Fed-batch saccharification and ethanol fermentation of Jerusalem artichoke stalks by an inulinase producing Saccharomyces cerevisiae MK01

Abstract

Jerusalem artichoke is a potential energy crop. While its tubers are being used to extract inulin, its stalks could be used for biofuels production. In this article, fed-batch saccharification and fermentation of Jerusalem artichoke stalks (JAS) was studied. Pretreatment with 2.0% (w/v) NaOH not only retained most inulin, the unique component in JAS, but also increased cellulose content from 42.3% to 58.2% due to the removal of lignin. Batch-feeding of both the pretreated JAS and cellulases was an effective strategy for saccharification, through which 115.8 g L⁻¹ total sugars including glucose, xylose, fructose and inulin were released from 20% (w/v) solids uploading under the supplementation of cellulases at 20 FPU (filter paper unit) g⁻¹ dry biomass. An inulinase-producing yeast Saccharomyces cerevisiae MK01 was developed by the cell surface display of inulinase for ethanol fermentation from the pretreated JAS under the fed-batch conditions, and 38.3 g L⁻¹ ethanol was produced at 96 h, with an ethanol yield of 0.361 g g⁻¹ total sugars consumed, about 71% of the theoretical yield of 0.511, indicating that JAS would be a promising feedstock for ethanol production.

---

1. Introduction

Fossil fuels such as crude oil and coal are not renewable or environmentally friendly, which presents a challenge for sustainable development. Biomass resources are one of the solutions. Currently, sugar and starch-based feedstocks are dominating biomass resources for this purpose,¹ but they are not sustainable taking into account the global population and increased demand for food supply. Indeed, plant genetics provides a tool to engineer cell wall structure to address the recalcitrant nature of lignocellulose, but it is not applicable for grain crops due to the risk of compromising grain yield. Therefore, dedicated energy crops that can grow well in marginal land without competing for arable land with grain production, and in the meantime be engineered for more amenable conversion are garnered attention.²

Jerusalem artichoke (JA) is a perennial crop of the Compositae family, which is tolerant to various environmental stresses such as drought, salt, pest invasion and infection of plant diseases.^3,4 Like switchgrass and Miscanthus selected as model herbaceous energy crops in the US and EU,^2,5 JA might also be developed as an energy crop due to its aforementioned agronomic traits. The major biomass of JA is from its tubers and stalks. Unlike grain crops that accumulate starch as carbohydrate, JA tubers contain inulin, a linear β-(1-2)-linked oligo-fructan with a glucose terminal, which can be used to extract inulin for food use.⁶ JA stalks (JAS) are another major biomass, but unfortunately related research is very limited. So far there are only two reports on ethanol production from JAS, in which pretreated JAS was mixed with grinded JA tubers.^7,8 Apparently, such a strategy compromised the advantage of JA tubers as the unique feedstock for inulin extraction. In addition, yeast strain used with these studies was from Kluyveromyces marxianus, a species not for ethanol fermentation in industry.

Like other lignocellulosic biomass, JAS is characterized by a complex composed mainly of cellulose entangled with hemicelluloses and lignin,⁹ making the cellulose recalcitrant to enzymatic hydrolysis.^10,11 Therefore, pretreatment is required to deconstruct the complex, which impacts all subsequent steps such as cellulose hydrolysis and microbial fermentation.¹² On the other hand, JAS contains inulin and its content depends on harvesting season and climate conditions, which should be ultimately remained after pretreatment for ethanol production. Nowadays, acid or alkali pretreatment has been extensively studied for cellulosic ethanol production. While acid pretreatment presents the advantage of solubilizing hemicelluloses, alkaline pretreatment leads to significant removal of lignin,¹³ but their impact on JAS pretreatment has never been reported.

