Accepted Manuscript

Case Study
Enhanced butanol production by solvent tolerance Clostridium acetobutylicum SE25 from cassava flour in a fibrous bed bioreactor
Han-guang Li, Xing-xing Ma, Qing-hua Zhang, Wei Luo, Ya-qing Wu, Xunhang Li
PII:
DOI:
Reference:

S0960-8524(16)31334-7
http://dx.doi.org/10.1016/j.biortech.2016.08.120
BITE 17090

To appear in:

Bioresource Technology

Received Date:
Revised Date:
Accepted Date:

7 July 2016
25 August 2016
26 August 2016

Please cite this article as: Li, H-g., Ma, X-x., Zhang, Q-h., Luo, W., Wu, Y-q., Li, X-h., Enhanced butanol production
by solvent tolerance Clostridium acetobutylicum SE25 from cassava flour in a fibrous bed bioreactor, Bioresource
Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.08.120

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Enhanced butanol production by solvent tolerance Clostridium acetobutylicum

SE25 from cassava flour in a fibrous bed bioreactor

3
4
5

Han-guang LIa , Xing-xing MAa, Qing-hua ZHANG a,*, Wei LUO b , Ya-qing WUb ,

Xun-hang LIb

7
8
9
10

College of Bioscience and Engineering, Jiangxi Agricultural University, Jiangxi

11

Engineering Laboratory for the Development and Utilization of Agricultural Microbial

12

Resources, Nanchang 330045, China

13

14

Biotechnology, Jiangnan University, Wuxi 214122, China

The Key Laboratory of Industrial Biotechnology, Ministry of Education, School of

15
16
17

Corresponding author at: College of Bioscience and Engineering, Jiangxi Agricultural

18

University, Jiangxi Engineering Laboratory for the Development and Utilization of

19

Agricultural Microbial Resources, Nanchang 330045, China.

20

Tel.: +86 791 83813459; fax: +86 791 83813459

21

E-mail address:lhg7886@sohu.com.

22
23
24

Abstract

To enhance the butanol productivity and reduce the material cost, acetone, butanol,

and ethanol fermentation by Clostridium acetobutylicum SE25 was investigated using

batch, repeated-batch and continuous cultures in a fibrous bed bioreactor, where cassava

flour was used as the substrate. With periodical nutrient supplementation, stable butanol

production was maintained for about 360 h in a 6-cycle repeated-batch fermentation

with an average butanol productivity of 0.28 g/L/h and butanol yield of 0.32 g/g-starch.

In addition, the highest butanol productivity of 0.63 g/L/h and butanol yield of 0.36

g/g-starch were achieved when the dilution rate were investigated in continuous

10

production of acetone, butanol, and ethanol using a fibrous bed bioreactor, which were

11

231.6 % and 28.6 % higher than those of the free-cell fermentation. On the other hand,

12

this study also successfully comfirmed that the biofilm can provide an effective

13

protection for the microbial cells which are growing in stressful environment.

14
15
16
17
18

19

Keywords: Clostridium acetobutylicum, Acetone-butanol-ethanol fermentation,

20

Immobilization, Biofilm, Cassava flour

21
22
23
24

1
2

1. Introduction
Due to the increasing concerns over the environmental issues associated with the

impact of petroleum fuel emissions and the decreasing of fossil fuel reserves,

production of alternative fuels such as butanol and ethanol by fermentation has drawn

worldwide attention (Zheng et al., 2013). Amongest these biofuels, butanol has many

advantages compared to ethanol, such as higher boiling point, higher energy content,

less corrosive and a reduced need to modify current combustion energies. Therefore,

butanol is regarded as one of the most appropriate candidates for biofuels. In addition,

butanol is a multipurpose chemical feedstock which has also been extensively used in

10

plastic and food industries. Commercial butanol fermentation processes have triggered

11

increasing attention by some companies (such as DuPont and BP) (Lepiz-Aguilar et al.,

12

2013, Li et al., 2016). Some solventogenic Clostridium, including Clostridium

13

saccharoperacetobutylicum, Clostridium saccharobutylicum, Clostridium beijerinckii

14

and Clostridium acetobutylicum, are used as butanol producing strains during acetone,

15

butanol, and ethanol (ABE) fermentation. These four species can be roughly divided

16

into two classes such as starch and sugar species according to the utilization of carbon

17

sources. Furthermore, it was believed C. acetobutylicum and C. beijerinckii are the

18

major butanol producing strains in present ABE fermentation industries (Keis et al.,

19

2001, Patakova et al., 2013). Butanol production in ABE fermentation processes can be

20

divided into two phases, i.e., acidogenic phase and solventogenic phase. Moreover,

21

some studies showed that the phase shift (from acidogenesis to solventogenesis) can be

22

trigged if the pH value is below 5 and the concentration of butyric acid is over 2 g/L

23

(Cheng et al., 2012, Li et al., 2014b). However, practical ABE fermentation processes

24

are often extremely complicated and difficult to being controlled. Consequently, further

system modifications for these processes are inevitably required (Chauvatcharin et al.,

1998).

