Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống
1
/ 16 trang
THÔNG TIN TÀI LIỆU
Thông tin cơ bản
Định dạng
Số trang
16
Dung lượng
460,55 KB
Nội dung
REVIEW ARTICLE
Plant a-amylaseinhibitorsandtheirinteractionwithinsect a-amylases
Structure, function and potential for crop protection
Octa
Â
vio L. Franco
1,2,3
, Daniel J. Rigden
1
, Francislete R. Melo
1,3
and Maria F. Grossi-de-Sa
Â
1
1
Centro Nacional de Recursos Gene
Â
ticos e Biotecnologia, Cenargen/Embrapa, Brasõ
Â
lia-DF, Brazil;
2
Universidade Cato
Â
lica de Brasõ
Â
lia,
Brasõ
Â
lia-DF, Brazil;
3
Depto. de Biologia Celular, Brasõ
Â
lia-DF, Brazil
Insect pests a nd patho gens (fungi, bacteria and viruses) are
responsible for severe crop losses. Insects f eed directly on
the plant tissues, while the pathogens lead to damage or
death of the plant. Plants have evolved a certain degree of
resistance through the production of defence compounds,
which may be aproteic, e.g. antibiotics, alkaloids, terpenes,
cyanogenic glucosides or proteic, e.g. chitinases, b-1,3-glu-
canases, lectins, arcelins, vicilins, systemins and enzyme
inhibitors. The enzyme inhibitors impede digestion t hrough
their action o n insect gut digestive a-amylasesand pro-
teinases, which play a key role in the digestion of plant
starch and proteins. The natural defences of crop plants
may be i mproved through the use o f transgenic t echnology.
Current research in the area focuses particularly on weevils
as these are highly dependent on starch for their energy
supply. Six dierent a-amylase inhibitor classes, lectin-like,
knottin-like, cereal-type, Kunitz-like, c-purothionin-like
and thaumatin-like could be used in pest control. These
classes of inhibitors show remarkable structural variety
leading to dierent modes of inhibition and dierent
speci®city pro®les against diverse a-amylases. Speci®city of
inhibition is an important issue as the introduced inhibitor
must not adversely aect t he plant's own a-amylases, nor
the nutritional value of the crop. Of particular interest are
some bifunctional inhibitorswith additional favourable
properties, such as proteinase inhibitory activity or chitin-
ase activity. The area has bene®ted from the recent deter-
mination of many structures of a-amylases, inhibitors and
complexes. These structures highlight the remarkable
variety in structural modes of a-amylase inhibition. The
continuing discovery of new classes of a-amylase inhibitor
ensures that e xciting discoveries remain to be made. In t his
review, w e summarize existing knowledge of insect a-am-
ylases, planta-amylaseinhibitorsandtheir interaction.
Positive results recently obtained for transgenic plants and
future prospects in the area are reviewed.
Keywords: a-amylase inhibitors; knottin-like; lectin-like;
thaumatin-like; Kunitz; cereal-type; bean weevils; bifunc-
tional in hibitors.
Insect pests and pathogens such as fungi, bacteria and
viruses are together, responsible for severe crop losses.
Worldwide, losses in agricultural production due to pest
attack are around 37%, with small-scale farmers hardest hit
[1]. Starchy leguminous seeds are an important staple food
and a source of dietary p rotein in many countries. These
seeds a re rich in protein, carbohydrate and lipid and
therefore suffer extensive predation by bruchids (weevils)
and other pests. The l arvae of the weevil b urrow into the
seedpods and seeds and the insects usually continue to
multiply during seed storage. The damage causes extensive
losses, especially if the seeds are stored for long periods.
In general, plants contain a certain degree of resistance
against insect predation, which is re¯ected in the limited
number o f insects capable of feeding on a given plant. This
resistance is the result of a set of defenc e mechanisms
acquired by plants during evolution [2]. It is only recently
that many secondary chemical compounds have been
de®nitively associate d withplant d efence, fo r e xample
through their synthesis in response t o pest or pathogen
attack. Plant defences are, however, incomplete, as bruchids
and o ther insects are able to infest seeds and different plant
tissues despite the presence of plant defence compounds.
Two f actors seem to have contributed to this phenomenon.
First, many plants suffer reductions in defence compounds
during domestication [3]. Thus, the selection of better-
tasting plants w ith better nutritional value has led, c oncom-
itantly, to crops that are more susceptible to predation.
Secondly, just as plants evolve defences, t heir predators
evolve means to evade those defence mechanisms; t his is the
Correspondence to O. L. Franco, Centro Nacional de Recursos
Gene
Â
ticos e Biotecnologia, Cenargen/Embrapa, S.A.I.N. Parque
Rural, Final W5, Asa Norte, Biotecnologia, Laboratory 05, CEP:
70770±900, Brasõ
Â
lia-DF, Brazil, Fax: + 5 5 6 1 340 3624,
Tel.: + 55 61 448 4705, E-m ail: ocfranco@cenargen.embrapa.br
Abbreviations: AAI, Amaranthus a-amylase inhibitor; a-AI1 and
a-AI2, a-amylaseinhibitors 1 and 2 from the common bean; AMY1
and AMY2, a-amylases from barley seeds; BASI, barley a-amylase
subtilisin inhibitor; BLA, Bacillus licheniformis a-amylase; CAI,
cowpea a-amylase inhibitor; CHFI, corn Hageman factor inhibitor;
HSA, human salivary a-amylase; LCAI, Lachrima jobi chitinase/
a-amylase inhibitor; PAI, pigeonpea a-amylase inhibitor; PPA, por-
cine pancreatic a-amylase; RASI, rice a-amylase/subtilisin inhibitor;
RBI, ragi bifunctional inhibitor; SIa1, SIa2 and SIa3, Sorghum
a-amylase inhibitors 1±3; TASI, triticale a-amylase/subtilisin inhibi-
tor; TMA, Tenebrio molitor a-amylase; WASI, wheat a-amylase
subtilisin inhibitor; ZSA, Zabrotes subfasciatus a-amylase.
