INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality Illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600 UMI®
NOTE TO USERS
Page(s) not included in the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received.
105
This reproduction is the best copy available.
UMI
MODE OF ACTION OF Cry2Aa, A BACILLUS THURINGIENSIS DUAL ACTIVE
INSECTICIDAL CRYSTAL PROTEIN
DISSERTATION
Presented in partial Fulfillment of the Requirements for the Degree
Doctor of Philosophy in the Graduate School of The Ohio State University
BY
Mongkon Audtho, B.S., M.S.
The Ohio State University
2001
Dissertation Committee: Approved by
Dr. Donald H. Dean, Adviser o\ .k
Dr. George A. Marzluf Adviser
Dr. Venkat Gopalan Department of Biochemistry UMI Number: 3011021
UMI
UMI Microform 3011021 Copyright 2001 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17. United States Code.
Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 ABSTRACT
Cry2Aa, a Bacillus thuringiensis 5-endotoxin, is specific to both dipteran and lepidopteran insects. Upon digestion by gypsy moth {Lymantria dispar) midgut proteases, the first 49 amino acids at the N-terminus are removed from the 63 kDa protoxin, to form a 58-kDa active fragment. The protease-resistant core is comprised of ValSO as the first amino acid, which is located at the loop between the aO and a1 helices of domain I. The toxin’s potency was lost when further in vitro midgut protease processing continued. The loss in toxicity was caused by proteolytic cleavage at the carboxyl end of Leu 144, which is on the loop between aS and a4 of domain I. To prevent the production of the non-toxic fragment, five mutant proteins, L144D, L144A, L144G, L144H, and LI44V, were constructed. All of the mutant proteins were highly resistant to the midgut proteases and chymotrypsin. However, the mutant proteins were as toxic to the insects as the wild type protein, indicating that the cleaved fragment is associated with the rest of the entire protein molecule in the midgut environment.
The cleavage was also protected when the toxin is bound to the midgut membrane. The role of the 49-amino acid length of the N-termina! sequence of
Cry2Aa was investigated. This short peptide sequence contains one extra a helix, called aO, which is located between Phe24 and His46. Gene deletion and site-directed mutagenesis results indicated that the presence of this a helix is
necessary for the formation of biological active inclusion of Cry2Aa. Deletion of the cry2Aa gene at the aO helix, and deletion of the entire N-terminus region led
to production of the non-soluble inclusion bodies that were non-toxic to both L.
dispar and A. quadrimacuiatus. However, expression cry2Aa gene that contains
the aO helix, but lacking the Val3 - Asn17 region, gave the soluble, and toxic
crystal inclusions. Tertiary structure of Cry2Aa reveals a number of
intramolecular and intermolecular interactions contributed by the amino residues
in aO. The intramolecular interactions might stabilize, and facilitate folding of the
Cry2Aa protoxin, as demonstrated by the observation that the melting
temperature of the Cry2Aa protoxin and the mutant protein containing aO were 7
°C higher than the active Cry2Aa fragment. Mutation of amino acids involved in
intramolecular interactions lowered the yield of the inclusion bodies. Loss in the
inclusion amount was also found when the residues involved in intermolecular
bondings were mutated.
Finally, results from intensive site-directed mutagenesis of amino acids on
loop 1 and loop 2 of domain II and bioassays against A. quadrimacuiatus and L
dispar revealed that specificity of Cry2Aa to dipteran insects was conferred by amino acids in loop 1, while the amino residues on loop 2 are responsible for
Lepidoptera specificity. The mutant proteins A2 (F 320-321AAA), F320A, and P321A were less toxic to A. quadrimacuiatus larvae. Change in toxicity was related to
change in competition binding affinity. However, the same mutant proteins were
as toxic to L dispar larvae as the wild type protein and change in binding
properties to L. d/spar BBMV was not observed. In contrast, the mutant proteins
obtained from the alterations in the region, i. e., A5 (DRE 383-5AAA),
D383V, R384A, and E385A, were less toxic to only L dispar. Results from
dissociation experiments indicated that these mutant proteins were not involved
in irreversible binding. The dramatic changes found in these mutant proteins
were reversible binding. Therefore, Initial binding of amino acids on loop 1 and
loop 2 of Cry2Aa might be a factor that determines the toxin dual specificity.
The obtained results concerning active fragment processing, solubility,
and specificity of Cry2Aa might reveal a fundamental mode of action of Cry2Aa
and other Cry toxins from Bacillus thuringiensis. Active fragments of Cry toxins
seem to contain all seven a-helices to retain their biological activity. The results
also indicated that formation of the bioactive crystal in the bacterial cells requires
not only the active fragment portion, but also the extra-portion of the protoxins.
These peptide portions might facilitate both protein folding and assembly of the
protoxin molecules. Finally, the results obtained from the Cry2Aa-receptor
binding and toxin specificity experiments might help in engineering more potent
and more specific Cry toxins, which will be used to control the certain target
insects.
iv Dedicated to my father ACKNOWLEDGMENTS
I wish to express my sincere appreciation to my advisor, Dr. Donald H.
Dean for his guidance and constant inspiration throughout my graduate study. I am especially grateful to him for his patience and understanding the situation I have had during staying in this school.
I gratefully acknowledge Dr. George Marla, and Dr. Venkat Gopalan for their helpful suggestion and constructive criticism on my dissertation, and their serving in my committee.
I thank Dr. Tipvadee Attathom, and Dr. Supat Attathom, who gave me a chance to continue my studies in the US. I would like to thank the Royal Thai
Government for financial support during the early period of my study.
I thank Dr. Daniel Ziegler, from Bacillus Stock Center, for providing some financial support and information about Bacillus thuringiensis.
I gratefully acknowledge Dr. Richard Morse and Dr. Robert Stroud,
University of California, San Francisco, for kindly providing the very helpful
information of Cry2Aa structure.
I thank Dr. Algimantas Valaitis, USDA, Delaware, Ohio, for protein
sequencing.
VI I thank Frank Martin and Win McKlane, from USDA, Massachusetts, for kindly providing gypsy moths.
I thank April Curtiss to her assisting in almost everything during my staying in this lab. I also thank Joy, Chung-Wen Wu for her assistance in protein purification and gypsy moth bioassays.
I thank Mohd Amir Abdullah for mosquito samples, and Norapan Kittivat for Cry toxin alignment diagrams.
I thank Dr. Oscar Alzate for biophysics experimental data and Dr. Mark
Foster for the protein unfolding studies facilities.
I am grateful to Dr. Mi K. Lee, Dr. Taek You for some suggestion and making a friendly environment in this lab.
Many thanks to my former and present laboratory colleagues for their help and best wishes: Dr. Jeremy Jenkins, Sang Soo Oh, Sylvia Liu, Marry Beth
Dunn, and Carroll Ziegler.
Special thanks go to Tara Grove for her friendly help in everything.
Finally, I thank my family members who took care of my father during 5 years of my staying in the US.
VII VITA
May 13, 1966 ...... Born-Mukdahan, Thailand
1984 - 1988 ...... 8 . S. Biology
Khon Kaen University
Khon Kaen, Thailand
1988 - 1992 ...... M.S., Biochemistry
Mahidol University
Bangkok, Thailand
1992 -1 9 9 5 ...... Research associate
National Center for Genetic
Engineering and Biotechnology
Bangkok, Thailand
1995 - 1997 ...... Fellowships
Royal Thai Government
1997 - 2001 ...... Graduate teaching Associate or
Graduate Research Associate
Department of Biochemistry
The Ohio State University
Columbus, Ohio
VIII PUBLICATIONS
1. Audtho, M., A. Tassanakajon, V. Boonsaeng, S. Piankijagum, and S.
Panyim. 1995. Simple non radioactive DNA hybridization method for
identification of sibling species of Anopheles dirus (Diptera: Culicidae)
complex. J. Med. Entomol. 32: 107-111.
2. Audtho, M., A. P. Valaitis, O. Alzate, and D. H. Dean. 1999. Production
of chymotrypsin-resistant Bacillus thuringiensis Cry2Aa1 5-endotoxin by
protein engineering. Appl. Environ. Microbiol. 65: 4601-4605.
FIELD OF STUDY
Major Field: Biochemistry
IX TABLES OF CONTENTS
Page
Abstract ...... ii
Dedication ...... v
Acknowledgments ...... vi
Vita ...... viii
Publications ...... ix
Table of Contents ...... x
List of Tables ...... xiii
List of Figures ...... xv
Chapters:
1 Introduction...... 1
1.1 Bacillus thuringiensis and cry genes ...... 1
1.2 Classification of insecticidal crystal proteins of
Bacillus thuringiensis...... 4
1.3 Sequence similarity and structures of crystal proteins ...... 6
1.4 Mode of action and domain function of crystal proteins ...... 8
1.6 Receptors for Cry proteins ...... 14
1.6 Membrane insertion and pore formation ...... 17
X 1.7 Specificity determining Region of Crystal Proteins ...... 19
1.8 The effect of mutations at specificity determining
region on toxicity ...... 22
Correlating stability and toxicity of Bacillus thuringiensis Cry2Aa toxin against gypsy moth {Lymantria dIspar) larvae ...... 25
2.1 Summary ...... 25
2.2 Introduction...... 27
2.3 Materials and methods ...... 29
2.4 Results ...... 34
2.5 Discussion...... 45
The role of the aO helix of Bacillus thuringiensis Cry2Aa protoxin in solubility and toxicity against gypsy moth {Lymantria dispar), and mosquito {Anopheles quadrimacuiatus) larvae ...... 50
3.1 Summary ...... 50
3.2 Introduction...... 52
3.3 Materials and methods ...... 57
3.4 Results ...... 64
3.5 Discussion...... 74
Dual specificity of Bacillus thuringiensis Cry2Aa to gypsy moth
Lymantria dispar) and mosquito {Anopheles quadrimacuiatus) larvae
is determined by different initial binding sites on loop land loop 2 of its domain II ...... 79
4.1 Summary ...... 79
xi 4.2 Introduction...... 81
4.3 Materials and methods ...... 84
4.4 Results ...... 90
4.4 Discussion...... 119
List of references ...... 125
XII LIST OF TABLES
Table Page
1.1 Insect specificity determining regions of Cry toxins ...... 23
2.1 Toxicity against L. dispar of Cry2Aa protoxin and protein
fragments from the processing of larvalmidgut extract ...... 37
2.2 Toxicity of Cry2Aa and the mutant proteins against L. dispar
larvae ...... 44
3.1 Oligonucleotides for site-directed mutagenesis and inverted
polymerase chain reaction ...... 60
3.2 Toxicity of Cry2Aa, its mutant, and N-terminus deleted proteins
Against gypsy moth {Lymantria dispar), and
Anopheles quadrimacuiatus larvae ...... 66
3.3 The role of amino acids of the N-terminus of Cry2Aa protoxin
in formation of intermolecular and intramolecular
hydrogen bonds and salt bridges ...... 67
4.1 Toxicity of Cry2Aa and its mutant proteins to gypsy moth (L. dispar),
and mosquito {A. quadrimacuiatus) larvae ...... 96
4.2 Toxicity of the mutant proteins from alanine-scaning of the
amino acids on loop 1 of Cry2Aa ...... 98
xiii 4.3 Toxicity of the mutant proteins from alanine-scanning
of the amino acids on loop 2 of Cry2Aa ...... 100
4.4 Effect of mutations at D383, R384, and E385 on toxicity
against L d/spar larvae ...... 101
4.5 Effects of mutations of F320, and P321 on toxicity against
A. quadrimacuiatus larvae ...... 104
4.6 Relative toxicity and binding affinity of Cry2Aa mutant proteins
to L dispar, and A. quadrimacuiatus...... 108
XIV LIST OF FIGURES
Figure Page
1.1. Phylogram demonstrating amino acid sequence similarity
among Cry and Cyt proteins ...... 5
1.2. Position of conserved blocks among Cry protein ...... 7
1.3 Structures of CrylAa, Cry2Aa, and CrySAa toxins ...... 9
1.4. Tertiary structure-guided sequence alignment across
domain I of Cryl Aa, Cry2Aa, and CrySAa ...... 10
1.5 Tertiary structure-guided sequence alignment across
domain II of Cryl Aa, Cry2Aa, and CrySAa ...... 11
1.6 Tertiary structure-guided sequence alignment across
domain III of Cryl Aa, Cry2Aa, and CrySAa ...... 12
1.7 Sequence of events characterizing the mode of
action of B.t. Cry toxins...... 13
2.1 10% SDS-polyacrylamide gel electrophoresis of the processing
products of Cry2Aa and Cryl Aa by gypsy moth larval midgut
extract at different periods of incubation ...... S6
XV 2.2 Autoradiography of L. dispar larval midgut, and the digested
Cry2Aa after feeding on diet contaminated with
^^^l-labeled Cry2Aa for 1 h...... 39
2.3 Digestion of Cry2Aa protoxin and its mutants by gut extract
and chymotrypsin ...... 42
3.1 Protein structure of Cry2Aa protoxin shown in ribbon and
space-filled model ...... 56
3.2 Amino acid sequence of the amino terminus of Cry2Aa, and
its mutant proteins ...... 58
3.3 Solubility of Cry2Aa, mutant proteins and its N-terminus-deleted
protein in 50 mM NaaCOa pH 10.5 ...... 65
3.4 Thermal unfolding of the Cry2Aa protoxin, Cry2Aa active
fragment, and D5 ...... 71
4.1 A Structures of Cryl Aa and Cry2Aa shown in ribbon ...... 91
4 .IB Structures of CrylAa and Cry2Aa shown in space-fill ...... 92
4.10 Structures of CrylAa and Cry2Aa showing the position
of 367pp|369 Qf CrylAa, and Cry2Aa and amino
acids residues on loop 1 and loop 2 of both proteins ...... 93
4.2 Sequence alignment of loop 1 and loop 2 regions of
CrylAa and Cry2Aa ...... 94
4.3 Diagram showing triple alanine replacement of amino acid
residues on the loop 1, and loop 2 regions, of Cry2Aa ...... 95
XVI 4.4 Digestion of the A2 and A5 protein by proteases from midgut
juice of the L dispar, and A. quadrimacuiatus larvae ...... 105
4.5 Saturation binding of ’^®l-labeled Cry2Aa to L dispar BBM\/
as a function of the concentration of BBMV ...... 107
4.6 Binding of ’^®l-iabeled Cry2Aa toxin to L. dispar BBMV in the
presence of increasing concentrations of nonlabeled Cry2Aa,
A2, F320A, P321A, N322A, A5. R384A, and E385A ...... 109
4.7 Binding of ^^®l-labeled Cry2Aa toxin to A. quadrimacuiatus BBMV
in the presence of increasing concentrations of nonlabeled Cry2Aa,
A2, F320A, P321A, N322A, A5, R384A, and E385A ...... 110
4.8 Saturation binding of ^^®l-labeled Cry2Aa to A. quadrimacuiatus
BBMV as a function of the concentration of BBMV ...... 113
4.9 Binding of ’^®l-labeled Cry2Aa toxin to A. quadrimacuiatus BBMV
in the presence of increasing concentrations of nonlabeled Cry2Aa,
A2, F320A. P321A, A5, R384A, and E385A ...... 114
4.10 Binding of ’^®l-labeled Cry2Aa toxin to A. quadrimacuiatus BBMV
in the presence of increasing concentrations of nonlabeled Cry2Aa,
F320C, F320S. F320Y, P321H, and P321N ...... 115
4.11 Dissociation of bound ^^®l-labeled toxins from L d/spar BBMV ...... 118
4.12 Dissociation of bound ^^®l-labeled toxins from
A. quadrimacuiatus BBMV...... 119
XVII CHAPTER 1
INTRODUCTION
1.1 Bacillus thuringiensis and Cry Genes
Bacillus thuringiensis {B.t.) is a gram-positive bacterium, and has been found in many habitats, including insects, soil, stored- product dust, coniferous leaves, and insect hemocoel (Stephen, 1952; Delucca etal., 1984; Martin and
Travers, 1989; Smith and Couche, 1991). The role of this bacterium in the ecosystem is not well understood. It has been a matter of contention as to whether this microorganism should be considered as an entomopathogen, as a phyloplane inhabitant, or just as a soil microorganism.
S. thuringiensis produces an enormous variety of intracellular proteinaceous crystalline inclusion bodies known as 5-endotoxin. The crystal is composed of proteins, called crystal (Cry) proteins. Although most Cry proteins are produced during sporulation, some, most notably CrySAa, are produced during vegetative growth phase. In some strains of S. thuringiensis, a number of pesticide proteins unrelated to the Cry proteins are also produced during vegetative growth. These include proteases, chitlnases, secreted vegetative insecticidal proteins (VIPs), and an antifungal compound (Lovgren etal., 1990;
Stabb et al., 1994; Estruch et al., 1997; Wiwat et al., 2000). However, among a number of virulence factors produced by this microorganism. Cry proteins are the most prominent factor that allows the development of the bacteria in dead or weakened insect larvae.
S. thuringiensis strains have genomic size of 2.4 to 5.7 million base pairs, and have several extrachromosomal elements, some of them circular and other linear. (Carlson etal., 1994). Most Cry proteins are encoded by large plasmids, and some by genomic DNA (Gonzalez et al., 1981). The first cry gene, crylAa, was cloned in 1981 by Schnepf and Whiteley (Schnepf and Whitelev, 1981). The gene was expressed in E. coll, and the protein extracts were demonstrated to be toxic to tobacco hornworm {Manduca sexta). Since then, hundreds of B. thuringiensis strains have been isolated and a number of cry genes have been cloned. To date, more than a hundred cloned c/y genes have been reported
(Chckmore et al., 1998).
The products from cry gene expression can account for 20 to 30% of the dry weight of the cells. The very high level of crystal protein synthesis is controlled by a variety of the regulatory mechanisms, which may be broadly outlined at three levels: transcriptional, post-transcriptional, and post-translational levels. Most cry genes are sporulation-specific genes. However, low-level transcription of some cry genes has been detected during vegetative stage
(Poncet et al., 1997). The development of sporulation is controlled by the successive activation of sigma factors, which bind the core RNA polymerase to direct the transcription from sporulation-specific promoter (Morgan, 1993). In contrast, expression of the cry3Aa gene is not controlled by sporulation-specific promoter, but by a promoter that is recognized by the primary sigma factor of vegetative cells, (Agaisse and Lereclus, 1994).
One of the post-transcriptional factors that contribute to high expression of the Cry proteins is the long half-life of cry mRNA. The half-life of cry mRNA is about 10 min, at least five fold greater than the half-life of an average bacterial mRNA (Glatron and Rapoport, 1972). It was reported that the putative transcriptional terminator of crylAa gene (a stem-loop structure) acts as a positive retroregulator (Wong and Chang, 1986). Therefore, it is likely that the crylAa transcriptional terminator increases the half life of cry mRNA by protecting it from exonucleolytic degradation from the 3’ end.
Post-translational mechanisms also contribute to the stability of the cry gene products. Cry proteins spontaneously form crystalline inclusions in the bacterial cells. The crystal shapes can be bipyramidal, cuboidal, flat rectangular, irregular, spherical and rhomboidal, depending on the compositions of the protoxins. The ability of the protoxin to form crystals in the cells may decrease their susceptibility to premature proteolytic degradation by cellular proteases
(Almond and Dean, 1993). However, the crystals have to be solubilized rapidly and efficiently in the insect midgut to become biological active. Several Cryl proteins produced in E. coli or B. subtilis were able to form crystals (Shivakumar et al., 1986; Ge et al., 1990). Because the proteins are able to form identical crystals in B. thuringiensis, B. subtilis, and E. coll, it is likely that specific host factors are not required for the protein assembly.
1.2 Classification of Crystal Proteins
In 1989, Hofte and Whiteley proposed the first systematic nomenclature and classification of the B. thuringiensis crystal proteins. This system is based on protein structure deduced from the DNA sequences as well as their host specificity. The cry genes have been divided into four major classes and several subclasses. The four major groups are Lepidoptera-specific (Cryl), Lepidoptera and Diptera-specific (Cryll), Coleoptera-specific (Crylll), and Diptera-specific
(CryIV). The system provided a useful framework for classifying the set of known genes. However, inconsistencies existed in the original scheme due to the attempt to accommodate genes that were highly homologous to known genes but did not encode a toxin with similar insecticidal spectrum.
In 1998, Crickmore et al. (Crick more et al., 1998) proposed a new nomenclature for the cry genes. This system is based on the degree of evolutionary divergence as estimated by phylogenetic tree algorithms (Fig. 1.1).
In primary rank of the new nomenclature, the Roman numerals have been changed for Arabic numerals. The change from function-based to a sequence- based nomenclature allows closely related toxins to be ranked together and removes the necessity for researchers to bioassay each new protein against a Primary Rank Secondary Rank Tertiary Rank CrylAb
CrylAc
•CrylDa ■CrylDb 8^1:: :8R;iig
■CrylCa •CrylCb ■CryIBb - r < :8q;}g: :8^ilg ■Cry7Aa •Cry7Ab :8^g5S r • CrySBa :#
■ Cry4Aa ■Cry4Ba •CrylOAa I L_ -CrylSAa -CrylSBa -Cry20Aa •Cryl6Aa -Cryl7Aa -CrySAa - CrySAc
Main Cry -Cryl3Aa Lineage - Cryl4Aa -Cry2Aa -Cry2Ab -Cry2Ac -CrylSAa -C ry llB a -C ryllB b -C ryllA a - CyUAa -CytlAb Cyt -C ytlB a Lineage -Cyt2Ba -Cyt2Bb -Cyt2Aa Outlying -CrylSAa Cr/ - Cry6Aa Lineages I 10 20 30 40 50 60 70 80 90 Percent Ammo Acid Sequence Identity
Fig. 1.1. Phylogram demonstrating amino acid sequence similarity among Cry and Cyt proteins. The vertical bars demarcate the four levels of nomenclature ranks (Crickmore etal., 1998). growing series of organisms before assigning it a name.
1.3 Sequence Similarity and Structures of Crystal Proteins
There are five conserved blocks among most of the Cry toxins (Fig. 1.2)
(Schnepf et al., 1998). ). It is unclear whether the sequence conservation in each block region reflects any functional significance. However, according to similarity of the protein sequences, Cry proteins might be categorized into 4 groups. The group consisting of Cryl, Cry3, Cry4, Cry7 to CrylO, Cry16, C ry l7, Cry19, and cry20 contains all five of the core blocks. A second group consisting of CryS,
Cry12 to Cry14, and Cry21 contains recognizable homologs of blocks 1, 2, 4, and
5, with more variability in block 1. Block 3 is completely absent from this second group. For the first two groups, when the protoxin possesses the C-terminal extension, blocks 6 , 7, and 8 are present (Fig. 1.2). The third similarity group consisting of Cry2, Cryl 1, and Cry 18, possessed only conserved block 1, and truncated variant of the block 2, and lack homologs of the other blocks. However, the other proteins, Cry 6 , Cryl 5, Cry22, Cytl, and Cyt2, have no recognizable
homologs to the conserved blocks seen in the three groups.