Saccharomyces cerevisiae has been used for ethanol fermentation from sugar- and starch-based feedstock for a long history, but this species is not efficient for ethanol production from inulin, presenting a necessity for engineering it with inulinase-producing ability for ethanol production from inulin in JAS. Heterologous expression of cellulases in S. cerevisiae by surface display has been developed for ethanol production from lignocellulosic biomass through consolidated bioprocessing strategy,^14,15 but unfortunately it has not been successful due to the intrinsic complexity of cellulases involving synergetic effect of different enzyme components as well as heterogeneous reaction with cellulose hydrolysis. In contrast, inulinase is simple and enzymatic hydrolysis of inulin is homogeneous. Therefore, surface display of inulinase in S. cerevisiae maybe an effective strategy for ethanol production from inulin in JAS.

Although high-solids loading has potential in increasing ethanol titer to reduce energy consumption with product recovery and stillage treatment,¹⁶ there are many drawbacks such as poor mixing and mass transfer due to extremely high viscosity.¹⁷ Fed-batch hydrolysis has been tested as a feasible strategy for addressing these problems.^18,19 On the other hand, the batch-feeding of cellulases was also an effective strategy for cellulose hydrolysis.²⁰

In this study, we explored conditions for pretreatment and enzymatic hydrolysis of JAS to enhance sugar yield. Yeast strain expressing inulinase was developed by surface display for ethanol fermentation from the pretreated JAS.

2. Materials and methods

2.1 Pretreatment

Raw JAS provided by Dalian Tianma Group Co. Ltd. (Dalian, China) was milled and sieved with 28 mesh screen. For acid pretreatment, 10.0% (w/v) feedstock was soaked in 0.5–5.0% (v/v) H₂SO₄ at 121 °C for 1 h. Then the biomass was filtered, washed to neutral pH, and dried at 50 °C for 24 h. For alkali pretreatment, 0.5–5.0% (w/v) NaOH was used and other conditions were same as those applied in the acid pretreatment. Solids recovery was calculated as the percentage of the total solids after pretreatment over the initial biomass.

2.2 Feeding strategies for the pretreated JAS and cellulases

Cellulases used in this study were generous gift from Youtell Biotechnology Co. Ltd. (Shandong, China) with an activity of 150 FPU mL⁻¹. Seven strategies illustrated in Table 1 were developed for feeding the pretreated JAS and cellulases. The total biomass loading was 20% and cellulases were supplemented at 20 FPU g⁻¹ (dry biomass). The saccharification was performed in 100 mL reaction volume with 50 mmol L⁻¹ citric acid buffer (pH 4.86) at 50 °C and 150 rpm for 48 h. The supernatant was then collected by centrifugation at 6000 × g for 10 min, which was subsequently analyzed by HPLC.

Table 1 Feeding strategies for JAS and cellulases^a

Run	Feeding time (h)
	0	12	24	36
a Biomass loading (g)/cellulases (FPU) supplemented.
Control	20/400	0/0	0/0	0/0
A-a	15/400	5/0	0/0	0/0
B-a	10/400	10/0	0/0	0/0
C-a	5/400	5/0	5/0	5/0
A-b	15/300	5/100	0/0	0/0
B-b	10/200	10/200	0/0	0/0
C-b	5/100	5/100	5/100	5/100

---

2.3 Strain development

The industrial strain S. cerevisiae from Angel Yeast (Hubei, China) for ethanol fermentation in China was used as the host to be engineered with INU1 from K. marxianus ATCC8554 by cell surface display. The primers used in the strain development were given in Table 2.

