A kinetics modeling study on the inhibition of glucose on cellulosome of Clostridium thermocellum

Bioresource Technology 190 (2015) 36–43

Contents lists available at ScienceDirect

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

A kinetics modeling study on the inhibition of glucose on cellulosome of Clostridium thermocellum Pengcheng Zhang, Buyun Wang ⇑, Qunfang Xiao, Shan Wu School of Environmental Engineering, Wuhan Textile University, Wuhan 430073, PR China

h i g h l i g h t s  A kinetics model was built to study the inhibition of glucose on cellulosome.  An experiment was built to verify the kinetics model.  Glucose was proven to be an inhibitor for cellulosome by kinetics study.  Parameters in model were discussed.  Adsorbent for glucose could enhance cellulose hydrolysis.

a r t i c l e

i n f o

Article history: Received 15 February 2015 Received in revised form 11 April 2015 Accepted 13 April 2015 Available online 22 April 2015 Keywords: Cellulose hydrolysis Kinetics Glucose Cellulosome Inhibition

a b s t r a c t A simplified kinetics model was built to study the inhibition of glucose on cellulosome of Clostridium thermocellum. Suitable reaction conditions were adopted to evaluate the model. The model was evaluated at different temperatures and further with various activated carbon additions as adsorbent for glucose. Investigation results revealed that the model could describe the hydrolysis kinetics of cellulose by cellulosome quite well. Glucose was found to be an inhibitor for cellulosome based on the kinetics analysis. Inhibition increased with the increase in temperature. Activated carbon as adsorbent could lower the inhibition. Parameters in the model were further discussed based on the experiment. The model might also be used to describe the strong inhibition of cellobiose on cellulosome. Saccharification of cellulose by both cellulosome and C. thermocellum could be enhanced efficiently by activated carbon addition. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Enzymatic hydrolysis was one of the key steps for biochemical conversion of lignocellulosic biomass to biofuel and bioproducts using the sugar platform (Demain et al., 2005). Aerobic cellulase had been widely used in application and was investigated repeatedly (Philippidis et al., 1993). Aerobic cellulase could not degrade lignocellulose individually (Dashtban et al., 2009). Some anaerobic bacteria could produce a large enzyme complex called cellulosome (Lynd et al., 2002; Waeonukul et al., 2012) which contained a variety of enzymes and could degrade lignocelluloses alone (Demain et al., 2005; Lynd et al., 2002; Tachaapaikoon et al., 2012). However, not as aerobic cellulase, kinetics study on cellulosome was very few because of the intricate structure and catalytic mechanism. Existing kinetics studies mainly focused on the relationship between bacterial growth and glucose accumulation (Linville et al., ⇑ Corresponding author. Tel.: +86 15071321619. E-mail address: [email protected] (B. Wang). http://dx.doi.org/10.1016/j.biortech.2015.04.037 0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.

2013) which could not described the relationship between cellulosome and products accumulation directly and were inefficient in conditions optimizing. Clostridium thermocellum was one of the most well-known and potent anaerobic cellulolytic bacteria which produced cellulosome (Demain et al., 2005). Cellulosome of C. thermocellum had been studied repeatedly in gene (Koeck et al., 2013; Lee et al., 2010; Waeonukul et al., 2012), molecular structure (Sajjad et al., 2012; Ye et al., 2010) and improvement (Islam et al., 2013; Reed et al., 2014; Waeonukul et al., 2013). Because cellulosome contained many enzymes and might be affected by the end-products such as glucose, cellobiose and ethanol simultaneously (Berlin et al., 2006; Ellis et al., 2012; Waeonukul et al., 2012), study on inhibition of end-products on cellulosome was difficult. So, inhibition of endproducts on cellulosome was normally evaluated by tolerance method (Berlin et al., 2006; Linville et al., 2013; Waeonukul et al., 2012). However, tolerance method could only demonstrate the inhibition indirectly. The demand on kinetics study which could demonstrate the inhibition directly was eager.

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

37

In this experiment, inhibition of glucose, an end-product, on cellulosome from C. thermocellum was studied. A mathematical model developed for the cellulase in simultaneous saccharification and fermentation process was modified to illustrate the inhibition. An elaborately programmed experiment was developed to evaluate the feasibility of the model.

culture medium contained 4% carbon resource and other compositions were the same as seed medium. Cellulose hydrolysis medium contained 7% carbon resource and other compositions were the same as seed medium. All incubations were carried out in anaerobic condition and 120 rpm at 60 °C which was the optimum temperature for growth and hydrolysis.

2. Methods

2.3. Preparation of cellulosome

2.1. Organism, enzyme and materials

2.3.1. Preparation of crude cellulosome Cellulosome was prepared from producing cellulosome culture medium with 4% filter paper as carbon resource for 3 days at 60 °C. Cell-free supernatant (prepared by centrifugation at 3000 rpm for 15 min) was precipitated with ammonium sulfate to remove some impurities. Then, a crude cellulosome enzyme pellet was obtained by centrifugation at 10,000 rpm, 4 °C for 15 min. Crude enzyme was dissolved in sodium phosphate buffer (pH 7.0, in minimum volume). The concentrated crude enzyme was then recovered with dialysis and concentration (with polyethylene glycol 20,000, 4 °C).

