Introduction

Worries regarding to the current crisis of climate change and depletion of fossil fuels make the utilization of bioethanol as an attractive option for combating both global warming and less dependence on fossil fuels. A further increase in bioethanol production may come from lignocellulosic biomass, such as agricultural residues and wood. Lignocellulose from waste products is abundant, inexpensive, and renewable. Cellulose, which is a homopolysaccharide composed of β-D-glucopyranose units linked by β-1,4 glycosidic bonds, comprises around 40-50 wt.% of plant biomass. One of the best known cellulose decomposers is termite, which is able to hydrolyze 74-99% of the cellulose ingested. Taking advantage of this high efficiency with which termites digest cellulose, and due to the essentiality of β-glucosidase catalyzing the hydrolysis of cellobiose or cello-oligomers into the fermentable sugar, glucose, in the process of cellulose degradation, the objective of this work is to heterologously express, purify, characterize, and compare three endogenous glycoside hydrolase family 1 (GH1) β-glucosidases from termites. Further comparison was done with the commercial β-glucosidase Novozym 188. In addition, the effect of addition of termite β-glucosidases or Novozym 188 to Celluclast 1.5 L on the degradation of Avicel was studied. This work might introduce alternative and efficient enzymes to the bioethanol production field.

Chapter 1. Heterologous expression and production of endogenous β-glucosidases from termites

Despite the high efficiency of termites in the degradation of cellulosic materials, the production level is too low to be used for commercial exploitation. In this work I have used the expression systems of Aspergillus oryzae and Pichia pastoris to produce three endogenous β-glucosidases from termites. G1NkBG, an enzyme found in the salivary gland of the lower termite Neotermes koshunensis, was expressed in A. oryzae, a filamentous fungus that can produce and secrete large amounts of proteins and has a Generally Recognized as Safe (GRAS) status. Since production of two other enzymes, G1sgNtBG1 and G1mgNtBG1 from the salivary gland and midgut, respectively, of the higher termite Nasutitermes takasagoensis, was not successful in A. oryzae due to premature polyadenylation of transcripts, P. pastoris was used to express them. P. pastoris has been widely used to express heterologous proteins due to its easy manipulation and low cost of production. Both G1sgNtBG1 and G1mgNtBG1 were successfully expressed in a new vector called pBGP3, which is introduced in Chapter 4.

Chapter 2. Purification and characterization of β-glucosidases

G1NkBG was purified by ammonium sulfate precipitation followed by anion exchange, hydrophobic, and gel filtration chromatographies. G1sgNtBG1 and G1mgNtBG1 were purified by affinity (Ni2+-NTA) chromatography. In the hydrolysis of lignocellulose, it is well-known that most β-glucosidases are strongly inhibited by the end-product, glucose, which may restrict the whole degradation process of cellulose. Differently from the majority of β-glucosidases, G1NkBG was not only resistant to glucose inhibition, but its activity against p-nitropheny β-D-glucopyranoside was stimulated by glucose: more than 90% of its maximum activity was retained in the presence of 1 M glucose, and with 0.2 M glucose the activity was stimulated by 1.3-fold. Although G1sgNtBG1 and G1mgNtBG1 did not show any stimulation by glucose, they were also very resistant to glucose inhibition. G1mgNtBG1 showed relatively high optimum temperature at 65°C. Regarding the thermostability, G1mgNtBG1 displayed more than 88% and 64% of its maximum activity, respectively, after 5 h of incubation at 55°C and 60°C. All β-glucosidases studied were active on cello-oligosaccharides. Having β-glucosidases resistant to glucose inhibition, with high thermostability, and active on cello-oligosaccharides, is very interesting for biotechnological applications, and they can be useful in bioethanol production.

