JOURNALOF BIOSCIENCE AND BIOENGINEERWG

Vol. 89, No. 1, 40-46. 2000

Characterization of a Bioflocculant Produced by Citrobacter sp.-

TKF04 from Acetic and Propionic Acids
MASANORI FUJITA,‘* MICHIHIKO IKE,’ SHINYA TACHIBANA,’ GO KITADA,’ SHIN MYOUNG KIM,’
AND ZENSUKE INOUE*
Department of Environmental Engineering, Osaka University, 2-l Yamadaoka, Suita, Osaka 565-0871 I and Research
Center, Takuma Co. Ltd., I-2-1, Shinhama, Arai-cho, Takasago, Hyogo 676-8540,= Japan

Received 15 June 1999/Accepted 4 October 1999

A bacterial strain, TKFW, capable of producing a bioflocculant from acetic and/or propionic acids was
isolated from a biolilm formed in inside a kitchen drain. It was identified as a Citrobacter based on its mor-
phological and physiological characteristics and the partial sequences of its 16s rRNA. TKFO4 produced the
bioflocculant during the logarithmic phase of growth, and the optimum temperature and pH for the biofloc-
culant production were 30°C and 7.2-10.0, respectively. It could utilize some organic acids and sugars for its
growth as the sole carbon sources when yeast extract was supplemented; however, only acetate and propionate
were found to be good substrates for the bioflocculant production. The crude bioflocculant could be recovered
from the supernatant of the culture broth by ethanol precipitation and dialysis against deionized water. It was
found to be effective for flocculation of a kaolin suspension, when added at a final concentration of l-10 mg/l,
over a wide range of pHs (2-8) and temperatures (approximately 3-95’C), while the co-presence of cations
(Na+, K+, Ca2+, Mg2+, Fe’+, AP+ or Fe3+) did not enhance the flocculating activity. It could efficiently floc-
culate a variety of inorganic and organic suspended particles, including kaolin, dlatomite, bentonite, activated
carbon, soil and activated sludge. It contained glucosamine as the major component, and the molecular weight
was estimated to be between 232 and 440 kHa by gel filtration. The observation that the flocculating activity was
completely lost following chltinase treatment and its analysis with a Fourier transform infrared spectrometer
suggested that the bioflocculant is a biopolymer structurally-similar to chitin or chitosan.
[Key words: bioflocculant, Citrobacter sp. acetic acid, propionic acid, flocculating activity]

