! IS!
$4M plan."-
1, "Chi?“ (4.: ‘

. 332.3

 

 

 

 

.v, . .v L ..F. . .5. 5.2 . .
. Assrmégi L319 ii .412}... p .. , = i,

This is to certify that the
dissertation entitled

GENOMIC APPROACHES TO HEAR1WOOD FORMATION
IN HARDWOOD TREE SPECIES,
BLACK LOCUST (ROBINIA PSEUDOACACIA L.)

presented by

JAEMO YANG

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Plant Breeding and Genetic -

 

 

 

Forestry
r 1 Ma' rofessor’s Signature

x/g/ogz

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c'JCIRCIDatoDuopss-pjs

 

 

GENOMIC APPROACHES TO HEARTWOOD FORMATION IN HARDWOOD
TREE SPECIES, BLACK LOCUST (ROBINIA PSEUDOACACIA L.)

By

J aemo Yang

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Plant Breeding and Genetics - Forestry

2004

ABSTRACT

GENOMIC APPROACHES TO HEARTWOOD FORMATION IN HARDWOOD
TREE SPECIES, BLACK LOCUST (ROBINIA PSEUDOACACIA L.)

By
Jaemo Yang

Trees comprise over 90% of the terrestrial biomass of the earth and serve as a primary
feedstock for biofuel, fiber, solid wood products, and various natural compounds. Most
commercially important tree crops produce heartwood. The presence of heartwood is the
major determining factor for wood quality and influences the way in which specific
woods are utilized. Understanding the molecular mechanisms of heartwood formation is
of great commercial and keen scientific interest. Despite the long history of study on
wood formation, our knowledge of this unique biological process is limited. It is
difficult to experimentally observe the processes in sapwood that lead to heartwood
formation. Forward genetic approaches are hampered by long generation times and
poorly defined tree populations. Comparative molecular genetic studies have limited use
because most model organisms do not undergo secondary woody growth. The changes
during the transition from sapwood to heartwood are complex and involve integration of
many metabolic processes in wood cells. A genomics-based approach provides a unique
opportunity for understanding the molecular biology of heartwood and the processes by
which its extractives are produced and stored.

In order to gain insight on the molecular regulations of heartwood and its extractive
formation, I carried out global examination of gene expression profiles across the trunk
wood of black locust (Robinia pseudoacacia L.) trees. Of the 2,915 expressed sequenced

tags (ESTs) that were generated and analyzed, 55.3% showed no match to known

sequences. Cluster analysis of the ESTs identified a total of 2,278 unigene sets, which
were used to construct cDNA rnicroarrays. Microarray hybridization analyses were then
performed to survey the changes in gene expression profiles of trunk wood. In addition
to the regional gene expression profiles in trunk wood, seasonal gene expression
changes were studied in the sapwood-heartwood transition zones that are in charge of
heartwood formation. Furthermore, in this study, plant allantoinase genes involved in
ureide pathway were investigated; in general, plants have the pathway, and these genes
are differentially expressed in developmental and environmental conditions. Therefore,
the impact of this study is expected to expand our knowledge of heartwood formation far

beyond a single hardwood species.

DEDICATION

This dissertation is dedicated to my wife, Youngmi Kim, my parents, Seungsoon Yang
and Sunsoon Lee, my parents in law, Dongho Kim and Soonja Lee, and my children,

J iwoo and Joanne.

ACKNOWLEDGEMENTS

I specially thank Dr. Kyung-Hwan Han, my advisor and professor, for his
continuous support, expertise, and for the opportunity I had to pursue my graduate
program under his guidance. His patience and generosity in terms of time and resources
were invaluable and helped me throughout the study. I thank also to my other committee
members, Dr. Amy Iezzoni, Dr. Wayne Loescher and Dr. Kenneth Sink for their helpful
comments, encouragement, and positive criticism.

Special thanks go to Dr. Daniel Keathley and Dr. D. Pascal Kamdem, for their
invaluable support, friendship, and help for my research. I extend my sincere thanks to
Dr. Jeff Landgrof for his help with microarray printing, Paul Bloese for help with tree
sampling, Dr. Jae-Heung Ko for help with valuable discussion, and Merilyn Ruthig for
her help with manuscript editing. Many others deserve my acknowledgements and thanks
including Dr. David MacFarlane, Sunchung Park, Costas Prassinos, and colleagues and
staff members of the Department of Forestry at Michigan State University.

Finally, I would like to take the opportunity to say my special thank, love and
gratitude to my wife, my parents, my parents in law, sisters, brothers and my children that

have been there for me and always an continuous source of inspiration.

TABLE OF CONTENTS

 

 

 

 

 

 

LIST OF TABLES - - - vii
LIST OF FIGURES - viii
INTRODUCTION - _ 1
CHAPTER 1 -- _ -- 13
Introduction ................................................................................................................... 14
Materials and Methods .................................................................................................. 16
Results ........................................................................................................................... 22
Discussion ...................................................................................................................... 48
Literature Cited .............................................................................................................. 55
CHAPTER 2 60
Introduction ................................................................................................................... 61
Materials and Methods .................................................................................................. 64
Results ........................................................................................................................... 68
Discussion ...................................................................................................................... 85
Literature Cited .............................................................................................................. 92
CHAPTER 3 - -- - 97
Introduction ................................................................................................................... 98
Materials and Methods ................................................................................................ 101
Results ......................................................................................................................... 108
Discussion .................................................................................................................... 122
Literature Cited ............................................................................................................ 127
CONCLUSION - 131

 

vi

LIST OF TABLES

Table 1-1. The functional classification of EST clones .................................................... 26
Table 1-2. The redundancy of EST clones based on contig analysis ................................ 30
Table 1-3. Summary of up-regulated or down-regulated genes ....................................... 34
Table 1-4. Up-regulated transcripts in the bark and cambial zone ................................... 37
Table 1-5. Up—regulated transcripts in the sapwood ......................................................... 40
Table 1-6. Up-regulated transcripts in the transition zone ................................................ 41
Table 2-1. Summary of correlation coefficients of ratios from TZS versus TZF ............. 69
Table 2-2. Functional classification of up-regulated clones in transition zones ............... 73
Table 2-3. Categorized genes whose expression was up-regulated in summer (TZS). 75

Table 2-4. List of up-regulated genes in transition zones harvested in fall (TZF) ........... 80

vii

LIST OF FIGURES

Figure 1-1. Cross-section of a stem from a mature Robinia pseudoacacia tree ............... 23
Figure 1-2. Functional classification of the ESTS from three libraries ............................ 27
Figure 1-3. Scatter plots of microarray hybridization results ........................................... 32
Figure 1-4. Cluster analysis of expression ratios from SWS or TZS vs. BCS ................. 44
Figure 1-5. Confirmation of microarray Data with antisense northern blot analysis ....... 47
Figure 2-1. Distribution of mean ratios of expression from T28 and TZF samples ........ 72
Figure 2-2. Metabolic pathway leading to phenylpropanoids .......................................... 84
Figure 3-1. Catabolism of ureide in plants ...................................................................... 100
Figure 3-2. Alignment of allantoinase amino acid sequences ........................................ 109
Figure 3-3. Functional complementation of a yeast allantoinase mutant ....................... 111

Figure 3-4. Identification of T-DNA insertional mutants of Arabidopsis allantoinase .. 112

Figure 3-5. Dendrogram and similarity of allantoinases and dihydroorotases ............... 115
Figure 3-6. Differential expression of black locust allantoinase .................................... 117
Figure 3-7. RpALN gene expression in trunk wood of black locust .............................. 120

Figure 3-8. Nitrogen influence on the expression of allantoinase genes ........................ 121

viii

INTRODUCTION

General information.

Sapwood is gradually converted to inactive heartwood as its water columns of
conducting vessels break due to freezing, wind vibration, tension, wood boring insects, and
other factors (Mauseth, 1998). The ultimate fate of the cavity of the broken vessels affects
different properties of wood. Trees adopt a mechanism to seal off the empty columns. The
adjacent wood parenchyma cells undergo numerous metabolic changes to produce and
accumulate in the vessels large quantity of heartwood extractives such as phenolic
compounds, lignin, and aromatic substances. These substances inhibit microbial growth. In
this process, which occurs in late fall in temperate zones, one annual ring is converted to
heartwood each year (Mauseth, 1998).

Formation of heartwood is a form of senescence that is accompanied by a variety of
alterations in metabolic conditions. Although the events of senescence have been studied at
the molecular level during leaf senescence (Miller et al., 1999; Wingler et al., 1998), seed
germination (Cercos et al., 1999), nodule development (Matamoros et al., 1999) and floral
senescence (Breeze et al., 2004), the cell maturation and death events occurring during
heartwood formation have been difficult to study because of the location and timing of the
events. Therefore, analysis of global gene expression patterns during the transition may
offer a powerful means of identifying the mechanisms controlling this process.

In practical sense, the presence of heartwood is a determining factor for wood
quality and influences the utilization of wood in many different ways. It also adversely

affects forest management decisions by making it difficult to predict the quality of wood

available for utilization from forest inventory. Understanding the biology of heartwood
formation can provide to control the contributing factors for its formation and thus to
make choices of the most suitable foresz practices. However, no study on the molecular

biology of heartwood formation has been reported.

Timing of heartwood formation.

In the temperate zones, heartwood formation occurs in late summer to late fall and
at the beginning of dormancy when temperatures are sufficiently high (above 5°C) for the
required cellular reactions (Hillis, 1987). Earlier studies have indicated that heartwood
formation occurs at the time of cambial dormancy in pine (Shain and Mackay, 1973a),
walnut and cherry (Nelson, 1978). Studies of the cytology and coloration of the
extractives suggested that heartwood formation commences in mid-summer and
continues in the fall and winter seasons in sugi (Nobuchi et al., 1984a) and black locust

(Nobuchi et al., 1984b).

Metabolic activity in transition zone.

During the transformation from sapwood to heartwood, cells undergo metabolic
changes that result in increased synthesis of secondary products. These involve the
consumption of storage carbohydrates and their conversion into heartwood substances
such as phenolic compounds (Hillis, 1987; Magel et al., 1991; Magel et al., 1994).
Observations lead to the suggestion that heartwood extractives are synthesized in situ in
the sapwood-heartwood transition zone from the breakdown of starch or from soluble

sugars (Magel et al., 1994; Magel and Hubner, 1997). The transition is tightly connected

with the degradation of storage lipids and the accumulation of hydrolysis products and
intermediates (Hillinger et al., 1996a and b). Light and electron microscopic studies have
found that during sapwood-heartwood transition, storage material such as starch is
consumed (Datta and Kumar, 1987; Nair, 1988; Nobuchi et al., 1987). Starch hydrolyzed
at the transition zone represents a primary source of hydroxycinnamic acid and flavonoid
synthesis (Magel et al., 1994). A period of enhanced metabolic activity has been found at
this zone in heartwood forming species such as Acacia (Baqui and Shah, 1985), black
locust (Magel et al., 1991), oak (Ebermann and Stich, 1985), and walnut (Nelson et al.,
1981). Maximum oxygen consumption was observed in the sapwood adjacent to the
heartwood of black locust, suggesting increased metabolic activity (H611 and Lendzian,
1973). Also, the formation and accumulation of heartwood phenolics coincided with the
transformation of sapwood to heartwood in black locust (Magel et al., 1994). Hillis and
Hasegawa (1963) noted that, 19 days after labeled glucose was administered to the
cambial region of Eucalyptus sieberi, it was converted to labeled extractives formed at
the heartwood periphery. This conversion was also observed in the transition zone of
Sugi (Higuchi et al., 1969).

Several enzymes have shown increased activity in the transition zone. Elevated
levels of phenol oxidizing enzymes have been observed in the transition zone of
Eucalyptus polyanthemos (Hillis and Yazaki, 1973). A marked peroxidase activity was
observed in the transition zone of Eucalyptus elaeophora (W ardrop and Cronshaw, 1962)
and F agus sylvatica (Dietrichs, 1964). In addition, Baqui et a1. (1979) found that
succinate dehydrogenase was significantly active only in the transition zone of Melia

azedarach. Adenosine triphosphatase, which is implicated in many energy-consuming

cellular processes, and lipase were also active in the transition zone. Other enzymes
reported to be highly active in transition zone include catechol oxidase, glucose-6-
phosphate dehydrogenase, malic dehydrogenase, maltase, and amylase (Hillis, 1987).
Furthermore, Magel et a1. (1991) have shown that two key enzymes for flavonoid
biosynthesis (chalcone synthase and phenylalanine ammonia-lyase) are highly active in
the sapwood-heartwood transition zone of black locust. They investigated the seasonal
activities of phenylalanine ammonia-lyase (PAL) and chalcone synthase (CHS) in xylem
of black locust. They found that PAL was active in the youngest wood near the cambium
in April and September but that it was active in the transition zone in all seasons, while
CHS was active only in the intermediate wood. From these results they suggested that
PAL in the youngest woods is involved in lignin biosynthesis but in flavonoid synthesis
in intermediate wood, while CHS is only involved in flavonoid synthesis. In addition, the
stimulus for biosynthesis of robinetin, a flavonoid with strong antimicrobial activity,
apparently occurs in the transition zone in Instsia species (Hillis, 1996).

There are three possible explanations for the increased enzyme activity in the
transition zone of mostly dead cells (Hillis, 1987): 1) the enzymes are produced by the
heartwood inhabiting microorganisms; 2) the enzymes are encoded by host parenchyma
cells and remain active after the host cell death; or 3) the enzymes are made by living
host parenchyma cells after heartwood formation. Shain and Mackay (1973b) provided
indirect evidence indicating that two phenol-oxidizing enzymes were produced by host

parenchyma cells and remain active after host cell necrosis.

Analysis of global gene expression pattern

Plant genomics changes and facilitates the way I acquire and utilize new knowledge
on fundamental biological processes in plants (W albot, 1999). DNA microarray technology
provides a simple and economical way to explore gene expression patterns on a genomic
scale (Brown and Botstein, 1999; DeRisi et al., 1996; Duggan et al., 1999; Heller et al.,
1997; Schena et al., 1996, 1998). The use of DNA microarrays provide not only the global
view of the changes, but also an unique opportunity to identify genes whose expression is
altered during the formation of heartwood and its extractives. Furthermore, the genome-
wide expression data obtained from DNA microarray hybridization can be analyzed using
standard statistical algorithms to arrange genes according to similarity in gene expression
pattern (Eisen et al., 1998). Such cluster analysis and display system facilitate the

identification of transition zone-specific genes and the study of their expression patterns.

Black locust (Robinia pseudoacacia L.)

For this study, I choose to work with black locust (Robinia pseudoacacia L.).
Although it is not the best species of economic importance, black locust represents one of
the most extensively studied hardwood species with regard to heartwood formation. This
multipurpose legume has great potential to become an economic species of choice, given
biotechnological improvement can be made to alleviate a few inherent problems such as
susceptibility to stem borer and stem crookedness. Black locust is the most favorable
model species for the proposed research for several reasons: 1) it is a heartwood-forming
species with clear distinction between distinct sapwood and heartwood. Poplar has been a

popular model hardwood species for many forest biology studies. However, its utility as a

model species is limited for this project as poplar form less heartwood; 2) chemical and
biochemical studies of heartwood formation have been extensively documented in black
locust (Kamdem et a1, 1996); 3) it produces and accumulates large amount of extractives
in its heartwood, which renders characteristic durability of its wood. Black locust is
currently being used as a model for the metabolic engineering study for decay resistance
(Kamdem 1994); 4) black locust has relatively small genome size (637 Mbp) and enable

genetic transformation (Han et al., 1999).

The ureid pathway

Allantoinase (allantoin amidohydrolase, EC 3.5.2.5), present in a wide variety of
bacteria, fungi, and plants, as well as a few animals, such as fish and amphibians,
catalyzes the hydrolysis of allantoin to allantoic acid in ureid metabolism (Noguchi et al.,
1986, Vogels et al., 1966). Ureide metabolism serves different roles and is evolutionarily
distinct in plants, animals, and microorganisms. The ureide pathway in animals functions
primarily in salvage or excretion of nitrogen from purines (Campbell and Bishop, 1970;
Stryer, 1988). In microorganisms, ureide degradation can provide nitrogen from a variety
of sources in the external environment (Cooper, 1980). Ureides have been detected in
various plants (Tracey 1961), and ureide metabolism has been studied on legume species
(Lukaszewski et al., 1992; Bell and Webb, 1995). In addition, Allantoinase has been
isolated from Pseudomonas (Jansenn et al., 1982), from mackerel and frog liver (Noguchi
et al., 1986) and from soybean seed (Webb and Lindell, 1993), and the gene encoding
allantoinase has been cloned from yeast (Buckholz and Cooper, 1991) and from bullfrog

(Hayashi et al., 1994). Yet plant allantoinase gene has not been cloned.

My research hypothesis

1) Ray parenchyma cells at the sapwood-heartwood transition zone undergo a form of
programmed cell daeth.

2) In the transition zone, reserve materials and phtotosynthesis products are converted to
secondary metabolites, which may serve as defense chemicals.

3) Heartwood formation is orchestrated by a number of enzymes invloved in the
breakdown of carbone sorces, secondary metabolite biosynthesis, and senescence.

4) In the transition zone, the genes encoding such enzymes are thought to be induced
during the certain season.

5) Allantoinase gene involved in ureide pathway generally exists in plant kingdom.

6) Nitrogen regulates the expression of allantoinase genes.

Literature Cited

Baqui, S., Shah, J., Pandalai, R., and Kothari, I. (1979). Histocherrrical changes during
transition from sapwood to heartwood in Melia azedarach. Indian J Exp Biol 17:
1032-1037.

Baqui, S., and Shah, J. (1985). Histoenzymatic studies in wood of Acacia auriculiformis
Cunn. during heartwood formation. Holzforshchung 39: 311-320.

Bell, JA, and Webb, MA (1995). Immunoaffinity purification and comparison of
allantoinases from soybean root nodules and cotyledons. Plant Physiol 107: 435-441.

Breeze, E., Wagstaff, C., Harrison, E., Bramke, 1., Rogers, H., Stead, A., Thomas, B., and
Buchanan-Wollaston, V. (2004). Gene expression patterns to define stages of post-
harvest senescence in Alstroemeria petals. Plant Biotech J 2: 155-168.

Brown, PO, and Botstein, D. (1999). Exploring the new world of the genome with DNA
microarrays. Nat Genet 21: 33-37.

Buckholz, R.G., and Cooper, T.G. (1991). The allantoinase (DALI) gene of
Saccharomyces cerevisiae. Yeast 7: 913-923.

Campbell, J.W., and Bishop, SH. (1970). Nitrogen metabolism in molluscs. In JW
Campbell, ed, Comparative Biochemistry of Nitrogen Metabolism, Vol 1. Academic
Press, London, UK, pp 103-206.

Cercos, M., Santamaria, S., and Carbonell, J. (1999). Cloning and characterization of
TPE4A, a thiol-protease gene induced during ovary senescence and seed germination
in pea. Plant Physiol 119: 1341-1348.

Cooper T.G. (1980). Selective gene expression and intracellular compartment: two means
of regulating nitrogen metabolism in yeast. Trands Biochem Sci 5: 332-334.

Datta, S., and Kumar, A. (1987). Histochernical studies of the transition from sapwood to
heartwood in Tectona grandis. IAWA Bull 8: 363-368.

DeRisi, J., Penland, L., Brown, P. 0., Bittner, M. L., Meltzer, P. S., Ray, M., Chen, Y.,
Su, Y. A., and Trent, J. M. (1996). Use of a cDNA microarray to analyse gene
expression patterns in human cancer. Nat Genet 14: 457-460.

Dietrichs, H. (1964). Das Verhalten von Kohlenhydraten bei der. Holzforschung l8: 14-
24.

Duggan, D. J., Bittner, M., Chen, Y., Meltzer, P., and Trent, J. M. (1999). Expression
profiling using cDNA microarrays. Nat Genet 21: 10-14.

Eberrnann, R., and Stich, K. (1985). Distribution and seasonal variation of wood
peroxydase activity in oak (Quercus rubur). Wood Fiber Sci 17: 391-396.

Eisen, M.B., Spellman,‘P.T., Brown, PO, and Botstein, D. (1998). Cluster analysis and
display of genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863-
14868.

Han, K.-H., Gordon, M., and Keathley, D. (1999). Genetic transformation of black locust
(Robinia pseudoacacia L.). In Biotechnology in agriculture and forestry (Y. Bajaj,
ed.). Berlin Hiedelberg New York, Springer-Verlag, pp. 273-282.

Hayashi, S., Jain, S., Chu, R., Alvares, K., Xu, 3., Erfurth, F., Usuda, N., Rao, M.S.,
Reddy, S.K., Noguchi, T., Reddy, J.K., and Yeldandi, A.Y. (1994). Amphibian

allantoinase. Molecular cloning, tissue distribution, and functional expression. J Biol
Chem 269: 12269-12276.

Heller, R. A., Schena, M., Chai, A., Shalon, D., Bedilion, T., Gilmore, J ., Woolley, D. E.,
and Davis, R. W. (1997). Discovery and analysis of inflammatory disease-related
genes using cDNA microarrays. Proc Natl Acad Sci USA 94: 2150-2155.

Higuchi, T., Onda, Y., and Fujimoto, Y. (1969). Biochemical aspects of heartwood
formation with special reference to the site of biogenesis of heartwood compounds.
Wood Res 48: 15-30.

Hillinger, C., Holl, W., and Ziegler, H. (1996a). Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. 1. Radial
distribution of lipid classes. Trees 10: 366-375.

Hillinger, C., Holl, W., and Ziegler, H. (1996b). Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. 11. Radial
distribution of lipases and phospholipases. Trees 10: 376-381.

Hillis, W. (1987). Heartwood and tree exudates. Berlin, New York, Springer-Verlag.

Hillis, W., and Hasegawa, M. (1963). The formation of polyphenols in trees. 1.
Administration of 14C-glucose and subsequent distribution of radioactivity.
Phytochem 2: 195-199.

Hillis, W., and Yazaki, Y. (1973). Wood polyphenols of Eucalyptus polyanthemos.
Phytochem 12: 2969-2977.

Hillis, W. E. (1996). Formation of robinetin crystals in vessels of Intsia species. IAWA J
17: 405-419.

Holl, W., and Lendzian, K. (1973). Respiration in the sapwood and heartwood of Robinia
pseudoacacia. Can J Bot 52: 727-734.

Janssen, D.B., Srnits, RA, and van der Drift, C. (1982). Allantoinase from Pseudomonas
aeruginosa. Purification, properties and immunochemical characterization of its in
vivo inactivation. Biochim Biophys Acta, 718: 212-219.

Kamdem, D., Dawson, A., and Wanli, M. (1996). Polyphenols content of black locust
heartwood by HPLC. In Forest Products Society meeting. Chaleston, SC.

Kamdem, D. P. (1994). Fungal decay resistance of aspen blocks treated with heartwood
extracts. For Prod J 44: 30-32.

Lukaszewski, K.M., Blevins, D.G., and Randall, DD. (1992). Asparagine and boric acid
cause allantoate accumulation in soybean leaves by inhibiting manganese-dependent
Plant Physio]. 99: 1670-1676.

Magel, E., Drouet, A., Claudot, A., and Ziegler, H. (1991). Formation of heartwood
substances in the stemwood of Robinia pseudoacacia L. 1. Distribution of

phenylalanine ammonium-lyase and chalcone synthase across the trunk. Trees 5: 203-
207.

Magel, E., J ay-Allemand, C., and Ziegler, H. (1994). Formation of heartwood substances
in the stemwood of Robinia pseudoacacia L.: 11. Distribution of nonstructural
carbohydrates and wood extractives across the trunk. Trees 8: 165-171.

Magel, E., and Hubner, B. (1997). Distribution of phenylalanine ammonia lyase and
chalcone synthase within trunks of Robinia pseudoacacia L. Bot Acta 110: 314-322.

Matamoros, M. A., Baird, L. M., Escuredo, P. R., Dalton, D. A., Minchin, F. R., Iturbe-
Ormaetxe, I., Rubio, M. C., Moran, J. E, Gordon, A. J., and Becana, M. (1999).
Stress-induced legume root nodule senescence. Physiological, biochemical, and
structural alterations. Plant Physio] 121: 97-112.

Mauseth, J. (1998). Botany: an introduction to plant biology. Jones and Bartlett
Publishers. Sudbury, Massachusetts.

Miller, J. D., Arteca, R. N ., and Pel], E. J. (1999). Senescence-associated gene expression
during ozone-induced leaf senescence in Arabidopsis. Plant Physiol 120: 1015-1024.

Nair, M. (1988). Wood anatomy and heartwood formation in Neem (Azadirachata indica
A. Juss.). Bot J Linn Soc 97: 79-90.

Nelson, N. (1978). Xylem ethylene, phenol-oxidising enzymes and nitrogen and
heartwood formation in walnut and cherry. Can J Bot 56: 626-634.

Nelson, N., Rietveld, W.J., and Isebrands, J. (1981). Xyleme ethylene production in five
black walnut families in the early stages of heartwood formation. For Sci 27: 537-
543.

10

Nobuchi, T., Matsuno, H., and Harada, H. (1984). Relationship between heartwood
phenols and cytological structure in the transition zone from sapwood to heartwood of

sugi (Cryptomeria japonica). In Pacific Regional Wood Anatomy Conference
(IAWA/IUFRO), Tsukuba, Japan, pp. 132-134.

Nobuchi, T., Sato, T., Iwata, R., and Harada, H. (1984). Season of heartwood formation
and the related cytological structure of ray parenchyma cells in Robinia
pseudoacacia. Mokuzai Gakkaishi 30: 628-636.

Noguchi, T., Fujiwara, S., and Hayashi, S. (1986). Evolution of allantoinase and
allantoicase involved in urate degradation in liver peroxisomes. A rapid purification
of amphibian allantoinase and allantoicase complex, its subunit locations of the 2

enzymes, and its comparison with fish allantoinase and allantoicase. J Biol Chem
261: 4221-4223.

Nobuchi, T., Tokuchi, N., and Harada, H. (1987). Variability of heartwood formation and
cytological features in broadleaved trees. Mokuzai Gakkaishi 33: 596-604.

Schena, M., Shalon, D., Heller, R., Chai, A., Brown, P. O., and Davis, R. W. (1996).
Parallel human genome analysis: microarray-based expression monitoring of 1000
genes. Proc Nat] Acad Sci USA 93: 10614-10619.

Schena, M., Heller, R. A., Theriault, T. P., Konrad, K., Lachenmeier, E., and Davis, R.
W. (1998). Microarrays: biotechnology’s discovery platform for functional genomics.
Trends Biotechnol 16: 301-306.

Shain, J., and Mackay, J. (1973a). Seasonal fluctuations in respiration of aging xylem in
relation to heartwood formation in Pinus radiata. Can J Bot 51: 7 37-741.

Shain, L., and Mackay, J. (1973b). Phenol-oxidizing enzymes in the heartwood of Pinus
radiata. For Sci 19: 153-155.

Stryer, L. (1988). Biochemistry, Ed 3. WH Freeman, New York, pp 618-623

Tracey, M.V. (1961). Urea and Ureides. In K Peach K, M Tracey, eds, Modern Methods
of Plant Analysis, Vol 4. Springer Verlag, Heidelberg, pp 119-141

Vogels, G.D., Trijbels, F., and Uffink, A. (1966). Allantoinases from bacterial, plant and
animal sources. 1. Purification and enzymatic properties. Biochim Biophys Acta 122:
482-496

Walbot, V. (1999). Genes, genomes, genomics. What can plant biologists expect from the
1998 national science foundation plant genome research program? Plant Physiol 119:
1 151-1 156.

Wardrop, A., and Cronshaw, J. (1962). Formation of phenolic substances in ray
paranchyma of angiosperms. Nature 193: 90-92.

11

Webb, M.A., and Lindell, J .S. (1993). Purification of allantoinase from soybean seeds
and production and characterization of anti-allantoinase antibodies. Plant Physiol.
103: 1235-41.