(Received 28 A ugust 2001, accepted 6 N ovember 2001)
Eur. J. Biochem. 269, 397±412 (2002) Ó FEBS 2002
phenomenon of host-parasite coevolution, as described by
Ehrlich & Raven [4]. Among thes e means are detoxi®cation
or excretion of the defence compound, or simple adaptation
of the predator so that the toxin no longer has a ny effect.
The relationship between leguminous s eed plants a nd seed
weevils provides an excellent example of host-parasite
coevolution. The seeds are rich in defence compounds so
that the majority of possible predators cannot eat them, yet
seed weevils thrive on the same see ds.
Plant defence compounds include a ntibiotics, alkaloids,
terpenes, cyanogenic glucosides and some proteins. Among
these proteins are chitinase and b-1,3glucanase enzymes,
lectins, arcelins, vicilins, systemins and enzyme inhibitors
[5±8]. The enzyme inhibitors act on key insect gut digestive
hydrolases, the a-amylasesand proteinases. Several k inds of
a-amylase and proteinase inhibitors, present in seeds and
vegetative organs, act to regulate numbers of phytophagous
insects [ 9±11]. a-Amylaseinhibitors are a ttractive candidates
for the control of seed w eevils as these insects are highly
dependent on starch as an energy source. The u se of
a-amylase inhibitors, t hrough plant genetic engineering, for
weevil control w ill be the focus of this review. The properties
of insecta-amylasesand available inhibitors will be
reviewed and issues affec ting their speci®city of interaction
addressed.
INSECT a -AMYLASES
a-A mylases ( a-1,4-glucan-4-glucanohydrolases, E C 3.2.1.1)
constitute a family of endo-amylases that catalyse the
hydrolysis of a-
D
-(1 ® 4)-glucan linkages in starch c ompo-
nents, glycogen and o ther carbohydrates. The enzyme plays
a key role in carbohydrate metabolism of microorganisms,
plants and animals. Moreover, several insects, especially
those similar t o the seed wee vils that feed on starchy seeds
during larval and/or adult stages, depend on their
a-amylases for survival. Research on starch digestion as a
target for control of starch-dependent insects was stimulated
in recent years after results showed that a-amylase inhibitors
from Ph aseolus v ulgaris seeds are detrimental to the
development of cowpea weevil Callosobruchus maculatus
and Azuki bean weevil Callosobruchus chinensis [12,13].
The c arbohydrate digestion of bruchid w eevils, such as
the Mexican bean weevil Zabrotes subfasciatus and the
cowpea w eevil C. maculatus, occurs mainly in the lumen of
the midgut. High enzymatic a ctivities against starch,
maltose, maltodextrins and galactosyl oligosaccharides were
found in the luminal ¯uid, while only aminopeptidase
activity was predominantly associated with gut membrane
[14]. In the yellow mealworm Tenebrio molitor,the
a-amylases are synthesized in anterior midgut cells and
packed in the Golgi area into secretory vesicles that undergo
fusion, as they migrate to the cell apex. At the same time, t he
cell apex undergoes s tructural disorganization with t he
disappearance of cell organelles. Eventually, t he apical
cytoplasm with the large amylase-containing membranous
structure is d ischarged into the midgut lumen. After
extruding the apical c ytoplasm, the cell a pparently remains
functional, as cells are f ound to lack the cell apex, but have
all the other normal ultra structural features [15].
To validate insect a-amylasesastargetsforcropprotec-
tion, it is important to research their variety and understand
how the expression o f different forms i s controlled. Studies
in this area are at an early stage, although s ome important
observations have been made. The presence of different
forms of a-amylases in th e insect midgut lumen h as been
observed in C. maculatus and Z. subfasciatus [14,16]. Pat-
terns of a-amylase expression vary in Z. subfasciatus fed o n
different diets, apparently in response to the presence of
antimetabolic proteins such as a-amylase inhibitors, rather
than as a response to structural differences in the starch
granules. Bean bruchids, such as the Mexican bean weevil
larvae, also have the ability to modulate t he concentration
of a-glucosidases anda-amylases when reared on different
diets [14].
Although the sequences of several i nsect a-amylases are
known [17,18], the only three-dimensional insect a-amylase
yet d etermined is that of the T. molitor enzyme (TMA). This
enzyme is well adapted t o the slightly acidic physiological
environment of the larval midgut with a pH optimum of 5.8
for the cleavage of starch [19]. The structure of TMA
comprises a single polypeptide chain of 471 amino-acid
residues, one calcium ion, one chloride ion and 261 water
molecules (Fig. 1 [20]). The protein folds i nto t hree distinc t
domains, named A, B a nd C (Fig. 1). Domain A, the major
structural unit of TMA is composed of two segments
(residues 1±97 and 160±379; green in Fig. 1) and forms a
(b/a)
8
-barrel; an eight-stranded, parallel b-barrel embraced
by a c oncentric circle of e ight helical segments (seven
a helices and one 3
10
-helix). This domain contains the
Fig. 1. Ribbon thr ee-dimensional structure o f Tenebrio molitor a-amy-
lase (PDB code 1tmq). ThedomainA,BandCarecolouredingreen,
red and o r ange, r espec tively. Th e s tructure c ont ains on e c alcium ion
(yellow) and one chloride ion (cyan). The ®gure was made using
MOLSCRIPT
2 [148] as were all other ®gures ex cept Fig. 2A.
398 O. L. Franco et al. (Eur. J. Biochem. 269) Ó FEBS 2002
catalytic site and the ligand b inding residues [20,21].
