Agarose hydrolysis by two-stage enzymatic process and bioethanol production from the hydrolysate

Accepted Manuscript Title: Agarose hydrolysis by two-stage enzymatic process and bioethanol production from the hydrolysate Author: Young Bin Seo Juyi Park In Young Huh Soon-Kwang Hong Yong Keun Chang PII: DOI: Reference:

S1359-5113(16)30041-1 http://dx.doi.org/doi:10.1016/j.procbio.2016.03.011 PRBI 10643

To appear in:

Process Biochemistry

Received date: Revised date: Accepted date:

8-9-2015 12-3-2016 14-3-2016

Please cite this article as: Seo Young Bin, Park Juyi, Huh In Young, Hong Soon-Kwang, Chang Yong Keun.Agarose hydrolysis by two-stage enzymatic process and bioethanol production from the hydrolysate.Process Biochemistry http://dx.doi.org/10.1016/j.procbio.2016.03.011 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.

1

Highlights

2 3

● This is the first report on enzymatic agarose hydrolysis process development with no acid

4

pretreatment.

5

● In two-stage process, 88% of agarose was degraded to galactose and anhydrogalactose.

6

● The resulting hydrolysate was used for ethanol production without separation process.

7

● Bioethanol could be produced with no inhibition of cellular activity observed.

8

1

1

Agarose hydrolysis by two-stage enzymatic process and bioethanol production from the

2

hydrolysate

3 4

Young Bin Seoa,†, Juyi Parka, In Young Huha, Soon-Kwang Hongb, and Yong Keun

5

Changa,* a

6 7

Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon 305-701, Korea

8

b

9

†

Division of Bioscience and Bioinformatics, Myung-Ji University, Yongin 449-728, Korea

Current address: SK innovation co., 325, Exporo, Yuseong-gu, Daejeon, 305-712, Korea

10 11 12

*Corresponding author:

13

Tel: +82-42-350-3927, Fax: +82-42-350-3910

14

E-mail: [email protected]

15

2

1

Abstract

2

A two-stage enzymatic process was developed for agarose hydrolysis without acid

3

pretreatment. In the first stage, agarose was hydrolyzed to produce neoagarobiose using an

4

optimized dosage of AgaG1 and DagB at pH 7.0 and 40 °C. Agarose gelation was avoided by

5

momentarily elevating the reaction temperature for the first 10 min, instead of employing a

6

pretreatment step with acid prior to the enzyme reaction. In the second stage, neoagarobiose

7

was further hydrolyzed to produce galactose using neoagarobiose hydrolase (NABH) at pH

8

7.0 and 35 °C. The overall yield of galactose from agarose was 88% of the theoretical

9

maximum. The crude galactose solution produced from agarose hydrolysis was used directly

10

for ethanol production by yeast to demonstrate its potential. In overall, 20 g/L of agarose

11

could be converted to 3.71 g/L of ethanol.

12 13

Key words: agarose; enzymatic hydrolysis; galactose; bioethanol

14

3

1

1. Introduction

2 3

Ethanol can be produced by fermentation from various types of biomass including starch,

4

lignocelluloses, and marine algae biomass. The bioethanol produced from marine algae has

5

several advantages over that based on starch and lignocellulosic biomass. It does not compete

6

with human foods, does not cause forest denudation, has a low content of lignocellulose, and

7

can fix a larger amount of CO2 per unit mass [1,2].

8

The main constituent of red algae, one of the marine algae biomass, is agarose. Agarose is a

9

linear chain of alternating residues of 3-O-linked β-D-galactopyranose and 4-O-linked 3,6-

10

anhydro-α-L-galactose [3]. Agarose can be hydrolyzed chemically or enzymatically. During

11

chemical hydrolysis, galactose from the galactosyl residue is retained, while the

12

anhydrogalactose formed is further converted to 5-hydrooxymethyl furfural (5-HMF) and

13

then to levulinic acid (LA), both of which are known to be toxic to many microorganisms

14

including yeast to lower fermentation efficiency [4,5]. Therefore, the resulting crude

15

galactose solution could be used as the medium for ethanol production only after a

16

pretreatment step, for example, by nanofiltration and electrodialysis for their removal [6]. In

17

contrast, an enzymatic hydrolysis process does not produce 5-HMF and LA, and leaves 3,6-

18

anhydro-L-galactose intact. The resulting crude galactose solution, containing no toxic

19

compounds, can be used directly to produce ethanol by fermentation.

