Enzymatic Hydrolysis of Industrial Derived Xylo ... - ACS Publications

Subscriber access provided by TUFTS UNIV

Article

Enzymatic hydrolysis of industrial derived xylo-oligomers to monomeric sugars for pilot plant scale glycol production Jinguang Hu, Joshua A. Davies, Yiu Ki Mok, Bryan Gene, Quak Foo Lee, Claudio Arato, and Jack N Saddler ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.6b02008 • Publication Date (Web): 12 Oct 2016 Downloaded from http://pubs.acs.org on October 16, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sustainable Chemistry & Engineering is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

The enzymatic hydrolysis of industrial derived xylo-oligomers to monomeric sugars for potential chemical/biofuel production a,b

Jinguang Hu , Joshua Daviesb, Yiu Ki (Tony) Moka, Bryan Geneb, Quak Foo Leeb, Claudio Aratob and Jack N. Saddler*a a

Forest Products Biotechnology/Bioenergy Group, Department of Wood Science, Faculty of Forestry

The University of British Columbia, 2424 Main Mall, Vancouver BC, Canada V6T 1Z4 b

S2G BioChemicals, 4250 Wesbrook Mall, Vancouver BC, Canada V6T 1W5

* Corresponding author: E-mail: [email protected]; Tel: (+01) 604-822-2434

Keywords: xylo-oligomers; enzymatic hydrolysis; bio-glycol; biorefinery

Abstract: Commercial grade, xylo-oligosaccharide rich, water soluble streams, obtained after hydrothermal pretreatment of wheat straw, were assessed for their potential as sugar feedstocks to make glycol. When acid and enzymatically based hydrolysis processes were compared it appeared there was considerable potential to further optimize the enzymatic approach of hydrolyzing the oligomers to pure xylose. Various commercial enzyme cocktails and their synergistic cooperation were assessed over a range of combinations and hydrolysis conditions. An optimized “enzyme cocktail”, at low protein loadings, could hydrolyze more than 80 % of the oligomers to xylose within 3 hours. After 24 hrs, all of xylo-oligomers were hydrolyzed to xylose. A moderately adjusted pH of 4.3 ensured fast and efficient hydrolysis without the need for agitation. The advantage of an enzymatic as opposed to an acid based approach to hydrolysis was evidenced when the process was scaled up to the 300 L. These include the use of a cheaper and simpler infrastructure, improved xylose recovery and a much easier xylose purification process. This indicated the potential for further enzyme recycle and reuse, further enhancing the economic attractiveness of an enzyme based approach.

Introduction: One of the goals of a successful “biorefinery” process is to biochemically convert all the polysaccharides within the lignocellulosic biomass, the hemicellulose as well as the cellulose, to sugars, so they can subsequently be used as the feedstock for the production of a range of chemicals and fuels.1,2 Due to the recalcitrant nature of biomass, a pretreatment step is often required to disrupt the highly organized cell wall structure before enzymatic hydrolysis of the cellulosic component. However, it has been shown that “compromise” conditions of pretreatment severity have to be used, to ensure both best recovery of the

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

hemicellulose sugars while still enhancing enzyme accessibility to the cellulose fraction. Although various industrially-relevant pretreatment processes (steam explosion, diluted acid pretreatment, hydrothermal pretreatment etc.) can successfully fractionate/solubilize large portions of the hemicellulose from the lignocellulosic biomass,3,4 much of this dissolved hemicellulose remain in an oligomeric form which is not amenable to further downstream processing or use.5,6 The structure and composition of hemicellulose differs between plant species and cell tissues.7,8 For example, the major hemicellulose in hardwoods and annual plants are xylans (β-1,4-xylosyl backbone with arabinose, acetyl and uronic acid side chains) whereas mannans (β-1,4-mannosyl backbone and/or glucosyl-mannosyl backbone with galactose side chains) are more abundant in softwoods.7,9 In addition, the hemicellulose derived oligomers from the same biomass have been shown to differ in their degree of substitution and polymerization, depending on the pretreatment strategy that was used.4,10 Therefore, the rapid and efficient breakdown of hemicellulose derived oligosaccharides to monomeric sugars still needs to be fully resolved.

