1...,1

Ivl:tvbvt

-

 

 

.Jﬂu

s: .32...

3..”

$4.... . z

1:,

 

....r|.=

(-

.

3.11..

‘wd‘km... ......:..m.. ...

135.. PM a. .61 . Q,
fan:
.13. a .,

t! , I II
3“: .3, l

. no.»

is {a
E

. zfu§4
in]
Huh‘ .3.
.. .5...
i... . .

E
dump! 1"

s 4‘“ «

Inf.

Ila
KIA-rt... Du
. at
I .

 

. .653 ‘_
mew 6’9- * LIBRARY
Michigan State

University

 

 

 

 

This is to certify that the

dissertation entitled
Ammonia Fiber Explosion (AFEX) Treatment of Corn Stover

&
Effects of AFEX Treatment on Plant-Produced Heterologous
Cellulase
presented by

Farzaneh Teymouri

has been accepted towards fulfillment
of the requirements for

 

 

Ph .D . degree in Chemical Engineering

v Major professor

Date YIIF‘03

 

MS U i: an Afﬁrmatiw Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

am 3 3 mar

 

Jet i a 2099

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p. 15

 

AMMONIA FIBER EXPLOSION (AFEX) TREATMENT OF CORN STOVER
&
EFFECTS OF AF EX TREATMENT ON PLANT -PRODUCED
HETEROLOGOUS CELLULASE

By

Farzaneh Teymouri

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Department of Chemical Engineering and Material Science

2003

ABSTRACT

AMMONIA FIBER EXPLOSION (AFEX) TREATMENT OF CORN STOVER
&
EFFECTS OF AFEX TREATMENT ON PLANT-PRODUCED HETEROLOGOUS
CELLULASE
By

Farzaneh Teymouri

The ammonia ﬁber explosion (AFEX) process treats lignocellulosic biomass with
liquid ammonia under pressure followed by explosive pressure release to enhance
conversion of structural carbohydrates (cellulose and hemicellulose) to fermentable
sugars. The potential of ammonia ﬁber explosion (AFEX) treatment for enhancing the
enzymatic conversion of lignocellulosic biomass is well recognized. However,
optimizing the process conditions and parameters such as ammonia loading, moisture
content of biomass, temperature and residence time is necessary for maximum
effectiveness of this process. The effectiveness of the AFEX treatment was measured by
the extent of enzymatic hydrolysis of the treated corn stover and also by the amount of
ethanol produced during simultaneous sacchariﬁcation and fermentation (SSF) process
from these samples. Approximate optimal pretreatment conditions for corn stover were
found to be temperature, 90°C; kg of ammonia: kg of dry corn stover, 1:1; moisture
content of corn stover, 60% (dry weight basis [dwb]); and residence time (holding at
target temperature), 5 minutes. Approximately 97% of the theoretical glucose yield was
obtained during enzymatic hydrolysis of the optimal treated corn stover using 60 ﬁlter
paper unit (FPU) of cellulase enzyme lg of glucan. The ethanol yield of optimally AFEX-

treated corn stover has been increased up to 2.2 times over that of an untreated sample.

The cost of enzymes used for sacchariﬁcation of cellulosic residues is dominant in
the overall bioconversion process. One way to decrease this cost is to produce these
enzymes in transgenic plants. In an effort to produce heterologous cellulases in transgenic
corn plant, a research group from the Crop and Soil Sciences Department (at Michigan
State University) and I constructed several different plasmid DNAs to target these
enzymes to different compartments (cytosol or chloroplasts) of transgenic corn plants.

We have also investigated the effects of AFEX pretreatment, employing a range
of treatment temperatures, moisture contents and ammonia loadings on the activity of
plant-produced heterologous cellulase. The plant materials included transgenic tobacco
plants expressing El (endoglucanase from Acidothermus cellulolyticus). The El activity
was measured in untreated and AFEX-treated tobacco leaves to investigate the effects of
the treatment on the activity of this enzyme. The maximum observed percent of retention
of activity was about 36% for transgenic tobacco plant treated at 60°C, 0.5:] ammonia

loading ratio, and 60% moisture content (dwb).

DEDICATION

Dedicated to my parents, Rahimeh Arabi and Hossein Teymouri, who have
continually encouraged and supported me throughout my entire life. To my husband,
Seyed Nasrollah Alavi, whose love, support, encouragement and personal sacriﬁces help
me to succeed in my endeavors. To my son Amir and my daughter Ayda, who are my

inspirations.

iv

ACKNOWLEDGEMENTS

I would like to thank Dr. Bruce Dale, for excellent guidance, encouragement and
support throughout my project.

I am very grateful for the advice of my guidance committee members: Dr. Daina
Briedis, Dr. Jonathan D. Walton and specially Dr. Mariam Sticklen for her guidance,
support, and encouragement and for use of her laboratory and equipment.

I am very thankful to Dr. Hasan Alizadeh, Dr. Shahina Bano Maqbool and
Lizbeth Laureano—Perez for their suggestions, comments and technical assistance.

I greatly appreciate the help and friendship of my colleagues in the laboratory: Dr.
Anwaar Ahmad, Dr. Yahia El-Maghraby, Dr. Donald Stuart Warkentin, Hesham Oraby,
Robab Sabzikar, Mitra Sticklen and Majed Hammud.

Furthermore, I must thank the Chemical Engineering and Crop and Soil Sciences

ofﬁce staff for all of their help.

TABLE OF CONTENT

List of Tables ....................................................................................... viii
List of Figures .......................................................................................... x
Chapter 1: General Introduction .................................................................... 1
Chapter 2: Literature Review ....................................................................... 6
Biomass Pretreatment ............................................................................ 9
Cellulose Hydrolysis ............................................................................ ll
Fermentation of Sugars to Ethanol ............................................................ 14
Biomass Pretreatment in Livestock Producing Industry .................................... 14
Chapter 3: Ammonia Fiber Explosion of Com Stover ......................................... 16
Abstract ........................................................................................... l6
Ammonia Fiber Explosion Treatment ........................................................ 18
Enzymatic Hydrolysis of AFEX-Treated Samples .......................................... 28
High Performance Liquid Chromatography (HPLC) ............................... 32
Results and Discussion .................................................................. 32
Simultaneous Sacchariﬁcation and Fermentation (SSF) ................................... 55
Gas Chromatography (GC) ............................................................. 58
Results and Discussion .................................................................. 58
Solubilization of Com Stoner During Enzymatic Hydrolysis ............................ 63
Different cellulase enzyme loading .......................................................... 65
Effects of longer AFEX treatment time ..................................................... 67
Conclusion ....................................................................................... 71

vi

Future work ...................................................................................... 72

Chapter 4: Constructing plasmid DNA containing cellulose genes & Effects of AFEX on

the activity of the plant-produced heterologous cellulase ..................................... 73
Abstract .......................................................................................... 74
Literature Review ............................................................................... 75
Introduction ...................................................................................... 81
Plasmid construction ........................................................................... 86
Effects of the AFEX treatment on the activity of cellulase .............................. 93
Results and Discussion ...................................................................... 104
Summary and Future work .................................................................. 108

Chapter 5: The pea (Pisum sativum L.) rch transit peptide directs the Alcaligenes
eutrophus polyhydroxybutyrate enzymes into the maize (Zea mays L.) chloroplasts... 110

Abstract ........................................................................................ 1 1 1
Introduction .................................................................................... 1 12
Results .......................................................................................... 122
Discussion ...................................................................................... 128
Appendix A ......................................................................................... 136
Appendix B ......................................................................................... 148
Bibliography ....................................................................................... 154

vii

LIST OF TABLES

Table 3.1. Condition during an AFEX run ...................................................... 25
Table 3.2. AFEX treatment conditions in corn stover treatment .............................. 27
Table 3.3. Composition of corn stover (provided by NREL) ................................. 30

Table 3.4. An example of an enzymatic hydrolysis recipe and related calculations ....... 31
Table 3.5. AFEX treatment conditions in corn stover treatment .............................. 50

Table 3.6. Effect of increasing the AFEX treatment time from 5 to 10 minutes in glucose

and xylose conversion (at 15 FPU/g of glucan, after 168 hr) ................................. 68
Table 3.7. Effect of increasing the AFEX treatment time from 5 to 15 minutes in glucose
and xylose conversion (at 15 FPU/g of glucan, after 168 hr) .................................. 69
Table 4.1. Best Synergistic Endo/Exo Cellulase Pairs Tested ................................. 77
Table 4.2. Maximum observed expression levels (percent of total soluble protein) for
Bland Elcd enzyme in transgenic tobacco ...................................................... 78
Table 4.3. CBHI plasmid features ................................................................ 82
Table 4.4. Conditions used in heat treatment of transgenic tobacco plant ................... 99
Table 4.5. Conditions during heat treatment of transgenic tobacco plants ................. 100
Table 4.6. Conditions in ammonia treatment of transgenic tobacco plants ................ 101
Table 4.7. Conditions during one of the ammonia treatment runs ........................... 102
Table 4.8. AFEX treatment conditions for transgenic tobacco plants ...................... 103
Table 1. Comparison of rch transit peptide sequence of maize and pea .................. 129

Table 2. Comparison of the ﬁrst 24 amino acid of Rubisco small subunit sequence in
maize and pea ...................................................................................... 130

viii

Table A. 1. Summary of AFEX treatment conditions for alfalfa and tobacco plants in all
the runs ammonia loading ratio was 1:1 ........................................................ 140

Table B. 1. Retention time of sugars on the HPLC ............................................ 149

ix

LIST OF FIGURES

Figure 2.1. Cellulose structure ...................................................................... 6
Figure 2.2. Some monomers of hemicellulose .................................................... 8
Figure 3.1. Stepwise schematic diagram of the process ....................................... 17
Figure 3.2. Detail design of AFEX vessel’s lid .................................................. 20
Figure 3.3. Whole assembly of AFEX unit ...................................................... 21
Figure 3.4 Schematic diagram of laboratory AFEX apparatus ................................ 22
Figure 3.5 Ammonia sample cylinders ........................................................... 23
Figure 3.6. Physical appearance of AFEX-treated and untreated corn stover ............... 26
Figure 3.7. Effects of ammonia loading on glucan conversion at 40% moisture content
(dwb) .................................................................................................. 34
Figure 3.8. Effects of ammonia loading on glucan conversion at 60% moisture content
(dwb) .................................................................................................. 35
Figure 3.9. Effects of ammonia loading on xylan conversion at 60% moisture content
(dwb) .................................................................................................. 36
Figure 3.10. Effects of Moisture content on the glucan conversion .......................... 38
Figure 3.11. Effects of moisture content on xylan conversion ................................ 39

Figure 3.12. Effect of temperature on glucan conversion at 0.721 ammonia loading ...... 40
Figure 3.13. Effect of temperature on glucan conversion at 1:1 ammonia loading. . .......41

Figure 3.14. Effect of temperature on glucan conversion at 1.3:] ammonia loading ...... 42

Figure 3.15. Effect of temperature on xylan conversion at 0.7:1 ammonia loading ........ 43
Figure 3.16. Effect of temperature on xylan conversion at 1:1 ammonia loading .......... 44
Figure 3.17. Effect of temperature on xylan conversion at 1.3:1 ammonia loading ........ 45

Figure 3.18. Glucose concentration vs hydrolysis time at enzyme loading of 60 FPU/g of

glucan ................................................................................................. 47

Figure 3.19. Normalized hydrolysis proﬁle for AFEX treated corn stover (60% moisture,
1:1 ammonia loading ratio, 90°C) at enzyme loading of 60 F PU/g of glucan .............. 48

Figure 3.20. Xylose concentration vs hydrolysis time at enzyme loading of 60 F PU/g of
glucan ................................................................................................. 49

Figure 3.21. Effects of moisture content on glucan and xylan conversion, all the run were
kept at the set temperature for 5 min ............................................................. 52

Figure 3.22. Effects of moisture content on glucan and xylan conversion, all the run were
kept at the set temperature for 5 min .............................................................. 53

Figure 3.23. Effects of temperature on glucan and xylan conversion, all the run were kept
at the set temperature for 5 min ................................................................... 54

Figure 3.24. Picture of SFF ﬂasks aﬁer 168hr of fermentation “Images in this dissertation

are presented in color” ............................................................................... 57
Figure 3.25. SSF proﬁle for glucose concentration ............................................. 59
Figure 3.26. SSF proﬁle for xylose concentration .............................................. 60
Figure 3.27. Ethanol concentration vs. SSF time ................................................ 61
Figure 3.28. Theoretical yield of ethanol after 96 hr of SSF ................................. 62

Figure 3.29. Pictures of enzymatic hydrolysis of AFEX treated and untreated samples
with 60 FPU /g of glucan: a) At 0 hr; b) After 168 hr of hydrolysis ““lmages in this
dissertation are presented in color” ................................................................ 64

Figure 3.30. Conversion of glucan and xylan vs treatment temperature A) at enzyme
loading 15 and 6OFPU/g of glucan; B) at enzyme loading 7.5 and 15 F PU/g of glucan...66

Figure 3.31. SSF results of some of the AFEX runs with longer residence time ........... 70
Figure 4.]. Different compartments in plant cell ................................................ 80
Figure 4.2. CBHI plasmid construct .............................................................. 83
Figure 4.3. pBluescript 11 SK +/- ................................................................. 87
Figure 4.4. pSMF 8 construct ...................................................................... 88
Figure 4.5. pSMF9 construct ..................................................................... 88

xi

Figure 4.6. pBRlO—l 1, pSMF] l, pSMF12, and pSMF 13 constructs ........................ 90

Figure 4.7. pSMF 14 construct ..................................................................... 91
Figure 4.8. pSMF 15 constructs ................................................................... 92
Figure 4.9. Schematic representation of Elcd expression cassette ............................ 94

Figure 4.10. Transgenic and nontransgenic tobacco plants grown in

greenhouse ........................................................................................... 95
Figure 4.1 1. Transgenic tobacco plants growing in the
greenhouse ............................................................................................ 96

Figure 4.12. Effects of temperature on the activity of Elcd extracted from heat-treated
transgenic tobacco plant ........................................................................... 104

Figure 4.13. Effects of ammonia on the activity of Elcd extracted from ammonia treated
transgenic tobacco plants ......................................................................... 106

Figure 4.14. Activity retention of Elcd extracted from AFEX-treated transgenic tobacco
plants ................................................................................................ 107

Figure 1. Plasmid constructs used in maize transformation ................................. 116

Figure 2. Southern blots of independent transgenic maize lines containing pth (a), pth
(b) and pth (0) genes ............................................................................ 123

Figure 3. Northern blot analyses Of independent transgenic maize lines containing: pth
(a), pth (b) and pth (c) genes ............................................................... 125

Figure 4. Irnmunoblot analyses of independent transgenic maize lines expressing: PHBA
(a), PHBB (b) and PHBC (c) ..................................................................... 126

Figure 5. Immunoﬂuorescent localization of PHBC enzyme: Chloroplasts of transgenic
maize transformed with pth gene and treated with PHBC-antiserum (a); Chloroplasts of
control untransformed maize plants treated with PHBC-antiserum (b). “Images in this

dissertation are presented in color” ............................................................... 127
Figure A.l. Rubisco activity in AFEX-treated alfalfa ......................................... 143
Figure A.2. Activity of 25 ug of Rubisco from AFEX-treated alfalfa ...................... 145
Figure A.3. Activity of Rubisco in AFEX-treated tobacco .................................. 146
Figure B]. Calibration curve for cellobiose ................................................... 150

xii

Figure B.2. Calibration curve for glucose ...................................................... 150

Figure B.3. Calibration curve for xylose ....................................................... 151

Figure B.4. Calibration curve for galactose .................................................... 151
Figure B.5. Calibration curve for arabinose .................................................... 152
Figure B.6. Calibration curve for mannose ..................................................... 152
Figure B.7. Calibration curve for ethanol ....................................................... 153

xiii

Chapter 1

General Introduction

As the human population continues to increase, the worldwide demand for energy
and liquid fuels increases even faster. In the United States and much of the word the
largest single portion of this energy demand is derived from petroleum. The earth’s
petroleum supply is ﬁnite, and it is becoming less accessible. As a result, many countries
such as the United States, which are net importers of petroleum, seek to develop new
sources of energy to reduce their dependency on foreign oil and to improve their
economic strength. In addition to the economic issues, we are also faced with potentially
great environmental consequences if we don’t change our energy use patterns.

Utilization of solar energy and extracting energy from the carbon ﬁxed by

photosynthesis in plants offer an attractive solution to establish clean and sustainable
resources for both energy demands and raw material needs. It has been estimated that
1011-1012 tons of carbon are ﬁxed annually around the world by photosynthesis of higher
plants. Theoretically, this renewable resource can provide approximately 10 times our
current energy demand and 100 times our current food needs (Grohman et al., 1992).
The potential for using lignocellulosic biomass material to produce energy carriers such
as electricity, gases, and transportation fuels is well recognized. Biomass energy
produced in an efﬁcient and sustainable manner can offer numerous economic,
environmental and social beneﬁts compared with fossil fuels.

One liquid fuel that has the potential to match the convenient attributes of
petroleum fuels is ethanol produced from lignocellulosic biomass. A major problem in

the commercialization of this potential is the inherent resistance of lignocellulosic

materials to conversion to fermentable sugars (Walter, 2000). There are two main
techniques for producing fermentable sugars from lignocellulosic biomass acid
hydrolysis and enzymatic hydrolysis. The required high temperature for acid hydrolysis
results in signiﬁcant sugar degradation and, hence, low yields. Enzymatic hydrolysis is
attractive because of its speciﬁcity and absence of the competitive degradation, which
result in higher yields (Wyman, 1996). With regard to enzymatic hydrolysis, although the
yields are higher, the conversion rates are low and the enzyme is expensive. In order to
improve the efﬁciency of both acid hydrolysis and enzymatic hydrolysis, a pretreatment
step is necessary to make the cellulose fraction accessible to cellulases. In most
pretreatments, deligniﬁcation, removal of hemicellulose, and decreasing the crystallinity
of cellulose produce more accessible surface area for cellulases to react with cellulose
(Lamptey et al. ,1 986).

A major obstacle to the commercialization of enzymatic hydrolysis is the high
cost of the enzyme. One way to reduce this cost is to use much less enzyme per unit of
biomass hydrolyzed. It has been shown that this goal can be accomplished by optimizing
the biomass pretreatment conditions. The Ammonia Fiber Explosion (AFEX) technique
is a biomass pretreatment technique that has been shown to be an effective means of
increasing the yields of fermentable sugars from biomass. Previous work has shown that
the effective enzymatic hydrolysis of AFEX-treated biomass at enzyme loadings as low
as 5 F PU/g of dry biomass can be achieved by adjusting the pretreatment parameters
(Holtzapple et al., 1991 ).

In the present research one of our foci was to optimize the AFEX treatment

variables such as temperature, moisture content, ammonia loading, and residence time to

increase the efﬁciency of the enzymatic hydrolysis of AFEX-treated corn stover (includes
all above ground portions of a corn plant except the grain and cob) at lower enzyme
loading.

Another promising approach to reduce the enzyme cost is to genetically transform
plants with the genes coding for these enzymes, in order to produce the desired enzymes.
Transgenic plants may be an attractive and cost-effective alternative to microbial systems
for production of biomolecules. Advances in biotechnology are enabling plants to be
exploited as bioreactors for the production of proteins, carbohydrates, lipids, and
industrial enzymes (Halliwell and Halliwell, 1995). Field-grown plants have the potential
to produce biomolecules such as industrial enzymes in bulk quantities with minimal
inputs of raw materials and energy (Verwoerd et al., 1994). In collaboration with a
research group from the Crop and Soil Sciences Department at Michigan State
University, with the concept of producing cellulase in transgenic corn plant, three
different DNA plasmids containing cellulase genes were constructed.

Plant proteins and enzymes produced in transgenic crops have very high value,
therefore they may provide a signiﬁcant byproduct credit in biomass reﬁning. Extraction
and recovery of these proteins in their active form or even releasing active cellulase from
the plants during bioconversion could have a signiﬁcant impact on overall lignocellulose
conversion economics. As this technology continues to grow and improve the production
levels of biomolecules in plants, the development of downstream processing technology
to extract and to recover these biochemicals will determine progress. Therefore an
integrated pretreatment that improves protein recovery from biomass and increases the

digestibility of 1i gnocellulosic biomass may signiﬁcantly enhance the process economics.

Many pretreatments that increase the conversion of cellulose to fermentable sugars
operate under harsh conditions (e.g., steam explosion and various acid processes) that
tend to degrade the sugars, biomass proteins, and enzymes. However the AFEX process
operates under relatively mild conditions and has shown a good potential for protein
recovery from herbaceous crops with simultaneous conversion of cellulose and
hemicellulose to simple sugars. (De La Rosa et al., 1994)

The potential of the AF EX process in releasing active cellulase from transgenic
plants during the bioconversion of the transgenic plants was examined in this research.
Transgenic tobacco plants expressing cellulase El (endoglucanase from Acidothermus
cellulolyticus) were used for this investigation.

The potential of the AFEX process as a biomass treatment technique that is able
to preserve the biomass proteins in their bioactive form was explored in this study. We
approached this goal by investigating the effects of the AFEX process on the activity of
Rubisco, the most abundant plant protein in alfalfa plants.

The primary objectives of this research may be summarized as follows:

1) Investigate interplay between AFEX treatment conditions such as temperature,
moisture content, ammonia loading, and residence time and their effects on the AFEX
results.

2) Analyze AFEX-treated corn stover and evaluate the effects of the treatment on the
enzymatic conversion of the samples to fermentable sugars as well as amount of ethanol
produced during Simultaneous Sacchariﬁcation and Fermentation (SSF) process.

3) Optimize AFEX treatment conditions for efﬁcient conversion of corn stover to

fermentable sugars.

4) Construct a series of DNA plasmids containing cellulase genes to be used by another
research group to produce transgenic plants expressing these genes.
5) Investigate the effects of the AFEX treatment on the activity of plant-produced

proteins such as Rubisco and cellulase.

Chapter 2

Literature Review

The increase in the price of petroleum feedstock has created opportunities for the
development of combined biological and chemical processes that will produce liquid
fuels and chemicals from alternate feedstock such as biomass. The term “biomass”
generally refers to renewable organic material generated by plants through
photosynthesis. lignocellulosic biomass materials made up of three major organic
fractions cellulose, hemicellulose and lignin. Cellulose and hemicellulose together
compose about 65-75% of the overall biomass composition and signiﬁcant portion of the
remaining material is lignin (Wyman, 1996).

Cellulose is the world’s most plentiful naturally occurring organic compound.
Cellulose is an insoluble, unbranched, and linear polymer that is composed of many D-
glucose subunits, linked end to end with [3-1—4 glucosidic bonds (Figure 2.1). The general

formula for cellulose is (Cd-11005)“ where (n) is the number of glucose units in the

polymer molecule.
CH OH II on CII,OH
C———— -~C C -- ()

H 0/” 0\ C/O \. / OII II\ a" H ,9/ H \\c / 0x
00/ on H x‘ H OH H H
H OH (in—,0: H on

Figure 2.1: Cellulose structure

Cellulose chains are held together by hydrogen bonding and van der Waals forces,
and they associate to form insoluble elementary ﬁbrils that further aggregate to form
microﬁbrils about 25nm wide (Campbell et al.,1999). These bonds and forces aggregate
portions of the molecular chains into various degrees of lateral order ranging from perfect
geometrical packing of the crystal lattice (crystalline region) to a random condition
(amorphous region). The degree of crystallinity controls the chemical reactivity of
cellulose; the crystalline regions are essentially not accessible to reactant molecules.
However, crystallinity alone is not responsible for unreactive nature of cellulose.
Cellulose microﬁbrils are surrounded by layers of lignin and hemicellulose, which also

protect cellulose from degradation (W yman, 1996).