Various kinds of flocculants, typically inorganic alumi-
num or ferric salts and organic synthetic high polymers,
have been used for wastewater treatment, tap water
production and dredging/downstream processing tech-
niques in a variety of industrial fields (l-4). They are
utilized for various purposes according to their chemical
properties, taking their toxicity and environmental im-
pact into consideration. Among them, aluminum salts,
represented by polyaluminum chloride (PAC), have been
the most widely used so far for water and wastewater
treatment, however, health problems caused by alumi-
num salts, including Alzheimer’s disease, have been
reported recently (5-8). Production of large volumes of
sludge using aluminum salts is another problem (9). Poly-
acrylamide (PAA) derivatives, a major group of organic
synthetic polymers, are also used widely because of their
high performance and cost-effectiveness. However, PM
derivatives are not easily biodegraded in the natural
environment and the monomers derived from them have
both neurotoxicity and carcinogenicity (10, 11). There-
fore, a biodegradable and safe, i.e.; environmentally
friendly, flocculant is required to be developed as an
alternative in place of existing synthetic flocculants such
as PAC and PAA.
On the other hand, microbiologically-produced floc-
culants (bioflocculants) are generally expected to be readi-
ly biodegradable and harmless to the environment and
humans, indicating their potential to replace the existing
synthetic flocculants. Several types of bioflocculants have
been previously reported, and some of them exhibit
efficient flocculating activities comparable to those of the
* Corresponding author.
40
synthetic flocculants for not only inorganic but also
organic suspended particles (12-15). However, they have
the common problem of high production costs compared
with the synthetic flocculants, because relatively expen-
sive substrates such as glucose, fructose, sucrose and
L-glutamate are necessary for their production (2, 4,
16, 17). In order to reduce the production costs, we pro-
pose here to utilize lower-molecular, volatile fatty acids
(VFAs) such as acetic and propionic acids, which are
derived from anaerobic digestion or thermal treatment of
organic wastes including sewage sludge, as the substrates
for the bioflocculant production. In this study, a bacte-
rial strain Citrobacter sp. TKF04 which can produce a
bioflocculant from acetic and propionic acids as isolated
and characterized. The crude biofiocculant was recovered
from the culture broth after optimization of the batch
cultivation, and its functional and structural characteris-
tics were investigated.
MATERIALS AND METHODS
Media and culture conditions The basal medium
(BM) used in this study contained 1 .Og (NH4)*S04, 1 .Og
K2HP04, 0.05 g NaCl, 0.2 g MgS04.7H20, 1.0 g CaC&,
0.01 g FeC& and 0.1 g yeast extract in 1 1. Unless other-
wise stated, Acetate-Propionate (AP) medium containing
7.0 g sodium acetate and 3.Og sodium propionate in 1 1
of the BM (pH 7.2) was used as the medium for the
bioflocculant production.
The acetate and propionate concentrations in AP me-
dium were determined so as to represent the actual VFA
composition in anaerobically digested sewage sludge
(18).
VOL. 89, 2000 BIOFLOCCULANT
The effects of carbon source on the bioflocculant
production were investigated using the BM supplemented
with 1% (w/v) sodium acetate (AC. medium), sodium
propionate (Pr. medium), sodium lactate, sodium oleate,
sodium butyrate, hexadecane, methanol, ethanol, glu-
cose or lactose. To investigate the effects of the nitrogen
source, 0.1% (w/v) NaN03, NH&l, NHa03, urea, pep-
tone or meat extract was used instead of (NH&SO4 for
preparing the AC. or Pr. medium. The pH of all the
media was adjusted to 7.2 using 2 N NaOH or HCl. Cul-
tivation was performed in lOm1 of the medium in a test
tube or 200ml of the medium in a 500-ml Erlenmeyer
flask at 30°C on a reciprocal or a rotary shaker at 120
rpm, unless otherwise stated. For monitoring the bac-
terial growth, the optical density at 660nm was mea-
sured.
Screening for bioflocculant-producing bacteria and bac-
terial identification Bacterial strains were isolated
from a wide variety of environmental samples; 14 soil
samples from agricultural fields, forests and gardens,
nine biofilm sample from kitchen drains, nine activated
sludge samples including scum, four river water samples,
four sediment samples from domestic drainages and etc.,
using solid AP medium. They were cultivated in AP
medium for 3 to lOd, and the resultant fully grown
cultures were examined for their flocculating activity for
a kaolin clay suspension (see below). Bacterial strains
which showed considerable flocculating activities were
regarded as the bioflocculant-producing bacteria. For
taxonomical studies, morphological and physiological tests
based on Hasegawa’s procedure (19) were used, and iden-
tification was performed according to “Bergey’s Manual
of Systematic Bacteriology” (20). The partial sequences
of the 16s rRNA were determined as described by
Rochelle et al. (21). The API20E bacterial identification
kit (BioMerieux S.A, Marcy l’Etoile, France) was also