Wingler, A., von Schaewen, A., Leegood, R., Lea, P., and Quick, W. (1998). Regulation
of leaf senescence by cytokinin, sugars, and light. Plant Physiol 116: 329-335.

12

CHAPTER 1

Novel gene expression profiles define the metabolic and physiological processes
characteristic of wood and its extractive formation in a hardwood tree species,

Robinia pseudoacacia.

ABSTRACT
Wood is of critical importance to humans as a primary feedstock for biofuel, fiber, solid

wood products, and various natural compounds including pharmaceuticals. The trunk
wood of most tree species has two distinctly different regions: sapwood and heartwood.
In addition to the major constituents, wood contains extraneous chemicals that can be
removed by extraction with various solvents. The composition and the content of the
extractives vary depending on such factors as, species, growth conditions, and time of
year when the tree is cut. Despite the great commercial and keen scientific interest, little
is known about the tree-specific biology of the formation of heartwood and its
extractives. In order to gain insight on the molecular regulations of heartwood and its
extractive formation, 1 carried out global examination of gene expression profiles across
the trunk wood of black locust (Robinia pseudoacacia L.) trees. Of the 2,915 expressed
sequenced tags (ESTS) that were generated and analyzed in the current study, 55.3 %
showed no match to known sequences. Cluster analysis of the ESTS identified a total of
2,278 unigene sets, which were used to construct cDNA microarrays. Microarray
hybridization analyses were then performed to survey the changes in gene expression
profiles of trunk wood. The gene expression profiles of wood formation differ according
to the region of trunk wood sampled, with highly expressed genes defining the metabolic

and physiological processes characteristic of each region. For example, the gene

13

encoding sugar transport had the highest expression in the sapwood, while the structural
genes for flavonoid biosynthesis were up-regulated in the sapwood-heartwood transition
zone. This analysis also established the expression patterns of 341 previously unknown

genes.

Introduction

Wood is a unique renewable material produced by trees using solar energy
through a highly ordered developmental process involving cell division/expansion,
secondary cell wall synthesis/deposition, lignification, programmed cell death, and
heartwood formation (Fukuda, 1996). As a result of radial growth and differentiation, the
trunk wood of many tree species has two distinctly different regions: sapwood and
heartwood. Sapwood is the outermost portion of the xylem tissue and contains living
cells, whereas the heartwood is defined as the “dead” central core of the woody axis and
only provides passive support to the tree. Sapwood (young xylem) has three important
functions: to conduct sap (water, solutes, and gases) from the roots to all parts of the tree;
to provide structural support for the entire tree; and to serve as a reservoir for water,
energy, minerals, and solutes. On average, about 10% of the cells in the sapwood are
alive (Kozlowski and Pallardy, 1997). The living ray cells in sapwood serve as the source
of raw materials for secondary substances. The ray parenchyma may also serve as
communication channels radially from the cambium through the sapwood, while axial
parenchyma functions largely as a storage tissue.

As sapwood is gradually converted to inactive heartwood, the wood parenchyma

cells undergo numerous metabolic changes and produce large quantities of heartwood

l4

extractives such as phenolic compounds, lignin, and aromatic substances that accumulate
in the vessels (Magel, 2000). During that process, one annual ring is converted to
heartwood each year (Mauseth, 1998). The reserve materials in the parenchyma cells of
the sapwood are used for wood formation and the synthesis of heartwood extractives,
such as condensed tannins, terpenes, flavonoids, lignans, stilbenes, and tropolones
(Burtin et al., 1998; Hillinger et al., 1996a and b; Hillis, 1987; Magel et al., 1994 and
2000). The formation of heartwood is accompanied by a variety of alterations in
metabolic conditions such as senescence. Although the events of senescence have been
studied at the molecular level during leaf senescence (Miller et al., 1999; Wingler et al.,
1998), seed germination (Cercos et al., 1999), and nodule development (Matamoros et a1.
1999); the cell maturation and death events occurring during heartwood formation have
been difficult to study because of the location and timing of the events. The presence of
heartwood is a major determining factor for wood quality and influences the way in
which specific woods are utilized. Various wood properties, such as dimensional
stability, durability, pulpability, colors and hues, scents and beauty, are affected by
extractives. Furthermore, stem wood sequesters large amounts of atmospheric C02 into a
much slower turnover pool and consequently accounts for the largest proportion (20-
40%) of total ecosystem aboveground carbon in closed forests (Saxe et al., 1998).
Therefore, understanding the regulation of wood formation is of great commercial and
keen scientific interest.

In recent years, a genomics approach has been successfully used to examine
global gene expression patterns in developing xylem tissues of pine (Allona et al., 1998;

Lorenz and Dean, 2002) and poplar (Sterky et al., 1998; Hertzberg et al., 2001).

15

Although the information derived from those studies undoubtedly provided a powerful
means for studying the molecular mechanisms of this important differentiation pathway,
it is still insufficient to account for the complete process of wood formation. So far, there
has been no report on global examination of gene expression profiles inside trunk wood
of mature trees. In order to gain some insight into the transcriptional hierarchy of
heartwood and its extractive formation, I examined global gene expression profiles across
the stems of lO-year-old Robinia pseudoacacia trees by sampling bark, sapwood, and
sapwood-to-heartwood transition zone tissues. This report describes the first
comprehensive look at global gene expression profiles in trunk wood and provides

expression data for many genes of unknown function.

Materials and Methods

RNA isolation and cDNA library construction

Three cDNA libraries of bark/cambial region (BCS), sapwood region (SWS), and
transition zone (TZS) of trunk-disk of lO-year-old black locusts (Robinia pseudoacacia
L.) harvested in early summer (July 27, designated “8”) and one cDNA library (TZF)
from a black locust harvested at late fall (November 27, designated “F”) were constructed
using the SMART cDNA library construction kit (ATriplEx2 vector system, Clontech,
Palo Alto, CA). Mature trees (20 cm DBH) were fell using a chain saw and made into 25
cm—long logs. The logs were immediately placed on ice and brought back to a wood shop,
where thin cookies (~1 cm thick) were made using a table saw. The cookies were
immediately submerged in RNA extraction buffer (20 mM EDTA, pH 8.0; 50 mM Tris-

HC], pH 8.0; 0.2 % SDS; 10 mM B-mercaptoethanol ). While submerged, the trunk wood

16

sections (bark/cambial region, sapwood, and transition zone) were carved out by using
chisel and hammer, and washed with RNase Away solution (Invitrogen, Carlsbad, CA).
The isolated sections (~1 cm3 cubicles) were frozen in liquid nitrogen and stored at —80°C
until needed. For RNA isolation, the frozen samples were first ground in a blender and
then further ground to fine powder using mortar and pestle. The ground sample were first
passed through the shredder column of DNeasy Maxi kit, and then subjected to total
RNA isolation using the RNeasy Maxi kit (Qiagen, Hilden, Germany) and cleaned up by

Qiagen RNeasy Mini kit.

Nucleotide sequencing

The cDNAs that were directionally cloned were randomly picked and sequenced
to generate ESTS. The sequencing was carried out at the Center for Computational
Genornics and Bioinforrnatics at the University of Minnesota and at the Genomics
Technology Supporting Facility at Michigan State University. The sequencing results are

posted at the website (http://web.ahc.umn.edu/biodatalblacklocust/).

Sequencing analysis

Raw sequence files were produced from the trace files using the Phred trace-
processing program followed by the Phran base-calling program with a quality threshold
of 8-10 (Ewing et al. 1998). Sequence artifacts were trimmed by the removal of leading
and trailing vector sequences in the raw sequence. To obtain the best subsequence where
the "N" value is 4% or less of the total number of bases, the number of unknown or "N"

bases in a sequence of trimming, leading and trailing high-N sections was determined.

17

Sequence similarity analyses were completed using a number of database searches.
Databases include GenBank, National Center for Biotechnology Information (NCBI)
GenPept (Benson et al., 2000), Protein Information Resource (PIR) (Barker et al. 2000),
Swiss Institute of Bioinformatics SWISS-PROT (Bairoch et al., 2000), TrEMBL
(Bairoch et al., 2000) and the National Biomedical Research Foundation NRL3D (Barker

et al., 2000). The EST sequences were deposited in the dbEST of the GenBank database.

Contig analysis

In order to generate the unigene sets and contigs of sequencing data, EST data
was analyzed using DNA similarity algorithms and the assembly program,
Phred/Phrap/Consed (Green and Ewing, 1996). Phrap (“phragment assembly program”,
or “phil's revised assembly program”) was used for assembling shotgun DNA sequence
data, constructing a contig sequence as a mosaic of the highest quality parts of reads
(rather than a consensus) and providing extensive information about assembly (including
quality values for contig sequence) to assist trouble-shooting. Phrap was used in
conjunction with the base calls and base quality values produced by the basecaller, Phred;
and with the sequence editor/assembly viewer, Consed. Cross-match was based on a
"banded" version of SWAT, an efficient implementation of the Smith-Waterman
algorithm for comparing any two sets of (long or short) DNA sequences. To calculate
estimates for comparison among different libraries, I used a statistical method described
by Audie and Claverie (1997). This method was developed to calculate statistical
differences from different numbers of different libraries, and now it is being used for

“digital gene expression profiles.” Based on the method, [calculated significantly

18

different numbers within the same contig among the libraries, and I applied it to estimate

statistical differences within a category as described in Kirst et al., (2002).

PCR amplification of the insert cDNA and microarray printing

ATriplExZ vector sequences flanking the insert (5’-
AAGCAGTGGTATCAACGCAGAGT-3’ and 5’-
ATI‘CTAGAGGCCGAGGCGGCCGACATG-3’) were used to amplify selected EST
clones using polymerase chain reaction (PCR). The PCR products were precipitated in
ethanol and re-suspended in 3 x SSC (1 x SSC is 0.15 M NaCl and 0.015 M sodium
citrate). They were checked for quality by using gel electrophoresis to observe the
concentration and multiple bands. PCR products of 2,592 clones were arrayed from 384-
well microtiter plates, and DNA was spotted on superaldehyde (Telechem, Sunnydale,
CA) glass slides at a high density using an Omnigridder robot (Gene Machines, San
Carlos, CA) and 16 ArrayIt chipmaker 2 pins (Telechem). Slides were washed and
blocked according to the manufacturer’s protocol.

Each glass slide contained two replications of the entire array, each of which
consisted of 16 subarrays with 12 rows and 14 columns. Negative control genes, B-cell
receptor protein genes, including genes such as Myosin heavy chain gene, Myosin
regulation light chain2 and insulin-like growth factor gene, were printed on the top,

middle and bottom of each array.

Preparation of labeled cDNA probes

Total RNA (lug from each sample) was reverse-transcribed and amplified using

19

the SMAT system (Clontech). To reduce nonspecific PCR amplification, cDNAs were
amplified with the fewest cycles. Two micrograms of cDNA were labeled by the
incorporation of either Cy5 or Cy3-dCTP (Amersham-Phamacia, Piscataway, NJ) during
oligo-dT-primed primer extension in the presence of Klneow DNA polymerase
(Promega, Madison, WI) as described by Schaffer et al. (1999). The labeled probes were
purified using the QiaQuick PCR cleanup kit (Qiagen). The probe samples were
denatured by placing them in a 100°C water bath for 3 min, left at room temperature for
30 min, and then used for hybridization.

To minimize the inherent variability of the microarray assay (Lee et al., 2000) and
to ensure the reliability of the results, at least two microarray slides (four replicates) were
used to analyze the transcript expression of each sample pair. The first slide was probed
with cDNAs labeled with Cy-3 and Cy-5 deoxy CTP. To probe the arrays, cDNAs were
synthesized and amplified from bark/cambial region, sapwood, or transition zone and
labeled by Klenow-mediated incorporation of Cy-3-dCTP or Cy-5-dCTP, respectively.
By using independent RNA preparations, the second slide was hybridized by cDNAs
reverse labeled with Cy-3 and Cy-5 dCTP from each sample pair to overcome potential
artifacts caused by the dye-related differences in labeling efficiency, different laser
settings, and nonlinearity of photomultiplier tubes in the scanner. Thus, at least two, and
sometimes three or four, independent RNA preparations were made for each biological
sample and were used to prepare labeled probes. The hybridization signal from each of

the replicate ESTS were averaged and used for analysis.

20

Hybridization and washing of the DNA microarray

The labeled probes with either Cy3 or Cy5 fluorescent dye were hybridized to a
microarray slide in a total volume of 30 uL of hybridization buffer (3.4 x SSC, 0.32%
SDS, and 5 ug of yeast tRNA) for 16 hours at 65°C. The slide was then washed at room
temperature in l x SSC, 0.1% SDS for 10 min, in 1 x SSC for 10 min, and in 0.01 x SSC
for 10 min. The slide was centrifuged dry and scanned with a 428 Array scanner

(Affymetrix, Palo Alto, CA). Each microarray experiment was repeated twice.

Microarray Data Analysis

The data were analyzed with GenePix Pro3.0 (Axon Instruments Inc., Union City,
CA). The scanned data were normalized by using the Global Normalization method
(Hihara et al., 2001), in which the image data between Cy3 and Cy5 channels are
normalized by adjusting the total signal intensities of two images and the bad spots are
removed. The unreliable spots were removed by the following screening. Spots
containing clones that had poor amplification or multiple bands, as well as those that were
flagged due to a false intensity caused by dust or background on the array, were removed.
Spots with <65% of the spot intensity at >1.5-fold that of the background in both channels
were ignored (see Stanford Microarray Database Web site, httpzl/genome-
www5.stanford.edu/ MicroArray/ SMDI). Clones in one sample that had an average
induction greater than 2-fold in another were determined as up-regulated. Comparison of
the arrays was achieved using Microsoft Excel and Microsoft Access database. For
cluster analysis, Cluster and Treeview software were used (Eisen et al., 1998; available at

httpzllgenome-www4.stanford.edu/ MicroArray/SMD/restech.html ).

2]

Antisense northern blot analysis

I conducted “Antisense Northern blot analysis” which requires only minute
amounts of RNA. Conventional Northern blot analysis protocol requires a large amount
of RNA from my inner wood samples, typically transition zone, which are hard to isolate,
making those protocols difficult to perform. An alternative was to use antisense RNA
(aRN A) amplification method that has been successfully used in other microarray
analyses (Wang et al., 2000; Baugh et al., 2001; Dent et al., 2001; for protocol, see
Patrick Brown’s web site; http://cmgm.stanford.edul pbrown/ protocols/
ampprotocol_3.html). Antisense RNA was generated as described previously (Wang et
al., 2000). Briefly, aRNA was amplified using Message AmpTM aRNA kit (Ambion,
Austin, TX). I began by synthesizing first stand cDNAs from the total RNAs of seedling,
bark/cambial region, sapwood, or transition zone, and the first cDNAs were used as
templates for the synthesis of second cDNAs. Finally, aRN As from the second cDNAs
were generated by in vitro transcription (amplification). About 200ng of aRNAs were
separated in a formamide agarose gel, and transferred onto the nylon membrane using the
capillary transfer method. The membrane was then hybridized with an isotope-labeled

probe. The signal was exposed and detected on an X-ray film.

Results

Trunk Wood cDNA Library Construction and Sequencing
Trunk wood of mature trees can be divided into three main parts: bark/cambial
region, sapwood, and heartwood (Figure l-l). While sapwood contains living ray cells,

the cells in heartwood are dead and filled with extractives, which produce intense and

22

 

Figure 1-1. Cross-section of a stem from a mature Robinia pseudoacacia tree

(A) Cross section under daylight. HW, Heartwood; SW, sapwood; TZ, transition zone.
(B) Cross section under UV light. Bright fluorescence in the middle is due to the
flavonoids.

23

bright fluorescence under UV light. In order to analyze the gene expression patterns in
different sections of the trunk wood, I constructed four cDNA libraries from the
bark/cambial region (BCS), sapwood (SWS), and transition zone of ten-year-old black
locusts. The transition zone samples were collected both in the summer and fall (T28 and
TZF). Due to the large amounts of polysaccharides and phenolic compounds present in
the inner wood tissues, it was not possible to obtain a large enough quantity of pure
mRNA for conventional cDNA library construction. Nonetheless, I was able to construct
high quality (> 5 x 105 pfu) phagemid libraries using the PCR-based cDNA library
construction kit (Clontech). The phagemids from each library were converted into
plasmids by mass excision in E. coli. Over 3,600 individual clones were randomly
selected from all four libraries and sequenced using a 5’ vector sequencing primer
provided in the kit. After trimming vector sequences, clones containing high ambiguous
calls (high “N” percent on a DNA sequence) were removed. Finally, a total of 2,915 were
chosen for further analyses and sequencing: 895 clones from the BCS library, 999 clones
from the SWS library, 880 clones from the T28 library and 141 clones from the TZF
library. An average length of 448 bases was obtained and used for contig analysis and

database searches.

EST Analysis

The high redundancy of the mRN A in a tissue is approximately reflected in the
abundance of its corresponding cDNA in non-normalized libraries. The random
sequencing of cDNAs yields information about the ESTs (Adams et al., 1993). Sequence

similarities were found by searching various available databases including: GenBank,

24

National Center for Biotechnology Information (NCBI) GenPept, Protein Information
Resource (PIR), Swiss Institute of Bioinforrnatics SWISS-PROT, TrEMBL and the
National Biomedical Research Foundation NRL3D (see the website;
http://web.ahc.umn.edulbiodata/blacklocust/). Sequence similarities identified by the
BLAST program were considered statistically significant with a Poisson P value of S 10'
5. The 1,304 ESTS (44.7%) of the total 2,915 ESTS matched previously sequenced genes.
The 909 ESTS of the 1,304 ESTS had significant homology to previously identified
genes. The annotations of genes with similarities to an EST were used to assign a putative
identification to my EST. The 909 ESTS with similarity to known genes were classified
into 13 putative functional categories (Bevan et al., 1998, Covitz et al., 1998), which are
listed in Table 1-1. My libraries have a significantly high number of no hit clones,
especially in the SWS library (80.6%). As a matter of fact, the portion of no hit clones in
other libraries was also high, about 50 %. Allona et al. (1998) suggested that the length
and quality of cDNA sequences are correlated with the ability to identify similar
sequences in public databases. Recent analysis of Pinus taeda cDNA clones with longer
than 1,000 bases-read revealed that about 95% of the pine genes had homologous
sequences in Arabidopsis genome (Sederoff et al., 2002). However, the average
sequenced length of a SWS clone is not entirely different from the total sequenced length
(sequence length distribution data is found at
http://web.ahc.umn.edulbiodata/blacklocustl). Many long, high-quality sequences show
neither strong nor marginal similarity to sequences in the database, and some no hit
clones also show high redundancy (Table 1-2). Therefore, these no-hit clones may

represent novel plant genes, reflecting the uniqueness of my samples.

25

Table 1-1. The functional classification of EST clones

 

 

 

 

Catgory Total BCS SWS TZS TZF
Cell division
and cycle 7 4 0 3 0
Cell wall structure
and metabolism 16 5 7 3 1
Chromatin
and DNA metabolism 27 8 4 12 3
Cytoskeleton 10 6 3 0 1
Defense 90 46 5 35 4
Gene expression
7 2

and RNA metabolism 6 38 0 16 2
Membrane transport 48 33 1 11 3
Miscellaneous 1 75 73 36 51 1 5
Primary metabolism 134 46 19 56 13
Protein synthesrs 190 84 19 80 7
and processrng

Secondary and hormone 55 8 2 44 1
metabolism

Signal transduction 75 26 1 1 32 6
Vesrcular trafficking, 6 2 2 2 0
protein sort, secret

Unknown, Hypothetical 395 162 65 140 28
Hit clones 1304 541 194 485 84

(44.7) (60.4) (49.4) (55.1) (59.6)
No hit clones 1611 354 805 395 57
(55.3) (39.6) (80.6) (44.9) (40.4)

Total sequenced clones 2915 895 999 880 141

 

The bold number within a row indicates significantly different number of ESTS in the
library compared to all the others (P < 0.01). For example, the number of ESTs in “Gene
expression and RNA metabolism” category was statistically higher in BCS when
compared to SWS and TZS libraries. Likewise, the number of ESTS in “Secondary and
hormone metabolism” category was significantly higher in TZS than in BCS and SWS.

26

(A) BCS (B) sws (C) TZS

           

Figure 1-2. Functional classification of the ESTS from three libraries

Due to the high proportion of no-hit clones in the libraries, only those ESTs with
significant homology with previously reported sequences were included in this
classification.

l3 Cell dlvlslon and cycle

I Cell wall structure and metabollsm

I] Chromatin and DNA metabollsm

El Cytoskeleton

I Defense

Gene expresslon and RNA metabollsm
I Membrane transport

El Miscellaneous

I Primary metabollsm

I Proteln synthase and processlng

El Secondary and hormone metabollsm
El Slgnal transductlon

IVeslcular trafflcklng, proteln sort, secret
I Unknown, Hypothetical

27

The proportions of ESTS in each functional category differed for the EST libraries from
the three trunk zones characterized in this study (Figure 1-2.) These data, which were
calculated by excluding no-hit clones in all libraries, show a common overall trend, but
specific groups of genes were more or less highly represented in specific zones. Notable
are the higher representation of secondary and hormone metabolism genes in the TZS
library (9.1% versus 1.4% and 0.8% in the other libraries), cell wall and structural
metabolism genes in the SWS library (3.6% versus 0.6% and 0.9% in the other libraries),
and genes associated with membrane transport in the BCS library (6.1% versus 1.2% and

2.3% in the other libraries).

Contig Analysis

It was attempted to estimate the redundancy of my EST clones on the basis of the
contig analysis (Phrap assembly with the 70 % identical value of pairwise sequence). The
proportion of singletons was extremely high, representing a calculated level of singletons
at 70 % (2,051 out of 2,915 ESTS). Only 864 ESTS were identified in 228 contig sets. As
a result of the contig analysis, I obtained a total of 2,278 unigene sets that were submitted
to GenBank database (accession numbers 31642054 to Bl679372). The most frequently
presented gene in my ESTS encodes a hypothetical protein (At2g41250; 1.3 %) followed
by an auxin-repressed protein (1.1%) that is expressed in dormant tissues and repressed
by auxin treatment (Reddy and Poovaiah, 1990; Stafstrom et al., 1998). It is abundant in
all libraries except the sapwood library (Table 1-2). It is notable that the TZF library
(2.1%) showed a similar proportion when compared to the TZS library (1.5%). The

function of this gene has not yet been identified. The contig analysis identified three

28

types of metallothionein or metallothionein-like protein genes with different expression
patterns in the different zones of the trunk wood. For example, metallothionein (conti g
226) is the highest in the TZS library, but metallothionein-like protein (contig 223) is
only in the BCS library. Such differential expression of metallothionein genes has been
reported in Arabidopsis plants (Garcia-Hernandez et al., 1998). In addition to providing
physical support, trunk wood functions as a conduit for water, nutrient, and
photosynthates transport. Aquaporin and phloem-specific protein, which are related to a
membrane transport system or phloem, are abundant mainly in the bark/cambial region
library. Two types of aquaporin genes that are related to water transport were found in
my libraries Aquaporin 1 contig existed in the bark/cambial region library only, while
aquaporin 2 contig was present in the transition zone library as well as bark/cambial
region library. However, no contig for aquaporin genes were present in the sapwood
library. Because the sole abundance of transcript encoding phloem-specific protein exists
only in the bark/cambial region library, it is clear that my bark/cambial region sample
contained phloem regions. Eight out of the 20 contigs in Table 1-2 were library-specific.
In other words, the contigs were present only in one library and absent in the other two
libraries. For example, maturase and hypothetical protein (PIR: A05 191) are abundant
only in the SWS library. Hypothetical protein (At3g03150) and cytochrome b5 DIF-F are
highly abundant in the TZS library. Some contig sets contained only no hit clones
resulting from the source zone of the samples; for example contig 224 and 216 consist of
clones from two transition zone libraries, and clones of contig 213 are in the BCS library.
These library specific contigs can be expected because the typical character of each

library is dependent upon its location within the trunk wood.

29

Table 1-2. The redundancy of EST clones based on contig analysis

 

 

. Contig Total BCS SWS TZS TZF
Annotation
number (%) (%) (%) f/o) (%)
Hypothetical protein
(At2941250) 228 37 (1.3) 27 (3.0) 0 8 (0.9) 2 (1.4)
Auxin-repressed
protein 227 33 (1 .1) 17 (1.9) 0 13 (1.5) 3 (2.1)
Metallothionein 226 18 (0.6) 2 (0.2) 1 (0.1) 14 (1.6) 1 (0.7)
Maturase 225 13 (0.4) 0 9 (0.9) 4 (0.5) 0
No hit 224 11 (0.4) 0 0 10 (1.1) 1 (0.7)
Metallothionein-like 223 11 (0.4) 11 (1.2) 0 0 0
protein
Hypothetical protein
(“3903150) 222 11 (0.4) 0 0 11 (1.3) 0
Cytochrome BS DIF-F 221 10 (0.4) 0 0 9 (1.0) 1 (0.7)
Aquaporin 1 220 9 (0.3) 9 (1.0) 0 0 0
Aquaporin 2 219 8 (0.3) 4 (0.4) 0 3 (0.3) 1 (0.7)
Hypothetical protein
(A05191) 218 8 (0.3) 0 8 (0.8) 0 0
Phloem-specific
protein Vein 1 217 8 (0.3) 8 (0.9) 0 O 0
No hit 216 7 (0.2) 0 0 7 (0.8) 0
Metallothionein
Class-ll 215 7 (0.2) 2 (0.2) 0 3 (0.3) 2 (1.4)
Ubiquitin 214 7 (0.2) 6 (0.7) 0 1 (0.1) 0
No hit 213 7 (0.2) 7 (0.8) 0 0 0
No hit 212 7 (0.2) 6 (0.7) 0 0 1 (0.7)
Integral membrane
transport protein 211 6 (0.2) 2 (0.2) 0 4 (0.5) 0
Hypothetical protein
(075542) 210 6 (0.2) 5 (0.6) 0 1 (0.1) 0
Extensin 83 2 0 0

 

0 2 (1.4)

Letters in bold are significantly different from the numbers in the other libraries at P <
0.05. For example, Metallothionein-like protein (contig 223) and Aquaporin 1 (contig
220) are significantly more abundant in BCS zone than SWS and TZS. SWS library had
significantly higher number of ESTS in the contig 218 (hypothetical protein, PIRA05191)
than did BCS and TZS libraries. Contigs 222 (hypothetical protein At3g03150) and 216
(no hit clone) are significantly abundant in TZS compared to BCS and SWS libraries.

30

Microarray Experiments

In order to produce gene expression profiles related to wood formation and to
compare gene expression patterns from different regions of the inner wood from a mature
tree, I produced cDNA microarrays carrying ~2,580 unigenes from all four libraries and
conducted microarray hybridization experiments. This approach allowed me to examine
the expression changes of ~2,580 genes simultaneously and to search the expression
patterns of bark/cambial region, sapwood, and transition zone through the use of one-by-
one comparisons. I demonstrated that one-by-one experiments can be carried out for
specific expressed profiling on continuous samples. My approach involved the
comparison of one sample with two other samples and then the generated ratios from the
two different experimental sets were plotted. When all of the spots were plotted, I could
determine which genes were specifically expressed in Sample A, when compared to
Sample B and Sample C. Unlike a reference or loop design, an approach of this type will
serve as a one-by-one comparison for a small number of side-by-side samples in
transcript profiling studies.

In my microarray experiments, the need to obtain large amounts of RNAs proved
to be a challenge. Standard microarray protocols require isolating poly(A) RNA from
samples, but my wood sample is too difficult to isolate RNAs. So, the cDNA
amplification method was used, and I checked the reproducibility of my experiments. I
also confirmed that the amplification method efficiently generates highly reproducible
populations of cDNA. This method can be useful in transcript profiling studies with
limited amounts of RNA. To test the reproducibility of the two different experiments, the

expression ratio derived from one microarray experiment was compared to the expression

31

 

 

 

 

 

 

 

Figure 1-3. Scatter plots of microarray hybridization results

(A) The reproducibility of microarray experiments. Two hybridization experiments of
SWS versus BCS were conducted using a dye-swap with Cy5 and Cy3 labeled probes.
The natural log values of the Cy5-to-Cy3 ratios were plotted for the two replicates.