Hemicellulose is another major class of cell wall matrix polysaccharides.
Hemicelluloses can be subdivided into several groups on the basis of monosaccharide
composition and ratio: 1) arabinans; 2) xylans (homo- and heteropolysaccharides
including arabinoxylans, glucuronoxylans, arabinoglucuronoxylans); 3) glucans
(homopolysaccharides includind B-(l-3)-D-glucan callose, and heteropolysaccharides
such as arabinoglucans); 4) galactans (both homo- and heteropolysaccharides, e.g.
arabinoglactans); 5) mannans (homo- heteropolysaccharides represented by
glucomannans, galactomannas, and glucogalactomannanas); 6) fructosans and
glucofructosans; 7) galacturonans, and 8) mannuronans (Tarchevsky and Marchenko,
1991). The main hemicellulosic component of monocotyledonous and dicotyledonous
plants is represented by xylans with a back bone built of B-(l-4)-linked D-xylose residues
bearing branches such as D-glucuronic acid, glucose, arabinose, and galactose (some of

these sugars are presented in Figure 2.2). Unlike cellulose, which is a crystalline strong,

and resistant to hydrolysis, hemicellulose has an amorphous structure with little physical
strength. It can be hydrolyzed easily by hemicellulases and mild acid to produce simple

sugars (Tarchevsky and Marchenko, 1991).

 

H- =0 H— =0 H-C=0
-OH H- -OH 0°" l-IO-C-H "24)"
H0- -H 110- 41 0 H0- -H 0
'OH H‘ -OH H- ‘OH
'0” H- -OH H- '0”
Hz-OH coon Hz-OH
0’9'00093 D-glucuronic acid D-mannose
ft = H- =0
HO- 'H -0H
-OH ”0- -H
-0” -0H
“[0" 112-011
D-arabinose D‘XY'OSC

Figure 2.2. Some monomers of hemicellulose

Lignin is essentially a three-dimensional phenylpropane polymer with
phenylpropane units held together by ether and carbon ~carbon bonds. It possesses a high
molecular weight, and it is amorphous in nature. Lignin, hemicellulose and cellulose with
extensive cross linking among them comprise the very rigid cell wall matrix of the plant
(T archevsky and Marchenko, 1991).

The hydrolysis of polysaccharides in plant cell walls is far more challenging than
hydrolysis of storage polysaccharides (starch). Due to intimate and strong associations
with hemicellulose and lignin, cellulose cannot be liberated by simple mechanical

treatments used successfully on starch granules.

Biomass Pretreatments

A key aspect of the overall conversion system for biological fuel production is
the integration of plant substrates and pretreatments to provide easily hydrolysable
cellulose. These treatments usually cause the destruction or severe modiﬁcation of
hemicelluloses and lignin. In most of these treatments the plant tissues are ﬁrst converted
into smaller particles to aid penetration of acidic and enzymatic catalysts. Chipping,
shredding and milling steps are common to most treatments. Acid-catalyzed hydrolysis
processes are more tolerant of larger particles because small acid molecules diffuse
through cell walls and plant tissues much more rapidly than the relatively large enzyme
molecules. However, even in acid hydrolysis, the use of large particles will decrease the
yield of sugars (T orget et al., 1988).

For enzymatic hydrolysis, the minimal pretreatment requirement is the
production of ﬁne particles with a high surface area accessible to cellulase enzymes.
Small particles can be produced by explosive decompression of steam or gases, but to
penetrate cell walls and to make cellulose surfaces more accessible, we also need to
modify other components in plant cell walls such as lignin and hemicellulose, which are
recognized as major barriers to enzymatic hydrolysis (Grohmann et al., 1992).

There are three major classes of biomass treatments, which are described below.
Physical Treatment

The most common physical treatment is ball milling. This technique provides a
dramatic increase in the rates of hydrolysis and in the yields of fermentable sugars.
Shearing and compressive forces involved in milling reduce crystallinity, decrease the

mean degree of polymerization, and increase surface area (Abraham and Kurup, 1997).

But this physical treatment demands expensive electrical energy, which is a major
drawback for this technique (Holtzapple et al., 1991).
Chemical Treatment

Chemical treatment with strong acids or bases also effectively increases the
hydrolysis of cellulose. These chemicals are generally quite corrosive and expensive. In
addition, they are often toxic or inhibitory to microorganisms, so that they have to be
removed. These problems increase the expense and difﬁculty of such chemical treatments
(Schell and Duff, 1996).

Physicochemical Treatment

In physicochemical treatment both physical and chemical aspects are involved.
This technique has the advantage of physical treatment without the expense of high
energy use in ball milling. The two common examples of physicochemical treatment are
steam explosion and ammonia ﬁber explosion (AFEX).

In steam explosion, lignocellulosic materials are saturated with water under
pressure (300-500 psi) at elevated temperature (215-260°C). When the pressure is
released, the water evaporates rapidly and makes the cellulose ﬁbers separate, which in
turn increases the surface area for subsequent hydrolysis. Even though the technique is
effective, it demands considerable thermal energy. In addition, due to the high
temperatures, some of the sugars are degraded (Carrasco et al., 1994).

The ammonia ﬁber explosion process (AFEX) treats lignocellulosic biomass
materials with liquid anhydrous ammonia under pressure and moderate temperatures (60-

90°C) for a few minutes and then the pressure is rapidly released. The rapid

10

decompression expands the ﬁber structure of the biomass and signiﬁcantly increases the
surface area available for enzymatic attack (Dale and Moreira, 1983).

This pretreatment has been shown to greatly enhance the susceptibility of biomass
to enzymatic hydrolysis. The mechanism by which the enhancement occurs has not been
elucidated yet, but presumably involves changes in physical and chemical structure. It is
known that anhydrous ammonia in either liquid or gaseous form is a strong cellulose
swelling agent which can change the cellulose ﬁber structure from crystalline cellulose I
(native crystalline cellulose) to crystalline cellulose III (ammonia-treated cellulose).
Anhydrous ammonia penetrates cellulose and reacts with the hydroxyl groups after
breaking the hydrogen bonds (Barry et al., 1936). Through its penetration in to the
crystallites the distances between the cellulose chains increases. The cellulose III
crystallites obtained are smaller than those of cellulose I, 16-23A as against 50-80A
(Lewin and Roldan, 1971). Ammonia can react with the lignocellulosic biomass by
ammonolysis of the ester crosslinks of uronic acid with the xylan units, cleaving the
bonds linking hemicellulose and lignin, and cleaving the C-0 and C-C bonds of lignin
macromolecules to produce smaller soluble fragments (Chou, 1986).

Cellulose Hydrolysis

Bioethanol can be produced by converting the complex cellulosic and
hemicellulosic polysaccharides into simple sugars, which can be fermented to ethanol by
various microorganisms. Complex polysaccharides can be hydrolyzed by acid or
enzymes.

Dilute acid hydrolysis is inexpensive and relatively simple, but the high

temperatures, ranging from 140°C to 260°C, may result in the formation of sugar

11

degradation products. Concentrated acids also can be used to hydrolyze the
lignocellulosic material and, since the temperature is lower (100- 120°C), high sugars
yields are obtained with little degradation of products. However concentrated acid
hydrolysis requires later neutralization of the acid solution and recovery of the acid, both
of which are expensive processes (Schell and Duff, 1996). On the other hand, enzymatic
hydrolysis of cellulose is attractive because of its speciﬁcity and absence of competitive
degradation, which results in high yield of sugars.

Cellulase refers to a family of enzymes that act synergistically to hydrolyze
cellulose to glucose. Cellulases are widely distributed throughout the biosphere and are
most manifest in fungal and microbial organisms. Enzymatic degradation of cellulose to
glucose is generally accomplished by the synergistic action of three distinct classes of
cellulase enzymes: Endo-1,4-B-glucanases (EC 3.2.1.4), Exo-l,4-B-D—glucanases or 1,4-
B-D-glucan cellobiohydrolases (EC3.2.1.91), and B-D-glucosidases (EC 3.2.1.21) (Wilke
et al., 1983).

The rate and the extent of enzymatic conversion of biomass are limited, by end-
product inhibition of cellulase by glucose and cellobiose formed during the breakdown of
cellulose. Several methods have been proposed to overcome this inhibitory problem.
These include the use of high concentration of enzymes (Ishihara et al., 1991), the
elimination of sugars from the hydrolysate by ultraﬁltration (Tan et al., 1986), or the
simultaneous sacchariﬁcation and fermentation (SSF) of the substrate (Saddler et al.,
1982), and supplementation of cellulases with B-glucosidase that converts cellobiose into
glucose (Wyman et al., 1986, Spindler et al., 1988, Spinder et al., 1992; Breuil et al.,

1990). It has been shown that both the endoglucanases and cellobiohydrolases are

12

inhibited by increased concentrations of cellobiose, whereas B-glucosidase is more
sensitive to glucose accumulation. Although the glucanases are also inhibited by glucose,
the drop in the overall enzyme activity of the system due to glucose accumulation is still
relatively small compared with that associated with cellobiose accumulation (Breuil et
al., 1990; Holtzapple et al., 1990).

Furthermore, the accumulation of glucose was minimized by using the
simultaneous sacchariﬁcation and fermentation (SSF) process, developed by the Gulf Oil
Corporation (Emert and Ratzen 1980). In this process the glucose becomes simply an
intermediate, which lowers the glucose concentration than in any one-stage

sacchariﬁcation process, provides less inhibition to the rate of conversion of the cellulose

(Wyman, 1999).

13

Fermentation of sugars to ethanol

Simple sugars such as glucose can be fermented to ethanol using yeasts and other
microbes. The most common yeast used in the fermentation of glucose is Saccharomyces
cerevisiae. The stoichiometric equation of glucose fermentation to ethanol is:

C6H1206 —> 2C2H50H + 2C0;

Usually the fermentation proceed until the ethanol concentration reaches ~10%-
12%, when the activity of the yeast becomes inhibited by the ethanol. In practice, the
ethanol yield from sugar is about 46% by weight; as opposed to the 51.1% theoretical
yield (Wyman, 1999).

The development of simultaneous sacchariﬁcation and fermentation (SSF) was a
major breakthrough in cellulose utilization. In this process hydrolysis and fermentation
are combined in one vessel (Walter, 2000). The presence of yeast along with the enzyme
minimizes the inhibitory effects of sugars. The sugars produced during the process slow
down the action of cellulase enzyme, which in turn lowers the ethanol production.
Additional beneﬁts are that the number of fermentation vessels is reduced, which lowers
the capital cost of the process. Furthermore the presence of ethanol makes the mixture
less vulnerable to contamination by unwanted microorganisms.

Biomass pretreatment in livestock producing industry

High protein crops such as alfalfa and other forages (which contain lO°/o-20%
protein) are particularly important in the livestock industry (De La Rosa et al., 1994).
Feed-grade protein is worth four times more per unit weight than carbohydrates;

therefore, any effort to increase the digestibility of these feeds and at the same time

14

preserve their protein in their useable form might signiﬁcantly improve the economics of
livestock production. Many biomass pretreatments have been used to improve feed
quality by increasing the digestibility of the feed. Unfortunately, most of these
pretreatments use relatively high temperatures, which are detrimental to feed grade
protein in biomass.

Ammonia ﬁber explosion treatment has shown great potential for improving the
digestibility of biomass by ruminants (Weaver, 1998, Gollapalli, 2001). In addition to
enhancing substrate digestibility, the ammonia bound to the biomass during the reaction
can serve as a source of nitrogen for the animal. Only ruminants can use non-protein
nitrogen as a source of nitrogen for synthesis of protein, therefore less protein needs to be
added to the animal’s diet. AFEX treatment has also shown potential to be used as an
integrated process that allows more protein extraction from biomass as well as increasing
the digestibility of biomass compared to the untreated biomass. The quality of biomass
feed will be signiﬁcantly improved if the biomass proteins are preserved in useable form

after the treatment.

15

Chapter 3

Ammonia Fiber Explosion of Corn Stover

Abstract

The potential of ammonia ﬁber explosion (AFEX) treatment for enhancing the
enzymatic conversion of lignocellulosic biomass is well recognized. However,
optimizing the process conditions and parameters such as ammonia loading, moisture
content of biomass, temperature, and residence time are necessary for maximum
effectiveness of this process.

In this research our focus was to identify the AF EX process conditions to
maximize the enzymatic conversion of corn stover to fermentable sugars, which in turn
would result in higher ethanol production. The effectiveness of the AFEX treatment was
measured by the extent of enzymatic hydrolysis of the treated corn stover and also by the
amount of ethanol produced during simultaneous sacchariﬁcation and fermentation (SSF)
from these samples. The stepwise schematic diagrams of these processes are illustrated in

Figure 3.1 (a & b).

16

Figure 3.1. Stepwise schematic diagram of the process

 

 

 

 

 

 

 

Ammonia
.. Tl
AFEX I
Treatment
Corn AFEX
Stover treated
Corn
b)
Ammonia
——> #1313 X ———->
reatment
Corn AFEX
Stover treated
Corn

Water

Wash

Water

 

 

 

 

 

 

1.}.

 

 

 

 

 

 

 

Enzyme
. I Enzymatic I
Washed HydrolySIS
. AFEX Hydrolysate
l liquid
treated
(Fermentable
Com Stover ,
ReSIdual sugars)
Solids
Yeast Enzyme
SSF —> Ethanol
Washed
AFEX
treated
Corn Stover .
Resrdual
Solids

17

Ammonia Fiber Explosion treatment

Following materials were used in this experiment:

Liquid anhydrous ammonia from AGA (Lansing, MI)

Dried and milled corn stover (with 10% moisture): provided by National Renewable
Energy Laboratory (Golden, CO) (NREL).

Throughout this report ammonia to biomass ratio is presented as g of ammonia: g of dry
biomass, and moisture content is presented as g of moisture/g of dry biomass (dry weight
basis [dwb]).

Ammonia hazards

Ammonia as gas or liquid is highly soluble in water. The reaction between ammonia and
water generates ammonia hydroxide, a caustic compound. This makes ammonia a contact
hazard, as well as inhalation hazard. The gas or liquid can be absorbed by moisture of
skin or eyes and cause caustic burns. Anhydrous liquid ammonia is also dangerous
because of the high heat of vaporization, which may lead to skin freezing (Davis, 1987;
Kirk-Othmer, 1992). Therefore all the AFEX treatments were carried out in fume hood,

and safety glasses and heavy-duty gloves were worn during the experiment.

Description of the apparatus

A 300 ml stainless steel pressure vessel (PARR Instrument Co., IL), equipped
with pressure gauge, feeding port, and exhaust ball valve was used as the reactor in this
experiment. However, in order to have better control on the system, several modiﬁcations

were made. To uniformly distribute ammonia in the vessel, three holes 1/16 inch of

18

diameter, one inch apart were drilled in the feeding tube. Another modiﬁcation was to
install a thermocouple well to monitor the temperature during the experiment. The well is
located one half inch from the bottom of the vessel, between the center and the inner wall
of the vessel. The arrangement of the feeding port, exhaust valve, and thermocouple on
the lid, the whole assembly of AFEX unit, and the schematic diagram of the laboratory
AFEX apparatus are shown in Figure 3.2, Figure 3.3, and Figure 3.4 respectively.

In order to load the vessel with the desired amount of ammonia, three
precalibrated sample cylinders with different capacities (10, 20, and 30g of ammonia)
were made (Figure 3.5). The anhydrous ammonia used was supplied in a 501b cylinder
equipped with a dip tube to ensure liquid delivery. In order to reduce the heating time, the
liquid ammonia was preheated before loading. For this purpose a lecture bottle with 300g
ammonia capacity, equipped with a dip tube, was ﬁlled from the main ammonia cylinder

(501b cylinder) and maintained at about 35°C in a water bath.

l9

Figure 3.2. Detail design of AFEX vessel’s lid

 

Ammonia
feeding valve

 

 

 

 

 

Exhaust
valve

 

 

  
  
  

20

 

Pressure
gauge

 

 

 

Figure 3.3. Whole assembly of AF EX unit.

    
   
    
 

Lecture bottl

 

r.

3
.. v-Jl
\-

AF EX Unit

Figure 3.4. Schematic diagram of laboratory AFEX apparatus

pressure
gauge

value valve
. pressure
gauge
?I'

«e— ammonia tank

  
 
   
  

 

 

temperature controller , ; 49— I‘BBCtor

q— water bath heating mantle

 

 

 

 

 

 

x")

E) heater 0 I

 

 

 

22

Figure 3.5. Ammonia sample cylinders.

g
Q
9.;

 

Capacity: 10g of Ammonia Capacity: 20g of Ammonia Capacity: 30g of Ammonia

23

Experiment Procedure

To carry out the experiment, corn stover samples were moistened with distilled
water to the desired moisture content and allowed to equilibrate for 30 minutes. The
prewetted samples were placed in the pressure vessel. The vessel was topped up with
stainless steel pellets (approximately 1mm diameter) to occupy the void space and thus
minimize transformation of the ammonia from liquid to gas during loading, and then the
lid was bolted shut. Liquid ammonia is most effective in transforming cellulosic biomass.

The precalibrated sample cylinders were ﬁlled from the lecture bottle with the
desired amount of warm ammonia to charge to the system. To insure that the desired
amount of ammonia was delivered, the reactor vessel was weighed before and after
loading. The entire reactor assembly was placed in a 400W PARR heating mantle to
warm the unit to the desired temperature, and the vessel temperature was monitored using
a K type thermocouple. To avoid overheating, the reactor was taken out of the heater at
approximately 10°C prior to the target temperature, and, if needed, the unit was placed in
a bath of cold water to maintain the system at the set temperature. The pressure and the
temperature were monitored and recorded throughout the experiment. After the
experiment was completed the exhaust valve was rapidly opened to relieve the pressure
and accomplish the explosion. The treated corn stover samples were removed and
allowed to stand overnight in a fume hood to evaporate the residual ammonia. An
example of an AFEX run recipe and the conditions during the run (temperature and

pressure) is presented in Table 3.1.

24

Table 3.1. Conditions during an AFEX run

 

18 g of Untreated corn stover (containing 1.8 g of water)

 

7.8 g of distilled water was added for 60% moisture (dry weight basis)

 

16.5 g of liquid ammonia for 1:1 ammonia loading ratio; Ammonia temperature: 35°C

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Elapsed time, min. Temperature, °C Pressure, psi g Comment

0 50 90 Put in the heater
2 54 120

4 58 150

6 67 180

8 74 205

10 78 225

12 80 240 Taken out of the
14 83 255

16 86 260

18 88 255

20 91 245

22 91 220

23 91 195 Pressure released

 

 

 

 

25

 

Treatment of the corn stover with ammonia resulted in a considerable darkening
of corn stover compare with untreated sample, but did not change the macroscopic
appearance of the substrate (Figure 3.6). The treated samples were stored in sealed plastic

bags in the refrigerator.

Figure 3.6. Physical appearance of AF EX treated and unheated corn stover.

 

26

The corn stover samples were treated under different process variables
(temperature, moisture content, ammoniazbiomass level, and time) to investigate interplay
between AFEX treatment conditions and to optimize this treatment for corn stover. The

various treatment conditions in AFEX pretreatment of the corn stover are summarized in

Table 3.2.

Table 3.2. AFEX treatment conditions in corn stover treatment.

 

 

 

 

 

 

 

 

(g of NH3: g of dry Biomass) 20% Moisture" 40% Moisture 60% Moisture
0.7: l T=60°C T=60°C T=60°C
T=70°C T=70°C T=70°C
T=80°C T=80°C T=80°C
T=90°C T=90°C T=90°C
l: l T=60°C T=60°C T=60°C
T=70°C =70°C T=70°C
T=80°C T=80°C T=80°C
T=90°C T=90°C T=90°C
l .3: l T=60°C T=60°C T=60°C
T=70°C T=70°C T=70°C
=80°C T=80°C T=80°C
T=90°C T=90°C T=90°C

 

* All run were kept at the target temperature for 5 min.

** Moisture content was calculated on dry weight basis.

27

 

Enzymatic Hydrolysis of AFEX-Treated Samples

Following materials were used:
Untreated corn stover as control (from NREL)
AFEX-treated corn stover
Cellulase enzyme (provided by NREL, CAS 9012-548, activity: 28 FPU/ml; the activity
was measured based on NREL standard ﬁlter paper unit assay protocol, LAP-006)
B-glucosidase (Novozym 188) from Sigma (St. Louis, MO)
The following chemicals were provided from Sigma:
(It-cellulose (Cat. # C-8002), tetracycline, cycloheximide, sodium citrate and pure sugars:
glucose, xylose, mannose, galactose, cellobiose and arabinose.
Method

In this series of experiments we essentially followed the NREL standard biomass
enzymatic hydrolysis protocol (LAP-009). All the NREL standard protocols can be
obtained from the following website:
http://www.ott.doe.gov/biofuels/analytical methodshtml
Experimental Procedure

The pretreated corn stover along with the untreated sample was washed to
remove glucose and inhibitors. The samples were squeezed through miracloth with 22-
25um pore size (Calbiochem, La Jalla, Ca), and then the wash water was subjected to
sugar analysis using high performance liquid chromatography (HPLC). The samples must
be washed until the glucose concentration in the water wash falls below 0.1 g/l. For all

the samples the concentration of glucose and other monosaccharides (xylose, mannose,

28

galactose, and arabinose) in the wash water were either zero or less than 0.1 g/l after the
ﬁrst wash, which indicates that no monosaccharides are produced during the AFEX
treatment. The water wash was kept in the freezer for further enzymatic hydrolysis to
check for oligomers. The washed samples were stored in Petri dish (sealed with parafﬁn
ﬁlm) at 4°C for further analysis.

Total solids of the washed samples were determined according to NREL protocol

(LAP-001). One gram of washed samples was dried at 105°C overnight; the percent of

total solids on 105°C dry weight basis was calculated as follows (equation 3.1):

weight of dried sample
weight of sample before drying

% Total solids = X100 (3.1)

 

Based on the glucan content of the corn stover (the composition of the corn stover
sample was provided to us by NREL, Table 3.3), a sample equivalent to 0.25g of glucan
on a 105°C dry weight basis was weighed and added to the 25ml glass vial. To each vial
an appropriate amount of sodium citrate buffer, tetracycline and cycloheximide were
added. An amount of distilled water was added to bring the total volume in each via] to
25ml after addition of the enzymes in the following steps. The solution and the biomass
are assumed to have a speciﬁc gravity of 1.00g/ml. An example of an enzymatic
hydrolysis recipe and related calculations are presented in Table 3.4.

The contents of the vials were warmed to 50°C in the incubator. Then the

appropriate volume of cellulase equal to 15 or 60 FPU/g of glucan and the appropriate

29

volume of B-glucosidase equal to 40 IU/g of glucan were added to each vial. The vials

were incubated at 50°C with gentle rotation (75 rpm) for a period of 168 hr.