Table 2 Primers used in this study

Name	Sequence (5′–3′)
G-418-F	CCC̲A̲T̲A̲T̲G̲GTTTAGCTTGCCTCGTC
G-418-R	GAA̲G̲G̲C̲C̲T̲GTTTTCGACACTGGATG
SED1A-F	GAA̲G̲A̲T̲C̲T̲AAATTATCAACTGTCCTATTATCTG
SED1A-R	CCCA̲A̲G̲C̲T̲T̲TTATAAGAATAACATAGCAACACCA
INU1-F	GGA̲C̲T̲A̲G̲T̲ATGAAGTTCGCATACTCCCTCTTGC
INU1-R	GAA̲G̲A̲T̲C̲T̲GATCAAACGTTAAATTGGGTAACGT
PCR-F	TCCC̲C̲G̲C̲G̲G̲GAGTGAGGAACTATCGCATACCTGC
PCR-R	TCCC̲C̲G̲C̲G̲G̲GCAAATTAAAGCCTTCGAGC

---

The G418 resistant gene KanMX4 was used as the selective marker for screening transformants. The NdeI-StuI DNA fragment containing KanMX4 was amplified from the HO-poly-KanMX4-HO vector by PCR using the G418-F and G418-R primers.²¹ The fragment was then subcloned into the NdeI and StuI sites of pRS316 to obtain pRS316-KanMX4, which was used as the starting plasmid for the cell-surface expression of INU1. The inulinase gene (Genebank accession number X57202.1) fragment without the termination codon was amplified from K. marxianus genome by PCR with the primers INU1-F (SpeI) and INU1-R (BglII). The SED1 (Genebank accession number NM_001180385) fragment lacking of only the start codon was used as anchoring region, which was amplified from S. cerevisiae S288c genomic DNA by PCR using the SED1_A-F (BglII) and SED1_A-R (HindIII) primers, and designed to situate to the 3′-terminal of the INU1 fragment. The two fragments were inserted into the plasmid pRS425-β-glucosidase.^22,23 The INU1-SED1 fusion protein was inserted between the PGK1 promoter and the CYC1 terminator of pRS425-β-glucosidase, and the resulting plasmid was named pRS425-IS. The SacII–SacII DNA fragment encoding PGK_P-INU1-SED1_A-CYC1_T was cut from pRS425-IS, which was then subcloned into the SacII site of pRS316-KanMX4. The final plasmid was named pRS316-IS (Fig. 1), and the correctness of the INU1 fragment and the SED1 fragment was confirmed by sequencing (Sangon Biotech, China).

Fig. 1 Development of the inulinase expressing plasmid.

The plasmid pRS316-IS was transformed into S. cerevisiae Angel using the electroporation method following the instruction of the Bio-Rad Gene Pulser System (Bio-Rad, USA). The host stain transformed with the empty plasmid pRS316-KanMX4 containing KanMX4 resistant gene was used as the control. Positive transformants carrying INU1 were screened in the YPD agar medium containing 1% yeast extract, 2% peptone and 2% glucose supplemented with 2% agar and 300 μg mL⁻¹ G-418. To confirm the existence of the target gene, INU1 fragment was amplified by PCR with the primers shown in Table 2, which was confirmed by gel electrophoresis. The constructed strain was designated as S. cerevisiae MK01.

In addition to the INU1-bearing recombinant, S. cerevisiae XL engineered with xylose metabolic pathway by inserting a tandem PsmXR-PsXDH-ScXK expression cassette into the chromosome of S. cerevisiae for co-fermentation of glucose and xylose was also examined.²⁴

2.4. Simulated medium

Simulated medium was developed based on the composition analysis of JAS hydrolysate containing 6.5% glucose, 1.6% inulin and 1.1% xylose with 0.4% peptone and 0.3% yeast extract supplemented to evaluate ethanol fermentation performance of yeast strains. Yeast culture with OD₆₂₀ ∼ 0.5 was inoculated into 100 mL simulated medium in 250 mL Erlenmeyer flask, and fermentation was performed at 37 °C and 150 rpm for 48 h. All fermentations were carried out in duplicate.

