Previous Article | Next Article 
Applied and Environmental Microbiology, November 1999, p. 5009-5016, Vol. 65, No. 11
0099-2240/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.
Relative Importance of Trophic Group Concentrations
during Anaerobic Degradation of Volatile Fatty Acids
Ravi K.
Voolapalli and
David C.
Stuckey*
Department of Chemical Engineering and
Chemical Technology, Imperial College of Science, Technology and
Medicine, London SW7 2BY, United Kingdom
Received 4 May 1999/Accepted 26 August 1999
 |
ABSTRACT |
Although obligate syntrophic reactions cannot proceed without
hydrogenotrophs, it has been unclear from the literature whether potential improvements are achievable with higher concentrations of
hydrogenotrophs. In this study, the relative importance of formate-/H2-utilizing and acetate-utilizing trophic groups
in the anaerobic degradation of butyrate and propionate was assessed by
adding various proportions of these enriched cultures to a mixed
anaerobic seed inoculum. The improvement resulting from the additional
acetate-utilizing cultures was much greater than with
formate/H2 utilizers. Furthermore, formate/H2
utilizers did not improve propionate utilization significantly,
suggesting the importance of optimum utilization of hydrogenotrophic
capacity. During most of the volatile fatty acid (VFA) degradation
period, the system responded with characteristic hydrogen levels to
maintain the Gibbs free energy of oxidation approximately constant for both butyrate (
6 kJ) and propionate (14 kJ). These free-energy values were independent of methanogenic activity, as well as the volume
of the seed inoculum and the VFA concentrations present. By comparing
the experimental results with kinetic and mass transfer models, it was
postulated that the diffusional transfer of reducing equivalents was
the major limiting factor for efficient VFA degradation. Therefore, for
optimum utilization of the hydrogenotrophs, low acetate concentrations
are vital to enable the system to respond with higher
formate/H2 levels, thus leading to improved transfer of
reducing equivalents. Due to the small number of propionate utilizers
(and hence their limited surface area) and low bulk liquid
concentrations, the additional formate/H2 utilizers were of
minimal use for improving the degradation rate further. The butyrate
degradation rates strongly correlated with the cumulative activity of
hydrogenotrophs and acetotrophs over the experimental range studied,
indicating the need to model obligate syntrophic reactions as a
dependent function of methanogenic activity.
| |
INTRODUCTION |
In methanogenic environments,
acetogenesis is a key process in the mineralization of organic waste,
with propionate and butyrate being the major volatile fatty acids
(VFAs) produced (16). It is known from thermodynamics as
well as experimental studies that the rate of VFA degradation is
severely constrained by the free energy available for these organisms.
Since both hydrogen and acetate are the final products of VFA
degradation, their accumulation significantly lowers the free energy
available for the organisms carrying out these reactions (Table
1). Recently, formate has also been
suggested as an alternative electron carrier and was postulated to be
more important than hydrogen in suspended systems due to mass transfer
considerations (2, 30, 31, 33). Nevertheless, the maximum
hydrogen or formate concentrations that can be tolerated by the VFA
degraders are extremely low. As a result, while degrading VFAs, these
obligate syntrophs can only grow in the presence of a hydrogen- or
formate-consuming partner, usually a methanogen. A characteristic of
this cooperation is the equalization of growth rates with a tight
coupling of growth rates and rate-limiting substrate concentrations,
while the ratio of these partner organisms is dependent upon the type
of syntrophic reaction (13, 23, 29).
Since these syntrophic cultures act as a hypothetical composite
species, the rate of VFA degradation is proportional to the H2-consuming activity. Both Schmidt and Ahring (27,
28) and Dwyer et al. (5) demonstrated this by the
addition of hydrogen utilizers to the VFA degraders, as well as by
inhibiting the hydrogenotrophs. Although these studies highlighted the
critical importance of hydrogenotrophs for VFA degradation, they were
not entirely clear about the quantitative improvements possible with
further addition of hydrogenotrophs to mixed-culture inocula. For
example, Schmidt and Ahring (27, 28) observed substantial
improvements after addition of hydrogen utilizers to disintegrated
granules while improvements with propionate utilizers were much smaller
than those obtained with a butyrate-degrading culture, although
comparatively, propionate utilizers suffer more severe inhibition due
to H2. Since most of these product inhibition experiments
were done primarily with pure cultures of VFA degraders and methanogens
(1, 5), it would be useful to understand the influence of
methanogens on a mixed-culture digester inoculum, as the ratio of VFA
consumers to hydrogenotrophs could be different, and also this changes
with time under shock loads. Such quantification and understanding of
the rate-limiting mechanism would help in controlling digesters more
efficiently. Although acetate is a potential inhibitor of VFA
degradation, its role has often been underestimated. Similarly, most of
the designs and control strategies suggested in the literature are
concerned mainly with hydrogen and no attention has been paid to the
control of acetate (9, 22).
Therefore, the objective of this work was to gain a better
understanding of the relative contribution of acetate and
formate/H2 utilizers added to a mixed-culture digester seed
degrading VFAs. A second objective was to analyze the thermodynamic,
kinetic, and mass transfer implications of acetate regulation and its
impact on VFA degradation in order to improve digester design and control.
| |
MATERIALS AND METHODS |
Design of experiments.
