Fermn 2

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Bioresource Technology 100 (2009) 4262–4270

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Lactic acid production by Lactobacillus sp. RKY2 in a cell-recycle continuous


fermentation using lignocellulosic hydrolyzates as inexpensive raw materials
Young-Jung Wee a, Hwa-Won Ryu b,*
a
Department of Food Science and Technology, College of Natural Resources, Yeungnam University, Gyeongsan, Gyeongbuk 712-749, Republic of Korea
b
School of Biological Sciences and Technology, Chonnam National University, Gwangju 500-757, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Continuous lactic acid fermentations were conducted using lignocellulosic hydrolyzates and corn steep
Received 16 January 2009 liquor as inexpensive raw materials. Lactic acid concentrations decreased with increases in the dilution
Received in revised form 27 March 2009 rate, whereas the residual substrate concentrations increased. However, lactic acid yields were main-
Accepted 27 March 2009
tained at more than 0.90 g g1 over all cases experimented. The cell-recycle cultivation system exerted
Available online 24 April 2009
positive effects on fermentation efficiency, including volumetric productivity, which is attributable to
the retention of cells in the bioreactor. The cell-recycle continuous fermentation of lignocellulosic
Keywords:
hydrolyzates yielded a lactic acid productivity of 6.7 g l1 h1 for a dilution rate of 0.16 h1 using
Lactic acid
Lactobacillus
30 g l1 of corn steep liquor and 1.5 g l1 of yeast extract as nutrients. The productivity (6.7 g l1 h1)
Lignocellulosic hydrolyzates acquired by the cell-recycle continuous fermentation of lignocellulosic hydrolyzates was 1.6 times higher
Cell-recycle than the lactic acid productivity yielded in the continuous fermentation without cell-recycle system.
Continuous fermentation Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction of DL-lactic acid, but the fermentative route from renewable re-
sources results in optically active L- or L-lactic acid. However, D-lac-
Lactic acid is commonly recognized as one of the most versatile tic acid is known occasionally to be harmful to human metabolism,
organic acids, with a long history of usage for the preservation of and thus frequently causes acidosis and decalcification (Datta et al.,
foodstuffs. It has a broad range of applications in the food, textile, 1995; Wee et al., 2006a). Additionally, the fermentative route to
pharmaceutical, cosmetic, and chemical industries (Davison et al., lactic acid has become the subject of a great deal of attention from
1995; VickRoy, 1985). As lactic acid has two reactive functional researchers of chemical synthesis, due to the need for optically
groups, carboxyl (–COOH) and hydroxyl (–OH) groups, it may pure lactic acid for the synthesis of highly crystalline PLA
potentially be employed as a fundamental chemical for the (Södergård and Stolt, 2002). The limited nature of available petro-
production of several industrial chemicals, including ethyl lactate, leum resources and recent increases in the price of crude oil price
propylene glycol, 2,3-pentandione, propanoic acid, acrylic acid, should be expected to render the fermentative route to lactic acid
acetaldehyde, and dilactide (Varadarajan and Miller, 1999). Re- increasingly competitive.
cently, lactic acid consumption has increased significantly due to The raw material cost for the fermentative production of lactic
its applications in the polymer industry as a feedstock monomer acid usually accounts for 68% of total manufacturing cost (Tejayadi
for the manufacture of poly(lactic acid) (PLA), a well-known sus- and Cheryan, 1995). Åkerberg and Zacchi (2000) previously re-
tainable and compostable biopolymer (Datta et al., 1995; Litch- ported that the cost of raw materials for lactic acid fermentation
field, 1996). The global consumption of lactic acid has been account for more than 34% of total production cost. Thus, the
roughly estimated as 130,000–150,000 metric tons per year, and raw materials required for industrial lactic acid production have
this is expected to increase considerably in the near future, as to meet several requirements, including low cost, low levels of con-
the result of a rapid expansion in lactic acid-derived biodegradable taminants, rapid fermentation rate, high lactic acid yields, little or
polymer markets (Wee et al., 2006a). no byproduct formation, and consistent availability (Ryu et al.,
There are two routes by which lactic acid can be produced, 2003). Moreover, it is necessary to minimize the production cost
microbial fermentation and chemical synthesis. Chemical synthe- in order to satisfy the current market demand on lactic acid (Altaf
sis from petroleum resources consistently yields a racemic mixture et al., 2007). Lignocellulosic resources, such as wood and crop res-
idues, are generally considered to represent an attractive and inex-
pensive raw material for the microbial production of lactic acid, as
* Corresponding author. Tel.: +82 62 530 1842; fax: +82 62 530 1909. they are renewable, widely distributed, abundant, and cheap (John
E-mail address: hwryu@chonnam.ac.kr (H.-W. Ryu). et al., 2007; Ryu et al., 2003; Wee et al., 2006a).

0960-8524/$ - see front matter Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2009.03.074
Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270 4263

Lactic acid production depends strictly on cell growth, as fer-


mentation is associated with growth. However, the batch lactic
acid fermentation technique suffers from an end-product inhibi-
tion, as the lactic acid produced during fermentation inhibits cell
growth (Xu et al., 2006). In continuous cultures, cells can be
maintained at a constant physiological state and growth rate.
Membrane cell-recycle systems coupled with the repeated-batch
and continuous cultures have proven to be efficient in lactic acid
production (Gao et al., 2005; Kim et al., 2006; Kwon et al., 2001;
Tejayadi and Cheryan, 1995; Wee et al., 2006b). However, to the
best of our knowledge, there have been few reports thus far
regarding membrane cell-recycle fermentation techniques for
the continuous production of lactic acid from lignocellulosic
hydrolyzates and corn steep liquor as the primary nutrient
sources.
The principle objective of the current study was to continuously
produce lactic acid from lignocellulosic hydrolyzates and corn
steep liquor using Lactobacillus sp. RKY2 in a membrane cell-re-
cycle bioreactor system. In service of this objective, the continuous
culture systems with and without the membrane cell-recycle were
evaluated for the production of lactic acid using lignocellulosic
hydrolyzates and glucose as a carbon source, specifically in terms
of the volumetric productivities and the lactic acid yields. Fig. 1. Schematic diagram of the procedures for the preparation of lignocellulosic
hydrolyzates utilized in this study.

