~ Pergamon 0043-1354(94)00177-4 
War. Res. Vol. 29, No. 2, pp. 571-578, 1995 
Copyright ~) 1995 Elsevier Science Ltd 
Printed in Great Britain. All rights reserved 
0043-1354/95 $7.00 + 0.00 
SULFIDE PRODUCT INHIBITION OF DESULFOVIBRIO 
DESULFURICANS IN BATCH AND CONTINUOUS 
CULTURES 
S. OKABE*~, P. H. NIELSEN~, W. L. JONES and W. G. CHARACKLIS 
Center for Biofilm Engineering, 409 Cobleigh Hall, Montana State University, 
Bozeman, MT 59717, U.S.A. 
(First received September 1993; accepted in revised form June 1994) 
Abstract--Sulfide product inhibition kinetics for growth and activity of Desulfovibrio desulfuricans was 
investigated in batch and continuous cultures at pH = 7.0. A non-competitive inhibition model adequately 
described sulfide product inhibition kinetics. Inhibition coefficient (K,) for maximum specific growth rate 
(/~hx) was 251 mg 1-1 S in a batch experiment. Cell yield determined in a chemostat was reduced in half 
by a sulfide concentration fabout 250 mg 1-1 S, which was very close to the K i value for the batch growth. 
Maximum specific growth rate (/z ~h x) and cell yield (Yc/e~c) were strongly inhibited by high levels of sulfide 
concentrations, whereas pecific lactate utilization rate increased with increasing sulfide concentrations. 
The results indicated an increase in the relative energy needed for maintenance to overcome sulfide 
inhibition and uncoupling rowth from energy production. However, D. desulfuricans tosome extent could 
recover from the shock of high sulfide concentrations. Stoichiometry for catabolic reactions (energy 
producing) did not change at high sulfide concentrations, while anabolic reactions (cellular synthesis) were 
strongly inhibited by high sulfide concentrations. These results uggested that separation of sulfide product 
inhibition into growth (cell yield) and activity (substrate utilization rate) was important o incorporate 
the sulfide product inhibition kinetics in a variety of applications. 
Key words--DesulJbvibrio desulfuricans, ulfide product inhibition, non-competitive inhibition kinetics, 
activity, cell yield, specific growth rate 
NOMENCLATURE 
D = dilution rate (t ~) 
i = total sulfide concentration (M~ 1-3) 
Ki = inhibition coefficient (M s 1-3) 
KLa ¢ = half-saturation coefficient for lactate (M s 1-3) 
m = maintenance oefficient (M s M~ 1 t - 1 )
S = effluent lactate concentration (Ms 1-3) 
TS = total sulfide (Mi 1-3) 
Y~n[~a c = intrinsic ell yield coefficient on lactate (M,. M( I ) 
Yc:Lac = observed cell yield coefficient on lactate 
(M,. M~ I ) 
/~ = specific growth rate (t - i ) 
/2ma x: maximum specific growth rate in the absence of 
sulfide (t i ) 
/~  = maximum specific growth rate in the presence of 
sulfide (t - i ) 
INTRODUCTION 
Sulfate reducing bacteria (SRB) cause serious prob- 
lems in sanitary sewer systems and industrial water 
systems because of production of highly toxic and 
corrosive hydrogen sulfide gas. The corrosion of 
*Present address: Department of Civil and Environmental 
Engineering, Miyazaki University, Miyazaki, Japan. 
i'Author to whom all correspondence should be addressed. 
:~Present address: Environmental Engineering Laboratory, 
Aalborg University, DK-9000 Aalborg, Denmark. 
concrete sewers occurs as a result of hydrogen sulfide 
production from sulfate-rich sewage by the activities 
of SRB (Mori et al., 1992). Furthermore, problems 
are well known in the oil industry where SRB cause 
serious corrosion of installations, plugging of for- 
mations, and contamination of petroleum with H2S 
(souring). In these industries, mathematical models 
have been developed for predicting and controlling 
SRB activity. A range of environmental factors 
affects the growth and activity of SRB and other 
bacteria in these systems, and a more quantitative 
knowledge of the significance of those environmental 
factors is needed for improving the mathematical 
models. One important factor is sulfide, a metabolic 
product from sulfate reduction, which has been re- 
ported inhibiting SRB growth and activity. Very high 
sulfide concentrations have, for example, been found 
throughout a petroleum formation with concen- 
trations up to 1000 mg I-~ H2S in the produced water 
(Subcasky, 1991), and may appear in anaerobic treat- 
ments of sulfate-rich wastewater (Hilton and Archer, 
1988). Therefore, sulfide product inhibition may be 
expected to some extent in these industrial water 
systems. 
Several studies of sulfide inhibition have been 
performed on anaerobic wastewater treatment, 
571 
572 S. OKABE et al. 
~-  ~.~ ~ 
~ =~ - _ 
~ . . 