Two primary approaches, enzymatic and acid, have been used to hydrolyze hemicellulose derived oligomers to monomeric sugars. Although acid hydrolysis has been relatively well studied,6,11-13 to date limited research has investigated the potential of an enzymatic hydrolysis strategy5,14,15 to breaking down hemicellulose derived oligomers. In contrast to acid hydrolysis, one advantage of enzyme mediated hydrolysis is the ability to operate at milder and environmentally friendly conditions.5 However, as mentioned earlier, hemicellulose derived oligomers are often highly substituted, significantly impeding their accessibility to hemicellulase enzyme.7,9 Previous studies have shown that the synergistic cooperation of various hemicellulose backbone cleaving and debranching enzymes can significantly enhance the hydrolysis of these hemicellulose derived oligomers.5,16,17

Several biochemical-based second generation (advanced) biofuel plants have been operating in Europe (e.g. Inbicon, Beta Renewables), the USA (e.g. Abengoa, DuPont/Poet-DSM) and Brazil (e.g. GrandBio, Raizen).18 As agricultural residues (wheat straw, corn stover, sugar cane bagasse) are the primary feedstock for all of these plants large quantities of soluble xylo-oligomers will be available in the water soluble fraction of these pretreated materials.9 Although good progress continues to be made in the effective fermentation of xylose to ethanol, it has proven to still be challenging to achieve economically attractive xylose fermentation due to the typically low ethanol titer and the preference of microorganisms to first utilize hexoses (e.g. glucose and mannose) rather than pentoses (e.g. xylose and arabinose).19


Page 2 of 19

Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Alternatively, pentoses are preferred feedstocks for the production of ethylene glycol and glycerol through the retro-aldol reaction under the applied hydrogenation condition.20 Since the demand for chemicals such as ethylene and propylene glycol continues to grow annually, one company that is currently investigating the potential of manufacturing high-value chemicals from biomass derived pentose sugars (e.g. xylose and arabinose) is S2G Biochemicals who plans to produce bio-glycols from xylose.

One of the initial goals of the work reported here was to compare the acid and enzymatic approaches to producing pure xylose that could be used as a feedstock to make glycols. Subsequently, the hydrolytic potential of various commercial enzyme cocktails and their synergistic effects on the hydrolysis of two, commercially derived, hydrothermal pretreated wheat straw, xylo-oligomer rich streams were assessed, using statistic design comparisons (i.e. Fractional Factorial Design and Central Component Design). The influence of the hydrolysis condition (e.g. solid content, pH, and agitation) on the final xylose sugar yield was evaluated and the optimized enzyme combination was subsequently used to scale-up hydrolysis to the 300 L scale. The xylose obtained was further purified through S2G’s pilot scale sugar purification process. The enzymatic hydrolysis approach showed at least comparable and even more potential than did the traditional acid hydrolysis approach. The potential of enzyme reuse/recycle was also evaluated.

Results and Discussion: Assessment of the chemical composition of the commercial xylo-oligomer enriched streams before and after acid hydrolysis. The commercial xylo-oligomer enriched streams were derived from the pretreatment of wheat straw and were purchased from two independently available sources by S2G Biochemicals, denoted subsequently as Sample 1 and Sample 2. Both companies used variations of the autohydrolysis (hydrothermal pretreatment) process at compromised conditions, where the acetic acid derived from the biomass acted as the catalyst for hemicellulose solubilisation. As expected, the major components within the pretreatment liquids were oligomeric sugars, which were 27.2 and 23.6 g/L for Sample 1 and Sample 2, respectively (Table 1). Although the addition of higher concentrations of acid catalysts (e.g. H2SO4, SO2) at a higher severity pretreatment condition has been shown to hydrolyze more of the soluble oligomers to monomeric sugars, these more severe conditions also resulted in a considerable amount of sugar degradation (to HFM and/or furfural) as well as lignin condensation.3,4 This not only significantly reduced overall biomass carbohydrate recovery it also resulted in the production of more inhibitory components that were detrimental to the downstream enzymatic hydrolysis and fermentation processes.3,4 Therefore, milder pretreatment conditions are typically preferred in an industrial process, where significant amounts of hemicellulose are released in the oligomer form as indicated by Samples 1 and 2. It is also worth noting that, even though the two samples were purchased



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

from different commercial plants (likely meaning that different types of wheat straws, pretreatment processes, conditions and equipment were used), they were surprisingly similar in sugar composition (Table 1).

When the traditional acid hydrolysis was used to deconstruct the hemicellulose derived oligomers to monomers, although xylose was the major product (~19 g/L) (Table 1), it appeared that about 10 % of the original sugars were degraded during acid hydrolysis. It was also likely that that the original oligomers were mainly branched oligosaccharides (e.g. arabinoxylan and/or xyloglucan and/or glucomannan), as significant amounts of arabinose, galactose, glucose and mannose were also released after acid hydrolysis (Table 1).

