be seen. These results suggest
that a higher catabolic activity of bio®lm would favour
formation of a stronger bio®lm. It appears from Figs 1, 2, 3
and 4 that the shear stress not only in¯uences the bio®lm
structures, but also has a signi®cant effect on the energy
metabolism of the bio®lm.
DISCUSSION
Effect of shear stress on bio®lm structure
Figure 1 shows that a higher shear stress results in a thinner
and denser bio®lm. Similar phenomena have also been
reported in the literature (Chang et al. 1991; Vieira et al.
1993; Ohashi and Harada 1994; Chen et al. 1998). In
addition, bio®lm density was found to correlate very closely
with the self-immobilization strength of ®xed bacteria,
which was dependent on the shear stress exerted on the
ã 2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 90, 337±342
340
Y. LIU AND J.-H. TAY
bio®lm (Ohashi and Harada 1994; Chen et al. 1998). This,
and previous research, suggests that hydrodynamic conditions have a signi®cant in¯uence on the density of bio®lm. It
has been hypothesized in the literature that a physical
phenomenon is responsible for the observed relation
between detachment force and bio®lm structure (Vieira
et al. 1993). However, the results in Figs 2 and 4 exhibit an
interrelationship between bio®lm metabolism and shear
stress.
Response of bio®lm to shear stress
To date, the mechanism by which ®xed bacteria respond to a
shear stress and further produce a compact bio®lm has been
unclear in a metabolic sense. In Fig. 2, a remarkable
reduction in Yobs is observed when the shear stress, in
terms of tip velocity of the rotating disc, increases, while the
high shear stress on the bio®lm leads to a signi®cant increase
in speci®c INT-dehydrogenase activity. It has been shown
that the speci®c INT-dehydrogenase activity is very highly
correlated with the electron transport system that determines catabolic activity of micro-organisms (Trevors 1984;
Lopez et al. 1986). The higher catabolic activity implies that
the more organic carbon should go to carbon dioxide. In
fact, it has been demonstrated that growth yield is a function
of the ratio between carbon channeled into carbon dioxide
by catabolism and that converted to biomass through
anabolism in suspended culture (Liu 1999). Therefore, the
observed variations in growth yield and speci®c INTdehydrogenase activity in Fig. 2 may represent stimulated
catabolic activity compared with anabolism. This, in turn,
implies that a discrepancy between the rate of energy
production by catabolism and the rate of energy utilization
by anabolism would occur at high shear stress. A similar
Yobs variation pattern with detachment force has also been
observed in a steady state bio®lm airlift suspension reactor
(Kwok et al. 1998). These observations suggest that bio®lms
may indeed respond to shear stress by regulating metabolic
pathways associated with the substrate ¯ux ¯owing between
catabolism and anabolism.
Fig. 5 Effect of speci®c velocity of air on Yobs and rO2/rTOC in a
steady state, three-phase, ¯uidized bed reactor (Lertpocasombut 1991).
(m) Yobs; (n) rO2/rTOC
In this case, the speci®c velocity of air can re¯ect shear stress
in the reactor. Similar to Fig. 2, Fig. 5 shows that higher
shear stress results in a lower growth yield and a stimulated
DO consumption rate. In fact, the biochemical reactions
associated with bacterial metabolism result in an approximately linear relationship between oxygen utilization and
carbon dioxide production (Selna and Schroeder 1979). In a
metabolic sense, the rO2/rTOC ratio thus represents carbon
distribution in the metabolic network of bacteria. The
reduced growth ef®ciency, together with the increased rO2/
rTOC ratio, seems to suggest energy uncoupling between
catabolism and anabolism. A direct relationship between
Yobs and the rO2/rTOC ratio is further given in Fig. 6.
Similar to Fig. 3, it is observed again that Yobs is inversely
dependent on the catabolic activity of the bio®lm. As noted
above, oxygen utilization and cell production oppose one
another, and the more oxygen used in carbon dioxide
production, the fewer cells are produced. Results reported
by Trinet et al. (1991) and Rittmann et al. (1992) also
showed that the ratio of carbon dioxide-carbon to bio®lm-
Literature data
Lertpocasombut (1991) studied the kinetics of bio®lm
developed in a three-phase ¯uidized bed reactor coupled
to a quadrupole mass spectrometer, and determined the
growth yield, dissolved oxygen (DO) utilization rate (rO2)
and total organic carbon (TOC) removal rate (rTOC). The
effects of the speci®c velocity of air on the observed growth
yield and rO2/rTOC ratio are shown in Fig. 5 for steady state
bio®lm. In the study of Lertpocasombut (1991), only the
speci®c velocity of air varied in the experiments, while the
carrier concentration and water velocity were kept constant.