Tertiary structures of three crystal proteins have been determined by X-ray
crystallography. These proteins are Lepidoptera-specific CrylAa (Grochulski et
a/., 1995), Coleoptera-specific Cry3Aa (Li etal., 1991), and Diptera, Lepidoptera-
specific Cry2Aa (Morse et al., 2001). Surprisingly, Cry2Aa protein, although not
homologous to CrylAa and Cry3Aa, possesses a three dimensional structure Domain I Domain II Domain III
[|[| (sjlD 0 [8] Cryl A I I . É Cry3A Cry4 A Cry7A I # f CrySA I ^ f Cr,9A I I ■ ■! I C„10A CmiXrl II , 1 Cryl9A I I B ■ - Ml- -I CryZOA .-IB Ml-|-----1 I I ! \ \ Cryl5 A ! I # . _# 1 _ B Cryl7A [ . _J. #
[üj^g 0 lü 3 0 CrySAa I lD I] ■ ■ ~^l I ■ Bi ,B-Ji CryI2A t o m t m butil ! \ / / / Cry MA I ..%p TJHj: / / CryZlA I IL-j^ T \ ______\ \ \ CryI3A I .. JUl
111|2 var| CryZA I # IT" alt I alt a I CrylSA I ■ 100 ammo acids cryiiA I — c n r
Fig. 1.2. Position of conserved blocks among Cry proteins (Schnept etal., 1998).
Sequence blocks are shown as dark gray, light gray, or white to indicate high, moderate, or low degree of homology, respectively. similar to CrylAa and CrySAa (Fig. 1.3).
The structure determined from the full-length Cry2Aa protoxin reveals an additional a helix, called aO, at domain I. Even though Cry2Aa lacks the other conserved blocks 2, 3, 4, and 5, the tertiary structure-guided alignment of all three domains of this protein indicates that almost all of a-helices and p-sheets found in CrylAa and CrySAa, are present in its structure (Fig. 1.4 -1.6).
1.4 Mode of Action and Domain Function of Crystal Protein
A general scheme for the mode of action of B. thuringiensis toxins is summarized in Fig. 1.7. The proposed steps in this mechanism were based on the toxin structure and recent mechanistic studies (Knowles and Dow, 1993;
Knowles, 1994). The protoxin must be first ingested and solubilized in the insect midguts. Subsequently, the high pH environment of the midgut helps dissolve the protoxin inclusion. The released protoxins are then activated by insect proteases. Most of the Cry protoxins have molecular weight about 130-140 kOa.
The proteases cleave about one half of the C-terminus, and some residues at the
N-terminus, leaving about 60-kDa protease-resistant toxins. The active toxins then traverse the peritrophic membrane lining the insect midgut. The pore size of the membrane allows only monomeric forms of the toxin to reach the midgut
membrane (Yonivitz at a/., 1986). Next, the activated toxins bind to specific
receptors of the susceptible insects, located on the brush border membrane of
8 CrylAa Cry2Aa CrySAa
aO
■c i,S ^ \ h
Domain I Domain III
H « V : ' <■• ! ■
n h t . ï * X. i >, # ##% • r . 'J V >K * , - Domain II 6
Fig. 1.3. Structures of CrylAa, Cry2Aa, and CrySAa showing top-view and side- view of the molecules. Each toxin consists of 3 domains: domain I, a a-helix bundle; domain II, an assembly of three p-sheets; domain III, a p-sandwlch
(Modified from Li et al., 1991; Grochulski et al., 1995; Morse et al., 2001). The activated CrylAa and CrySAa toxins, not the full-length protoxins, were used in determination of their structures by X-ray crystallography, while the structure of
Cry2Aa was determined from the undigested, full-length protoxin. C ry2Aa AMDP FS FEHKS LOTIQKEWMNLKRTOHSLVVAPVVGTVSSFLLKKiGS LIG KR IL C ry iA b AHDP FS FQHKS LOTVQKEWTgwKKNNHSLYLOPIVGTVASFLLKKfGS LVGKRIL C ry lS A a PIDNNTtCSTDFTPINVMRTDPFRKKS TQELTREWTlwKENSPSLFTpklVGVVTSFLLQS.KK QATSFLL C ry lA a LSNPE VEVLGGERxj TGVTPlblSLSLTQFLUSE:V PGAGFV c r y lA a ETPNPTL EDLNYKEFLRMTADNNTBAUDSSTTKPVIDKGISVVGDLLGV/GFPFGGALVSFY
C ry jA a C ry lA a C ry lA a lOa u O a iiO b C ryZAa id 20 30 40 70 C ry lA a id 20 30
C ry2Aa ELWGIl FPSI TNL.MQOILRET IFLNQRLNTDTLARVNAEUi; UQANIREFNQQV iNFLNP TQNPVP C ry2Ab ELRNLlFPSi TNLMQOILRET IFLNQRLNTOTLARVNAELT LQANVEEFNRQV iNFLNP .SITS C ry lS A a TLTDLLFPNIIF 5SLTMEEIURAT lYVQERLOTOTANRVSQELV LKNNLTTFNDQV-OFLQ NRVGISP ■ A IID C ry lA a GLVDI GI FjGPSQMDAFLVQll ILINQRIEEFARNQAISRLE USNLYQIYAESF lEWEAD PTNPALR lE M R I crylAa NFLNT IP 5EDPWKAFMEQV ILHOQK IAOYAKNKALAELQI liQNNVEDYVSAL ISWQRNPVSSRNPHSQ RIRE
C ry2A a c r y l A a C ry lA a
C ry2A a id 110 120 130 140 C ry lA a id 90 100 110 120 1 3 0
♦ / * » * Cry2A a SVNTMQQ l FLNRLPQFQIQG QLLLLPLFAmAANMlILSFIR ^ILN A D E GISAATLRTYRDYLRNYTRDjSNYCINT Cry2Ab SV N TMQQ LFLNRLPQFQMQG q l l l l p l f a I a a n l L s F t R VILNADE GISAATLRTYROYLKNYTROI5NYCINT C ry lS A a StNTMQQLFVNRLPQFQVSC q v l l l p l f a I a a t l ] L t f i r i/IIN A O E « ptaqlntytryfkeyiae I s n y a l s t C ry lA a qfnomnsalttaipllavqn q v p l lS v y vJ a a n l I LSVLR VSVFGQR GFOAATINSRYNDLTRLIGN# rOYAVRW c r y lA a lfsqaesmfrnsmpsfaisg e v u f l t t y a I a a n t i Il f l l k AQIYGEE gyekediaefykrqlkltqe I roHCVKw
C ry2A a C ry lA a C ry lA a C ry2Aa qlAFRCLW t i^lhomuefrtymflnvfeyvsinslf C ry2A b QSAFKGLF TI^LHOMLEFRTYMFLNVFEYVSIWSLF 0 ILLVS C ry lS A a DOCFRTRf yPR NTtLEDMLQFKTFMTLNALDLVSIWSLL iLYVS C ry lA a NTGLERVK GPOSROWVRYNQFRRELTLTVLOIVALFSNVJSRRB p I (TVS c r y lA a NVGLDRLF GSSYESWVNFNRYRREMTLTVLDLIALFPLYJVR l Bp R IVKTE C ry2A a C ry lA a C ry lA a C ry2A a id 230 C ry lA a id 220 250 Fig. 1.4. Tertiary structure-guided sequence alignment across domain I of CrylAa, Cry2Aa, and CrySAa. Regions of high structural similarity are boxed. Salt bridges are designated with a “+” on the upper line. The striped bars above the sequence represent the conserved blocks. The sequences of Cry2Ab and CrylSA are also aligned to compare similarity to Cry2Aa (modified from Morse et al., 2001). 10 /wwx Cry2Aa ANLYASGSGPQQ TQSFTAQNWF rŸ^FQVNSf4YtLS(5ÏS'6f(lÙSITFPNIGG LPG Cpy2Ab ANLVASGSGPQQ TQSFTSQOWF FLYSLFQVNS “iYVLNCFSGARL 5NTFPNIVG LPG SrfTHALLAARVNY C rylS A a ANLYNIGONKVN EGAYPISYGF FFN ^IQ TKS VYVLSGVSGIGA RFTYSTVLGRVLHOOLKNI TTYVGGTQGPN C ry lA a REIYTNPVLENFD CSFRGMAQRIEQNI QPHL4DILNSITIYTDY HRiFNYWSCHQITA Cry3Aa ROVLTDPIVCVNNLRGYGTTFSNIE WYH KPHLPOYLHRIQFHTRFQP G YYG no: FNYWSGNYVST Cry2Aa C ry lA a CrySAa (U b uSa C ry2Aa id 280 300 310 320 330 C ry lA a id 270 280 290 300 310 C ry iA a se e VSSGLIC ATNL NH VFNCS rV L P P LST »FVRSWL >SGTDREGV ATST Cry2Ab se e ISSGOIC ASPF HQ 4FNCS TFLP PLLT>FVRSWL(>' SGSOREGV ATVT C rylS A a IGVt LSTTELCELKKQQQATRDSLQQVDFQFFriN C m lp n p ita >YFAYSL rES RYS SIGG C ry lA a SPVCFSGPE FAFPLFC NAGNAAP PVLVS GL :IFRTLS plyrriilgsgpnnqelf c ry lA a RPSIGSNDI ITSPFYC NASS EP /QNLEFN GEkVYRAVAkTNLAVWPS AVYS CryZAa C 3 > 4 C ry lA a A a 3 C ry lA a 3 S 3 ;14 i\7 r CryZAa id 350 360 370 380 390 C ry lA a id 330 3 4 0 350 360 370 380 C ry jA a TESFQ ARGNS FP )YFIRNrSGVPL /IRNEOLTRPlHYH Cry2Ab ARGNS FP )YFIRNISGVPL /VRHEDLRRPLHYN C rylS A a RKDVFKS P JYYITNISATVQ [NGENTDTTPLYFK C ry lA a TTN RGTVASL LOVlPPQONSVPPRAGFkHRLSHVTMLSQLAG c ry 3Aa TKVEFS NOQTDE YD5KRNV1GAVSW lOQLPPETTOEPLEKGV5HQLWYVMCPLM 3 Cry2Aa c r y lA a C ry lA a PIO Cry2Aa id 420 4 30 440 C ry lA a id 410 420 430 440 Cry2AA QIRNIESPSGTPCGAF AYLVSVH* RKNNlYAANENCTMI Cry2Ab EIRNIASPSGTPGGAP AYMVSVHh RKNNtHAVHENGSMI C rylS A a ENRPITSTRCVN AViAVYt RKANtAGTNQNCTMl C ry lA a AVYTIRAPTFSWCF RSAE FN c ry lA a GSRCTIPVLTW1 ASVO FF Cry2Aa C ry lA a C ry lA a l U i iil2d |(I2 C ry2Aa id 450 460 470 480 C ry lA a id 4 SO 460 Fig. 1.5. Tertiary structure-guided sequence alignment across domain II of CrylAa, Cry2Aa, and CrySAa. Regions of high structural similarity are boxed. Salt bridges are designated with a on the upper line. The striped bars above the sequence represent the conserved blocks. The sequences of Cry2Ab and CrylSA are also aligned to compare similarity to Cry2Aa (modified from Morse et a/., 2001). 11 wywwMF Wf4 CryZAL HLAPEDYTGSFT15 H ►fATOü'NftoTRfiTT? LRFEQSNTlrikRYTLRd LYLRVSSIGNSTIR CryZAb HLAPNOYTC: FTISPIHATQVN^JTHfTFIS EKFGN URFEQNNTr kRYTLRG lylrvssicnstir CrylSAa HQAPPOCTC■FTVSPLHPSA NTITSYI» ENYGN5 LHLKGQGY. HYMLSG RLVLRLSGAAN QIK CrylA a N i l pssqitqipctkstnlgs C r s v v » GPGFTG LRRTSPCqp TLRVNI CAPLS VRIRYASTTNLQFH c ry lA a NMI OSKKITqLPLVKAYKLQS CASVVlGPRFTG IQCTENGS \ kXIYVTP 3VSYS } kBR arihyastsqitft CryZAa CrylAa C rylA a |M2 ml CryZAa id 490 C rylA a id 470 CrylAa INGR VYTVS TNNDGVNONGARFFSOINI VTLDINVT ,NS GS rPFDLMNIMFVPT ILPPLY CrylAb INGR VYTA1 TNNDCVNONGARFFSOINI VPLOINVT .NS G rQFDLMNIMLVPTfF ISPLY CrylSAa s p t t ; lYAF TNNEGITDNGSKFFKOFAFST PFVI IVLYFEGV iSLDLMNLIFLPAI )OTPLY CrylAa lOGRf INQC^ SSGS NLQSGSFFRTVGFT 1PFNF SVFTLSAH /FNSGi JEVYIORIEFVPA ■VTFEA crylA a LDGAI FNOYV DKBINKGD TLTYNSF 5 GNNLQIGVT ILSAGFNLASF GIKVYIDKIEFIPvk CrylAa CrylAa CrylAa 1119 l l l O n i l 1'22 c ry lA a id 560 570 580 590 600 610 C rylA a id 540 550 560 570 580 590 Fig. 1.6. Tertiary structure-guided sequence alignment across domain III of CrylAa, Cry2Aa, and CrySAa. Regions of high structural similarity are boxed. Salt bridges are designated with a on the upper line. The striped bars above the sequence represent the conserved blocks. The sequences of Cry2Ab and CrylSA are also aligned to compare similarity to Cry2Aa (modified from Morse et a/., 2001). 12 Ingestion i Solubilization I Proteolytic activation I Receptor binding I Formation of ion channels I 1. Breakdown of the permeability barrier of the membrane 2. Cell lysis 3. Disruption of gut integrity I Death of the insect Fig. 1.7. Sequence of events characterizing the mode of action of 8. thuringiensis Cry toxins. 13 columnar cells. A conformational change may occur, followed by Irreversible insertion of the toxin into the membrane (Wolfersberger et al., 1986). Formation of a pore has been hypothesized to depolarize the membrane, breaking down high-energy potassium pump and elevating cytoplasmic pH (Harvey et a/., 1986; Wolfersberger, 1992). Cellular swelling caused by toxins prior to cell lysis appears to be consistent in all B. thuringiensis toxins (Bauer and Pankratz, 1992; Gill et al., 1992). Once midgut stops functioning as a barrier between the body cavity (hemocoel) and the gut, the larva eventually stops feeding and dies. 1.5 Receptors for Crystal Proteins It has been demonstrated that binding of Cry toxins to their receptors is a necessary step for the toxin activity and specificity. Many laboratory methods have been developed to assess the binding of the toxin to the receptors. The most commonly used approach employs ’^®l-labeled toxin to study its binding to brush border membrane vesicles (BBMVs) (Wolfersberger et al., 1987; Hofmann etal., 1988). Immunohistochemical staining (Bravo etal., 1992), surface plasma resonance (Masson etal., 1995a; Jenkins etal., 2000), and ligand blotting (Garczynski et al., 1991) of BBMV protein have also been used. Binding affinity of Cry proteins to BBMVs typically range from 0.1 to 10 mM (Gill etal., 1992). However, this binding affinity does not represent a true dynamic equilibrium dissociation constant (Kd), since the binding reaction is irreversible (Liang et al., 1995). The binding affinity of Cry protein to the purified 14 receptors using ligand blotting is in the range from 100 to 200 nM (Sangadala et al., 1994). The results indicate that the toxin interacts with more than one biological receptor for its activity. Moreover, membrane insertion is likely to occur after the toxin is bound to the membrane. Due to the available knowledge and limitation of the techniques used for receptor exploration, identification of the Cry toxin receptors has been focused on membrane proteins. To date, two types of membrane proteins from insect BBMVs have been identified as Cry toxin receptors: cadherin-like glycoprotein and the enzyme aminopepidase-N (APN). Recently, a different type of receptor for Cry protein was also reported (Valaitis at a/., 2001 ). Preliminary analysis indicates that it may be a glycosaminoglycan. Cadherin-like protein was purified from M. sexta and binding to Cryl A was observed (Vadlamudi etal., 1993). In 1995, the BT-Rigene, which encodes this protein, was cloned (Vadlamudi et al., 1995). The estimated molecular size from the gene sequence is 172 kDa, but the expressed products showed molecular mass at 210 kDa, due to N-glycosylation of the protein. The cloned receptor was able to bind to CrylAa, CrylAb, and CrylAc (Francis and Bulla, 1997). Both insect and mammalian cell cultures transfected with BT-Ri gene showed high affinity binding for Cryl A toxins (Keeton and Bulla, 1997). Recently, BtR175, a 180-kDa protein with sequence similar to BT-RI was purified from Bombyx marl (Nagamatsu etal., 1998). High affinity binding was found in this cloned receptor protein (lhara etal., 1998). 15 Binding of Cry protein to aminopeptidase N (APN) was first observed on ligand blot using M. sexta BBMVs (Knowles et al., 1991 ). The molecular mass of this metalloprotease enzyme is 120 kDa (Garczynski etal., 1991; Knight etal., 1995). When this protein was reconstituted into phospholipid vesicles, higher binding affinity of CrylAc to the vesicles was observed (Sangadala et al., 1994). The affinity of CrylAc for purified APN was 100 - 200 nM, approximately 100-fold lower than the binding affinity of the toxin for BBMVs (Lee et al., 1996). Despite the lower binding affinity, pore formation in lipid membrane was enhanced when APN was added (Sangadala et al., 1994; Lorence et al., 1997; Schwartz et al., 1997). Binding of CrylAc toxin to the receptor requires N-acetylglucosamine (GalNAc) moiety on the receptor, and the binding was inhibited when free GalNAc was added into the binding reaction (Knowles etal., 1991). Correlation between binding affinity and toxicity has not been well established. Several proteins with higher initial binding affinity and enhanced toxicity have been reported, for example, CrylAa, and CrylAb (Lu et al., 1994, Rajamohan et al., 1996b). However, this correlation was not found in the binding of CrylAc to purified APN (Jenkins et al., 1999). Different results were found in irreversible binding experiments (Liang etal., 1995, Rajamohan etal., 1996a). In the latter, a direct correlation between toxicity and the rate constant for irreversible bound toxin was observed. 16 1.6 Membrane Insertion and Pore Formation After binding to the receptors on the membrane, a conformational change might precede insertion of the toxin into the membrane. However, there has been no direct evidence showing that the Cry toxin inserts into the cell membrane. The insertion has been postulated by analyzing indirect evidence such as permeability of BBMVs (Sacchi et a/., 1986; Carroll and Ellar, 1993), electrical short circuit current measurements on isolated midgut (Chen et al., 1993), and channel formation on the planar lipid bilayers (Slatin et a/., 1990). By analogy with other membrane-inserting toxins (Lakey et a/., 1991 ; Choe et al., 1992) and with transmembrane ion channels, 20 amino acid-long amphiphatic a-helices in Cry proteins are the most likely to penetrate and form a pore. This led Hodgman and Ellar (Hodgman and Ellar, 1990) to propose a “penknife" model of Cry proteins after they identified a-helices which show a striking conservation of amphipathic character. In this model, helices a5 and a6 as a pair are most likely to form the pore and joined at the end of domain I. The rest of domain I is predicted to be furthest away from the membrane and therefore have to flip out of domain I like the opening of a penknife (Fig. 1 .SB). Alignment of other Cry sequences, including CrylAa and Cry2Aa, showed that these two helices are highly consen/ed among different Cry toxins. 17 B L___ Fig. 1.8. Mode! of insertion of Cry toxin into the insect plasma membrane. The a- helices in domain I are numbered from the N-terminus of the activated toxin. (A) One possible orientation of the toxin as it binds to its receptor. (B) The “penknife” model. (C) The umbrella model (modified from Knowles, 1994) 18 Li et al. (1991) proposed the so called umbrella model in which the a4-a5 hairpin insertion is followed by spreading of the remaining domain I bundled helices over the membrane surface like an "umbrella " (Fig. 1.8C). Mutagenesis in a5 generally decreased toxicity, whereas mutations in a3 or a6 showed little or no effect (Wu and Aronson, 1992: Aronson at a/., 1995; Uawithya at a/., 1998). Cross-linking of domain I helices by disulfide-bridge engineering showed that decreased flexibility of either a4 or aS blocks their ability to create an ion channel (Schwartz at a/., 1997). Studies of synthetic peptide helices showed that a5 could span artificial membrane (Gazit and Shai, 1995). In the refined umbrella model proposed by Schwartz (Schwartz et al., 1997), domain I swings away from domain II and III, followed by a4 and a5 integration into the membrane forming a tetrameric ion channel with a diameter of approximately 6 A (Fig. 1.9). 1.7 Specificity Determining Region of Crystal Proteins Despite the similarity among their sequences. Cry proteins of the same group show considerable differences in their activity spectrum. For examples, CrylAa is highly toxic to Bombyx mari, whereas CrylAc has very little activity to the insect (Ge at a/., 1989). On the other hand, CrylAc is the most toxic Cry protein against Trichopiusia ni and Haliothis virascans (Moar at a i, 1990; Ge at a i, 1991), followed by CrylAb and CrylAa. The most remarkable specificity is found in Cry2aa and Cry2Ab which are 87% identical n amino acid sequence. Only Cry2Aa is active against both mosquito and lepidopteran larvae, while 19 o-: Kilar rcsidu: = Hydrophobic residue Fig. 1.9. Hydropathic profile of the a4a5 hairpin and a working model for pore formation. (A) Helical wheel projections of both helices within the hairpin. The X represents the starting amino acid and the arrow indicates the direction of the a- carbon backbone. (B) Schematic top view representation of a tetrameric model of the pore formed by CrylAa toxin in a phospholipid membrane (modified from Schwartz et al. 1997). 20 Cry2Ab is toxic to only lepidopteran insects (Widner and Whiteley, 1989; Dankocsik etal., 1990). Recombination studies have been used to determine the region of Cry protein responsible for the toxin specificity. Homolog scanning mutagenesis revealed a region of CrylAa as a specific determination of 8. mori (Ge et a!., 1989). When this region was replaced by the homolog region of CrylAc, the activity against 8. mori was lost. When the reciprocal exchange was made, CrylAc gained toxicity to the level of CrylAa. The specificity determining region against 8. mori in CrylAa was located between amino acids residues 335 to 450, a relatively small region of domain II. However, a significantly larger region spanning residues 355 to 615 was found to be important for determining specificity against H. virescens. Widner and Whitely (1990) investigated the peptide region that determines the mosquitocidal activity of Cry2Aa by in vivo homologous recombination of cry2Aa and cry2Ab. The specificity-determining region of Cry2Aa against mosquito was located between residues 307 to 382. Liang and Dean (1994) used in vitro homologous scanning recombination in the putative domain II of Cry2Aa and Cry2Ab, and found the region responsible for mosquitocidal specificity is located between amino residues 278 and 340. However, the region involved in lepidopteran specificity is located in a broad region, between the residues 341 to 21 487. The identified specific determining regions of some Cry proteins are shown in Table 1.1. 1.8 The Effect of Mutations at Specificity Determining Regions on Toxicity As shown in Table 1.1 most of the specificity determining regions are located on domain II of the active toxin. Point mutations were also introduced to investigate the “contact point" of the interaction between the toxin and its receptors. The amino acids targeted for mutagenesis are on the loop regions, but some in the p-sheets are also investigated. CrySAa, a coleopteran-specific toxin, contains a long loop 1 region (Li et a/., 1991). Change in binding and toxicity was found when some amino residues in this region were mutated (Wu and Dean, 1996, Wu et al., 2000). Mutant derivatives with various combinations of substitutions (R345A, Y350F, Y351F) and deletion (R345A, AY350, AY351) exhibited higher toxicity against Tenebrio molitor. The tighter binding affinities of both proteins to the insect BBMVs ware also observed. The effect of mutations at the long loop 2 region of CrylAa, and CrylAb on toxicity were intensively investigated. Alanine substitution or deletion of residues 365 to 371 (LYRRIIL) of CrylAa abolished all toxicity for 6. mon (Lu et al., 1994). The loss in toxicity was related to loss in binding affinity of the mutant 22 Specificity CrylAa CrylAb CrylAc CrylC Cry2Aa References e. mori 332-345 Ge ef ai., 1991 H. virescens 258-646 Honee et a/., 1991 H. virescens 335-615 Ge et a i, 1991 T. ni 335-450 Ge et a i, 1991 S. littoralis 261-654 Honee et ai., 1991 1 » S. exiqua 448-559 Bosch et al., 1994 L dispar 341-412 Liang and Dean, 1994 A. aegypti 307-382 Widner and Whiteley, 1990 A. aegypti 278-340 Liang and Dean, 1994 Table 1.1. Insect specificity determining regions of Cry toxins. proteins to the insect receptors. The role of two positively charged residues 368rrr37o CrylAb was investigated by Rajamohan (Rajamohan et al., 1996c). Substitution of these three residues by alanine caused a loss of toxicity against M. sexta. The residues were apparently involved in initial binding of the toxin. Interestingly, mutations at F371 and N371 dramatically affect irreversible binding and toxicity of CrylAb (Rajamohan et a!., 1995). Amino acid residues on loop 3 of domain II of CrylAa and CrylAb were reported to be the initial binding sites for the lepidopteran insects (Rajamohan et a!., 1996b). Two amino residues, G349 and F440, appear to be necessary for the binding. The mutant proteins obtained from replacing these two residues with alanine were less toxic, and their binding affinities were affected. 