A hyper cellulolytic strain C. thermocellum ATCC 27405 was provided kindly by Huazhong University of Science and Technology. The strain had been adapted for growth on lignocellulose (stalk). B-glucosidase (9031-11-2, 345 IU/mL, purified) was kindly provided by Clean Energy Laboratory, Huazhong University of Science and Technology. Ramie (Boehmeria nivea) stalk which was the residual decorticated stem of ramie was from farm in Xianning City, Hubei Province, China. It was cut to 1  5 cm stick with chaff cutters and dried under the sun for 7 days. Composition of ramie stalk (by NREL method) consisted of 37.78%, 15.77%, 29.85% and 4.27% for cellulose, hemicellulose, lignin and moisture respectively. Filter paper was 1  6 cm Whatman No. 1 filter paper. Cotton yarn was 10 cm absorbent cotton yarn. Cotton yarn was washed by distilled water and dried, and then washed by alcohol, acetone and 75% alcohol sequentially, and then washed by distilled water and dried. Pitch-based activated carbon (AC) was kindly provided by Clean Energy Laboratory, Huazhong University of Science and Technology. The nominal glucose adsorption capacity of AC was 75 mg/g. Woody biochar (0.5 mm in diameter, sieved) was the by-production of gas production through a fast pyrolysis of sawdust at 800 °C in cyclone furnace with a 0.35 water flow equivalence ratio. Biochar was washed in 2% glucose solution with 2% w/v at 60 °C, 200 rpm for 3 h. After solid/liquid separation, liquid phase was removed and biochar was treated with another batch of glucose solution again. This procedure was repeated till glucose content in liquid phase remained constant. Biochar was then washed by deionized water till no glucose could be detected. Characteristics of AC and treated biochar (BC) were 1514 and 10.12 m2/g in SBET, 0.61 and 0.016 cm3/g in Vt, 0.44 and 0.0032 cm3/g in Vmic and 2.43 and 7.23 nm in Dp, respectively. 2.2. Media and growth conditions for C. thermocellum Seed medium was used for the proliferation of C. thermocellum. Producing cellulosome culture medium was used for producing cellulosome by C. thermocellum. Cellulose hydrolysis medium was used to investigate the hydrolysis of cellulose by C. thermocellum. C. thermocellum could hydrolyzate lignocellulosic biomass as carbon resource. In this experiment, carbon resource was cotton yarn, filter paper and ramie stalk. To eliminate lag phase, carbon resource in seed medium would be the same as that in producing cellulosome culture medium or cellulose hydrolysis medium. For example, to produce cellulosome on filter paper, strain would grow on seed medium with filter paper as carbon resource. After population growth in log phase for 4 h, 5 mL seed medium was incubated into producing cellulosome culture medium to produce cellulosome for kinetics investigation. Incubation from seed medium to cellulose hydrolysis medium was used to investigate the hydrolysis of cellulose. Seed medium contained 0.1% carbon resource, 0.05% KH2PO4, 0.001% FeSO47H2O, 0.05%, K2HPO4, 0.1% yeast extract, 0.02% MgSO47H2O, 0.01% peptone, 0.5% CaCO3 and 100% distilled water. PH of medium was adjusted to 7.0. The producing cellulosome

2.3.2. Characterization of cellulosome Cellulosome sample for characterization was prepared from crude cellulosome with Waeonukul’s method (2009). The concentrated crude enzyme preparation (20 mg) was applied onto a Sephacryl S-300 high-resolution (Amersham Biosciences, Piscataway, NJ, USA) column (0.9  48.5 cm), which was equilibrated with 50 mM phosphate buffer (pH 6.0) containing 150 mM NaCl and was eluted with the same buffer with a flow rate of 0.5 mL/min. The combined active fractions of the first peak (10 mg) from the gel filtration column were then applied onto a Avicel column (2 cm  8 cm), which was equilibrated with phosphate buffered saline (PBS) (250 mM phosphate buffer pH 6.0 containing 200 mM NaCl). After the cellulose-binding proteins were washed four times with a large amount of the same buffer until no more proteins were found in the eluent, the cellulosome was eluted with 1% (v/v) triethylamine (Sigma). The active fractions were combined and dialyzed. All isolation steps were carried out at 4 °C. Then, native molecular mass of cellulosome was estimated by mass spectrum. 2.3.3. Preparation of cellulosome for kinetics investigation Cellulosome sample for kinetics investigation was prepared from 3 g crude cellulosome by ion exchange chromatography using macroporous cation exchange resin 001X7 (in 4 cm  80 cm column). Elution was carried out in phosphate buffer (pH 4.0–10.0, 12 h and 5 mL/min). Protein was determined with Coomassie Brilliant Blue G250 (per 25 mL eluent). Cellulosome was determined by the reaction with filter paper (DNS method) and mass spectrum. Cellulosome sample was then recovered with dialysis and concentration. 2.4. Cellulosomal hydrolysis of cotton yarn Kinetics investigation on hydrolysis of cellulose by cellulosome was carried out on cotton yarn in 1 L sodium phosphate buffer (pH 7.0) containing 70 g cotton yarn and 2 g cysteine. Enzyme loading was 237 FPAU (U = glucose production in mg/min with 1 mL cellulosome) cellulosome and 2760 IU b-glucosidase. The solution was preserved at 4 °C for 1 h with rotation occasionally. Then, hydrolysis was carried out at 50, 55 and 60 °C with 160 rpm for 8 min. Glucose concentration in hydrolysate was detected every minute. Inhibition of glucose on cellulosome was evaluated by a mathematical model developed for the simultaneous saccharification and fermentation process. To confirm the inhibition, hydrolysis was also carried out with different amounts of AC and BC as adsorbent for glucose at 50 °C. In kinetics investigation, AC and BC were

38

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

DC Dt

sealed in PES semipermeable membrane with pore in 100 nm. Unit of AC was 1 g per package and that of BC was 3 g.

m¼

2.5. Adsorption of glucose onto AC and BC

where C was the glucose concentration. With the factors influencing reaction, m being:

To confirm the adsorption ability of AC and BC for glucose, adsorption investigation was carried out in 5 g/L glucose phosphate buffer solution (pH 7.0) at 50 °C, 120 rpm. Pseudo-first-order (1) and pseudo-second-order (2) equations were used to study the kinetics. Langmuir (3) and Freundlich (4) equations were used to evaluate the adsorption equilibrium.

log ðqe  qt Þ ¼ log qe  k1 t=2:303

ð1Þ

t=qt ¼ 1=k2 q2e þ t=qe

ð2Þ

where k was the equilibrium rate constant, qe was the amount of glucose adsorbed at equilibrium, qt was the amount of glucose on the surface of the adsorbent at time t (mg/g). In kinetics investigation, adsorbent loading was 1 g/L.

