Kobe University Repository : Thesis

学位論文題目Tit le

Engineering of high-cellulolyt ic Saccharomyces cerevisiae using cell-surface display technique: Towards consolidated bioprocessing(統合型バイオエタノール生産プロセスに資する高効率セルロース分解酵母の開発)

氏名Author Liu, Zhuo

専攻分野Degree 博士（工学）

学位授与の日付Date of Degree 2016-09-25

公開日Date of Publicat ion 2017-09-01

資源タイプResource Type Thesis or Dissertat ion / 学位論文

報告番号Report Number 甲第6745号

権利Rights

JaLCDOI

URL http://www.lib.kobe-u.ac.jp/handle_kernel/D1006745※当コンテンツは神戸大学の学術成果です。無断複製・不正使用等を禁じます。著作権法で認められている範囲内で、適切にご利用ください。

PDF issue: 2021-07-27

DOCTORAL THESIS

Engineering of high-cellulolytic Saccharomyces

cerevisiae using cell-surface display technique:

Towards consolidated bioprocessing

統合型バイオエタノール生産プロセスに資する

高効率セルロース分解酵母の開発

Department of Chemical Science and Engineering

Graduate School of Engineering

KOBE UNIVERSITY

LIU ZHUO July 2016

CONTENTS

Page

Introduction 1

Synopsis 13

Chapter I. 16

Combined cell-surface display- and secretion-based strategies for production of

cellulosic ethanol with Saccharomyces cerevisiae

Chapter II. 43

Engineering of a novel cellulose-adherent cellulolytic Saccharomyces

cerevisiae for cellulosic biofuel production

Chapter III. 72

Efficient ethanol production from crystalline cellulose using high-cellulolytic

Saccharomyces cerevisiae with optimized cellulase ratios on the cell surface

General conclusion 87

References 91

Acknowledgements 104

Publication lists 106

1

INTRODUCTION

Cellulosic biofuels

With the increasing worldwide industrialization, a steep rise for the demand of

petroleum-based fuels appeared. Currently, fossil fuels take up 80% of the primary

energy consumed in the world1, but the sources of these fossil fuels are becoming

exhausted and generating serious environmental issues, such as greenhouse gas

emissions and acid rains. Increasing energy demand also leads to an increase in crude

oil price, directly affected to global economic activity. Thus, many researchers are

focusing on the renewable, environment-friendly, cost-effective energy alternatives

for fossil fuels. In the past decades, biofuels (e.g., ethanol, methanol, biodiesel)

emerge as the most environment-friendly alternative fuels due to their renewability,

security of supply and generating acceptable quality exhaust gases2.

The feedstock for 1st generation biofuel production is starch/ sucrose-rich food

crops, such as corn, wheat, barley, and sugarcane3. Although the 1st generation

biofuels production is commercial in many countries (e.g., corn ethanol in US,

sugarcane ethanol in Brazil), the major obstacle toward its global commercialization

is the concern of “food-versus-fuel” debate4, that whether to use these starch/

sucrose-rich crops as fuel feedstock or as human food. On the contrary,

lignocellulosic biomass (e.g., sugarcane bagasse, newspaper, rice straw) is of great

interest as its low-price, large-scale availability, and no competition with food,

referring as the 2nd generation feedstock for biofuel production. Cellulose is

composed of β-1,4-linked glucose monomers, representing the most abundant

polysaccharide (up to 30-50%) in lignocellulosic biomass. After physical or

thermochemical pretreatments, cellulosic materials are enzymatically decomposed

into monosaccharides (glucose), and subsequently converted to desirable biofuels via

microbial fermentation. However, the extensive hydrogen linkages among glucose

2

molecules lead to a crystalline and strong matrix structure of cellulose 5. As a result, it

requires large amount of costly cellulases to break the recalcitrance of cellulose,

which makes it challenging for cellulose-based biofuels to be economically feasible 6.

Thus, the decrease of the need for enzyme addition may lead to a feasible path

towards the cost-competitive production of cellulosic ethanol.

Consolidated bioprocessing

Conventional technology for lignocellulosic biofuel production includes a

physical/chemical pretreatment step during which polymeric sugar fractions more

accessible to enzyme. Subsequent bioconversion process generally consists of four

steps: the enzyme production, the hydrolysis of cellulose and hemicellulose into

monosaccharides, and the fermentation of sugars into biofuels (Figure 1). Combining

two or more of these steps into one integrated process will greatly decrease the

processing cost. Consolidated bioprocessing (CBP), which allows the four biological

steps occurring in one reactor using a CBP microorganism7, contributes to reducing

the costs in capital investment, substance and utilities associated with microbial

enzyme production. It has been widely considered as the most promising way to

achieve the commercialization of cellulosic ethanol production7.

Figure 1 Schemes of conventional process and CBP process for lignocellulosic biofuel

production

3

The key to CBP is the engineering of a microorganism capable of both producing

saccharifying enzymes to degrade lignocellulosic biomass and efficiently fermenting

sugars into biofuels. Microorganisms used for CBP can be categorized into two

groups: cellulase producers and ethanol producers. Cellulase producers include

cellulolytic thermophilic bacteria (e.g., Clostridium thermocellum), cellulolytic fungal

(e.g., Trichoderma reesei) and some anaerobic filamentous fungi8. These

microorganisms contain high cellulolytic activities, however, their ethanol production

capacities and tolerances to ethanol are relatively low. On the contrary, ethanol

producer group (e.g., Saccharomyces cerevisiae, Zymomonas mobilis, Kluyveromyces

marxianus) are natively possesses superior fermenting capacity and tolerance to

ethanol, but lacks essential cellulolytic activities to degrade cellulose into glucose. To

resolve this problem, researches are working on expressing heterologous cellulases

derived from filamentous fungi or Clostridium species in yeast strains9, 10. Currently,

the production of cellulases in yeast strain follows two major strategies: secreting

enzymes into the fermentation broth or immobilzing enzymes on the cell exterior

using cell-surface display engineering.

Cell-surface display engineering

Cell-surface display engineering is a promising technique that uses microbial

functional components to locate enzymes or peptides on the cell exterior of

microorganisms. This technique endows the engineered strain with novel functions,

such as whole-cell biocatalysts, bio-adsorbents, biosensor, vaccine-delivery vehicles

and screening platforms 11. As shown in Figure 2, cell-surface display engineering has

been applied to the biorefinery of various types of waste materials. In the biorefinery

of sugar-/protein-rich waste to produce biofuels and biochemicals, hydrolytic

enzymes (e.g. cellulase, hemicellulose, and amylase) are immobilized on the surface

of yeast to perform biodegradation and bioconversion of sugars simultaneously. With

4

regard to waste oils, lipase-displaying yeast cells are employed as whole-cell

biocatalysts, which catalyze the transesterification of triglycerides with short-chain

alcohols into biodiesels. Moreover, yeast cells displaying metal-binding proteins on

the cell surface can also be used as bioadsorbents to retrieve heavy metals from

wastewater.

Cell-surface display systems have been successfully developed in various

microorganisms, such as Escherichia coli and Streptococcus gordonii12. Yeast is one

of the most suitable host strains for cell-surface display, because of its rigid cell walls

(around 110-200 nm wild), as well as useful platform for protein production, since

yeast allows the folding and glycosylation of expressed heterologous eukaryotic

proteins. The cell wall of yeast strain S. cerevisiae mainly constitutes cross-linked β-1,

3/1,6-glucans, mannoproteins, and chitin. Although the β-1,6-glucan takes up

quantitatively a minor component of the cell wall, it plays a central role in anchoring

cell-wall protein. Glycosylphosphatidylinositol (GPI) proteins are kind of

glucanase-extractable cell-wall proteins that are linked with β-1,6-glucan via the GPI

anchor, and play critical roles in the structural rigidity of cell walls (e.g., Cwp2p and

Sed1p) and cell functions (e.g., Flo1p for cell flocculation)13. During the secretory of

a GPI protein, GPI anchor is bound onto GPI-attachment signal on the membrane of

endoplasmic reticulum; while secretion signal directs GPI protein transported to cell

surface. On reaching plasma membrane, GPI protein is cleaved at GPI-anchor site by

a phosphatidylinositol-specific phospholipase C and subsequently released into cell

wall. GPI protein is then covalently bound with β-1,6-glucan via remnant

GPI-anchor13. Yeast cell-surface display technique is to fuse the targeted enzyme or

peptide to a GPI protein/ GPI anchoring domain, and this is anchored covalently on

the cell surface via the GPI anchor14. The characteristics and applications of GPI

5

anchoring domains in yeast strains have been summarized in Table 1.

Anchoring of enzymes or peptides on the cell surface to perform cellulose

degradation has the following advantages: (1) Proteins are produced and

auto-immobilized on the cell surface via easy cellpropagation, which can reduce the

expenditure and facilities needed for protein generation and enrichment37. (2) The

proteins that are anchored on solid surfaces (such as cell surface) are more stable than

free-form proteins under extreme conditions (e.g., high temperature and organic

solvents)38. Improved enzyme stability will facilitate long-term storage and recycling

in industrial processes. (3) Tethering of multiple synergetic enzymes on a single cell

significantly shortens the enzyme-to-enzyme distance16, preventing long-distance

mass transfer of substrates, especially in high-solid fermentation. (4) Monomers (e.g.,

glucose, xylose, and amino acid) that are liberated via enzymatic hydrolysis can be

immediately utilized by cells because of the proximity of the enzyme to cell surface.

As a result, the monomers in the extracellular environment are maintained at low

concentrations, which significantly reduce the risk of contamination (especially when

glucose is released from cellulose) and diminishes the repression effects of substrates

on microorganism. (5) The proteins that are immobilized on the cell surface can be

easily recollected through centrifugation or filtration, implying high potentials of

recycling.

6

7

Was

te S

ugar�

Was

te O

il�

Cel

lulo

se�

Cel

lula

se�

Prot

ease�

Lip

ase�

Met

al-b

indi

ng p

rote

in�

Anc

hor

prot

ein�

Pept

ide�

Yeas

t cel

l�

Glu

cose�

2Pyr

uvat

e�

2Ace

tald

ehyd

e�E

than

ol�

Was

te P

rote

in�

Was

te W

ater�

Met

al io

n�

Abs

orpt

ion� R

ecov

ery�

Cel

l mas

s�

Fatt

y ac

id a

lkyl

est

ers�

Methanol�

Tria

cylg

lyce

ride� Glycerol�

Am

ino

acid�

Glu

tath

ione

Figu

re 2

Sch

emes

for v

ario

us fo

rms o

f was

te b

iore

finer

y us

ing

cell-

surf

ace

disp

lay

engi

neer

ing.

8

Applications of cell-surface display technique in cellulosic ethanol production

Currently, cell-surface display technique has been widely applied in the ethanol

production from cellulosic materials. In order to degrade cellulose into glucose, at

least three kinds of synergistic cellulases are required, including β-glucosidase (BGL),

endoglucanase (EG), and cellobiohydrolase (CBH)8. Display of heterologous

synergistic cellulases derived from filamentous fungi or Clostridium species on the

surface of yeast enables it to directly generate biofuels from cellulosic substrates. As

shown in Table 2, Fujita et al. (2002) and Yanase et al. (2010) showed that BGL- and

EG-displaying yeast strains could efficiently convert β-glucan (linear, water-soluble

polysaccharides composed of six or seven β-1, 4-linked glucose residues) into ethanol,

with a high yield of more than 90% of the theoretical level. In comparison, phosphoric

acid swollen cellulose (PASC) is an insoluble cellulosic material mainly composed of

amorphous cellulose, and more resistant to enzymatic hydrolysis than β-glucan.

Although it has been demonstrated that cellulase-displaying yeasts are able to directly

produce ethanol from PASC, most of their ethanol yields were less than 60% (Table

2). This is because the cellulolytic activities on the cell surface are insufficient to

support the complete degradation of PASC.

In addition, as shown in Table 2, the current cellulase-displaying yeast strains

seem incapable of direct degrading lignocellulosic substrates in the absence of

exogenous cellulases, owing to the recalcitrance of lignocellulose. A recombinant S.

cerevisiae co-displaying BGL, EG and CBHII achieved 89% of the theoretical ethanol

yield from rice straw in the presence of 10 FPU/g-biomass commercial cellulase,

while it obtained an ethanol yield of 65% under 5 FPU/g-biomass cellulase39,

indicating that the cellulosic ethanol yield is strongly related to the support of

additional cellulase. As for natural lignocellulosic materials, crystalline cellulose

serves as the major form of cellulose. The highly rigid structures of crystalline

9

cellulose are resistant to enzymatic depolymerization, making it more difficult to be

degraded than the fluffy cellulosic materials (e.g., β-glucan and PASC). Thus,

improvement of the degradation ability towards crystalline cellulose is of great

importance in achieving CBP using cell-surface display engineering.

To solve these problems, much effort has been devoted to increasing the

enzyme-display efficiency or improving cellulase production. In order to alleviate the

conflict between cell-wall protein and displayed-protein, Kotaka et al. disrupted the

gene encoding cell-wall protein SED1, obtaining a 1.6-fold increase of enzyme

activity on the cell surface 40. In addition, the genes involved in the protein secretory

pathway of S. cerevisiae were over-expressed, achieving a 71% increase in CBH

production 41. However, these improvements only result in a limited increase of

cellulose degradation ability, which are still insufficient for the hydrolysis of

crystalline cellulose. Therefore, in the present study, three novel strategies have been

employed to improve the cellulose degradation ability of cellulase-displaying S.

cerevisiae: study of the most suitable strategy for cellulase production (Chapter I),

improvement of interactions between cell and cellulose (Chapter II), and optimization

of enzyme ratios that displayed on the cell surface (Chapter III).

Firstly, we investigated the suitable strategy for cellulase production from the

view of steric hindrance. The location of cellulase (immobilized on the cell surface or

secreted into medium) is of great importance in enzyme synergism and the mass

transfer of intermediate products. It has been reported that BGL is preferable to be

displayed on the cell surface rather than secreted because of the improved stability 38.

However, "display" systems may suffer from inefficiency of processive enzymes (e.g.,

CBH1 and CBH2), or cause steric restriction in the collision with cellulose. To date,

the best locations (or production strategies) for EG and CBH remained obscure. Thus,

10

in this study, EG and CBH1 were produced heterologously in a BGL-displaying S.

cerevisiae strain using cell-surface display, secretion, or a combined strategy. The

most suitable strategy for producing EG and CBH1 for cellulose degradation was

evaluated. (described in Chapter I)

Secondly, we investigated the interactions between cellulase-displaying cells and

cellulose during cellulose degradation process. Currently, large amounts of researches

have been pursued to improve the cellulolytic activities on yeast cell surface through

delicate design of surface-displayed scaffoldin42, increasing enzyme diversities43, or

increasing the expression level of cellulase genes. However, very limited reports

could provide the information of cellulose degradation mechanism via

surface-displayed cellulase, which correlates with the interactions of

enzyme-to-cellulose and cell-to-cellulose. In this study, we focus on the

cellulose-adherent characteristics of cellulase-displaying cells that could be used to

improve the cellulose hydrolysis efficiency, such as enhancement of cell-to-cellulose

interactions via altering the surface properties of cellulose. (described in Chapter II)

Thirdly, we engineered a recombinant yeast strain with high degradation ability

towards crystalline cellulose through optimizing of the enzyme ratio on the cell

surface. It is postulated that improvement of the crystalline-cellulose degradation

ability in yeast cell will enable it direct utilization of natural cellulosic materials. Most

current reports of cellulase-displaying yeast strain focus on enhancing the degradation

potential of amorphous cellulose (e.g., PASC) (Table 1). However, the enzymatic

hydrolysis pattern of crystalline cellulose, especially the optimal ratios of synergisitic

cellulases, is completely different from that of amorphous cellulose. It has been

reported that engineering of EG as the main proportion in total cellulase mixture is

preferable for amorphous cellulose degradation, while using CBHs as the major

11

component is to be assumed more effective in the degradation of crystalline

cellulose44. In the present study, we integrated multiple-copy of cellulase genes (EG,

CBH1, CBH2) into the genome of BGL-displaying yeast strain using cocktail

δ-integration method44. The transformants were screened based on their degradation

abilities toward Avicel. Moreover, the enzyme ratios that displayed on the cell surface

were determined via nano LC-MS/MS. (described in Chapter III)

12

13

SYNOPSIS Chapter I

Combined cell-surface display- and secretion-based strategies for production of

cellulosic ethanol with Saccharomyces cerevisiae

The production of cellulase is usually pursued by one of the two strategies:

displaying enzyme on the cell surface or secreting enzyme into the medium. However,

to our knowledge, the combination of the two strategies in a yeast strain has not been

employed. In this chapter, heterologous endoglucanase (EG) and cellobiohydrolase 1

(CBH1) were produced in a β-glucosidase (BGL) -displaying S. cerevisiae strain

using cell-surface display, secretion, or a combined strategy. Strains EG-D-CBH1-D

and EG-S-CBH1-S (with both enzymes displayed on the cell surface or with both

enzymes secreted to the surrounding medium) showed higher ethanol production (2.9

g/L and 2.6 g/L from 10 g/L phosphoric acid swollen cellulose, respectively) than

strains EG-D-CBH1-S and EG-S-CBH1-D (with EG displayed on cell surface and

CBH1 secreted, or vice versa). After 3-cycle repeated-batch fermentation, the

cellulose degradation ability of strain EG-D-CBH1-D remained 60% of the 1st batch,

at a level that was 1.7-fold higher than that of strain EG-S-CBH1-S. This work

demonstrated that placing EG and CBH1 in the same space (on the cell surface or in

the medium) was favorable for amorphous cellulose-based ethanol fermentation. In

addition, the cellulolytic yeast strain that produced enzymes by the cell-surface

display strategy performed better in cell-recycle batch fermentation compared to

strains producing enzymes via the secretion strategy.

14

Chapter II

Engineering of a novel cellulose-adherent cellulolytic Saccharomyces cerevisiae

for cellulosic biofuel production

The main obstacle toward the economic feasibility of cellulosic bioethanol

production is the recalcitrance of lignocellulose requiring large amount of costly

enzyme to break. To reduce the need for enzyme addition, we successfully engineered

a high-efficiency cellulolytic Saccharomyces cerevisiae by displaying four synergistic

cellulases (BGL, EG, CBH1, and CBH2) on the cell surface. The cellulase-displaying

yeast strain exhibited clear cell-to-cellulose adhesion and a “tearing” cellulose

degradation pattern; the adhesion ability correlated with enhanced surface area and

roughness of the target cellulose fibers, resulting in higher hydrolysis efficiency. The

engineered yeast could directly produced ethanol from rice straw despite a more than

40% decrease in the required enzyme dosage for high-density fermentation. Thus,

improved cell-to-cellulose interactions provided a novel strategy for increasing

cellulose hydrolysis, suggesting a mechanism for promoting the feasibility of

cellulosic biofuel production.

15

Chapter III

Efficient ethanol production from crystalline cellulose using high-cellulolytic

Saccharomyces cerevisiae with optimized cellulase ratios on the cell surface

Crystalline cellulose is the most rigid, degradation-resistant structure in natural

cellulosic materials, and is widely considered as the major obstacle to achieve the

efficient utilization of lignocellulose. To improve the degradation ability towards

crystalline cellulose in cellulolytic yeast strain, the optimal ratio of synergisitic

cellulases (BGL, EG, CBH1, and CBH2) displayed on the cell surface was

investigated in this chapter. We constructed a pool of cellulase-displaying yeast strain

with various enzyme ratios through cocktail δ-integration method. Strain A26, which

achieved the highest cellulosic ethanol yield (57%, corresponding to 1.46-fold of that

in control strain (cellulase ratio is 1:1:1:1)) without any metabolic burden on cell

growth was screened from the yeast pool. The cellulase ratios on the cell surface were

analyzed using nano LC-MS/MS. To our knowledge, this is the first work on relative

quantitation of four kinds of cellulase (BGL, EG, CBH1, and CBH2) anchored on

single yeast cell, and the first to optimize enzyme ratio towards crystalline cellulose

degradation.