20

Enzymatic hydrolysis of agarose is known to include three steps, in principle: (1) agarose

21

liquefaction by β-agarase I to neoagaro-oligosaccharides; (2) neoagaro-oligosaccharides

22

degradation by β-agarase II to neoagaoriobiose; and (3) neoagarobiose decomposition by α-

23

neoagarobiose hydrolase (NABH) to galactose and anhydrogalactose [7].

4

1

In a series of efforts by our research group to search for enzymes needed for agarose

2

degradation, the first involved AgaG1, screened from Alteromonas sp. GNUM1. This enzyme

3

is an endo-type β-agarase, which hydrolyzes agarose to neoagarobiose and neoagarotetroase.

4

It was expressed in E. coli BL21 [8-10]. The second enzyme, DagB, an exo-type β-agarase

5

which hydrolyzes agarose and neoagarotetraose to neoagarobiose, was screened from S.

6

coelicolor A3(2) and expressed in S. lividans TK24 [11]. Finally, a newly screened microbial

7

strain of Alcanivorax sp. A28-3 (KCTC12788BP) was found to produce NABH, although its

8

amino acid sequence is yet to be identified. When neoagarobiose was treated with the cell

9

lysate of Alcanivorax sp. A28-3, galactose and anhydrogalactose were produced as the final

10

products [12].

11

In this study, a two-stage process for the saccharification of agarose to galactose and

12

anhydrogalactose was developed and optimized. In the first step, agarose was degraded to

13

neoagarobiose using a mixture of AgaG1 and DagB, produced by a recombinant E. coli and a

14

recombinant S. lividans, respectively. In the second step, neoagarobiose was decomposed to

15

galactose and anhydrogalactose using the cell lysate of Alcanivorax sp. A28-3. The resulting

16

galactose solution was used by Saccharomyces cerevisiae KL17 (KFCC11493P), a galactose-

17

utilizing yeast strain newly screened by our group for ethanol production.

18

5

1

2. Materials and methods

2 3

2.1. Strains and media

4

E. coli BL21 and pMoPac1 were used as the host and plasmid for the expression of agaG1

5

gene, respectively [10]. E. coli BL21/pMoPac1-AgaG1 was grown on the Luria-Bertani (LB)

6

medium containing 35 mg/L of chloramphenicol.

7

S. lividans TK24, which was obtained from The John Innes Foundation (United Kingdom)

8

and pUWL201PW [13] were used as the host and vector for the expression of DagB,

9

respectively. The S. lividans TK24/pUWL201PW-DagB culture was performed in a R2YE

10

medium with no sucrose, containing per liter: 0.25 g K2SO4, 10.12 g MgCl2·6H2O, 10 g

11

glucose, 0.1 g casamino acids, 5 g yeast extract, 10 ml of 0.5% K2HPO4, 80 ml of 3.68%

12

CaCl2·2H2O, 15 ml of 20% L-proline, 100 ml of 5.73% N-tris(hydroxymethyl)methyl-2-

13

aminoethanesulfonic acid (TES; pH 7.2), and 2 ml of trace elements solution [14]. For stable

14

maintenance of the plasmid, 50 mg/L of thiostrepton was used [11].

15

Alcanivorax sp. A28-3 (KCTC12788BP) was used for the production of NABH. The strain

16

was cultured in an artificial sea water media (ASW-YP) containing per liter 6.1 g Trizma base

17

(pH 7.2), 12.3 g MgSO4, 0.74 g KCl, 0.13 g (NH4)2HPO4, 17.5 g NaCl, 0.14 g CaCl2, 0.2 g

18

yeast extract, 3.0 g bacto peptone, and 2.0 g agar.

19

Saccharomyces cerevisiae KL17 (KFCC11493P) was used for the production of ethanol, and

20

was cultured in an YPG medium containing per liter: 10 g yeast extract, 20 g peptone, and 10

21

g galactose, or in an YPEH medium containing per liter: 10 g yeast extract and 20 g peptone,

22

and actual enzyme hydrolysate of agarose.