Fig. 6 Relationship between Yobs and rO2/rTOC (Lertpocasombut
1991)
ã 2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 90, 337±342
RESPONSE OF BIOFILM TO SHEAR STRESS
carbon increased with increasing shear stress in a steady
state ¯uidized bed reactor. On the other hand, Vanderborght
and Gilliard (1981) observed that up to 89% of the organic
carbon was oxidized to carbon dioxide in the same type of
reactor. This evidence from the literature supports a shear
stress-induced metabolic shift in bio®lm culture.
Shear stress-induced proton translocation: an
hypothesis
In the aerobic oxidation process, ATP is generated by
oxidative phosphorylation in which process electrons are
transported through the electron transport system from an
electron donor (substrate) to a ®nal electron acceptor
(oxygen) (Mitchell 1961; Wolfe 1993). On the other hand,
the respiratory activity of aerobic bacteria couples with
proton translocation activity, and a clear linkage of oxygen
reduction to proton translocation has been shown (Babcock
and Wikstrom 1992). Thus, it appears that the respiratory
activity of aerobic bacteria can be indirectly determined by
the measurement of proton translocation activity, and the
INT can be used as arti®cial proton acceptor to quantify the
catabolic activity of bio®lm. In this case, the magnitude of
the speci®c INT-dehydrogenase and oxygen uptake activities can be used to describe the activity of proton
translocation across the cell membrane, i.e., the INTdehydrogenase and oxygen utilization activities would be
linked to the activity of proton diffusion across the cell
membrane.
As discussed earlier, shear stress-induced energy uncoupling can be seen as the increase in speci®c INTdehydrogenase activity and oxygen uptake rate, as shown
in Figs 2 and 5. In fact, under the energy uncoupling state
of cells, stimulated respiration activity was also observed in
suspended microbe cultures (Mayhew and Stephenson
1998).
Russell and Cook (1995) reported that futile cycles of
protons would operate whenever there was an imbalance
of catabolic and anabolic rates, and would be responsible
for most of the observed energy spilling which led to
lower growth ef®ciency. Further, a linear correlation
between the rate of energy spilling and ¯ux of protons
across the cell membrane was shown (Cook and Russell
1994). Therefore, it appears that the shear stress-induced
INT-dehydrogenase and oxygen uptake activities could
re¯ect facilitated proton diffusion across the cell membrane. It has been reported that acidi®cation is a common
response to the cells subjected to high shear stress in
suspended cell cultures (Meijer et al. 1993; Chen and
Huang 2000). In fact, Chen and Huang (2000) found that
high shear stress could induce proton elicitation in a
suspended Stizolobium hassjoo culture.
341
Effect of shear-induced proton translocation on
bio®lm structure
It has been reported that proton translocation would have the
following consequences in relation to surface characteristics
of cells. (i) High proton ¯ux across the cell membrane leading
to enhanced hydrophobic interaction of cells (Tay et al.
2000). (ii) Proton conductance across the cell membrane
could induce surface dehydration (Babcock and Wikstrom
1992; Deamer and Akeson 1994; Tay et al. 2000). (iii) The
proton translocating activity could induce the protonation
and membrane fusion of the cell surface (Liu 1989; Babcock
and Wikstrom 1992). (iv) The biovolume of cells would be
condensed by up to 50% when catabolic activity was
stimulated (Zubay 1984; Shen and Wang 1991).
In general, the surface of micro-organisms is negatively
charged under normal pH conditions. Obviously, increased
proton translocation activity would decrease the net negative
charge of the cell surface, and further promote cell±cell
interaction. As noted above, one of the most important
consequences of enhanced proton translocation is an
increase in hydrophobic characteristics of the cell surface.