2. Methods 2.3. Microorganism and inoculum preparation

2.1. Lignocellulosic feedstock and pretreatment Lactobacillus sp. RKY2 KCTC 10353BP was used throughout this
study as a homofermentative lactic acid bacterium (Kim et al.,
The oak wood chips used in this study as a lignocellulosic feed- 2006; Wee et al., 2004b, 2005). Stock cultures were maintained
stock were kindly provided by the Korea Institute of Energy Re- at 20 °C in the pre-culture media with 20% (v/v) glycerol.
search (KIER, Daejeon, Korea). The raw material composition was The cells (4%, v/v) from the stock cultures were transferred to
determined according to the method of Lu et al. (2009), and the 15 ml of sterile pre-culture media in a 20 ml serum bottle, and then
oak wood utilized in this study was composed of 49.3% (w/w) cel- incubated for 24 h at 36 °C and 200 rpm on a rotary shaker (KMC-
lulose, 25.9% (w/w) hemicellulose, and 21.7% (w/w) Klasson lignin. 8480SF; Vision Scientific Co., Daejon, Korea). After three or four
The wood chips (2  4 mm) were mixed with 0.5% (w/w) H2SO4, successive propagation steps, 3 ml of the final culture was trans-
and then incubated overnight at room temperature. A solid con- ferred to 40 ml of pre-culture media in a 50 ml serum bottle. This
centration of the preparation was adjusted to approximately 33% inoculum was then incubated at 36 °C for 12 h on the rotary shaker
(w/w). The impregnated wood chips were then charged into a prior to inoculation into the bioreactor (4%, v/v).
preheated 8-l exploder and the reaction was initiated by direct
injection of saturated steam. The exploder was maintained at a 2.4. Culture media
temperature of 215 °C for 5 min, followed by a rapid decompres-
sion and discharge of the wood material into a reservoir. The pre-culture medium for the inoculum preparation con-
This was finally utilized as a raw material for enzymatic tained the following ingredients (per liter): 30 g of glucose, 10 g
saccharification. of yeast extract, 2 g of (NH4)2HPO4, and 0.1 g of MnSO4. The initial
pH was adjusted to 6.0 prior to sterilization. The medium for con-
2.2. Enzymatic saccharification for the preparation of lignocellulosic tinuous fermentation was comprised of the following components
hydrolyzates (per liter): glucose or wood hydrolyzate, 30 g of corn steep liquor,
and 1.5 g of yeast extract. All of the components in the media for
The solid residue obtained after the steam explosion was sub- the continuous fermentations, except for glucose and wood
ject to enzymatic saccharification. Hydrolysis was conducted in a hydrolyzates, were identical for both the continuous fermentations
2-l Erlenmeyer flask using two commercial enzymes, Celluclast with and without cell-recycle. The quantity of glucose in both cases
1.5 L (cellulase; Novo Nordisk A/S, Bagsvrd, Denmark) and Nov- was adjusted to 50 g l1 and 75 g l1.
ozym 188 (b-glucosidase; Novo Nordisk A/S). Celluclast 1.5 L
(20 IU per g-residue) was added to a mixture of the solid residue 2.5. Ultrafiltration membrane bioreactor for cell-recycle cultivation
(525 g dry mass) and deionized water (1.2 l). The mixture was then
supplemented with Novozym 188 at a concentration of 30 IU per A 2.5-l bioreactor system (KF-2.5L; Kobiotech, Incheon, Korea)
g-residue. Enzymatic hydrolysis was carried out for 48 h at 50 °C was employed, and the working volume in the bioreactor was
and pH 4.8. The resultant hydrolyzates were then isolated by cen- maintained at 1 l. In the membrane cell-recycle experiments, a
trifugation at 15,000 g, and the supernatant was utilized as a med- polysulfone ultrafiltration (UF) module (SKUF-103-0830,
ium component. The quantity of glucose present in the resultant 300 mm  25 mm; SK Chemical, Suwon, Korea) containing 100
hydrolyzates was estimated as 88 ± 0.6 g l1, which was subject hollow-fiber membranes was coupled to the bioreactor. The
to dilution with an appropriate quantity of deionized water to ad- molecular weight cut-off (MWCO) of the membranes was 30 kDa,
just the sample to the desired sugar concentrations. A schematic and the effective surface area was 0.06 m2. The UF membranes
diagram of the entire process for the preparation of the wood were packed into an acrylic resin holder, which was equipped with
hydrolyzates is provided in Fig. 1. pressure gauges at the inlet and outlets. A Masterflex pump system
4264 Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270