used for preliminary identification.
Assay of flocculating activity Unless otherwise stat-
ed, the flocculating activity was measured according to
the method of Kurane et al. (1) using a suspension of
kaolin clay as a test material with minor modifications.
Kaolin clay (300 mesh, Kishida Chemical, Osaka) was
suspended in deionized water at a concentration of 5000
mg/l (kaolin suspension). One milliliter of the sample
(supernatant of the culture broth or the crude biofloc-
culant in deionized water) was added to 10ml of the
kaolin suspension in a test tube, and the reaction mixture
was stirred for 30 s with a vortex mixer and then allowed
to stand for 5 min. The optical density of the upper
phase at 550 nm (A) was measured with a spectrophotom-
eter. A control experiment in which 1 ml of deionized
water instead of the sample was added to the suspension
was performed in the same manner, and the optical
density was measured (B). The flocculating activity was
defined as {(B-A)/B} x 100 (%). The activity was expres-
sed as the mean value from triplicate determinations.
For investigating the effect of pH on the flocculating
activity, the pH of the kaolin suspension was adjusted
using HCl or NaOH. The effect of temperature was
studied using the kaolin suspension and the sample, the
temperature of which was adjusted to a definite value.
As for investigate the effect of various cations, kaolin
suspensions were prepared in 50 mM Tris-HCI (pH 7.0),
and 0.1 ml of a cation solution was added to the reac-
tion mixtures; NaCl, KCl, CaCl*. 2Hz0, MgC&. 6Hz0,
FeC12, AlC& .6H20 or FeC13.6Hz0 solutions as the
PRODUCED FROM ACETATE AND PROPIONATE 41
sources of Na+, K+, Ca2+, Mg2+, Fez+, A13+ or Fe3+
respectively. The suspensions of diatomite (Kishida
Chemical, Osaka), bentonite (300 mesh, Kishida Chemi-
cal, Osaka), activated carbon (powder, Kishida Chemi-
cal, Osaka), cellulose powder (>300 mesh, Advantec
Toyo Ltd. Co., Tokyo), dry yeast (Asahi Brewery Ltd.
Co., Tokyo), soil and activated sludge were also used as
the test materials for the flocculating tests in the same
manner as kaolin. To prepare the soil suspension, a
garden soil sample was stirred in deionized water, al-
lowed to stand for 5 min, and the upper phase was ob-
tained. For the assay using activated sludge, measure-
ment of the optical density of the reaction mixture was
performed after allowing the mixture to stand for 1 min.
Recovery, purification and characterization of the bio-
flocculant The bioflocculant produced by strain TKF04
was recovered from its culture supernatant obtained by
removing the cells by centrifugation (21,000 X g, 10 min).
The culture supernatant was mixed with 2 volumes of
chilled ethanol and left overnight at 4°C followed by
recovery of the precipitate by centrifugation. The
resultant precipitate was dissolved in deionized water,
the pH of which was adjusted to approximately 2 with
HCl, and dialyzed against deionized or ultra-pure water
using a membrane Seamless Cellulose Tubing 27/32
(Sankojunyaku Ltd. Co., Tokyo). The crude biofloc-
culant was recovered by lyophilization of the dialyzed
solution.
Further purification was performed by chromato-
graphies in aqueous solution. The pH of the crude
bioflocculant in 10 mM phosphate buffer was adjusted to
5, and it was applied to a Gigapite amphoteric column
(Seikagaku Kogyo Co., Tokyo; 3.2cmx 18 cm). It was
eluted with the same buffer at a flow rate of 28 ml/h,
and the fractions exhibiting kaolin-flocculating activity
were collected. After concentration using polyethylene
glycol, it was further loaded to a Sephadex 200HR
column (Pharmacia Co., Sweden; 3.2 cm x 120 cm) and
eluted with the same buffer at flow rate of 8.2ml/h. The
active fractions were collected and lyophilized to obtain
the purified bioflocculant. The molecular weight of the
active fraction was estimated gel filtration using the same
the Sephadex 200HR column with thyroglobulin (bovine)
(669 kDa), ferritin (440 kDa), catalase (232 kDa), lactate
dehydrogenase (140 kDa) and bovine serum albumin
(67 kDa) as the standard proteins (Pharmacia Co.).
To ash, organic carbon, nitrogen and phosphorus con-
tents of the crude bioflocculant were analyzed according
to the “Standard Methods” (22). The total sugar was
assayed by the phenol sulfuric acid procedure using
glucose, xylose and glucuronic acid as the reference
substances (23). An anthrone method was used to detect
hexose in terms of glucose (24). Hexosamines were ana-
lyzed by the Elson-Morgan method using glucosamine as
the standard (25). Protein content was assayed by the
methods of Lowry (26) and Bradford (27). The purified
bioflocculant was further characterized by thin-layer
chromatography (TLC). TLC analysis of its hydrolysate
(in 2 N HCl at 100°C for 4 h) was performed using a
silica gel 60 G (Kishida Chemical) with isopropylalco-
ho1 : acetone : lactic acid (2 : 2 : 1, w/w) as the develop-
ing phase, and the monosaccharide(s) in the hydrolysate
was detected by spraying HzS04 reagent followed by
heating at 100°C for 5 min. The bioflocculant was also
characterized using a Fourier transform infrared spec-
trometer (FT-IR: FT-710, Hobira). Digestion of the
42 FUJITA ET AL. J. BIOSCI. BIOENG.,