2.5 Analysis

A HPLC system (Waters 410, Waters, MA, USA) with the column Bio-Rad Aminex HPX-87H (300 mm × 7.8 mm, Hercules CA) and Waters 2414 refractive detector was used to analyze glucose, fructose, xylose and ethanol. The mobile phase was 5 mmol L⁻¹ H₂SO₄ at the flow rate of 0.5 mL min⁻¹. The lignocellulosic components were determined by the standard methods developed by the National Renewable Energy Laboratory (NREL),²⁵ and the cellulose, hemicelluloses and lignin contents were calculated using equations below:

Cellulose (%) = (glucose × 0.90)/biomass × 100% (1)

Hemicellulose (%) = (xylose × 0.88)/biomass × 100% (2)

Lignin (%) = (soluble & insoluble lignin)/biomass × 100% (3)

The degree of polymerization (DP) of inulin varied depending on the origin and season of harvesting of Jerusalem artichoke.²⁶ Herein inulin was quantified by fructose content and the DP was chosen as 5 (fructose : glucose = 4 : 1) in safe.²⁷

Inulin (%) = (fructose × 1.15)/biomass × 100% (4)

3. Results and discussion

3.1 Pretreatment of JAS with acid or alkali

The impact of H₂SO₄ or NaOH supplemented with different concentrations on the pretreatment of JAS was evaluated by measuring the contents of cellulose, hemicelluloses, lignin and inulin, which are summarized in Table 3.

Table 3 Evaluation of JAS pretreatment by H₂SO₄ or NaOH

	Solids recovery (%)	Cellulose (%)	Hemicelluloses (%)	Lignin (%)	Inulin (%)
Raw JAS	100.0	39.4 ± 1.5	14.9 ± 1.3	18.1 ± 1.1	18.3 ± 0.4
H2SO4 (%)
0.5	85.0 ± 1.8	42.2 ± 0.5	12.0 ± 1.1	19.2 ± 1.1	13.3 ± 0.1
1.0	83.1 ± 1.1	48.3 ± 2.0	8.9 ± 0.2	20.2 ± 1.6	11.2 ± 0.1
2.0	78.3 ± 1.8	55.4 ± 1.1	5.3 ± 0.3	21.8 ± 1.3	10.1 ± 0.1
3.0	70.2 ± 1.5	41.5 ± 0.9	3.3 ± 0.5	23.5 ± 0.8	8.4 ± 0.8
4.0	68.5 ± 1.5	38.7 ± 0.6	2.5 ± 1.1	26.0 ± 0.1	7.5 ± 0.1
5.0	63.7 ± 1.7	35.8 ± 1.9	2.1 ± 1.3	27.1 ± 1.7	4.2 ± 0.2
NaOH (%)
0.5	90.0 ± 1.1	42.3 ± 1.6	13.7 ± 2.5	13.7 ± 1.7	15.8 ± 0.5
1.0	87.1 ± 1.8	49.8 ± 0.7	12.3 ± 2.7	10.6 ± 1.5	13.4 ± 0.1
2.0	85.4 ± 2.1	58.2 ± 1.2	11.1 ± 1.9	7.1 ± 0.9	13.0 ± 0.6
3.0	75.5 ± 1.5	58.9 ± 1.4	10.8 ± 1.9	6.1 ± 1.3	10.7 ± 0.0
4.0	70.7 ± 1.1	60.7 ± 1.4	10.0 ± 1.4	5.9 ± 0.1	8.2 ± 0.1
5.0	66.5 ± 1.9	60.8 ± 1.4	9.9 ± 0.2	5.2 ± 0.1	4.9 ± 0.1

---

With the increase of H₂SO₄ from 0.5% to 5.0%, hemicelluloses, inulin and total solids decreased from 12.0%, 13.3% and 85.0% to 2.1%, 4.2%, and 63.7%, respectively, indicating that the acid pretreatment significantly hydrolyzed hemicelluloses and inulin. However, when H₂SO₄ supplementation was in the range from 0.2% to 2.0%, cellulose content was increased from 42.2% to 55.4%, but more H₂SO₄ supplementation significantly decreased cellulose content due to hydrolysis. Besides, lignin content in the pretreated JAS increased concomitantly from 19.2% to 27.1% due to the decrease of hemicelluloses. By comparing solids recovery, cellulose, hemicelluloses and inulin contents, 2.0% H₂SO₄ treatment showed the best result among all acid dosages.