NNO ~ b~ 
=N~-N 
¢~ r.) .< r.; 
0 0 0 
._ .~ ~.=..~ 
. -  ~ .  t . . -  e~ 
rr ~ ,,,, 'r ~ I 
¢~- ~V~ ~ ~ I t<  t< 
¢< 
O~ o'~ oq  ~t3 
g ~g g 
.~-~ --~ ~ ~ ~ 
..= 
especially on methane producing bacteria (Karhad- 
kar et aL, 1986; Koster et al., 1986; Hilton and 
Archer, 1988; Rinzema and Lettinga, 1988; Hilton 
and Oleskiewicz, 1989; Oleszkiewicz et aL, 1989; 
McCartney and Oleszkiewicz, 1991). The overall 
effect of sulfide on microbial sulfate reduction has 
been described qualitatively by several authors (Post- 
gate, 1984; Klemps et al., 1985; Shimada, 1987; 
Hilton and Oleskiewicz, 1989; Min and Zinder, 1990; 
McCartney and Oleszkiewicz, 1991; Reis et al., 1991; 
Okabe el al., 1992; Reis et al., 1992). The relevant 
literature is summarized in Table 1. However, no 
quantitative SRB sulfide product inhibition datum 
has been reported. Most of the data listed in Table ! 
are obtained from the batch experiment, inwhich pH, 
sulfide concentration, and limiting substrate cannot 
be maintained at the same levels over many gener- 
ations. Therefore, batch culture data reported in 
the literature must be interpreted cautiously. Sulfide 
toxicity is strongly dependent on pH, because the 
chemical equilibrium of sulfide species is pH depen- 
dent. At pH 8 most of total sulfide (TS) is in the HS- 
form, while at pH 6 most is in the n2s form. Molecu- 
lar hydrogen sulfide (H2S) has been found to be the 
major toxic form of sulfide because H2S can pass 
through the cell membrane (Speece, 1983; Reis et al., 
1991). 
The distinction between growth and activity 
becomes very important when the environmental 
conditions become xtreme and energy consumption 
shifts from growth to maintenance (Hunik et al., 
1990). Thus, it is speculated that under high sulfide 
conditions, the growth-associated H2S production 
will decrease, whereas total H2S production will 
remain relatively constant because the decrease in 
growth is counter balanced by increased nongrowth- 
associated H 2 S production caused by increased main- 
tenance nergy requirement. 
The goal of this paper is to describe quantitatively 
effects of sulfide on activity and growth of SRB, here 
exemplified by Desulfovibrio desulfuricans. The dis- 
tinction between growth and activity was clearly 
made in this study; growth was defined as biomass 
(cell) production rate and activity was defined as 
substrate utilization rate (e.g. lactate). 
THEORETICAL  B A C K G R O U N D  
In this study, it is assumed that sulfide is a non- 
competitive inhibitor, which decreases maximum 
specific growth rate (#max) but does not alter half- 
saturation coefficient (KLan). Non-competitive inhi- 
bition kinetics for lactate-limited cultures can be 
described as follows (Aiba et al., 1973) 
/Jmax SKi 
= (I) 
(Kt~ + S)(K, + i) 
where i =sulfide concentration (M~l-3), K i= 
inhibition coefficient (Mil-3), S=eff luent lactate 
Sulfide product inhibition kinetics 573 
concentration (M~ 1-3), and/~ = specific growth rate 
( t - ' ) .  
For S>>KL~c (e.g. in a batch culture), equation (1) 
can be simplified and linearized by plotting "" ~,h 1/12 max 
against i
1 1 i 
+ - -  (2 )  
inh •max Ki Pm~,, /-tm~,, 
where irih - -  #m~-  maximum specific growth rate in the 
presence of  sulfide (t - j) .  Thus, the x-interception 
of a best fit line gives -K i -  However, equation (2) 
cannot be applied to the case of S -  KL~ ¢ (i.e. in a 
chemostat culture). 
inh #m~, in a steady state chemostat at various sulfide 
concentrations can be determined based on a theor- 
etical construct extrapolated from the measured lac- 
tate concentration (S) to what i,h /~m~ would be based 
on the ratio of Kt~c to S as described rearranging 
equation (I) 
i.h _ D(KL~¢ + S) 
m~ - (3)  
S 
The value of KL~, has been determined previously in 
a similar type of  experiment (Okabe et al., 1992). 
In this study, steady state S's were determined at 
a constant dilution rate (D) and various sulfide 
concentrations. In this way, the effect of sulfide 
nn inh concentration ~. . /~  can be evaluated easily instead 
of conducting a series of experiments to determine 
dependence of S on D at various sulfide concen- 
trations. 
EXPERIMENTAL MATERIALS AND METHODS 
Microorganisms 
Desulfovibrio desulfuricans (ATCC 5575) was grown in 
Postgate medium G (Postgate, 1984) including Na-lactate 
(oL-lactic acid, SIGMA, L-1375) and Na2SO4. Trace el- 
ements and vitamins were added. The details of medium 
preparation and preculture of microorganisms have been 
described elsewhere previously (Okabe and Characklis, 
1992). 
