- .. . .H
« _;<--htv.vu—l;

'..'Mn1:
.
A I 19:

w
..

1:" ;:.7‘.'.::‘uu\..t \
-m. >4 '5‘ ‘
. "aunt.” .
... I... “fl"? nu-

._ .
. .11.
115.2 m.

rang.
‘

um:
:1‘m x «:ng. .fi

”via-.3:- v
f

,.

t
1
|

o

. v

 

c
“M“'ly
“W“

‘ a - . I ‘ -l
"i ‘ .lvvlr'fn‘t -
.. .

 

r»-
. ~<r:: :«

'_-_‘-x'1_r_'I’

 

ur
' " 4 'r pp.
n v. ~ . ‘ . . , , ’V “Unbr’,
V V -c.
1 mm , I ’. "u 'M
w”... ‘ ' ' . H r ’ ' t
1 A 't-p- ‘ .

.L‘.”

I v

. --— «um av

A. ’ ' ‘I'Auu—u. r-

 

 

'a 3
."..L.....,;
.';’. “y...”

In;

*— .”._ __ ___.—__———..

MICHIGAN TATE UN v sm UBRARI
ll!
l

ill I [ll unuIt'lllummmmtill

3 1 93 00877 0087

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the
dissertation entitled
Molecular analysis of the structure and function of two
chitin-binding proteins: barley and stinging nettle lectins.

presented by

David Ross Lerner

has been accepted towards fulfillment
of the requirements for

Doctoral degree inBotany and Plant Pathology

 

 

flax/W flea/24%

Major professor

W A fag

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove thie checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE '

LJ__
SE:
| l-l

:

 

    
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU le An Afflnnetlve ActloNEquel Opportunlty Institution
cmmem

——_—.———

MOLECULAR ANALYSIS OF THE STRUCTURE AND FUNCTION OF TWO
CHITlN-BINDING PROTEINS: HARLEY AND STINGING NEITLE LECTINS

By

DevidFioeeLemer

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

DepartmentofBotanyandPIantPathology

1992

b/J' woo

ABSTRACT

MOLECULAR ANALYSIS OF THE STRUCTURE AND FUNCTION OF TWO

CHITIN-BINDING PROTEINS: BARLEY AND STINGING NETTLE LECTINS

BY

David Ross Lerner

Plants possess a family of chitin-binding proteins containing one or more
conserved domains, 40 to 43 amino acids in length, responsible for binding to
chitin. Stinging nettle and barley Iectins consist of two and four repeated chitin-
binding domains, respectively. Besides binding chitin, both of these Iectins can
inhibit the growth of insect larvae in vitro. Nettle lectin, however, can also inhibit
the growth of some chitin-containing fungi. The objectives of this dissertation were
to isolate and characterize barley and nettle lectin cDNA clones and to investigate
the expression and structure of these Iectins as they relate to possible functions
in plant defense.

The barley lectin cDNA clone, isolated using a partial wheat germ agglutinin

(WGA) cDNA as a heterologous probe, was shown to be over 90% identical to

WGA. Expression of barley Iectin mRNA was localized to the adult root tips and
the embryonic root cap and coleorhiza with RNA gel blot analysis and in situ
hybridizations. Barley Iectin was also shown to be synthesized with a glycosylated
carboxyl-terminal propeptide not found in the mature protein.

A synthetic gene for nettle Iectin was prepared and used as the probe for
cloning the nettle Iectin cDNA. Although nettle Iectin is only 86 amino acids long,
the cDNA encoded a 372 amino acid polypeptide containing a putative signal
sequence, the lectin, and a carboxyI-terminal domain with chitinase activity and
similarity to other plant chitinases. Southern and northern blot analyses were used
to determine that nettle Iectin was encoded by a small multigene family expressed
in rhizomes and inflorescence.

To initiate investigations into the relationship between structure and
antifungal activity of chitin-binding proteins, modified barley Iectin (mBL) genes
encoding approximately two chitin-binding domains, the size of nettle lectin, were
constructed using oligonucleotide directed mutagenesis. Analysis of mBL
expressed in either E. coli or tobacco revealed that the mBL retained chitin-binding
activity.

The results of these investigations are consistent with the hypothesis that
plant chitin-binding proteins have defense-related functions. The potential for
modifying chitin-binding proteins for investigating basic biological problems and

practical applications has also been demonstrated.

Copyright by
DAVID ROSS LERNER
1992

To my parents who, through their love and example,

have shown me that anything is possible.

ACKNOWLEDGEMENTS

I would like to thank the members of my committee for their helpful
discussion throughout my work on this dissertation. I would also like to thank Dr.
John Olhrogge for help with the design and synthesis of the synthetic gene for
stinging nettle Iectin. Everyone in Natasha’s laboratory, past and present, deserve
thanks for their constant help with techniques, ideas, encouragement and
forebearance of numerous questions.

Natasha deserves a special acknowledgement for her constant
encouragement and support. She has often gone far beyond what was required
or expected to ensure my development as a productive researcher.

I am also indebted to my dear friends Neera Agrwal and David Rhoads kept
me amused, distracted and in shape. Finally, I must thank Yolanda for being with

me even when she was miles away.

vi

TABLE OF CONTENTS

List of Tables .............................................
List of Figures ............................................
Abbreviations .............................................
CHAPTER 1. lNTRODUCTION: Chitin-Binding Proteins and
Evidence for Their Role in Plant Defense ....................
Introduction .........................................
Proteins Containing Only Chitin-Binding Domains .............
Proteins Containing Chitin-Binding Domains Fused
to an Unrelated Domain ...........................
Protein Processing and Targeting .........................
Regulation of Gene Expression ...........................
In Vitro Activities ......................................
Conclusions .........................................
References ..........................................
CHAPTER 2: Cloning and Characterization of Root-Specific
Barley Lectin ........................................
Abstract ............................................
Introduction .........................................
Materials and Methods .................................
Results .............................................

Discussion ..........................................

vii

ix
x

xii

10
12
16
20
24

26

37
38
39
4o
45

57

References ..........................................

CHAPTER 3: The Gene for Stinging Nettle Lectin (Un‘ica
dioica Agglutinin) Encodes both a Lectin and a
Chitinase ...........................................
Abstract ............................................
Introduction .........................................
Materials and Methods .................................
Results .............................................
Discussion ..........................................
References ..........................................
CHAPTER 4: Size, Structure and Function Relationships
Chitin-Binding Proteins: Preliminary Studies on
Modifying Barley Lectin .................................
Introduction .........................................
Materials and Methods .................................
Results .............................................
Discussion ..........................................
References ..........................................
CHAPTER 5: Summary and Future Prospects ....................
APPENDIX A .............................................
Synthetic Stinging Nettle Lectin Gene Sequence and

Oligonucleotide Structure ...............................

viii

63

68
69
70
71
82
1 02

107

114

123
134
141
149

154

159

LIST OF TABLES

Table 1.1 Chitin-binding proteins properties and in vitro
activities ....................................... 6

Table 2.1 Differences in deduced amino acid sequence between
barley Iectin and WGA-B ........................... 48

Table 3.1 In vitro activities of chitin-binding proteins. ................ 118

Figure 1.1

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6
Figure 3.7

Figure 3.8

Figure 4.1

LIST OF FIGURES

The domain structure of chitin-binding and related proteins
in plants. ......................................

Nucleotide and deduced amino acid sequence of barley
Iectin cDNA clone BLc3. ...........................

In vitro translation and immunoprecipitation analysis
of poly A“ RNA and BLc3 transcripts. .................

Western blot analysis of native and Endo H treated
barley Iectin .....................................

Localization of barley Iectin mRNA by in situ hybridization. . . .
Northern analysis of poly A”r mRNA from root tips

and coleoptiles of 3-d-old seedlings and 15 to

25 dpa barley embryos. ...........................
Schematic diagram of the sequential ligation reactions

used to produce the synthetic stinging nettle Iectin

gene from 18 synthetic oligonucleotides. ...............

Nucleotide and deduced amino acid sequence of the stinging
nettle Iectin cDNA clone uda1. ......................

PCR analysis using reverse transcribed total RNA from
U. dioica as the template with primers from uda1. ........

Comparison of stinging nettle Iectin amino acid sequences. . .

Comparison of uda1 deduced amino acid sequence with the
deduced amino acid sequences of various chitinases. .....

RNA gel blot analysis of total RNA from U. dioica tissues .....
Southern blot analysis of U. dioica genomic DNA. .........

Chitinase assay of crude cell extracts from E. coli cells
expressing uda1 encoded domains. ..................

Gramineae Iectins and homologous proteins. .............

47

51

54

56

58

76

89
90

93
96

98

101

117

Figure 4.2 Graphical representation of mature barley Iectin and
stinging nettle Iectin. .............................. 121

Figure 4.3 Alignment of stinging nettle Iectin amino acid sequence
with the deduced amino acid sequence of barley Iectin

domain pairs. ................................... 126
Figure 4.4 Barley Iectin double deletion construction. ............... 129
Figure 4.5 Expression of fusion protein in E. coli. .................. 137
Figure 4.6 Chitin column affinity purified mBL expressed in E. coli. ..... 140
Figure 4.7 Expression of mBL in transgenic tobacco. ............... 143

xi

ABBREVIATIONS

Blc3 ......................... Barley Iectin cDNA clone #3

BSA ......................... Bovine serum albumin

CaMV ....................... Cauliflower mosaic virus

cDNA ........................ Complementary deoxyribonucleic acid
CTPP ........................ Carboxyl-terminal propeptide

dATP ........................ Deoxy-adenosine triphosphate

dGTP ........................ Deoxy-guanosine triphosphate

DPA ......................... Days post-anthesis

DTI’ ......................... Dithiothreitol

EDTA ........................ Ethylenediamine tetraacetic acid
EGTA ........................ Ethylene glycol-bis(B-aminoethyl ether)-

N,N,N’,N’-tetraacetic acid

Endo H ...................... Endoglycosidase H

FW ......................... Fresh weight

GIcNAc ...................... N-acetylglucosamine

HPLC ........................ High pressure liquid chromatography
HRGP ....................... Hydroxyproline rich glycoprotein
IPTG ........................ lsopropyI-B-D-thiogalactopyranoside
kb .......................... lfilobases

kDa ......................... Kilodaitons

xii

mBL ......................... Modified barley Iectin

MBP ........................ Maltose binding protein

mRNA ....................... Messenger ribonucleic acid

nt ........................... nucleotide

oligo ........................ oligodeoxyribonucleotide

PBS ......................... Phosphate buffered saline

PCR ......................... Polymerase chain reaction

pfu .......................... plaque forming unit

PMSF ........................ Phenylmethyl sulfonyl fluoride
PVP ......................... Polyvinyl pyrrolidone

RBV ......................... Remizol brilliant violet

rGTP ........................ Riboguanosine triphosphate

RT .......................... Room temperature

SDS ......................... Sodium dodecyl sulfate

SSC ......................... Standard sodium citrate

TE .......................... Tris (10 mM), EDTA (0.1 mM)
TMV ......................... Tobacco mosaic virus

Tris ......................... Tris(hydroxymethyl)amino methane
UDA ......................... Urtica dioica agglutinin

WGA ........................ Wheat germ agglutinin

win1 ......................... Wound inducible gene 1 (potato)
win2 ......................... Wound inducible gene 2 (potato)

xiii

CHAPTER 1

INTRODUCTION:

Chitin-Binding Proteins and Evidence for
Their Role in Plant Debnee

2
INTRODUCTION

A family of chitin-binding proteins in plants can be identified by the presence
of a distinct and conserved chitin-binding domain (figure 1, open boxes). This
domain is characteristically 40-43 amino acids in length and is rich in Gly and Cys
residues. The structure of the chitin-binding domain has been elucidated from
X-ray crystallography of wheat germ agglutinin (WGA) (Wright, 1977). WGA has
four tandem repeats of the distinct chitin-binding domain. Each chitin-binding
domain of WGA contains 8 Cys residues making up four disulfide bonds. This
maintains the tight globular structure of each domain since there are no distinct
regions of a-helix or B-sheet in the chitin-binding domains (for review see Wright
et al.,1991). Further studies of WGA crystallized with bound polysaccharides
revealed the specific amino acid residues directly involved in carbohydrate binding
(Wright, 1980). These residues and the Cys residues involved in disulfide bridges

are highly conserved in all chitin-binding proteins examined (Wright et at, 1991).

Chitin-binding proteins can be divided between proteins with multiple chitin-
binding domains repeated in tandem and proteins containing a single chitin-
binding domain. In both cases the chitin-binding domain is often fused to the
amino-terminal of a structurally unrelated domain (figure 1). Only proteins with
multiple chitin-binding domains (is. Gramineae Iectins) are able to agglutinate red
blood cells indicating the presence of multiple N-acetylglucosamine (GIcNAc), the

chitin monomer, binding sites.

.coszES m: :o o3m=m>m E Ban 9. Sn .2. meE cozmooa o5 >2 ocuocco . ogm cozmimoozo “.3572 52:33
m 2:950 582 Ste: 3355 695.0 coon Ho: go; 2353 635 .8 3:3 8:3 Emu .moEozooE E9». 298
twosome coon mm: c.5823 EmEou czoe. ommomcsom new 335:. .mBoLo_E_Em-o< 64206 3:20 So cocoacem Eon
oEEm peoaoou 9: So: 259: 33.935 coon 9m: 3553:» 52.00 493m 9 55 c265» 2m 5055 Lo 320

some he mEmEop 95:828. moxom .35... c. «582.. “.329. can @535:on no 233:» 52:25 2:. . P 2:9".—

     

 

 

 

 

 

 

 

 

gin-C avg-.10

Egg-a no.5!

 

 

,Ww/A/flx///////////////////////////////////

 

 

 

 

 

 

 

for; 255.09

“Zr—hum..— ”(gas

3E9.-
4388.35

V In. 8030....
2%....
8223510 .. g0

“mgr—58.3

25W..— MHEZ 02.92:.”

02-0”... ”(NZ—go

OZEPOm—n. DWFSNE OZ< >1..S_<u_ OZ_DZ_m-ZF.IO WI...

5

The wide distribution of both classes of chitin-binding proteins in monocots
and dicots indicates conserved and essential functions for these proteins. In spite
of this, the endogenous functions have yet to be conclusively demonstrated.
There are many indications, however, that these proteins are involved in plant
defense against microbial pathogens or insects (Chrispeeis and Raikhei, 1991).
In this review i will describe the chitin-binding proteins, their processing, regulation
and in vitro activities (T able 1) with particular emphasis on properties related to

possible plant defense functions.

PROTEINS CONTAINING ONLY CHIT'lN-BINDING DOMAINS

Proteins comprised solely of chitin-binding domains are of special interest
since their properties and characteristics are not influenced by the presence of an
unrelated domain. In all cases these proteins are highly resistant to heat and
proteases, and retain chitin-binding activity after heating to 70°C for 15 min.
These proteins range in size from 3 kDa monomeric polypeptides to 36 kDa

homodimers containing 8 chitin-binding domains.

Gramineae Leciins
The Gramineae Iectins are, perhaps, the most studied proteins containing
chitin-binding domains. These iectins have been isolated from wheat, barley, rye,

rice and a number of other Gramineae species (for review see Raikhei and Lerner,

Table 1. Chitin—binding proteinspropertleeandhvib'oactivitiee

6

 

PROTEIN

    

 

MOL. WT.:
MATURE/

# CHITIN-
BINDING

CARBOXYL-
TERMINAL

IN VITHO
ACTIVITIES

 

 

 

 

 

 

 

 

 

 

PROPROTEIN DOMAINS

Gramineae 18 kDa/23 kDa 4/monomer None Agglutination

Iectins 36 kDa 8/dimer insecticidal

Stinging 8.5 kDa/ 2 Chitinase Agglutination

nettle Iectin (38 kDa)1 Fungicidal
Insecticidal

Class I 30 kDa 1 Chitinase Fungicidal

chitinase

Hevein 4.5 kDa/ 1 PR4—Iike Fungicidal

20 kDa

Solanaceae 30 to 68 kDa 22/monomer HRGP-like Agglutination

Iectins

Ac-AMP 3 kDa 2/391 ? Fungicidal

 

‘Molecular weight shown in parenthesis is deduced from sequence data.
2Based on biochemical data only.

 

7
1991). In addition, cDNA clones have been isolated for the three WGA isolectins

(Raikhei and Wilkins, 1987; Smith and Raikhei, 1989), barley Iectin (Lerner and
Raikhei, 1989), and rice Iectin (Wilkins and Raikhei, 1989). Mature Gramineae
Iectins are 36 kDa homodimers with four chitin-binding domains in each monomer.
Biochemical studies, including subunit exchange experiments (Peumans at al.,
1982), show that the Gramineae Iectins, and in particular the cereal Iectins, are
very similar. Carbohydrate binding sites of these Iectins have been localized to the
cleft between the subunits with both monomers contributing residues which
interact directly with the ligand. Although dimer formation is apparently essential
for the Gramineae Iectin ligand binding, Wright and coworkers (1991) suggest that
dimer formation is not required for saccharide binding in other chitin-binding
proteins.

In all cases studied, the Gramineae Iectins are expressed in specific cell
layers in both embryos and root tips of adult plants (for review see Raikhei and
Lerner, 1991). Immunolocalization (Mishkind at al., 1983) and in situ hybridization
to mRNA (Lerner and Raikhei, 1989; Wilkins and Raikhei, 1989; Raikhei et al.,
1988) allowed the determination that root tip expression is restricted to the root
cap cells and outer layer of cells at the root tip. These studies also show that
Gramineae Iectins are expressed in tissues which protect the embryonic leaves,
the coieoptile, and roots, the coleorhiza, in the seed and while germinating.
Expression of the Gramineae iectins in the coleorhiza and progenitor root cap cell
of the embryo was found in all species studied (Mishkind et al., 1983). Gramineae

Iectin expression in the coleoptiles, however, varies from species to species

8
(Mishkind et al., 1983). In rice, the lectin is expressed throughout the coleoptile.

Rye contains Iectin in the single outermost and innermost cell layers, while in
wheat, WGA is present only in the outermost cell layer of the coleoptile. Barley
Iectin is not found in the coieoptile at all (Mishkind et al., 1983 and Lerner and
Raikhei, 1989). Interestingly, root cap cells, the coleoptiie and the coleorhiza are
all considered to be protective tissues and are the first to contact soil microbes as

the roots and germinating seedlings grow through the soil.

Stinging Nettle Lectin

Stinging nettle Iectin (Un‘ica dioica agglutinin, UDA) is a 8.5 kDa chitin-
binding protein which, like the Gramineae Iectins, can agglutinate red blood cells
(Peumans etal, 1984). Unlike the Gramineae Iectins however, stinging nettle Iectin
does not effectively bind the chitin monomer, GIcNAc, which may account for its
lower agglutination specific activity (Shibuya et al., 1986). The amino acid
sequence of stinging nettle Iectin, as determined by analysis of tryptic fragments
(Chapot at al., 1986) and the cloned cDNA (Lerner and Raikhei, 1992), shows two
chitin-binding domains in tandem. Proteins with multiple chitin-binding domains
therefore are not restricted to the Gramineae. This may be important evolutionarily
especially if the domain duplication occurred prior to divergence into monocots
and dicots as suggested by Wright, et al. (1991). Also, like the Gramineae Iectins,
stinging nettle Iectin is found in the seeds (Lerner and Raikhei, 1992) and in

underground tissues, in this case the roots and rhizomes (Peumans, at al., 1984).

9

Localization of stinging nettle Iectin to cell layers within these tissues has not been

determined.

Hevein

Hevein is perhaps the archetypical chitin-binding protein because it consists
of exactly one 43 amino acid chitin-binding domain. This domain, when found in
other plant proteins, has been referred to as the hevein domain’. Hevein
accumulates in the vacuole-like iutoid bodies of specialized cells, laticifers, of
rubber trees (Hevea brasiliensis) (Waiq'ono at al., 1975). These cells form an
anastomosing system through the trunk, stems and leaves of rubber trees and are

responsible for latex synthesis (Esau, 1960).

Amarmfhue caudatus Antimicrobial Peptides (Ac-AMP1 and Ao-AMPZ)

Two very small, 30 and 31 amino acid, chitin-binding proteins have been
isolated from the seeds of amaranth (Amaranflws caudatus) (Broekaert et al.,
1992). Peptide sequencing data indicates Ac-AMP1 and 2 are identical except for
an additional Arg residue at the carboxyl-terminus of Ac-AMP2. Thus Ac-AMP1
and Ac-AMP2 may be derived from a single precursor polypeptide by differential
post-translational processing or degradation during the isolation procedure. These
chitin-binding proteins appear to be truncated variants of the chitin-binding
domain. Alignment of the Ac-AMP peptide sequences and chitin-binding domains
from other proteins shows that approximately 50% of the residues are identical

with the insertion of a 4 residue gap in the Ac-AMP sequence. Wright and

10

coworkers (1991) refer to residues 3 to 30 of a consensus chitin-binding domain
as the "inner conserved core" and define it as responsible for ligand binding. Ac-
AMPs amino acid sequence corresponds to this "inner conserved core". It is not
surprising therefore that the shortened chitin-binding domains of Ac-AMPs do bind

chitin.

PROTEINS CONTAINING CHITlN-BINDING DOMAINS FUSED TO AN
UNRELATED DOMAIN

Class I Chltinases

The class i chitinases contain a single 40 to 43 amino acid chitin-binding
domain linked by a short hinge region to the chitinase catalytic domain (for review
see Boiler, 1988). The chitin-binding domain generally gives these proteins a high
isoelectric point. Because of this, these chitinases are often referred to as the
basic chitinases. Proteins containing a homologous chitinase catalytic domain by
itself, the acidic or class II chitinases are expressed from genes which do not
encode a chitin-binding domain. Thus the class II chitinases are not derived from
the class I chitinases by removal of the chitin-binding domain. The class I
chitinases are expressed in many plant tissues including leaves, flowers, seeds,
and roots (Shinshi et al., 1987; Lotan, 1989). These chitinases have been identified

in virtually all plants examined (for reviews see Linthorst, 1991; Bol at al., 1990).

11
Prostinglng Nettle Lectin

The stinging nettle Iectin mRNA encodes a polypeptide extending 174 amino
acids beyond the 86 amino acids encoding the lectin’s two chitin-binding domains
(Lerner and Raikhei, 1992). This carboxyI-terminal domain has approximately 45%
homology with class I and class II chitinase catalytic domains. The identity of
prostinging nettle Iectin’s carboxyl-terminal domain with these other chitinase
catalytic domains, however, is considerably less than the identity between the class
I and class II chitinases. In spite of this, chitinase activity is observed when the 37
kDa prostinging nettle Iectin or the 28 kDa carboxyI-terminal domain alone is
expressed in E. coli (Lerner and Raikhei, 1992). It is not known, however, whether
prostinging nettle Iectin exists as a stable protein or is just a transient precursor

of stinging nettle Iectin in the plant.

Proheveln and Potato Wound Inducible Genes, wini and wlnll

Like stinging nettle lectin, the hevein mRNA encodes a polypeptide
extending far beyond (141 amino acids) the single chitin-binding domain (43 amino
acids) which makes up hevein (Broekaert at al., 1990). Highly homologous
polypeptides are also encoded by two wound inducible genes from potato, wini
and winll (Stanford at al., 1989). The 20 kDa prohevein, 4 kDa hevein and 16 kDa
hevein carboxyl-domain are all found as major protein constituents of rubber tree
latex (Lee et al., 1991) although these polypeptides have not been identified in
potato. Proteins with homology to the hevein carboxyI-domain alone have now

been identified as the pathogenesis-related protein from tobacco, PR-4, and

12
tomato, PR-P2 (Linthorst et al., 1991 ). Cloning the PR-4 and PR-P2 cDNAs reveals

that they are encoded by mRNAs which do not encode the chitin-binding domains

(Linthorst et al., 1991).

Solanaceae Lectlns

Chitin-binding proteins from potato tubers, tomato fruits, and thorn-apple
seeds have been identified which contain a Cys and Gly rich, chitin-binding,
domain linked to a highly glycosylated hydroxyproline-rich domain (Broekaert at
al., 1987; Allen, at al., 1978; Nachbar, et al., 1980). Although little protein
sequence data are available, the amino acid composition and chitin-binding activity
of these proteins suggests they contain chitin-binding domains homologous to
other chitin-binding proteins. In addition, the agglutinating activity exhibited by
these lectins suggests they contain multiple chitin-binding domains. Amino acid
and carbohydrate compositions of the hydroxyproline-rich domain indicate that this
region is very similar to other hydroxyproline-rich glycoproteins (HRGPs) such as
extensin. These unique lectins are the only chitin-binding proteins known to

contain a high amount of carbohydrate.