TABLE 1. Effect of carbon source on the bioflocculant production

Carbon source Cell growth Floccula;%

activity
0
Acetic acid 95.6
Propionic acid in 98.4
Butyric acid -i N.A.
Lactic acid -I- 13.6
Oleic acid -7 18.5
Hexadecan i N.A.
0 20 40 60 80 100 0 20 40 60 80 100
Methanol -- N.A.
Ethanol N.A.
Incubation time (h)
Incubation time (b) Glucose t 12.9
Lactose t 37.0
FIG. 1. Effects of cultivation temperature on the cell growth (A)
and bioflocculant production @) by Cifrobacter sp. TKFO4. Symbols: -, No growth; +, OD 6M)=0.1-0.5; t+, OD,o=0.5-1.5; N.A., no
A, 20°C; w , 3o”c; 0, 37°C. activity (< 10%).

bioflocculant with chitinase (Sigma Chemical Co., St.
Louis, MO, USA), chitosanase (Sigma Chemical Co.) or
cellulase (Sigma Chemical Co.) was carried out at 25°C
for 2 h in a phosphate buffer. Control experiments were
performed under the same conditions without the addi-
tion of the enzymes. The analyses were conducted in tri-
plicate, and the average values were obtained.
RESULTS
Screening and identification of a bioflocculant-produc-
ing bacterium TKF04 Out of the 1564 bacterial
strains isolated from various environmental samples, 104
bacterial colonies could flocculate kaolin clay when
grown in AP medium. Among them, the bacterial strain
TKF04 isolated from a biofilm inside a kitchen drain
showed the highest flocculating activity against kaolin,
and is therefore used for further studies. Strain TKF04
was gram-negative, rod-shaped, motile, catalase-positive,
and oxidase-negative and produced acids from glucose
under anaerobic conditions. The identification using the
API20E kit classified this strain as Citrobacter freundii
with a 94.7% probability (identification code no.
3204572). The partial 16s rRNA sequences of this strain
(more than 600 bases each from both ends) were com-
pared to other sequences in the GenBank using BLAST,
and we found a strong but not complete homology to
those of C. freundii, too. Taking the results of other
physiological tests into consideration (data not shown),
strain TKF04 was identified as a Citrobacter sp.
Bioflocculant production by Citrobacter sp. strain
TJ.U?04 The flocculating activity of the culture broth
of strain TKF04 increased during the logarithmie phase
of growth in AP medium, thereafter the activity re-
(4 100
mained stable for about 1 d and showed a considerable
decrease afterwards (see courses in Figs. 1 and 2). When
the maximum cell growth was reached (about 1.0-1.5
g/Z), acetate in the medium was completely consumed
while some of propionate still remained. The flocculating
activity of the culture suspension was almost the same
s that of its cell-free supernatant, indicating that the
bioflocculant is extracellular.
To optimize the culture conditions for the biofloc-
culant production, the effects of temperature and initial
pH of the AP medium on the cell growth and the floc-
culating activity were investigated. The bioflocculant
production was found to considerably depend on the
culture temperature, and the optimum temperature was
found to be 3O”C, although that for the cell growth was
37°C (Fig. 1). On the other hand, the initial pH of the
medium showed little effect on the bioflocculant produc-
tion within the pH range of 7.2-10.0 in which strain
TKF04 could grow, while the flocculating activity was
considerably decreased during the late-log phase (Fig. 2).
Furthermore, the effects of the medium components on
the bioflocculant production were investigated. Since
yeast extract was found to be necessary for the biofloc-
culant production in a preliminary study, 0.1 g/l of yeast
extract was added to the test media throughout this
study. The effects of the carbon source are shown in
Table 1. Although strain TKF04 could grow on some of
the tested carbon sources including glucose and lactose,
only acetate and propionate were effective for the biofloc-
culant production. The effects of the nitrogen source
were also investigated when acetate or propionate was
used as a carbon source (Table 2). Ammonium as the
inorganic nitrogen source was found to be favorable for
TABLE 2. Effects of nitrogen source on the flocculant production
in AC. and Pr. media
Acetate (AC. *) Propionate (Pr.*)
Nitrogen source Cell Flocculating Cell Flocculating
growth activity growth activity
(!?A)
.I ",
CM) \, ",

NaC03 - 20.0 - 12.0

NH&l t 96.6 96.8
NH.,N03 + 96.8 - 96.9
Urea + 89.4 +- 95.2
0 20 40 60 80 100 0 20 40 60 80 100 86.2
Pepton t 93.8 t

Incubation time (h) Incubation time (h) Meat extract + 50.6 t 44.4
NH&O& tt 95.0 -t 97.6
FIG. 2. Effects of initial pH of the medium on the cell growth (A)
and bioflocculant production (B) by Citrobacter sp. strain TKF04. Acetate (AC.*): Acetate (7.Og/f) was added to the BM as the sole
Symbols: 0, pH 5.0; A, pH 6.0; W , pH 7.0; 0 , pH 7.2; A, pH 8.0; q , carbon source. Propionate (Pr.*): Propionate (3.0 g/l) was added to
pH 9.0; 0, pH 10.0. the BM as the sole carbon source.
VOL. 89, 2000 BIOFLOCCULANT PRODUCED FROM ACETATE AND PROPIONATE 43

0
0.01 0.1 1 10 100 1000
Concentration (mgll) o-

FIG. 3. Effects of concentration of the crude bioflocculant on the 0 20 40 60 80 100