Investigations to explore the effect of NaOH pretreatment on JAS were also performed. With the increase of NaOH from 0.5% to 5.0%, significant decreases of lignin from 13.7% to 5.2%, inulin from 15.8% to 4.9% and solids recovery from 90.0% to 66.5% were observed, but the change in hemicelluloses content was comparatively small. These results showed that NaOH pretreatment of the JAS could effectively increase cellulose content from 42.3% to 60.8%. The optimal condition was 2.0% NaOH based on overall evaluation on solid recovery and cellulose and inulin contents.

Comparing to H₂SO₄, NaOH pretreatment removed lignin and enriched cellulose and hemicelluloses more effectively with less inulin lost. Thus, 2.0% NaOH pretreatment was considered to be more suitable for JAS.

3.2 Saccharification of the pretreated JAS

Fig. 2(a) illustrates major sugars detected in the hydrolysate of JAS pretreated by 2% NaOH and saccharified with strategies illustrated in Table 1. As can be seen, the control with all biomass and enzyme loaded at the beginning produced the lowest total sugars of 57.3 g L⁻¹, since the mixing was poor due to the high solid uploading, and extremely viscous slurry was observed until 20 h. By applying the biomass feeding strategies (A-a, B-a and C-a), total sugars were increased by 25.1%, 53.4% and 40.4%, respectively, compared to that detected in the control. When the fed-batch of cellulases (A-b, B-b and C-b) was further applied, total sugars were increased by 17.2%, 48.6% and 10.3% compared to that detected with the fed-batch of the pretreated JAS only. The time-course of glucose releasing from the hydrolysis of the cellulose component during the process illustrated in Fig. 2(b) further supported the effectiveness of the saccharification strategy B-b.

Fig. 2 Major sugars detected at 48 h for JAS pretreated by 2% NaOH and hydrolyzed with strategies illustrated in Table 1 (a) and time-course of glucose releasing from the hydrolysis of the cellulose component (b).

For the batch feeding of cellulases, the decrease of their activities was alleviated. However, the fed-batch saccharification strategy C-b with four equally feeding of the feedstock and enzyme at 0, 12, 24 and 36 h performed worse than the strategy B-b with the twice loading of the feedstock and enzyme due to short time for the enzymatic hydrolysis and low enzyme dosage at the early saccharification stage. It is clearly indicated that total sugars produced was dependent on the synergistic feeding of the biomass and enzyme. Twice feeding of both biomass and enzyme with 10% initial biomass loading was validated as a more efficient strategy for the enzymatic hydrolysis, through which 83.7 g L⁻¹ glucose, 11.2 g L⁻¹ xylose and 4.3 g L⁻¹ fructose were obtained from 200 g L⁻¹ biomass. It is interesting that 16.4 g L⁻¹ inulin was also detected after 48 h saccharification.

3.3 Evaluation of yeast strains

As can be seen in Fig. 3(a), inulinase activity was significantly enhanced in S. cerevisiae MK01, and the inulinase activity as high as 6000 U g⁻¹ (DCW) was detected at 96 h with the engineered yeast strain, about 4 folds of that observed with the control, while the xylose-metabolizing strain S. cerevisiae XL exhibited even poorer inulinase activity than that with the control. Meanwhile, no significant difference was observed in growth between S. cerevisiae MK01 and the control (Fig. 3(b)).

Fig. 3 Inulinase activities (a) and growth profiles (b) of S. cerevisiae MK01, S. cerevisiae XL and the control strain S. cerevisiae Angel.