Comparison of major commercial enzyme preparations ability to hydrolyze the oligomers presents in water-soluble, hemicellulose rich commercial fractions. We next assessed the hydrolytic potential of various commercial enzymes and their interactions during hydrolysis of the oligomer rich liquids. Although both of the commercially derived water soluble fractions contained similar amounts and composition of sugars (Table 1), the sample 1 was used for further work because more of this material was available. Initially, six (6) potential enzymes preparations were screened including, Cellic HTec (mainly family 10 endo-xylanase), Multifect Xylanases (mainly family 11 endo-xylanase), β-xylosidases, Novozyme 188 (mainly β-glucosidase), and Cellic CTec2 and CTec3 (cellulase mixture with high hemicellulase and other activities) were selected. Cellic HTec and Multifect xylanase preparations were initially screened because of the large amount of xylooligomers that had been detected in these hemicellulose rich, water soluble fractions (Table 1). Synergistic cooperation between family 10 (enriched in Cellic HTec) and 11 endo-xylanases (enriched in Multifect xylanase) has also been observed during hydrolysis of certain xylan substrates.21,22 β-xylosidases and β-glucosidase (Novozyme 188) were selected as they have been shown to efficiently break down small oligomers, especially xylobiose and cellobiose respectively, which were highly inhibitory towards xylanase and cellulase enzymes.17 The selection of Cellic CTec2 and CTec3 was due to their broad activities on a wide spectrum of biomass polysaccharides, their higher thermostability, and higher tolerance to the inhibitory components such as monomer/oligomer sugars and pretreatment derived phenolics.23,24

A Fractional Factorial Design (FFD) was initially used to screen the essential enzyme preparations for their potential to hydrolyze the oligomers within the hemicellulose rich, water-soluble fractions. A half Fractional Factorial Design (FFD) for six (6) variables (enzyme preparations) created a total of 35 samples, namely 32 different combination plus three (3) central points. The hydrolysis design and their


Page 4 of 19

Page 5 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


response (monomer sugar concentration) at three (3) time points (3 h, 8 h, and 24 h) were described in supplementary data Table S1-S3. It appears that, by choosing an optimized enzyme combination, effective xylo-oligomeric sugar hydrolysis (>90 %, based on the original xylooligomers concentration ~22.5 g/L) can be achieved in a relatively short amount of time (8 hours), as highlighted in red in Table S1-S3. Although close to theoretical xylo-oligomer hydrolysis (100%) could be reached after 24 hours with some enzyme combinations, a target of 90-95 % conversion after 8 hours of hydrolysis is likely more commercially attractive in an industrial process. Thus, subsequent FFD Analysis of Variance (ANOVA) analysis was based on the xylose concentration obtained after 8 hours of hydrolysis, as described in Table S4. The high F-value (30.67) and low P-value (0.032) indicated that the model adequately fit the experimental data. The model’s coefficient of determination R2 was also calculated to be 0.9742, indicating that the model was good to 97 % of data variability. As the P-value for factor of CTec2, CTec3, BX, and their interactions were all significant (p < 0.05), the enzyme components of CTec2, CTec3, and BX were used for the following experiments.

Although the effectiveness of BX (β-xylosidases) for xylooligomer hydrolysis was anticipated, the ability of CTec2 and CTec3 (even better than the Cellic HTec and Multifect xylanase) in hydrolyzing the xylooligosaccharides was unexpected. It appeared that these “advanced cellulase mixture” also contain enzyme activities that are able to effectively hydrolyze the xylooligomers derived from biomass pretreatment. This might have been anticipated as xylooligomers have been shown to be quite inhibitory towards canonical cellulase enzymes25.

Optimization of the “enzyme cocktail” used for the hydrolysis of the hemicellulose rich, water soluble fractions. We next assessed CTec2 and CTec3 for further possible enzyme cocktail optimization using a Central Component Design (CCD). The combination and dosages of the two variables and their response (xylose concentration) after eight hours of hydrolysis is summarized in supplementary data Table S5. During data analysis, when a quadratic model was employed, the ANOVA analysis indicated that the model adequately fit the experimental data (P-value < 0.0005, R2 > 0.97) (Table S6). As mentioned previously, the response surface model (Figure 1) was based on the xylose concentration after eight hours of hydrolysis of Sample 1. Although CTec3 alone resulted in reasonable xylooligomer hydrolysis (e.g. resulting in ~19 mg/ml xylose and ~80% xylooligomer conversion (figure 1A)), xylose production plateaued even when high enzyme loadings were used. However, the synergistic cooperation between the CTec3 and CTec2 mixtures resulted in even higher xylose yields (e.g. resulting in ~21 mg/ml xylose and ~95% xylooligomer conversion (figure 1A)) when using lower enzyme loadings. We and other groups have shown that, by optimising enzyme synergism, fast and efficient hydrolysis of various