(B) The BCS-specific expression pattern. The X-axis is the log-scaled ratio of gene
expression for the experiment of BCS over SWS, and Y-axis is the log-scaled ratio of
gene expression for the experiment of BCS over TZS.

(C) The SWS-specific expression pattern. X-axis is the log-scaled ratio of SWS over
BCS, and Y-axis is the log-scaled ratio of SWS over TZS.

(D) The TZS-specific expression pattern. X-axis is the log-scaled ratio of TZS over BCS,
and Y axis is the log-scaled ratio of TZS over SWS. All number values of the log-scaled
ratio of gene expression are average values of data points generated from each
experiment. Abbreviations: BC, bark/cambial zone; SW, sapwood; TZ, transition zone.

32

ratio from the other, for all of the normalized clones (Figure 1-3A). The scatter plot
shows that the gene expression ratio from the two experiments was remarkably similar
(the coefficient of determination R2 = 0.96). Furthermore, the coefficients of
determination between other experiments were sufficiently high (> 0.91). These results
are similar to those presented in other papers, in which the same or similar amplification
protocol was used (Livesey et al., 2000, Hertzberg et al., 2001). Furthermore, using
“antisense northern blot” analysis, I confirmed the microarray data as well as the EST
results. Thus, the microarray signals from the replicates in my study were highly

reproducible and the conclusions derived from this analysis are considered reliable.

Difierential Gene Expression across the Stem

In order to investigate distinct differences in gene expression profiles among the
three regions of the trunk wood (bark/cambial zone, sapwood, and sapwood-heartwood
transition zone) during active tree growth, I compared differentially expressed genes in
the three zones. The expression ratios for each zone in comparison to the other two zones
were examined using regression analysis. Figure 1-3B shows the scatter plot and
regression line of the expression ratios for the bark/cambial region versus the sapwood or
transition zone libraries. The trend line of the scatter plot had a slope of 0.401, with R2 =
0.2956. These values indicate a positive correlation, but with little of the variation in the
data explained by the regression. On the basis of that graph, I identified the genes that
were up or down regulated in BCS tissue using the cutoff values of an expression ratio

greater than two-fold or less than 0.5. Of all arrayed cDNAs, 292 clones were

33

Table 1-3. Summary of up-regulated or down-regulated genes

 

 

 

Up-regulation Down-Elation
Functional categorL BC SW TZ BC SW TZ
Cell division and cycle 0 0 0 0 0 0
Cell wall structure and metabolism 0 0 0 1 0 O
Chromatin and DNA metabolism 0 0 1 1 0 0
Cytoskeleton 2 0 0 O 0 2
Defense 6 0 2 2 0 6
Gene expression and RNA metabolism 4 0 0 4 0 3
Membrane transport 7 1 0 0 0 5
Miscellaneous 1 6 0 1 3 0 1 6
Primary metabolism 8 1 2 1 0 5
Protein synthase and processing 2 O 1 0 0 5
Secondary and hormone metabolism 1 0 8 3 1 0
Signal transduction 1 0 2 1 0 0
Vesicular trafficking, protein sort, secret 1 0 0 0 0 0
Unknown, Hypothetical 22 0 20 7 3 14
No hit 26 1 37 40 0 12
Total 96 3 74 63 4 68

 

34

more highly expressed in the bark/cambial region than in sapwood. The expression of
174 clones in the bark/cambial region was higher than those in the transition zone.
Ninety-six of the clones were considered as the BCS-specific clones (Tables 1-3 and 1-4).
As expected, metallothionein-like protein (contig 223; containing all 11 BCS clones),
phloem-specific protein genes (contig 217; containing all 8 BCS clones), and a no-hit
clone in contig 213, which is a contig set composed of all 7 BCS clones, were highly
expressed in bark/cambial region when compared to other zones. These results
corroborate with those of EST redundancy analysis. In addition, genes encoding
photosynthesis-related proteins, such as Photosystem 11 10K proteins, were highly
expressed in the bark/cambial region. In addition, 63 clones were specifically down
regulated in bark/cambial region, when compared to sapwood and transition zone (Table
1-3). I compared each functional category to find out how many tissue-specific genes
were in each category. Table 1-3 shows the proportion of functional categories of up or
down regulated genes in bark/cambial region. Like the EST results, the proportion of the
genes categorized as involving membrane transport was high relative to other functional
categories in the bark/cambial region. When considering the physiological significance of
the region in tree growth and development, the proteins encoded by the genes in each of
these categories might be targets of special interest for biotechnological improvement of
trees.

Similarly, the expression ratio of the SWS was compared to both the BCS and
TZS (Fig 1-3c) and the number of up and down regulated genes was calculated. Unlike
the trendline of the bark/cambial region and transition zone scatter plots; the scatter plot

of sapwood had a negative slope (0.401), indicating a negative correlation. Many of the

35

clones that were highly expressed in sapwood vs. bark/cambial region were down
regulated in the sapwood vs. transition zone comparison. For example, the expression
ratio of the PR-10 gene in the sapwood region vs. bark/cambial region was 6.7, but the
ratio in the sapwood region vs. the transition zone was 0.5. Similarly, the expression ratio
of the phloem-specific protein gene in the sapwood region versus the bark/cambial region
was 0.2. However, in contrast, the ratio in the sapwood region versus transition zone was
4.3, showing that the expression relationships vary for the different proteins in the three
regions. In addition, the number of clones specifically up or down regulated in the
sapwood region was very small (3 or 4), but when the expression in the sapwood region
was compared to bark/cambial region or transition zone, the number of up or down
regulated clones increased dramatically (395 up regulated for sapwood vs. bark/cambial
region; 177 up regulated for sapwood vs. transition zone, and 291 down regulated for
sapwood vs. bark/cambial region; 226 down regulated for sapwood vs. transition zone).
This suggests that the sapwood region plays the role of a bridging zone between
bark/cambial region and transitional zone. Table 1-5 shows the list of up-regulated genes
in the sapwood compared to bark/cambial region and transition zone. In other words,
only three ESTS (T280226, SWS0332, and SWS0562) had sapwood-specific up-
regulation. Interestingly, the transcript of a sugar transport protein, which plays key roles
in source-sink relationships (Lalonde et al., 1999), was highly expressed in sapwood. The
highly redundant EST (contig 218) in the sapwood library is also highly expressed in the
sapwood when compared with the bark/cambial region and transition zone (at the ratio of

2.2 and 1.7, respectively).

36

Table 1-4. Up-regulated transcripts in the bark and cambial zone

 

 

Clone ID GenBank Acc. Annotation BCISW“ 8C/1‘Z*
Cytoskeleton

CLSOO35 81677465 Actin depolymerizing factor 5 3.2 3.9
CLSO977 81678088 (SMC)-like 5.1 3.3
Defense

CLSOlOO 81677437 Superoxide Dismutase (Cu-Zn) 2.4 2.4
CL80239 81677578 Metallothionein-like protein 5.4 3.5
CLS0595 81677862 Superoxide Dismutase (Cu-Zn) sod8 2 2.1
CLS0801 81677956 Metallothionein 2.8 2.5
CLS0867 81678223 Glyoxalase 3.2 4.1
CLS0960 81678073 Proteinase Inhibitor 3.2 2.4
Gene expression and RNA metabolism

CLS0340 81677663 Transcriptional regulator, putative 2.8 2.3
CLS0439 81677739 Zinc Finger Protein ID] 2.2 2.7
CLSO673 81678286 Transcription Factor like Protein 4.3 4.4
SWS 1456 81679312 Zinc Finger Protein IDl 2.2 2.7
Membrane transport

CLSOOZS 81677455 Lectin precursor, Bark Agglutinin I 7.7 3.8
CLSOOSI 81677477 Vacuolar V-H subunit E [Citrus limon] 2.2 2.2
CL80052 81677478 Aquaporin 7.5 3.8
CLS0197 81677550 Lectin 2.1 2.9
CLSO488 81677780 Tonoplast Intrinsic Protein, delta type 7 5.9
CLSO929 81678048 Lectin-related polypeptide 4.7 3.3
CLS0950 81678064 Lectin like protein (hypothetical) 3.3 2.4
Miscellaneous

CL80022 81677452 Phloem-specific protein Veinl 5.3 10.1
CLSOO68 81677490 Chlorophyll a/b binding protein 4.6 2.5
CLSOO81 81677501 MTNS Gene Precursor 3.5 6.7
CLS0158 81677521 Phloem-specific protein Veinl 8.8 5.5
CLSO274 81677608 Trypsin Inhibitor (Serine Proteinase Inhibitor) 3.1 2.9
CLSO376 81677690 Photosystem 11 10K protein 2.6 4.2
CLSO423 81677726 Photosystem 11 Protein X precursor 8.1 4.9
CLSO479 81677773 Phloem-specific protein Veinl 7.1 4.9
CLSOS36 81677820 Core protein 2.8 2.1
CLSOS64 81677840 Photosystem 11 10K protein 3 2.1
CLS0603 81677869 Auxin-repressed protein 3.1 3.1
CLSOSOS 81677959 Photosystem 1 Reaction Centre Subunit V1 5 3.7
CLS0819 81677971 Trypsin Inhibitor 3.5 2.6
CLSO824 81677975 Magnesium Chelatase 4.1 3.3
CLSO952 81678066 Specific Tissue Protein 2 6.3 2.7
CLS 1025 816781] 1 Ripeninfielated protein 4.2 3

 

 

* The ratio was estimated as average value from data points.

37

Table 1-4. (Cont’d)

 

 

Clone ID GenBank Acc. Annotation BCISW“ 8CfI‘Z*
Primary metabolism

CLSOOl7 81677447 Copper Amine Oxidase precursor 4.4 5.4
CLSO714 81678307 Lipid Transfer Protein 4.2 3.6
CLSO745 81677908 Alcohol Dehydrogenase l 6.2 3.8
CLS0813 81677966 Alcohol Dehydrogenase 7 3.6 3
CLSO930 81678049 Acetoacyl-CoA-thiolase 2.3 2.9
CLSO956 81678070 Blue copper protein 2.5 2.1
CLSlOlS 81678039 Lipid Transfer Protein 4.5 2.9
1280305 81642632 Epoxide Hydrolase 4.7 2.3
Protein synthesis and processing

CLSO811 81677964 Ribosomal Protein S16 protein 3.7 2.3
CLSO973 81678084 Peptidylprolyl Isomerase; Cyclophilin (Cyp) 2.4 2
Secondary and hormone metabolism

CLSO373 81677687 Monooxygenase 3.5 2.6
Signal transduction

CLSO377 81677691 Protein Phosphatase 2C-like 3.2 2.3
Vesicular trafficking, protein sorting, secretion

CLSO783 81677940 ER etention receptor Erd2 2.3 2.8
Unknown

CL80032 81677462 Unknown Protein 3.3 3.1
CLSOO71 81677493 Unknown Protein 2.3 2.1
CLSOl86 81677541 Unknown Protein, hypothetical 2.6 2.8
CL80227 81677572 Unknown Protein 4 3.3
CL80273 81677607 Unknown Protein, hypothetical 3.8 4.8
CL80297 81677624 Unknown Protein, hypothetical 2.6 2.5
CLSO324 81677649 Unknown Protein, hypothetical 2.8 2.5
CL80333 81677657 Unknown Protein, hypothetical 2.5 2.2
CLSO342 81677665 Unknown Protein, hypothetical 2.9 2.8
CLSO362 81677678 Unknown Protein, hypothetical 3.3 2.2
CLSO406 81677712 Unknown Protein, hypothetical 3.5 3.4
CLS0493 81677785 Unknown Protein, hypothetical 3.1 2.9
CL80537 81677821 Unknown Protein, hypothetical 2.5 2.3
CLSO630 81677890 Unknown Protein, hypothetical 10.7 3.1
CLSO740 81677903 Unknown Protein, hypothetical 4.3 2.9
CLSO776 81677934 Unknown Protein 3.3 2.7
CL30800 81677955 Unknown Protein 4.1 2.9
CLSO854 81678208 Unknown Protein, hypothetical 2.3 2.1
CLSlOl4 81678038 Unknown Protein, hypothetical 2.4 2.7
CLSlO37 81678119 Unknown Protein 3 2.8
CLSl 100 81678177 Unknown Protein, hypothetical 3.3 2.4
CLSl 1 14 81678186 Unknown Protein, hypothetical 2.5 2.7

 

* The ratio was estimated as average value from data points.

38

Table 1-4. (Cont’d)

 

 

Clone ID GenBank Acc. Annotation 8C/SW* 8012*
No hit

CLSOO36 81677466 No hit 3.6 2.4
CL80055 81677481 No hit 4.4 2.1
CL80060 81677485 No hit (contig 213) 5.8 3.9
CLSOO66 81677489 No hit 6.3 3.2
CLSO201 81677554 No hit 3.4 2.6
CL80211 81677559 No hit (contig 212) 4.3 2.4
CLSO253 81677591 No hit 5.2 4.3
CLSO349 81677670 No hit 9.4 12.3
CL80366 81677681 No hit 4.9 2.5
CLSO378 81677692 No hit 3.8 5.5
CLSO426 81677729 No hit 2.1 2
CLSO461 81677758 No hit 2.1 4.3
CL80568 81677843 No hit 2.8 2.7
CLSO608 81677873 No hit 2.1 2
CL80712 81678306 No hit 2.9 2.3
CLSO739 81677902 No hit 4 3.3
CLSO827 81678187 No hit 6.5 4.8
CL80840 81678197 No hit 2.8 3.6
CL80884 81678235 No hit 8 2.9
CLSO982 81678092 No hit 2.6 2
CL81048 81678129 No hit 2.5 2.5
CL81095 81678170 No hit 2.4 2
SWSOS75 81678741 No hit 2.2 3.6
SWSO655 81678790 No hit 2.6 2.6
T280560 81642188 No hit 3.3 3.2
T281309 81642865 No hit 3.1 2.5

. _... n. are!

 

 

* The ratio was estimated as average value from data points.

39

In

Table 1-5. Up-regulated transcripts in the sapwood

 

 

Clone ID GenBank Acc. Annotation SW/8C* SW/TZ*
Membrane transport
1280226 81642581 Sugar transport protein 2.8 3
No hit
SWSO332 81678553 No hit 6.8 2.3
Primary metabolism
Acetyl-CoA Carboxylase Carboxyl
SW80562 Transferase 2.4 3.3

 

* The ratio was estimated as average value from data points.

There were dramatic differences in gene expression patterns in the transition zone
relative to the other two comparisons. As shown in Figure l-3D, the gene expression
ratios in the transition zone versus the other two zones showed a more definite
relationship, indicating a higher correlation and lessened variability (slope: 0.6999 and
R2: 0.5123). Genes with low expression ratios in the transition zone verses bark/cambial
region were also low at the ratio of transition zone verses sapwood. In addition, genes
that were highly expressed in the transition zone compared to bark/cambial region also
had a higher expression in the transition zone than in sapwood region. Of 357 genes that
had a higher expression in the transition zone than in either the bark/cambial region or
sapwood region, 75 genes were specifically up regulated and 68 genes were down
regulated in transition zone (Tables 1-3 and 1-6). Tables 1-3 and 1-6 list the proportion of
functional categories of up-regulated genes in transition zone. The results of the EST
analysis and microarray results show that the proportion of secondary and hormone

metabolism is relatively high (10.7%) in the transition zone. In addition, three T28 clones

40

 

Table 1-6. Up-regulated transcripts in the transition zone

 

 

Clone ID GenBank Acc. Annotation 12180“ TZJSW“
Chromatin and DNA metabolism

1280924 81642324 SAP] Protein 3.1 3.2
Defense

T280124 81642508 PR-lO Protein 5.3 3.6
T280357 81642069 PR-lO Protein 3.2 2.1
Miscellaneous

T281380 81642911 NAM-like protein 2.1 2.7
Primary metabolism

T280097 81642647 Proline Oxidase 2.9 2.4
T280519 81642157 Cytochrome 85 DIF-F 6.3 5.1
Protein synthesis and processing

T280133 81642514 eIF4F chain p28 4.8 6.8
Secondary and hormone metabolism

T280021 81642439 Chalcone-Flavone Isomerase 4.4 4.5
T280108 81642499 Naringenin 3-Dioxygenase 6.5 5.5
1280312 81642638 Flavonoid 3’,5’-Hydroxylase 4.6 4.8
T280424 81642100 Chalcone Synthase 5 4.2
1280751 81642292 Flavonoid 3’-hydroxylase 3.4 2.6
1280854 81642412 Chalcone Flavone Isomerase 2.6 2.3
T280870 81642387 Dihydroflavonol 4-reductase 7.4 5.7
1280962 81642432 Chalcone Reductase 5.5 4.]
Signal transduction

12F0100 81642916 Ser/Thr-specific protein kinase 2.2 2
1280163 81642537 GTP-binding Protein, ras-like 5.2 4.7
Unknown

CL81022 81678108 Unknown Protein 3.4 3.1
SW80040 81678344 Unknown Protein 3.5 5.2
12FOOOI 81642988 Unknown Protein 2.2 2.9
TZFOOIO 81642996 Unknown Protein 2.1 3
TZFOOZO 81642927 Unknown Protein 2.3 2.8
12F0137 81643021 Unknown Protein 2.6 2.3
12F0160 81643042 Unknown Protein 2.2 3.]
1280046 81642463 Unknown Protein (contig 222) 4.9 6.2
1280295 81642621 Unknown Protein 3.1 3.8
1280445 81642109 Unknown Protein 6.5 3.9
1280547 81642176 Unknown Protein 3.2 3.1
1280826 81642402 Unknown Protein 2 2.1
1280828 81642404 Unknown Protein 2.7 2
1280984 81642764 Unknown Protein 5.1 5.3
1281012 81642649 Unknown Protein 6.7 5.7
T281095 81642715 Unknown Protein 5.6 6
1281161 81642738 Unknown Protein 2.1 2.3
1281232 81642802 Unknown Protein 3.2 3.2
1281287 81642850 Unknown Protein 2.4 2.6
1281375 81642907 Unknown Protein 2.5 2.6

 

* The ratio was estimated as average value from data points.

41

 

Table 1-6. (Cont’d)

 

 

Clone ID GenBank Acc. Annotation 12/8C* TZJSW*
No hit

SWSOO33 81678339 No hit 2.2 3.2
SW80663 81679126 No hit 2.6 2.2
8W80732 81678837 No hit 2.2 2.1
SW80757 81678855 No hit 2.6 2
SW80820 81679148 No hit 3.3 3.3
SW80855 81679178 No hit 3.5 2.7
SWSO931 81678931 No hit 2.7 3.6
SW80983 81679030 No hit 3.5 2.2
8W81086 81679045 No hit 2.2 2.6
SW81 123 81679072 No hit 2 4
12F0086 81642977 No hit 2.2 2.3
12F0109 81643000 No hit 2.1 2.9
1280038 81642455 No hit (contig 224) 4.3 4.6
1280145 81642532 No hit 6.2 5.4
1280161 81642544 No hit 5.8 4.5
1280278 81642608 No hit 3.7 3.3
1280362 81642070 No hit 3.3 3.4
1280388 81642081 No hit 2.] 2.5
T280392 81642084 No hit 6.2 6.6
1280437 81642106 No hit 3.9 2.6
1280440 81642107 No hit 7.2 7.6
1280550 81642179 No hit 6.8 7.7
1280617 81642214 No hit (contig 216) 7.2 11.2
1280660 81642240 No hit 2.2 2.]
1280837 81642379 No hit 4.1 3.9
1280903 81642337 No hit 3.4 2.9
1280922 81642430 No hit 2.1 2.1
1280944 81642356 No hit 6.3 10.4
1280953 81642359 No hit 4.2 3.4
1280973 81642756 No hit 3.6 3
1280986 81642766 No hit 3.7 4.]
1281067 81642697 No hit 2.2 3.3
1281112 81642719 No hit 2.2 2.5
1281226 81642796 No hit 3.8 5.4
1281268 81642831 No hit 2.9 3
1281331 81642873 No hit 3.1 3
1281334 81642879 No hit 2 3.4

 

* The ratio was estimated as average value from data points.

42

 

(one unknown and two no-hit genes) that are highly redundant in the contig sets (contig
222, 224 and 216) were also highly expressed in transition zone. These results, when
combined with the EST analysis and microarray data, show that secondary metabolism-
related genes are up-regulated in transition zone. No-hit clones were found in a high
proportion in this region as well, possibly indicating the relatively unstudied nature of
this unique plant tissue zone. Interestingly, the bark/cambial region and the transition
zone showed opposite gene-expression patterns, with most of the down-regulated genes '-
in transition zone being up regulated in the bark/cambial region. For example, phloem-
specific protein VEINl was up regulated in bark/cambial region, but it was down-

regulated in transition zone. Such contrasting gene expression is not unexpected

 

considering the functional differences of the two regions. Transition zone is the region
where many metabolic changes occur in the establishment of the inner wood. In fact,
there was a diversity of genes classified into secondary and hormone metabolism as well
as a remarkable increase in the proportion of genes related to secondary metabolism in
the 128 library. This observation indicates that the transcript expression pattern of
transition zone is highly related to secondary metabolism involving the flavonoid

biosynthesis pathway.

Identification of the wood formation-associated genes

To visualize the inner-wood gene expression patterns that could potentially
identify the wood formation-related genes, 1 performed hierarchical clustering of the
arrayed genes based on microarray results. Data from the 2,580 clones on my cDNA

microarray were clustered from the data of two different experiments:

43

 

METJA ATP-binding protein
Chalcone Synthase
Unknown protein
Dihydroflavonol 4-reductase
Cytochrome 85 DIF-f

c-rnyc binding protein
Proline Oxidase
Auxin-induced protein
GTP-binding protein
Naringenin 3-Dioxygenase
LDOX

Chalcone Flavone Isomerase

<0" (3)

 

I'fi- _

GTP-binding protein

Light inducible protein

. ' Alcohol Dehydrogense
Photosystem II protein
Metallothionein-like protein

' Phloem-specific protein Vein 1
Photosystem l Rx center
Lectin-related polypeptide

~ monooxygenase

 

Figure 1-4. Cluster analysis of expression ratios from SWS or TZS vs. BCS
(A) The subcluster of highly expressed genes in inner wood (the sapwood and transition
zone) from total hierarchical clustering display.

(8) The subcluster of high expression genes in the bark/cambial zone from total
hierarchical clustering display. Only hit EST clones were included for cluster analysis.
8C (C2); bark/cambial zone, 8W; sapwood, and T2; transition zone

 

SWS versus BCS and T28 versus 8C8. Information from clustering not only allows for
the identification of related expression patterns of different genes but also shows
expression patterns of individual genes over different experiments. Figure 1-4A
represents genes that show higher expression in inner wood (i.e., sapwood and transition
zone) than in bark/cambial region. Many up-regulated genes in inner wood are not
characterized, but there are some interesting genes related to secondary metabolism
(chalcone synthase, chalcone flavone isomerase and dihydroflavonol 4-reductase),
primary metabolism (proline oxidase, cytochrome 85 DIF—F), and signal transduction (c-
myc binding protein and GTP-binding protein ras-like). This observation suggests that
these genes may be proximately induced in inner wood and their expression in inner
wood may account for the heartwood formation in trunk wood. On the contrary, down-
regulated genes are monooxygenase, alcohol dehydrogenase, metallothionein-like

protein, phloem-specific protein, and so on (Figure 1-48).

Confirmation of M icroarray Results

Even though the data generated from microarray experiments are reproducible,
the data should also be confirmed by northern blot analysis, western blot analysis, or RT-
PCR analysis (Seki et al., 2001; Perez-Amador et al., 2001; Yu et al., 2002). In order to
compare the expression patterns of each region based on the same reference, I conducted
the second microarray experiment using two probes labeled from the RNA populations of
the target sample and the seedling control. Due to the nature of trunk wood samples,
where only about 10% of the cells are live and the presence of wood extractives makes

the isolation of RNA difficult, I was unable to obtain a large amount of mRN A for

45

conventional northern blot analysis. Accordingly, I performed the approach of “Antisense
northern blot analysis” as well as EST results. Expression patterns in the microarray
experiment were confirmed by antisense northern blot analysis (Figure 1-5) using Histon
H 3.2 as control. Considering the data derived from ESTs, microarray, and “antisense
northern blot” analyses; the gene expression profiles reported here are considered

reliable.

46

(A) Microarray (B) Antisense Northern Blot

 

SD BC SW TZ SD BC SW TZ
Phloem specific 2...... .
protein 1 2.8 0.1 0.1
Sugar transport
protein 1 0.3 1.2 0.2
Ser/T hr specrflc 1 5.5 45 63

 

protein kinase
WRKY-like protein 1 11 64 56

Auxin-induced protein
(TZS1 033)

PR-10 1 0.1 0.7 1.7

 

1 0.6 2.4 4.6

Cytochrome b5 DlF-F 1 1.1 3.5 12
Flavonoid 3,5-

hydroxylase 1 2-8 ' 23
Naringenin 3-

dioxygenase 1 8.6 10 50
No hit (T280038,

conti9224) 1 0.5 9.2 45
Histon H 3.2 1 0.8 1.1 1.0

Figure 1-5. Confirmation of microarray Data with antisense northern blot analysis
(A) The expression ratio of the selected genes in microarray analysis with each sample
(8C ; bark/cambial zone, SW; sapwood, and 12; transition zone) versus seedling (SD).
Smaller than 1.0 means down-regulation in the target sample compared to the seedling
control (sd). Greater than 1.0 indicates up-regulation in the target sample. (8) The
expression patterns of the selected genes in seedling, bark/cambial zone, sapwood, and
transition zone by antisense northern blot analysis. Each sample lane contained the equal
amount of 200 ng of aRNAs.

47

 

ITT ,.

Discussion

In addition to the major constituents (i.e., cellulose, hemicellulose, and lignin),
wood contains many extractives including tannins and other polyphenolics, coloring
matter, essential oils, fats, resins, waxes, gum starch, and other secondary metabolites.
These extractives make wood a veritable chemical storehouse that provides many organic
compounds including biocides for biological control of insects and diseases, adhesives,
biofuels, industrial oils, preservatives, pharmacologically active compounds, and rubber.
Little is known about the molecular basis for such chemical diversity in heartwood. In
order to gain insight on the temporal, spatial, and developmental regulation of the genes
involved in the processes of heartwood and its extractive formation, I carried out global
examination of gene expression profiles across the stems of mature heartwood-forming
trees. It is very difficult to obtain high quality and quantity mRNA from the limited
number of live inner wood cells impregnated with extractives. In this study, 1
characterized the transcriptional profile of black locust wood trunk through the
generation of 2,915 ESTs. Assembly of 2,915 ESTS estimated the maximum number of
unique genes represented in this set to be 2,278. Because this analysis was performed on
5’end sequences that may arise from multiple non-overlapping segments of the same
cDNA, the true number of unique genes may be overestimated. Of the 1,304 matched
ESTs that were analyzed, the largest number (30 %) were uncharacterized genes,
categorized as hypothetical or unknown proteins. Many of the known transcripts
belonged to groups related to housekeeping genes, such as those involved in protein
synthesis and processing. The low representation of genes associated with defense in the

SWS library and with secondary and hormone metabolism in both the BCS and SWS

48

 

libraries are also notable considering the physiological importance of the regions. Thus,
the libraries from the different trunk regions have distinctive characteristics based on the
position of the sample in the wood. They show a common overall trend of the proportion
of genes in the various categories, but with differential expression in some categories that
highlights the unique metabolic characteristics of each zone.