Chapter 3. Comparative analysis on hydrolytic activities of purified β-glucosidases from termites with those of commercially available cellulases

For complete hydrolysis of cellulose, synergistic action of three different enzymes known as endoglucanase, cellobiohydrolase, and β-glucosidase is needed. In industry two kinds of commercial cellulases are usually used in the hydrolysis of cellulose; Celluclast 1.5 L from Trichoderma reesei and Novozym 188 from Aspergillus niger. While the content of the former is mainly endoglucanases and cellobiohydrolases, the latter is mainly composed of β-glucosidases. In this Chapter the thermostability and glucose-resistance of β-glucosidases from termites and Novozym 188 were compared. In addition, the cooperation of G1NkBG, G1sgNtBG1, G1mgNtBG1, and Novozym 188 with Celluclast 1.5 L in the hydrolysis of Avicel was examined. The results showed that G1mgNtBG1 was more thermostable than Novozym 188. All β-glucosidases from termites tested were far more glucose-tolerant than Novozym 188. When added to Celluclast 1.5 L, G1mgNtBG1 produced more reducing sugars than Novozym 188, G1NkBG, and G1sgNtBG1. This suggests that G1mgNtBG1 more efficiently removed accumulated cellodextrins in the reaction mixture which could have inhibited the activity of cellulase components in Celluclast 1.5 L. Hence, the results further suggest that G1mgNtBG1 serves as an enzyme that shows better synergism with other cellulolytic components.

Chapter 4. Vector construction for expression of N-tagged heterologous proteins in P. pastoris

Purification of proteins usually requires laborious works. Very often more than one chromatography step is needed, and setting/employing the best conditions for each steps is time-consuming and can lead to low yields and increased costs of production. With these issues in mind, recombinant proteins are usually expressed as fusions with extra sequences added at either N- or C-terminal end, which allows one-step purification by affinity chromatography. Hexahistidine tag is often used due to the high affinity of polyhistidines to the immobilized metal ion matrices such as Ni2+, Zn2+, and Co2+. It is difficult, however, to predict which of the N- or C-terminal addition of tags is compatible with the correct folding, detection, and purification of the protein expressed. Hence having several options of vectors is helpful in determining the best way to express and purify expressed proteins. For this purpose, a sequence for the addition of N-terminal c-Myc and 6×His tags was inserted into pBGP1, a vector for the expression of C-terminal tagged proteins in P. pastoris, giving origin to pBGP2. Next, the redundant C-terminal tags in pBGP2 were removed to generate pBGP3. To test the effectiveness of pBGP3, three β-glucosidases (G1NkBG, G1sgNtBG1, and G1mgNtBG1) and two endoglucanases (RsEG and NtEG) from termites were expressed using this vector. Western blot analysis with anti-c-Myc antibody confirmed the production of the enzymes. Purification was performed in one-step by Ni2+-affinity chromatography as confirmed by SDS-PAGE analyses. These results demonstrate the efficacy of pBGP3 for the expression and one-step purification of heterologous proteins.

Conclusion

In parallel with an increasing demand for bio-based fuels, there is also a further interest for new technologies that improve lignocellulose hydrolysis. One way to reach such achievement is the research on cellulolytic enzymes which facilitate the degradation of cellulose. Particularly, as discussed in this work, β-glucosidase is a key enzyme in this process. I have worked on three endogenous β-glucosidases from termites, G1NkBG, G1sgNtBG1, and G1mgNtBG1. All of them were more glucose-tolerant than the majority of β-glucosidases including Novozyme 188, and active on cello-oligosaccharides. Moreover, G1mgNtBG1 was thermostable at 60°C. G1NkBG, G1sgNtBG1, and G1mgNtBG1 produced more reducing sugars than Celluclast 1.5 L alone, suggesting that termite β-glucosidases were effective in alleviating the inhibitory effect of cellodextrins on cellulases. Hence, they are potential enzymes that can be used as a supplement in the hydrolysis of lignocellulosic biomass.