(Cole-Parmer Instrument, Vernon Hills, IL, USA) equipped with a filtration module. The working volume was maintained at 1 l by
PTFE diaphragm pump head was utilized to supply the driving supplying the fresh media in balance with the permeate flow from
force for transmembrane flux, and a three-channel peristaltic the membranes and by pumping out the excess volume in the bio-
pump (Teledyne Isco, Lincoln, NE, USA) was employed to feed reactor. Throughout the experiments, the cultured samples were
and remove the culture broth. taken from the reactor every 6 h, for the measurement of cell
Once one set of experiments was completed, the membrane growth and the concentrations of glucose and lactic acid. The stea-
module was cleaned. The cleaning procedure was conducted using dy state in the continuous culture was considered to reach after
a mixture of 0.1 M NaOH and 200 ppm NaOCl, prepared with being replaced by at least three volumes equal to that in the biore-
deionized water. During the cleaning operation, the cleaning solu- actor and when three successive measurements of the concentra-
tion was replaced by a newly prepared one whenever it turned tur- tion of cells, glucose, and lactic acid were similar with a
bid. After the cleaning solution clarified, the sterile deionized water maximum deviation of 5%. The conventional continuous fermenta-
was used to flush the membrane system, in order to remove resid- tion was conducted in a similar manner, without coupling the UF
ual NaOH and NaOCl solution remaining in the membrane module. membrane module to the bioreactor.
After cleaning, the membrane filtration system was again coupled
to the bioreactor, and the UF membrane bioreactor was operated 2.7. Analytical methods
again for further experiments.
The density of the cells was monitored through optical density
2.6. Fermentation measurements at 660 nm using a Shimadzu UV-1700 spectropho-
tometer (Kyoto, Japan). For the determination of dry weight, the
The temperature of the culture broth was maintained at 36 °C, cells were centrifuged at 13,000 g for 20 min using a bench-top
the agitation speed was maintained at 200 rpm, and the pH was centrifuge (Vision Scientific Co.), washed twice with deionized
automatically adjusted at 6.0 by the addition of 10 M NaOH solu- water, and dried at 105 °C until a constant weight was achieved
tion. After 12 h of batch fermentation, the overall system was (24 h). The cell density was then converted to dry weight (g l1)
switched to continuous culture mode by operating the membrane using an appropriate calibration curve.

Fig. 2. Effect of dilution rates on lactic acid concentration, residual glucose, dry cell weight, and productivity in glucose media (a, 50 g l1 glucose; b, 75 g l1 glucose) by the
conventional continuous culture of Lactobacillus sp. RKY2 (-d-, lactic acid; -s-, residual glucose; -j-, dry cell weight; -h-, productivity).
Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270 4265

Lactic acid concentration was analyzed using a high perfor- 3. Results


mance liquid chromatography (HPLC) system (Young Lin
Instruments, Anyang, Korea) with an Aminex HPX-87H ion-exclu- 3.1. Continuous production of lactic acid using glucose
sion column (300  7.8 mm; Bio-Rad, Hercules, CA, USA) and a
UV detector adjusted to 210 nm. The injection volume of the sam- Continuous fermentations using glucose were conducted in or-
ple was 20 ll. The column, which was maintained at 35 °C, was der to compare them with those using the lignocellulosic hydroly-
then eluted with 5 mM H2SO4 solution at a flow rate of zate, and the conventional continuous fermentations without cell-
0.6 ml min1 for 15 min. The lactic acid retention time under these recycle system were also conducted for comparison with the mem-
conditions was 12.78 ± 0.06 min. brane cell-recycle continuous culture system. The conventional
Glucose concentration was enzymatically analyzed through a continuous culture fermentation using 50 g l1 glucose as constant
glucose oxidase–peroxidase method using an assay kit (Asan Phar- feed sugar content was conducted at dilution rates of between 0.04
maceutical, Seoul, Korea). The red colored dye released after two and 0.28 h1. The concentration of dry cells, residual glucose, lactic
serial reactions catalyzed by glucose oxidase and peroxidase was acid, and productivity as a function of dilution rate are shown in
determined via absorbance measurements at 505 nm. Fig. 2a. The highest cell concentration (4.1 g l1) was achieved at
Each of the assays was performed in triplicate and mean values the lowest dilution rate (0.04 h1). The concentration of cells and
were presented. Dilution rate was defined as the ratio of the feed lactic acid decreased to 1.6 g l1 and 12.4 g l1, respectively, with
flow rate and the culture volume. Lactic acid yield and substrate increasing dilution rate. The concentration of residual glucose in-
conversion were calculated by the following equation, respectively. creased to 36.3 g l1 with increasing dilution rate. However, the
dilution rates evaluated here did not exert significant effects on
Lactic acid producedðgÞ the lactic acid yields, and an average value was achieved of more
Lactic acid yield ðg=gÞ ¼
Glucose consumedðgÞ than 0.90 g g1. Lactic acid productivity increased with increasing
  dilution rate, achieved a maximum level of 3.9 g l1 h1 at a dilu-
Residual glucoseðgÞ tion rate of 0.20 h1, and decreased with further increases in the
Substrate conversion ð%Þ ¼ 1  100
Initial glucoseðgÞ dilution rate.

Fig. 3. Effect of dilution rates on lactic acid concentration, residual glucose, dry cell weight, and productivity in glucose media (a, 50 g l1 glucose; b, 75 g l1 glucose) by the
cell-recycle continuous culture of Lactobacillus sp. RKY2 (-d-, lactic acid; -s-, residual glucose; -j-, dry cell weight; -h-, productivity).
4266 Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270