Batch culture experiment 
Sulfide inhibition kinetics were determined in batch 
experiments using 500 ml Erlenmeyer flasks. Flasks were 
equipped with butyl rubber stoppers fitted with an injection 
port and a gas exchange port. 400 ml of the pH adjusted 
(pH = 7.0). Postgate medium G (Postgate, 1984) containing 
500 mg 1- ~ of lactate and 800 mg 1 - t of sulfate was dispensed 
into each Erlenmeyer flask and autoclaved at 121°C 
for 20 min. Trace elements and vitamin solutions were 
aseptically added after cooling. Na2S' 9H20 solution was 
prepared separately inan air-tight bottle. After autoclaving, 
the pH of the sulfide solution was adjusted to 7.0 in the 
air-tight bottle using a sterile syringe and sterile 1.0 N 
HCI and NaOH solutions. Then appropriate volumes of the 
sulfide solution were added to the autoclaved culture 
medium to obtain the designed total sulfide concentrations 
(0, 70, 220, 350, 700mgl -~ S). After addition of sulfide 
solution, the pH of the culture medium was remeasured and 
readjusted using sterile syringes. A sterile oxygen-free Na- 
dithionite solution was added to the sulfide-free medium to 
provide the required negative redox potential for growth 
(the final concentration of NazS204 was approximately 
30 rag/l). D. desulfuricans taken from an actively growing 
chemostat culture was inoculated into each flask and incu- 
bated at 35°C. The inoculum size from the chemostat was 
adjusted to obtain an initial cell count of approximately 
I × 10 7 cells ml -I. 
Continuous culture experiment 
The effect of sulfide on the activity and growth was 
determined in a lactate-limited chemostat operated at a 
constant dilution rate (D) of 0.20 h -~. The chemostat was 
equipped with a butyl rubber wall growth scraper continu- 
ously rotated by an electric motor to prevent wall growth. 
Constant pH (+0.1 unit) and temperature (_.+0.5°C) were 
maintained using a pH control system with sterile 1.0 N HCI 
and NaOH solutions and thermoregulator, respectively. 
Various concentrations of sulfide solutions with pH roughly 
adjusted to 7.5, were separately prepared and autoclaved, 
then fed to the chemostat by a peristaltic pump and speed 
controller (Cole-Parmer, Chicago, IU). A fine pH adjust- 
ment was conducted in the chemostat. Na-dithionite 
(Na2 $2 O4) was added to the sulfide-free medium to provide 
the required negative redox potential. Redox potential was 
monitored continuously using a redox probe (Orion) with 
a saturated calomel reference lectrode during the course 
of the experiments. The reported redox potential was 
standarized to a hydrogen redox couple. The details of 
chemostat set up protocol have been described elsewhere 
previously (Okabe and Characklis, 1992). 
Analytical methods 
At steady state, chemostat effluent samples were obtained 
for the following analyses: (1) total organic arbon (TOC); 
(2) soluble organic arbon (SOC); (3) total bacterial counts 
and cell size; (4) sulfate; (5) sulfide; (6) lactate; (7) acetate. 
The samples for SOC, lactate, and acetate were prepared by 
centrifugation at20,000 g and 4°C for 20 min using stainless 
steel centrifuge tubes. The obtained supernatant was sub- 
jected to these analyses. The sample for sulfate analysis 
was fixed with 1% (w/v) ZnAc solution to remove sulfide 
species and then filtered by Sterile Acrodisc filter (pore size 
0.2/~m, Gelman Science, No. 4192). Sulfate and acetate 
concentrations were measured using an ion chromatograph 
(DIONEX A1-450). Total dissolved sulfide concentration 
(TS, in mg l- ~ S) in the liquid phase was measured using the 
methylene blue method described previously by Cline 
(1969). Total cell number and cell size were measured using 
an image analyzer (Cambridge/Olympus Quantment 10) by 
the epifluorescence method escribed previously by Hobbie 
et al. (1977). Samples (variable volumes) for cell counts were 
obtained irectly from the chemostat, homogenized, stained 
with 0.02% acridine orange solution, and fixed on Nucle- 
pore filters (0.2ttm pore size, 25mm dia.). Cell number 
was reported as the mean of more than 20 measurements 
along the filter. Cellular carbon concentration was deter- 
mined converting cell size determined by an epifluorescence 
technique to cellular carbon using various factors (Okabe 
and Characklis, 1992). The details of the rest of the chemical 
analytical methods have been described elsewhere pre- 
viously (Okabe and Characklis, 1992). 