PROTEIN PROCESSING AND TARGETTNG

Chitin-binding proteins are secretory proteins which are synthesized on the

ER and processed through the Golgi apparatus before accumulation in the

13
vacuole or cell wall (Chrispeels and Raikhei, 1991; Meins et al., 1992). The

analysis of cDNA clones of chitin-binding proteins indicates they all encode a
putative amino-terminal signal sequence which would target the nascent

polypeptides to the ER.

Carboxyl-Tenninal Extensions and Vacuolar Targeting

In addition to the signal sequence, the cDNAs of several chitin-binding
proteins encode a short extension beyond the carboxyl-terminus of the mature
protein. The Gramineae lectins, for example, have well characterized carboxyl-
terminal propeptides (CTPP’s). The WGA isolectins and barley Iectin have 15
amino acid CTPP’s which are almost identical in sequence (for review see Raikhei
and Lerner, 1991). Rice Iectin contains a CTPP which differs in length (26 amino
acids) and amino acid sequence from the cereal Iectin CTPPs (Wilkins and Raikhei,
1989). A common feature of the Gramineae Iectin CTPPs is an asparagine-Iinked
glycosylation site which is utilized to yield a glycosylated precursor (Mansfield et
al., 1988). This glycosylated precursor is processed into the smaller,
unglycosylated, mature polypeptide. The tobacco class I chitinase is also
synthesized with a carboxyl-terminal extension which is not found in the mature
protein (J.-M. Neuhaus, Botanisches Institut der Universitat Basel, personal com-
munication). The chitinase extension is only 6 amino acids long and does not
contain a glycosylation site. Comparison of cloned class I chitinases indicates that

they all encode similar regions 6 to 10 amino acids long.

14
The CTPP of barley Iectin and the carboxyl-terminal extension of the

tobacco class I chitinase have recently been shown to be the vacuolar targeting
signals for these proteins. Bednarek et al. (1990) demonstrated that barley Iectin
is secreted when expressed in tobacco plants without the CTPP while the wild type
barley Iectin is faithfully targeted to the vacuole. The CTPP of barley Iectin
(Bednarek and Raikhei, 1991) and the carboxyl-terminal extension of the tobacco
class I chitinase (Neuhaus at al., 1991b) are sufficient for redirecting a normally
secreted protein, cucumber class III chitinase, to the vacuole when expressed in
tobacco. Interestingly, the secreted class II chitinases, which do not possess the
amino-terminal chitin-binding domain, also do not have the carboxyl-terminal
extension encoded in their cDNA’s. Mauch and Staehelin (1989) suggest that the
vacuolar localization of the class I chitinase and 81,3-9Iucanase in bean leaves is
directly related to their role as a last line of the plant’s defense against invading
pathogens.

Two other chitin-binding proteins, prostinging nettle Iectin and prohevein,
have carboxyl domains longer than homologous proteins which do not contain the
chitin-binding domain. Like the class I chitinases, the stinging nettle Iectin chitinase
domain is 10 amino acids longer than the class II chitinases. The intra cellular
localization and potential processing of this extension have not yet been
investigated. As stated earlier, the carboxyl-domain of prohevein is homologous
to tobacco PR-4 which does not contain a chitin-binding domain. Similar to the
chitinases, the vacuolar prohevein is 14 amino acids longer than the secreted PR-

4. The actual carboxyl-terminus of prohevein, however, is still unknown. Thus, a

15

pattern is emerging wherein chitin-binding proteins containing an extensive
carboxyl-domain also have a short carboxyl-terminal extension which may be
responsible for their targeting to the vacuole. Homologous proteins lacking the
chitin-binding domain and carboxyl-terminal extension are found in the apoplastic
space outside the plasma membrane. It is important to note that chitin-binding
domains by themselves are not sufficient for targeting or redirecting proteins to the

vacuole (Bednarek and Raikhei, 1991; Neuhaus at al., 1991 b).

Additional Processing Events

Several chitin-binding proteins undergo internal proteolytic cleavage events.
Approximately 85% of the rice Iectin monomer (18 kDa) is cleaved into 8 and 10
kDa polypeptides (Stinissen et al., 1983). Rice Iectin also contains an additional
pair of Cys residues (Wilkins and Raikhei, 1989) not found in the other Gramineae
lectins, which form a disulfide bridge between the 8 and 10 kDa polypeptides
(Wright et al., 1991).

Prohevein undergoes an internal cleavage to release hevein from the
carboxyl-domain (Lee at al., 1991). During this processing a 6 amino acid ’spacer'
region between the two domains is lost. it is interesting to note that, similar to rice
lectin, apparently not all of the prohevein is cleaved and all three polypeptides
(prohevein, hevein and the carboxyl-domain) are found in the rubber tree latex
(Lee et al., 1991).

Presumably, prostinging nettle Iectin is also processed to release the 8.5

kDa lectin from the carboxyl-terminal chitinase domain. This processing event,

16

however, has not been demonstrated experimentally. In light of the accumulation
of the 8.5 kDa lectin to high levels in the rhizomes and the processing events
demonstrated for prohevein, it is likely that prostinging nettle Iectin undergoes a

specific proteolytic cleavage as well.

REGULATION OF GENE EXPRESSION

Developmental Regulation

Many chitin-binding proteins are expressed in specific tissues. As a result,
their expression is, in part, developmentally regulated. The Gramineae Iectins, for
example, are expressed in developing embryos and accumulate as the embryos
mature. Raikhei at al. (1988) found WGA mRNA is first detectable at 10 days post
anthesis (DPA) and reaches maximum levels at 30 to 40 DPA. The WGA mRNA
level later decreases as the seeds desiccate and increases slightly as the seeds
germinate, due to increased lectin expression in the growing root tips (Raikhei at
al., 1988). Stinging nettle lectin is also found in seeds and not in leaves or stems
(Lerner and Raikhei, 1992) indicating developmental control over its accumulation.
interestingly, stinging nettle Iectin mRNA and protein are not found in aerial stems
but accumulate to high levels in stems which have been buried to generate
rhizomes (Lerner and Raikhei, 1992). The developmental conversion of these
stems to rhizomes must, therefore, induce expression of the lectin. The class I

chitinases have been found in many different plant tissues and developmental

17

regulation of expression has been particularly noted by higher mRNA levels in

older leaves (Shinshi et al., 1987) and floral tissues (Lotan et al., 1989).

Hormonal Regulation

Exogenously applied plant growth regulators, abscisic acid (ABA) and
ethylene, have been shown to enhance the levels of message and protein for
several chitin-binding proteins. Gramineae lectin accumulation in embryos and
root tips of adult plants is increased by ABA (Raikhei et al., 1986; Triplett and
Quatrano, 1982). For example, WGA accumulation in root-tips is increased 2 to
3 fold by exogenous application of 10 mM ABA (Raikhei et al., 1986). Endogenous
ABA also accumulates in developing wheat embryos beginning just before the
developmental increase in WGA accumulation (Raikhei at al., 1987). This is
consistent with the hypothesis of ABA control over WGA expression. Exogenously
applied ABA (50 uM) also causes a several-fold increase in Hevein message in
rubber tree latex (Broekaert at al., 1990). Hevein and chitinase have been shown
to accumulate upon application of ethephon which converts to ethylene. Hevein
mRNA in rubber tree leaves, stems, and latex increased several-fold in the
presence of 0.1% ethephon (Broekaert at al., 1990). Broglie at al. (1986) found a
30-fold increase in chitinase activity of bean seedlings following exposure to

1mg/ml ethephon.

18
StreeslnducedExpreeelon

Environmental stresses often result in the increased production of ABA or
ethylene by the plant. It is not surprising, therefore, that osmotic stress and
wounding can also cause the accumulation of chitin-binding proteins. The effects
of osmotic stress on expression of chitin-binding proteins have been examined
only for the Gramineae lectins. WGA and ABA accumulate when seedlings are
stressed by air drying or chemically with mannitol (0.5 M) or polyethylene glycol
(180 g/l) (Cammue et al., 1989). Under these conditions WGA levels increase up
to 10-fold, but only in tissues at the root tips where the protein is normally found
(Cammue at al., 1989). Similar results have been found for expression of chitin-
binding proteins after wounding. The proteins accumulate to higher than normal
levels in tissues where they are otherwise expressed. Within 24 hours of
wounding, dramatic increases in hevein message in rubber tree leaves is seen.
A slightly lesser increase in hevein mRNA accumulation is found in rubber tree
stems and latex (Broekaert et al., 1989). The results for latex expression are
somewhat difficult to interpret since latex is collected by ’wounding’ the rubber
tree. Because of this, the high levels of expression seen in ’unwounded' latex may
also be due to wounding. As implied by their names, the hevein homologues in
potato, wound inducible mRNAs 1 and 2 (MM and win2) accumulate in wounded
potato plants (Stanford et al., 1989). Interestingly, these two genes seem to have .
arisen through gene duplication and show differential expression in potato tissues
after wounding (Stanford at al., 1989). Specifically, win2 will accumulate in

wounded tubers while win1 does not. Both mRNAs accumulate in wounded leaves

19

and stems but not in wounded roots. Class I chitinases have also been shown to
accumulate in leaf tissue in response to wounding (Hedrick et al., 1988; Parsons
at al., 1989). Wounded stinging nettle leaves do not, however, accumulate the
stinging nettle lectin mRNA which also encodes a chitinase (Lerner and Raikhei,

unpublished results).

Induction by Plant Pathogens

Although plants contain no chitin, insect pests and many fungal and
bacterial pathogens do. Plant responses to these organisms by the accumulation
of chitin-binding proteins are strongly suggestive of the role chitin-binding proteins
may play in plant defense. Tobacco mosaic virus infection causes the systemic
induction of the classical acidic pathogenesis-related (PR) proteins, and the class
I chitinases (Legrand at al., 1987). The levels of tobacco class I chitinase, for
example, increase 20-fold after 3 days of infection by tobacco mosaic virus (T MV)
(vogeli-Lange at al., 1988). In addition, the tobacco class I chitinase mRNA
increases 6-fold after infection by the bacterial pathogen Pseudomonas tabaci and
increases ZOO-fold after infection by the fungal pathogen Phytophmora parasifica
var nicotianae (Meins and Ahl, 1989). In bean (Phaseolus vulgaris), infection of
hypocotyls with the fungus Col/etotrichum Iindemuthianum causes an increase in
class I mRNA which peaks at approximately 7 days after inoculation (Hedrick et
al., 1988). A fungal cell wall elicitor causes an increase of the class I chitinases in
suspension cultured cells of tobacco and rice (Hedrick stat, 1988; Zhu and Lamb,

1991 respectively). WGA accumulation is also induced 2- to 5-fold in root-tips by

20

infection with several different fungi and fungal cell wall elicitors (Cammue at al.,
1990).

Many insect pests cause wounding-like iriuries as they eat the plants. It is
likely, therefore, that hevein, the potato win genes, and the class i chitinases will
accumulate in response to insects since they are induced in wounding

experiments.
IN VITRO AC‘ITVITIES

Activities due to the Chitin-Binding Domains

Many chitin-binding proteins are made up solely of chitin-binding domains
and activities observed for these proteins can be attributed to these domains.
Besides binding chitin, a number of these proteins agglutinate red blood cells and
are therefore termed lectins. Agglutination of red blood cells requires the proteins
to crosslink the cells by binding to extracellular oligosaccharides. Two binding
sites are required for crossiinking, although a single chitin-binding domain will
potentially bind the same oligosaccharides. Stinging nettle lectin and the
Gramineae lectins, which have 2 and 4 chitin-binding domains respectively, are
able to agglutinate red blood cells. Potato lectin, which also agglutinates blood
cells, should have multiple chitin-binding domains, although no direct evidence for
this is available. All chitin-binding proteins tested which contain a single chitin-

binding domain, however, do not show agglutination activity.

21

Perhaps more significant is the finding that several proteins containing only
chitin-binding domains can inhibit the growth of fungi in vitro. Stinging nettle lectin,
hevein, and the Ac-AMPs have been shown to retard the growth of a number of
chitin-containing fungi (Broekaert et al., 1989; Van Parijs at al., 1991; and Broekaert
et al., 1992 respectively). A striking feature of all these proteins is their small size.
Stinging nettle lectin, the largest of these proteins, is only 8.5 kDa while hevein is
4 kDa and the Ac-AMPs are approximately 3 kDa. The Gramineae lectins, on the
other hand, form dimers of 36 kDa and do not show antifungal activity
(Schlumbaum et al., 1986). It is tempting to speculate that the small size of the
chitin-binding proteins with antifungal activity allows them to penetrate into, and
interact with, the fungal wall where the Gramineae lectins cannot. An estimate of
fungal wall pore size supports this hypothesis (Money, 1990). This study predicts
that proteins larger than 15 to 20 kDa will not be able to pass through the fungal
cell wall.

Several chitin-binding proteins have also been tested for their effect on
insects. WGA, rice lectin and stinging nettle lectin are all able to retard the growth
of cowpea weevil larvae in vitro (Murdock at al., 1990; Huesing, at al., 1991 b).
WGA and rice lectin are similar in activity and two to four times more effective per
mole than stinging nettle Iectin (Huesing at al., 1991a). In addition, WGA has been
shown to be deleterious to two major corn pests, the European corn borer and the
Southern corn borer when added to their diet (Czapla at al., 1991). While low
doses of these lectins don’t kill the insects, they delay larval development time by

approximately 1.5 days per 0.1% (w/w) increase in lectin concentration (Huesing

22

et al., 1991 b). This may be significant since plants and their insect pests often
exist in an equilibrium where a delay in insect development may shift the balance

' in favor of the plant.

Activities due to the carboxyl-Terminal Domains

The chitinase activities observed for the class I chitinases and prostinging
nettle lectin are the most obvious in vitro activities which can be attributed to the
carboxyl-domain of chitin-binding proteins. In addition, the antifungal activity
exhibited by the class I chitinases differs from that observed for stinging nettle
lectin, hevein or Ac-AMPs. Microscopic examination reveals that the fungal hyphal
tips burst in the presence of chitinase while the other chitin-binding proteins slow
the growth of the hyphae and cause a characteristic swelling just behind the
hyphal tip (Broekaert at al., 1989; Broekaert et al., 1992; Van Paris at al., 1991).

No function or activities have been ascribed to the carboxyl-terminal domain
of hevein, win1 or win2. This domain, however, is highly conserved in plants as
divergent as rubber tree, potato, tobacco and tomato suggesting an important
function within the plant.

The Solanaceous lectins have a domain reminiscent of other hydroxyproline
rich glycoproteins such as extensin. These HRGPs are thought to be involved in
strengthening the plant cell wall, especially in response to pathogens (for review
see Bowles, 1990). It is likely that the Solanaceous lectins have similar roles

(Casaiongué and Pont Lezica, 1985).

23
Synergistic Effects

A synergistic enhancement of antifungal activity between stinging nettle
lectin and chitinase is one of the most interesting findings from in vitro activity
experiments of chitin-binding proteins. Broekaert at al. (1989) found that although
stinging nettle lectin and chitinase would inhibit fungal growth individually,
combinations of the two acted cooperatively. The potential for enhancing plant
resistance against fungal pathogens by constitutive expression of class I chitinases
has been addressed by two groups. Broglie and coworkers (1991) found that
tobacco (Nicotiana tabacum cv. Xanthi) and canola (Brassica napus cv. Westar)
expressing a bean chitinase gene under control of the cauliflower mosaic virus 358
promoter had an increased ability to survive in soils infected with Rhizoctania
solani and showed delays in disease symptom development. These transgenic
tobacco plants however were only slightly resistant to Cercospora nicotiana. The
other group (Neuhaus et al., 1991a) found that high levels of a tobacco (N.
tabacum) class i chitinase gene expressed in Nicotiana sylvestris also did not
enhance resistance to C. nicotianae. Thus, these studies are inconclusive as to
the effectiveness of engineering fungal resistance in plants by overexpressing
chitinase alone. Since stinging nettle lectin acts synergistically with chitinase to
inhibit fungal growth in vitro, the possibility exists that overexpression of
prostinging nettle lectin, with its lectin and chitinase domains, will enhance
resistance to fungal pathogens not seen with the chitinase alone. Another
possibility is that stinging nettle lectin will work cooperatively with endogenous

chitinases in the transgenic plant.

24
Ac-AMPs do not act synergistically with either chitinase or stinging nettle

lectin in inhibiting fungal hyphae (Broekaert er al., 1992). Unlike stinging nettle
lectin, however, the Ac-AMPs antifungal activity is strongly inhibited by divalent
cations indicating that these proteins may act in different ways. The chitinases
have also been found to act synergistically with glucanases for fungal inhibition
(Mauch at al., 1988). In this case it is likely that the two enzymes act on different

polysaccharides in the fungal cell wall to give the enhanced antifungal effect.

CONCLUSIONS

The biological function of chitin-binding proteins in plants have yet to be
conclusively demonstrated. There is, however, a wealth of circumstantial evidence
for functions in plant defense. First there is the lack of chitin in plants and its
presence in many plant pathogens and pests. Second, the conservation of amino
acid sequence and chitin-binding activity within this family of proteins indicates an
essential role in the plant’s interactions with organisms containing chitin. Especially
noteworthy is the distinct pattern of Cys residues and conservation of residues
directly involved in chitin-binding. Fusion of the chitin-binding domain to a number
of structurally unrelated domains suggests this domain may be genetically
recombined with other polypeptides by the plant to create new antimicrobial

proteins. The finding that the carboxyl-terminal domain of prohevein is

25

homologous to tobacco and tomato pathogenesis-related proteins suggests
additional activities may be found for chitin-binding proteins.

Localization of the chitin-binding proteins in the plants also suggests they
may interact with plant pathogens. The Gramineae lectins are expressed in the
tissues which first expand into new areas of soil as the embryo germinates and the
root system grows. The intracellular localization of the Gramineae lectins and the
class i chitinases in the vacuole suggests that these proteins act after the invasion
of the plant tissue which releases the contents of the cell. Stinging nettle lectin is
also localized in underground tissues and seeds. Ac-AMPs are also found in
seeds. Similar tissue specific localization of similar proteins in divergent plant
species suggests important roles for these proteins.

Induction of chitin-binding protein expression by wounding, microbial
infection, and the plant ’stress’ hormones also indicates these proteins act as part
of the plant’s defense mechanisms. Chitinases, Gramineae lectins and hevein are
all induced under these conditions. Finally, and perhaps most importantly, are the
in vino antifungal and insecticidal activities shown for chitin-binding proteins. Every

chitin-binding protein tested exhibits insecticidal, antifungal or both activities.

26
REFERENCES

Alien, AK, Desai, MN, and Neuberger, A. (1978). Properties of potato lectin and

the nature of its glycoprotein linkages. Biochem. J. 171:665-674.

Bednarek, S.Y. and Raikhei, NV. (1991). The barley Iectin carboxyl-terminal
propeptide is a vacuolar protein sorting determinant in plants. Plant Cell

3:1195-1206.

Bednarek, S.Y., Wilkins, T.A., Dombrowski, J.E., and Raikhei, NV. (1990). A
carboxyl-terminal propeptide is necessary for proper sorting of barley lectin

to vacuoles of tobacco. Plant Cell 21145-1155.

Bol, J.F., Linthorst, H.J.M. and Cornelissen, B.J.C. (1990). Plant pathogenesis-
related proteins induced by virus infection. Annu. Rev. Phytopathol.

28:113-138.

Boiler, T. (1988). Ethylene and the regulation of antifungal hydrolases in plants.

Oxford Surveys of Plant Mol. And Cell. Biology 52145-174.

Bowles, DJ. (1990). Defense-related proteins in higher plants. Annu. Rev.

Biochem. 5&873-907.

27
Broekaert, W.F., Allen, AK, and Peumans, W.J. (1987). Separation and partial

characterization of isolectins with different subunit compositions from Datura

stramonium. FEBS Lett. 220116-120.

Broekaert, W.F., Lee, H.-I., Kush, A., Chua, N.-H., and Raikhei, NV. (1990).
Wound-induced accumulation of mRNA containing a hevein sequence in
laticifers of rubber tree (Hevea brasiliensis). Proc. Natl. Acad. Sci. USA

87:7633-7637.

Broekaert, W.F., Marién, W., Terras, F.R.G., De Bolle, M.F.C., Proost, P., Van
Damme, J., Dillen, L, Claeys, M., Rees, S.B., Vanderleyden, J., and
Cammue, EPA. (1992). Antimicrobial peptides from Amaranthus caudatus
seeds with sequence homology to the cysteine/glycine-rich domain of

chitin—binding proteins. Proc. Natl. Acad. Sci. USA, submitted.

Broekaert, W.F., Van Paris, J., Leyns, F., Joos, H., and Peumans, W.J. (1989). A
chitin-binding lectin from stinging nettle rhizomes with antifungal properties.

Science 245:1100-1102.

Broglie, K., Chet, l., Holliday, M., Cressman, R., Biddle, P., Knowlton, S., Mauvais,
C.J., and Broglie, R. (1991). Transgenic plants with enhanced resistance

to the fungal pathogen Rhi'mctonia solani. Science 254:1194-1197.

28
Broglie, K.E., Gaynor, J.J., and Broglie, RM. (1986). Ethylene-regulated gene

expression: Molecular cloning of the genes encoding an endochitinase

from Phaseolus vulgaris. Proc. Natl. Acad. Sci. USA 83:6820-6824.

Cammue, B.P.A., Broekaert, W.F., Kellens, J.T.C., Raikhei, N.V., and Peumans,
W.J. (1989). Stress-induced accumulation of wheat germ agglutinin and

abscisic acid in roots of wheat seedlings. Plant Physiol. 91:1432-1435.

Cammue, B.P.A., Broekaert, W.F., and Peumans, W.J. (1990). Wheat germ
agglutinin in wheat seedling roots: induction by elicitors and fungi. Plant

Cell Reports 92264-267.

Casaiongué, C. and Pont Lezica, R. (1985). Potato lectin: A cell-wall glycoprotein.

Plant Cell Physiol. 28:1533-1539.

Chapot, M.P., Peumans, W.J., and Strosberg, AD. (1986). Extensive homologies

between lectins from non-Ieguminous plants. FEBS Lett. 195:231-234.

Chrispeels, M.J. and Raikhei, NV. (1991). Lectins, Iectin genes, and their role in

plant defense. Plant Cell 3:1-9.

29
Czapla, TH. and Lang, BA. (1991). Effects of plant lectins on the larval

development of European corn borer (Lepidoptera: Pyralidae) and
Southern corn rootworm (Coleoptera: Chrysomelidae). J. Econ. Entomol.

83:2480-2485.

Esau, K (1960). Anatomy of seed plants. John Wiley and Sons, Inc., New York,
NY.

Hedrick, S.A., Bell, J.N., Boiler, T., and Lamb, C.J. (1988). Chitinase cDNA cloning
and mRNA induction by fungal elicitor, wounding, and infection. Plant

Physiol. 86:182-186.

Huesing, J.E., Murdock, LL, and Shade, R.E. (1991a). Effects of wheat germ

isolectins on development of cowpea weevils. Phytochemistry 30:785-788.

Huesing, J.E., Murdock, LL., and Shade, RE. (1991 b). Rice and stinging nettle
lectins: Insecticidal activity similar to wheat germ agglutinin.

Phytochemistry 303565-3568.

Lee, H.-I., Broekaert, W.F., Raikhei, NV. (1991). Co-and post-translational
processing of the hevein preproprotein of latex of the rubber tree (Hevea

brasiliensis). J. Biol. Chem. 26621594415948.

30
Legrand, M., Kauffmann, S., Geoffroy, P., and Fritig, B. (1987). Biological function

of pathogenesis-related proteins: Four tobacco pathogenesis-related

proteins are chitinases. Proc. Natl. Acad. Sci. USA 84:6750-6754.

Lerner, DR. and Raikhei, NV. (1989). Cloning and characterization of root-

specific barley lectin. Plant Physiol. 91:124-129.

Lerner, DR. and Raikhei, NV. (1992). The gene for stinging nettle lectin (Urtica
dioica agglutinin) encodes both a lectin and a chitinase. J. Biol. Chem.

submitted.

Linthorst, H.J.M. (1991). Pathogenesis-related proteins of plants. Crit. Rev. Plant

Sci. 10:123-150.

Linthorst, H.J.M., Danhash, N., Brederode, F.T., Van Kan, J.A.L., De Wit, P.J.G.M.,
and Bol, JP (1991). Tobacco and tomato PR proteins homologous to win
and pro-hevein lack the "hevein" domain. Moi. Plant-Microbe Interact.

4:586-592.

Lotan, T., Ori, N., and Fluhr, R. (1989). Pathogenesis related proteins are

developmentally regulated in tobacco flowers. Plant Cell 1:881 -887.