Fig. 2b shows the effects of dilution rate on dry cell weight, cose concentrations continuously increased. Dilution rates in ex-
residual glucose, lactic acid, and productivity under conditions of cess of 0.16 h1 also resulted in a continuous reduction in the
conventional continuous fermentation using 75 g l1 glucose as lactic acid concentrations. The cell concentrations continuously
constant feed sugar content. The dilution rate was increased from increased and the highest dry cell weight (24.4 g l1) was
0.04 to 0.24 h1. At the higher dilution rate of 0.24 h1, residual achieved at a dilution rate of 0.36 h1. The yields of lactic acid
glucose level had reached 64.4 g l1 and considerable reductions were not significantly influenced by the dilution rates and values
in cell concentration (1.7 g l1) and lactic acid (10.3 g l1) were ob- of more than 0.93 g g1 were achieved in all of the cases
served. The highest volumetric productivity of lactic acid investigated.
(5.2 g l1 h1) was obtained at a dilution rate of 0.12 h1. For all Fig. 3b shows the effects of dilution rate on dry cell weight,
the dilution rates studied, the yields of lactic acid were in excess residual glucose, lactic acid, and productivity when the membrane
of 0.96 g g1. cell-recycle continuous culture fermentation was conducted using
Membrane cell-recycle continuous culture fermentation using 75 g l1 glucose as the constant feed sugar content at a dilution
50 g l1 glucose as the feed sugar content was conducted at dilu- rate from 0.04 to 0.24 h1. The residual glucose concentrations in-
tion rates between 0.04 and 0.36 h1. Fig. 3a shows the influences creased with increases in the dilution rate, but the lactic acid con-
of dilution rate on dry cell weight, residual glucose, lactic acid, centrations decreased. In this case, the cell-recycle system yielded
and productivity. The maximum lactic acid concentration the maximum dry cell weight (28.3 g l1) at the highest dilution
(48.1 g l1) was achieved at the lowest dilution rate utilized in rate (0.24 h1). The volumetric productivity of lactic acid increased
this set of experiments (0.04 h1). In this case, however, a low vol- with increases in the dilution rate up to 0.16 h1 and then slightly
umetric productivity of lactic acid (1.9 g l1 h1) was observed. decreased at higher dilution rates. Therefore, the highest lactic acid
The highest level of lactic acid productivity (11.4 g l1 h1) was productivity (8.8 g l1 h1) was achieved with a dilution rate of
achieved with a dilution rate of 0.28 h1. With dilution rates be- 0.16 h1. As was determined in the previous investigations, the lac-
tween 0.04 and 0.16 h1 the feed glucose was almost completely tic acid yields were maintained at greater than 0.90 g g1 at all the
consumed, but at dilution rates beyond 0.16 h1 the residual glu- dilution rates studied.

Fig. 4. Effect of dilution rates on lactic acid concentration, residual glucose, dry cell weight, and productivity in wood hydrolyzate media (a, 50 g l1 glucose; b, 75 g l1
glucose) by the conventional continuous culture of Lactobacillus sp. RKY2 (-d-, lactic acid; -s-, residual glucose; -j-, dry cell weight; -h-, productivity).
Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270 4267

3.2. Continuous production of lactic acid using lignocellulosic parameter reached 1.4 and 6.9 g l1, respectively. The volumetric
hydrolyzates productivity of lactic acid increased from 2.6 to 4.3 g l1 h1 in a
dilution rate range of 0.04–0.12 h1, but decreased when the dilu-
The conventional continuous culture fermentation without cell- tion rate exceeded 0.12 h1. It dropped to 1.7 g l1 h1 at the high-
recycle system using lignocellulosic hydrolyzates containing est dilution rate (0.24 h1). Lactic acid yields in excess of 0.91 g g1
50 g l1 glucose as the constant feed sugar content was conducted were achieved at all the experimental dilution rates.
at dilution rates between 0.04 and 0.28 h1. The effects of dilution The effects of dilution rate on dry cell weight, residual glucose,
rate on residual glucose, lactic acid production, cell growth, and lactic acid, and productivity when the membrane cell-recycle con-
productivity are demonstrated in Fig. 4a. The cell concentrations tinuous culture fermentation using lignocellulosic hydrolyzates
decreased with increases in the dilution rates, and the maximum containing 50 g l1 glucose as the constant feed sugar content
dry cell weight (3.5 g l1) was achieved at a dilution rate of was conducted in a dilution rate range between 0.04 and
0.04 h1. The residual glucose concentration increased with 0.36 h1 are provided in Fig. 5a. The residual glucose concentra-
increasing dilution rates, whereas the lactic acid concentration tions increased with increases in the dilution rate, but the lactic
was reduced. In this case, the highest volumetric productivity of acid concentration was reduced. In this case, the cell-recycle sys-
lactic acid (3.1 g l1 h1) was achieved at a dilution rate of tem yielded the maximum dry cell weight (21.4 g l1) at the high-
0.16 h1, and the yields of lactic acid ranged between 0.90 and est dilution rate (0.36 h1). The volumetric productivity of lactic
0.96 g g1 on the basis of the consumed glucose. acid increased from 1.7 to 6.3 g l1 h1 with dilution rates in the
Fig. 4b shows the effects of dilution rate on dry cell weight, range of 0.04–0.24 h1, but then decreased when the dilution rate
residual glucose, lactic acid production, and productivity when exceeded 0.24 h1. It eventually dropped to 4.0 g l1 h1 at the
the conventional continuous culture fermentation was conducted highest dilution rate (0.36 h1). Under these conditions, the lactic
using wood hydrolyzates containing 75 g l1 glucose as the con- acid yields ranged from 0.91 to 0.98 g g1.
stant feed sugar content at a dilution rate between 0.04 and The membrane cell-recycle continuous cultivation was con-
0.24 h1. Both the dry cell weight and lactic acid were reduced ducted using lignocellulosic hydrolyzates containing 75 g l1 glu-
with increases in the dilution rate, and the lowest values of each cose as the constant feed substrate content, and the results are