Domain B is globular and is inserted into domain A. It is
formed by several extended segments and a short a helix
(residue 98±159; red i n Fig. 1). This domain forms a c avity
against the b barrel of domain A in which the calcium ion is
bound. This cation is of fundamental importance for the
structural integrity o f a-amylases [20,22]. Finally, domain C
is located exactly opposite t o domain B on the o ther side of
domain A. The C domain comprises the C-terminal residues
380±471 (orange in Fig. 1) and forms a separated folding
unit, exclusively made of b sheet. Eight of the 10 strands fold
into a b sandwich structure with ÔGreek keyÕ topology. The
conservation of the interface of A and C domains among
a-amylases from different sources suggests an important
role for e nzyme activity, stability and folding [ 20]. In the
porcine pancreatic a-amylase (PPA), the interface between
C and A domains contains a secondary starch-binding site,
occupied by maltose in one crystal structure [23], but it
remains to be seen if this is also the case for TMA.
TMA, in co mmon with almost all determined a-amylase
structures (the exception being that of calcium-deplete d
Bacillus licheniformis a-amylase (BLA) [ 21], contains a
calcium ion at a c onserved position (yellow in Fig. 1). The
removal of the calcium ion in BLA causes local disorder
around the Ca
2+
-binding s ite, resulting in an inactive
enzyme [24]. The calcium-binding site of TMA is l ocated at
the interface between domains A and B ( Fig. 1), near to
the catalytic centre. The Ca
2+
ion is important for activity
due its contact with His189. This histidine r esidue interacts
with the fourth sugar o f the substrate, bound in the active
site, forming a hinge between the catalytic-site and the
Ca
2+
-binding site [20]. TMA crystal structures also contain
a chloride anion (cyan in Fig. 1). The chloride m ay be
capable of allosterically activating T MA [19] due to its
proximity to a water molecule, which probably initiates
substrate cleavage [25]. The nucleophilicity of this water
molecule might be enhanced by the negative charge of the
chloride anion.
The enzymatic mechanism of a-amylases h as not yet
been completely elucidated. It is likely that d ifferent
a-amylases have a similar mechanism of action with
catalytic residues conserved among all the enzymes [26,27].
Three acidic side chains in PPA (Asp197, Glu233 and
Asp300) (Fig. 2B), corresponding to Asp185, Glu222 and
Asp287 in TMA [20] are direc tly involved in catalysis [28].
The polysacchari de-binding groove of a-amylases can
accommodate at least six sugar units, as observed
crystallographically in PPA [28], and cleavage occurs
between the third and the fourth pyranose residues. The
reaction is believed to proceed by a double displacement
mechanism [27].
a -AMYLASE INHIBITORS
Nonproteinaceous inhibitors
The class of non proteinaceous inhibitors contains diverse
types of organic compounds s uch a s a carbose, isoacarbose,
acarviosine-glucose, hibiscus acid and the cyclo dextrins
(Fig. 2 A). T he two h ibiscus a cid forms, puri®ed from
Roselle tea (Hibiscus sabdaria), the acarviosine-glucose,
the isoacarbose and a-, b-andc-cyclodextrins are highly
active against porcine and human pancreatic a-amylase
(PPA and HPA) [29±32]. The inhibitory activity of these
compounds against a-amylase s i s due in part to their cyclic
structures, which resemble a-amylase substrates and there-
fore bind to a-a mylase catalytic sites. In previous X-ray
crystallography studies [33], t hree a-cyclodextrins molecules
I-III bound to PPA. a-Cyclodextrin I and II bound near to
the catalytic binding cleft, while a-cyclodextrin III biound at
Fig. 2. The structures of nonproteinaceous a-amylasesand acarbose green bound to the catalytic site of PPA. (A) Nonproteinace ous a-amylase
inhibitors. (B ) T he structure of ac arbose g reen bound to t he catalytic site of PPA. Only enzym e r esid ues m akin g h ydrogen bonds (dashed line s) or
hydrophobic contacts with the i nhibitor are shown. The t hree catalytic acidic r esidu es are labeled.
Ó FEBS 2002 Planta-amylaseinhibitors (Eur. J. Biochem. 269) 399
an accessory site. The a-andb-cyclodextrin are not
hydrolyzed to any signi®cant extent b y a-amylases, except
by fungi a mylases [34,35]. In contrast, h uman saliva
a-amylase (HSA) and PPA are capable of hydrolysing
c-cyclodextrin [36]. The cyclodextrin mechanism of PPA
inhibition is pH-, temperature- and substrate-dependent.
When amylose is used as substrate, the inhibition is of the
competitive type, bu t when m altopentose is u sed, the
inhibition becomes noncompetitive [37]. PPA inhibition by
acarbose, in contrast, is noncompetitive, irrespective of
substrate [ 38]. The s tructure of PPA wit h acarbose, a
pseudotetrasaccharide (Fig. 2B), bound to the active site
has been determined [39]. Two linked identical acarbose
fragments occupy the PPA active site (coloured green in
Fig. 2B), hindering substrate hydrolysis. These acarbose
fragments are bonded to residues from the a-amylase active
site by a hydrogen-bonding network (Fig. 2B). No dis-
placement o f a carbose-binding residues is r equired f or
acarbose binding, compared to their positions in the empty
enzyme structure, enhancing the effectiveness of t he inhibi-
tion [39]. The valienamine ring (Fig. 2A) of acarbose is
considered to be crucial in the inhibition mechanism of
a-glucosidases, a-amylasesand other amylolytic enzymes
[40,41]. Its unsaturated structure and half-chair conforma-
tion are r eminiscent of a planar oxocarbonium ion,
proposed as either a transition state or an intermediate
during the hydrolytic pathway of g lucosidases [42]. The
properties of nonproteinaceous inhibitors make them
interesting in the ®eld of medicine, both f or treatment [43]
and in diagnostic procedures [44]. Nevertheless, the use of
nonproteinaceous inhibitors for production of insect resis-
tant transgenic plants is much more dif®cult. The produc-
tion of acarbose o r organic acids i n plants is very complex
and several metabolic pathways are involved. Hence, the
presence of multiple expressed transgenes would be r equired
in order to confer protection. In this area, the proteinaceous
inhibitors, coded by a single gene, are more suitable.