Experiments were performed in 160-ml
serum bottles in batch mode. Fixed volumes of inocula (20 ml) from a
digester (operating at steady state with 4 g of chemical oxygen
demand (COD) per liter of sucrose-based feed at a 15-day HRT) were
supplemented with varying volumes of enriched formate and acetate
cultures into 30% CO2-70% N2-flushed serum
bottles. The concentration of volatile suspended solids in this
inoculum was 1.67 g/liter, while the Sauter mean diameter of sludge
flocs operating with a similar feed was ~18 µm (34, 35).
The enrichment cultures were added in portions of 10 to 30 ml by volume
(see Tables 2 and 3). In two treatments, 30 ml of an enriched formate
and acetate culture mixture was added at ratios of 2:1 and 1:2. To
assess the initial population of VFA utilizers, in some treatments the
inoculum volume was also varied. The total volume of the culture medium
was 55 ml, and this was achieved by adding various proportions of the anaerobic medium of Owen et al. (19). After preparation of
the culture medium, the serum bottles were inoculated in a 35°C water bath and left for 2 days to exhaust the residual carbon source in the
cultures. The batch experiment was started by spiking the serum bottles
with 5 ml of a concentrated VFA solution as sodium salt to result in a
final concentration of approximately 2,500 mg/liter in the bottle. The
degradation was then monitored by measuring the headspace hydrogen and
the liquid-phase VFAs.
Analytical methods.
The VFAs were analyzed by
high-performance liquid chromatography, and the gas-phase hydrogen was
measured by using GMI's exhaled-H2 monitor
(34). All experimental treatments were done in duplicate, and the mean value was reported. The deviation observed between duplicates was, in general, less than 20% (mostly less than 10%), but
at lower VFA concentrations (<200 mg/liter), 50 to 100% deviations were sometimes observed. The Gibbs free-energy values were calculated by using the standard free-energy values after correcting for temperature (3, 36).
Enrichment cultures. (i) Formate.
Enriched formate cultures
were selected for several reasons: firstly, because both hydrogen and
formate are possible electron transfer carriers for syntrophic
degradation, and growth on formate is relatively simple and the
activity can be quantified more easily; secondly, because in suspended
systems the transport of formate seems to be more important than that
of hydrogen (2, 30, 31). Finally, during step shock load, as
well as pulse load, experiments, use of this sludge fed with 4 g
of COD per liter resulted in greater amounts of formate than hydrogen
being accumulated (35). Nevertheless, the formate-enriched
cultures used in this study were able to use both H2 and
formate. Throughout this work, it was assumed that the formate
utilizers will use H2 if it is the main mode of electron
transport (at least at the same rate as formate). This appears to be a
valid assumption, as Schmidt and Ahring (27, 28) did not
find any differences in the stimulation effect on either butyrate or
propionate degradation with cultures that could use only hydrogen or
both formate and hydrogen.
The formate cultures were enriched in 160-ml serum bottles through
serial dilution (5-ml inocula in 45 ml of anaerobic medium)
from a
mixed-culture inoculum obtained from an anaerobic baffled
reactor
treating a sucrose-based feed (
34). Formic acid (50
mg) was
injected daily into the serum bottles as a 1-ml concentrated
solution.
The concentrated formic acid solution was prepared by
adding 2 ml of
formic acid (BDH) to 50 ml of a filter-sterilized
effluent from a
digester treating a sucrose-based feed. The COD
of this effluent was
less than 400 mg/liter (much lower than the
formic acid COD in 1 ml,
i.e., a COD of 17 versus 0.4 mg/liter),
and this solution was used to
avoid any nutrient deficiency. After
10 days of batch operation, the
seed from the serum bottles was
diluted to 10% with anaerobic medium
and the enrichment was repeated
four times in this manner. The typical
daily gas production was
24 ml, with almost stoichiometric amounts of
CH
4. The ability
of these organisms to consume hydrogen was
tested by adding a
5% H
2-30% CO
2-65%
N
2 gas mixture, and methane production was measured
by gas
chromatography.
(ii) Acetate.
The acetate enrichments were developed in a
4-liter stirred tank reactor while continuously feeding acetate at an
HRT of 8 days. The feed consisted mainly of sodium acetate, while
peptone and meat extract were added as nutrient sources. Five liters of feed contained 27.33 g of sodium acetate, 3.0 g of peptone,
1.0 g of meat extract, 0.6 g of
K2HPO4, 1.62 g of
KH2PO4, 1.803 g of
Na2HPO4, 0.0357 g of CoCl2 · 6H2O, 0.2355 g of FeCl2 · 4H2O, 0.01125 g of MnCl2 · 4H2O, 0.01125 g of Na2MoO4 · 2H2O, 0.0135 g of NiCl2 · 6H2O, 0.05 g of cysteine, and 1.25 g of
Na2S. A nutrient solution consisting of S4 (3 ml/liter of
feed) and vitamin solution S7 (10 ml/liter of feed) (19) was
also added. All of the trace nutrients were injected separately daily,
while the feed, buffer, peptone, and meat extract were autoclaved and
fed continuously. To quantify the contamination due to VFA and formate
utilizers, separate serum bottles were set up with spikes of formate,
butyrate, and propionate for both enriched acetate and formate
cultures. The formate utilization capacity (milligrams of formate per
day per milliliter of culture) of enriched acetate utilizers was at least 10 times lower than the enriched formate utilizer activity (during a 24-h period). Less than 30 mg of acetate was consumed by
enriched formate utilizers in 7 days. Approximately 300 and 70 mg of
butyrate was consumed during 5 days of operation, and 580 and 840 mg of
propionate was consumed in a 16-day period by enriched acetate and
formate cultures, respectively (on a 1-liter basis).