Table 3.3. Composition of the Corn Stover (provided by NREL)

 

 

 

 

 

 

 

 

 

 

 

 

 

Component Percentage (based on dry basis)
Glucan 36.1
Xylan 21.4
Arabinan 3.5
Mannan 1.8
Galactan 2.5
Lignin 17.2
Protein 4.0
Acetyl 3.2
Ash 7.1
Uronic Acid (est) 3.6
Non-structural Sugars 1.2
Total 101.6

 

 

 

 

30

Table 3.4. An example of an enzymatic hydrolysis recipe and related calculations

 

 

 

 

 

 

 

 

 

 

 

 

Washed sample (wet) 1.0482 g
Dried sample at 105°C 0.2972 g
Solid content 28.37 %
Sample loaded in vial for 0.25 g of 2.44 g
glucan
Sodium citrate buffer (0.1M, pH 4.8) 1.25 ml
Tetracycline (10 mg/ml in 70% 0.1 ml
ethanol)
Cycloheximide (10 mg/ml in distilled 0.075 ml
water)
Cellulase (28 FPU/ml) 0.535 ml for 60 FPU or 0.134 ml for 15 FPU/g
of glucan
B-glucosidase 0.05 ml for 40 IU/ g of glucan
Distilled water 20.95 ml
Total 25 ml

 

 

A lml aliquot was removed at each predetermined time interval (0, 3, 6, 24, 48,

72, and 168hr). The samples were centrifuged; the supematants were ﬁltered and

subjected to sugars analysis using HPLC.

31

 

High Performance Liquid Chromatography (HPLC)

The sugars were analyzed in a BioRad (Richmond, CA) High Performance Liquid
Chromatograph using an Aminex HPX87P column (HPLC Carbohydrates Analysis
Column) at 85°C and a BioRad Deashing Cartridge as a guard column. The mobile phase
used was degassed HPLC water at a ﬂow rate of 0.6 ml/min. The injection volume used
was 20 ILL and the run time was 20 minutes.

Standard solutions of pure sugars: glucose, xylose, mannose, galactose, cellobiose and
arabinose were individually run on the HPLC to determine their retention time and

calibrate the system (calibration curves are presented in Appendix B).

Results and Discussion

To optimize the AFEX treatment conditions for maximum enzymatic conversion
of corn stover, the corn stover samples were treated at different process conditions (e.g.,
temperature, moisture content, ammonia loading ratio, and residence time).

The effects of each variable on the conversion of glucan to glucose and xylan to
xylose through enzymatic hydrolysis are illustrated in the following ﬁgures. All
enzymatic hydrolysis experiments were performed in duplicate. The difference within
each duplicate was from 0.2% to 6% (observed in 30 runs). In the results the glucan
conversion to glucose and xylan conversion to xylose have been calculated as follows
(equation 3.2 and 3.3):

[Glucose]+ 1.053[Cellobiose]

, X100% (3.2)
1.1 1 lfg [Biomass]

 

% Glucan Conversion =

32

% Xylan Conversion =

where:

[Glucose]
[Xylose]
[Cellobiose]

[Biomass]

fg
fr

1.111

1.136

Effects of ammonia to biomass ratio on the enzymatic hydrolysis of AFEX treated corn

stover

Figure 3.7 and Figure 3.8 show the effect of ammonia to biomass ratio (0.5:1,
0.7: 1, 1:1, and 1.3:1 g of NH3: g of dry biomass) on the subsequent enzymatic hydrolysis
of AFEX-treated corn stover at two temperatures (80 and 90°C) with different moisture
contents (40% and 60% dwb). Glucan conversion increased with increasing ammonia
loading at any temperature and moisture content and attained a maximum value at a mass
ratio of 1:1 (ammoniazdry biomass). Xylan conversion also showed the same trend as
glucan conversion in response to ammonia loading (Figure 3.9). Ammonia at this loading

ratio provided maximum overall enhancement of reactivity during pretreatment. It is

[Xylose]
1.136f, [Biomass]

 

X100% (3.3)

glucose concentration in hydrolysates (g/L)

xylose concentration in hydrolysates (g/L)

cellobiose concentration in hydrolysates (g/L)

Dry biomass concentration at the beginning of the fermentation
(g/L)

Glucan fraction in dry biomass (g/g)

Xylan fraction in dry biomass (g/ g)

Converts glucan to equivalent glucose

Converts xylan to equivalent xylose

33

known that ammonia can react with lignocellulose by ammonolysis of the ester crosslinks
of some uronic acids with the xylan units and cleaving the bond linkages between
hemicellulose and lignin (Wang et al., 1967). However it is evident from Figures 3.7, 3.8,
and 3.9 that further increases in ammonia loading decreased glucan conversion. It is
possible that extra liquid ammonia plasticizes the cellulose and thereby reduces the
disruptive effect of the pressure release (O’Conner, 1972). Based on these observations
the ammonia ratio 1:1 is the optimum ammonia loading ratio for AFEX treatment of corn
stover. Previous work also recognized the same ammonia ratio as the optimum ratio for

AFEX treatment of com ﬁber (Moniruzzaman et al., 1997).

Figure 3.7. Effects of ammonia loading on glucan conversion at 40% moisture

content (dwb).

      
    
 
   
   

Conditions:
Cellulase loading: 60 FPU/ g of glucan
l68hr of enzymatic hydrolysis

40% Moisture content (dwb)

Average of two runs

1 00
90
80
70
60
50
40
30
20
1 0

0

% Conversion of Glucan

9090, 1:1
909C, 1 3 1
8090, 0.5:1
809C, 0 7 1

8090, 1:1
80°C, 1.3:1

'- r.
I" ".
O O
o” 0'
°é 8
O O

Treatment Temperature, Ammonia loading ratio

34

% Conversion of Glucan

1 00
90
80
70
60
50

30
20
1 0

  
 

  

Conditions:
Cellulase loading: 60 FPU/ g of glucan
l68hr of enzymatic hydrolysis

60% Moisture content (dwb)

Average of two runs

 
 
    
   

90°C, 1:1
90°C, 1.3:1
80°C, 0.5:1
80°C, 0.7:1

80°C, 1:1
80°C, 1.3:1

Untreated
a-cellulose

‘- 1'.
In N.
O O
o" '
2: g
en 0

Treatment Temperature, Ammonia loading ratio

Figure 3.8. Effects of ammonia loading on glucan conversion at 60% moisture

content (dwb).

35

% Convorslon of Xylan

Figure 3.9. Effects of ammonia loading on the xylan conversion at 60% moisture
content (dwb).

Conditions:

Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

60% Moisture content (dwb), 90°C
Average of two runs

 

0.7:1 1 :1 1.3:1

Kg of ammonia: kg of dry biomass

36

Eﬁects of moisture content on the enzymatic hydrolysis of AFEX treated corn stover
Effects of different moisture contents (20%, 40% and 60% dwb) on the
subsequent enzymatic hydrolysis (after 168hr) are presented in Figure 3.10. Glucan
conversion increased with increasing moisture content at any temperature. Even though at
higher moisture contents ammonia is more diluted, but apparently the afﬁnity of
ammonia for biomass components (e.g., cellulose, hemicellulose) is still sufﬁciently
strong so that the ammonia reacts adequately with the biomass. Previous studies have
shown that the moisture in the biomass allows formation of ammonium hydroxide, which
hydrolyzed hemicellulose and thereby enhance the overall effect of AFEX treatment
(Dale et al., 1985). As these data show, the glucan conversion is still increasing with
moisture content so with this series of run we were not able to identify the optimum
moisture content for AFEX treatment of corn stover. Therefore another set of the AFEX
runs of corn stover with higher moisture contents were conducted. The results of these

runs are presented later in this chapter.

37

% Conversion of Glucan
8 8 8 8 8

  
   
   
   
   

 

Conditions:
Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

1:1 ammonia loading ratio

Average of two runs

388

O

i.

N

5

Its
ass;

Moisture content % (dwb), Temperature

40%, 80°C

iiii
§§§

Figure 3.10. Effects of moisture content on glucan conversion.

38

Figure 3.11 summarizes the effects of different moisture contents on xylan
conversion at different temperatures with 1:1 ammonia loading ratio. At a 90°C
temperature, this ﬁgure shows the same trend as we observed in Figure 3.10. The xylan
conversion was enhanced with increasing moisture content from 20, to 40, and to 60%.
As these data demonstrate, at 70 and 80°C the xylan conversion was increased with

increasing the moisture from 20 to 40% but at 60% moisture a decline was observed.

Figure 3.11. Effects of moisture content on xylan conversion

100
90 Conditions:
Cellulase loading: 60 FPU/ g of glucan

80 168hr of enzymatic hydrolysis
5'; 1:1 ammonia loading ratio
2' 70 Average of two runs
a
;> so
‘6
r: 50
.2
E 40
5
° 30
O

20

10

 

0

iii. iii iii

Moisture content % (dwb), Temperature

39

% Converelon of Glucan
o 3 8 '6’ 8 S 8 3‘ 8 8 8

Effects of temperature on the enzymatic hydrolysis of AFEX treated corn stover
The effect of temperature on glucan conversion was explored at different
ammonia loading levels and different moisture contents. The results are illustrated in

Figure 3.12, 3.13, and 3.14.

Figure 3.12. Effect of temperature on glucan conversion at 0.7:1 ammonia loading

d

Conditions:

Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

0.7:1 ammonia loading ratio

Average of two runs

 

40%, 60°C
40%, 70°C
40%, 80°C

90°C
60%, 60°C
60%, 70°C
60%, 80°C

i
i

Untreated
e-eelluloee

Moisture content % (dwb), Temperature

96 Conversion of Glucan
asses

100

3‘88

d
O O
I L

Figure 3.13. Effect of temperature on glucan conversion at 1:1 ammonia loading.

 

 

   

 

 

Conditions:
Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

1:1 ammonia loading ratio
Average of two runs

 

 

 

 

 

 

 

I i
S 2
5 i

Moisture content % (dwb), Temperature

41

Figure 3.14. Effect of temperature on glucan conversion at 1.3:1 ammonia loading

  
   
   
     

 

Conditions:
Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

1.3:1 ammonia loading ratio

Average of two runs

% Conversion of Glucan
a a s s a a a a a s '8‘

it“. ,§.=

Moisture content % (dwb), Temperature

Data in these three ﬁgures (3.12, 3.13, and 3.14) illustrate increasing temperature
at any ammonia loading level dramatically enhanced the conversion of the glucan to
glucose. Pretreatment temperature is a very important variable, it determines the amount
of ammonia vaporized during the explosive ﬂash, inﬂuences system pressure and affects
the rate of reaction occurring. At higher temperature more ammonia vapors ﬂash, thereby
greater disruption of biomass ﬁber structure might be achieved. Also higher temperature
accelerates the chemical reactions such as alkaline hydrolysis of hemicellulose. Figures
3.15, 3.16, and 3.17 show the effect of treatment temperature on the xylan conversion at
different ammonia loading levels. With one exception, xylan conversion increased as the

treatment temperature increased. At 60% moisture and 0.7:1 armnonia loading ratio the

42

xylan conversion was increased as the temperature rose to 70°C, but at higher
temperature a slight decrease in xylan conversion was observed.

Since both glucan and xylan conversions are still increasing with temperature so
we investigated higher temperature (100 and 110°C) to identify the optimum temperature

for AFEX treatment of corn stover. The results are presented later in this chapter.

Figure 3.15. Effects of temperature on xylan conversion at 0.7:1 ammonia loading.

100

Conditions:

Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

0.7:1 ammonia loading ratio

Average of two runs

Conversion of Xylan (96)
o 3 3 8 8 8 8 3‘ 8 8

 

40%, 60°C
40%, 70°C
40%, 80°C
60%, 60°C

i
i

40%, 90°C
60%, 80°C
60%, 90°C
Untreated

Moisture content % (dwb), Temperature

43

Xylan Conversion (96)

88838

Figure 3.16. Effects of temperature on xylan conversion at 1:1 arrunonia loading.

100

3‘88

—e
00

Conditions:

Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

1:1 ammonia loading ratio

Average of two runs

 

i i

Moisture content % (dwb), Temperature

40%, 70°C
40%, 60°C
90°

iii
§§

00%, 90°C
Untreated

Conversion of Xylan (96)

Figure 3.17. Effects of temperature on xylan conversion at 1.3:1 ammonia

loading.

Conditions:

Cellulase loading: 60 FPU/ g of glucan
168hr of enzymatic hydrolysis

1.3:1 ammonia loading ratio

Average of two runs

 

i

i

Moisture content % (dwb), Temperature

40% 60°C
40%, 70°C
40%, 80°C
60%. 60°C
60%, 70°C
60%, 80°C
60%, 90°C

Untreated

45

Figure 3.18 shows the hydrolysis proﬁles for selected AFEX-treated and
untreated corn stover samples at enzyme loading levels of 60 FPU/g of glucan. In all
cases, the curves that describe sugar production vs time have a similar shape. The AFEX-
treated material showed a consistently higher degree of digestibility than the untreated
sample. Figure 3.18 also shows that the initial rate of digestion of treated material was
higher than untreated. Some of the AFEX treatments approximately double (almost
triples in one case) the yield of glucose from cellulose under these conditions versus
untreated corn stover.

Figure 3.19 shows glucose production vs time for one of our best AFEX runs
(60% moisture, 90°C and 1:1 ammonia ratio), with the data normalized to the 3-day
yield. The 3-day glucose yield is a convenient measure of enzymatic susceptibility of the
AFEX treated biomass. The normalized curve allows determination of glucose yield at
times other than 3 day. Based on this ﬁgure about 96% of the 3-day glucose was released
in 24h and about 97% of glucose was released in 48 hr. These data demonstrate the high

digestibility of the AFEX treated corn stover.

46

Glucose Concentration (gll)

 

all
&

 

us
N

.5
O

 

 

 

 

 

 

 

 

 

 

 

 

k—i/i/I
, / N A +mmm
, //'\* e’ Eggs
. / // ..
, M /\°/”

48 72 96 120 144 168 192
Hydrolysis Time (h)

Figure 3.18. Glucose concentration vs hydrolysis time at enzyme loading of

60 FPU/g of glucan

47

Figure 3.19. Normalized hydrolysis proﬁle for AFEX treated corn stover (60% moisture,

1:1 ammonia loading ratio, 90°C) at enzyme loading of 60 FPU/g of glucan

 

120

 

100 c

f
/
., /
,/

0 0.5 1 1.5 2 2.5 3 3.5
Hydrolysis time (day)

0

 

 

 

Normalized reducing glucose yield
(% of 3-d yield)
8

 

 

 

 

48

Figure 3.20 shows the xylose production proﬁle for selected AFEX-treated and
untreated corn stover samples at an enzyme loading level of 60 FPU/g of glucan. In all
cases, AFEX-treated material showed a consistently higher degree of xylose production
than the untreated sample. Figure 3.20 also shows that the initial rate of xylose
production from treated material was higher than untreated. Some of the AFEX
treatments approximately quadruple the yield of xylose versus untreated stover. These

ﬁndings once again prove that the AFEX treatment has the ability to increase

hemicellulose hydrolysis.

 

12

.. /{/,

 

 

 

 

 

 

 

 

 

 

 

 

3
5 8
:8: +60%, 90°C, 1:1
‘ 6 L +40%. cone, 1:1
g \¥ +40%, eo-c, 1:1
8 N -O-Untreated 7
g 4 / —r
>.
x A

2 ‘10

o I I I I Y T T

0 24 48 72 96 120 144 168 192
Hydrolysis Time (h)

Figure 3.20. Xylose conversion vs hydrolysis time at enzyme loading of 60

FPU/g of glucan

49

AFEX treatment with higher temperatures, higher moistures, and

longer treatment times

Since in our ﬁrst trials of AFEX treatment of corn stover, both glucan and xylan
conversion were increasing with temperature and moisture content, we were not able to
top out on the treatment temperature and the moisture content. Therefore some AFEX
runs with higher temperature (100 and 110°C), higher moisture (70 and 80% dry weight
basis) and longer treatment times (10 and 15 min) all at 1:1 ammonia loading ratio were
conducted to establish the optimum conditions for AFEX treatment of corn stover. The

AFEX treatment conditions are summarized in Table3.5.

Table 3.5. AFEX treatment conditions in corn stover treatment.

 

 

 

 

 

 

 

 

Treatment time (min) 60% Moisture" 70% Moisture 80% Moisture
5 T=90°C T=90°C T=90°C
T=100°C T=100°C T=100°C
T=110°C T=110°C T=110°C
10 =90°C =90°C T=90°C
T=100°C T=100°C T=100°C
T=110°C T=110°C T=110°C
15 T=90°C T=90°C =90°C
T=100°C T=100°C T=100°C
T=110°C T=110°C T=110°C

 

* Ammonia loading ratio of 1:1 was used in all the runs.
** Moisture content was calculated on dry weight basis.

50

 

The AFEX treatment and the enzymatic hydrolysis were conducted as described
before. The enzyme loading level of 15 FPU/g of glucan was used in the following
enzymatic hydrolysis. The hydrolysis results are presented in following ﬁgures.

Figure 3.21 compares the effects of the different moisture contents on the glucan
and xylan conversion, at 90°C treatment temperature and 1:1 ammonia loading ratio. As
this ﬁgure shows further increases in moisture content after 60% (70 and 80% dry weight
basis) did not improve either glucan or xylan conversion. Dilution of ammonia at higher
moisture content may reduce the afﬁnity of ammonia for biomass components (e.g.,
cellulose, hemicellulose). Figure 3.22 shows the effects of different moisture content on
both glucan and xylan conversion at 100 and 110°C with 1:1 ammonia loading ratio. This
ﬁgure also shows the same trend as we observed in Figure 3.21. Based on these data we
selected 60% dry weight basis moisture content as the optimum moisture for AFEX

treatment of corn stover.

51

Figure 3.21. Effects of moisture content on glucan and xylan conversion, all runs were

kept at the set temperature for 5 min.

 

Conditions:

1:1 ammonia loading ratio
90°C, 15 FPU/g of glucan
168hr hydrolysis

Average of two runs

 

 

 

 

 

     
  
        

   
  

 

 

 

 

     
   

 

 

100
II] Glucan conversion
90 * EX an conversion
V
c 80
2
>1
x 70
1: 2
= a
W 60 :
c :
s 5,, , - - ~52?
_ e I
3 r3322 1 13%: '33???
0 Maw. Villa ,0»; ”0’1
l ”at. In». pm ”we
- 40 - . we». rum ,0»; III!‘
e I ”a”. In». ”m ”we
we». rum i ”m men
I: _ m0. gun. ”m ”In.
a are "as: 453:: as: are
- 30 4 um I we». run. we” ”In
a ”I” we”. we». pm ”I”.
III” I M View (”a ”I”
o ”I” we». '0”. [I’ll ”I”.
> 20 ”m ”a”. we». a»; mu
= : w: I Is": 399:: are: an:
o ”m l we». term w»; m».
0 am . re M I” 4 I” up».
am we». we». a»; new
10 — a»; rate. run; we» ”a”.
”m , ”a. ”m a» mm
at» we». rum (”’4 new
are l are we M” are
0 g Illll 'llll: ’IIII: we”; I’ll}:

20% 40% 60% 70% 60%

Moisture content % (dwb)

52

Figure 3.22. Effects of moisture content on glucan and xylan conversion, all runs were

 

    

 

 

 

 

 

 

 

 

k tat the set tem erature for 5 min. , _
ep p Conditions:
1:1 ammonia loading ratio
15 FPU/g of glucan
168hr hydrolysis
lIDGlucan conversion Average of two runs
tax Ian conversion
100 : H
r:
i: 90 -
E
E. 80 -
2 so - £355 355% ”"
”a ,0»
3 ea- 5%: a
50 - II”: we;
3 5559- $935 ”"
"a: w”
9 4o - an as: 59':
o tar-- as:
c 30 — ”$5: 55555 i’”‘
2 5555:- 5555‘ v”
M»: ”a
B 20 - me. gem
0 I’ll- Illl III
> W:- I
8 10 ~ are:
. ,...- .
o g .(Q’;
9 .9 9 9 9 9
O O O O O O
O O O I- 1- I-
!- 1- s- 1- 1- I-
g .\° as" g :5 ai
'9 I? 8 '9 I? 8

Moisture content % (dwb). Treatment temperature

53

Figure 3.23. Effects of temperature on glucan and xylan conversion; all runs were kept at

the set temperature for 5 min.

 

       
  

 

Conditions:

1:1 ammonia loading ratio
80 -_ 15 FPU/g ofglucan

168hr hydrolysis

Average of two runs

 

  
 

 

 

sss'ssl
seal

     

;

 

or»

.‘ssssl
““‘
\\\\\
at»:

s
\
2

z

$§§S§$§SSS§S§S$§$§SS

\\\.\.\.\.\.\.\\\.\

111 Glucan Conversion
EX an Conversion

§§§

RR

§§§§§§§s

ssssssssssssssssssss§

teats“

Conversion of Glucan and Xylan ('/o)
8

\\'""

s§§

ssss‘
\\\\\
sss§§§§§

\\\\\\\\\\\\\\\\\\

'\'\'\'\'\'\'\'\'\'\'\'\'\'\'\'\'\'\‘\'\‘\'\’\'§\'\\'\'\\‘\"
s

a

   

\

 

 

$555 :I .
$553 :0 '
55$: 3%:

60°C 70°C 80°C 90°C 1 00°C 1 1 0°C

.\

     
  

Treatment Temperature

As Figure 3.23 shows increasing temperature from 90° to 100 and 110°C
improved the conversion of glucan to glucose only by about ~ 2-3%. Apparently ﬁrrther
increases in treatment temperature do not have much additional beneﬁcial effect. The
ultimate goal of the AFEX treatment is to increase the ethanol yield of the fermentation
process by increasing the digestibility of the biomass. Therefore some of the above runs
that showed higher glucan and xylan conversion were chosen for further SSF analysis.
The best treatment temperature was selected based on the fermentation results, not just

enzymatic hydrolysis results.

54

Simultaneous Saccharification and Fermentation (SSF)

Following materials were used:
Untreated corn stover
AFEX treated corn stover
Cellulase enzyme (provided by NREL, CAS 9012-548, activity: 28 FPU/ml )
B-glucosidase (Novozym 188) from Sigma (St. Louis, MO)
Saccharomyces cerevisiae D5A (provided by NREL)
Method
SSF experiments were conducted according to NREL standard protocol (LAP-008).
Preparation of inoculum

A ﬂask containing YPD (yeast extract, pepton, dextrose) was inoculated with one
stock vial of Saccharomyces cerevisiae D5A (provided by NREL) and incubated for 14hr
in a rotary incubator shaker at 45°C and 130 rpm. Based on optical density (CD. at
600nm) of the inoculum the amount of culture needed to inoculate the SSF ﬂask for a
starting CD. of 0.5 was calculated.
Experimental procedure

The pretreated corn stover along with the untreated sample was washed and
squeezed through miracloth. The wet samples were kept in sealed plastic bags in
refrigerator until next day for the SSF experiment. Total solids of the washed samples
were determined according to NREL protocol (LAP-001).