近年、地球温暖化の進行や石油価格高騰などの諸問題の深刻化、また化石燃料枯渇への懸念を受け、再生可能エネルギーであるバイオマスからのエタノール等の生産に期待が集まっている。現在、米国やブラジルを中心にトウモロコシやサトウキビなどからバイオエタノールが生産されているが、それら可食性の作物を燃料生産に利用することには食糧との競合という大きな問題が潜在している。この問題の解決のためには、非可食性の木質バイオマスからのエタノール生産技術を確立することが決定的に重要であるが、植物細胞壁由来の多糖からなる木質バイオマスを糖化することは容易ではなく、実用化に至っていない。一方、自然界では様々な生物が木質バイオマス、特にセルロースを分解してエネルギー源としており、中でもシロアリは極めて高効率にセルロースを資化できることが知られている。本研究は新たな木質バイオマス利用技術の確立を目指し、シロアリ由来の3種類のGHF1 β-グルコシダーゼの異種生産系の確立、精製および性質の解析を行ったものである。論文は序章、結果を述べた4つの章、および総括と展望を記した終章からなる。

第一章では3つのβ-グルコシダーゼ、即ちコウシュンシロアリのG1NkBG(以下NkBG)、タカサゴシロアリのG1sgNtBG1とG1mgNtBG1(以下各々sgNtBGとmgNtBG)の異種生産について述べている。3つの酵素とも麹菌Aspergillus oryzaeの発現システムを利用して生産を試みたが、NkBGのみ生産が認められた。sgNtBGとmgNtBGが生産されない原因を調べたところ、これらを導入した麹菌株ではそのコード領域の途中でポリAが付加された未成熟mRNAが作られていることがわかった。興味深いことに、両遺伝子を酵母Pichia pastorisに導入したところ問題なく生産されたことから、麹菌とP. pastorisにおいてはポリA付加シグナルの認識機構がかなり異なるものと考えられた。

第二章では3つのβ-グルコシダーゼの精製と性質の解析について述べている。NkBGは麹菌培養上清から硫酸アンモニウム沈殿と数種のクロマトグラフィーを行うことにより精製を行った。sgNtBGとmgNtBGはそのN末端に付加した6×Hisタグを利用してNi2+-NTAカラムによる精製を行った。精製酵素を用いて各々のVmaxとKm値をp-nitrophenyl-β-D-glucopyranoside (pNPG) を基質に調べたところ、NkBGはそれぞれ16 U/mg、と0.23 mg/ml、sgNtBGは0.8 U/mgと0.2 mg/ml、mgNtBGは8.0 U/mgと0.2 mg/mlであった。

一般にβ-グルコシダーゼはその活性が反応産物であるグルコースによって阻害されることが知られている。3つのβ-グルコシダーゼについてグルコース感受性を調べたところ、興味深いことにNkBGの活性は0.1~0.6 Mグルコースの存在下でやや亢進し、1 Mグルコース存在下でも92%の活性を保持することがわかった。sgNtBGとmgNtBGの活性はグルコースによって濃度依存的に阻害されたが、それでも一般的なβ-グルコシダーゼに比べ高いグルコース耐性を示した。

第三章ではシロアリ由来の3つのβ-グルコシダーゼと、市販のβ-グルコシダーゼ剤であるNovozyme 188との比較を行っている。その結果、耐熱性、グルコース耐性などの点においてmgNtBGが優れた性質を持つことが分かった。また、市販のセルラーゼ剤であるCelluclastを用いた結晶性セルロース分解においても、mgNtBGがNovozyme 188に比べより高い協調的効果を示すことがわかった。

第四章ではP. pastorisを用いた異種タンパク質生産において有用なベクターpBGP3の作製について述べている。pBGP3は発現タンパク質のN末端にmycエピトープタグと6×Hisタグが付加されるように設計されたベクターであり、発現タンパク質は出芽酵母Saccharomyces cerevisiae由来のプレプロβ-ファクターとの融合タンパク質として生産され、この配列を利用して分泌経路に送られる。プレプロβ-ファクター配列は分泌系路上で切断され、目的タンパク質のみが分泌される。第一章におけるsgNtBGとmgNtBGの生産にはこのベクターが用いられており、その有用性が実証された。

以上、本論文はシロアリ由来の3種類のβ-グルコシダーゼの異種生産と精製、性質の解析、およびそれらが示すバイオマス分解促進効果を示したものであり、学術的・応用的に貢献するところが少なくない。よって審査委員一同は本論文が博士(農学)の学位論文として価値あるものと認めた。