Presently, the major obstacles associated with ABE fermentation include conversion

efficiency of substrate to product, product toxicity (especially butanol) to the producing

strains, the potential for culture degeneration, and the ability to utilize cheaper raw

materials or renewable agricultural wastes as the substrates, which resulted in the

production of butanol from ABE fermentation less competitive when compared with the

petroleum-based butanol production (Qureshi et al., 2014). Furthermore, economic

analysis of ABE fermentation processes showed that the separation cost would be

10

reduced and led to an economically viable process if the final butanol concentration has

11

been slightly increased (Ting et al., 2012). Therefore, in order to enhance the economic

12

competitiveness of biobutanol production, the improvement of substrates, microbial

13

strains and processes are the key issues of priority research. Diverse methods such as

14

solvent tolerant microorganisms screening, genetic engineering, metabolic engineering,

15

modern fermentation techniques and downstream processing have been constantly used

16

to surmount the above hampers (Ezeji et al., 2010). Amongst those methods, fed-batch

17

fermentation coupling with an immobilization technology, is a relatively simple and an

18

economic bioprocess, and worth to maximize the butanol production as exhibited herein

19

this study.

20

In traditional ABE fermentation, corn starch or glucose were mostly used as the

21

major feedstocks. As the case stands, the fermentation substrate cost, which account for

22

60-70 % of the total production cost, is the most influential factor in ABE fermentation

23

(Qureshi and Blaschek 2000). Therefore, it is necessary to explore inexpensive raw

24

materials for making ABE fermentation more economically attractive. Cassava, a

non-cereal starchy crop, can widely grow in harsh climates and poor soils. In 2010, the

worldwide yield of cassava was 242 million tons, especially in China, the cassava

output was about 7.3 million tons which accounted for about 3.0 % of the current global

cassava production (Li et al., 2015). In addition, cassava is regarded as a nonfood crop

in China, and the concerns for food security would thus be minimized. Therefore,

cassava represents an alternative cheap substrate for ABE fermentation, which is

attractive and promising in both geographical and economic considerations.

8
9

In this study, in order to enhance the butanol productivity and reduce the production
cost, continuous butanol production using a high butanol tolerant strain C.

10

acetobutylicum SE25 with cassava flour as the substrate was carried out in a fibrous bed

11

bioreactor (FBB) system. AquaMats AO was firstly used as an immobilized material for

12

the ABE fermentation processes. And the effect of various dilution rates on solvent

13

production in continuous ABE fermentation was then investigated. In addition, the

14

correlation between biofilm and butanol tolerance in C. acetobutylicum was also

15

evaluated. To the best of the authors knowledge, little literatures has been reported on

16

the correlation between biofilm and butanol tolerance in C. acetobutylicum.

17

Consequently, the results presented herein also would provide a valuable reference for

18

further elucidating the solvent tolerance mechanisms of other microorganisms.

19
20

2. Materials and methods

21

2.1 Microorganism and medium

22

The solvent tolerance strain C. acetobutylicum SE25, derived from C. acetobutylicum

23

PW12, was isolated from the waste water of Daqing Oilfield Company in North China

24

and used as the producing strain. The cell suspension was routinely maintained in 20 %

(v/v) of steriled glycerol at -80 C in screw-capped bottles. SE25 cell suspension in a

glycerol tube was heat-shocked for 90 s at 80 C and then cooled to room temperature

on ice. The heat shocked cell suspension was transferred into an anoxic sterilized

tryptone-yeast extract-acetate medium (TYA medium) at 37 C for 16-18 h. Following

growth, 10-15 mL of activated cultures was inoculated into 250 mL screw-capped

bottles containing 100 mL of fermentation medium (Li et al., 2014a).

The developed P2 medium contained the following substances in per liter of distilled

water: 60 g glucose, 3 g yeast extract, 1 mL of filter-sterilized stock solutions (vitamin:

0.1 g thiamin, 0.001 g biotin, 0.1 g para-aminobenzoic acid, mineral: 1 g MnSO4H2O, 1

10

g NaCl, 20 g MgSO47H2O, 1 g FeSO47H2O, and buffer: 220 g ammonium acetate, 50

11

g K2HPO4, 50 g KH2PO4) was added into 1 L of P2 medium. Minerals and vitamins

12

were filter-sterilized through 0.22 m sterilized membrane filters. Buffer, nitrogen and

13

carbon sources were separately sterilized by autoclaving at 121 C for 15 min (Li et al.,

14

2013).

15

The developed TYA medium comprised of the following substances in per liter of

16

distilled water: 40 g glucose, 6.0 g tryptone, 2.0 g yeast extract, 0.2 g MgSO4H2O, 0.01

17

g FeSO4, 3.0 g CH3COONH4, 0.75 g K2HPO4 and 0.75 g KH2PO4. The medium was

18

sterilized by autoclaving at 121 C for 15 min (Loyarkat et al., 2013).

19
20
21

2.2 Cassava flour pretreatment and hydrolysis

The cassava flour was pretreated and hydrolyzed according to the method reported by

22

Yang et al. (2015). Required amounts of cassava flour were immersed in distilled water

23

to obtain a suspension of 80 g/L cassava flour, which was used for subsequent

24

enzymatic hydrolysis. Enzymatic hydrolysis process consisted of two steps: liquefaction

and saccharification. CaCl2 (1 g/L), previous to liquefaction, was added into cassava

flour suspensions to obtain a final concentration of 40 mg-Ca2+/L. At the same time, the

pH was adjusted to 6.5 using 2 M HCl or 2 M NaOH. For liquefaction, the processes

were carried out in flasks which were placed in a bath shaker (150 rpm), and -amylase

was added at a dosage of 8 U/g-cassava. The enzymatic reaction was carried out for 45

min at 100 C, followed by cooling down to 60 C on ice and a decrease in pH to 4.5

with 2 M HCl. Subsequently, the process of saccharification was performed by adding

-glucoamylase (120 U/g-cassava) for 4 h at 60 C, and the enzyme was then

inactivated by heating at 80 C for 5 min after saccharification. The hydrolysis liquid

10

was sterilized at 121 C for 15 min at the final pretreatment and hydrolysis procedure.