31
Mauch, F., Mauch-Mani, B., and Boiler, T. (1988). Antifungal hydrolases in pea

tissues: ll. Inhibition of fungal growth by combinations of chitinase and 8-

1,3-glucanase. Plant Physiol. 88:936-942.

Mauch, F. and Staehelin, A. (1989). Functional implications of the subceilular
localization of ethylene-induced chitinase and B-1,3-glucanase in bean

leaves. Plant Cell 1:447-457.

Mansfield, MA, Peumans, W.J. and Raikhei, NV. (1988). Wheat-germ agglutinin

is synthesized as a glycosylated precursor. Planta 173:482-489.

Meins, F.Jr. and Ahl, P. (1989). Induction of chitinase and 6-1 ,3-glucanase in
tobacco plants infected with Pseudomonas tabaci and Phytophthora

parasitica var. nicotlanae. Plant Science 81:155-161.

Meins, F.Jr., Neuhaus, J.-M., Sperisen, C., and Ryals, J. (1992). The primary
structure of plant pathogenesis-related glucanohydrolases and their genes.
in: Genes Involved in Plant Defense (Boiler, T. and Meins, F.Jr., eds).

‘ Springer Verlag, Wien-New York. In press.

Mishkind, M.L., Palevitz, B.A., Raikhei, N.V., Keegstra, K. (1983). Localization of
wheat germ agglutinin-like lectins in various species of the gramineae.

Science 220:1290-1292.

32
Money, N. (1990). Measurement of pore size in the hyphal cell wall of Ash/ya

bisexualis. Exp. Mycol. 14:234-242.

Murdock, L.L., Huesing, J.E., Nielsen, S.S., Pratt, RC, and Shade, RE. (1990).
Biological effects of plant lectins on the cowpea weevil. Phytochemistry

29:85-89.

Nachbar, M.S., Oppenheim, JD, and Thomas, JD. (1980). Lectins in the US.
diet. Isolation and characterization of a lectin from tomato (Lycopersicon

esculentum). J. Biol. Chem. 225:2056-2061.

Neuhaus, J.-M., Ahl-Goy, P., Hinz, U., Flores, S., and Meins, F.Jr. (1991a). High-
level expression of a tobacco chitinase gene in Nicotiana sylvestn's.
Susceptibility of transgenic plants to Cercospora nicotiana infection. Plant

Mol. Biol. 16:141-151.

Neuhaus, J.-M., Sticher, L, Meins, F.Jr., and Boiler, T. (1991 b). A short C-terminal
sequence is necessary and sufficient for the targeting of chitinases to the

plant vacuole. Proc. Natl. Acad. Sci. USA 88:10362-10366.

Parsons, T.J., Bradshaw, H.D.Jr., and Gordon, MP. (1989). Systemic
accumulation of specific mRNAs in response to wounding in poplar trees.

Proc. Natl. Acad. SCI. USA 86:7895-7899.

33
Peumans, W.J., Stinissen, HM, and Carlier, AR. (1982). Subunit exchange

between lectins from different cereal species. Planta 154:568-572.

Peumans, W.J., De Ley, M., and Broekaert, W.F. (1984). An unusual lectin from

stinging nettle (Urtica dioica) rhizomes. FEBS Lett. 177:99-103.

Raikhei, N.V., Bednarek, S.Y., and Vlfilkins, TA (1988). Cell-type-specific
expression of a wheat-germ agglutinin gene in embryos and young

seedlings of Triticum aestivum. Planta 178:406-414.

Raikhei, N.V., Hughes, D.W. and Galau, GA (1987). An enzyme-immunoassay for.
quantitative analysis of abscisic acid in wheat. In: Molecular Biology of
Plant Growth Control (Fox, J.E. and Jacobs, M. eds.), Alan R. Liss, New

York, pp 197-207.

Raikhei, N.V., Palevitz, BA, and Haigler, CH. (1986). Abscisic acid control of
lectin accumulation in wheat seedlings and callus cultures. Plant Physiol.

80:167-171.

Raikhei, NV. and Lerner, DR. (1991). Expression and regulation of lectin genes

in cereals and rice. Develop. Genetics. 12:255-260.

Re

Si

Sm

Star

34
Raikhei, NV. and Wilkins, TA (1987). Isolation and characterization of a cDNA

clone encoding wheat germ agglutinin. Proc. Natl. Acad. Sci. USA.

84:6745-6749.

Schlumbaum, A., Mauch, F., vogeli, U., and Boiler, T. (1986). Plant chitinases are

potent inhibitors of fungal growth. Nature 324:365-367.

Shibuya, M, Goldstein, i.J., Shafer, J.A., Peumans, W.F., and Broekaert, W.F.
(1986). Carbohydrate-binding properties of the stinging nettle (Urli'ca

dioica) rhizome lectin. Arch. Biochem. Biophys. 249:215-224.

Shinshi, H., Mohnen, D., and Meins, F.Jr. (1987). Regulation of a plant
pathogenesis-related enzyme: inhibition of chitinase and chitinase mRNA
accumulation in cultured tobacco tissues by auxin and cytokinin. Proc. Natl.

Acad. Sci. USA 84:89-93.

Smith, J.J. and Raikhei, NV. (1989). Nucleotide sequences of cDNA clones
encoding wheat germ agglutinin isolectinsAand D. Plant Mol. Bol. 13:601-

603.

Stanford, A., Beven, M., and Northcote, D. (1989). Differential expression within
a family of novel wound-induced genes in potato. Mol. Gen. Genet.

215:200-208.

35
Stinissen, H.M., Peumans, W.J., and Carlier, AR. (1983). Two-step processing of

in vivo synthesized rice lectin. Plant Mol. Biol. 2:33-40.

Triplett, BA. and Quatrano, RS. (1982). Timing, localization, and control of wheat
germ agglutinin synthesis in developing wheat embryos. Develop. Biol.

91 :491 -496.

Van Parijs, J., Broekaert, W.F, Goldstein, U., and Peumans, W.J. (1991). Hevein:
an antifungal protein from rubber-tree (Hevea brasiliensis) latex. Planta

1 83:258-264.

vogeli-Lange, R., Hansen-Gehri, A., Boiler, T., and Meins, F.Jr. (1988). induction
of the defense-related glucanohydrolases, 8-1,3—glucanase and chitinase,
by tobacco mosaic virus infection of tobacco leaves. Plant Science 54:171-

1 76.

Walujono, K., Scholma, R.A., Beintema, J.J., Mariono, A., and Hahn, AM. (1975).
In Proceedings of the international rubber conference Vol. 2, pp. 518-531.

Rubber Research Institute Malaysia, KuaLa Lumpur.

Wilkins, TA. and Raikhei, NV. (1989). Expression of rice lectin is governed by two
temporally and spatially regulated mRNAs in developing embryos. Plant

Cell 1:541-549.

36
Wright, CS. (1977). The crystal structure of wheat germ agglutinin at 2.2A

resolution. J. Mol. Biol. 111:439-457.

Wright, CS. (1980). Crystallographic elucidation of the saccharide-binding mode
on wheat germ agglutinin and its biological significance. J. Moi. Biol.

141 :267-291 .

Wright, H.T., Sandrasegaram, G., and Wright, CS. (1991). Evolution of a family
of N-acetyiglucosamine binding proteins containing the disulfide rich domain

of wheat germ agglutinin. J. Mol. Evol. 33:283-294.

Zhu, Q. and Lamb, C.J. (1991). Isolation and characterization of a rice gene

encoding a basic chitinase. Mol. Gen. Genet. 226:289-296.

CHAPTER 2

Cloning and Characterization of Root-Specific Bariey Lectin

[ This chapter is reprinted with permission from:
Lerner, DR. and Raikhei, NV. (1989).

Plant Physiology 91 :124-129 ]

37

ABSTRACT

Cereal lectins are a class of biochemically and antigenically related proteins
localized in a tissue-specific manner in embryos and adult plants. To study the
specificity of lectin expression, a barley (Hordeum vulgare L.) embryo cDNA library
was constructed and a clone (BLc3) for barley lectin was isolated. BLc3 is 972
nucleotides long and includes an open reading frame of 212 amino acids. The
deduced amino acid sequence contains a putative signal peptide of 26 amino acid
residues followed by a 186 amino acid polypeptide. This polypeptide has 95%
sequence identity to the antigenically indistinguishable wheat germ agglutinin
isolectin-B (WGA-B) suggesting that BLc3 encodes barley lectin. Further evidence
that BLc3 encodes barley lectin was obtained by immunoprecipitation of the in
vitro translation products of BLc3 RNA transcripts and barley embryo poly A+ RNA.
In situ hybridizations with BLc3 showed that barley Iectin gene expression is
confined to the outermost cell layers of both embryonic and adult roots tips. On
Northern blots, BLc3 hybridizes to a 1.0 kilobase mRNA in poly At RNA from both
embryos and root tips. We suggest, on the basis of immunoblot experiments, that
barley lectin is synthesized as a glycosylated precursor and processed by removal
of a portion of the carboxyl terminus including the single N-linked glycosylation

site.

39
INTRODUCTION

Lectins are a class of proteins with very specific carbohydrate binding
properties. Many of the plant lectins are well characterized in their sugar binding
specificities and, in some cases, the crystalline structure of the protein is known
(Etzler, 1985). In spite of this, the biological significance of plant lectins remains
elusive (Etzler, 1985). The Gramineae lectins all specifically bind
N-acetylglucosamine and are closely related antigenically and biochemically
(Stinissen et al., 1983). These lectins are especially interesting because of their
unique patterns of expression in specific cell layers of embryonic organs and in the
root tips of adult plants (Mishkind et al., 1983). We are investigating how these
specific patterns of lectin accumulation are regulated and what this may indicate
about possible roles of these proteins within the plant.

Lectins from wheat, barley and rye (cereal lectins) are all dimers with 18 kDa
subunits which are synthesized as 23 kDa precursors (Stinissen at al., 1985).
Wheat germ agglutinin (WGA) is the best characterized cereal lectin. Wheat
(Triticum aestivum L.), however, is a hexaploid with each diploid genome
contributing an antigenically indistinguishable isolectin (isolectins A, B and D)
(Stinissen et al., 1983). Functional dimers of WGA isolectins form in vivo by
random association of the isolectin monomers (Peumans et al., 1982). Direct
sequencing of all three isolectin genes has revealed greater than 90% sequence

identity between them (Smith and Raikhei, 1989a). These features of the WGA

40

system have made molecular and cellular studies of individual isolectin expression
particularly difficult.

In this study, we have circumvented the difficulties of the WGA system by
studying barley lectin. Barley, a diploid, contains a lectin shown to be antigenically
indistinguishable from WGA (Stinissen et al., 1983). The lectins are so similar that
active heterodimers containing wheat and barley lectin subunits can be formed in
vitro (Peumans et al., 1982). Barley lectin accumulates in the embryonic and adult
root tips, but unlike in wheat, rye and rice; no lectin is found in the coleoptile
(Mishkind et al., 1983). By studying Iectin expression in barley, we avoid the
possible complications of discerning coleoptile-specific versus root-specific
regulatory elements and differential expression of isolectins. A barley lectin cDNA
clone, BLc3, was isolated from a barley embryo lambda gt10 library. Using this
clone as an in situ hybridization probe we localized lectin mRNA in the embryonic
and adult root tips. We also present evidence that the barley lectin precursor is

glycosylated and undergoes carboxyl terminal processing to produce the mature

polypeptide.

MATERIALS AND METHODS

PlarltMatorIal

Barley (Hordeum vulgare L. var. Betzes) was grown in soil under growth

chamber conditions with a 13 h light cycle (440 uE/mzls) or under greenhouse

41

conditions. Developing grains were harvested at 15 to 25 days post-anthesis
(dpa) and stored at -70°C for RNA isolation or used directly for in situ
hybridization and protein extraction. For isolation of root tips and coleoptiles,
barley grains were surface sterilized with 10% commercial bleach for 20 min, rinsed
with sterile distilled water and germinated on Whatman #1 filter paper over 0.7%

agar for 3 d.

RNA Isolation and Northern Blot Analysis

Total RNA was isolated from developing embryos and root tips (3 to 5 mm)
or whole coleoptiles of 3-d-old seedlings by the method of Finkelstein and Crouch
(Finkelstein and Crouch, 1986). Polyadenylated RNA (poly A+ RNA) was purified
by oligo-deoxythymidine (oligo-dT) cellulose affinity chromatography using the
method of Silflow et al. (Silflow et al., 1979) except that the poly A+ RNA was
eluted at room temperature. For Northern analysis, poly A+ RNAs were separated
electrophoretically, 2 ug per lane, on 2% agarose gels containing 6% formaldehyde
and transferred to Immobilon N (Millipore, Bedford, MA) as previously described
(Raikhei and Wilkins, 1987). The pre-hybridizations and hybridizations were
performed as previously described (Raikhei and Wilkins, 1987) except that the
amounts of SDS and salmon sperm DNA were increased to 0.1% and 250 ug/ml
respectively. The blots were probed with WGA-B cDNA or BLc3 labeled with

a-‘P-ATP by the random primers method (Feinberg and Vogelstein, 1983).

42

CloningandSequenclngacDNAforBarleyLectln

Poly At RNA used as a template for cDNA was prepared as described
(Mansfield et al., 1988). The poly A+ RNA was examined for lectin mRNA by
Northern analysis using a partial cDNA clone for WGA-B (Raikhei and Wilkins,
1987) as the probe. The presence of full-length, translatable barley lectin mRNA
was demonstrated by in vitro translation followed by immunoprecipitation with anti-
WGA antiserum (Mansfield et al., 1988). The cDNA synthesis reaction was primed
with oligo-dT, and the second strand was synthesized using a modification of the
Gubler and Hoffman (1983) method with the Bethesda Research Laboratories
(Gaithersburg, MD) cDNA Synthesis System. The cDNA was ligated into lambda
gt10 (Stratagene, San Diego, CA) with EcoRI linkers (New England Biolabs,
Beverly, MA) and packaged in vitro using Gigapack Gold (Stratagene). Plaques
(5 x 10’) were screened with 32P-labeled WGA-B cDNA at low stringency
hybridization conditions (Vlfilkins and Raikhei, 1989) and positive plaques were
purified at high stringency (Raikhei et al., 1988) with the same probe. Inserts from
purified plaques were subcloned into the EcoRI site of pUC 119 Moira and
Messing, 1987) and sequenced by the dideoxynucleotide chain termination
method (Sanger et al., 1977) using a-“S-dATP in place of a-‘P-dATP and 7-
deaza-dGTP in place of dGTP (Mizusawa et al., 1986). The complete sequence
of both strands of one clone, designated BLc3, was obtained by sequencing
overlapping deletions generated by T4 DNA polymerase (Dale and Arrow, 1987).
Sequence analysis was performed with Microgenie (Beckman, Fullerton, CA) and

Editbase software (courtesy of N. Nielsen, Purdue Univ., West Lafayette, IN).

43
In vitro Translations and lmmunopreclpitatlon of BLc3 RNA Transcripts

To generate RNA transcripts, BLc3 was subcloned into the EcoRI site of
Bluescript KS+ (Stratagene). The construct, designated szBLc3, was linearized
with Xhol or Xbal for sense or antisense RNA transcripts, respectively. For
increased efficiency of translation, ’capped’ transcripts were generated using an
RNA Transcription lfit (Stratagene) according to the manufacturer’s protocol with
the modifications described below. Capping analog, 0.5 mM m7GpppG
(Pharmacia, Piscataway, NJ), and 0.05 mM rGTP were initially used and 2 aliquots
of rGTP were added at 10 min intervals to concentrations of 0.30 mM and 0.55 mM
rGTP, respectively.

Two (:9 of ’capped’ sense transcripts or 10 pg barley embryo poly A+ RNAs
were translated in a rabbit reticulocyte lysate (Promega, Madison, WI) using 50 uCi
ass-methionine (T ranass-label; ICN Biomedicals, Irvine, CA) per reaction. The in
vitro translation products were immunoprecipitated (Hondred et al., 1987) using
anti-WGA antiserum (Mansfield et al., 1988). Samples were carboxyamidated with
2.4 M iodoacetamide at 37°C for 30 min to optimize resolution of the lectins
(Raikhei et al., 1984). Translation products were analyzed by SDS-PAGE on 12.5%

acrylamide gels and visualized by fluorography.

Analysis of Barley Lectin Synthesized In Vim
Barley embryos (300), 15 to 25 dpa, were isolated onto moistened 3MM
paper. Embryos were then incubated in 0.1 mM ABA for 4 h at room temperature

to enhance lectin synthesis (T riplett and Quatrano, 1984). Acid soluble protein was

44
extracted and affinity-purified on immobilized GIcNAc as previously described

(Mansfield et al., 1988). Affinity purified lectin, from 100 embryos, was digested
with 10 mUnits Endo-fl-N-acetylglucaminidase H (Endo H, Calbiochem, San Diego,
CA) at 37°C for 18 h. Samples were lyophilized, carboxyamidated, separated on
SDS-PAGE, as above, and electroblotted onto nitrocellulose (T owbin et al., 1979).
Lectin was detected immunologically with anti-WGA antiserum or anti-WGA-B 172-
186, an antiserum specific for the 15 amino acid pro-peptide at the carboxyl

terminus of pro-WGA (Smith and Raikhei, 1989b).

In Stu Hybridization

For use as in situ hybridization probes, a“‘S-UTP-labeled sense and
antisense RNA transcripts were produced from linearized szBLc3. Labeled
transcripts were partially hydrolyzed with alkali to an average size of 150
nucleotides for increased efficiency of hybridization to mRNA in the tissue sections.
Barley embryos (15 to 25 dpa) and 3-d-old root tips from growing seedlings were
cryosectioned to 8 pm and processed as previously described (Raikhei et al.,

1 988).

45
RESULTS

Isolation and cliaracterhation of barley cDNA clone BLc3.

Eight putative barley lectin clones were isolated from the unamplified barley
embryo cDNA library. The 972 nucleotide sequence for one of these clones,
designated BLc3 (Figure 1), was determined from overlapping sequential deletions.
BLc3 contains a start codon at nucleotides 16-18 initiating a 212 amino acid open
reading frame (calculated mol wt = 21,208 D). Amino acid residues -26 to -1
make up a putative signal sequence (figure 1, broken underline). The cleavage
site for the signal sequence predicted by the method of von Heijne (1986) matches
the amino terminus predicted by sequence identity to mature WGA-B. This
putative signal sequence is followed by a 186 amino acid protein with high
percentages of Cys (17%) and Gly (22%) and low percentages of His (0.5%), Met
(1%), Arg, lle, Phe, Trp, and Val (1.5% each). A single potential site for Asn-linked
glycosylation, Asn-Ser-Thr, is found at residues 206 through 208 (figure 1, marked
with asterisks). The deduced amino acid sequence of BLc3 is 95% identical to
that of WGA-B. Table 1 lists the amino acid differences between BLc3 and WGA-
B. The coding region is followed by two consecutive TGA termination codons
(marked with squares) and a 321 nucleotide 3’ untranslated region. Four putative
polyadenylation signals (Figure 1, underlined) are located at positions 688 and 754
(AATAAT), and at positions 832 and 946 (AATATA). Since an extensive poly At

tail is not found however, the exact 3’ end of the barley lectin mRNA is unknown.

46

Flgure1. Nucleotideanddeducsdamlnoaddsequencecfbarleyleclln
cDNA clone BLc3. The deduced amino acid sequence is from the first
methionine residue and numbered along the right margin. The putative
signal sequence (broken underline) and carboxyl-terminal extension
(double underline) are presumably no present in the mature protein.
The single potential asparagine-linked glycosylation site is designated
with asterisks. Two stop codons at the end of the coding region are
indicated with squares. The four potential polyadenylation signals are
underlined. An extensive poly(A+) tail is not present, so the actual site

of polyadenylation is unknown.

76

151

226

301

376

451

526

601

676

751

826

901

‘47

A Is It ti §. 1. A. A. L A. L. A A A. A 1. L A. E 4A -7
CAGAAAACAAGAAGGATGAAGATGATGAGCACCAGGGCCCTCGCTCTCGGCGCGGCCGCCGTCCTCGCCTTCGCG

A I A A A Q R C G E Q G S N H E C P N N L C C S 19
GCGGCGACCGCGCACGCCCAGAGGTGCGGCGAGCAGGGCAGCAACATGGAGTGCCCCAACAACCTCTGCTGCAGC

0 Y G Y C G M G G D Y C G K G C Q N G A C Y T S K 44
CAGTACGGGTACTGCGGCATGGGCGGCGACTACTGCGGCAAGGGCTGCCAGAACGGCGCCTGCTACACCAGCAAG

R C G T Q A G G K T C P N N H C C S Q N G Y C G F 69
CGCTGCGGCACTCAGGCCGGCGGCAAGACATGCCCTAACAACCACTGCTGCAGCCAGTGGGGTTACTGCGGCTTC

G A E Y C G A G C Q G G P C R A D I K C G S Q A G 94
GGCGCCGAGTACTGCGGCGCCGGCTGCCAGGGCGGCCCCTGCCGCGCCGACATCAAGTGCGGCAGCCAGGCCGGC

G K L C P N N L C C S Q N G Y C G L G S E F C G E 119
GGCAAGCTTTGCCCCAACAACCTCTGCTGCAGCCAGTGGGGTTACTGCGGCCTCGGCTCCGAGTTCTGCGGCGAG

G C Q G G A C S T 0 K P C G K A A G G K V C T N N 144
GGCTGCCAGGGCGGTGCTTGCAGCACCGACAAGCCGTGCGGCAAGGCCGCCGGCGGCAAAGTTTGCACCAACAAC

Y C C S K N G S C G I G P G Y C G A G C Q S G G C 169
TACTGCTGCAGCAAGTGGGGATCCTGTGGCATCGGCCCGGGCTACTGCGGCGCAGGTTGCCAGAGCGGCGGCTGC
'k * *

D G V F A E A I A A N S T L V A i ll -
GACGGTG TGATGATCTTGCTAATGGCAGTAT

TATTGCAACGACGAATAATCCGTGGCAGTTTTGTTGCCACGTACGGTCTCCCTTCACTTACTTTTAGCACTAGTC

 

CTTAATAATTCTCCAGCCTTGCAATATGACGTGCAGGTTGCTACATGCATGGACATATTGCAGTGAGAAGTACTG

 

TGTGGCAATATAGGGTGTACTATTGTTGCCACAAATTTAGTTCTTTCTTGTTACGTACGTACAGTTGTCAGGATG

CATGCATCCCCGTTGTAATGTTGGAGTACTCCATGATTTCGTTGCAAIAIATATATTGCCATGAGTCTAAAG

48

Table 1. Differences in deduced amino acid sequence between barley lectin
and WGA-B

 

 

 

AMINO ACID BLc3 WGA-B
POSITION
C 'v I . I'

41 Tyr TH)

48 ' Thr Ser

64 Tip Tyr
1 39 Lys Arg
1 79 Ala Thr
1 84 Val Leu

N n- n rv iv i i n
9 Asn Gly

66 Tyr His

1 23 Gly Asn

135 Ala A50

49
To verify that BLc3 encodes barley lectin, BLc3 RNA transcripts and barley

embryo poly At RNA were each translated in vitro. The products were then
immunoprecipitated with anti-WGA antiserum and resolved on SDS-PAGE. In vitro
translation of BLc3 RNA transcripts produced a protein of M, 21 kDa (figure 2, lane
1). A M, 21 kDa polypeptide was also specifically immunoprecipitated from in vitro
translation products of embryo poly A+ RNA (figure 2, lane 2). These M,’s agree
well with the mol wt of 21.2 kDa calculated from the deduced amino acid

sequence.

Post-translational Modifications of Barley Lectin

To investigate the in vivo synthesis of barley lectin, Western blots of affinity
purified lectin from developing barley embryos were probed with anti-WGA
antiserum. Affinity-purified barley lectin contained two polypeptides of M, 18 kDa
and M, 23 kDa (figure 3, lane 1). Mature barley lectin has the same mobility as
purified WGA (M, 18 kDa; figure 3, lane 2). The M, 23 kDa protein is most likely
the barley lectin precursor (Stinissen et al., 1985). In vivo labeling studies with
barley embryos also show a M, 23 kDa band after immunoprecipitation with anti-
WGA antiserum (data not shown). In addition, pulse labeling studies in wheat
have shown that WGA is also synthesized as a M, 23 kDa precursor (Mansfield et
al., 1988). Based on the 95% amino acid sequence identity with WGA-B and the
evidence presented above, we will refer to the M, 23 kDa form as the barley lectin
precursor. The barley lectin precursor migrates more slowly, M, 23 kDa, on SDS-

PAGE than predicted from the deduced amino acid sequence alone (21.2 kDa).