16

Chapter I

Combined cell-surface display- and secretion-based strategies for

production of cellulosic ethanol with Saccharomyces cerevisiae

17

Introduction

Due to the limitations in fossil fuel supplies and environmental issues, bioethanol

derived from lignocellulosic materials has recently gained increased attention1, 5.

Saccharomyces cerevisiae is the most commonly used microorganism for ethanol

production, but lacks essential cellulolytic enzyme activities to degrade cellulose into

glucose52. To resolve this problem, the construction of recombinant yeast strains

capable of producing heterologous cellulases, including β-glucosidase (BGL),

endoglucanase (EG), and cellobiohydrolase (CBH), has been pursued over the last

two decades10, 39.

Currently, the production of cellulases follows two major strategies: displaying

enzymes on the cell surface or secreting enzymes into the fermentation broth. The

glycosylphosphatidylinositol (GPI) anchoring system enables the display of various

kinds of enzymes on the cell surface14. The cell-surface display strategy increases the

effective concentration of enzymes, and promotes a greater degree of synergy16. In

addition, glucose liberated from cellulose in proximity to the cell surface is

immediately taken up, thereby minimizing the risk of contamination or product

inhibition53. Furthermore, immobilizing enzymes on the cell surface enables the

re-use of enzymes and cells in multi-batch fermentations, which reduces the cost of

yeast propagation and that of supplementation with extraneous enzymes54, 55. In

contrast, secreting enzymes into the medium recreates the “free enzyme system”,

which is similar to the cellulase system of filamentous fungi. The quantity of secreted

enzymes is limited only by the production capacity of cells, not by physical

restrictions, such as the incorporation capacity of yeast cell wall associated with

cell-surface display56. Moreover, free cellulases can penetrate into the secondary cell

walls of plant biomass57, increasing the accessibility of cellulose, which was reported

as the critical factor in enzymatic hydrolysis58.

18

Thus, each strategy has both advantages and disadvantages. The selection of an

optimal strategy for enzyme production should be based on the characteristics of a

given enzyme and its reaction mechanism. It has been reported that

cell-surface-displayed BGL exhibited higher efficiency in cellobiose usage than

secreted BGL because of the improved stability caused by immobilization on the cell

wall38. However, “display” systems may suffer from inefficiency of processive

enzymes (e.g., CBH), leading to decreased hydrolysis efficiency compared to free

enzyme systems56. Thus the combination of cell-surface display and secretion

strategies into one recombinant yeast strain was expected to achieve improved

hydrolysis of cellulose compared to either single strategy of displaying or secreting

cellulases. Such a combined strategy may allow the various kinds of enzymes to be

produced in their most appropriate location, assembling the advantages of the two

strategies into one system of enzyme production.

Cellobiohydrolase I (CBH1) is the major component (~ 60%) of the total

cellulolytic protein of the cellulase system of Trichoderma reesei7. CBH1 acts by

hydrolyzing from the reducing end of crystalline cellulose fibers in a progressive

manner. Recently, CBH1 was reported as the main contributor to overall cellulose

degradation, with other enzymes synergistically enhancing its hydrolytic efficiency59.

Although CBH1 has been heterologously expressed and secreted in S. cerevisiae, the

relatively low titer60, 61 and low specific activity62 of secreted CBH1 has limited the

study of co-expression of CBH1 with other cellulolytic enzymes. Recently, CBH1

originating from Talaromyces emersonii fused with the T. reesei C-terminal

carbohydrate-binding module (CBM) was efficiently expressed in S. cerevisiae with a

yield of 100-200 mg/L, which is approximately 20-fold higher than the expression

levels of T. emersonii CBH1 reported elsewhere63. In separate work, immobilization

19

of enzyme on the cell surface was reported to improve the stability of the enzyme38.

However, to our knowledge, there has been no previous report of displaying CBH1 on

the cell surface of yeast strain.

In the present study, EG and CBH1 were produced heterologously in a

BGL-displaying S. cerevisiae strain using cell-surface display, secretion, or a

combined strategy. The most suitable strategy for producing EG and CBH1 for

cellulose degradation was evaluated. Direct conversion of cellulose into ethanol was

conducted by cellulolytic yeast strains and then applied in cell-recycle batch

fermentation for further evaluation. To our knowledge, the work reported here is the

first study on displaying CBH1 on the yeast cell surface and the first study on the

feasibility of combining the cell-surface display and secretion strategies in one yeast

strain for heterologous cellulase production. We believe that this work will

significantly increase our knowledge of how to engineer optimal yeast strains for

biofuel production from cellulosic biomass.

Materials and methods

Microbial strains and media

The relevant features and sources of the yeast strains used in this study are listed

in Table I-1. Strain Escherichia coli NovaBlue (Novagen, Inc., Madison, WI, USA)

was used for the propagation of the plasmids. Bacterial cells were grown at 37 °C in

Luria-Bertani broth (10 g/L tryptone, 5 g/L yeast extract, and 5 g/L sodium chloride)

containing 100 mg/L ampicillin. Haploid yeast S. cerevisiae BY4741 (Life

Technologies, Carlsbad, CA, USA) was used for cellulase production. Yeast

strainswere screened and pre-cultivated in synthetic dextrose (SD) medium (6.7 g/L of

yeast nitrogen base without amino acids (Difco Laboratories, Detroit, MI, USA) and

20 g/L of glucose) supplemented with the appropriate amino acids in a shaker

20

21

incubator (150 rpm) at 30 °C, and then aerobically cultivated at 30 °C in YPD

medium (20 g/L peptone (Bacto-PeptoneTM, Difco Laboratories), 10 g/L yeast extract

and 20 g/L glucose). Ethanol fermentation was performed in YP medium (10 g/L

yeast extract and 20 g/L peptone) containing either 10 g/L PASC or 10 g/L β-glucan

from barley (Megazyme, Bray, Ireland). PASC was prepared from Avicel PH-101

(Fluka Chemie GmbH, Buchs, Switzerland), as previously described 9.

Plasmid and strain construction

The plasmids and primers used in this study are summarized in Table I-2 and

Table I-3, respectively. To construct the plasmid pRDH225, the gene T. reesei EG2

was cloned as a 1277 bp PacI/AscI fragment from pRDH14764 into pBZD265 to

form pRDH225. To construct pRDH226, the T. emersonii CBH1 encoding gene

containing a domain encoding a carboxy-terminal CBM originating from the T. reesei

CBH1 was amplified using Phusion hi-fidelity polymerase (Thermo Scientific) as

directed by the manufacturer from pMI529 63 as template with primers TeCBH1-L

and TeCBH1-R. The resulting 1567 bp fragment was cloned as a PacI/AscI fragment

into pBZD2 to form pRDH226.

The integrative plasmids for cell-surface display with the SED1 anchor were

constructed as follows: the DNA fragment encoding EG2 from T. reesei was

amplified from plasmid pRDH225 by PCR using the primers TrEG2-F and TrEG2-R.

The cell-surface display cassette, which includes the I2 region (the 3’ non-coding

region between gene YFL021W and YFL020C, used for integration), LEU2, SED1

promoter, SED1 anchoring region, and SAG1 terminator was amplified from plasmid

pIL2GA-SS using primers P-EG2 and EG2-A. Two DNA fragments were connected

by the isothermal assembly method66, generating the plasmid pIL2-EGD. To construct

the secretion expression cassette without the SED1 anchoring region, amplification

22

was performed using the plasmid pIL2GA-SS as a template with primers P-EG2 and

EG2-T. Primers TrEG2-F and TrEG2-R2 were used for amplifying the fragment

encoding EG. The resulting plasmid from the combination of two DNA fragments

was named pIL2-EGS.

Table I-2 Characteristics of the integrative plasmids used in this study

Plasmid Relevant features References

pRDH225 KanMX, expression of T. reesei EG2 gene This study

pRDH226 ZeoR, expression of T. emersonii CBH1 gene This study

pIL2GA-SS LEU2, display of R. oryzae glucoamylase [67]

pIU5GA-SS URA3, display of R. oryzae glucoamylase [67]

pIBG-SS HIS3, display of A. aculeatus BGL1 [25]

pIL2-EGD LEU2, display of T. reesei EG2 This study

pIL2-EGS LEU2, secretion of T. reesei EG2 This study

pIU5-CBH1D URA3, display of T. emersonii CBH1 This study

pIU5-CBH1S URA3, secretion of T. emersonii CBH1 This study

R. oryzae, Rhizopus oryzae; A. aculeatus, Aspergillus aculeatus; T. reesei, Trichoderma

reesei; T. emersonii, Talaromyces emersonii; BGL1, β-glucosidase 1; EG2, endoglucanase 2;

CBH1, cellobiohydrolase 1

The construction of CBH1 integrative plasmids was performed by a process

similar to the above description. For the cell-surface display plasmid, the DNA

fragment of CBH1 from T. emersonii was amplified from plasmid pRDH226 using

primers TeCBH1-F and TeCBH1-R, and fused with the PCR product amplified from

plasmid pIU5GA-SS by primers P-CBH1 and CBH1-A. The resulting plasmid, which

was named pIU5-CBH1D, can integrate into I5 region (the 3’ non-coding region of

gene YLL055W and YLL054C). Plasmid pIU5-CBH1S is the integrative plasmid with

the secretion expression cassette, connected by the segment of CBH1-encoding gene

(primes TeCBH1-F and TeCBH1-R2) and the PCR products amplified from plasmid

23

pIU5GA-SS (primers P-CBH1 and CBH1-T for secretion expression cassette).

Table I-3 The primers used in this study

Plasmids were transformed into S. cerevisiae BY4741 using lithium acetate as

described68. The transformants were identified using colony PCR to check the

integration of the cellulase gene expression cassettes (primers I2-F and I2-R for I2

insertion of EG-encoding cassette, and primers I5-F and I5-R for I5 insertion of

CBH1-encoding cassette). Transformants with one copy of the cassette were selected

for subsequent experiments.

Primers Sequence (5’-3’)

TeCBH1-L GACTTTAATTAAAATGCTAAGAAGAGCTTTACTATTG

TeCBH1-R GACTGGCGCGCCTTACAAACATTGAGAGTAGTATGGG

TrEG2-F AATACGTTCGCTCTATTAAGATGAACAAGTCTGTTGCTCCATTG

TrEG2-R GTTGATAATTTACTCGAGCCTAACTTTCTAGCCAAACATGAAGAAACC

TrEG2-R2 CTCAATGTACTAACTGTACATTATAACTTTCTAGCCAAACATGAAGAAAC

TeCBH1-F AATACGTTCGCTCTATTAAGATGCTAAAGAAGAGCTTTACTATTGAGC

TeCBH1-R GTTGATAATTTACTCGAGCCCAAACATTGAGAGTAGTATGGGTTT

TeCBH1-R2 CTCAATGTACTAACTGTACACTACAAACATTGAGAGTAGTATGGGTTT

P-EG2 GGAGCAACAGACTTGTTCATCTTAATAGAGCGAACGTATTTT

EG2-A CATGTTTGGCTAGAAAGTTAGGCTCGAGTAAATTATCAACTGTCC

EG2-T GTTTGGCTAGAAAGTTATAATGTACAGTTAGTACATTGAGTCTAAATA

P-CBH1 AGTAAAGCTCTTCTTAGCATCTTAATAGAGCGAACGTATTTT

CBH1-A CATACTACTCTCAATGTTTGGGCTCGAGTAAATTATCAACTGTCC

CBH1-T ACTACTCTCAATGTTTGTAGTGTACAGTTAGTACATTGAGTCTAAATA

I2-F GAAGCCGCGAGTACGAACAATGATG

I2-R TGGTATTTTCGTGAGCAAACCCAAC

I5-F CATTGAAGAAGGGAAAGTGGTAACC

I5-R TCCCTCTCTAATCTGGGTGAGAC

rt-ACT1-F TGGATTCCGGTGATGGTGTT

rt-ACT1-R TCAAAATGGCGTGAGGTAGAGA

rt-EG-F GGTTGTTTGTCTTTGGGTGCTTAC

rt-EG-R AATTGAGCATTTGTTGGACCACCTT

rt-CBH1-F CAACTTACTGTCCAGACGACGAAAC

rt-CBH1-R AAGGAAGAACCAGAGGAGGTAACAC

24

Quantification of the transcription level of cellulase-encoding genes by real-time

PCR

The transcription levels of the cellulase- encoding genes were quantified by

real-time PCR as described previously69. The PCR primers BGL 761F and BGL

858R44 were used to determine the transcription level of gene BGL1. Primers rt-EG-F

and rt-EG-R were used for the EG2 gene and primers rt-CBH1-F and rt-CBH1-R

were used for the CBH1 gene. Transcription levels of the target genes were

normalized to the housekeeping gene ACT1 (primers rt-ACT1-R and rt-ACT1-F).

Yeast cell growth assay

To measure cell growth, the parent strain and the engineered strains were

cultivated individually in SD medium at 150 rpm for 24 h at 30 °C. The pre-cultured

medium was inoculated into 5 mL YPD medium in a L-shaped vitreous tube at the

initial OD660 of 0.05 and cultivated at 30 °C. The value of the OD660 was measured

once hourly using a TVS062CA Bio-photorecorder (Advantec Toyo, Tokyo, Japan).

The value of the OD660 was taken as an indicator of cell growth.

Ethanolic fermentation

Recombinant yeast strains were pre-cultivated in SD medium for 24 h, then

inoculated into YPD medium and aerobically cultured in YPD medium at 30 °C for

72 h. Cells were harvested by centrifugation at 3,000 × g for 10 min at 4 °C, and then

washed twice with sterile distilled water. The wet cell pellet was weighed and then

resuspended in 20 mL YP medium containing 10 g/L PASC or β-glucan from barley

at an initial cell concentration of 150 g wet cells/L (PASC) or 50 g wet cells/L

(β-glucan). Ethanol fermentation was performed at 37 °C for 96 h with 200 rpm

agitation in 100 mL closed bottles, each equipped with a siliconized tube and check

valve (Sanplatec Corp., Osaka, Japan) as a CO2 outlet under the oxygen-limited

25

conditions. The ethanol concentration in the fermentation medium was determined

using a gas chromatograph (model GC-2010; Shimadzu, Kyoto, Japan), as described

previously48.

In the cell recycle batch fermentation, after the 96-h batch fermentation

described above, cells were collected by centrifugation at 8,000 × g for 10 min at 4 °C.

The pelleted cells were inoculated into fresh YP medium supplemented with 10 g/L

PASC. The fermentation was repeated three times sequentially under the

oxygen-limited conditions.

To measure PASC amount in fermentation, the fermentation broth (including the

cells and residual PASC) was sterilized at 121 °C, 20 min (to terminate glucose

consumption by yeast cells) and then cooled to room temperature. Sterilized medium

was incubated with 3 FPU/g-biomass PASC commercial cellulase (Cellic CTec2;

Novozymes Inc., Bagsvaerd, Denmark) for 2 h at 50 °C. After the hydrolysis reaction,

the supernatant was obtained by centrifugation at 8,000 × g, 10 min, 4 °C. Glucose

concentration in the supernatant was measured by the Glucose CII kit (Wako Pure

Chemical Industries, Ltd., Osaka, Japan) and taken as the amount of PASC remnant.

Enzyme assay

At the 0-h and 96-h time points of ethanol fermentation, fermentation medium

was assayed for PASCase, and individual cellulase activities. PASCase activity

represents the PASC degradation ability of all enzymes present. Fermentation broth

was added into a final concentration of 5 g/L PASC in 50 mM sodium citrate buffer

(pH 5.0) and 100 mM methyl glyoxal (Nacalai Tesque, Inc., Kyoto, Japan); the

methyl glyoxal prevents the assimilation of glucose by yeast cells70. The reaction was

performed at 50 °C for 4 h using a heat block (Thermo Block Rotator SN- 06BN;

26

Nissin, Tokyo, Japan) with shaking at 35 rpm, and the supernatant was collected by

centrifugation for 10 min at 8,000 × g at 4 °C to remove cells and debris. The amount

of glucose in the supernatant was measured by the Glucose CII kit. One unit of

PASCase activity (U/mL) was defined as the amount of enzyme needed to produce 1

µmoL of glucose per minute at 50 °C, pH 5.0.

The medium of PASC fermentation was used for the BGL, EG, and CBH1

activity assays. The BGL and EG activities were determined as previously described25.

One unit of the BGL activity was defined as the enzyme amount required for

production of 1 µmoL p-nitrophenol (pNP) in 1 min at 30 °C (U/mL). One unit of EG

activity was defined as the absorption at 590 nm of released blue dye in 1 h at 38 °C

(U/mL). p-nitrophenyl-β-lactopyranoside (pNPL, Sigma Co. Ltd, St. Louis, MO, USA)

was used for the measurement of CBH1 activity as previously described71. One unit of

CBH1 activity (U/mL) was defined as the enzyme amount required for production of

1 µmoL pNP in 1 min at 50 °C.

Results

Construction of yeast strains

In this study, the haploid yeast strain S. cerevisiae BY4741 was used as the host

strain for the heterologous expression of cellulase genes. The plasmids containing

gene expression cassettes are listed in Table I-2. All the gene expression cassettes

included the SED1 promoter and the SAG1 terminator. The secretion signal peptide of

BGL1 was derived from Rhizopus oryzae glucoamylase, while EG2 and CBH1 were

produced with their native secretion signals. The GPI-anchoring region used for

cell-surface display was constructed using the full length S. cerevisiae SED1 gene to

display cellulases on the cell surface. It has been reported that the combination of

SED1 promoter and SED1 anchoring region in a gene cassette enables highly efficient

27

immobilization of enzyme into the cell wall25. Each cellulase gene was integrated into

a separate internal open reading frame (ORF) region (I2 region for EG2 gene and I5

region for CBH1 gene) and confirmed by polymerase chain reaction (PCR). Figure I-1

shows the cellulase production scheme of the recombinant yeast strains constructed in

this study. Strain BY-BG-SS, which displayed Aspergillus aculeatus BGL1 on the

cell surface, was reported previously25. The T. reesei EG2 gene expression cassettes,

with and without the SED1 anchoring region, were integrated into the genome of

strain BY-BG-SS to yield strains EG-D and EG-S, respectively. Next, the expression

cassettes of the T. emersonii CBH1 gene, with and without the SED1 anchoring region,

were integrated into the genome of strain EG-D to yield EG-D-CBH1-D and

EG-D-CBH1-S or into the genome of strain EG-S to yield EG-S-CBH1-D and

EG-S-CBH1-S. The engineered yeast strains in this study are listed in Table I-1.

Figure I-1 Schematic description of the recombinant yeasts strains constructed in this study.

In addition, the transcription levels of cellulase genes were determined after 72 h

of cultivation. The transcription levels of the BGL1, EG2, and CBH1 genes, each of

28

which were under the control of a SED1 promoter, were similar among all

transformants (Figure I-2).

Figure I-2 Relative transcription levels of cellulolytic enzyme-encoding genes in

recombinant yeast strains. Gene ACT1 was used as the internal standard. The relative

transcription levels were shown normalized to the level observed in strain EG-D-CBH1-D,

whose relative transcription level was defined as 1. For each strain, data are presented as the

mean ± SD from three independent experiments.

Effect of multiple gene expression on cell growth

Cell growth was profiled to determine the metabolic burden caused by the

expression of heterologous cellulase genes. Each of the engineered strains was

inoculated into liquid YPD media and cultivated aerobically for 72 h at 30 °C. The

host strain S. cerevisiae BY4741 was used as a reference strain. As shown in Figure

I-3, no apparent difference in cell growth was observed between the host strain and

any of the recombinant yeast strains.