11

The sterilized hydrolysis liquefied was then used as the medium for ABE fermentation.

12

-amylase (20,000 U/mL) and -glucoamylase (100,000 U/mL), which were used as the

13

enzymatic treatment, were obtained from Boli Biological Products Co. Ltd, Taizhou,

14

China.

15
16

2.3 Preparation of fibrous bed bioreactor (FBB) and operation

17

2.3.1 The construction of FBB

18

The FBB reactor was prepared as reported by (Silva and Yang 1995) and the schematic

19

drawing of FBB system was shown in Fig. 1. AquaMats AO, which was purchased from

20

Zhongyu water ecological technology Co., Ltd in South China, was used as the

21

immobilization carrier. AquaMats AO (300 mm 260 mm) and stainless steel wire were

22

coiled into a spiral cylindrical. At the same time, the clearance of each ring of stainless

23

steel wire mesh was 3 mm. The glass column reactor ( 60 mm 300 mm) was filled

24

till to reach the height of 25-30 mm by ceramic tube glass. Afterwards, the stainless

steel wire mesh was put into the glass column reactor. This reactor system was

designated as FBB. The FBB and the bioreactor were separately autoclaved for 30 min,

and then aseptically connected with the fermentation tank by a hosepipe after

sterilization.

5
6
7

2.3.2 Repeated-batch fermentation

About 300 mL of activated growing cells (16-18 h) were inoculated into a 5 L

bioreactor (LiFlus GX, Biotron Corp, China) containing 2.7 L medium. The FBB

system was purged with nitrogen to remove trace amounts of oxygen before and after

10

inoculation. To immobilize cell onto the AquaMats AO, the fermentation broth was

11

re-circulated through the FBB when the optical density (OD) reached 3.0 (after 16-24 h).

12

Furthermore, the re-circulation flow was controlled at 30 mL/min by a peristaltic pump,

13

until the production of butanol no longer increased. The old broth was then drained and

14

replaced with a fresh fermentation medium to promote the cells in the FBB continually

15

growing. Thereafter, 60 mL/min of re-circulation flow was be used. Before starting a

16

new batch, the fermentation broth in the FBB system was completely drained out. Then,

17

3.0 L of fresh fermentation medium was pumped into the bioreactor and circulated

18

through the FBB. The fermentation time of each batch was kept at about 48 h. A total of

19

6 batches were carried out with cassava flour as the substrates. The fermentation

20

kinetics was further investigated in a repeated batch mode at 37 C without pH control.

21

Broth samples were collected periodically from the bioreactor for the analyses of

22

product concentrations.

23
24

2.3.3 Continuous fermentation

Continuous fermentation using cassava flour as substrate was also performed in the

FBB system. The process of continuous fermentation was carried out in a bioreactor

without pH control. Throughout the performance, the fermentor was operated at 50 rpm

and maintained at 37 C. The initial inoculation amount and the start-up time of

peristaltic pump was the same as the repeated-batch fermentation. Furthermore, 30

mL/min of re-circulation flow was conducted. In order to investigate the effect of

dilution rate on ABE fermentation performance, the dilution rates of bioreactor were

controlled at 0.04 h-10.06 h -1 and 0.10 h -1, respectively. In addition, the volume of the

inflow of the feeding fermentation medium was the same as the outflow volume.

10

Continuous fermentation was started by feeding fresh fermentation medium at certain

11

flow rates. Samples for monitoring the performance of fermentation were collected at a

12

regular time interval.

13
14

2.4 Cell morphology studies using scanning electron microscopy (SEM)

15

The effect of butanol stress on the morphology of strains SE25 and PW12 was

16

investigated using SEM. After 48-60 h of incubation in the presence of 20 g/L of

17

butanol, bacterial cells were harvested at late exponential phase and washed twice with

18

0.9 % of NaCl solution. Bacteria were then fixed by immersion in 2.0 % of

19

glutaraldehyde at 4 C for 2-4 h. Following that, the specimens were washed twice with

20

0.9 % of NaCl solution and dehydrated by a series of progressive ethanol concentration

21

at 30, 40, 50, 60, 70, 80, 90 and 100 % (v/v). Each processing time was 15 min and

22

dried at room temperature. The specimens in 100 % ethanol were further mounted on

23

SEM stubs and coated with gold for 90 s using gold sputter (Zhang et al., 2012).

24

Morphological changes was observed under an S-4800 (Hitachi Co., Japan) SEM

1
2
3

2.5 Analytical methods

The optical density was measured at 600 nm using a spectrophotometer (Spectronic

200, Thermo scientific). The concentrations of starch and glucose were determined

according to the method described by Thang et al. (2010).