24 CHAPTER 2 CORRELATING STABILITY AND TOXICITY OF BACILLUS THURINGIENSIS Cry2Aa TOXIN AGAINST GYPSY MOTH {LYMANTRIA DISPAR) LAVAE 2.1 SUMMARY Cleavage of Cry2Aa protoxin (MW 63 kDa) from Bacillus thuringiensis by midgut juice of gypsy moth (L. dispar) larvae resulted in two major protein fragments: a 58-kDa fragment which was highly toxic to the insect, and a 49-kDa fragment which was not toxic. In the midgut juice, the protoxin was processed into a 58-kDa toxin within 1 minute, but after digestion for 1 h, the 58-kDa fragment was further cleaved within domain I, resulting in the protease-resistant 49-kDa fragment. Both the 58-kDa and non-toxic 49-kDa fragments were also found in vivo when ’^®l-labeled toxin was fed to the insects. N-terminal sequencing revealed that the protease cleavage sites are at the C-terminus of Tyr49 and Leu 144 for the active fragment and the smaller fragment, respectively. To prevent the production of the non-toxic fragment during midgut processing, five mutant proteins were constructed by replacing Leu 144 of the toxin with Asp (L144D), Ala (L144A), Gly (L144G), His (L144H), and Val (LI44V), using a pair of 25 complementary mutagenic oligonucleotides in polymerase chain reaction (PCR). All of the mutant proteins were highly resistant to the midgut proteases and chymotrypsin. Digestion of the mutant proteins by insect midgut extract and chymotrypsin produced only the active 58-kDa fragment, except L144H that was partially cleaved at Leu 144. 26 2.2 INTRODUCTION During sporulation, Bacillus thuringiensis, a group of gram-positive bacteria, produce crystalline insecticidal proteins (Cry proteins) which are toxic against a range of insect groups. Recently, a new revised nomenclature has been proposed to 120 Cry toxins based on amino acid homology of the proteins (Crickmore et al., 1998). The mechanism of action of most Cry proteins consists of three major steps; solubilization and activation of protoxin in the insect midgut (Tojo and Aizawa, 1983), binding of the activated fragment to midgut receptor (Bravo et al., 1992; Hofmann et al., 1988), and insertion of the toxin into the midgut apical membrane, resulting in destruction of membrane potential (Harvey and Wolfersberger, 1979; Giordana etal., 1993). The molecular sizes of most Cry proteins are around 130-140 kDa (Hofte and Whiteley, 1989). After digestion by trypsin-like enzymes in the midgut (Tojo and Aizawa, 1983; Milne and Kaplan, 1993), the active fragment is produced. In general, the protease-resistant core protein is in the range of 60-70 kDa (Aronson et al., 1986; Hofte and Whiteley, 1989) covering three domains of the active toxin (Li etal., 1991; Grochulski etal., 1995). Deletion of the c/y genes beyond the active sequences completely abolished the toxic activity of the gene products (Schnepf and Whiteley, 1985; Wabiko et al., 1986). Proteolytic processing of Cry protein by midgut proteases is reported to generate active toxins of varying potency and specificity (Haider et al., 1986). However, digestion of Cryl A by diamondback moth {Plutella xylostella) midgut 27 extract generated a core that lacks a-helix 1 in domain I of Cry1 A (Ogiwara et al., 1992). Similarly, digestion of Cry3A with chymotrypsin caused a nick in the region between a-helix 3 and a-helix 4 of domain I (Carroll at a/., 1997). The protease cleavage inside domain I was also found in Cry9Ca1 and more stable protein was produced after removal of the trypsin-cleaved residue (Lambert at a/., 1996). Cry2Aa, a 633-amino acid toxin with molecular weight of 63 kDa, was originally described by Yamamoto as a dipteran- and lepidopteran-active protein (Yamamoto and McLaughlin, 1981; Yamamoto, 1983; Donovan atal., 1988). Even though the sequence of this Cry protein shows rather limited homology to the other Cry proteins, its structure has recently been determined and observed to be similar to CrylAa and Cry3A, consisting of three distinct domains (Morse at a/., 2001). English at al. (1994) reported the distinct binding and ion channels formed by Cry2Aa in Halicovarpa zaa, indicating a unique mode of action among the other Cry proteins. Here we report that Cry2Aa is rapidly cleaved at a single position in domain I by midgut enzymes of gypsy moth {Lymantria dispar). The cleavage product is not toxic against the insect. Several mutant proteins were constructed by removing the amino acid targeted by the proteases. All of these mutants were able to release the active fragment, which is more resistant to protease digestion than the wild type toxin. 28 2.3 MATERIALS AND METHODS All enzymes and reagents were purchased from Boehringer Mannheim Biochemicals (BMB) unless otherwise stated. Bacterial Strains and Plasmids E. coli strain BL21 and plasmid pDL103 (Liang and Dean, 1994) and pOS4201 (Ge etal., 1989), which contain cry2Aa and crylAa gene, respectively, were obtained from our laboratory stocks and from the Bacillus Genetic Stock Center, The Ohio State University. Gut Juice Preparation Fourth-instar larvae of L. cf/spar were dissected and the midguts were recovered. The midguts were centrifuged at 15,000 g for 30 min and the supernatant was collected and used as the midgut extract. Protein Purification E. coli harboring cry2Aa gene was cultured in LB-broth containing 100 ug/ml of ampicillin at 37"C for 72 h. Inclusion bodies were purified from the bacteria as previously described (Liang and Dean 1994). The bacterial cells from a 500 ml culture were resuspended in 100 ml lysis buffer (15% sucrose, 50 mM EDTA, 50 mM Tris pH 8.0) containinglO pg/ml lysozyme at 37°C for 2 h, followed by sonication and washing three times in 100 ml of 2% Triton X-100, 0.5 M NaCI, 29 five times in 100 ml of 0.5 M NaCI, and three times in distilled water. The crystal protein was solubilized in 100 mM NaaPOA pH 12 at 37“C for 1 h. The soluble protein samples were then dialysed in 50 mM NazCOa pH 10.5. Protein concentration was determined by BOA method (Pierce) and gel densitometer (Kodak Digital Science™ Electrophoresis Documentation and Analysis System). Protein Digestion Digestion reactions using the midgut extract were performed at room temperature as indicated. The reaction was stopped by adding to final concentration IX of Complete™ (BMB), a protease inhibitor, 2 pM of E-64 (a papain inhibitor), 0.4 N NaOH. SDS-loading buffer was added and the samples were boiled. The protein samples were neutralized with equal molar of HCI and then separated by 10% SDS-polyacrylamiue yol electrophoresis (SDS-PAGE) (Laemmli, 1970). Chymotrypsin digestion was performed by using 2% (w/w) of the enzyme to the protein. The reaction was stopped by adding Complete™ to final concentration of IX , and boiling in SDS-loading buffer. The molecular weights of the digested products were determined by Kodak Digital Science™ID Image Analysis Software. 30 Autoradiography and in vivo Cleavage of Cry2Aa The protoxin of Cry2Aa was iodlnated with lODO-BEAD (Pierce) as previously described (Lee et al., 1992). Twenty-five micrograms of the toxin were iodinated with 1 mCi of ^^®I-Nal. Each overnight-starved fourth-instar larva of gypsy moth was fed on diet supplimented with 1 pg/cm^of ^^^l-labeled protein. After feeding for 1 h, the insect was longitudinally dissected. The peritrophic membrane and contents were gently removed from the midgut membrane. The midgut tissue was washed in 50 ml of binding buffer (150 mM NaCI, 8 mM Na 2HP0 4 , 2 mM KH2PO4, pH 7.4) for 10 min twice. The peritrophic membrane with its contents and midgut were separately placed on a piece of Whatman filter paper, and then vacuum dried before autoradiography for 48 h. To determine the midgut-processed fragments in vivo, another group of the insects were separately treated with ^^^l-labeled toxin as described above. After dissection, the midgut contents in the peritrophic membrane were centrifuged at 15,000g for 30 min. The supernatant was diluted to 1:10 with distilled water before adding Complete™, E-64, and NaOH as decribed above. After boiling in SDS-loading buffer, the protein samples were neutralized with HCI and analyzed by gel electrophoresis. The bound toxin on the midgut tissue was extracted by boiling the whole gut tissue in 200 pi solution containing IX Complete™, 2 pM E-64 and IX SDS-loading buffer. The samples were centrifuged at 15,000g for 15 min and the supernatant was analyzed by gel electrophoresis. The gel was vacuum dried and autoradiographed for 48 h. 31 Site-directed Mutagenesis Double-primer site-directed mutagenesis was performed by using double stranded pDL103 as a template DNA. The method used was a modification of the QuikChange™ site-directed mutagenesis method (Stratagene). Two complementary oligonucleotides were used as the mutagenic primers, 5’- CAAAACCCTGTTCCTCACTCAATAACTTCTTCG-3' to replace Leu 144 by His, and 5 -CGAAGAAGTTATTGAGNCAGGAACAGGGTTTTG-3' to replace Leu 144 by Ala, Asp, Gly, and Val. The mutagenic double-stranded DNA was amplified by the polymerase chain reaction (PCR) (Mullis and Faloona, 1987). Pwo was used as a DNA polymerase enzyme in the reaction. The temperature and time for annealing, extension, and denaturing were 48°C for 1 min, 68°C for 10 min, and 95°C for 30 sec, respectively. The reactions were performed for 15 cycles. PCR products were treated with Dpn\ for 1 h to digest methylated parental DNA template before transforming E. coli BL21. N-Terminal Sequencing of the Protein Fragments The protein samples from the midgut juice-digestion reaction were separated on 10% SDS-PAGE and then transferred by electrophoresis to Immobilon-P polyvinylidene difluoride membrane (Bio-Rad). The peptide fragments were analyzed by Edman degradation using an Applied Biosystems Model 477A pulsed liquid-sequencer equipped with an on-line HPLC for PTH amino acid analysis at the USDA Forest Service facility (Delaware, Ohio). 32 Bioassays Gypsy moth diet (F9631B, BloServ) was used In all experiments. The protein samples were treated as indicated in the results and the digestion products were examined by SDS-PAGE just prior to bioassay experiments. The 58-kDa fragment was prepared by digestion of the protoxin in the diluted (1:50 v/v) gut extract for 10 min, the reaction was stopped by adding Complete ™ and stored at 4 °C. The 49-kDa fragment was prepared by the same procedure except the digestion was carried out for 16 h. The proteins were added to the surface of the diet in Multiwell-24 plates. 50% lethal concentration (LCso) was determined after 5 days of intoxication of the insect neonates. 50% growth inhibition dose (IDso) was determined by using second-instar larvae. The weight of each larva was measured before transfer to the toxin-contaminated diet, and 5 days after intoxication. Growth inhibition effect was considered positive when the larvae failed to gain weight. The data was analyzed by Probit method (Raymond, 1985). 33 2.4 RESULTS Digestion of Cry Proteins by Midgut Extract Proteolysis of Cry proteins after adding loading buffer has been reported by Chôma and Kaplan (1990). At high concentrations of L. d/spar gut extract (greater than 1:10 v/v), we also observed degradation of the protein after adding the protein-loading buffer. The degradation might be from the digestion of SDS- induced denatured toxin proteins by SDS-resistant proteases in the insect midgut juice. Therefore, both proteinase inhibitors and NaOH were added before mixing the protein samples with protein-loading buffer. Within 1 min, at 1:10 (v/v) of L dispar gut extract to protein solution, at room temperature, Cry2Aa protoxin was digested to a 58-kDa fragment. After 1 h, the toxin was further digested to a 49-kDa fragment (Fig. 2.1). The sequences from the amino terminus of the Cry2Aa protoxin, the 58-kDa fragment and that of the 49-kDa fragment are shown in Table 2.1. Amino acid sequencing revealed that Val50 and S e ri45 of the protein were indeed at the N-termini of the 58-and 49-kDa fragments, respectively. For comparison, CrylAa protoxin was processed into a 60-kDa fragment within 1 min. Within this short period of digestion smaller protein fragments were also found in the gel (results not shown). Further digestion for 1 h reduced the presence of the small fragments (Fig. 2.1, lane 6). This fragment was stable and remains present in the solution after digestion in the midgut extract 16 h. 34 Fig. 2.1. 10% SDS-polyacrylamide gel electrophoresis of the processing products of Cry2Aa and CrylAa by gypsy moth larval midgut extract at different periods of incubation. The protein solutions were mixed with the gut extract at 10:1 (v/v) ratio and incubated at room temperature. The reactions were stopped by adding protease inhibitors described in the methods. Lanes: 1, molecular weight markers (phosphorylase b, 97 kDa; bovine serum albumin, 66 kDa; aldolase 39 kDa; triosephosphate isomerase, 26 kDa); 2, Cry2Aa protoxin; 3, Cry2Aa digested for 1 min; 4, Cry2Aa digested for 1 h; 5, C rylA al protoxin; 6, CrylAa digested for 1 h; 7, CrylAa digested for 16 h. Arrowhead indicates a trace amount of the 58-kDa fragment. 35 1 2 3 4 5 6 7 kDa 97 I! 66 39 26 Fig. 2.1. 36 Proteins N-terminal LCso* (ng/cm^) ID5o*(ng/cm^) Sequences Cry2Aa protoxin MNNVLNSGRT 5.6 (2.1 -8 .7 ) 26.9(17.2-38.2) (63 kDa) 58-kDa fragment VAPWGTVSSFL 4.7 (1.3-7.5) 30.4(19.1 -48.5) 49-kDa fragment SITSSVNTMQ ND*ND Table 2.1. Toxicity against L c/Zspar larvae of Cry2Aa protoxin and protein fragments from the processing of larval midgut extract. ^ Lethal concentration for 50% insect neonates, n = 40. ^ Inhibition dose for the 50% second-instar larvae, n = 12. * No death or growth inhibition was observed in the range of 10 - 5,000 ng/cm^ toxin concentrations. Confidence intervals (95%) are given in parentheses. 37 Cleavage of Cry2Aa in vivo To investigate whether the 58-kDa and 49-kDa fragments were produced in vivo, ^^®l-labeled Cry2Aa was applied to the diet. After feeding on toxin- contaminated diet for 1 h, we found diffusion of the toxin throughout the food tract (Fig. 2.2A-1), and the toxin was also found on the midgut membrane (Fig. 2.2A- 2). More bound toxin was detected in the anterior and the middle regions of the midgut. After intoxication for 16 h, the toxin was equally detected on all midgut regions (results not shown). The 49-kDa fragment was found in the midgut fluid (Fig. 2.2B, lane 2), while the 58-kDa fragment and small amounts of 49-kDa fragment were found on the midgut membrane (Fig. 2.2B, lane 3). Toxicity of the Digested Cry2Aa to L dispar Larvae LCso of Cry2Aa protoxin against the insect neonate larvae was 5.6 ng/cm^, and similarly, LCso of the 58-kDa fragment was 4.7 ng/cm^ (Table 2.1). The toxicities of these two proteins were not significantly different. The results were confirmed by using 50% growth inhibition (IDso) measurement on second-instar larvae. The IDso of the protoxin and the 58-kDa fragment were 26.9 and 30.4 ng/cm^, respectively. In contrast to protoxin and the 58-kDa fragment, the 49-kDa fragment was not toxic against the insect in the treatment concentration up to 5,000 ng/cm^. This 49-kDa fragment was insoluble and found as the precipitation form (result not shown). It was easily separated from solution by centrifugation. 38 B 1 2 3 97 66 I 39 - 26 Fig. 2.2. Autoradiography of L. d/spar larval midgut, and the digested Cry2Aa after feeding on diet contaminated with ^^^l-labeled Cry2Aa for 1 h. A: the peritrophic membrane with its contents (1) and midgut tissue after removal of peritrophic membrane (2). Note: anterior regions of both peritrophic sac and midgut tissue are arranged upward. B; lane 1, ^^^l-labeled Cry2Aa before applied to the insect; 2, ^^®l-labeled Cry2Aa extracted from midgut fluid; 3, ^^®l-labeled Cry2Aa extracted from midgut membrane of the same insect as in lane 2. 39 However, we still found a trace amount of 58-kDa protein co-precipitated with this fragment (Fig. 2.1, lane 4, arrowhead). This contaminating active fragment might be the protein that caused growth inhibition and lethality when a higher concentration of the 49-kDa fragment was applied to the insects. Mutagenesis Site-direct mutagenesis was used to replace Leu 144 of Cry 2Aa1 with different amino acids. The reverse primer was designed to have His at this position because we expected that the residue in this position might have the same properties as His161 in the loop between a3 and a4 of Cry3A. The "spin" codon designed in the forward primer allowed other small amino acids, Ala, Asp, Gly and Val, to replace Leu 144 so that the stability and toxicity of the new mutant proteins could be studied. Using these two oligonucleotides, 5 mutants were produced: Leu144Asp (L144D), Leu 14 Ala (L144A), Leu144Gly (L144G), Leu144His (L144H), and Leu144Val (LI44V). All mutants gave high expressed protoxins and these proteins were more stable than the wild type protoxin (Fig. 2.3). Protease Resistance of the Mutant Proteins Digestion of L144D, LI44V, L144A and L144G protoxins with 10:1 protein solution volume per gut juice volume produced only a 58-kDa fragment. Stability of these mutants was also found in gut juice digestion for 16 h (results not 40 Fig. 2.3. Digestion of Cry2Aa protoxin and its mutant derivatives by gut extract (A) and chymotrypsin (B). 10 pg of the protein were incubated with midgut extract at a ratio of 10:1 (v/v) of protein solution to gut extract or 2% by mass of chymotrypsin. Lane 1, molecular weight marker; Lanes 2, 4, 6, 8, 10, 12 are Cry2Aa, L144D, L144H, LI44V, L I44A and L144G protoxin, respectively. Lanes 3, 5, 7, 9, 11, and 13 show digestion products of the proteins after 1 h incubation of Cry2Aa wild type protein, L144D, L144H, LI44V, L I44A and L144G, respectively. 41 kDa 1 2 3 4 5 6 7 8 9 10 11 12 13 gj - - — — - - B kDa 1 2 3 4 5 6 7 8 9 10 11 12 13 97 ^ “ w # m # 39 ^ ^ '■* - (TH I 1 Fig. 2.3. 42 shown). Digestion of L144H yielded two protein fragments, 58-kDa and 49-kDa (Fig. 2.3A, lane 7). The remaining 58-kDa fragment indicates some higher degree of protease resistance of the L144H protein. Digestion of these proteins with chymotrypsin produced the same results as with midgut extract digestion (Fig. 2.3B). Toxicity of the Mutant Proteins Bioassay of the mutant proteins treated with the midgut extract and chymotrypsin against second-instar larvae of gypsy moth is shown in Table 2.2. Even though all mutant proteins were more resistant to protease cleavage, compared to the wild type protein, they did not show higher toxicity. We tested toxicity of the protoxin and confirmed the data by treating the proteins with either gut extract or chymotrypsin before applying to the insect diets. LCso of all mutant proteins were the same as the wild type, between 2.8 - 8.8 ng/cm^. 43 Proteins LCjo (ng/cm') Protoxin Gut juice treated" Chymotrypsin treated* Cry2Aa 5.6 (2.1 -8.7) ND* ND* L144A 3.1 (0.9-5.8) 5.2(1.1 -9.5) 5.8 (4.2-7.1) L144D 6.4 (3.5-9.1) 4.2 (0.5 - 7.7) 6.9 (2.9 - 10.5) L144G 4.8 (1.3 -8.1) 8.0 (5.2- 11.4) 7.7 (2.1 - 12.8) L144H 8.8 (4.5- 12.3) 18.4 (9.7-25.3) 19.6(10.4-29.1) LI 44 V 6.3 (4.7 - 7.4) 2.8 (0.7-5.2) 7.1 (5.0-9.7) Table 2.2. Toxicity of Cry2Aa and the mutant proteins against L. d/spar larvae. * ND, not detectable. No mortality was observed in the range of 5 - 5,000 ng/cm^ protein concentrations. ® The proteins were digested with 1:50 (v/v) midgut extract, or *’2%(w/w) of chymotrypsin for 1 h before the treatment to the insect larvae. Confidence intervals (95%) are given in parentheses, n = 40. 