Product inhibition model.
To compare the influence of
trophic groups on butyrate degradation, a product inhibition model
based on thermodynamics was selected from the literature
(11). It was expected, from the work of Dwyer et al.
(5), that the butyrate consumers would initially grow
exponentially but that after a short time, the butyrate consumption
would be linear. By neglecting the growth of the organisms during the
linear period, the butyrate consumption rate
(rB) can be described (11):
|
(1)
|
where
= [A]
2[H
2]
2/[B],
K = 2.1e
8 (equilibrium constant), and the rate
of acetate accumulation is
|
(2)
|
where
max_A and
max_B are the maximum acetate uptake and VFA
uptake (in the absence of thermodynamic
inhibition) and the term
/K
accounts for product inhibition due
to acetate and hydrogen. [B] and
[A] are the molar butyrate and
acetate concentrations, while
H
2 is the hydrogen partial pressure
in atmospheres.
KSA and
KSB are the
half-saturation constants
for acetate and butyrate (30 and 7 mg/liter),
and
Ath is the threshold
acetate concentration
(0.6 mg/liter) (
21). H
2 metabolism was
not
incorporated into the model equations; instead, the dynamic
variation
of hydrogen was incorporated through a polynomial equation
fitted to
the experimentally measured H
2 values over time. The
model
equations were integrated by using the modelling and simulation
package
gPROMS, and
max_A and
max_B were estimated
by using a parameter
estimation package, gEST (
18).
| |
RESULTS AND DISCUSSION |
Butyrate degradation.
Figure 1
shows the butyrate and acetate dynamics during the batch degradation of
butyrate with the addition of enriched formate and acetate cultures in
comparison to the control (experiments 2, 6, and 9 in Table
2). The theoretical-model results are
also presented in the same graphs as continuous lines. It can be seen from Fig. 1 that there was a significant improvement in VFA
degradation, in comparison to the control, when acetate and formate
utilizers were added, and this improvement was much greater in the
presence of acetate utilizers than with formate utilizers. Butyrate was consumed completely in the presence of acetate utilizers after only
135 h, whereas it was nearly 500 mg/liter in the control and 160 mg/liter with formate utilizers, even after 160 h. Figure 2 shows the headspace H2
variation in the control and in the presence of acetate enrichment
cultures, which were quite similar (maximum H2 of 26 to 28 Pa). In contrast, with formate utilizers, the maximum H2
levels reached were only around 15 Pa. However, these H2
levels did not exactly reflect the observed butyrate degradation rates, as the highest rates were noted with the acetate utilizers. This is in
contrast to the results of Schmidt and Ahring (27, 28), who
suggested that the rate of VFA degradation is inversely related to
headspace hydrogen levels. Nevertheless, these hydrogen levels resulted
in similar free-energy values for all of these treatments, which were
independent of methanogenic activity and VFA concentrations. During most of the linear butyrate consumption period, the
free-energy values were between 10 and 6 kJ and were almost
stable at 6 kJ after 100 h (Fig. 2). However, this time period
would have been much shorter if hydrogen mass transfer limitations were
accounted for. Pauss et al. (20) suggested that with poorly
soluble compounds like hydrogen, the liquid-phase concentrations will
often be significantly higher than the headspace levels due to the slow
mass transfer from the liquid to the gas phase. Interestingly, these
stable 
G' values were quite similar in all of the treatments and
close to the values reported in the literature (5, 10).
Similar "energetic homeostasis" has also been reported earlier for
the syntrophic H2 producer during ethanol oxidation
(29). Under energetic homeostasis, the H2
producers work close to the thermodynamic limit of their survival
through close feedback control of the product concentrations
(5).
View larger version (15K):
[in this window]
[in a new window]
|
FIG. 1.
Butyrate degradation and acetate accumulation in the
control and with the addition of formate and acetate utilizers (see
experiments 2, 6, and 9 in Table 2). The simulation results are shown
as continuous lines from 65 h onward.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 2.
Headspace hydrogen and Gibbs free-energy variations in
serum bottles during the batch degradation of butyrate with the control
and the addition of formate and acetate utilizers (see experiments 2, 6, and 9 in Table 2).
|
|
Variation of max with formate/H2 and
acetate removal.
The equilibrium thermodynamic model (equation 1)
described the experimental data quite closely (Fig. 1). However, as the
G' values during the degradation period were almost the same, the model predicted the different responses by varying the
max value of butyrate degradation.
Figure
3 shows the variation in
max_B in the bottles with formate and acetate
utilizer addition and the control.
The increase in the formate (Table
2, column 6, the numerical
values of experiments 4 to 6 minus the
control experiment 2) and
acetate (Table
2, column 7, the numerical
values of experiments
7 to 9 minus the control experiment 2) maximum
uptake rates over
the control, after the additions, is plotted on the
abscissa;
the
max for formate uptake was determined
experimentally before
the commencement of these experiments. The
max values for both
acetate and butyrate were obtained
from the parameter estimation
results (columns 7 and 8, Table
2). From
the slopes of the regression
lines, it can be seen that the improvement
in
max_B is quite small when the number of
formate/H
2 utilizers is increased
in comparison to the
acetate utilizers (
max_B values
used for
acetate enrichments were modified to compensate for slight
formate
contamination). From the thermodynamics, it could be expected
that
lowering of the formate and hydrogen levels would improve
VFA
degradation, as acetate is only inhibitory at very high levels,
comparatively. It was surprising that the improvement with increasing
amounts of acetate utilizers was much greater than with the
formate/H
2 utilizers. Interestingly, the
max_B values fitted through
parameter
estimation (equation 1) correlated reasonably well with
the predicted
max_B values for all of the batch experiments
(Table
2). The average error observed between fitted and estimated
max_B values was less than 6%, with a
correlation of
0.97 (with 12 datum points). This result clearly
demonstrates
the functional dependency of VFA degradation activity on
both
of the methanogens and the importance of acetate regulation.