Solvent concentration was analyzed by internal standard method using gas

chromatography (GC-2010, Shimadzu Scientific Instruments, Japan) equipped with a

flame ionization detector (FID) operated at 240 C, along with capillary column

(PED-20M, 30 m, 0.32 mmID,0.4 m df). The column and injector temperature were

10

maintained at 210 C and 280 C, respectively. The nitrogen gas and isobutanol were

11

used as carrier gas and internal standard. The injection volume was 0.4 L. Acetate and

12

butyrate were evaluated using the HPLC system (DIONEX, U-3000, USA) equipped

13

with an organic acid analysis column (Aminex HPX-87H, 300 mm 7.8 mm, Hercules,

14

CA, USA) operated at 50 C, a refractive index detector operated at 210 nm and room

15

temperature, and 5 mmol/L of H2SO4 solution was used as the mobile phase (0.6

16

mL/min). The injection volume was 20 L. Solvent yield was calculated as solvent

17

produced divided by the sugar utilized. Solvent productivity was defined as the solvent

18

production achieved (g/L) divided by the fermentation time (h) (Qureshi et al., 2007).

19
20
21

3. Results and discussion

3.1 Effects of Aquamats-AO addition on ABE fermentaion

22

Aquamats-AO, a material firstly made by Meridian Aquatic Technology. L. L. C, is a

23

kind of immobilized carrier for microbial growth. Aquamats-AO, which is made by

24

super knitting technology, belongs to the inert synthetic polymer, and it has a uniform

10

and reasonable pore structure. Aquamats-AO has high specific surface areas (1m2

aquamats-AO can probably provide 245 m2 surface area), and they can provide great

adsorption surface area for microbial growth. In addition, the working life of

Aquamats-AO can maintain for a long time (no less than 14 years). Consequently, it had

been extensively utilized in water ecological environment restoration and water

pollution prevention and control field all over the world since 1995 (Yu et al., 2010). On

the other hand, to the best of our knowledge, literatures scarcely reported the water

purification materials of Aquamats-AO is been used in ABE immobilized fermentation

till date. Previous study indicated that cell immobilization can enhance the efficiency of

10

traditional fermentation process because cells can deposit on support matrix and further

11

increase the cell amount of per bioreactor volume (Survase et al., 2012). Therefore, to

12

enhance the production of butanol, the Aquamats-AO was used as the immobilization

13

carrier in this study. In order to investigate the effects of addition of Aquamats-AO on

14

ABE fermentaion, different addition amount of Aquamats-AO ranging from 0.6 to 2.5 g

15

were added into the fermentation medium. The fermentation performance was carried

16

out in a 250 mL of screw-capped bottle containing 150 mL of fermentation medium.

17

The initial pH and fermentation time were 6.0 and 84 h, respectively.

18

The results showed that 20.02 0.87 g/L of total solvent and 13.38 0.73 g/L of

19

butanol were obtained, without immobilization carrier supplementation, respectively.

20

When the immobilization carrier amount was below 8.0 g/L (1.2 g/150 mL), an increase

21

of the total solvent and butanol concentration was observed. However, when the amount

22

of immobilization carrier exceeded 8.0 g/L, a decrease of the total solvent and butanol

23

concentration occurred. At the optimal amount of immobilization carrier (8.0 g/L), the

24

total solvent and butanol concentration were 23.28 1.31 g/L and 15.15 0.79 g/L,

11

respectively, which were 16.3 % and 13.2 % higher than those of the control group.

Therefore, it could reasonably infer from these results that the immobilization

fermentation performance with low additive quality of immobilization carrier may be

reduce due to a relatively small adsorption surface area, which further reduces the

immobile and protective function for the producing strains. On the other hand, if a

higher addition amount of immobilization carrier was performed, the solvent production

may also be reduced due to the high amount of solid carrier, which limits the mass

transfer for the whole fermentation system flow and nutrient absorption. Therefore, an

optimum amount of 8.0 g/L of the immobilization carrier was selected for subsequent

10

fermentation experiments.

11
12
13

3.2 Kinetics of free-cell and immobilized cell fermentation

Previous studies indicated the fermentation performance can be improved by the

14

addition of immobilized carrier (Aquamats-AO) into the fermentation broth. To further

15

investigate the difference between free-cell and immobilized cell fermentation, free-cell

16

and immobilized cell fermentation were all carried out in a 5-L bioreactor.

17

The results for the different residual glucose, pH and solvent were shown in Fig. 2,

18

where cassava flour was used as the substrate. The immobilized cell fermentation

19

kinetics, such as the consumption rate of the reducing sugar and descent speed of the pH,

20

was consistent with the free-cell fermentation before the first 18 h. The reason for this

21

phenomenon might due to the un-working of the peristaltic pump. Consequently, the

22

fermentation environment was almost the same. However, the difference of

23

fermentation kinetics between the free and immobilized cell fermentation were observed

24

after 18 h of incubation. When the peristaltic pump started to work, the fermentation

12

broth and cells were flowed into the FBB system, and the cells were then gradually

adsorbed onto the immobilized carrier. Therefore, it could provide a direct contact

between the immobilized cells and nutrients, so the diffusion problems would be

minimized (Kilonzo et al., 2011). When the immobilized cell fermentation was

performed, the descent speed of the pH and consumption rate of the reducing sugar

were higher than those of the free-cell fermentation. Furthermore, the highest

concentration of butanol and total solvent was obtained at 60 h and 78 h for

immobilized cell and free-cell fermentation, respectively. The highest butanol and total

solvent productivity of 0.19 g/L/h and 0.29 g/L/h were obtained when free-cell

10

fermentation was performed. Contrarily, if the immobilized cell fermentation was used,