The results of my contig analysis show that functional categorization also presents
the specified function of each tissue region. A direct proportion comparison of functional
categories within each library could not be completed reliably, due to the high percent of
no hit clones in the SW8. So, I excluded the portion of no-hit clones and compared the
proportions of functional categories in each library. All three libraries contained similar
percents of genes in their unknown function groups, as well as their housekeeping
function category, which is related to protein synthesis and processing. When compared
with other libraries, the proportion of membrane transport- related genes was high in the
BCS, but the category of secondary and hormone metabolism membrane was highly
scored in the 128.

It is predicted that the bark/cambial region library would contain many genes
categorized into the cytoskeleton, vesicle trafficking, cell division and cycle functioning
categories because this region includes actively growing and differentiating cells.
However, even though representatives of these classes were found in the library, the
proportions were not significantly different from those of other libraries. Additionally, I
found that this library has a comparatively large number of genes classified into
membrane transport such as: aquaporin, aquaporin-like protein PIP2, plasma membrane

intrinsic protein, plasma membrane integral protein ZmPIP2-7, and delta type tonoplast

49

 

intrinsic protein. Many transcripts (e.g., the genes encoding phloem-specific proteins) of
unknown function or no hit were also highly expressed in this region. When considering
the cell growth and division in this region, the proteins encoded by the genes in each of
these categories are of interest for future analysis.

The sapwood sample included developing-xylem cells. Like the ESTS derived
from the developing-xylem cells of pine and popular, my sapwood library contained a
higher concentration of those genes involved in cell wall synthesis, when compared to the
other two libraries. However, this library possessed only a few transcripts coding for
enzymes involved in the synthesis of lignin. Instead, there were clones corresponding to
cell wall structural proteins including: extensin-like proteins, proline-rich proteins,
leucine-rich repeat proteins, arabinoxylan arabinofuranohydrolase isoenzyme AXAH-II,
and pectinacetylesterases. These results indicate that sapwood gene expression patterns
more closely resemble the characteristics of inner wood gene expression than that of
developing-xylem. This finding parallels the fact that sapwood is a part of inner wood
and has ray cells, which remain alive and maintain their metabolic activity. Based on
microarray results few genes are specifically expressed in the sapwood when compared
with bark/cambial region and transition zone. One possible explanation for the small
number of specifically expressed genes in sapwood is that this region might play bridging
roles between bark/cambial region and transition zone.

As in the EST results, genes involved in secondary metabolism were highly
expressed in transition zone. Recent research has shown a specific example of this in
black walnut (Juglans nigra), where flavonoid biosynthesis was up-regulated in the

transition zone (Beritognolo et al., 2002). In the July samples of Juglans nigra, both

50

 

chalcone synthase (CH8) and flavonoid 3’-hydroxylase (F3H) reached their maximum
levels of expression in the transition zone, while phenylalanine ammonia lyase (PAL) had
increased expression in the outer sapwood. Its activity was much higher in the transition
zone than the sapwood of black locust (Magel and Hilbner, 1997). In the current study, its
expression was increased only in the sapwood (8-fold), but no change in the transition
zone. This may reflect the difference in the genotypes used in the studies, sampling time,
age, or environmental conditions where the trees were grown. This suggests that the basis
for the heartwood chemical diversity among different trees may be at the transcriptional
level. My study showed that CH8 was up-regulated 5-fold in the transition zone as
compared to the bark/cambium region, 4-fold compared to the sapwood, and the level of
F3H was increased about 3-fold in the transition zone. These results corroborate with the
gene expression patterns observed in black walnut (Beritognolo et al., 2002). The
expression of dihydroflavono] 4-reductase (DFR) could not be detected by northern blot,
only by RT-PCR in the trunk of walnut, and showed no significant change across the
stem sections. However, it was up-regulated 7- and 6-fold in the transition zone of black
locust as compared to the bark/cambium region and the sapwood, respectively. This
differential expression of the structural genes may explain the difference in the heartwood
chemical profiles of the two species. The high expression of cytochrome 85 in the
transition zone may be required for flavonoid biosynthesis during heartwood formation.
This gene is known to enhance the activity of flavonoid 3’,5’-hydroxy1ase, which
catalyzes the 3', 5'-hydroxy1ation of dihydroflavonols, the precursors of purple
anthocyanins (de Vetten, et al., 1999). I found that the expression level of 85 DIF-F

(accession number 81642157) was up-regulated 6- and 5-fold in the transition compared

51

to the bark/cambium region and the sapwood, respectively. Accordingly, the expression
of flavonoid 3’,5’-hydroxylase gene was dramatically increased in the transition zone.
The high level expression of the genes related to flavonoid biosynthesis in the transition
zone fits well with the darkening of the heartwood that is mediated by these genes. The
pathogenesis-related class 10 (PR-10), which is a protein related to defense, was also
highly expressed in the transition zone, as well as the sapwood region. WRKY, which is a
family of plant-specific zinc-finger-type transcriptional factors, was expressed in trunk
wood at low levels. Eulgem et al. (1999) reported that the promoter of PR-10 gene is
regulated by WRKY proteins as an early defense response in parsley. In addition, Ser/Thr
specific protein kinase genes and auxin-induced protein genes were also highly expressed
in the transition zone.

The stems of growing trees are characterized by low oxygen and high carbon
dioxide (C02) concentrations (Carrodus and Triffett, 1975). Under such conditions,
Magel (2000) suggested that the products of the accelerated oxidative pentose-phosphate
pathway might be used for the synthesis of phenolics. My microarray has two pentose-
phosphate pathway genes, transaldolase and fructose bisphosphate aldolase. Both of those
genes were up-regulated in the sapwood. The plant hormone ethylene has been suggested
as an important regulator of heartwood formation (Nilsson etal., 2002). This view is
further supported by the fact that ethylene production was greater in the transition zone
than in the outer sapwood (Nelson, 1978) and it stimulates the activity of important
enzymes for polyphenol biosynthesis (Roberts and Miller, 1983; Ingermarsson, 1995).
Neither of the ethylene receptor gene (accession number 81677627) and the ethylene-

responsive small GTP-binding protein (accession number 81677741) on my microarray

52

showed changes in their expression levels across the stem.

One important question in the study of heartwood formation is the source of the
carbon-skeletons used for the biosynthesis of heartwood extractives. The evidence
gathered so far support the hypothesis that the substrates are derived from imported
carbohydrates (Magel and Hubner, 1997; Hauch and Magel, 1998; Magel, 2000). The
microarray analysis showed that the gene encoding sucrose transporter (accession number
81642581) was up-regulated 2.8-times in the sapwood, suggesting increased activity of
sucrose transporter in the region. Carbohydrates have been shown to be distributed across
trunk wood (Magel et al., 1994; Uggla et al., 2001), and there are many reports
confirming that sugar transporters play a role in the cell-to—cell and long-distance
distribution of sugars throughout the plant (Williams et al., 2000). The high expression of
sugar transport protein genes in the sapwood region may indicate that carbohydrates from
source tissues are transported to inner wood by the sugar transport proteins. During
summer months, starch is accumulated in the sapwood and its accumulation correlates
with enhanced sucrose synthase (SuSy) activities in the sapwood (Magel, 2000).
Especially, the activity of SuSy increases dramatically at the sapwood-heartwood
transition zone, which may lead to enhanced degradation of sucrose in the region where
the synthesis and accumulation of phenolic heartwood extractives occur (Magel et al.,
1994; Magel and Huber, 1997; Magel, 2000). The activity of SuSy has been proposed as
a measure for sink strength in tissues with extensive synthesis of phenolic compounds
(Magel, 2000). In this case, my data add additional evidence for the view that heartwood
extractives are synthesized at the transition zone using imported carbohydrates, not

translocated via the phloem and the wood rays to the heartwood (Steward, 1966).

53

In summary, I report the first comprehensive study of gene expression profiles
deep inside trunk wood of a hardwood tree. My ESTS present unique gene sets that are
expressed in uncharted plant tissues. Using DNA microarray analysis, I have identified
genes associated with inner wood formation and profiled gene expression patterns in
trunk wood. These genes will assist future investigations to unravel the molecular
mechanisms regulating the formation of inner wood. Resolving the dilemma of achieving
greater environmental protection of forest ecosystems while meeting the increasing
demand for forest utilization necessitates gaining a fundamental understanding of the
biochemical processes involved in tree growth and deve10pment. The findings described

here will provide a platform for such attempts.

54

Literature Cited

Adams, M., Kerlavage, A., Fields, C., and Venter, J. 1993. 3,400 new expressed sequence
tags identify diversity of transcripts in human brain. Nat Genet 4: 256-267.

Allona, 1., Quinn, M., Shoop, E., Swope, K., Cyr, S., Carlis, J., Riedl, J., Retzel, 8.,
Campbell, M., Sederoff, R., and Whetten, R. 1998. Analysis of xylem formation in
pine by cDNA sequencing. Proc Natl Acad Sci USA 95: 9693-9698.

Audic, S. and Claverie, J-M. 1997. The significance of digital gene expression profiles.
Genome Research 7: 986-995.

Bairoch, A. and preiler, R. 2000. The SWISS-PROT protein sequence database and its
supplement TrEMBL in 2000. Nucleic Acids Res. 28: 45-48.

Barker, W., Garavelli, J., Huang, H., McGarvey, P., Orcutt, 8., Srinivasarao, G., Xiao,
C., Yeh, L.-S.L., Ledley, R., Janda, J., Pfeiffer, F., Mewes, H.-W., Tsugita, A. and
Wu, C. 2000. The Protein Information Resource (PIR). Nucleic Acids Res. 28: 41-44.

Baugh, L., Hill, A., Brown, 8., and Hunter, C. 2001. Quantitative analysis of mRNA
amplification by in vitro transcription. Nuccleic Acid Res. 29: e29.

Benson, D., Karsch-Mizrachi, 1., Lipman, D., Ostell, J ., Rapp, 8., and Wheeler, D. 2000.
GenBank. Nucleic Acids Res. 28: 15-18.

Beritognolo, 1., Magel, E., Abdel-Latif, A., Charpentier, J., Jay-Allemand, C., and
Breton, C. 2002. Expression of genes encoding chalcone synthase, flavanone 3-
hydroxylase, and dihydroflavonol 4-reductase correlates with flavanol accumulation
during heartwood formation in Juglans nigra L. Tree Physiol 22: 291-300.

Bevan, M., Bancroft, 1., Bent, 8., Love, K., Goodman, H., Dean, C., Bergkamp, R.,
Dirkse, W., Van Staveren, M., Stickema, W., Drost, L., Ridley, P., Hudson, S.A.,
Patel, K., Murphy, 6., Piffanelli, P., Wedler, H., Wedler, E., Wambutt, R.,
Weitzenegger, T., Pohl, T.M., Terryn, N., Gielen, J ., Villarroel, R., and Chalwatzis,
N. 1998. Analysis of 1.9 Mb contiguous sequence from chromosome 4 of
Arabidopsis thaliana. Nature 391: 485-488.

Burtin, P., Jay-Allemand, C., Charpentier, J., and Janin, G. 1998 Natural wood colouring
process in Juglans sp. (J. nigra, J. regia and hybrid J. nigra 23 x J. regia) depends on
native phenolic compounds accumulated in the transition zone between sapwood and
heartwood. Trees 12: 258-264.

Carrodus, 8., and Triffett, A. 1975. Analysis of respiratory gases in woody stems by mass
spectrometry. New Phytol. 74: 243-246.

55

Cercos, M., Santamaria, S., and Carbonell, J. 1999. Cloning and characterization of

TPE4A, a thiol-protease gene induced during ovary senescence and seed germination
in pea. Plant Physiol 119: 1341-1348.

Covitz, P., Smith, L., and Long, S. 1998. Expressed Sequence Tags from a Root-Hair-
Enriched Medicago truncatula cDNA library. Plant. Physiol. 117: 1325-1332.

Dazzo, F., and Hubbell, H. 1975. Cross-reactive antigens and lectins as determinants of
symbiotic specificity in the Rhizobium—clover association. Appl. Microbiol. 30: 1017-
1033.

de Vetten, N., ter Horst, J., van Schaik, H.-P., de Boer, A., Mol, J., and Koes, R. 1999. A
cytochrome b5 is required for full activity of flavonoid 3',5'-hydroxylase, a
cytochrome P450 involved in the formation of blue flowers. Proc Nat] Acad Sci USA
96: 778-783

Dent, G., O’Dell, D, and Eberwine, H. 2001. Gene expression profiling in the amygdala:
An approach to examine the moleculat substrates of mammalian behavior. Physiol. &
Behavior 73: 841-847.

Eisen, M., Spellman, P., Brown, R, and Botstein, D. 1998. Cluster analysis and display of
genome-wide expression patterns. Proc. Natl. Acad. Sci. USA 95: 14863-14868.

Eulgem, T., Rushton, P., Schmelzer, E., Hahlbrock, K., and Somssich, I. 1999 Early
nuclear events in plant defense signaling: rapid gene activation by WRKY
transcription factors. EMBO J. 18: 4689-99.

Ewing, 8., Hillier, L., Wendl M., and Green, P. 1998. Base-calling of automated
sequencer traces using phred. 1. Accuracy assessment. Genome Res. 8: 175-85.

Fukuda, H. (1996). Xylogenesis: initiation, progression, and cell death. Ann Rev Plant
Physiol Plant Mol Biol 47, 299-325.

Garcia-Hernandez, M., Murphy, A., and Taiz, L. 1998. Metallothioneins 1 and 2 have
distinct but overlapping expression patterns in Arabidopsis. Plant Physiol. 118: 387-
397.

Hauch, S. and Magel, E. 1998. Extractable activities and protein content of sucrose-
phosphate synthase, sucrose synthase and neutral invertase in trunk tissues of Robinia
pseudoacacia L. are related to cambial wood production and heartwood formation.
Planta 207: 266-274

Hertzberg, M., Aspeborg, H., Schrader, J., Andersson, A., Erlandsson, R., Blomqvist, K.,
Bhalerao, R., Uhlén, M., Teeri, T., Lundberg, J., Sundberg, 8., Nisson, P., and
Sandberg, G. 2001. A transcriptional roadmap to wood formation. Proc. Natl. Acad.
Sci. USA 98: 14732-14737.

56

 

Hihara, Y., Kamei, A., Kanehisa, M., Kaplan, A., and Ikeuchi, M. 2001. DNA
Microarray Analysis of Cyanobacterial Gene Expression during Acclimation to High
Light, Plant Cell 13: 793-806

Hillinger, C., Holl, W., and Ziegler, H. 1996a. Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. 1. Radial
distribution of lipid classes. Trees 10: 366-375.

Hillinger, C., Holl, W., and Ziegler, H. 1996b. Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. 11. Radial
distribution of lipases and phospholipases. Trees 10: 376-381.

Hillis W. 1987. Heartwood and the exudates, Spinger, Berlin, Heidelberg, New York.

Ingemarsson, 8. 1995. Ethylene effects on peroxidases and cell growth patterns in Picea
abies hypocoty] cuttings. Physiol. Plant. 94: 21 1-218

Kirst, M., Johnson, A., Retzel, E., van Zyl, L., Craig, D., Li, 2]., Whetten, R., Baucom,
C., Ulrich, E., Hubbard, Kristy, and Sederoff, R. (2002). Quantitative inference in
functional genomics of loblolly pine (Pinus taeda L.) using ESTs and microarrays. In
Proc. the 10th IAPTC&B Congress. (Ed) I. Vasil. Kluwer Academic Publishers,
Dordrecht, Netherlands.

Kozlowski, T. and Pallardy, S. (1997). Physiology of woody plants. San Diego,
Academic Press.

Lalonde, S., Boles, E., Hellmann, H., Barker, L., Patrick, J., Frommer, W., and Ward, J.
1999. The dual function of sugar carriers: transport and sugar sensing. Plant Cell 11:
707-726.

Lee, M., Kuo, F., Whitmore, G., and Sklar, J. 2000. Importance of replication in

microarray gene expression studies: Statistical methods and evidence from repetitive
cDNA hybridizations. Proc. Nat]. Acad. Sci. USA 97: 9834—9839.

Livesey, F., Furukawa, T., Steffen, M., Church, G., and Cepko, C. 2000 Microarray
analysis of the transcriptional network controlled by the photoreceptor homeobox
gene Crx. Current Biology 10:301-310.

Lorenz, W. and Dean, J. 2002. SAGE profiling and demonstration of differential gene

expression along the axial developmental gradient of lignifying xylem in loblolly pine
(Pinus taeda). Tree Physiology 22: 301-310.

Magel, E., Jay-Allemand, C., and Ziegler, H. 1994. Formation of heartwood substances
in the stemwood of Robinia pseudoacacia L.: 11. Distribution of nonstructural
carbohydrates and wood extractives across the trunk. Trees 8: 165-171.

Magel, E. and Hubner, 8. 1997 Distribution of phenylalanine ammonia lyase and
chalcone synthase within trunks of Robinia pseudoacacia L. Bot. Acta 110: 314-322

57

 

In] ..

Magel, E. 2000. Biochemistry and physiology of heartwood formation. In Molecular and
Cell Biology of Wood Formation. Eds. R. Savidge, J. Barnett and R. Napier. 8108
Scientific Publishers, Oxford, pp363-376.

Matamoros, M., Baird, L., Escuredo, P., Dalton, D., Minchin, F., Iturbe—Ormaetxe, 1.,
Rubio, M., Moran, J., Gordon, A., and Becana, M. 1999. Stress-induced legume root
nodule senescence. Physiological, biochemical, and structural alterations. Plant
Physiol 121: 97-112.

Mauseth, J. 1998. Botany: an introduction to plant biology. Jones and Bartlett Publishers.
Sudbury, Massachusetts.

Miller, J., Arteca, R., and Pell, E. 1999. Senescence-associated gene expression during
ozone-induced leaf senescence in Arabidopsis. Plant Physiol 120: 1015-1024.

Nelson, N. 197 8. Xylem ethylene, phenol oxidising enzymes and nitrogen and heartwood
formation in walnut and cherry. Can. J. Bot. 56: 626-634

Nilsson, M., Wikman, S., and Eklund, L. 2002. Induction of discolored wood in Scots
pine (Pinus sylvestris). Tree Physiology 22: 331-338

Perez-Amador, M., Lidder, P., Johnson, M., Landgraf, J., Wisman, E., and Green, P.
2001. New molecular phenotypes in the dst mutants of Arabidopsis revealed by DNA
microarray analysis. Plant Cell 13: 2703-2717.

Reddy, A., and Poovaiah, 8. 1990. Molecular cloning and sequencing of a cDNA for an
auxin-repressed mRNA: correlation between fruit growth and repression of the auxin-
regulated gene. Plant Mol Biol 14: 127-136.

Roberts, L. and Miller, A. 1983 Is ethylene involved in xylem of differentiation? Plant
Physiol. 6: 1-24

Saxe, H., Ellsworth, D. S., and Heath, J. 1998. Tree and forest functioning in an enriched
C02 atmosphere. New Phytol. 139: 395-436.

Schaffer, R., Landgraf, J., Accerbi, M., Simon, V., Larson, M., and Wisman, E. 2001.
Microarray Analysis of Diurnal and Circadian-Regulated Genes in Arabidopsis. Plant
Cell 13: 113-123.

Sederoff, R., Kirst, M., Johnson, A., Retzel, E., Whetten, R., Vasques-Kool, J. &
O’Malley, D. 2002. Homology of expressed genes in loblolly pine (Pinus taeda L.)
with Arabidopsis thaliana. The 10th IAPTC&8 Congress, February 23-28, 2002,
Orlando, FL.

Seki, M., Narusaka, M., Abe, H., Kasuga, M., Yamaguchi-Shinozaki, K., Caminci, P.,
Hayashizaki, Y., and Shinozaki, K. 2001. Monitoring the Expression Pattern of 1300
Arabidopsis Genes under Drought and Cold Stresses by Using a Full-Length cDNA
Microarray. Plant Cell 13: 61-72.

58

Stafstrom, J., Ripley, 8., Devitt, M., and Drake, 8. 1998. Dormancy-associated gene
expression in pea axillary buds. Cloning and expression of PsDRMl and PsDRM2.
Planta 205: 547-552.

Sterky, F., Regan, S., Karlsson, J., Hertzberg, M., Rohde, A., Holmberg, A., Amini, 8.,
Bhalerao, R., Larsson, M., Villarroel, R., Van Montagu, M., Sandberg, G., Olsson,
O., Teeri, T. T., Boerjan, W., Gustafsson, P., Uhlen, M., Sundberg, 8., and
Lundeberg, J. 1998. Gene discovery in the wood-forrning tissues of poplar: analysis
of 5, 692 expressed sequence tags. Proc Natl Acad Sci U S A 95: 13330-13335.

Steward, C. 1966. Excretion and heartwood formation in living trees. Science 153: 1068-
1074.

Uggla, C., Magel, E., Moritz, T., and Sundberg, 8. 2001. Function and dynamics of auxin
and carbohydrates during early wood/late wood transition in Scots pine. Plant Physiol
125: 2029-2039

Wang, E., Miller, L., Ohnmacht, G., Liu, 8., and Marincola, F. 2000. High-fidelity
mRNA amplification for gene profiling. Nat. Biotech. 18: 457-459.

Williams, L., Lemoine, R., and Sauer, N. 2000. Sugar transporters in higher plants - a
diversity of roles and complex regulation. Trens in Plant Science 5: 283-290

Wingler, A., von Schaewen, A., Leegood, R., Lea, P., and Quick, W. 1998. Regulation of
leaf senescence by cytokinin, sugars, and light. Plant Physiol 116: 329-335.

Yu, Q., He, M., Lee, N., and Liu, E. (2002). Identification of MYC-mediated death
response pathways by microarray analysis. J Biol Chem 277: 13059-13066.

59

 

CHAPTER 2

Seasonal changes in gene expression at the sapwood-heartwood transition zone of

black locust (Robinia pseudoacacia) revealed by cDN A microarray analysis

ABSTRACT
Heartwood is a determining factor of wood quality and understanding the biology of

heartwood may allow us to control its formation. Heartwood formation is a form of
senescence that is accompanied by a variety of metabolic alterations in ray parenchyma
cells at the sapwood—heartwood transition zone. Although senescence has been studied at
the molecular level with respect to primary growth, the cell maturation and death events
occurring during heartwood formation have been difficult to study because of their
location and timing. Analysis of global gene expression patterns during the transition
from sapwood to heartwood may offer a powerful means of identifying the mechanisms
controlling heartwood formation. Previously, I developed cDNA microarrays carrying
2,567 unigenes derived from the bark/cambium region, sapwood and transition zone of a
mature black locust tree. Here, I describe the use of these microarrays to characterize
seasonal changes in the expression patterns of 1,873 genes from the transition zone of
mature black locust trees. When samples collected in summer and fall were compared,
569 genes showed differential expression patterns: 293 genes were up-regulated (>
twofold) in summer (July 5) and 276 genes were up-regulated in fall (November 27).
More than 50% of the secondary and hormone metabolism-related genes on the
microarrays were up-regulated in summer. Twenty-nine out of 55 genes involved in
signal transduction were differentially regulated, suggesting that the ray parenchyma cells

located in the inner-most part of the trunk wood react to seasonal changes. I established

60

the expression patterns of 349 novel genes (previously unknown or no-hit), of which 154

were up-regulated in summer and 195 were up-regulated in the fall.

Introduction

Most tree species have a dark-colored zone called heartwood in the central parts
of their trunks. Besides the difference in color, heartwood is distinguished from the
surrounding pale-colored sapwood by features such as a greater content of extractives, a
lower water content and parenchyma cell death. The presence of heartwood is the major
factor determining wood quality. Many characteristics of wood, including dimensional
stability, durability, pulpability, color, hue, scents and beauty, are affected by the
presence of various organic substances classified as extractives. Identifying changes in
gene expression during the transition from sapwood to heartwood may help us control
heartwood formation.

Heartwood results from cell death and contains no living cells. Studies of
heartwood formation have focused on metabolic changes and extractive formation in a
narrow zone, called the transition zone (12), adjacent to the heartwood. During the
transition from sapwood to heartwood, parenchyma cells of the sapwood undergo
metabolic changes that result in heartwood formation and increased synthesis of
secondary products such as condensed tannins, terpenes, flavonoids, lignans, lipids,
stilbenes and tropolones (Hillis, 1987; Magel et al., 1991, 1994; Hillinger eta1., 1996a,
1996b; Burtin et al. 1998). There is much evidence that heartwood extractives are
synthesized in situ in the sapwood-heartwood transition zone from the breakdown of

starch or from soluble sugars (Kumar and Datta, 1987; Nobuchi et al., 1987a, 1987b;

6]

Magel etal., 1994; Hillinger et al., 1996a, 1996b; Magel and Htlbner, 1997). Starch
hydrolyzed at the transition zone is a primary source of hydroxycinnamic acid and
flavonoids (Magel et al., 1994). A period of enhanced metabolic activity has been found
at this zone in heartwood-forming species, such as Acacia (Baqui and Shah; 1985), black
locust (Robinia pseudoacacia L.; H011 and Lendzian, 1973; Magel et al., 1991), oak
(Quercus robur L.; Ebermann and Stich, 1985) and walnut (Juglans nigra L.; Nelson et
al., 1981). It has also been found that formation and accumulation of heartwood phenolics
coincides with the transformation of sapwood to heartwood in black locust (Magel et al.,
1994). Hillis and Hasegawa (1963) noted that, 19 days after labeled glucose was
administered to the cambial region of Eucalyptus sieberi L. A. S. Johnson, it was
converted to labeled extractives formed at the heartwood periphery. This conversion was
also observed in the transition zone of sugi (Cryptomeria japonica D. Don; I-liguchi et al.,
1969).

Several enzymes show high activity in the transition zone. Elevated activities of
phenol oxidizing enzymes have been observed in the transition zone of Eucalyptus
polyanthemos Schauer(1-Iillis and Yazaki, 1973). High peroxidase activity was observed
in the transition zone of Eucalyptus elaeophora F. Muell. (W ardrop and Cronshaw, 1962)
and F agus sylvatica L. (Dietrichs, 1964). Baqui et al. (1979) found that succinate
dehydrogenase, adenosine triphosphatase and lipase had enhanced activity only in the
transition zone of Melia azedarach L. Other enzymes reported to be highly active in the
transition zone include catechol oxidase, glucose-6-phosphate dehydrogenase, malic
dehydrogenase, maltase and amylase (Hillis, 1987). Furthermore, Magel et al. (1991)

have shown that two key enzymes for flavonoid biosynthesis (chalcone synthase (CH8)

62

and phenylalanine ammonia-lyase (PAL)) are highly active in the sapwood—heartwood
transition zone of black locust; PAL was active in the youngest wood near the cambium
in April and September and active in the transition zone in all seasons, whereas CH8 was
active only in the intermediate wood. The stimulus for biosynthesis of robinetin, a
flavonoid with strong antimicrobial activity, apparently occurs in the transition zone in
Intsia species (Hillis, 1996).

In temperate zones, heartwood formation occurs in late summer and continues
through late fall and the beginning of dormancy, provided temperatures are above 5 °C
(Hillis, 1987). Heartwood formation occurs at the time of cambial dormancy in radiata
pine (Pinus radiata D. Don) (Shain and Mackay, 1973), walnut and cherry (Nelson,
1978). Studies of the cytology and coloration of extractives suggest that heartwood
formation commences in midsummer and continues in the fall and winter seasons in sugi
(Nobuchi et al., 1987a) and black locust (Nobuchi et al., 198 7b). Beritognolo et al. (2002)
studied seasonal changes in the expression of several flavonoid biosynthetic genes during
heartwood formation in Juglans nigra and demonstrated that flavonol accumulation in
the sapwood—heartwood transition zone was correlated with transcript levels of key
flavonoid structural genes.