Typically,
VFA degradation is modelled with a separate
max value although
it is known that VFA degraders are
obligate syntrophs, and hence
the VFA consumption rate is
stoichiometrically related to their
product removal rate. As these
results show, the maximum VFA uptake
rate is governed by the capacity
of both the acetate and hydrogen
utilizers and, in fact, varies over
time as the concentration
of these organisms changes with growth.
Moreover, equation 1 is
adequate only in describing the system behavior
with acetate or
hydrogen inhibition (
11). However, it cannot
describe the dynamics
of the variations in methanogenic activity (Table
2). Hence,
it is important to incorporate these changes in order to
describe
the kinetic capacity of VFA utilizers more accurately under
shock
loads, as well as under normal conditions. The same argument can
probably be used to explain why propionate degradation is considered
to
be the rate-limiting step under shock loads (
17) while the
kinetic turnover rates under normal conditions suggest that acetate
degradation is the rate-limiting step (
12).
Comparison of simulated and experimental responses.
The
greater improvements with acetate utilizers can be explained as due to
either improved kinetics of formate/H2 uptake or reduced
diffusional resistance. The following section analyses both of these
possibilities through a simple theoretical model.
Since the H
2/formate concentrations are extremely low, it
was assumed that these concentrations could be estimated from
thermodynamic
relationships. This is a reasonable assumption, as many
syntrophs
work close to a characteristic free-energy level under
steady-state
conditions and even in batch operation (
5,
10,
29); in
our experiments, this value was around
6 kJ. By
analogy, the
analysis presented for hydrogen should also be applicable
to formate.
Under energetic homeostasis,
![−&Dgr;G<SUB>C</SUB>−&Dgr;G<SUB>0</SUB><SUP>′</SUP>=RT·<UP>ln</UP><FR><NU>[H<SUB>2</SUB>]<SUP>2</SUP>[A]<SUP>2</SUP></NU><DE>[B]</DE></FR>](/content/vol65/issue11/fulltext/5009/img004.gif)
|
(3)
|
where
G
c is the stable minimum free energy of the
VFA degrader,
G
0' is the standard free energy of
reaction,
R is the
gas constant, and
T is the
temperature, and
![⇒[H<SUB>2</SUB>]=<FR><NU><RAD><RCD>[B]·K</RCD></RAD></NU><DE>[A]</DE></FR>](/content/vol65/issue11/fulltext/5009/img005.gif)
|
(4)
|
where
|
(5)
|
The hydrogen levels estimated through equation 4, in general,
represent the qualitative trend of the headspace measurements
and are
quite similar during the later stages of the experiment.
By neglecting
hydrogen accumulation, the rate of butyrate degradation
is related to
hydrogen uptake under kinetic as well as mass transfer
limiting
conditions as follows:
|
(6)
|
|
(7)
|
where
h is the average mass transfer coefficient for
the diffusional transfer of reducing equivalents in a unit volume of
reactor and
H2th is the threshold hydrogen
concentration of the
methanogens (0.3 Pa) (
21). The acetate
accumulation is computed
by using equation
2.
Figure
4 presents the simulated batch
degradation profiles in the control and the treatments with formate and
acetate utilizer
addition under three different controlling mechanisms.
The model
parameters (
h and
max values) used
in these simulations were
estimated separately for each mechanism by
using the control data,
while the influence of formate/H
2
and acetate utilizer addition
was simulated by proportionately
increasing these parameters using
their activity data (Table
2). Figure
4A was simulated by assuming
that hydrogen was the main mode of
electron transport and hydrogen
kinetics were the limiting factor
(
KSH = 5 µm) (
25). In contrast
to
the experimental results of all of the batch degradation tests,
the
simulated degradation rates increased more with the addition
of
H
2/formate utilizers than with the addition of acetate
utilizers.
View larger version (14K):
[in this window]
[in a new window]
|
FIG. 4.
Simulation results of butyrate degradation in the
control and the treatments with formate and acetate utilizer addition
under kinetic and mass transfer limitations. (A) H2 as the
main mode of transport. (B) Formate as the main mode of transport. (C)
Diffusion of H2/formate as the rate-limiting step.
|
|
The predictions were more similar to the experimental data if formate
was the electron transport mechanism between syntrophic
partners (Fig.