23

6

1

2.2. Preparation of enzyme solutions

2

2.2.1 AgaG1

3

E. coli BL21/pMoPac1-AgaG1 culture was performed in baffled Erlenmeyer flask. Cells

4

from the cell stock were used to inoculate 10 ml of LB medium for the seed culture. The

5

culture was performed at 37 °C and 200 rpm. After overnight culture, 2% of the seed culture

6

broth was inoculated into 200 ml of LB medium. When the culture OD600 reached 1.2, the

7

temperature was decreased to 18 °C and then AgaG1 expression was induced using 1.0 mM

8

of IPTG. The culture was performed for 8 h at 18 °C after induction. The harvested cells were

9

disrupted by using a homogenizer (Model 500, Fisher scientific, USA) to recover AgaG1

10

inside the cells. Cell debris was removed by centrifugation for 20 min at 5000 g-force. The

11

supernatant was used for the hydrolysis of agarose.

12 13

2.2.2 DagB

14

S. lividans TK24/pUWL201PW-DagB was grown in a baffled Erlenmeyer flask. Cells from

15

the cell stock were used to inoculate 10 ml of R2YE medium for the seed culture. The culture

16

was performed at 28 °C and 200 rpm. After 2 days of seed culture, 2% of the seed culture

17

broth was inoculated into 200 ml of R2YE medium. After 2 days of culture, cells were

18

removed by centrifugation for 20 min at 5000 g. The supernatant was used for the hydrolysis

19

of neoagarotetraose.

20 21

2.2.3 NABH

22

Alcanivorax sp. A28-3 was cultivated in an Erlenmeyer flask. Cells from the cell stock were

23

used to inoculate 10 ml of ASW-YP medium for the seed culture at 28 °C and 200 rpm. After

24

1 day of seed culture, 2% of the seed culture broth was inoculated into 200 ml of ASW-YP 7

1

medium. After 2 days of culture, the harvested cells were disrupted by using a homogenizer

2

(Model 500, Fisher scientific, USA) to recover NABH inside the cells. Cell debris was

3

removed by centrifugation for 20 min at 5000 g. The supernatant was used for the hydrolysis

4

of neoagarobiose.

5 6

2.3. Two-stage hydrolysis of agarose

7

In the first stage, 10 or 20 g/L of agarose (Lonza, Switzerland) was hydrolyzed to

8

neoagarobiose by using a mixture of AgaG1 and DagB, both of which had the same optimum

9

condition of pH 7.0, and 40 °C [10,11]. The enzyme reaction was performed pH 7.0, 40 °C,

10

and 100 rpm for 12 h in an Erlenmeyer flask containing 100 mL of reaction mixture. Tris-HCl

11

buffer at 50 mM was used. To avoid the problem of agarose gelation, the reaction was started

12

at 46 °C. The temperature was lowered to 40 °C after 10 min, when gelation was no longer a

13

problem. In the second stage, neoagarobiose produced in the first stage was treated with the

14

NABH solution at pH 7.0, 35 °C, and 100 rpm for 12 h in a flask containing 100 ml of

15

reaction mixture. Tris-HCl buffer was used.

16 17

2.4. Ethanol production

18

Saccharomyces cerevisiae KL17 (KFCC11493P) was used. It had been newly screened from

19

the soil by our group, and exhibited efficient uptake of galactose to produce ethanol [15].

20

Seed culture was carried out in 10 ml of the YPG medium. The main culture was inoculated

21

with 2% seed culture broth. As required, 100 ml of YPG medium, YPEH medium, or YPGH

22

(YPG media + 0.5% of 5-HMF) medium was used for the main culture. All the yeast cultures

23

were carried out at 30 °C and 200 rpm.

24

8

1

2.5. Analysis

2

The agarase activity was measured by the dinitrosalicylic acid (DNS) method as previously

3

described [10]. One unit of enzyme activity was defined for both AgaG1 and DagB as the

4

amount of enzyme liberating 1 μmole of reducing sugar per minute, at pH 7.0 and 40 °C. One

5

unit of NABH activity was defined as the amount of enzyme liberating 1 μmole of galactose

6

per minute, at pH 7.0 and 35 °C. Neoagarobiose and neoagarotetraose were quantified by

7

using a high-performance liquid chromatograph (HPLC) (Waters, USA), with an Asahipak

8

NH2P-50 4E column (250 × 4.6 mm, Shodex, Japan) and an ELSD detector (Sedex 75,

9

Sedere, France) under conditions described previously [16]. Galactose and ethanol were

10

quantified by using an YSI 2700 Select Biochemistry Analyzer (YSI, USA). The agarose

11

conversion to neoagarobiose and neoagarobiose conversion to galactose, were estimated from

12

the amount of these two compounds produced by the reaction using the following conversion

13

factors.