RESULTS 
Batch experiments 
Maximum specific growth rate in the presence of 
sulfide (/1 ~h ) of D. desulfuricans was determined from 
the cell doubling times (td) at various total sulfide 
concentrations in batch cultures containing excess 
substrates (Fig. 1). The initial growth rates at various 
sulfide concentrations were determined by linear in- 
terpolation of the data in Fig. 1 because total sulfide 
concentration, substrate concentrations (i.e. lactate 
and sulfate), and the pH of culture medium changed 
574 S. OKABE et al. 
1.0E+09: 
~. 1.0E+08 
E 
~ 1.0E+07 
1'o ~ a'o ,/o 5o 
Tlrne (hour) 
Initia Sulfide Concentration 
--=-- 0 mg/L --*-- 70 mg/L + 224 mg/L 
342 mg/L --at- 693 mg/L 
60 
Fig. 1. Effect of sulfide on growth of D. desulfurtcans 
batch cultures: temperature = 35°C, pH = 7.0. 
in 
during the incubation. Cell numbers for initial sulfide 
concentrat ions of 0, 70, and 224 mg 1- ~ S leveled off 
after 30 h due to depletion of lactate. For data 
interpretation pH and total sulfide concentrations, 
doubl ing times, and corresponding maximum specific 
growth rates (#i,]ahx) are presented in Table 2. 
The maximum specific growth rates " ~,h, ~,[/max ) deter- 
mined in the batch cultures never exceeded 0.20 h-J 
at each total sulfide concentration, which was signifi- 
cantly lower than values of continuous culture. 
The inhibition coefficient for maximum specific 
growth rate, Ki, was determined to be 251 mgl - t  S 
using the non-competit ive inhibit ion model [equa- 
tion (2)] as shown in Fig. 2. A high square of 
correlation coefficient (r 2) indicates that the non- 
competitive inhibit ion model adequately described 
sulfide inhibition. 
Continuous experiments 
Response to step changes in sulfide concentration. 
D. desulfuricans growing at a constant dilution rate 
(D) of 0.20 h ~ at 35°C and pH = 7.0 was continu- 
ously exposed to high total sulfide concentrations 
to examine the response to step changes in sulfide 
concentrat ion in terms of growth and activity. 
Sulfide solutions were fed continuously into the reac- 
tor to maintain constant otal sulfide concentrations 
during the experimental period. The initial total 
dissolved sulfide concentrations in both exper- 
iments were about 20 mg 1-' S. Addit ion of sulfide 
solutions was started at time zero. Cell concentra- 
tion decreased about 25% after 8 h exposure to 
Table 2. Results of sulfide ffect on the maximum specific growth rate 
of D. desulfuricans in batch cultures. Initial lactate and sulfate 
concentration were 500 and 800 mgl '  respectively 
TS (mgl IS) pH 
Initial End Initial End t d (h) ,u';h~ (h ') 
0 81 6.96 6.90 4.0 0.17 
70 161 6.95 6.82 3.5 0.20 
224 303 7.30 7.06 5.5 0.13 
342 390 7.00 6.95 7.0 0.10 
693 729 6.95 6.80 12.0 0.058 
20. 
A .c | 
!,2 
~8 
E 
J 
0 .4oo .ao0-a0o-lb0 o 16o a6o a0o Ah0 s0o e~o r00 
Total Sulfide Concentration (mg-S/I.) 
Fig. 2. Determination of K i value in the batch culture. 
Theoretical plot of the reciprocal of maximum specific 
growth rate in the presence of sulfide ~nh (I//~m~x) as a function 
of sulfide concentration. K i of 251 mg 1-1 S was determined 
(r 2 = 0.96). 
133___ 10mgl  ~ S, but recovered gradually to the 
original level after about 40 h [Fig. 3(a)]. The activity 
expressed as fractional lactate util ization [(Si - S)/Si] 
was relatively constant during the experimental 
period as effluent lactate concentrat ion increased 
slightly only after sulfide addition. Cell concentrat ion 
decreased 64% by exposure to 212_  23 mgl  -~ S. 
However, it recovered and approached a steady 
14 
8 
~8 
(a)  
Ca~ll Number 
4 
Lactate Cone. 
2 
-20 0 20 40 60 80 
Time (hour)  
30 
2o ~ 
q,l 
0 ~ 
lO0 
14 
10 
8 
~ 4 
2 
-20 
(b) 
Cell Number 
20 40 60 80 
Time (hour) 
40 
30 ;~ 
o 
2o ~ 
I00 ~ 
Fig. 3. Typical response of cell concentration and 
effluent lactate concentration to exposure to (a) 
133+10mgl - IS  and (b) 212_23mgl  -IS: D=0.20h -I, 
temperature= 35~'C, pH = 7.0. Error bars represent he 
standard eviation of mean (n = 2). 
Sulfide product 
state value, which was about 55% of the original 
cell concentration. A spike increase in effluent 
lactate concentration was observed after sulfide 
addition. Afterwards the effluent lactate concen- 
tration stabilized in the range of 0-5.0 mgl -~ even 
though the cell concentration was reduced by 45%. 
The culture approached steady state after about 40 h 
in both cases. Therefore, all samples in later exper- 
iments were taken at about 40h after the sulfide 
concentration was changed. Standard deviations 
of duplicate measurements of all cell counts were 
within 20% of their respective mean values, which 
indicating that the accuracy of cell count is reliable. 
Redox potential was below -180 mV during both 
experiments. 