Proteinaceous inhibitors
Proteinaceous a-amylaseinhibitors are found in microor-
ganisms, plants and animals [5,45±47]. In plants, proteina-
ceous inhibitors are m ainly p resent in cereals such as wheat
Triticum aestivum [46,48,49], barley Hordeum vulgareum
[50], s orghum So rghum b icolor [51], rye Secale cereale [47,52]
and rice Oryza sativa [53] but also in leguminosae such as
pigeonpea Cajanus cajan [54], cowpea Vigna ungu iculata [55]
and bean P. vulgaris [56,57]. These inhibitors have showed
monomeric molecular m asses of 5 kDa [51], 9 kDa [55] and
13 kDa [49], homodimeric and heterodimeric masses of
26 kDa [49,57] and tetrameric masses of 50 kDa [58].
Different planta-amylaseinhibitors exhibit different spec-
i®cities against a-amylases from diverse sources (Table 1).
Determination of speci®city o f inhibition is the important
®rst step towards the discovery of an inhibitor that could be
useful for generating insect-resistant transgenic plants. In
some cases, the a-amylaseinhibitors act only against
mammalian a-amylases or, on the contrary, just against
insect a-amylases. In the latter case, this provides a highly
speci®c potential weapon in plant defence. a-AI2, AAI and
some wheat inhibitors are among those naturally possessing
favourable in hibition pro®les (Table 1). However, in gen-
eral, a-amylaseinhibitors inhibit several a-amylases from
different sources. I n these cases, an improved understanding
of the structural b ases for inhibition pro®les (as discussed
later) may e nable the rational design of mutants with more
desirable characteristics. A s proposed by Richardson [59],
a-amylase inhibitors may b e conveniently classi®ed by their
tertiary structure (Table 2) into six classes: lectin-like,
knottin-like, cereal-type, Kunitz-like, c-purothionin-like
and thaumatin-like.
Lectin like a-amylase inhibitors
a-AIs has been puri®ed and characterized from different
accessions and varieties of the common bean P. vulgaris,
including the white, red and black kidney beans [58,60±
62]. The best-characterized isoform, known as a-AI1, was
cloned and identi®ed as an a-amylase inhibitor homolo-
gous to phytohemagglutinin (PHA) [63]. A second variant
of a-AI, called a-AI2, is f ound in some wild accessions of
the common bean [56]. These two allelic variants have
different inhibition speci®cities. a-AI1 inhibits PPA a s well
as the a-amylases of the C. maculatus and C. chinensis,
but it does not inhibit the a-amylases of the Z. subfas-
ciatus (ZSA). In contrast, a-AI2 does no t inhibit the ®rst
three amylases mentioned above but it does inhibit ZSA
[56].
To reach their active mature form, comprising t wo
noncovalently bound glycopeptide subunits, a and b,of7.8
and 14 kDa, respectively [57,64], bean a-AIs are post-
translationally modi®ed. The proteolysis leading to the
activation o f a-AI1 has been studied by mass spectrometr y.
A simple cleavage at the car boxyl side of Asn77, presum-
ably by an Asn-speci®c seed protease of previously
demonstrated importance in legume lectin processing [65],
is made and Asn79 is removed apparently by the action of
a carboxypeptidase. Furthermore, 19 residues at the
C-terminus of the b chain of a-AI1 are clipped. a-AI2
shows s imilar c leavages to a-AI1, but a s omewhat d ifferent
glycosylation pattern [57]. Both variant inhibitors in their
mature form have a h eterotetrameric structure of two
a chains and two b chains [58,66] and are highly glycosy-
lated [57]. a-AIs contain glycans attached to Asn63 and
Asn67 [57] but these m ay not be necessary for inhibitory
activity [67].
A third isoform, a-AIL (also known as a-AI3), isolated
from P. vulgaris cv Rico 23 is a single-chain a-a mylase
inhibitor-like protein completely inactive towards all a-am-
ylases tested [68]. This protein, as with the insecticidal
isoform Grp29 [69], may represent an evolutionary inter-
mediate between phytohaemagglutinnins, arcelins and
a-amylase inhibitors [68,157,158]. Interestingly, another
noncleaved member o f this i nhibitor group speci ®cally
inhibits fungal a-amylases [70] and additionally possesses
hemagglutination a ctivity, showing that these two activities
are not mutually exclusive and that cleavage probably is not
a prerequisite for a-amylase inhibition.
The formation of the inhibitor±enzyme complex for this
class of a-amylaseinhibitors is pH-, time- a nd concentra-
tion-dependent [56,62]. One heterotetramer of a-AI1 binds
to and inhibits two molecules of PPA with K
D
10
)10
M
[58]. To e lucidate the i nhibitory mechanism o f these
inhibitors, the structures of the common bean a-AI1 in
complex with PPA [71] and TMA [72] have been deter-
mined. Structural analysis demonstrated that two h airpin
400 O. L. Franco et al. (Eur. J. Biochem. 269) Ó FEBS 2002
loops of a-AI1 (residues 29±46 and 171±189) were inserted
into the TMA reactive site (Fig. 3A), blocking substrate
binding and establishing a hydrogen bond network with the
residues of the substrate-binding region. The catalytic
residues are str ongly bonded to the inhibitor residues
Tyr186 and Tyr37 that occupies the catalytic pocket. When
compared to results obtained in the PPA±a-AI1 complex,
the s trong contacts in the c atalytic clefts are highly
conserved and only slight modi®cations occur in the
extended protein±protein interaction [71,72]. Bean amylase
inhibitors have been extensively used in transgenic p lants
due to their insecticidal properties.