44 2.5 DISCUSSION It is commonly believed that the products from the proteolytic cleavage of the Cry protoxins in the insect larval gut are the active fragments that play the major roles in binding to the receptors on the midgut columnar epithelial cells (Hofmann etal., 1988; Lee etal., 1992; Rajamohan etal., 1996a; Rajamohan et al., 1996b) and in irreversibly inserting into the cell membrane and destroying the electric potential on the apical midgut membrane (Harvey and Wolfersberger, 1979; Giordana etal., 1993). Furthermore, the proteolytic activation of the protoxins was also reported as an important factor for the mechanism of resistance of the insects to S. thuringiensis toxins (Oppert et al., 1997). Since it was observed that the intact CrySA and chymotrypsin-treated protein exhibited different affinity binding (Martfnez-Ramlirez and Real, 1996), it is important to investigate which portion of the polypeptides from the proteolytic process play the role in toxicity. Structural analysis of the trypsinized CrylAa and CrySA revealed three domains in both Cry proteins of which 7 helices are the main secondary structure in domain I, and p-sheets and loops are the main secondary structures in both domain II and domain III (Li etal., 1991; Grochulski etal., 1995). This led to the hypothesis that all Cry proteins may exhibit three-domain structures (Grochulski et al., 1995). The protease-resistant core polypeptide starts from lie 29 of CrylAa, CrylAb, and CrylAc, and Asp58 of CrySA. Both residues are located just before a1 on domain I. Further removal of amino acids from the N-terminus 45 of the peptide leads to the loss of toxicity of the proteins (Schnepf and Whiteley, 1985). The three-dimensional structure of Cry2Aa has been solved (Morse et al., 2000). Compared to the structures of CrylAa and Cry3A, there Is one more a- hellx In domain I of Cry2Aa, designated aO, based on the nomenclature of Cry3A structure given by LI et al. (1991). Tyr49 Is on the loop between aO and a1. Removal of aO from the mutant protoxIn by mIdgut protease or chymotrypsin generates the protease-reslstant core that Is comparable to the trypsinized Cry3A and CrylAa that starts from a1 . Another cleavage site In Cry2Aa Is Leu 144 that Is on the loop before a4. The cleavage after Leu 144 was expected as an activity of chymotrypsln-llke proteases In the Insect gut extract. This cleavage was found during digestion of Cry2Aa with chymotrypsin (Fig. 2.3B, lane 3). Unlike NIcolls et al. (1989) who reported the N-termlnal sequences of these protease-reslstant cores using limited proteolysis by chymotrypsin, we prepared the protein fragments by using Insect mIdgut extract. The obtained sequences were the same for both digestions, Indicating the prevalence of the enzyme In the Insect mIdgut. Carroll et al. (1997) reported that cabbage white butterfly {Piehs brassicae) gut extract and chymotrypsin cleaved Cry3A at the N-termlnus of Arg158 and Hls161, respectively. These residues are aligned around the loop region between a3 and a4. However, because the small component Is still associated with the major component, the nicked toxin Is soluble and as active as 46 the protoxin. Different results were found in experiments with Cry9Ca1 (Lambert et al., 1996). Cry9Ca1 contains a protease cleavage site at Arg164, but trypsinization of this protein generated a nontoxic 55-kOa fragment without the small additional N-terminal fragment. In contrast to CrySA and Cry9Ca1, Cry2Aa cleaved by gut protease or by chymotrypsin at the carboxyl terminus of Leu 144 tends precipitate. We digested Cry2Aa in different buffers with pH ranging from 7 to 12, but the precipitation was consistently observed. The insoluble form might be an aggregation of broken Cry2Aa through the broken domain I after losing aO- a3 helices. SDS-PAGE analysis of the precipitate showed that almost all of it was the nontoxic 49-kDa fragment, contaminated by a trace amount of 58-kDa (Fig. 2.1, lane 4, arrowhead). The absence of the 49-kDa fragment from proteolytic processing of our mutant proteins provides further support to our postulate that the 58-kDa peptide is the active fragment against L. dispar \s. Leu 144 was replaced so that the cleavage site was eliminated. Indeed, mutant derivatives in which Leu 144 was altered were highly resistant to proteolysis by both chymotrypsin and gut juice extracts (Fig. 2.3). They were stable in midgut extracts for 16 h (results not shown). Initially, we expected that these more resistant proteins might show higher toxicity against L. dispar. But the results from Table 2.2 indicate that their toxicities were not significantly different from that of the wild type protein. The observation of diminished protease susceptibility of the Leu 144 mutants, compared to the wild type toxin, and the similar toxicity of all these 47 protoxins lead us to propose that the process of solubilization, proteolytic processing and binding (or insertion) of Cry2Aa must happen more rapidly in vivo than in vitro. We further propose that the cleavage of the protoxin to produce the 58-kDa toxin in vivo occurs more rapidly than the cleavage of the 58-kDa toxin to produce the 49-kDa protein. This evidence is obvious since higher amounts of the 58-kDa fragment were found on the midgut membrane (Fig. 2.2B, lane 3), indicating that this fragment is bound to the midgut membrane and protected by the membrane components before being cleaved at Leu 144. Another possible mechanism is that binding of the 58-kDa fragment to the receptor renders a conformational change in such a way that the second cleavage site, Leu 144, is away from the active site of the proteases. The rapid toxicity of Cry2Aa against H. zea larvae was previously observed by English et al. (1994) who reported that Cry2Aa-fed larvae of this insect stopped feeding about 1 min after intoxication and showed morbidity within 4 min. Experiments performed with the other Cry proteins also showed that the apical cell membrane electrical potential response was detected within a few minutes after adding the toxins (Liebig ef a/., 1995; Peyronnet et al., 1997). The results from our experiments also imply that the toxin binds and inserts very rapidly in the insect midgut membrane before proteolytic cleavage by chymotrypsin-liked proteases of the midgut enzymes. In vivo and in vitro cleavage at Leu 144 of Cry2Aa might produce different forms of the processed fragments. Digestion at Leu 144 in the midgut 48 environment, which is more viscous and contains much more "biochemical carriers", might only create a nick on the protein molecule, leaving the active fragment. The “nicked” fragment of wild type Cry2Aa is as toxic as the 58-kDa fragment produced by the mutants. This might be an explanation of why both mutant proteins and wild type Cry2Aa showed the same degree of toxicity. On the other hand, nicking at Leu 144 during in vitro digestion might cause dissociation of a1 to a3 helices from the rest of the protein molecule. Making chymotrypsin-resistant Cry proteins by site-directed mutagenesis is an approach that facilitates the production of active fragments which could be used in an investigation of the mode of action of these toxins. In addition, this strategy can be used in production of more stable proteins. Furthermore, we expect that this approach could be used to stabilize other labile Cry proteins like Cry20Aa1, which loses its mosquitocidal activity due to degradation (Lee and Gill, 1997). 49 environment, which is more viscous and contains much more "biochemical carriers", might only create a nick on the protein molecule, leaving the active fragment. The "nicked" fragment of wild type Cry2Aa is as toxic as the 58-kDa fragment produced by the mutants. This might be an explanation of why both mutant proteins and wild type Cry2Aa showed the same degree of toxicity. On the other hand, nicking at Leu 144 during in vitro digestion might cause dissociation of a1 to a3 helices from the rest of the protein molecule. Making chymotrypsin-resistant Cry proteins by site-directed mutagenesis is an approach that facilitates the production of active fragments which could be used in an investigation of the mode of action of these toxins. In addition, this strategy can be used in production of more stable proteins. Furthermore, we expect that this approach could be used to stabilize other labile Cry proteins like Cry20Aa1, which loses its mosquitocidal activity due to degradation (Lee and Gill, 1997). 49 CHAPTER 3 THE ROLE OF THE aO HELIX OF BACILLUS THURINGIENSIS Cry2Aa PROTOXIN IN SOLUBILITY AND TOXICITY AGAINST GYPSY MOTH (LYMANTRIA DISPAR), AND MOSQUITO (ANOPHELES QUADRIMACULATUS) LARVAE 3.1 SUMMARY Upon digestion by insect midgut proteases, the first 49 amino acids from the N-terminus of Cry2Aa protoxin from Bacillus thuringiensis are removed to form a 58-kDa active toxin. The short N-terminus consists of an a helix, known as aO, which is located before the a1 of the active toxin. In the present study, we investigated the role of the aO helix in the stability of Cry2Aa protoxin, the solubility and toxicity of Cry2Aa crystalline inclusions formed in E. coli cells. The results from gene deletion and site-directed mutagenesis indicate that the presence of this a helix is necessary for the formation of biological active inclusion of Cry2Aa. The D5 construct, which contains DNA coding for the aO helix, but which has the Val3 - Asn17 region deleted, was expressed and its inclusions were dissolved in the carbonate buffer pH 10.5 and as toxic to 50 Lymantria d/spar and Anopheles quadhmaculatus as wild type Cry2Aa. In contrast, expression of the D7 construct, In which the Val3 - Tyr49 DNA region was deleted, produced Insoluble and non-toxic Inclusions. However, cleavage of the first 49 amino acids from the presolublllzed protoxIn by chymotrypsin did not affect protein toxicity. Thermal unfolding experiments showed that the melting temperature (Tm) of both soluble D5 and Cry2Aa protoxIn, which contain the aO helix. Is 60°C, while the chymotrypsln-actlvated Cry2Aa exhibits a Tm of 53°C. Therefore, the aO helix contributes to the stability of the toxin molecule. Stability of the Cry2Aa protoxIn might correlate with the crystallization mechanism since mutation of the amino acids which are Involved In either Intramolecular salt bridges or hydrogen bondings, such as Lys29, Glu37, Tyr41, Lys42, Arg43, and Asp45, significantly lower the amount of the crystal yield. Furthermore, the lower crystal yield or loss of the product was also observed when the amino acids that are Involved In Intermolecular hydrogen bondings, such as Lys42, Arg43, Thr44, Hls46, Ser47, and Tyr49, were mutated. The Intramolecular Interaction between the amino acids In the aO helix and the rest of the toxin, and the Interaction of among protoxIn molecules In the crystalline Inclusion seem to be a factor that facilitates the crystallization mechanism of the Cry2Aa protoxIn. 51 3.2 INTRODUCTION The self-assembly of the crystal (Cry) protoxin to form proteinacous crystalline inclusions is the mechanism that Bacillus thuringiensis uses to minimize proteolytic cleavage of the protoxin (Almond and Dean, 1993; Agaisse and Lereclus, 1995). An inclusion may be comprised of one or more different types of Cry protoxins, and may be dissolved at alkaline pH environment of the insect midgut. There is usually one inclusion per cell, but two or more has also been observed. The inclusion shape can be bipyramidal, cuboidal, ovoid, or amorphous (Aronson and Fitz-James, 1976; Bechtel and Bulla, 1976; Mikkola at a/., 1982). However, it is not known how the protoxin assembles in the crystalline inclusion and why it is easily dissolved in the high pH solution. Based on molecular sizes, crystal (Cry) proteins may be divided into two groups, the large protoxin, with a molecular mass of 130-140 kDa; and the small protoxin, with a molecular mass of 70 - 80 kDa (Hdfte and Whitley, 1989). Cryl proteins are among the first group, while Cry2 and Cry3 are among the latter. The assembly of the Cry protoxin in the inclusion might involve interaction of the structurally- and size-related protoxin molecules through salt bridges and disulfide bonds, which are prevalent in the C-terminal half of the large protoxins (Li et al., 1991 ; Du et al., 1995). Dissociation of the protein molecules from the inclusion might be caused by the weakness of these forces due to alkaline pH. Re-formation of this biologically active crystal can be induced in vitro by removing the dénaturant or lowering pH (Bernhard, 1986; Dai and Gill, 1993). 52 After ingestion and solubilizaton in the insect midgut, about one half of the large protoxin molecule, mainly the C-terminus, is removed by the midgut proteases to generate a 55 to 65-kDa activated toxin (Chestukhina et al., 1982; Chôma eta!., 1990). The processing also includes removal of some amino residues at the amino terminus (Bietlot et a!., 1989). In contrast, the small protoxins are processed to the modest extent by the insect proteases. The processed region is mostly limited to the amino terminus. For example, the first 49 amino acids of Cry2Aa and the first 57 amino acids of Cry3A protoxin were proteolytically cleaved (McPherson et a!., 1988; Audtho et a!., 1999). The function of each portion of the protoxin in bacterial cells is largely unknown. The large C-terminal portion of the large protoxin is believed to play an important role in stabilization and crystallization of the protoxin. However, an explanation at the molecular level for the role of this domain has not been presented. It was found that a truncated mutant of Cry1C, which consists of only the active fragment and the N-terminal portion could form crystals in the bacterial cells. These were soluble in high pH buffer and as active as the protoxin (Park et al., 2000). Therefore, the C-terminal portion might not exclusively be the portion responsible for crystallization and solubility of the Cry protoxin. The three-dimensional structure of Cry2Aa was solved (Morse et al., 2001). Unlike the structure of CrylAa and Cry3A (Li et al., 1991; Grochulski et al., 1995), which was solved from protease-activated toxin, the structure of Cry2Aa was obtained using the protoxin. Although the primary sequences of 53 Cry2Aa, CrylAa and Cry3A have low homology, they share a similar overall tertiary structure. Cry2Aa is unique in having an extra a helix, named aO, located before a1 (Fig. 3.1), which is the first a helix of the active fragment of Cry protoxin. We reported that this helix was not involved in toxicity and specificity of the protein because removing this peptide from the protoxin by chymotrypsin digestion did not affect the protein function (Audtho et al., 1999). The structure showed that there are a number of exposed charge residues on this a helix. Crystallographic studies revealed that these residues are involved in both intramolecular and intermolecular interaction of the protoxin molecules. However, the role of this a helix in Cry2Aa has not been investigated. In this study, a number of mutations at the N-terminus of the cry2Aa gene were made. The mutated amino acid residues could be categorized into two groups: those which form intramolecular salt bridges or hydrogen bondings between the N-terminus and the rest of the protein molecule; those which form intermolecular hydrogen bondings among the protoxin molecules. We found that the presence of the aO helix in Cry2Aa protoxin is necessary for the crystallization and solubility of the protein. Deletion of the aO helix resulted in decreased solubility of the inclusion and stability of the protoxin. Additionally, mutations of the amino acid residues which are involved in either intramolecular or intermolecular interaction dramatically affect the yield of the crystal inclusion. 54 Fig. 3.1. Protein structure of Cry2Aa protoxin shown in ribbon and space-filled model. The top-viewed, and side viewed models are shown in A and B panel, respectively. Depicted in dark color is the N-terminal region of the protoxin (Metl to Tyr49), which is removed by proteases of the insect midgut. The positive- charged and negative-charged residues in the N-terminus are also indicated. 55 vÿ.Tr-;Vs À .V o D14 K36 B iC' \ 0 1 4 ^ 1 56 3.3 MATERIALS AND METHODS Mutagenesis and Deletion of Cry2Aa Cry2Aa gene from pDL103 (Liang and Dean, 1994) was transferred to pGEM-3Z(+) (Promega) by digesting pDL103 with SamH I and Hind III. The cry2Aa gene fragment was then inserted into pGEM-3Z(+) at SamH I and Hind III sites. The obtained construct, pGEM 103-9, was used to produce single-stranded DMA for mutagenesis experiments according to the Kunkel method (Kunkel, 1985). The mutated residues are shown in Fig. 3.2. In this method, a dut ', ung ' Escherichia coli strain, CJ236, was transformed with pGEM 103-9. After infection with the M13K07 helper phage, uracil-containing single-stranded DNA (U-DNA) was extracted and used as DNA template. The mutagenic oligonucleotides used for producing D1, D2, 03 are shown in Table 3.1. In addition to the desired mutation, additional silent changed were introduced to create new restriction sites in the mutant progeny. After annealing the U-DNA and the mutagenic oligonucleotides, T4 DNA polymerase and T4 Ligase was added and the reaction was allowed to proceed at 37°C for 2 h. The newly synthesized DNA was used to transform E. coli DH5aF’. The mutant DNA was screened by restriction enzyme digestion. The restriction enzymes designed for the mutant D1, D2 DNA primers were Sal I, and for D3 was Xba I. The DNA sequences of the obtained constructs were verified by DNA sequencing. 57 1 10 20 30 40 50 aOa aOb a l n ) 0 ) ŒD + T + -r + + +4- + ++ T Cry2Aa MNNVLMSGRTTICDAYNVVAHDPFSFEHKSLDTIQKEWMEWKRTDHSLYVAPVV DI MNNVLNSGRTTICDAYNVDAHDPFSFEHKSLDTIQKEWMEWKRTDHSLYVDPVV D2 MNNVLNSGRTTICDAYNVVAHDPFSFEHKSVDTIQKEWMEWKRTDHSLYVAPVV D3 MNNVLNSGRTTICDAYNVVAHDPFSFEHKSLDTIQKEWMEWKRTDHSRYVAPVV 04 MNNVLNSGRTTICDAYNVVAHDPFSFEHKS------RYVAPVV 05 MNN------VDAHOPFSFEHKSLOTIQKEWMEWKRTOHSLYVAPVV 06 MNNVLMSGRTTICOAYN------VOTIQKEWMEWKRTOHSLYVAPVV 07 MNN------VDPW Fig. 3.2. Amino acid sequence of the amino terminus of Cry2Aa and the mutant derivatives prepared for this study. The residue number and the tertiary structure of the protein sequence are shown on the top. The activated toxin of Cry2Aa starts from Val 50 (Audtho et a i, 1999). The mutant proteins, D I, D2, and D3, were constructed to examine the toxicity and normal production of the protein in E. CO//after replacing V a il9 and AlaSI by Asp, Leu31 by Val, and Leu48 by Arg. The deleted Cry2Aa, D4, 05, 06, and 07 were obtained by inverted polymerase chain reaction (Triglia at a i, 1988). The charged residues are designed as "+" on the upper line. 58 Deletion forms of Cry2Aa gene, D4, D5, 06, were constructed by inverted polymerase chain reaction technique (Triglia et al., 1988), using a pair of DNA primers listed in Table 3.1. Briefly, 5 ng of double stranded DNA template, 5 pmoles of each DNA primer, 10 nmoles ofdNTP, 1 unit of Pwo DNA polymerase were used in 50 ^1 PCR reaction. The temperature and time for annealing, extension, and dénaturation were 45°C for 1 min, 72°C for 4 min, and 90°C for 30 seconds, respectively. The reaction was performed for 35 cycles. The PCR product was purified by using QIAquick PCR purification kit (QIAGEN) and digested by the corresponding restriction enzymes. The restriction enzymes designed for D4 was Xba I, and for D5, D6, and D7 was Sal I. The digested PCR products were precipitated and used in ligation reaction at 14°C for 2 h before transforming E. coli DH5aF’. The mutant genes were verified by DNA sequencing. Automated DNA sequencing was performed using the BigDye™ Terminator Cycle Sequencing Ready Reaction (PE Applied Biosystems) following the manufacturer’s manual. The mutants shown in Table 3.3 were constructed by using the primers that have the “built in" restriction enzyme site. Therefore, only a certain amino acid was coded by each restriction site sequence. The mutants were screened by restriction enzyme digestion and then by DNA sequencing. 59 Mutant Proteins Oligonucleotides DI 5' GAT GCG TATAAT GTCGAC GCC CAT CATCCATTT 3 ' 5' GATCATAGTTTATATGTC GAC CCTGTA GTC GG 3 D2 5' GAACAT AAA TCAGTC GAT ACC ATCCAA AAA G 3 ' D3 5' AGAACA GATCATTCT AGA TAT GTA GCT CCT 3 ' D4 5' GGTATC TCT AGA TTT ATG TTC AAA ACTAAATGG ATG 3 5' GATCAT TCT AGA TAT GTAGCTCCTGTA GTC GGA 3 ' D5 5' TCCACT ATT GTCGACATTATTCATATAAAATTC CTC 3 5' TATAATGTC GAC GCC CAT CAT CCA T T T AGT T T T GAA 3 06 5' ATG GGC GTC GAC ATT ATACGCATC ACA AAT AGT TG 3 t 5' AAA TCAGTC GAT ACC ATCCAA AAA GAA TGG ATG GAG 3 07 5' TCCACT ATT GTC GAC ATTATT CAT ATAAAATTC CTC 3 5' TTATAT GTC GAT CCTGTA GTC GGAACT GTG TCT 3 ' Table 3.1. Oligonucleotides for site-directed mutagenesis and inverted polymerase chain reaction. The silent changes were introduced to create new restriction site in the mutagenic DNA primers as indicated by the underlines. The DNA sequence of the constructs were confirmed by DNA sequencing. 60 Protein Expression and Protein Purification Transcription of cry2Aa gene in pGEM 103-9 is under control of 17 RNA polymerase promoter. Transformation of E. coli BL21 (DE3) pLysS (Promega) with each construct was performed to accomplish T7-driven overexpression. Protein purification was performed according to Liang and Dean method (Liang and Dean, 1994). Briefly, E. coli containing each desired construct was grown in 500 ml Terrific Broth culture at 37°C for 48 h. The bacterial cells were collected and lysed by sonication. The lysate was resuspended and washed three times in 50 ml each of 0.5 M NaCI, 2% Triton X-100 and then three times in 0.5 M NaCI. The pellet was washed in 50 ml distilled water twice and resuspended in distilled water. Three ml of 50 mM NagCOs pH 10.5 were added in 5 grams of the pellet and the suspension was incubated at 37°C for 3 h to solubilize the protein. The suspension was then centrifuged at 12,000g for 15 min. The protein in the supernatant was considered as soluble protein. The pellet obtained after centrifugation was also determined for protein concentration by SDS-PAGE. Measurement of Protein Concentration and SDS-PAGE of Soluble and Crystal Proteins The concentration of protein soluble in the NaaCOa buffer was measured by using Coomassie Protein Assay Reagent (Pierce). The concentration of the protein in crystal form was measured by comparison of the intensity of the crystal protein bands in SDS-PAGE with the intensity of soluble protein band by using 61 Labworks™ Image Acquisition and Analysis Software (UVP, Inc.). Briefly, 40 ^1 of soluble protein solution was mixed with 10 |al 5X protein loading buffer and boiled for 5 min. For the crystal protein, the pellet was directly boiled in IX protein loading buffer for 5 min and then centrifuged at 12,000g for 10 min. The protein was separated in 10% polyacrylamide gel electrophoresis (Laemmli, 1970). The intensity of each protein band was scanned and the concentration of the active fragment band was calculated. Bioassays Forced-feeding was performed in third in-star larvae of gypsy moth (L. dispar). Serial dilution of the crystalline inclusion was prepared by resuspending the inclusion in distilled water. The soluble protein was diluted in 50 mM NazCOs pH 9.5. 2 pi of different toxin concentration of the soluble toxin or water- suspended inclusion were injected through the insect mouth. The insects were then allowed to feed on the normal diet. LCso was calculated from the percentage of the dead insects after incubation for 5 days. Probit analysis was used for data analysis (Raymond, 1985). Bioassay of A. quadrimaculatus was performed on two-day-old larvae of the mosquito. The larvae were placed into water containing different concentration of the toxin inclusion. The number of the dead larvae was monitored after incubation at 37°C for 24 h. 62 Thermal Unfolding of Cry2Aa and Deleted Protein Wild type Cry2Aa, chymotrypsin-digested L144G Cry2Aa (Audtho et al., 1999), and the D4 toxin were purified by HPLC using Sephadex G200 column. 