14

The amount of agarose converted to neoagarobiose:

15

agarose converted (g) = neoagarobiose produced (g) / fbiose/agarose

16

where fbiose/agarose = 1.06 (= 324 / 306).

17

Eq. (1)

The amount of neoagarobiose converted to galactose:

18

neoagarobiose converted (g) = galactose produced (g) / fgalactose/biose

19

where fgalactose/biose = 0.56 (= 180 / 324).

9

Eq. (2)

1

3. Results and discussion

2 3

3.1. Preparation of enzyme solutions

4

By cultivating E. coli BL21/pMoPac1-AgaG1, an intracellular AgaG1 activity of 5.9 U/ml-

5

broth was obtained. The cell lysate, prepared as described earlier, was concentrated about 10

6

fold by ultrafiltration and then a required amount was used for each experiment. From S.

7

lividans TK24/pUWL201PW-DagB, an extracellular DagB activity of 0.75 U/ml was

8

obtained. The cell-free broth was ultra-filtered to concentrate the DagB, before being used for

9

experiments. By cultivating Alcanivorax sp. A28-3, an intracellular NABH activity of 0.025

10

U/ml-broth was obtained. The cell lysate, prepared as described earlier, was concentrated

11

about 20 fold before being used for experiments.

12 13

3.2. Optimization of AgaG1 and DagB dosage for neoagarobiose production

14

As mentioned earlier, the final products of AgaG1 from agarose hydrolysis were

15

neoagarotetraose and neoagarobiose, while that from DagB was neoagarobiose only [10,11].

16

Theoretically, agarose can be hydrolyzed to neoagarobiose only by DagB. However, the main

17

problem with DagB is that it (an exo-type enzyme) is not efficient in degrading oligomers

18

because it attacks only one end of the chain. On the other hand, AgaG1 (an endo-type

19

enzyme) was found more efficient than DagB in attacking oligomers larger than

20

neoagarotetraose at the same dosage, as shown in Fig. 1. For this reason, the use of a mixture

21

of these two enzymes was considered an efficient way to produce neoagarobiose from

22

agarose and at the same time, to minimize the dosage of DagB, which is very expensive to

23

produce. In such systems, the main contribution of AgaG1would be the conversion of agarose

24

to neoagarotetraose, which would be subsequently degraded to neoagarobiose by DagB. 10

1

Fortunately, such simultaneous usage of these enzymes was possible because both of them

2

were found to have the same optimum conditions (pH 7.0 and 40 °C) [10,11].

3

In determining the dosages of these two enzymes required for complete hydrolysis of 10 g/L

4

agarose to neoagarobiose in 12 h, it was assumed for the convenience of analysis that agarose

5

hydrolysis occurred in the following sequence:

6

Assumption 1) Neoagarotetraose is generated from agarose by AgaG1, and

7

Assumption 2) DagB degraded neoagarotetraose to produce neoagarobiose.

8

First, the kinetics of neoagarotetraose hydrolysis by DagB was identified. The rate of

9

neoagarotetraose hydrolysis was measured for different concentrations of DagB (Fig. 2a).

10

The neoagarotetraose degradation rate was observed to follow first-order kinetics (Fig. 2b) as

11

represented by Eq. (3).

12

d[N4] / dt = –f([E]) [N4]

Eq. (3)

13

where [N4] is concentration of neoagarotetraose, and f([E]) the first-order reaction rate

14

coefficient as a function of DagB, [E].

15

The equation was integrated with [N4] = [N4]0 at t = 0 to give

16

ln[N4] = ln[N4]0 – f([E])t

Eq. (4)

17

By plotting the experimental data, based on Eq. (4) (Fig. 2b), f([E]) was found to linearly

18

dependent on DagB dosage as presented in Fig. 2c and Eq. (5).