4B). The
KS and threshold formate concentration
used were 0.22 mM and 10 µM (
26). With formate utilizers,
the
simulated degradation rate was much higher during the initial
period. However, this soon reduced dramatically due to a faster
reduction of the concentration term (equation 6). In contrast,
acetate
regulation helped the system to respond with higher formate
levels
through lower acetate concentrations, thereby enhancing
butyrate
degradation. Figure
4C shows the simulated batch degradation
profiles
obtained by assuming that diffusion of formate/hydrogen
was the
rate-limiting step. The addition of formate utilizers
improves the mass
transfer rates by reducing the interbacterial
distance, thus
influencing the mass transfer coefficient value,
h. However,
the improvements with the additional formate utilizers
were not
substantial, as the interbacterial distance is reduced
by the cubic
root of the number of organisms. For example, to
double the VFA
degradation rate, the number of formate utilizers
required is eight
times the initial mass. In the above simulation,
the mass transfer
coefficient was increased by 1.3 times to account
for the addition of
formate utilizers while with acetate utilizers
the same value as the
control was used. It can be seen from Fig.
4C that the improvement
obtained with hydrogen utilizers quickly
diminished and almost reached
the control, whereas acetate regulation
helped the system to reach
higher fluxes through an improved concentration
gradient, even at lower
butyrate
levels.
Qualitatively, the models based on either formate kinetics or diffusion
explained the experimental data. However, neither
model predicted the
experimental data well with a unique set of
parameters, as the
mechanisms of transport and the concentrations
at the cell surfaces
were unknown. Moreover, the thermodynamic
relationship used in the
simulation study (equation 4) assumed
a constant [H
+]
(pH) but this changes with butyrate degradation, thus affecting
the
equilibrium
concentrations.
It was found that to obtain a good fit with the kinetic model, a very
low
KS value (0.003 µM) was necessary. The
max_H values fitted to the control and the
formate and acetate utilizer
additions were 0.39, 0.51, and 0.67 mmol
of H
2/liter · h, respectively.
If enhanced kinetics
were the reason for the improvements in VFA
degradation, the
max_H value obtained with formate utilizer
addition should have been twice that of the control, and with
acetate
utilizers it should have been the same as the control.
Interestingly,
these values can be explained in a more rational
way with mass transfer
limitations; i.e., doubling of the formate
utilizers improved butyrate
degradation 1.3 times, whereas the
improvements obtained with acetate
utilizers were inversely related
to their average acetate accumulation
during the degradation period
(500 mg/liter with acetate utilizers
versus 850 mg/liter in the
control).
The analysis presented here is subject to the limitations that batch
systems are not stationary and their conditions vary
over time. In the
present study, the growth of biomass during
the linear degradation
period was not considered, although as
a percentage change it should
not be significant. Ideally, the
experiments carried out as described
above should have been done
in continuous culture since the data might
have been more reliable
and easier to interpret. Furthermore, in batch
culture, the cell-cell
interactions could have changed when
artificially altering the
activity by dilution, whereas this problem
would be less important
in continuous culture systems. Ultimately,
batch systems were
used since the workload involved using continuous
cultures would
have been prohibitive. Nevertheless, this should not
undermine
the analysis here as butyrate degradation was almost linear
with
time (correlation coefficients obtained for seven datum points
were >0.97) and hence it could be restricted to a small time period.
Furthermore, the experimentally estimated improvement coefficients
predicted reasonably well for all of the other treatments tested
(Table
2), suggesting that acetate utilizers had a greater influence
than
formate utilizers. During the exponential phase, approximately
250 mg
of butyrate was consumed; with this, the hydrogen utilizer
yield could
have increased by 10 to 20% over the initial control
amount (assuming
that the yield is independent of the hydrogen
concentration). However,
the actual yield could have been lower
than this due to the
concentration dependency of the Gibbs free
energy (
15).
Nevertheless, the improvement obtained with formate
utilizers was in
the same range as that of Schmidt and Ahring
(
28), in whose
study the batch degradation period was relatively
short (<10 h).
Furthermore, if mass transfer limitations controlled
butyrate
degradation, propionate should be influenced even more
as propionate
degradation is feasible at much lower
concentrations.
Propionate degradation.
Figure 5
shows the propionate and acetate dynamics during the batch degradation
of propionate with the addition of enriched formate and acetate
cultures in comparison to the controls (experiments 1 to 4, 7, and 10 in Table 3). Propionate degradation was
quite similar to butyrate consumption; after a brief lag phase,
propionate was consumed in a linear fashion. The linear degradation
gave a 0.98 correlation (with five points) for all of the treatments tested (except treatment 1, where four points were used from time 257 h, for which the correlation obtained was 0.91). The rates of
degradation during the linear period were not substantially different
in any of the treatments tested (Table 3), but the greatest
improvements were observed in the treatments with acetate utilizers
added or doubling of the inoculum size (treatments 4, 9, and 12). The
addition of formate utilizers (treatments 5 to 7) resulted in a
response similar to that of the control (treatment 2), but this
decreased slightly at higher concentrations. The improvement obtained
by increasing the initial inoculum size by four times was only a factor
of 1.7. This was not entirely surprising, as such minimal improvements
were reported earlier with H2/formate utilizer addition
(27, 28). The headspace hydrogen values were quite similar
during the entire period, with the average values ranging from 6 to 9 Pa. Similar to butyrate degradation during the linear degradation
period, the free-energy values in all of the treatments were quite
constant (approximately 14 kJ; Table 3).
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 5.
Propionate degradation and acetate accumulation with
different inoculum sizes and treatments with formate and acetate
utilizer addition. Symbols:  , control (10 ml);  , control (20 ml);
 , control (30 ml);  , control (40 ml);  , plus formate utilizers
(30 ml);  , plus acetate utilizers (30 ml). Linear degradation rates
were noted from 218 h.
|
|
Unlike butyrate degradation, the improvements did not increase linearly
with acetate and formate utilizer addition. Figure
6 shows the propionate degradation
profiles with the addition
of various proportions of acetate and
formate utilizers; for comparison,
control data are also shown. It can
be seen from these graphs
that the improvements were greater with
smaller acetate or formate
culture additions, while higher
concentrations of these microorganisms
started reducing the stimulatory
effect. These differences were
not due to experimental errors, as
duplicate samples gave almost
identical responses, with an average
deviation of less than 10%.