11

the maximum productivity of butanol and total solvent were 0.24 g/L/h and 0.37 g/L/h,

12

respectively, which were 26.3 % and 27.6 % higher than those of the free-cell

13

fermentation. The butanol yield from starch was 0.27 g/g-starch and 0.28 g/g-starch,

14

with immobilized cell and free-cell fermentation, respectively. On the other hand, the

15

fermentation could be ceased after being incubated for 60 h and 78 h with immobilized

16

cell and free-cell fermentation. Therefore, the fermentation time for immobilized cell

17

fermentation was shortened 18 h compared with free-cell fermentation. The butanol

18

productivity (0.24 g/L/h) and the fermentation time for immobilized cell fermentation

19

were superior to those obtained with free-cell fermentation. Therefore, these results

20

indicated that the immobilized cell fermentation through FBB system is a promising

21

technique for the improvement of ABE fermentation performance.

22
23
24

3.3 Repeated-batch fermentation in FBB continuous reactor

According to the (Kittithanesuan and Phisalaphong 2015), the fermentation time

13

could be shortened and the butanol productivity and yield could also be enhanced when

the method of the repeated-batch fermentation was performed. Therefore, ABE

fermentation by SE25 immobilized on Aquamats-AO was further investigated in 6

batches of repeated-batch fermentation, and the experimental results were illustrated in

Fig. 3 where cassava flour was used as the substrate.

As can be seen, the production of butanol and total solvent were relatively stable

during the 6-cycle repeated-batch fermentation. Specifically, the butanol and total

solvent concentration were fluctuated ranging from 14.00 to 15.50 g/L and from 20.00

to 23.60 g/L, respectively. Furthermore, 89.54 g/L of butanol and 133.16 g/L of total

10

solvent were obtained at the end of the fermenation. The inital pH was 6.0 for the first

11

batch (Batch 1) and about 5.6 for consecutive batches (Batches 2, 3, 4, 5 and 6). The

12

consequent lower pH could be explained that a small amount of last fermenation broth

13

was kept in the FBB system, together with some living cells. Therefore, due to the

14

above phenomenon, the fermentation time for consecutive batches (Batches 2, 3, 4 and

15

5) was shorter than that of the first batch (Batch 1). On the other hand, the average

16

butanol productivity and yield from starch were 0.28 g/L/h and 0.32 g/g-starch,

17

respectively, which were 47.4 % and 14.3 % higher than those of the free-cell

18

fermentation (section 3.2).

19

The genetic stability of the producing strains is one of the basic conditions for a

20

promising industrial strain. Therefore, an indirect evidence of this ability is sustained

21

secretion objective product in different generations (Hua et al., 2010). In this study, the

22

producing strain SE25 was roughly equivalent to successive five generation cultures.

23

Furthermore, the results from this case showed that the production of total solvent and

24

butanol still maintained relatively stable. The reasons could be explained by two

14

primary factors. On the one hand, the producing strain SE25 owned the advantage in

genetic stability. On the other hand, the cells in the FBB system were in the dynamic

balance of adsorption and supersession. Particularly, the activated cells were continually

adsorbed into the Aquamats-AO. The death cells would come adrift from the adsorption

material due to the periodic flow of the fermentation broth. Therefore, the cells in the

FBB system always maintained a higher vitality. At the same time, the concentrations of

solvent were at high levels in this system, and some solvent tolerance strains were

probably isolated. Thus, repeated-batch fermentation in the FBB continuous reactor

could also be used as an effective adaptive evolution method to enhance the tolerance of

10

producing strains and butanol produciton.

11
12
13

3.4 Continuous production of ABE using FBB system

To improve solvent productivity, continuous fermentation is usually performed in

14

industrial ABE fermentation (Ni et al., 2013). Meanwhile, it has been reported that

15

substrate feed rate had a significant influence on cell growth and solvent production in

16

continuous ABE fermentation process. Therefore, the dilution rate was often

17

investigated in continuous fermentation (Yen and Li 2011). The residence time of

18

fermentation medium in a fermentor was inversely related to dilution rate. In other

19

words, the fermentation medium could be stayed for relatively long time in the FBB

20

system when lower dilution rate was performed. Therefore, the optimum use of

21

fermentation medium would be realized. At higher dilution rates, the residence time of

22

fermentation medium was relatively shortened, and the fermentation medium was then

23

not used to the full (Survase, van Heiningen et al. 2012). In this case, the effect of

24

different dilution rates on substrate yield and solvent production was investigated. The

15

results from the continuous production of ABE by SE25 using cassava flour as the

substrate in FBB system were shown in Fig. 4.