29

Figure I-3 Time-course profiles of cell growth using host strain BY4741 and recombinant

yeast strains in YPD medium. Each strain was inoculated in YPD medium to an initial OD660

of 0.05 and then cultured aerobically at 30 °C, 150 rpm for 72 h. For each strain, data are

presented as the mean ± SD from three independent experiments.

Direct ethanol production from cellulosic materials

Barley β-glucan and phosphoric acid swollen cellulose (PASC) were utilized as

fermentation substrates. β-glucan is a linear, water-soluble polysaccharide composed

of 6 or 7 β-1,4-linked glucose residues49. PASC, which is derived from phosphoric

acid treatment of Avicel PH-101, is an insoluble cellulosic material with more

amorphous regions and a lower degree of crystallinity compared to Avicel72, 73.

As depicted in Figure I-4A, ethanol production from 10 g/L β-glucan was

performed using strains EG-D and EG-S. Yeast strains were cultivated in YPD

medium for 72 h; cells then were collected by centrifugation and inoculated into

fermentation medium at an initial cell concentration of 50 g wet cells/L. The ethanol

fermentation was conducted under oxygen-limited conditions at 37 °C for 24 h.

30

Ethanol production by strain EG-D initiated immediately after the start of

fermentation and reached a maximum of 4.1 g/L after 6 h of fermentation. In contrast,

strain EG-S exhibited a long lag phase before the start of ethanol production; no

ethanol was detected until 9 h of fermentation.

Figure I-4 Time course of direct ethanol production from cellulosic materials by recombinant

strains EG-D and EG-S. (A) Ethanol production from β-glucan. (B) Ethanol production from

PASC. For each strain and time point, data are presented as the mean ± SD from three

independent experiments.

The fermentation abilities of EG-D and EG-S also were evaluated by performing

direct ethanol production from 10 g/L PASC (Figure I-4B). The fermentation was

conducted under oxygen-limited conditions at 37 °C for 96 h with an initial cell

concentration of 150 g wet cells/L. Ethanol production by strain EG-D peaked at 1.5

g/L ethanol at 72 h, while the production by strain EG-S peaked at 0.6 g/L at 48 h.

These results revealed that locating EG on the cell surface improved the ethanol

production from both soluble and insoluble cellulosic materials.

31

Figure I-5 Time course of direct ethanol production from PASC by recombinant strains

EG-D-CBH1-D, EG-D-CBH1-S, EG-S-CBH1-D, and EG-S-CBH1-S. For each strain and

time point, data are presented as the mean ± SD from three independent experiments.

To investigate the most suitable strategy for EG and CBH1 production, direct

ethanol production from 10 g/L PASC was evaluated using recombinant strains

EG-D-CBH1-D, EG-D-CBH1-S, EG-S-CBH1-D, and EG-S-CBH1-S. As shown in

Figure I-5, after 96 h of fermentation, ethanol production by strains EG-D-CBH1-D

and EG-S-CBH1-S peaked at 2.9 g/L and 2.6 g/L, respectively. Ethanol production by

strain EG-D-CBH1-S peaked at 2.3 g/L at 96 h while strain EG-S-CBH1-D peaked at

1.2 g/L at 24 h. To further characterize the fermentation capacity of our constructs,

the strains were compared by evaluating the PASCase and individual cellulase

enzyme activities in PASC fermentation, and by testing the strains in cell-recycle

batch fermentation.

Enzyme activity in direct ethanol production from PASC

The cellulose degradation ability of cellulolytic strains is considered as one of

the critical factors in the conversion of cellulose into ethanol. In this study, PASCase

32

Figure I-6 PASCase activities and cellulase activities of the cellulolytic S. cerevisiae strains

in PASC fermentation. (A) PASCase activity at 0 h of the fermentation. (B) PASCase activity

at 96 h of the fermentation. (C) Cellulase activity at 0 h of the fermentation. (D) Cellulase

activity at 96 h of the fermentation. For each strain and time point, data are presented as the

mean ± SD from three independent experiments.

33

activity represents the PASC-degradation capability of the cellulolytic yeast strains.

The PASCase activity at 0 h and 96 h of ethanol production from 10 g/L PASC was

investigated. As shown in Figures I-6A and I-6B, PASCase activity was highest

(among the four recombinant yeast strains) in EG-D-CBH1-D at both 0 h (50.9

mU/mL) and 96 h (53.6 mU/mL). The PASCase activity of strain EG-S-CBH1-S

increased from 26.9 mU/mL at 0 h to 47.8 mU/mL at 96 h. By contrast, the PASCase

activity of strain EG-S-CBH1-D decreased after 96 h of fermentation, exhibiting the

lowest activity (34.1 mU/mL) compared with other recombinant yeast strains. The

activity of BGL, EG, and CBH1 from yeast strains at 0 h and 96 h of PASC

fermentation also was determined, as illustrated in Figures I-6C and I-6D. BGL

activity was similar among the four recombinant strains and remained similar (within

a given strain and among different strains) during the fermentation. EG activity at 96

h was highest in strain EG-D-CBH1-D (24.2 U/mL) when compared with that in the

other cellulolytic yeast strains. Initial (0 h) CBH1 activity was highest in strain

EG-D-CBH1-D (4.7 U/mL), and lowest in strain EG-S-CBH1-S. After 96 h of

fermentation, CBH1 levels appeared to rise similarly in the four strains, achieving

activities of ~ 7-10 U/mL.

Cell-recycle batch fermentation

To investigate the efficiency of strains EG-D-CBH1-D and EG-S-CBH1-S in a

continuous process, cell-recycle fermentation was conducted under anaerobic

conditions (Figure I-7). Recombinant yeast cells were collected for recycling and 10

g/L PASC was added into the fermentation medium at the beginning of each run. In

the first batch, 69.3% and 55.9% of PASC (corresponding to 6.9 g/L and 5.6 g/L

PASC) was converted into 2.9 g/L and 2.5 g/L ethanol by strain EG-D-CBH1-D and

strain EG-S-CBH1-S, respectively. In the 3rd batch, 41.8% of PASC was consumed

34

0 10

20

30

40

50

60

70

80

90

100

0

0.5 1

1.5 2

2.5 3

3.5

0 24

48

72

96

12

0 14

4 16

8 19

2 21

6 24

0 26

4 28

8

PASC (%)

Ethanol (g/L)

Tim

e (h

)

EG-D

-CB

HI-

D e

than

ol

EG-S

-CB

HI-

S et

hano

l EG

-D-C

BH

I-D

PA

SC

EG-S

-CB

HI-

S PA

SC

Cyc

le 1�

Cyc

le 2�

Cyc

le 3�

Figu

re I

-7 T

hree

cyc

les

of C

RB

F us

ing

reco

mbi

nant

S. c

erev

isia

e st

rain

s E

G-D

-CB

H1-

D a

nd E

G-S

-CB

H1-

S. P

ASC

(%)

indi

cate

s th

e pe

rcen

tage

of

resi

dual

PA

SC n

orm

aliz

ed t

o th

e in

itial

con

cent

ratio

n in

the

res

pect

ive

cycl

e. T

he i

nitia

l

amou

nt o

f PA

SC in

eac

h cy

cle

was

def

ined

as

100%

. For

eac

h st

rain

and

tim

e po

int,

data

are

pre

sent

ed a

s th

e m

ean

± SD

from

thre

e in

depe

nden

t exp

erim

ents

.

35

by EG-D-CBH1-D, corresponding to ~60% of the consumption in the 1st batch.

In contrast, the consumption of PASC in strain EG-S-CBH1-S decreased from

55.9% to 19.4% after 3-cycle repeated fermentation. Correspondingly, the final

ethanol titer generated by strain EG-D-CBH1-D was 1.7-fold higher than that

generated by strain EG-S-CBH1-S in the 3rd batch. These results suggested that

associating EG and CBH1 with cells facilitated retention of cellulolytic activity even

after three cell-recycles.

Discussion

In this study, we integrated heterologous EG and CBH1 genes into the genome of

a BGL-displaying S. cerevisiae, permitting the production of EG and CBH1 via

cell-surface display, secretion, or a combined strategy. The recombinant strains that

produced EG and CBH1 in the same space (on the cell surface or in the medium)

showed superior performance in the production of cellulosic ethanol. To our

knowledge, this is the first report on combining cell-surface display and secretion

strategies in a single yeast strain for heterologous cellulase production.

The benefits of attaching BGL to the yeast cell wall have been reported

previously38. However, suitable strategies for EG and CBH production by cells

remained obscure. In the present study, direct ethanol production from β-glucan and

PASC was performed using cellulolytic yeast strains. Specifically, the production of

ethanol was compared to investigate the suitable strategy for producing various kinds

of cellulases. When β-glucan was used as the fermentation substrate, the ethanol

production rate of strain EG-D was apparently higher than that of EG-S. Notably,

strain EG-D was able to convert β-glucan into ethanol immediately after the start of

fermentation, indicating that the displayed BGL and EG were successfully transferred

36

into the fermentation medium with the cell inoculum. In contrast, strain EG-S

converted β-glucan into ethanol after a 9-h lag, consistent with the need for this strain

to accumulate (via secretion) EG in the medium following inoculation. This

phenomenon demonstrated the advantage of an EG-display system to the cellulose

fermentation process, since display permitted early onset of the production of ethanol.

Compared with β-glucan, PASC contains a higher degree of polymerization and

crystallization, rendering this substrate more difficult to degrade by BGL and EG.

Indeed, fermentation with PASC yielded apparently lower rates of ethanol production

than those seen upon fermentation with β-glucan. The decline of ethanol observed in

PASC fermentation is probably due to the consumption via yeast cells when cellulases

could not hydrolyze sufficient glucose from PASC. Notably, the rate of ethanol

production from PASC by EG-D appeared higher than that by EG-S.

Subsequently, the CBH1 gene was successfully expressed along with BGL and

EG genes in S. cerevisiae. The T. emersonii CBH1 used in this study was fused with

the CBM of T. reesei CBH163; use of the CBM has been shown to enhance the

adsorption of enzyme to its substrate and modify substrate surfaces to facilitate

enzymatic hydrolysis74. We observed that co-production of CBH1 with displayed EG

enhanced ethanol production by up to 2-fold, demonstrating that the heterologous

CBH1 that was present assisted in the degradation of PASC. In the display strategy,

the CBH1 gene was expressed via the SED1 expression cassette and anchored on the

cell surface of S. cerevisiae using the SED1 anchoring domain25. It has been reported

that the expression level of the SED1 gene was highly induced in the stationary phase

by various environmental stresses, such as ethanol75. In the present work, expression

of both CBH1 and EG gene by cell-surface expression cassettes (in strain

EG-D-CBH1-D) permitted a doubling of CBH1 activity at 96 h vs. 0 h of

fermentation, confirming the utility of the stress-induced SED1 expression cassette.

37

However, we note that the SED1 anchoring domain was fused to the C-terminus of

the CBH1-CBM chimera protein; N- and C-terminal of CBM were fused with CBH1

and SED1 anchoring domain respectively, which may hinder the function of the CBM.

Alternatively, an N-terminal anchoring domain, such as the N-terminus flocculation

functional domain of Flo1p, may be more suitable for displaying the chimeric CBH1

containing a C-terminal CBM14. Nonetheless, to our knowledge, the present work

represents the first report on displaying a CBH1 on the cell surface of yeast strain.

In a previous report, a yeast strain displaying BGL, EG, and cellobiohydrolase 2

(CBH2) on the cell surface yielded higher ethanol production than a strain secreting

the corresponding enzymes in free form76. Immobilized BGL on the cell wall is

considered more appropriate for cellulosic ethanol production compared to free BGL

38; our use of cell-surface-displayed BGL is presumably one of the reasons for the

elevated ethanol yields in the present study. Ethanol production by strain

EG-S-CBH1-S was similar to that of strain EG-D-CBH1-D. To understand this

interesting result, the mechanism of the enzymatic hydrolysis of cellulose should be

taken into account. EG can randomly cleave the amorphous regions of cellulose to

produce oligosaccharides and provide free chain ends for CBH activity. Then CBH1

can initialize cleavage from the free chain ends, degrading crystalline cellulose into

cellobioses in a processive manner 77. The binding of EG and CBH1 onto the

cellulose surface, along with the processive movement of CBH1, are considered

essential steps in the degradation of cellulose 77. Besides enzymes, the presence of a

living microorganism is also important to the hydrolysis of cellulose. Generally, the

radius of a spherical yeast cell is around 2 µm 78, which is nearly 400-fold larger than

the hydrodynamic radius of a cellulase protein (≥ 5 nm) 79. The immobilization of

enzymes onto the cell wall may block the movement of processive enzymes (e.g.,

38

CBH1) or cause steric restriction in the collision with cellulose. As shown in Figure

I-8A, in the case of strain EG-D-CBH1-D, three different cellulases were displayed on

the cell-surface in a relatively close proximity (e.g., CBH1-BGL distance) compared

to the strains constructed by the other strategies. Such co-localization is expected to

increase the occurrence of synergistic interactions among cellulases16 and to facilitate

the transportation of glucose into cells. However, due to the size of the cell-enzyme

complex, the penetration of EG and CBH1 into the internal space of cellulose is

expected to be limited; such enzyme penetration has been reported as an important

determinant of hydrolytic rate58 80. Additionally, the processive movement of CBH1

may be retarded due to its immobilization on the cell. In previous pre-steady-state

analyses of CBH1 activity on cellulose, stalling of the processive movement of CBH1

was reported to lead to lower specific activity 81, 82. By contrast, EG and CBH1 in

strain EG-S-CBH1-S were produced as free forms (Figure I-8D), a strategy that was

expected to decrease steric hindrance and to increase the chance of collision with the

substrate. Although the cellulases appeared to accumulate in fermentation with

EG-S-CBH1-S (rising from 27 mU/mL at 0 h to 48 mU/mL at 96 h), the cellulolytic

enzyme activity was still lower than that of EG-D-CBH1-D. Additionally, the

diffusion efficiency of enzymes might affect the ability to degrade cellulose,

especially in a substrate with higher viscosity (e.g., PASC). Thus, co-locating EG and

CBH1 in the same space (on the cell-surface or in the medium) is favorable for

amorphous cellulose-based ethanol fermentation.

By contrast, ethanol production levels were apparently lower in strains

EG-D-CBH1-S and EG-S-CBH1-D. We infer that yeast cells attach to the surface of

cellulose to facilitate the collision between immobilized cellulases and cellulose, but

39

that this proximity may block the access of secreted EG or CBH1 to the surface of the

substrate (Figure I-8B, I-8C). For instance, the cellulase activities of strain

EG-D-CBH1-S at 96 h were higher than those of EG-S-CBH1-S, while the PASCase

Figure I-8 The effect of different locations of EG and CBH1 on the conversion of

cellulose into ethanol. A EG-D-CBH1-D. B EG-D-CBH1-S. C EG-S-CBH1-D. D

EG-S-CBH1-S. The diagrams suggest the multiple factors involved in the degradation of

cellulose, such as the distance between synergistic enzymes (CBH1-BGL distance), the effect

of cell-surface display on the processive movement of CBH1 (Retarded/ Un-retarded CBH1

movement), and the steric restriction during the binding of cellulases to cellulose surface

(Enzyme binding path)

activity was lower than that of strain EG-S-CBH1-S, effects that may be due to the

steric hindrance mentioned above. Additionally, the hydrolysis efficiency of strain

EG-S-CBH1-D appeared to be further decreased, presumably by the retarded

40

movement of immobilized CBH1 on the cell surface, as evidenced by the lower

PASCase activity and ethanol production obtained in strain EG-S-CBH1-D.

Interestingly, the individual enzyme activities in strain EG-D-CBH1-D were raised

while PASCase activity kept constant. We assume that in the case of cell-surface

displayed cell, only the enzymes located in the contacting region between yeast cell

and cellulose can involve in cellulose degradation. Although the individual enzyme

activities in overall cell were raised, the improvement of cellulolytic activity in

“attaching region” is limiting. In contrast, most of free cellulases can bind to cellulose

and participate in hydrolysis; the increase of individual enzyme activities can be fully

reflected on improvement of cellulolytic activity. This may also explain why the

overall enzyme activity of strain EG-S-CBH1-S (secretion system) observably rose

(77%) along with the increase of individual activities. Nonetheless, our results do

indicate that the involvement of the microorganism in the synergism of cellulases may

affect their hydrolysis efficiency towards cellulose. Moreover, the cellulose

degradation process perhaps involves more complex interactions, such as between

microbial cells and cellulose. Francisco et al. displayed Cex CBH on the surface of E.

coli and found that engineered E. coli exhibited specific adhesion capacity toward

cellulose 83. Such cell-to-cellulose adhesion pattern may lead to a distinct mechanism

from the “ablative mechanism” 57 of free cellulases to break down cellulose.

From an industrial point of view, the reuse of yeast cells through various rounds

of fermentation may be important in the ethanol production process 84. Cell-recycle

batch fermentation (CRBF) is a semi-continuous operation strategy using high

densities of recycled cells to produce ethanol continuously. In previous reports, the

CRBF of lignocelluloses was performed by the addition of large amount of

commercial cellulases, a step that represents one of the main bottlenecks for

commercialization85, 86. The application of a cellulolytic yeast strains to CRBF is

41

expected to decrease the need for the addition of costly commercial enzymes 39. Thus,

in this study, the direct ethanol production from PASC using recycled cellulolytic

cells was conducted without the addition of commercial cellulases. After a 3-batch

recycle fermentation, our strain EG-D-CBH1-D retained 60% of the PASC

degradation ability that the strain had in the 1st batch. This retention of activity was

1.7-fold higher than that seen in strain EG-S-CBH1-S. The apparently lower

degradation ability in strain EG-S-CBH1-S likely can be attributed to the loss of

cellulases in each cycle, as most secreted enzymes were separated from cells during

the cell collection step at the beginning of each batch. Although the amount of

enzyme produced was sufficient for saccharification in the first batch, the ability to

generate cellulases in each new cycle apparently declined in subsequent batches. In

contrast, the enzymes immobilized on strain EG-D-CBH1-D showed more consistent

activity than the activities in strain EG-S-CBH1-S during the recycling. Khaw et al.

reported that the ethanol production rate of a yeast strain producing displayed

α-amylase was maintained during a number of repetitions87. Consequently, the

construction of cellulolytic yeast by cell-surface display strategy is more applicable to

CRBF compared with the secretion strategy. Matano et al. even reported that with the

addition of extraneous cellulases (10 FPU/g-biomass), the fermentation ability of a

cellulase-displaying yeast strain remained constant after 5-cycle repeated-batch

fermentation 46. However, the required amount of additional cellulase was still too

high, precluding economic feasibility for an industrial process. In future work, we

propose to further improve the cellulolytic activity of yeast strains and apply the

cell-surface display strategy to ethanol production from lignocellulosic biomass (e.g.,

rice straw) with addition of trace amounts of exogenous cellulases.

Conclusion

In this study, we investigated suitable strategies for heterologous production of

42

EG and CBH1 production in the engineering of cellulolytic S. cerevisiae strains. We

demonstrated that cell-surface display system enhanced the production rate of ethanol.

Placement of EG and CBH1 in the same space (both on the cell surface or both

secreted into the medium) was favorable for ethanol production from amorphous

cellulose. In addition, a cellulolytic yeast strain producing cellulases via cell-surface

display was more effective in the cell-recycle batch fermentation than a strain

producing secreted cellulases.

43

Chapter II

Engineering of a novel cellulose-adherent cellulolytic Saccharomyces

cerevisiae for cellulosic biofuel production

44

Introduction

As the demand for fossil fuels increases and atmospheric CO2 levels continue to

rise, biofuels have attracted increasing attention as sustainable and renewable

alternative energy sources88. Lignocellulose is a potential resource for biofuel

generation because of its low cost and large-scale availability89. However, the need

for high-dosages of costly commercial cellulases in the saccharification process

makes it challenging for cellulose-based biofuels to be economically feasible6.