Fig. 4 is the results of ethanol fermentation with simulated medium. S. cerevisiae MK01 grew faster than others, but no significant difference was observed in glucose consumption since all yeast strains almost completely consumed glucose within 24 h. However, S. cerevisiae MK01 consumed 13.9 g L⁻¹ inulin, while less inulin (2.1–8.9 g L⁻¹) was consumed by other strains. In case of xylose conversion, S. cerevisiae XL showed better performance than other strains, with 6.3 g L⁻¹ xylose consumed. It is clear that strains consumed more total sugars produced more ethanol. The highest ethanol of 34.2 g L⁻¹ was produced by S. cerevisiae MK01. If concerning xylose conversion, S. cerevisiae XL was better, but for JAS hydrolysate containing more inulin than xylose from the hydrolysis of hemicelluloses, S. cerevisiae MK01 took advantages in converting more sugars that conferred by its improved inulinase activity for more efficient ethanol production. Therefore, S. cerevisiae MK01 was more suitable for ethanol production from JAS.

Fig. 4 Ethanol fermentation of the simulated medium by different yeast strains.

3.4 Ethanol fermentation from JAS

Unit operations were integrated for ethanol fermentation from JAS: 2% NaOH pretreated JAS was saccharified with the loading strategy B-b for the biomass and enzyme, and the fermentation was carried out by S. cerevisiae MK01 at 37 °C for 96 h. The purpose of saccharification was to release sugars quickly at elevated temperature for the enzymatic hydrolysis, and in the meantime reduce viscosity with high biomass loading that caused diffusion and mixing problems with the fermentation process.²⁸ The total sugars with the process were analyzed to be 115.8 g L⁻¹. The experimental results are shown in Fig. 5.

Fig. 5 Ethanol fermentation from JAS hydrolysate by S. cerevisiae MK01 (2% NaOH pretreated JAS was pre-saccharified under 50 °C for 48 h with the biomass and cellulase loading strategy B-b).

Yeast cells grew exponentially until 24 h, and then experienced a stationary phase for the next 24 h followed by the death phase. It was observed that glucose was quickly consumed. Meanwhile, inulin was gradually hydrolyzed by the inulinase displayed onto the surface of S. cerevisiae MK01, and fructose was released consistently, but was consumed as glucose was depleted. However, S. cerevisiae MK01 is not a good xylose utilizing strain and 8.5 g L⁻¹ xylose was left at the end of the fermentation. Ethanol increased rapidly at the first 20 h as glucose was consumed. The maximum ethanol production of 38.3 g L⁻¹ was achieved at 96 h, with 9.7 g L⁻¹ total sugars unconverted, making the ethanol yield of 0.361 g g⁻¹ total sugars consumed, about 71% of the theoretical yield of 0.511. Since xylose was the major residual sugar, it is expected that ethanol yield would be further improved with the development of yeast strains expressing both genes for inulinase production and xylose metabolism.

Ethanol production from JA tubers was explored in the 1980s to address the oil crisis.^29,30 As a result, JA was grown in large quantities by Midwestern US farmers,³¹ but this project was interrupted shortly when cheap oil came back, and the farmers suffered big economic loss. Rocked oil prices at the beginning of the millennium, particularly in 2008 when oil prices hit the unprecedented record of USD147 per barrel, highlighted the importance of biofuels and feedstocks for biofuels production again.^32,33

Although JA could be developed as an energy crop with advantages over other lignocellulosic biomass, how to process it without a repeat of the historic tragedy presents a challenge. We herewith propose a roadmap for the biorefinery of JA in Fig. 6, and compare its advantages with other lignocellulosic biomass in Table 4.

Fig. 6 Roadmap for the biorefinery of JA biomass.