In fact, there is strong evidence that the hydrophobicity of
the cell surface is an important af®nity force in the selfimmobilization and attachment of cells (Marshall and
Gruickshank 1973; Pringle and Fletcher 1983; van Loosdrecht et al. 1987). It had been reported that bio®lm density is
proportionally related to the self-immobilization strength of
®xed bacteria, which, in turn, is determined by the shear
stress on the bio®lm surface (Ohashi and Harada 1994; Chen
et al. 1998). Recent research on microbe granulation has
shown that the proton translocation-induced dehydration of
the cell surface could facilitate and strengthen the cell±cell
interaction and further, result in a high density of the
microbial community (Tay et al. 2000; Teo et al. 2000). The
proton translocation-strengthened structure of the microbe
community had also been observed in suspended cultures.
Low et al. (2000) reported that in media supplemented with
an organic protonophore, para-nitrophenol (pNP), cell
agglomerates formed dense ¯ocs compared with pNP-free
cultures. This suggests that shear stress-induced proton
translocation could facilitate and strengthen the cell±cell
interaction and further, result in the high density of
adhering cells, as observed in Fig. 4.
It appears from the above discussion that when shear
stress on the bio®lm is high, the bio®lm community would
have to regulate its metabolic pathway so as to maintain a
balance with the external detachment force through consuming non-growth-associated energy. As Russell and Cook
(1995) noted, energy spilling is probably not a fortuitous act,
and may be a mechanism for protecting bacteria from
stressful environmental conditions and potentially toxic
schemes of substrate metabolism.
ã 2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 90, 337±342
342
Y. LIU AND J.-H. TAY
REFERENCES
APHA (1995) Standard Methods for the Examination of Water and
Wastewater 19th edn. Washington, D.C.: American Public Health
Association.
Babcock, G.T. and Wikstrom, M. (1992) Oxygen activation and the
conservation of energy in cell respiration. Nature 356, 301±309.
Chang, H.T., Rittmann, B.E., Amar, D., Heim, R., Ehrlinger, O. and
Lesty, Y. (1991) Bio®lm detachment mechanisms in a liquid
¯uidized bed. Biotechnology and Bioengineering 38, 499±506.
Chen, M.J., Zhang, Z. and Bott, T.R. (1998) Direct measurement of
the adhesive strength of bio®lms in pipes by micromanipulation.
Biotechnology Techniques 12, 875±880.
Chen, S.Y. and Huang, S.Y. (2000) Shear stress effects on cell growth
and L-DOPA production by suspension culture of Stizolobium
hassjoo cells in an agitated bioreactor. Bioprocess Engineering 22, 5±12.
Cook, G.M. and Russell, J.B. (1994) Energy-spilling reactions of
Streptococcus bovis and resistance of its membrane to proton
conductance. Applied and Environmental Microbiology 60, 1942±1948.
Deamer, D.W. and Akeson, M. (1994) Role of water in proton
conductance across model and biological membranes. In Biomembrane Electrochemistry ed. Blank, M. and Vodynoy, I. pp. 3±25.
Washington: ACS.
Kwok, W.K., Picioreanu, C., Ong, S.L., van Loosdrecht, M.C.M., Ng,
W.J. and Heijnen, J.J. (1998) In¯uence of biomass production and
detachment force on bio®lm structures in a bio®lm airlift suspension
reactor. Biotechnology and Bioengineering 58, 400±407.
Lertpocasombut, K. (1991) Epuration carboneÂe par ®lm biologique
mince dans un reÂacteur aÁ lit ¯uidise triphasique. PhD Thesis,
Institut National des Sciences AppliqueÂes-Toulouse, France.
Liu, S.S. (1989) A new function of proton pumps in membrane fusion.
In Current Biochemical Research in China ed. Zhou, C.L. pp.
161±171. New York: Academic Press.
Liu, Y. (1999) Model of dissolved organic carbon distribution for
substrate-suf®cient continuous culture. Biotechnology and Bioengineering 65, 474±479.
Lopez, J.M., Koopman, B. and Bitton, G. (1986) INT-dehydrogenase
test for activated sludge process control. Biotechnology and Bioengineering 28, 1080±1085.
Low, E.W., Chase, H.A., Milner, M.G. and Curtis, T.P. (2000)
Uncoupling of metabolism to reduce biomass production in the
activated sludge process. Water Research 34, 3204±3212.
Marshall, K.C. and Gruickshank, R.H. (1973) Cell surface hydrophobicity and the orientation of certain bacteria at interfaces. Archive
of Microbiology 91, 29±40.