FIGURE2. lnvltrotranslatlonand Irnmunopreclpltallonanalyslsofpoly
At RNA and BLca transcripts. Poly At RNA isolated from 15 to 25 dpa

developing embryos (lane 1) and BLc3 RNA transcripts (lane 2) were
translated in vitro using rabbit reticulocyte lysate and 35S-methionine.
Translation products were immunoprecipitated with anti-WGA antiserum,
separated on SDS-PAGE and visualized with fluorography. A single
product with M, 21 kDa was immunoprecipitated in each case indicating

BLc3 encodes the barley lectin.

 

51

 

52
Since the polypeptide deduced from the clone BLc3 includes the only potential

glycosylation site at the carboxyl-terminus, we investigated whether this
glycosylation site was utilized. Affinity-purified protein from developing barley
embryos was treated with Endo-B-N-acetyiglucaminidase H. Endo H will
specifically cleave high mannose oligosaccharides linked to Asn residues. The
smaller size of a protein after Endo H digestion would confirm the presence 3 high
mannose, N-linked glycan. In this experiment we used an antiserum specific for
the carboxyl-terminal portion of pro-WGA, anti-WGA-B 172-186 (Smith and Raikhei,
1989b). Binding of anti-WGA-B 172-186 to pro-barley lectin was expected since
there are only 2 conservative amino acid differences between the pro-peptide of
WGA-B and the last 15 residues encoded by BLc3 (Table I). Anti-WGA-B 172-186
detected the M, 23 kDa precursor band but failed to recognize mature barley Iectin
in the sample (figure 3, lane 3). This provides further evidence that the M, 23 kDa
band represents pro-barley lectin. Endo H digestion of affinity-purified barley Iectin
reduced the size of pro-barley lectin by M, 3 kDa (fig. 3, lane 4), indicating the

presence of a high-mannose oligosaccharide.

Cellular Localization and Temporal Expression of Barley Lectin

The spatial distribution of barley lectin mRNA was determined by in situ
hybridization with BLcS antisense RNA transcripts. Barley lectin mRNA was
localized to the coleorhiza, outer cell layers of the radicles, and the root caps of
the developing embryo (figure 4a and b). Lectin mRNA was also found in the root

tip and root cap of 3-d-old seedlings (figure 4c and d). Lectin mRNA was not

53

FIGURES. Wememblotanalyslsofnatlveand EndoHtreatedbarley
Iectin. Isolated barley embryos, 15 to 25 dpa, were treated with 0.1 mM
ABA (4 h) to enhance Iectin expression. Barley Iectin was affinity
purified from acid extracted protein and resolved on SDS-PAGE prior to
transfer onto nitrocellulose. Western blots were probed with either anti-
WGA antiserum, lanes 1 and 2; or anti-WGA-B 172-186, lanes 3, 4 and
5. Anti-WGA-B 172-186 is an antiserum specific for the pro-peptide of
pro-WGA. Barley Iectin has a M, 23 kDa putative precursor and a M, 18
kDa mature form, lane 1. Commercial WGA, lane 2, contains only the
mature lectin, M, 18 kDa. Anti-WGA-B 172-186 detects only the M, 23
kDa pro-barley lectin, lane 3. Treatment for 18 h at 37 °C with Endo H
changes the M, of pro-barley lectin to M, 20 kDa, lane 4. Anti-WGA-B
172-186 does not detect commercial WGA since no pro-WGA is present,

lane 5.

 

 

54

 

4

5

55

FIGURE 4. Localization of bariey Iectin mRNA by In situ hybridization.
Barley embryos, 15 to 25 dpa, and root tips from 3-d-old seedlings were
cryosectioned to 8 um and probed with BLc3 antisense RNA transcripts.
Silver grains developed in the autoradiographic emulsion appear as
bright areas with darkfield optics. Phase contrast micrograph of
developing embryo, panel A, shows the coleorhiza (C), radicles (R), and
embryonic root cap (RC). Darkfield micrograph of the same section,
panel B, localizes barley Iectin mRNA in the cells of the coleorhiza, the
outer cell layer of the radicle and the root cap. Phase contrast, panel
C, and Darkfield, panel D, micrographs of root tips from germinating
seedlings show specific hybridization of the probe to the root tip and

particularly the root cap (RC). Scale bar, 50 um. Magnification, 400x.

 

56

 

57

detected in the primordial leaves, coleoptile or scutellum of the embryo (data not
shown). Sense BLcS RNA transcripts, used to monitor non-specific binding of
labeled nucleic acids to the sections, did not bind significantly to any tissue (data
not shown).

To determine if barley lectin mRNAs from embryos and adult roots were the
same size, Northern blot analysis was performed (figure 5). A 1.0 kb mRNA was
detected in poly A+ RNA from both tissues (figure 5, lanes 1 and 2). No
detectable lectin mRNA was found in coleoptiles of 3-d-old seedlings (figure 5, lane

3).

DISCUSSION

Our goal is to gain an understanding of the mechanisms controlling the
specificity of expression observed in the cereal Iectins. Previous work shows that
de novo synthesis of both lectin mRNA (Raikhei et al., 1988) and protein (Raikhei
et al., 1984) is responsible, at least in part, for the pattern of accumulation seen for
WGA expression. These data suggest that transcriptional control accounts for
some of the observed specificity. The pattern of lectin expression found in cereals
is species specific. However, only barley Iectin is expressed solely in the adult and
embryonic roots. As an initial step in understanding this root-specific expression,

a cDNA clone for barley lectin, BLc3, was isolated and characterized.

58

FIGURE 5. Northern analysis of poly At mRNA from root tips and
coleoptiles of 3-d-old seedlings and 15 to 25 dpa barley embryos. Poly
A” RNA was separated on a formaldehyde/agarose denaturing gel,
immobilized on nitrocellulose and hybridized at high stringency with 3“P-
labeled cDNA clone BLc3. BLc3 hybridizes to a 1.0 kb mRNA from both
embryos, lane 1, and root tips, lane 2. No hybridization to coleoptile

poly A+ RNA, lane 3, was observed.

1 Rb-

59

60
ComplementaryDNACloneBLcSEncodesBarieyLectin

BLcS was shown to encode barley Iectin by in vitro translation experiments
followed by immunoprecipitation of the products. As shown previously, barley
lectin and WGA are immunologically indistinguishable (Stinissen et al., 1983).
Thus, anti-WGA antiserum should immunoprecipitate in vitro translation products
of BLcS. Our results show a M, 21 kDa polypeptide was immunoprecipitated by
anti-WGA antiserum. These data were supported by in vitro translation and
immunoprecipitation of barley embryo poly At RNA. Here, a single M, 21 kDa
band was also immunoprecipitated by anti-WGA antiserum. The identical M, of the
immunoprecipitated products from both sources indicates that BLc3 probably
contains the entire coding region of barley Iectin.

Analysis of the amino acid sequence encoded by BLc3 provides further
evidence that BLc3 encodes barley Iectin. The amino acid composition, rich in Gly
and Cys while poor in several other amino acids, is characteristic of the cereal
Iectins (Peumans et al., 1982). in addition, there were only 10 differences (95%
sequence identity) between WGA-B and the deduced amino acid sequence of
barley lectin (Table I). Six of these differences were conservative substitutions
(Microgenie, Beckman); making the structural similarity even greater. The striking
sequence identity found between BLc3 and WGA-B explains the immunological
similarity (Stinissen et al., 1983) and the agglutinating activity of WGA/barley lectin
heterodimers (Peumans et al., 1982).

The translated sequence of BLcs is given from the first methionine codon.

It is unknown, however, which of the initial methionine residues (~26, -24 or -23) is

61

used to initiate translation in vivo. The coding region of BLc3 begins with a typical
tripartite signal sequence (residues -26 to -1) characteristic of secretory proteins.
This signal sequence was expected in a full length clone since previous studies
have localized cereal and rice lectins to the vacuoles/protein bodies (Mansfield et
al., 1988; Mishkind et al., 1982; Stinissen et al., 1985). The predicted cleavage site
(von Heijne, 1986) for the signal sequence corresponds exactly to the amino
terminus of mature WGA. These data support the hypothesis that Gln #1 is the
amino-terminus of the mature barley lectin although the actual terminus is

unknown.

Glycosyiation and Cleavage of a Propeptlde from Probarley Lectin

The results presented in this paper showed that the precursor for barley
lectin (M, 23 kDa) is larger than predicted from the cDNA sequence (mol. wt. 21.2
kDa). Pro-barley Iectin was found to be Endo H sensitive and therefore
glycosylated with a high mannose glycan. This glycan would account for only part
of the additional size of the precursor. In these experiments a polyclonai
antiserum, anti-WGA-B 172-186, specific for pro-WGA was used (Smith and
Raikhei, 1989b). Anti-WGA-B 172-186 specifically recognized pro-barley lectin and
deglyCosylated pro-barley lectin, but did not bind mature barley lectin. This makes
anti-WGA-B 172-186 an especially powerful tool for investigating modifications of
the carboxyl-terminal end of barley Iectin. The results of these experiments allow
us to tentatively assign the carboxyl-terminus of mature barley Iectin as Gly #171,

although the actual terminal residue is unknown. Furthermore, mature barley lectin

62
has the same M, as WGA on SDS-PAGE and the region surrounding the carboxyl-

terminus of mature WGA is identical in barley Iectin (Table I). Thus, based on our
results with anti-WGA-B 172-186 and sequence identity with WGA-B, the carboxyl
terminal portion of the barley lectin precursor (double underlined in figure 1) is
probable absent in mature barley lectin. WGA (Mansfield et al., 1988), rice lectin
(Wilkins and Raikhei, 1989) and fi-glucanase (Shinshi et al., 1988), have also been
shown to be synthesized as glycosylated precursors and undergo carboxyl-

terminai processing of the polypeptide.

Temporal and Cellular Localization of Barley Lectin

In situ hybridization experiments show barley lectin mRNA is localized to the
root tip of the adult plant and the analogous structures in the embryo. As might
be expected, this pattern of expression coincides with that for lectin accumulation
(Mishkind et al., 1983). WGA-B mRNA shows a similar pattern of expression
(Raikhei et al., 1988), however, recent data showing greater than 90% identity
between wheat isolectin mRNAs (Smith and Raikhei, 1989a) will make precise
analysis of individual isolectin expression difficult. Furthermore, the complicated
pattern of WGA accumulation in different genotypes of wheat remains unexplained
(Raikhei eta/"1986). The barley Iectin system, devoid of isolectin complications,
is therefore more amenable to the study of root tip-specific protein expression.
The cDNA for barley Iectin presented in this paper will provide a valuable tool for
the isolation of gene promoter sequences for barley Iectin and characterization of

the cis-elements involved in root-tip-specific expression.

63
REFERENCES

Dale, R.M.K. and Arrow, A. (1987). A rapid single-stranded cloning, sequencing,

insertion, and deletion strategy. Methods Enzymol. 155:204-214.

Etzler, ME. (1985). Plant lectins: molecular and biological aspects. Annu. Rev.

Plant Physiol. 38:209-234.

Feinberg, AP. and Vogelstein, B. (1983). A technique for radiolabelling DNA
restriction endonuclease fragments to high specific activity. Anal. Biochem.

132:6-13.

Finkelstein, RR. and Crouch, ML. (1986). Rapeseed embryo development in
culture on high osmoticum is similar to that in seeds. Plant Physiol. 81 :907-

912.

Gubler, U. and Hoffman, B.J. (1983). A simple and very efficient method for

generating cDNA libraries. Gene 25: 263-269.

Hondred, D., Wadle, D.-M., Titus, DE. and Becker, WM. (1987). Light stimulated
accumulation of the peroxisomal enzymes hydroxypyruvate reductase and
serine: glyoxylate aminotransferase and their translatable mRNAs in

cotyledons of cucumber seedlings. Plant Mol. Biol. 9:259-275.

64
Mansfield, MA, Peumans, W.J. and Raikhei, NV. (1988). Wheat-germ agglutinin

is synthesized as a glycosylated precursor. Planta 173:482-489.

Mishkind, M.L., Raikhei, N.V., Palevitz, BA. and Keegstra, K. (1982).
immunocytochemicai localization of wheat germ agglutinin in wheat. J. Cell

Biol. 92:753-764.

Mishkind, M.L., Palevitz, B.A., Raikhei, NV. and Keegstra, K. (1983). Localization
of wheat germ agglutinin-like lectins in various species of the gramineae.

Science 220:1290-1292.

Mizusawa, S., Nishimura, S. and Seela, F. (1986). Improvement of the dideoxy
chain termination method of DNA sequencing by use of deoxy-7-
deazaguanosine triphosphate in place of dGTP. Nucl. Acids Res. 14:1319-

1 324.

Peumans, W.J., Stinissen, H.M., Carlier, AR. (1982). Isolation and partial
characterization of wheat-germ-agglutinin-like lectins from rye (Secale

cereals) and barley (Hordeum vulgare) embryos. Biochem. J. 203:239-243.

Peumans, W.J., Stinissen, HM. and Carlier, A.R. (1982a). A genetic basis for the

origin of six different isolectins in hexaploid wheat. Planta 154:562-567.

65
Peumans, W.J., Stinissen, HM. and Carlier AR (1982b). Subunit exchange

between lectins from different cereal species. Planta 154:568-572.

Raikhei, N.V., Bednarek, S.Y. and Lerner, DR. (1988). In situ RNA hybridization
in plant tissues. In: Plant Molecular Biology Manual (Geivin, SB. and
Schilperoort, R.A., eds), Section B9, Kluwer Academic Publishers,

Dordrecht, The Netherlands, pp 1-32.

Raikhei, N.V., Bednarek, S.Y. and Wilkins, TA. (1988). Ceil-type-specific
expression of a wheat-germ agglutinin gene in embryos and young

seedlings of Triticum aestivum. Planta 176:406-414.

Raikhei, N.V., Mishkind, ML. and Palevitz, BA. (1984). Characterization of a wheat

germ agglutinin-like lectin from adult wheat plants. Planta 162:55-61.

Raikhei, N.V., Palevitz, BA. and Quatrano, RS, (1986). Pattern of wheat germ
agglutinin accumulation in different genotypes of wheat. In: Lectins (Bog-
Hansen, TC. and van Driessche, E. eds). vol V, Walter de Gruyter & Co.,

Berlin, pp 75-81.

Raikhei, NV. and Wilkins, TA. (1987). Isolation and characterization of a cDNA
clone encoding wheat germ agglutinin. Proc. Natl. Acad. Sci. USA

84:6745-6749.

66
Sanger, F., Nicklen, S. and Coulson, AR. (1977). DNA sequencing with chain-

terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467.

Shinshi, H., Wenzler, H., Neuhaus, J.-M., Felix, G., Hofsteenge, J. and Meins, F.
(1988). Evidence for N- and C- terminal processing of a plant defence-
related enzyme: Primary structure of tobacco prepro-fl-1,3-giucanase.

Proc. Natl. Acad. Sci. USA 85:5541-5545.

Silflow, C.D., Hammett, JR. and Key, J.L (1979). Sequence complexity of
polyadenylated ribonucleic acid from soybean suspension cultured cells.

Biochem. 18:2725-2731.

Smith, J.J. and Raikhei, N.V. (1989a). Tools to study the post-translational

modification of WGA. Plant Physiol. 896:102.

Smith J.J. and Raikhei, N.V. (1989b). Production of an antibody specific for the

propeptide of wheat germ agglutinin. Plant Physiol. 91:473-476.

Stinissen, H.M., Chrispeels, M.J. and Peumans, W.J. (1985). Biosynthesis of lectin

in roots of germinating and adult cereal plants. Planta 164:278-286.

67

Stinissen, H.M., Peumans, W.J., Carlier, AR. (1983). Occurrence and
immunological relationships of lectins in gramineous species. Planta

159:105-111.

Towbin, H., Staehelin, T. and Gordon, J. (1979). Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and

some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.

Triplett, BA. and Quatrano, RS. (1984). Timing, localization and control of wheat
germ agglutinin synthesis in developing wheat embryos. Develop. Biol.

91 :491 -496.

Vieira, J. and Messing, J. (1987). Production of single-stranded plasmid DNA

Methods Enzymol. 15323-11.

von Heijne, G. (1986). A new method for predicting signal sequence cleavage

sites. Nucl. Acids. Res. 14:4683-4690.

Wilkins, TA. and Raikhei, NV. (1989). Expression of rice lectin is governed by two
temporally and spatially regulated mRNAs in developing embryos and

young seedlings. Plant Cell 12541-549.

CHAPTER 3

TheGeneforStlngingNettIeLectln(UrtlcedoIceAgqutlrfln)
EncodesbcthaLectlnandaChltlnase.

[ The text and figures in this chapter have been
submitted for publication to the Journal of Biological

Chemistry ]

69
ABSTRACT

Chitin-binding proteins are present in a wide range of plant species,
including both monocots and dicots, even though these plants contain no chitin.
To investigate the relationship between in vitro antifungal and insecticidal activities
of chitin-binding proteins and their unknown endogenous functions, the stinging
nettle lectin (Urtica dioica agglutinin, UDA) cDNA was cloned Using a synthetic
gene as the probe. The nettle lectin cDNA clone contained an open reading frame
encoding 374 amino acids. Analysis of the deduced amino acid sequence
revealed a 21 amino acid putative signal sequence and the 86 amino acids
encoding the two chitin-binding domains of nettle Iectin. These domains were
fused to a 263 amino acid carboxyl extension with partial identity to a chitinase
catalytic domain. The authenticity of the cDNA clone was confirmed by deduced
amino acid sequence identity with sequence data obtained from tryptic digests of
the stinging nettle lectin, RNA gel blot and PCR analyses. RNA gel blot analysis
also showed the nettle lectin message was present primarily in rhizomes and
inflorescence (with immature seeds) but not in leaves or stems. Chitinase
enzymatic activity was found when the chitinase-like domain alone or the chitinase-
like domain with the chitin-binding domains were expressed in E. coli. This is the
first example of a chitin-binding protein with both a duplication of the 43 amino
acid chitin-binding domain and a fusion of the chitin-binding domains to a

structurally unrelated domain, the chitinase domain.

70
INTRODUCTION

Chitin, a polymer of l3-1,4-N-acetylglucosamine, is found in the cell wall of
many fungi, the exoskeleton and digestive tract of some insects and in some
nematodes. It is curious then that plants, which contain no chitin, express a family
of chitin-binding proteins with a conserved chitin-binding domain. These plants
include both monocots and dicots ranging from wheat and barley to tobacco and
rubber trees. Chitin-binding proteins are secretory proteins which may be involved
in plant defense (see Chrispeels and Raikhei, 1991 for review). Several genes and
proteins from this family have been isolated and characterized. Some of them, for
example the Gramineae lectins, have been shown to have insecticidal activity by
in vitro experiments (Huesing et al., 1991 a). Others, such as the class Ichitinases
and hevein, possess antifungal activity in in vitro experiments (Van Parijs et al.,
1991). Recently, a chitin-binding lectin isolated from the rhizomes of stinging nettle
(Urtica dioica), Urfica dioica agglutinin (UDA), has been shown to possess both
antifungal and insecticidal activities (Broekaert et al., 1989 and Huesing et al.,
1991 b, respectively).

The rhizomes of stinging nettle serve as underground storage tissues and
as a source of vegetative meristems for regeneration of shoots. For both these
functions rhizomes would require mechanisms for protection against soil
pathogens. Nettle lectin has been found to accumulate to high levels (1 gm/kg)
in the rhizomes (Peumans et al., 1984a). This small, 8.5 kDa protein is made up

of two 43 amino acid glycine and cysteine rich domains, possesses several

71

isoforms (Van Damme et al., 1988) and has been shown to have homology to the
chitin-binding domains of other proteins (Chapot et al., 1986).

In this paper we report the isolation of a cDNA clone encoding stinging
nettle lectin using a synthetic gene as the probe. The cDNA clone was found to
encode a putative signal sequence adjacent the two nettle Iectin chitin-binding
domains in the amino-terminal portion of the deduced amino acid sequence. In
addition, a long and unexpected carboxyl-terminal domain with partial identity to
a chitinase catalytic domain was observed. The validity of the clone was
addressed by several approaches and the predicted properties of both domains

were analyzed by expression in E. coli.

MATERIALS AND METHODS

Plant Material

Urtica dioica plants were grown under standard greenhouse conditions. U.
dioica seed was from Dr. W.J. Peumans (Kathoiieke Univ. Leuven, Belgium) and
Egessa (FRG). Rhizomes were generated by cutting stems off at the base of the
plant, burying these stems under 1 to 2 cm of soil, and growing them under
standard greenhouse conditions. These rhizomes were tested for the presence
of stinging nettle lectin by agglutination of trypsin-treated rabbit erythrocytes

(Peumans et al., 1984b).

72
General Molecular Biology Methods

Unless stated otherwise, all general molecular biology methods were done
as per Molecular Cloning: A Laboratory Manual (Maniatis et al. 1982). Chemicals
were from Sigma (St. Louis, MO) and restriction enzymes were from Boehringer

Mannheim (Indianapolis, IN).

RNA isolationsand RNAGeI BlotAnalysls

Rhizome and root tissue was washed free of soil, blotted dry, frozen in
liquid N2 and stored at -80° C. Frozen tissue was ground in liquid N2, mixed with
2.6 ml/gm extraction buffer [0.2 M Tris-HCI pH 9.0, 0.4 M NaCl, 0.025 M EGTA, 1%
SDS, 0.5% PVP-40) and ground further with a poiytron for 4 min. Proteinase K (0.5
mg/ml) was added and incubated at 37°C with constant mixing for 60 min. The
extract was then mixed with BaCl (0.075 M) and KCl (0.035 M), incubated on ice
15 min, centrifuged at 16,000xg for 10 min at 4°C and the supernatant filtered
through Miracloth (Calbiochem, San Diego, CA). RNA in the supernatant was
precipitated with 2 M LiCl at -20°C overnight and collected by centrifugation at
5,000xg for 20 min at 4°C. The pellet was resuspended in TE (10 mM Tris-HCI pH
8, 0.1 mM EDTA), the RNA reprecipitated with sodium acetate and ethanol and
stored at -20° C. Poly(A)+RNA was isolated from total RNA with oligo-dT
sepharose (Silflow et al., 1979). Total RNA from all other tissues was isolated by
the method of Nagy et al. (1988).

RNA gel blot analysis was performed by separating 20 ug total RNA per

lane and 5 ul RNA ladder (BRL, Gaithersburg, MD) as size standards on a 2%

73
agarose/formaldehyde gel as previously described (Raikhei and Vlfilkins, 1987).

32P-Iabelled probes were prepared with random primers (Feinberg and Vogelstein,
1983), the unincorporated nucleotides were removed with Nuctrap columns
(Stratagene, San Diego CA). Medium stringency washes were done with 2x SSC
with 0.1% SDS at 65° C. High stringency washes were done with 0.2x SSC with

0.1% SDS at 65°C.

Design ofthe SynthetIcGeneforStlnging Nettle Lectin

The amino acid sequence of stinging nettle Iectin as determined by Chapot
et al. (1985) and Steven Michnick (at Harvard University, personal communication)
was used as a template for designing the synthetic gene. Since the synthetic
gene was intended for expression in transgenic tomato, codon usage was derived
primarily from a consensus cf tomato genes (Genebank, version 59, 3/89).
Codons from highly conserved amino acid residues found in other chitin-binding
proteins were also used. In addition, where four G residues in a row were
unavoidable, such as tryptophan (T 66) next to glycine (GGX), the guanidine
residues were split between two adjacent oligonucleotides since these sequences
are difficult to synthesize accurately (Gait et al., 1980). Complementary pairs of
oligonucleotides were designed to contain one 5’ and one 3’ eight base overhang.
This greatly reduced the potential for mispaired ligations compared to pairs with
two 5’ or 3' overhangs. To facilitate cloning, restriction endonuclease recognition
sites were included in the synthetic gene flanking the coding region for stinging

nettle lectin. In addition, three extraneous base pairs were added to each end of

74

the synthetic gene to prevent exonuclease activity during construction from

obliterating restriction sites at the extreme ends of the gene.

Synthetic Gene Construwon

Oligonucleotides were synthesized using fi-cyanoethyl chemistry on an
Applied Biosystems DNA synthesizer (model 3803, Foster City, CA) by the
Macromolecular Structure Facility (Michigan State Univ., East Lansing, MI).
Purification of the oligonucleotides was by gel electrophoresis (Sambrook, et al.
1989) or HPLC using a Zorbax Bioseries oligo column (Dupont, Wilmington, DE).
Purified oligonucleotides were desalted with SEPAK C-18 columns Meters
Associates, Bedford, MA) as recommended by the manufacturer.

One nmol of each purified oligonucleotide, except those at the terminal 5’
ends of the synthetic gene, were phosphorylated with T4 polynucieotide kinase
(New England Biolabs, Beverly, MA) (Sambrook et al., 1989) and separated from
free nucleotides with Nuctrap push columns (Stratagene, La Jolia, CA), using
10 mM Tris, pH 8.0.