flocculating activity. Symbols: 0, crude bioflocculant; A, PAA; 0, Temperature(%)
PAC.
FIG. 5. Effects of reaction temperature on the flocculating activ-
the bioflocculant production. The optimum concentra- ity. The crude bioflocculant was added to the reaction mixture at the
tion of (NH&SO4 as the inorganic nitrogen source was final concentration of 1 mg/l.
l.Og/i (data not shown). Organic nitrogen sources like
urea and peptone appeared to be able to replace ammo- The effects of pH of the reaction mixture on the
nium to a certain extent. kaolin-flocculating activity were investigated at the crude
Recovery of the crude bioflocculant Extracellular bioflocculant concentrations of 1 and lOmg/l (Fig. 4).
bioflocculant produced by strain TKF04 could be success- The flocculating activity was maintained at high levels in
fully recovered in a crude form by ethanol precipitation an acidic pH range (approximately 2-6 and 2-8 at the
and its dialysis (see Materials and Methods), whereas flocculant concentrations of 1 and lOmg/l, respectively),
other recovery methods including precipitation with however, it dropped with the increase of pH. The effects
ammonium sulfate, extraction with solvents and acid of reaction temperature were also studied, and it was
precipitation were not useful for the recovery of the found that the effective flocculation of kaolin occurred
bioflocculant. From 1 I of the cell-free culture of strain in the temperature range of 3-95°C (Fig. 5). The effects
TKF04, which was obtained under an optimized cultiva- of various cations on the flocculating activity are shown
tion conditions for the bioflocculant production (AP in Fig. 6. The flocculating activity was not enhanced by
medium, pH 7.2, 3O”C, 2 d), approximately 200mg of the addition of any cations including Ca2+. High concen-
the crude bioflocculant was recovered as a white powder. trations of cations led to a decrease of the flocculating
Properties of the flocculating activity of the crude bio- activity.
flocculant The flocculation properties of the biofloc- Furthermore, the flocculating activity of the crude
culant were examined using the crude sample obtained bioflocculant (10 mg/l) for a variety of inorganic and or-
by the above-mentioned method. The effect of the con- ganic suspended particles was investigated in comparison
centration of the crude bioflocculant on the kaolin-floc- with that of PAA (1 mg/[) and PAC (lOOmgIl) (Table
culating activity is shown in Fig. 3, compared with those 3). When added at lOmg/l, the crude bioflocculant
of PAC and PAA (average molecular weight; 9-10 mil- could efficiently flocculate all the tested materials except
lion, Kishida Chemical, Osaka). More than 90% of the for dry yeast. It was particularly effective for kaolin,
flocculating activity was observed at a concentration diatomite and soil particles.
ranging from 1 to lOmg/l. This activity is comparable Chemical properties of the bioflocculant Results of
to or a slightly lower than that of PAA and much higher the chemical analyses of the crude bioflocculant are sum-
than that of PAC. marized in Table 4. The ash content of the crude biofloc-
culant was 24%, indicating the incomplete removal of
inorganic salts by dialysis. The total sugar content was
10% as determined by the phenol sulfuric acid method,
and the hexose content determined by the anthrone
method was about 6%. On the other hand, hexosamine
was determined by the Elson-Morgan reaction as a
TABLE 3. Flocculating activity of the crude bioflocculant for
various suspended particles compared with those of PAC and PAA
Suspension Biofloccuiant PAC PAA
Diatomite 19.1 41.4 83.1
Cellulose powder 42.0 N.A. N.A.
Activated carbon powder 82.1 N.A. 88.6
Bentonite 96.1 N.A. N.A.
0 2 4 6 8 10 12 Kaolin 96.7 48.7 73.0
Soil 95.1 17.3 82.8
PH Dry yeast 22.5 N.A. 12.4
Activated sludge 74.0 64.5 N.A.
FIG. 4. Effects of pH on the flocculating activity. Symbols: crude
bioflocculant, 1 mg/l ( l ); 10 mg/l (0). N.A., No activity (< 10%).
44 FUJITA ET AL. J. BIOSCI. BIOENG.,

0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 loo 0.0001 0.001 0.01 0.1 1

cation concentrtion (mrq Cation concentration (mrvfl cation concentration (mM)

FIG. 6. Effects of addition of cations on the flocculating activity. Symbols: (A) monovalent cations: l , none; A, KCl; 1, NaCl; (B) divalent
cations: 0 , none; A, CaC&; W , MgC&; + , FeC12; (C) trivalent cations: 0 , none; A, AlClr; n , FeClr.