Effect of sulfide on cell yield. Cell yield was deter- 
mined in a chemostat at various sulfide concen- 
trations (Fig. 4). The cell yield was maximum at 
108 mgl -~ S, after that it decreased with increasing 
sulfide concentration. The cell yield was reduced in 
half at about 250mgl-~S (Fig. 4). The effluent 
lactate concentration, however, remained relatively 
constant with increasing total sulfide concentration 
up to 332 mg 1- I S and increased to only 22.2 mg l- 
at 437mgl ~S (Fig. 4), indicating that activity ex- 
pressed as fractional actate utilization [ (S i -  S)/Si] 
was not affected in this range of sulfide concentration. 
These results uggested that specific lactate utilization 
rate increased when D. desulfurieans grew in high 
sulfide concentrations. 
Effect of sulfide on growth rate. Maximum specific 
- inh x growth rates U~m,x) were determined at various sul- 
fide concentrations u ing equation (3) with the efflu- 
ent lactate concentrations and gLa c of 2.35 mgl 
(Okabe and Characklis, 1992) (Table 3). The maxi- 
mum specific growth rate in the presence of sulfide 
(/li,~hx) decreased from 0.33 h -1 at low levels of TS 
to 0.21 h -~ at high levels of TS. /t~x determined in 
the chemostat was higher than the one determined 
in the batch study at a corresponding sulfide 
concentration. 
0.05 
0.04 
0.03  
0.02 
_ 0.01 
0 
5O 
v 
i i i , i 
100 200 300 400 
Total Sulfide Concentration (sg-S/L) 
40 
30 
20 
10 
500 
Fig. 4. Effect of sulfide concentration on cell yield and 
effluent lactate concentration: D = 0.20 h L tempera- 
ture = 35°C. Error bars represent the standard eviation of 
mean (n = 2). 
inhibition kinetics 575 
Table 3. The effect of sulfide on the maximum specific 
growth rate in a chemostat (mean + SD) 
Effluent lactate 
TS (mgl IS) concentration i,h (h t) ,Umal 
26.1 + 1.0 3.6_+0.2 0.33 
108.2 + 3.8 4.3 + 0.1 0.31 
157.9_+ 19.0 6.1 _+0.3 0.26 
259.1 -+ 16.9 6.7-+0.8 0.24 
284.4 -+ 7.1 7.6 -+ 0.2 0.25 
332.0 -+ 8.6 5.7 ± 0.1 0.26 
378.5-+ 13.0 9.8_+0.1 0.22 
437.5 ± 7.8 22.2 _+ 0.4 0.21 
Stoichiometry. Stoichiometry of sulfate reduction 
was determined at each sulfide concentration to 
elucidate whether the presence of sulfide results in an 
alteration of the metabolic pathway(s) for sulfate 
reduction. The stiochiometric ratios for catabolic 
(energy producing) reactions uch as lactate oxidized 
to sulfate reduced, CO 2 produced, and acetate pro- 
duced were independent of sulfide concentration 
(Table 4). The ranges of 0.42-0.51 for SO~-/lactate, 
0.81-0.94 for acetate/lactate, and 1.01-1.12 for 
CO2/lactate approached the theoretical ratios of 
0.47, 0.94, and 0.94, respectively (Okabe and 
Characklis, 1992). However, the stoichometric ratio 
for anabolic (cell synthesis) reactions, cell/lactate (i.e., 
cell yield), decreased with increasing total sulfide 
concentrations. 
DISCUSSION 
Effects of sulfide on cell yield and specific lactate 
utilization 
It was very important o separate the effect of 
sulfide inhibition into an effect on cell yield (growth) 
and on lactate utilization rate (activity) for a correct 
interpretation of experimental data. Cell yield de- 
creased ramatically from 0.036 (g cell)(g lactate)- 
at low levels of TS to 0.011 (g cell)(g lactate) ~ at 
437mg1-~S, while fractional lactate utilization 
(shown as effluent lactate concentration i Fig. 4) 
decreased at TS of only 437 mg 1 -l S. This decrease in 
cell yield may have been a result of a large part 
of the energy source being used for maintenance 
and uncoupling growth from energy production. 
Cell yield is known to be dependent on ~, the 
maintenance coefficient (m), and the intrinsic cell 
Table 4. Steady state stoichiometry of the continuous culture of 
D. desulturicans exposed to various total sulfide concentrations at a 
constant dilution rate of 0.20 h ~ (mean ± SD) 
Stoichiometric ratio (M/M) 
Total sulfide 
(mgl t S) SO] /Lac Ac/Lac. CO2/Lac. Cell/Lac. 
26.1 -+ 1.0 0.42 0.85 1.03 0.148 
108.2 ___ 3.8 0.50 0.81 1.08 tL 174 
157.9 -+ 19.0 0.49 0.84 1.05 0.128 
259.1 + 16.9 0.46 0.88 1.12 0.083 
284.4 + 7.1 0.51 0.91 1.01 0.067 
332.0 + 8.6 0.51 0.88 1.08 0.076 
378.5 -+ 13.0 0.48 0.83 1.09 0.065 
437.5 + 7.8 0.47 0.94 1.01 0.043 
576 S. OKABE et al. 
12 
10. 
