Abstract
Nanocellulose made by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-catalyzed oxidation, described as TEMPO-oxidized cellulose nanofibers (TOCNs), has a high density of negative charges on its surface. Its use in microbial fermentation systems is expected to benefit microbial process stability. In particular, microbial stability is strongly required in acetone–butanol–ethanol (ABE) fermentation associated with the solvent-extraction process of butanol production. Here, TOCNs derived from oil palm empty fruit bunches pulp were added to extractive ABE fermentation media containing glucose as a main source, which can be potentially obtained from biomass by saccharification. Then, microbial fermentation was carried out using free or immobilized bacterial cells, to produce butanol from glucose. The presence of TOCNs induced higher total butanol production in broth by improving the growth environment of Clostridium saccharoperbutylacetonicum N1-4, which was used as the butanol-producing strain. Microscopic analysis revealed that the spider-web-like TOCN network helped to entrap bacterial cells in alginate beads, by ionic crosslinking of TOCNs and alginates via Ca2+ ions, to increase stability of bacterial cells in the composite gel beads. The addition of TOCNs to fermentation media had significant positive effects on the total butanol yield.
Download PDF
Full Article
Nanocellulose from Oil Palm Biomass to Enhance Microbial Fermentation of Butanol for Bioenergy Applications
Novitri Hastuti,a,b Rizki Fitria Darmayanti,c,d Safrina Dyah Hardiningtyas,e,f Kyohei Kanomata,a Kenji Sonomoto,c Masahiro Goto,e and Takuya Kitaoka a,*
Nanocellulose made by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-catalyzed oxidation, described as TEMPO-oxidized cellulose nanofibers (TOCNs), has a high density of negative charges on its surface. Its use in microbial fermentation systems is expected to benefit microbial process stability. In particular, microbial stability is strongly required in acetone–butanol–ethanol (ABE) fermentation associated with the solvent-extraction process of butanol production. Here, TOCNs derived from oil palm empty fruit bunches pulp were added to extractive ABE fermentation media containing glucose as a main source, which can be potentially obtained from biomass by saccharification. Then, microbial fermentation was carried out using free or immobilized bacterial cells, to produce butanol from glucose. The presence of TOCNs induced higher total butanol production in broth by improving the growth environment of Clostridium saccharoperbutylacetonicum N1-4, which was used as the butanol-producing strain. Microscopic analysis revealed that the spider-web-like TOCN network helped to entrap bacterial cells in alginate beads, by ionic crosslinking of TOCNs and alginates via Ca2+ ions, to increase stability of bacterial cells in the composite gel beads. The addition of TOCNs to fermentation media had significant positive effects on the total butanol yield.
Keywords: Nanocellulose; Fermentation; Biobutanol; Alginate; Cell immobilization; Oil palm empty fruit bunch
Contact information: a: Department Agro-Environmental Sciences, Graduate School of Bioresource and Bioenvironmental Sciences, Faculty of Agriculture, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan; b: Forest Products Research and Development Center, Research Development and Innovation Agency, Ministry of Environment and Forestry, Jalan Gunung Batu 5, Bogor, West Java 16610, Indonesia; c: Department of Innovative Science and Technology for Bio-industry, Graduate School of Bioresource and Bioenvironmental Sciences, Faculty of Agriculture, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan; d: Department of Chemical Engineering, Faculty of Engineering, University of Jember, Jalan Kalimantan, Kampus Tegal Boto, Jember, East Java 68121, Indonesia; e: Department of Applied Chemistry, Graduate School of Engineering, Faculty of Engineering, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan; f: Department of Aquatic Products Technology, Faculty of Fisheries and Marine Sciences, Bogor Agricultural University, Jalan Raya Dramaga, Bogor, West Java 16680, Indonesia; * Corresponding author: tkitaoka@agr.kyushu-u.ac.jp
INTRODUCTION