Kidney bean crude extracts containing lectin-like
a-amylase inhibitors originally found in vivo,wereusedas
starch blockers in the early 1980 s, for the control of human
noninsulin-dependent diabetes mellitus and obesity
[43,44,73,74]. Those early attempts were unsuccessful due
to the undes irable presence of P HAs and proteinase
inhibitors in the extract. Later work with puri®ed a-AI
in diabetic patients met with more success [73]. More
Table 1. Activity of amylase inhibitors from dierent plant sources aga inst mammalian andinsect a-amylases. Low activity represents less than 4 0% of
the maximum activity.
Inhibitors Source
Inhibitory activities
References
Mammalian Insect
a-AI1 P. vulgaris PPA Callosobruchus maculatus
Callosobruchus chinensis
Diabrotica virgifera virgifera
Hypothenemus hampei
Tenebrio molitor
[12,18,31,149]
a-AI2 P. vulgaris None activity Zabrotes subfasciatus [56,150]
Wheat T. aestivum PPA and HSA Diabrotica virgifera virgifera [49,151]
Extract Lygus hesperus
Lygus lineoralis
0.19 T. aestivum PPA and HSA Diabrotica virgifera virgifera
Callosobruchus maculatus
Zabrotes subfasciatus
Acanthoscelides obtectus
Tenebrio molitor
Sitophilus oryzae
Tribolium castaneum
[18,46,49,97]
0.53 T. aestivum HSA and PPA (low) Tenebrio molitor
Callosobruchus maculatus
Zabrotes subfasciatus
Acanthoscelides obtectus
[46,90,97]
0.28 T. aestivum PPA and HSA Tenebrio molitor [97]
WRP25 T. aestivum None Sitophilus oryzae
Tribolium castaneum
Tenebrio molitor
Callosobruchus maculatus
Zabrotes subfasciatus
[46,49]
WRP26 T. aestivum None Tenebrio molitor
Sitophilus oryzae
Tribolium castaneum
Callosobruchus maculatus
[46,49]
WRP27 T. aestivum None Tenebrio molitor (low)
Sitophilus oryzae
[49]
1,2 and 3 S. cereale HSA Tenebrio molitor [52]
BIII S. cereale HSA and PPA Zabrotes subfasciatus
Acanthoscelides obtectus
[47]
AAI A. hypochondriacus None activity Tenebrio molitor
Hypothenemus hampei
Prostephanus truncatus
[77,82,149,152]
CAI V.unguiculata None Callosobruchus maculatus (low) [55]
PAI C. cajan HSA and PPA Helicoverpa armigera (low) [54]
Zeamatin Z mays None activity Tribolium castaneum
Sitophilus zeamais
Rhyzoperta dominica
[112,113]
SIa1, SIa2 S. bicolor HSA (low) Locusta migratoria [51]
and SIa3 Periplaneta americana
Ó FEBS 2002 Planta-amylaseinhibitors (Eur. J. Biochem. 269) 401
recently this c lass of inhibitors has b een us ed for i ts
insecticidal prope rties t o p rotect seeds for insect predation
[13,75,76].
Knottin-type a-amylase inhibitors
The a-amylase inhibitor from Amaranthus hypocondriacus
seeds (AAI) is the smallest proteinaceous inhibitor of
a-amylases yet described, with just 32 residues and three
disul®de b onds [77]. The structure o f its inhibitor, as
determined by NMR [78,79], contains a knottin fold; three
antiparallel b strands and a characteristic disul®de to pology.
It revealed structural similarity to other proteins such a s the
proteinase inhibitor from Cucubirta maxima [80], charybdo-
toxin and conotoxins [81].
AAI s peci®cally inhibits insecta-amylasesand is inactive
against mammalian a-amylases ([77]; Table 1). The struc-
ture of its inhibitor in complex revealed that inhibition, as
with the lectin-like inhibitors, is through blockage of the
catalytic site ([82], Fig. 3B). The inhibitor binds in the active
site crevice interacting with catalytic residues f rom the A
and B domains of TMA (Fig. 1 [ 82]). The residue Asp287,
one of the catalytic residues of its enzyme, forms a salt
bridge directly with Arg7 of AAI. The other two enzymatic
catalytic residues, as well other c onserved residues involved
in substrate recognition and orientation, are connected to
AAI via an intricate water-mediated hydrogen-bonding
network [82]. The TMA±AAI complex is characterized by a
high complementarity of the interaction surfaces (Fig. 3B).
Structural comparisons of the inhibitor structure in solution
[79] to the X-ray structure of A AI bounded to TMA [82]
demonstrate that both backbone and side conformations
are only slightly adjusted on formation of the complex [79].
The speci®c activity of AAI against insect a-a mylases makes
it an attractive candidate for the development of insect-
resistant transgenic plants.
Cereal-type a-amylase inhibitors
a-Amylase inhibitors of the cereal family are composed of
120±160 amino-acid residues forming ®ve disul®de bonds
(Table 2) [46,83,84]. These inhibitors are also known as
sensitizing agents in humans upon repeated exposure,
causing allergy, dermatitis and baker's asthma assoc iated
with cereal ¯our [85,86]. N-glycosylation is involved in the
reactivity of the most reactive allergen, an inhibitor from rye
[87].