1=qe ¼ 1=kl qm ce þ 1=qm

ð3Þ

log qe ¼ logkf þ 1=n  log ce

ð4Þ

where qe was amount of adsorbate adsorbed per specific amount of adsorbent (mg/g), ce was equilibrium concentration (mg/L), qm was amount of adsorbate required to form a monolayer (mg/g), kl, kf and 1/n were equation parameters. 2.6. Adsorbent effect on saccharification of cellulose by C. thermocellum To confirm the enhancement of adsorbent on saccharification of cellulose by C. thermocellum, different amounts of AC was added into cellulose hydrolysis medium. Incubation was carried out in facultative anaerobic condition (removing the seal for 0.5 h every 4 h). After incubation for 72 h, saccharification efficiency was investigated. 2.7. Calculation

m¼

kc at uec KE    ð1  K L LÞ ðke þ ec Þ 1 þ KBB þ KC K E þ E

ð7Þ

where at was the surface area of cellulose available for cellulosome adsorption; u was the reactivity coefficient of cellulose; kc was the specific rate of cellulose hydrolysis; ke was the equilibrium constant for cellulosome adsorption to cellulose; KL was constant for cellulosome and b-glucosidase adsorption to lignin; and KB, K, and KE were the inhibition constants for cellulosome and b-glucosidase by cellobiose (B), glucose (C), and ethanol (E) respectively. In this investigation, no fermentation for ethanol would happen and the reaction was carried out on pure cellulose. Expression (7) could be simplified as:

A

m¼

1 þ kBB þ KC



ð8Þ

where A was:

A¼

kc at uec ðke þ ec Þ

ð9Þ

In a short reaction period, enzyme deactivation was assumed negligible and ec could be looked as a constant. Likewise, at, u, kc and ke were also thought to be constants. Because kc was expressed as the biomass loss in reaction, A would be a negative constant in the kinetics model. Definitely, absolute value of A would decrease with increase in ke if adsorption of cellulosome onto cellulose became effective. Being cell-free, cellobiose was released from cellulosome directly into hydrolysate. With excess b-glucosidase, cellobiose could be assumed negligible. So, expression (8) could be further simplified as:

m¼

Because high deviation was produced in weight loss method, only FPA of cellulosome and saccharification of cellulose were calculated by weight loss method (for five repeated detection). Kinetics and adsorption investigations were perform by glucose concentration with DNS method (in triple experiments, deviation lower than 5%). Pseudo-first-order calculation was performed by SPSS 21.0 and other calculations were carried out in EXCEL 2010.

ð6Þ

A 1 þ KC

ð10Þ

Eq. (10) could further be converted to:

C¼K

KA

m

ð11Þ

Based on Eqs. (6 and 11), inhibition constant (K) for glucose on cellulosome and constant A could be obtained from the slope of the straight line, which plotted based on C versus 1/m.

3. Results and discussion

3.2. Enzyme and cellulose biomass

3.1. Kinetics modeling of cellulose hydrolysis by cellulosome

The purified cellulosome prepared by gel filtration was estimated by mass spectrum to be 1540 kDa according to Waeonukul’s method (2009). The molecular mass was higher than that in some researches, probably due to more subunits of the enzyme (Demain et al., 2005), because the bacterium was initially adapted on stalk. Though gel filtration could purify cellulosome enough for characterization investigation, purified cellulosome production in this method was too low for kinetics investigation. The purified cellulosome for kinetics investigation was prepared by ion exchange chromatography. After crude enzyme was applied onto ion exchange chromatography, three peaks containing protein appeared (data not shown) and proteins in two peaks (the first and second) exhibited enzymatic activity. The second peak (also the major peak at pH 5.6) was proofed to be cellulosome by mass determination (1540 kDa) and used for the concentration of cellulosome. Mass spectrometry revealed that there was no other

Existing models focused on the cellulose hydrolysis (Lynd et al., 2002), glucose accumulation and cellobiose accumulation (Mosier et al., 1999). No kinetics model had been developed to study the inhibition of products on cellulosome. The mathematical model developed for the simultaneous saccharification and fermentation process (Philippidis et al., 1993) was used to support further simplification in this investigation. The portion of this model that described the kinetics of cellulose hydrolysis consisted of the following mass balance equation:

dC c ¼ m dt

ð5Þ

where Cc was the cellulose mass at time t; m was the rate of the hydrolytic action of cellulosome (ec). Within a short reaction time, m could also be described as the variation in glucose concentration:

39

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

protein in the cellulosome sample prepared from ion exchange chromatography. Optimum temperature and pH for cellulosome were 60 °C and 7.0 respectively. Thermostability (enzymatic activity P95%) of cellulosome could last for 2 h at 60 °C. After concentration, FPA of cellulosome was 296.02 U. The high temperature resistant b-glucosidase was produced by Aspergillus oryzae and its optimum temperature and pH were 50 °C and 7.0. Though thermostability (enzymatic activity P95% by pNPG) of b-glucosidase could only last for 30 min at 60 °C, it met the requirement for kinetics investigation. In some investigations, avicel and CMC were chosen as cellulose substrate (Shao et al., 2011). However, avicel and CMC were not the cellulose used in application at all. Cotton, paper and stalk constituted the majority of cellulosic biomass in application (Lynd et al., 2002). The high cellulose content in cotton and paper could completely satisfy the requirement of kinetics investigation. Stalk was another important lignocellulosic biomass in application. Ramie stalk as a biomass waste in textile industry was easy to collect and produced in great amount. However, the complex compositions made stalk not suitable for kinetics investigation because lignin and the hydrolysates of hemicellulose would pose serious influence. So, ramie stalk was used only in the bacterial hydrolysis investigation. Major products of cellulosome on cellulose consisted of cellobiose and cellodextrins (Demain et al., 2005; Ye et al., 2010). Cellobiose was a strong inhibitor for cellulosome (Waeonukul et al., 2012) and would affect kinetics investigation heavily. If cellulosome posed low activity on cellulose, cellobiose production would also be low. So, with excess b-glucosidase, cellobiose in reaction could be converted into glucose in time and the impact of cellobiose on kinetics investigation could be eliminated. Hydrolysis of cotton yarn by cellulosome produced on filter paper was suitable for kinetics investigation. Because crystallinity of cotton was much higher than that of paper, cellulosome produced on filter paper exhibited low efficiency on cotton yarn. Also, for the high crystallinity of cotton, glucose produced by b-glucosidase directly from cellulose was negligible in the short reaction time (Demain et al., 2005; Kumar and Wyman, 2009). Thus, all reducing sugar in reaction could be thought as glucose which was converted completely from the cellobiose broken down from cotton yarn by cellulosome. 3.3. Hydrolysis kinetics at various temperatures Glucose concentrations at various hydrolysis times were presented in Table 1. Relationship between C and 1/m at various temperatures were presented in Fig. 1 and the calculated parameters of the investigation fitting to the Eq. (11) were also presented in Table 1. With high relative coefficient, the presented kinetics model could describe the inhibition of glucose on cellulosome well. K and A values were presented in Table 1. K decreased and A increased with the increase in temperature. 3.4. Hydrolysis kinetics with various amounts of adsorbent additions 3.4.1. Adsorption of glucose onto AC and BC Based on the research on structure of cellulosome, inhibition of glucose on cellulosome was thought to be at least partly caused by

the adsorption of glucose onto the active site which was responsible for decomposition of cellulose (product competitive inhibition) (Lynd et al., 2002). Theoretically, competition with other adsorbents could lower the adsorption of glucose onto cellulosome and consequently lower the inhibition of glucose on cellulosome. In this experiment, AC and BC were used as adsorbent for glucose and the adsorption characteristics were presented in Table 2. Qes were lower than the experimental values. It made pseudo-firstorder equation not suitable to describe the adsorption kinetics. Though the Qes were much higher than experimental values, pseudo-second-order equation might be used to describe the adsorption on AC and BC with high R2S. Adsorption equilibrium might be described by Langmuir and Freundlich equations. Because high 1/ns met the long time to build equilibrium, Freundlich equation might be more suitable than Langmuir equation. AC and BC might lower the inhibition of glucose on cellulosome by competition. At the same time, because AC and BC needed much more time to build adsorption equilibrium for glucose than that of kinetics investigation, deviation in glucose concentration produced by adsorption of glucose onto AC or BC was negligible. Thus, AC and BC might be suitable adsorbents to study the hydrolysis kinetics. Adsorption of cellulosome onto AC or BC might produce heavy impact. PES semipermeable membrane which could prevent enzymes from adsorbing onto Ac or BC was used to seal AC and BC. 3.4.2. Hydrolysis with AC and BC Hydrolysis kinetics investigation with AC and BC as adsorbent for glucose was performed. The hydrolysis without adsorbent was used as control. Results were presented in Table 3 and Fig. 2. In the kinetics investigation, C and 1/m were highly correlated with each other. As those in investigation at various temperatures, calculated K and A values varied with the variation in amount of adsorbent. K increased and A decreased with the increase in amount of AC. 3.5. Parameters in kinetics model 3.5.1. K K was named as inhibition constant for cellulosome by glucose in the model. According to the modeling of Eq. (7) and the principle of competitive inhibition, K was definitely highly related to dissociation of glucose from cellulosome. An infinite K meant no inhibition. However, inhibition happened when a K value could be achieved from the kinetics model. Former researches could only demonstrate the inhibition indirectly by tolerance of cellulosome to glucose (Mosier et al., 1999). In this experiment, inhibition was confirmed directly. Value of K was not the glucose concentration which would inhibit reaction completely. However, K could be used to evaluate inhibition in statistics. Among a series of experiments under different conditions, smaller K meant that the dissociation between glucose and cellulosome was more difficult and consequently, inhibition was stronger. As an inhibitor, glucose must be adsorbed onto cellulosome to produce inhibition. Higher temperature would lead to more effective adsorption of glucose onto cellulosome. So, in Table 1, K decreased with higher temperature. It meant that glucose would inhibit cellulosomal activity

Table 1 Glucose (mg/L) with reaction time (min) at various temperatures and calculated parameters. Time (min)

1

2

3

4

5

6

7

8

R2

K (mg/L)

A

50 °C 55 °C 60 °C

28.64 ± 0.71 29.31 ± 0.21 32.34 ± 0.46

51.91 ± 0.37 53.19 ± 0.27 54.52 ± 0.39

71.56 ± 0.48 73.23 ± 0.52 74.85 ± 0.13

87.69 ± 0.55 88.46 ± 0.93 89.14 ± 0.02

101.84 ± 0.41 104.95 ± 1.02 105.00 ± 0.19

113.75 ± 0.32 114.37 ± 0.87 115.24 ± 0.11

123.32 ± 0.31 124.14 ± 0.74 124.93 ± 0.15

131.29 ± 0.72 133.91 ± 1.11 134.11 ± 0.07

0.87 0.84 0.90

11.26 10.01 9.51

95.69 113.82 121.03

40

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

Fig. 1. Relationship between C and 1/m at various temperatures.