Black locust is one of the most extensively studied hardwood species with regard
to heartwood formation (Magel et al., 1991, 1994; Hillinger et al., 1996a, 1996b; Hauch
and Magel, 1998). Its sapwood is yellowish and narrow, whereas the heartwood ranges
from green to greenish yellow, dark yellow or golden brown. I chose to work with black
locust because it is a heartwood-forming species with clearly distinguished sapwood and

heartwood zones. In addition, black locust has a relatively small genome (637 Mbp),

63

making it a suitable species for genomic experiments. I previously isolated mRNA from
the inner wood of mature black locust trees (> 10 years old) and analyzed a large number
of expressed sequence tags (2,915 ESTS) (Yang et al., 2003). I then constructed cDNA
microarrays carrying 2,567 unigenes identified from the EST analysis (a list of the
unigenes is available at http://forestry.msu.edu/biotech/Projects.htm). To gain an
increased understanding of the genetic regulation of heartwood formation, 1 report here a
series of microarray hybridization experiments performed to examine changes in gene
expression in the sapwood—heartwood transition zone of black locust during active

growth (July) and during the transition from active growth to dormancy (November).

Materials and Methods

Plant material

In summer (July 5) and fall (November 27), two mature 10-year-old (20—cm
diameter at breast height) black locust trees growing at the Tree Research Center on the
campus of Michigan State University were felled and cut into 25-cm-long logs. The logs
were immediately placed on ice and transported to a wood shop where ~1-cm-thick disks
were cut with a table saw. The disks were immediately submerged in RNA extraction
buffer (50 mM Tris-HCl buffer, pH8.0, containing 20 mM EDTA, 0.2% SDS and 10 mM
2-mercaptoethanol). The sapwood—heartwood transition zone, located 2—4 cm inside the
outer edge of each disk, was carved out with a chisel and hammer that had been
prewashed with RNase Away solution (Invitrogen, Carlsbad, CA). The isolated sections
(~l cm 3 cubes) of the transition zone sampled in summer (128) and fall (12F) were

frozen in liquid nitrogen and stored at —80 °C until use.

Microarray preparation

The construction of microarrays carrying 2,567 unigenes that were expressed in
the bark/cambium, sapwood and transition zone of a mature black locust has been
described previously (Yang et al., 2003). Briefly, the PCR-amplified products of the
selected unigenes were precipitated with ethanol, re-suspended in 3 x SSC (1 x SSC is
0.15 M NaCl and 0.015 M sodium citrate) and checked for quality by gel electrophoresis.
The PCR products were arrayed from 384-well microplates and spotted on superaldehyde
glass slides (Telechem, Sunnydale, CA) at a high density with an Omnigridder robot
(Gene Machines, San Carlos, CA) and a 2-pin 16 ArrayIt chipmaker (Telechem). Each
slide contained duplicate spots for each gene. Slides were rinsed twice in 0.2% SDS for 2
min, and washed in distilled water. To denature the DNA, the slides were transferred to
boiling water for 2 min. After denaturing the DNA, the slides were dipped in blocking
solution (1.5 g Na8H4 in 450 ml of phosphate buffered saline to which 133 ml of 100%
ethanol was added) for 5 min to reduce free aldehydes, and then washed with 0.2% SDS

solution and distilled water.

Hybridization and washing of the DNA microarray

For total RNA isolation, frozen 128 and 12F samples were ground in a blender
and reground to fine powder with a mortar and pestle. The ground samples were then
passed through the shredder column of a DNeasy Maxi kit subjected to total RNA
isolation using the RNeasy Maxi kit (Qiagen, Hilden, Germany), and cleaned up with a
Qiagen RNeasy Mini kit. The cDN As were synthesized in reverse transcription reactions

using 1 ug of total RNA and then PCR-amplified using the SMART system (ClonTech,

65

Palo Alto, CA) and the fewest possible cycles (15—18 cycles) to minimize generation of
non-specific PCR products. To generate probes for array hybridization, 2 pg of the PCR-
amplified cDNAs were labeled by incorporation of either CyS- or Cy3-dCTP
(Amersham-Pharmacia, Piscataway, NJ) during oligo-dT-primed primer extension in the
presence of Klenow DNA polymerase (Promega, Madison, WI), as described by Schaffer
et a]. (2001). The labeling reactions were purified with the QiaQuick PCR cleanup kit
(Qiagen). Probe samples were denatured at 100 °C for 3 min, left at room temperature for
30 min, and then used for hybridization. Probes labeled with either Cy3 or Cy5
fluorescent dye were hybridized to a microarray slide in a total volume of 30 u] of
hybridization buffer (3.4 x SSC, 0.32% SDS, and 5 ug of yeast tRNA) for 16 h at 65 °C.
Three replicate slides were hybridized for each sample pair, including one dye swap
experiment in which the labeling of the sample in each pair was reversed on the second
slide by using independent RNA preparations. The third slide was a repeat of the first
hybridization to check the reliability of the results. Each slide was then washed at room
temperature in 1 x SSC, 0.1% SDS for 10 min, in l x SSC for 10 min, and in 0.01 x SSC
for 10 min. The slides were centrifuge-dried and scanned with a 428 Array scanner

(Affymetrix, Palo Alto, CA).

Microarray data analysis

Microarray data were analyzed with GenePix Pro 3.0 (Axon Instruments, Union
City, CA). The scanned data were normalized by the Global Normalization method
(Hihara et al., 2001), which normalizes the image data between Cy3 and Cy5 channels by

adjusting the total signal intensities of two images and removing unreliable spots. The

66

unreliable spots were discarded based on the following screening. Spots containing
clones that had poor amplification, or multiple bands, as well as those that were flagged
because of a false intensity caused by dust or background on the array, were removed.
Spots with < 65% of the spot intensity at > 1.5-fold that of the background in both
channels were ignored (see Stanford Microarray Database Web site, httpzl/genome-
www5.stanford.edu/Micro-Array/SMD/). Clones in one sample that had an average
induction greater than twofold in another were determined as up-regulated. Data
management and analyses were carried out with Microsoft Excel and Microsoft Access
database. After normalization, I calculated the means and coefficients of variation for the
observed signal intensities in each channel and the ratio of signals from three replicates.
All data were then converted to base 2 logarithm. Pairwise comparisons were made by
simple linear correlation analysis to evaluate variations within and between microarrays
and within and between slides (Fernandes et al., 2002, Swidzinski et al., 2002). To
determine the strength of the relationship between replicates, the Pearson correlation

coefficient (r) was calculated.

Antisense Northern blot analysis

Antisense RNA (aRN A) was generated as described by Wang et al. (2000).
Briefly, aRNA was amplified with a Message Amp aRNA kit (Ambion, Austin, TX). I
began by synthesizing first stand cDNAs from the total RNA of seedling, bark/cambial
region, sapwood, or transition zone, and the first cDNAs were used as templates for the
synthesis of second cDNAs. Finally, aRN As from the second cDNAs were generated by

in vitro transcription (amplification). About 200 ng of aRN A was amplified, separated in

67

a formamide agarose gel, and transferred onto a nylon membrane by the capillary transfer
method. The membrane was then hybridized with a labeled probe. The signal was
exposed and detected on X-ray film. Histon H3.2 gene was used as a control, which was

similarly expressed in both 128 and 12F based on my microarray results.

Results

Microarray experiments

Correlations between replicate microarrays were high (Table 2-1), confirming that
PCR amplification generated highly reproducible populations of cDNAs. Even the lowest
r value obtained, 0.86, was well within the range of significance (P < 0.05). Hybridization
in dye reversal experiments was more variable between replicate microarrays than
between replicate elements within the same slide (cf. Fernandes et al., 2002). In my
analysis, the highest correlation coefficient (r = 0.96) was between duplicate spots on the
same slide. The r-value of two experiments in which the same dye was used to label each
sample was higher than the r-value of the two-dye reversal experiments. Furthermore,
two experiments using the same dye to label each sample were less variable than two-dye
reversal experiments. These data imply that replication reliability can be affected by dye
reversion and that hybridizations should be replicated at least three times, with at least
one replicate being a dye swap experiment. Statistically, to detect up-regulated genes (>
twofold increase) on the microarrays, the log2 (128/12F) should fall at least two
standard errors away from zero (no change in log2 expression) when considering the
three replicate arrays (Swidzinski et al., 2002). Thus, the log2 standard deviation (SD) for

each gene, averaged across three replicate arrays, must be less than or equal to 0.5.

68

Table 2-1. Summary of correlation coefficients of ratios from TZS versus TZF

Numbers 1 and 2 in parentheses represent duplicate 1 and duplicate 2 of the same slide.
Abbreviations: 128 or T8, transition zone harvested on July 5; 12F or TF, transition

zone harvested on November 27; Cy3, Cy3-deoxyCTP; and Cy5, Cy5-deoxyCTP.

 

 

 

Replication 1 Replication 2 Replication 3
TS-Cy3 TS-Cy3 TS-Cy5 TS-Cy5 TS-Cy3 TS-Cy3
VS. VS. VS. VS. VS.
TF-Cy5(1) TF-Cy5(2) TF-Cy3(1) TF-Cy3(2) TF-Cy5(1) TF-Cy5(2)
TZS-Cy3
VS. - 0.942 0.91 0.89 0.92 0.92
TZF-Cy5 (1)
TZS-Cy3
VS. - 0.92 0.91 0.94 0.93
TZF-Cy5 (2)
TZS-Cy5
VS. - 0.93 0.91 0.93
TZF-Cy3 (1)
TZS-Cy5
VS. - 0.86 0.91
TZF-Cy3 (2)
TZS-Cy3
VS. - 0.96
TZF-Cy5 (1)

 

69

In 1,873 genes out of 2,580 considered in the analysis, 90% of the means had a SD < 0.5.

These 1,873 genes were selected for further analyses.

Difi‘erential gene expression between 728 and TZF

Analysis of the microarray data revealed significant changes in transcript levels of
the selected 1,873 genes. Figure 2-1 shows that the number of genes was normally
distributed in log2 (TZS/ TZF). Only the genes for which the expression varied more than
twofold when averaged across six data points (two duplicates per slide and three
replications) for each treatment were considered to represent significant changes in gene
expression (Girke et al., 2000, Reymond et al., 2000). Genes for which the log2
(TZS/12F) was more than one were considered as highly expressed genes in 128.
Conversely, genes in which the log2 (128/12F) was less than -1 were considered as
highly expressed genes in TZF or as being down-regulated in T28. About 30% of the
1,873 genes selected displayed significant expression changes. A total of 293 genes were
up-regulated in 128 and 276 genes were highly expressed in 12F when compared with
128 (Table 2-2). About 570 genes were differentially expressed between summer and
fall.

To assess seasonal changes (from summer to fall) in gene expression patterns of
the transition zone, I classified the up-regulated genes in 128 or 12F into 14 functional
categories (Table 2-2). The percentage of up-regulated genes to total genes in each
sample was similar: 15.6% for TZS (293 genes out of 1,873 total genes), and 14.7% for
TZF (276 genes out of 1,873 total genes). The percentages of up-regulated genes to total

genes for several categories were similar in the T28 and TZF groups; however, the

70

percentages for some categories varied between the two groups. For example, the
percentage of the genes in the ‘unknown’ or ‘no-hit’ category was higher in 12F than in
128. Conversely, the percentage of genes classified in the ‘cell division and cycle’,
‘defense’ and ‘secondary and hormone metabolism’ categories was higher in 128 than in
12F. Although these findings provide insight into the characteristics of each sample

group, I caution that the numbers of genes included in the ‘cell division and cycle’ and

‘vesicular trafficking, protein sort and secretion’ categories were only six and five, F"
respectively. The high proportion of secondary and hormone metabolism-related genes 5
among the T28 up-regulated genes indicates distinctive features of the 128 sample. :
Additionally, the portion of up-regulated genes related to ‘signal transduction’ in each :
L.,

sample (5.4% in 128, 4.7% in TZF) was higher than the percentage values calculated in
the Total (2.9%). Furthermore, 53% of the genes within the ‘signal transduction’ category

were differentially expressed.

71

160*

140-

120- —-.

1co-

 

Number of genes
8

 

 

 

 

 

 

 

 

 

 

 

 

-2 -1 0
L092 (TZS/TZF)

Figure 2-1. Distribution of mean ratios of expression from TZS and TZF samples
Total RNAs from the transition zone sampled on July 5 (T28) and on November 27
(TZF) were labeled with fluorescence dyes (Cy3 or Cy5) and hybridized with the
microarrays. The ratio for each clone is the mean of three replicates.

72

Table 2-2. Functional classification of up-regulated clones in transition zones
Numbers are of selected clones for microarray analysis after normalization. Numbers in
parentheses indicate the percentage of each number per total number.

 

 

 

Catego_ry Total TZS TZF

Cell division and cycle 6 (0.3) 4 (1.4) 0 (0.0)
Cell wall structure and metabolism 13 (0.7) 2 (0.7) 1 (0.4)
Chromatin and DNA metabolism 26 (1.4) 7 (2.4) 3 (1 .1)
Cytoskeleton 6 (0.3) 0 (0.0) 1 (0.4)
Defense 43 (2.3) 13 (4.4) 7 (2.5)
Gene expression and RNA metabolism 55 (2.9) 7 (2.4) 8 (2.9)
Membrane transport 19 (1 .0) 4 (1.4) 2 (0.7)
Miscellaneous 85 (4.5) 22 (7.5) 15 (5.4)
Primary metabolism 107 (5.7) 31 (10.6) 22 (8.0)
Protein synthase and processing 74 (4.0) 15 (5.1) 8 (2.9)
Secondary and hormone metabolism 31 (1.7) 16 (5.4) 1 (0.4)
Signal transduction 55 (2.9) 16 (5.4) 13 (4.7)
Vesicular trafficking, protein sort, Secretion 5 (0.3) 2 (0.7) 0 (0.0)
Unknown, No hit 1348 (72) 154 (33) 195 (71)
Total clones 1873 293 276

 

73

Genes up-regulated in summer (125')

A tOtal of 293 genes were up-regulated in 128 (139 known and 154 unknown/no
hit genes) (Table 2-3). Three out of four cell division and cycle-related genes that were
up-regulated in 128 were SKPl (suppressor of kinetochore protein) genes. Two cell wall
biosynthesis-related genes were highly expressed in T28. Stress-inducible genes such as
pathogenesis-related class 10 gene (PR-10), and genes coding for aluminum-induced
protein, metallothionein and dehydration induced protein, were highly expressed in 128.
It is notable that the gene coding for aquaporin-like protein PIP2 was up-regulated in
128, as were genes coding for dehydration-induced protein. Two genes coding for auxin-
repressed protein and one auxin-induced protein were up-regulated in 128, suggesting
that the expression of these genes could be differentially regulated in the transition zone.
The gene for sucrose synthase, which is known to be involved in the production of
substrates for heartwood formation in the transition zone (Shah et al., 1981, Nobuchi et
al., 1982), and the gene for sugar transport protein were 3.7- and 3.9-fold up-regulated in
128, respectively. Other 128 up-regulated genes in the ‘primary metabolism’ category
included those encoding cytochrome 85 DIF-F (2.8-fold), phenylalanine ammonia lyase
(PAL, 3.3-fold), adenosine kinase-like protein (4.4-fold) and glycine
hydroxymethyltransferase (5.1-fold). Sixteen of the 31 genes in the ‘secondary and
hormone metabolism’ category were up-regulated in T28. Nine genes were involved in
the flavonoid pathway. Two lignin biosynthesis-related genes, genes encoding Caffeoyl-
CoA o-methyltransferase and Caffeic acid o-methyl-transferase, were up-regulated in
T28. Two genes encoding glutathione S-transferase were more highly expressed in T28

than in 12F. Two genes encoding monooxygenase and one gene for fumarylacetoacetase

74

Table 2-3. Categorized genes whose expression was up-regulated in summer (TZS)

 

 

CIOIICID GenBank Acc. Annotation TZSIIZF
Cell division and cycle
CL80043 81677470 Skpl 2.3
CL80092 81677511 Calcineurin 8 Subunit 2.2
CLSO877 81678229 SKPl 3.0
1280380 81642078 SKPl Homolog 2.8
Cell wall structure and metabolism
CL80298 81677625 Pectinesterase-like protein 2.0
CL80529 81677813 Poly galacturonase-like protein 2.3
Chromatin and DNA metabolism
CL80083 81677503 HMGl/2-like Protein (881] Protein) 2.4
CLSO418 81677721 Histone H4 3.5
CL80521 81677806 Histone H281 2.3
CL80549 81677829 DNA Topoisomerase VI subunit 8-1ike protein 2.2
CL80968 81678080 DNA-Binding Protein, hypothetical 2.2
1280239 81642590 Histone H2A 3.1
T280517 81642156 DNA Polymerase Accessory Protein 2.2
Defense
CL80021 81677451 Dehydration-induced protein 2.7
CL80176 81677536 L-ascorbate peroxidas, cytosolic 5.2
CL80243 81677582 Superoxide Dismutase (Cu-Zn) 2.2
CL80450 81677748 Metallothionein 2.2
CL80513 81677800 Disease Resistance Protein-like 2.3
CL80609 81677874 Peroxidase pde precursor 2.5
1280036 81642453 Aluminum-induced Protein 2.9
T280124 81642508 PR-10 Protein 2.0
1280304 81642629 Metallothionein 2.1
T280357 81642069 PR-lO Protein 4.2
1280535 81642166 Dehydration-induced Protein 2.4
DEHYDRATION STRESS-INDUCED
T280539 81642170 PROTEIN. 5.0
1280643 81642228 Cystein Proteinase Inhibitor 2.3
Gene expression and RNA metabolism
CL80034 81677464 Glycine-rich RNA-binding protein 2.4
CL80074 81677494 Transcription factor 2.2
CLSO394 81677705 Zinc Finger protein DOF 2.7
CLSO497 81677788 C-myc binding protein MM-l-like protein 4.4
CL80515 81677802 RNA polymerase II subunit-like protein 2.6
CL80842 81678199 RNA polymerase II subunit 2.0
SW80968 81678966 Poly(A)-8inding Protein 2.1
Membrane transport
CL80285 81677616 Plasma Membrane Intrinsic Protein 3.4
CL81026 81678112 Secretory Canier-Associated Membrane Protein 2.3
1280226 81642581 Sugar Transport Protein 3.9
1280255 81642596 Aquaporin-like protein PIP2. 3.2

 

75

Table 2-3 (Cont’d)

 

 

CloneID GenBank Acc. Annotation TZS/TZF
Miscellaneous

CLSOO81 81677501 MTNS Gene Precursor 3.3
CL80162 81677524 Germin-like Protein 2.7
CLSO274 81677608 Trypsin Inhibitor (Serine Proteinase Inhibitor) 2.3
CL80464 81677761 auxin-repressed protein 2.8
CL80536 81677820 Core protein 2.9
CLSO669 81678285 P1 starvation-induced protein 2.1
CLSO778 81677935 DnaJ Protein homolog 2.6
CLSO786 81677943 Ribulose-bisphosphate carboxylase 2.2
CL81025 81678111 Ripening related protein 2.2
1280314 81642642 Pollen Specific Protein 2.1
1280319 81642056 Oxygen-evolving Enhancer Protein 3.0
1280363 81642071 Tumor-related Protein 2.8
1280422 81642099 Integrase 2. 1
1280473 81642130 auxin-repressed protein 2.7
1280542 81642171 Famesylated protein GMFPS 2.6
1280662 81642242 Gerrnin-Like Protein 3.2
1280717 81642278 MAGO NASHI Protein 2.4
1280815 81642398 DnaJ Protein homolog 2.2
T280928 81642348 Famesylated protein 2.]
1280964 81642747 erythrocyte-binding protein 2.4
1281033 81642665 Auxin-induced protein 2.]
1281380 81642911 NAM (no apical meristem)-1ike protein 2.9
Primary metabolism

CL80031 81677461 NADH Dehydrogenase (hYPOthetical) 2.6
CL80279 81677611 GDP-FUCOSE SYNTHETASE 2.0
CL80350 8167767 1 Enolase 2. l
CL80478 81677772 Adenosine kinase-like protein 4.4
CL80484 81677776 PHOSPHATIDYLINOSITOL 3-KINASE 2.2
CLSO633 81677893 GDP-D-Mannose-4,6-Dehydratase 3.4
CL80813 81677966 Alcohol Dehydrogenase 7 2.4
CL80834 81678193 S-Adenosyl-L-Methionine Synthetase 3.5
CLSO861 81678214 Inorganic pyrophosphatase-like protein 2.]
CL80913 81678259 Soluble Inorganic Pyrophosphatase 2.2
CLSO930 81678049 Acetoacyl-CoA-thiolase 2.6
CL80935 81678052 Ally] Alcohol Dehydrogenase-like protein 2.8
CL80956 81678070 Blue copper protein 2.9
CL80992 81678099 Sucrose Synthase 3.7
CL81064 81678145 Thioredoxin 4.3
CLS 1078 81678150 Glycine hydroxymethyltransferase 5.1
SW80965 81678963 dTDP-D-Glucose 4,6-Dehydratase 2.5
1280048 81642465 Cytochrome 85 Reductase 2.5
1280054 81642472 Xyloglucan endo-l,4-beta-D-glucanase 3.8
1280094 81642644 ATPase (Ht-transporting), ATP synthase 2.6
1280171 81642542 Alpha-Amylase/Subtisin Inhibitor 2.2

 

76

Table 2-3 (Cont’d)

 

 

CloneID GenBank Acc. Annotation 128/12F
Primary metabolism
1280193 81642554 Citrate Synthase 2.6
1280511 81642151 UDP Glucose 4-Epimerase 2.]
1280519 81642157 Cytochrome 85 DIF-F 2.8
1280533 81642165 Acyl-CoA-Binding Protein 2.]
1280657 81642237 Glyceraldehide 3-Phosphate Dehydrogenase 2.3
1280703 81642268 Phosphoenolpyruvate Carboxylkinase 3.2
1280919 81642344 BISPHOSPHATE NUCLEOTIDASE 2.7
128097] 81642754 DNA-dependent ATPase 2.1
1281265 81642829 Glutamine Synthetase PR-2 2.2
1281352 81642887 Phenylalanine Ammonia Lyase 3.3
Protein synthesis and processing
CL80093 81677512 Eukaryotic translation initiation factor eIF4E 3.7
CL80216 81677564 Polyubiquitin 2.7
CL80272 81677606 Legumain precursor (vicilin peptidohydrolase) 2.5
CL80625 81677885 Chaperonin 2.4
CLSO769 81677928 Polyubiquitin 2.3
CL80886 81678236 Ferredoxin [2Fe-28] II 2.4
CL80916 81678261 Ubiquitin extension protein 3.]
C1..81002 81678027 Proteasome beta, 208 subunit P8G1 2.3
CLS 1009 81678033 Papain-like Cystein Proteinase Isoforml 2.3
12F127 81643016 subtilisin-like proteinase (EC 3.4.2l.-) 2.7
T280133 81642514 eIF4F chain p28 2.4
1280364 81642072 Ubiquitin-like Protein 2.2
1280530 81642162 Translation Elongation Factor l-beta 2.6
1280739 81642284 Immunophilin 2.3
1281359 81642899 Formyltransferase purU 2.0
Secondary and hormone metabolism
CL80302 81677629 Leucoanthocyanidin Dioxygenase-like Protein 2.4
1280021 81642439 Chalcone-Flavone Isomerase 2.7
T280063 81642476 Isoflavone Reductase 3.0
1280108 81642499 Flavanone 3-hydroxylase 3.3
1280121 81642505 Monooxygenase 2.6
1280216 81642565 Monooxygenase 2.8
1280312 81642638 Flavonoid 3’,5’-Hydroxylase 3.]
1280424 81642100 Chalcone Synthase 2.8
1280608 81642208 Caffeoyl-CoA O-methyltransferase 3.0
1280672 81642248 Fumarylacetoacetase 2.3
1280721 81642279 Glutathione S-Transferase 2.7
1280751 81642292 Flavonoid 3’-hydroxylase 2.4
1280799 81642363 Glutathione S-Transferase 3.5
1280870 81642387 Dihydroflavonol 4-reductase 3.5
1280962 81642432 Chalcone Reductase 2.8
1281298 81642857 caffeic acid O-methyltransferase-likefiotein 3.9

 

77

Table 2-3 (Cont’d)

 

 

CloneID GenBank Acc. Annotation TZSII‘ZF
Signal transduction

CL80213 81677561 Casein kinase 1] alpha subunit 2.1
CL80214 81677562 CSF-l PROTEIN 2.1
CL80252 81677590 ADP-ribosylation factor 2.3
CL80300 81677627 Ethylene receptor homolog 2.0
CLSO377 81677691 Protein Phosphatase 2C-like 4.1
CL80503 81677793 14-3-3 protein homolog Vfa-1433b 2.3
CLSO899 81678246 Light-inducible protein ATLSl 2.6
128011] 81642495 Serine/I'hreonine Protein Kinase 2.7
1280163 81642537 GTP-binding Protein, ras-like 3.9
1280224 81642579 Casein Kinase 2.0
1280258 81642600 ADP-Ribosylation Factor 2.2
1280311 81642643 Nucleoside Diphosphate Kinase 1 4.0
1280353 81642065 Phosphoglycerate Kinase 1 3.9
1280548 81642177 GTP-Binding Protein 3.]
1280745 81642288 ADP-Ribosylation Factor 2.6
1281034 81642667 Signal peptidase 18 kDa subunit 2.3
Vesicular trafficking, protein sorting, secretion

CLSO775 81677933 Dynamin-like protein 4 (GTP-binding proteins) 2.4
1281114 81642720 GTPase Activating Protein-like (GAP) 2.5

 

78

were also among genes up-regulated in 128. Signal transduction-related genes were
differentially expressed in the transition zone depending on the sampling time. For
example, two genes for GTP-binding protein were up-regulated in 128, but five genes

for GTP-binding protein were up-regulated in 12F or down-regulated in T28.

Genes up-regulated in fall (72F)

Seventy-one percent of the 276 12F up-regulated genes are either of unknown
function or no-hit compared with 53% of the 128 up-regulated genes, indicating that
more novel genes are up-regulated in the fall than in the summer (Table 2-4). The gene
encoding DNA mismatch repair protein (accession no. 81679071) had 6.7-fold higher
expression in 12F compared with 128. Two genes encoding cysteine protease inhibitor
(81678216 and 81642350) were up-regulated in 12F, whereas another gene (81642228)
was highly expressed in T28. Eight genes categorized in ‘gene expression and RNA
metabolism’ were up-regulated in 12F compared with only six in 128. A gene encoding
reverse transcriptase, in particular, was highly up-regulated (4.9-fold) compared with
128. Considering the 12F sampling time (November 27), these data suggest that gene
expression activity in the transition zone continues during the dormancy period.
Furthermore, many of the primary metabolism-related genes were also up-regulated in
TZF. For example, expression of the gene for acetyl-CoA carboxylase carboxyl
transferase (81678729) was 16.5-fold higher in 12F than in 128. Other up-regulated
genes in the primary metabolism category included those encoding ATP synthase (7.2-
fold), cytochrome C oxidase (4.2-fold), epoxide hydrolase (5.9-fold),

ethanolaminephosphotransferase (4.8-fold), and starch phosphorylase (3.2-fold).