40 pg of purified protein sample were loaded in a Thomas UV cuvette, containing 1.0 M guanidium hydrochloride (GnHCI) in 50 mM NazCOs buffer pH 9.5. The absorbance was detected by HP spectrophotometer with a diode array detector. The temperature was increased form 22.4 to 76 °C, at 2°C intervals with 2 min for equilibration after the proper temperature was reached. The absorbance spectra were followed at 222, 250 and 280 nm. Buffer and GnHCI were used for obtaining the base line. The data collected at 280 were used to analyze the unfolding behavior of the aromatic amino acids. 63 3.4 RESULTS Protein Expression Various mutant derivatives of Cry2Aa which were successfully overexpressed are listed in Table 3.2 and Table 3.3. Wild type Cry2Aa and mutants, D I , D2, D3, D5, and C13L were highly expressed in E. coli BL21 (DE3) pLysS, which contains the T7 bacteriophage gene I that encodes T7 RNA polymerase. The amount of the crystalline inclusion of the various mutant proteins was slightly lower than that of the wild type Cry2Aa (Fig. 3.3, lanes 3, 5, 7 ,1 3 , and 15). The lower amount of the inclusion was also observed in the D4 and D5 proteins (Fig. 3.3, lanes 9 and 11). We constructed mutants in which the amino acid residues involved in intramolecular and intermolecular interaction were altered. The results of the crystal yields and toxicity of these mutant proteins to both L. dispar and A. quadrimaculatus are shown in Table 3.3. Except C13L, the crystal yield of all mutant proteins was much lower than the wild type protein. Mutation of Lys42 and Arg43, which are likely involved in both intramolecular and intermolecular bondings, dramatically lowered the crystal yield (Table 3.3). Solubility and Toxicity of the Crystalline Inclusions All three mutant proteins, 01 (V I90, A510), 02 (L31V), 03 (L48R), showed the same solubility behavior as the wild type Cry2Aa (Fig 3.3, lanes 1-8, 64 M W 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 kDa 117 66 39 I Fig. 3.3. Solubility of Cry2Aa, mutant proteins and its N-terminus-deleted protein in 50 mM NaaCOa pH 10.5. Protein molecular standard is MW. Lanes 1, 3,5, 7, 9 ,11, 13, and 15 are the directly boiled crystalline inclusions of the Cry2Aa protein, D I, D2, D3, 04, D5, 06, and 07, respectively. Lanes 2, 4, 6, 8, 10,12, 14, and 16 corresponded to the soluble protein in crude lysates of E. co//which overexpressed Cry2Aa, D I, D2, D3, D4, D5, D6, and D7, respectively. 65 Table 3.2. Toxicity of Cry2Aa, its mutant, and N-terminus deleted proteins against gypsy moth {Lymantria dispar) and Anopheles quadrimaculatus larvae. ^ Crystal inclusions of the toxin were suspended in distilled water. Soluble proteins were in the supernatant of the 50 mM NaaCOa buffer after solubilization at 37°C for 3 h and centrifugation at 12,000g for 10 min. Lethal dose of the toxin against the third-instar larvae of gypsy moth by using force-feeding method. •^Toxicity was not observed at the toxin concentration up to 10,000 ng/larva. Toxicity was not observed at the toxin concentration up to 100,000 ng/m l. 95% confidence limit is shown in parentheses. 66 Proteins Sources * L D so (ng/larva)^ LCso (ng/ml) against L. dispar against A. quadrimaculatus Cry2Aa Crystal 6.9 (5.4 - 9.6) 41.2 (29.0 - 54.1) Cry2Aa Soluble 0.7 (0.4 - 2.1) - DI Crystal 8.6 (5.4 - 15.9) 33.2 (24.0 - 42.9) DI Soluble 1.2 (0.5 - 1.9) - D2 Crystal 10.5 (8.2 - 13.8) 28.5 (9.7 - 43.9) D2 Soluble 0.4 (0.08 - 0.9) - D3 Crystal 12.7 (7.0 - 22.7) 39.5 (24.8 - 53.6) D4 Crystal 6,778.8 (4,891.2- 9,235.1) ND= D4 Soluble 83.1 (61.3 - 133.7) - D5 Crystal 20.6 (7.4 - 38.9) 51.7 (16.4 - 87.9) D5 Soluble 3.3 (0.8 - 5.7) - D6 Crystal 5,977.3 (3,211.4 - 8,931.8) ND D6 Soluble 99.8 (69.5 - 149.4) - D7 Crystal ND** ND D7 Soluble -- Table 3.3. The role of amino acid residues of the N-terminus of Cry2Aa protoxin in formation of intermolecular and intramolecular hydrogen bonds and salt bridges, and the effect of mutation on the crystal yields and toxicity against L. dispar and A. quadrimaculatus larvae. ® Mutation was performed by using mutagenic DNA primer containing a certain restriction enzyme site. Toxicity against both L dispar and A. quadrimaculatus was observed. Moreover, the toxicity was comparable to wild type Cry2Aa. The high resolution structure does not implicate this residue in hydrogen bonding. ^ Not determined 68 Residues Intermolecular Intramolecuar Mutant Proteins^ Crystal yields Toxicity* Bonding Bonding Cry2Aa wild type -- +++++ Yes Cys 13 No': No C13L +++++ Yes His 21 No Asn 472 ND ND Asp 22 No Asn 472 ND ND ND Lys 29 No Glu 37. Asp 420 K29I ++ Yes o> Glu 37 No Lys 29 E37G ++ Yes < o Trp41 No Tyr417 W41L + Yes Lys 42 Glu 479 Glu 96 K42I -- Arg 43 Asn 480, Val 504 Glu 40 R43G -- Thr44 Gin 503 No T44S + Yes Asp 45 No Arg 375 (2 H bonds) D45G -- His 46 Glu 479 No H46I + Yes Ser 47 Glu 479 No E479A + Yes Tyr49 Ala 477 No Y49D + Yes Table 3.3. 11-12). The protein concentration obtained in 3 ml NagCOs buffer was about 2 mg/ml (data not shown). In addition, the results from forced-feeding experiments indicated that all of the above proteins were as toxic as wild type Cry2Aa against third-instar larvae of the gypsy moth, ranging from 6.9 to 12.7 ng/cm^ for the crystalline inclusions, and about 0.7-3.3 ng/cm^ for the soluble forms (Table 3.2). The inclusions of all mutant proteins were toxic to A. quadrimaculatus larvae. The LDso value of the wild type protein was 41.2 ng/ml, which were in the same confidence limits as the D I, D2, and D3 proteins (Table 3.2). Despite their lower crystal yields, mutation of the residues involved in intermolecular and intramolecular interactions did not change the solubility properties of the inclusion. The toxicity of these mutant proteins was not significantly different from the wild type Cry2Aa protein (Table 3.3). Solubility and Toxicity of the Deleted Cry2Aa Even though all deleted mutant delivatives were well overexpressed in E. coli, as shown in the inclusion form (Fig. 3.3, lanes 9,11,13, and 15), only the D5 protein was soluble in the carbonate buffer (Fig. 3.3, lane 12). Very low solubility in the buffer was found in the D4, and D6 proteins, and no solubility was found in the D7 deleted protein (Fig. 3.3, lanes 15-16). The concentration of the deletion derivatives of Cry2Aa bands obtained from the protein gel scanning for D4 and D6 was approximately 0.01 mg/ml, about 100-200 times lower than the soluble protein from the wild type toxin or the mutant proteins (Fig. 3.3, lanes 10, 70 and 14). Bioassay of the third instar L dispar larvae showed that the D5 protein was as toxic as the wild type Cry2Aa (Table 3.2). The LCso value of the soluble D4 protein against the insect was 83.1 ng/larva, which is about 118 times less toxic than the soluble wild type protein. The same result was also found in the soluble form of the D6 protein, which showed a toxicity of 99.8 ng/larva (Table 3.2). Unlike soluble forms, the inclusions of D4, D6 and D7 showed very low toxicity. The toxicity of the D4 and D6 was much less than that of the wild type Cry2Aa, i.e., 6,779 ng/larva, and 5,977 ng/larva, respectively. The toxicity of the inclusion of the D7 protein was not observed even thought up to 10,000 ng/cm^ of the protein was applied. Except for the 05 protein, which is as toxic as wild type Cry2Aa, none of the inclusions from the deleted protein, even at 100 pg/ml, were toxic against the mosquito larvae (Table 3.2). Stability of Deleted Toxin, Active Fragment, and Cry2Aa Protoxin The change in the absorbance at 280 nm, which is the maximal absorbance for Tyrosine, indicates the sharp conformational change of Cry2Aa, according to the change in temperature (Fig. 3.4). In this experimental condition, the temperature above 65 °C could denature all protoxin molecules. However, the Cry2Aa protoxin was more stable than the active fragment which is missing the first 49 amino acid from its N-terminus (Audtho ef a/., 1999). The melting 71 temperature (Tm) of the protoxin was 59.7 °C, while that of the active fragment was only 53.0 °C (Fig. 3.4). Surprisingly, the D5 deleted protein, which lacks 17 amino acids from the N-terminus, but still contains aO, showed approximately the same melting point as of the protoxin. 72 100 - 90 - 80 - % 70 - 2 a c 60 - D 10 ■ 0 —r- —r- - 1 45 5055 60 65 75 Temperature, °C Fig. 3.4. Thermal unfolding of the Cry2Aa protoxin, Cry2Aa active fragment, and the D5 mutant protein. The active fragment of Cry2Aa was obtained by HPLC- based purification of a reaction mixture in which the Cry2Aa protoxin was digested with chymotrypsin. 73 3.5 DISCUSSION It has been proposed that crystalline Inclusion bodies of Bacillus thuringiensis are attributable to massive production of the Cry protoxIn under Its strong promoter systems and stable mRNAs (Glatron and Rapoport, 1972; Wong and Chang, 1983; Baum and Malvar, 1995). Over-expresslon of recombinant protein by recombinant DNA technology also produces aggregated proteins known as Inclusion bodies (Kane and Hartley, 1989). It Is not know exactly how the Inclusion bodies are formed, but It Is thought that the protein Is partially or Incorrectly folded. The aggregated recombinant protein Is usually Insoluble and generally requires the use of strong denaturing agents. This can be a problem where native folded protein Is required. Unlike Inclusion bodies formed by other proteins, the Inclusions formed by Cry protein can be easily dissolved In any buffer at the pH >9, Indicating the unique assembly of each protoxIn subunit In the Inclusion. Our results showed that the presence of N-termlnal sequence Is necessary for the formation of the active biological Inclusion of Cry2Aa. The expressed wild type Cry2Aa protoxIn when dissolved In the carbonate buffer pH 10.5 Is toxic against both L dispar and A. quadrimacuiatus. However, Its mutant derivatives show different degrees of solubility and toxicity. Deletion of 14 amino acids (mutant D5; Val 4 - Asn 17) from the N-termlnal part of the Cry2Aa protoxIn slightly lowered the crystal Inclusion yield (Fig. 3.3, lane 11). The protein was normally dissolved In carbonate buffer (Fig. 3.3, lanes 12), and as toxic as 74 wild type protoxin (Table 3.2). The inclusion yield obtained from the D4, D6, and D7 proteins was as high as that from the wild type protein (Fig. 3.3, lanes 9,13, and 15), but lost in solubility was found. Therefore, the aO helix might play a role in the solubility of the inclusion of the Cry2Aa protoxin. Three-dimensional structure of Cry2Aa reveals several salt bridges and hydrogen bonding formed between amino acid residues at the N-terminus and the rest of the Cry2Aa molecule (Morse et al., 2001). As shown in Table 3.3, despite the short length of the N-terminus, the intramolecular salt bridges formed by the amino acid residues in this portion account for about one-fourth of the total number of salt bridges in the entire molecule. These include: Lys29 - Asp420, Lys42 - Glu96, and Asp45 - Arg375. Additionally, there are hydrogen bondings formed by several amino acids located on this short sequence, i. e., His21 - Asn472, Asp22 - Asn472, Lys29 - Asp420, Trp41 - Tyr417, and Asp45 - Arg375. The 7 °C increase in melting temperature of the D5 mutant protein and the protoxin compared to the chymotrypsinized Cry2Aa (Fig. 3.4) is likely a result of interaction of amino acids on aO with the other residues. The interaction of these amino acid residues might play a role in stabilizing the fold of the domain I a-helix bundle, enhancing the packing of domain III against domain I, and stabilizing the crystal (Morse at a/., 2001). Surprisingly, deletion of 49 amino acids of the N-terminus at the genetic level (mutant D7) resulted in the loss of solubility of the protein inclusion. Without the N-terminal domain, the protein might be incorrectly folded and could aggregate to form inclusion bodies in a 75 manner similar to that of heterologously overexpressed recombinant proteins in E. coli. On the other hand, after the protoxin is correctly folded and dissolved in the solution, the N-terminus can be removed by protease digestion without disturbing its toxicity (Audtho eta!., 1999). Cry2Aa contains twice as many intermolecular hydrogen bonds as CrylAa and Cry3A (Li etal., 1991; Grochulski eta!., 1995; Morse etal., 2001). The large number of intermolecular interaction of Cry2Aa protoxin in the inclusion might contribute to the formation of the sturdy crystals (Yamamoto and McLaughlin, 1981; English etal., 1994). Crystalline inclusion of Cry2Aa protein in bacterial cells is formed by interaction among protoxin molecules at the physiological pH, mainly through interactions of charged residues. There is only one cysteine residue in the N-terminus of the protoxin. Surprisingly, the crystal yield and solubility of the C13L mutant protein was not different from the wild type protein, indicating that this residue is not required for the crystallization process of Cry2Aa. There are a number of charged residues located at the aO helix (Fig. 3.1). These residues are exposed and accessible for the other protoxin molecules. Surprisingly, eight out of the total twenty six intermolecular hydrogen bondings of the entire protoxin, are formed by the amino acids in the aO helix (Table 3.3). The protoxin can form inclusions at low pH, and dissociate when the “holding forces" are weakened at high pH. The process of “association and dissociation" of the protoxin in crystal formation is reversible since the crystal formation of the 76 presolubilized Cry2Aa protoxin could be readily induced in vitro, by lowering pH of the buffer to below 8 (result not shown). However, at the same pH, the crystal of the protease-active Cry2Aa was not formed, indicating that the residues on the removed N-terminus may be involved in this process. The inducing crystal formation by lowering pH has been reported for other Cry protoxins (Bernhard, 1986; Dai and Gill, 1993). To establish a correlation between stability and crystal formation of the Cry2Aa protoxin, mutations of some amino acids that form intramolecular bondings were made (Table 3.3). These include Lys29, Glu37, Trp41, Lys42, Arg43, and Asp45. Lower crystal yield was found in mutation of Lys29, Glu37 and Trp41, while crystal formation was not observed for the mutation at Lys42, Arg43, and Asp45 (Table 3.3). The crystal yields of the these mutant proteins were very low, and the amount of the soluble form of the mutant proteins was insufficient for monitoring the changes in melting temperature caused by loss in a number of salt bridges and hydrogen bondings. Moreover, breaking salt bridges, Lys42 - Glu96, and Asp45 - Arg375, might destabilize the protoxin and eventually resulted in loss of the crystalline inclusions (Table 3.3). Correlation between solubility and toxicity of the Cry protein inclusion was observed (Jaquet et al., 1987; Aronson et al., 1991; Aronson, 1995). Highly toxic inclusions were solubilized at pH 9 to 10.5 (Bulla et al., 1977; Gill ef a/., 1992). It was found that inclusion solubilized only at pH 12 was not toxic to the insects (Pietrantonio and Gill, 1992; Du etal., 1994). Similar results were found in our 77 experiments. Inclusions of the Cry2Aa mutant proteins with the lower solubility at pH 10.5 were less toxic to insects (Fig. 3.2, and Table 3.2). Toxicity was not detected for either insect with the D7 protein, which failed to dissolve in the solution. The solubilized D4 and D6 proteins were toxic to the insect, but still less than the wild type protein. The lower toxicity observed might be caused by contamination of the “improperly folded" protein that was also released from the inclusion. This unfolded protein was easily destroyed during trypsin digestion. In contrast, the wild type protein exhibited much higher degree of protease resistance (result not shown). Cry2Aa is toxic to insects from both lepidoptera and diptera groups (Yamamoto and MaLaughlin, 1981). Despite its lower solubility, toxicity to a variety of insects was found (English et al., 1994). The high pH of the lepidopteran and dipteran midguts might facilitate solubility of this protein (Dadd, 1975; Dow, 1992). Unlike the alkaline pH environment midgut of the lepidopteran and dipteran, the pH of coleopteran midgut is about 4.5 to 5 (Koller et a/., 1992). However, more experiments have to be performed to determine whether lack in toxicity of Cry2Aa to some groups of insects, (e.g., coleopteran), is due to loss in receptor binding capability or loss of solubility of the toxin. 78 CHAPTER 4 DUAL SPECIFICITY OF BACILLUS THURINGIENSIS Cry2Aa TO GYPSY MOTH {LYMANTRiA DISPAR) AND MOSQUITO (ANOPHELES QUADRIMACULATUS) LARVAE IS DETERMINED BY DIFFERENT INITIAL BINDING SITES ON LOOP 1 AND LOOP 2 OF ITS DOMAIN II 4.1 SUMMARY Cry2Aa from Bacillus thuringiensis is toxic to both lepidopteran and dipteran insects. The protein consists of three domains and possesses an overall tertiary structure that is similar to CrylAa and CrySAa. Intensive alanine substitution study of both loops of domain II showed that toxicity of Cry2Aa to mosquito (Anopheles quadrimaculatus), and gypsy moth (Lymantria dispar), was affected by mutation of amino acids on loop 1 and loop 2, respectively. A2, the protein of which three amino residues on loop 1, were substituted by alanine was 270 times less toxic to A. quadrimaculatus, but as toxic to L. dispar as wild type toxin. A5, which ^op 2 was replaced by alanine, was as toxic to A. quadrimaculatus as wild type toxin but 140 times less toxic to L. dispar. Individual alanine scanning of each amino residue also indicated that the 79 mutant proteins F320A and P321A were less toxic to the mosquito larvae, while the R384A, and E385A were less toxic to the gypsy moth. The results from competition binding studies indicated that reduction in toxicity to each insect of the mutant proteins correlated to loss in initial binding to the brush border membrane vesicles (BBMV) from the insects. Our results also showed that irreversible binding was not affected by mutation of these amino residues. Therefore, specificity of Cry2Aa to the insects is probably determined by the initial binding of amino acids on its loop 1 and loop 2 of domain II. 80 4.2 INTRODUCTION Insecticidal crystal (Cry) proteins from Bacillus thuringiensis have been used as a safer alternative to chemical insecticides in sprays and in a number of transgenic crops (Dean and Adang, 1992). Upon ingestion by the insects, the protoxins in the crystals are dissolved and activated by insect midgut proteases. The activated toxins then bind to specific receptors on the midgut epithelial cells, and are believed to insert into the cell membrane, opening or forming pores that ultimately leading to cell lysis and eventually death of the insects. The tertiary structures of three Cry proteins, CrylAa, CrySAa, and Cry2Aa, have been solved (Li, etal., 1991; Grochulski et a i, 1995; Morse et a i, 2001). They share similar three-domain structures. Study of the structure-function relationship indicates that domain I is involved in pore formation in the cell membrane (Gill et a i, 1992; Wu and Arsonson, 1992; Alcantara et ai., 2001). Domain II has been shown to influence binding to the insect brush border membrane vesicles (Ge et ai., 1991 ; Lee et a i, 1992; Rajamohan et a i, 1995, 1996a, 1996b, 1996; Wu and Dean, 1996). Domain III was also reported to be involved in receptor binding in certain toxins (Lee et a i, 1995; de Maagd et ai., 1996a, 1996b, Jenkins et a i, 2000), as well as involved in ion channel activity (Chen et a i, 1993; Wolfersberger ef a/., 1996, Schwartz et a i, 1997c) The toxin-receptor binding is considered to be a critical step for specificity of Cry proteins toward certain groups of insects. To locate the region responsible for specificity-determination of Cry toxins, the highly variable domain II region has 81 been intensively studied. Homolog-scanning experiments in CrylAa showed that the Bombyx mori specificity region of the protein lies within its domain II (Ge et al., 1989; Ge etal., 1991; Lee etal., 1992). The amino acid residues involved in this binding are on loop 2 of the protein (Lu etal., 1994). The region containing these amino acids is believed to be the contact point in receptor binding since mutation of the amino acid residues of loop 2 of both CrylAa and CrylAb affected both reversible and irreversible binding to brush border membrane vesicles at Lymantria d/spar (Rajamohan etal., 1 9 9 5 ,1996c). Similarly, mutation of amino acid residues on loop 1 and loop 2 of CrySAa either increased or decreased toxicity of the toxin (Wu and Dean, 1996; Wu etal., 2000). The roles of domain III in receptor binding were reported in several Cryl proteins. The first report was focused on CrylAa and was published by Lee et al. (1995). In this study, domain III of CrylAa was replaced with the same domain from Cryl Ac. The hybrid protein exhibited the Cryl Ac binding properties and bound to gypsy moth aminopepidase N (APN). It was found later that the lectin like domain III of Cryl Ac binds to an W-acetylgalactosamine (GalNAc) moiety of the APN (de Maagd, et al., 1996; Burton et al., 1999; Jenkins et al., 2000). The binding was completely inhibited by pre-incubating the toxin with GalNAc (Masson et al., 1995). Domain III of CrylC was also found to be a specificity determinant for Spodoptera exigua, and substitution of domain III of CrylAb by domain III of CrylC yielded a hybrid protein which displays enhanced toxicity (de Maagd etal., 1996, de Maagd etal, 2000). 