Table 2 Adsorption of glucose onto AC and BC. Adsorption kinetics qt (mg/g)

t (min) AC BC

30 46.3 27.5

Pseudo-first-order equation

60 48.5 29

90 49.2 30.2

120 50.8 31.4

150 52.1 32.2

180 52.8 32.7

210 53.5 32.8

240 54

270 54.2

Pseudo-second-order equation

qe (mg/g)

k1

R2

qe (mg/g)

k2

R2

53.21 32.41

1.18 1.44

0.71 0.75

344.83 204.08

4.70E-04 8.29E-04

0.99 0.99

Adsorption equilibrium qe (mg/g)

Loading (g/L) AC Loading (g/L) BC

1 54.2 3 32.8

Langmuir equation

2 53.2 6 31.7

3 53 9 31

4 52.7 12 30.6

5 52.5 15 30.4

Freundlich equation

qm (mg/g)

kl

R2

1/n

kf

R2

163.93

9.91E-05

0.87

0.679

0.176

0.87

1428.57

4.74E-06

0.92

0.99

0.007

0.92

Table 3 Glucose (mg/L) with reaction time (min) at various adsorbent additions (/L) and calculated parameters. Time (min) *

Control 1 g AC 2 g AC 3 g AC 4 g AC 5 g AC BC** * **

1

2

3

4

5

6

7

8

R2

K

A

28.64 ± 0.37 29.63 ± 0.23 29.61 ± 0.56 29.59 ± 0.01 29.57 ± 0.41 29.57 ± 0.12 29.53 ± 0.51

51.91 ± 0.24 52.74 ± 0.21 52.77 ± 0.42 52.64 ± 0.02 52.62 ± 0.38 52.59 ± 0.05 52.67 ± 0.74

71.56 ± 0.21 72.69 ± 0.09 72.61 ± 0.37 72.49 ± 0.10 72.47 ± 0.21 72.41 ± 0.07 72.58 ± 1.41

87.69 ± 0.24 89.93 ± 0.18 89.83 ± 0.33 89.73 ± 0.05 89.69 ± 0.37 89.63 ± 0.07 89.81 ± 1.26

101.84 ± 1.27 102.92 ± 0.38 102.92 ± 0.25 102.81 ± 0.04 102.73 ± 0.25 102.62 ± 0.05 102.90 ± 2.37

113.75 ± 0.21 115.54 ± 0.11 115.49 ± 0.35 115.40 ± 1.95 115.23 ± 0.38 115.13 ± 0.18 115.14 ± 5.21

123.32 ± 3.24 124.49 ± 0.98 124.45 ± 1.13 124.24 ± 0.75 124.15 ± 2.33 123.96 ± 0.45 124.37 ± 0.77

131.29 ± 1.42 132.80 ± 0.33 132.74 ± 1.28 132.51 ± 1.72 132.34 ± 1.54 132.15 ± 1.22 132.55 ± 4.29

0.87 0.87 0.87 0.87 0.87 0.87 0.87

11.26 13.71 13.88 14.03 14.11 14.34 13.64

95.69 77.60 76.39 75.16 74.51 72.87 77.91

The hydrolysis without adsorbent. BC addition was 15 g/L.

more at higher temperature. However, it could be also noticed that glucose concentration increased with increase in temperature. Definitely, K was only a factor influencing reaction rate and glucose accumulation was also determined by other factors. On the other hand, factors lowered adsorption of glucose onto cellulosome could reduce inhibition. Competition with other adsorbents would make the adsorption difficult. So, K values increased with the increase in AC addition (Table 3). Inhibition of glucose on cellulosome was lowered by adsorbent. With more AC, inhibition was lowered more according to K value. Because adsorption capacity of BC was lower than that of AC, improvement in treatment with BC was low. All the glucose productions with AC were higher than that in control. However, glucose accumulation

in solution decreased slightly with increase in AC. It might be caused by adsorption of glucose onto AC. The more, ratio of K to C might also illustrate inhibition. According to this viewpoint, glucose inhibition on cellulosome was much higher than that on fungal cellulase and a little lower than that on b-glucosidase (Philippidis et al., 1993). It met the usually observed result that very low glucose accumulation could be achieved in hydrolysate of cellulose by anaerobic bacterium (Ellis et al., 2012). However, for the complex components in cellulosome, relationship between glucose and cellulosome should be more intricate than that on fungal cellulase. More researches should be carried out to achieve the statistical analysis on the ratio.

41

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

160

C (mg/L)

120

Contorl 5g AC 4g AC 3g AC 2g AC 1g AC BC

80

40

0 0

0.05

0.1

0.15

1/v (min L/mg)

Fig. 2. Relationship between C and 1/m with various adsorbent additions (/L, BC addition was 15 g/L, control was the hydrolysis without adsorbent).