8 
8- 
8 
"~ 2- 
0 • , 
0 5O 1(~ 1,~ 200 250 300 3~ 400 450 
Total Sulfide Concentration (rag-S/L) 
Fig. 5. Effect of sulfide concentration on maintenance 
coefficient (m). The maintenance oefficient was calculated 
based on equation (4) using i,t, _ Y~/L~¢ --0.03 g cell (g lactate)- l 
(Okabe et al., 1992). 
yield intr (Yc/LJ and is expressed as follows (Bulthuis, 
1989; Characklis, 1990) 
1 m 1 
- -  =-  + int------V-" (4) 
Yc/Lac D Yc/Lac 
The maintenance nergy was calculated using the 
observed cell yield (Yc/~c) in this study and the 
intrinsic cell yield of ~ntr Yc/Lac = 0.03 g cell (g lactate)- l 
measured in a similar type of experiment without 
sulfide inhibition (Okabe et al., 1992) (Fig. 5). 
The calculated maintenance oefficient increased ra- 
matically at total sulfide concentrations above 
200 mg 1 -~ S. 
A similar effect of sulfide has been observed in 
other types of bacteria. Mountfort and Asher (1979) 
reported that the biomass yield, the specific growth 
rate, and the intracellular levels of adenosine tri- 
phosphate (ATP) decreased with increasing sulfide 
level when Methanosarcina barkeri strain DM was 
grown in a batch system with methanol as carbon 
and energy source. Furthermore, they observed an 
increase in specific methane production rate. They 
concluded that the decrease in biomass yield resulted 
from an increase in the maintenance coefficient and 
uncoupling rowth from energy production. Hobson 
and Millis (1990) also reported that the maintenance 
energy requirements for a mixed culture grown in a 
two-stage chemostat with phenolics increased with 
increasing sulfide concentration, possibly due to the 
need to repair damaged cell membranes. 
Although data derived from batch culture exper- 
iments are often problematic, a degree of sulfide 
inhibition on cell yield in the chemostat agreed 
reasonably with the inhibition coefficient for #~nhmax 
(K~-----251 mgl -~ S) determined in the batch study, 
which is dependent on cell yield. The cell yield 
determined in the chemostat was reduced 50% at a 
sulfide concentration near 250 rag/l, even though K~ 
could not be determined in the chemostat study 
because equation (2) could not be used since the 
chemostat culture was lactate limited (S--KL,~). 
This degree of sulfide inhibition on cell yield also 
agreed with the result of the recovery from sulfide 
shock which showed that the cell number was re- 
duced to 55% of the original value by exposure to 
212_ 23 mgl - t  S [Fig. 3 (b)]. These results indicate 
that separation of sulfide inhibition into growth (cell 
yield) and activity (substrate utilization rate) is essen- 
tial to interpret experimental data correctly. 
Effect of sulfide on specific growth rate 
The maximum specific growth rates in the presence 
of sulfide (#~x) determined in the batch culture were 
significantly lower than the continuous culture data 
at each corresponding sulfide concentration. In our 
previous studies (unpublished ata), lower growth 
rates have been observed frequently in batch cultures, 
which is a normal observation for most SRB (Okabe 
and Characklis, 1992). This discrepancy of the values 
inh of/~max between the batch and the continuous exper- 
iment is attributed to several factors. Firstly, inh #ma~ for 
the continuous experiment was determined from the 
measured lactate concentration to what inh g~ would 
be based on the ratio of KL,¢ to S [equation (3)], which 
is yield independent. In contrast, ~nh #m~ for the batch 
experiment was determined by directly measuring the 
increase in cells, which is yield dependent. Specific 
growth rate (#) is directly proportional to the lactate 
utilization rate unless maintenance nergy require- 
ment and intrinsic cell yield change [equation (4)]. 
If maintenance energy requirement increases due to 
sulfide inhibition, specific lactate utilization conse- 
quently increases. As a result of the increase in 
maintenance energy requirement, or in other words, 
the increase in specific lactate utilization, lactate 
concentration i the chemostat remained relatively 
constant and low, even though cell concentration 
decreases. Thus, specific growth rate determined in 
the chemostat under sulfide inhibition was higher 
than that observed in the batch experiment. 
Secondly, the discrepancy is in part attributed to 
the selection of less susceptible microorganisms to
high sulfide concentrations and the recovery from 
sulfide shock in the continuous culture as demon- 
strated in Fig. 3(a) and (b). Another contributing 
factor for the low cell yield could be that death and 
lysis rates may increase at high sulfide levels. 
Final ly,  sulfide precipitates Fe in the batch exper- 
iment, so that Fe and other trace nutrients (not the 
carbon and energy source) may become the rate 
limiting factor. Moreover, pH, sulfide concentration, 
and limiting factor cannot be maintained at the same 
levels over many generations. If sulfide concentration 
changes as they grow, the pH shifts, or vice versa, 
another stress is imposed on the population which 
also requires more maintenance energy. 