The e xogenous wheat a-amylase inhibitor coded 0.19
[46,49] and the bifunctional inhibitor from Indian ®nger
millet R BI [88,89], are the most studied inhibitors from this
family. T he a-amylase inhibitor 0.19, named according to its
gel electrophoretic mobility relative to b romophenol blue,
inhibits a-amylase s f rom birds, Bacilli, insects and mammals
(Table 1). Its inhibition of human salivary a-amylase is
characterized by K
i
0.29 n
M
[160]. It has 124 amino-acid
residues and is homologous to RBI [ 90]. Mass spectrometry
results [46] clearly demonstrated the presence of homodi-
mers of 0.19 with smaller quantities of other multimers, in
accord with sedimentation [49] and X-ray crystallography
results [91]. Nevertheless, some cereal inhibitors act as
monomers, such the wheat inhibitors 0.28, W RP25, WRP26
and W RP27 [46]. T he dimeric a-amylase inhibitor 0.19 was
crystallized [92] and its structure d etermined. It con tains ®ve
Table 2. Dierent structural c lasses of a-amylase inhibitors, based on a c lassi®cation by Richardson [59].
Structural class Source and references
Residue
numbers
Disul®de
bonds CATH
b
code and Family SCOP fold Names
Legume lectin type Common beans [31,56] 240±250 5 2.60.120.60
c
Lectin
Concanavalin A-like
lectins/glucanases
a-A11 & a-A12
Knottin type Amaranth [78,79] 32 3 ND
a
Knottins AA1
Cereal type Wheat [46], barley [59] &
Indian ®nger millet [153]
124±160 5 1.10.120.10
Cereal inhibitor
Bifunctional inhibitor/
lipid-transfer protein/
seed storage 2S albumin
0.19, 0.53, 0.28
WRP25, WRP26,
WRP27 & RBI
Kunitz type Barley [102], wheat [99] & rice [100] 176±181 1±2 2.80.10.50.6
Proteinase inhibitor
b-Trefoil BASI, WASI & RASI
Thaumatin type Maize [112,113,156] 173±235 5±8 2.60.110.10
Sweet tasting protein
Osmotin
Thaumatin-like proteins
Zeamatin
c-Purothionin type Sorghum [51] 47±48 5 3.30.30.10
Antibacterial protein
Knottins SIa1, SIa2&SIa3
a
This inhibitor class was not classi®ed by CATH program.
b
Orengo et al. [154].
c
1, 2, and 3 from CATH code represents mainly a helix, mainly b sheets and mixed a helix and b sheets, respectively.
402 O. L. Franco et al. (Eur. J. Biochem. 269) Ó FEBS 2002
a-helices arranged in an up-and-down manner, satisfying
favorable packing modes, with al l 10 cysteine residues
forming disul®de bonds [91].
The bifunctional a-amylase/trypsin inhibitor (RBI) is
another prototype of the cereal i nhibitor f amily. T his
inhibitor is a stable monomer of 122 amino acids with ®ve
disul®de bonds, which is resistant to urea, guanidine
hydrochloride and thermal denaturation [93]. This bifunc-
tional inhibitor p resents a three-dimensional structure very
similar t o that of 0.19 inhibitor [91], with a globular fold
with four a helices in a simple Ôup-and-downÕ topology and
a small antiparallel b sheet ([94]; Fig. 3C). Like other
inhibitors of this class, it can competitively inhibit a variety
of a-amylases, including PPA and TMA. This latter
enzyme is inhibited with a K
i
15 2 n
M
[88,89,95].
As the structure of RBI±TMA complex reveals, the
inhibitor binds to the enzyme active site, once again
impeding substrate binding (Fig. 3 C). Two RBI segments
are responsible for the interactionwith TMA. Segment 1,
comprising the N-terminal residues Ser1±Ala11 and the
residues Pro52±Cys55, protrudes like an a rrow head i nto
TMA's substrate-binding groove and directly targets the
active site of the enzyme. The N-terminus forms the arrow
tip, adopting a 3
10
-helical conformation in complex. The
Ser1 residue makes several hydrogen bonds with the
catalytic Asp185 and Glu222 from enzyme while Val2
and S er5 from i nhibitor i nteract with the third conserved
acidic residue of the catalytic site, Asp287. The second
binding segment comprises several r esidues, which form a
collar around the upper part of the arrow head an d stabilize
Fig. 3. The various known modes o f a-amylase inhibition. A s tandard c olou ring sc he me is used with enzyme d rawn i n m age nta and inhibitor helix,
strand and coil drawn in red, cyan and yellow, respectively. The calcium ion common to all structures is drawn i n orange. The structures are (A)
TMA bound to a-AI1 (PDB code 1vi w) (B) TMA-AAI (1clv) (C) TMA-RBI ( 1tmq) and (D) AMY2-BASI ( 1ava). In (D) the calcium ion bound at
the enzyme±inhibitor i nterface (see text for details) is d rawn in green.
Ó FEBS 2002 Planta-amylaseinhibitors (Eur. J. Biochem. 269) 403
the complex by further interactions with this enzyme [88].
Tests with a peptide containing the N-terminal 10 residues
of RBI showed that only segment 1 is necessary for
a-amylase inhibition. Synthetic peptides containing muta-
ted N -terminal R BI seq uences demonstrated different
inhibitory potentials [89]. Alam et al.[89]alsoshowedthat
this inhibitor binds to soluble amylase substrate, reducing
the apparent af®nity of the enzyme f or the s ubstrate.
Inhibitor±substrate interactions could explain the differ-
ences in the type o f inhibition observed for d ifferent
substrates with the same enzyme [37,89]. E xploiting the
overall structural s imilarity between RBI and 0.19, molec-
ular models of 0.19±TMA and 0.19±HSA complexes have
been constructed [46].
Multiple genes encode the members of the cereal-type
inhibitor f amily [96]. Their different sequences yield a
remarkable array of inhibition speci®city p ro®les (Table 1
[46,52]). In wheat, some a-amylases i nhibitors genes may
be silent or expressed at a much lower level [96]. It can
be envisaged that the lack of the pertin ent predator in
the e cological niche could silent th e respective inhibitor
gene [97].