79

Table 2-4. List of up-regulated genes in transition zones harvested in fall (TZF)

 

 

Clone ID GenBank Acc. Annotation 12F/128
Cell wall structure and metabolism
SW81028 81678995 Beta-glucosidase precursor 2.6
Chromatin and DNA metabolism
CL80968 81678080 DNA-Binding Protein, hypothetical 2.0
SW81024 81678992 histone H1.41 5.4
SW81122 81679071 DNA Mismatch Repair Protein 6.7
Cytoskeleton
128011 81642920 cytoskeletal protein homolog 3.6
Defense
SW80593 81679110 N ADPH-Cytochrome P450 Reductase 2.1
CL80974 81678085 ABC Transporter-like protein 2.3
SW81009 81678982 Elicitor-responsive gene 2.4
T280931 81642350 Cystein Proteinase Inhibitor 2.4
CLSO863 81678216 Cystatin (cysteine proteinase inhibitor) 3.4
CL80240 81677579 wound-inducible basic protein 3.6
1280462 81642120 Oxygenase, pathogen-induced 3.8
Gene expression and RNA metabolism
SWS 1456 81679312 Zinc Finger Protein 1D] 2.3
SW80664 81679127 RNA Polymerase 2.5
SW80786 81678876 RNA Helicase ATP-Dependent 2.6
CLSlOll 81678035 RING-H2 Zinc finger protein 2.7
SW80307 81678530 RNA polymerase beta’ subunit 2.8
CL80258 81677595 heat shock transcription factor-like protein 3.5
1280302 81642630 Glycine-rich RNA Binding Protein 3.7
SW80180 81678403 Reverse Transcriptase 4.9
Membrane transport
CL80025 81677455 Lectin precursor, Bark Agglutinin I 2.5
SWSIOOS 81678978 Tonoplast Intrinsic Protein alpha 3.5
Miscellaneous
CL80517 81677803 Tripeptidyl-Peptidase I 2.0
12F094 81642983 nonstructural protein 1 2.]
1280351 81642064 ATAF2 Protein 2.2
SW81013 81678984 81 self-incompatibility 2.3
1280103 81642503 NOI Protein 2.3
12F013 81642922 POLYPROTEIN 2.6
SW80481 81678670 AINTEGUMENTA-LIKE PROTEIN 2.6
CL80369 81677684 Synaptobrevin homolog F10N7.40 2.7
128029] 81642616 RP42 2.9
SW80193 81678438 Altered response to gravity 3.5
1281138 81642726 Steroid-Binding (progesterone-binding) Protein 3.9
Protein Translocation Complex sec6l Gamma
SW80990 81679034 Chain 3.9
SWSO615 81678759 SUCROSE CARRIER SCAl. 4.7
CL80975 81678086 Mitochondrial processing peptidase 4.9
SW80064 81678369 Maturase-like Protein 7.2

 

80

Table 2-4 (Cont’d)

 

 

Clone 1D GenBank Acc. Annotation 12F/128
Primary metabolism

SWSO982 81678976 Lipid Transfer Protein 2.0
CL80089 81677508 dTDP-D-Glucose 4,6-Dehydratase 2.0
CLSO888 81678221 Methyltransferase 2.3
SWS 1286 81679214 NADH Dehydrogenase (ubiquinone) 2.3
1281 144 81642729 Outer membrane lipoprotein-like 2.3
CL81041 81678124 Granule-bound glycogen (starch) synthase 2.3
CL81062 81678144 Methyltransferase 2.4
1280305 81642632 Epoxide Hydrolase 2.5
SW81067 81679019 Aminotransferase-like protein 2.5
SWSO616 81679124 NADH-Plastoquinone Oxidoreductase 2.6
1280502 81642145 UDP Glucose 4-Epimerase 2.6
1280287 81642622 Lipid Transfer Protein 2.6
SW80564 81678731 ATPase, Cadmium-transporting 2.8
12F050 81642948 GDSL-motif lipase/acylhydrolase 2.8
CLSO760 81677920 Thioredoxin-like proteins 2.9
12F150 81643032 cytochrome b5 - common tobacco 3.0
1281266 81642830 Starch Phosphorylase 3.2
1280958 81642329 Cytochrome c oxidase subunit 6b-l 4.2
SWSO401 81678642 Ethanolaminephosphotransferase 4.8
1280266 81642604 Epoxide Hydrolase 5.9
SWSl451 81679307 ATP Synthase beta chain 7.2
SW80562 81678729 Acetyl-CoA Carboxylase Carboxyl Transferase 16.5
Protein synthesis and processing

SWSO868 81679188 Ubiquitin-Conjugating Enzyme 2.3
1280297 81642624 Arninoacylase l 2.4
SWSO90] 81678907 Histidine-tRNA ligase 2.4
CL80417 81677720 Ribosomal Protein 812, Chloroplast 308 3.1
SW80983 81679028 Heat Shock Protein 3.7
CLSO264 81677601 ClpP-like Protease 3.7
SW80197 81678442 ClpP-like Protease 4.4
1281286 81642848 Heat Shock Protein 70 4.4
Secondary and hormone metabolism

CLSO373 81677687 Monooxygenase 2.6
CL80631 81677891 Auxin-binding protein T85 precursor 2.0
1281040 81642673 Protein Phosphatase 2.1
CL80984 81678094 GTP-binding protein sral 2.2
CL80433 81677735 Calcium-dependent protein kinase 2.2
SW81034 81679000 Light-inducible protein ATLSl 2.2
CL80941 81678057 GTP-binding protein GTP13 2.4
SWSO952 81678950 Protein Kinase PVPK-l 2.8
CL80441 81677741 small GTP-binding protein 3.0
SWSO491 81678678 WD-Repeat Protein-like 3.7
1280467 81642124 GTP-binding Protein 4.0
1280026 81642444 GTP-Binding Protein SUP] 4.7
CL80869 81678222 Calmodulin-related protein 6.2
SWSO339 81678559 Receptor Kinase 9.4

 

81

Many protein synthesis and process-related genes were also up-regulated in TZF,
including genes encoding two ClpP-like proteases (4.7- and 3.7-fold), two heat shock
proteins (3.7- and 4.4-fold), and ribosomal protein 812 (3.1-fold). Most importantly, a
similar number (13) of signal transduction genes were up-regulated in 12F compared
with 16 genes in 128. Expression of a gene encoding for receptor kinase (81678559) was
increased 9.4-fold in 12F compared with 128. Other up-regulated signal transduction
genes include those encoding for calmodulin-related protein (6.2-fold), four GTP-binding
proteins (2.2- to 4.7-fold) and WD-repeat protein-like (3.7-fold). However, only one
gene, encoding for monooxygenase (81677687), in the ‘secondary and hormone
metabolism’ category was up-regulated in TZF. It had been thought that biosynthesis of
heartwood extractives (i.e., mostly secondary metabolites) was most active in fall, rather
than in summer when a total of 16 genes in the category were up-regulated. Overall, the
12F gene expression data strongly suggest that many biological activities (e.g., gene

expression, protein synthesis and signal transduction) occur after trees enter dormancy.

Confirmation of microarray results

To further support my microarray results, and to confirm the seasonal expression
of the secondary metabolism related genes, antisense Northern blot was carried out on 10
selected genes (the genes encoding PAL, C4H, CH8, CHR, CHI, F3H, F3_’H, F3’,5’H,
DFR and IFR) that are involved in the phenylpropanoid metabolic pathway. When the
expression of these genes was compared at three seasonal points (spring, sampled on
March 5; summer, sampled on July 5; fall, sampled on November 27), the expression of

all selected genes was highest in summer (Figure 2-2), which corroborates the microarray

82

results. In addition, the flavonoid metabolism-related genes were differentially expressed
according to the sampling season with highest expression levels in the summer samples
and similar expression levels in the spring and fall samples. These results suggest that the
flavonoid pathway is activated and secondary metabolites are produced during the

summer.

83

Phenylalanine
(PAL '-
Cinnamate
C4H E l
p- oumarate ......
4CL """"" ; Lignins
p-Coumaroyl-CoA """
lCHS 5'
CHR  
4, 2’, 4’, 6’- -Tetrahydroxycha|cone
4, 2’ ,4’-Trihydroxychalcone
ICHI r I
Naringenin ............. .. Flavones
Liquiritigenin

F3H
F3’H

 

 

 

 

 

 

 

 

 

 

 

 

F3’ 5’H - .
Dihydroflavonol -------- > Flavonols
:DFR '

v
Anthocyanins

 

 

Figure 2-2. Metabolic pathway leading to phenylpropanoids

84

Discussion

Transverse trunk sections of mature trees usually contain two zones: a pale-
colored outer zone, the sapwood, and a dark-colored inner zone, the heartwood. The pale-
colored sapwood contains storage material (starch and lipids), whereas the dark color of
the heartwood reflects the presence of various organic substances, the so-called
extractives. The nature of the extractives determines the specific heartwood color and
durability. In black locust, which is known for its resistance to wood decay (Scheffer et
al., 1944), most of the extractives are phenolics. Although the biochemistry of heartwood
has been studied, gene expression studies of heartwood formation are limited (Magel,
2000; Beritognolo et al., 2002; Yang et al., 2003). My work provides the first
comprehensive analysis of the global gene expression changes occurring in the transition
zone of a mature tree from active growth (July 5) to the start of dormancy (November
27).

During the transition from sapwood to heartwood, the ray parenchyma cells in the
sapwood—heartwood transition zone undergo a form of programmed cell death (Magel,
2000). The reserve materials in these cells are converted to secondary metabolites, which
may serve as defense chemicals. This complex and slow event of heartwood formation is
presumably orchestrated by enzymes involved in the breakdown of storage materials,
secondary metabolite biosynthesis, and senescence. The genes encoding such enzymes
are thought to be induced in the transition zone at the time of heartwood formation. In the
temperate zones, heartwood formation occurs in late summer to late fall and at the
beginning of dormancy provided temperatures are above 5 °C (Hillis, 1987). Earlier

studies have indicated that heartwood formation occurs at the time of cambial dormancy

85

in pine (Shain and Mackay, 1973), walnut and cherry (Nelson 1978). Studies of the
cytology and coloration of the extractives suggest that heartwood formation commences
in midsummer and continues in the fall and winter seasons in sugi (Nobuchi et al. 198 7a)
and black locust (Nobuchi et al. 1987b). Based on my survey of global gene expression
changes from active growth (July 5) to the start of dormancy (November 27), I found that
more than 50% of the secondary and hormone metabolism-related genes on the
microarray were expressed at a twofold or greater level in July (128) compared with
November (TZF). In addition, 31 genes in the ‘primary metabolism’ category were up-
regulated in TZS, but only 22 genes were up-regulated in TZF. The expression levels of
53% (29 out of 55) of the signal transduction-related genes were differentially expressed
in the transition zone, suggesting that the cells located in the innermost part of the trunk
are affected by environmental conditions, such as seasonal change.

Higuchi (1997) proposed that the physiological processes leading to the transition
from sapwood to heartwood occur in three main steps: (1) alterations in the metabolism
of living parenchyma cells activated by hormonal (e.g., ethylene) and physical factors,
such as reduction of water content; (2) increased metabolic activity and synthesis of
heartwood extractives; and (3) cellular death, liberation of extractives stored in the
vacuole, oxidation and polymerization of extractives by chemical or enzymatic processes.
Among the 128 up-regulated genes, there are many representative genes that support this
proposed scheme of heartwood formation. For example, three SKPI (suppressor of
kinetochore protein) genes, a component of ubiquitin 1i gase complex, were up-regulated
in 128. The function of SKPl protein is unclear, but plant SKPl proteins involved in

proteolysis play a role in cell differentiation, programmed cell death and defense

86

mechanisms. The SKPI gene is related to the regulation of proteolysis in different stages
of the cell cycle (Bai et al., 1996; Deshaies, 1999). The mutant of Arabidopsis SKPl,
ask] -1, shows reduced auxin responses, abnormal flowers and reduced lateral roots (Gray
et al., 1999; Zhao et al., 2001). In tobacco, the NbSKPI gene plays an important role in
the N gene-mediated resistance response to tobacco mosaic virus (Liu et al., 2002). The
high level of expression of SKPI and SKPI homolog in 128 suggests that the transition
zone undergoes programmed cell death during summer. However, because most of the
SKPl proteins studied are involved in post-translational regulation of gene expression, a
proteomics-based investigation should provide more insights into the role of black locust
SKPl protein during heartwood formation. The pathogenesis-related class 10 (PR-10),
which is a protein related to defense, was also highly expressed in 128. The PR-IO genes
are induced, not only by pathogen infections, but also by a large variety of environmental
stresses (Dubos and Plomion, 2001; McGee et al., 2001; Borsics and Lados, 2002).
Metallothioneins (MTs) are ubiquitous low molecular weight cysteine-rich proteins and
polypeptides of extremely high metal and sulfur content, and are required for heavy metal
tolerance in animals and fungi. In plants, the transcripts of metallothionein genes
accumulate during leaf senescence as well as heavy metal detoxification (Buchanan-
Wollaston, 1997). There are three genes that are induced under dehydration conditions.
One of them is homologous to ERDIS, a dehydration-induced gene in Arabidopsis
(Kiyosue et al., 1994). High peroxidase activity has been observed in the transition zone
of various wood species: Juglans (Nelson, 1977), Picea (Estebauer et al., 1978), Quercus
(Ebermann and Stich, 1982 and 1985), Paulownia (Ota et al., 1991) and Larix (Korori et

al., 1998). Dehon et a]. (2001 and 2002) noted that peroxidases in walnut trunk were

87

involved in heartwood formation and its brown coloration. The up-regulation of an
ethylene receptor homologous gene (accession no. 81677627) suggests that ethylene
signaling is involved in heartwood formation. Nelson et a]. (1981) reported that ethylene
production is related to heartwood formation in black walnut. In addition, ethylene is an
important factor controlling heartwood formation in pine (Nilsson et al., 2002). The high
level of expression of these genes related to environmental stress, defense or senescence
indicates that, during the summer, the transition zone is subjected to various stressful
conditions such as dehydration and senescence, and that the ray parenchyma cells in the
transition zone undergo heartwood formation through the expression of these genes.
Heartwood formation requires degradation of carbohydrate storage materials at
the transition zone to generate energy and substrates necessary for biosynthesis of
heartwood extractives. Biochemical studies have shown that storage material is
consumed during the sapwood—heartwood transition (Shah etal., 1981; Nobuchi et al.,
1982, 1987a, 1987b; Nair, 1988; Kumar and Datta, 1989). Carbohydrates stored at the
transition zone may be used by the cells to produce phenolics (Magel et al., 1991, 1994).
Magel (2000) proposed that the sucrose synthase pathway is associated with the
biosynthesis of phenolic extractives during heartwood formation and, therefore, the gene
coding for sucrose synthase can be considered a marker gene of heartwood formation.
Because the bulk of phenolic compound synthesis is dependent on imported carbon, the
sugar transport system should be involved in heartwood formation. The genes encoding
carbohydrate transporter proteins and sucrose degradation enzymes were highly
expressed during the period of heartwood formation (Magel, 2000). I found that these

genes were up-regulated during summer rather than in late fall, suggesting that

88

carbohydrate transport and break down occur during summer in the trunk wood of black
locust.

Biosynthesis of secondary metabolites is a major process that leads to
accumulation of heartwood extractives. Cytochrome 85 DIF-F gene was highly
expressed in the 128 sample. This gene product stimulates the activity of specific Cyt
P4503 such as a flavonoid 3’,5’-hydroxylase (F3,5’H) in petunia (de Vetten et al., 1999).
Many secondary metabolites are toxic to the cells that produce them and are, therefore,
sequestered in the vacuole once they are made. Programmed cell death of the ray
parenchyma cells at the transition zone may involve the release of secondary metabolites
stored in cell vacuoles. Glutathione S-transferase (GST) has been shown to play a role in
vacuolar sequestration of flavonoids (Mueller et al., 2000), and G818 may also serve as
binding proteins for anthocyanin to sequestrate flavonoids into vacuoles. Two of the three
genes encoding GST on the microarrays were up-regulated in 128.

In black locust, the activities of two key enzymes of the phenylpropanoid
biosynthetic pathway (PAL and CH8) increase in the transition zone compared with the
outer sapwood (Magel et al., 1991; Magel and Hubner, 1997). These genes are more
highly expressed in the transition zone than in the bark/cambial zone and sapwood region
(Yang et al., 2003). Beritognolo et a]. (2002) studied seasonal changes in the expression
of several flavonoid biosynthetic genes during heartwood formation in Juglans nigra and
found that flavonol accumulation at the sapwood-heartwood transition zone was
correlated with transcript levels of key flavonoid biosynthetic genes such as those
encoding CH8, DFR, and F3H. I compared the expression of these genes among three

seasons (spring, summer and fall). Expression of the secondary metabolism-related genes

89

was highest in summer, suggesting that the biosynthesis of phenylpropanoids and other
heartwood extractives in the ray parenchyma cells at the transition zone is most active
during summer. Biosynthesis of flavonoid and anthocyanin has been extensively studied
(Forkmann and Martens, 2001), but the biosynthetic pathways for many natural products
in heart wood remain to be elucidated. For example, even though more chemical analyses
have been carried out on the heartwood flavonoids of Leguminosae, to which black locust
belongs, than on heartwood flavonoids of any other plant family (Rowe, 1989), the
specific flavonoid biosynthetic pathway of this family has not been established and no
explanation has been found for the absence of 5-hydroxylation in this family. Three
major flavonoids in black locust heartwood are dihydrorobinetin (3,7,3’,4’,5’-
pentahydroxyflavanonol), robinetin (3,7,3’,4’,5’-pentahydroxyflavone) and
leucorobinetidin (tetrahydroxy-flavan-3,4-diol) (J .A. Duke, http://www.ars-

grin. gov/dukel).

When studying heartwood formation, it is crucial to select the periods
characterized by high activities in extractive accumulation and heartwood expansion. In
R. pseudoacacia, these processes appear to be seasonally regulated and linked to tree
phenology. Accumulation of extractives and radial expansion of heartwood follow the
period of cambial activity and are more marked in late summer and autumn than in
spring. A year-round sampling of black locust trees has just been completed to monitor
the temporal dynamics of heartwood formation. Analyses of extractives and transcription
profiles are currently being carried out on these samples.

Microarray analysis was performed to examine changes in global gene expression

patterns in the sapwood—heartwood transition zone of mature black locust trees between

90

the termination of active growth and the onset of dormancy. I identified candidate genes
that are likely to be involved in heartwood formation, established expression patterns of
many unknown genes, and provided insights into seasonal changes in the molecular
events leading to heartwood expansion and extractive formation. Although these findings
need to be substantiated at the protein or phenotypic level before biologically meaningful
conclusions can be drawn, they indicate the potential for bio-technological improvement
of forest trees for production of value-added forest products (e.g., decay resistance, novel

pharmaceuticals) and disease and insect resistance.

9]

Literature Cited

Alizadeh, A., M.8. Eisen, R.E. Davis, C. Ma, 1.8. Lossos, A. Rosenwald, J.C. Boldrick,
H. Sabet, T. Tran and X. Yu. 2000. Distinct types of diffuse large 8-cel] lymphoma
identified by gene expression profiling. Nature 403: 503-511.

Bai, C., P. Sen, K. Hofmann, L. Ma, M. Goebl, J.W. Harper and 8.]. Ellwdge.1996.
SKPl connects cell cycle regulators to the ubiquitin proteolysis machinery through a
novel motif, the F-box. Cell 86: 263-274.

Baqui, S. and J. Shah. 1985. Histoenzymatic studies in wood of Acacia auriculiformis
Cunn. during heartwood formation. Holzforshchung 39: 311-320.

Baqui, S., J. Shah, R. Pandalai and I. Kothari. 1979. Histochemica] changes during
transition from sapwood to heartwood in Melia azedarach. Indian J. Exp. Biol. 17:
1032-1037.

Beritognolo, 1., E. Magel, A. Abdel-Latif, J.P. Charpentier, C. Jay-Allemand and C.
Breton. 2002. Expression of genes encoding chalcone synthase, flavanone 3-
hydroxylase, and dihydroflavonol 4-reductase correlates with flavanol accumulation
during heartwood formation in Juglans nigra L. Tree Physiol. 22: 291-300.

Borsics, T. and M. Lados. 2002. Dodder infection induces the expression of a
pathogenesis-related gene of the family PR-lO in alfalfa. J. Exp. Bot. 53: 1831-1842.

Buchanan-Wollaston, V. 1997. The molecular biology of leaf senescence. J. Exp. Bot.
48: 181-199.

Burtin, P., C. Jay-Allemand, J .P. Charpentier and G. Janin. 1998. Natural wood colouring
process in Juglan sp. (J. nigra, J. regia and hybrid J. nigra x J. regia) depends on
native phenolic compounds accumulated in the transition zone between sapwood and
heartwood. Trees 12: 258-264.

de Vetten, N., J. ter Horst, H.-P. van Schaik, A. de Boer, J. Mo] and R. Koes. 1999. A
cytochrome b5 is required for full activity of flavonoid 3',5'-hydroxylase, a

cytochrome P450 involved in the formation of blue flowers. Proc. Natl. Acad. Sci.
USA 96: 778-783.

Dehon, L., L. Mondolot, M. Durand, C. Chalies, C. Andary and JJ. Macheix. 2001.
Differential compartmentation of o-diphenols and peroxidase activity in the internal
sapwood of walnut tree (Juglans nigra L.). Plant Physiol. Biochem. 39: 339—344.

Dehon, L., J. Macheix and M. Durand. 2002. Involvement of peroxidases in the
formation of the brown coloration of heartwood in Juglans nigra. J. Exp. Bot. 53:
303-311. '

92

Deshaies, R.J. 1999. SCF and cullin/ring H2-based ubiquitin ligases. Annu. Rev. Cell
Biol. 15: 435—467

Dietrichs, H. 1964. Das Verhalten von Kohlenhydraten bei der. Holzforschung 18: 14-24.

Dubos, C. and C. Plomion. 2001. Drought differentially affects expression of a PR-10
protein, in needles of maritime pine (Pinus pinaster Ait.) seedlings. J. Exp. Bot. 52:
1 143-1154.

Ebermann, R. and K. Stich. 1982. Peroxidase and amylase isoenzymes in the sapwood
and heartwood of trees. Phytochemistry 21: 2401—2402.

Ebermann, R. and K. Stich. 1985. Distribution and seasonal variations of wood
peroxydase activity in oak (Quercus robur). Wood Fiber Sci. 17: 391—396.

Eisen, M8. and PO. Brown. 1999. DNA arrays for analysis of gene expression. Methods
Enzymol. 303: 179-205.

Estebauer, H., D. Grill and M. Zotter. 197 8. Peroxidase in needles of Picea abies (L.)
Karst. Biochemica Physiologic et Pflanzenphysiologie 172: 155-159.

Fernandes, J ., V. Brendel, X. Gal, 8. La], V. Chandler, R. Elumalai, D. Galbraith, E.
Pierson and V. Walbot. 2002. Comparison of RNA expression profiles based on

maize expressed sequence tag frequency analysis and micro-array hybridization. Plant
Physiol. 128: 896-910.

Forkmann, G. and S. Martens. 2001. Metabolic engineering and applications of
flavonoids. Current Opinion Biotechnol. 12: 155-160.

Girke, T., J. Todd, 8. Ruuska, J. White, C. Benning and J. Ohlrogge. 2000. Microarray
analysis of developing Arabidopsis Seeds. Plant Physiol. 124: 1570-1581.

Golub, T.R., D.K. Slonim, P. T amayo, C. Huard, M. Gaasenbeek, J.P. Mesirov, H.
Coller, M. Loh, J .R. Downing, and MA. Caligiuri. 1999. Molecular classification of

cancer: Class discovery and class prediction by gene expression monitoring. Science
286: 531-537.

Gray, W.M., J.C. del Pozo, L. Walker, L. Hobbie, E. Risseeuw, T. Banks, W.L. Crosby,
M. Yang, H. Ma and M. Estelle. 1999. Identification of an SCF ubiquitin-ligase

complex required for auxin response in Arabidopsis thaliana. Genes Dev. 13: 1678-
1691.

Hauch, 8. and E. Magel. 1998. Extractable activities and protein content of sucrose-
phosphate synthase, sucrose synthase and neutral invertase in trunk tissues of Robinia
pseudoacacia L. are related to cambial wood production and heartwood formation.
Planta 207: 266-274.

Higuchi, T. 1997. Biochemistry and Molecular Biology. Springer Verlag: Berlin.

93

Higuchi, T.,Y. Onda and Y. Fujimoto. 1969. Biochemical aspects of heartwood formation
with special reference to the site of biogenesis of heartwood compounds. Wood Res.
48: 15-30.

Hillinger, C., W. H611 and H. Ziegler. 1996a. Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. 1. Radial
distribution of lipid classes. Trees 10: 366-375.

Hillinger, C., W. H611 and H. Ziegler. 1996b. Lipids and lipolytic enzymes in the
trunkwood of Robinia pseudoacacia L. during heartwood formation. H. Radial
distribution of lipases and phospholipases. Trees 10: 376-381.

Hillis, W.E. 1987. Heartwood and tree exudates. Berlin, New York, Springer-Verlag.

Hillis, W.E. 1996. Formation of robinetin crystals in vessels of Intsia species. IAWA J.
17: 405-419.

Hillis, W. E. and M. Hasegawa. 1963. The formation of polyphenols in trees. 1.
Administration of l4-C-glucose and subsequent distribution of radioactivity.
Phytochem. 2: 195-199.

Hillis, W. E. and Y. Yazaki. 1973. Wood polyphenols of Eucalyptus polyanthemos.
Phytochem. 12: 2969-2977.

H611, W. and K. Lendzian. 1973. Respiration in the sapwood and heartwood of Robinia
pseudoacacia. Can J. Bot. 52: 727-734.

Kiyosue, T., K. Yamaguchi-Shinozaki and K. Shinozaki. 1994. ERDIS, a cDNA for a
dehydration-induced gene from Arabidopsis thaliana. Plant Physiol. 106: 1707.

Korori S.A.A., H. Pichomer and R. Ebermann. 1998. Seasonal alterations of peroxidase
and catalase isoenzymes in branches and seeds on three different species of Larix.
Plant Peroxidase Newsletter 3: 12—17.

Kummer, A. and S. Datta. 1989. Histochemica] and histoenzymological changes during
heartwood formation in a timber tree Shorea robusta. Proc. Indian Sci. (Plant Sci.)
99: 21-27.

Liu, Y., M. Schiff, G. Serino, X. Deng and S. Dinesh-Kumar. 2002. Role of SCF
ubiquitin-ligase and the COP9 signalosome in the N gene-mediated resistance
response to tobacco mosaic virus. Plant Cell 14: 1483-1496.

Magel, E. 2000. Biochemistry and physiology of heartwood formation. In Molecular and
Cell Biology of Wood Formation. Eds. R. Savidge, J. 8amett and R. Napier. 8108
Scientific Publishers, Oxford, pp363-376.

Magel, E. and 8. Hilbner. 1997. Distribution of phenylalanine ammonia lyase and
chalcone synthase within trunks of Robinia pseudoacacia L. Bot. Acta. 110: 314-322.

94

Magel, E., A. Drouet, A. Claudot and H. Ziegler. 1991. Formation of heartwood
substances in the stemwood of Robinia pseudoacacia L. 1. Distribution of

phenylalanine ammonium-lyase and chalcone synthase across the trunk. Trees 5: 203-
207 .

Magel, E., C. Jay-Allemand and H. Ziegler. 1994. Formation of heartwood substances in
the stemwood of Robinia pseudoacacia L.: 11. Distribution of nonstructural
carbohydrates and wood extractives across the trunk. Trees 8: 165-171.

McGee, J., J. Hamer and T. Hodges. 2001. Characterization of a PR-10 pathogenesis-

related gene family induced in rice during infection with Magnaporthe grisea. Mol.
Plant Microbe Interact. 14: 877-886.

Mueller, L., OD. Goodman, R.A. Silady and V. Walbot. 2000. AN9, a petunia
glutathione S-transferase required for anthocyanin sequestration is a flavonoid-
binding protein. Plant Physiol. 123: 1561-1570.

Nair, M. 1988. Wood anatomy and heartwood formation in Neem (Azadirachata indica
A. Juss.). Bot. J. Linn. Soc. 97: 79-90.

Nelson, N.D., W.J. Rietveld and J .G. Isebrands. 1981. Xylem ethylene production in five
black walnut families in the early stages of heartwood formation. For. Sci. 27: 537-
543.

Nelson, ND. 1977. Xylem ethylene, phenol-oxidizing enzymes, and nitrogen and
heartwood formation in walnut and cherry. Can. J. Bot. 56: 626—634.

Nelson, ND. 1978. Xylem ethylene, phenol-oxidising enzymes and nitrogen and
heartwood formation in walnut and cherry. Can. J. Bot. 56: 626-634.