Effect of sulfide on stoichiometry 
The stoichiometric ratios for catabolic reactions in 
this study did not change at high levels of TS. Thus, 
the same amount of energy (ATP) was generated 
Sulfide product 
from substrate-level-phosphorylation nd electron 
transport system (i.e. reduction of sulfate to sulfide). 
However, lower cell yields were observed in this study 
at high levels of sulfide, suggesting an increase in 
maintenance energy and uncoupling rowth from 
energy production. Minor changes in the anabolic 
products would not have been detected in this study 
because of the precision of the analytical methods 
used in this study. 
Effect of wall growth 
Hill and Robinson (1975) reported that a possible 
wall growth in a chemostat culture exerts significant 
effects on the cell concentration a d substrate utiliz- 
ation. Thus, the possible wall growth increases 
specific substrate utilization rate as a result of the 
decrease in suspended cell concentration. To evaluate 
the effect of wall growth on experimental results 
obtained in this study, the concentration of cells 
attached to the possible surfaces in the reactor system 
during an experimental period (about 2 weeks) 
was measured. The total numbers of attached 
cells were measured using the epifiuorescence t ch- 
nique at the end of the experiment and found to be 
2.4 x 104 cells cm -2, which is approximately 10% of 
an average total suspended cell concentration for the 
reactor with surface area of 613 cm 2. Thus, the effect 
of wall growth was regarded as insignificant in this 
experiment. 
Implication in sulfide inhibition model 
The kinetics of D. desulfuricans growth on lactate 
as a function of sulfide concentration are presented 
in Fig. 6 based on experimentally determined rate 
coefficients (Okabe and Characklis, 1992; Okabe 
et al., 1992) including the inhibition coefficient of 
251 mgl ~S determined in the batch experiment. 
These results suggest hat under high sulfide con- 
centrations uch as in the petroleum formation 
[0-1000mgl -t H2S (Subcasky, 1991)] and waste- 
water treatment systems (Hilton and Archer, 1988), 
sulfide inhibition kinetics should be considered in 
inhibition kinetics 577 
biological sulfide production models. Although other 
nutritional and physical conditions are suitable for 
SRB growth, SRB growth may be strongly inhibited 
by high sulfide concentrations. Furthermore, unless a 
change in specific sulfate reduction rate with increas- 
ing sulfide concentrations is taken into account, it is 
not possible to estimate the resulting sulfide pro- 
duction from growth rate data only. For instance, if
SRB growth is monitored as change in total SRB cell 
count, it may underestimate H2S production under 
inhibitory levels of sulfide concentration because the 
specific H2S production rate may be elevated at 
higher sulfide levels even through bacterial growth 
rate and biomass production rate decreased by sulfide 
inhibition. 
CONCLUSIONS 
Effects of sulfide on growth and activity of D. 
desulfuricans were quantitatively investigated in this 
paper. It can be concluded that separation of sulfide 
inhibition into cell yield (growth) and activity (lactate 
utilization rate) must be emphasized based on the 
following specific conclusions: 
1. Cell yield and maximum specific growth rate 
decreased with increasing sulfide concentrations, 
whereas pecific lactate utilization rate increased. 
2. Inhibition coefficient (Ki) for maximum specific 
growth rate was 251 mg I ~ S in the batch experiment. 
3. Cell yield determined in the chemostat was 
reduced in half by a sulfide concentration f approxi- 
mately 250 mg 1- ~ S, which was very closed to the K~ 
value determined in the batch culture. 
4. D. desulfuricans growing in the chemostat could 
recover from the shock of high sulfide concentrations. 
5. A non-competitive inhibition model adequately 
described sulfide inhibition of D. desulfuricans in the 
batch experiment. 
Acknowledgements--The authors gratefully acknowledge 
support from the cooperative agreement ECD-8907039 
between Montana State University and the National Science 
Foundation and from the Center Industrial Associates, 
0.4 
0.35 
0.3 
0.25 
0,2 
0.15 
• g 0. 1 
m 0.05 
1:0 
I:50 
1:100 
I:150 
1:200 
I =250 
110 ;0 3'0 ;0 50 
Lactate Concentration (mg/L) 
60 
Fig. 6. Results of model simulation. Inhibition coeffi- 
cient (K~) was determined from experimental data: K~ = 
251 mg 1 -J S,/lm, ~ = 0.34 h- 1 KL~¢ = 2.35 mg 1- i (Okabe and 
Characklis, 1992; Okabe et al., 1992). 
REFERENCES 
Aiba S., Humprey A. E. and Mills N. F. (1973) In: 
Biochemical Engineering, 2nd edn, p. 97. Academic Press, 
New York. 
Bulthuis B. A., Koningstein G. M., Stouthamer A. H. and 
Van Verseveld H. W. (1989) A comparison between 
aerobic growth of Bacillus licheniformis in continuous 
culture and partial-recycling fermentor, with contri- 
butions to the discussion on maintenance energy demand. 
Arch. Microbiol. 152, 499-507. 
Characklis W. G. (1990) Kinetics of microbial transform- 
ations. In: Biofilms (Edited by Characklis W. G. and 
Marshall K. C.) Wiley, New York. 