Fig. 5. Effect of dilution rate on lactic acid concentration, residual glucose, dry cell weight, and productivity in wood hydrolyzate media (a, 50 g l1 glucose; b, 75 g l1
glucose) by the cell-recycle continuous culture of Lactobacillus sp. RKY2 (-d-, lactic acid; -s-, residual glucose; -j-, dry cell weight; -h-, productivity).
4268 Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270

shown in Fig. 5b. Lactic acid concentrations changed from 69.4 to The residual glucose concentrations decreased with increasing
21.2 g l1, with a gradient of dilution rates from 0.02 to 0.24 h1, dilution rates at all initial glucose or wood hydrolyzate concentra-
respectively. The residual glucose concentration increased from tions. Higher substrate concentrations resulted in higher residual
2.1 to 52.0 g l1 with increases in the dilution rate from 0.02 to glucose concentrations, and this is likely attributable to osmotic ef-
0.24 h1. The cell concentration increased from 6.2 to 19.0 g l1 fects on the microorganisms (Roukas and Kotzekidou, 1998). These
as the dilution rate increased from 0.02 to 0.24 h1. The volumetric results are generally consistent with the findings of Bustos et al.
productivity of lactic acid reached a maximum (6.7 g l1 h1) at a (2007), who studied the continuous production of lactic acid and
dilution rate of 0.16 h1. The yields of lactic acid from the con- surfactants using hemicellulosic trimming vine shoot hydrolyzates.
sumed glucose remained almost constant at dilution rates in ex- They reported that the residual xylose concentrations increased
cess of 0.91 g g1. continuously with increasing dilution rates because the fermenta-
tion broth had insufficient time to ferment. Similar results were
previously reported by González-Pajuelo et al. (2005) and Shene
4. Discussion and Bravo (2007). The cell-recycle system utilized in this study re-
sulted in reductions in the residual glucose content for all initial
Lactic acid is a commercially viable organic chemical with a substrate concentrations and dilution rates. For example, as is
broad range of applications, but its production costs still need to shown in Figs. 4b and 5b, the residual glucose concentration at a
be reduced, as the cost of raw materials for lactic acid production dilution rate of 0.16 h1 decreased from 47.9 to 30.9 g l1 due to
typically account for 40–70% of the total manufacturing cost the coupling of the membrane module to the bioreactor.
(Tejayadi and Cheryan, 1995; Wee et al., 2006b). The production Lactic acid concentration decreased with increasing dilution
of lactic acid from carbohydrates still exerts a negative impact on rates. The maximum lactic acid concentration was achieved at
the economics of the process, thus rendering biologically-derived the lowest dilution rate for all cases assessed in this study. Higher
lactic acid more expensive than chemically-derived lactic acid. lactic acid concentrations were typically observed at lower dilution
Therefore, several attempts have been made to produce lactic acid rates, and the cell-recycle fermentation method generally induced
using starchy and lignocellulosic resources in order to lower the an increase in the concentration of lactic acid. The lactic acid con-
production costs (John et al., 2007; Melzoch et al., 1996, 1997; Red- centration reached a maximum (69.8 g l1) at a dilution of 0.04 h1
dy et al., 2008; Wee et al., 2004a, 2006b). However, to the best of with the use of 75 g l1 glucose as the feed sugar. As shown in
our knowledge, there have been relatively few reports in the liter- Fig. 5a, when the cell-recycle continuous culture fermentation
ature establishing a cell-recycle continuous fermentation system method was conducted using wood hydrolyzate as the feed sugar,
for lactic acid production using wood hydrolyzate media combined the maximum lactic acid concentration (69.4 g l1) was attained
with corn steep liquor as an inexpensive raw material. The Lactoba- with a dilution rate of 0.04 h1 at an initial substrate (wood hydro-
cillus sp. RKY2 utilized in this study has been described as a poten- lyzate) concentration of 75 g l1.
tial homolactic producer, as it was shown that this strain is capable The volumetric productivity of lactic acid ranged between 1.7
of producing lactic acid from crop residues, including rice bran and and 11.4 g l1 h1, depending on the initial substrate concentration
wheat bran (Yun et al., 2004). and dilution rate employed in this work. The highest levels of lactic
In this study, the effects of dilution rates on the fermentation acid productivity (11.4 g l1 h1) were achieved at a dilution rate of
kinetics of Lactobacillus sp. RKY2 were evaluated, and the conven- 0.28 h1 with 50 g l1 initial glucose concentration and the cell-re-
tional continuous cultivation method was compared with the cycle cultivation method (Fig. 3a), whereas the lowest levels of
membrane cell-recycle continuous technique. Glucose and wood productivity (1.7 g l1 h1) were observed at a dilution rate of
hydrolyzates were used as the carbon source, and corn steep liquor 0.24 h1 with 75 g l1 of initial glucose and conventional continu-
was utilized as the primary nitrogen source. The concentration of ous cultivation (Fig. 2b). When the wood hydrolyzates containing
cells and lactic acid decreased with increases in the dilution rate 75 g l1 glucose were utilized as the constant feed sugar, the lactic
for all initial substrate concentrations in the conventional continu- acid production rate was achieved of a maximum (4.3 g l1 h1) at
ous fermentation method. However, the cell concentration in- 0.12 h1 without cell recycling. The dilution rate necessary to at-
creased with increases in the dilution rate when the cells were tain the highest levels of lactic acid productivity was slightly in-
retained in the bioreactor by coupling the bioreactor to an ultrafil- creased from 0.12 to 0.16 h1 by coupling the membrane cell-
tration membrane module, which resulted in a significant increase recycle system to the bioreactor. Bustos et al. (2007) previously re-
in the lactic acid production rate. Xu et al. (2006) observed similar ported achieving a lactic acid production rate of 3.1 g l1 h1 via
results from a cell-recycle continuous cultivation of Lactobacillus the continuous culture of Lactobacillus pentosus using trimming
paracasei. When the cell-recycle continuous fermentation was con- vine shoots hydrolyzates. Melzoch et al. (1997) reported achieving
ducted using 100 g l1 of glucose, the highest dry cell weight, a maximum lactic acid productivity of approximately 30 g l1 h1
37.7 g l1, was noted at a dilution rate of 0.20 h1. Generally, the for Lactobacillus casei and Lactobacillus lactis using corn cob
cell concentration obtained from wood hydrolyzate media under hydrolyzates. However, their results were obtained using the med-
similar experimental conditions was approximately 10–15% lower ia containing a large amount of expensive nitrogen sources, such as
than that observed with glucose media. This result suggests that yeast extract (20 g l1) and peptone (10 g l1). Although the high-
some inhibitory compounds contained in the lignocellulosic est lactic acid productivity obtained using wood hydrolyzates
hydrolyzates exert a negative effect on the fermentation kinetics, was 6.7 g l1 h1 in our study (Fig. 5b), it is interesting to note that
as this was observed in both experimental cases in this study, the productivity was achieved using the media containing inex-
namely the conventional (Figs. 2a and 4a) and the cell-recycle pensive raw materials such as lignocellulosic hydrolyzates and
(Figs. 3a and 5a) continuous fermentations. The lignocellulosic corn steep liquor.
hydrolyzates have to be detoxified in order to minimize these Table 1 provides the lactic acid yield and substrate conversion
inhibitory effects prior to fermentation, as some of the by-products at different initial substrate concentrations and dilution rates. For
released during the pretreatment are known to be toxic to the all of the initial substrate concentrations and dilution rates, lactic
microorganisms and known to inhibit their metabolism (Mussatto acid yields were maintained at more than 0.90 g g1. However,
and Roberto, 2004). However, our results indicate that the cell-re- the substrate conversion decreased with increases in the initial
cycle cultivation system may prove to circumvent these inhibitory substrate concentrations and dilution rates. For example, with
effects to some degree. wood hydrolyzates containing 75 g l1 glucose, the substrate
Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270 4269