major component (approximately 30%). However, the
protein content could not be determined by either the
Lowry or the Bradford method due to occurrence of
precipitation during the analyses. The aqueous solution
did not show any peak of absorbance at 280 nm that is
specific for protein (data not shown).
On the gel filtration chromatograms, a single peak
showing kaolin-flocculating activity appeared between
the molecular weight markers of 232 and 440 kDa, and
the molecular weight of the bioflocculant was estimated
to be about 320 kDa (Fig. 7). The bioflocculant was fur-
ther purified by gel filtration, and its infrared radiation
(IR) spectrum was investigated. The IR spectrum of the
partially purified bioflocculant (Fig. 8) showed a trans-
mission pattern similar to that of chitin and chitosan,
both of which exhibit a hydroxyl band at 3450cmp1,
an amide band at 1655-1550cm-l, and an amine band
at 1630-1550 cm-l as characteristic bands (28). The TLC
analysis of the bioflocculant hydrolysate showed only
one obvious spot with an Rf value of 0.627. This Rf
value was almost the same as that of o-glucosamine
(R,=O.629) which was used as the reference chemical.
The hydrolyzates of authentic chitin and chitosan also
showed the spots with similar Rf values 0.635 and 0.633,
respectively. To confirm the presence of chitin-and/or
chitosan-like structure, the crude bioflocculant was treat-
ed with cellulase, chitinase and chitosanase, which result-
ed in the complete or considerable loss of the flocculat-
BF
61 kDa
440kDa
1 11 21 31 42
Fraction no.
FIG. 7. Estimation of molecular weight of the bioflocculant by
gel filtration. BF* indicates the bioflocculant which is evaluated by the
scale on the right side. Dotted line indicates the reference proteins
used which are evaluated by the scale on the left side.
ing activity against kaolin (Table 5).
Based on these results, it was verified that the sugar of
this bioflocculant was glucosamine.
DISCUSSION
To date, many studies on the microbial production of
flocculating substances have been reported from different
viewpoints. Previous-reports showed that, microbio-
logically-produced bioflocculants are generally high-
molecular-weight polymers, and have been identified or
presumed to be proteins (29, 30), glycoproteins (3), poly-
saccharides (16, 31), glycolipids (32), cellulose (33),
DNA (34) or complex hetero-polymers (35). For the
production of these biopolymer flocculants, sugars such
as glucose, fructose or sucrose (2, 4, 16, 17), casein, L-
glutamate or citrate are usually required as the main
substrate(s), leading to the problem of high production
costs. In an attempt to reduce the cost, Kurane et al. (13)
determined ethanol to be the cheapest substrate for the
bioflocculant production by Rhodococcus erythropolis
S-l, which was originally isolated as a bioflocculant-
producing strain using sugars. In this study, we pro-
posed the production of a bioflocculant using VFAs (acetic
and propionic acids), which can be easily produced from
various organic wastes to reduce the production cost, and
a bacterial strain we isolated, Citrobacter sp. TKFM,
which can utilize acetic or propionic acids as the sole
carbon sources for bioflocculant production. Within the
range of our reference search, this is the first report on
-I 0.4
bioflocculant production from VFAs.
I During the cultivation of strain TKFM in AP medi-
2 um, the flocculating activity increased in parallel with its
cell growth, indicating that bioflocculant was accumulat-
ed extracellularly in the medium during the phase when
it was actively growing. This suggests that the biofloc-
TABLE 4. Summary of the results of chemical analysis of the
crude bioflocculant
Ash 24.0%
Total sugar 10.0%
Hexose 5.7%
52 Hexosamine 29.4%
Protein N.D.
Elements
Carbon 31.4%
Nitrogen 5.0%
Phosphorus 1.6%
N.D., Protein content could not be determined.
VOL. 89, 2ooo BIOFLOCCULANT
4000 3000 2000 1000 doo.0
Wave number (cm-*)
FIG. 8. FT-IR analysisof the partially purified bioflocculant.
culant was not produced by cell autolysis but by biosyn-
thesis (36, 37). After the flocculating activity reached
a maximum, it remained unchanged for about 1 d, fol-
lowed by a considerable decline. Therefore, the biofloc-
culant seems to be degraded by TKF04 itself when the
cultivation is extended for a certain period.
It was clarified that the bioflocculant produced by
TKF04 from VFAs has a satisfactory level of flocculat-
ing activity. Based on the effective concentration of the
crude bioflocculant for kaolin-flocculation, its flocculat-
ing activity is comparable to or slightly lower than that
of PAA and much higher than that of PAC. When it
was added to the reaction mixture at 10 mg/l, efficient
kaolin removal could be performed in the pH range of
2-8 and temperature range of 3-95”C, and it could floc-
culate a variety of organic and inorganic particles includ-
ing activated sludge and soil. These results suggest that
the bioflocculant can be successfully applied for the
clarification of a very wide range of water/wastewaters
under various environmental conditions. Furthermore, it
has the merit that it does not require any addition of
cations such as Ca2+ to enhance stimulating the flocculat-