Because cellulosome of C. thermocellum broke down cellulose with cellobiose as main product and the extracellular b-glucosidase activity was low (Demain et al., 2005), the model could also describe the cellobiose inhibition approximately when hydrolysis happened without b-glucosidase (removing b-glucosidase in Section 2.4). The result was presented in Table 4. As repeatedly proven, cellobiose was a strong inhibitor for cellulosome according to the Ks. The low cellobiose concentration and K both revealed that cellobiose was a much stronger inhibitor compared with glucose. However, to describe the inhibition of cellobiose on cellulosome accurately using the model in this paper, more elaborately experiments should be programmed to eliminate the impact by other products such as glucose and cellodextrins in the future.

3.5.2. A According to Eq. (9), value of A depended on the surface area of cellulose available for cellulosome adsorption (at), the reactivity coefficient of cellulose (u), the specific rate of cellulose hydrolysis (kc), the hydrolytic action of cellulosome (ec) and the equilibrium constant for cellulosome adsorption to cellulose (ke). Definitely, A was an indicator of the available cellulose for cellulosome. As K, A was an indicator in statistics. The greater the absolute value of A, the more cellulose was available for cellulosome. In application, methods increasing available cellulose for cellulosome such as excess cellulose substrate loading and pretreatment were usually adopted to accelerate hydrolysis by C. thermocellum (Arantes and Saddler, 2011; Pallapolu et al., 2011; Shao et al., 2011; Waeonukul et al., 2012). If characteristics of cellulose remained constant, A was definitely highly related to the cellulosome adsorption to cellulose (ke). Decrease in ke would lead to increase in absolute value of A, and vice versa. Because it was thought to be adsorbed onto the same site for cellulose in cellulosome (Demain et al., 2005), glucose would definitely lower the adsorption of cellulose to cellulosome. When adsorbent for glucose was added, adsorption of cellulosome onto cellulose would increase. It would lead to the decrease in the absolute value of A (Table 3). When temperature increased, adsorption of glucose on cellulosome increased, adsorption of cellulosome

onto cellulose would be weakened. It would lead to the increase in the absolute value of A (Table 1). According to Eq. (11), at a specific hydrolysis rate (m), less accumulated glucose concentration meant more cellulosome adsorption to cellulose, i.e., less available cellulose for cellulosome, and vice versa. It was illustrated clearly in Figs. 1 and 2. In other words, to keep the cellulosome adsorption to cellulose and hydrolysis rate effective, glucose concentration must be lowered. In application, impact of glucose on reaction between cellulose and cellulosome would be aggravated for the high glucose content. Existing efforts mainly focused on the pretreatment on cellulose substrate (Lee et al., 2011), i.e., increasing the absolute value of A. According to kinetics investigation, definitely, removing glucose from hydrolysate in time could also increase the absolute value of A and improve the reaction rate. It should be also a feasible method in application. Though cellobiose could be utilized in other methods, hydrolysis of cellulose by cellulosome for ethanol production was mainly using glucose platform. Existing models normally focused on the glucose accumulation in hydrolysis of cellulose. These models could described the relationship between glucose production and time or cellulosome concentration directly. However, they normally required a long experimental time and were inefficient in condition optimization. The model in this paper could not predict the final glucose accumulation. However, with the parameters, this model could determine the optimum reaction conditions in a short time. So, combination of former models and investigation in this paper should be efficient in improving utilization of cellulose in application.

3.5.3. C and m Reaction rate had an inverse proportionality with glucose concentration according to Eq. (11). It meant that the increase in C would lead to the decrease in m which was clearly illustrated in Figs. 1 and 2. Because A was considered as a constant at the initial reaction stage and K was a constant, m was mainly influenced by C. In Fig. 1, it could be found that decrease in reaction rate was faster at high temperature than that at low temperature according to the

Table 4 Cellobiose (mg/L) with reaction time (min) at various temperatures and calculated parameters. Time (min)

1

2

3

4

5

6

7

8

R2

K (mg/L)

A

50 °C 55 °C 60 °C

3.25 ± 0.12 3.47 ± 0.12 3.64 ± 0.13

6.14 ± 0.11 6.32 ± 0.14 6.53 ± 0.12

8.32 ± 0.17 8.47 ± 0.21 8.63 ± 0.01

10.21 ± 0.09 10.38 ± 0.11 10.54 ± 0.16

11.87 ± 0.02 12.11 ± 0.13 12.29 ± 0.07

13.31 ± 0.21 13.41 ± 0.04 13.54 ± 0.01

14.29 ± 0.02 14.45 ± 0.02 14.64 ± 0.17

15.07 ± 0.24 15.21 ± 0.01 15.41 ± 0.01

0.76 0.78 0.79

3.26 3.53 3.57

3.30 3.00 3.05

42

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43

slopes. Though an increased initial reaction rate could be achieved at high temperature according to the initial glucose concentration in Table 1, reaction rate would decrease quickly with the fast glucose accumulation. Final glucose production was highly related to variation in reaction rate. In application, temperature should be investigated in detail according to the kinetics. On the other hand, according to Fig. 2, adsorbent made the decrease in reaction rate slow down. It made increase in glucose accumulation faster than that in control (Table 3). Adsorbent could improve the hydrolysis definitely. However, different from K and A, the relationship between C and m (Fig. 2) became irregular because of the adsorption caused by adsorbent. 3.6. Cellulose saccharification by C. thermocellum with adsorbent addition As adsorbent, AC and BC, could lower the glucose concentration in hydrolysate. According to the analysis of the kinetics model, it might enhance the hydrolysis greatly by lowering the inhibition of glucose and improving the adsorption of cellulose onto cellulosome. Experiment was designed to eliminate side reactions such as ethanol and CO2 fermentation (Chinn et al., 2008; Geng et al., 2010; Islam et al., 2013) (50 °C and facultative anaerobic condition). Results were presented in Table 5. It could be found that with adsorbent, hydrolysis of cellulose could increase greatly. This method was also effective for lignocellulose (ramie stalk). Adsorbent addition could be a feasible method to enhance the hydrolysis and reducing sugar production in application. Adsorption capacity of BC was lower than that of AC. Consequently, enhancements by BC were lower than those by AC. However, because AC was expensive, BC might be feasible in application. 3.7. Some key points (1) A kinetics model was developed to study the inhibition of glucose on cellulosome. It revealed the relationship between glucose concentration and hydrolysis rate at a specific glucose concentration or a specific time. This relationship was described by Eq. (10). Eq. (11) was used to facilitate calculation. It did not mean that glucose accumulation in a long term depended on the hydrolysis rate at a specific time. (2) Aim of this paper was to study the inhibition of glucose on cellulosome by kinetics. However, value of K was not an indicator in quantity. In practical application, end-products should be evaluated together to find the major inhibitor.