82 Cry2Aa is comprised of 633 amino acids and is toxic to both dipteran and lepidopteran insects (Yamamoto and McLaughlin, 1981; Donovan etal., 1988). The protein is 87% identical to Cry2Ab, which is not toxic to diptera larvae (Widner and Whiteley, 1989). Widner and Whiteley (1990) identified a region in domain II of Cry2Aa that was responsible for dipteran specificity, but the lepidopteran-specific region was still uncertain. Liang and Dean (1994) found that the dipteran-specific region of Cry2Aa was located between amino acids 278 to 340, while the amino acid residues 341 to 412 were involved in specificity against Lepidoptera. However, individual amino acid residues responsible for specificity to each group of insects were not determined. In the present study, we investigated the effect of mutation of amino acid residues on the loop 1 and loop 2 region of domain II on toxicity and binding properties of Cry2Aa. The results showed that mutation of amino acids on loopi affect toxicity and binding to mosquito {A. quadrimaculatus) larvae, but do not affect toxicity to gypsy moth (L. dispar) larvae. On the other hand, mutations of amino acids on the loop 2 region affect toxicity to only gypsy moth larvae. Binding studies showed that change in toxicity correlates with change in initial binding affinity of the mutant proteins to the receptors. However, change in irreversible binding was not observed in these mutant proteins, indicating that the initial binding step by different regions of Cry2Aa to its receptors might be a factor in determining specificity of the toxin to both insect groups. 83 4.3 MATERIALS AND METHODS Site-Directed Mutagenesis of Cry2Aa gene The cry2Aa gene construct, pGEM103-9, was made by subcloning the cry2Aa gene from pDL103 (Liang and Dean, 1994) into the pGEM-3Z(+) vector (Promega). Uracil-containing single-stranded DNA was obtained by transforming E. coli CJ 236 with pGEM 103-9. Site-directed mutagenesis to substitute the amino acids on loop 1 and loop 2 of domain II with alanine was performed by following the instruction in the manufacturer’s manual (Muta-Gene M l 3 in vitro mutagenesis kit; Bio-Rad). Automated DNA sequencing was carried out using a DNA sequencing kit (PE Applied Biosystems), following the manufacturer’s instructions. Replacement of F320, P321, D383, R384, and E385 by different amino acid residues was performed by using the DNA primers designed to encode a variety of the amino acids (Audtho et a/., 1999). The DNA primers used for F320, P321, D383, R384, and E385 were: 5’TCTATTACCTNCCCTAATATTGGT 3’, 5’ATTACCTTCCNTAATATTGGTGGT TTA 3’, 5’ TCAGGTACAGNNCGAGAGGGCGTT 3’, 5’GGTACAGATNNTGAG GGCGTTGC 3’, and 5’ACAGATCGTNNGGGCGTTGCTACC 3’, respectively. 84 Protein Expression and Purification Cry2Aa gene and its derivatives were expressed in E. coli BL21 (DE3) pLysS (Promega). Crystal inclusion bodies were purified as described (Lee et a!., 1992). The purified crystal proteins were solubilized in 50 mM NaaCOa pH 10.5 at 37 °C for 3 h. Protein concentration of the protoxin was measured by Coomassie Protein Assay Reagent (Pierce), and the protein purity was examined by separating the protein in 10% SDS-polyacrylamide gel electrophoresis (Laemmli, 1970). The concentration of the protein in crystal inclusion was measured by comparing the intensity of the crystal protein bands in SDS-PAGE to that of the soluble protein band by using Labworks™ Image Acquisition and Analysis Software (UVP, inc.). Briefly, 40 pi of soluble protein solution was mixed with 10 pi 5X protein loading buffer and boiled for 5 min. For the crystal protein, the pellet was directly boiled in IX protein loading buffer for 5 min and then centrifuged at 12,000 g for 10 min. The protein was separated in 10% polyacrylamide gel electrophoresis. The intensity of each protein band was scanned and the concentration of the active fragment band was determined. For binding experiments, the soluble protoxin of Cry2Aa and mutant proteins were activated by digestion with 2% (w/w) trypsin at 37 °C for 3 h. Further protein purification was performed using size-exclusion chromatography on an ÂKTA Explorer System (Pharmacia Biotech AB, Sweden). 85 Toxicity Assays L. dispar eggs were kindly supplied by Frank Martin (Otis Methods Development Center, U. S. Department of Agriculture, Beltsville, Md.). Toxicity of the toxin to the gypsy moth larvae was examined by surface contaminated methods as described (Rajamohan etal., 1996a). Briefly, serial dilution of the toxin in 50 mM NaaCOa buffer pH 10.5 was added onto the artificial gypsy moth diet (BioServ) in 24-well tissue culture plates. After the diet was air-dried, ten of the gypsy moth neonates were placed in each well. Mortality of the insects was monitored after incubation for 5 days. Bioassays were repeated at least 3 times. Lethal concentration (LC 50) of the toxin was calculated by using Probit analysis (Raymond, 1985). Toxicity assays against A. quadrimaculatus were performed on two-day old larvae. Six mosquito larvae were placed into 2 ml of water containing food and different concentrations of the crystal protein. Each toxin concentration was assayed in duplicate. Mortality of the larvae was monitored after allowing the larvae to feed for 24 h. Bioassay experiments were repeated 3 times. LC 50 values were determined by Probit analysis. Preparation of Midgut Proteases and In vitro Protein Digestion The midgut fluid of gypsy moth larvae was prepared as described (Audtho at a!., 1999). In brief, fourth-instar larvae of L d/spar were dissected and the midguts were recovered. The midguts were centrifuged at 15,000 g for 30 min 86 and the supernatant was collected and used as the midgut extract. The mosquito midgut fluid was prepared as described by Dai and Gill (1993), with minor modification. The mosquito midguts were dissected from the fourth-instar larvae and centrifuged at 15,000g for 30 min. The supernatant was used as the source of midgut proteases. In both midgut-protease digestion reactions, 100 pg/ml of the toxin solution was mixed with 100:1 (v/v) gypsy moth midgut fluid, or 10:1 (v/v) mosquito midgut fluid, and incubated at room temperature for 1 h. The reaction was stopped by adding to final concentration IX of Complete™ (BMB), a protease inhibitor, 2 i^M of E-64 (a papain inhibitor). The digestion products were determined by 10% SDS-PAGE (Laemmli, 1970). BBMV Preparation Brush border membrane vesicles (BBMV) of gypsy moth larvae were prepared from fourth-instar larvae following the differential magnesium precipitation method (Wolfersberger et al., 1987). BBMV of A. quadrimaculatus was prepared from fourth-instar whole larvae as described (Siva-Filha at ai, 1999). The pellet from the final step of BBMV preparation was resuspended in binding buffer pH 8.0, and the BBMV concentration was measured in protein concentration units using Coomassie Protein Assay reagent (Pierce). 87 lodination of the Proteins Protoxin of Cry2Aa and mutant protein were activated by 2% (by mass) trypsin at 37 °C for 3 h. lodination of the toxins was carried out as described (Rajamohan etal., 1996a). In brief, 25 pg of the activated toxin was iodinated using one lODO-BEAD (Pierce) and 1 mCi of Na^^®l as instructed by the manufacturer (Pierce). A 2.0-ml Excellulose column (Pierce) was used to separate the free iodine from the labeled toxin. The specific activity of Cry2Aa, A2, F320A, P321A, N322A, AS, R384A, and E385A were 0.36, 0.24, 0.51, 0.43, 0.18, 0.27, 0.33 and 0.39 mCi/mg, respectively. Saturation Binding Assays The saturation assays were performed with fixed amount of labeled toxin, but varying amounts of BBMV. One nM of the ^®l-labeled toxins were incubated with increasing amount of BBMV (10 to 1,000 pg/ml of L dispar BBMV, and 10 to 500 pg/ml of A. quadrimaculatus BBMV) in binding buffer (8 mM NazHPÛA, 2 mM KH2PO4, 150 mM NaCI, pH 8.0) at room temperature for 1 h. The bound toxins were separated from the unbound by centrifugation at 13,500 g for 15 min. The pellets were washed three times with binding buffer and counted in a gamma counter (Beckman). Total binding was obtained from the binding series without non-labeled toxin. Non-specific binding was obtained from the binding experiments in the presence of excess amount (1000 nM) of the unlabeled toxin. 88 Specific binding for each data point was obtained by subtracting total binding by non-specific binding. Competition Binding Assays Homologous and heterologous competiton assays were performed as described by Lee et al. (1992) with minor modifications: BBMV (10 pg) of L. d/spar larvae were incubated with 1 nM ^^®l-labeled toxins in 100 pi binding buffer at room temperature for 1 h. Free labeled toxins were separated from the BBMV- bound toxins by centrifugation. The BBMV pellets were washed three times in binding buffer. The final pellet was counted in a gamma counter (Beckman). Samples were assayed in duplicate for each data point, and the binding data was analyzed using the LIGAND program (Munson and Rodbard, 1980). Kcom. a homologous competition constant, was calculated based on the relative binding affinity. Competition assays of ^^^l-labeled toxins and the A. quadrimaculatus BBMV were performed as described above, except 5 pg of the mosquito larval BBMV were used in each binding reaction. Dissociation Binding Assays 1 nM of ^^^l-labeled toxin was incubated with 10 pg of the L. d/spar BBMV, or 5 pg of the A. quadrimaculatus BBMV, for 60 min to achieve saturation binding. After association binding, the mixtures were diluted two-fold in binding 89 buffer containing 1000 nM final concentration of the corresponding unlabeled toxin. The reaction was stopped at different time points by centrifugation. The pellets were washed three times with binding buffer and the final pellet was counted in a gamma counter (Beckman). 4.4 RESULTS Mutagenesis and Expression of Loop 1 Mutant Proteins According to the structure of Cry2Aa shown in Fig. 4.1A-4.1C, and Fig. 4.2, amino acids on loop 1 and its proximity were replaced by alanine. There were 4 mutant derivatives, A1 (SIT 317.318AAA), A2 (FPN320-322AAA), A3 (IGG323- 325AAA), and A4 (LPG 326-328AAA), in each of which three amino acids were replaced by alanine (Fig. 4.3). The residues, starting from 8317 to H333, were also individually mutated. The A1, A2, and A3 proteins were highly expressed in £. coli, while no crystal protein was obtained from the A4 construct. The soluble forms of the T330A, T331A, T332A, and H333A were not stable during the trypsin digestion (result not shown). Therefore, a toxicity test of the product from these constructs was not performed (Table 4.1 and 4.2). Replacing F320, and P321 by some amino acids using “spin codon" DNA primers was also successful. DNA isolated from twenty colonies of £. coli was sequenced. Six different mutations were found, F320C, F320S, F320Y, P321H, 90 CrylAa Cry2Aa \ Loop 2 ^ Loop I Loop 2 ^ Loop I Fig. 4.1 A. Structures of CrylAa and Cry2Aa shown in ribbon with loop 1 and loop 2 as indicated. 91 CrylAa CrylAa n : \ ’■■ '-V L . i V. V, C , t K. V.. \v V 1318 S390 D308 F320 V 3 8» -> -l L326 U 7 1 ^ ^ G372 G374 G324 Fig. 4.1B. Structures of CrylAa and Cry2Aa shown in space-fill. Amino acid residues on loop 1 and loop 2 are indicated. Note that I 318 and T319 are located at the end of p2. 92 CrylAa Cry2Aa ' '■■-'i'- ■* ' i'. Yk- E38S R368 G386 L371 G374 Fig. 4.1C. A spacefilling representation of the tertiary structures of CrylAa and Cry2Aa showing the position of of CrylAa, and 383(^^^385 cry2Aa. The indicated residues in the CrylAa protein were reported to be an initial binding site for lepidopteran insects (Lu et al., 1994) 93 p2 P3 Cry2Aa m ^ m CrylAa " " Cry2Aa 305 NYILS6ISGTRLSITFPNIG6LPGSTTTHS 334 CrylAa 296 MDILNSITITYTDV------HRGFNY 315 B P7 Cry2Aa : CrylAa : Cry2Aa : 379 DSGTDREGV----- ATSTNW 393 CrylAa : 363 SPLYRRIILGSGPNNQELFVL 383 Fig. 4.2. Sequence alignment of loop 1 (A) and loop 2 (B) regions of CrylAa and Cry2Aa. The p-strands p2, pS, p6, and p7, are shown on the upper lines. 94 P2 P3 Cry2Aa 305 NYILS6IS6TRLSITFPMIGGLP6STTTHS 334 A1 305 NYILSGISGTRLAAAFPNIGGLPGSTTTHS 334 A2 305 NYILSGISGTRLSITAAAIGGLPGSTTTHS 334 A3 305 NYILSGISGTRLSITFPNAAALPGSTTTHS 334 A4 305 MYILS6ISGTRLSITFPNIGGAAASTTTHS 334 B P6 Cry2Aa 379 DSGTDREGV-- -- ATSTNW 393 A5 379 DSGTAAAGV-- -- ATSTNW 393 A6 379 DSGTDRE^------A ^ T N W 393 A7 379 DSGTDREGV-- -- ATAAAW 393 Fig. 4.3. Diagram showing triple alanine replacement of amino acid residues on the loop 1 (A), and loop 2 (B) regions, of Cry2Aa. The sequence of the Cry2Aa protein Is shown on the top line. A1 to A7 are the mutants. 95 Proteins LCso (ng/cm^) LCso (ng/ml) L dispar A. quadrimaculatus Cry2Aa 6.8 (3.1 -9 .4 ) 38.3 (20.1 -6 4 .3 ) A1 4.2 ( 2 . 4 - 6 .8 ) 36.7(17.7-60.8) A2 3.7 (1.3-5.5) 5,899.7(3.211-9,455) A3 7.4 (3.4-12.8) 27.9(15.1 -52.6) A4 ND*ND AS 956.5 (678-1,213) 30.8(11.2-59.7) A6 12.8 (4.2-23.1) 42.3 (20.0 - 68.5) A7 ND ND Table 4.1. Toxicity of Cry2Aa and its mutant proteins to gypsy moth (L. dispar), and mosquito {A. quadrimaculatus) larvae. * Stable crystal proteins were not obtained from the A4 and >45 constructs and their toxicity was therefore not determined. Confidence intervals (95%) are given in parentheses. 96 Table 4.2. Toxicity of the mutant derivatives in which the amino acid residues on loop 1 of Cry2Aa were replaced with alanine. Note that the crystal proteins from T330A, T331A, T332A, and H 333A are not stable during trypsin digestion. Confidence intervals (95%) are given in parentheses. 97 Proteins LCso (ng/cm^) LCso (ng/ml) against L. dispar against A. quadrimaculatus Cry2Aa 6.8 (5.2 - 9.6) 38.3 (20.1 - 64.3) S317A 9.7 (4.0-17.3) 61.8 (43.5-82.9) I318A 8.0 (6.4-10.7) 33.8(17.9-47.5) T319A 7.5 (5.8-10.2) 33.2 (24.0 - 42.9) F320A 10.3 (6.0-17.1) 4,202.1 (2,550 - 8,545) P321A 7.0 (5.6-11.6) 3,112.8 (2,124-5,899) N322A 7.1 (6.5-8.0) 28.5 (9.7-43.9) I323A 12.6(10.4-15.6) 22.8 (4.8 - 36.3) G324A 13.7(10.1 -15.7) 51.7(16.4-87.9) G325A 15.2(12.4-18.7) 39.5 (24.8 - 53.6) L326A 6.3 (5.1 -8 .2 ) 56.4 (22.6 - 93.4) P327A 5.8 (4.3-7.5) 76.9 (56.1 -102.0) G328A 6.9 (3.4-12.2) 62.2 (37.3-90.8) S329A 8.5 (7.6-9.7) 29.5 (20.5 - 37.8) T330A Unstable protein ND T331A Unstable protein ND T332A Unstable protein ND H333A Unstable protein ND 98 P321N, and P321R. All constructs, except P321R, produced stable crystal proteins. Expression of Loop 2 Mutant Proteins Fifteen mutants, i.e., A5 (DREsas-sasAAA), A 6 (GVAT386-389AAAA), A7 (STN390-392AAA), and individual residues (T382A to W393A) at or proximal to loop 2, were successfully constructed. The A5 and A 6 proteins were highly expressed, but no crystal protein was produced by the A 7 construct. For individually mutated residues, all proteins were stable during trypsin digestion, except D383A, S390A, and W393A (Table 4.3). Another eleven mutations at D383, R384, and E385 were obtained from “spin codon" DNA primers. These were D383G, D383V, D383E, R384H, R384P, R384L, R384N, R384D, E385G, E385R, and E385N. However, not all constructs produced stable crystal proteins. The stable proteins were obtained only by D383V, D383E, R384L, R384D, E385G, and E385N (Table 4.4). Toxicity Assays of the Mutant Proteins In L. dispar Larvae All of the mutant proteins from the mutation of the amino acids on loop 1, and loop 2 were toxic to L. dispar larvae. The LCso of the wild type Cry2Aa was 6.8 ng/cm^, which was not significantly different from the other mutant proteins (Table 4.1, 4.2,4.3). Loss in toxicity toward the insect larvae was found in A5 (DRE383-385AAA), R384A, and E385A, with the LCso at 956.5 ng/cm^ (Table 4.1), 99 Proteins LCso (ng/cm^) LCso (ng/ml) against L dispar against A. quadrimaculatus Cry2Aa 6.8 (5 .2 -9 .6 ) 38.3 (20.1 -6 4 .3 ) T382A 14.8(11.5-17.4) 44.1 (1 1 .4 -1 5 0 .3 ) D383A Unstable protein ND* R384A 52.1 (4 0 .2 -7 1 .8 ) 38.1 (1 7 .9 -7 0 .2 ) E385A 38.3 (27.7-51.3) 60.7 (28.7-122.3) G386A 5.8 (4.5 - 8.0) 65.2 (37.2-176.1) V387A 13.0 (8 .5 -1 8 .7 ) 48.2 (23.0-110.3) A388G ND ND T389A 5.4 (4.1 - 7 .1 ) 60.1 (21.1 -1 6 5 .4 ) S390A Unstable protein ND T391A 7.7 (4.8-16.9) 41.1 (29.0-54.1) N392A 6.1 (3.8-12.6) 57.2 (29.3-175.0) W393A Unstable protein ND Table 4.3. Toxicity of the mutant proteins from alanine-scanning mutagenesis of the amino acid residues on loop 2 of Cry2Aa. The proteins from the mutants D383A, A388G, S390A, and W333A, were not stable. •Toxicity was not determined. Confidence intervals (95%) are given in parentheses. 100 Proteins LCso (ng/cm^) D383A ND* D383G ND D383V 30.9 (24.1 -4 0 .8 ) D383E 5.0 (2.9-10.1) R384A 52.1 (4 0 .2 -7 1 .8 ) R384H ND R384P ND R384L 29.6 (20.4 - 50.5) R384N ND R384D 102.6 (67.8-150.7) E385A 38.3 (27.7-51.3) E385G 47.5 (33.8-65.8) E385R ND E385N 12.1 (8.1-17.7) Table 4.4. Effect of mutations at D383, R384, and E385 on toxicity against L. dispar larvae. * Toxicity was not determined since overexpression of the corresponding constructs resulted in unstable proteins. 101 52.1 ng/cm^ and 38.3 ng/cm^ (Table 4.3), respectively. Slightly less toxicity was observed in I323A, G324A, and G325A with the toxicity values at 12.6 ng/cm^, 13.7 ng/cm^, and 15.2 ng/cm^, respectively (Table 4.2). However, the confidence limits of these proteins tend to overlap with the other mutant proteins (Table 4.2), and therefore, a binding study of these proteins was not performed. The mutant proteins obtained from substitution of D383, R384, and E385 by different amino acids showed different degree of toxicities, depending on the identity of the replacing residues. Replacing D383 by Val (D383V) led to loss in toxicity, while the D383E protein retained the wild type toxicity (Table 4. 4). Substituting Arg384 by Asp (R384D) produced a dramatically less toxic protein with LCso of 102.6 (Table 4.4). Less toxic protein was also found as a consequence of the E385G mutation. However, replacing Glu385 by Asn (E385N) produced a protein as toxic as the wild type, with the LCso at 12.1 ng/cm^ (Table 4.4). Toxicity Assays of the Mutant Proteins against A. quadrimaculatus Larvae The mutant derivatives used in bioassays against L. dispar larvae were applied to the mosquito {A. quadrimaculatus) larvae. LCso of the wild type Cry2Aa against the mosquito larvae was 38.3 ng/ml (Table 4.1). LCso of all mutant proteins, except A2 (FPN 320-322AAA), F320A, and P321A, ranged from 22.8 to 76.9 ng/ml (Table 4.1, 4.2, 4.3), which were not significantly different from the wild type protein. The A2 (FPN 320-322AAA), F320A, and P321A proteins were 102 far less toxic to the mosquito larvae with LC 50 values of 5, 899.7 ng/ml (Table 4.1), 4,202.1 ng/ml, and 3,112.8 ng/ml respectively (Table 4.2), Less toxic proteins against the mosquito larvae were found when F320 was substituted by Cys, or Ser. LC 50 values of F320C, and F320S were 3,822 and 2,876 ng/ml, respectively (Table 5). However, F320Y was as toxic as the wild type protein, with LC 50 of 10.6 ng/ml (Table 4.5). In vitro Protease Digestion To test whether loss in toxicity of the A2 protein to A. quadrimaculatus larvae, and of the A5 protein to L dispar larvae was caused by proteolytic digestion of the insect midgut proteases, both proteins were incubated with the diluted midgut fluid. The final product of the A2 protein after digestion in the midgut fluid of the L. dispar larvae and A. quadrimaculatus larvae were approximately 58 kDa (Fig. 4.4, lanes 3, and 4), which are the size of the Cry2Aa active fragment reported by Audtho at ai. (1999). The molecular masses of the digested products of the A5 protein in L d/spar and A. quadrimaculatus midgut fluid were also about 58 kDa (Fig. 4.4, lanes 6 and 7). Saturation Binding and Effect of Mutation on Binding to L dispar BBMV Saturation binding was observed when increasing amount of L. dispar BBMV was incubated with fixed amount of ^^®l-labeled toxin (Fig. 4.5). The bound toxin was saturated when 100 pg /ml of the BBMV was used. Therefore, the 103 Proteins LCso (ng/ml) F320A 4,202.1 (2,550.2-8,545.3) F320C 3,822.3(1,741.6-6,762.4) F320S 2,876.2(1,330.4-5,442.8) F320Y 10.6 (5.1-18.5) P321A 3,112.8 (2,124.2-5,899.0) P321H 9.1 (4.4-18.5) P321N 1,523.7 (836.7 - 2,668.4) P321R ND* Table 4.5. Effects of mutations of F320, and P321 on toxicity against A. quadrimaculatus larvae. * Stable protein was not obtained from the P321R construct. 95% confidence intervals are given in parentheses. 104 NOTE TO USERS Page(s) not included In the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received. 105 This reproduction is the best copy available. UMI the same amount of both labeled toxin and BBMV were used In all binding experiments. The results from association binding study of Cry2Aa, A2, F320A, P321A, N322A, A5, R384A, and E385A showed that there is a correlation between relative toxicity of the mutant proteins to the binding affinity, Kcom (Table 4.6). A2, F320A, P321A, and N322A competed for binding with Cry2Aa to the insect BBMV the same way that the wild type protein did (Fig. 4.6). On the other hand, R384A, and E385A slightly compete with the wild type protein. A lower competition capacity was found in the A5 protein (Fig. 4.6). The binding affinity values of the A5, R384A, and E385A proteins are 91.33, 50.56, and 39.08 nM (Table 4.6), approximately 8, 4, and 3 times, respectively, higher than the wild type Cry2Aa protein. The role of D383, R384, and E385 on binding of Cry2Aa to the L dispar BBMV was determined by altering the identity of these residues. A lower binding competition with the wild type protein was found in D383V, R384L, R384D, and E385G, while D383E, and E385N showed similar binding behavior to wild type Cry2Aa (Fig. 4.7). 106 14000 12000 10000 8000 6000 4000 Total binding Non-specific binding 2000 Specific binding 2000 400 600 800 1000 1200 L cf/spar BBMV concentration (pg/ml) Fig. 4.5. Saturation binding of ^^®l-labeled Cry2Aa to L dispar BBMV as a function of the concentration of BBMV. 1 nM of labeled Cry2Aa was incubated with different concentration of BBMV. Total binding is determined from the CPM of the bound toxin to BBMV without 500 nM cold toxin in the binding reaction. Non-specific binding is obtained from the CPM of the bound toxin in the presence of 500 mM cold toxin. Specific binding is calculated from the difference of total binding and non-specific binding. 