All SSF experiments were performed in duplicate and all appropriate controls
(SSF ﬂasks without any substrate or with Ot-cellulose as the substrate) were included in

the SSF experiment. The difference within each duplicate was from 2% to 7% (observed

55

in 20 runs). Each SSF ﬂask was loaded with 6% w/w glucan, 1% w/v yeast extract, 2%
w/v peptone, 0.05 M citrate buffer (pH 4.8), an appropriate amount of cellulase enzyme
for 15 FPU/g of glucan and an appropriate amount of D5A inoculum (starting 0D. 0.5).
The SSF ﬂasks were equipped with water traps to maintain anaerobic conditions. The
SSF ﬂasks were incubated at 45°C with gentle rotation (130 rpm) for a period of 168hr.
At each predetermined time interval (0, 3, 6, 24, 48, 72, 96 and 168hr) a 2m]
aliquot was removed aseptically. The samples were centrifuged and the supematants were
ﬁltered and subjected to sugars analysis using HPLC and to ethanol analysis using gas
chromatography (GC). At the last time point, a sample from each SSF ﬂask was streaked
on an YPD plate to check for contamination. In this series of SSF experiments no
contamination was observed. After 168 hr the physical appearance of SSF solid residues
of the AFEX-treated corn stover samples was very different compared to that of untreated
corn stover sample. After 168hr of SSF the AFEX-treated samples were mostly
solubilized and had lost their rigid structure, while there still was considerable rigid solid

in the untreated sample. These differences are illustrated in Figure 3.24.

56

Figure 3.24. Picture of SSF ﬂasks aﬁer 168hr of fermentation.

 

 

Untreated corn stover AF EX—treated corn stover

57

Gas Chromatography (GC)

The fermentation samples were analyzed for ethanol by gas chromatography using model
GC 17 (Shimadzu). The injection temperature was 240°C and the detector temperature
was 255°C. The column was ﬁrst maintained at 80 °C up to 3 min followed by a
temperature program at 15°C/min up to 125 °C for 6 min. The carrier gas was helium and
ethanol was used as external standard for calibration. The calibration curve is presented
in Appendix B.

The fermentation samples were also analyzed for sugars by HPLC. The results obtained
from GC and HPLC are presented in following ﬁgures.

Results and Discussion

The % theoretical ethanol yield or % cellulose (glucan) conversion was calculated
using equation 3.4:

[Ethanol ] f — [Ethanol I,
0.5 1( f *[Biornass]* 1.1 11)

 

% Cellulose Conversion = X100% (3.4)

where:

[ Ethanol If Ethanol concentration at the end of the fermentation (g/L) minus any
ethanol produced from the enzyme and medium

[ Ethanol Io Ethanol concentration at the beginning of the fermentation (g/L) which
should be zero

[Biomass] Dry biomass concentration at the beginning Of the fermentation (g/L)

f Cellulose fraction of dry biomass (g/ g)

0.51 Conversion factor for glucose to ethanol based on stoichiometric

biochemistry of yeast.

58

1.111 Converts cellulose to equivalent glucose

As Figure 3.25 shows, throughout the SSF process, glucose produced by the cellulase
enzyme was almost completely consumed by the yeast and converted to ethanol in all the
selected runs (the high concentration of glucose at 0hr is due to YPD medium in the SSF
ﬂask). On the other hand, the concentration of xylose produced consistently increased
throughout the SSF process (Figure 3.26). Our Saccharomyces cerevisiae D5A yeast does
not have the ability to utilize xylose and to convert it to ethanol. The data in Figure 3.25
also show similar results as we observed in our enzymatic hydrolysis, earlier in this
chapter. These data show that in some cases the AFEX treatment approximately

quadruples the yield of xylose versus untreated corn stover.

1L

 

.e
N

 

.5

 

 

9
on

+60%, 90°C, 5min
43-60%, 100°C, 5min
-)(— untreated

 

 

 

P
or

 

 

Glucose Concentration (gll)
.°
&

 

P
to

 

 

\l

 

 

O

20 40 60 80 100 1 20
SSF Time (h)

O

Figure 3.25. SSF proﬁle for glucose concentration

59

 

 

i

 

 

b

 

 

 

 

 

 

 

 

 

 

 

3
C
.9 —
ﬂ
8
‘5 y// -°-60%, 90°C,1:1,5mln
8 3 *6096,100°C,1:1,5mln
g +untreated
0
q, 2
U}
2 ‘—/\{
>.
X 1 V
o I I W I I
0 20 40 60 80 100 120
SSF Time (h)

Figure 3.26. SSF proﬁle for xylose concentration

Figure 3.26. shows the time proﬁle of ethanol production during the SSF hydrolysis of
AFEX-treated and untreated corn stover. The AFEX-treated samples consistently
produced higher amount of ethanol throughout the entire course of the SSF process. The
rate of ethanol production was quite rapid during the ﬁrst 6 hr of the fermentation. All of
the samples attained the maximum amount of ethanol after 96hr. As seen in Figure 3.27,
run treated at 90°C, 1:1 ammonia loading ratio, and 60% moisture content produced more
than twice as much as ethanol compared to the untreated sample. The data presented in
Figure 3.28 also shows the AFEX treatment under these conditions had the highest

ethanol yield as a percent of theoretical.

Even though the enzymatic hydrolysis of sample treated at 90°C, 1:1 ammonia
loading ratio, and 60% moisture content showed lower glucan conversion compared to
sample treated at 100°C, 1:1 ammonia loading ratio, and 60% moisture content, the
results presented in ﬁgure 3.27and 3.28 indicate that this run produced higher amount of
ethanol. The higher temperature may have produced some inhibitory material that
affected the performance of the yeast and reduced SSF productivity. Since the main
objective is higher production of ethanol, we selected 90°C as the best treatment

temperature for AFEX treatment of corn stover.

Figure 3.27. Ethanol concentration vs. SSF time

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1e
14 "

r:

3. -~ ~l

: 12

.2 ~III 4

“

a 10

:.-.

8

o a

C

8 6- 1g 8

'6

5 4 M

5 +60%, 90°C, 1:1, 5min

III 2 *untreated
1 +60%, 100°C,1:1,5mln
0' I I I j l l I
024487296120144168192

SSFtime(h)

61

% Theoretical Yield of Ethanol

Figure 3.28. Yield of ethanol as percent of theoretical after 96 hr of SSF

 

3????

 

O

 

 

 

 

 

 

i
I:

60%, 90°C,
5min
60%, 100°C,
5min
10min
untreated

Moisture content % (dwb), Temperature, residence time

62

 

Solubilization of corn stover during enzymatic hydrolysis

To investigate the solubilization of the AFEX treated and untreated corn stover, a
series of the enzymatic hydrolysis vials were set up, using the selected AFEX treated
samples and untreated sample with an enzyme loading of 60 FPU/g of glucan (as
described above). The vials were incubated at 50°C with gentle rotation (75rpm) for a
period of 168hr. No sample was taken during the entire period of 168hr. At the end of
168hr of hydrolysis the solid residues were collected by squeezing the samples through
miracloth. The moisture content of the collected solid was calculated as described before.

The % of solubilization was calculated according to equation (3.5).

 

% Solubilization = 1-[ ma] 801” aft“ 1681" (g) ] (3.5)

Total Solid at the begining (g)

The AFEX treated samples were solubilized up to 66% (condition of sample: 60%
moisture, 1:1 ammonia loading ratio, and 90°C) while the untreated sample was
solubilized only 45%. The different amount of solubilization can easily be noticed by the
appearance of the enzymatic hydrolysis vials after 168hr of hydrolysis. These differences

are visible in Figure 3.29.

63

Figure 3.29. Pictures of AFEX-treated and untreated samples after treatment with 60

FPU/g of glucan; a) At 0 hr; b) Aﬁer 168 hr of hydrolysis

8)

 

H—J Untreated
corn stover

AF EX treated corn stover

1))

 

Untreated

H‘J corn stover

AFEX treated corn stover

64

Different cellulase enzyme loading

The cost of enzyme used for sacchariﬁcation of cellulosic residues is dominant in
the overall bioconversion process. One way to decrease this cost is to use lower amount
of enzyme per kg of biomass in enzymatic hydrolysis. Therefore as part of our study we
performed enzymatic hydrolysis on some of the AFEX-treated corn stover with three
different cellulase loadings (7.5, 15, and 60 FPU/g of glucan). As Figure 3.30A
demonstrates increasing cellulase loading from 15 to 60 FPU/g of glucan, did not make
much difference in glucan or xylan conversion of AFEX-treated samples. These data also
show that at higher temperature these differences are diminished, for example at 90°C,
1:1 ammonia loading, and 60% moisture content both 15 and 60 FPU result in almost the
same amount of glucan and xylan conversion. Figure 3303 presents the results of
enzymatic hydrolysis of AFEX-treated corn stover at enzyme loading of 7.5 FPU/g of
glucan compared with 15 FPU/g of glucan. These data show that the AFEX treatment
makes possible over 88% conversion of glucan and xylan to glucose and xylose at

enzyme loading as low as 7.5 FPU/g of glucan (equal to 2.75 FPU/ g of dry biomass).

65

Figure 3.30. Conversion of glucan and xylan vs treatment temperature A) at enzyme

loading 15 and 60 F PU/g of glucan; B) at enzyme loading 7.5 and 15 FPU/g of glucan

A)

%Conversion of Glucan or Xylan

 

120
Conditions:

5 min AFEX treatment
100 168 hr hydrolysis

l:| Ammonia: Biomass
60 % Moisture content

 

   

  

llll Glucan Conversion (a) 15 FPU
Glucan Conversion (it) 60 F PU
a Xylan Conversion Vii l5 FPIl W
E Xylan Conversion a 60 l’l’ll

 

 

 

 

   

 

   

 

 

 

 

 

 

 

3)
§ 120
S
3, 100
><
'6
c 80
G
U
.3
0 60
I.-
o
E 40‘
2
9
c 20‘
o
o

0

 

 

   

80
60
’////////A

40
20 ‘

0

60 60 70 70 80 80 90 90
Temperature (”C)
i fonamons; IGlucan Conversion @ 15 FPU
5 min AFEX treatment @ 90°C ’ﬁ‘ ”Gluca" Convemion @ 7-5 FPU
168 hrs hydrolysis g Xylan Conversion @ 15 FPU
1:1 Ammonia: Biomass E Xylan ConverSion @ 7-5 FPU
I i7/////, W
60 60 70 70 80 80

Moisture Content (%)

66

Effects of longer AFEX treatment time

In an effort to explore the effects of longer treatment time on the hydrolysis
results, a series of AFEX runs (all at 1:1 ammonia loading ratio) were conducted; the
system was kept at the target temperature for 5, 10 or 15 min. The conditions of these
runs and the effect of increasing the AFEX treatment time from 5 to 10 minutes or from 5
to 15min in glucan or xylan conversion are summarized in Tables 3.6 and 3.7 As these
tables show increasing the treatment time in some cases had a positive effect and in some
cases a negative effect on the conversion of glucan and xylan. Since these effects were
very different in each set of AFEX runs, we were not able to ﬁnd any correlation among
the treatment time and other conditions of the AFEX runs. Therefore to be able to choose
the best treatment time, we selected some of the runs, which showed greater glucan and
xylan conversion with increasing time, for further SSF experiments. The results are
shown in Figure 3.31 As seen in this ﬁgure all the runs (with one exception) with longer
treatment time showed lower ethanol yield. However, the hydrolysis results of these runs
all showed greater glucan and xylan conversion (Table 3.6 and 3.7). Perhaps during
longer treatment times some inhibitory materials were produced which in turn reduced
the yield of the fermentation. Based on these ﬁndings we selected 5 minutes as the best

treatment time for the AFEX treatment of corn stover.

67

Table 3.6 Effect of increasing the AFEX treatment time from 5 to 10 minutes in glucose

and xylose conversion (at 15 FPU, after 168hr).

 

Temperature

90°C

100°C

110°C

 

Sugar

Glucose
Conversion

Xylose

Conversion

Glucose
Conversion

Xylose

Conversion

Glucose
Conversion
Xylose
Conversion

 

60 %

-6.9%
-4 %
4.14 %

7.8 %

68

Moisture Content
70 %
3.2%
4.97%
‘ 155%
14.5 %
’26 %

25.8 %

 

Table 3.7 Effect of increasing the AFEX treatment time from 5 to 15 minutes in

glucose and xylose conversion (at 15 FPU, after 168hr).

 

 

Moisture Content W

Temperature Sugar

.. 60% . 70% 8.6 .%. ._.!

90°C

100°C

110°C

Glucose
Conversion

Xylose
Conversion

Glucose
Xylose
Conversion

Gin...

Conversion

Xylose
Conversion

Conversion ‘

7.5 %

21.6%

19.2% W

8.7 %

16%

11.9%

.13. % .- .. ... - I . . .

0.0 %

266%. - ..

12.5 %

“I‘m.”

-4.7 % f

-39.. % . - .

12.4 %

6.9 a

16.8 %

17.6 «7.

 

 

 

69

% Ethanol yield

Figure 3.31. SSF results of some of the AFEX runs with longer residence time.

 

 

100

 

 

 

60..

40‘

20-

 

60%, 90%, 5min
60%, 90°C, 10mln
70%, 90°C, 5min
70%, 90°C. 10mln
60%. 100°C, 5min
60%, 100°C, 10min
70%, 100°C, 5min
70%, 100°C, 10min

Moisture Content, Treatment Temperature, Residence Time

70

 

Conclusion

Our major ﬁndings and conclusions are as follows:

Our AF EX unit works essentially as designed and gives results within the
expected ranges based on previous studies.

Even without mechanical mixing, ammonia distribution appears to be uniform in
this 300 ml reactor, but there is substantial non-isothermal behavior, which is hard
to avoid in a lab unit.

Optimum conditions are quite broad. However, highest glucan and xylan
conversion and ethanol yield from AFEX-treated corn stover were achieved at:
1:1 kg of ammonia / kg of dry biomass, moisture content 60% (dwb), temperature
90°C, and residence time 5 min.

It appears possible, at least within some limits, to achieve similar hydrolysis
results by increasing temperature while reducing ammonia levels, that is, we can
“trade off’ these two parameters.

Increasing temperature and moisture content enhance the AFEX treatment.
Enzymatic hydrolysis of the com stover treated under optimal AF EX conditions
showed almost 97% glucan conversion and 68% xylan conversion versus 29 %
and 16% for untreated corn stover respectively (with enzyme loading of 60 FPU/
g of glucan).

Lowering the cellulase loading from 60 FPU/ g of glucan to 7.5 or15 F PU/ g of
glucan did not make appreciable differences in glucan or xylan conversion of
AFEX treated corn stover. These results are very important in terms of process

economics

71

o AFEX treatment doubled the amount of ethanol production compared to
untreated corn stover in the SSF process (S. cerevisiae strain that we used in our

SSF process does not utilize xylose).

Future Work

The AFEX treatment, unlike high temperature or acidic treatment produces
signiﬁcant amount of xylose. Existing cellulase mixtures have been developed to
hydrolyze these acid/high temperature-treated lignocellulosic material and are probably
not optimal for AFEX-treated materials. Therefore supplementing the enzymatic
hydrolysis system with xylanase might help to achieve maximum hydrolysis of plant cell
wall polymers.

Enzymatic hydrolysis of AFEX-treated material, produces large amount of xylose
therefore, using microorganisms capable of utilizing xylose as well as glucose in the SSF
process could increase ethanol production yield.

The cost of enzyme used for sacchariﬁcation of lignocellulosic biomass is
dominant in the overall bioconversion process; therefore, using lower cellulase (less than

7.5 FPU/g of glucan) deserves attention.

72

Chapter 4

Constructing Plasmid DNA containing Cellulase Genes

&
Effects of AFEX on the activity of the plant-produced heterologous

cellulase

73

Abstract

A critical parameter affecting the economic feasibility of lignocellulosic
bioconversion is the production of inexpensive and highly active cellulases in bulk
quantity. A promising approach to reduce enzyme costs is to genetically transform plants
with the genes of these enzymes, thereby producing the desired enzymes in plants
themselves. Extraction and recovery of active proteins or releasing active cellulase from
the plants during bioconversion could have a signiﬁcant positive impact on overall
lignocellulose conversion economics.

In an effort to produce cellulases in transgenic corn plant, a research group from
the Crop and Soil Sciences Department (at Michigan State University) and I constructed
several different plasmid DNAs to target these enzymes to different compartments
(cytosol or chloroplasts) of transgenic corn plants.

We have also investigated the effects of AFEX pretreatment, employing a range
of treatment temperatures, moisture contents and ammonia loadings on the activity of
plant-produced heterologous cellulases. The plant materials included transgenic tobacco
plants expressing E1 (endoglucanase from Acidothermus cellulolylicus). The El activity
was measured in untreated and AF EX-treated tobacco leaves to investigate the effects of

the treatment on the activity of this enzyme.

74

Literature Review

Enzymatic degradation of cellulose to glucose is generally accomplished by the
synergistic action of three distinct classes of cellulases (Wilke et al., 1983):

o Endo-1,4-B-glucanases (EC 3.2.1.4), which act randomly on the interior of the
cellulose chain to generate new chain ends.

0 Exo-1,4-B-D-glucanases or 1,4-B-D-glucan cellobiohydrolases (EC 3.2.1.91),
which act on the nonreducing ends of the cellulose chain and liberate D-
cellobiose from 1,4-B-glucans.

o B-D-glucosidases (EC 3.2.1.21), which act to release D-glucose units from
cellobiose.

Many naturally occurring cellulose-degrading fungi and bacteria have been
reported so far. More than 60 cellulase-producing fungi, representing the soﬁ-rot, brown-
rot and white-rot (Robson and Chambliss, 1989) and 46 unique bacterial producers of
cellulases have been reported (Coughland and Ljungdahl, 1988). The most studied and
well characterized cellulosic fungus is the ﬁlamentous soft rot fungus Trichoderma reesei
which is also the most widely used in industrial cellulase system (Mandels and Stemberg.
1976). Another well characterized and useful cellulase source is Acidolhermus
celIqulyticus (Mandels and Stemberg, 1976).

The enzyme production step is important. Preliminary economic analysis shows
that about 6% of the ﬁnal ethanol cost in a simultaneous sacchariﬁcation fermentation
(SSF) process and 43.4% of ﬁnal costs in a separate hydrolysis and fermentation (SHF)
process can be attributed to enzyme costs (Hinman et al., 1992). Therefore any effort to

reduce cellulase costs can have a positive impact on the ﬁnal cost of bioethanol. One way

75

10 161

try I:

cry:
of 1
em
In
co
g1

dc-

5}

to reduce this cost would be to ﬁnd the most effective combination of cellulases and then
try to produce these enzymes in transgenic plants in a large scale.

It is well established that cellulase is a multicomponent enzyme complex and that
crystalline cellulose is hydrolyzed by synergism of cellulase components. The synergism
of these enzymes has been investigated by many researchers. Different combinations of
endoglucanases and exoglucanases isolated from heterologous systems show different
levels of synergy (Thomas et al., 1995). Therefore it is important to determine the best
combination that produces the greatest synergistic effect and offers the highest amount of
glucose production per unit of time at equivalent enzyme loading. The synergetic effect is
deﬁned as the ratio of glucose released when both enzymes are used versus the sum using

equivalent amount of enzymes in separate individual reaction.

Synergism Ratio == (glucose ,ndmmy (glucosecndo+ glucoseexo)

A ratio greater than 1 means a synergistic effect between the two enzymes. A
values around 1 indicates no synergistic effect and a ratio less than 1 indicates some
degree of interference between the two enzymes. Baker et al., (1994) tested the
synergistic effect of 16 different endo/exo binary pairs. Table 4.1 shows the four best
synergistic pairs of the tested pairs. These results show that the most synergistic pair is
the combination of cellobiohydrolase, CBHI, from T. reesei and endoglucanase, E1, from
A. cellulolyticus. This observation highlights the potential value of using these enzymes

in future recombinant systems and using the genes of these enzymes to transform plants.

76

Table 4.1. Best Synergistic Endo/Exo Cellulase Pairs Tested.

 

Endo/Exo Pair Synergism Ratio

 

A. cellulolyticus E1 & 2.74

T. reesei CBHI

 

T. fusca E5 & 2.61

T. reesei CBHI

 

M. bispora rEndo A & 2.04

T. reesei CBHI

 

T. neapolitana rEndo B & 1.90
T. reesei CBHI
*Adopted from Baker et al. (1994)

 

 

 

 

The therrnotolerant endoglucanase, E1, which is secreted from A. cellulolyticus,
demonstrates temperature activity characteristics that are useful for high temperature
sacchariﬁcation and, like other bacterial cellulases, shows a high speciﬁc activity. The El
enzyme has three domains a cellulose binding domain, a linker region and a catalytic
domain. Prior studies have shown that the catalytic domain by itself is sufﬁcient for
efﬁcient degradation of cellulose in vitro (Ziegelhoffer et al., 2001).

Testing the effect of truncation of E1 enzyme (E1 with only catalytic domain) on
the protein expression level, a research group from University of Wisconsin transformed
tobacco plants with both truncated El (Elcd) enzyme and the whole El enzyme, and as
the results show (Table 4.2) the expression level of Elcd was much higher than the

expression level of El (Ziegelhoffer et al., 2001).

77

Table 4.2. Maximum observed expression levels’ (percent of total soluble

protein) for Eland Elcd enzyme in transgenic tobacco.

 

Compartment El Elcd

 

Cytosol 0.00065 0.014

 

Chloroplast 0.007 0.07

 

Apoplast 0.33 0.58

 

 

 

 

 

*Adopted from Ziegelhoffer et al., 2001

There are many unfavorable codons (with respect to plants) in the linker region. It
contains frequent proline residues, which are encoded by rare (in plant) CCG codons. In
Elcd expression constructs a stop codon was introduced at the end of the catalytic
domain, effectively removing the linker and the cellulose binding domains. This research
group made series of plasmids to target E1 and Elcd enzymes to different cellular
compartments of plant cells such as the apoplast, chloroplast, and cytosol. Their results
showed that the expression level of E1 or Elcd in apoplast was much higher compared to
the expression level of E1 in cytosol and chloroplast. These results are summarized in
Table 4.2 (Ziegelhoffer et al., 2001). In this study, the accumulation of E1 or Elcd did
not affect normal growth of the plants

Expression of recombinant proteins in plant cells is dependent upon many factors
such as transcriptional, posttranscriptional, and posttranslational factors (Dai et al., 2000).
Besides factors inﬂuencing transcription and translation efﬁciency, recombinant protein
accumulation as well as stability strongly depend on the compartment of the plant cell

chosen for expression (Conard and Fiedler 1998). As mentioned above, Ziegelhoffer et

78

et al., 2001 showed that for El enzyme, subcellular targeting had signiﬁcant influence on
the expression level of this enzyme in tobacco plants. The abundance of mRNA in this
study was very similar in tobacco plants transformed with either cytosol, chloroplast or
apoplast constructs, therefore, the barrier to higher expression in the cytosol and
chloroplast is post transcriptional. There is evidence that there are distinct molecular
chaperon systems in targeted compartments to translocate or fold proteins (Boston et al.,
1996). The action of the unfolded protein with these molecular chaperon systems would
yield different rates of protein folding. Therefore targeting the E1 to the apoplast is
thought to promote correct protein folding leading to higher stability and accumulation.
In addition to these factors the pH of each compartment also can play a very important
role in the stability of enzyme. Each compartment of the plant cell has a different pH.
therefore targeting the enzyme to the compartment with closer match to its favorable pH
can improve the stability of the enzyme. As Figure 4.1 shows the apoplast pH is
approximately 5.5 which is very similar to the ideal pH for E1, whereas the cytosol pH is
7.5 and chloroplast pH is 7 at daytime and 8 at nighttime (Sander and Bethke, 2000).
These facts can indicate that the barrier to higher expression in cytosol and chloroplast

might be because of their pH, which is much higher than the ideal pH for E1cd.