Although two leading enzyme companies (Genencor and Novozymes) have

significantly reduced cellulase prices (to 15-20 cents per gallon of ethanol produced),

these prices are still 5- to 10-fold higher than those of the amylases used for

starch-based biofuel production90. Thus, development of a microorganism that is

capable of both producing cellulases and fermenting resultant sugars into biofuels is a

promising approach to alleviate the economic burden imposed by the need for

commercial enzymes7.

Saccharomyces cerevisiae has been reported to have a superior capacity for

converting glucose into ethanol91. However, yeast lacks the cellulolytic enzymes

needed to degrade cellulose into glucose, such as β-glucosidase (BGL),

endoglucanase (EG), and cellobiohydrolase (CBH). Several engineered S. cerevisiae

strains capable of producing heterologous cellulases have been reported9, 10. However,

these engineered strains are unable to degrade crystalline cellulose effectively

(achieving only 8-23% hydrolysis efficiency)92, 93. This is due to insufficient

production (or secretion) of CBH63, as CBHs are responsible for the degradation of

crystalline cellulose and have been considered key forces in disrupting the recalcitrant

structure of cellulose59.

Up to now, cellulolytic S. cerevisiae strains have been constructed primarily via

45

either secretion or cell-surface display56. Secretion system, which releases enzymes

into extracellular environment, is the most common route of cellulase production in

recombinant strains. Although free enzymes can easily penetrate into the secondary

cell wall of plant cells57, they are incapable of being recycled usage during an

industrial process56. In comparison, a cell-surface display system permits

immobilization of enzymes on the cell surface through glycosylphosphatidylinositol

(GPI) anchoring of proteins14. Immobilization of numerous cellulases on a given

microbial cell provides an effective increase of local enzyme concentration; facilitates

synergistic interactions among enzymes; and, most importantly, enables re-utilization

of enzymes in repeated fermentation, improving the economic feasibility of the

process37. The cellulose degradation mechanisms of free-form cellulases and

cellulosomes (complexed cellulase system) have been studied intensively in the last

few decades57, 94, however, few investigations have focused on the mechanisms

employed by cell-surface-displayed enzymes (non-complexed cellulase system).

Cellulosic ethanol production using cellulase-displaying cells is gaining increased

attention37, but information on how these cells digest cellulose to fermentable sugars

is still unclear.

To achieve commercial scale, high bioethanol productivity will need to be

obtained from lignocellulosic waste (e.g., rice straw). However, most studies select

only pure/model cellulose (e.g., phosphoric acid swollen cellulose (PASC) or Avicel)

as an experimental substrate37, a model that is significantly different to real word

ethanol fermentation from lignocellulosic feedstock. Here, we report the successful

engineering of a cellulose-adherent S. cerevisiae strain producing heterologous BGL,

EG, CBH1, and CBH2 via cell-surface display, permitting direct ethanol production

from a realistic lignocellulosic substrate (rice straw). The interactions between

engineered yeast cells and cellulose were imaged by scanning electron microscopy

46

(SEM) to demonstrate clearly the dynamic mechanisms of cell-to-cellulose adhesion.

This work provided valuable information on how to increase cellulose hydrolysis

efficiency by enhancing cell-to-cellulose interactions for various types of substrate,

and therefore may lead to a feasible path towards the cost-competitive production of

cellulosic ethanol.

Materials and methods

Media and materials

Escherichia coli NovaBlue was grown in Luria-Bertani broth at 37 °C, and 100

mg/L ampicillin was added to the medium when required. Yeast strains were screened

on synthetic dextrose (SD) agar plates (6.7 g/L of yeast nitrogen base without amino

acids and 20 g/L of glucose) supplemented with appropriate amino acids. Yeast

strains were pre-cultured at 30 °C for 72 h in yeast extract-peptone (YP) medium (10

g/L yeast extract and 20 g/L peptone) containing 20 g/L glucose (YPD). Cellulosic

ethanol production was carried out at 37 °C in YP media containing different

cellulosic materials depending on the purpose of different fermentation experiments.

PASC was prepared from Avicel PH-101 (Fluka Chemie GmbH, Buchs, Switzerland)

as described previously9. Rice straw was pretreated with a liquid hot water method

(130-300 °C under a pressure of less than 10 MPa) and then subjected to 4 cycles of

milling using a CMJ01 nano-mech reactor (Techno Eye, Tokyo, Japan)95; the

resultant biomass was designated MC6. The composition of MC6 was 43% (w/w)

glucan, 2% (w/w) xylan, 42.3% (w/w) ash and lignin, and 12.7% (w/w) other

materials39.

47

48

Plasmid and strain constructions

The plasmids and primers used in this study are summarized in Table II-1 and

Table II-2, respectively. The plasmid pRDH227 was constructed by removing the C.

lucknowense open reading frame as a 1468 bp PacI/AscI fragment from the plasmid

pMU78463 and cloning it into the corresponding sites of pBHD197. Other plasmids

were constructed by connecting DNA fragments using the isothermal assembly

method66. For plasmid pDI9-CBH1D, four PCR products were connected. The four

PCR products included the following: I9 region fragments (the 3’ non-coding region

between gene YOR191W and YOR192C, amplified from the genome of S. cerevisiae

BY4741 using primers pairs I9a-M-F + I9a-O-R and I9b-O-F + I9b-C1-R; ori-ampR

fragment (from plasmid pIU-CBH1D) using primers O-I9a-F + O-I9b-R; MET15

fragment (from plasmid pRS401) using primers M-I9a-F + M-C1-R; and T. emersonii

CBH1-encoding surface-display cassette (from plasmid pI5-CBH1D) using primers

C1-M-F + C1-I9b-R. To construct plasmid pDI9-CBH2D, a PCR-amplified C.

lucknowense CBH2-encoding gene (from plasmid pRDH227, using primers C2-F +

C2-R) and the vector backbone containing the enzyme-display cassette (from plasmid

pDI9-CBH1D, using primers D-C2-F + P-C2-R) were linked together. Similarly, the

CBH2 gene (primers C2-F and C2-R2) was ligated with the vector backbone

containing the enzyme-secretion-cassette (from plasmid pDI9-CBH1D, using primers

D-C2-F2 + P-C2-R), to yield plasmid pDI9-CBH2S. In addition, the CBH2 gene,

enzyme-display cassette, and I5 region (the 3’ non-coding region of gene YLL055W

and YLL054C, amplified from plasmid pI5-CBH1-D using primers D-C2-F + P-C2-R)

were connected to yield plasmid pIU5-CBH2D.

Plasmids were transformed into S. cerevisiae BY4741 using lithium acetate as

described68, and integrated into either the I5 or I9 region by homologous

recombination. The transformants were identified using colony PCR to check for the

49

integration of cellulase genes (using screening with primers I9-F + I9-R for I9 region

integration or primers I5-F + I5-R for I5 region integration).

Table II-2 PCR primers used in this study

Primers Sequence (5’-3’)

I9a-M-F ATTAATGAATCGGCCAACGCTGGATATGACTGTGTTGTTGCTGATA

I9a-O-R GGGGGCGGAGCCTATGGAAAAACGCCAGCAACGCGG

O-I9a-F AAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCC

O-I9b-R GCACTTTTCGGGGAAATGTGCGCGGAACCCCTATTTGTTTATTTTTC

I9b-O-F AAACAAATAGGGGTTCCGCGCACATTTCCCCGAAAAGTGCCACC

I9b-C1-R TTTTCACCGTCATCACCGAAGGGCCCATGGCTAGGTGT

M-I9a-F ACACACCTAGCCATGGGCCCTTCGGTGATGACGGTGAAAA

M-C1-R TTTCACACCGCATAGATCCGACTTGTGAGAGAAAGTAGGTTTAT

C1-M-F ACCTACTTTCTCTCACAAGTCGGATCTATGCGGTGTGAAATAC

C1-I9b-R CAACAACACAGTCATATCCAGCGTTGGCCGATTCATTA

C2-F AATACGTTCGCTCTATTAAGATGGCCAAGAAGTTGTTCATTACC

C2-R GTTGATAATTTACTCGAGCCGAATGGTGGATTTGCGTTCGTTAAC

C2-R2 CTCAATGTACTAACTGTACATTAGAATGGTGGATTTGCGTTCG

D-C2-F CGAACGCAAATCCACCATTCGGCTCGAGTAAATTATCAACTGTCC

P-C2-R ATGAACAACTTCTTGGCCATCTTAATAGAGCGAACGTATTTT

D-C2-F2 ACGCAAATCCACCATTCTAATGTACAGTTAGTACATTGAGTCTAAATA

I9-F AAGAAGAAATCCGTGCTTACACATT

I9-R GCTATCCCATGCAAAGATTGTCAACG

I5-F CATTGAAGAAGGGAAAGTGGTAACC

I5-R TCCCTCTCTAATCTGGGTGAGAC

rt-CBH1-F CAACTTACTGTCCAGACGACGAAAC

rt-CBH1-R AAGGAAGAACCAGAGGAGGTAACAC

rt-CBH2-F AGAAGTCCCTAGTTTCCAATGGCTT

rt-CBH2-R CGGCCTTATTCAAAGCTCTAACCTG

rt-ACT1-F TGGATTCCGGTGATGGTGTT

rt-ACT1-R TCAAAATGGCGTGAGGTAGAGA

50

Cell growth assay

To measure the growth of yeast cells, parent strain S. cerevisiae BY4741 and

recombinant strains were cultivated individually in SD medium at 30 °C with shaking

at 150 rpm for 24 h. The resulting pre-cultures were inoculated into 5 mL YPD

medium at an initial optical density (OD660) of 0.05 and cultivated at 30 °C with

shaking at 70 rpm. The value of the OD660 was measured once hourly using a

TVS062CA Bio-photorecorder (Advantec Toyo, Tokyo, Japan). The value of the

OD660 was taken as an indicator of cell growth.

Quantitative real-time PCR

The transcription levels of the cellulase-encoding genes were quantified as

described previously 69. Primers rt-CBH1-F and rt-CBH1-R were used to determine

the transcription level of the CBH1 gene; primers rt-CBH2-F and rt-CBH2-R were

used to determine the transcription level of the CBH2 gene. Transcription levels of the

target genes were normalized to those of the housekeeping gene ACT1 (tested using

primers rt-ACT1-R and rt-ACT1-F).

Scanning electronic microscopy and optical microscopy

Yeast cells were resuspended in phosphate buffer (pH 5.0) at a concentration of

30 g wet cells/L. Subsequently, 1% (w/v) PASC or Avicel was added to the cell

suspension, which then was incubated at 37 °C for 2 h. Cellulose fibers were fixed

with 4% paraformaldehyde and 4% glutaraldehyde (GA) in 0.1 M cacodylate buffer

(pH 7.4) at 4 °C. Thereafter, fibers were fixed with 2% GA in cacodylate buffer

overnight. The samples were additionally fixed with 1% tannic acid at 4 °C for 2 h.

After the fixation the fibers were washed with cacodylate buffer 4 times, followed by

post fixation with 2% osmium tetroxide in cacodylate buffer for 4 h. The samples next

were dehydrated in a graded series of ethanol solutions (50, 70, 90, and 100%), and

51

then were substituted into tert-butyl alcohol and dried by vacuum freeze drying. After

drying, the samples were coated using an osmium plasma coater (NL-OPC80NS,

Nippon Laser & Electronics Laboratory, Nagoya, Japan). The samples were

visualized using a scanning electron microscope (JSM-6340F; JEOL Ltd., Tokyo,

Japan) at an acceleration voltage of 5 kV. For observation using optical microscopy,

post-incubation samples were directly applied onto microscope slides and imaged.

Ethanol production from cellulosic materials

Fermentations were performed at 37 °C under oxygen limited conditions at an

agitation speed of 200 rpm in 100-mL closed bottles equipped with a bubbling

CO2 outlet and a stir bar. Pre-cultivated cells in YPD medium were centrifuged and

washed twice by sterilized water. The collected cells were inoculated into 20 mL YP

medium containing 20 g/L PASC, 10 g/L Avicel, or 100 g/L MC6. Initial cell

densities were adjusted to approximately 150 g wet cells/L. Yeast cell wet weight was

determined by weighing a cell pellet that was harvested by centrifugation at 1,000 × g

for 5 min. To evaluate cellulase dosages in the SSF process, commercial enzyme

(Novozymes Cellic CTec2; Novozymes Inc., Bagsvaerd, Denmark) was added into

medium at 0-h of fermentation at enzyme concentrations of 0, 0.2, 0.6, 1.0, 1.4, 1.8,

2.2 FPU/g-biomass. Measurement of the filter paper cellulase units (FPU) of CTec2

was based on the standard NREL analytical procedure 98 performed at 37 °C. The

ethanol concentrations in the fermentation medium were determined using a gas

chromatograph, as described previously 48.

Cellulolytic activity assay

At the 0-h and 96-h time points of ethanol production from 100 g/L MC6,

fermentation medium was assayed for cellulolytic activities. Cellulolytic activity

represents the degradation ability of all enzymes present in the fermentation broth.

52

Fermentation broth was added into 50 mM sodium citrate buffer (pH 5.0) containing a

final concentration of 1% (w/v) MC6 and 100 mM methyl glyoxal (Nacalai Tesque,

Inc., Kyoto, Japan); the methyl glyoxal prevented the assimilation of glucose by yeast

cells 70. The reaction was performed at 37 °C using a heat block (Thermo Block

Rotator SN- 06BN; Nissin, Tokyo, Japan) with shaking at 35 rpm, and the supernatant

was collected by centrifugation. The amount of glucose in the supernatant was

determined by the Glucose CII kit (Wako Pure Chemical, Osaka, Japan). One unit of

cellulolytic activity (expressed as U/L) was defined as the amount of enzyme needed

to produce 1 µmoL of glucose per minute at 37 °C, pH 5.0.

Results

Heterologous expression of cellulases in S. cerevisiae

The relevant features of the recombinant yeast strains used in this study are listed

in Table II-1. Four codon-optimized genes that encode Aspergillus aculeatus BGL1,

Trichoderma reesei EG2, Talaromyces emersonii CBH1, and Chrysosporium

lucknowense CBH2 were expressed and used for assembly of enzyme cocktails in S.

cerevisiae BY4741. Cellulase enzyme encoding genes were all expressed under the

control of the SED1 promoter, which is highly induced during the stationary-phase

growth 75. The secretion signal peptide of BGL1 was derived from Rhizopus oryzae

glucoamylase, while EG2, CBH1, and CBH2 were produced with their native

secretion signals. The schemes of engineered cellulolytic yeast strains producing

enzymes via cell-surface display or secretion are illustrated in Figure II-1. In the strain

expressing cell-surface-displayed proteins, the enzymes were encoded fused to the

N-terminus of Sed1, a S. cerevisiae cell wall protein rich in threonine/serine residues

that contains a putative GPI attachment signal. Sed1p becomes the most abundant cell

wall protein during the stationary phase of growth75. The GPI-anchor is attached to

the Sed1-enzyme chimeric proteins in the endoplasmic reticulum and subsequently

53

Cel

l-sur

face

dis

play

syst

em�

CB

H1

SS

ASE

D1

T SA

G1

P SE

D1

CB

H2

SS

ASE

D1

T SA

G1

P SE

D1

EG

SS

A

SED

1 T S

AG

1 P S

ED

1

BG

L SS

A

SED

1 T S

AG

1 P S

ED

1

Pyru

vate�

Ace

tald

ehyd

e�

CO

2�

ATP�

AD

P�

Eth

anol�

NA

D+�

NA

DH�

Glu

cose�

CB

H1�

EG�

CB

H2�

BG

L�

Glu

cose�

Cel

lulo

se�

Sacc

haro

myc

es c

erev

isia

e ce

ll�

SED

1 an

chor�

Secr

etio

n sy

stem�

Pyru

vate�

Ace

tald

ehyd

e�

CO

2�

ATP�

AD

P�

Eth

anol�

NA

D+�

NA

DH�

Glu

cose�

CB

H1

SS

T SA

G1

P SE

D1

CB

H2

SS

T SA

G1

P SE

D1

EG

SS

T S

AG

1 P S

ED

1

BG

L SS

P S

ED

1 A

SED

1 T S

AG

1

Figu

re I

I-1

Engi

neer

ing

cellu

loly

tic S

. cer

evis

iae

for

cellu

losi

c et

hano

l pro

duct

ion

via

eith

er c

ell-s

urfa

ce d

ispl

ay o

r

secr

etio

n of

enz

ymes

. P S

ED1,

SED

1 pr

omot

er;

SS,

secr

etio

n si

gnal

; A S

ED1,

SED

1 an

chor

ing

regi

on;

T SAG

1, SA

G1

54

transferred to the cell surface through an S. cerevisiae secretory pathway. Upon

reaching outer layer of the cell wall, chimeric proteins are covalently bound to

β-1,6-glucan via the GPI anchor99. In contrast, a chimeric protein lacking the Sed1

domain passes through cell wall and is released into the extracellular medium

(secretion system). Metabolic burden often occurs due to the expression of

heterologous protein genes, resulting in inhibition of cell growth100. However, in the

present study, the growth rates of the recombinant strains were similar to those of the

parent strain (Figure II-2), suggesting that production of the heterologous enzymes did

not impose an obvious metabolic burden.

Figure II-2 Time-course profiles of cell growth using parent strain BY4741 and recombinant

yeast strains in YPD medium. Each strain was inoculated in YPD medium to an initial OD660

of 0.05 and then cultured aerobically at 30 °C, 70 rpm, for 72 h. For each strain, data are

presented as the mean ± SD from three independent experiments.

Assembly of high-efficiency enzyme cocktail

To develop a high-efficiency cellulase system, various cellulase combinations of

some or all of our candidate proteins were engineered, including BGL, EG, CBH1,

0.01

0.1

1

10

100

0 24 48 72

Cel

l gro

wth

(OD

660

)�

Time (h)�

BY4741

EG-D-CBH1-D-CBH2-D

EG-S-CBH1-S-CBH2-S

EG-D-CBH1-D-CBH1-D

EG-D-CBH2-D

EG-D-CBH1-D

55

and CBH2. All enzyme gene combinations were expressed in S. cerevisiae and

produced through cell-surface display technique. Cellulosic ethanol production using

different recombinant strains was investigated. Pure celluloses consisting of PASC

(amorphous type) or Avicel (crystalline type) were used as substrates. Two

recombinant strains designated EG-D-CBH1-D and EG-D-CBH2-D (i.e., containing

combinations of BGL+EG+CBH1 and BGL+EG+CBH2, respectively) were designed

as “minimum enzyme cocktails”. As shown in Figure II-3, a maximal ethanol

production of 2.3 g/L (equivalent to 23% of the theoretical value) was achieved in

both engineered strains using PASC as the substrate (Figure II-3a). In contrast, strain

EG-D-CBH1-D showed a higher ethanol production rate than strain EG-D-CBH2-D

when fermenting Avicel (Figure II-3b), suggesting that the T. emersonii CBH1 is

more effective than C. lucknowense CBH2 in the degradation of crystalline cellulose.

Figure II-3 Time course of direct ethanol production from cellulosic materials. Ethanol was

produced from 20 g/L PASC (a) and from 10 g/L Avicel (b). Data are presented as the means

and standard deviations of triplicate measurements.