107 Proteins L. dispar A. quadrimaculatus Relative toxicity* Kcom (n M f Relative toxicity* Kcom(nM) Cry2Aa 1.0 12.23 + 2.17 1.0 17.51 ±2 .5 5 A2 0.6 11.97 + 3.34 70.6 252.55 + 5.29 F320A 1.5 14.51+2.58 109.7 121.85 + 7.54 P321A 1.1 11.77 + 1.37 81.2 60.93 + 5.33 N322A 1.0 13.94 + 2.50 0.7 15.21 +1.01 A5 140.7 91.33 + 2.21 1.4 17.29 + 3.22 R384A 7.7 50.56 + 4.99 1.0 22.74 + 6.94 E385A 5.6 39.08 + 3.32 1.6 19.02 + 4.09 Table 4.6. Relative toxicity and binding affinity of Cry2Aa mutant proteins to L dispar and A. quadrimaculatus. ^ Relative toxicity yielded by wild-type toxin LCso/mutant toxin LCso- ^ Dissociation constant calculated from homologous competition assays. 108 100 - c 1 1 a m N322A R384A 0.1 10 100 1000 10000 Competitor concentration (nM) Fig. 4.6. Binding of ^^®l-labeled Cry2Aa toxin to A. quadrimaculatus BBMV in the presence of increasing concentrations of nonlabeled Cry2Aa, A2, F320A, P321A, N322A, A5, R384A, and E385A. Binding is expressed as a percentage of the amount of bound toxin upon incubation with labeled toxin alone. 109 120 100 c 80 4 X 2 1 60 - Cry2Aa D383V m D383E R384L 40 - R384D E385G E385N 20 - 0.1 10 100 1000 10000 Competitor concentration (nM) Fig. 4.7. Binding of ^^®l-labeled Cry2Aa toxin to L. dispar BBMV in the presence of increasing concentrations of unlabeled Cry2Aa, D383V, D383E, R384L, R384D, E385G, and E385N. Binding is expressed as a percentage of the amount of bound toxin upon incubation with labeled toxin alone. 110 Saturation Binding and Effect of Mutation on Binding to A. quadrimaculatus BBMV Cry2Aa specifically binds to the A. quardimaculatus BBMV (Fig. 4.8). When 1 nM of the labeled Cry2Aa was applied, saturation binding was accomplished at 50 ng/ml of the mosquito BBMV (Fig. 4.8). The results from association binding experiments using 50 pg/ml BBMV showed that binding of the same mutant proteins to A. quadrimaculatus BBMV is different from L. dispar BBMV. In experiments with A. quadrimaculatus BBMV, A5, R384A, and E385A apparently competed for binding with the wild type protein, while A2, F320A, and P321A showed lower binding affinity (Fig. 4.9). The binding affinity of A2, F320A, and P321A are 252.55, 121.85, and 60.93 nM (Table 4.6), which are about 14, 7, and 4 times, respectively, higher than the wild type protein. The effect of replacing F320 and P321 with other amino acids on binding to the mosquito BBMV was also determined. Substituting F320 with Tyr (F320Y) bound to the membrane similarly to the wild type protein, but replacing it with Ser (F320S) led to a lower binding affinity (Fig. 4.10). A lower binding affinity was also observed in the P321N protein (Fig. 4.10). The Effect of Mutation on Dissociation Binding After saturation binding was achieved, and the higher concentration of the corresponding non-labeled toxin was added, dissociation binding of each mutant protein was observed at different time points. A significant change in dissociation 111 Fig. 4.8. Saturation binding of ^^^l-labeled Cry2Aa to A. quadrimaculatus BBMV as a function of the concentration of BBMV. 1 nM of labeled Cry2Aa was incubated with different concentrations of BBMV. Total binding is determined from the CPM of the bound toxin to BBMV without 500 nM cold toxin in the binding reaction. Non-specific binding is obtained from the CPM of the bound toxin in the presence of 500 nM cold toxin. Specific binding is calculated from the difference of total binding and non-specific binding. 112 3000 - 2500 - 2000 - O 1500 - 1000 - — Total binding — Non-specific binding 500 - — A — Specific binding 0 100 200 400 500300 600 A. quadrimaculatus BBMV concentration (pg/ml) Fig. 4.8. 113 120 100 - 80 - X g ■o c 60 3 2Aa W T O CD F320A 40 P321A R384A 20 0.1 10 100 1000 10000 Competitor concentration (nM) Fig. 4.9. Binding of ^^®l-labeled Cry2Aa toxin to A. quadrimaculatus BBMV in the presence of increasing concentrations of nonlabeled Cry2Aa, A2, F320A, P321 A, A5, R384A, and E385A. Binding is expressed as a percentage of the amount bound toxin upon incubation with labeled toxin alone. 114 120 100 - 80 - c I 1 60 - m 40 - Cry2Aa F320S F320Y 20 - P321N 0.1 1 10 1000 10000100 Competitor concentration (nM) Fig. 4.10. Binding of ^^®l-labeled Cry2Aa toxin to A. quadrimaculatus BBMV in the presence of increasing concentrations of nonlabeled Cry2Aa, F320C, F320S, F320Y, P321H, and P321N. Binding is expressed as a percentage of the amount bound toxin upon incubation with labeled toxin alone. 115 to L. dispar BBMV was not observed in the mutant proteins (A5, R384A, and E385A) (Fig. 4.11 ). A similar result was found in the dissociation binding of A2, F320A, and P321A to the BBMV from A. quadrimaculatus larvae (Fig. 4.12). 116 120 100 - c 1 TJ C 60 - o ÛO 40 - A5 R384A E385A Cry2Aa 20 - 0 20 40 60 80 100 120 140 Post incubation time (min) Fig. 4.11. Dissociation of bound ^^®l-labeled toxins from L. cf/spar BBMV. 10 ng of L. dispar BBMV were incubated with 1 nM ^^®l-labeled Cry2Aa, A5, R384A, and E385A for 60 min. After association reaction, 1000 nM of the corresponding nonlabeled toxins were added to the samples and post-binding incubation was continued. Binding is expressed as a percentage of the amount of bound compared with the amount bound at 0 min post-incubation. 117 120 100 - c SO- 1 1 60 - m 40 - Cry2Aa A2 F320A P321A 20 - 0 20 40 60 80 100 120 140 Post incubation time (min) Fig. 4.12. Dissociation of bound ^^®l-labeled toxins from A. quadrimaculatus BBMV. 10 fiQ of 4. quadrimaculatus BBMV was incubated with 1 nM ’^®l-labeled Cry2Aa, A2, F320A, and P321A for 60 min. After association reaction, 1000 nM of the corresponding nonlabeled toxins were added to the samples and post binding incubation was continued. Binding is expressed as a percentage of the amount of bound compared to the amount bound at 0 min post-incubation. 118 4.5 DISCUSSION The molecular mechanism that confers biological specificity of 8 . thuringiensis 5-endotoxin has been intensively investigated in order to produce more specific insecticides that target only a certain group of pest insects. A more specific toxin is pesticide that can reduce indiscriminate environmental consequences caused by insecticide application. This study focused on the determination of amino acids in Cry2Aa that are involved in the dual specificity of the protein to Diptera and Lepidoptera. The overall structure of the loop 1 and loop 2 in CrylAa and Cry2Aa are strikingly similar. In Cry2Aa, loop 1 is longer, while loop 2 is shorter (Fig. 4.1A- 4.1 C). Several amino acids on these 2 loops of both proteins are exposed, indicating the possible binding contacts for protein-receptor interaction. The specificity-determining region of several Cryl proteins has been located primarily in domain II for several lepidopterans (Ge etal., 1989: Schnepf ef a/., 1990; Masson etal., 1994; Rajamohan etal., 1996a, 1996b). To identify specificity- determining residues of Cry2Aa, 17 amino residues located on loop 1 (and its proximal region) and 14 residues on loop 2 (and its proximal regioin) were mutagenized. The mutant proteins were then tested for their toxicity to gypsy moth (L. dispar) and mosquito (A. quadrimaculatus), and their binding properties to the insect BBMV. It is apparent that toxicity of Cry2Aa to A. quadrimaculatus is determined by amino acid residues on loopi of domain II, while toxicity to L. dispar is 119 determined by amino acids on loop 2. The change in binding and toxicity of Cry2Aa to A. quadrimaculatus was found in the triple mutation, A 2 (FPN320- 2AAA). This mutant protein was 270-fold less toxic to the mosquito larvae (Table 4.6). Loss in toxicity of the A2 protein was not due to proteolytic cleavage because the toxin was stable during digestion with the gut fluid extracted from A. quadrimaculatus larvae (Fig. 4.4, lane 4). The results from competition binding experiments showed that this mutant protein bound less tightly to the mosquito BBMV (Fig. 4.9), with 14-fold greater binding affinity (Kcom). An individual mutation indicated that N322 might not be involved in toxicity or receptor binding since a change in both properties was not observed in N322A (Table 4.2 and Fig. 4.6). In contrast, only F320A and P321A exhibited lower toxicity and lower binding affinity to the A. quadrimaculatus larvae (Tables 4.2 and 4.6). Loss in toxicity of the A2, F320A, and P321A proteins was also observed when these proteins were applied to Anopheles gambiae (results not shown). Substitution of F320 with different amino acid residues indicated that a large hydrophobic residue is preferred at this position. F320Y was as toxic as the Cry2Aa protein, while the wild type toxicity was not retained when this residue was replaced by a small amino acid such as alanine, cysteine and serine (Table 4.5). The residue P321 might contribute to the structural binding region for mosquito receptors. Among 4 amino acids used, Ala, His, Asn and Arg, replacing P321 by only His (F321H) could yield the wild type toxicity (Table 4.5). 120 Our results are in agreement with Liang and Dean (1994) who found that replacing a peptide fragment of domain II from amino acids 278 to 340 of Cry2Aa by the corresponding fragment from Cry2Ab caused loss in mosquitocidal toxicity of the Cry2Aa protein. Comparable results were found in CrySAa (Wu and dean, 1996). Mutation of amino acids on loop 1 of CrySAa affected toxicity to yellow mealworm {Tenebrio molitor) and Colorado potato beetle {Leptinotarsa decemlineata). The similarity between CrySAa and Cry2Aa is that they both have the long loop 1 and short loop 2 regions (Li et al., 1991; Morse et al., 2001). The most dramatic change in toxicity of CrySAa was found when mutations were simultaneously introduced on three amino acids in loop 1 (RS45A, YS50F, and YS51F). Surprisingly, deletion of YS50 and YS51 (RS45A, YS50, YS51) could increase toxicity and binding affinity of CrySAa to the insects (Wu etal., 2000). However, there is uncertainty whether this region is a specificity-determining region for Coleoptera. It should be mentioned that the residues FS20 and PS21 are also present in Cry2Ab, which is not toxic to mosquitoes. Specificity determination for mosquito might require another contact region that is present in only Cry2Aa, not Cry2Ab. The region might facilitate Cry2Aa to bind to its receptor through amino acids on loop 1. This might include more than one binding step as found in binding of Cryl Ac to its receptor proposed by Jenkins et al. (2000). The latter protein is proposed to bind to the receptor by docking of its domain III with 121 GalNAc moiety on the receptor, followed by binding of the higher affinity domain II region. The amino acids responsible for toxicity of Cry2Aa to L. dispar are located on loop 2 of the toxin. Loss in toxicity to gypsy moth larvae was observed when residues D383, R384, and E385 were mutated. Dramatic change in toxicity was observed in the ASmutant (DRE 383-5AAA) (Table 4.1). Even though crystal protein from D383A was not obtained, the lower toxicity was still found in R384A, and E385A (Table 4.1). The same mutant protein samples were applied to A. quadrimaculatus, but a change in toxicity was not observed (Table 4.3). Surprisingly, amino acid alignment showed that the position of Cry2Aa is approximately in the same position as the residues 367pp|369 CrylAa (Fig. 4.10, and 4.2). These residues in CrylAa were reported to be involved in reversible binding of CrylAa to 8 . mori BBMV (Lu et al., 1994). Mutation or deletion of these residues dramatically affected both toxicity and binding affinity to the insect receptors. Examination of the similar residues, 368rrr37o Cryl Ab was also reported (Rajamohan at ai., 1996). In the latter, replacing these three residues by alanine abolished its potency toward M. sexta and H. virescens. It is remarkable that similar residues in these three proteins are aligned in the same position. To demonstrate what kind of amino acids can be used to replace this lepidopteran specific region, each residue was mutagenized using “spin codon". Unstable proteins were found in several mutations (Table 4.4). These residues 122 are located at the tip of the p 6 strand (Fig. 4.2) and may contribute to the structural integrity of the protein. However, some mutant proteins were still obtained. When D383 was substituted with Glu (D383D), the wild type toxicity was retained, suggesting that a charged residue might be required for this position (Table 4.4). A more interesting substitution was found in R384. Substitution of this residue by a negatively charged amino acid (R384D) significantly reduced the protein’s toxicity (Table 4.4). The less selective substitution might be at E385. E385N showed the wild type toxicity, while E385G was less toxic, indicating that the binding of toxin and receptors is not strictly through charge-charge interaction. Another result that is in agreement among mutation of the residues mentioned above in CrylAa, CrylAb, and Cry2Aa is their role in reversible and irreversible binding. The residues Q^yiand 368ppp37o c^ylAb, were shown to play a role in reversible binding (Lu eta/., 1994; Rajamohan eta/., 1996). The same binding property was found for the residues ^w^pgsas Cry2Aa (Fig. 4.9). Any mutation of these amino residues that affected toxicity exhibited changes in binding affinity (Table 4.4, Fig. 4.7). Therefore, correlation between binding affinity and toxicity of the proteins should be established. The residues ^^^DRE^^^ of Cry2Aa were not involved in irreversible binding. It was demonstrated that N372 and 1375 of CrylAb were involved in irreversible binding to /W. sexta (Rajamohan eta/., 1995). Homology modeling of CrylAb with CrylAa (Grochulski et a/., 1995) shows that N372 is located on loop 123 2, at the tip of domain II, while 1375 is closer to the p7 strand (data not shown). Interestingly, the structurally comparable residues are on loop 1 of Cry2Aa (Fig. IB ). These might be 1323, G324, G325, or L326. However, further study is required to determine whether these residues are involved in irreversible binding as found in CrylAb. In summary, the amino acid residues responsible for dual-specificity of Cry2Aa to A. quadrimaculatus and L. d/spar were identified. The residues contributing to dipteran specificity are located on loop 1 of domain I. Alteration of F320 and P321 significantly affect toxicity and receptor binding. Interestingly, the region responsible for lepidopteran specificity is distantly located on loop 2 of the domain II. These results might be useful for engineering Cry2Aa and the other Cry proteins to target only a certain group of insects. 124 LIST OF REFERENCES Alcantara, E. P., 0. Alzate, M. K. Lee, A. Curtiss, and D. H. Dean. 2001. Role of the a-hellx seven of Bacillus thuringiensis CrylAb 6-endotoxin in membrane insertion, structural stability, and ion channel acitivity. Biochemistry 40: 2540-2547. Agaisse, H., and D. Lereclus. 1994. Structural and functional analysis of the promoter region involved in full expression of the crylllA toxin gene of Bacillus thuringiensis. Mol. Microbiol. 13: 97-107. Agaisse, H., and D. Lereclus. 1995. How does Bacillus thuringiensis produce so much insecticidal crystal protein? J. Bacteriol. 21: 6027-6032. Almond, B. D., and D. H. Dean. 1993. Structural stability oi Bacillus thuringiensis 6-endotoxin homolog-scanning mutants determination by susceptibility to proteases. Appl. Environ. Microbiol. 59: 2442-2448. Aronson, A. I., and P. Fitz-James. 1976. Structure and morphogenesis of the bacterial spore coat. Bacteriol. Rev. 40: 360-402. Aronson, A. I., W. Beckman, and P. Dunn. 1986. Bacillus thuringiensis and related insect pathogens. Microbiol. Rev. 50:1-24. 125 Aronson, A I., E.-S. Han, W. McGaughey, and D. Johnson. 1991. The solubility of inclusion proteins from Bacillus thuringiensis is dependent on protoxin composition and is a factor in toxicity to insects. Appl. Environ. Microbiol. 57: 981-986. Aronson, A. 1995. The protoxin composition of Bacillus thuringiensis insecticidal inclusions affects solubility and toxicity. Appl. Environ. Microbiol. 61: 4057-4060. Aronson, A. I., D. Wu, and C. Zhang. 1995. Mutagenesis of specificity and toxicity regions of a Bacillus thuringiensis protoxin gene. J. Bacteriol. 177: 4059-4065. Audtho, M., A. P. Valaitls, O. Alzate, and D. H. Dean. 1999. Production of chymotrypsin-resistant Bacillus thuringiensis Cry2Aa1 6-endotoxin by protein engineering. Appl. Environ. Microbiol. 65: 4601-4605. Baum, J. A, and T. Malvar. 1995. Regulation of insecticidal crystal protein production in Bacillus thuringiensis. Mol. Microbiol. 18:1-12. Bechtel, D. B., and L. A. Bulla. 1976. Electron microscopic study of sporulation and parasporal crystal formation in Bacillus thuringiensis. J. Bacteriol. 127: 1472-1481. Bernhard, K. 1986. Studies on the delta-endotoxin of Bacillus thuringiensis van tenebrionis. FEMS Microbiol. Lett. 33: 261-265. Bletlot, H. P., P. R. Carey, M. Pozsgay, and H. Kaplan. 1989. Isolation of carboxyl-terminal peptides from proteins by diagonal electrophoresis: 126 application to the entomocidal toxin from Bacillus thuringiensis. Anal. Biochem. 181: 212-215. Bosch, D., B. Schipper, H. van der Kleij, R. A. de Maagd, and W. J. Stiekema. 1994. Recombinant Bacillus thuringiensis crystal proteins with new properties: possibilities for resistance management. Bio/Technology 12: 915-918. Bulla, L. A., Jr., K. J. Kramer, and L. I. Davidson. 1977. Characterization of the entomocidal parasporal crystal of Bacillus thuringiensis. J. Bacteriol. 130: 375-383. Burton, S. L , D. J. Ellar, J. Li, and D. J. Derbyshire. 1999. N- acetylgalactosamine on the putative insect receptor aminopeptidase N is recognised by a site on the domain III lectin-like fold of a Bacillus thuringiensis insecticidal toxin. J. Mol. Biol. 287: 1011-1022. Bravo, A., K. Hendrickx, S. Jensens, and M. Peferoen. 1992. Immunocytochemical analysis of specific binding of Bacillus thuringiensis insecticidal crystal proteins to lepidopteran and coleopteran midgut membranes. J. Invert. Pathol. 60: 247-253. Carlson, C. R., D. A. Caugant, and A. B. Kolsto. 1994. Genotypic diversity among Bacillus cereus and Bacillus thuringiensis strains. Appl. Environ. Microbiol. 60: 1719-1725. 127 Carroll, J. D., and D. J. Ellar. 1993. An analysis of Bacillus thuringiensis 5- endotoxin on insect-midgut-membrane permeability using a light scattering assay. Eur. J. Biochem. 214: 771-778. Carroll, J., D. Convents, J. Van Damme, A. Boets, J. Van RIe, and D. J. Ellar. 1997. Intramolecular proteolytic cleavage of Bacillus thuringiensis CrySA 5-endotoxin may facilitate its coleopteran toxicity. J. Invert. Pathol. 70: 41-49. Chen, X. J., M. K. Lee, and D. H. Dean. 1993. Site-directed mutations in a highly conserved region of Bacillus thuringiensis 5-endotoxin affect inhibition of short circuit current across Bombyx mori midguts. Proc. Natl. Acad. Sci. USA 90: 9041-9045. Chestukhina, G. G., L. I. Kostina, A. L. MIkkallova, S. A. Tyurin, F. S. Klepikova, and V. M. Stepanov. 1982. The main feathers ot Bacillus thuringiensis delta-endotoxin molecular structure. Arch. Microbiol. 132: 159-162. Choe, S., M. J. Bernett, G. Fujli, P. M. G. CurmI, K. A. Kantardjleff, R. J. Collier, and D. Eisenberg. 1992. The crystal structure of diphtheria toxin. Nature 357: 216-222. Chôma, C. T., and H. Kaplan. 1990. Folding and unfolding of the protoxin from Bacillus thuringiensis: evidence that the toxic moiety is present in an active conformation. Biochemistry 29:10971-10977. 128 Crickmore, N., D R. Zeigler, J. Feitelson, E. Schnepf, J. Van Rie, D. Lereclus, J. Baum, and D.H. Dean. 1998. Revision of the nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiol. Mol. Biol. Rev. 62: 807-813. Dadd R. H. 1975. Alkalinity within the midgut of mosquito larvae with alkaline- active digestive enzymes. J. Insect Physiol. 21: 1847-1853. Dai, S.M., and S. S. Gill. 1993. In vitro and in vivo proteolysis of the Bacillus thuringiensis subsp. israelensis CrylVD protein by Culex quinquefasciatus larval midgut proteases. Insect Biochem. Mol. Biol. 23: 273-283. Dankocsik, C., W. P. Donovan, and C. S. Jany. 1990. Activation of a cryptic crystal protein gene of Bacillus thuringiensis subspecies kurstaki by gene fusion and determination of the crystal protein insecticidal specificity. Mol. Microbiol. 4: 2087-2094. de Maagd, R. A., P. L. Bakker, L. Maason, M. J. Adang, S. Sangadala, W. Stiekema, and D. Bosch. 1996. Different domains of Bacillus thuringiensis 5-endotoxins can bind to insect midgut membrane proteins on ligand blots. Mol. Microbiol. 31: 463-47. de Maagd, R. A., M. S. G. Kwa, H. van der Kiel, T. Yamamoto, B. Schipper, J. M. Vlak, W. Stiekema, and D. Bosch. Domain III substitution in Bacillus thuringiensis CrylA(b) results in superior toxicity for Spodoptera exigua and altered membrane protein recognition. Appl. Environ. Microbiol. 62: 1537-1543. 129 de Maagd, R.A., M Weemen-Hendriks, W. Stiekema, and D. Bosch. 2000. Bacillus thuringiensis delta-endotoxin CrylC domain III can function as a specificity determinant for Spodoptera exigua in different, but not all, Cryl- Cry1C hybrids. Appl. Environ. Microbiol. 66:1559-1563. Delucca, A. J., II, M. S. Palmgren, and H. de Barjac. 1984. A new serovar of Bacillus thuringiensis from grain dust: Bacillus thuringiensis serovar colmeri (serovar 21). J. Invertebr. Pathol. 43: 437-438. Donovan, W. P., C. C. Dankocsik, M. P. Gilbert, W. C. Gawron-Burke, R. R. Groat, and B. C. Carton. 1988. Amino acid sequence and entomocidal activity of the P2 crystal protein. An insect toxin from Bacillus thuringiensis var. kurstaki. J. Biol. Chem. 263: 561-567. Dow, J. A. T. 1992. pH gradient in lepidopteran midgut. J. Exp. Biol. 172: 355- 375. Du, C., P. A. W. Martin, and K. W. Nickerson. 