Complementary pairs of oligonucleotides, 200 pmol each, were annealed
by heating to 80°C then cooling slowly to room temperature. The annealed pairs,
100 pmol each, were then ligated in sequential reactions (Figure 1). Initially, the
outermost pairs of annealed oligonucleotides were mixed with the adjacent pairs
in separate reactions which included 60 units T4 DNA Iigase (New England
Biolabs), 200 mM Tris-HCI (pH 7.6), 50 mM MgCl2, 50 mM dithiothreitoi (DTT) and

500 ug/ml bovine serum albumin (BSA) and then incubated for 30 min at 15° C.

75

.m_m>_mcm Scenes.» .8 mp 635 85 pound: can scan 65 So once 65 is: oocoEmoa mega Eoom um
“332.59 mm; econ 255:5 sauce. :2 9F E265 mm scam 0:65;» 65 Co mace e5 :5: “32mm: >=£Eo=oem
e52, 325263595 .6 mafia 66.352 .2288. 3938 E uefieccm 99$ .65 .54 + ms. .32 + E. .6:
325203596 poufltocomoza outta Co 255 EmucoEoEEoU .aopzosoEogo 255:: 9 Set econ 580.

0.22. 9.553 0.35:5 o5 cos—coco 8 no»: «cocoa! c339. 32.5.53 65 .o Esau... ozoEocom .9 2:2“.

SYNTHETIC STINGING NETTLE LECTIN GENE UGATION SCHEME

3: 2 2:": £2":
—>
+ 2|: all:
—r
.— — + 2|:
—c
2|: .,.
:Ifi
LI.
4.
ID 0
”I' + I
—c
+ ”I

P

 

 

76

G I? l?
-c
N II“ II.N
'- v-

mllo
'- v-
“ C
v- 1-
sls II
1- v-
” *
v- 1-
v- v-
I- IN
'- v-
+
o s
+
d” ”
.. I.
al.
“II.

77

For subsequent ligation reactions the next adjacent pairs of annealed
oligonucleotides were added to the previous reaction along with 30 units T4 DNA
ligase. Tris, MgCla, DTT, and BSA were also added to bring concentrations to
initial values and the reactions incubated for 30 min at 15°C. Finally, the two
halves of the synthetic gene were ligated together with an additional 60 units of T4
DNA ligase for 2.5 h at 15°C. The ligation reactions were stopped by heating to
70°C and the products precipitated with sodium acetate and ethanol. To assess
the reaction products 10% of the reaction mix was removed at the end of each
ligation step, end labeled with y-‘QP-ATP, resolved on a 12.5% non-denaturing
polyacrylamide gel, and visualized with autoradiography (data not shown). The
synthetic gene was phosphorylated as described above for ligation to the cloning
vector, pUC119.

Restriction digestion by EcoRI of the synthetic gene and pUC119 vector
were performed using Boehringer Mannheim enzymes following the manufacturers
instructions. Vector dephosphorylation reactions were performed with calf
intestinal phosphatase (Sigma, St. Louis, M0) for 30 min at 55° C. The synthetic
stinging nettle lectin gene (approximately 2.5 pmol) and vector (0.25 pmol) were
ligated as above with 100 units Iigase for 12 h at 15° C. The ligation products were
cloned by standard methods and the sequence of the synthetic gene confirmed

by dideoxy chain termination sequencing (Sanger et al., 1977).

78
Compiernentary DNA Ubrary Construction and Screening

A stinging nettle rhizome lambda gt10 cDNA library was constructed with
U. dioica rhizome poly(A)+RNA and EcoRI adapters (Promega, Madison WI) as
previously described (Lerner and Raikhei, 1989). 5 x 105 plaque forming units
(pfu) of the amplified library (1.5 x 105 original pfu) were screened at medium
stringency (2X SSC + 0.1% SDS at 65 °C) with random-primer labelled synthetic
stinging nettle lectin gene. Positive plaques were purified to homogeneity and the
lambda DNA isolated. Due to loss of the EcoRI adapter sites during construction
of the library, insert DNA was isolated by PCR with primers just flanking the EcoRI
site of lambda gt10 (5’-AGCAAGTTCAGCCTGGTTAA-3’ and
5’-TTATGAGTTATTTCTTCCAGG-3’). PCR products were phosphorylated with T4
polynucieotide kinase and ligated into the Smal site of pUC119 for cloning and
sequence analysis. The complete sequence of cDNA clones was obtained with
dideoxy chain termination sequencing of deletions generated by the method of
Dale and Arrow (1987).

Independent isolates of nettle Iectin cDNAs were obtained from a separate
and unamplified nettle rhizome cDNA library prepared as described above. A
random primer a‘iP-labeled uda1 fragment (nt #267 to 1317) was used to probe
approximately 5 x 105 plaques. Plaques were purified and insert cDNAs partially

sequenced as described above.

79
Polymerase Chain Reaction DNA Amplification

Amplification reactions were carried out with deoxynucleotides, buffers and
enzyme concentrations as recommended by the enzyme manufacturer. Taq
polymerase (Amplitaq, Perkin Elmer/Cetus, Norwalk, CT) was used for amplifying
inserts from the lambda gt10 library. Replinase (NEN/Dupont, Wilmington, DE)
was used for amplification of reverse transcribed RNA (see below). Reactions
were carried out on a Perkin-Elmer thermocycler with an initial denaturation step
of 94°C for 4 min followed by 30 cycles of 94°C for 1 min, 65°C for 2 min, 72°C
for 3 min. A final polymerization step of 72°C for 7 min was added after the 30
cycles. Reaction products were purified from unincorporated nucleotides with PCR
purification columns (Qiagen, Studio City, CA). For subcloning, the reaction
products were size fractionated on 1% low melting point agarose gels (SeaKem

LE, FMC, Rockland, ME).

Reverse Transcription with PCR

First strand cDNA was generated from 1 ug U. dioica total rhizome RNA
using 10 ng oligo-dT primer and the cDNA synthesis kit from BRL (Gaithersburg,
MD) following the manufacturers instructions in a 20 ul reaction. First strand cDNA
was then used as the template for the PCR with 5 ul reverse transcription reaction
per 25 ul PCR reaction. Primers for the PCR were identical or complementary to
the stinging nettle lectin cDNA clone as specified: #1: 5’-TCTGCCGTAGTGAT-

CATG-3' (nt #40 to 57), #2: 5’-AGCGGTACTGGCATTTGC-3’ (nt #348 to 329),

80
#3: 5’-ATGGTAGCTGTAGAAGC-3’ (nt #495 to 479), #4: 5’-GTCGCAGTACCTCT-

TGTA-3’ (nt #1044 to 1027).

Southern Blot Analysis

Genomic DNA was isolated by the method of Dellaporta et al. (1983) and
40 ug were cleaved with BamHI (16, 80 or 160 units), Hindlll (160 units), or Xbal
(160 units) using standard reaction conditions for 20 hours. The reaction products
were extracted with phenol, phenol/CHCI3/lsoamyl alcohol (2522421), CHClsfisoamyl
alcohol (24:1), precipitated with sodium acetate and ethanol (Sambrook et al.,
1989) and resuspended in 10mM Tris, pH 8.0 with 0.1 mM EDTA. This digested
DNA was redigested, extracted and precipitated as above then size separated on
a 0.8% agarose gel and capillary blotted onto transfer membrane (Nitroplus,
Micron Separations lnc., Westboro, MA). azP-labelled uda1 was prepared and
used to probe blots as described for RNA gel blots. Blots were washed at high
stringency (0.2x SSC + 0.1% SDS at 65° C) and exposed to autoradiographic film

for visualization of bound probe.

Sequence Analysis and Comparisons

Sequence analysis and comparisons were performed with MICROGENIE
software (Beckman, Fullerton, CA) and EDITBASE software (courtesy of N. Nielsen,
Purdue University, West Lafayette, IN). Amino acid sequence of the chitinase
genes was deduced from the nucleotide sequence in Genebank. Amino acid

sequence alignments were done manually.

81
Expression ofthe uda1 Encoded Domains in E. coli

Oligonucleotide directed mutagenesis (Dale and Arrow, 1987) was used to
insert restriction sites into the cDNA for convenient cloning into the 17 RNA
polymerase E. coli expression vector pET3a (Studier et al., 1990). One
oligonucleotide (#1: 5’-GGTCTAGTGTCGGCATATGCAGAGGTGCGGAAGC-3’) was
used to insert an Ndel site at the beginning of the region encoding the mature
stinging nettle lectin. A second oligonucleotide (#2: 5’-GCCAGTACCGC-
TGCTAACATATGATCGGCAACGTCGTCG-3’) was used to insert a stop codon
immediately following the sequence encoding mature stinging nettle lectin along
with an Ndel site at the beginning of the chitinase-like region. A third
oligonucleotide (#3: 5’-TGTTGCGGCGTAAAACATATGCTAGTCCTCCCCAAG-3’)
was used to place an Ndel site just following the stop codon for the open reading
frame. Pairs of these oligonucleotides were then used in separate mutagenesis
reactions to construct Ndel bound inserts encoding the mature stinging nettle
Iectin (#1 and #2), the chitinase-like region (#2 and #3), and the entire open
reading frame without the putative signal sequence (#1 and #3). Mutagenized
nettle lectin cDNAs were sequenced to confirm mutagenesis and check for errors
introduced during mutagenesis. These inserts were ligated into the Ndel site of
pETSa for expression of the polypeptides in E. coli. E. coli, BL21(DE3) (Studier,
et al., 1990), cultures containing the constructs were grown to OD“ 0.5, induced
with 100 uM isopropyl B-D-thiogalactoside (IPTG) and incubated for 5 h. Cells
were collected by centrifugation and resuspended in 1 ml extraction buffer (50 mM

Tris (pH 8.0), 1 mM EDTA, and 100 mM NaCI) per gm cells. Crude lysate was

82
prepared by freezing the cell suspension at -20°C, thawing at 37°C for 20 min.,

and removing the cell debris by centrifugation at 25000xg for 15 min. (Studier, et

al., 1990).

Chitinase Assay

The colorimetric chitinase assay was performed by the method of \Mrth and
Wolf (1990). Briefly, two parts crude extract, diluted to 1 ug/ul protein with
extraction buffer, was mixed with one part 0.2 M sodium acetate (pH 3.6) and one
part chitin dye substrate (carboxymethyl-chitin-Remazol Brilliant Violet 5R
(CM-chitin-RBV) from Loewe Biochemica GmbH, Otterfing, FRG) and then
incubated at RT for 12 h. Undegraded substrate was precipitated with one part
1 M HCI for 10 min on ice and removed by centrifugation at 2500xg for 10 min.
Absorbance at 550 nm was used to measure soluble RBV released by the
chitinase activity in the extracts. A blank was prepared with extraction buffer in

place of the E. coli extract.

RESULTS

To further our studies on chitin-binding proteins and their endogenous
functions the stinging nettle lectin cDNA was cloned. An initial attempt to use the
barley Iectin cDNA (Lerner and Raikhei, 1989), another chitin-binding protein, as

a heterologous probe was unsuccessful due to insufficient hybridization specificity

83
as determined by RNA gel blot analysis (data not shown). Additional attempts to

use degenerate oligonucleotides as a probe or generate a specific probe by PCR
were also unsuccessful. Finally, we decided to synthesize a gene encoding the

reported stinging nettle Iectin amino acid sequence.

Assemblyand Cloning oftheSyntheticStlnging NettleLecfin Gene

The synthetic gene was assembled by sequential ligation reactions of
annealed oligonucleotide pairs in two separate reactions starting at each end of
the gene. The two half-genes were then ligated together (Figure 1). This strategy
was used to limit the number of potential mismatched ligations. Denaturing
polyacrylamide gel electrophoresis indicated that both complete and partial ligation
products were present at each step. The final ligation reaction products were
restricted with EcoRI and ligated into appropriately restricted and
dephosphorylated pUC119. Sequence analysis showed that several synthetic
gene clones were missing single base pairs but others were accurately
synthesized and assembled. One of these correctly synthesized clones was used

for further experiments.

IsolationcfacDNACloneforStingIngNettieLectin

The synthetic gene was used as the probe to screen an amplified stinging
nettle rhizome cDNA library. Eight positive plaques were purified through
successive rounds of screening and their inserts isolated via PCR due to loss of

the EcoRI restriction sites during library construction. Isolated inserts were

34

blunt-end ligated into the Smal site of pUC1 19 and sequenced. Only one clone,
uda1, encoded nettle lectin amino acid sequence and was completely sequenced
in both directions. Uda1 was approximately 72% identical to the synthetic gene
used as the probe. Although the deduced amino acid sequence for stinging nettle
lectin was only 86 amino acids long, uda1 encoded an open reading frame of 372
amino acids. The deduced amino acid sequence included a putative signal
peptide (23 amino acids), the two chitin-binding domains of nettle Iectin (86 amino
acids), a ’spacer’ domain (19 amino acids), and a 244 amino acid carboxyl-

terminal domain (Figure 2).

Deterrnlning the Authenticity of uda1

The unexpected structure of uda1 prompted us to use several methods to
verify that the isolated cDNA corresponded to the actual message for stinging
nettle lectin. First, RNA gel blot analysis of stinging nettle rhizome total RNA was
performed to analyze the hybridization pattern with cDNA clone fragments
corresponding to the stinging nettle lectin encoding domain or the carboxyl-
terminal domain. A single band of approximately 1.3 kb in size was detected with
each probe (data not shown). These results also indicated that a near full length
cDNA clone had been isolated since the hybridization was to a message of
approximately the same size as the cDNA.

Next, PCR was uSed to analyze whether the hybridization seen in the RNA
gel blot analysis was in fact due to a single mRNA species and not different

messages of approximately the same size. Reverse transcribed total rhizome RNA

85

6:253: ciao: a >: oemeE m. 50::
mocmocom Eon o:.Em because 65 :. 2.» :o_pm_>moo>_o “335736933 53:82. 65:5 6:... .586. v.36: usaczm
oLBmE so team 65 9 6283mm. 623:5: cc: mocesoom @2523: 65 26.3 665.. Be no.0: o:.Em 32660
665.52: m. 5:9.» :o.um.>:som>_oa .555: 65 cc: 6E9: @532 :ooo 9.... Co 655.59: 5:: 65 «a «team

5.5.3:: on... . Pep: o:o.o (200 case. 23¢: usage 2: So 8:253 Eon o:.Ea census: :5 £53.32 .N :52“.

86

»»m»

cow»

am»—

one»

00»

who

chm

00¢

00~

No—

<<0¢»»0»¢» 0»¢000¢»»0 »¢»»¢»<¢»0

I
0¢»00<¢0<¢ ¢¢» 000

> 0
0»0 <00
oo~

_ >
0»¢ 0»0

3 a
00» 0»»

a m
0¢¢ 0»»

> »
0»0 »0¢

> 0
000 <00
on»

000 0<0

0 4
00¢ 0»0

> m
0»0 00¢

» a

00¢ 000
»

a 0

00¢ 0<0

<

0~n
4 3 0 0
0»0 0»¢ 0<0 00»

0 a z u
»00 <¢¢ 0¢¢ <00

¢ » a a
000 00¢ 0¢¢ 0»»

a a » 4
0¢¢ 0<0 00¢ 0»0
oo—

» » 2 ~
»0¢ ¢0¢ »<¢ 0»¢
00»

0 a m 0
0»0 0<0 <0¢ 0¢¢

m a 0 0
00¢ 000 »0» 0<0

¢ > a A
»00 0»0 000 000

0 0 a 0

00» 00» 000 0<0
on

0 0 > a

000 00» 0¢» 000

¢ 0 > 0
<00 00» 0»0 ¢»0

¢ > a x
000 »»0 »00 0»¢

> a u >
0¢» 00¢ 0¢¢ 0¢»

~ 0 z ¢
0»¢ <0¢ 0¢¢ 000
omN

m _ ~ a
<0» ¢»< 0»¢ <00
ONN

a 0 a 0
0¢0 0»¢ 000 <00

2 0 0 0
»<0 »0» »»0 000

m » ¢ 0
000 ¢0< 000 »»0

a a a m
0¢¢ 000 0¢0 00»
00

0 a > m
<00 »00 0»0 00»
00

a 0 0 a
0¢0 000 00» 000

m m 0 0
0<0 00» 0<0 000

0 > ¢ 8
»00 0»0 000 0»¢

»00»»»0 <000¢»0a¢fl »¢<¢»¢»»0¢ 00¢»»0»00¢ »0»»»»¢0»0 »0»¢»¢»¢0¢
»¢»00»¢»»0 »000»<<0»» 0»¢00000<0 <0<000¢»<0 »0»0»00¢»0 <0¢<»0¢»0¢ »0»0»000»0 ¢<0000»00»

own 0mm
0 0 a a m w 0 0 a » z 0 m 3 > u a 0 a 0 > 0
¢¢0 0»¢ 000 <¢0 0»» 0¢0 00» <0» <00 »0¢ 0¢¢ »<0 0»» 00» 0¢» 0¢¢ 0»» 0<0 »<0 <00 »¢» 00¢
0»M can
a 0 0 m m m a > ¢ 5 0 a z 0 0 ¢ a m _ ~ 2 0
»»» <00 0»¢ 00¢ <0» »0» <0< »»0 000 »»» 0<0 0<0 0¢0 000 »»» »00 000 00¢ »»¢ »»¢ 0¢¢ 000
00~ DNN . cow
¢ 0 z ¢ 3 0 > ~ 0 z 0 0 a a 3 0 z a 0 8 3 m
000 00» 0¢¢ 000 »<¢ 0»0 0»0 »»¢ 0<0 »<0 00» <0» »00 <¢¢ 0¢¢ 0<0 0¢0 0¢0 00» 0»¢ 00» 0»»
0¢~ OMN
0 u m > 0 0 A z a _ a a m 0 n < a 0 ¢ 0 0 >
»<0 0¢¢ <¢0 ¢»0 0»» 0<0 »00 0¢¢ 0¢0 »»¢ 0»0 0¢0 0<0 <00 »»¢ 000 <¢0 000 000 »»0 <00 0¢»
0»~ 00m
x a m > u x 0 ¢ ¢ 0 a 3 0 0 0 » 0 u 0 z m »
<00 »00 00¢ 0¢» <¢¢ <¢¢ 000 000 »00 00» »00 00» 0<0 00» 00» 00¢ »0» 0»» 0<0 »<¢ 0<0 »0¢
on» 0»»
3 < z a z 0 ¢ < > m a x c » ¢ > o 0 _ » 0 a
00» <00 »<0 »00 0¢¢ <¢0 0»0 0<0 »0» 00¢ <¢0 000 <0» ¢0¢ <00 <¢0 00» 0»¢ »<0 000 0»0 0»»
0m» 00»
m _ z ¢ 4 0 ¢ a u 0 0 ¢ ¢ > 0 u 0 z > 0 > 5
»<0 <00 0»¢ 00¢ 000 0»» »00 <00 0»» 00» 0<0 000 000 ¢»0 0»0 0»» 00» 0¢0 0¢» 00¢ 0¢» 0»»
ON» 0»» Imp»
a x a 0 a 4 a w » > > > a 0 n m z ¢ » m _ 0
0»» 0»¢ 0<0 0¢0 0»» 0»0 000 0<0 00¢ 0»0 0»0 0»0 0¢¢ 000 0»¢ 00» 0¢¢ 000 00¢ 00» »¢¢ 000
00 oh
0 0 0 c > o 0 u m 0 m 0 > 0 a 0 0 0 0 3 0 x
00» 00» 00» 000 0¢» 0¢0 00» <¢¢ 00¢ 000 00» 00» 0¢» 0¢0 0¢¢ 000 000 »00 00» 00» 000 0<0
on 00
a a 0 > ¢ ¢ 0 0 a z o m a w 0 m 3 0 a z m 0
»00 0¢¢ <00 ¢»0 »00 000 000 00» 000 0<0 0<0 00» 000 0<0 000 00¢ 00» 00» 0¢¢ 0¢¢ 0<0 00»
ON 0»
0 3 0 3 ~ 0 0 0 3 0 ¢ a 0 » 0 0 0 0 a m 0 0
00» 00» 000 00» 0»¢ 00¢ 00» 00» 00» 0»» 000 000 »0» 00¢ »00 000 000 000 <¢0 00¢ <00 00»
0»- 0m.
¢ 0 0 a u > > ¢ 0 a a a z x z
»00 00» 00» 0»¢ 0»¢ 0»0 ¢»0 000 »0» ¢»» »»» 00¢ 0»¢ 0»¢ 0»¢ 0 <<<<0<<¢0¢ ¢»0¢»<u»¢¢

87

was used as the template for the PCR. Primers for the reactions corresponded
to: the beginning of the open reading frame (Figure 3, #1), the end of the stinging
nettle lectin encoding domain (Figure 3, #2), the beginning of the carboxyl-terminal
domain (Figure 3, #3), and the end of the carboxyl-terminal domain (Figure 3. #4).
An elevated annealing temperature (65° C) was used for the reactions to obtain
highly specific products. Figure 3 shows that each set of primers produced a
single predominant PCR product as detected by an ethidium bromide stained
agarose gel. The size of each product was exactly as predicted from the nettle
lectin cDNA clone. Primers #1 and #2 generated a 303 bp product, primers #1
and #3 a 451 bp product and primers #1 and #4 a 995 bp product. The PCR
products from each reaction were then subcloned into pUC119 and partially
sequenced. In all cases the sequence matched that of uda1. In addition, three
partial cDNA clones were independently isolated from an unamplified U. dioica
rhizome cDNA library using uda1 as a probe. Sequence obtained from these
clones matched uda1 sequence (data not shown). The amino acid sequence,
RNA gel blot, PCR, and clone sequence data all indicated that an authentic nettle

lectin cDNA was isolated.

Comparison of udal Deduoed Amino Acids to Published Amino Acid Sequence
of Stinging Nettle Lectin

The deduced amino acid sequence of stinging nettle lectin in uda1 matched
68 of the 72 residues (94%) reported by Chapot et al. (1986) (Figure 4). Three of

the differences (Ala (52) for Gln, Pro (57) for Thr, and Asp (61) for Leu) were within

Flgure3. PCRanalyslsuslngreversetrensoflbedtotal RNAfrom U.
doles as the template with primers from uda1. A) Diagram showing
positions of the primers on the cDNA clone. B) PCR products were
separated by electrophoresis on a 1% agarose gel containing 0.5 ug/ml
ethidium bromide and visualized with UV light. Lanes 2, 3, and 4 are
total PCR products generated with the primers indicated. Hindlll cut

lambda DNA with Alul cut pBR322 are the size standards (lane 1).

89

A. NEIT LE LECTIN cDNA

+5

2 3
+ ‘-

*d

 

  
  

signal "spacer" 3' untranslated region
sequence

chitin binding domains chitinase-like domain

B.

1 2 3 4

2.32 -
2.03 - ~

 

90

l 10 20 30 40
uda1 (deduced): QR/CGSQGGGGTCPALN CCSI WGWCGDSEPYCGR/TCENK/CNSGER/
Netti e 1 ecti n: QR/CGSQGGGGTCPALB/CCSI WGWCGDSEPYCG-l CNSGER/
(trypti c fragments)
50 60 70 80
uda1 (deduced): SDHR/CGAAVGNPPCGQDR/CCSVHGNCGGGNDYCSGSK/CQYR/C
Netti e 1 ecti n: CGAQVGNPlCGQIiR/CCSVHGNCGGGNDYC-/ CQYR/

(tryptic fragments)

Figure 4. Comparison of stinging nettle Iectin amino acid sequences. Alignment

of amino acid sequence from Edman degradation of stinging nettle lectin tryptic

digests (Chapot et al., 1986) with the deduced amino acid sequence of udat.

Trypsin cleavage sites are indicated with a slash (l). Non-identical amino acids are

double underlined. lncompletely sequenced fragments are indicated with a dash

H at the carboxyl end.

91

the sequence of a single tryptic fragment. The fourth unmatched amino acid was
Trp (16) for Arg in the N-terminal sequence. In addition, 14 residues not identified
by peptide sequencing are present in the deduced amino acid sequence encoding
stinging nettle lectin. Nine of these residues are predicted to be completely
contained within two tryptic fragments which were not sequenced by Chapot et al.
(1986). The remaining five residues would be present at the carboxyl-terminal

ends of tryptic fragments only partially sequenced by Chapot et al. (1986).