ing activity, thus it is cost-effective. On the whole, it can
be concluded that the bioflocculant exhibits flocculating
activity comparable or superior to that of the existing
synthetic flocculants.
The above-mentioned unique properties of the biofloc-
culant produced by TKF04 led us to partially character-
ize its chemical structure. By gel filtration, the molecular
weight of the bioflocculant was estimated to be more
than 232 kDa (about 320 kDa), indicating that it is a
high-molecular-weight biopolymer. It contained hexosa-
mine as a major component, and a neutral sugar was
also detected by its chemical analyses. On the other
hand, the protein content could not be determined. The
content of organic nitrogen (approximately 4%) could
be mostly accounted for by hexosamine. The FT-IR anal-
ysis of the bioflocculant showed a transmission pattern
similar to that of chitin and chitosan. The TLC analysis
tentatively identified the main component of the biofloc-
culant as glucosamine, which is also a major component
of chitin and chitosan. Furthermore, the flocculating
activity completely disappeared by chitinase treatments and
was considerably reduced by chitosanase and cellulase
treatments. All these results suggest that this biofloc-
culant has a structure similar to that of chitin or its
derivatives, e.g., chitosan, and that the flocculating activ-
ity may depend on the structure. The observation that
its flocculating activity decreased in an alkaline pH range
similar to that of chitosan may support this inference.
PRODUCED FROM ACETATE AND PROPIONATE 45
TABLE 5. Residual flocculating activity of the purified
bioflocculant after enzyme digestion
Reaction time (h)
Enzyme 2h
treatment
Cellulase Chitinase Chitosanase
Before 97.4% 94.5% 94.8%
After 53.7% 7.8% 38.1%
Control 97.2% 94.4% 94.2%
To completely identify the chemical structure of the
bioflocculant, further detailed studies are required.
It has been reported that chitosan is produced by
fungi such as Mucor rouxii (38), Absidia butleri (39),
Abisidia coerulea (40), Abisidia atrospora and Gon-
gronella butleri (28). However, the production of chito-
san or other chitin high-molecular-weight chitin derivates
by bacteria has not been reported within the range of
our reference search, although low-molecular-weight
chitin oligosaccharides production by Rhizobium legumi-
nosarum has been reported (41). Thus, this is the first
report of the bacterial production of a biopolymer floc-
culant which has a structure similar to chitin or its deriva-
tives .
ACKNOWLEDGMENTS
The present study was supported by the Proposal-Based Ad-
vanced Industrial Technology R and D Program from New Energy
and Industrial Development Organization (NEDO) of Japan. This
research was also supported in part by the Japan Waste Research
Foundation. We are grateful to Dr. Masaaki Morikawa, Depart-
ment of Material and Life Sciences, Osaka University, for his great
help in the 16s rRNA sequencing.
REFERENCES
1. Kurane, R., Takeda, K., and Suzuki, T.: Screening for and
characteristics of microbial flocculants. Agric. Biol. Chem., 50,
2301-2307 (1986).
2. Kurane, R., Toeda, K., Takeda, K., and Suzuki, T.: Culture
conditions for production of microbial flocculant by Rhodococ-
cus erythropolis. Agric. Biol. Chem., 50, 2309-2313 (1986).
3. Nakamura, J., Miyashiro, S., and Hirose, Y.: Screening, isola-
tion, and some properties of microbial cell flocculants. Agric.
Biol. Chem., 40, 377-383 (1976).
4. Takagi, H. and Kodowaki, K.: Floccculant production by
Paecilomyces sp. taxonomic studies and culture conditions for
production. A&c. Biol. Chem., 49, 3151-31.57 (1985).
5. KowaB. N. W.. Pendlebum. W. W.. Kessler. J. B.. Perl. D. P..
- , , I I
and Beal, M. F.: Aluminium-induced neurofibrillary degenera-
tion affects a subset of neurons in rabbit cerebral cortex, basal
forebrain and upper brainstem. Neuroscience, 29, 329-337
(1989).
6. Kruck, T. P. A. and McLachlan, D. R. C.: Aluminum as a
pathogenic factor in senile dementia of the alzheimer type: ion
specific chelation. Prog. Clin. Biol. Res., 317, 1155-1167
(1989).
7. Master, C. L., Multhaup, G., Simms, G., Pottgiesser, J., Mar-
tins, R. N., and Beyreuther, K.: Neuronal origin of a cerebral
amyloid: neurofibrillary tangles of Alzheimer’s disease contain
the same protein as the amyloid of plaque cores and Blood
vessels. EMBO J., 4, 2757-2763 (1985).
8. Pullen, R. G. L., Candy, J. M., Morris, C. M., Taylor, G.,
Keith, A. B., and Edwardson, J. A.: Gallium-67 as a potential
marker for aluminum transport in rat brain: implifications for
Alzheimer’s disease. J. Neurochem., 55, 251-259 (1990).
9. Ndabigengesers, A. and Subba Narasiah, K.: Quality of water
treated by coagulation using Moringa oleifera seeds. Wat. Res.,
32, 781-791 (1998).
10. DearlIeld, K. L., Albernatby, C. O., Ottley, M. S., Brantner,
46 FUJITA ET AL.
J. H., and Hayes, P. F.: Acrylamide: its metabolism, develop-
mental and reproductive effects, genotoxicity, and carcinogenic-
ity. Mut. Res., 195, 45-77 (1988).
11. Vanhorick, M. and Moens, W.: Carcinogen-mediated induc-
tion of SV40 DNA amplification is enhanced by acrylamide
in Chinese hamster CO60 cells. Carcinogenesis, 4, 1459-1463
(1983).
12. Lee, S. H., Lee, S. O., Jang, I(. L., and Lee, T. H.: Microbial