These differences can be explained through
mass transfer limitations
during syntrophic reactions.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 6.
Propionate degradation with different amounts of formate
(top) and acetate (bottom) utilizers. Symbols: , control (20 ml);
, plus 10-ml enrichment; , plus 20-ml enrichment; , plus 30-ml
enrichment.
|
|
Mass transfer during syntrophic reactions.
Theoretically, the
rate of obligate syntrophic compound degradation is dependent on the
kinetic capacity of the syntrophic partner if diffusion of intermediate
metabolites is not the limiting factor. However, Boone et al.
(2) suggested that both formate and hydrogen utilizers
suffer severe mass transfer limitations when their bulk concentrations
are very low, thus leading to underutilization of their kinetic
capacity. Their mathematical model predicted sharp concentration
gradients of formate or hydrogen around the methanogens responsible for
their catabolism and, hence, the diffusion of these substrates. Formate
or H2 concentrations are lowest at the cell surface of the
methanogen and increase to bulk liquid concentrations at a distance of
about 10 µm from the cell, assuming a single-cell diameter of 1 µm
with dilute cell concentrations (106 to 107
cells/ml) (2). By comparing the literature (31)
with similar feed strength and methane production rates, the cell
concentration in the present study could be between 109 and
1010 cells/ml (average distance of less than 10 µm). At
these high concentrations, the cells can no longer be assumed to exist
as individual cells and the metabolite fluxes will vary significantly due to the influence of neighboring H2-/formate-consuming
or -producing cells. Figure 7 shows the
local flux variations that could develop when a methanogen is
surrounded by formate, H2, or acetate consumers at low and
high cell concentrations. In Fig. 7A, the electron flux will be uniform
for all of the methanogens as the distance between the cells is large
compared to the size of the concentration boundary layer (10 µm). In
contrast, at shorter interbacterial distances, the flux will be higher
with methanogens that are close to the VFA consumers compared to the
cells which are farthest away from them. Hence, at higher cell
concentrations, the diffusional transfer of electrons from VFA
consumers, and hence the reduced flux from the bulk liquid, becomes the
rate-limiting factor, thus severely constraining the stimulatory
effects obtainable with additional methanogens. Using the same
rationale, it could be hypothesized that the layered structure of
aggregates suggested by Macleod et al. (14) is more
efficient for VFA consumption as low acetate concentrations could be
maintained with minimal interference to syntrophic partners due to
aceticlastic or other organisms' presence.
View larger version (13K):
[in this window]
[in a new window]
|
FIG. 7.
Possible variations in electron fluxes from VFA consumer
to methanogens (open symbols) when interbacterial distances are large
(A) and small (B).
|
|
Grotenhuis et al. (
8) reported that in granular sludge the
highest percentage of hydrogenotrophs was present when treating
propionate-fed, compared to ethanol- or sucrose-fed, granules
due to
unfavorable propionate thermodynamics. Since the cell surface
area of
propionate oxidizers is limited, addition of extra methanogens
is of
little use as the mass transfer rates are low due to the
low
concentration gradient from the bulk liquid to the methanogens.
In the
present study, the greatest improvements (25%) were noted
when acetate
utilizers were added, and this was despite the fact
that, unlike
butyrate, the influence of acetate on propionate
degradation was less
dramatic, suggesting that mass transfer was
the main limiting factor
for further improvements. But this was
reduced to 15% at higher
concentrations of aceticlastic methanogens,
possibly due to the
shielding of the electron flux from producer
to consumer by
aceticlasts. Typically, the enriched cultures tended
to have higher
KS values than their syntrophically growing
methanogenic
counterparts. The reduced propionate degradation rates
with higher
concentrations of formate utilizers may be again due to the
shielding
of the electron flux by less efficient methanogens. In the
present
case, there was not much acetate production due to slow
propionate
degradation. However, Dong et al. (
4) reported
that the propionate
degradation rate was almost doubled after addition
of acetate
utilizers.
In contrast, with butyrate, a 30% improvement in degradation was
possible by doubling of the methanogens, and this was probably
due to a
reduced interbacterial distance. The arguments for mass
transfer
limitations presented in this report can be further strengthened
by the
observation that in a recent modelling study involving
syntrophic
perturbation experiments, the investigator had to consider
local
concentration variations through mass transfer to improve
the model
prediction capability (
6). Similarly, the half-saturation
constant for H
2 in the absence of mass transfer limitations
was
found to be very low (
7), suggesting that for most of
the time
H
2 utilizers are saturated kinetically. However,
this observation
contradicts the experimental results of Kasper and
Wuhrmann (
12),
who reported that in normal digesters the
capacity of H
2 utilizers
was only exploited to a maximum of
1%. Nevertheless, if mass transfer
is the main limitation for
H
2 metabolism, it provides a better
explanation for both a
low
KS value and a high surplus activity
of
H
2 utilizers. However, the possibility that different
enzymes
(depending on the H
2 concentration
[
24]), are responsible for
such variations should not
be ruled
out.
| |
ACKNOWLEDGMENTS |
R.K.V. to acknowledges the financial support of the Association
of Commonwealth Universities and Engineers India Ltd., New Delhi, for
granting him study leave.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Department of
Chemical Engineering and Chemical Technology, Imperial College of
Science, Technology and Medicine, London SW7 2B4, United Kingdom.