As can be seen from Fig. 4 (a) and (b), after 60 h of continuous operation, a relatively

stable solvent production was observed at the same dilution rate. Increasing the dilution

rate from 0.04 h-1 to 0.10 h-1 resulted in a decreased solvent production and yield. When

dilution rate was controlled at 0.04 h-1, the concentration of butanol and total solvent

were 10.72 g/L and 14.21 g/L, respectively, and the butanol productivity and yield from

starch were 0.43 g/L/h and 0.36 g/g-starch, respectively, which were 126.3 % and 28.6 %

higher than those of the free-cell fermentation (section 3.2). However, 6.28 g/L of

10

butanol and 8.15 g/L of total solvent were obtained at a dilution rate of 0.10 h-1, and the

11

butanol productivity and yield from starch were 0.63 g/L/h and 0.31 g/g-starch,

12

respectively, which were 231.6 % and 10.7 % higher than those of the free-cell

13

fermentation (section 3.2). Therefore, it was found that the butanol productivity has

14

been greatly improved when compared to the batch fermentation. On the one hand, the

15

cells always maintained a higher vitality in this condition, and the inoculation times

16

could be then reduced. Consequently, the higher reactor productivity was achieved. On

17

the other hand, the ABE fermentation process could be divided into two phases, i.e., the

18

acidogenic phase and the solventogenic phase. The acidogenic phase was found being

19

associated with cell growth, while there was no association with cell growth in the

20

solventogenic phase. Therefore, the continuous production of ABE using FBB system

21

could lead to a higher butanol productivity. Furthermore, at lower dilution rates, a

22

higher solvent concentration was obtained, whereas at higher dilution rates, acid (acetic

23

and butyric) concentation was increased. It is suggested that the residence time is

24

important for solvent production, where sufficient residence time is needed to permit the

16

ABE fermentation process to transfer into the solventogenic phase, and these results are

in accordance with the conclusion reported by Huang et al. (2004).

The final concentrations of acetic, butyric and residual sugar would increase as the

dilution rate was enhanced. In addition, the production of acetone and ehtanol was

slightly decreased with an increase of dilution rate. This phenomenon comfirmed the

sugar utilization rate and acids reassimilation rate would be decreased when a higher

dilution rate was performed. On the other hand, the different cell density (OD600) was

observed when dilution rates were controlled at different levels. Furthermore, the

average cell density was increased with an increase of the dilution rate. The maximum

10

average cell density was obtained at the dilution rate of 0.10 h-1. Therefore, it was

11

suggested that the higher dilution rate would be beneficial to cell growth, but in contrast

12

more solvents were achieved if the dilution rate was kept at a lower level.

13
14

3.5 Solvent tolerance test by butanol accumulation

15

Organic solvents are known to be terrifically toxic to microbial cells. The relationship

16

between cell morphology and organic solvent tolerance has been intensely investigated,

17

but it was mainly concentrated on the change of the cell size and specific surface area

18

(Neumann et al., 2005). To the best of our knowledge, literatures on the correlation

19

between biofilm and butanol tolerance in C. acetobutylicum has scarcely been reported

20

till date. Therefore, in order to investigate the effect of biofilm on butanol tolerance, the

21

strains SE25 and PW12 were grown in a development P2 agar plate medium (containing

22

20 g/L butanol) at 37 C for 48-60 h, and the morphological changes were observed

23

using SEM.

24

The results showed that the morphological changes of strains SE25 and PW12 could

17

be observed in the presence of butanol. Strain SE25 had a typical rod shape and their

cell surface was relatively smooth without membrane indentations. At the same time,

the biofilm was also found being kept intact without any fracture. However, when

strain PW12 was exposed to the same condition, some morphological changes could

be observed: (1) both cell surfaces became rough and membrane had indentations, and

(2) biofilm had been damaged or completely disappeared. These results clearly

confirmed that cells adapt towards stressful environment (butanol stress) through

modifying their phenotypic and physiological characteristics. The biofilm not only can

provide a good protection for cell, but also can be used as a permeation barrier which

10

controls the material enter or out of cells, and then enhances the ability of adapting

11

adverse environment. At the same time, the ability of adapting the butanol stress of

12

strain PW12 was much less than that of strain SE25 (date not show). Therefore,

13

according to the above results, the biofilm plays a key role in improving the ability of

14

butanol stress resistance for C. acetobutylicum, and thereby there is a close

15

relationship between biofilm and solvent (butnaol) tolerance.

16

To facilitate commerial ABE fermentation, enhancement of the butanol productivity

17

and reducing material cost are promising ways. In the present study, it was suggested

18

that cassava flour could be used as a perspective substrate for ABE fermentation. A

19

novel approach was developed to dispose of the problem of low butanol productivity,

20

followed by enhancing the solvent production in immobilized C. acetobutylicum SE25

21

with continuous production of ABE. Furthermore, due to the protection of the biofilm,

22

the butanol tolerance of the strain SE25 has been greatly improved compared with the

23

strain PW12. The above results showed that this novel and effective way for

24

immobilized cell fermentation is a promising technique for isolating high butanol

18

tolerance strains and improvement of butanol production.

2
3
4

4. Conclusion
The FBB system coupling continuous ABE fermentation has been explored to

enhance the butanol production. In summary, the highest butanol productivity of 0.63

g/L/h and butanol yield of 0.36 g/g-starch were achieved after the immobilization

carrier and the dilution rate were optimized, which were 231.6 % and 28.6 % higher

than those of free-cell fermentation. Therefore, this report demonstrated that the FBB

system could be an effective and promising approach for enhancing the butanol

10

productivity. Furthermore, it was also successfully comfirmed that the biofilm can

11

provide an effective protection for microbial cells when they were grown in stressful

12

environments.