Table 4 Biorefinery of JA over other lignocellulosic biomass

Biomass		Land usage	Advantages
Jerusalem artichoke	Potential energy crop	Marginal lands	Inulin extraction from tubers for food use, and modified to address the recalcitrance of stalks
Switchgrass and Miscanthus	Energy crops	Marginal lands	Modified to address their recalcitrance
Corn stover, rice/wheat straw and so on	Agricultural residues	No	No way to address their recalcitrance

---

JA tubers present potentials for producing value-added products to credit its biorefinery, but other lignocellulosic biomass is lacking of this advantage. Even inulin extracted directly from JA tubers without significant energy input compared to the pretreatment of lignocellulosic biomass and recovery of ethanol from fermentation broth can be sold in the food market at much higher prices than that for ethanol as a biofuel, needless to say functional products and specialities that can be derived from inulin.^34,35 Therefore, JA tubers should not be used for producing bulk commodities such as fuel ethanol. On the other hand, JAS with some inulin seems to be better than other lignocellulosic biomass for ethanol production through microbial fermentation in case significant progress is made in engineering strains with pentose utilization.

4. Conclusions

Experimental results suggest that the pretreatment with 2.0% (w/v) NaOH could remove lignin effectively from JAS and consequently increased cellulose content from 42.3% to 58.2%. Fed-batch for the pretreated feedstock and cellulases was proven to be a promising strategy for efficient saccharification, which produced 115.8 g L⁻¹ total sugars from 20% (w/v) biomass uploading. The inulinase-producing yeast S. cerevisiae MK01 was evaluated as a suitable strain at present to ferment the JAS hydrolysate, which produced 38.3 g L⁻¹ ethanol at 96 h from 115.8 g L⁻¹ total sugars, making the ethanol yield of 0.361 g g⁻¹ total sugars consumed, about 71% of the theoretical yield of 0.511. With the development of yeast strains engineered with inulinase production and pentose metabolism as well, ethanol production from JAS is expected to be further improved, although many challenges are still ahead. Therefore, JAS is a promising feedstock for ethanol production, making JA a potential energy crop.

Acknowledgements

This work was financially supported by National High Tech R & D Program of China with a project reference number of 2012AA101805 and National Natural Science Foundation of China with a project reference number of 51561145014. M. Mahfuza Khatun is grateful to Chinese Scholarship Council for financial support of her PhD study. The assistance from Liang Xiong, Cheng Cheng and Md. Asraful Alam with strain development and experiment is highly appreciated.