1994. Comparison of disulfide contents and solubility at alkaline pH of insecticidal and noinsecticidal Bacillus thuringiensis protein crystals. Appl. Environ. Microbiol. 60: 3847- 3853. English, L , H. L. Robbins, M. A. Von Tersch, C. A. Kulesza, D. Ave, D. Coyle, C. S. Jany, and S. L. Slatin. 1994. Mode of action of CryllA: a Bacillus thuringiensis delta-endotoxin. Insect Biochem. Molec. Biol. 24: 1025-1035. 130 Estruch, J. J., G. W. Warren, M. A. Mullins, G. J. Nye, J. A. Craig, and M. G. Koziel. 1997. Vip3A, a novel Bacillus thuringiensis vegetative insecticidal protein with a wide spectrum of activities against lepidopteran insects. Proc. NatI.Acad. Sci. USA 93: 5389-5394. Francis, B. R., and L. A. Bulla, Jr. 1997. Further characterization of BT-R.^, the cadherin-like receptor for CrylAb toxin in tobacco hornworm {Manduca sexta) midguts. Insect Biochem. Molec. Biol. 27: 541-550. Garczynski, S. P., J. W. Grim, and M. J. Adang. 1991. Identification of putative brush border membrane binding protein specific to Baciilus thuring\ens\s 5- endotoxin by protein blot analysis. Appl. Environ. Microbiol. 57 : 2816-2820. Gazit, E., and Y. Shai. 1995. The assembly and organization of the a5 and a7 helices from the pore-forming domain of Baciilus thuringiensis 6-endotoxin J. Biol. Chem. 270 : 2571-2578. Ge, A. Z , N. I. Shivarova, and D. H. Dean. 1989. Location of the Bombyx mori specificity domain on a Bacillus thuringiensis 6-endotoxin protein. Proc. Natl. Acad. Sci. USA 86: 4037-4041. Ge, A. Z , R. M. Pfister, and D. H. Dean. 1990. Hyperexpression of a Bacillus thuringiensis delta-endotoxin-encoding gene in Escherichia coli: properties of the product. Gene 93: 49-54. Ge, A. Z., D. Rivers, R. Milne, and D. H. Dean. 1991. Functional domains of Bacillus thuringiensis insecticidal crystal proteins: refinement of Heliothis 131 virescens and Trichoplusia ni specificity domains on CrylA(c) J. Biol. Chem. 266: 17954-17958. Giordana, B., M. Tasca, M. Villa, C. Chlantore, G. M. Hanozet, and P. Parent!. 1993. Bacillus thuringiensis subsp. aizawai 5-endotoxin inhibits the KVamino acid cotransporters of lepidopteran larval midgut. Comp. Biochem. Physiol. 106C: 403-407. Gill, S. S., E. A. Cowles, and P. V. Pietrantonio. 1992. The mode of action of Bacillus thuringiensis endotoxm. Ann. Rev. Entomol. 37: 615-636. Glatron, M. F, and G. Rapoport. 1972. Biosynthesis of the parasporal inclusion of Bacillus thuringiensis: half-life of its corresponding messenger RNA. Biochimie 54:1291-1301. Gonzalez, J. M., Jr., H. T. Dulmage, and B. C. Carlton. 1981. Correlation between specific plasmids and 5-endotoxin production in Bacillus thuringiensis. Plasmid 5: 351-365. Grochulski, P., L. Masson, S. Borisova, M. Pusztai-Carey, J-L Schwartz, R. Brousseau, and M. Cygler. 1995. Bacillus thuringiensis CrylA(a) insecticidal toxin: crystal structure and channel formation. J. Mol. Biol. 254: 447-464. Haider, M. Z , B. H. Knowles, and D. J. Ellar. 1986. Specificity of Bacillus thuringiensis var. colmeri insecticidal 5-determined by differential proteolytic processing of the protoxin by larval gut proteases. Eur. J. Biochem. 156: 531-540. 132 Harvey, W. R., and M. G. Wolfersberger. 1979. Mechanism of Inhibition of active potassium transport in isolated midgut of Manduca sexta by Bacillus thuringiensis endotoxin. J. Exp. Biol. 83: 293-304. Harvey, W. R„ M. Cioffi, and M. G. Wolfersberger. 1986. Transport physiology of lepidopteran midgut in relation to the action of Bt delta toxin. Foundation for the international Colloquium on Invertebrate Pathology, 11-14. Hodgman, T. C., and D. J. Ellar. 1990. Model for the structure and function of the Baciiius thuringiensis 5-endotoxin determined by compilational analysis. DNA Sequence 1: 97-106. Hofmann, C., H. Vanderberggen, H. Hofte, J. Van Rie, S. Jansens, and H. Van Mellaert. 1988. Specificity of Baciilus thuringiensis 5-endotoxin is correlated with the presence of high-affinity binding sites in the brush border membrane of target insect midguts. Proc. Natl. Acad. Sci. USA 85: 7844-7848. Hofte, H., and H. R. Whiteley. 1989. Insecticidal crystal proteins of Bacillus thuringiensis. Microbiol. Rev. 53: 242-255. Honée, G., D. Convents, J. Van Rie, S. Jansens, M. Peteroen, and B. Visser. 1991. The C-terminal domain of the toxic fragment of a Bacillus thuringiensis crystal protein determines receptor binding. Mol. Microbiol. 5: 2799-2806. lhara, H. T. Uemura, M. Masuhara, S. Ikawa, S. Sugimoto, A. Wadano, and M. Himeno. 1998. Purification and partial amino acid sequences of the 133 binding protein from Bombyx mori for CrylAa 8 -endotoxin of Bacillus thuringiensis. Comp. Biochem. Biophysiol. 120:197-204. Keeton, T. P., B. R. Francis, W. S. A. Maaty, and J. A. Bulla, Jr. 1998. Effects of midgut-protein-preparative and ligand binding procedures on the toxin binding characteristics of BT-R1, a common high-affinity receptor in Manduca sexta for CrylA Bacillus thuringiensis toxins. Appl. Environ. Microbiol. 64: 2158-2165. Jaquet, F., R. Mutter, and P. Lüthy. 1987. Specificity of Bacillus thuringiensis delta-endotoxin. Appl. Environ. Microbiol. 53: 500-504. Jenkins, J. L , M. K. Lee, S. Sangadala, M. J. Adang, and D. H. Dean. 1999. Binding of Bacillus thuringiensis Cry 1 Ac toxin to Manduca sexta aminopeptidase-N receptor is not directly related to toxicity. FEBS Lett. 462: 373-376. Jenkins, J. L., M. K. Lee, A. P. Valaltis, A. Curtiss, and D. H. Dean. 2000. Bivalent sequential binding model of a Bacillus thuringiensis toxin to gypsy moth aminnopeptidase N receptor. J. Biol. Chem. 275: 14423-14431. Kane, J. F, and D. L. Hartley. 1989. Formation of recombinant protein inclusions in Escherichia coii. Tibtech. 6: 95-101. Knight, P. J. K., Knowles, B. H., and D. J. Ellar. 1995. Molecular cloning of an insect aminopeptidase N that serves as a receptor for Bacillus thuringiensis CrylA(c) toxin. J. Biol. Chem. 2 7 0 : 17765-17770. 134 Knowles, B. H., P. J. K. Knight, and D. J. Ellar. 1991. N-acetyl galactosamine is part of the receptor in insect gut epithelia that recognizes an insecticidal protein from Bacillus thuringiensis. Proc. R. Soc. Lond. B 245: 31-35. Knowles, B. H. 1994. Mechanism of action of Bacillus thuringiensis insecticidal 6-endotoxin. Adv. Insect Physiol. 24: 275-308. Koller, C. N., Bauer, L. S., and R. M. Hollingworth. 1992. Characterization of the pH-mediated solubility of Bacillus thuringiensis var. san diego native 6- endotoxin crystals. Biochem. Biophys. Res. Commun. 184: 692-699. Kunkel, T.A 1985. Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82: 488-492. Laemmll, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685. Lambert, B., L. Buysse, 0. Decock, S. Jansens, C. Plens, B. Saey, J. Seurinck, K. Van Audenhove, J. Van Rie, A. Van Vllet, and M. Peferoen. 1996. A Bacillus thuringiensis insecticidal crystal protein with a high activity against members of the family Noctuidae. Appl. Environ. Microbiol. 62: 80-86. Lee, M. K., R. E. Milne, A. Z Ge, and D. H. Dean. 1992. Location of Bombyx mori receptor binding on a Bacillus thuringiensis 5-endotoxin. J. Biol. Chem. 267: 3115-3121. 135 Lee, M. K., B. A. Young, and D. H. Dean. 1995. Domain III exchanges of Bacillus thuringiensis CrylAa toxins affect binding to different gypsy moth midgut receptors. Biochem. Biophysic. Res. Commun. 216: 306-312. Lee, M. K., T. H. You, A. Curtiss, and D. H. Dean. 1996a. Involvement of two amino acid residues in the loop region of Bacillus thuringiensis Cry 1 Ac toxin in toxicity and binding to Lymantria dispar. Biochem. Biophys. Res. Commun. 229:139-146. Lee, M. K., T. H. You, B. A. Young, A. P. Valaltls, and D. H. Dean. 1996b. Aminopeptidase N purified from gypsy moth BBMV is a specific receptor for Bacillus thuringiensis CrylAc toxin. Appl. Environ. Microbiol. 62: 2845- 2849. Lee, H-K., and S. S. Gill. 1997. Molecular cloning and characterization of a novel mosquitocidal protein gene from Bacillus thuringiensis subsp. fukuokaensis. Appl. Environ. Microbiol. 63: 4664-4670. Li, J., J. Carroll, and D. J. Ellar. 1991. Crystal structures of insecticidal 5- endotoxin from Bacillus thuringiensis at 2.5 A resolution. Nature 353: 815- 821. Liang, Y., and D. H. Dean. 1994. Location of a lepidopteran specificity region in insecticidal crystal protein CryllA from Bacillus thuringiensis. Molec. Microbiol. 13: 569-575. Liang, Y., S. S. Patel, and D. H. Dean. 1995. Irreversible binding kinetics of Bacillus thuringiensis CrylA 5-endotoxin to gypsy moth brush border 136 membrane vesicles is directly correlated to toxicity. J. Biol. Chem. 270: 24719-24724. Liebig, B., D. L. Stetson, and D. H. Dean. 1995. Quantification of the effect of Bacillus thuringiensis toxins on short-circuit current in the midgut of Bombyx mori. J. Insect. Physiol. 41: 17-22. Lovgren, A., M-Y. Zhang, A. Engstrom, G. Dalhammar, and R. Landén. 1990. Molecular characterization of immune inhibitor A, a secreted virulence protease from Bacillus thuringiensis. Mol. Microbiol. 4: 2137- 2146. Lorence, A., A. Drazon, and A. Bravo. 1997. Aminopeptidase dependent pore formation of Bacillus thuringiensis CrylAc toxin on Tricoplusia ni membrane. FEBS Lett. 414: 303-307. Lu, H., F. Rajamohan, and D. H. Dean. 1994. Identification of amino acid residues of Bacillus thuringiensis 6-endotoxin CrylAa associated with membrane binding and toxicity to Bombyx mori. J. Bacteriol. 176: 5554- 5559. Martin, P. A. W., and R. S. Travers. 1989. Worldwide abundance and distribustion of Bacillus thuringiensis isolates. Appl. Environ. Microbiol. 55: 2437-2442. Martinez-Ramirez, A. C., and M. D. Real. 1996. Proteolytic processing of Bacillus thuringiensis CrylllA toxin and specific binding to brush-border 137 membrane vesicles of Leptinotarsa decemlineata (Colorado potato beetle). Pesticide Biochem. Physiol. 54: 115-122. Masson, P. J., A. Mazza, J. L. Gringgorten, D. Baines, V. Anelunias, and R. Brousseau. 1994. Specificity domain localization of Bacillus thuringiensis insecticidal toxins is highly dependent on the bioassay system. Mol. Microbiol. 14: 851-860. Masson, L , A. Mazza, R. Brousseau, and B. Tabashnik. 1995a. Kinetics of Bacillus thuringiensis toxin binding with brush border membrane vesicles from susceptible and resistant larvae of Plutella xylostella. J. Biol. Chem. 270 : 11887-11896. Masson, L., Y. Lu, A. Mazza, R. Brousseau, and M. J. Adang. 1995b. The CrylA(c) receptor purified from Manduca sexta displays multiple specificities. J. Biol. Chem. 270 : 20309-20315. McPherson, S. A., F. J. Perlack, R. L. Fuchs, P.G. Marrone, P. B. Lavrik, and D. A. Fischhoff. 1998. Characterization of the coleopteran-specific protein gene of Bacillus thuringiensis var. tenebrionis. Bio/Technology 6: 61-66. Milne R., and H. Kaplan. 1993. Purification and characterization of trypsin-like digestive enzyme from spruce budworm Choristoneura fumiferana responsible for the activation of delta endotoxin from Bacillus thuringiensis. Insect Biochem. Molec. Biol. 23: 663-673. Moar, W. L., L. Masson, R. Brousseau and J. T Trumble. 1990. Toxicity to Spodoptera exigua and Trichoplusia ni of individual PI protoxins and 138 sporulated cultures of Bacillus thuringiensis subsp. kurstaki HD-1 and NRD-12. Appl. Environ. Microbiol. 56: 2480-2483. Moran, C. P. 1993. RNA polymerase and transcription factors, p. 653-667. In A. L. Sonenshein, J, A. Hoch, and R. Losick (ed.), Bacillus subtilis and other gram-positive bacteria. American Society of Microbiology, Washington, D.C. Morse et al., 2001. Structure of Cry2Aa (in press) Mullis, K.B., and F. A. Faloona. 1987. Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Method. Enzymol. 155: 335-350. Nagamatsu, Y., T. Koike, K. Sasaki, A. Yoshimoto, and Y. Furukawa. 1999. The cadherin-like protein is essential to specificity determination and cytotoxic action of the Bacillus thuringiensis Insecticidal CrylAa toxin. FEBS Lett. 460: 385-390. Nicoils, C. N., W. Ahmad, and D. J. Ellar. 1989. Evidence for two different types of insecticidal P2 toxins with dual specificity in Bacillus thuringiensis subspecies. J. Bacteriol. 171 : 5141-5147. Oglwara, K., L. S. Indraslth, S. Asano, and H. Hori. 1992. Processing of 5- endotoxin from Bacillus thuringiensis subsp. kurstaki HD-1 and HD-73 by gut juices of various insect larvae. J. Invert. Pathol. 6 0 :121-126. Oppert, B., K. J. Kramer, R. W. Beeman, D. Johnson, and W. H. McGaughey. 1997. Proteinase-mediated insect resistance to Bacillus thuringiensis toxins. J. Biol. Biochem. 272: 23473-23476. 139 Park, H-W., D. K. Bideshi, and 8. A. Federici. 2000. Molecular genetic manipulation of truncated Cry1C protein synthesis in Bacillus thuringiensis to improve stability and yield. Appl. Environ. Microbiol. 66: 4449-4455. Peyronnet, O., V. Vachon, R. Brousseau, 0. Baines, J-L. Schwartz, and R. Laprade. 1997. Effect of Bacillus thuringiensis toxins on the membrane potential of lepidopteran insect midgut cells. Appl. Environ. Microbiol. 63: 1679-1684. Pietrantonio, P. V., and S. S. Gill. 1992. The parasporal inclusion of Baciiius thuringiensis subsp. shandongiensis: characterization and screening for insecticidal activity. Invertebr. Pathol. 59: 295-302. Poncet, S., E. Dervyn, A. Klier, and G. Rapoport. 1997. SpoGA represses transcription of the cry toxin gene in Baciilus thuringiensis. Microbiology 143:2743-2751. Rajamohan, F., E. Alcantara, M. K. Lee, X. J. Chen, A. Curtiss, and D. H. Dean. 1995. Single amino acid changes in domain II of Baciilus thuringiensis CrylAb 6-endotoxin affect irreversible binding to Manduca sexta midgut membrane vesicles. J. Bacteriol. 177: 2276-2282. Rajamohan, F., 0. Alzate, J. A. Cotrill, A. Curtiss, and D. H. Dean. 1996a. Protein engineering of Bacillus thuhngiensis 6-endotoxin: mutations at domain II of CrylAb enhance receptor affinity and toxicity toward gypsy moth larvae. Proc. Natl. Acad. Sci. 93: 14338-14343. 140 Rajamohan F., S. R. A. Hussain, J. A. CotrIII, F. Gould, and D. H. Dean. 1996b. Mutations at domain II, loop 3, of Bacillus thuringiensis CrylAa and CrylAb delta-endotoxins suggest loop 3 is involved in initial binding to lepidopteran midguts. J. Biol. Chem. 271: 25220-25226. Rajamohan F., J. A. Cotrill, F. Gouid, and D. H. Dean. 1996c. Role of domain II loop 2 residues of Bacillus thuringiensis CrylAb 5-endotoxin on reversible and irreversible binding to Manduca sexta and Heiiothis virescens. J. Biol. Chem. 271: 2390-2396. Raymond, M. 1985. Présentation d'un programme d'analyse log probit pour micro-ordinateur. Cah. ORSTOM Ser. Entomol. Mer. Parasitol. 22:117- 121. Sacchi, V. F., P. Parenti, G. M. Hanozet, B. Giordana, P. Lathy, and M. G. Wolfersberger. 1986. Bacillus thuringiensis ioxm inhibits K^-gradient- dependent amino acid transport across the brush border membrane of Pieris brassicae midgut cells. FEBS Lett. 204: 213-218. Sangadala, S., F. S. Walters, and L. H. English, and M. J. Adang. 1994. A mixture of Manduca sexta aminopeptidase and phosphatase enhances Bacillus thuringiensis insecticidal CrylA(c) toxin binding and Rb -K efflux in vitro. J. Biol. Chem. 269: 10088-10092. Schnepf, H. E., and H. R. Whiteley. 1981. Cloning and expression of the Baciilus thuringiensis crystal protein gene in Escherichia coli. Proc. Natl. Acad. Sci. USA 78: 2893-2897. 141 Schnepf, H. E., and H. R. Whiteley. 1985. Delineation of a toxin-encoding segment of a Bacillus thuringiensis crystal protein gene. J. Biol. Chem. 260: 6273-6280. Schnepf, H. E., K. Tomczak, J. P. Ortega, and H. R. Whiteley. 1990. Specificity-determining regions of a lepidopteran-specific insecticidal protein produced by Bacillus thuringiensis. J. Biol. Chem. 265: 20923- 20930. Schwartz, J. L., Y. Lu, P. Sohnlein, R. Brousseau, R. Laprade, L. Masson, and M. J. Adang. 1997a. Ion chanel formed in planar lipid bilayers by Bacillus thuringiensis toxins in the presence of Manduca sexta midgut receptors. FEBS Lett. 412: 270-276. Schwartz, J. L., M. Juteau, P. Grochulski, M. Cygler, G. Préfontaine, R. Brousseau, and L. Masson. 1997b. Restriction of intramolecular movements within the CrylAa toxin molecule of Bacillus thuringiensis through disulfide bond engineering. FEBS Lett. 410: 397-402. Schwartz, J. L., L. Potvin, X. J. Chen, R. Brousseau, R. Laprade, and D. H. Dean. 1997c. Single site mutations in the conserved alternating arginine region affect ionic channels formed by CrylAa, a Bacillus thuringiensis toxin. Appl. Environ. Microbiol. 63: 3979-3984. Silva-Fiiha, M. H., C. Nielsen-Leroux, and J-F. Charles. 1999. Identification of the receptor for Bacillus sphaericus crystal toxin in the brush border 142 membrane of the mosquito Culex pipiens (Diptera: Culicldae). Insect Biochem. Molec. Biol. 29: 711-721. Shivakumar, A. G., G. J. Gundling, T. A. Benson, D. Casuto, M. F. Miller, and B. B. Spear. 1986. Vegetative expression of the 5-endotoxin genes of Bacillus thuringiensis subsp. kurstaki in Bacillus subtilis. J. Bacteriol. 166: 194-204. Smith, R. A., and G. A. Couche. 1991. The phylloplane as a source of Bacillus thuringiensis yanants. Appl. Environ. Microbiol. 57: 311-315. Stabb. E. V., L. M. Jacobson, and J. Handelsman. 1994. Zwittermycin A- producing strains of Bacillus cereus from divers soils. Appl. Environ. Microbiol. 60: 4404-4412. Slatin, S. L., C. K. Abrams, and L. English. 1990. Delta-endotoxins form cation-selective channels in planner lipid bilayers. Biochem. Biophys. Res. Commun. 169: 765-772. Stephens, J. M. 1952. Disease in codling moth larvae produced by several strains of Bacillus cereus. Can. J. Zool. 30: 30 - 40. Tojo, A, and K. Aizawa. 1983. Dissolution and degradation of Bacillus thuringiensis 5-endotoxin by gut juice protease of the silkworm Bombyx mori. Appl. Environ. Microbiol. 45: 576-580. Triglia, T., M. G. Peterson, and D. J. Kemp. 1988. A procedure for in vitro amplification of DNA segments that lie outside the boundaries of known sequences. Nucleic Acids Res.16: 8186-8191. 143 Uawithya, P., T. Tuntitipawan, G. Katzenmeier, S. Panyim, and C. Angsuthanasombat. 1998. Screening charged positions in helix 4 of the Bacillus thuringiensis Cry4B toxin: Elimination of larvicidal activity. Biochem. Mol. Biol. Int. 44: 825-832. Vadlamudi, R. K., T. H. Ji, and L. A. Bulla, Jr. 1993. A specific binding protein from Manduca sexta for the insecticidal toxin of Bacillus thuringiensis subsp. Berliner. J. Biol. Chem. 268: 12334-12340. Vadlamudi, R. K., E. Weber, I. Ji, T. H. Ji, and L. A. Bulla, Jr. 1995. Cloning and expression of a receptor for an insecticidal toxin of Bacillus thuringiensis. J. Biol. Chem. 270: 5490-5494. Valaitis, A. P., J. L. Jenkins, M. K. Lee, and 0. H. Dean. 2001. Isolation and analysis of gypsy moth BTR-270, a novel brush border membrane molecule that binds Bacillus thuhngiensis C rylA toxins with high affinity. In preparation. Wabiko, H., K. C. Raymond, and L. A. Bulla, Jr. 1986. Bacillus thuringiensis entomocidal protoxin gene sequence and gene product analysis. DNA. 5: 305-314. Widner, W. R., and H. R. Whiteley. 1989. Two highly related Insecticidal crystal proteins of Bacillus thuringiensis subsp. kurstaki possess different host range specificities. J. Bacteriol. 171: 965-974. 144 Widner, W. R., and H. R. Whiteley. 1990. Location of the dlpteran specificity region in a lepidopteran-specific crystal protein from Bacillus thuringiensis. J. Bacteriol. 172: 2826-2832. Wiwat, C., S. Thaithanun, S. Pantuwatana, and A. Bhumiratana. Toxicity of chitinase-producing Bacillus thuringiensis ssp. kurstaki HD-1(G) toward Plutella xylostella. J. Invert. Pathol. 76: 270-277. Wolfersberger, M. G., C. Hofmann, and P. Lathy. 1986. Interaction of Bacillus thuringiensis delta-endotoxin with membrane vesicles isolated from lepidopteran larval midgut, p. 237-238. In Falmagne P. (ed.), Bacterial protein toxins. Guslav Fischer Verlag, Stuttgard, Germany. Wolfersberger, M. G., P. Luthy, A. Maurer, P. Parenti, V. Sacchi, B. Giordana, and G. Hanozet. 1987. Preparation and partial characterization of amino acid transporting brush border membrane vesicles from the larval midgut of the cabbage butterfly {Pieris brassicae). Comp. Biochem. Physiol. 86 : 301-308. Wolfersberger, M. G. 1992. V-ATPase-energized epithelia and biological insect control. J. Ex[. Biol. 172: 377-386. Wolfersberger, M. G., X. J. Chen, and D. H. Dean. 1996. Site-directed mutations in the third domain of Bacillus thuringiensis 8 -endotoxin CrylAa affects its ability to increase the permeability of Bombyx mori midgut brush border membrane vesicles. Appl. Environ. Microbiol. 62: 269-282. 145 Wong, H. C., and S. Chang. 1986. Identification of a positive retroregulator that stabilizes mRNAs in bacteria. Proc. Natl. Acad. Sci. USA 83: 3233 - 3237. Wu, D., and A. I. Aronson. 1992. Localized mutagenesis defines regions of the Bacillus thuringiensis 5-endotoxin involved in toxicity and specificity. J. Biol. Chem. 267: 2311-2317. Wu, S. -J., and D. H. Dean. 1996. Functional significance of loops in the receptor binding domain of Bacillus thuringiensis CrylllA 5-endotoxin. J. Mol. Biol. 255: 628-640. Wu, S.-J. 0 . N. Koller, D. L. Miller, L. S. Bauer, and D. H. Dean. 2000. Enhance toxicity of Baciiius thuringiensis Cry3Aa 5-endotoxin in coleopterans by mutagenesis in a receptor binding loop. FEBS Lett. 473: 227-232. Yamamoto, T., and R. McLaughlin. 1981. Isolation of a protein from the parasporal crystal of Bacillus thuringiensis var. kurstaki toxic to the mosquito larva, Aedes taeniorhynchus. Biochem. Biophys. Res. Commun. 103: 414-421. Yamamoto, T. 1983. Identification of entomocidal toxins of Baciiius thuringiensis by high performance liquid chromatography. J. Gen. Microbiol. 129: 2595-2603. 146 Yunovitz, H., B. Sneh, S. Schuster, U. Oron, M. Broza, and A. Yawetz. 1986. A new sensitive method for determining the toxicity of a highly purified fraction from delta-endotoxin produced by Bacillus thuringiensis var. entomocidus on isolated larval midgut of Spodoptera littoralis (Lepidoptera, Noctuidae). J. Invert. Pathol. 48: 223-231. 147