79

Figure 4.1. Different compartments in plant cell (adobted from Sander and Bethke,

2000). _ . Cytolsol
ﬁucleus l lMltochondnon l pH=7.5

I n2

 

 

 

 

 

  

 

Apoplast
pH~5.5

 

 

 

 

 

Chloroplast
Daytime pH=7 Vacuole
Nighttime pH=8 pH—5.5

 

 

 

 

 

 

Dai et a1. (1999) have also successfully transformed tobacco plants with T. reesei
exo—cellobiohydrolase I (cbhl) gene. In this study the accumulation of CBHI did not
affect normal growth of the transgenic plants.

These successes have shown the feasibility of cellulase production in plants and
are the important ﬁrst steps in the development of crop plants as production system for

cellulases.

80

 

Introduction

In ethanol production enzymatic hydrolysis of cellulose to glucose is a very
attractive route, because nearly theoretical yields of glucose are possible (Wyman, 1999;
Wyman, 1995). Cellulases are currently produced from microorganisms by expensive
large-scale fermentation. The cost of enzyme-based processes has been reduced by about
a factor of four (Wyman, 1999), and additional opportunities have been identiﬁed that
may improve the technology (Lynd et al., 1996). However, the costs of production of
enzymes through microbial systems are still very high and tend to dominate the
economics of enzyme-based bioconversion processes. These costs might conceivably be
reduced by genetically transforming plants with cellulase genes to produce the desired
enzymes, and perhaps even releasing active cellulases from the plants during
bioconversion.

Transgenic plants are an attractive and cost-effective alternative to microbial
systems for production of biomolecules (Goddijn, 1995). Advances in biotechnology are
enabling plants to be exploited as bioreactors for the production of proteins,
carbohydrates (Kidd and Devorak, 1994; Stark et al., 1992), lipids (Graybum et al., 1992;
Poirier et al., 1995), and industrial enzymes (Austin et al., 1994; Pen et al., 1992; Pen et
al., 1993) in bulk quantities with minimal inputs of raw materials and energy. As this
technology continues to grow and improve the production levels of biomolecules in
plants, the development of downstream processing technology to extract and to recover
these biochemicals will increasingly determine progress in this area.

With an ultimate goal of production of cellulases in transgenic corn plants, several

different DNA plasmids were made through our collaboration with a research group from

81

the Crop and Soil Sciences department of Michigan State University. These plasmids
were designed to target the cellulases to cytosol, chloroplast, or apoplast of transgenic
corn plants.

Since A. cellulolyticus endoglucanase (El) and T. reesei exo-cellobiohydrolase I
(CBHI) demonstrate a high synergistic activity, this pair seemed a good choice for initial
studies of the effect of recombinant cellulase expression in plants. To test the effect of
sub-cellular targeting on the accumulation of CBHI enzyme, cth was fused to rbc‘S for
chloroplast-targeting, and for cytosol targeting the sequence encoding leader peptides of
(:th was removed. In all plasmids the expression of cth was driven by the rice rch
promoter and terminated with either pin3’ or nosB’ region (Fig. 4.2). The plasmid

structures and their features are summarized in Table 4.3.

Table 4.3. CBHI plasmid features

 

Plasmid Construct Plasmid features

 

PSMF13 rch-cth—pin3’ rch leaf-speciﬁc promoter driving
cellulase cDNA of T. reesei.

Expressing CBHI in cytosol.

 

pSMF 14 rch-TP-cth-pin3’ The rch TP targets cellulase of T.

reesei into maize chloroplasts.

 

 

pSMFlS rch-TP—syn-cth-pin3’ The rch TP targets modiﬁed
cellulase of T. reesei into maize

chloroplasts.

 

 

 

 

82

Figure 4.2. CBHI plasmid construct

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pSMFl3
_| mg -l cth -1 pin 3’
pSK
2.1 kb 1.7 kb 1 kb
pSMFl4
—-| rch l—Eﬁ cth '— Pin 3’
2.1 kb. 0.14 kb 1.7 kb 1 kb
pSMFIS
——-| 1'ch Syn-cth l— nos 3’ l..—
’ * ’ pB1221
2.1 kb 0.14 kb 1.5 kb 0.28 kb

Abbreviations in consutructs:

rch = Full-length rice rch promoter region.

cbhl= Cellulase cDNA isolated from the T. reesei.

Syn-ebb]: Synthetic cth gene.

pin 3’ = Potato protease inhibitor 11 gene 3’ non-coding region.

nos 3’ = 3’ region from Agrobacterium nopaline synthase gene.

TP = rice rch transit peptide.

83

Plant proteins and enzymes produced in transgenic crops have very high value;
therefore recovering and utilizing valuable bioactive plant proteins and enzymes in an
overall process for producing fuels and chemicals from biomass might improve the
economics of lignocelulosic conversion.

To process cellulase-containing transgenic plants, the following technical options
can be envisioned:

1) Harvest the cellulase-containing transgenic plants while they are green and grind
them in a suitable buffer to produce an enzyme concentrate, which can later be
used for the enzymatic hydrolysis of pretreated biomass. Depending on the
production level of this enzyme in transgenic plants, the need for externally added
cellulases in the enzymatic hydrolysis step might be avoided or minimized. The
debris also can be combined with other pretreated biomass to increase the
production of fermentable sugars.

2) Harvest the cellulase-containing transgenic plants at the end of the growing
season (dry), grind this material to release enzyme and then combine them with
pretreated biomass for ethanol production.

3) Harvest the cellulase-containing transgenic plants at the end of the growing
season, treat them with a biomass pretreatment technique to rupture cells to
facilitate the release of enzyme, and then combine with additional pretreated
biomass for enzymatic hydrolysis.

The success of any of these approaches depends on having a sufﬁciently high level of
cellulase in the transgenic plant to make hydrolysis effective. Each of these options might

ﬁnd use under particular circumstances.

84

A high yield recovery of plant proteins and enzymes from biomass depends upon
extensive cell maceration; the more cell wall that is disrupted the more protein will be
recovered (Carroad et al., 1981). Therefore a pretreatment that disrupts plant cells can be
useful in protein recovery process.

An integrated pretreatment that improves the protein recovery and increases the
conversion of cellulose and hemicellulose to fermentable sugars may signiﬁcantly
enhance the biomass process. Many pretreatments that increase the conversion of
cellulose to fermentable sugars operate under harsh conditions that tend to degrade the
sugars, proteins, and enzymes. However, the data presented in Appendix A have shown
that under limited treatment conditions, the AF EX process not only increases the
conversion of cellulose and hemicellulose to simple sugars, it also allows the recovery of
plant proteins in their native functional form.

In this study we have explored the third option that was mentioned above, by
investigating the potential of using AFEX treatment as an integrated pretreatment to
release active cellulase from a transgenic plants while it increases the digestibility of the

biomass.

85

Plasmid construction

Material and Methods

Molecular biology techniques were essentially those of Sambrook et al., 1989 or
those of commercial suppliers of enzymes and other products. In all of the CBHI-
constructs, cth expression was driven by the rice rubisco small subunit (rch) promoter
and terminated with pin3’ or nosB’ region. The CHBI-constructs were prepared as
follows:

1. The plasmid pRRl, containing the rice rch (Yong et al., 1987), was kindly
provided by Dr. Ray Wu of Cornell University. The 2.1-kb EcoRI/EcoRV fragment
containing rch was excised from pRRl. The excised fragment was inserted into the
EcoRl/EcoRV site in the poly-linker of pBluescript H SK +/- (Figure 4.3) from
Stratagene (La Jolla, CA) to use as a promoter in our CBHI constructs. This plasmid with

rch promoter is called pSMF8 (Figure 4.4).

86

Figure 4.3. pBluescript 11 SK +/-

AmpiCiIIin

 

 

pBluescript II SK+/- é

 

my \

BssH ll
T7 ‘

Kpnl
Apal
Ecomomwmmi
Xhol
Sall/Hincll/Accl
Bsp|061 (Clal)
Hindlll

EcoRV

EcoRl

Psll

Smal

BamHl

Spel

X bal

Natl

Eagl

Sacll

BstXl

Sacl

mt

BssHll

 

 

87

Figure 4.4. pSMF8 construct

 

 

 

 

EcoRV EcoRI
___|._ rch _|___
pSK
2.1 kb

2. The plasmid pB210-5A containing cth was obtained from the National
Renewable Energy Laboratory (NREL) Golden, Colorado. To clone cth down stream of
the rch promoter in pSMF8, the 1.7-kb SalI/Xhol fragment containing cth was cleaved
from pBZlO-SA (Shoemaker et al., 1983). The digested plasmid was electrophoresed in
an agarose gel, and the 1.7 kb fragment was puriﬁed. The cleaved fragment was treated
with DNA polymerase I (Klenow enzyme) to produce blunt ends. pSMF8 was digested
with BamHI (a unique site) and then blunt ends were generated for cloning cbhl. Then
cth fragment was blunt-end ligated with pSMF8/ BamHI and used to transform
Escherichi coli XLlBlue bacterial cells. The transformed cells were plated on LB agar
media containing ampicillin for selection of transformed cells. Colonies were picked and

screened for ones containing the correct construct. This cloning yielded the plasmid

 

 

pSMF9 (Figure 4.5).
Figure 4.5. pSMF9 construct.
EcoRV EcoRI PstI Spel, NotI
__l_ rch l l cbh I I

 

 

 

 

 

 

 

pSK
2.1 kb 17 kb

88

3. To complete the heterologous gene expression cassette, the pin3’ transcription
termination nucleotide sequence was inserted at the 3’ end of cbhl in pSMF9. pin3’ is the
3’ region of the gene coding for protease inhibitor [1 from potato. This terminator was
excised from plasmid pBRlO-l I, which was obtained from Dr. Ray Wu, Cornell
University. To add pin3’ gene to pSMF9 without disturbing the rest of the plasmid, we
needed to excise pin3’ as a PstI fragment from plasmid pBRlO—l 1. According to the
pBRlO—ll map (Figure 4.6) there is a PM] site downstream of pin 3’ sequence. To
generate an additional PstI site upstream of pin3’sequence, a 70 bp Natl/Xhol fragment
containing multi- cloning site sequences from pSK vector was ligated with pBRlO-ll/
NotI/XhoI. pBRlO-ll with additional multicloning sites was named pSMFll (Figure
4.6). Then the pSMFl l/PstI fragment containing pin3’ terminator sequences was cloned
in pSK at PstI site. This plasmid was named as pSMF 12 (Figure 4.6). Then this pin 3’
terminator was cleaved from pSMF12 with NotI enzyme as lkb fragment and cloned
downstream of cbhl in pSMF9. This construct was named pSMFl3 (Figure 4.6). Using

this construct for transformation should result the expression of cbhl in the cytosol.

89

Figure 4.6. pBRlO—l 1, pSMFl 1, pSMF12, and pSMF13 constructs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pBRlO-ll:
NotI, XhoI, BamHI PstI
— ActinS’ CryIA(b) I pin3’ .l. 358 bar _ nos3’
pBY505
pSMF 11:
SmaI, PstI, HindIII, XbaI PstI
— Actin5’ CryIA(b) I pin3’ .I. 358 p bar _. nos3’
pBYSOS
pSMF12:
Pstl, NotI PstI, NotI
_|__ pin3’ ..__l_
pSK
lkb
pSMF13:
EcoRV EcoRI, Aer PstI, BerI Spel, NOII NOtI
__|_ rch cbh] pin3’ _l_
pSK
2-1kb 1.7 kb lkb

90

4. To target the CBHI enzyme to the chloroplast we needed to insert a transit peptide
(TP) sequence into the plasmid pSMF13 between the rch promoter and the cbhl. Transit
peptide was isolated from rch sequence from pRRl. Since there were not any restriction
sites inside the rch promoter to remove the transit peptide sequences, we ampliﬁed that
region by PCR. The pSMF 13 map shows there is a unique site Aer at the end of the
rch sequences (promoter) and there is a unique site BerI available upstream of the ﬁrst
ATG sequences in the cth. Therefore, PCR primers were generated using sequences for
these two sites and the TP sequences. The transit peptide was PCR-ampliﬁed and the
product (145bp) was puriﬁed and conﬁrmed by sequencing. This PCR-ampliﬁed TP was
inserted in pSMF13 between the promoter and the cbhl as sticky ends using directional
ligation using Aer and BerI restriction sites. This construct was named pSMF 14

(Figure 4.7). This construct should result in chloroplast accumulation of CBHI.

Figure 4.7. pSMF14 construct

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EcoRV EcoRI, AvrII PstI, BerI SpeI, NotI Notl
_|_ rch I TP l cbh] pin3’ _.|_
pSK
2-1kb 0.14 kb 1.7 kb 1 kb
PCR Primers:

SPlF 5’CCGCCTAGGCGCATGGCCCCCTCCGT3’

SP3R 5 ’CGCTGTACACGCACCTGATCCTGCC3 ’

91

5. Dai et al., (1999) have also tried to use cbhl from T. reesei to transform tobacco
plants. Because of the relatively high difference between the GC content of tobacco and
cth, to ensure higher expression of this gene in tobacco plants they codon-modiﬁed the
original cth. The GC content in their synthetic cth (Syn-cbhl) is around 43%, which is
much closer to tobacco GC content than the original cth with 55% GC content. But
since the GC content of monocots (maize) is higher (~65%) than dicots (tobacco), the
original gene with higher GC content is probably more suitable for maize (Murrary et al.,
1988). However, we have used their synthetic cth along with the original cth to
compare their expression levels in maize plants.

Plasmid pZD408 containing Syn-cbhl was obtained from Dr. Dai. pZD408 was
digested with NcoI and made blunt-ended. This blunt-ended vector was digested with
HindIII to remove the CaMV35S promoter. The large fragment (4.5 kb) was isolated, and
dephosphorylated. Plasmid pSMF 14 was digested with BSrGI and its ends ﬁlled in with
Klenows. Then this fragment was digested with HindIII. This removed rch promoter
with TP from pSMF14 plasmid, which was then ligated with pZD408 fragment. The
resulted clone was named pSMF15 (Figure 4.8). This plasmid should direct the synthetic
CBHI to chloroplast.

Figure 4.8. pSMF 15 construct.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EcoRV EcoRI, AvrII PstI, BerI Spel, NotI NotI
_l__ rch TP Syn—cbhl nos3’ J—
pB 1 121
2-1 kb 0.14 kb 1.5 kb 0.28 kb

92

Effects of the AFEX treatment on the activity of cellulase

Abstract

The potential of the AFEX process as a biomass treatment while preserving the
proteins in their bioactive form, was explored as part of our research. We approached this
goal by investigating the effects of the AFEX process on the activity of Rubisco in alfalfa
plants. The results of this investigation are presented in Appendix A. Based on data
presented in Appendix A, chapter 3 and previous work (De La Rosa et al., 1994) we
believe that, under a limited set of conditions, the AFEX treatment has the potential to be
used as an integrated pretreatment to increase the digestibility of biomass, improve
extraction of protein from biomass, and at the same time to preserve proteins in their
active forms.

Thus, there is a legitimate reason to hope that industrial enzymes such as cellulases,
produced in transgenic plants, can also survive the AFEX pretreatment and be recovered
in their useable form during biomass reﬁning. This possibility was explored in this
chapter by treating transgenic tobacco plants, expressing cellulase, with the AFEX

process.

93

Material and Methods
Plant material

Seeds for transgenic tobacco plants expressing Elcd (catalytic domain fragment
of E1 endo-l,4-B-glucanase from Acidothermus cellulolyticus) were obtained from Dr.
Sandra Austin-Phillips at the University of Wisconsin-Madison.

Figure 4.9 shows the plasmid that has been used by this group to transform

tobacco plants to express Elcd in apoplast.

VSPB
CaMV 358p a 61 Cd

 

 

nos I

 

Figure 4.9. Schematic representation of Elcd expression cassette. CaMV35sz
Cauliﬂower mosaic virus promoter; nos: The nopaline synthase transcription termination
signal; e1 cd: Catalytic domain of e1; VSPB: Soybean vegetative storage protein [3 leader

sequence to target the protein to apoplast (Ziegelhoffer et al., 2001).

94

Transgenic plants were grown, harvested (leaves) and dried at Michigan State
University greenhouses. As Figure 4.10 and Figure 4.11 show the expression of Elcd did

not cause any obvious phenotypic effect in plants.

 

 

 

 

Nontransgenic tobacco plant Transgenic tobacco plant

Figure 4.10. Transgenic and nontransgenic tobacco plants grown in greenhouse

95

Figure 4.11. Transgenic tobacco plants growing in the greenhouse.

 

 

 

 

 

 

96

Identiﬂing expressing E ch plants

After seedlings developed their ﬁrst true leaves, samples are removed for enzyme
assay to identify transgenic plants.
Fifty milligrams of leaf samples were homogenized in grinding buffer (50 mM NaOAc,
pH 5.5, 100 mM NaCl, 10% (v/v) glycerol, 0.5 mM ethylenediaminetetraacetic acid
(disodium salt), 1 mM phenylmethylsulfonyl ﬂuoride, 1 mg/l aprotinin, 1 mg/l leupeptin,
1 mg/l pepstatin) at a ratio of 2 pl per mg of sample (fresh weight). Samples were
homogenized with the aid of a power drill using 1.5 ml microcentrifuge tubes and plastic
pestles (Kontes). Soluble extract was recovered from insoluble debris after centrifugation
at 15000 x g for 5 min. A solution of saturated ammonium sulfate was added to the
extracts to achieve a ﬁnal concentration of 2.7 M ammonium sulfate (about 70%
saturation at 0°C). After incubation on ice for 30 minutes, the resulting precipitate was
recovered by centrifugation at 15000 x g for 5 min. The ammonium sulfate pellet was
resuspended in 5 u] of grinding buffer for each 2111 of starting crude extract.
Appropriate dilutions of plant extracts were assayed for Elcd activity using a 96-well
plate format. The reaction buffer contains 50 mM NaOAc pH 5.5, 100 mM NaCl, 0.5
mM 4-methylumbelliferyl B-D—cellobioside (MUC) (from Sigma). MUC hydrolysis by
EllElcd releases the ﬂuorescent product, 4-methylumbelliferone (hex = 360 nm, Lem =
465 nm). Each well contained 1 to 411.1 of the sample to be assayed and 100 pl reaction
buffer. Plates were covered with adhesive lids to prevent evaporation and incubated for
30 min at 65° C. The reaction was terminated by the addition of 100 pl of stop mix (0.15
M glycine, pH 10.0) and ﬂuorescence was determined with a Spectra Max Gemini

(Molecular Devices, Sunnyvale, CA) at 465 nm using an excitation wavelength of 360

97

nm. Fluorescence values were compared to values obtained with 12 to 240 pg of puriﬁed
Elcd (kindly provided by Steven R. Thomas, National Renewable Energy Laboratory,
Golden, Co), a range of enzyme concentrations that yields a linear response. A series of
4-methylumbelliferone standards (4 to 160 pmol) was also included. For all transgenic
plant samples, Elcd activity was determined by subtracting the background contributed
by W38 control extracts (done in parallel). The production level of Elcd in transgenic
tobacco plants was up to 2.5 % of total soluble protein. The plants that were identiﬁed as
expressing Elcd were grown to maturity, and their leaves were harvested, dried, and
collected for AFEX treatment. Since each plant had different Elcd levels of expression,
before AFEX treatment the dried leaves were ground and well mixed to insure the

homogeneity of the samples.

Heat treatment of the transgenic tobacco plants expressing E1 at

The total moisture content of the dried transgenic tobacco plant material was
measured as described in chapter 3. The samples were prewetted to the desired moisture
content 30 min prior to each treatment. The prewetted samples were placed in the
pressure vessel. The vessel was topped up with stainless steel pellets (approximately
1mm diameter) to occupy the void space, then the lid was bolted shut. The entire reactor
assembly was placed in a 400W PARR heating mantle to warm the unit to the desired
temperature. To avoid overheating, the reactor was taken out of the heater at
approximately 10°C prior to the target temperature, and if needed the unit was placed in a
bath of cold water to maintain the system at the set temperature. The pressure and the

temperature were monitored and recorded throughout the experiment. The system was

98

kept in the target temperature for 5 minutes. Since there was not any ammonia in the
system, no pressure increase was observed during the experiment. After the experiment
was completed the heat-treated samples were removed and kept at 4°C until further

analysis. Table 4.4 shows the summary of the conditions used in this set of experiments.

Table 4.4. Conditions used in heat treatment of transgenic tobacco plant

 

Temperature Moisture content (% dry weight basis)

 

60°C 20
40

60

 

70°C 20
40

60

 

80°C 20
40

60

 

90°C 20
40

60

 

 

 

 

99

Table 4.5 shows the related calculations and the conditions of the system during

one of the heat treatment runs of transgenic tobacco plant material.

 

 

 

 

 

 

 

 

 

10g transgenic tobacco plant material (containing 1g water)
6g distilled water was added for 60 % moisture content (dry weight basis)
Elapsed time, min Temperature, °C Pressure, psi Comments
0 28 0 Placed in heater
2 36 0
4 45 0
6 62 0 Taken out of heater
8 69 0
12 70 0 End of the experiment

 

 

 

 

 

 

Table 4.5. Conditions during heat treatment of transgenic tobacco plants.

Ammonia treatment of the transgenic tobacco plants expressing Elcd

The prewetted samples were placed in the pressure vessel. The vessel was topped
up with stainless steel pellets to occupy the void space and then the lid was bolted shut.
The precalibrated sample cylinders were ﬁlled from the lecture bottle with the desired
amount of warm ammonia to charge the system. To insure that the desired amount of
ammonia was delivered, the reactor vessel was weighed before and after loading. Without

heating, the vessel was left in the hood for 10 min; during the course of the experiment

100

only 10 to 20 psi pressure increase was observed. Since during the experiment some of
the liquid ammonia converted to gaseous ammonia, in all of the experiments the system
temperature was decreased, in some of them down to 28°C. Table 4.6 shows the

summary of the applied conditions in ammonia treatment of transgenic tobacco plants.

 

Ammonia loading Moisture content

g of NH3: g of dry biomass (% dry weight basis)

 

0.5:1 20

40

60

 

0.7:1 20

4O

60

 

1:1 20

40

6O

 

 

 

 

Table 4.6. Conditions of ammonia treatment of transgenic tobacco plants

101

An example of related calculations and the conditions during the ammonia

treatment is presented in Table 4.7.