Nelson, N.D., W.J. Rietveld and J. Isebrands. 1981. Xylem ethylene production in five
black walnut families in the early stages of heartwood formation. For. Sci. 27: 537-
543.

Nilsson, M., S. Wikman and L. Eklund. 2002. Induction of discolored wood in Scots pine
(Pinus sylvestris) Tree Physiol. 22: 331-338.

Nobuchi, T., K. Kuroda, R. Iwata and H. Harada. 1982. Cytological study of the seasonal
features of Sugi (Cryptomeria japonica D. Don). Mokuzai Gakkaishi 28: 669-976.

Nobuchi, T., K. Takai and H. Harada. 1987a. Distribution of heartwood phenols in the
trunk of Sugi (Cryptomeria jaonica D. Don) and partial characterization of heartwood
formation. Mokuzai Gakkaishi 33: 88-96.

Nobuchi, T., N. Tokuchi and H. Harada. 1987b. Variability of heartwood formation and
cytological features in broadleaved trees. Mokuzai Gakkaishi 33: 596-604.

95

Perou, C.M., T. Sorlie, M.8. Eisen, M. Van de Rijn, 8. Jeffrey, C.A. Rees, J.R. Pollock,
D.T. Ross, H. Johnsen and LA. Akslen. 2000. Molecular portraits of human breast
tumors. Nature 406: 747-752.

Pinkel, D., R. Segraves, S. Sudar, S. Clark, 1. Poole, D. Kowbel, C. Collins, W.-L. Kuo,
C. Chen and Z. Zhai. 1998. High resolution analysis of DNA copy number variation
using comparative genomic hybridization to microarrays. Nat. Genet. 20: 207-211.

Reymond, P., H. Weber, M. Damond and 8. Farmer. 2000. Differential gene expression

in response to mechanical wounding and insect feeding in Arabidopsis. Plant Cell 12:
707-720.

Rowe, J. 1989. Natural products of wood plants 1. Berlin, New York, Springer-Verlag.
Schaffer, R., J. Landgraf, M. Accerbi, V. Simon, M. Larson and E. Wisman, E. 2001.
Microarray analysis of diurnal and circadian-regulated genes in Arabidopsis. Plant
Cell 13: 113-123.

Scheffer, T., H. Lachmund and H. Hopp. 1944. Relation between hot-water extractives
and decay resistance of black locust wood. J. Agri. Res. 68: 415-426.

Schena, M., D. Shalon, R.W. Davis and PO. Brown. 1995. Quantitative monitoring of

gene expression patterns with a complementary DNA microarray. Science 270: 467-
470.

Shah, J., S. Baqui, R. Pandalai and K. Patel. 1981. Histochemical changes in Acacia
nilotica during transition from sapwood to heartwood. IAWA Bulletin 2: 31-36.

Shain, J. and J. Mackay. 1973. Seasonal fluctuations in respiration of aging xylem in
relation to heartwood formation in Pinus radiata. Can J. Bot. 51: 737-741.

Swidzinski, J.A., L.J. Sweetlove and CJ. Leaver. 2002. A custom microarray analysis of
gene expression during programmed cell death in Arabidopsis thaliana. Plant J. 30:
431-446.

Wang, 8., L.D. Miller, G.A. Ohnmacht, ET. Lin and RM. Marincola. 2000. High-
fidelity mRNA amplification for gene profiling. Nat. Biotech. 18: 457-459.

Wardrop, A. and J. Cronshaw. 1962. Formation of phenolic substances in ray
paranchyma of angiosperms. Nature 193: 90-92.

Yang, J ., S. Park, D.P. Kamdem, D.E. Keathley, E. Retzel, C. Paule, V. Kapur and K.-H.
Han. 2003. Novel insight into trunk-wood gene expression profiles in a hardwood tree
species, Robinia pseudoacacia. Plant Mol. Biol. 52: 935-956.

Zhao, D., Q. Yu, M. Chen and H. Ma. 2001. The ASK] gene regulates 8 function gene
expression in cooperation with UFO and LEAF Y in Arabidopsis. Development 128:
2735-2746.

96

CHAPTER 3

Functional characterization of allantoinase genes from Arabidopsis and a

nonureide-type legume black locust

ABSTRACT

The availability of nitrogen is a limiting factor for plant growth in most soils. Allantoin
and its degradation derivatives are a group of soil heterocyclic nitrogen compounds that
play an essential role in the assimilation, metabolism, transport, and storage of nitrogen in
plants. Allantoinase is a key enzyme for biogenesis and degradation of these ureide
compounds. Here, I describe the isolation of two functional allantoinase genes, AtALN
(Arabidopsis allantoinase) and RpALN (Robinia pseudoacacia allantoinase), from
Arabidopsis and black locust (Robinia pseudoacacia). The proteins encoded by those
genes were predicted to have a signal peptide for the secretory pathway, which is
consistent with earlier biochemical work that localized allantoinase activity to
microbodies and endoplasmic reticulum (Hanks et al., 1981). Their functions were
confirmed by genetic complementation of a yeast mutant (dal 1) deficient in allantoin
hydrolysis. The absence of nitrogen in the medium increased the expression of the genes.
In Arabidopsis, the addition of allantoin to the medium as a sole source of nitrogen
resulted in the up-regulation of the AtALN gene. The black locust gene (RpALN) was
differentially regulated in cotyledons, axis, and hypocotyls during seed germination and
seedling growth, but was not expressed in root tissues. In the trunk wood of a mature
black locust tree, the RpALN gene was highly expressed in the bark/cambial region, but

had no detectable expression in the sapwood or sapwood-heartwood transition zone. In

97

addition, the gene expression in the bark/cambial region was up-regulated in spring and

fall when compared with summer, suggesting its involvement in nitrogen mobilization.

Introduction

Nitrogen is a key component of plant metabolism and its availability is often a
limiting factor for plant growth in most soils. Heterocyclic nitrogen compounds (i.e.
purines, pyrimidines, and their degradation products) represent major sources of soil
organic nitrogen (Schulten and Schnitzer, 1998). Among these, allantoin (ALN) and its
degradation product allantoic acid (ALA) are nitrogen-rich organic compounds with a
C:N ratio of 1:1, and they play an essential role in the assimilation, metabolism, transport,
and storage of nitrogen in plants (Schubert and Boland, 1990). In addition, they serve as
effective carriers of the biologically fixed nitrogen in ureide-type legumes, and provide
nitrogen storage with minimal expense of reduced carbon. For example, these compounds
constitute 70% to 80% (w/v) of the organic nitrogen in the xylem sap of nodulated
soybean (Glycine max; McClure and Israel, 1979). In a tree legume (black locust,
Robinia pseudoacacia), about 3% (w/w) of xylem nitrogen was ureide (Atkins et al.,
1991). Furthermore, ALN and ALA are central metabolites in the seasonal nitrogen cycle
of perennial species such as maple (Acer spp.) and comfrey (Symphytum ofiicinale; for
review, see Schubert and Boland, 1990), indicating their key role in nitrogen storage and
cycling. In legumes, the organic nitrogen fixed by soil bacteria (e.g. Rhizobium sp.) can
be transported to the other parts of the plant as amides (Asn and Gln; amide-type legume)
or as ureides (ALN and ALA; ureide-type legume).

The major route for ALN biogenesis is the purine oxidation pathway (also called

98

ureide pathway and ALN degradation pathway; Figure 3-1). The first step in this pathway
is the degradation of nucleic acid purine moieties (adenine and guanine) to uric acid.
After two consecutive oxidation reactions by urate oxidase and hydroxy-isourate
hydrolase, the uric acid is converted to ALN (Raychaudhuri and Tipton, 2002).
Allantoinase (ALN amidohydrolase, EC 3.5.2.5) catalyzes the hydrolysis of ALN to form
allantoic acid, which is a key reaction for biogenesis and the degradation of ureides
(Vogels et al., 1966; Noguchi et al., 1986). The resulting ALA is then further metabolized
to ammonium, urea, and glyoxylate (Munoz et al., 2001). Allantoate degradation can be
catalyzed by allantoate amidohydrolase (EC 3.5.3.9) or allantoate amidinohydrolase (EC
3.5.3.4) to produce ureidoglycolate, which is metabolized to glyoxylate by
ureidoglycolate urea-lyase (EC 4.3.2.3) or ureidoglycolate amidohydrolase (EC 3.5.3.19).
Allantoate amidohydrolases and ureidoglycolate amidohydrolases generate ammonium,
whereas allantoate amidinohydrolase and ureidoglycolate urea-lyase release urea. This
pathway serves different roles and is evolutionarily distinct in plants, animals, and
microorganisms. It is primarily used for salvage or excretion of nitrogen from purines in
animals (Campbell and Bishop, 1970; Stryer, 1988). On the other hand, microorganisms
use it to extract nitrogen from a variety of sources in the external environment (Cooper,
1980). Although its main function in animals is nitrogen excretion, many legurninous
plant species use this pathway to store and recycle nitrogen. The pathway for ureide
degradation in nonlegume plants has not been well documented. Recently, Desimone et
al. (2002) demonstrated that Arabidopsis, a nonlegume model species, could take up and
use ALN as a sole nitrogen source. This requires enzymatic degradation of the mobilized

ALN inside the cells. The first step of the catabolism is catalyzed by allantoinase.

99

Xanthine

Xanthine dehyd rogenase
or Xanthine oxgenase

 

v
Uric acid

Uricase
Allantoin
Allantoinase
Allantoate
Allantoate amidOhydrolase
Ureidoglycolate

l Ureidoglycolase
Glyoxylate

Figure 3-1. Catabolism of ureide in plants

100

Allantoinase is present in a wide variety of organisms, including animals, bacteria, fungi,
and plants (for review, see Schubert and Boland, 1990). It is important for ureide
biogenesis and degradation. Despite the long history of allantoinase study in plants
(mainly legumes), my understanding of the physiological significance of the enzyme in
plant growth and development is still limited. Molecular analysis of allantoinase genes is
the first step in my approach to elucidate the relevance of ALN for plant nutrition. The
genes encoding allantoinase have been cloned from yeast (Saccharomyces cerevisiae;
Buckholz and Cooper, 1991) and bullfrog (Rana catesbeiana; Hayashi et al., 1994), but
no plant allantoinase genes have been characterized yet. In this report, I isolated and
characterized functional plant allantoinase genes from a tree legume (black locust,
Robinia pseudoacacia) and Arabidopsis (nonlegume model species). The black locust
allantoinase gene was discovered as a part of my trunk wood expressed sequence tag
(EST) project, and the Arabidopsis gene was identified by a homologous search with the
black locust gene. Their functions were confirmed by genetic complementation of a yeast
mutant defective in allantoinase function. Differential expression of the genes was
investigated with regard to the stage of development, seasonal variation, presence or

absence of nitrogen, and different tissues.

Materials and Methods

Plant Materials and Culture Conditions
Arabidopsis (Colombia ecotypes) and black locust (Robinia pseudoacacia)
seedlings were grown in petri dishes under light conditions (16 h of light per day)at 25°C.

For black locust seed germination and seedling growth, seeds were heated in boiling

lOl

water for 1 min, soaked in sterile water for 1 h, and transferred into glass jars containing
0.7% (w/v) agar (without nitrogen) or 0.7% (w/v) agar plus 10 mm ammonium nitrate
(with nitrogen). For Arabidopsis plants, seeds were subjected to cold treatments for 3 din
sterile water, and then were transferred onto Murashige and Skoog (1962) medium with
2% (w/v) Glc and 10 mm ammonium nitrate or 10 mm ALN (ALN medium) as a

nitrogen source. All procedures were performed under sterile conditions.

Cloning, Multiple Alignment, and Phylogenetic Analysis

I used the cDNA library prepared from the 4- to 6-d-old seedlings of black locust
to obtain the full-length cDNA clone of black locust allantoinase. About 100,000 plaques
were screened with the partial cDNA clone (GenBank accession no. 81643000) and were
labeled with [a-32P]dCTP. Five positive phages were randomly selected and inserts were
recovered as pBIuescript SK phagemids according to the manufacture’s instruction
(Stratagene, Cedar Creek, TX). All of them were sequenced and shown to be identical
clones. I used the sequenced black locust full-length cDNA clone to search for other plant
homologous clones on the National Center for Biotechnology Information
(http://www.ncbi.nlm.nih. gov) and the TIGR (http://www.tigr.org) sequence databases.
The full-length cDNA clone of Arabidopsis allantoinase gene (TAIR clone no. 3989737)
was obtained from TAIR. The multiple alignment of amino acid sequences was carried
out by CLUSTALW (Thompson et al., 1994) using default parameters, and the
phylogenetic tree was created by DRAWTREE (PHYLIP unrooted phylogenetic tree,

Felsenstein, 1989).

102

Identification of ataln T-DNA Insertional Mutant

Seven T-DNA insertion mutants (SALK-000325, SALK-000327, SALK-000282,
SALK-013427, SALK-013985, SALK-013986, and SALK-142607) of the AtALN gene
generated by the Salk Institute Genomic Analysis Laboratory (http://signal.salk.edu/)
were obtained from the Arabidopsis Biological Resource Center (Ohio State University,
Columbus). The seeds were planted on kanamycin plates and the resulting kanamycin-
resistant plants were transferred to soil and seeds were harvested separately from
individual plants. Subsequently, the seeds were screened on Murashige and Skoog media
supplied with 10 mm ALN as a sole nitrogen source. After an additional phenotypic
selection, I randomly chose one mutant line (ataln m2-1) that did not grow well on ALN
media for further analyses. To confirm the mutant line as a homozygote, PCR was
performed with the genomic DNA of ataln m2-1 mutants using gene-specific oligo-
nucleotides (forwardl, 5’-AGAGATATGGAGAGAAC1TI‘ GC1T-3’; forward2, 5’-
GGAAGGAGACATI‘GATATGCTGAG-3’; reverse, 5 ’ -
AAG1TAAGTAGTI‘GCAAG1TGCAG—3’) or one gene-specific primer and one left-
border-specific primer. As expected, the progenies of the homozygous mutant line
showed the mutant phenotype and kanamycin resistance with no segregation. The mutant
line did not have any noticeable morphological alterations or growth defects under the

normal condition.

RNA Isolation
Using the Trizol reagent (Invitrogen, Carlsbad, CA), total RNAs were isolated

from seedlings of Arabidopsis and black locust. RNA extraction from inner wood was

103

performed as previously described (Yang et al., 2003). Briefly, mature trees (20 cm
diameter at breast height) were felled using a chain saw and were made into 25-cm-long
logs. The logs were immediately placed on ice and brought back to a wood shop, where
thin cookies (approximately 1 cm thick) were made using a table saw. The cookies were
immediately submerged in RNA extraction buffer (20 mM EDTA, pH 8.0, 50 mM Tris-
HC], pH 8.0, 0.2% [w/v] SDS, and 10 mM B-mercaptoethanol). While submerged, the
trunk wood sections (bark/cambial region, sapwood, and transition zone) were carved out
by using a chisel and hammer, and were washed with RNase Away solution (Invitrogen).
The isolated sections (approximately l-cm 3 cubicles) were frozen in liquid nitrogen and
stored at -80°C until needed. For RNA isolation, the frozen samples were first ground in
a blender and then further ground to a fine powder using a mortar and pestle. The
grounded sample was first passed through the shredder column of DNeasy Maxi kit
(Qiagen, Hilden, Germany), and was then subjected to total RNA isolation using the

RNeasy Maxi kit (Qiagen) and cleaned up by RNeasy Mini kit (Qiagen).

Northern Blots

Total RNA (5 or 10 ug) was separated on a 1.2% (w/v) agarose gel containing 1X
MOPS buffer with 2.2 M formaldehyde at 60 V for 3.5 h. The RNA was then washed
twice in water and transferred by capillary action to a nylon membrane (Hybond N+;
Amersham Pharmacia Biotech, Piscataway, NJ). Membranes were prehybridized for 2 h
at 42°C in 50% (v/v) formamide, 6X SSC, 5X Denhardt’s solution, 0.5% (w/v) SDS, and
20 pg mL'l sonicated salmon sperm DNA. Hybridizations were performed overnight at

65°C after adding the randomly primed 32 P-labeled cDNA probe in the prehybridization

104

buffer. Membranes were washed at 65°C for 10 min each with washing solution I (2.0X
SSC and 0.1% [w/v] SDS), washing solution H (0.5X SSC and 0.1% [w/v] SDS), or
washing solution HI (0.1X SSC and 0.1% [w/v] SDS). Random-labeled probes were
made with a kit (Rediprime 11; Amersham Pharmacia Biotech) using an internal fragment

from the full-length cDNA of AtALN and RpALN 1.

RT-PCR

One microgram of total RNA, isolated from wild-type and mutant plants, was
reverse transcribed using a random primer in a 20 IIL reverse transcription reaction as
recommended by the manufacturer (SuperScript H reverse transcriptase; Invitrogen). One
microliter of the first-strand cDNA reaction was used as a template for PCR amplification
with specific oligo-nucleotide primers for AtALN (forward, 5’-
AGAGATATGGAGAGAACTITGCIT-3 ’; reverse, 5 ’ -
AAG1'I’AAGTAG1'I’GCAAG1'I’GCAG-3’). As an internal control, plant 183 primers
(Ambion, Austin, TX) were used. The PCR products were loaded on ethidium bromide-

stained agarose gels.

Antisense Northern Blot

Antisense northem-blot analysis was performed as previously described (Yang et
al., 2003). For the generation of antisense RNAs, total RNAs (2 ug) were amplified using
Message AmpTM aRNA kit (Ambion). I began by synthesizing first stand cDN As from
the total RNAs of seedling, bark/cambial region, sapwood, or transition zone, and the

first cDNAs were used as templates for the synthesis of second cDNAs. Finally, aRN As

105

from the second cDNAs were generated by in vitro transcription (amplification). About
200 ng of aRN As was separated in a formamide agarose gel and transferred onto the
nylon membrane using the capillary transfer method. The membrane was then hybridized
with an isotope-labeled probe. The signal was exposed and detected on an X-ray film.
Histon 3.2, which showed constitutive expression in black locust tissues (Yang et al.,

2003), was used as a control.

Genetic Complementation of Yeast (Saccharomyces cerevisiae) Mutant with RpALN and
AtALN Genes

To conduct the functional complementation test of RpALN and AtALN in yeast, 1
found the yeast allantoinase (DALI) gene deletion mutant on the Saccharomyces
Genome Deletion Project web page (http://wwwsequence.stanford.edu/ group/
yeast_deletion_project/ deletions3.html), and bought the homozygous diploid strain of
dalI deletion mutant (item no. 4035962) from American Type Culture Collection
(Manassas, VA). The homozygous diploid of the mutant is in the background of 8Y4743
(8Y4741: MATa his3A1 leu2A0 metl 5A0 ura3A0/8Y4742: MAT_his3Al leu2A0
lysZAO ura3A0). The deficiency of the mutant, using ALN as a sole nitrogen source, was
determined on solid plates containing different concentrations of ALN as sole nitrogen
sources. Optimal selective conditions were found at 2 g L'l. The dalI mutant was
transformed with an expression vector pYES2 (Invitrogen) harboring a full-length cDNA
of RpALN or AtALN gene. PCR products generated from the full-length cDNA of
RpAIJV or AtALN gene were digested by BamHI and Xbal and ligated into the BamHI-

XbaI site of plasmid pYES2. At the first, transforrnants were selected on solid plates that

106

were composed of a yeast nitrogen base, 2% (w/v) dextrose, 0.5% (w/v) ammonium
sulfate, and a synthetic complete supplement mixture minus uracil (Qbiogene, Carlsbad,
CA). RpALN or AtALN transformants were then confirmed by PCR analysis using gene-
specific primers. After uracil selection, each transforrnant was reselected on a solid plate
containing yeast nitrogen base, 2% (w/v) Gal, 1% (w/v) raffinose, and 0.2% (w/v) ALN
(Sigma, St. Louis) as a nitrogen source, and 1/100-fold reduced synthetic complete
supplement mixture minus uracil to supply Leu and His. The colonies capable of growing
were reselected under the same conditions. As a negative control, the mutant was
transformed with the empty vector. After a single colony from each transformed mutant
was chosen, growth complementation tests were performed using ALN as a sole nitrogen

source (Desimone et al., 2002).

Protein Extraction and Enzyme Assay

Crude extracts for ALN assays of RpALN and AtALN were prepared according
to the method described by Bell and Webb (1995) with slight modification. Plant samples
(0.5-1.0 g) were ground with 4X volumes of 20 mM Tris-HCl and 500 mM NaCl, pH
7.5, 100 pM dithiothreitol, 10 mM EDTA, 0.1% (w/v) polyvinyl-polypyrrolidone, and a
mixture of protease inhibitors, including phenylmethylsulfonyl fluoride (2 mM), aprotini
(10 pg mL"), leupeptin (10 pg mL'l), and pepstatin (10 pg mL") in a chilled mortar, with
a pestle. Extracts were centrifuged at 30,000g for 30 min at 4°C and were filtered through
two layers of Miracloth (Calbiochem, San Diego), and the supernatant was collected.
Then, the supernatant was filtered through 0.2-Inn filters (Millipore, Bedford, MA) to

remove particulate matter. Extracts were kept on ice or -20°C until they were used. The

107

 

assay for allantoinase activity was based on the procedure reported by Schubert (1981)
and Romanov et a]. (1999). The enzyme extract was incubated for 20 min at room
temperature with 10 mM ALN in 50 mM Tris-HCl buffer. The pH was set to 7.5.
Reaction solution (0.5 mL) was then treated with 0.125 mL of 0.2 N HCl, and was heated
in a boiling water bath for 4 min to degrade ALA to glyoxylate and urea. After cooling
for 30 sec, 0.4 mL of concentrated HCl and 0.125 mL of potassium ferricyanide solution
(16 mg mL") were added and incubated at 37°C for 30 min. The total reaction solution
was immediately applied to measure the A520 arising from the final product of this
reaction, glyoxylate diphenylformazan. All assays were replicated at least twice, and
controls were included to account for endogenous ALA in the enzyme extract and non-

enzymatic degradation of ALN.

Results

Allantoinase Genes from Black Locust and Arabidopsis

In my previous EST analysis from the trunk wood of black locust (Yang et al.,
2003), I found an 881 that was homologous to other reported allantoinase genes such as
those found in Escherichia coli (Kim et al., 2000), yeast (Buckholz and Cooper, 1991),
and bullfrog (Hayashi et al., 1994). To obtain a full-length cDNA clone, 1 used 4- to 6-d-
old seedlings to construct a cDNA library, and then screened them using the EST as a
probe. The full-length cDNA, named RpALN, was comprised of 1,820 bp with an open
reading frame of 1,536 bp, which encodes a 56,375-D protein with 512 amino acid

residues. The protein was predicted by TargetP V1.0

108

 
 

Tomato Mas--oxxs A LPLPLS-PLPYPDPS --------- xxnpssD-csn
Arabidopsis MERTLLQWRL p LALvanrsrrrasp --------- asaoounxcsn
Black locust MDQ--vaa PJLALLVSFLVFPYLQDSYKAQLSPPIRLPGDB-CSL
Rice . nanxxxxcn p LAVAAALAAALLYRA ------------ prsxvsoa
Yeast ------------- uprnxrrsnavrrua ----------------------
Bullfrog ------------- unansxpenmrrpo ----------------------
R.coli ------------------ MSFDLIIKNG ----------------------
Tomato Basra ssxarvrp ”or: 31x arr assnwavusa-rr N
Arabidopsis Henri SSKRIVTPE GLIS :s Bvx II xsvnwaxsoasnv ID
Black locust HRKY SSKRIVTPQGIII BIN BI IIBGYGKQGNS-HQB ID
Rice PPRYSNLRSLLSYEKAEENPF rasp ----- RGGEGRADQRDRRRGLPI
Yeast auxpxr ----- IvvsrssorILDVLs --s er--Brrxrsrarnzu
Bullfrog sxrsvfinmroxgrrslcmnsn xrs naweuxavrscaxnno-
B.coll --------- TVILB EARVVDIAVK xraxro ------ onnonxxsfinn
Tomato YRISF‘VVMPGID H'HLDDSPSPGREWEGF

Arabidopsis YGE-VEMPGLID HVHLDDPGRSBWEGFPS KAAAAGGITT VDMPLNS
Black locust YGE-VVMPGLID HVHLDEPGRTEWEGFD RAAAAGGETT vnan N
Rice LPAPATGGGLR--HEHLDBPGRABWEGFS RAAAAGGITTLVDMPLNS
Yeast: VSPCTILPGL osnvrrr. EPGRTSEWEGF QAAEGGETT VDMPL A
Bullfrog vonnvvnficlrnpava LAAAAGGIT-IVDMPLNS
B.coli ASGLV SPGI -

 
  
 
 

 

Tomato -P8 VQAAEGRV VDVGFWGGLVPE 'E
Arabidopsis FPS JVDVGFWGGLVPD ‘L
Black locust pr EAAEKKL vnvcrwchBpB 'L

  

arcs ps DAAKDKLIVDVGFWGGLVPE L
Yeast IPP rnv NFRI EAAEGQMWCDVGFWGGLVPH---
Bullfrog pp r vrurar QAAKRQC vnvnrwchIPD---

 

8.coli P DRASIB P-AAKGILTID AQLGGLVSY---

Figure 3-2. Alignment of allantoinase amino acid sequences

Black locust putative allantoinase, Arabidopsis putative allantoinase and tomato putative
allantoinase were compared with published allantoinases from E. coli, yeast and bullfrog.
Completely conserved amino acids in all species are marked in black; identical amino
acids in dark gray; similar amino acids in light gray; different amino acids in white. The
conserved residues responsible for metal biding are indicated with asterisks.

MD

(httpzllwww.cbs.dtu.dk/services/TargetP/; Nielsen et al., 1997; Emanuelsson et al., 2000)
to contain a signal peptide for the secretory pathway and a cleavage site at the 30th amino
acid. The RpALN sequence information was used to carry out BLASTN searches against
public databases (National Center for Biotechnology Information,
http://www.ncbi.nlm.nih.gov and The Institute for Genomic Research [TIGR],
http://www.tigr.org). I found an Arabidopsis homologous gene at the locus (accession no.
AT4G04955). The BLASTX analysis with the RpALN amino acid sequence further
confirmed its homology with an E value of 0.0 (BLASTX score = 670). I obtained a full-
length cDNA clone (The Arabidopsis Resource [TAIR] clone no. 3989737) for this gene
from TAIR (http://www.Arabidopsis.org). The size of the inserted gene was 1,552 hp
with an open reading frame of 1,518 bp, encoding a protein with 506 amino acid residues
and a calculated molecular mass of 55,438 D. Again, this protein was predicted to have a
signal peptide for the secretory pathway and a cleavage site at the 26‘h amino acid. The
Arabidopsis genome has only one copy of this gene. The deduced amino acid sequences
of the RpALN and AtALN share highly conserved residues that are responsible for metal

binding that were found in other reported ALN proteins (Figure 3-2).

Function of RpALN and AtALN

After cloning the RpALN and AtALN, 1 confirmed their functions via genetic
complementation of a yeast mutant (dal 1) deficient in ALN hydrolysis. This dalI mutant
grows slowly with ALN as a sole nitrogen source because no enzymatic degradation of
ALN to urea can occur (Winkler et al., 1988). However, both transformants carrying

RpALN or AtALN cDNA gene grew well (Figure 3-3), suggesting the enzymes encoded

110

by the plant genes could catalyze the hydrolysis of ALN to allantoic acid. Recently,
Desimone et al. (2002) have shown that an Arabidopsis ureide permease (AtUPSl) could

genetically complement an ALN uptake—deficient yeast mutant.

1:10 1:100 1:1000

Control
(pYESZ)

AtALN

RpALN

 

Figure 3-3. Functional complementation of a yeast allantoinase mutant

The dall mutant was transformed with the empty vector (pYESZ), with pYESZ carrying
AtALN, or with pYESZ carrying RpALN. Three cell dilutions (1:10, 1:100, and 1:1,000)
were plated on yeast minimal medium containing 0.1% allantoin as sole nitrogen sources
and incubated for 5 days at room temperature.