References

1. F. W. Bai, W. A. Anderson and M. Moo-Young, Biotechnol. Adv., 2008, 26, 89–105.
2. E. A. Heaton, R. B. Flavell, P. N. Mascia, S. R. Thomas, F. G. Dohleman and S. P. Long, Curr. Opin. Biotechnol., 2008, 19, 202–209.
3. Z. Huang, L. Zhao, D. Chen, M. Liang, Z. Liu, H. Shao and X. Long, PLoS One, 2013, 8, e62085.
4. X. Y. Ma, L. H. Zhang, H. B. Shao, G. Xu, F. Zhang, F. T. Ni and M. Brestic, J. Med. Plants Res., 2011, 5, 1272–1279.
5. X. Ge, D. M. Burner, J. Xu, G. C. Phillips and G. Sivakumar, Biotechnol. J., 2011, 6, 66–73.
6. S. J. Kays and S. F. Nottingham, Biology and chemistry of Jerusalem artichoke: Helianthus tuberosus L., CRC press, Taylor and Francis Group, New York, 2008.
7. S. Kim and C. H. Kim, J. Renewable Energy, 2014, 65, 83–91.
8. S. Kim, J. M. Park and C. H. Kim, Appl. Biochem. Biotechnol., 2013, 169, 1531–1545.
9. A. T. W. M. Hendriks and G. Zeeman, Bioresour. Technol., 2009, 100, 10–18.
10. H. Jørgensen, J. B. Kristensen and C. Felby, Biofuels, Bioprod. Biorefin., 2007, 1, 119–134.
11. M. E. Himmel, S. Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos, J. W. Brady and T. D. Foust, Science, 2007, 315, 804–807.
12. M. Galbe and G. Zacchi, Biomass Bioenergy, 2012, 46, 70–78.
13. P. Alvira, E. Tomás-Pejó, M. Ballesteros and M. Negro, Bioresour. Technol., 2010, 101, 4851–4861.
14. A. Kondo and M. Ueda, Appl. Microbiol. Biotechnol., 2004, 64, 28–40.
15. Q. Xu, A. Singh and M. E. Himmel, Curr. Opin. Biotechnol., 2009, 20, 364–371.
16. R. Koppram, E. Tomás-Pejó, C. Xiros and L. Olsson, Trends Biotechnol., 2014, 32, 46–53.
17. Y. Gao, J. Xu, Z. Yuan, Y. Zhang, Y. Liu and C. Liang, Bioresour. Technol., 2014, 167, 41–45.
18. A. Rudolf, M. Alkasrawi, G. Zacchi and G. Lidén, Enzyme Microb. Technol., 2005, 37, 195–204.
19. E. Varga, H. B. Klinke, K. Réczey and A. B. Thomsen, Biotechnol. Bioeng., 2004, 88, 567–574.
20. K. Hoyer, M. Galbe and G. Zacchi, Biotechnol. Biofuels, 2010, 3, 14,  DOI:10.1186/1754-6834-3-14.
21. W. P. Voth, J. D. Richards, J. M. Shaw and D. J. Stillman, Nucleic Acids Res., 2001, 29, e59.
22. J. M. Galazka, C. Tian, W. T. Beeson, B. Martinez, N. L. Glass and J. D. H. Cate, Science, 2010, 330, 84–86.
23. K. Inokuma, T. Hasunuma and A. Kondo, Biotechnol. Biofuels, 2014, 7, 8,  DOI:10.1186/1754-6834-7-8.
24. Q. Zuo, X. Q. Zhao, L. Xiong, H. J. Liu, Y. H. Xu, S. Y. Hu, Z. Y. Ma, Q. W. Zhu and F. W. Bai, Biochem. Biophys. Res. Commun., 2013, 440, 241–244.
25. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton and D. Crocker, Laboratory Analytical Procedure (LAP), NREL, Golden, CO, USA, 2011.
26. L. Kocsis, P. Liebhard and W. Praznik, J. Agric. Food Chem., 2007, 55, 9401–9408.
27. P. Dysseler, D. Hoffem, J. Fockedey, B. Quemener, J. F. Thibault and P. Coussement, Determination of inulin and oligo-fructose in food products (modified AOAC dietary fiber method), Complex carbohydrates in foods, ed. S. S. Cho, L. Prosky and M. Dreher, Marcel Dekker Inc., New York, 1999, pp. 213–227.
28. Z. H. Liu, L. Qin, J. Q. Zhu, B. Z. Li and Y. J. Yuan, Biotechnol. Biofuels, 2014, 7, 167.
29. A. Margaritis and P. Bajpai, Biotechnol. Bioeng., 1982, 24, 1473–1482.
30. A. Margaritis and P. Bajpai, Biotechnol. Bioeng., 1982, 24, 1483–1493.
31. J. A. Amato, The Great Jerusalem Artichoke Circus: The Buying and Selling of the Rural American Dream, University of Minnesota Press, 1993.
32. W. J. Yuan, X. Q. Zhao, L. J. Chen and F. W. Bai, Biotechnol. Bioprocess Eng., 2013, 18, 721–727.
33. J. Q. Gao, W. J. Yuan, L. Kong, R. J. Xiang and S. J. Zhong, Biomass Bioenergy, 2015, 80, 10–16.
34. Y. X. Hu, J. Zhang, C. W. Yu, Q. Li, F. Dong, G. Wang and Z. Y. Guo, Int. J. Biol. Macromol., 2014, 70, 44–49.
35. Z. Y. Guo, Q. Li, G. Wang, F. Dong, H. Y. Zhou and J. Zhang, Carbohydr. Polym., 2014, 99, 469–473.

---

Footnote

† M. Mahfuza Khatun and Yong-Hao Li contributed equally to this work.

---