Cellulose is the key material for the forthcoming sustainable society because it is renewable and the most abundant biomass on Earth (Sadeghifar et al. 2017). Nanocellulose, a crystalline bundle of cellulose macromolecules with nanometer-order width, has recently emerged as a promising nanomaterial for various practical applications because of its physicochemical properties such as high aspect ratio, elasticity, transparency, thermal stability, chemical durability, and biodegradability (Fukuzumi et al. 2009, 2010; de Mesquita et al. 2010). While woody biomass is the major cellulose resource, non-woody biomass, such as oil palm empty fruit bunches (OPEFB), a large-scale byproduct of oil palm plantations in Southeast Asia, is another attractive raw material for production of nanocellulose because it contains 60% cellulose (Azrina et al. 2017). Various types of nanocellulose are produced from cellulosic raw materials by physical disintegration, chemical/enzymatic treatment, or combined processing (Jonoobi et al. 2011; Nechyporchuk et al. 2016; Chen et al. 2017; Hastuti et al. 2018). Nanocellulose produced by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-catalyzed oxidation, namely “TEMPO-oxidized cellulose nanofibers” (TOCNs), have attracted considerable attention since they require low energy for production and afford the narrowest reported nanofibers (Isogai et al. 2011). TOCNs bear anionic carboxylate groups on their surfaces as the result of TEMPO-mediated selective oxidation of surface-exposed primary alcohols to carboxylates (Saito and Isogai 2004). These carboxylates induce a zeta potential as low as –70 mV on the solid surfaces, resulting in high dispersibility of TOCNs in water by electrical repulsion (Okita et al. 2010), and high dispersant effects for various suspended solid materials (Li et al. 2015).
Although nanocellulose has been extensively studied for developing reinforcing agents in composite plastics and films (Goetz et al. 2009; Yan et al. 2017), enhancers in emulsion systems (Kalashnikova et al. 2012; Hu et al. 2015), functional hydrogels for environmental remediation (Jin et al. 2015; Dwivedi et al. 2017), and carriers for metal catalysts and enzyme immobilization (Azetsu et al. 2013; Sulaiman et al. 2014; Uddin et al. 2017), its application in microbial systems remains limited. Yu et al. (2016) reported the effects of the size of nanocellulose on microalgal flocculation and lipid metabolism; cellulose nanofibrils effectively induced microalgal flocculation via a mechanical interaction based on geometric properties such as nanocellulose morphology and hydrogen bonding. Sun et al. (2014, 2015) observed aggregation of Pseudomonas fluorescens and Escherichia coli K12 in the presence of nanocellulose; they found that bacterial aggregation and adhesion to solid surfaces were significantly affected by the surrounding solution chemistry. The electrostatic interaction promoted by the charged cellulose nanocrystals could give rise to clustering, phase separation, and rapid aggregation of negatively charged bacteria (Larsen et al. 2009; Sun et al. 2012, 2014). Clustering and phase separation of bacteria are very important for product recovery in microbial biofuel production.
Practical applications of microbial systems in the production of biofuels from renewable resources are attracting much attention for environmental and economic reasons (Zheng et al.2009). Biobutanol produced by microbial fermentation is of significant interest as demand for butanol as an industrial intermediate is rapidly increasing, and it can be used directly in gasoline engines without any modification and/or substitution (Xue et al. 2017). Butanol is the most attractive biofuel alternative to ethanol because it has numerous desirable properties, including lower vapor pressure, higher calorific value, less corrosive properties, and a non-hygroscopic nature (Dürre 2007). Butanol can be produced from renewable resources through acetone–butanol–ethanol (ABE) fermentation (Lee et al. 1995) using Clostridia as a high-performance butanol-producing strain (Tashiro et al. 2004). The metabolic pathway of typical ABE fermentation of butanol-producing Clostridia strain (Jones and Woods 1986) is summarized in Fig. 1.
Fig. 1. Metabolic pathway of acetone–butanol–ethanol (ABE) fermentation in butanol-producing Clostridia strain. Several enzymes involved in ABE fermentation are described in italics. These abbreviations are: AK, acetate kinase; PTA, phosphotransacetylase; CoAT, CoA transferase; PTB, phosphotransbutyrylase; BK, butyrate kinase; BADH, butyraldehyde dehydrogenase; BDH, butanol dehydrogenase.