Table 5 Glucose* production (C g/L) and weight loss (wl%) in hydrolysis by C. thermocellum with adsorbent additions (/L).

Control** 1 g AC 2 g AC 3 g AC 4 g AC 5 g AC 15 g BC

* **

C Wl C Wl C Wl C Wl C Wl C Wl C Wl

Ramie stalk

Cotton yarn

Filter paper

2.08 ± 0.02 7.26 ± 1.24 2.53 ± 0.03 9.29 ± 0.95 2.98 ± 0.13 9.87 ± 1.25 3.69 ± 0.11 10.99 ± 1.17 3.71 ± 0.02 12.43 ± 1.26 4.15 ± 0.11 12.39 ± 2.09 3.52 ± 0.14 11.37 ± 1.01

1.85 ± 0.01 4.95 ± 0.21 2.28 ± 0.02 6.27 ± 0.15 2.87 ± 0.02 7.75 ± 0.09 3.38 ± 0.01 9.37 ± 0.04 4.01 ± 0.03 11.57 ± 0.10 4.07 ± 0.01 14.15 ± 0.07 3.29 ± 0.04 13.24 ± 0.07

0.37 ± 0.03 2.25 ± 0.47 0.68 ± 0.02 3.19 ± 0.51 0.74 ± 0.03 4.17 ± 0.32 1.25 ± 0.02 5.07 ± 0.15 1.49 ± 0.01 5.89 ± 1.12 1.74 ± 0.03 6.34 ± 0.76 1.43 ± 0.15 5.42 ± 0.88

All the reducing sugar was calculated as glucose. Without adsorbent addition.

(3) A was thought to be a constant in this model. However, A would vary if the experiment lasted a long time or cellulose was not ample. In application, attention should be paid to this point. (4) Adding glucose might be also helpful to investigate the kinetics. However, diffusion rate of glucose might be a heavy influence factor. So, investigation was not carried out in this way. 4. Conclusion By kinetics study, glucose was proven to be an inhibitor for cellulosome produced by C. thermocellum, an anaerobic bacterium. With high glucose concentration, impact of glucose on cellulosome would aggravate. Glucose concentration should be an important factor affecting glucose accumulation rate in hydrolysis of cellulose by anaerobic bacterium. Besides the treatments increasing the surface area of cellulose available for cellulosome, methods that could decrease the inhibition of glucose on cellulosome might be effective in enhancing cellulose saccharification by anaerobic bacterium. Acknowledgement This research was supported by Hubei NFC 2014CFB758. References Arantes, V., Saddler, J.N., 2011. Cellulose accessibility limits the effectiveness of minimum cellulase loading on the efficient hydrolysis of pretreated lignocellulosic substrates. Biotechnol. Biofuels 4, 3–16. Berlin, A., Balakshin, M., Gilkes, N., Kadla, J., Maximenko, V., Kubo, S., Saddler, J., 2006. Inhibition of cellulase, xylanase and b-glucosidase activities by softwood lignin preparations. J. Biotechnol. 125, 198–209. Chinn, M.S., Nokes, S.E., Strobel, H.J., 2008. Influence of moisture content and cultivation duration on Clostridium thermocellum 27405 end-product formation in solid substrate cultivation on Avicel. Bioresour. Technol. 99, 2664–2671. Dashtban, M., Schraft, H., Qin, W., 2009. Fungal bioconversion of lignocellulosic residues; opportunities & perspectives. Int. J. Biol. Sci. 5, 578–595. Demain, A.L., Newcomb, M., Wu, J.H.D., 2005. Cellulase, Clostridia, and Ethanol. Microbiol. Mol. Biol. Rev. 69, 124–154. Ellis, L.D., Holwerda, E.K., Hogsett, D., Rogers, S., Shao, X., Tschaplinski, T., Thorne, P., Lynd, L.R., 2012. Closing the carbon balance for fermentation by Clostridium thermocellum (ATCC 27405). Bioresoure Technol. 103, 293–299. Geng, A., He, Y., Qian, C., Yan, X., Zhou, Z., 2010. Effect of key factors on hydrogen production from cellulose in a co-culture of Clostridium thermocellum and Clostridium thermopalmarium. Bioresoure Technol. 101, 4029–4033. Islam, R., Özmihçi, S., Cicek, N., Sparling, R., Levin, D.B., 2013. Enhanced cellulose fermentation and end-product synthesis by Clostridium thermocellum with varied nutrient compositions under carbon-excess conditions. Biomass Bioenergy 48, 213–223. Koeck, D.E., Wibberg, D., Koellmeier, T., Blom, J., Jaenicke, S., Winkler, A., Albersmeier, A., Zverlov, V.V., Pühler, A., Schwarz, W.H., Schlüter, A., 2013. Draft genome sequence of the cellulolytic Clostridium thermocellum wild-type strain BC1 playing a role in cellulosic biomass degradation. J. Biotechnol. 168, 62–63. Kumar, R., Wyman, C.E., 2009. Access of cellulase to cellulose and lignin for poplar solids produced by leading pretreatment technologies. Biotechnol. Prog. 25, 807–819. Lee, C.Y., Yu, K.O., Kim, S.W., Han, S.O., 2010. Enhancement of the thermostability and activity of mesophilic Clostridium cellulovorans EngD by in vitro DNA recombination with Clostridium thermocellum CelE. J. Biosci. Bioeng. 109, 331– 336. Lee, J.M., Venditti, R.A., Jameel, H., Kenealy, W.R., 2011. Detoxification of woody hydrolyzates with activated carbon for bioconversion to ethanol by the thermophilic anaerobic bacterium Thermoanaerobacterium saccharolyticum. Biomass Bioenergy 35, 626–636. Linville, J.L., Rodriguez Jr, M., Mielenz, J.R., Cox, C.D., 2013. Kinetic modeling of batch fermentation for Populus hydrolysate tolerant mutant and wild type strains of Clostridium thermocellum. Bioresoure Technol. 147, 605–613. Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorius, I.S., 2002. Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66, 506– 577. Mosier, N., Hall, P., Ladisch, C., Ladisch, M., 1999. Reaction kinetics, molecular action, and mechanisms of cellulolytic proteins. In: Tsao, G.T., Brainard, A.P., Bungay, H.R., Cao, N.J., Cen, P., Chen, Z., Du, J., Foody, B., Gong, C.S., Hall, P., Ho, N.W.Y., Irwin, D.C., Iyer, P., Jeffries, T.W., Ladisch, C.M., Ladisch, M.R., Lee, Y.Y.,