Table 4.7. Conditions during one of the ammonia treatment runs

 

 

 

 

 

 

 

 

 

 

16 g of transgenic tobacco plant (containing 1.6g water)

4.16 g distilled water was added for 40 % moisture content (dry weight basis)
Temperature of N H3 was 33°C; 10 g NH; was added for 0.7:1 ammonia level
Elapsed time min Temperature °C Pressure psi Comment

0 34 95 No heating

2 32 100

4 3 l 105

6 29 105

8 28 100

10 26 100 The pressure was released

 

 

 

 

 

 

AFEX treatment of transgenic tobacco plants expressing [5ch

The prewetted samples with desired moisture contents were placed in the pressure
vessel and AFEX treated exactly as described in chapter 3. The treated samples were
allowed to stand in a fume hood overnight to evaporate the residual ammonia, and then
kept at 4°C for further analysis. The samples were AFEX treated under various

conditions, which are summarized in Table 4.8.

102

Table 4.8. AFEX treatment conditions for transgenic tobacco plants

 

 

 

 

 

 

 

g of NH3: g ofdry biomass 20 % 40 % 60 %
Moisture Moisture Moisture
0.5:1 T: 60°C T= 60°C T= 60°C
T=70°C =70°C T=70°C
T=80°C T=80°C T=80°C
0.7:1 T: 60°C T= 60°C T= 60°C
T=70°C T=70°C T=70°C
T=80°C =80°C T=80°C

 

 

Measuring the activity of Elcd enzyme in transgenic tobacco plant after treatments
The treated and untreated samples were ground and sieved to 40 mesh prior to the

Elcd activity assay. The activity was measured as described above.

103

Results and Discussion

Transgenic tobacco plants (dried leaves) were treated under different range of
temperatures (Table 4.3) or under different levels of ammonia (Table 4.5) to assess the
individual effect of each AFEX variable on the Elcd enzymatic activity. All the results
are the mean of two replicates. The effects of the different temperatures are presented in

Figure 4.12.

 

 

 

 

 

 

 

 

 

compared to untreated

 

% Retentlon of E1 ed actlvlty

 

 

 

899999999g99°
§88822288e88§
=\\\\\\\°\\\\~2
goooooooocooc

NQONQONQCDNQO

Moisture Content% (dwb), Treatment Temperature

Figure 4.12. Effects of temperature on the activity of Elcd extracted from heat-treated

transgenic tobacco plants.

104

Ziegelhoffer er al. (2001) tested the stability of apoplast-targeted Elcd in
transgenic tobacco plants. In this study the apoplast-targeted Elcd enzyme extracted
from tobacco plant along with the puriﬁed microbial Elcd were subjected to different
temperatures (60°C-90°C) for 10 min. Both enzymes showed similar thermal stability
throughout the experiment and as expected they both showed great heat resistance. Their
results showed that at 60°C up to 95%, at 70°C up to 90%, at 80°C up to 80%, and at
90°C up to 40% of both enzyme activities were retained. But as seen in Figure 4.12, our
heat stability test showed very different results from what Ziegelhoffer et al. (2001)
reported. The Elcd extracted from transgenic tobacco plant treated at 60°C showed
maximum 67% and at 70°C showed maximum 46% activity retention compared to Elcd
extracted from untreated transgenic tobacco plant. The Elcd extracted from transgenic
tobacco plant treated at 80°C and 90°C was almost totally inactivated. Based on these
results and the previous study (Ziegelhoffer et al., 2001), we believe that the Elcd
extracted from transgenic plants is therrnostable if it is heated in the extraction buffer
(which has the enzyme’s ideal pH 5.5) and its thermal stability drastically decreases if the

expressing plants are heated directly.

105

 

The effect of ammonia on the activity of the Elcd enzyme was examined at

different ammonia loading levels (Table 4.5) and the results are presented in Figure 4.13.

§

90
80
70
60
50
40
30
20
1 0

% Retention of E1 cd activity
compared to untreated

 

‘3 o\°

o
‘5 N
g .:
c m
3 a

0.5:1, 40%
0.5:1, 60%
0.7:1, 20%
0.7:1, 40%
0.7:1, 60%
1 1, 20%
1:1, 40%
1 1, 60%

Kg of ammonia: kg of dry biomass, Moisture content % (dwb)

Figure 4.13. Effects of ammonia on the activity of Elcd extracted from ammonia-treated

transgenic tobacco plants.

Figure 4.13 shows, at any moisture content, as the ammonia loading increases the
percent of Elcd activity retention decreases and at 1:1 loading ratio the Elcd enzyme was
almost totally inactivated. The highest percent of activity retention (49%) was observed

at 0.5:1 ammonia loading with 40% of moisture.

106

The transgenic tobacco plants expressing Elcd were AFEX treated under different
conditions (Table 4.7). The conditions of these experiments were selected based on the
results of our heat and ammonia treatment of transgenic tobacco plants; we chose the

conditions that showed higher enzymatic activity retention for E1cd. The results of these

experiments are presented in Figure 4.14.

 

100a
90-
80*
70-
60*
50-
40:
30‘
20~
10*
o-

 

 

 

 

 

 

 

 

 

 

     

  

 

 

 

 

 

     

% Retention of E1 ed avtlvity
compared to untreated

Untreated
0.5:1, 20%,60°C
0.5:1, 40%, 60°C
0.5:1, 20%, 70°C
0.5:1, 40%, 70°C

0.7:1, 20%,60°C
0.7:1, 40%,60°C
0 7 1, 20%,70°C
0.7:1, 40%,70°C

Kg of ammonia: kg of dry biomass, Moisture content % (dwb), Temperature

Figure 4.14. Activity retention of Elcd extracted from AFEX-treated transgenic tobacco

plants.

Elcd enzyme showed a better survival rate with each heat or ammonia treatment
(up to 67% and 49% respectively), but the combination of these conditions in AFEX
treatment caused a drastic loss in the activity of Elcd extracted from AFEX-treated

transgenic tobacco plants. The maximum observed activity retention in AFEX-treated

107

 

transgenic tobacco plant was only 35% at 60°C, 0.5:] ammonia loading and 40%
moisture.
Summary and Future work

Our research collaborator (Dr. Sticklen’s group from Michigan State University)
has used plasmids pSMF13, pSMF14, and pSMF15 for maize transformation to target
CBHI to different organelles of maize plants. In addition to these plasmids they have also
used a series of plasmid containing the elcd (provided by Dr. Austin-Phillips) to target
Elcd to the cytosol, chloroplast, or apoplast. The transformed plants are still in the
regeneration step and have not been tested yet. But based on the prior work (Ziegelhoffer
et al. (2001)) we are expecting to observe similar results in terms of subcellular targeting
effects on enzyme expression level, which is higher in apoplast than in the cytosol or
chloroplast.

As we mentioned before, the pH of each compartment might have a signiﬁcant
effect on the stability of any particular heterologous protein targeted to each of these
organelles. Considering this fact, one can suggest that one way to optimize the expression
of any foreign proteins in transgenic plants is to direct them to the organelle with a pH
most similar to their ideal pH.

Vacuoles are very conspicuous organelles with pH around 5.5. In some plants the
vacuole may occupy as much as 95% of the plant cell volume (Spremulli, 2000). Since
the ideal pH for both El and CBHI is around 5.5, directing these enzymes to vacuoles
and comparing these results with prior work could provide important insights that can

help to improve the production of these enzymes in transgenic plants.

108

 

Originally, pretreatment techniques were developed for various end uses of
lignocellulosic biomass with emphasis on ethanol production. The pretreatments have
been used to enhance enzymatic hydrolysis of biomass to increase ethanol production and
improve lignocellulose conversion. Advances in biotechnology are enabling plants to
become economically important systems for producing heterologous proteins such as
cellulases. Therefore expanding the application of biomass pretreatment in releasing and
recovering these proteins could have a signiﬁcant impact on bioreﬁnery economics. In
this research we examined the potential of AFEX treatment in this regard. The data show
that the maximum Elcd activity retention in AFEX-treated transgenic tobacco plant
(under 60°C, 0.5:] ammonia loading and 40% moisture conditions) was only 35%, and
the glucan conversion of the corn stover treated under the same conditions was only 47%
of theoretical yield with 60 FPU/ g of glucan of cellulase (chapter 3). Based on these
ﬁndings, it is our Opinion that AFEX pretreatment is not a suitable option for releasing
cellulase from transgenic plants. Considering the fact that other biomass pretreatments
(e.g. steam explosion, acid treatment) operate under even harsher conditions, it is
reasonable to postulate that none of these pretreatments are suitable choice for this
purpose. Using biomass pretreatment for releasing cellulase from transgenic plants, was
only one our three technical options; therefore, exploring the potential of other options

(mentioned in the introduction) deserves attention in future work.

109

Chapter 5

The pea (Pisum sativum L.) rch transit peptide directs the Alcaligenes
eutrophus polyhydroxybutyrate enzymes into the maize (Zea mays L.)
chloroplasts

Heng Zhong, Farzaneh Teymouri, Brad Chapman, Shahina Bano Maqbool, Robab
Sabzikar, Yahia El-Maghraby, Bruce Dale and Mariam B. Sticklen, (2003). Plant Science

In Press

110

Abstract

With a concept of producing polyhydroxybutyrate (PHB), a biodegradable thermoplastic
polymer, in maize chloroplasts, over 200 independent transgenic maize plants were
produced. Five different constructs pPHBA, pPHBB, pPI-IBC and pBY520 or pJSlOl
were used in transformation experiments. PHB constructs contained phb genes from
Alcaligenes eutrophus fused with the pea rch transit peptide plus the ﬁrst 24 amino
acids of the mature rch protein. Plasmids BY52O and JSlOl contained the selectable
marker bar gene linked with abiotic stress-related gene either hval or mtld. Southern
blots on putative transgenic plants conﬁrmed that the transgenic plants harbor transgenes
from all ﬁve constructs. Transcription of transgenes was conﬁrmed by northern blot
analyses. Western blot analyses using total cellular protein and protein from isolated
chloroplasts conﬁrmed that all three genes produced their corresponding enzymes, and all
PHB enzymes were targeted into maize chloroplasts. Immunoﬂuorescent localization on
both transgenic and non-transgenic maize chloroplasts treated with PHBC-antiserum also

conﬁrmed the targeting of PHBC enzyme to the chloroplasts of transgenic maize.

Key words: Chloroplasts; Maize transformation; PHB; rch - transit peptide

Abbreviation: Rubisco = ribulose 1-5, bisphosphate carboxylase

lll

1. Introduction

Polyhydroxybutyrate (PHB), ﬁrst characterized in Alcaligenes eutrophus, is a
biodegradable polymer, which can be completely degraded by enzymatic activities in the
soil [1]. There are three enzymes needed for production of PHB polymer. These enzymes
include 3-ketothiolase (PHBA), acetoacetyl-CoA reductase (PHBB), and
polyhydroxybutyrate synthase (PHBC). Genes encoding these three enzymes were
isolated from A. eutrophus and were transferred to the nuclear genome of Arabidopsis
thaliana [l], where granules of PHB were produced in the cytoplasm of leaf mesophyll
cells of this plant. However severe growth inhibition and yield reduction were observed
in this research. In a follow up experiment, Nawrath et al. [2] used the pea rch transit
peptide and the ﬁrst 24 amino acid coding sequence for targeting all three PHB enzymes
into the A. thaliana chloroplasts. In this follow up experiment, the PHB polymer was
produced at a high level in the chloroplast without any reduction in plant growth and/or

seed yield.

Use of transit peptide (TP) of Rubisco small subunit (rch) has become a routine
practice to transport proteins into chloroplasts of transgenic plants. Corbin et a1. [3] used
the transit peptide from the Arabidopsis 1A Rubisco small subunit gene plus transit
peptidase cleavage site to target cholesterol oxidase into chloroplasts of transgenic
tobacco plants. In another event Comai et al. [4] used the pea rch transit peptide plus the
ﬁrst 24 amino acid coding sequence for targeting the 5-enolpyruvyl 3-phosphoshikimate
(EPSP) protein into the chloroplasts of transgenic tobacco plants. Although literatures
have shown that the transit peptide of pea is sufﬁcient to direct foreign polypeptides both

in vitro and in vivo in dicot [5,6] and monocot [7] plants, Comai et al. [4] showed that

112

chloroplast uptake also requires the ﬁrst 24 amino acids of the mature rch protein in
addition to the transit peptide. In all of the above experiments, genes were transferred into

the plant nuclear genome while enzymes were targeted into the chloroplasts.

Since, the chloroplast genome has limited coding capacity, many chloroplast
proteins are synthesized in cytoplasm by cytoplasmic ribosomes in the form of larger
precursors. The precursors consist of an amino terminal extension referred to as a transit
peptide and a carboxyl terminal portion. The import of the protein from cytoplasm to
chloroplasts is initiated via binding of the precursor to a proteinaceous receptor followed
by translocation across the membrane via an energy requiring mechanism. Inside the
chloroplast, a stromatic protease cleaves the transit peptide, leaving the mature protein
[8]. The most abundant protein in the chloroplast is Rubisco. The large subunit of
Rubisco is encoded in chloroplast DNA but the small subunit is a nucleocytoplasmic

product that is imported post translationally into the chloroplast [9].

Over the past few years progress in plastid transformation technology has showed
the feasibility of chloroplast transformation. Chloroplast transformation opens the
possibility of expressing multiple genes or entire biosynthetic pathway in a single
transformation event [10]. Plastid transformation technology in higher plants was ﬁrst
developed for tobacco plants [1 1]. But it has been shown that extending this technology

to other species especially important crops such as maize is still challenging [12].

Here, we chose to transform PHB constructs (PHBA, PHBB and PHBC) into
maize nuclear genome and targeted the PHB enzymes inside the chloroplast. This is the

ﬁrst report about targeting of the PHB enzymes into a monocot chloroplast using a dicot

113

 

transient peptide plus the ﬁrst 24 amino acids of the Rubisco small subunit mature
protein.

2. Materials and methods

2.1. Plant materials

Honey N Pearl sweet corn genotype was used in our transformation experiments
on the basis of its best response in vitro regeneration ability and fertility following our

previous work [13,14].

Shoot tip clumps were obtained in vitro from the apical meristems of 7-day-old
maize seedling as described [13]. Shoot tips with one to three visible leaf primordia were
selected under a stereomicroscope (Carl Zeiss, Germany) for microprojectile
bombardment. To obtain a better exposure of shoot meristems to the bombardment, leaf
primordia were physically removed and discarded. These explants were placed in a
circular area of about 1.5 cm diameter in a Petri dish containing phytagel-solidiﬁed

culture medium prior to bombardment.

2.2. Plasmids

Three plasmids: PHBA, PHBB and PHBC were used to transfer phb genes into
maize shoot apical meristems (Fig. l). The expression of all three genes were driven by
the CaMV3SS promoter and terminated with nos 3’ region [1]. Each of these plasmids

was co-transformed with either pBY520 or plSlOl individually. The plasmid BY520

114

contains selectable marker bar gene and abiotic stress-related hvaI gene [15]. The
plasmid J S 101 contains the bar gene and another abiotic stress-related mtlD gene [16]. In
both plasmids, bar gene expression was driven by the CaMV35S promoter and
terminated with nos 3’ region [17]. In pBY520 and pJSlOl, the hval and mtlD coding
regions were driven by a region of 1.3 kb upstream of the rice Actl translation codon and

terminated with pin 11 3’ region.

Abbreviation in Figure 1:
CaMVBSS: Cauliﬂower Mosaic Virus 35S promoter; TP: Transit peptide of the
small subunit of Rubisco of pea plus the ﬁrst 24 amino acids of rch mature
protein; pth: gene of Alcaligenes eutrophus encoding the 3-ketothiolase; pth:
gene of Alcaligenes eutrophus encoding the acetoacetyl-COA reductase; pth:
gene of Alcaligenes eutrophus encoding the PHB synthase; Actl: rice Actl gene
promoter, which includes the 5’ intron region; bar: phosphinothricin acetyl
transferase (selectable marker/herbicide resistance) gene; hvaI: LEA3 gene from
barley; mtlD: mannitol l-P dehydrogenase gene from Ecoli; pin: potato protease
inhibitor 11 gene 3’ non-coding region; nos: 3’ region from Agrobacterium

nopaline synthase gene.

115

Figure l. Plasmid constructs used in maize transformation.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HindIII EcoRI
pPHBA I I
C MV35 TP — , ~
3 [1th nos pBIlZl
0.8kb 0.25kb 1.3 kb 0.28 kb
HindIII EcoRI
pPHBB I l
C MV35 ~ TP .
a pm ""8 P131121
0.8 kb 0.25 kb 0.8 kb 0.28 kb
HindIII EcoRI
pPHBC l
C MV35 — TP th -»
a p “"8 pB1121
0.8 kb 0.25 kb 19 kb 0.28 kb
pBY520
XhoI EcoRI,

 

41‘ Actl I've! -

 

 

 

 

 

pin — CaMV3 ’ bar nos —L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pSK
lkb 0.9kb 0.8kb 0.59kb 0.28k
1.3kb
pJSlOl
XhOI EcoRI,
‘1' Actl mtlD pin CaMV35 bar - nos -—|—
1.3 kb 1.5kb 0.9 kb 0.8 kb 0.59kb 0.28 kaSK

116

2.3. Microprojectile bombardment

Plasmid DNA was precipitated onto tungsten particles following the protocol
described previously [14]. The bombardments were carried out using a biolistic particle
acceleration device (PDS lOOO/He, Du Pont,Wilmington, DE) under a chamber pressure
of 26 mm of Hg at different distances. The distance between the rupture disc to the
macrocarrier was 1.5 cm, to the stopping screen was 2.0 cm and to the target was 6.5 cm.
The meristems were bombarded twice with 1.0 pm DNA coated tungsten particles as

described [14] with a density of 150 u g per shot and 1550 psi acceleration pressure.

2.4. Selection and regeneration of transfonnants

Bombarded shoot tips were cultured on the MS medium [18] containing 2 mgl'l
N6-Benzyladenine (BA) and 0.5 mgl'l 2,4-Dichlorophenoxyacetic acid (2,4-D) for 4
weeks. The clumps then were divided and subcultured on the same medium
supplemented with 3 mgl‘l of glufosinate ammonium for selection for 4 weeks. The green
clumps were selected, divided and subcultured on the same medium containing 5 mgl‘l
glufosinate ammonium at 4-week intervals. Plant regeneration was obtained by
transferring the selected shoot-tip clumps to MS medium containing 0.5 mgl'l BA, 0.5
mgl'1 Indole-3-butyric acid (BA), and 10 mgl’l glufosinate ammonium. Regenerated

shoots (5-10 cm tall) were rooted on MS medium containing 1 mgl'l [BA and 10 mgl’l

117

glufosinate ammonium. Plantlets were transferred to pots containing a soil mixture of 1:1

(v/v) peat:perlite and grown to maturity in a green house.

2.5. DNA isolation, PCR and Southern blot analyses

Genomic DNA from glufosinate ammonium resistant plantlets was extracted as
described [19]. PCR analyses were carried out using speciﬁc primers for pth, pth and

pth. The sequences of the primers were as follows: for pth, forward primer 5’-

GCGTCAAGCCGGAGCAGGTG-3', reverse primer 5'—GATCGTGGCCAGCGGGG

TCAGG-3’; for pth, forward primer 5’-CCGGCGGCATGGGTGGTATC-3’, reverse

primer 5’-CGGCCTI‘GGCGGTGGAGTAGTTG-3’; for pth, forward primer 5’-CTGC
CGCG'ITCTACCTGCI‘CAATG-3’ and reverse primer 5'-CACCGTATGTCCC TGCT C

CACCAC-3’. The DNA was subjected to PCR using a GeneAmp PCR system (Perkin
Elmer, Norwalk, CT) for 3 min at 94 °C following by 40 cycles of 30 s at 94 °C, 1 min at
62 °C and 1 min at 72 °C and ending with 5 min at 72 °C. Ampliﬁed DNA was

electrophoresed in a 0.8% agarose gel and visualized under UV.

A modiﬁed C-TAB method [20] was used to isolate high molecular weight DNA
from control as well as putative transgenic plants. Southern blot hybridization was
performed as described [21]. Twenty p g of plant DNA from each sample was restriction
digested overnight with either HindIlI alone or in combination with EcoRI. DNA was

electrophoresed and transferred onto a piece of pre-wetted Hybond N+ nylon membrane

118

(Amersharn, Piscataway, NJ). Gene-speciﬁc probes were generated using EcoRI/Smal
digest of pPHBA, pPHBB and pPHBC to isolate fragments containing pth-, pth- and
pth— coding sequences. Restriction fragment was puriﬁed using the QIAquick kit
(QIAGEN, Valencia, CA), and labeled with a-[32P]-dCTP using Rad Prime labeling kit
(GIBCO, Rockville,MR) according to the manufacturer’s instructions. Membranes were
hybridized at 68 °C overnight using speciﬁc probes and analyzed by autoradiography on

Kodak X-OMAT ﬁlm at —80 °C.

2.6. RNA isolation, northern blot analyses

Total RNA was isolated using Extract-A-PlantTM RNA Isolation Kit (Clontech,
Palo Alto, CA) and treated with DNAase I (Gibco) to eliminate any traces of genomic

DNA.

For northern blots, 20 ug of RNA were fractionated in 1.2% agarose-
formaldehyde denaturing gels and blotted onto Hybond-N nylon membranes (Amersharn)
as described [22]. Gene speciﬁc probes used for Southern analyses were also used for

northern hybridization.

2.7. Protein isolation and western blot analyses

Total leaf protein extraction: One gram of fresh leaf tissue from each sample was
pulverized in liquid nitrogen and extracted in 0.4 ml of protein extraction buffer

(Phosphate-buffered saline (PBS), 10% glycerol, 10 mm DDT). The concentration of

119

total soluble protein in each sample was measured according to Bradford [23] using

bovin serum albumin as standard.

Extraction of Chloroplast protein: The intact chloroplasts were isolated as
described [24]. Intact chloroplasts were lysed for protein extraction by resuspension in 40
pl of deionized water and incubated for 3 min at 4°C. Then 40 pl of 2X protein extraction
buffer was added. The concentration of total soluble protein in each sample was

measured according to Bradford [23] using bovin serum albumin as standard.

Seventy pg of total leaf protein or chloroplast protein from each sample were
loaded side by side in SDS polyacrylamide gels lanes. The electrophoresed gels were
transferred to the nitrocellulose membrane (Amersham) using Semi-Dry Transfer Cell
(Bio-Rad, Hercules, CA). The membranes were incubated with primary antibody raised
against each of the PHB enzymes (kindly provided by Dr. Poirier) in recommended
dilution and then with alkaline phosphatase conjugated anti-rabbit IgG (Sigma, St. Louis,
MO) as the secondary antibody following by washing steps according to standard
procedures [22]. Protein bands were visualized in alkaline phosphotase substrate buffer

(Promega).

Transgenic Arabidopsis (homozygous) plants, expressing pth, pth, and pth
genes (provided by Dr. Somerville) were used as positive controls in our western blots.

Proteins from these plants were isolated as described by Nawrath et al. [2].

2.8. Immunoﬂuorescent localization and visualization of the PHB enzymes in

chloroplasts

120

Intact chloroplasts of non-transgenic and transgenic plants were isolated as
described [24]. The chloroplasts were resuspended in 200 pl of cold buffer B (0.3 M
sucrose, 50 mM Tris- HCL, 20 mM disodium EDTA, pH 8.0) and incubated with the
primary antiserum raised against the PHBC enzyme for 3 h, then centrifuged at 1000g for
5 min. The pellet was washed with buffer B and incubated with Alexa Fluor® 488 goat
anti-rabbit IgG (Molecular Probes, Eugene, OR) as the ﬂuorescent probe for 2 h. After
centrifugation at 1000g for 5 min, the pellet was washed with buffer B and pelleted again
then the pellet was resuspended in 50 pl of buffer B. Ten pl of the suspension was loaded
on a glass slide and left for 5 min to dry out. A drop of Prolong Antifade (Molecular
Probes) was added to the dry sample, and then the sample was covered with a coverslip.
The slides were viewed with a Zeiss 210 confocal laser-scanning microscope with a ﬁlter

to detect Alexa ﬂuor 488.