Phone: 0171-594 5591. Fax: 0171-594 5629. E-mail:
d.stuckey{at}ic.ac.uk.
| |
REFERENCES |
| 1.
|
Beaty, P. S., and M. J. McInerney.
1989.
Effects of organic acid anions on the growth and metabolism of Syntrophomonas wolfei in pure culture and defined consortia.
Appl. Environ. Microbiol.
55:977-983[Abstract/Free Full Text].
|
| 2.
|
Boone, D. R.,
R. L. Johnson, and Y. Liu.
1989.
Diffusion of the interspecies electron carriers hydrogen and formate in methanogenic ecosystems, and implications in the measurement of KM for H2 or formate uptake.
Appl. Environ. Microbiol.
55:1735-1741[Abstract/Free Full Text].
|
| 3.
|
Conrad, R., and B. Wetter.
1990.
Influence of temperature on energetics of hydrogen metabolism in homoacetogenic, methanogenic, and other anaerobic bacteria.
Arch. Microbiol.
155:94-98.
|
| 4.
|
Dong, X.,
C. M. Plugge, and A. J. M. Stams.
1994.
Anaerobic degradation of propionate by a mesophilic acetogenic bacterium in coculture and triculture with different methanogens.
Appl. Environ. Microbiol.
60:2834-2838[Abstract/Free Full Text].
|
| 5.
|
Dwyer, D. F.,
E. Weeg-Aerssens,
D. R. Shelton, and J. M. Tiedje.
1988.
Bioenergetic conditions of butyrate metabolism by a syntrophic, anaerobic bacterium in coculture with hydrogen-oxidizing methanogenic and sulfidogenic bacteria.
Appl. Environ. Microbiol.
54:1354-1359[Abstract/Free Full Text].
|
| 6.
|
Furumai, H.
1997.
Modeling of hydrogen behavior and reduced product formation following perturbation with hydrogenic substrate in a methanogenic reactor.
Water Sci. Technol.
36:255-262.
|
| 7.
|
Giraldo-Gomez, E.,
S. Goodwin, and M. S. Switzenbaum.
1992.
Influence of mass transfer limitations on determination of the half saturation constant for hydrogen uptake in a mixed culture CH4 producing enrichment.
Biotechnol. Bioeng.
40:768-776.
|
| 8.
|
Grotenhuis, J. T. C.,
M. Smit,
C. M. Plugge,
Y. Xu,
A. A. M. van Lammeren,
A. J. M. Stams, and A. J. B. Zehnder.
1991.
Bacteriological composition and structure of granular sludge adapted to different substrates.
Appl. Environ. Microbiol.
57:1942-1949[Abstract/Free Full Text].
|
| 9.
|
Harper, S. R., and F. G. Pohland.
1986.
Recent developments in hydrogen management during anaerobic biological wastewater treatment.
Biotechnol. Bioeng.
28:585-602.
|
| 10.
|
Hickey, R. F., and M. Switzenbaum.
1991.
Thermodynamics of volatile fatty acid accumulation in anaerobic digesters subject to increases in hydraulic and organic loading, Res.
J. Water Pollut. Control Fed.
63:141-144.
|
| 11.
|
Hoh, C. Y., and R. Cord-Ruwisch.
1996.
A practical kinetic model that considers end product inhibition in an anaerobic digestion process by including the equilibrium constant.
Biotechnol. Bioeng.
51:597-604.
|
| 12.
|
Kasper, H. F., and K. Wuhrmann.
1978.
Kinetic parameters and relative turnovers of some important catabolic reactions in digesting sludge.
Appl. Environ. Microbiol.
36:1-7[Abstract/Free Full Text].
|
| 13.
|
Kreikenbohm, R., and E. Bohl.
1986.
A mathematical model of syntrophic cocultures in the chemostat.
FEMS Microbiol. Ecol.
38:131-140.
|
| 14.
|
Macleod, F. A.,
S. R. Guiot, and J. W. Costerton.
1990.
Layered structure of bacterial aggregates produced in an upflow anaerobic sludge bed and filter reactor.
Appl. Environ. Microbiol.
56:1598-1607[Abstract/Free Full Text].
|
| 15.
|
McCarty, P. L.
1971.
Energetics and bacterial growth in organic compounds in aquatic environments, p. 71-94.
Marcel Dekker Inc., New York, N.Y
|
| 16.
|
McCarty, P. L.
1982.
One hundred years of anaerobic treatment, p. 3-22.
In
D. E. Hughes, D. A. Stafford, B. I. Wheaty, W. Badder, G. Lettinga, E. J. Nyns, and W. Verstrate (ed.), Anaerobic digestion 1981. Elsevier, Amsterdam, The Netherlands
|
| 17.
|
McCarty, P. L., and F. E. Mosey.
1991.
Modelling of anaerobic digestion processes (a discussion of concepts).
Water Sci. Technol.
24:17-33.
|
| 18.
|
Oh, M., and C. C. Pantelides.
1996.
A modelling and simulation language for combined lumped and distributed parameter systems.
Comput. Chem. Eng.