However, microbial butanol production has suffered from severe product inhibition in fermentation processes and high cost of product recovery, resulting in low butanol productivity. Studies have been conducted to overcome these obstacles by integrating a butanol separation step, such as cell immobilization, extractive fermentation, pervaporation, or perstraction (Barton and Daugulis 1992; Qureshi and Maddox 1995, 2005). Among these, extractive fermentation (liquid-liquid extraction) has great potential to increase the product titer if one can determine the most appropriate solvent for butanol selectivity that is compatible with the cells producing the butanol (Qureshi and Maddox 1995; Ishizaki et al. 1999). In extractive fermentation, the broth is in contact with an extracting solvent; therefore, some inhibitory products become dissolved in the solvent, resulting in reduction of inhibitory effects on the culture (Roffler et al. 1987). However, the efficiency of this method depends on the affinity of solutes for the extraction solvent and the mixing ratio of the phases (de Jesus et al. 2019). Extractive fermentation may result in inactivation of cells due to the extensive exposure of cells to extraction solvents and product toxicity (Ishii et al. 1985). Therefore, advanced technology to immobilize cells has been investigated to overcome such problems by using silica gel, pumice, and Ca-alginate as microbial carriers and has been applied in microbial fermentation (Napoli et al. 2010; Pereira et al. 2014). However, to the best of our knowledge, no study has reported an effective strategy using nanocellulose to improve microbial stability to enhance butanol productivity.
Fig. 2. Overview of use of 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-oxidized cellulose nanofibers (TOCNs) in microbial biobutanol production.
Herein, microbial stability in biobutanol production was addressed by using in-broth TOCNs, which were produced from OPEFB waste. The characteristics of raw pulp OPEFB and the resultant TOCNs are described in Table S1 and Figs. S1–S4 in the Appendix. TOCNs have surface anionic carboxylate groups with high density (up to 1.5 mmol g–1) and aspect ratio more than 40. The characteristics of TOCNs from company made from wood are described in Fig. S5; these can promote electrostatic repulsion of negatively charged bacteria producing biobutanol, which is important in microbial system stability by preventing bacterial aggregation and then enhancing microbial dispersibility (Fukuzumi et al. 2009; Sun et al. 2012). The strategy is illustrated in Fig. 2. The addition of TOCNs improved microbial stability and increased butanol production. The addition of TOCNs in extractive fermentation of biobutanol significantly improved cell growth and the total butanol concentration after 96-h fermentation. This new application of nanocellulose obtained from the low-cost agricultural residue OPEFB in bioalcohol production is expected to expand the potential of natural nanomaterials from biomass.
EXPERIMENTAL
Materials
Microorganism inoculation
Clostridium saccharoperbutylacetonicum N1-4 ATCC 13564 was used in this study. One milliliter of spore suspension of C. saccharoperbutylacetonicum was transferred from sand stock and refreshed in 9 mL of fresh potato glucose medium (10% v/v) (Ishizaki et al. 1999). The spore suspension was heat-shocked in a water bath at 100 °C for 1 min, and then refreshed at 30 °C for 24 h anaerobically using an Anaeropack (Mitsubishi Gas Chemical, Co., Inc., Tokyo, Japan). The refreshed culture broth was inoculated into tryptone–yeast extract–acetate (TYA) medium (Tashiro et al. 2004). After inoculation, the culture broth was incubated anaerobically at 30 °C for 15 h using an Anaeropack, and then used as a seed culture.
Methods
Extractive fed-batch fermentation with free cells
Fed-batch extractive fermentation with free cells was performed in a 25-mL portion of TYA broth and a 50-mL portion of extractant, which was composed of a 1:1 (v/v) mixture of oleyl alcohol and tributyrin; the volume ratio of extractant (Ve) to broth (Vb) was thus 2.0 (Ve/Vb = 2.0). In brief, seed culture (2.5 mL) was inoculated into a 25-mL portion of TYA medium, containing glucose (2.5 mL of 50 g/L solution), calcium carbonate (2.5 mL of 3 g/L solution), and TOCNs (5.0 mL of 0.5% w/v aqueous dispersion, 25 mg of TOCNs in dry weight; see the Supporting Information for preparation of TOCNs) in a 500-mL Erlenmeyer flask closed with a rubber seal, as an aqueous phase. The flask was then sparged with nitrogen gas for 10 min to obtain anaerobic conditions. The fed-batch cultures were grown at 30 °C in a shaker (100 rpm) and by feeding 1 g of glucose powder at 24, 48, and 72 h after seeding and maintained the anaerobic condition by sparging with N2 after glucose feeding.