19

f([E]) = 0.021 + 2.02 [E] (hr-1)

Eq. (5)

20

In the previous work, 10 g/L of agarose was completely degraded by AgaG1 to 3.8 g/L and

21

6.4 g/L of neoagarobiose and neoagarotetraose, respectively [10]. This meant that 6.4 g/L of

22

neoagarotetraose needed to be hydrolyzed by DagB when dealing with 10 g/L agarose. Using

23

Eq. (4) and (5), the required DagB dosage for the complete hydrolysis of 6.4 g/L

24

neoagarotetraose in 12 h was calculated to be 0.133 U/ml. 11

1

Second, different dosages of AgaG1, along with 0.133 U/ml of DagB, were applied to

2

degrade 10 g/L of agarose to the final product, neoagarobiose. As shown in Fig. 3, the

3

neoagarobiose production rate increased monotonically as the AgaG1 dosage increased (in

4

the range tested: 0.375–1.500 U/ml). However, the dosage effect was negligible at > 1.250

5

U/ml. When the AgaG1 dosage was 1.25 or 1.50 U/ml, the agarose hydrolysis was nearly

6

complete in 9 h. With 0.875 U/ml of AgaG1, the reaction ended at 12 h, whereas with 0.375

7

U/ml and 0.625 U/ml of AgaG1, the reaction was not completed in 12 h. This was clear from

8

a noticeable amount of neoagarotetraose remaining in the reaction mixture (Table 1). Unless

9

otherwise mentioned, 0.875 U/ml of AgaG1 and 0.133 U/ml of DagB were used in the

10

subsequent experiments.

11 12

3.3. Two-stage process of enzymatic hydrolysis of agarose

13

In the first stage, 0.875 U/mL AgaG1 and 0.133 U/mL DagB were used for the hydrolysis of

14

10 g/L agarose, at pH 7.0 and 40 °C for 12 h. In the second stage, NABH was used for the

15

degradation of neoagarobiose to galactose and anhydrogalactose, under the optimum

16

conditions of pH 7.0 and 35 °C [12]. Fig. 4 shows the time profiles of galactose,

17

neoagarobiose, and neoagarotetraose during the first stage, when 9.7 g/L of neoagarobiose

18

was produced from 10 g/L agarose. This indicated that 92% of the agarose added was

19

converted to neoagarobiose. In the second stage, the galactose concentration increased

20

monotonically with reaction time, and its final concentration at 12 h was 5.2 g/L. This

21

corresponded to 88% of the galactose yield from agarose.

22

In the two-stage enzymatic process, the most critical step was the first stage of agarose

23

liquefaction (primarily done by AgaG1), lowering the viscosity or gelation tendency of the

24

agarose solution. It was reported that an increase in the solution viscosity caused decreasing 12

1

enzyme activity due to decreased motion of the enzymes in solution [17]. To avoid gelation

2

or high-viscosity problems, Kim et al. proposed an agarose pretreatment method with mild

3

acid (e.g., acetic acid) to degrade agarose partially before enzymatic hydrolysis [18].

4

However, it is known that acids cleave agarose bonds randomly [19], generating products

5

which β-agarases cannot recognize, and which could cause a significant decrease in yield.

6

Moreover, the use of acid entails a neutralization step to adjust the pH to the optimum level

7

for the subsequent step of enzymatic hydrolysis. Such neutralization generates a significant

8

amount of salts, which could potentially inhibit microbial activity in the subsequent

9

fermentation step. In this study, it was found that enzymatic hydrolysis of agarose could be

10

performed with no acid pretreatment. Instead, the gelation problem was avoided by keeping

11

the temperature at 45 °C for 10 min at the very beginning of the reaction in the first stage

12

considering the gelling temperature is known to be 43–45 °C. Then, the temperature was

13

lowered and maintained at 40 °C for the rest of the reaction. It was expected that agarose

14

molecules were randomly cleaved mainly by the action of AgaG1 with a rapid decrease in the

15

viscosity and thus gelation tendency during the 10-min period of operation at 45 °C [10]. In

16

other words, the gelation problem could be avoided at the cost of briefly sacrificing enzyme

17

activity, which was expected to cause no significant loss in the overall efficiency of the

18

process. In fact, the AgaG1 activity was only about 5% less than that at the optimum

19

temperature of 40 °C. Moreover, operation at 45 °C for a period as short as 10 min was

20

expected to result in no significant permanent deactivation of AgaG1. It was found in the

21

previous study that AgaG1 activity could recover to about 95% of its original level after

22

exposure to 45 °C for 10 min [10].