Subsequently, an additional copy of the CBH1-encoding gene was added to

BGL+EG+CBH1 cocktail, yielding strain EG-D-CBH1-D-CBH1-D. The transcription

level of CBH1 in strain EG-D-CBH1-D-CBH1-D was approximately 1.8-fold that of

0

1

2

3

4

5

6

7

8

0 24 48 72 96

Eth

anol

(g/L

)�

Time (h)�

a�

EG-D-CBH1-D EG-D-CBH2-D EG-D-CBH1-D-CBH1-D EG-D-CBH1-D-CBH2-D

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 24 48 72 96

Eth

anol

(g/L

)�

Time (h)�

b�

56

strain EG-D-CBH1-D (Figure II-4). In PASC and Avicel fermentations, strain

EG-D-CBH1-D-CBH1-D produced ethanol at 3.1 g/L (30% of the theoretical value)

and 0.56 g/L (11% of the theoretical value), respectively, corresponding to 1.3- and

1.9-fold the levels seen in strain EG-D-CBH1-D (Figure II-3). To further increase the

cellulose degradation efficiency, a strain (designated EG-D-CBH1-D-CBH2-D) that

contained genes encoding BGL, EG, CBH1, and CBH2 was designed; this strain

generated an ethanol titer of 6.7 g/L from PASC (66% of the theoretical value) and

1.4 g/L from Avicel (27% of the theoretical value), which are 2.9-fold and 4.5-fold

higher than strain EG-D-CBH1-D. These results suggested that introduction of a

second copy of the CBH1-encoding gene only slightly improved the cellulolytic

activities, while the combination of both CBH1- and CBH2-encoding loci yielded a

greater enhancement in the efficiency of cellulose hydrolysis.

Figure II-4 Relative transcription levels of CBHI and CBH2 genes in recombinant yeast

strains. Gene ACT1 was used as the internal standard. The relative transcription levels are

shown normalized to the level observed in strain EG-D-CBH1-D-CBH2-D, whose relative

transcription level was defined as 1. For each strain, data are presented as the mean ± SD

from three independent experiments.

0

0.5

1

1.5

2

2.5

CBH1 CBH2

Rel

ativ

e ex

pres

sion

leve

l (fo

ld)��

EG-D-CBH1-D-CBH2-D

EG-D-CBH1-D

EG-D-CBH2-D

EG-D-CBH1-D-CBH1-D

EG-S-CBH1-S-CBH2-S

57

Performance of enzyme-secreting and -displaying cells in cellulose degradation

We next compared the cellulose fermenting efficacy of S. cerevisiae strains

expressing the EG, CBH1, and CBH2 cellulases via either cell-surface display or

secretion. As depicted in Figure II-5, the cellulase-displaying cells

(EG-D-CHB1-D-CBH2-D) showed better ethanol yield (63% of the theoretical yield)

from PASC than was seen from cellulase-secreting cells (EG-S-CBH1-S-CBH2-S, 54%

of the theoretical yield), which is in good agreement with previous studies 101.

Figure II-5 Time course of direct ethanol production using recombinant strains

EG-D-CBH1-D-CBH2-D and EG-S-CBH1-S-CBH2-S. Ethanol was produced from 20 g/L

PASC (a) and from 10 g/L Avicel (b). Data are presented as the means and standard

deviations of triplicate measurements.

Observation of enzyme-secreting and -displaying cells in cellulose degradation

Based on the SEM observation shown in Figure II-6a, PASC appeared to form a

more sponge-like surface material than Avicel. After incubation of yeast cells with

cellulose, numerous cellulase-displaying cells (EG-D-CBH1-D-CBH2-D) remained

attached to the PASC surface, while very few cellulase-secreting cells

0

1

2

3

4

5

6

7

8

0 24 48 72 96

Eth

anol

(g/L

)�

Time (h)�

a�

EG-D-CBH1-D-CBH2-D EG-S-CBH1-S-CBH2-S

0

0.4

0.8

1.2

1.6

2

2.4

0 24 48 72 96

Eth

anol

(g/L

)�

Time (h)�

b�

58

(EG-S-CBH1-S-CBH2-S) were retained on the cellulose surface (Figure II-6b).

Figure II-6 SEM micrographs of the interactions between cellulolytic S. cerevisiae cells and

cellulosic materials. Interactions with PASC (a, b) and with Avicel (c, d). Cellulolytic cells

(30 g/L) were incubated with 1% cellulosic materials (PASC or Avicel) for 2 h, and the

cellulosic materials were used for SEM imaging. Scale bars are 10 µm.

EG-D-CBH1-D-CBH2-D� EG-S-CBH1-S-CBH2-S�

A PASC�

c Avicel �

b PASC�

d Avicel�

10 µm� 10 µm�

10 µm� 10 µm�

59

Notably, the adhesion between displayed cells and sponge-like cellulose (PASC)

was clearly observed in Figure II-7, showing that the displayed cells are tightly bound

to cellulose filaments. Unlike the enzyme-secreting strain, which digests the cellulose

via free-form enzymes (such that yeast cell need have no direct interaction with

cellulose), the enzyme-displaying strain apparently mediates cellulose degradation via

attachment to the cellulose surface, which presumably significantly shortens the

distance between microbe and substrate. However, interestingly, no obvious

difference in ethanol titer between enzyme-displaying and -secreting cells was

observed in Avicel fermentation (around 32% Avicel conversion, Figure II-5b), which

may reflect lower binding efficiency of displayed cells to Avicel than to PASC

(Figure II-6). We hypothesize that rough, sponge-like structure of PASC provides

greater surface area and fine structures, thereby facilitating adhesion of displaying

cells and promoting degradation of the substrate. To our knowledge, this is the first

report on observation of the adhesion between cellulolytic yeast cells and cellulose.

To better understand the dynamic degradation mechanisms employed by

displaying or secreting cells, the morphological changes of celluloses after 24 h

degradation were monitored (Figure II-8). Cellulose fibers (both PASC and Avicel)

were observed to be disrupted into small, irregular pieces. In contrast, the cellulose

fibers digested by secreting cells appear to be evenly ablated from the outer layer of

microfibers, which is consistent with a previous report 57.

60

1 µm�

1 µm�

Figu

re I

I-7

SEM

mic

rogr

aphs

of

the

inte

ract

ions

bet

wee

n PA

SC a

nd E

G-D

-CB

H1-

D-C

BH

2-D

cel

ls.

The

obse

rvat

ion

cond

ition

s wer

e th

e sa

me

as th

ose

used

in F

igur

e II

-6. S

cale

bar

s ar

e 1

µm.

61

Figure II-8 Micrographs of PASC (a, b) and Avicel (c, d) degradation process by cellulolytic

S. cerevisiae. Cellulolytic cells (30 g/L) were incubated with 1% cellulosic materials for 24 h

at 37 °C. Cellulosic materials were sampled and directly applied to microscope slides for

observation. Scale bars are 10 µm. The arrows indicate the different cellulose degradation

tendencies of cellulolytic yeasts: strain EG-D-CBH1-D-CBH2-D tended to conduct

degradation via tearing cellulose fibers into smaller pieces, while EG-S-CBH1-S-CBH2-S

equably eroded cellulose from the outer surface.

a PASC�

cAvicel�

b PASC�

d Avicel�

EG-D-CBH1-D-CBH2-D� EG-S-CBH1-S-CBH2-S�

62

Targeted pre-treatment of rice straw

To verify our hypothesis, that higher ethanol yield can be obtained using rough,

sponge-like cellulose as a substrate that provides more surface area and finer structure

for adhesion by enzyme-displaying yeast cells, the surface properties of cellulose

obtained from agricultural waste (specifically, from rice straw) were altered. Liquid

hot water (LHW) pretreatment has been reported to disrupt recalcitrant microstructure

and remove 35–60% of the lignin and all of the hemicellulose from lignocellulose

materials102; separately, milling has been reported to reduce particle size and

crystallinity of lignocellulose, resulting in increased surface area for enzymatic attack 103. In the present study, rice straw was treated with LHW plus 4 cycles of milling,

and the resultant biomass was designated MC6. The morphological characteristics of

rice straw and MC6 were observed by SEM (Figure II-9aI, II). Unprocessed rice straw

exhibited a flat, rigid, and compact surface, while the surface of MC6 was rough and

sponge-like. The average diameters of MC6 particles were about 20- to 50-fold

smaller than those of rice straw. These results suggested that LHW and milling

pretreatment significantly altered the surface characteristics of rice straw into a more

sponge-like structure with increased surface area.

It was also observed that more cellulolytic cells adhere to MC6 fibers than to

unprocessed rice straw (Figure II-9aI’, II’), a result that correlated with the obtained

ethanol concentrations (Figure II-9b). These results suggested that rough, sponge-like

cellulose particles facilitated cell-to-cellulose adhesion, thereby improving cellulose

degradation.

63

Figure II-9 Comparison of cellulosic feedstock rice straw with MC6 for cellulosic ethanol

production. (a) SEM micrographs of the surface structures on cellulosic materials (I, II);

observation of adhesion between strain EG-D-CBH1-D-CBH2 and cellulosic materials using

optical microscopy (I’, II’). (b) Ethanol yields at 96 h of fermentation from 25 g/L rice straw

and MC6, respectively. Data are presented as the means and standard deviations of triplicate

measurements.

I’�

I�

a�Rice straw�

III�

II’�

MC6�

I� II�

0 0.2 0.4 0.6 0.8

1 1.2 1.4 1.6 1.8

Rice straw MC6

Eth

anol

(g/L

)

b

64

Evaluating the feasibility of high-density cellulosic ethanol production

To test the feasibility of high-density cellulosic ethanol production using strain

EG-D-CBH1-D-CBH2-D, fermentation was conducted using 100 g/L MC6 as the

substrate. Only 1.3 g/L ethanol (7% of the theoretical yield) was produced by

enzyme-displaying cells after 96 h fermentation, while no ethanol was obtained by

wild type strain (BY4741). Considering the low ethanol yields in both two strains

(Figure II-9b), the inclusions of small amounts of exogenous commercial enzyme

(C-Tec2) were still required to support hydrolysis of MC6. To address the economic

feasibility of fermentation of MC6 using the enzyme-displaying strain, the potential

diminution of C-Tec2 supplementation by using displaying cells was assessed. As

shown in Figure II-10a, the ethanol yield from EG-D-CBH1-D-CBH2-D fermentation

of MC6 was 7-fold increased by addition of trace amount of exogenous C-Tec2 (0.2

FPU/g-biomass). Specifically, the inclusion of 1.0 FPU/g-biomass C-Tec2 together

with displaying cells provided a high cellulosic ethanol yield of 18 g/L, equivalent to

80% of the theoretical value. The result represents an approximately 44% decrease of

the required enzyme dosage compared to that required for M6 fermentation by the

wild-type strain. Additionally, although approximately half of the C-Tec2 activity was

lost after 96 h of fermentation, the cellulolytic activities from displaying cells

(indicated as the activity difference between EG-D-CBH1-D-CBH2-D and BY4741

fermentations) persisted (Figure II-10b), demonstrating the feasibility of recycling of

cellulase-displaying yeast cells.

65

0 2 4 6 8 10

12

14

16

18

20

22 0

0.2

0.4

0.6

0.8

1 1.

2 1.

4 1.

6 1.

8 2

2.2

2.4

Ethanol (g/L)�

Cel

lula

se a

dditi

on (F

PU/g

bio

mas

s)�

EG-D

-CB

H1-

D-C

BH

2-D

B

Y47

41

0 10

20

30

40

50

60

70

80 0

0.2

0.4

0.6

0.8

1 1.

2 1.

4 1.

6 1.

8 2

2.2

2.4

Cellulolytic activity (mU/mL)�

Cel

lula

se a

dditi

on (F

PU/g

bio

mas

s)��

EG-D

-CB

H1-

D-C

BH

2-D

0 h

B

Y47

41 0

h

EG-D

-CB

H1-

D-C

BH

2-D

96

h B

Y47

41 9

6 h

a�b�

Figu

re I

I-10

Eva

luat

ions

of c

omm

erci

al c

ellu

lase

add

ition

in e

than

ol p

rodu

ctio

n fr

om 1

00 g

/L M

C6.

(a)

Eth

anol

prod

uctio

n at

96

h of

ferm

enta

tion

with

the

addi

tion

of 0

, 0.2

, 0.6

, 1.0

, 1.4

, 1.8

, and

2.2

FPU

/g-b

iom

ass

cellu

lase

(C-T

ec2)

. (b)

Cel

lulo

lytic

act

iviti

es in

the

ferm

enta

tion

med

ia a

t 0 a

nd 9

6 h

of f

erm

enta

tion

with

the

resp

ectiv

e

stra

ins.

Dat

a ar

e pr

esen

ted

as th

e m

eans

and

sta

ndar

d de

viat

ions

of t

riplic

ate

mea

sure

men

ts.

66

Discussion

One of the main bottlenecks impeding the widespread production of cellulosic

bioethanol is the large amount of costly commercial enzymes required for hydrolysis

of cellulose7. As described in this work, a cellulose-adherent S. cerevisiae that

displayed BGL, EG, CBH1, and CBH2 on the cell surface was constructed through a

series of rational designs. It was demonstrated that the cellulase-displayed yeast strain

employed a cell-to-cellulose adhesion and a “tearing” pattern as parts of its

cellulose-degradation mechanism, which differed from those of strains producing

free-form enzyme. Appropriate pretreatment of lignocellulose materials aiming to

enhancing cell-to-cellulose interactions dramatically enhanced cellulosic ethanol yield.

The resultant cellulose-adherent S. cerevisiae may significantly reduce the need for

exogenous enzyme, potentially alleviating the bottleneck in commercial production of

cellulosic bioethanol.

A crucial point in assembly of the cellulase cocktail is to maximize the

synergistic actions among enzymes. Both CBH1 and CBH2 are known to act as

exoglucanases, initiating cleavage of cellulose chains from reducing and non-reducing

ends, respectively8. In this study, T. emersonii CBH1 appeared more important than C.

lucknowense CBH2 in the synergism with BGL and EG (especially in Avicel

fermentation), but we found that providing both CBH1 and CBH2 in enzyme cocktail

was more effective on improving cellulolytic activities than increasing the proportion

of CBH1 (here, by doubling the gene dose of the CBH1 encoding locus). This is

because CBH2 can synergistically enhance the hydrolysis efficiency of CBH1, that it

diminished the bumpy surface on cellulose, apparently preventing CBH1 from getting

stuck during processive movement104. To date, a maximum of three kinds of

cellulases (non-complexed cellulase system) have been simultaneously displayed on

one single cell, because of the difficulty in displaying stable and functional

67

heterologous proteins on the microbial surface. In the present work, we provided the

first report (to our knowledge) of tethering four types of heterologous cellulases to the

surface of a single yeast cell. The resulting strain exhibited significantly higher

ethanol yield from both amorphous cellulose and crystalline cellulose compared to the

cellulolytic yeast strains previously described in relevant studies.

Most reports have assessed enzyme production strategies using amorphous

cellulose as substrate, and the cellulosic ethanol yield obtained with cell-surface

display systems was generally higher than that obtained with secretion systems 42, 105.

However, we proved that the performance of each system is actually

substrate-dependent. The higher ethanol yields in enzyme-displaying systems are

obtained only using amorphous cellulose as substrate, indicating that there is likely

unique interactions between displaying cells and substrate, differing from those that

occur in secreting cell. Cellulase-displaying cells degrade cellulose via tight

attachment onto cellulose filaments, while cellulase-secreting cells did not exhibit

obvious interactions with substrate. One of the reasons for the cell-to-cellulose

adhesion is the high affinity of the carbohydrate binding domain (CBD) in cellulase

towards cellulose106, and anchoring cellulase on cell surface results in a higher affinity

of cells towards cellulose. Increased adhesion of Escherichia coli to cotton fibers has

been demonstrated via anchoring the CBD from CBH onto the cell surface 83. We also

found that the adhesion between displaying cells and cellulose correlates with the

hydrolysis efficiency of cellulose. Electron microscopy (Figure II-7) clearly illustrated

that the swollen, rough microfibers of PASC can encircle the cellulase-displaying

cells, thereby enhancing the adhesion between cells and cellulose, which may explain

the higher ethanol yield obtained from PASC using displaying cells. The

cellulose-adherent characteristic of enzyme-displaying cells is expected to shorten the

distance between cells and substrates, thereby facilitating the mass transfer of

68

hydrolysis products, particularly in high-density fermentation.

Notably, two distinct cellulose-degradation mechanisms are proposed based on

the morphological changes of cellulose caused by both displayed and secreted cells

(Figure II-8). The results indicated that the cell-surface display system tends to

increase the accessible surface area for cell-adhesion by tearing microfibers from

cellulose particles. In contrast, free (secreted) enzymes equably erode cellulose

surface and diminish particle size, consistent with a previous report that free enzymes

employ an ablative, fibril-sharpening mechanism during cellulose degradation107. A

potential explanation for the different mechanisms is the distinct enzyme-to-substrate

interactions represented by the two systems. It has been known that free enzymes

repeatedly associate and dissociate with cellulose to avoid getting stuck during

processive movement81. In contrast, owing to cell-to-cellulose adhesion, the

processive movements of surface-displayed enzymes are slowed down; enzymes are

entrapped on cellulose and digestion proceeds without repeated cycles of association

and dissociation with the substrate. As a result, the displayed cellulases are more

likely to carry out deeper directed digestion of cellulose particles than are free

enzymes, resulting in the splayed morphology observed in Figure II-8.

Natural lignocellulose (e.g., rice straw) is usually rigid and lacks the rough,

intricate surface structures presented in PASC. Thus, to increase the adhesion between

natural lignocellulose and cellulase-displaying yeast, the surface structures of rice

straw were broken into more “sponge-like” materials via pretreatment by LHW and

milling. In our work, cellulase-displaying yeast cells preferred to attach to MC6 and

exhibited dramatically higher hydrolysis rates than those seen with untreated rice

straw, presumably because the MC6 has more surface area with favorable sites for

cell-to-cellulose adhesion. Thus, appropriate pretreatment of lignocellulose, which

69

adopts the cell-to-cellulose adhesion pattern in combination with cellulase-displaying

yeast, is expected to effectively promote the degradation efficiency. Although the

optimal biomass pretreatment combined with free-cellulase-mediated saccharification

has been extensively studied103, the pretreatment process suitable for

surface-displayed cellulases still remains obscure. Pretreatment by steam explosion

has been reported to be capable of producing pores with 3-nm to 1-µm diameters on

the surface of lignocellulose108, facilitating the access of free enzymes (approximately

5.1 nm in diameter) to the interior of cellulose particles109. However, the

enzyme-displaying cells (around 3 µm) would not be able to enter pores of this size;

such cells would instead be expected to digest the substrate primarily by “shaving”

cellulose fibers from the external surface of substrate particles. As a result, such a

“pore-punching” pretreatment process may be not ideal for cell-surface display

hydrolysis systems. On the contrary, the method of ammonia fiber expansion (AFEX)

pretreatment can effectively increase the surface roughness of lignocellulose materials 110, which likely will be of greater use in combination with cell-surface display

systems.

Natural lignocellulose is commonly considered as the most rigid substrate

resistant to digestion, and impossible to degrade by cellulolytic yeast strains in the

absence of exogenous enzymes. Even with addition of exogenous enzymes, natural

lignocelluloses rarely provide high ethanol yields (over 80% of theoretical value)

when used as the substrate for simultaneous saccharification and fermentation (SSF) 111. In the reported SSF for bioethanol production mediated by S. cerevisiae, cellulase

loadings of 15-30 FPU/g-biomass are generally used, depending on the specific

substrate 112, 113. Recently, Matano et al. reported the use of cellulase-displayed yeast

strain to diminish cellulase dosage to 10 FPU/g-biomass in ethanol production from

LHW-pretreated rice straw 46. Surprisingly, the engineered cellulose-adherent S.

70

cerevisiae in the present study is capable of converting pretreated rice straw into

ethanol without commercial enzyme addition. To our knowledge, this is the first

report of direct ethanol production from nature lignocellulose using a cellulolytic

yeast strain, presenting a new potential towards making efficient use of lignocellulose.