Extractive fed-batch fermentation with immobilized cells
A solution of sodium alginate (4% w/v) was prepared in boiling water and autoclaved at 115 °C for 15 min. Precultured cells (2.5 mL) and TOCNs (0.5% w/v, 25 mg dry weight) were added into sterilized saline water (20 mL, 0.85% NaCl). The resultant suspension was mixed with an equal volume of 4% sodium alginate solution (final concentration of sodium alginate, 2%). The mixture was dropped into a 3% CaCl2 solution using a syringe with continuous stirring to form alginate beads containing cells and TOCNs. The resultant beads were recovered by filtration. Cell-containing alginate beads without TOCNs were also prepared as a control. The diameters of beads were approximately 5 mm. Extractive fed-batch fermentation with immobilized cells was performed as described above (extractive fed-batch fermentation with free cells).
Analytical methods
Cell density was determined by measuring the optical density of the suspension at 562 nm (OD562) using a UV/vis spectrophotometer (Bio-Spec, Shimadzu, Kyoto, Japan) after diluting the samples. The dry cell weight (DCW) was calculated as previously reported (Yoshida et al. 2012),
DCW = 0.301 × OD562 × D – 0.0008 (1)
where D is the dilution ratio.
The total butanol concentration [BuOH]Total was defined as the total amount of butanol produced in all the phases per broth volume (g/L-broth), and calculated as follows (Darmayanti et al. 2018),
[BuOH]Total = [BuOH]b Vb + [BuOH]c Vc + [BuOH]e Ve) / Vb (2)
where [BuOH]b and Vb are the butanol concentration (g/L) in the broth and the volume (L) of the broth, [BuOH]c and Vc are the butanol concentration in the cell beads and the volume of the beads, and [BuOH]e and Ve are the butanol concentration in the extractant and the volume of the extractant, respectively. Vc the volume of alginate beads, is 1.5 mL for every measurement because six beads were crushed in aqueous sodium citrate solution in a total volume of 1.5 mL.
The concentration of glucose was measured in the supernatant liquid obtained by centrifugation of broth by using a high-performance liquid chromatograph (US-HPLC-1210, JASCO, Tokyo, Japan) equipped with a refractive index detector and SH-1011 column (Shodex, Tokyo, Japan). Aqueous H2SO4 (0.05 mM) was used as the mobile phase (1.0 mL/min, 50°C), using an injection volume of 20 μL. The concentration of butanol was measured in supernatant obtained by centrifugation of both extractant and broth using a gas chromatograph (6890A, Agilent Technologies, Palo Alto, CA, USA) equipped with a flame ionization detector and a 15-m capillary column (INNOWAX 19095N-121, Agilent Technologies) in previously reported conditions (Tashiro et al. 2004).
The distribution coefficient (Kd) of butanol between the extractant (oil phase) and broth (aqueous phase) was calculated using Eq. 3.
Kd = [BuOH]e / [BuOH]b (3)
The treatments of free TOCNs and the presence of TOCNs were evaluated with analysis of variance (ANOVA) reported as p-values. Findings with p-values less than 0.05 suggested that differences were statistically significant.
Microscopic analysis
Samples for scanning electron microscopy (SEM) analysis were prepared as follows. Samples of free cells were prepared by collecting a 1.5-mL portion of broth, which was centrifuged at 120 rpm for 20 min to obtain cells as the precipitate. For immobilized cells, several alginate beads were collected from the broth and cut into smaller pieces. A 500-µL portion of broth from free cells or the cut beads containing immobilized cells were fixed by 2% formaldehyde (300 µL) and phosphate buffer (1 mL) at pH 5.5 and 4 °C, overnight. Half of the fixed cells were collected and washed with deionized water and centrifuged at 120 rpm for 10 min.
The precipitate was stained with 1% OsO4 solution for 4 h and washed with deionized water. The specimen was washed successively with 50, 70, 80, and 99.5% ethanol, each for 5 min. The specimens were freeze-dried and observed using an SU-8000 apparatus (Hitachi, Tokyo, Japan) at the Center of Advanced Instrumental Analysis, Kyushu University.