23

In this study, after the NABH treatment a galactose yield of 88% was obtained, which was

24

significantly higher than that (79%) reported by the group using acid pretreatment [18]. In 13

1

addition, they calculated the yield based on the data obtained by the DNS method only, with

2

no actual measurement of the galactose concentration. This carried a high probability of

3

overestimation.

4 5

3.4. Ethanol production

6

For the preparation of agarose hydrolysate for ethanol production, a higher concentration of

7

agarose (20 g/L) was used to obtain a more concentrated galactose solution than before with

8

10.0 g/L galactose. Other than galactose, the other required nutrients (such as yeast extract)

9

were added to the enzyme hydrolysate of agarose to make YPEH medium for ethanol

10

production by Saccharomyces cerevisiae KL17. The galactose concentration in the YPEH

11

medium (9.2 g/L) was slightly lower than that of the original agarose hydrolysate solution

12

(10.0 g/L) due to the dilution caused by the nutrient supplementation step. For the purpose of

13

comparison, a synthetic galactose medium containing 10.0 g/L galactose (YPG) and YPG

14

containing 5 g/L 5-HMF (YPGH) were also tested for ethanol production. As shown in Fig. 5,

15

the galactose consumption rate and ethanol production rate in the YPEH medium showed

16

patterns similar to those of the YPG medium, but showed a pattern of high inhibition in the

17

YPGH medium. For the YPG medium, 9.75 g/L of galactose was consumed in 12 h to

18

produce 3.85 g/L of ethanol, providing an ethanol yield of 0.39 (g/g) (77% of theoretical

19

yield). When the YPEH medium was used, 8.76 g/L of galactose was completely metabolized

20

to produce 3.71 g/L of ethanol with an ethanol yield of 0.42 (g/g) (83% of theoretical yield).

21

As an example, yeast culture was performed using the crude galactose solution produced by

22

the two-stage enzymatic process, to demonstrate its usability as a carbon source for the

23

fermentation processes. As shown in Fig. 5, the YPEH medium prepared from the crude

24

galactose solution was found to offer a good environment for the growth of Saccharomyces 14

1

cerevisiae KL17, and for ethanol production. In the YPEH medium, galactose was

2

metabolized completely, and the ethanol yield was 83% of the theoretical yield, while in YPG,

3

a synthetic medium made from pure galactose, the yield was 77% of theoretical. The higher

4

amount of ethanol produced with YPEH medium than YPG medium was considered due to

5

its complex nature. It was believed to contain, as originated from a crude hydrolysate solution,

6

various unidentified materials that might have been utilized by the yeast as carbon source in

7

addition to galactose. The possibility that the existence of anhydrogalactose in the YPEH

8

medium could have positively contributed to ethanol production was slim because

9

Saccharomyces cerevisiae is known to be unable to metabolize anhydrogalactose [7].

10

Both galactose consumption and ethanol production significantly declined in YPGH medium

11

due to the toxic effects of 5-HMF, one of the major toxic byproducts from the chemical

12

hydrolysis of agarose (Fig. 5). For this reason, the chemical hydrolysate of agarose needed to

13

undergo a detoxification step before being used for ethanol fermentation [6], while enzyme

14

hydrolysate containing no inhibitory byproducts, and could be used directly, as done in this

15

study.

16

Galactose has several applications. The ongoing and most important one is for the synthesis

17

of D-tagatose, a low-calorie sweetener [20]. Galactose is also used for the synthesis of

18

isopropyl β-D-1-thiogalactopyranoside also [21]. As mentioned earlier, anhydrogalactose is

19

formed together with galactose during enzymatic hydrolysis. Unfortunately, there have been

20

no reports of anhydrogalactose utilization by microorganisms, although indirect evidence was

21

observed by our group that some marine microorganisms metabolize it (data not given). No

22

cases of anydrogalactose application have been reported, either. It would be a significant

23

contribution to enhanced efficiency of substrate utilization if genetically engineered

24

microorganisms could be developed that metabolize anhydrogalactose.