Moreover, nearly 80% of ethanol theoretical yield was achieved in the presence of

only 1 FPU/g-biomass C-Tec2, a quantity that represents a 44% decrease of enzyme

loading compared with that required in non-cellulolytic S. cerevisiae and a much

lower enzyme dosage compared with other reports. Meanwhile, the enzymes

immobilized on the cell surface appeared higher stability during the fermentation than

did commercial enzymes. This observation is in agreement with the previous report

that tethering of enzymes to solid supports can increase the protein stability under

non-optimal reaction conditions such as low temperature and organic solvent

composition 114. The retention of activities in displayed enzymes also demonstrates the

feasibility of using cell-surface display in cell-recycling processes. By constructing a

novel cellulose-adherent S. cerevisiae with surface-displayed cellulases, we

successfully demonstrated the concept of enhancing cell-to-cellulose interactions to

effectively alleviate the requirement for high doses of supplemental enzymes in

cellulosic ethanol production. Further improvement of cell-to-cellulose adhesion (e.g.,

increased display of carbohydrate binding domains) combined with the increased

cellulolytic activities on yeast surface is expected to provide a viable model for

implementation of economically feasible cellulosic ethanol production.

Conclusion

In this chapter, we engineered a high-efficient cellulolytic S. cerevisiae with

displaying four different synergistic cellulases (BGL, EG, CBH1, and CBH2) on the

cell surface simultaneously. The engineered yeast cells exhibited clear

cell-to-cellulose adhesion and a “cellulose-tearing” pattern during cellulose

71

degradation. Notably, this strain could directly produce ethanol from rice straw; it

requires only 1.8 FPU/g-biomass commercial cellulases to achieve ethanol yield of 80%

from 100 g/L rice straw, cutting down more than 40% of the required enzymes under

high-dense fermentation. Our study also provides a novel strategy that improves

cellulose hydrolysis via enhancement of cell-to-cellulose interactions.

72

Chapter III

Efficient ethanol production from crystalline cellulose using

high-cellulolytic Saccharomyces cerevisiae with optimized cellulase

ratios on the cell surface

73

Introduction

Production of biofuels to replace the exhausting fossil fuels is gaining increasing

prominence worldwide. Cellulosic biomass presents the most renewable feedstock for

sustainable production of biofuels because of its abundance, low-price, and favorable

environmental properties115. Cellulose is composed of long-chain of glucose

monomers via β-1,4-glycosidic bonds. The extensive hydrogen linkages among chains

lead to a rigid structure of mostly well-ordered crystalline domains and small part of

disordered amorphous domains80. The crystallinities in natural cellulosic materials are

90-100% in plant-based fibers and 60-70% in wood-based fibers116. The crystalline

cellulose is resistant to the access of reactant and enzyme, whereas amorphous

cellulose is readily digestible6. It has been postulated that the efficient hydrolysis of

crystalline cellulose will enable the commercialization of cellulosic biofuels

production.

Efficient degradation of cellulose requires a synergistic reaction of the

cellulolytic enzymes including β-glucosidase (BGL), endoglucanase (EG), and

cellobiohydrolase (CBH). EG randomly degrades the non-crystalline region of the

cellulose; CBH attacks the exposed end of cellulose chain, degrading crystalline

cellulose into cellobioses; BGLdegrades cellooligosaccharide into glucose117. To date,

a variety of cellulolytic biofuel-producing microorganisms have been constructed via

expressing heterologous cellulases in a strategy of either cell-surface display or

secretion9, 50, such as displaying cellulases on the surface of Saccharomyces

cerevisiae. Although these microbes are capable of well degrading amorphous

cellulose (e.g., phosphoric acid swollen cellulose (PASC)), few of them can

efficiently hydrolyze crystalline cellulose (usually 10-30% hydrolysis of cellulose)63,

92. Liu et al. obtained a recombinant S. cerevisiae with BGL, EG, CBH1 and CBH2

activities, which achieved an ethanol yield of 66% from PASC, but only a yield of 27%

74

from Avicel (microcrystalline cellulose)118. This is because most of current researches

are focusing on overexpression of cellulase genes25 or improving the secretory level

of cellulases41, while few of them working on the optimization of enzyme ratios in

recombinant cellulolytic strains. The optimal ratio of synergisitic cellulase is one of

the most critical determinants in the hydrolysis of crystalline cellulose, and it differs

depending on the types of cellulose (e.g., resource, proportion of crystalline and

amorphous domains). Yamada et al. constructed a variety of cellulolytic yeast strains

containing different cellulase ratios (BGL, EG, CBH2). They found that the strains

with high degradation abilities towards amorphous cellulose generally used EG as the

principal proportion in total cellulase mixture44. Nonetheless, due to the preferences

of cellulase to substrate (EG prefers amorphous domain, while CBH targets on

crystalline domain)115, CBH activity should be more important than EG activity

during the degradation of crystalline cellulose. Thus, engineering of a recombinant

strain with the optimal cellulase ratio (perhaps taken CBH as the major proportion) is

assumed to achieve the efficient utilization of cellulosic biomass. Currently, the

optimal cellulase ratio that immobilized on cell surface for crystalline cellulose

degradation remains obscure.

In this chapter, to engineer the cellulase-displaying yeast pools with various

enzyme ratios, we employed cocktail δ-integration method44 to incorporate

multi-copy of cellulase genes (EG, CBH1, and CBH2) into the genome of a

BGL-displaying S. cerevisiae simultaneously. The cocktail-strains exhibiting higher

Avicel degradation abilities than strain EG-D-CBH1-D-CBH2-D (the strain

containing single-copy of each cellulase gene, constructed in Chapter II) were

selected to analyze the enzyme ratios on the cell surface using nano LC-MS/MS. To

our knowledge, this is the first work on relative quantitation of four kinds of cellulase

(BGL, EG, CBH1, and CBH2) anchored on one yeast cell, and the first to investigate

75

the optimal enzyme ratio for crystalline cellulose degradation, which therefore may

lead to a feasible path towards the efficient utilization of natural cellulosic materials.

Materials and methods

Strains and media

All yeast strains used in this study (Table III-1) are derived from strain S.

cerevisiae BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0). The Escherichia coli

strain NovaBlue (Novagen, Inc., Madison, WI, USA) was used for maintenance and

amplification of plasmid DNA. Bacterial cells were grown at 37 °C in Luria-Bertani

broth (10 g/L tryptone, 5 g/L yeast extract, and 5 g/L sodium chloride) containing 100

mg/L ampicillin. Yeast strains were pre-cultivated in synthetic dextrose (SD) medium

(6.7 g/L of yeast nitrogen base without amino acids (Difco Laboratories, Detroit, MI,

USA) and 20 g/L of glucose) supplemented with appropriate amino acids in a shaker

incubator (150 rpm) at 30 °C, and then aerobically cultivated at 30 °C in YP medium

(20 g/L peptone, 10 g/L yeast extract) and 20 g/L glucose. Ethanol fermentation was

performed in YP medium containing either 10 g/L Avicel PH-101 (Fluka Chemie

GmbH, Buchs, Switzerland) or 100 g/L MC6 as carbon sources. MC6 was prepared

from rice straw, as previously described95.

Plasmid construction and cocktail δ-integration

The vectors and primers used in this study are listed in Table III-1 and Table

III-2, respectively. The δ-integrative plasmid pδEG-D was constructed as followed:

the δ DNA fragment was amplified from plasmid pδW by PCR using primers δ-F and

δEG-R. The cell-surface display cassette, which includes SED1 promoter, SED1

anchoring region, and SAG1 terminator was amplified from plasmid pIL2-EGD using

primers SEG-F and S-R. Two DNA fragments were linked together by the isothermal

assembly method66 to yield the plasmid pδEG-D. The constructions of plasmids

76

77

pδCBH1-D and pδCBH2-D were performed by a process similar to the above

description. The δ DNA fragment was amplified from plasmid pδW using primers δ-F

and δC1-R for the construction of pδCBH1-D (primers δ-F and δC2-R for

pδCBH2-D), and fused with the PCR product amplified from plasmid pIU5-CBH1D

via primers SC1-F and S-R to construct pδCBH1-D (amplified from pDI9-CBH2D via

primers SC2-F and S-R to construct pδCBH2-D).

Table III-2 PCR primers used in this study

Primers Sequence (5’-3’)

δ-F TCACTGGATCTGCATTGAGAAATGGGTGAATGTTGAG

δEG-R CCACAGTTGATGTGGTGTTGGAATAGAAATCAACTATCATC

δC1-R GCTGTGGCAGGTTGTTGTTGGAATAGAAATCAACTATCATC

δC2-R TCTCTTTTATAGGAATGTTGGAATAGAAATCAACTATCATC

SEG-F ATTTCTATTCCAACACCACATCAACTGTGGGAATAC

SC1-F ATTTCTATTCCAACAACAACCTGCCACAGCTTTTCAA

SC2-F ATTTCTATTCCAACATTCCTATAAAAGAGAAGCGTATAAAAC

S-R TTCACCCATTTCTCAATGCAGATCCAGTGAGCG

Plasmids were transformed into S. cerevisiae BY4741 using lithium acetate as

described68. In cocktail δ-integration method, identical amounts of three δ-integrative

plasmids (over 20 µg of each plasmid), pδEG-D, pδCBH1-D, and pδCBH2-D, were

mixed and transformed simultaneously. The transformants with the highest

cellulolytic activities were selected based on their degradation abilities towards

crystalline cellulose. Cell growth of parent strain S. cerevisiae BY4741 and

recombinant strains were measured as described by Liu et al96.

Avicelase activity assay

The Avicelase activity of yeast cells was determined to indicate their degradation

ability towards microcrystalline cellulose. Recombinant colonies were picked up and

78

cultivated at 30 °C for 24 h in SD medium. The culture was then inoculated into YPD

medium, and grew at 30 °C, 300 rpm for 72 h. Yeast cells were subsequently

collected via centrifuge and washed twice by sterilized water. The Avicelase activities

of transformants were performed in a 48-well microplate. Each well involves 1%

Avicel solution, 100 mM methylglyoxal (Nacalai Tesque, Inc., Kyoto, Japan) and 100

g wet cells/L. The enzymatic reaction was carried out at 50 ºC, 300 rpm for 24 h. The

Avicelase activity of strain EG-D-CBH1-D-CBH2-D was used as control.

Methylglyoxal has been reported to inhibit the transportation of glucose into cells119.

The yeast cell surface glucose sensors Rgt2 and Snf3 act as glucose receptors that

detect extracellular glucose and generated an inductive signal for gene expression of

glucose transporters. Methylglyoxal can inactivate these glucose sensors, blocking the

production of glucose transporter119. As a result, the glucose generated from Avicel

degradation is remained in supernatant, rather than being assimilated by yeast cells.

After the reaction, the supernatant was obtained by centrifugation at 8,000 × g, 10 min,

4 °C. Glucose concentration in the supernatant was measured by the Glucose CII kit

(Wako Pure Chemical Industries, Ltd., Osaka, Japan). The principle of Glucose CII

kit is followed: the α-D-glucose existing in the sample is converted rapidly to

β-D-glucose under the action of mutarotase. β-D-glucose is oxidized by glucose

oxidase (GOD) to produce hydrogen peroxide, which is subsequently converted into

red pigment using peroxidase (POD). The absorbance of pigment is measured at 505

nm. One unit of Avicelase activity (U/mL) was defined as the amount of enzyme

needed to produce 1 µmoL of glucose per 24 h at 50 °C, pH 5.0.

Ethanol production from cellulosic materials

Fermentations were performed at 37 °C at an agitation speed of 200 rpm in 30

mL glass serum bottles, which are sealed using rubber stoppers with a needle for the

removal of CO2 produced during fermentation. Pre-cultivated cells in YPD medium

79

were centrifuged and washed twice by sterilized water. The collected cells were

inoculated into 10 mL YP medium containing 10 g/L Avicel, or 100 g/L MC6.

Pre-cultures of yeast strains were inoculated with an average cell concentration of

150 g wet cells/L. The ethanol concentrations in the fermentation medium were

determined using a gas chromatograph, as described by Yamada et al48.

Results

Screening of transformants with increased cellulose degradation abilities

Three cellulase genes, EG, CBH1, and CBH2 were introduced into a

BGL-displaying yeast (strain BY-BG-SS, described in Chapter I) simultaneously

using different amino acid markers (Figure III-1). As a result, a pool of recombinant

yeasts with various enzyme ratios was constructed. Then the strains with optimized

enzyme ratios were screened from over 500 numbers of transformants based on their

degradation abilities toward Avicel (namely, Avicelase activity). Twenty-nine

transformants exhibiting higher Avicelase activities than that of control strain

EG-D-CBH1-D-CBH2-D were illustrated in Figure III-2. The Avicelase activities in

δ-integrated strains were mostly over 20% higher than that in single-integrated strain

(Control). Strain B6 showed the highest Avicelase activity (97 U/mL), corresponding

to 1.5-fold the level seen in control strain. Alternatively, the Avicelase activity of

strain A26 (89 U/mL) was appropriately 43% higher than that of control strain. These

results demonstrate that it is efficient to reach high cellulolytic activities in yeast cells

through expression of multiple-copy cellulase gene. The enzyme ratios on the cell

surface are still under analysis.

80

….�

Avi

cel�

Mea

sure

cel

l gro

wth

and

A

vice

l fer

men

tatio

n�

Coc

ktai

l of E

G, C

BH

1,

CB

H2

expr

essi

on p

lasm

ids�

Qua

ntita

tion

of su

rfac

e-di

spla

yed

cellu

lase�

Pick

up

posi

tive

colo

nies�

Scre

enin

g�

Nan

o L

C-M

S/M

S�

Pool

of c

ellu

loly

tic

yeas

t str

ains�

BG

L-d

ispl

ayin

g y

east

stra

in�

Met

hylg

lyox

al�

Glu

cose�

BG

L�

EG�

CB

H1�

CB

H2�

Yeas

t cel

l�

Figu

re I

II-1

Sch

emat

ic d

escr

iptio

n of

the

con

stru

ctio

n an

d sc

reen

ing

of h

igh-

cellu

loly

tic y

east

stra

ins

in t

his

stud

y. B

GL,

β-gl

ucos

idas

e ; E

G, E

ndog

luca

nase

; CB

H1,

Cel

lobi

ohyd

rola

se 1

; CB

H2,

Cel

lobi

ohyd

rola

se 2

.

81

50

60

70

80

90

100

110

Contro

l A1 A15

A16

A18

A23

A24

A26

A27

B6 B12

B15

B19

B20

B22

B23

B24

B26

B30

B37

C1 C4 C5 C6 C9 C11

C18

C20

C21

C25

Avicelase activity �U/mL)�

Coc

ktai

l δ-in

tegr

ated

str

ains

Figu

re I

II-2

The

Avi

cela

se a

ctiv

ities

of c

ockt

ail δ

-inte

grat

ion

trans

form

ants

. The

Avi

cela

se a

ctiv

ity o

f stra

in E

G-D

-CB

H1-

D-C

BH

2-D

(sin

gle-

copy

int

egra

ted

stra

in)

was

tak

en a

s co

ntro

l. Th

e A

vice

lase

act

ivity

of

yeas

t ce

lls i

ndic

ates

the

ir de

grad

atio

n ab

ility

tow

ards

mic

rocr

ysta

lline

cel

lulo

se. O

ne u

nit o

f Avi

cela

se a

ctiv

ity w

as d

efin

ed a

s the

am

ount

of e

nzym

e ne

eded

to p

rodu

ce 1

µm

ol o

f glu

cose

per

24 h

at 5

0 °C

, pH

5.0

(U/m

L). V

alue

s ar

e pr

esen

ted

as th

e m

ean

± SD

from

thre

e in

depe

nden

t exp

erim

ents

.

82

Cell growth and cellulosic ethanol production

Generally, overloaded expression of heterologous proteins in yeast cell leads to

decreases in cell growth and ethanol fermentation capacity, referred as the metabolic

burden. Thus, the growth deficiency associated with metabolic burden should be

considered in the development of cellulolytic yeast strains. A cell growth comparison

of cocktail δ-integrated strains and control strain is presented in Table III-3. Aerobic

conditions were used to fully understand the underlying metabolic effects associated

with recombinant protein production. Although most of the δ-integrated strains

showed limited metabolic burden on cell growth compared with reference strain, six

transformants exhibited significant decrease in cell growth. The maximum specific

growth rate (µmax) of strains B12, B26 (0.35 h-1) are significantly lower than that of

control (0.44 h-1); the periods of lag phase in strains B24 (3 h), B26 (4 h), and B37 (4

h) are longer than in the other strains (2 h). Notably, although strain A24 showed

similar values of µmax and lag phase to those of control strain, it still yielded a

relatively low biomass concentration after 12 h cultivation, because its increase speed

of specific growth rate (µ) was slower than the control in exponential phase.

Subsequently, to obtain a high-cellulolytic yeast strain without growth deficiency

associated with metabolic burdens, the cellulosic ethanol yields of cocktail

δ-integrated strains were investigated. As shown in Figure III-3, most of

transformants gained higher ethanol yields from Avicel than the single-integrated

control strain (39%). Strain A26 achieved the highest cellulosic ethanol yield (57%,

corresponding to 1.46-fold of control strain) without any metabolic burden on cell

growth. On the contrary, some growth-deficient strains appeared much lower ethanol

yield than the control, indicating that metabolic burdens probably influence the

ethanol fermentation capacities in yeast cells. Interestingly, growth-deficient strain

A24 (with an ethanol yield of 55%) showed 1.41-fold of reference strain, suggesting

83

Table III-3 Growth of cocktail δ-integrated strains. Cell growth was measured under aerobic

condition in YPD medium. Mean values from triplicate experiments with standard deviations

are shown.

Strains Lag phase

(h)

Specific growth rate

(µmax/ h)

12 h biomass concentration

(OD660)

Control 2 0.44±0.02 1.83±0.07

A1 2 0.40±0.01 1.62 ±0.19

A15 2 0.41±0.02 1.65±0.05

A16 2 0.38±0.01 1.31±0.17

A18 2 0.44±0.02 1.82±0.18

A23 2 0.42±0.02 1.60±0.03

A24 2 0.40±0.02 0.92±0.05

A26 2 0.44±0.01 1.73±0.10

A27 2 0.41±0.02 1.65±0.09

B6 2 0.47±0.02 2.05±0.13

B12 2 0.35±0.02 1.11±0.09

B15 2 0.49±0.04 2.08±0.25

B19 2 0.44±0.02 2.00±0.33

B20 2 0.43±0.01 1.83±0.07

B22 2 0.41±0.01 1.61±0.12

B23 2 0.44±0.02 1.89±0.22

B24 3 0.45±0.02 1.00±0.16

B26 4 0.35±0.01 0.40±0.38

B30 2 0.43±0.02 1.91±0.04

B37 4 0.40±0.02 0.93±0.12

C1 2 0.41±0.02 1.98±0.13

C4 2 0.46±0.01 2.03±0.05

C5 3 0.42±0.03 1.51±0.07

C6 2 0.45±0.04 2.05±0.06

C9 2 0.45±0.01 2.00±0.17

C11 2 0.42±0.02 1.97±0.19

C18 2 0.46±0.02 2.00±0.24

C20 2 0.45±0.01 1.94±0.03

C21 2 0.44±0.02 2.30±0.15

C25 2 0.41±0.02 1.50±0.09

84

that its cellulolytic activities on the cell surface is possibly high enough to make up

the deficiency of metabolic burden. The amount of cellulase displayed on yeast

surface will be evaluated via nano LC-MS/MS in the further experiment. Considering

of heterologous protein expression and metabolic burdens, strain A26 emerges as the

best one among the cocktail δ-integration strains.

Figure III-3 Cell growth and cellulosic ethanol yields of cocktail δ-integrated strains.