Table 1
Lactic acid yield and substrate conversion at different initial substrate concentrations and dilution rates using Lactobacillus sp. RKY2.

Substrate Dilution rate (h1) Mode of fermentation


Continuous w/o cell-recycle system Continuous w/cell-recycle system
Lactic acid yield (g g1) Substrate conversion (%) Lactic acid yield (g g1) Substrate conversion (%)
Glucose 50 g l1 0.04 0.98 ± 0.02 100 ± 1.0 0.96 ± 0.02 100 ± 0.8
0.08 0.95 ± 0.01 78.8 ± 0.4 0.97 ± 0.02 100 ± 0.4
0.12 0.96 ± 0.02 59.1 ± 1.1 0.96 ± 0.03 99.8 ± 0.4
0.16 0.93 ± 0.01 49.6 ± 0.5 0.95 ± 0.01 98.0 ± 0.5
0.2 0.94 ± 0.02 41.4 ± 1.2 0.96 ± 0.02 96.2 ± 1.0
0.24 0.95 ± 0.03 32.6 ± 1.5 0.94 ± 0.02 92.5 ± 1.1
0.28 0.90 ± 0.01 27.4 ± 0.6 0.94 ± 0.03 86.4 ± 1.6
0.32 – – 0.95 ± 0.03 74.5 ± 1.3
0.36 – – 0.93 ± 0.02 65.2 ± 0.9
Glucose 75 g l1 0.04 0.96 ± 0.03 90.6 ± 1.4 0.93 ± 0.01 100 ± 0.4
0.08 0.97 ± 0.04 76.2 ± 2.1 0.90 ± 0.01 95.3 ± 0.5
0.12 0.99 ± 0.05 57.9 ± 2.4 0.90 ± 0.01 92.0 ± 0.4
0.16 0.96 ± 0.02 42.6 ± 1.1 0.92 ± 0.01 80.1 ± 0.6
0.2 0.96 ± 0.04 30.9 ± 2.3 0.90 ± 0.01 64.4 ± 1.1
0.24 0.98 ± 0.05 14.1 ± 2.6 0.92 ± 0.02 49.8 ± 1.2
Lignocellulosic hydrolyzates 0.04 0.96 ± 0.02 94.9 ± 1.0 0.98 ± 0.05 88.4 ± 2.3
(glucose 50 g l1) 0.08 0.96 ± 0.03 76.8 ± 1.6 0.98 ± 0.03 87.2 ± 1.3
0.12 0.91 ± 0.01 57.1 ± 0.6 0.95 ± 0.02 84.7 ± 0.9
0.16 0.95 ± 0.02 41.6 ± 1.1 0.92 ± 0.03 78.5 ± 1.7
0.2 0.99 ± 0.05 29.4 ± 2.7 0.94 ± 0.01 66.2 ± 0.4
0.24 0.90 ± 0.01 22.6 ± 0.8 0.92 ± 0.01 56.5 ± 0.6
0.28 0.90 ± 0.01 12.2 ± 0.5 0.93 ± 0.02 44.4 ± 0.8
0.32 – – 0.91 ± 0.02 36.5 ± 1.0
0.36 – – 0.92 ± 0.01 24.2 ± 0.5
Lignocellulosic hydrolyzates 0.04 0.96 ± 0.02 90.6 ± 0.9 0.95 ± 0.04 97.3 ± 2.1
(glucose 75 g l1) 0.08 0.94 ± 0.01 71.5 ± 0.4 0.94 ± 0.01 91.5 ± 0.2
0.12 0.91 ± 0.02 53.0 ± 1.2 0.96 ± 0.03 74.7 ± 1.4
0.16 0.95 ± 0.03 36.1 ± 1.6 0.95 ± 0.02 58.7 ± 1.0
0.2 0.92 ± 0.01 22.3 ± 0.6 0.91 ± 0.01 44.3 ± 0.6
0.24 0.91 ± 0.01 10.1 ± 0.4 0.92 ± 0.01 30.6 ± 0.4