25

15

1

4. Conclusions

2

A two-stage enzymatic process was developed and optimized for agarose hydrolysis using a

3

mixture of AgaG1 and DagB in the first stage, and NABH in the second stage. In the first

4

stage, the agarose gelation problem was avoided by an adjustment of reaction temperature

5

requiring no pretreatment with acid. Agarose could be hydrolyzed to galactose with a yield of

6

88% of theoretical maximum. The crude galactose solution produced from agarose hydrolysis

7

was used directly for ethanol production by yeast, without cell-growth inhibition, to

8

demonstrate its potential as a carbon source for fermentation.

9 10

Acknowledgements

11

This work was supported by the New & Renewable Energy Technology Development

12

Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP)

13

grant funded by the Korean government Ministry of Knowledge Economy (No.

14

2009301009001B) and the Advanced Biomass R&D Center (ABC) of the Global Frontier

15

Project funded by the Ministry of Education, Science and Technology (ABC-2014-047211).

16 17

16

1

References

2 3

[1] Field, C.B., Behrenfeld, M.J., Randerson, J.T., Falkowski, P. 1998. Primary production of

4

the biosphere: Integrating terrestrial and oceanic components. Science, 281(5374),

5

237-240.

6 7 8 9

[2] Smith, S.V. 1981. Marine macrophytes as a global carbon sink. Science, 211(4484), 838840. [3] Duckwort.M, Yaphe, W. 1971. Structure of Agar .1. Fractionation of a complex mixture of polysaccharides. Carbohydrate Research, 16(1), 189-197.

10

[4] Klinke, H.B., Thomsen, A.B., Ahring, B.K. 2004. Inhibition of ethanol-producing yeast

11

and bacteria by degradation products produced during pre-treatment of biomass.

12

Applied Microbiology and Biotechnology, 66(1), 10-26.

13 14

[5] Yaphe, W. 1959. Effect of acid hydrolyzates of agar on the growth of Escherichia coli. Canadian Journal of Microbiology, 5(6), 589-93.

15

[6] Kim, J.H., Na, J.G., Yang, J.W., Chang, Y.K. 2013. Separation of galactose, 5-

16

hydroxymethylfurfural and levulinic acid in acid hydrolysate of agarose by

17

nanofiltration and electrodialysis. Bioresource Technology, 140, 64-72.

18

[7] Chi, W.J., Chang, Y.K., Hong, S.K. 2012. Agar degradation by microorganisms and agar-

19

degrading enzymes. Applied Microbiology and Biotechnology, 94(4), 917-930.

20

[8] Chi, W.J., Park, D.Y., Seo, Y.B., Chang, Y.K., S.Y., L., Hong, S.K. 2014. Cloning,

21

expression, and biochemical characterization of a novel GH16 β-agarase AgaG1 from

22

Alteromonas sp. GNUM-1. Applied Microbiology and Biotechnology, 98(10), 4545-

23

4555.

24

[9] Kim, J., Hong, S.K. 2012. Isolation and characterization of an agarase-producing bacterial 17

1

strain, Alteromonas sp. GNUM-1, from the west sea, Korea. Journal of Microbiology

2

and Biotechnology, 22(12), 1621-1628.

3

[10] Seo, Y.B., Lu, Y., Chi, W.J., Park, H.R., Jeong, K.J., K., H.S., Chang, Y.K. 2014.

4

Heterologous expression of a newly screened β-agarase from Alteromonas sp.

5

GNUM1 in Escherichia coli and its application for agarose degradation. Process

6

Biochemistry, 49(3), 430-436.

7

[11] Temuujin, U., Chi, W.J., Chang, Y.K., Hong, S.K. 2012. Identification and biochemical

8

characterization of Sco3487 from Streptomyces coelicolor A3(2), an exo- and endo-

9

type beta-agarase-producing neoagarobiose. Journal of Bacteriolgy, 194(1), 142-149.

10

[12] Seo, Y.B. 2014. Enzymes screening and process development for agarose hydrolysis. Ph.

11

D. Thesis, Korea Advanced Institute of Science and Technology, Daejeon, South

12

Korea.

13

[13] Doumith, M., Weingarten, P., Wehmeier, U.F., Salah-Bey, K., Benhamou, B., Capdevila,

14

C., Michel, J.M., Piepersberg, W., Raynal, M.C. 2000. Analysis of genes involved in

15

6-deoxyhexose biosynthesis and transfer in Saccharopolyspora erythraea. Molecular

16

and General Genetics, 264(4), 477-485.