Kunitz-like a-amylase inhibitors
The Kunitz-like a-amylaseinhibitors contain around 180
residues and four cystines (Table 2). They are present in
cereals such as barley [98], wheat [99] and rice [100]. The
best-characterized a-amylase inhibitor from the Kunitz class
is the barley a-amylase/subtilisin inhibitor (BASI), a
bifunctional double-headed inhibitor with a fast tight
inhibitory reaction with cereal a-amylase AMY2
(K
i
0.2 2 n
M
) and serine prote inases of the subtilisin
family [50 ,101]. The structure of BASI [102] r evealed two
disul®de bonds and a b trefoil topology (Fig. 3D) shared
with the homologous wheat a-amylase subtilisin inhibitor
(WASI [103]), the Erythrina cara trypsin inhibitor [104]
andthericinBchain[105].
In the cases of a-AI1, AAI and RBI, inhibition involves
the i nsertion of inhibitor l oops into the a-amylase active site,
thereby establishing a network of hydrogen bonds with
catalytic and substrate-binding residues (Fig. 3A±C). The
mechanism of i nhibition of barley a-amylase 2 ( AMY2) by
BASI [102] is different, i n t hat the inhibitor does not interact
directly with any catalytic acidic residues of the enzyme.
Nevertheless, this inhibitor interacts strongly with both the
A and B domains near the catalytic site, through the
formation of 12 hydrogen bonds, two salt bridges and
multiple van der Waal's contacts, and thereby prevents
substrate access (Fig. 3D). A c avity at the enzyme±inhibitor
interface contains a trapped calcium ion whose presence is
suggested to electrostatically enhance the network of water
molecules a t t he complex interface and thereby raises the
stability of the complex.
BASI is involved in regulating the degradation of seed
carbohydrate, preventing the e ndogenous a-amyla se 2 from
hydrolysing starch during premature sprouting [41]. Ad di-
tionally, it p rotects t he seeds against exogenous proteinases
and a-amylases produced by various pathogens a nd pests
[106]. BASI i nhibits t he endogenou s enzyme w ith a stoichi-
ometry of 1 : 1 [107], but, interestingly, is unable to inhibit
barley a-amylase 1 (AMY1), which bears 74% sequence
identity to AMY2 [101].
Thaumatin-like a-amylaseinhibitors type
The t haumatin-like inhibitors are proteins with molecular
masses of 22 kDa with signi®cant sequence similarity to
pathogenesis-related group 5 (PR-5) proteins and to thaum-
atin, an intensively sweet protein from Thaumatococcus
danielli fruit [108,109].
The best-ch aracterized inhibitor from t his class is zeam-
atin, a bifunctional inhibitor from Zea mays that is
homologous to the sweet protein t haumatin. Zeamatin has
a total of 13 bstrands, 11 of which form a bsandwich at the
core of protein ([110]; Fig. 4A). Several loops extend from
this inhibitor core and are secured by one or more of the
Fig. 4. The structural classes of a-amylaseinhibitors whose modes of
inhibition are not yet known (A) zeamatin (PDB code 1 dl5) and (B)
SIa1 (1gpt). The coordinates of SIa1 have n o t be en deposited in t he
PDB so that (B) shows the structure of the homolo gous c-thionin [124].
404 O. L. Franco et al. (Eur. J. Biochem. 269) Ó FEBS 2002
eight disul®de bon ds. E lectrostatic modelling o f z eamatin
reveals an electrostatically polarized surface, heavily popu-
lated w ith Arg and L ys residues [110]. This maize inh ibitor is
not sweet, despite its similarity to thaumatin, probably d ue
to changes in a putative receptor-binding site [111]. Zeam-
atin was a ble to inhibit porcine pancreatic trypsin and
digestive a-amylases of the insects T. castaneum, Sitophilus
zeamays and Rizopherta d ominica [112,113]. Other proteins
from this class, such as the thaumatin-like p roteins R and S
from barley seeds, did not show any inhibitory activity
against trypsin or a-amylases [114] despite their highly
similar N-terminal sequences. Zeamatin is mainly known
for its antifungal activity, but this is not related to inhibition
of hydrolyic enzymes as this protein does not inhibit f ungal
a-amylases [112] and fungi do not contain trypsin. Zeamatin
binds to b-1,3-glucans [115] and permeabilizes fungal-cells
leading to cell d eath [116] but the a ntifungal mode of action
of this protein is still a matter o f d ebate. For t hese
properties, zeamatin could be used as a medical agent,
acting on vaginal murine candidosis cells [117] or in
transgenic plants, increasing their resistance against pests
and pathogens.
c-Purothionins-like a-amylase inhibitors
The a-amylaseinhibitors of this family have 47 or 48
residues, are sulfur-rich and form part of the c-thionin
superfamily (Table 2). Members of this superfamily are
involved in plant defence through a remarkable variety of
mechanisms: m odi®cation o f m embrane p ermeability
[118,119], i nhibition of protein synthesis [120] and protein-
ase inhibition [121] Inhibition of insecta-amylases has bee n
observed for three isoforms from Sorghum bicolor called
SIa-1, S Ia-2 an d S Ia-3 [51]. These molecules strongly
inhibited the digestive a-amylases of guts of locust and
cockroach, poorly inhibited a-amylases from A. oryzae and
human saliva and failed to inhibit the a-amylases from
porcine pancreas, barley and Bacillus sp. [51]. T he structure
of SIa-1 has been solved by NMR [122] revealing a a + b
sandwich structure [123] with a nine-residue helix packed
tightly a gainst the sheet (Fig. 4B). T he hel ix is h eld i n place
by two disul®de bridges, which link sequential turns of the
helix to residues 41 and 43 in the middle of strand b3, the
so-called cysteine-stabilized helix (CSH) motif [122]. As
expected from sequ ence comparison, the structure is similar
to those of w heat c1-purothionin [ 124] and scorpion toxins
[122].