The standard deviations were calculated from three independent analyses.

conversion was reduced from 90.6% to 10.1% when the dilution Table 2 compares the results of continuous lactic acid fermenta-
rate was increased from 0.04 to 0.24 h1. The largest reduction in tion using lignocellulosic hydrolyzates with the results reported in
the substrate conversion (from 94.9% to 12.2%) was observed with several studies in the relevant literature. Although Melzoch et al.
the conventional continuous fermentation using wood hydroly- (1997) previously reported the highest lactic acid productivity
zates containing 50 g l1 glucose. However, it is interesting to note (13.1 g l1 h1) using corn cob hydrolyzates, they used yeast ex-
that the substrate conversion was also improved by coupling the tract and peptone as nitrogen sources, which is not economically
cell-recycle system to the bioreactor. The substrate conversions feasible, because this significantly increases the production cost
in the utilization of wood hydrolyzates were found to be less than of lactic acid. Bustos et al. (2007) reported that a lactic acid produc-
those of refined glucose. This was possibly due to the presence of tivity of 3.1 g l1 h1 was achieved with the cell-recycle cultivation
inhibitory compounds in the wood hydrolyzates (Wee et al., of L. pentosus using trimming vine shoots hydrolyzates and corn
2004a). steep liquor. However, in our study, the highest level of lactic acid

Table 2
Comparison of lactic acid fermentations in continuous culture systems using lignocellulosic hydrolyzates.

Microorganism Raw materials Fermentation Dilution Lactic acid Lactic acid Productivity Reference
type rate (h1) concentration (g l1) yield (g g1) (g l1 h1)
Lactobacillus sp. RKY2 Wood (glucose 50 g l1) Cell-recycle 0.24 26.1 0.92 6.3 This work
Corn steep liquor (30 gl1)
Yeast extract (1.5 gl1)
Lactobacillus sp. RKY2 Wood (glucose 75 g l1) Cell-recycle 0.16 42.0 0.95 6.7 This work
Corn steep liquor (30 g l1)
Yeast extract (1.5 g l1)
Lactobacillus pentosus Vine shoots (11.1 g l1 glucose Conventional 0.20 15.5 0.70 3.1 Bustos et al. (2007)
and 18 g l1 xylose)
Corn steep liquor (10 g l1)
Yeast extract (10 g l1)
Lactobacillus casei Corn cobs (50 g l1 glucose) Cell-recycle 0.08–0.30 34.1 0.68 3.4 Melzoch et al. (1997)
Yeast extract (20 g l1)
Peptone (10 g l1)
Lactobacillus lactis Corn cobs (65–150 g l1 glucose) Cell-recycle 0.08–0.30 108.9 0.73 13.1 Melzoch et al. (1997)
Yeast extract (20 g l1)
Peptone (10 g l1)
4270 Y.-J. Wee, H.-W. Ryu / Bioresource Technology 100 (2009) 4262–4270