Comparison ofthe uda1 DeducadAmlnoAcldSequencewltl'l ClonedChltlnases

Further analysis of the deduced amino acid sequence from the carboxyl-
terminal domain revealed extensive similarity with the deduced amino acid
sequence of cloned chitinases (figure 5). This similarity extends beyond the chitin-
binding domains to include the chitinase catalytic domain. As expected. stinging
nettle lectin contains two chitin-binding domains (86 amino acids) whereas the
class I chitinases and hevein possess only a single chitin-binding domain
(approximately 43 amino acids). The carboxyl-terminal domain encoded by uda1
has 40—46% identity (approximately 60% with conservative substitutions) with the
catalytic domain of the cloned chitinases (Figure 5). The ’spacer’ domain between
the chitin-binding and chitinase-like domains has no similarity to previously
published sequence for stinging nettle lectin or to the deduced amino acid
sequence of the cloned class I chitinases. This ’spacer’ region does, however,

contain the only potential asparagine(N)-linked oligosaccharide recognition site (N-

92

.500» :E :6
Ems... 00000.2 63520 >250 2000» . .3 as 086% 00000.). .= 322:0 800:0... “.000» . .E 8 335$ 50000—2
63:25 Each. “.000» :3. .8 2.090. 0000 :2 635:5 530 “K03 in 8 23:30. 0h »0 :2 ._ mmmczfo ooomno...
2000» .5530. :00ox 63520 939... "mBnEac :23302 $383 xcmmmcmw o5 :_ 3:03am» 658.03:
So: 09:00 9.25 30:33.» Bum o:_E< .3. camp 0 53> pouocop Sm 30:033.» 030:: 05 :_ 33 0:0 3
:28 m 53> 065:0: 2m 225533 Eon 0560 0252630.". .. : 6:: .85.? o 53> 090:2. 2m :3: 3 .8552
mass o:_E< .Btszfi be 6903 56:9: o5 6232: 3 33 53> >=m=:mE 250% 225 30:63:60 63855
Ba 3588 5298 322:0 “Em 3209 Leeann. .mEmEoc 0:659:25 .mmocmaaom 3:03. 359:0 623520

32.2, 3 «8:33» Eu: o:_Ea boosts: 25 53> 3:260» Eon o:_Eu nonstop pan: 3 533955 .0 2:0.“—

93

5.53.85... ..u._ ._ . .13.}.895 55:25 5.5...
.8.......-...8.5.... .. ... . . .5... .555 2 59.525 82.8.
.xasa..dz.u.2.-§:S.8 0.... . 5.5.8.58... 55:25 «58..
.a. 282:5...128 ... ..o a.....::..:..8>..m8..8 55:25 25..
283555.158 .5... ......m...._:...8>.. .82... 2 55:25 82.8.
...S 3.25.3. .-....8 ...o .... . . . .m . _ .. . . 582.558 ”9...: .5 2:9.
3:5555.55.53.652553-13%.... .58. 325..

ovm 000 own 08 con
tilt-l:- .:s§. 95-55.55 -----v

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

528..-: :5: ..86...&59...<.m.- .99....7 ...c.:_.<._. ._.... — .... .....__:_.o.._.:..:.v<...._._ 55:25 5.53
8.8..- _ ...8 ...82258...:m. -..v....:.. .....z. 2. . .. 5.. . . .5 ...... ... .2. .............. : 5555 828.
8-5.... 2.5...a...o.um..m..5—.....:<Em32...: 3:251... 3.5. ....u. H... .r........... 5.5.5358... 05:25 5.8..
8.8..- ..2. 5......5.3.3522:..-...8...:::..:..m.2....x.._..._:...u...8_ .5..._...o>8....:&.......- 55:25 25..
8.8.7.2»..57.15.58.53.25-:8...:::.:..H::...._.. ... 5.2.5... ......_ ............:_vo...._”_ 255.225 828.
Baez-...»...B... ....o.mm......5....::-...8...:..::.:..:1...3:12:55... ...... ....S......:.am.&. 59525850..
:555:5.5¥..:¢z<<mi ...... 25.2uma-355955155289:39.5285559.5555628.55.353.935... .58.. 320..
as 8... 2... 8w 8... 3“ on 8.. 2m 08
I 55.8 5.?gz225 .-
$55.5.»:5935...uxzo........v§m=2.2.773... . _. .05... 3.2.95.5...” 55:25 5...:
-895- ... .23..-23.-u...u..=8..:. ....a..m...o:... ......o . 2:5... axing—62:50.8 : “95:25 89.8.
$550....»....2625.....qu......._...§..E..o......r.a . _ ..woz......::. 3:5... 55:25 5.5..
-25... ......::.......25_ .....8 .2: ............_......_.....o.._ 5.5.2.3....aaxam8-......a ----------- 55:25 .25
é58.......»:.2.w...<25_. 1.1:... ...._.._.....m...................... ....85.....y......$:.85-..8a&8.. ---------- 255.25 83.8»
5.55512... 5.95.5 5.50:1... 5.1.2.5............z...58.9....5.....32388-88558. -------- 55:25 028..
52255.9(...5- 5552525.: 5.52.85.5.5535..§505-55558552357252855553 223 32...
8. 8. 2. 8. a... a: on. 8. o: 8. 8
; 55.8 6.3-5.225 \A--- 53.0 --v
3.21.52. 2.; .52.. 05:25 5:53
752.55.553.82! 2 55:25 888.
3.. -u5..<>3..._u.w 2.8.. _:5;...35535...:..m..:5:... 55:25 538..
08 -oa...9.w_...o ...8 53...; »mo....as..u.£¥.:..§¥ 55:25 25..
08.28.52... 53.1.5“ ....5........vu.zm...:5.mm.75...: 35:25 828—
08.52.53.253. .8 ..x... ...... ..3m..3....;.:..5xz..i .9525 2:9,.
26338858use...muuaowwuazosiouxxomfiogxz5- 53588.. 38.39355885mmou §m.d>52mm....><m:§t 523 32”....
8 2 S 8 on 8 S . 2- 8-

 

 

: 0:580 0232—0 1350 \Aiii 0020300 #205 IV

94

S-T) at deduced amino acid residue #100. it remains to be determined whether

the stinging nettle lectin precursor is glycosylated or not.

Tissue Distribution of Stinging Nettle Lectin in u. dioica

Total RNA isolated from rhizomes, lower and upper stem sections,
inflorescence with immature seeds, and leaves were examined for the presence
of stinging nettle lectin message. Relatively high steady state levels of the 1.3 kb
nettle lectin mRNA were detected in the rhizomes and the inflorescence containing
immature seeds (Figure 6. lanes 1 and 4). In addition, longer exposures of the
autoradiograms revealed much lower levels of stinging nettle lectin message in the
RNA isolated from the upper portion of the stem. No hybridization was detected

with the RNA samples from the lower stem portion or leaf.

Southern Blot Analysis

Previous studies have reported that stinging nettle Iectin in rhizomes of
stinging nettle is a complex mixture of isolectins (Van Damme et al., 1988).
Southern blot analysis of U.dioica genomic DNA using uda1 as a probe was
performed to analyze whether the reported isolectins could be due to multiple
nettle lectin genes. Figure 7 shows several hybridizing bands of genomic DNA
cleaved with BamHI, Hindlll and Xbal. Lanes 1, 2 and 3 contain DNA digested with
increasing amounts (16, 80 and 160 units) of BamHl. The identical pattern of
hybridizing bands indicates that the digestion in lane 1 was complete and that lane

3 contains at least a 10-fold excess of enzyme. The DNA in lanes 4 and 5 were

95

FlgureB. RNAgelblotanalysleoftotalRNAfrom Maiden”. 20
ug of total RNA (lanes 2-6) was electrophoresed on a 2%
agarose/formaldehyde denaturing gel transferred to nitrocellulose.
hybridized with 32P-labeled uda1, washed at high stringency (0.2X SSC
+ 0.1% SDS at 65° C), and visualized with autoradiography. Lanes 2-6,
rhizomes, lower stem sections, upper stem sections. inflorescence (with

immature seeds) and leaves, respectively.

96

Lower Upper Inflores-
l Rhizome Stem Stem cence Leaf

 

97

Figure 7. Southern blot analysis of U. dioica genomlc DNA. 40 ug
genomic DNA was digested two times 20 h with 16 U BamHl (lane 1).
80 U BamHl (lane 2), 160 U BamHl (lane 3), 160 U Hindlll (lane 4) or
160 U Xbal (lane 5). 5 ug per lane of digested DNA (lanes 1-5) or
undigested DNA (lane 6) was electrophoresed on a 0.8% agarose gel,
transferred to nitrocellulose, hybridized with 32F labeled uda1, washed
at high stringency (0.2x SSC + 0.1% SDS at 65°C), and visualized with

autoradiography.

98

1 2 3 4 5 6

23.1-

9.42 it: «It»
l6.56- 4“

l4'37-
l.

   

2.32 -
[2.03 -

 

 

’~———’—'

99

digested with 160 units of Hindlll and Xbal respectively. Since uda1 contains one
BamHl restriction site but has no sites for Hindlll or Xbal and multiple hybridizing
bands were seen in each digest, these data suggests that nettle Iectin is encoded
by a small multigene family. Thus, some of the nettle lectin isolectins found in the
rhizomes may be encoded by independent genes rather than the result of

posttranslational processing.

Expression of the uda1 Encoded Domains in E._o_gli

To assess the properties of the domains encoded by uda1, constructs were
prepared encoding the chitin-binding domain alone, the chitinase-like domain
alone, or the chitin-binding domain with the chitinase-like domain. These
constructs were placed under the control of the T7 RNA polymerase which, in turn
was controlled by the lacUV5 promotor and induced by the presence of lPTG
(Studier, et al., 1990). Expression of the chitinase-like domain or the full coding
region, minus the putative signal sequence, in E. coli results in chitinase activity in
crude cell extracts. Figure 8 shows a comparison of the levels of chitinase activity
in the crude cell extracts measured by the colorimetric assay of Wirth and Wolf
(1990). Both the chitinase domain alone (Figure 8, #3) and the chitinase domain
with the chitin-binding domain (Figure 8, #4) have considerably higher activity per

ug extract protein than the vector alone (Figure 8, #2).

100

Figure 8. Chitinase assay of crude cell extracts from E. oofi cells
expressing uda1 encoded domains. Colorimetric chitinase assays were
done with crude cell extracts from IPTG induced E. coli cells with the
vector pET3a (vector), pET3a containing the chitinase-like encoding
domain of uda1 (chitinase domain), or pETSa containing the chitin-
binding and the chitinase-like encoding domains of uda1. Optical
density readings were normalized to reactions containing extraction
buffer instead of the cell extracts. All values are the average of three
independent reactions with the error bars representing standard

deviations.

101

  

O

. 3mm.
9
O 0

w
9

O

‘.‘

 

“

 

900 OowoiOiOq «

9 99
%flflfiflfi. %fiv
999999 99
999999 99
999999 99
999999 99
%fififlfifl.?§%
99993‘ 43%
.9 9999

9’. D

 

 

.9“.

O

O

9900
O
09
99
900
09906
O.

.I

i.

99
9V

  

   

......» i

    

x€3€39
99 99
99 99
99 99
99 99
.0 O.
99 99
O. .9
33% .3%
999 99

     

               

 

 

    

 

 

  

  

99
v“

0.040‘
69990

"

    

  

   

 

 

 

(L50 P

(L45 -

(L40 ~

(L35 —

(L30 b

O'D°550

(L25 -

(L20 —

(L15 -

(L10 F

(L05 -

(LOO

BOTH

DOMAIN DOMAINS

VECTOR CHITINASE

BUFFER

1 02
DISCUSSION

In this paper we have presented the nucleotide and deduced amino acid
sequence of a cDNA clone for stinging nettle lectin. Initially, barley Iectin cDNA
was used as a heterologous probe for stinging nettle rhizome RNA gel blots.
Results from these experiments showed that the heterologous probe would not
specifically hybridize to a nettle rhizome message. Next, PCR was used in an
attempt to generate a specific probe for the stinging nettle lectin gene. This
strategy involved synthesizing redundant oligonucleotides encoding all possible
codons for several regions of the amino acid sequence. Specific PCR products
of the predicted size were obtained from some of these reactions. Subcloning and
sequencing these PCR products, however, revealed that with all primer
combinations, only primer/primer products had been made. In retrospect it is
clear that these results were due to the highly G/C rich nature of the stinging nettle
Iectin gene. In addition to the base composition of the primers, it is obvious from
the deduced amino acid sequence of uda1 that a tryptic fragment, to which some
of the primers were made, had been previously arranged incorrectly (Dr. Stephen
Michnick, Harvard University). Thus, PCR primers to this region would lead to
DNA synthesis away from the opposing primers and result in no amplification.

Stinging nettle lectin is a small, 86 amino acid protein with 72 amino acid
residues known from tryptic fragment sequencing (Chapot et al., 1986). Although
the complete amino acid sequence was not available, homology between stinging

nettle lectin and other chitin-binding proteins and positioning of the tryptic

103

fragments (Steven Michnick, Harvard University, personal communication)
permitted designing a synthetic gene for use as a hybridization probe. This
synthetic gene showed a high level of hybridization specificity by RNA gel blot
analysis (data not shown) which allowed for screening the cDNA library.

The cDNA clone isolated had an unexpected structure. Along with the
anticipated region encoding the two chitin-binding domains of stinging nettle lectin
and the putative signal sequence, the open reading frame possessed an additional
263 amino acids. Four lines of evidence show that the cDNA isolated was the
authentic stinging nettle lectin gene. First, the deduced amino acids of the stinging
nettle lectin encoding domain of uda1 match 68 of 72 amino acids sequenced
from stinging nettle lectin tryptic fragments (Chapot et al. 1986) (Figure 4).
Second, RNA gel blot analysis showed that both the stinging nettle lectin and the
chitinase encoding regions hybridized to a single mRNA species similar to the size
of uda1 (1300 bp). Third, PCR analysis of reverse transcribed total rhizome RNA
shows that the stinging nettle lectin encoding domain and the carboxyl-terminal
domain exist on the same mRNA species (Figure 3). Finally, independently
isolated stinging nettle lectin cDNA clones from an unamplified library matched the
sequence of uda1.

Although the extended structure of the uda1 open reading frame was
unexpected, it was not entirely surprising. All of the previously cloned genes for
chitin-binding proteins seem to be the result of either chitin-binding domain

duplications, as in the Gramineae lectins, or fusions of a single chitin-binding

104

domain with an unrelated domain, as in hevein and the class I chitinases (see
Chrispeels and Raikhel, 1991 for review). The stinging nettle Iectin is the first
example of a chitin-binding protein gene resulting from both a domain duplication
and fusion with an unrelated domain. Differences between the deduced amino
acid sequence of uda1 and previous amino acid sequence of stinging nettle lectin
tryptic fragments (Chapot et al., 1986) are most probably due to variations
between isolectins. Stinging nettle lectin has been shown to have up to seven
isoforms (Van Damme et al., 1988). It is also possible that mistakes arose in
amino acid sequencing since the majority of the differences lie on a single tryptic
fragment.

The chitinase domain of uda1 remains enigmatic both as to its endogenous
function and processing. The deduced amino acids for the carboxyl-terminal
domain possess approximately 60% similarity with the catalytic domain of other
cloned chitinases. These chitinases, however, are over 90% similar to each other.
Where studied, it is known that the class I chitinases do not undergo a processing
event to release the chitin-binding domain from the chitinase catalytic domain.
Presumably, this type of processing event does occur with the uda1 encoded
protein since stinging nettle lectin, when isolated from the plant, is a very small
protein (8.5 kDa, 86 amino acids) containing only the two chitin-binding domains.
Lee et al. (1991) have shown that the hevein preproprotein is processed in vivo to
yield the chitin-binding protein (hevein, 5 kDa) and a polypeptide of unknown
function (16 kDa). The Gramineae lectins have also been shown to undergo post-

translational processing of the carboxyl-terminal propeptide with 15 to 27 amino

105

acids removed (see Raikhei and Lerner, 1991 for review). Differences between the
uda1 encoded ’spacer’ region and the ’hinge’ region of the cloned class I
chitinases suggests that there may be differences in their functions. The chitinase
’hinge’ region tends to be short, 8 to 11 residues, and rich in proline and glycine
while the ’spacer’ region of the uda1 deduced amino acid sequence was longer,
19 residues, and contained only one proline. Since the class I chitinases have
never been shown to undergo a cleavage to remove the chitin-binding domain it
is likely that this ’spacer’ region in the stinging nettle lectin proprotein contains the
recognition site for proteolytic cleavage. Future studies will address processing
of the stinging nettle lectin proprotein. Future studies will also include intracellular
localization of the stinging nettle lectin and the chitinase domain in stinging nettle
(U. dioica) and transgenic plants. The possibility of separate targeting signals for
the two domains is intriguing. Vacuolar targeting domains have recently been
identified for barley lectin and the tobacco class I chitinase (Bednarek et al., 1990,
and Neuhaus, et al., 1991b, respectively). In both cases a short carboxyl terminal
propeptide (CTPP) has been shown to be necessary and sufficient for targeting
secretory proteins to the vacuole (Bednarek and Raikhel, 1991, and Neuhaus, et
al., 1991b). Comparison of the ’spacer’ region encoded in the nettle lectin cDNA
shows no sequence homology with the CTPP of barley lectin except for the
presence of an N-linked glycosylation site. The glycosylation site, however, is not
required for correct targeting of barley lectin (Wilkins, et al., 1990). In addition, the
chitinase domain encoded by uda1 extends beyond the carboxyl-terminus of the

secreted class II chitinases. Again this is similar to the CTPP of the vacuolar,

106

class I, chitinases but no distinct sequence similarity can be found. Expression of
uda1 in transgenic plants will provide an experimental system to study targeting
of the nettle lectin domains.

Expression of three uda1 constructs in E. coli using a T7 RNA polymerase
expression system showed that uda1 encodes a protein with chitinase enzymatic
activity. Expression of constructs encoding the chitinase domain alone or the
chitin-binding domain with the chitinase domain both show marked chitinase
activity.

The potential plant defense functions of the chitin-binding proteins presents
an intriguing opportunity for enhancement of plant disease resistance by
overexpression of these proteins in transgenic plants. It has been shown that
increasing the chitinase activity in transgenic plants by overexpression of a
chitinase in tobacco had no discernable effect on fungal resistance (Neuhaus et
al., 1991 a). However, Broekaert et al. (1989) has shown that stinging nettle lectin
and chitinase have different modes of action for inhibition of fungal growth in vitro.
In addition, stinging nettle lectin acts synergistically with chitinase for inhibiting
fungal growth in vitro. These data suggest that expression of stinging nettle lectin
in transgenic plants may enhance resistance to fungal pathogens where
expression of chitinase did not. In addition, this is an exciting opportunity to
analyze whether the in vitro synergistic antifungal effect between stinging nettle

lectin and chitinase is observed in the transgenic plants.

1 07
REFERENCES

Bednarek, S.Y., Wilkins, T.A., Dombrowski, J.E., and Raikhel, NV. (1990). A
carboxyl-terminal propeptide is necessary for proper sorting of barley lectin

to vacuoles of tobacco. Plant Cell 221145-1155.

Bednarek, S.Y. and Raikhel, NV. (1991). The barley lectin carboxyl-terminal
propeptide is a vacuolar protein sorting determinant in plants. Plant Cell

321195-1206.

Broekaert, W.F., Lee, H.-l., Kush, A., Chua, N.-H., and Raikhel, NV. (1990).
Wound-induced accumulation of mRNA containing a hevein sequence in
laticifers of rubber tree (Hevea brasiliensis). Proc. Natl. Acad. Sci. USA

87:7633-7637.

Broekaert, W.F., Van Parijs, J., Leyns, F., Joos, H., and Peumans, W.J. (1989). A
chitin-binding lectin from stinging nettle rhizomes with antifungal properties.

Science 245:1100-1102.

Broglie, K.E., Gaynor, J.J., and Broglie, RM. (1986). Ethylene-regulated gene
expression: Molecular cloning of the genes encoding an endochitinase

from Phaseolus vulgaris. Proc. Natl. Acad. Sci. USA 83:6820-6824.

108
Cammue, B.P.A., Broekaert, W.F., Kellens, J.T.C., Raikhel, N.V., and Peumans,

W.J. (1989). Stress-induced accumulation of wheat germ agglutinin and

abscisic acid in roots of wheat seedlings. Plant Physiol. 91:1432-1435.

Chapot, M.P., Peumans, W.J., and Strosberg, AD. (1986). Extensive homologies

between lectins from non-leguminous plants. FEBS Lett. 195:231-234.

Chrispeels, M.J. and Raikhel, NV. (1991). Lectins, lectin genes, and their role in

plant defense. Plant Cell 321-9.

Dale, R.M.K., and Arrow, A. (1987). A rapid single-stranded cloning, sequencing,

insertion, and deletion strategy. Methods Enzymol. 155:204-214.

Dellaporta, S.L., Wood, J., and Hicks, J.B. (1983). A plant DNA minipreparation:

version ll. Plant Mol. Biol. Rep. 1:19-21.

Feinberg, AP. and Vogelstein, B. (1983). A technique for radiolabelling DNA

restriction endonuclease fragments to high specific activity. Anal. Biochem.

132:6-1 3.

109
Gait, M.J., Popov, S.G., Singh, M., and Titmas, RC. (1980). Rapid synthesis of

oligodeoxyribonucleotides V. Further studies in solid phase synthesis of
oligodeoxyribonucleotides through phosphotriester intermediates. Nucl.

Acids Symp. Ser. 7:243-257.

Gaynor, J.J. (1988). Primary structure of an endochitinase mRNA from Solanum

tuberosum. Nucl. Acids Res. 16:5210.

Huesing, J.E., Murdock, LL, and Shade, R.E. (1991a). Effects of wheat germ

isolectins on development of cowpea weevils. Phytochemistry 30:785-788.

Huesing, J.E., Murdock, LL, and Shade, RE. (1991 b). Rice and stinging nettle
lectins: Insecticidal activity similar to wheat germ agglutinin.

Phytochemistry 30:3565-3568.

Leah, R., Tommerup, H., Svendsen, l., and Mundy, J. (1991). Biochemical and
molecular characterization of three barley seed proteins with antifungal

properties. .J. Biol. Chem. 266:1564-1573.

Lee, H.-I., Broekaert, W.F., Raikhel, NV. (1991). Co-and post-translational
processing of the hevein preproprotein of latex of the rubber tree (Hevea

bras/liensis). J. Biol. Chem. 26621594445948.

110
Lerner, DR. and Raikhel, NV. (1989). Cloning and characterization of root-specific

barley Iectin. Plant Physiol. 91:124-129

Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982) Molecular Cloning: A
Laboratory Manual, 1st Edition. Cold Spring Harbor Laboratory. Cold

Spring Harbor, NY.

Mishkind, M.L., Palevitz, B.A., Raikhel, N.V., Keegstra, K. (1983). Localization of
wheat germ agglutinin-like lectins in various species of the gramineae.

Science 220:1290-1292.

NaQY. F., Kay, S.A., and Chua, N.-H. (1988). Analysis of gene expression in

transgenic plants. Plant Mol. Biol. Manual 34:1-29.

Neuhaus, J.-M., Ahl-Goy, P., Hinz, U., Flores, S., and Meins, F.Jr. (1991a). High-
level expression of a tobacco chitinase gene in Nicotiana sylvestris.
Susceptibility of transgenic plants to Cercospora nicotiana infection. Plant

Mol. Biol. 16:141-151.

Neuhaus, J.-M., Sticher, L., Meins, F.Jr., and Boiler, T. (1991 b). A short C-terminal
sequence is necessary and sufficient for the targeting of chitinases to the

plant vacuole. Proc. Natl. Acad. Sci. USA 88:10362-10366.

111
Parsons, T.J., Bradshaw, H.D.Jr., and Gordon, MP. (1989). Systemic

accumulation of specific mRNAs in response to wounding in poplar trees.

Proc. Natl. Acad. Sci. USA 86:7895-7899.

Payne, G., Ahl, P., Moyer, M., Harper, A., Beck, J., Meins, F., and Ryals, J. (1990).
Isolation of complementary DNA clones encoding pathogenesis-related
proteins P and Q, two acidic chitinases from tobacco. Proc. Natl. Acad. Sci.

USA 87298-102.

Peumans, W.J., De Ley, M., and Broekaert, W.F. (1984a). An unusual lectin from

stinging nettle (Urfica dioica) rhizomes. FEBS Lett. 177:99-103.

Peumans, W.J., Nsimba-Lubaki, M., Carlier, AR, and Van Driessche, E. (1984b).

A lectin from Bryonia dioica root stocks. Planta 160:222-228.

Raikhel, NV. and Lerner, DR. (1991). Expression and regulation of lectin genes

in cereals and rice. Develop. Genetics. 12:255-260.

Raikhel, NV. and Wilkins, TA. (1987). Isolation and characterization of a cDNA
clone encoding wheat germ agglutinin. Proc. Natl. Acad. Sci. USA

84:6745-6749.