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 19

oligomeric/polymeric carbohydrates can be achieved.27,30 The regression equation for xylo-oligomer hydrolysis in Sample 1 (Figure 1A) was: “Xylose = +19.92 + 0.79 * CTec 2 + 3.52 * CTec3 - 0.16 * CTec 2 * CTec3 - 0.87 * CTec 2^2 - 2.03 * CTec3^2”.

(1)

Similar results were observed when this enzyme optimization strategy was applied to Sample 2 using CTec2 and CTec3 as the essential enzyme preparations (data not shown). The slightly different regression equation used for xylo-oligomer hydrolysis of Sample 2 (Figure 1B) was: “Xylose = + 15.66 + 0.20 * CTec 2 + 0.75 * CTec3 - 0.11 * CTec 2 * CTec3 - 0.32 * CTec 2^2 – 0.87 * CTec3^2”.

(2)

Based on minimizing the CTec3 and CTec2 dosages while maximizing xylo-oligomer hydrolysis, the solution provided by the model was “5.3 mg CTec2 + 6.7 mg CTec3” per gram of oligomers for Sample 1 (Figure 1A) and “7.2 mg CTec2 + 7.2 mg CTec3” per gram of oligomers for Sample 2 (Figure 1B) with a final xylose concentration of 21.1 g/L (93 % xylo-oligomer conversion) and 15.8 g/L (85 % xylooligomer conversion) after 8 hour enzymatic hydrolysis, respectively. It appeared that the enzymatic hydrolysis approach resulted in a better xylose recovery as compared to the traditional acid hydrolysis process for Sample 1 (Figure 1A and Table 1), and this advantage would be even more significant if a longer enzymatic hydrolysis time (24 hours) was applied where the theoretical xylose recovery (~22 g/L) was achieved without sugar degradation (Data not shown). However, the optimized enzyme cocktail for Sample 2 resulted in only ~85 % xylooligomer hydrolysis (15.8 g/L, Figure 1B), which was slightly lower than what was obtained after acid hydrolysis process (16.5 g/L, Table 1). This might be due to the higher acid soluble lignin content of Sample 2 (1.1 g/L and 4.5 g/L for Sample 1 and Sample 2, respectively) which has been shown inhibited the enzyme activities during the enzymatic hydrolysis of pretreated biomass.24,26

The influence of hydrolysis condition on xylose yields. As hydrolysis conditions such as the solids content, pH and agitation would likely affect the capital costs of a scaled-up process, we next assessed their influences on the hydrolysis of Sample 1 when using the optimized enzyme combination. To assess the possible influence of the solids content, Sample 1, which had a solids content of 6.8 %, was further concentrated to a range of solid contents using reverse osmosis. As expected, hydrolysis performance was gradually reduced by an increase in the initial solids content (Figure 2A) with xylose hydrolysis yields decreasing from 93 % to 80 % after 8 hours of hydrolysis when the initial solid content was doubled (~13 % w/v). This further decreased to only 65 % when the initial solid content reached 18 % (w/v) (Figure 2A).


Page 7 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


As shown previously, this reduced enzyme performance was likely caused by the increased amount of inhibitors (e.g. oligo/monomer sugars and phenolics) that were present.25

The initial pH of Sample 1 was pH 3.8 at room temperature (20 ºC). We had previously adjusted the pH of all of the liquids to pH 5.0 (using NaOH) as this was the optimized pH for most of the commercial enzyme preparations used to hydrolyze insoluble biomass substrates.27 Since the enzymes may have a different preference on liquid/soluble samples and pH adjustments also consume considerable amounts of NaOH, the possible influence of the pH on the hydrolytic performance of the optimized enzyme mixture (Figure 2B) was next assessed. It appeared that adjusting the liquid pH to 4.3 was acceptable as similar xylose hydrolysis yield was achieved over a pH range of 4.3 to 5.0. But a lower pH value (

Enzymatic Hydrolysis of Industrial Derived Xylo ... - ACS Publications

Recommend Documents