13
14
15
16
17
18
19
20
21
22
23
24

19

Acknowledgements

This work was supported by the National Natural Science Foundation of China

(Grant No. 21466014), the Fundamental Research Funds for the Central Universities

(Grant No. JUSRP51504), the Doctoral Starting up Foundation of Jiangxi Agriculture

University (Grant No. 9232305387), and the Training Program of Innovation and

Entrepreneurship for Undergraduates from Jiangxi Agriculture University (Grant No.

201610410026).

8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

20

References

1.

Chauvatcharin, S.., C. Siripatana., T. Seki., M. Takagi., T. Yoshida. 1998.

Metabolism analysis and on-line physiological state diagnosis of acetone-butanol

fermentation. Biotechnol. Bioeng. 58(6): 561-571.

2.

Cheng, C.-L.., P.-Y. Che., B.-Y. Chen., W.-J. Lee., L.-J. Chien., J.-S. Chang. 2012.

High yield bio-butanol production by solvent-producing bacterial microflora.

Bioresour. Technol. 113: 58-64.

3.

Ezeji, T.., C. Milne., N. D. Price., H. P. Blaschek. 2010. Achievements and

perspectives to overcome the poor solvent resistance in acetone and

9
10

butanol-producing microorganisms. Appl. Microbiol. Biotechnol. 85(6):

11

1697-1712.

12

4.

Hua, X.., J. Wang., Z. Wu., H. Zhang., H. Li., X. Xing., Z. Liu. 2010. A salt tolerant

13

Enterobacter cloacae mutant for bioaugmentation of petroleum-and

14

salt-contaminated soil. Biochem. Eng. J. 49(2): 201-206.

15

5.

Huang, W.-C.., D. E. Ramey., S.-T. Yang. 2003. Continuous production of butanol

16

by Clostridium acetobutylicum immobilized in a fibrous bed bioreactor.

17

Proceedings of the Twenty-Fifth Symposium on Biotechnology for Fuels and

18

Chemicals, Held May 4-7, in Breckenridge, CO, Humana, Press, PP. 887-898.

19

6.

Keis, S.., R. Shaheen., D. T. Jones. 2001. Emended descriptions of Clostridium

20

acetobutylicum and Clostridium beijerinckii, and descriptions of Clostridium

21

saccharoperbutylacetonicum sp nov and Clostridium saccharobutylicum sp nov.

21

Int J Syst Evol Micr. 51: 2095-2103.

1
2

7.

Kilonzo, P.., A. Margaritis., M. Bergougnou. 2011. Effects of surface treatment and

process parameters on immobilization of recombinant yeast cells by adsorption to

fibrous matrices. Bioresour. Technol. 102(4): 3662-3672.

8.

Kittithanesuan, N.., M. Phisalaphong. 2015. Enhanced acetone-butanol production

from sugarcane juice by immobilized Clostridium acetobutylicum (ATCC 824) on

thin-shell silk cocoons. Biotechnol. Bioprocesss. Eng. 20(3): 599-607.

9.

Lepiz-Aguilar, L.., C. E. Rodriguez-Rodriguez., M. Laura Arias., G. Lutz. 2013.

Acetone-Butanol-Ethanol (ABE) Production in Fermentation of Enzymatically

9
10

Hydrolyzed Cassava Flour by Clostridium beijerinckii BA101 and Solvent

11

Separation. J. Microbiol. Biotechnol. 23(8): 1092-1098.

12

10.

Li, H.-g.., W. Luo., Q.-y. Gu., Q. Wang., W.-j. Hu., X.-b. Yu. 2013. Acetone,

13

butanol, and ethanol production from cane molasses using Clostridium beijerinckii

14

mutant obtained by combined low-energy ion beam implantation and

15

N-methyl-N-nitro-N-nitrosoguanidine induction. Bioresour. Technol. 137: 254-260.

16

11.

Li, H.-g.., W. Luo., Q. Wang., X.-b. Yu. 2014a. Direct fermentation of gelatinized

17

cassava starch to acetone, butanol, and ethanol using Clostridium acetobutylicum

18

mutant obtained by atmospheric and room temperature plasma. Appl. Biochem.

19

Biotechnol. 172(7): 3330-3341.

20
21

12.

Li, H.-g.., F. K. Ofosu., K.-t. Li., Q.-y. Gu., Q. Wang., X.-b. Yu. 2014b. Acetone,
butanol, and ethanol production from gelatinized cassava flour by a new isolates

22

with high butanol tolerance. Bioresour. Technol. 172: 276-282.

1
2

13.

Li, H.-g.., Q.-h. Zhang., X.-b. Yu., L. Wei., Q. Wang. 2016. Enhancement of

butanol production in Clostridium acetobutylicum SE25 through accelerating phase

shift by different phases pH regulation from cassava flour. Bioresour. Technol. 201:

148-155.

14.

Li, T.., Y. Yan., J. He. 2015. Enhanced direct fermentation of cassava to butanol by

Clostridium species strain BOH3 in cofactor-mediated medium. Biotechnol.

Biofuels. 8(1): 1-12.

15.

Loyarkat, S.., B. Cheirsilp., K. Umsakul. 2013. Direct Conversion of Sugars and

10

Organic Acids to Biobutanol by Non-growing Cells of Clostridium spp. Incubated

11

in a Nitrogen-Free Medium. Appl. Biochem. Biotech. 171(7): 1726-1738.

12

16.

Neumann, G.., Y. Veeranagouda., T. B. Karegoudar., . Sahin., I. Musezahl., N.