112
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A

Laboratory Manual, 2nd Edition. Cold Spring Harbor Laboratory, Cold

Spring Harbor, New York.

Sanger, F., Nicklen, S., and Coulson, A.R. (1977). DNA sequencing with chain-

terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467.

Shinshi, H., Mohnen, D., and Meins, F.Jr. (1987). Regulation of a plant
pathogenesis-related enzyme: Inhibition of chitinase and chitinase mRNA

accumulation in cultured tobacco tissues by auxin and cytokinin. Proc. Natl.

Acad. Sci. USA 84:89-93.

Silflow, C.D., Hammett, JR. and Key, J.L. (1979). Sequence complexity of
polyadenylated ribonucleic acid from soybean suspension cultured cells.

Biochem. 18:2725-2731.

Studier, F.W., Rosenberg, A.H., Dunn, J.J., and Dubendorff, J.W. (1990). The use
of T7 RNA polymerase to direct expression of cloned genes. Methods

Enzymol. 185260-89.

Van Damme, E.J.M., Broekaert, W.F., and Peumans, W.J. (1988). The Urtica dioica

agglutinin is a complex mixture of isolectins. Plant Physiol. 66:598-601.

113
Van Parijs, J., Broekaert, W.F, Goldstein, l.J., and Peumans, W.J. (1991). Hevein:

an antifungal protein from rubber-tree (Hevea brasi/iensis) latex. Planta

1 83:258-264.

Walujono, K., Scholma, R.A., Beintema, J.J., Mariono, A., and Hahn, AM. (1975)
In Proceedings of the International Rubber Conference, Vol. 2. Rubber

Research Institute Malaysia, KuaLa Lumpur, pp. 518-531.

Wilkins, T.A., Bednarek, S.Y., and Raikhel, NV. (1990). Role of propeptide glycan
in post-translational processing and transport of barley lectin to vacuoles

in transgenic tobacco. Plant Cell 22301-313.

Wirth, S.J., Wolf, GA. (1990). Dye-labelled substrates for the assay and detection

of chitinase and lysozyme activity. J. Microbiol. Methods 122197-205

CHAPTER 4

Size, Structure and Function Relationships of Chitin-Binding Proteins:
Preliminary Studies on Modifying Barley Lectin.

114

1 15
INTRODUCTION

Chitin-binding proteins provide a unique opportunity to examine the
structural features responsible for their activity. These proteins have common
structural elements (figure 1) and chitin-binding activity but differ in their ability to
inhibit the growth of fungi and insect larvae in vitro (Table 1). Chitin-binding
proteins are related by the presence of a conserved chitin-binding domain. This
domain has been duplicated in tandem in the Gramineae lectins and in stinging
nettle lectin. The mature Gramineae lectins are dimers with each subunit
consisting of four copies of the 43 amino acid chitin-binding domain. Stinging
nettle Iectin is approximately half the size of the Gramineae lectin monomer
consisting of two tandem repeats of the chitin-binding domain. Other chitin-
binding proteins, prohevein and the class I chitinases for example, contain only
one copy of the chitin-binding domain which has been fused to a structurally
unrelated domain.

The Gramineae lectins, especially wheat germ agglutinin (WGA), are the
most studied of the chitin-binding proteins. Biochemically and immunologically
these proteins are very similar (Peumans et al., 1982a, Mishkind et al., 1983).
Subunit exchange experiments, for example, have shown that WGA and barley
lectin monomers are interchangeable in the active dimer (Peumans et al., 1982b).
Wheat is a hexaploid plant with three distinct diploid genomes. Each of these
genomes contributes a separate. isolectin which differs in only a few amino acids

(Wright and Raikhel, 1989). Barley lectin is very similar to the WGA isolectins also

116

Figure 1. Gramineae lectins and homologous proteins. A graphical
representation of the domain structure of the Gramineae lectins and
some related proteins. Open boxes represent the 43 amino acid,
cysteine and glycine rich chitin-binding domain found throughout this
family of proteins. Grey boxes represent the unrelated domain fused to
the chitin-binding domain in hevein and the potato wound inducible
mRNAs, win1 and win2. The black boxes represent the chitinase
catalytic domain which is fused to the chitin-binding domain in the class

1, basic, chitinases.

117

 

oz; «2.50m

 

 

 

mm<zEIO 00030...

 

mm<zEIO Z<mm

 

 

«25> Obs—.2

 

 

.25, 05.5.0;

 

 

z_m>wI

 

 

 

25m.— NEZ

 

 

 

25m... NOE

 

25m.— >m..E<m

 

 

 

>>

wz_m._.0m_n_ mac—0040.20... DZ< $2....qu m<mz=2<cmv

heal
and
old.
this
9 to

Bible
1838
835

, l
l

118

Table 1. In vitro activities of chitin-binding proteins.

 

 

Protein Antifungal Insecticidal Reference
Activity Activity

Gramineae - - +++ Huesing et al., 1991a,b

lectins Schlumbaum ef al., 1986

Nettle lectin +++ ++ Broekaert et al., 1989
Huesing et al., 1991b

Hevein ++ NA Van Parijs et al., 1991

Class I +++ NA Schlumbaum et al., 1986

chitinases

 

 

 

Relative specific activity for inhibition of growth is represented by +++,
highly active; + +, moderately active; +, some activity; -, no activity; NA, data not
available.

119

differing by only a few amino acids (Lerner and Raikhel, 1989). Because of the
close similarity between the Gramineae lectins, in particular WGA and barley,
structural information obtained for WGA is likely to be applicable throughout the
Gramineae Iectins.

These Gramineae Iectins bind N-acetylglucosamine (GIcNAc), GIcNAc
oligomers, and chitin, the GIcNAc polymer (for review see Goldstein and Hayes,
1978). The active ligand-binding protein is a dimer with two primary and two
secondary binding sites in the cleft between the two subunits (figure 2).
Elucidation of the amino acid residues directly involved in binding chitin has been
achieved through the crystallization of WGA in the presence of the disaccharide
GIcNAc-GIcNAc (Wright, 1980; for review see, Reeke and Becker, 1988). Four
amino acids from one subunit (Ser62, Tyr64, Hi366, Tyr73) and one amino acid
(Glu115) from the other subunit interact directly with the disaccharide (Wright,
1980).

Stinging nettle lectin is similar to the Gramineae Iectins in its ability to
agglutinate red blood cells indicating the presence of two ligand binding sites.
Uke the Gramineae lectins, mature stinging nettle Iectin consists only of chitin-
binding domains. Stinging nettle lectin has been reported to contain up to 85
amino acids when the amino acid composition is examined (Peumans et al., 1984;
Van Damme ef al., 1988). Only 72 amino acids, however, have been sequenced
from tryptic fragments and subsequently aligned (Chapot et al., 1986, and Dr.
Steve Michnick, personal communication). Amino acid composition and SDS-

PAGE analyses indicate that stinging nettle Iectin is approximately 8.5 kDa in size

120

Figure 2. Graphical representation of mature barley lectin and stinging
nettle Iectin. The three dimensional structure of barley lectin is known
from crystallographic data of WGA (Wright, 1980). The chitin-binding
domains are designated A, B, C, D beginning at the amino-terminal of
the protein. The domains of the two monomers are designated with a
1 or 2 respectively. The active, carbohydrate binding protein, is a
homodimer with the binding sites in the cleft between the two
monomers. ’P’ and ’8’ represent the primary and secondary
carbohydrate binding sites, respectively. The structure for stinging nettle
lectin is inferred from that of WGA (model 1) and biochemical evidence
suggesting that the active stinging nettle lectin is a monomer (model

2) (Peumans et al., 1984).

121

BARLEY LECTIN AND STINGING NETTLE LECTIN
3-DIMENSIONAL STRUCTURE

BARLEY LECT IN

S'flNGING NEITLE LECTIN
Model 1

STINGING NETTLE LECT IN
Model 2

 

122

(Peumans et al., 1984). Alignment of the stinging nettle Iectin amino acid
sequence with other chitin-binding domains shows that the sequenced amino
acids contain the essential elements of two chitin-binding domains in tandem.
These include fourteen cysteine residues and amino acids homologous to those
involved in binding chitin in the Gramineae Iectins. Furthermore, both the
Gramineae lectins and stinging nettle lectin can retard the growth of insect larvae
in vitro (Huesing et al., 1991a and b). Stinging nettle lectin and the Gramineae
lectins differ both in size and in the ability to inhibit the growth of fungi. The
Gramineae lectins show no antifungal activity (Schlumbaum et al., 1986) while
stinging nettle lectin can inhibit the growth of a variety of fungi (Broekaert at al.,
1989). Although the endogenous functions of the Gramineae lectins and stinging
nettle Iectin are not known, the in vitro activities suggest that these proteins play
a role in plant defense.

The goals of this study were to examine the relationships between protein
size and in vitro antifungal activity of these chitin-binding proteins. One hypothesis
is that the Gramineae Iectins exhibit no antifungal activity because they are too
large to interact with the fungal cell walls in the same way as stinging nettle Iectin.
Measurements of fungal cell wall porosity (Money, 1990) indicate that Gramineae
lectins, especially as active 36 kDa dimers, will not be able to move through the
wall. Stinging nettle Iectin however, at half the size of the Gramineae lectins, may
be able to permeate the fungal cell wall where it could interact with the chitin to
inhibit fungal growth. To test this hypothesis, in vitro mutagenesis of the barley

lectin cDNA was used to construct a modified barley lectin (mBL) gene encoding

123

a polypeptide resembling stinging nettle lectin in size. This mBL was shown to

retain the ability to bind chitin when expressed in E. coli or tobacco.

MATERIALS AND METHODS

Plant Material

Tobacco plants (Nicotiana tabacum) were maintained in sterile conditions
by bimonthly nodal transfers on Mirashige and Skoog salt mix (BRL/Gibco,
Gaithersburg, MD) with 1.5% sucrose and 0.7% phytagar (BRL/Gibco). Plants
were grown in GA7 vessels (Magenta Corp., Chicago, IL) in a growth chamber at
28°C with a 16 h light cycle. For generation of seeds from transgenic plants, the
GA7 lids were set slightly open to allow to plants to acclimatize to humidity
conditions of the air for several days. Plants were then transferred to soil, covered
with plastic wrap and grown at RT in the laboratory with a 16 h light cycle for

several days before transfer to the greenhouse.

General Molecular Techniques

Except where explicitly stated all molecular biology techniques were done
according to olecular lonin ° bor Manua 1" (Maniatis et al., 1982) or
2"d (Sambrook et al., 1989) editions. Chemicals were from Sigma Corp. (St. Louis,
MO), and restriction enzymes from Boehringer Mannheim Biochemicals

(Indianapolis, IN).

124

Mutagenesis Strategy

To design the mBL the amino acid sequence of barley lectin was aligned
with the known amino acid sequence for stinging nettle Iectin (Chapot et al., 1986)
(figure 3). Additional information was provided by Dr. Steven Michnick (personal
communication) to arrange the tryptic fragment sequences of Chapot. According
to the above sources, mature stinging nettle Iectin was between 72 and 85 amino
acids in length with 72 amino acids of known sequence. This corresponded to
approximately two chitin-binding domains of barley lectin. Alignment of the known
stinging nettle lectin with the barley lectin amino acid sequence showed the
presence of one full and one truncated chitin-binding domains with the truncated
domain containing all five amino acids required for chitin-binding. Therefore, mBL
was design to contain 72 amino acids corresponding to domain B and
approximately two thirds of domain C. These domains were chosen because they
contain the primary carbohydrate binding sites of the mature barley lectin (figure

3).

Oligonucleotide Directed Mutagenesis

Barley lectin cDNA clone blcs (Lerner and Raikhel, 1989), in pUC1 19, which
had previously been altered to add Xbal sites external to the EcoRI sites (Wilkins
et al., 1990) was used as the template for the mutagenesis. Mutagenesis was
performed using the "MUTA-GENE‘“ phagemid in vitro mutagenesis kit (BioRad

laboratories, Richmond, CA). Four oligonucleotides were used to form single

125

.xmtoumm
cm 5 35:2. 2m 9553 28253.3 5 co>_o>c_ 289:. “”23 oEEm 582 5:3 £28 m 3 coumcafioc
9m 325533 How oEEm czamZomcoo ..onE>m _ 9: >2 cmuocmp 9m mEom 93:5 .8352 .223 55:20 582

>233 ho cocoa—om» Eon 05:3 cognac 2: 5.3 3:263 Eon oEEu 582 0.59: 9:95» .o «55:93 .0 2:9".

12E5

« awn-on advooa onuvon mawuuaum an"

Omzb “MOO MOOMmmm4003U3HmuuqumOBOUOOOmUOmO
.

o noocmaooduowumwuGOmuzumooazzao>x00¢<xuomx
c nan-on «sauna magnum an“

« nan-av nave-A cave-a madness» an. a nan-on aqua-a onus-a masseuse an“
WWOZWOWW3WE>wwwmndowamzw>0<ww macmzw MMWW uwwumwmdww3w¢Hwwwquwwfiowwefimwwma
. . _ . u . . «u . . u .u . . . . _ ....»
UOOUOmOOG¢OOH0m0HUOmUSMmOON2280>M0044M00mxnam OdUOOOOfiOOhmmUAUONGZOmOOszmOAXOO<OmUOKH
.C
c addloo danced hoauln Au. n adnlon advocd hcdunn An“

N ndnlcv advuou Guava: madmawun AI“ a nadlov afluuod cauvca madma49u Au—
wan2w0mw3wm>wwwmqoowamzw>w¢wwm Hemxw mwww xwwwmmmtww3w4awwwquwwwcwwufimwwwa

«« _ . u. . n . . .«. u u . .. . . n . u u . .«.
GBmO¢OOOOUQUOMflmUAOOHOSUmUOA22m0A290¢OmuoxHa anmUUOOUCUOHfidUh00N030m00222h08fi00¢0800¢£

9 a t a a
n uddloo aduOOA redunn A». n awnlon dwaOOH hoauun an“
a addloo 64900H Guava: madmadfiu wu_ a uwllov awuoou cau»cu unquudvn AI.

wwszomfi3wm>mwwmqoowwmzw>fi¢UwMNOm3w muww wwwwmmmdwwswzHmwomnflmWBOOOWmewmw
. u ..n . u u .. . .n .. ”u . .u u .. . . ..
admOmUOOOU¢UONMdUh00HUSOmOOmzzmvawiddfiuomxmBHOCUZOUUXUUNnowzcouomamooazszNSZmUOMUOKO
¥ {ii
a awnlov :«uoon hududn Ann M addloa danced hududn Au“

$235.00 02.025-22.50
2_._.Um._ m._._:_.m2 02.62....m 02< 2:.0m._ >m._m<m "—0 202143550

127
deletion mutants of the barley lectin cDNA: 1) 5’-GACCGCGCACGCCCAGCGCTG-

CGGCACTCAGG-S’ to delete domain A (Ant #97-225); 2) 5’—GTCGACTCTAGAATI'-
CCAGCGCTGCGGCACTCAGG-3’, to delete the signal sequence and domain A
(Ant #24 to 225); 3) 5’-CGGCTCCGAGTTCTGCGGTGTC‘ITCGCCGAGG-3’, to
delete the last third of domain C and domain D (Ant #445 to 603); and 4) 5’-CGG-
CTCCGAGTTCTGCTGATGATCTTGCTAATGGC-3’, to delete the last third of

domain C, domain D, and the carboxyl terminal propeptide (Ant #445-651).

Bartey Lectin Gene Double Deletion Mutants

To generate double deletion constructs of barley lectin, Hindlll restriction
sites within the pUC119 polylinker and at nt #379 were utilized as shown in figure
4. The Hindlll fragment from a single deletion construct, made with oligonucleotide
#1 or #2 above, was used to replace the unmutagenized Hindlll fragment from a
single deletion construct made with oligonucleotide #3 or #4. In this way three
modified barley lectin (mBL) genes were assembled. For expression in E. coli, the
mBL gene encoded just domain 8 and two thirds of domain C (DL111, figure 4).
The second mBL gene, for expression in transgenic plants, encoded the barley
lectin signal sequence, domain B, two thirds of domain C and the carboxyl terminal

extension (DL108, figure 4).

Overexpreaion in E. coli

The mBL gene, DL111, was ligated into the EcoRI site of the maltose

binding protein (MBP) expression vector le821 (New England Biolabs, Beverly,

128

Figure 4. Barley lectin double deletion construction. Single deletion
barley lectin constructs were generated by oligonucleotide directed
mutagenesis. The Hindlll (H) fragment containing a deletion was used
to replace the corresponding fragment in the second deletion mutant.
Using this general method two mBL genes were constructed. The first,
for expression in E. coli(DL111) contained the region encoding domain
8, two thirds of domain C followed by a stop codon and the 3’
untranslated region. The second construct (DL108) was designed for
expression in transgenic tobacco. DL108 contained the 5' untranslated
region and the region encoding the signal sequence immediately
followed by the regions encoding domain B, two thirds of domain C and
the 18 amino acid carboxyl-terminal propeptide along with the 3’

untranslated region.

129

BARLEY LECTIN DOUBLE DELETION CONSTRUCTION

Y
”I. l

—r—-

 

 

 

 

 

SINGLE DELETION MUTANT #1
me
i : i
II I
H H
SINGLE DELETION MUTANT 02

 

 

 

W” W“

VV=
ll

DOUBLE DELETION MUTANT

 

 

 

 

 

BARLEY LECTIN: UNMODIFIED

 

 

%ésesEziAlBICI".

DL111: MODIFIED FOR W EXPRESSION

 

{BIC}

 

 

DL1“: MODIFIED FOR PLANT EXPRESSION

+5221 8 l 0 %

 

 

130

MA) and cloned into the E. coli strain MV1193. Induction, amylose column
purification and Factor Xa cleavage of E. coli expressed proteins was performed
according to New England Biolabs protocols. Briefly, one liter of LB,
supplemented with 100 ug/ml ampicillin to select for le821 derivatives, was
inoculated with 10 ml of a saturated overnight culture. The culture was incubated
at 37°C to an optical density at 595 nm (0.0.595) of 0.3 to 0.4, induced with 100
uM IPTG and incubated at 37°C for 2 h. Cells were collected by centrifugation,
resuspended in 50 ml lysis buffer (10 mM sodium phosphate pH 7.2, 30 mM NaCl,
10 mM EDTA, 10 mM EGTA, 0.25% Tween 20), frozen and/or stored at -20° C,
thawed at 37°C and sonicated to release protein from the cells. The soluble
protein extract was cleared by centrifugation at 9,000 xg for 30 min, mixed 1:1 with
amylose column buffer (10 mM sodium phosphate pH 7.2, 0.5 M NaCl, 10 mM 6-

mercaptoethanol, and 1 mM EGTA) with 0.25% Tween 20.

Amylase Column Purification of MBP-Fusion Protein

Soluble protein was mixed with approximately 3 ml rehydrated amylose
column matrix (New England Biolabs) and incubated 10 to 16 h at 4° C. Unbound
protein was removed and saved (effluent fraction). The amylose beads were
collected in a 5 ml Ouick-Sep column (lsolab, OH), washed with 15 ml column
buffer with 0.25% Tween 20 followed by 25 ml column buffer alone. Bound
proteins were eluted from the amylose column with 10 mM maltose in 0.8 to 1 ml
fractions. Fractions containing protein, as shown by 0.0.230 above baseline, were

pooled (eluant fraction), diluted to <200 ug/ml protein with dialysis buffer (100 mM

131
NaCl, 10 mM Tris-HCI pH 8.0, and 1 mM EGTA) and dialyzed extensively at 4°C

against the dialysis buffer. Eluant and Effluent fraction were concentrated in
Centricon 30 microconcentrators and stored at 4° C. For release of mBL from the
maltose binding protein, fusion protein was mixed with 1.5% (w/w) Factor Xa (New
England Biolabs) in 100 mM NaCl, 20 mM Tris-HCI pH 8.0, and 2 mM CaCl2 and
incubated 16 h at RT (Factor Xa cleavage experiments were done by Antje Heese,

Michigan State University).

Chitin Column Purification of mBL Constructs

Affinity purification on chitin was performed according to Peumans et al.
(1984). Chitin column matrix, provided by Dr. Willem Broekaert, was resuspended
in phosphate buffered saline (PBS). Swollen chitin was washed extensively with
PBS, regenerated with 6 M guanidine-HCI with 0.1 M NaOH, and again washed
extensively with PBS. Concentrated amylose column eluate containing 50 ug
protein was diluted to 1 ml with PBS and applied to a 5 ml column with 250 chitin
bed volume. The column was sealed at both ends and rocked 2 h at 4° C.
Column effluent containing unbound protein was collected, the column washed
with 10 ml PBS and bound protein eluted with 6 M guanidine-HCI with 0.1 M
NaOH. Fractions were dialyzed (3500 molecular weight cut-off dialysis membrane)
extensively against H20, lyophilized and resuspended in SDS-PAGE sample buffer
(Chitin-column purification of mBL expressed in E. coli was done by Antje Heese,

Michigan State University).

132
Denaturing Polyactylamlde Gel Electrophoresis (SDS-PAGE) and Western Blotting

SDS-PAGE was performed essentially as described (Laemmli, 1970) using
15% acrylamide resolving gels. Samples were boiled 2 min immediately prior to
electrophoresis. Semi-dry electrophoretic transfer of protein from the gel onto
Immobilon P (Millipore, Bedford, MA) membrane was accomplished with the semi-
dry electroblotter (Millipore) according to the manufacturer’s recommended
procedures. Current density was increased from 10% to the maximum over four
steps (10, 30, 60, 90%) 15 min each to enhance transfer of the small proteins
(mBL is approximately 8.5 kDa). After transfer, remaining protein binding sites on
the transfer membrane were blocked with saturation buffer (3% gelatin and 1%
BSA in tris buffered saline, 150 mM NaCl with 20 mM Tris-HCI pH 8) 1 h at RT and
8 h at 4° C. Bound mBL was detected with anti-WGA antiserum and visualized
with alkaline phosphatase-linked secondary antibody using nitroblue tetrazolium

as the substrate as described (Blake ef al., 1984; Mansfield et al., 1988).

Microsequencing mBL Expressed in E. call

For sequencing small quantities of mBL bound to transfer membrane
different SDS-PAGE and electroblotting conditions were required. Electrophoresis
was done as described above except specially pure, recrystallized SDS (from Dr.
J. Leykem, MSU Macromolecular Structure Facility) was used and time allowed for
polymerization of the acrylamide gels was increased to ensure its complete
polymerization. The resolving gel was polymerized for 10 to 12 h at 4°C and the

stacking gel was polymerized for at least 4 h at RT. A CAPS/methanol buffer

133

system (Matsudaira, 1987) was used in a tank electroblotter (Biorad) for the
transfer of protein from the polyacrylamide gel to the Immobilon P transfer
membrane. This buffer system was required since residues from the Tris-
glycine/methanol normally used interfere with accurate determination of the amino
acid sequence. After transfer to immobilon P, the outermost lanes were removed
and the mBL in these lanes detected with anti-WGA antiserum as described above.
The stained mBL in these lanes was used as a guide for the precise location of the
mBL in the unstained lanes. The mBL in the unstained lanes was cut from the
remaining membrane and used for sequencing. The amino acid sequence of mBL
was determined by Dr. J. Leykem (MSU-Macromolecular Structure Facility) using
an Applied Biosystems gas phase sequenator and reverse phase HPLC as
described by Hewick et al. (1981) (SDS—PAGE and electroblotting of mBL for

microsequencing was done by Antje Heese, Michigan State University).

Expresslon of mBL in Transgenic Tobacco

Modified barley lectin construct DL108 was ligated into the Xbal site of the
plant expression vector pGA643 and transformed into E. co/iMV1193. This put the
mBL gene under control of the CaMV 35$ promoter allowing for constitutive
expression in the transgenic plant. The plasmid was mobilized to Agrobacferium
tumafaciens by triparental mating and leaf disk transformations of tobacco were
performed as previously described (Wilkins et al., 1990). Transformed
regenerated plants were selected for resistance to kanamycin and mBL expressed

in the transgenic tobacco was detected by western blot analysis of leaf extract

134

affinity purified on a chitin column. One gm leaf tissue was ground to a fine
powder in liquid N2, mixed with 2 ml extraction buffer (PBS with 2 mM PMSF and
1% PVP-40) and centrifuged at 15,000 xg 10 min at 4° C. The supernatant was
filtered through Miracloth (Calbiochem, La Jolla, CA) and stored on ice. Chitin
column affinity chromatography and western blot analysis were performed as

described above.