Single-integrated strain EG-D-CBH1-D-CBH2-D was used as control. Ethanol was produced

from 10 g/L Avicel. The condition of cell growth measurement is the same with that in Table

III-3. Data are presented as the means and standard deviations of triplicate measurements.

Control

A18�

A23�A24�

A26

B6

A16

B24�B26�

B37�

B12

C5

C25�

15

20

25

30

35

40

45

50

55

60

65

0.2 0.7 1.2 1.7 2.2

Cel

lulo

sic

etha

nol y

ield

(%)�

Cell growth (12 h biomass concentration)�

85

Discussion

In this study, a series of recombinant yeast strains with high degradation abilities

towards crystalline cellulose were developed via optimizing the ratio of three types of

cellulase displayed on the cell surface. Notably, we obtained a strain A26 producing a

cellulosic ethanol titer corresponding to 57% of the theoretical yield. To our

knowledge, this is the highest ethanol yield from crystalline cellulose using

cellulolytic yeast among the current reports84, 92.

Currently, much effort has been done to engineer yeast strains to directly convert

cellulose, especially the crystalline cellulose into bioethanol, but the cellulolytic

activities in these strains are still insufficient. Fan et al. designed a delicate

miniscaffoldin on the surface of S. cerevisiae, achieving an ethanol yield of 27% from

10 g/L Avicel84. In addition, a yeast strain expressing a cocktail of cellulase genes was

constructed92; but this strain could only generate an ethanol yield of 12% from Avicel.

In contrast, using cocktail δ-integration method, several kinds of cellulase genes are

integrated into yeast chromosomes simultaneously in one step, and strains with high

cellulolytic activity (that is, expressing the optimum ratio of cellulases) can be easily

obtained. In Figure III-2, around one-second of the twenty-nine selected transformants

produced ethanol yields of over 45%. Especially, strain A26 exhibited an ethanol

yield of 57%, demonstrating the effectiveness of this method.

However, the production of large amounts of heterologous protein leads to

obvious growth deficiency in recombinant strains. We found that the lag phases of

some transformants were extended, indicating that the adaptive capacity of these

strains to cultivation environment declined; on the other hand, the µmax of some

transformants were decreased. A possible explanation for the lower µmax value could

reside in the metabolic burden from the overloaded heterologous protein expression.

Generally, the metabolic burden in yeast cells generates a decrease in the maximum

86

specific growth rate, biomass yield and respiratory capacity as well as an increase in

maintenance requirement120. The generation of metabolic burden has been attributed

to a broad variety of causes, such as the cellular stress associated with protein folding

and degradation of misfolded protein in the endoplasmic reticulum (ER)121, and

competition for resources to synthesize native protein and heterologous enzyme122.

These stresses would lead to a drain on the energy metabolism, which have negatively

impacted on the µmax of recombinant strain. The relationships between cell growth

(Table III-3) and Avicelase activities (Figure III-2) also support this hypothesis, that

the strains with higher cellulolytic activities (e.g., strain A16, A24, and B26) also

possessed defects in cell growth. Notably, growth-deficient strain A24 showed higher

cellulosic ethanol yield than control, indicating that its intracellular burden probably

emerged as the cell-growth-related stress in ER, rather than competing energy

resource (e.g., NADPH) with glycolysis pathway. On the contrary, strain B6 exhibited

the highest cellulolytic activities among transformants and similar cell growth to

control, but its ethanol yield is relatively low. It is postulated that the increased carbon

flux (more available glucose caused by improved cellulose degradation ability) in B6

did not flow into the synthesis of ethanol owing to the effect of metabolic burden.

Another potential reason for the growth deficiency is the random, vast

integration of genes into the δ-regions. δ-Integration simultaneously occurs on

multiple chromosomes due to the presence of around 425 δ-regions dispersed

throughout the yeast genome123. The random integration of genes into δ-regions

possibly up-/ down-regulates the expression level of some critical genes involved in

cell propagation. Currently, no obvious growth deficiency caused by δ-integration has

been observed in other reports, except that the stability of the integrated gene varied

with the integration sites on the genome after 50 generations of subculture124.

87

Many researchers have attempted to reduce the metabolic burden derived from

protein expression in yeast strains, such as employing of a non-growth related

expression system for heterologous protein gene expression125, or releasing of ER

stress through over-expression of chaperone and ubiquitin126. Canonaco et al. reported

the functional expression of arginine kinase in S. cerevisiae, setting up an intracellular

pool of phosphoarginine to maintain constant intracellular ATP levels upon

fluctuating energy demands; the engineered strain conferred a clear reduce of lag

phase period in growth127. However, owing to the disproportionate complexity of

intracellular activities, more detail information including metabolic flux balance

between amino acid biosynthesis pathway and glycolysis pathway, and the genes

against cellular stress sensitivity should be clarified to break the limitation of

metabolic burdens.

Conclusion

In this chapter, we successfully constructed a pool of cellulolytic S. cerevisiae

with diverse enzyme ratios on the cell surface using cocktail δ-integration method.

Strain A26 achieved the highest cellulosic ethanol yield from crystalline cellulose

(57%, corresponding to 1.46-fold of that in control strain (cellulase ratio is 1:1:1:1))

without any metabolic burden on cell growth. This study provided the insight that

optimization of cellulase ratios displayed on the yeast surface would significantly

increase its degradation ability towards crystalline cellulose, and ultimately result in

the overall viability of low-cost, sustainable cellulosic ethanol production.

88

GENERAL CONCLUSION

In this thesis, to achieve low-cost, environment-friendly degradation of cellulose

for bioethanol production, novel cellulolytic yeast strains were developed using

cell-surface display technique. Investigation of the suitable strategy for cellulase

production, optimization of enzyme ratio on the cell surface, and improvement of

cell-to-cellulose interactions were used to improve the cellulose degradation ability in

yeast strains. A summary of the results from each chapter is listed in Table IV. The

following can be concluded:

1. It was demonstrated that placement of EG and CBH1 in the same space (both

on the cell surface or both secreted into the medium) was favorable for ethanol

production from amorphous cellulose.

2. The cellulolytic yeast constructed via cell-surface display strategy was able to

accelerate the production rate of ethanol from cellulosic materials; it appeared

more effective in the cell-recycle batch fermentation than the strain

constructed via secretion strategy.

3. Increase of the enzyme variety (co-display of BGL, EG, CBH1, and CBH2) on

the cell surface could efficiently enhance the synergism between cellulases, as

a result, significantly improving the efficiency of cellulose hydrolysis.

4. Through cocktail δ-integration method, the enzyme ratio on the cell surface

was optimized, yielding a 1.46-fold higher cellulosic ethanol yield than the

unoptimized strain (enzyme ratio of 1:1:1:1).

5. The cellulase-displaying yeast strain exhibited clear cell-to-cellulose adhesion

and a “tearing” cellulose degradation pattern; the adhesion ability correlated

89

with enhanced surface area and roughness of the target cellulose fibers. Our

engineered strain could directly produce ethanol from rice straw, as well as

cutting down more than 40% of the required enzymes under high-dense

fermentation

6. This thesis provides a novel strategy to improve the efficiency of cellulose

hydrolysis: enhancement of the interactions between displayed-cellulase and

cellulosic substrate based on the mechanism of cell-to-cellulose adhesion.

This thesis presents some new understanding of the cellulose degradation

mechanism via cellulase-displaying yeast strain. The following is recommended for

future work.

1. Carbohydrate binding domain (CBD) has been report to possess high affinity

towards cellulose. Therefore, display of CBD on yeast surface is expected to

enhance the interactions of cell-to-cellulose as well as the cellulose

degradation ability.

2. Although the optimal biomass pretreatment method combined with

free-cellulase-mediated saccharification has been extensively studied, the

pretreatment method suitable for surface-displayed cellulases still remains

obscure, which will undoubtly improve the efficiency of cellulose hydrolysis.

Thus, the corresponding suitable pretreatment approach needs to be studied in

the future.

3. The combination of cell-surface display technique with metabolic engineering

aimed at redesigning microbial metabolic pathway will enable the production

of advanced biofuels and biochemicals from cellulose, presenting a new

opportunity in the biorefinery of cellulosic biomass.

90

91

REFERENCE

1. Nigam PS, Singh A. Production of liquid biofuels from renewable resources. Prog

Energy Combust Sci 37, 52-68 (2011).

2. Lynd LR, Van Zyl WH, McBride JE, Laser M. Consolidated bioprocessing of

cellulosic biomass: an update. Curr Opin Biotechnol 16, 577-583 (2005).

3. Ajanovic A. Biofuels versus food production: Does biofuels production increase food

prices? Energy 36, 2070-2076 (2011).

4. Naik SN, Goud VV, Rout PK, Dalai AK. Production of first and second generation

biofuels: A comprehensive review. Renew Sust Energ Rev 14, 578-597 (2010).

5. Limayem A, Ricke SC. Lignocellulosic biomass for bioethanol production: current

perspectives, potential issues and future prospects. Prog Energy Combust Sci 38,

449-467 (2012).

6. Himmel ME, et al. Biomass recalcitrance: engineering plants and enzymes for

biofuels production. Science 315, 804-807 (2007).

7. van Zyl WH, Lynd LR, den Haan R, McBride JE. Consolidated bioprocessing for

bioethanol production using Saccharomyces cerevisiae. Adv Biochem Eng Biotechnol

108, 205-235 (2007).

8. Hasunuma T, Okazaki F, Okai N, Hara KY, Ishii J, Kondo A. A review of enzymes

and microbes for lignocellulosic biorefinery and the possibility of their application to

consolidated bioprocessing technology. Bioresour Technol 135, 513-522 (2013).

9. Den Haan R, Rose SH, Lynd LR, van Zyl WH. Hydrolysis and fermentation of

amorphous cellulose by recombinant Saccharomyces cerevisiae Metab Eng 9, 87-94

(2007).

10. Yamada R, Taniguchi N, Tanaka T, Ogino C, Fukuda H, Kondo A. Direct ethanol

production from cellulosic materials using a diploid strain of Saccharomyces

cerevisiae with optimized cellulase expression. Biotechnol Biofuels 4, 8 (2011).

92

11. Wu CH, Mulchandani A, Chen W. Versatile microbial surface-display for

environmental remediation and biofuels production. Trends Microbiol 16, 181-188

(2008).

12. Lee SY, Choi JH, Xu Z. Microbial cell-surface display. Trends Biotechnol 21, 45-52

(2003).

13. Orlean P. Architecture and biosynthesis of the Saccharomyces cerevisiae cell wall.

Genetics 192, 775-818 (2012).

14. Kondo A, Ueda M. Yeast cell-surface display-applications of molecular display. Appl

Microbiol Biotechnol 64, 28-40 (2004).

15. Hara KY, et al. Development of a glutathione production process from proteinaceous

biomass resources using protease-displaying Saccharomyces cerevisiae. Appl

Microbiol Biotechnol 93, 1495-1502 (2012).

16. Bae J, Kuroda K, Ueda M. Proximity effect among cellulose-degrading enzymes

displayed on the Saccharomyces cerevisiae cell surface. Appl Environ Microb 81,

59-66 (2015).

17. Kuroda K, Ueda M. Bioadsorption of cadmium ion by cell surface-engineered yeasts

displaying metallothionein and hexa-His. Appl Microbiol Biotechnol 63, 182-186

(2003).

18. Liu W, Zhao H, Jia B, Xu L, Yan Y. Surface display of active lipase in

Saccharomyces cerevisiae using Cwp2 as an anchor protein. Biotechnol Lett 32,

255-260 (2010).

19. Kambe-Honjoh H, Ohsumi K, Shimoi H, Nakajima H, Kitamoto K. Molecular

breeding of yeast with higher metal-adsorption capacity by expression of

histidine-repeat insertion in the protein anchored to the cell wall. J Gen Appl

Microbiol 46, 113-117 (2000).

20. Mormeneo M, Andrés I, Bofill C, Díaz P, Zueco J. Efficient secretion of Bacillus

subtilis lipase A in Saccharomyces cerevisiae by translational fusion to the Pir4 cell

wall protein. Appl Microbiol Biotechnol 80, 437-445 (2008).

93

21. Matsumoto T, Fukuda H, Ueda M, Tanaka A, Kondo A. Construction of yeast strains

with high cell surface lipase activity by using novel display systems based on the

Flo1p flocculation functional domain. Appl Environ Microbiol 68, 4517-4522 (2002).

22. Sato N, Matsumoto T, Ueda M, Tanaka A, Fukuda H, Kondo A. Long anchor using

Flo1 protein enhances reactivity of cell surface-displayed glucoamylase to polymer

substrates. Appl Microbiol Biotechnol 60, 469-474 (2002).

23. Xu L, et al. Preparation of a promising whole cell biocatalyst of Geotrichum sp.

lipase and its properties. J Chem Technol Biotechnol 87, 498-504 (2012).

24. Nam KT, Lee YJ, Krauland EM, Kottmann ST, Belcher AM. Peptide-mediated

reduction of silver ions on engineered biological scaffolds. ACS Nano 2, 1480-1486

(2008).

25. Inokuma K, Hasunuma T, Kondo A. Efficient yeast cell-surface display of exo-and

endo-cellulase using the SED1 anchoring region and its original promoter. Biotechnol

Biofuels 7, (2014).

26. Ni X, Yue L, Chi Z, Li J, Wang X, Madzak C. Alkaline protease gene cloning from

the marine yeast Aureobasidium pullulans HN2-3 and the protease surface display on

Yarrowia lipolytica for bioactive peptide production. Mar Biotechnol (NY) 11, 81-89

(2009).

27. Song H, et al. Construction of a whole-cell catalyst displaying a fungal lipase for

effective treatment of oily wastewaters. J Mol Catal B: Enzym 71, 166-170 (2011).

28. Pan X-X, et al. Efficient display of active Geotrichum sp. lipase on Pichia pastoris

cell wall and its application as a whole-cell biocatalyst to enrich EPA and DHA in

fish oil. J Agric Food Chem 60, 9673-9679 (2012).

29. Han Z-l, Han S-y, Zheng S-p, Lin Y. Enhancing thermostability of a Rhizomucor

miehei lipase by engineering a disulfide bond and displaying on the yeast cell surface.

Appl Microbiol Biotechnol 85, 117-126 (2009).

30. Liang X-x, et al. Quantitative evaluation of Candia antarctica lipase B displayed on

the cell surface of a Pichia pastoris based on an FS anchor system. Biotechnol Lett 35,

367-374 (2013).

94

31. Yan YJ, Xu L, Dai M. A synergetic whole-cell biocatalyst for biodiesel production.

RSC Adv 2, 6170-6173 (2012).

32. Jin Z, Han SY, Zhang L, Zheng SP, Wang Y, Lin Y. Combined utilization of

lipase-displaying Pichia pastoris whole-cell biocatalysts to improve biodiesel

production in co-solvent media. Bioresour Technol 130, 102-109 (2013).

33. Moura MVH, da Silva GP, de Oliveira Machado AC, Torres FAG, Freire DMG,

Almeida RV. Displaying lipase B from Candida antarctica in Pichia pastoris using

the yeast surface display approach: Prospection of a new anchor and characterization

of the whole cell biocatalyst. PloS One 10, e0141454 (2015).

34. Wang Q, Li L, Chen M, Qi Q, Wang PG. Construction of a novel Pichia pastoris

cell-surface display system based on the cell wall protein Pir1. Curr Microbiol 56,

352-357 (2008).

35. Yanase S, et al. Direct ethanol production from cellulosic materials at high

temperature using the thermotolerant yeast Kluyveromyces marxianus displaying

cellulolytic enzymes. Appl Microbiol Biotechnol 88, 381-388 (2010).

36. Kim SY, Sohn J-H, Pyun Y-R, Yang IS, Kim K-H, Choi E-S. In vitro evolution of

lipase B from Candida antarctica using surface display in hansenula polymorpha. J

Microbiol Biotechnol 17, 1308-1315 (2007).

37. Hasunuma T, Kondo A. Development of yeast cell factories for consolidated

bioprocessing of lignocellulose to bioethanol through cell surface engineering.

Biotechnol Adv 30, 1207-1218 (2012).

38. Pack SP, Park K, Yoo YJ. Enhancement of β-glucosidase stability and

cellobiose-usage using surface-engineered recombinant Saccharomyces cerevisiae in

ethanol production. Biotechnol Lett 24, 1919-1925 (2002).

39. Matano Y, Hasunuma T, Kondo A. Display of cellulases on the cell surface of

Saccharomyces cerevisiae for high yield ethanol production from high-solid

lignocellulosic biomass. Bioresour Technol 108, 128-133 (2012).

40. Kotaka A, Sahara H, Kuroda K, Kondo A, Ueda M, Hata Y. Enhancement of

95

β-glucosidase activity on the cell-surface of sake yeast by disruption of SED1 J

Biosci Bioeng 109, 442-446 (2010).

41. Van Zyl JHD, Den Haan R, Van Zyl WH. Over-expression of native Saccharomyces

cerevisiae exocytic SNARE genes increased heterologous cellulase secretion. Appl

Microbiol Biotechnol 98, 5567-5578 (2014).

42. Tsai SL, DaSilva NA, Chen W. Functional display of complex cellulosomes on the

yeast surface via adaptive assembly. ACS Synth Biol 2, 14-21 (2013).

43. Nakatani Y, Yamada R, Ogino C, Kondo A. Synergetic effect of yeast cell-surface

expression of cellulase and expansin-like protein on direct ethanol production from

cellulose. Microb Cell Fact 12, 66 (2013).

44. Yamada R, Taniguchi N, Tanaka T, Ogino C, Fukuda H, Kondo A. Cocktail

δ-integration: a novel method to construct cellulolytic enzyme expression

ratio-optimized yeast strains. Microb Cell Fact 9, 32 (2010).

45. Nakanishi A, Kuroda K, Ueda M. Direct fermentation of newspaper after

laccase-treatment using yeast codisplaying endoglucanase, cellobiohydrolase, and

β-glucosidase. Renew Energ 44, 199-205 (2012).

46. Matano Y, Hasunuma T, Kondo A. Cell recycle batch fermentation of high-solid

lignocellulose using a recombinant cellulase-displaying yeast strain for high yield

ethanol production in consolidated bioprocessing. Bioresour Technol 135, 403-409

(2013).

47. Nakanishi A, et al. Effect of pretreatment of hydrothermally processed rice straw

with laccase-displaying yeast on ethanol fermentation. Appl Microbiol Biotechnol 94,

939-948 (2012).

48. Yamada R, Nakatani Y, Ogino C, Kondo A. Efficient direct ethanol production from

cellulose by cellulase- and cellodextrin transporter-co-expressing Saccharomyces

cerevisiae. AMB Express doi: 10.1186/2191-0855-3-34. , (2013).

49. Fujita Y, Ito J, Ueda M, Fukuda H, Kondo A. Synergistic saccharification, and direct

fermentation to ethanol, of amorphous cellulose by use of an engineered yeast strain

codisplaying three types of cellulolytic enzyme. Appl Environ Microb 70, 1207-1212

96

(2004).

50. Baek S-H, Kim S, Lee K, Lee J-K, Hahn J-S. Cellulosic ethanol production by

combination of cellulase-displaying yeast cells. Enzyme Microb Technol 51, 366-372

(2012).

51. Fujita Y, et al. Direct and efficient production of ethanol from cellulosic material

with a yeast strain displaying cellulolytic enzymes. Appl Environ Microb 68,

5136-5141 (2002).

52. Garvey M, Klose H, Fischer R, Lambertz C, Commandeur U. Cellulases for biomass

degradation: comparing recombinant cellulase expression platforms. Trends

Biotechnol 31, 581-593 (2013).

53. Hasunuma T, Kondo A. Consolidated bioprocessing and simultaneous

saccharification and fermentation of lignocellulose to ethanol with thermotolerant

yeast strains. Process Biochem 47, 1287-1294 (2012).