20:611-633.
|
| 19.
|
Owen, W. F.,
D. C. Stuckey,
J. B. Healy, Jr.,
L. Y. Young, and P. L. McCarty.
1979.
Bioassay for monitoring biochemical methane potential and anaerobic toxicity.
Water Res.
13:485-492.
|
| 20.
|
Pauss, A.,
G. Andre,
M. Perrier, and S. R. Guiot.
1990.
Liquid-to-gas mass transfer in anaerobic processes: inevitable transfer limitations of methane and hydrogen in the biomethanation process.
Appl. Environ. Microbiol.
56:1636-1644[Abstract/Free Full Text].
|
| 21.
|
Pavlostathis, S. G., and E. Giraldo-Gomez.
1991.
Kinetics of anaerobic treatment: a critical review.
Crit. Rev. Environ. Control
21:411-490.
|
| 22.
|
Poels, J.,
P. van Assche, and W. Verstraete.
1985.
Influence of H2 stripping on methane production in conventional digesters.
Biotechnol. Bioeng.
27:1692-1698.
|
| 23.
|
Powell, G. E.
1984.
Equalisation of specific growth rates for syntrophic associations in batch culture.
J. Chem. Technol. Biotechnol.
34B:97-100.
|
| 24.
|
Reeve, J. N.,
R. M. Morgan, and J. Nolling.
1997.
Environmental and molecular regulation of methanogenesis.
Water Sci. Technol.
36:6-7.
|
| 25.
|
Robinson, J. A., and J. M. Tiedje.
1982.
Kinetics of hydrogen consumption by rumen fluid, anaerobic digestor sludge, and sediment.
Appl. Environ. Microbiol.
44:1374-1384[Abstract/Free Full Text].
|
| 26.
|
Schauer, N. L.,
D. P. Brown, and J. G. Ferry.
1982.
Kinetics of formate metabolism in Methanobacterium formicium and Methanospirillum hungatei.
Appl. Environ. Microbiol.
44:540-554[Abstract/Free Full Text].
|
| 27.
|
Schmidt, J. E., and B. K. Ahring.
1993.
Effects of hydrogen and formate on the degradation of propionate and butyrate in thermophilic granules from an upflow anaerobic sludge blanket reactor.
Appl. Environ. Microbiol.
59:2546-2551[Abstract/Free Full Text].
|
| 28.
|
Schmidt, J. E., and B. K. Ahring.
1995.
Interspecies electron transfer during propionate and butyrate degradation in mesophilic granular sludge.
Appl. Environ. Microbiol.
61:2765-2767[Abstract].
|
| 29.
|
Seitz, H. J.,
B. Schink,
N. Pfennig, and R. Conrad.
1990.
Energetics of syntrophic ethanol oxidation in defined chemostat cocultures 1. Energy requirement for H2 production and H2 consumption.
Arch. Microbiol.
155:82-88.
|
| 30.
|
Stams, A. J. M.
1994.
Metabolic interactions between anaerobic bacteria in methanogenic environments.
Antonie van Leeuwenhoek
66:271-294[Medline].
|
| 31.
|
Stams, A. J. M., and X. Dong.
1995.
Role of formate and hydrogen in the degradation of propionate and butyrate by defined suspended cocultures of acetogenic and methanogenic bacteria.
Antonie van Leeuwenhoek
68:281-284[Medline].
|
| 32.
|
Thauer, R. K.,
K. Jungermann, and K. Decker.
1977.
Energy conservation in chemotrophic anaerobic bacteria.
Bacteriol. Rev.
41:100-180[Free Full Text].
|
| 33.
|
Thiele, J. H., and J. G. Zeikus.
1988.
Interaction between hydrogen- and formate-producing bacteria and methanogens during anaerobic digestion, p. 132-159.
In
Handbook on anaerobic fermentations. Marcel Dekker, Inc., New York, N.Y
|
| 34.
|
Voolapalli, R. K., and D. C. Stuckey.
1998.
Stability enhancement of anaerobic digestion through membrane gas extraction under organic shock loads.
J. Chem. Technol. Biotechnol.
73:153-161.
|
| 35.
| Voolapalli, R. K., and D. C. Stuckey. The
utility of hydrogen monitoring for methanogenic reactors influence of
formate production and H2 kinetics. Submitted for
publication.
|
| 36.
|
Westermann, P.
1994.
The effect of incubation temperature on steady state concentrations of hydrogen and volatile fatty acids during anaerobic degradation in slurries from wetland sediments.
FEMS Microbiol. Ecol.
13:295-302.
|
Applied and Environmental Microbiology, November 1999, p. 5009-5016, Vol. 65, No. 11
0099-2240/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Kim, J. R., Dec, J., Bruns, M. A., Logan, B. E.
(2008). Removal of Odors from Swine Wastewater by Using Microbial Fuel Cells. Appl. Environ. Microbiol.
74: 2540-2543
[Abstract]
[Full Text]
-
Satoh, H., Miura, Y., Tsushima, I., Okabe, S.
(2007). Layered Structure of Bacterial and Archaeal Communities and Their In Situ Activities in Anaerobic Granules. Appl. Environ. Microbiol.
73: 7300-7307
[Abstract]
[Full Text]
-
Chauhan, A., Ogram, A., Reddy, K. R.
(2004). Syntrophic-Methanogenic Associations along a Nutrient Gradient in the Florida Everglades. Appl. Environ. Microbiol.
70: 3475-3484
[Abstract]
[Full Text]
Top
Abstract
Introduction