RESULTS

To investigate the relationship between size and in vitro activities of chitin-
binding proteins, modified barley lectin (mBL) genes were constructed. Mature
barley lectin monomers are 18 kDa in size, consisting of four 43 amino acid chitin-
binding domains. Stinging nettle Iectin consists of approximately two chitin-binding
domains containing up to 85 amino acids. The amino acid sequence for only 72
residues of stinging nettle lectin was known at the beginning of this study.
Modified barley lectin genes were constructed to encode polypeptides
approximating the size of the stinging nettle lectin (figure 2). For expression in E.
coli the mBL gene DL111 encoded the 72 amino acids from the start of domain
B (figure 3). For expression in transgenic tobacco the second mBL gene, DL108,
contained the same 72 amino acids along with the BLc3 5’ untranslated region for

correct initiation of translation and the regions encoding the signal sequence and

135

carboxyl-terminal propeptide for correct targeting of the mBL to the vacuole (figure

3).

Expression of mBL in E. coli

The mBL gene designed for expression in E. coli, DL111, was ligated into
the EcoRI site of the maltose binding fusion protein expression vector le821.
Induction with IPTG of the maltose binding protein-fusion proteins, MBP-lacZa (47
kDa, vector without insert) or MBP-mBL (47 kDa), was detected by western blot
analysis of identical SDS-PAGE gels using anti-MBP (figure 5, panel a) or anti-WGA
antibodies (figure 5, panel b). As expected anti-MBP recognized both 47 kDa
fusion proteins, MBP-mBL (panel a, lane 2) and MBP-lacZa control (panel a, lane
7). Anti-WGA specifically recognized MBP-mBL (panel b, lane 2) but did not bind
MBP-lacZa (panel b, lane 7). The 47 kDa fusion proteins were not detected in
uninduced cells (panels a and b, lanes 1 and 6). The MBP-mBL and MBP-lacZa
fusion proteins were mostly present in the soluble fraction (panels a and b, lanes
4 and 9) although due to technical errors little MBP-mBL was detected with the
anti-WGA antiserum on this blot (panel b, lane 4). Some MBP-lacZa was detected
in the insoluble fraction (panel a, lane 8) but no insoluble MBP-mBL could be
detected (panels a and b, lane 3). The 47 kDa fusion proteins specifically bound
to amylose and were eluted from an amylose affinity column with 20 mM maltose
(panels a and b, lane 5; MBP-lacZa data not shown). Native MBP (40 kDa) was
also detected using anti-MBP antiserum (panel a, lane 5). A cross-reactive E. coli

protein of approximately 35 kDa along with several other less abundant proteins

136

Figure 5. Expression of fusion proteln in E. coli. MBP-mBL (lanes 1-5)
or the vector without the mBL, le821 (lanes 6-9) was expressed in E.
coli following induction with lPTG. lmmunoblot analysis of identical gels
was performed with anti-MBP antiserum (panel a) or anti-WGA antiserum
(panel b) and visualized by alkaline phosphatase linked secondary
antibodies. Uninduced cells (lanes 1 and 6) show cross reactive E. coli
proteins. Induced cells (lanes 2 and 7) show the MBP-mBL fusion
protein specifically reacts with each antiserum. Lanes 3 and 8 contain
the insoluble protein fraction. Lanes 4 and 9 contain the soluble protein
fraction which was used for amylose column affinity purification. Affinity
purified MBP-mBL (1 ug) eluted from the amylose column with 20 mM
. maltose is shown in lane 5. All lanes, except #5, contain protein extract

from 5x10“ cells (figure courtesy of Antje Heese).

a DL111 p|H821
ikoa'12 3 4 5"6 7 s 9j

66.2- , . __ _,
242.7- H "I! R -'

31.0-

 

21.5-
: b DL111 le 821
jkoar12 3 43% 789'
66.2-
42.7- ~°-'- r‘

.. —« -- —— —- —
(31.0-

21.5-

138

were often detected with anti-WGA but were most insoluble (panel b, lanes 3 and
8) and did not bind to the amylose column (panel b, lane 5).

The 8 kDa mBL could be released from the amylose column purified fusion
protein, MBP-mBL, by cleavage with Factor Xa (figure 6). Factor Xa cleaved (lanes
1-3) proteins were affinity purified on chitin columns. Proteins in the column
effluent (lane 1), wash (lane 2) and eluate (lane 3) were detected by western blot
analysis with anti-WGA antibodies. The released mBL and the MBP-mBL fusion
protein were both capable of binding to chitin (data for fusion protein not shown).
Chitin column eluate of Factor Xa cleaved protein contains the 8 kDa mBL released
from the fusion protein (lane 3). The efficiency of cleavage by Factor Xa, however,
was too low to allow for the production of large quantities of mBL. Storage of the
fusion protein at 4°C for an extended period (is. several months) also caused the
degradation of the MBP moiety but not the mBL moiety of the fusion protein (data
not shown).

To confirm that the 8 kDa protein detected after cleavage with factor Xa was
the intact mBL, microsequencing of the amino terminal residues of this protein was
performed by the MSU macromolecular structure facility. Five cycles of Edman
degradation detected GIn-Arg-Xm-Gly-Glu at the amino terminal end of the protein.
The four identified amino acids confirmed the sequence of the mBL. This
technique cannot identify Cys residues which further confirms the mBL amino acid

sequence since mBL has Cys as the third residue.

139

Figure 6. Chitin column affinity purified mBL expressed in E. coli.
Amylose column purified protein from cells containing the vector
(le821) with the mBL gene (pDL111) was visualized with anti-WGA
antiserum. Protein fractions were digested with Factor Xa before affinity
chromatography on chitin columns. The chitin column effluent (lane 1)
and wash (lane 2) do not contain the MBP-mBL fusion protein or the
mBL released with Factor Xa. The chitin-binding MBP-mBL (47 kDa)
and mBL (8 kDa) eluted from the column with 6 M guanidine + 0.1 N
NaOH (lane 3). Lane 4 contains 0.75 ug MBP-mBL fusion protein for

size comparison.

140

‘ IMMUNOBLOTANALYSISOFTHEBARLEYLECTIN
CONSTRUCTEXPRESSEDINEcd

kDa

49.4-

27.5-

16.9-

8.1-

141
Expression of mBL in Tobacco Plants

For expression in tobacco plants the mBL gene (DL108) was introduced
into pGA643 under control of the constitutive CaMV 35$ promotor. This vector
also contains the selectable marker nptll which imparts kanamycin resistance to
the transformed plants. Six kanamycin resistant plants were screened for the
presence of active mBL by immunoblot detection of the 8 kDa protein in chitin
column eluate (figure 7). Although a number of abundant cross-reactive chitin-
binding proteins are deteCted by anti-WGA antiserum, the 8 kDa mBL could be
detected in several transgenic plants, especially DL108-5 (lane 4). This protein
was not a native tobacco protein since it could not be detected in chitin column
eluate of leaf extract from an untransformed plant (lane 7) or a plant transformed
with the vector pGA643 alone (lane 9). Protein from a transgenic plant expressing
the full length barley lectin was affinity purified on chitin and provided a positive
control for the western blot (lane 8). The levels of expression of the mBL in
transformed tobacco plants, however, was very low. Plants produced by selfing

the transformants also failed to produce high levels of mBL (data not shown).

DISCUSSION

A family of chitin-binding proteins is found in a wide variety of plants. These
proteins all contain a Cysteine and Glycine rich' domain of approximately 43 amino

acids. The chitin-binding domain has been fused to an unrelated domain in

142

Figure 7. Expression of mBL in transgenic tobacco. lmmunoblot
detection with anti-WGA antiserum of leaf extract proteins which bound
to a chitin column. Lanes 1-6 are extracts from separate kanamycin
resistant plants transformed with DL108. The 8 kDa chitin-binding mBL
can be seen in several lanes and was most abundantly present in a
single plant (lane 4). The control lanes contain extracts from an
untransformed plant (lane 7), a plant expressing large quantities of the
full length barley lectin gene (lane 8), and a plant transformed with the

vector, pGA643, without any insert (lane 9).

obicl
Jund
iya‘n

mBL

kDa

' 27.5-

143

IMMUNOBLOT ANALYSIS OF THE BARLEY LEC'HN
CONSTRUCT EXPRESSED IN TRANSGENIC TOBACCO

TRANSGENIC PLANTS CONTROLS

49.4-

16.9-
8.1-

 

144

prohevein and the class I chitinases. Other proteins in this family are made up of
the chitin-binding domains alone. The very small protein hevein, 4 kDa, is made
up of a single chitin-binding domain and can be isolated from rubber tree latex by
its affinity to chitin. Because of this, chitin-binding domains are often referred to
as ’hevein’ domains. Stinging nettle lectin has approximately two chitin-binding
domains repeated in tandem while the Gramineae lectins, such as barley lectin,
consist of four chitin-binding domains in tandem. The chitin-binding proteins are
thought to play roles in plant defense since plants contain no chitin whereas many
plant pathogens and pests do (Chrispeels and Raikhel, 1991). Recently several
groups have shown that several proteins which consist solely of the chitin-binding
domains (hevein, stinging nettle lectin, and the Gramineae lectins) exhibit in vitro
antifungal and/or insecticidal activities (Table I). The Gramineae lectins can inhibit
the growth of insect larvae (Huesing et al., 1991 a) but do not show any antifungal
activity (Schlumbaum et al., 1986). Stinging nettle lectin, which at 8.5 kDa is
approximately half the size of the Gramineae lectin monomers, inhibits the growth
of both insect larvae (Huesing et al., 1991 b) and some fungi (Broekaert et al.,
1989). Hevein can also inhibit the growth of some fungal hyphae (Van Paris at al.,
1991) but has not been tested for insecticidal activity.

To examine the relationship between protein size and these in vitro activities,
the barley lectin gene was modified to encode a protein approximately the size of
the nettle lectin. At the onset of this study only partial sequence data for the
stinging nettle lectin was known. Chapot et al. (1986) determined the location of

72 stinging nettle lectin amino acids residues on 6 tryptic fragments. The

145

sequence of these residues was further confirmed by Dr. Steven Michnick who
also determined the order of the tryptic fragments (personal communication). This
amino acid sequence data was aligned with the barley lectin cDNA deduced amino
acid sequence (figure 4). When aligned with the barley lectin domains containing
the primary ligand-binding sites (B and C), the 72 amino acids of stinging nettle
lectin sequence contains residues homologous to those directly involved in
Gramineae lectin ligand binding. Thus the mBL gene was designed to encode the
72 amino acids from the start of domain B. For expression in transgenic plants
the mBL gene also encoded the barley lectin signal sequence, for translocation
into the ER, and carboxyl-terminal propeptide, for targeting the mBL to the
vacuole.

The mBL was expressed in E. coli to quickly determine if an active, chitin-
binding protein would be produced. In addition, large quantities of mBL could be
obtained simply by increasing the size of the bacterial culture. By using the
maltose binding fusion protein expression system the fusion protein could be
affinity purified on an amylose column (figure 5). This was an important design
consideration since it was thought the mBL would very likely be inactive when
produced in E. coli and require a denaturation/renaturation procedure to isolate
a chitin-binding protein. Once the fusion protein was collected the mBL could be
isolated by cleavage with the protease Factor Xa and the MBP removed by its
affinity to amylose.

It is striking that both the MBP-mBL fusion protein and mBL alone could

bind to the chitin column. Several important conclusions can be drawn from these

146

data. First, it is possible to modify the barley lectin, eliminating half of the protein,
while retaining its ability to bind chitin (Figure 6). Second, the mBL expressed in
bacteria does not require denaturation/renaturation to be active. Disulfide bonds
of eukaryotic proteins are not always assembled correctly when expressed in E.
coli (T suji et al., 1987). The mBL should have seven disulfide bonds which,
presumably, are required for correct conformation into the chitin-binding protein.
Finally, the finding that a carboxyl-linked mBL, in the MBP-mBL fusion protein, can
impart chitin-binding activity to the MBP polypeptide (data not shown) is important
for the potential of adding chitin-binding activity to other proteins. All known
proteins containing the chitin-binding domain and an unrelated domain, ie hevein
and the class I chitinases, have the chitin-binding domain at the amino-terminal
end of the protein. Hevein linked to the carboxyl-terminal of MBP has also been
shown to bind chitin (Lee et al., 1991) indicating this versatility may be a general
property of the chitin-binding domains. The potential for future protein engineering
to provide chitin-binding activity to other unrelated proteins is enhanced by this
property since placement of the chitin-binding domains could be at either the
amino or carboxyl ends of the polypeptide.

The bacterial expression vector used to express the mBL is designed to
encode a Factor Xa cleavage site between the MBP and the protein of interest.
Amino-terminal sequencing of the mBL released by Factor Xa was used to confirm
that protease cleavage had occurred correctly. The quantity of Factor Xa required,
however, was prohibitively expensive for collecting large amounts of the mBL.

Because of this restraint, antifungal assays of the E. coli expressed mBL were not

147

continued. Preliminary antifungal assays performed with the MBP-mBL fusion
protein did not show significant antifungal activity (data not shown).

Expression and analysis of mBL in transgenic plants was the ultimate goal
of this investigation. According to the WGA model for chitin-binding the mBL
would have to be correctly synthesized, processed, and assembled into the dimer
in order to possess chitin-binding activity. As shown in figure 7 several of the
kanamycin resistant, transgenic tobacco plants expressed the 8 kDa chitin-binding
mBL. Again these data demonstrate that, working from the known biochemical,
crystallographic, and sequence data of the chitin-binding proteins, it is possible to
’engineer' a protein only 1/2 the size of the native barley lectin but retaining chitin-
binding activity when expressed in tobacco. The active, chitin-binding mBL in the
transgenic plants was detectable, however, only with the highly sensitive western
blot analysis. This level of expression was too low for use in the in vitro antifungal
assay. In addition, before enhanced fungal resistance in the transgenic plants
would be expected, mBL should accumulate to the level required for stinging
nettle’s in vitro antifungal activity (50 ug/ml or approximately 50 ug/gm FVV). Since
the amount of the mBL in one gm leaf tissue was barely detectable with the
western blot analysis it was clear that enhanced fungal resistance due to the mBL
could not be expected.

Future studies will include enhancing the expression of the mBL gene in
transgenic tobacco by use of a stronger promotor/enhancer system. In addition,
recent cloning of the stinging nettle cDNA (Lerner and Raikhel, 1992) has provided

more accurate and complete sequence information for this protein. These data will

148

be used to construct a second generation of mBL genes which will be expressed
in transgenic plants. These studies will allow us to determine if the antifungal
activity shown for nettle lectin but not for the Gramineae lectins, including barley
lectin, is primarily due to the size of the active protein. Although the goal of
determining whether the mBL exhibits antifungal activity was not achieved, this
study has demonstrated the feasibility of substantially modifying the barley lectin

while retaining the chitin-binding activity.

1 49
REFERENCES

Blake, M.S., Johnston, KH., Russel-Jones, G.J., and Gotshclich, EC. (1984). A
rapid, sensitive method for detection of alkaline phosphatase-conjugated

anti-antibody on western blots. Anal. Biochem. 136:175-179.

Broekaert, W.F., Van Parijs, J., Leyns, F., Joos, H., and Peumans, W.J. (1989). A
chitin-binding lectin from stinging nettle rhizomes with antifungal properties.

Science 245:1100-1102.

Chapot, M.P., Peumans, W.J., and Strosberg, AD. (1986). extensive homologies

between lectins from non-leguminous plants. FEBS Lett. 195:231-234.

Chrispeels, M.J. and Raikhel, NV. (1991). Lectins, lectin genes, and their role in

plant defense. Plant Cell 321-9.

Goldstein, l.J. and Hayes, CE. (1978). The lectins: Carbohydrate-binding
proteins of plants and animals. Adv. Carbohydr. Chem. Biochem. 35:120-

340.

Hewick, R.M., Hunkapiller, M.W., Hood, LE. and Dreyer, W.J. (1981). A gas-liquid

solid phase peptide an protein sequenator. J. Biol. Chem. 256:7990—7997.

150
Huesing, J.E., Murdock, LL, and Shade, R.E. (1991a). Effect of wheat germ

isolectins on development of cowpea weevil. Phytochemistry 30:785-788.

Huesing, J.E., Murdock, LL, and Shade, R.E. (1991b). Rice and stinging nettle
lectins: Insecticidal activity similar to wheat germ agglutinin.

Phytochemistry 30:3565-3568.

Laemmli, UK. (1970). Cleavage of structural proteins during the assembly of the

head of bacteriophage T4. Nature 227:680-685.

Lee, H.-l., Broekaert, W.F., and Raikhel, NV. (1991). Co- and post-translational
processing of the hevein preproprotein of latex of the rubber tree (Hevea

brasi/iensis). J. Biol. Chem. 266:15944-15948.

Lerner, DR. and Raikhel, NV. (1989). Cloning and characterization of root-specific

barley lectin. Plant Physiol. 91 :124-129.

Lerner, DR. and Raikhel, NV. (1992). The gene for stinging nettle lectin (Urtica
dioica agglutinin) encodes both a lectin and a chitinase. J. Biol. Chem., in

press.

151
Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982). Molecular cloning: A

laboratory manual, 1" edition. Cold Spring Harbor Laboratory. Cold Spring

Harbor, NY.

Mansfield, MA, Peumans, W.J., and Raikhel, NV. (1988). Wheat germ agglutinin

is synthesized as a glycosylated precursor. Planta 173:482-489.

Matsudaira, P. (1987). Sequence from picomole quantities of protein
electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem.

262: 1 0035-1 0038.

Mishkind, M.L., Palevitz, BA, and Raikhel, NV. (1983). Localization of wheat germ
agglutinin-like lectins in various species of the gramineae. Science

20: 1 290-1 292.

Money, NP. (1990). Measurement of pore size in the hyphal cell wall of Ash/ya

bisexualis. Experimental Mycology 14:234-242.

Peumans, W.J., De Ley, M., and Broekaert, W.F. (1984). An unusual lectin from

stinging nettle (Urtica dioica). FEBS Lett. 177299-103.

152
Peumans, W.J., Stinissen, H.M., and Carlier, A.R. (1982a). Isolation and partial

characterization of WGA-like lectins from Rye (Secale cereals) and barley

(Hordeum vulgare) embryos. Biochem. J. 206239-243.

Peumans, W.J., Stinissen, H.M., and Carlier, A.R. (1982b). Subunit exchange

between lectins from different cereal species. Planta 154:568-572.

Reeke, G.N. Jr. and Becker, J.W. (1988). Carbohydrate-binding sites of plant

Iectins. Curr. Top. in Microbiol. and lmmunol. 139235-58.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular cloning: A
laboratory manual, 2"d edition. Cold Spring Harbor Laboratory Press. Cold

Spring Harbor, NY.

Schlumbaum, A., Mauch, F., vogeli, U., and Boiler, T. (1986). Plant chitinases are

potent inhibitors of fungal growth. Nature 324:365-367.

Tsuji, T., Nakagawa, R., Sigimoto, N., and Fukugara, K. (1987). Characterization
of disulfide bonds in recombinant proteins: Reduced human interleukin 2

in inclusion bodies and its oxidative refolding. Biochemistry 26:3129-3134.

153
Van Parijs, J., Broekaert, W.F., Goldstein, l.J., and Peumans, W.J. (1991). Hevein:

An antifungal protein from rubber tree (Hevea brasi/iansis) latex. Planta

1 83:258-264.

Van Damme, E.J.M., Broekaert, W.F., and Peumans, W.J. (1988). The Urtica
dioica agglutinin is a complex mixture of isolectins. Plant Physiol. 86:598-

601.

Wilkins, T.A., Bednarek, S.Y., and Raikhel, NV. (1990). Role of propeptide glycan
in post-translational processing and transport of barley lectin to vacuoles

in transgenic tobacco. Plant Cell 22301-313.

Wright, 08. (1980). Crystallographic elucidation of the saccharide-binding mode
on wheat germ agglutinin and its biological significance. J. Mol. Biol.

1 41 :267-291 .

Wright, CS, and Raikhel, NV. (1989). Sequence variability in three wheat germ
agglutinin isolectins: Products of multiple genes in polyploid wheat. J. Mol.

Evol. 28:327-336.

CHAPTER 5

SUMMARY AND FUTURE PROSPECTS

154

155

The data presented in this dissertation support the hypothesis that Barley
Iectin and stinging nettle lectin are involved in plant defense. Barley lectin
expression was localized to the root-tips and the coleorhiza of the embryo by in
situ hybridizations. This indicates that its function may relate to interactions with
chitin containing soil microbes. Expression of stinging nettle lectin mRNA in the
seeds and rhizomes, which also contact the soil, was reminiscent of barley lectin
expression. The finding that the stinging nettle Iectin mRNA encoded an
unexpected chitinase domain was also suggestive of a plant defense function for
this protein. This is particularly the case since stinging nettle lectin and chitinase
act synergistically to inhibit fungal growth during in vitro experiments (Broekaert,
1989). Data was also presented which showed that barley lectin can be
substantially modified and still bind chitin when expressed in either E. coli or
tobacco. This was the case after deletion of over half the protein and fusion to the
carboxyl-terminus of a maltose binding protein indicating that the chitin-binding
proteins are quite amenable to molecular manipulations. Design and construction
of artificial genes encoding chitin-binding domains alone or chitin-binding domains
fused to unrelated domains is now a reality.

The unexpected structure of the protein encoded by the stinging nettle
Iectin mRNA leads us to a number of questions. First the processing of the
prostinging nettle lectin polypeptide needs to be addressed. Production of
prostinging nettle Iectin domain specific antibodies will provide the tools necessary

for determining the fate of the individual prostinging nettle Iectin domains. There

156

are many interesting questions concerning these processing events. Is the
prolectin processing tissue specific or developmentally regulated? It will be
particularly interesting to determine if expression or processing of the stinging
nettle lectin is influenced by ABA, ethylene, wounding or pathogens. Also, do the
38 kDa prostinging nettle lectin or the 28 kDa chitinase domain accumulate in the
plant? If so, then it will be important to determine the tissue specificity,
intracellular localization and whether the chitinase domain is active. Answering
these questions will help in understanding the endogenous function of the stinging
nettle lectin.

Stinging nettle Iectin may be useful in engineering plants with enhanced
fungal or insect resistance. It will be important to determine if processing of
prostinging nettle Iectin in transgenic plants is the same as in the stinging nettle
plant. Perhaps the most probing question is whether stinging nettle lectin, or the
proprotein, can enhance the plant’s resistance to microbial pathogens or insect
pests. The possibility exists that stinging nettle lectin will act in concert with
endogenous chitinases in the plant. With the expression of large quantities of
stinging nettle lectin and barley lectin, with various modifications, in transgenic
plants it should be possible to determine why stinging nettle lectin inhibits fungal
growth but barley lectin does not.

Chitin-binding proteins in plants are a widespread and varied family of
proteins with potential functions in plant defense. Work presented in this
dissertation has broadened the variety of protein structures we can expect within

this family. In addition, these investigations have opened up new avenues of

 

'—

157

research by showing the variability possible through molecular manipulations of
chitin-binding protein genes. These are, perhaps, the first steps toward a general
understanding of the endogenous functions of chitin-binding proteins and their

potential uses in genetically engineered crops.

APPENDK A

SYNTHETIC STTNGING NETTLE LECTTN GENE
SEQUENCE AND OLIGONUCLEOTIDE STRUCTURE

158

159

APPENDIX A

 

.m hh<uhuw<co<<hbo<ph<<u<<h<whu<hh<uob00huu<u<uu<uuu

 

 

.n <<hu<thuuhh<<oh<<b...whp<Tu<op<<pwo<oc<wopwhowpowo

 

a 03x . Laos“ .

 

<h¢<<uwohod<<u<uop<<ohhuhuuou<<whhoocphpuu<<uhho<ou

 

 

 

 

 

»<0»huum<uohhwhwo<hbu<<u<oooohLu<<uou<<<ou-o<<uwuw

 

 

»uuuu<hUhhhuhuuwohuu<ou<hnrpoh<0howu<huuuphuuu<u<<p<

 

 

 

 

<¢¢uw~<u<<<o<oouw<owbum»<.u<u<pc<uuw»<uoo<<ooohoh»<p

 

 

 

 

hwoouheuh<oohuo<u<ou<pouuu<<<has»ou<<u<uuh<<o<ou<uo
(8988(882huu<oupopoup<u£wohpc<uu<oohasheo?M»8puopuu

 

 

 

 

 

 

 

 

 

du<¢mqpuuadddddqddfiAdAJJAdddddddqdAddd<h»uuuuouo<uu

uubhu<<owuoohow<uw.<<u»0h<oouuhow<w<uuhh<<cuoo090huo

 

 

 

 

Zoom » g u 85

52535 as :owfiszowSO <9: 02532.5

I’IICHIGRN STRTE UNIV. LIBRRRIES
illilliiilllillliiliiillliIilliiilllililillllllllllillllll
31293008770087