flocculant from Arcuudendron sp. TS-49. Biotech. Lett., 17,
95-199 (1995).
13. Kurane, R., Hatamochi, K., Kakuno, T., Kiyohara, M., Hirano,
M., and Taniguchi, Y.: Production of a bioflocculant by
Rhodococcus erythropolis S-l grown on alcohols. Biosci. Bio-
tech. Biochem., 58, 428-429 (1994).
14. Yokoi, H., Shiraki, M., Hirose, J., Hayashi, S., and Takasaki,
Y.: Flocculation properties of xanthan produced by Xanthomo-
nas cumpestris. Biotechnol. Techniq., 10, 789-792 (1996).
15. Nam, J. S., Kwon, G. S., Lee, S. O., Hwang, J. S., Lee, J. D.,
Yoon, B. D., and Lee, T. H.: Bioflocculant produced by Asper-
gillus sp. JS-42. Biosci. Biotech. Biochem., 60, 325-327 (1996).
16. Kurane, R. and Nohata, Y.: Microbial flocculation of waste
liquids and oil emulsion by a bioflocculant from Alcaligenes
latus. Agric. Biol. Chem., 55, 1127-1129 (1991).
17. Toeda, K. and Kurane, R.: Microbial flocculant from Al-
caligenes cupidus KT201. Agric. Biol. Chem., 55, 2793-2799
(1991).
18. Jones, P. H. and Szekely, J.: Studies of the production of vola-
tile fatty acids in the anaerobic digestion of municipal sludge.
Water Pollut. Res. J. Canada, 21, 58-70 (1986).
19. Hasegawa, T.: Biseibutsu no bunrui to doutei. Gakkai Shup-
pan Center, Tokyo (1985).
20. Peter, H. A. S., Nicholas, S. M., Sharpe, M. E., and Holt,
J. G.: Bergey’s manual of systematic bacteriology, vol. 1. Wil-
liams and Wilkins Co., Baltimore (1986).
21. Rocelle, P. A., Will, J. A. I(., Fry, J. C., Jenkins, G. J. S.,
Parkes, R. J., Turley, C. M., and Weightman, A. J.: Extrac-
tion and amplification of 16s rRNA genes from deep marine
sediments and seawater to assess bacterial community diversity,
p. 219-239. Nucleic acids in the environment. Springer,
Berlin (1995).
22. American Public Health Association: Standard methods for
the examination of water and wastewater, 16th edn. American
Public Health Association, American Water Works Association
and Water Pollution Control Federation, Washington, D.C.
(1985).
23. Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and
Smith, F.: Calorimetric method for determination of sugars
and related substances. Anal. Chem., 28, 350-356 (1956).
24. Honda, S., Kakei, K., and Okada, K.: Convenient method for
the gas chromatographic analysis of hexosamines in the
presence of neutral monosaccharides and uranic acids. J. Chro-
matogr., 176, 367-373 (1979).
25. Lewy, G. A. and McABen, A.: The N-acetylation and estima-
tion of hexosamines. Biochem. J., 73, 127-132 (1959).
J. BIOSCI. BIOENC;.,
26. Lowery, 0. H., Rosebrough, N. J., Farr, A. L., and Randall,
R. J.: Protein measurement with the Folin phenol reagent. J.
Biol. Chem., 193, 265-275 (1951).
27. Bradford, M. M.: A rapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem., 72, 248-254 (1976).
28. Yokoi, H., Aratake, T., Hirose, J., Hayashi, S., and Takasaki,
Y.: Chitosan production from shochu distillery wastewater by
Funguses. J. Ferment. Bioeng., 85, 246-249 (1998).
29. Takeda, M., Kurane, R., Koizumi, J., and Nakamura, I.: A
protein bioflocculant produced by Rhodococcus erythropolis.
Argric. Biol. Chem., 55, 2663-2664 (1991).
30. Takeda, M., Koizumi, J., Matsuoka, H., and Hikuma, M.:
Factors affecting the activity of a protein bioflocculant pro-
duced by Nocardia amarae. J. Ferment. Bioeng., 74, 408-409
(1992).
31. Toeda, K. and Kurane, R.: Microbial flocculant from AI-
caligenes cupidus KT201. Agric. Biol. Chem., 55, 2793-2799
(1991).
32. Kurane, R., Hatamochi, K., Kakuno, T., Kiyohara, M.,
Kawaguchi, K., Mizuno, Y., Hirano, M., and Tanignchi, Y .:
Purification and characterization of lipid bioflocculant pro-
duced by Rhodococcus erythropolis. Biosci. Biotech. Biochem.,
58, 1977-1982 (1994).
33. Napoli, C., Dazzo, F., and Hubbell, D.: Production of cellu-
lose microfibrils by Rhizobium. Appl. Microbial., 30, 123-131
(1975).
34. Sakka, K. and Takahashi, H.: DNA as a flocculation factor in
Pseudomonas sp. Agric. Biol. Chem., 45,2869-2876 (1981).
35. Nakamura, J., Miyashiro, S., and Hirose, Y.: Purification and
chemical analysis of microbial cell flocculant produced by
Aspergillus sojae AJ7002. Agric. Biol. Chem., 40, 619-624
(1976).
36. Kurane, R., Takeda, K., and Suzuki, T.: Screening for and
characteristics of microbial flocculants. Agric. Biol. Chem., 50,
2301-2307 (1986).
37. Nakamura, J., Miyashiro, S., and Hirose, Y.: Conditions for
production of microbial cell flocculant by Aspergillus sojae
AJ7002. Agric. Biol. Chem., 40, 1341-1347 (1976).
38. White, S. A., Farina, P.R., and Fultoa, I.: Production and
isolation of chitosan from Mucor rouxii. Appl. Environ.
Microbial., 38, 323-328 (1979).
39. Kobayashi, T., Takiguchi, Y., Shimahara, K., and Sannan, T.:
Distribution of chitosan in Absidia strains and some properties
of the chitosan isolated. Nippon Nogeikagaku Kaishi, 62,
1463-1469 (1988). (in Japanese)
40. McGahren, W. J., Perkinson, G. A., Growich, J. A., Leese,
R. A., and Ellestad, G. A.: Chitosan by fermentation. Proc.
Biochem., 19, 88-90 (1984).
41. Spaink, H.P., Wiifjes, A.H.M., van der Drift, K.M.G.M.,
Haverkamp, J., Thomas-Oates, J. E., and Lugteaberg, B. J. J.:
Structure identification of metabolites produced by the NodB
and NodC proteins of Rhizobium leguminisarum. Mol. Micro-
biol., 13, 821-831 (1994).