13

Kabelitz., U. Kappelmeyer., H. Heipieper. 2005. Cells of Pseudomonas putida and

14

Enterobacter sp. adapt to toxic organic compounds by increasing their size.

15

Extremophiles. 9(2): 163-168.

16

17.

Ni, Y.., Z. Xia., Y. Wang., Z. Sun. 2013. Continuous butanol fermentation from

17

inexpensive sugar-based feedstocks by Clostridium saccharobutylicum DSM

18

13864. Bioresour. Technol. 129: 680-685.

19

18.

Patakova, P.., M. Linhova., M. Rychtera., L. Paulova., K. Melzoch. 2013. Novel

20

and neglected issues of acetone-butanol-ethanol (ABE) fermentation by clostridia:

21

Clostridium metabolic diversity, tools for process mapping and continuous

23

fermentation systems. Biotechnol Adv. 31(1): 58-67.

1
2

19.

Qureshi, N.., H. P. Blaschek. 2000. Economics of butanol fermentation using

hyper-butanol producing Clostridium beijerinckii BA101. Food Bioprod Process.

78(C3): 139-144.

20.

Qureshi, N.., B. C. Saha., M. A. Cotta. 2007. Butanol production from wheat straw

hydrolysate using Clostridium beijerinckii. Bioprocess. Biosyst. Eng. 30(6):

419-427.

21.

Qureshi, N.., V. Singh., S. Liu., T. Ezeji., B. Saha., M. Cotta. 2014. Process

integration for simultaneous saccharification, fermentation, and recovery (SSFR):

9
10

Production of butanol from corn stover using Clostridium beijerinckii P260.

11

Bioresour. Technol. 154: 222-228.

12

22.

Silva, E. M.., S.-T. Yang. 1995. Kinetics and stability of a fibrous-bed bioreactor

13

for continuous production of lactic acid from unsupplemented acid whey. J.

14

Biotechnol. 41(1): 59-70.

15

23.

Survase, S. A.., A. van Heiningen., T. Granstrm. 2012. Continuous bio-catalytic

16

conversion of sugar mixture to acetone-butanol-ethanol by immobilized

17

Clostridium acetobutylicum DSM 792. Appl. Microbiol. Biotechnol. 93(6):

18

2309-2316.

19

24.

Thang, V.., K. Kanda., G. Kobayashi. 2010. Production of

20

AcetoneButanolEthanol (ABE) in Direct Fermentation of Cassava by

21

Clostridium saccharoperbutylacetonicum N1-4. Appl. Biochem. Biotech. 161(1-8):

24

157-170.

1
2

25.

Ting, C.., J. Wu., K. Takahashi., A. Endo., H. Zhao. 2012. Screened

Butanol-Tolerant Enterococcus faecium Capable of Butanol Production. Appl.

Biochem. Biotech. 168(6): 1672-1680.

26.

Yang, T.., Z. Rao., B. Kimani., M. Xu., X. Zhang., S.-T. Yang. 2015. Two-step

production of gamma-aminobutyric acid from cassava powder using

Corynebacterium glutamicum and Lactobacillus plantarum. J. Ind. Microbiol.

Biotechnol. 42(8): 1157-1165.

27.

Yen, H. W.., R. J. Li. 2011. The effects of dilution rate and glucose concentration

10

on continuous acetone-butanol-ethanol fermentation by Clostridium

11

acetobutylicum immobilized on bricks. J. Chem. Technol. Biotechnol. 86(11):

12

1399-1404.

13

28.

Yu, Y.., Q. Zhu., Q. Fang. 2010. Application of the ecological technology of

14

Aquamats artificial aquatic in the treatment of rural wastewater in Hangzhou.

15

Technology of Water Treatment. (05): 118-120.

16

29.

Zhang, H.., H. Chong., C. Ching., H. Song., R. Jiang. 2012. Engineering global

17

transcription factor cyclic AMP receptor protein of Escherichia coli for improved

18

1-butanol tolerance. Appl. Microbiol. Biotechnol. 94(4): 1107-1117.

19

30.

Zheng, J.., Y. Tashiro., T. Yoshida., M. Gao., Q. Wang., K. Sonomoto. 2013.

20

Continuous butanol fermentation from xylose with high cell density by cell

21

recycling system. Bioresour. Technol. 129: 360-365.

25

1
2
3

Figure captions

Fig. 1 Schematic diagram of the FBB system used for immobilized fermentation

Fig. 2 Performance comparisons of ABE fermentation using free cell (solid line) and in

immobilized cell (dashed line) FBB continuous reactor by SE25

Fig. 3 Curves of repeated-batch fermentation in FBB continuous reactor by SE25

Fig. 4 Effects of different dilution rates on ABE fermentation in FBB continuous reactor

9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

26

1
2
3
4

Fig. 1

5
6

27

1
2
3
4

Fig. 2

5
6

28

1
2
3
4

Fig. 3

5
6

29

2
3
4
5

Fig. 4

6
7

30

We consider the highlights of this paper as follows:

1. AquaMats AO was used as the immobilization carrier in ABE fermentation.

2. The butanol-producing ability of the producing strain SE25 was fairly stable.

3. The FBB system is a promising approach for enhancing the butanol productivity.

4. The correlation between the biofilm and solvent tolerance was investigated.

6
7

31