Cline J. D. (1969) Spectrophotometric determination of
hydrogen sulfide in natural waters. Limnol. Oceanog. 14, 
454-458. 
Hill G. A. and Robinson C. W. (1975) Substrate inhibition 
kinetics: phenol degradation by Pseudomonas putida. 
Biotechnol. Bioengng 17, 1599-1615. 
578 S. OKABE et al. 
Hilton B. L. and Oleskiewicz J. A. (1989) Sulfide-induced 
inhibition of anaerobic digestion. J Environ. Engng 114, 
1377-139 I. 
Hilton M. G. and Archer D. B. (1988) Anaerobic diges- 
tion of a sulfate-rich molasses wastewater: Inhibition of 
hydrogen sulfide production. Biotechnol. Bioengng 31, 
885-888. 
Hobbie J. E., Daley R. J. and Jasper S. (1977) Use of 
nucleopore filters for counting bacteria by fluorescence 
microscopy. Appl. Envir. Microbiol. 33, 1225-1228. 
Hobson M. J. and Millis N. F. (1990) Chemostat studies of 
a mixed culture growing on phenolics. J. Wat. Pollut. 
Control Fed. 62, 684-4591. 
Hunik J. H., Hamelers H. V. M. and Koster I. W. (1990) 
Growth-rate inhibition of acetoclastic methanogens by 
ammonia and pH in poultry manure digestion. Biol. 
Wastes 32, 285-297. 
Karhadkar P. P., Audic J. M., Faup G. M. and Khanna P. 
(1986) Sulfide and sulfate inhibition of methanogenisis. 
Wat. Res. 21, 1061-1066. 
Klemps R., Cypionka H., Widdel F. and Pfenning N. (1985) 
Growth with hydrogen, and further physiological charac- 
teristics of Desulfotomaculum species. Arch. Microbiol. 
143, 203-208. 
Koster I. W., Rinzema A., De Vegt A. L. and Lettinga G. 
(1986) Sulfide inhibition of the methanogenic activity 
of granular sludge at various pH-levels. War. Res. 20, 
1561-1567. 
Lehninger A. L. (1982) Enzymes. In Principles of Biochem- 
istry (Edited by Lehninger A. L.), pp. 207-247. Worth, 
New York. 
McCartney D. M. and Oleszkiewicz J. A. (1991) Sulfide 
inhibition of anaerobic degradation of lactate and acetate. 
Wat. Res. 25, 203-209. 
Min H. and Zinder S, H. (1990) Isolation and characteriz- 
ation of a thermophilic sulfate-reducing bacterium Desul- 
fotomaculum thermoacetoxidans sp. nov. Arch. Microbiol. 
153, 399-404. 
Mori T., Nonaka T., Tazaki K., Koga M., Hikosaka Y. 
and Noda S. (1992) Interactions of nutrients, moisture, 
and pH on microbial corrosion of concrete sewer pipes. 
War. Res. 26, 29-37. 
Mountfort D. O. and Asher R. A. (1979) Effect of inorganic 
sulfide on the growth and metabolism of Methanosarcina 
barkeri strain DM. Appl. Envir. Microbiol. 37, 670-475. 
Okabe S. and Characklis W. G. (1992) Effects of tempera- 
ture and phosphorous concentration  microbial sulfate 
reduction by Desulfovibrio desulfuricans. Biotechnol. Bio- 
engng 39, 1031 1042. 
Okabe S., Nielsen P. H. and Characklis W. G. (1992) 
Factors affecting microbial sulfate reduction by Desul- 
fovibrio desulfur&ans in continuous culture: limiting nutri- 
ents and sulfide concentration. Biotechnol. Bioengng 40, 
725-734. 
Oleszkiewicz J. A., Marstaller T. and McCartney D. M. 
(1989) Effects of pH on sulfide toxicity to anaerobic 
processes. Envir. Technol. Lett. 10, 815-822. 
Postgate J. R. (1984) In Sulfate-Reducing Bacteria, 2nd edn. 
Cambridge Univ. Press. 
Reis M. A., Almeida J. S., Lemos P. C. and Carrondo J. T. 
(1992) Effect of hydrogen sulfide on growth of sulfate 
reducing bacteria. Biotechnol. Bioengng 40, 593-600. 
Reis M. A., Lemos P. C., Almeida J. S. and Carrondo J. T. 
(1991) Evidence for the intrinsic toxicity of H2S to 
sulphate-reducing bacteria. Appl. Microbiol. Bioteehnol. 
36, 145-147. 
Rinzema A. and Lettinga G. (1988) The effect of sulphide 
on the anaerobic degradation of propionate. Envir. Tech - 
nol. Left. 9, 83-88. 
Shimada K. (1987) Removal of heavy metals from mine 
wastewater using sulfate-reducing bacteria. J. War. 
Waste, Jpn 31, 52-56. 
Speece R. E. (1983) Anaerobic biotechnology for indus- 
trial wastewater treatment. Envir. Sci. Technol. 17, 
416A-427A. 
Subcasky W. (1991) Personal communication.