The cells were observed using a confocal laser scanning microscope (LSM 700, Carl Zeiss AG, Oberkochen, Germany). A 500-µL portion of culture medium containing free cells was collected in a 1.5-mL plastic tube and washed with phosphate buffer at pH 5.5 by centrifugation.
The precipitate was stained with 4′,6-diamidino-2-phenylindole (DAPI), and washed again with phosphate buffer by centrifugation. The cells in alginate beads were treated in the same manner after destroying the beads by treatment with 500 µL 0.2 M sodium citrate for 2 h. A 20-µL portion of prepared cells in phosphate buffer was put on an observation glass and observed under a 405-nm laser.
RESULTS AND DISCUSSION
Extractive Fed-batch Fermentation with Free Cells
The fermentation behavior of free cells was monitored in terms of DCW, glucose consumption, and butanol production for 96 h in the presence of OPEFB-derived TOCNs (op-TOCN), and compared with the results using wood-derived TOCNs (w-TOCN), and in the absence of TOCNs (control). As shown in Fig. 3, the cells drastically increased by 48 h and then stayed in a stationary phase until 96 h in the presence of both op-TOCN and w-TOCN; in contrast, cell growth was almost negligible over 96 h in the absence of TOCNs (Fig. 3A).
Fig. 3. Time course of extractive fermentation using free Clostridium saccharoperbutylacetonicum N1-4 cells. Glucose was fed every 24 h. (A) Dry cell weight (DCW), (B) glucose concentration, and (C) total butanol concentration. Filled circles: oil palm empty fruit bunches (OPEFB)-derived TOCNs (op-TOCN); open circles: wood-derived TOCNs (w-TOCN); triangles: control, no TOCNs.
An increase of cell density in the presence of op-TOCN and w-TOCN is in good accordance with the glucose consumption, in which fed glucose was efficiently consumed over 48 h in the presence of op- and w-TOCNs (Fig. 3B). Glucose consumption became slower after 48 h of fermentation, corresponding to the stationary phase of the cultures. In contrast, very little glucose was consumed in the absence of TOCNs. Butanol production was drastically improved in the presence of both op- and w-TOCNs, reaching 30 g/L total butanol concentration after 96-h fermentation, while the butanol concentration was almost negligible in the absence of TOCNs (Fig. 3C). It is noteworthy that the total butanol concentrations in the presence of op- and w-TOCNs were higher (28.8 and 30.3 g/L-broth after 96 h, respectively) than that reported in previous work (24.2 g/L-broth) (Darmayanti et al. 2018) in which extractive fermentation was carried out by the free cell method using a large ratio of extractant to broth (Ve/Vb = 5.0) and C. saccharoperbutylacetonicum N1-4 was used as butanol-producing strain. The distribution coefficients of butanol (Kd) in the present TOCN systems were also higher (3.98 for op-TOCN and 4.97 for w-TOCN) than in the previous work (Kd = 3.14).
The time course of cell growth is in good accordance with the general features of the metabolic pathway of butanol production by C. saccharoperbutylacetonicum N1-4, which consists of two phases, acidogenesis and solventogenesis (Jones and Woods 1986). The growth behavior in the first 48 h corresponds to acidogenesis, during which cells rapidly grow while producing acetic acid and butyric acid. Solventogenesis occurs in the stationary phase, during which the cells reassimilate the previously excreted acids to form acetone, butanol, and ethanol (Shinto et al. 2007).
Enhancement of butanol production with TOCNs could be explained by increased dispersibility of bacteria during fermentation, which improves the microenvironment of cells. According to the Derjaguin–Landau–Verwey–Overbeek theory, dispersibility of suspended solids such as bacteria is dominated by repulsive electrical double layer forces (Larsen et al. 2009). The surface of both TOCNs and bacterial cells are negatively charged through carboxylate and phosphate groups, respectively. Therefore, their anionic nature causes improved colloidal stability of the bacterial system by preventing flocculation between bacteria (Sun et al. 2012). In another aspect, carboxylate groups in TOCNs were possibly involved in calcium binding, resulting in keeping repulsion between bacterial cells, even in the presence of Ca2+ ions. In this work, the increased dispersibility presumably enhanced bacterial growth, as illustrated in Fig. 4, resulting in higher butanol production.