121

3. Results

Over 5,000 plantlets containing a total of 240 independent transformation events
resistant to glufosinate ammonium were recovered from maize meristem clumps

bombarded with one of the three phb genes, the bar selectable marker gene and either

mtlD or hvaI genes.

The PCR analysis on glufosinate ammonium resistant selected plantlets indicated
the presence of the phb genes (pth, pth and pth) in the analyzed transgenic plants
(data not shown). Further, conﬁrmation on phb genes integration into maize genome was
made by Southern blot analyses (Fig. 2). Southern blots using DNAs digested with
HindIII revealed the integration of 3-4 copies of phb genes (Lane H; Fig. 2) and the
DNAs digested with HindIII-EcoRI conﬁrmed the presence of complete phb gene

cassettes into the maize genome. (Lane D; Fig. 2).

122

Figure 2. Southern blots of independent transgenic maize lines containing pth (a),
pth (b) and pth (c) genes. P: plasmid DNA: a) PHBA, b) PHBB and c) PHBC; C:
control (untransformed maize plant); U: undigested; D: digested with HindIII/EcoRI; H:

digested with only HindIII; lanes 5- 16: independent transgenic maize lines.

P C T1-21 T1-25 T1-32 T1-18
UDUDDHUDHUDHUDHU

- ,e .§‘I-ge

 

a. “4"
ﬁt-
‘9'
C "" .7. "
a an - *

 

 

 

P C JR1-3O JR1-12 JR1-29 JR1-9
UDUDDHU DHUDHUDHU

 

 

 

 

 

 

 

b -- - to -— “Q 'r i - i
u.- “ ._.. - ’
w
«I. v - U
2.2kb
_’ .. __ _. .. a.
P C 185—6 131-3 JSS-S J -1
U D U D D H U D H U D H U D H U
nu ﬂ
3.3kb , V u " p , "'
——> no U a w. a. a".
w _ a
U at t.

 

 

 

123 4567 8910111213141516

123

Northern blot analyses conﬁrmed the presence of mRNA for the phb genes at
detectable levels in the Southern blot positive transgenic lines (Fig. 3). Western blot
analyses conﬁrmed the PHBA, PHBB and PHBC enzymes syntheses as the result of
translation of the pth, pth and pth genes (Fig. 4). Visual comparison was performed
between the total protein isolated from chloroplasts (lane C; Fig. 4) and total protein from
the whole leaf tissue (lane T; Fig. 4) based on the band intensity of each of the PHB
enzymes. The whole—leaf protein and the chloroplast protein extracts from each
transgenic plant showed that the level of the PHB enzymes in the chloroplast extracts
were much higher than the PHB enzymes in the whole-leaf tissue extracts (Fig. 4).
Immunoﬂuorescent localization on both transgenic and non-transgenic maize chloroplasts
treated with the PHBC-antiserum also conﬁrmed the targeting of the PHBC enzyme to

the chloroplasts of transgenic maize (Fig. 5).

124

Figure 3. Northern blot analyses of independent transgenic maize lines containing: pth
(a), pth (b) and pth (c) genes. C: untransformed maize plant; lanes 1-6: independent

transgenic maize lines.

\3) «\3‘) ($5514 (‘Y\% (“3,“) «\A'b C

may‘ “’1': w

        

 

 

,e ,3 .5 ,5“ ,0 :9
85" 83‘ a" 3 85" s" -C .,

 

125

Figure 4. Irnmunoblot analyses of independent transgenic maize lines expressing: PHBA

(a), PHBB (b) and PHBC (c). M: marker; P: positive protein (transgenic Arabidopsis); C:

chloroplast protein; T: total leaf tissue protein; lanes 2-5: independent transgenic maize

lines.

kDa

33

29

68

29

18

43

18

 

 

T1 ~21 T1 -25 Untransformed
M C ' T}. 1 C T C T P
m... a .,
w a. inn “ "‘ M
W

 

 

 

 

 

 

 

 

‘9

 

 

 

 

 

 

JR1-30 JR1-29 JR1-30 JR1-29 Untransformed
M 9 9 T r c "“1" 1?
“Wart-v» * -‘-,
JSS-G JS1-3 Untransformed
M C T C T C T P
I. w ‘- w gun—n- «hm-0a. “L
1 2 3 4 5 6 7 8

126

PHBA
41 kDa

PHBB
26 kDa

 

Figure 5. Immunoﬂuorescent localization of PHBC enzyme: Chloroplasts of transgenic
maize transformed with pth gene and treated with PHBC-antiserum (a); Chloroplasts of
control untransformed maize plants treated with PHBC-antiserum (b). (Please note that
green dots are showing the PHBC enzyme, red shows chloroplast, and yellow dots are the

overlaps of PHB enzymes with grana (contains chlorophyll) of chloroplasts).

 

127

4. Discussion

Here, we have expanded the application of the maize meristem transformation
system and the bar gene selection scheme formulated in our laboratory, which proved to

be effective in linked to unlinked co-transformation of maize [14,25].

It has been shown that the best way to optimize a metabolic pathway in plants is
to direct the foreign proteins to organelles where substrate is most available [26]. Acetyl
Co-A is the substrate needed in the pathway of the PHB synthesis. Since, there is a high
amount of acetyl-coA substrate in chloroplast [2], we chose to direct the PHB enzymes

into maize chloroplasts.

We used a transit peptide to target the PHB enzymes into the maize chloroplasts.
As expected, chloroplasts of the transgenic plants showed higher level of the PHB
enzymes accumulation as compared to the PHB accumulation in whole-leaf tissue (Fig.
4). The localization of the PHBC enzyme in chloroplasts was conﬁrmed using

immunofluorescent microscopy (Fig. 5).

According to Young et al. [27], the similarity between dicot and monocot rch
transit peptides is very low (24-38%). But on the other hand, the similarity between the
first 24 amino acids of rch mature protein of dicot and monocot is relatively high (66-
79%). The result of our report suggests that in this ChlorOplast protein targeting, the low
similarity (only 21%) between rch transit peptide of pea and maize may not be as
important as the role of their high similarity (79%) in their ﬁrst 24 amino acids of rch

mature protein (Table 1; Table 2).

128

as a

 

 

 

:2.

% uq>moozmpsmesmh 8;

t. o m _ as:

3. z > m m AM? a o H 2 m x q o o dips... > .«..-.«.,_ 8m

3. acozm>204mmemmm<>ﬁqm<§as:

NN Mam... o o m m «.m m. may at, <..m.-.W m m Hr»! - m_ «2%.. .§

mmmeAOOmm<.>.<_H<m.m.<22>em<zhwas:

.88. 8:268 Boa oEE< 28E

 

 

 

 

«on. can 332 Mo 8533 @233 amen: moat mo :oﬂSQEoU .— 933.

129

Table 2. Comparison of the ﬁrst 24 amino acid of Rubisco small subunit sequence in

Maize and Pea

 

 

 

Plant Amino acid sequence Total
Maize' 'ﬁ'6"v°-W'_ﬁ'§ A YR)”; N3 KH'Kuﬁmém'i:Imémfmf'i’mﬁmfz S T : 15; 24
Pea” MQVWPPIGKKKFETLSYLPPLTRID: 24

 

oooooooooooooooooo

IIIII

:-.---o-o—-—o-o-o---c- ooooooooooooooooooooo

 

 

Since the production of the PHLB biodegradable plastic for commercial use in

microbes is very costly [1], a promising alternative to this is to produce PHB polymer in

transgenic plants. We chose maize because it is one of the major US. crops, a G4 plant

with large chloroplasts, and a more rapid rate of photosynthesis than 03 species.

The targeting of the PHB enzymes in maize using the pea rch transit peptide is

our ﬁrst report towards the production of the PHB biodegradable plastic in maize. The

crossbreeding of the transgenic plants for combining the three phb loci into a single

maize genotype followed by backcrossing of the three phb loci into elite open-pedigree

maize inbred lines is in progress.

130

 

Acknowledgement

We wish to thank Dr Chris Somerville for providing the PHB constructs and
seeds of transgenic Arabidopis, Dr. Ray Wu for pBY520 and pJSlOl constructs, and Dr.
Yves Poirier for the gift of polyconal antiserum to each of the PHB enzymes. We would
also like to thank Dr. Joan Whallon for viewing the chloroplast slides by confocal laser-

scanning microscope and Dr. Barbara Sears for the review of this manuscript.

This research was funded by the Consortium for Plant Biotechnology Research

(CPBR) and The MSU Intramural Research Grants Program (IRGP).

131

 

References

[1]

[2]

[3]

[4]

[5]

[6]

Y. Poirier, D.E. Dennis, K. Klomparens, C. Somerville, Polyhydroxybutrate, a
biodegradable thermoplastic, produced in transgenic plants, Science 256 (1992)

520-523.

C. Nawrath, Y. Poirier, C. Somerville, Targeting of polyhydroxybutyrate
biosynthetic pathway to the plastids of Arabidopsis thaliana results in high level of
polymer accumulation, Proc. Natl. Acad. Sci. USA 91 (1994) 12760-12764.

D.R. Corbin, R.J. Grebenok, T.E. Ohnmeiss, J .T. Greenplate, J .P. Purcell,
Expression and chloroplast targeting of cholesterol oxidase in transgenic tobacco

plants, Plant Physiology 126 (2001) 1116-1128.

L. Comai, N. Lasrson-Kelly, J. Kiser, C. Mau, A.R. Pokalsky, C.K. Shewmaker, K.
McBirde, A. Jones, D.M. Stalker, Chloroplast transport of ribulose bisphosphate
carboxylase small subunit-S-enolpyruvyl 3-phosphoshikimate synthase chimeric

protein requires part of the mature small subunit in addition to the transit peptide,

The Journal of Biological Chemistry 263 (1988) 15104—15109.

G. Van den Broeck, M.P. Timko, A.P. Kausch, A.R. Cashmore, M. Van Montagu,
L. Herrera-Estrella, Targeting of a foreign protein to chloroplast by fusion to the

transit peptide from the small subunit of ribulose 1,5-bisphosphate carboxylase,

Nature 313 (1985) 358—363.

T.H. Lubben, K. Keegstra, Efﬁcient in vitro import of a cytosolic heat shock protein

into pea chloroplasts, Proc. Natl. Acad. Sci. USA 83 (1986) 5502-5506.

132

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

TH. Teeri, G.K. Patel, K. Aspegren, V. Kauppinen, Chloroplast targeting of
neomycin phosphotransferase II with a pea transit peptide in electroporated barely

mesophyll protoplasts, Plant Cell Report 8 (1989) 4 187-190.

N. Raikhel, M.J. Chn'peels, Protein sorting and vesicle trafﬁc. In: B. Buchanan, W.
Gruissem, R. Jones, (Eds.), Biochemistry & Molecular Biology of Plants,

American Society of Plant Physiologist, 2000, pp. 160-201.

L. Spremulli, Protein synthesis, assembly, and degradation, In: B. Buchanan, W.
Gruissem, R. Jones, (Eds.), Biochemistry & Molecular Biology of Plants,
American Society of Plant Physiologist, 2000, pp. 412—454.

B. De Cosa, W. Moar, S.B. Lee, M. Miller, H. Daniel], Overexpression of the Bt
cry2Aa2 operon in chloroplasts leads to formation of insecticidal crystals, Nature

Biotechnol. 19 (2001) 71-74.

Z. Svab, P. Hajdukiewicz, P. Maliga, Stable transformation of plastids in higher
plants, Proc Natl Acad Sci USA 87 (1990) 8526-8530.
R. Bock, R. Hagemann , Plastid genetics: manipulation of plastid genomes and

biotechnological applications, Prog. Bot. 61 (1999) 76-90.

H. Zhong, C. Srinivasan, M.B. Sticklen, In-vitro morphogenesis of corn (Zea mays
L.). I. Differentition of multiple shoot clumps and somatic embryos from shoot tips,

Planta 187 (1992) 483-489.

H. Zhong, B. Sun, D. Warkentin, S. Zhang, R. Wu, T. Wu, M.B. Sticklen, The
competence of maize shoot meristems for integrative transformation and inherited

expression of transgenes, Plant Physiology 110 (1996) 1097- 1107.

133

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

D. Xu, X. Duan, B. Wang, B. Hong, T.D. Ho, R. Wu, Expression of a late
embryogenesis abundant protein gene hvaI, from barley confers tolerance to water

deﬁcit and salt stress in transgenic rice, Plant Physiology 110 (1996) 249-257.

C.A. Lee, M.H. Saier, Use of cloned mtl genes of Escherichia coli to introduce mtl
deletion mutations into the chromosome, Journal of Bacteriology 153(2) (1983)

685-692.

X. Deping, X. Duan, B. Wang, B. Hong, T.H.D. Ho, R. Wu, Expression of a late
embryogenesis abundant protein gene, HVAl, from barley confers tolerance to

water deﬁcit and salt stress in transgenic rice, Plant Physiol. 110 (1996) 249-257.

T. Murashige, F. Skoog, A revised medium for rapid growth and bioassays with

tobacco tissue cultures, Physiol. Plant. 15 (1962) 473-497.

K. Edwards, C. Johnstone, C. Thompson, A simple and rapid method for the
preparation of plant genomic DNA for PCR analysis, Nucl. Acids. Res. 19 (1991)

15-29.

8.0. Rogers, A.J. Bendich, Extraction of DNA from milligram amount of fresh,

herbarium, and mummiﬁed plant tissues, Plant Mol. Biol. 5 (1985) 69-76.

E.M. Southern, Detection of speciﬁc sequence among DNA fragments separation

by gel electrophoresis, J. Mol. Biol. 98 (1975) 503-517.

J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular cloning laboratory manual

secound edition, 1989.

134

[23]

[24]

[25]

[26]

[27]

M. Bradford, A rapid and sensitive method for quantitation of microgram
quantities of protein utilizing the principle protein-dye binding, Anal. Biochem. 72

(1976) 248-254.

S. Roderrnel, Isolation of maize chloroplasts and chloroplast DNA, In: M. Feeling,
V. Walbot, (Eds.), The Maize Handbook, Ed 1. Springer Verlag, New York, 1994,

pp. 556-558.

H. Zhong, S. Zhang, D. Warkentin, B. Sun, T. Wu, R. Wu, M.B. Sticklen, Analysis
of the functional activity of the 1.4-kb 5’-region of the rice actin 1 gene in stable

transgenic plants of maize (Zea mays L.), Plant Science 116 (1996) 73-84.

S. Herminghaus, P.H. Schreier, J.E.G. McCarthy, J. Landsmann, J. Botterrnan, J.
Berlin, Expression of a bacterial lysine decarboxylase gene and transport of the

protein into chloroplast of transgenic tobacco, Plant Mol. Biol. 17 (1991) 475-486.

X. Yong, P. Zeguo, C. Yixin, W. Rui, Cloning and analysis of histone 3 gene and
rubisco small subunit gene of rice, Scientia Sinica (Series B) XXX (7) (1987) 706-

719.

[28] RM. Ewing, The Rubisco small subunit gene family in the C4 plant maize:

structure and expression, Department of Plant Biology. Oxford, Oxford University

(1996).

135

Appendix A

Effects of Ammonia Fiber Explosion (AFEX) on activity of

Rubisco

136

Effects of Ammonia Fiber Explosion (AF EX) on activity of Rubisco

Abstract

Ammonia ﬁber explosion treatment has shown a great potential to be used in
integrated processing that allows more protein extraction from biomass as well as
increasing the digestibility of biomass compared to untreated material. It would be
beneﬁcial to determine whether AF EX treatment affects the native functional properties
of the extracted proteins. In this research we have chosen ribulose-1,5-bisphosphate
carboxylase/oxygenase (Rubisco) enzyme from alfalfa and tobacco plant as our model
protein to investigate the effects of AFEX treatment on the functional properties of
protein. These effects were evaluated by measuring and comparing the activity of

Rubisco enzyme in AFEX treated and untreated alfalfa and tobacco samples.

Introduction

Production of fuel and chemicals from lignocellulosic biomass such as
agricultural crops and residues, has received a great deal of attention. These
lignocellulosic materials have also been investigated as potential animal feedstuffs. A
main obstacle in utilizing lignocellulosics for either feedstuffs or fuel/chemical
production is the relatively unreactive nature of cellulose. Many biomass pretreatments
have been used to increase the digestibility of the biomass to improve feed quality and
also increase the yield of fermentable sugars in biomass enzymatic hydrolysis (Weaver,
1998; Gollapalli, 200]). In addition to cellulose, hemicellulose and lignin, biomass

contains protein. Protein is potentially the most valuable fraction of biomass (De La Rosa

137

 

 

et al. 1994); therefore it provides an important byproduct credit in biomass reﬁning.
However, little attention has been focused on protein recovery. Plant proteins are
comparatively fragile component of biomass; their potential value can be decreased by
harsh processing steps. A fully integrated biomass process, which improves protein
recovery while preserving protein native functional properties and increasing the
conversion of cellulose and hemicellulose to fermentable sugars may signiﬁcantly
enhance biomass process economics. Unfortunately most of the biomass pretreatments
operate under harsh conditions (e.g., steam explosion and various acid processes) that
tend to degrade the proteins. However, the AF EX process operates under relatively mild
conditions and has shown good potential for protein recovery from herbaceous crops with
simultaneous conversion of cellulose and hemicellulose to simple sugars (De La Rosa el
al. 1994). However, the quality of the recovered protein from AF EX treated biomass, in
terms of their native functional properties such as enzyme activity has never been
examined.

The predominant protein in leaf tissue is Rubisco, the key enzyme in
photosynthesis and photorespiration. Rubisco catalyses the initial reactions of both the
reductive pentosephosphate and photorespiration cycles. Rubisco is the most abundant
soluble protein in the chloroplast and probably the most abundant protein on Earth
(Malkin and Niyogi, 2000). Therefore, in this report we have chosen Rubisco as our plant
protein model to investigate the potential of AFEX treatment in recovering bioactive
protein from biomass. This potential was evaluated by measuring and comparing the

activity of Rubisco enzyme in AFEX treated and untreated alfalfa and tobacco samples.

138

 

Material and Methods
Plant material

Dried alfalfa plant materials (harvested in north central Kansas in spring 2001)
were provided by Wenger manufacturing Company. Plant materials were AF EX treated
as described in chapter 3. The treated samples were leﬁ in a fume hood overnight to
evaporate the residual ammonia. The next day the treated and untreated samples were
ground and sieved to 40 mesh to ensure well mixed samples. Samples were kept in sealed
plastic bag in the refrigerator for ﬁrrther analysis.

In addition to alfalfa, we also studied Rubisco in tobacco plants. Tobacco plants
were grown, harvested and dried at the Crop and Soil Sciences greenhouses at Michigan
State University. The dried tobacco leaves were treated by the AFEX process and stored
as described above. The AFEX process conditions in alfalfa and tobacco treatment are

summarized in Table A. 1.

139

 

Table A.1. Summary of AF EX treatment conditions for alfalfa and tobacco plants

In all the runs ammonia loading ratio was 1:1

 

 

 

 

 

 

 

 

 

 

 

Sample# Temperature Moisture Content (%) Plant material
1 Untreated Alfalfa
2 60°C 60 Alfalfa
3 70°C 60 Alfalfa
4 80°C 60 Alfalfa
5 90°C 60 Alfalfa
6 Untreated Tobacco
7 60°C 60 Tobacco
8 70°C 60 Tobacco
9 80°C 60 Tobacco
1 0 90°C 60 Tobacco

 

 

 

 

 

 

140

Chemicals

All the required chemicals for Rubisco extraction, Rubisco activity and Enzyme
Linked lmmunosorbent (ELISA) assay were provided from Sigma (St. Louis, MO)
Rubisco activity assay

Rubisco was extracted from plant materials according to Metodiev and
Demirevska-Kepova (1992) with some modiﬁcations and Rubisco activity was measured
as the rate of I“C incorporation into acid-stable products in a 1 min assay at 25°C
according to Catt and Millard, (1988).

Fifty milligram of AFEX-treated and untreated alfalfa samples were ground to a
ﬁne powder with mortar and pestle. The powder was homogenized with the aid of a
power drill using 1.5 ml microcentrifuge tubes and plastic pestles (Kontes), in 4 m1
extraction buffer (0.1 M Tris-HCI (pH 7.8), 0.1 mM EDTA, 1.5% polyvinyl-pyrrolidone-
4OT, 5 mM 2-mercaptoethanol, 1 mM phenylmethylsulfonylfonylfuoride (PMSF)) for
one minute. The extraction buffer was prepared to be COz-free by gently boiling the
distilled water for 15 min prior to addition of reagents, all the solutions were also bubbled
with nitrogen gas to make sure the system is 02-free. The homogenate was ﬁltered
through two layers of cheesecloth. The extract was brought to 37% saturation with solid
ammonia sulfate (226g/liter of extract) and after standing for 1 hour was centriﬁrged at
9000 g for 30 min. The supernatant solution was brought to 50% saturation with solid
ammonia sulfate (92.5g/1iter). After centrifugation as described above, the precipitate was
dissolved in 1 ml of extraction buffer. The extracted Rubisco was used in activity assay
and ELISA assay. The concentration of total soluble protein in each sample was

measured according to Bradford (1976) using bovin serum albumin as the standard. Total

141

activity was measured by adding an appropriate volume of each sample (equal to 50pg of
total protein) to a sufﬁcient volume of the activation mixture consisting of 100 mM
NaHCO; and 200 mM Mng to bring to a ﬁnal concentration to 10 mM NaHCO3 and 20

mM MgC12. The mixture was incubated for 15 min at 25°C to activate any unactivated

Rubisco. Then 50 pl of activated enzyme solution was added to 450 u] of assay buffer
consisting of 0.1 M Tris-HCI (pH 8), 20 mM MgC12, 1 mM EDTA, l3mM NaH'4CO3 and
0.5 mM ribulose biphosphate at 25°C. The reaction was terminated after 60 s with 100 p1
of 2 N HCI. The samples were dried then acid-stable I"C-radioactivity determined with a
1500 TRI CARB liquid scintillation counter (PACKARD, Downer Grove, IL). The
activity was determined based on the comparison of the readings with the standard curve
that was made for different concentration of NaH“CO3,

ELISA assay

An enzyme linked immunosorbent assay (ELISA) has been used to measure
Rubisco concentration. The assay relied upon Rubisco in crude extract of plant material
competing with pure Rubisco for a solid phase antibody, so the antigen was detected by a
decrease in reaction product over a high background. ELISA method provides accurate
and easy determination of the quantity of Rubisco in plant leaves (Catt and Millard,
1988).

In this study ELISA assay was conducted as described by Catt and Millard,
(1988). The Rabbit anti-Rubisco antiserum used in this assay was provided by Dr.
Sticklen, Michigan State University. The amount of Rubisco in the leaf extract was
calculated using a standard curve, which was plotted using pure Rubisco (from Sigma)

diluted in PBS.