Arabidopsis Mutant ataln

Consisting of 15 exons and 14 introns, the AtALN gene is located on chromosome
4 of the Arabidopsis genome (Figure 3-4A). To examine the phenotypic consequence of
the AtALN gene mutation, I screened T-DNA insertion mutants received from the Salk
Institute Genomic Analysis Laboratory (http:l/signa1.salk.edu/). First, single-copy T-

DNA insertion was confirmed by 3:1 (resistantzsensitive) segregation in kanamycin

Ill

 

 

 

d

A

O N § 0) a O N
I l

1

Enzyme actIVIty
(pmolelmin*mg protein)

 

 

0.1
WT ataln m2-1

Figure 3-4. Identification of T-DNA insertional mutants of Arabidopsis allantoinase

112

selection. From the SALK-000327 line, I identified one mutant line (ataln m2-1) that did
not grow well on Murashige and Skoog medium (Murashige and Skoog, 1962), which
was composed of 10 mm ALN, instead of ammonium nitrate (Figure 3-4 8). 1 confirmed
the T-DNA insertion in the Arabidopsis genome using PCR analysis with the gene-
specific primer sets, which failed to detect stable AtALN transcripts in the mutant plants
when compared with wild-type plants (Figure 3-4 A and C). As expected, the progenies
of the homozygous mutant line showed the mutant phenotype and kanamycin resistance
with no segregation. No allantoinase activity was detected in the mutant plants. However,
wild-type seedlings had an allantoinase activity of about 10 pmol Inin‘l mg'1 protein
when cultured on ALN media (Figure 3-4D). This level was higher than the activity of
wild-type seedlings (1 pmol min'l mg'1 protein) and the black locust seedlings (5 pmol
min'l mg'1 protein) cultured on Murashige and Skoog media, suggesting that the AtALN
gene may be regulated by exogenous nitrogen conditions. In the presence of 10 mm
ALN, the mutant seedlings were smaller when compared with the wild-type seedlings.
Furthermore, the leaf of the mutant was colored pale green, and the mutant cotyledons
showed very little red. Notably, wild-type seedlings and mutant seedlings that were
subjected to poor nitrogen conditions (less than 10 pM ammonium nitrate) showed
similar phenotypes. Regardless of the nitrogen levels in the media, the ataln mutant
growth patterns did not differ from those of the wild-type seedlings previously reported
by Desimone et al. (2002). However, when the mutant seedlings were transferred from
the ALN media onto Murashige and Skoog media containing 10 mm ammonium nitrate,
the seedlings recovered, leaf color returned to dark green, and the plants grew normally

and produced flowers.

113

 

Phylogenetic Analysis

This Arabidopsis ALN (AtALN) has been annotated as a putative dihydroorotase
(DHO) or DHO-like protein. Because ALN and DHO, along with hydantoinase and
dihydropyrimidinase, belong to the same superfamily of cyclic amidohydrolases (EC
3.5.2.-), this annotation is not surprising. Furthermore, many plant ALN gene sequences
were not correctly annotated (mainly as “unknown” or “expressed protein”). Amino acid
sequence alignment of RpALN identifies seven plant ALNs (Arabidopsis [Arabidopsis
thaliana], barrel medic [Medicago truncatula], potato [Solanum tuberosum], rice [Oryza
sativa], soybean [Glycine max], and tomato [Lycopersicon esculentum]) with high levels
of amino acid sequence similarity (Figure 3-5). To examine the structural relationship of
ALNs and DHOs, I constructed an unrooted phylogenetic tree with the seven plant ALNs
and four plant DHOs (Figure 3-5). First, the analysis clearly showed that the two
enzymes were unmistakably different at the amino acid level. The two enzyme groups
(ALNs and DHOs) were distinctly separated in the phylogenetic tree, ALNs having very
low similarity to DHOs (less than 12% similarity at the amino acid level). Within the
same enzyme group, it was obvious that the proteins from the same family had the most
structural similarity to each other. For instance, black locust ALN was most similar to
that of soybean (GmALN; 82% similar) and barrel medic (MtALN; 86% similar), just as
potato ALN was almost identical to the tomato ALN (95% similar). Furthermore, potato
DHO (StDHO) was the most similar to tomato DHO (LeDHO; 98% similar).

In this report, 1 confirmed the function of putative ALNs from two species (black
locust and Arabidopsis) and I suggest that the six plant sequences, which are currently

annotated as “unknown” proteins, are ALNs based on sequence homology and

114

 

(A)

$0)
Q
9°

~o
o
v“
Ba . \
We, meme po
x. " tato
09 0
0° 7

 

 

 

 

 

 

 

 

 

 

 

 

 

\ a 3
6°C} 6? 993 93
-Q 0-
), O
a? ‘9.
m
Allantoinases Dihydroorotases
(B)
Allantoinase (ALN) Dlhyctoorotase (DHO)
BLBMTMPTATRCTMPTATRC
Soybean
LSB) 92 76 59 56 62 49 19 19 4 9
33'3““"“ 199 86 66 64 67 55 19 19 6 9
fifie'md" 199 66 63 67 57 11 11 7 s
5 Tomato
< (TM) 199 95 67 55 11 11 9 8
PP?“ 199 67 52 11 11 7 19
Arabldopsls
(AT) 199 57 19 19 6 3
Rice
IRC) 199 s 11 7 6
HR“ 199 98 77 77
0 Potato
3 (PT) 199 77 77
Arabidopsis
(AT) 199 76

 

 

 

 

 

 

Figure 3-5. Dendrogram and similarity of allantoinases and dihydroorotases

115

phylogenetic relationship. Allantoinase belongs to a superfamily of metal-containing
amidohydrolases and is predicted to have a common three-dimensional fold (Holm and
Sander, 1997). Especially, the conserved. residues (Figure 3-2, asterisks) responsible for
metal biding were found in all reported ALNs. Recently, the recombinant E. coli ALN
was shown to have the highest activity levels when observed in cell extracts from cultures

supplemented with zinc and cobalt (Mulrooney and Hausinger, 2003).

RpALN Gene Expression

During seed germination and seedling growth, nitrogen metabolism is a critical
factor for the survival of the tree in a natural environment. To determine the RpALN
expression pattern during early development of black locust, northem-blot analysis was
carried out using total RNAs isolated from cotyledons, axis, hypocotyls, and roots in
different seed germination stages. In black locust, an axis emerges from the seed coat at 1
d post-imbibition. The axis continues to grow and shows the line of root and shoot
junction at 4 DAG. At 6 DAG, cotyledons open, become wrinkled, and true leaves
sprout. In the axis and hypocotyls, the RpALN gene expression level peaked at 2 to 4
DAG, and then dropped at 6 DAG. On the other hand, its expression in cotyledon was the
highest at 6 to 9 DAG (Figure 3-6A) and reduced to the basal level at 19 DAG. No
significant signal was detected in root tissues (4—6 DAG). The differential expression of
the gene during the seed germination and seedling growth was also observed at the level
of enzyme activity. The allantoinase activity was the highest in axis/hypocotyledon at 4

DAG and in cotyledon at 6 DAG (Figure 3-68).

116

 

A

DAG 0 2 4 6
Tissue C A CACHCRCHCR

RpALN "Irv“.wnc‘
rRNA - "" "

 

   

65 I Cotyledon

 

6.0 C] Hypocotyl
Root

2.8

 

1.2 as

Enzyme activity
(pmole/min”mg protein)
9
@

O-‘Nw-hUION
I

 

 

 

 

 

 

 

4 DAG 6 DAG

Figure 3-6. Differential expression of black locust allantoinase

(A) Northern blot analyses of RpALN in black locust seedlings. RpALN mRNA is
differentially expressed in seedling tissues. According to the number of days after
germination, total RNA (10 pg) from the indicated tissues was electrophoresed and
blotted. Then, the RNA was probed with [anP]dCTP-labeled RpALN cDNA. Each lower
panel presents the corresponding agarose gel stained with ethidium bromide, showing
ribosomal RNAs.

(8) Enzyme assay of allantoinase of seedling tissues. Total proteins were extracted from
three tissues of 4 day-old or 6 day-old seedlings: cotyledon (black bar), hypocotyl (white
bar) and root (hatched bar). Results represent means of two independent experiments.

117

Trunk wood serves as a conduit for long-distance transport of nutrient and
represents one of the largest sink tissues in trees. The expression pattern of RpALN gene
was examined in the bark/cambial zone, sapwood, and sapwood-heartwood transition
zones of mature black locust trees (Figure 3-7). The bark/cambial zone is the outer most
region, the sapwood is the bright-colored middle region, and the heartwood is the dark,
inner region. The transition zone, the interphase between sapwood and heartwood, is
where the carbohydrates are converted to heartwood extractives. Because most of the F‘—
cells in trunk wood are dead, the quantity and quality of mRNA that can be extracted
from such tissue is limited. To overcome this problem, I used antisense northem-blot

analysis, which included an in vitro transcription process to amplify mRN As (Yang et al.,

 

 

2003). My results showed that the bark/cambial region had the highest expression level
(Figure 3-7A). 1 then examined the seasonal regulation of RpALN gene expression using
bark/cambial zone tissues collected in spring, summer, and fall. In temperate deciduous
trees, leaf nitrogen is translocated to perennial tissues (e. g. trunk wood) during the fall
and the stored nitrogen is remobilized for spring shoot growth (Taylor and May, 1967;
Ryan and Bormann, 1982). I found that the expression level of RpALN gene was higher
in the spring and fall samples compared with that of the summer sample (Figure 3-78),
suggesting that the RpALN gene may be involved in nitrogen resorption and storage

during the fall and resupply during the spring growth.

Effects of Nitrogen on RpALN and AtALN Gene Expression
To test whether nitrogen affects allantoinase gene expression, 1 carried out

northem-blot analysis using black locust and Arabidopsis seedlings that were cultured on

118

Murashige and Skoog medium supplemented with or without nitrogen. The absence of
nitrogen in the medium appears to induce the expression of RpALN gene (Figure 3-8A)
during seedling germination and growth (2 and 4 DAG). In Arabidopsis seedlings, the
absence of nitrogen or the presence of ALN as a sole nitrogen source dramatically

increased AtALN gene expression (Figure 3-88).

119

 

so so sw rz
RpALN arr-Q

Histon H321- . ...i

B
SP SM FL

RpALN lt- r, lit-msggi
rRNA

 

Figure 3-7. RpALN gene expression in trunk wood of black locust

(A) Antisense northern blot analysis of RpALN mRNA in seedling (SD), bark/cambial
zone (8C), sapwood (SW), and transition zone (12) by antisense northern blot analysis.
Histon H 3.2 transcript levels were indicated as a control. Each sample lane contained an
equal amount of 200 ng of aRN As.

(8) Northern blot analysis of RpALN mRNA using three different seasons of trunk wood.
RpALN mRNA is differentially expressed in bark/cambial zones according to spring (SP),
summer (SM) and fall (FL). Total RNA (10 pg) from the indicated season samples was
separated and blotted, and the mRNA was then detected with [a32P]dCTP-labeled
RpALN cDNA. Lower panel presents the corresponding agarose gel stained with
ethidium bromide, showing ribosomal RNAs.

120

 

ZDAG 4DAG
-N +N -N +N

RpALN *mfl ‘

MS -N +A
AtALN “t”

Figure 3-8. Nitrogen influence on the expression of allantoinase genes

(A) Expression of RpALN mRNA under nitrogen conditions. Total RNAs were isolated
from 2 day-old or 4 day-old seedlings grown under nitrogen absent condition (-N) or
supplied with 10 mM ammonium nitrate (+N).

(8) Northern blot analysis of AtALN mRNA in different nitrogen conditions. Total
RNAs were isolated from 8 day-old seedlings cultured on MS medium supplemented
with nitrogen (MS), without nitrogen (-N) or with 10 mM allantoin instead of the
nitrogen sources (+A). Total RNA (10 pg) from the indicated season samples was
separated and blotted, and the mRNA was then detected with [a32P]dCTP-labeled
RpALN or AtALN cDNA. Each lower panel presents the corresponding agarose gel
stained with ethidium bromide, showing ribosomal RNAs.

 

121

Discussion

Biochemical analyses have shown that ureides such as ALN and ALA are present
in higher plants, including ureide-type legume species such as soybean (Tracey, 1961;
Atkins et al., 1991; Cheng et al., 2000). Desimone et al. (2002) demonstrated that even
non-legume Arabidopsis can take up and use ALN as a nitrogen source. Here, I cloned
and characterized two full-length cDNA clones (AtALN and RpALN) encoding
allantoinase, further confirming that the ALN degradation pathway exists in Arabidopsis
as well as in amide-type legumes (e.g. black locust). To my best knowledge, this is the
first report on the cloning of functional allantoinase genes in plants. Using these two
ALN gene sequences in a BLAST search against public databases, 1 identified additional
ALN genes from various plant species including barrel medic, potato, soybean, potato,
rice, and tomato. Multiple sequence alignment showed that all of the ALNs share high
sequence similarity with known ALNs from other non-plant organisms (bullfrog, E. coli,
and yeast), and identified characteristic conserved residues for metal binding. On the
other hand, their sequence similarity with plant DHOs was very low (5 11%). Plant ALN
genes appear to be expressed abundantly in various tissues, suggesting that the ureide
pathway is universal among plant species. In fact, Desimone et al. (2002) demonstrated
that Arabidopsis could uptake and use ALN as a sole nitrogen source. The BLAST search
against EST databases and the rice genome sequence database on the TIGR identified 15
ESTS from developing stems, developing leaves, drought-treated plants of barrel medic,
12 ESTS from dormant tuber, mature tuber, leaves/petioles of potato, and 31 ESTS from
most tomato organs. Other plant species for which ALN genes were identified include

rice, cotton (Gossypium hirsutum), soybean, and barley (Hordeum vulgare).

122

 

Until now, plant allantoinase genes have been mis-annotated as putative,
unknown, or DHO. Particularly, the current Arabidopsis genome annotation calls AtALN
a putative DHO or DHO-like protein. At least three pieces of evidence confirm that the
RpALN and AtALN genes encode allantoinase. First, both of the genes were able to
genetically complement a yeast mutant defective in ALN hydrolysis. Second, an
Arabidopsis mutant (ataln mZ-l) with T-DNA insertion in the 12th exon of AtALN was
not able to use ALN as sole source of nitrogen, whereas wild-type plants grew on ALN
medium. Third, a phylogenetic tree constructed with seven plant ALNs and four plant
DHOs showed that the two enzyme groups were unmistakably different at the amino acid
level.

Seed germination and seedling growth are the developmental stages that can
affect the survival of the plant in its natural ecosystem. Developmental and metabolic
programs are accomplished during the early stage of plant life (Howell, 1998;
Holdsworth et al., 1999; Eastmond et al., 2000). Different metabolic pathways are carried
out in each tissue of the seedling during its growth and development. For example, a
nutrient source tissue (e.g. cotyledon) performs metabolisms that generate nutrients
transported to sink tissues, whereas a sink source tissue (e.g. axis) conducts metabolisms
involved in cell growth. Thus, an axis cell likely demands more nitrogen and carbon than
does a cotyledon cell. It is necessary even for the same metabolism to be differentially
controlled depending on the tissues involved. Enzyme activities required for the
production of ureides were observed in young soybean seedlings (Polayes and Schubert,
1984). My results showed that black locust seedlings differentially express allantoinase

genes depending on tissue type and developmental stages. Although these findings

123

 

suggest that allantoinase may be differentially regulated according to the nitrogen
demand, it is not known whether the allantoinase degradation pathway plays any
nutritional role in seedling development and growth.

Many plant purine biosynthetic enzymes appear to be localized in organelles
(Smith and Atkins, 2002). Previous reports indicate that allantoinase is localized in
microbodies of plants and animals (for review, see Schubert and Borland, 1990). Hanks
et al. (1981) reported that allantoinase was localized in the endoplasmic reticulum (ER).
It has been suggested that this ER localization may play a role in the export of ALA from
the cell. However, the ER localization of allantoinase has not been independently
confirmed. In this regard, my finding that both of the two plant allantoinases have
putative transit sequences is interesting. However, further protein localization studies,
using green fluorescent protein (GFP) fusion proteins in transgenic plants, can provide
direct evidence for the intracellular localization of allantoinase. The cloning of the two
functional ALN genes can facilitate such studies.

Signals derived from nitrogen treatment or nitrogen shortage are certainly
involved in triggering widespread changes in gene expression and metabolic pathways.
Cooper et al. (1990) have shown that in yeast, the ureide pathway is transcriptionally up-
regulated under the nitrogen-deficient conditions. It appears that allantoinase genes are
down-regulated in nitrogen-rich conditions, but up-regulated in nitrogen-limiting
conditions. For example, the yeast allantoinase gene (dall) was up-regulated by nitrogen
depletion as Well as amino acid starvation (Gasch, et al., 2000; also web supplement at
httpzl/genome-www.stanford.edu/yeast stress/index.shtml). In Arabidopsis, cDNA

microarray data available from the Stanford Microarray Database (http:l/genome-

124

 

www5.stanford.edu/MicroArray/SMDI) showed that the ALN gene is down-regulated by
nitrogen treatment, potassium nitrate versus no treatment (experiment ID nos. 3787 and
3789), the potassium chloride versus ammonium chloride (experiment ID nos. 12308 and
12309), and the potassium chloride versus potassium nitrate (experiment ID nos. 10849
and 10851). In addition, previous studies showed that nitrate treatment decreased the
nodule mass and nitrogen-fixing activity, which subsequently resulted in the reduction of
ureide content in the xylem sap in soybean (Israel and McClure, 1980; McNeil and
LaRue, 1984) and cowpea (Pate et al., 1980) plants. My data showing nitrogen-induced
down-regulation of ALN gene expression confirm these earlier physiological and
biochemical observations.

In addition to the developmental regulation of ALN, a differential gene expression
among different tissue types and seasonal variation might offer some insight on nitrogen
metabolism in perennial woody plant species. In the current study, I found that black
locust ALN genes had the highest expression levels in bark/cambium tissue compared
with sapwood, sapwood-heartwood transition zone, or seedlings. In the fall, nitrogen in
the leaves is mobilized and transported to storage tissues such as bark and xylem ray cells
(Kang and Titus, 1980; Chapin and Kedrowswki, 1983). The stored nitrogen is then
remobilized to support the spring growth (Taylor and May, 1967; Ryan and Bormann,
1982). It is notable that the RpALN gene expression was up-regulated in spring and fall
compared with summer, again supporting the previous physiological observations. This
seasonal differential expression of RpALN gene strongly suggests that RpALN may be
involved in the seasonal cycle of nitrogen resorption, storage, and recycling in temperate

deciduous trees. Furthermore, the cloning of two plant ALN genes will facilitate further

125

 

studies on the molecular and biochemical basis for nitrogen uptake, metabolism, and

recycling in plants.

126

 

Literature Cited

Atkins CA, Storer PJ, Young EB (1991) Translocation of nitrogen and expression of
nodule-specific uricase (nodulin-35) in Robinia pseudoacacia. Physiol Plant 83: 483-
491.

Bell JA, Webb MA (1995) Immunoaffinity purification and comparison of allantoinases
from soybean root nodules and cotyledons. Plant Physiol 107: 435-441.

Buckholz RG, Cooper TG (1991) The allantoinase (DALI) gene of Saccharomyces
cerevisiae. Yeast 7: 913-923.

Campbell JW, Bishop SH (1970) Nitrogen metabolism in molluscs. In JW Campbell, ed,
Comparative Biochemistry of Nitrogen Metabolism, Vol 1. Academic Press, London,
UK. PP 103-206.

Chapin F8, Kedrowski RA (1983) Seasonal changes in nitrogen and phosphorus fractions
and autumn retranslocation in evergreen and deciduous taiga trees. Ecology 64: 376-
391.

Cheng XG, Nomura M, Takane K, Kouchi H, Tajima S (2000) Expression of two uricase
(Nodulin-35) genes in a non-ureide type legume, Medicago sativa. Plant Cell Physiol
41: 104-109.

Cooper TG (1980) Selective gene expression and intracellular compartment: two means
of regulating nitrogen metabolism in yeast. Trands Biochem Sci 5: 332-334.

Cooper TG, Ferguson D, Rai R, Bysani N (1990) The GLN3 gene product is required for
transcriptional activation of allantoin system gene expression in Saccharomyces
cerevisiae. J Bacteriol 172: 1014-1018.

Desimone M, Catoni E, Ludewig U, Hilpert M, Schneider A, Kunze R, T egeder M,
Frommer W8, Schumacher K (2002) A novel superfamily of transporters for

allantoin and other oxo derivatives of nitrogen heterocyclic compounds in
Arabidopsis. Plant Cell 14: 847-856.

Eastmond PJ, Germain V, Lange PR, Bryce JH, Smith SM, Graham IA (2000) Post-
germinative growth and lipid catabolism in oilseeds lacking the glyoxylate cycle.
Proc Nat] Acad Sci USA 10: 5669-5674.

Emanuelsson O, Nielsen H, Brunak 8, von Heijne G (2000) Predicting subcellular
localization of proteins based on their N-terrninal amino acid sequence. J Mol Biol
300: 1005-1016.

Felsenstein J. (1989) PHYLIP -- Phylogeny Inference Package (Version 3.2). Cladistics
5: 164-166.

127

 

Gasch AP, Spellman PT, Kao CM, Carmel-Hare] O, Eisen M8, Storz G, Botstein D,
Brown PO (2000) Genomic expression programs in the response of yeast cells to
environmental changes. Mol. Biol. Cell 11: 4241-4257.

Hanks JF, Tolbert NE, Schubert KR (1981) Localization of enzymes of ureide
biosynthesis in peroxisomes and microsomes of nodules. Plant Physiol 68: 65-69.

Hayashi S, Jain S, Chu R, Alvares K, Xu 8, Erfurth F, Usuda N, Rao MS, Reddy SK,
Noguchi T, Reddy JK, Yeldandi AY (1994) Amphibian allantoinase. Molecular
cloning, tissue distribution, and functional expression. J Biol Chem 269: 12269-
12276.

Holdsworth M, Kurup S, McKibbin R (1999) Molecular and genetic mechanisms
regulating the transition from embryo development to germination. Trends Pharmacol
Sci 4: 275-280.

Holm L, Sander C (1997) An evolutionary treasure: unification of a broad set of
amidohydrolases related to urease. Proteins 28: 72-82.

Howell SH (1998) Seedling development. In Molecular Genetics of Plant Development.
Cambridge University Press, Cambridge, UK, pp 83-102.

Israel DW, McClure PR (1980) Nitrogen translocation in the xylem of soybeans. In FI‘
Corbin, ed, World Soybean Research Conference 11: Proceedings, Westview Press,
Boulder, Colorado, pp 111-127.

Kang SM, Tirus JS (1980) Qualitative and quantitative changes in nitrogenous
compounds in senescing leaf and bark tissues of apple. Physiol Plant 50: 285-290.

Kim GJ, Lee DE, Kim HS (2000) Functional expression and characterization of the two

cyclic amidohydrolase enzymes, allantoinase and a novel phenylhydantoinase, from
Escherichia coli. J Bacteriol 182: 7021-7028.

McClure PR, Israel DW (1979) Transport of nitrogen in the xylem of soybean plants.
Plant Physiol 64: 411-416.

McNeic DL, LaRue TA (1984) Effect of nitrogen source on ureides in soybean. Plant
Physiol 66: 720-725.

Mulrooney SB, Hausinger RP (2003) Metal ion dependence of recombinant Escherichia
coli allantoinase. J Bacteriol 185: 126-34.

Mufioz A, Piedras P, Aguilar M, Pineda M (2001) Urea is a product of ureidoglycolate
degradation in chickpea. Purification and characterization of the ureidoglycolate urea-
lyase. Plant Physiology 125: 828-834.

Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with
tobacco tissue cultures. Physiol Plant 15: 473-479.

128

 

Nielsen H, Engelbrecht J, Brunak 8, von Heijne G (1997) Identification of prokaryotic

and eukaryotic signal peptides and prediction of their cleavage sites. Prot. Eng. 10: l-
6.

Noguchi T, Fujiwara S, Hayashi S (1986) Evolution of allantoinase and allantoicase
involved in urate degradation in liver peroxisomes. A rapid purification of amphibian
allantoinase and allantoicase complex, its subunit locations of the 2 enzymes, and its
comparison with fish allantoinase and allantoicase. J Biol Chem 261: 4221-4223.

Pate JS, Atkins CA, White ST, Rainbird RM, Woo KC (1980) Nitrogen nutrition and
xylem transport of nitrogen in ureide-producing grain legumes. Plant Physiol 65: 961-
965.

Polayes DA, Schubert KR (1984) Purine synthesis and catabolism in soybean seedlings.
The biogenesis of ureides. Plant Physiol 75: 1104-1110.

Raychaudhuri A, Tipton PT (2002) Cloning and expression of the gene for soybean
hydroxyisourate hydrolase. Localization and implications for function and
mechanism. Plant Physiol 130: 2061-2068.

Romanov V, Merski MT, Hausinger RP (1999) Assays for allantoinase. Analytical
Biochemistry 268: 49-53.

Ryan DF, Bormann PH (1982) Nutrient resorption in northern hardwood forests.
8ioScience 32: 29-32.

Schubert KR, Boland MJ (1990) The ureides. In B] Miflin, PJ Lea, eds, The
Biochemistry of Plants, Vol. 16. Academic Press, San Diego, pp 197-283.

Schulten HR, Schnitzer M (1998) The chemistry of soil organic nitrogen: A review. Biol
Fertil Soils 26: 1-15.

Shubert KR (1981) Enzymes of purine biosynthesis and catabolism in Glycine max. Plant
Physiol 68: 1115-1122.

Smith PMC, Atkins CA (2002) Purine biosynthesis. Big in cell division, even bigger in
nitrogen assimilation. Plant Physiol 128: 793-802

Stryer L (1988) Biochemistry, Ed 3. WH Freeman, New York, pp 618-623

Taylor 8K, May LH (1967) Nitrogen metabolism, translocation and recycling in apple
trees. Hort. Rev. 4: 204-246

Thompson JD, Higgins DG, Gibson 1] (1994) CLUSTAL W: improving the sensitivity

of progressive multiple sequence alignment through sequence weighting, position-
specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673-4680

129

 

WT

Tracey MV (1961) Urea and Ureides. In K Peach K, M Tracey, eds, Modern Methods of
Plant Analysis, V014. Springer Verlag, Heidelberg, pp 119-141

Vogels GD, Trijbels F, Uffink A, (1966) Allantoinases from bacterial, plant and animal
sources. 1. Purification and enzymatic properties. Biochim Biophys Acta 122: 482-
496

Winkler RG, Blevins DG, Polacco JC, Randall DD (1988) Ureide catabolism in nitrogen-
fixing legumes. Trends Biochem Sci 13: 97-100

Yang J, Park S, Kamdem DP, Keathley DE, Retzel E, Paule C, Kapur V, Han K-H (2003)
Novel gene expression profiles define the metabolic and physiological processes
characteristic of wood and its extractive formation in a hardwood tree species,
Robinia pseudoacacia. Plant Mol Biol 52: 935-956.

130

 

1)

2)

3)

4)

5)

6)

7)

8)

9)

CONCLUSIONS

In this study, ESTS present unique gene sets that are expressed in uncharted deep
inside trunk wood.

EST analysis shows that the libraies from the different trunk regions have
distinctive characteristics.

The gene expression profiles are different based on the regions of trunk wood, and
highly expressed genes represnt the character of each region.

Depending on season, the gene expression patterns are different.

The genes related to secondary metabolism are highly expressed in the summer
season when compared with other seasons.

Allantoinase gene is not specific in ureid-type legume,but general in plant
kingdom.

The black locust allantoinase gene is differentially regulated in seedling accoding
to the nitrogen demand.

The allantoinase gene is involved in nitrogen mobilization in black locust.

Negatively, nitrogen regulates the expression of allantoinase genes.

131