P. Zhang et al. / Bioresource Technology 190 (2015) 36–43 Mosier, Mühlemann, H.M., Sedlak, M., Shi, N.Q., Tsao, G.T., Tolan, J.S., Torget, R.W., Wilson, D.B., Xia, L. (Eds.), 2010, Recent Progress in Bioconversion of Lignocellulosics, Vol. 65. Springer, Berlin Heidelberg, pp. 23–40. Pallapolu, V.R., Lee, Y.Y., Garlock, R.J., Balan, V., Dale, B.E., Kim, Y., Mosier, N.S., Ladisch, M.R., Falls, M., Holtzapple, M.T., Sierra-Ramirez, R., Shi, J., Ebrik, M.A., Redmond, T., Yang, B., Wyman, C.E., Donohoe, B.S., Vinzant, T.B., Elander, R.T., Hames, B., Thomas, S., Warner, R.E., 2011. Effects of enzyme loading and b-glucosidase supplementation on enzymatic hydrolysis of switchgrass processed by leading pretreatment technologies. Bioresour. Technol. 102, 11115–11120. Philippidis, G.P., Smith, T.K., Wyman, C.E., 1993. Study of the enzymatic hydrolysis of cellulose for production of fuel ethanol by the simultaneous saccharification and fermentation process. Biotechnol. Bioeng. 41, 846–853. Reed, P.T., Izquierdo, J.A., Lynd, L.R., 2014. Cellulose fermentation by Clostridium thermocellum and a mixed consortium in an automated repetitive batch reactor. Bioresour. Technol. 155, 50–56. Sajjad, M., Khan, M.I.M., Zafar, R., Ahmad, S., Niazi, U.H.K., Akhtar, M.W., 2012. Influence of positioning of carbohydrate binding module on the activity of endoglucanase CelA of Clostridium thermocellum. J. Biotechnol. 161, 206–212. Shao, X., Jin, M., Guseva, A., Liu, C., Balan, V., Hogsett, D., Dale, B.E., Lynd, L., 2011. Conversion for Avicel and AFEX pretreated corn stover by Clostridium thermocellum and simultaneous saccharification and fermentation: insights

43

into microbial conversion of pretreated cellulosic biomass. Bioresour. Technol. 102, 8040–8045. Tachaapaikoon, C., Kosugi, A., Pason, P., Waeonukul, R., Ratanakhanokchai, K., Kyu, K., Arai, T., Murata, Y., Mori, Y., 2012. Isolation and characterization of a new cellulosome-producing Clostridium thermocellum strain. Biodegradation 23, 57– 68. Waeonukul, R., Kyu, K.L., Sakka, K., Ratanakhanokchai, K., 2009. Isolation and characterization of a multienzyme complex (cellulosome) of the Paenibacillus curdlanolyticus B-6 grown on Avicel under aerobic conditions. J. Biosci. Bioeng. 107, 610–614. Waeonukul, R., Kosugi, A., Tachaapaikoon, C., Pason, P., Ratanakhanokchai, K., Prawitwong, P., Deng, L., Saito, M., Mori, Y., 2012. Efficient saccharification of ammonia soaked rice straw by combination of Clostridium thermocellum cellulosome and Thermoanaerobacter brockii b-glucosidase. Bioresour. Technol. 107, 352–357. Waeonukul, R., Kosugi, A., Prawitwong, P., Deng, L., Tachaapaikoon, C., Pason, P., Ratanakhanokchai, K., Saito, M., Mori, Y., 2013. Novel cellulase recycling method using a combination of Clostridium thermocellum cellulosomes and Thermoanaerobacter brockii b-glucosidase. Bioresour. Technol. 130, 424–430. Ye, X., Rollin, J., Zhang, Y.-H.P., 2010. Thermophilic a-glucan phosphorylase from Clostridium thermocellum: cloning, characterization and enhanced thermostability. J. Mol. Catal. B: Enzym. 65, 110–116.