142

Result and Discussion

In all the experiments, untreated samples were used as controls and were run
parallel to the other samples. The results are demonstrated in the following ﬁgures; all
data are the mean of two replicates. In these ﬁgures the activity is reported as the amount
of I4C incorporation into acid-stable products in 1min.

Figure A.1 presents the activity of Rubisco in SOug of total soluble protein

extracted from AFEX treated alfalfa samples along with the untreated sample.

Figure A.1. Rubisco activity in AF EX-treated alfalfa

 

P
5

Conditions:
60% Moisture content (dwb)
Ammonia loading ratio 1:1

 

P
u
an
HH

 

H4

 

H4

 

 

 

P
u

 

 

 

 

 

 

P
N
a:

 

 

 

 

 

 

P
d
m
H

 

 

 

 

 

P
d
I]

 

 

nmole 0714c fixed in one minute
per 50 ug of total soluble protein
0
N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Control alfatfa 60°C 70°C 80°C 90°C

Treatment Temperature

143

As this ﬁgure demonstrates, for the ﬁrst two runs slight decreases (5 and 8%,
respectively) were observed in Rubisco activity. But as the temperature rose to 80 and
90°C the activity dropped by 57 and 72%, respectively. These large drops in activity
might be caused by both high temperature and the longer time that the AFEX unit had to
be heated to reach to those high temperatures. Apparently these factors have damaged the
protein.

As the result of the intense disruptive effect of the AFEX treatment on the cell
walls of plant, the AFEX treatment has been able to improve the protein recovery from
plant material (De La Rosa et al., 1994). Therefore, to properly integrate above results we
had to distinguish between increased protein extraction and stability of the activity. In
fact, part of the observed activity in Figure A.l might be due to the presence of a larger
amount of Rubisco in the 50pg of total soluble protein that were subjected to the activity
assay, and not an indication of the Rubisco activity retention. To clarify this issue we
measured the amount of Rubisco in the samples by the ELISA method and then an
appropriate volume of the leaf extract to equal 25pg of Rubisco was subjected to the
activity assay. Even for the AF EX-treated samples that the retention of the Rubisco
activity was very low we were able to measure the amount of the Rubisco by ELISA
assay, apparently enough of the 3 dimensional structure of Rubisco had remained to react

with ELISA assay.

144

 

As Figure A.2 shows, 25ug of Rubisco extracted from alfalfa treat at 60°C, and
70°C (with 60% moisture and 1:1 ammonia loading) lost about 12% and 24% of the
activity respectively compared to 25ug of Rubisco extracted from untreated alfalfa.
Whereas, the activity assay of SOpg of total protein extracted from the same samples
showed only 5 and 8% lost of activity respectively compared to the SOug of total protein
from untreated sample (Figure 1). These differences clariﬁed that in fact, part of the
observed activity in Figure l, was due to the presence of a larger amount of Rubisco in
the 50pg of total soluble protein and actual percentage of Rubisco activity retention for
alfalfa samples treated at 60°C or 70°C with 60 % moisture, 1:] ammonia loading ratio

were about 88% and 76%, respectively.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5
0.45 Conditions: —
8 o 4 IL 60% Moisture content (dwb) H
:3: ' I Ammonia loading ratio 1:1
.. 1
E a 0.35
O o I
s 2 1
.5 E 0.3
E "6 0 25
c s
g :3 0.2
g a 0.15
g I
i: 0.1 4.
0.05
0
control 60°C 70°C 80°C 90°C

Treatment Temperature

Figure A.2. Activity of 25 pg of Rubisco from AFEX-treated alfalfa.

145

Figure A.3 presents the activity of 25 pg of Rubisco extracted from AFEX treated
and untreated tobacco samples. This ﬁgure also shows the same trend as observed in
Figure A.2, indicating that higher percentage of Rubisco activity survived at lower

temperature vs higher temperature.

Figure A.3. Activity of Rubisco in AF EX-treated tobacco

 

 

Conditions:
60% Moisture content (dwb) —-
Ammonia loading ratio 1:1

 

NH

 

 

 

 

 

 

 

 

 

P
b
w

 

 

 

 

 

 

 

P
N
H

 

 

 

nmole of 14C fixed in one minute
per 25ug of Rubisco
O
u

 

 

 

 

 

 

p
a

 

 

 

 

 

 

 

 

 

 

 

 

 

Control tobacco 60°C 70°C 80°C 90°C

Treatment Temperature

146

 

Conclusion

Assuming 100% activity for untreated sample, the percentage of Rubisco activity
retention for alfalfa samples treated at 60°C or 70°C with 60 % moisture, 1:1 ammonia
loading ratio was about 88% and 76%, respectively. The enzymatic hydrolysis results
(after 168hr with 60 FPU/g of glucan) of corn stover treated at the same conditions
showed 64% and 68% glucan conversion respectively versus 29% glucan conversion for
untreated corn stover (chapter 3). Based on the results of this chapter, chapter 3 and
previous work (De La Rosa et al., 1994) we believe that the AFEX treatment, under at
least some conditions, has potential as an integrated pretreatment to increase the
digestibility of the biomass, to improve extraction of protein from the biomass and at the
same time to preserve plant protein in its bioactive form.

Thus there is a legitimate reason to hope that industrial enzymes such as
cellulases, which produced in transgenic plants, can also survive the AF EX pretreatment,
and be recovered in their useable from during biomass reﬁning. This possibility was
explored in chapter 4, by treating transgenic tobacco plants, expressing cellulase, with

ammonia ﬁber explosion process.

147

 

Appendix B

HPLC and GC calibration curves

148

 

High Performance Liquid Chromatography (HPLC)

The sugars were analyzed in a BioRad (Richmond, CA) High Performance Liquid
Chromatograph using an Aminex HPX87P column (HPLC Carbohydrates Analysis
Column) at 85°C and a BioRad Deashing Cartridge as a guard column. The mobile phase
used was degassed HPLC water at a ﬂow rate of 0.6 mllmin. The injection volume used
was 20 ILL and the run time was 20 minutes.

Standard solutions of pure sugars: glucose, xylose, mannose, galactose, cellobiose and
arabinose were individually run on the HPLC to determine their retention time (Table

B. l) and calibrate the system.

Table B.l. Retention time of sugars on the HPLC

 

 

 

 

 

 

 

Sugars Retention time, min
Cellobiose 10.416
Glucose 12.766
Xylose 13.866
Galactose 14.65
Arabinose 15.8
Mannose 16.73

 

 

 

 

149

 

Area Count

Area Count

14

12

10

30

25

20

15

10

Figure B.1. Calibration curve for celloboise

 

y = 1 .3374x
R2 = 1

 

 

 

/

 

/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concentration gll

150

2 3 4 5 6 7 8 10
Concentration gll
Figure 8.2. Calibration curve for glucose
y = 1.203”
R2 = 0.9999
5 10 15 20 25

 

Area Count

Area Count

Figure B.3. Calibration curve for xylose

 

 

 

 

 

 

 

 

 

 

30
y = 1.9533):
n’ = 0.9909
25
20 //
15
10
5
o I l l I Y I 1 T 1
0 2 4 8 8 10 12 14 18 18
Concentration gll
Figure B.4. Calibration curve for galactose
14

 

 

2 -
" y
10

 

 

 

 

 

 

 

 

 

6

4

2

0 r . . . f .
0 1 2 3 4 5 6

Concentration gll

151

 

 

Area Count

Area Count

14

12

10

14

12

10

Figure B.5. Calibration curve for arabinose

 

y = 1 .5516x
a” = 0.9994

 

 

/

 

 

 

 

 

 

 

 

2 3 4 5 c 7
concentration gll

Figure B.6. Calibration curve for mannose

 

 

 

 

 

 

 

 

 

3 4 5 6 7 0
Concentration gll

152

 

 

 

Area Count

Gas Chromatography (GC)

The fermentation samples were analyzed for ethanol by gas chromatography using model
GC 17 (Shimadzu). The injection temperature was 240°C and the detector temperature
was 255°C. The column was ﬁrst maintained at 80 °C up to 3 min followed by a

temperature program at 15°C/min up to 125 °C for 6 min. The carrier gas was helium and

ethanol was used as external standard for calibration (Figure B.7).

Figure B.7. Calibration curve for ethanol (retention time was 1.66min)

 

 

 

 

 

 

 

 

2500
y = 197.24X
n’ = 0.9990

2000 /

1m /

500
0 r I . r .
0 2 4 6 8 10 12

Ethanol Concentration gll

153

BIBLIOGRAPHY

154

 

BIBLIOGRAPHY

Abraham M. and Kurup GM. (1997). “Pretreatment studies of cellulose wastes for

optimization of cellulase enzyme activity”. Applied Biochemistry and Biotechnology,
62(2-3): 201-21 1.

Austin 8., Koegel R.G., Matthews D., Shahan M., Strau RI. and Burgess R. (1994). “An
overview of a feasibility study for the production of industrial enzymes in transgenic
alfalfa” Ann. NY Acad. Sci. 721:234-244.

Baker J.O., Thomas S.R., Andy W.S., Nieves RA. and Himmel ME. (1994). “The
cellulase synergistic effect: binary and ternary system” Presented at the symposium on
enzyrnic degradation of insoluble polysaccharides, The annual American chemical
society meeting, San Diego, CA.

Barry A.J., Peterson EC. and King A.J. (1936). J. Amer. Chem. Soc. 58: 333

Boston R.S., Viitanen RV. and Vierling E. (1996). “ Molecular chaperones and protein
folding in plants” Plant Mol. Biol. 32: 191-222.

Bradford M. (1976). “ A rapid and sensitive method for quantitation of microgram
quantities of protein utilizing the principle protein-dye binding” Anal. Biochem. 72:248-
254.

Breuil C., Chan M. and Saddler J .N. (1990). “Comparison of the hydrolytic activity of
commercial cellulase preparation” Applied Microbiology and Biotechnology 34: 31-35.

Campbell N.A., Reece J.B. and Mitchell LG, (1999). Biology ﬁfth edition,
Benjamin.Cummings pp 102-205.

Carrasco J.B., Saiz M.C., Navarro A., Soriano P., Saez F. and Martinez J.M. (1994).
“Effects of dilute acid and steam explosion pretreatments on the cellulose structure and

kinetics of cellulose fraction hydrolysis by dilute acids in lignocellulosic materials.”
Appl. Biochem. Biotechnol., 45/46: 23.

155

 

Carroad P.A., Anaya-Serrano H., Edwards RH. and Kohler GO. (1981). “ Optimization
of Cell Disruption for alfalfa Leaf Protein Concentration (Pro-Xan) Production” J. Food
Sci. 46(2): 383-386.

Catt J. W. and Millard P. (1988). “The measurement of Ribulose 1,5-bisphosphate
Carboxylase/Oxygenase concentration in the leaves of potato plants by enzyme linked
immunosorbtion assay”, journal of Experimental Botany 39(199):157-164.

Chou Y.T. (1986). “Supercritical ammonia pretreatment of lignocellulosic materials”,
Eighth Symposium on biotechnology for fuels and chemicals, 17:19-32.

Conard U. and Fiedler U. (1998). “ Compartment-speciﬁc accumulation of recombinant
immunomodulation of physiological functions and pathogen activity” Plant Mol. Biol.
38: 101-109.

Coughland MP. and Ljungdahl LG. (1988). “In Biochemistry and Genetics of Cellulose
Degradation” edited by J .P. Aubert, P. Beguin, and J. Millet. New York: Academic
Press, pp. 11-30.

Dai Z., Hooker B.S., Anderson DB. and Thomas SR. (2000). “ Improved plant-based
production of El endoglucanase using potato: expression optimization and tissue
targeting” Molecular Breeding 6: 277-285.

Dai Z., Hooker B.S., Quesenberry RD. and Gao J. (1999). “ Expression of Trichodenna
reesei Exo-cellobiohydrolase I in transgenic tobacco leaves and calli” Applied
Biochemistry and Biotechnology 77/79: 689-699.

Dale BE. and Moreira M.J. (1982). “A Freeze Explosion Technique for Increasing
Cellulose Hydrolysis” Biotech. Bioeng. Symp. 12:31-43.

Dale BE. and Moreira M.J. (1983). “A freeze explosion technique for increasing
cellulose hydrolysis” Biotechnol. Bioeng. Symp. NO 12:31.

Dale B.E., Henk LL. and Shiang M. (1985). “ Fermentation of Lignocellulosic Materials
Treated by Ammonia Freeze Explosion” Dev. Ind. Microbiol. 26:223-233.

De la Rosa L.B., Dale B.E., Reshamwala S.T., Latimer V.M., Stuart ED. and Shawky
B.T. (1994). Applied Biochemistry and Biotechnology 45/46: 483-497.

156

Davis D. (1987). “Prevention Reference Manual: chemical speciﬁc”, Volume 4: control
of accidental releases of ammonia (SCAQMD), US. Environmental protection Agency,
Research Triangle park.

Goddijn OJM. And Pen J. (1995). “ Plants as bioreactors” Trends Biotechnol 13:379-387.

Gollapalli L. E. (2001), “Predicting digestibility of ammonia ﬁber explosion (AFEX)
treated rice straw” M.S. thesis, Chemical Engineering Department, Michigan State
University, East Lansing, Michigan

Graybum W.S., Collins GB. and Hildebrand D.F.(1992). “Fatty acid alteration by a A9
desaturase in transgenic tobacco tissue” Bio/technology 10: 675-678.

Grohmann K., Wyman CE. and Himmel ME. (1992), “Potential for Fuels from Biomass
and Wastes” in Emerging Technology for Materials and Chemicals from Biomass. Eds.
R.M. Rowell; T.P. Schultz and R. Narayan, American Chemical Society, pp350-392.

Halliwell N. and Halliwell G. (1995). “Biotechnological aspects of lignocellulose and
biomass degradation” Outlook on Agriculture 24(4): 219-225.

I-Iinman N .D., Wright J.D., Hoagland W. and Wyman CE. (1992). “Xylose fermentation,
an economic analysis” Appl. Biochem. Biotech. 20121: 391

Holtzapple M.T., Jun J., Ashok G., Patibandla S.L. and Dale BE. (1991). “The Ammonia
Freeze Explosion (AFEX) Process. A Practical lignocellulose Pretreatment” Appl.
Biochem. Biotech. 28/29: 59-74.

Holtzapple M., Cognate M., Shu Y. and Hendrickson C. (1990). “Inhibition of
Trichodenna reesei cellulase by sugars and solvents” Biotechnology and Bioengineering
36: 275-287.

Ishihara M., Uemura S., Hayashi N. and Shimizu K. (1991). “Sernicontinuous enzymatic
hydrolysis of lignocellulosics” Biotechnology and Bioengineering 37: 948-954

Kidd, G. and Devorak, J. (1994). “Treholose is a sweet target for agbiotech”
Bio/Technology 12: 1328-1329.

157

 

Kirk-Othmer D. (1992). “Encyclopedia of Chemical Technology” 4th edition, John Wiley
and Sons, New York.

Lamptey J ., Moo-Young M., and Robinson C.W., (1986). “Pretreatment of
lignocellulosic for bioconversion applications: process options” in Biotechnology and
renewable energy (Edited by Moo-Young M., Laptey 1.), pp 46-56. Elsevier applied
science publishers, London and New York.

Lewin M. and Roldan LG. (1971). “The effect of liquid anhydrous ammonia in the
structure and morphology of cotton cellulose” J. Polymer Sci.: Part C 36: 213-229.

Lynd L.R., Elander R.T., Wyman CE. (1996). “Likely features and costs of mature
biomass ethanol technology” Applied Biochemistry and Biotechnology 57/58:?41-61.

Malkin R. Niyogi K., (2000), “Photosynthesis” In Biochemistry and Molecular Biology
of Plants. Edited by Buchanan B., Gruissem W., Joues R. pp 110-158, American Society
of plant physiologists, Rockville, Maryland.

Mandels M. and Stemberg D. (1976).“Recent advances in cellulase technology” Ferment.
Technol. 54: 267-286.

Metodiev M. and Demirevska-Kepova K. (1992). ‘Rubisco quantitation in leaves of
different barley varieties by Enzyme-linked Irnmunosorbent Assay” Journal of
Experimental Botany 43(247): 155-158.

Moniruzzaman M., Dale B.E., Hespell RB. and Bothast R.J. (1997), “Enzymatic
Hydrolysis of High-Moisture Corn Fiber Pretreated by AFEX and Recovery and

Recycling of the Enzyme Complex” Applied Biochemistry and Biotechnology 67:113-
126.

Murray E., Lotzer J. and Eberle M. (1988). “ Codon usage in plant genes” Nucleic Acids
Research 17(2): 477-498.

O’Conner J. J. (1972). “Ammonia Explosion Pulping, a new ﬁber separation process”
Tappi 55(3): 353-358.

158

Pen J ., Molendijk L., Quax W.J., Sijmons P.C., van Ooyen A.J.J., van den Elzen P.J.M.,
Rietveld K., and Hoekema A. (1992). “Production of active Bacillus Licheniforrnis
alpha-amylase in tobacco and its application in starch liquefaction” Bio/Technology 10:
292-296.

Pen J ., Verwoerd Theo C., van Paridon P.A., Beudeker R.F., van den Elzen, Geerse K.,
van den Klis J .D., Versteegh H.A.J., van Ooyen A.J.J., and Hoekema A. (1993).
“Phytase-containing transgenic seeds as a novel feed additive for improved phosphorus
utilization” Bio/Technology 11: 81 1-814.

Poirier Y., Nawrath C. and Somerville C. (1995). “Production of polyhydroxyalkanoates,
a family of biodegradable plastic and elastomers, in bacteria and plants” Bio/Technology
13: 142-150.

Robson L.M and Chambliss OH. (1989). “Cellulase of bacterial origin” Enzyme
Microb. Technol. 11: 626-644.

Saddler J.N., Brownell H.H., Clermont LP. and Levitin N. (1982). “Enzymatic
hydrolysis of cellulose and various pretreated wood fractions” Biotechnology and
Bioengineering 24: 1389-1402.

Sambrook J ., Fritsch EF. and Maniatis T. (1989). Molecular cloning: A laboratory
Manual, 2ml ed., Cold Spring Harbor Laboratory Press, Plainview, NY.

Sander D. and Bethke P. (2000). “Membrane Transport”, In Biochemistry and Molecular
Biology of Plants. Edited by Buchanan B., Gruissem W., Joues R. pp 110-158, American
Society of plant physiologists, Rockville, Maryland

Schell DJ. and Duff B. (1996). Review of pilot plant programs for bioethanol

conversion, in Handbook on Bioethanol production and utilization, edited by C.E.
Wyman, PP 381-394.

Shoemaker 8., Watt k., Tsitovsky G., and Cox R. (1983). “Characterization and
properties of cellulases puriﬁed from Trichoderma ressei strain L27” Biotechnology 1:
687-690.

Spindler DD, Wyman CE., Grohmann K., Mohagheghi A. (1988), “Simultaneous
sacchariﬁcation and fermentation of pretreated wheat straw to ethanol with selected yeast
strains and B-glucosidase supplementation. Appl. Biochem. Biotechnol. 21:529-540.

159

Spindler DD, Wyman CE., Grohmann K., Philippidis GP.(1992), “ Evaluation of the
cellobiose-fermenting yeast Brettanomyces custersii in the simultaneous sacchariﬁcation
and fermentation of cellulose. Biotechnol. Lett. 14(5): 403-407.

Spremulli L. (2000), “ Protein Synthesis, Assembly, and Degradation”, In Biochemistry
and Molecular Biology of Plants. Edited by Buchanan B., Gruissem W., Joues R. pp 110-
158, American Society of plant physiologists, Rockville, Maryland

Stark, D., Timmermann, K.P., Barry, G. F., Preiss, J. and Kishore G. M. (1992). “
Regulation of the amount of starch in plant tissues by ADP glucose purophorylase”
Science 258: 287-292.

Tan L.U.L., Yu E.K.C., Mayers P., and Saddler J.N. (1986). “Column cellulose
hydrolysis reactor: cellulase absorption proﬁle” Applied Microbiology and
Biotechnology 25: 256-261.

Tarchevsky I. A. and Marchenko G. N. (1991). “Cellulose: Biosynthesis and Structure”
Springer-Verlag New York, NY, pp 9-31.

Thomas S.R., Laymon R.A., Chou Y.CH., Tucker M.P., Vinzant T.B., Andy W.S., Baker
J .O., Nieves R.A., Mielenz J .R. and Himmel ME. (1995). “ Initial approaches to artiﬁcial
cellulase systems for conversion of biomass to ethanol” in Enzymatic degradation of
insoluble carbohydrates, edited by J .N . Saddler and M.H. Penner pp 208-236.

Torget R., Himmel M.E., Wright J .D. and Grohmann K. (1988). “Initial design of a dilute
sulfuric acid pretreatment process for aspen wood chips” Appl. Biochem. Biotechnol.
17/18: 89-104.

Verwoerd T.C., van Paridon P.A., van Ooyen A.J.J., van Lent J.W.M., Hoekema A. and
Pen J ., (1994). “ Stable accumulation of Aspergillus niger phytase in transgenic tobacco
leaves. Plant Physiology 109:1 199-1203.

Walter A. (2000). “New technologies for modern biomass energy carriers”, in Industrial
uses of biomass energy, edited by Rosillo-Calle F., Bajay S.V. and Rothman H., pp 200-
253, Taylor & Francis.

Wang P. Y., Bolker H. I., and Pruves C. B. (1967). “Uronic acid ester groups in some
softwoods and hardwoods” Tappi 50(3): 123-124.

160

Weaver J. (1998), “Extrusion processing for ammonia ﬁber explosion (AFEX)” M.S.
thesis, Chemical Engineering Department, Michigan State University, East Lansing,
Michigan

Wilke C.R., Maiorella B., Sciamanna A., Tangnu K., Wiley D., Wong H., (1983). “
Theory of enzymatic hydrolysis in Enzymatic hydrolysis of cellulose” pp 3-15.

Wyman CE. 1995. Economic fundamentals of ethanol production from lignocellulosic
biomass. In Enzymatic Degradation of Insoluble Carbohydrates, ACS Symposium Series
618.

Wyman CE., Spindler DD.,Grohmann K., Lastick SM.(1986). “Simultaneous
sacchariﬁcation and fermentation of cellulose with the yeast Brettanomyces clausenii.
Biotechnology Bioenergy 17: 221-238.

Wyman Ch.E. (1996). “Ethanol production from lignocellulosic biomass: overview” in
handbook on Bioethanol: Production and Utilization, edited by Wyman Ch.E., Taylor and
Francis, Washington DC.

Wyman Ch.E. (1999).“Biomass ethanol: Technical progress, opportunities, and
commercial challenges”, Annu. Rev. Energy Environ. 24: 189-226.

Yong X., Zeguo P., Yixin C., Rui W. (1987). “Cloning and analysis of histone 3 gene and
rubisco small subunit gene of rice” Scientia Sinica (Series B) XXX (7): 706-719.

Ziegelhoffer T., Raasch A. and Austin-Phillips S. (2001). “Dramatic effects of truncation

and sub-cellular targeting on the accumulation of recombinant microbial cellulase in
tobacco” Molecular Breeding 8: 147-158.

161

      

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

lam/ill