productivity, 6.7 g l1 h1, was achieved with the cell-recycle cul- John, R.P., Nampoothiri, K.M., Pandey, A., 2007. Fermentative production of lactic
acid from biomass: an overview on process developments and future
ture fermentation method using wood hydrolyzates and corn steep
perspectives. Appl. Microbiol. Biotechnol. 74, 524–534.
liquor. Most of the continuous fermentation techniques with high Kim, H.O., Wee, Y.J., Kim, J.N., Yun, J.S., Ryu, H.W., 2006. Production of lactic acid
productivities reported in the literature have been accompanied by from cheese whey by batch and repeated batch cultures of Lactobacillus sp.
high levels of expensive nitrogen sources. Thus, a reduction of RKY2. Appl. Biochem. Biotechnol. 131, 694–704.
Kwon, S., Yoo, I.K., Lee, W.G., Chang, H.N., Chang, Y.K., 2001. High rate continuous
yeast extract concentration coupled with the use of lignocellulosic production of lactic acid by Lactobacillus rhamnosus in a two-stage membrane
hydrolyzates and corn steep liquor, as in our study, should be ex- cell-recycle bioreactor. Biotechnol. Bioeng. 73, 225–234.
pected to significantly reduce the total manufacturing costs. Litchfield, J.H., 1996. Microbiological production of lactic acid. Adv. Appl. Microbiol.
42, 45–95.
Lu, X., Zhang, Y., Angelidaki, I., 2009. Optimization of H2SO4-catalyzed hydrothermal
5. Conclusion pretreatment of rapeseed straw for bioconversion to ethanol: focusing on
pretreatment at high solids content. Bioresour. Technol. 100, 3048–3053.
Melzoch, K., Votruba, J., Schwippel, J., Rychtera, M., Hábová, V., 1996. Lactic acid
The results of this study indicate that the cell-recycle cultiva- production in a continuous culture using lignocellulosic hydrolysate as a
tion method may improve fermentation efficiency, including volu- substrate: identification of physiological model. Folia Microbiol. 41, 211–215.
Melzoch, K., Votruba, J., Hábová, V., Rychtera, M., 1997. Lactic acid production in a
metric productivity, in a continuous fermentation system using
cell retention continuous culture using lignocellulosic hydrolyzate as a
inexpensive raw material. The highest levels of volumetric produc- substrate. J. Biotechnol. 56, 25–32.
tivity when using either glucose or wood hydrolyzates were Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of diluted-acid
achieved via the cell-recycle cultivation method. With lignocellu- lignocellulosic hydrolyzates for use in fermentative processes: a review.
Bioresour. Technol. 93, 1–10.
losic hydrolyzates, the lactic acid productivity was maximum Reddy, G., Altaf, Md., Naveena, B.J., Venkateshwar, M., Kumar, E.V., 2008. Amylolytic
(6.7 g l1 h1) at a dilution rate of 0.16 h1 under the cell-recycle bacterial lactic acid fermentation – a review. Biotechnol. Adv. 26, 22–34.
continuous fermentation conditions, in which the lactic acid con- Roukas, T., Kotzekidou, P., 1998. Lactic acid production from deproteinized whey by
mixed cultures of free and coimmobilized Lactobacillus casei and Lactococcus
centration was 42.0 g l1. As compared to the reports in the litera- lactis using fedbatch culture. Enzyme Microb. Technol. 22, 199–204.
ture regarding continuous fermentation using lignocellulosic Ryu, H.W., Yun, J.S., Wee, Y.J., 2003. Lactic acid. In: Pandey, A. (Ed.), Concise
hydrolyzates, the maximum lactic acid productivity and concen- Encyclopedia of Bioresource Technology. The Haworth Press, New York, pp.
635–644.
trations observed in this study are more economically feasible than Shene, C., Bravo, S., 2007. Whey fermentation by Lactobacillus delbrueckii subsp.
has been reported in related studies. The lactic acid productivity Bulgaricus for expolysaccharide production in continuous culture. Enzyme
levels and economic feasibility of the process can be improved fur- Microb. Technol. 40, 1578–1584.
Södergård, A., Stolt, M., 2002. Properties of lactic acid based polymers and their
ther using more economical raw materials, such as crop residues
correlation with composition. Prog. Polym. Sci. 27, 1123–1163.
with high nitrogen contents. Tejayadi, S., Cheryan, M., 1995. Lactic acid from cheese whey permeate. Productivity
and economics of a continuous membrane bioreactor. Appl. Microbiol.
Biotechnol. 43, 242–248.
References Varadarajan, S., Miller, D.J., 1999. Catalytic upgrading of fermentation-derived
organic acids. Biotechnol. Prog. 15, 845–854.
Åkerberg, C., Zacchi, G., 2000. An economic evaluation of the fermentative VickRoy, T.B., 1985. Lactic acid. In: Moo-Young, M. (Ed.), Comprehensive
production of lactic acid from wheat flour. Bioresour. Technol. 75, 119–126. Biotechnology, Vol. 3. Pergamon Press, New York, pp. 761–776.
Altaf, Md., Naveena, B.J., Reddy, G., 2007. Use of inexpensive nitrogen sources and Wee, Y.J., Yun, J.S., Park, D.H., Ryu, H.W., 2004a. Biotechnological production of L(+)-
starch for L(+) lactic acid production in anaerobic submerged fermentation. lactic acid from wood hydrolyzate by batch fermentation of Enterococcus
Bioresour. Technol. 98, 498–503. faecalis. Biotechnol. Lett. 26, 71–74.
Bustos, G., Torre, N., Moldes, A.B., Cruz, J.M., Domínguez, J.M., 2007. Revalorization Wee, Y.J., Yun, J.S., Park, D.H., Ryu, H.W., 2004b. Isolation and characterization of a
of hemicellulosic trimming vine shoots hydrolyzates through continuous novel lactic acid bacterium for the production of lactic acid. Biotechnol.
production of lactic acid and biosurfactants by L. Pentosus. J. Food Eng. 78, Bioprocess Eng. 9, 303–308.
405–412. Wee, Y.J., Kim, J.N., Yun, J.S., Ryu, H.W., 2005. Optimum conditions for the biological
Datta, R., Tsai, S.P., Bonsignore, P., Moon, S.H., Frank, J.R., 1995. Technological and production of lactic acid by a newly isolated lactic acid bacterium, Lactobacillus
economic potential of poly(lactic acid) and lactic acid derivatives. FEMS sp. RKY2. Biotechnol. Bioprocess Eng. 10, 23–28.
Microbiol. Rev. 16, 221–231. Wee, Y.J., Kim, J.N., Ryu, H.W., 2006a. Biotechnological production of lactic acid and
Davison, B.E., Llanos, R.L., Cancilla, M.R., Redman, N.C., Hillier, A.J., 1995. Current its recent applications. Food Technol. Biotechnol. 44, 163–172.
research on the genetics of lactic acid production in lactic acid bacteria. Int. Wee, Y.J., Yun, J.S., Kim, D., Ryu, H.W., 2006b. Batch and repeated batch production
Dairy J. 5, 763–784. of L(+)-lactic acid by Enterococcus faecalis RKY1 using wood hydrolyzate and
Gao, M.T., Michiteru, K., Rie, G., Hirokazu, T., Makoto, H., Tadashi, H., 2005. corn steep liquor. J. Ind. Microbiol. Biotechnol. 33, 431–435.
Development of a continuous electrodialysis fermentation system for Xu, G.Q., Chu, J., Wang, Y.H., Zhuang, Y.P., Zhang, S.L., Peng, H.Q., 2006. Development
production of lactic acid by Lactobacillus rhamnosus. Process Biochem. 40, of a continuous cell-recycle fermentation system for production of lactic acid by
1033–1036. Lactobacillus paracasei. Process Biochem. 41, 2458–2463.
González-Pajuelo, M., Adrade, J.C., Vasconcelos, I., 2005. Production of 1, 3- Yun, J.S., Wee, Y.J., Kim, J.N., Ryu, H.W., 2004. Fermentative production of dl-lactic
propanediol by Clostridium butyricum VPI 3266 in continuous cultures with acid from amylase-treated rice and wheat brans hydrolyzate by a novel lactic
high yield and productivity. J. Ind. Microbiol. Biotechnol. 32, 391–396. acid bacterium, Lactobacillus sp.. Biotechnol. Lett. 26, 1613–1616.

You might also like