17 18

[14] Kieser H, B.M., Buttner MJ, Chater FK, Hopwood DA. 2000. Practical Streptomyces genetics. The John Innes Foundation, Norwich.

19

[15] Kim, J.H., Ryu, J., Huh, I. Y., Hong, S.K., Kang, H.A., Chang, Y.K. 2014. Ethanol

20

production from galactose by a newly isolated Saccharomyces cerevisiae KL17.

21

Bioprocess and Biosystems Engineering, 37, 1871-1878.

22

[16] Kazlowski, B., Pan, C.L., Ko, Y.T. 2008. Separation and quantification of neoagaro- and

23

agaro-oligosaccharide products generated from agarose digestion by beta-agarase and

24

HCl in liquid chromatography systems. Carbohydrate Research, 343(14), 2443-2450. 18

1 2

[17] Jacob, M., Schmid, F.X. 1999. Protein folding as a diffusional process. Biochemistry, 38(42), 13773-13779.

3

[18] Kim, H.T., Lee, S., Kim, K.H., Choi, I.G. 2012. The complete enzymatic saccharification

4

of agarose and its application to simultaneous saccharification and fermentation of

5

agarose for ethanol production. Bioresource Technology, 107, 301-306.

6

[19] Yang, B., Yu, G.L., Zhao, X., Jiao, G.L., Ren, S.M., Chai, W.G. 2009. Mechanism of mild

7

acid hydrolysis of galactan polysaccharides with highly ordered disaccharide repeats

8

leading to a complete series of exclusively odd-numbered oligosaccharides. Febs

9

Journal, 276(7), 2125-2137.

10

[20] Rhimi, M., Chouayekh, H., Gouillouard, I., Maguin, E., Bejar, S. 2011. Production of D-

11

tagatose, a low caloric sweetener during milk fermentation using

12

isomerase. Bioresource Technology, 102(3), 3309-3315.

L-arabinose

13

[21] Carlsson, U., Ferskgard, P.O., Svensson, S.C.T. 1991. A simple and efficient synthesis of

14

the inducer IPTG made for inexpensive heterologous protein-production using the lac-

15

promoter. Protein Engineering, 4(8), 1019-1020.

16

19

1

Figure Legends

2

Fig 1. Agarose degradation by AgaG1 or DagB for 9 hours.

3

Fig 2. Effect of DagB dosage on neoagarotetraose degradation rate at pH 7.0 and 40 °C. a.

4

neoagarotetraose concentration profiles b. semi-log plot of neoagarotetraose concentration

5

profiles c. dependency of reaction rate coefficient on DagB dosage

6

Fig 3. Effect of AgaG1 dosage on neoagarobiose formation from agarose at pH 7.0 and 40 °C

7

at a fixed DagB dosage of 0.133 U/ml.

8

Fig 4. Two-stage enzymatic hydrolysis of agarose: In the first stage, 10 g/L of agarose was

9

hydrolyzed to neoagarobiose by using AgaG1 and DagB mixture at pH 7.0 and 40 °C. In the

10

second stage, neoagarobiose was hydrolyzed to galactose by using NABH at pH 7.0 and

11

35 °C.

12

Fig 5. Ethanol production by Saccharomyces cerevisiae KL17. YPG: medium containing

13

synthetic galactose (control), YPEH: medium based on enzymatic hydrolysate, YPGH:

14

medium containing galactose and 5-HMF.

15 16

Table Legend

17

Table 1. Neoagarobiose and neoagarotetraose formation from agarose hydrolysis by DagB

18

and AgaG1.

19

20

1

Table 1. AgaG1 dosage (U/ml)

DagB dosage (U/ml)

Reaction time (h)

Neoagarobiose produced (g/L)

Neoagarotetraose produced (g/L)

0.375

0.133

12

9.56 ± 0.25

0.58 ± 0.05

0.625

0.133

12

9.84 ± 0.02

0.30 ± 0.02

0.875

0.133

12

9.97 ± 0.03

0.12 ± 0.00

1.250

0.133

9

10.02 ± 0.02

0.12 ± 0.02

1.500

0.133

9

10.05 ± 0.00

0.16 ± 0.02

2

21

*Graphical Abstract

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5