54. Sanda T, Hasunuma T, Matsuda F, Kondo A. Repeated-batch fermentation of

lignocellulosic hydrolysate to ethanol using a hybrid Saccharomyces cerevisiae strain

metabolically engineered for tolerance to acetic and formic acids. Bioresour Technol

102, 7917-7924 (2011).

55. Kondo A, et al. High-level ethanol production from starch by a flocculent

Saccharomyces cerevisiae strain displaying cell-surface glucoamylase. Appl

Microbiol Biotechnol 58, 291-296 (2002).

56. den Haan R, van Rensburg E, Rose SH, Görgens JF, van Zyl WH. Progress and

challenges in the engineering of non-cellulolytic microorganisms for consolidated

bioprocessing. Curr Opin Biotechnol 33, 32-38 (2015).

57. Resch MG, et al. Fungal cellulases and complexed cellulosomal enzymes exhibit

synergistic mechanisms in cellulose deconstruction. Energy Environ Sci 6, 1858-1867

(2013).

58. Arantes V, Saddler JN. Access to cellulose limits the efficiency of enzymatic

hydrolysis: the role of amorphogenesis. Biotechnol Biofuels 3, (2010).

97

59. Hu J, Gourlay K, Arantes V, Van Dyk JS, Pribowo A, Saddler JN. The accessible

cellulose surface influences cellulase synergism during the hydrolysis of

lignocellulosic substrates. ChemSusChem 8, 901-907 (2015).

60. Den Haan R, Mcbride JE, Grange DlCL, Lynd LR, Van Zyl WH. Functional

expression of cellobiohydrolases in Saccharomyces cerevisiae towards one-step

conversion of cellulose to ethanol. Enzyme Microb Technol 40, 1291-1299 (2007).

61. Takada G, Kawaguchi T, Sumitani J, Arai M. Expression of Aspergillus aculeatus No.

F-50 cellobiohydrolase I (cbhI) and beta-glucosidase 1 (bgl1) genes by

Saccharomyces cerevisiae. Biosci Biotechnol Biochem 62, 1615-1618 (1998).

62. den Haan R, Kroukamp H, van Zyl J-HD, van Zyl WH. Cellobiohydrolase secretion

by yeast: current state and prospects for improvement. Process Biochem 48, 1-12

(2013).

63. Ilmén M, et al. High level secretion of cellobiohydrolases by Saccharomyces

cerevisiae. Biotechnol Biofuels 4, 30 (2011).

64. Brevnova E MJ, et al. Yeast expressing saccharolytic enzymes for consolidated

bioprocessing using starch and cellulose. PCT/US2011/039192(WO/2011/153516)

12-8-2011 3-6-0110.

65. McBride JE BE, et al. Yeast expressing cellulases for simultaneous saccharification

and fermentation using cellulose. PCT/US2009/065571 5-27-2010.

66. Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO. Enzymatic

assembly of DNA molecules up to several hundred kilobases. Nat Methods 6,

343-345 (2009).

67. Inokuma K, Yoshida T, Ishii J, Hasunuma T, Kondo A. Efficient co-displaying and

artificial ratio control of α-amylase and glucoamylase on the yeast cell surface by

using combinations of different anchoring domains. Appl Microbiol Biotechnol 99,

1655-1663 (2014).

68. Chen D-C, Yang B-C, Kuo T-T. One-step transformation of yeast in stationary phase.

Curr Genet 21, 83-84 (1992).

98

69. Ismail KSK, Sakamoto T, Hatanaka H, Hasunuma T, Kondo A. Gene expression

cross-profiling in genetically modified industrial Saccharomyces cerevisiae strains

during high-temperature ethanol production from xylose. J Biotechnol 163, 50-60

(2013).

70. Kalapos MP. Methylglyoxal in living organisms: chemistry, biochemistry, toxicology

and biological implications. Toxicol Lett 110, 145-175 (1999).

71. Yamada R, et al. Effective saccharification of kraft pulp by using a cellulase cocktail

prepared from genetically engineered Aspergillus oryzae. Biosci Biotechnol Biochem,

1-4 (2015).

72. Bunterngsook B, Eurwilaichitr L, Thamchaipenet A, Champreda V. Binding

characteristics and synergistic effects of bacterial expansins on cellulosic and

hemicellulosic substrates. Bioresour Technol 176, 129-135 (2015).

73. Andersen N, Johansen KS, Michelsen M, Stenby EH, Krogh KBRM, Olsson L.

Hydrolysis of cellulose using mono-component enzymes shows synergy during

hydrolysis of phosphoric acid swollen cellulose (PASC), but competition on Avicel.

Enzyme Microb Technol 42, 362-370 (2008).

74. Reyes-Ortiz V, et al. Addition of a carbohydrate-binding module enhances cellulase

penetration into cellulose substrates. Biotechnol Biofuels 6, (2013).

75. Shimoi H, Kitagaki H, Ohmori H, Iimura Y, Ito K. Sed1p is a major cell wall protein

of Saccharomyces cerevisiae in the stationary phase and is involved in lytic enzyme

resistance. J Bacteriol 180, 3381-3387 (1998).

76. Yanase S, et al. Ethanol production from cellulosic materials using cellulase‐

expressing yeast. Biotechnol J 5, 449-455 (2010).

77. Griggs AJ, Stickel JJ, Lischeske JJ. A mechanistic model for enzymatic

saccharification of cellulose using continuous distribution kinetics I:

depolymerization by EGI and CBHI. Biotechnol Bioeng 109, 665-675 (2012).

78. Peckys DB, Mazur P, Gould KL, de Jonge N. Fully hydrated yeast cells imaged with

electron microscopy. Biophys J 100, 2522-2529 (2011).

99

79. Ju X, Grego C, Zhang X. Specific effects of fiber size and fiber swelling on biomass

substrate surface area and enzymatic digestibility. Bioresour Technol 144, 232-239

(2013).

80. Lynd LR, Weimer PJ, Van Zyl WH, Pretorius IS. Microbial cellulose utilization:

fundamentals and biotechnology. Microbiol Mol Biol Rev 66, 506-577 (2002).

81. Cruys-Bagger N, et al. Pre-steady-state kinetics for hydrolysis of insoluble cellulose

by cellobiohydrolase Cel7A. J Biol Chem 287, 18451-18458 (2012).

82. Kurasin M, Valjamae P. Processivity of cellobiohydrolases is limited by the substrate.

J Biol Chem 286, 169-177 (2011).

83. Francisco JA, Stathopoulos C, Warren RA, Kilburn DG, Georgiou G. Specific

adhesion and hydrolysis of cellulose by intact Escherichia coli expressing surface

anchored cellulase or cellulose binding domains. Biotechnology (N Y) 11, 491-495

(1993).

84. Fan L-H, Zhang Z-J, Yu X-Y, Xue Y-X, Tan T-W. Self-surface assembly of

cellulosomes with two miniscaffoldins on Saccharomyces cerevisiae for cellulosic

ethanol production. PNAS 109, 13260-13265 (2012).

85. Jin M, Gunawan C, Uppugundla N, Balan V, Dale BE. A novel integrated biological

process for cellulosic ethanol production featuring high ethanol productivity, enzyme

recycling and yeast cells reuse. Energy Environ Sci 5, 7168-7175 (2012).

86. Watanabe I, Miyata N, Ando A, Shiroma R, Tokuyasu K, Nakamura T. Ethanol

production by repeated-batch simultaneous saccharification and fermentation (SSF)

of alkali-treated rice straw using immobilized Saccharomyces cerevisiae cells.

Bioresour Technol 123, 695-698 (2012).

87. Khaw TS, Katakura Y, Koh J, Kondo A, Ueda M, Shioya S. Evaluation of

performance of different surface-engineered yeast strains for direct ethanol

production from raw starch. Appl Microbiol Biotechnol 70, 573-579 (2006).

88. Sheehan J. Engineering direct conversion of CO2 to biofuel. Nat Biotechnol 27,

1128-1129 (2009).

100

89. Lynd LR, Wyman CE, Gerngross TU. Biocommodity Engineering. Biotechnol Prog

15, 777-793 (1999).

90. Schubert C. Can biofuels finally take center stage? Nat Biotechnol 24, 777-784

(2006).

91. Piskur J, Rozpedowska E, Polakova S, Merico A, Compagno C. How did

Saccharomyces evolve to become a good brewer? Trends Genet 22, 183-186 (2006).

92. Chang J-J, et al. Assembling a cellulase cocktail and a cellodextrin transporter into a

yeast host for CBP ethanol production. Biotechnol Biofuels 6, 19-31 (2013).

93. Wen F, Sun J, Zhao H. Yeast surface display of trifunctional minicellulosomes for

simultaneous saccharification and fermentation of cellulose to ethanol. Appl Environ

Microbiol 76, 1251-1260 (2010).

94. Morais S, Shterzer N, Lamed R, Bayer EA, Mizrahi I. A combined cell-consortium

approach for lignocellulose degradation by specialized Lactobacillus plantarum cells.

Biotechnol Biofuels 7, 112 (2014).

95. Sasaki K, et al. Mechanical milling and membrane separation for increased ethanol

production during simultaneous saccharification and co-fermentation of rice straw by

xylose-fermenting Saccharomyces cerevisiae. Bioresour Technol 185, 263-268

(2015).

96. Liu Z, et al. Combined cell-surface display- and secretion-based strategies for

production of cellulosic ethanol with Saccharomyces cerevisiae. Biotechnol Biofuels

8, 162 (2015).

97. Kroukamp H, den Haan R, van Wyk N, van Zyl WH. Overexpression of native PSE1

and SOD1 in Saccharomyces cerevisiae improved heterologous cellulase secretion.

Appl Energy 102, 150-156 (2013).

98. Adney B, Baker J. Measurement of cellulase activities. Laboratory analytical

procedure 6, 1996 (1996).

99. Moukadiri I, Armero J, Abad A, Sentandreu R, Zueco J. Identification of a

mannoprotein present in the inner layer of the cell wall of Saccharomyces cerevisiae.

101

J Bacteriol 179, 2154-2162 (1997).

100. Glick BR. Metabolic load and heterologous gene expression. Biotechnol Adv 13,

247-261 (1995).

101. Goyal G, Tsai S-L, Madan B, DaSilva NA, Chen W. Simultaneous cell growth and

ethanol production from cellulose by an engineered yeast consortium displaying a

functional mini-cellulosome. Microb Cell Fact 10, 89 (2011).

102. Li XZ, Lu J, Zhao J, Qu YB. Characteristics of corn stover pretreated with liquid hot

water and fed-batch semi-simultaneous saccharification and fermentation for

bioethanol production. Plos One 9, (2014).

103. Alvira P, Tomas-Pejo E, Ballesteros M, Negro MJ. Pretreatment technologies for an

efficient bioethanol production process based on enzymatic hydrolysis: A review.

Bioresour Technol 101, 4851-4861 (2010).

104. Valjamae P, Sild V, Pettersson G, Johansson G. The initial kinetics of hydrolysis by

cellobiohydrolases I and II is consistent with a cellulose surface-erosion model. Eur J

Biochem 253, 469-475 (1998).

105. Tsai SL, Oh J, Singh S, Chen R, Chen W. Functional assembly of minicellulosomes

on the Saccharomyces cerevisiae cell surface for cellulose hydrolysis and ethanol

production. Appl Environ Microbiol 75, 6087-6093 (2009).

106. Wang AA, Mulchandani A, Chen W. Specific adhesion to cellulose and hydrolysis of

organophosphate nerve agents by a genetically engineered Escherichia coli strain

with a surface-expressed cellulose-binding domain and organophosphorus hydrolase.

Appl Environ Microbiol 68, 1684-1689 (2002).

107. Brunecky R, et al. Revealing nature's cellulase diversity: the digestion mechanism of

Caldicellulosiruptor bescii CelA. Science 342, 1513-1516 (2013).

108. Meng XZ, Foston M, Leisen J, DeMartini J, Wyman CE, Ragauskas AJ.

Determination of porosity of lignocellulosic biomass before and after pretreatment by

using Simons' stain and NMR techniques. Bioresour Technol 144, 467-476 (2013).

109. Wang QQ, et al. Evaluations of cellulose accessibilities of lignocelluloses by solute

102

exclusion and protein adsorption techniques. Biotechnol Bioeng 109, 381-389 (2012).

110. Li C, et al. Influence of physico-chemical changes on enzymatic digestibility of ionic

liquid and AFEX pretreated corn stover. Bioresour Technol 102, 6928-6936 (2011).

111. Paulova L, Patakova P, Branska B, Rychtera M, Melzoch K. Lignocellulosic ethanol:

Technology design and its impact on process efficiency. Biotechnol Adv 33,

1091-1107 (2014).

112. Lopez-Linares JC, Romero I, Cara C, Ruiz E, Moya M, Castro E. Bioethanol

production from rapeseed straw at high solids loading with different process

configurations. Fuel 122, 112-118 (2014).

113. Ko JK, et al. Ethanol production from rice straw using optimized aqueous-ammonia

soaking pretreatment and simultaneous saccharification and fermentation processes.

Bioresour Technol 100, 4374-4380 (2009).

114. Lupoi JS, Smith EA. Evaluation of nanoparticle-immobilized cellulase for improved

ethanol yield in simultaneous saccharification and fermentation reactions. Biotechnol

Bioeng 108, 2835-2843 (2011).

115. La Grange DC, Den Haan R, Van Zyl WH. Engineering cellulolytic ability into

bioprocessing organisms. Appl Microbiol Biotechnol 87, 1195-1208 (2010).

116. Thygesen A, Oddershede J, Lilholt H, Thomsen AB, Stahl K. On the determination of

crystallinity and cellulose content in plant fibres. Cellulose 12, 563-576 (2005).

117. Yamada R, Hasunuma T, Kondo A. Endowing non-cellulolytic microorganisms with

cellulolytic activity aiming for consolidated bioprocessing. Biotechnol Adv 31,

754-763 (2013).

118. Liu Z, et al. Engineering of a novel cellulose-adherent cellulolytic Saccharomyces

cerevisiae for cellulosic biofuel production. Sci Rep 6, 24550 (2016).

119. Roy A, Hashmi S, Li Z, Dement AD, Cho KH, Kim J-H. The glucose metabolite

methylglyoxal inhibits expression of the glucose transport genes by inactivating the

cell surface glucose sensors Rgt2 and Snf3 in yeast. Mol Biol Cell, mbc. E15-11-0789

(2016).

103

120. van Rensburg E, den Haan R, Smith J, van Zyl WH, Görgens JF. The metabolic

burden of cellulase expression by recombinant Saccharomyces cerevisiae Y294 in

aerobic batch culture. Appl Microbiol Biotechnol 96, 197-209 (2012).

121. Travers KJ, Patil CK, Wodicka L, Lockhart DJ, Weissman JS, Walter P. Functional

and genomic analyses reveal an essential coordination between the unfolded protein

response and ER-associated degradation. Cell 101, 249-258 (2000).

122. Snoep JL, Yomano LP, Westerhoff HV, Ingram LO. Protein burden in Zymomonas

mobilis: negative flux and growth control due to overproduction of glycolytic

enzymes. Microbiol 141, 2329-2337 (1995).

123. Dujon B. The yeast genome project: what did we learn? Trends Genet 12, 263-270

(1996).

124. Lee FW, Da Silva NA. Sequential delta-integration for the regulated insertion of

cloned genes in Saccharomyces cerevisiae. Biotechnol Prog 13, 368-373 (1997).

125. Gorgens JF, van Zyl WH, Knoetze JH, Hahn-Hagerdal B. The metabolic burden of

the PGK1 and ADH2 promoter systems for heterologous xylanase production by

Saccharomyces cerevisiae in defined medium. Biotechnol Bioeng 73, 238-245

(2001).

126. Kim MD, Han KC, Kang HA, Rhee SK, Seo JH. Coexpression of BiP increased

antithrombotic hirudin production in recombinant Saccharomyces cerevisiae. J

Biotechnol 101, 81-87 (2003).

127. Canonaco F, Schlattner U, Pruett PS, Wallimann T, Sauer U. Functional expression

of phosphagen kinase systems confers resistance to transient stresses in

Saccharomyces cerevisiae by buffering the ATP pool. J Biol Chem 277, 31303-31309

(2002).

104

ACKNOWLEDGEMENT

I would like to express my sincere gratitude and appreciation to the followings:

l My supervisors, Professors Akihiko Kondo and Tomohisa Hasunuma, for

their assistance, guidance, advice and ideas. They gave me a wonderful

opportunity to study in this top-level laboratory. They indicated me new

perspectives and inspired me to challenge more attractive goals. I will

always remember their encouragement and enthusiasm.

l Professor Shih-Hsin Ho and Associate Professor Chiaki Ogino for their

invaluable suggestions and kind support during the preparation of this

thesis.

l Assistant Professor Kentaro Inokuma for his warm guidance on

experimental manipulations.

l Associate Professor Kengo Sasaki, for his kind supply of MC6 for the

experimental usage.

l Professor Willem H. van Zyl and Riaan den Haan, for their constructive

comments and revisions on my manuscripts.

l Professor Xinqing Zhao, Fengwu Bai, and Jo-Shu Chang, for their sincere

concerns on my thesis and helpful guidance for my future career.

l Mr. Takahiro Bamba, Musashi Takenaka, Kenta Morita, Ms. Mami

Matsuda, and all the members of Professor Kondo’s laboratory for their

technical assistance and encouragement.

l My friends, Iwamoto Ryo, Xiao Xue and Shan Wang, for their

encouragements and supports to complete my Ph.D. thesis.

105

l My parents, Juan Zhai and Dawei Liu, for their understanding, financial

support and deeply care on me. Without my family’s support, I would not

have been able to devote myself to the research and complete my thesis.

The present work was supported by Special Coordination Funds for Promoting

Science and Technology, Creation of Innovative Centers for Advanced

Interdisciplinary Research Areas (Innovative Bioproduction Kobe), MEXT, Japan, the

National Research Foundation (NRF) of South African’s Research Chair (SARChi)

for Biofuels, and through an international collaborative project supported by Japan

Society for the Promotion of Science (JSPS) and NRF of South Africa.

Liu Zhuo

Department of Chemical Science and Engineering

Graduate School of Engineering

Kobe University

106

PUBLICATION LISTS

Introduction

Liu Z, Ho SH, Hasunuma T, Chang JS, Ren NQ, Kondo A. (2016) Recent

advances in yeast cell-surface display technologies for waste biorefinery

(Review), Bioresourse Technology, S0960-8524 (16) 30435-7

Chapter I.

Liu Z, Inokuma K, Ho SH, den Haan R, Hasunuma T, van Zyl WH, Kondo A.

(2015) Combined cell-surface display- and secretion-based strategies for

production of cellulosic ethanol with Saccharomyces cerevisiae, Biotechnology

for Biofuels, 8: 162

Chapter II.

Liu Z, Ho SH, Sasaki, K, den Haan R, Inokuma K, Ogino C, van Zyl WH,

Hasunuma T, Kondo A. (2016) Engineering of a novel cellulose-adherent

cellulolytic Saccharomyces cerevisiae for cellulosic biofuel production,

Scientific Reports, 6: 24550

Chapter III.

Liu Z, Ho SH, Inokuma K, den Haan R, van Zyl WH, Hasunuma T, Kondo A.

Efficient ethanol production from crystalline cellulose by engineering of a

high-cellulolytic Saccharomyces cerevisiae: Towards consolidated bioprocessing.

under preparation.

107

Doctor Thesis, Kobe University

“Engineering of high-cellulolytic Saccharomyces cerevisiae using cell-surface display

technique: Towards consolidated bioprocessing”, 107 pages

Submitted on August 29th, 2016

The date of publication is printed in cover of repository version published in Kobe

University Repository Kernel.

© LIU ZHUO

All Right Reserved, 2016