Mayhew, M. and Stephenson, T. (1998) Biomass yield reduction: is
biochemical manipulation possible without affecting activated sludge
process ef®ciency. Water Science and Technology 38, 137±144.
Meijer, J.J., Hoopen, T., Luyben, H.J.G. and Libbenga, A.M. (1993)
Effects of hydrodynamic stress on cultured plant cells: a literature
survey. Enzyme Microbial Technology 15, 234±238.
Mitchell, P. (1961) Coupling of phosphorylation to electron and
hydrogen transfer by a chemiosmotic type of mechanism. Nature
191, 144±148.
Moreira, J.L., Alves, P.M., Aunins, J.G. and Carrondo, M.J.T. (1995)
Hydrodynamic effects on BHK cells grown as suspended natural
aggregates. Biotechnology and Bioengineering 46, 351±360.
Nollert, M.U., Diamond, S.L. and Mclntire, L.V. (1991) Hydrodynamic shear stress and mass transport modulation of endothelial cell
metabolism. Biotechnology and Bioengineering 38, 588±602.
Ohashi, A. and Harada, H. (1994) Adhesion strength of bio®lm
developed in an attached-growth reactor. Water Science and Technology 29, 281±288.
Pringle, J.H. and Fletcher, M. (1983) In¯uence of substratum
wettability on attachment of fresh bacteria to solid surface. Applied
and Environmental Microbiology 45, 811±817.
Rittmann, B.E. (1982) The effect of shear stress on bio®lm loss rate.
Biotechnology and Bioengineering 24, 501±506.
Rittmann, B.E., Trinet, F., Amar, D. and Chang, H.T. (1992)
Measurement of the activity of a bio®lm: effects of surface loading
and detachment on a three-phase liquid ¯uidized bed reactor. Water
Science and Technology 26, 585±594.
Russell, J.B. and Cook, G. (1995) Energetics of bacterial growth:
balance of anabolic and catabolic reactions. Microbiology Review 59,
48±62.
Selna, M.W. and Schroeder, E.D. (1979) Response of activated sludge
processes to organic transient II-stoichiometry. Journal of Water
Pollution Control Federation 51, 150±157.
Shen, T. and Wang, J.Y. (1991) Biochemistry. Beijing: High Education
Press.
Tay, J.H., Xu, H.L. and Teo, K.C. (2000) Molecular mechanism of
granulation. I: H+ translocation-dehydration theory. Journal of
Environmental Engineering 126, 403±410.
Teo, K.C., Xu, H.L. and Tay, J.H. (2000) Molecular mechanism of
granulation. II: Proton translocating activity. Journal of Environmental Engineering 126, 411±418.
Trevors, J.T. (1984) The measurement of electron transport system
(ETS) activity in freshwater sediment. Water Research 18, 581±584.
Trinet, F., Heim, R., Amar, D., Chang, H.T. and Rittmann, B.E.
(1991) Study of bio®lm and ¯uidization of bioparticles in a threephase ¯uidized-bed reactor. Water Science and Technology 23,
1347±1354.
Trulear, M.G. and Characklis, W.G. (1982) Dynamics of bio®lm
processes. Journal of Water Pollution Control Federation 9,
1288±1301.
Vanderborght, J.P. and Gilliard, C. (1981) BOD removal in ¯uidized
beds: high biomass concentration or high oxygen transfer. In
Biological Fluidized Bed Treatment of Water and Wastewater ed.
Cooper, P.F. and Atkinson, B. pp. 383±389. Chichester: Ellis
Horwood.
van Loosdrecht, M.C.M., Lyklema, J., Norde, W., Schraa, G. and
Zehnder, A.J.B. (1987) The role of bacterial cell wall hydrophobicity
in adhesion. Applied and Environmental Microbiology 53, 1893±1897.
Vieira, M.J., Melo, L.F. and Phinheiro, M.M. (1993) Bio®lm
formation: Hydrodynamic effects on internal diffusion and structure.
Biofouling 7, 67±80.
Wolfe, S.L. (1993) Molecular and Cellular Biology. Wadsworth
Publications Co. Belmont, California, USA.
Zubay, G. (1984) Biochemistry. Addison-Wesley Publishing Company,
New York, USA.
ã 2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 90, 337±342