BIFUNCTIONAL INHIBITORS
As the a bove d iscussion highlights, bifunctional a-amylase/
proteinase inhibitors are relatively c ommon (Table 3). As
inhibition of predatory insect digestive proteinases is
another attractive route for plant protection, the co mbina-
tion of a-amylaseand proteinase inhibition is potentially
very useful. It is therefore important to know whether
simultaneous inhibition of proteinase anda-amylase is
possible for these inhibitors.
In the c ase of RBI, w ith two ind ependent inhibition sites,
formation of a stable a-amylase±RBI±trypsin complex h as
been observed [95]. The N-terminal site is responsible for
a-amylase inhibition, as previously discussed, while on the
opposite side a canonical substrate-like trypsin inhibitor
region is present [89,94,125,126]. In this inhibitor, the
exposed trypsin-binding loop is located between two
a-helices, contains the residues Gly32±Tyr37 and also
contains Arg34 that confers trypsin-speci®city. Modelling
of the TMA±RBI±trypsin complex (Fig. 5; [88]), con®rms
Table 3. Activity o f bifunctional inhibitors from dierent plant sources against m ammalian andinsect a-amylases. Low a ctivity represents a pprox-
imately 40% of a total activity.
Inhibitor Source
a-Amylase inhibition
References Other activity
Insect Fungal Plant Mammalian
BASI Barley + ND + ± [50,102] Subtilisin Inhibition
RASI Rice + ND + ND [53,100] Subtilisin Inhibition
WASI Wheat ND ND + ND [99,103] Subtilisin Inhibition
CHIF Maize + ND ND ND [128] Trypsin Inhibition
Zeamatin Maize + ± ND Low [112,113] Trypsin Inhibition
RBI Finger millet + ND ND + [88] Trypsin Inhibition
LCAI Jobi tear's seeds + ND ND ± [132] Chitinase
Fig. 5. A model of a ternary complex fo rmed by TMA (mage nt a), RB I
(blue) and pan creatic bovine trypsin (green ). Constructed as described in
the text.
Ó FEBS 2002 Planta-amylaseinhibitors (Eur. J. Biochem. 269) 405
that no steric clashes prevent simultaneous inhibition of
both enzymes. The homologous bifunctional corn H ag-
eman factor inhibitor (CHIF) inhibits mammalian trypsin,
Factor XIIa (Hageman factor) of the co ntact pathway of
coagulation as well a-amylases from several insects [127]. Its
structure has been solved revealing a similar proteinase
inhibitory site to that in RBI [128]. As w ith RBI, a-amylase
inhibition requires the N-terminal region [129].
Kunitz-like inhibitors are also bifunctional, this time
possessing inhibitory activity against the subtilisin c lass pf
proteinases. Among them, the BASI and WASI are the best
characterized, and both structures have been determined by
X-ray crystallography [102,103]. It has been suggested that
WASI has t wo different sites because t he activity against
a-amylases is retained after incubation with proteinases K
[130]. A complex of WASI±proteinase K showed that a
loop containing the Gly66 and Ala67 is crucial to proteinase
inhibition [131]. This class o f bifunctional inhibitors inhibits
insect and endogenous planta-amylases but not mamma-
lian a-amylases [53,100,101]. They are also inactive against
different classes of proteinases such as trypsins and chym-
otrypsins [53]. BASI is deposited during grain ®lling and
therefore found in the seed prior to AMY2, which is
synthesized de novo during germination. This inhibitor has
been proposed to control t he activity of AMY2 in the case
of premature sprouting or t o act in plant defence [159]. The
inhibitor R ASI could also help to regulate seed development
by inhibiting a development-speci®c a-amylase [53]. This
a-amylase inhibition speci®city is in agreement with a dual
role in starch control of the storage tissues at plant
developmental stages and as defensive agents in response
to pest attack.
Zeamatin represents a third bifunctional a-amylase/
proteinase inhibitor [112,113] with a known crystal structure
[110]. H owever, t he observation of trypsin inhibition, albeit
weak, was only made r ecently [ 113] and it remains to be seen
whether or not zeamatin possesses independent sites for
proteinase anda-amylase inhibition.
In addition to bifunctional a-amylase/proteinase inhibi-
tion, a single report has found chitinase activity present in
an insecta-amylase inhibitor isolated from Lachrima jobi
seeds [132]. Chitinase activity is another recognized plan t
defence [7] so that this double activity i s of great potential
biotechnological interest. However, further characterization
of the inhibitor is clearly required.
ISSUES OF a -AMYLASE INHIBITOR
SPECIFICITY
In order to be of practical use for the production of
transgenic plants, a-amylaseinhibitors should have appro-
priate speci®city pro®les. On the one hand, they should
ideally be eff ective against t he full range of potential
predatory insects. However, they must not interfere with the
action of endogenous a-amylases, which are of demonstrated
importance in, for example, germination [41]. They should
also lack activity against the mammalian enzymes, although
this is in general a lesser issue as cooking would denature
any inhibitors before ingestion. These simple considerations,
in combination with biochemical data in the literature
(Tables 1 and 3), already highlight some inhibitors as more
promising candidates than others. For example, the cereal
bifunctional a-amylase/subtilisin inhibitors have strong
af®nity for plant enzymes [107] deriving from their role in
regulation of starch metabolism, and are therefore less
favoured. The known a-amylaseinhibitors selective for
insect enzymes and inactive against mammalian enzymes
include WRP25, WRP26 a nd WRP27 from the cereal-type
class [46], the Amaranthus a-amylase inhibitor [77] and
zeamatin [113]. As well as differences in af®nity for broad
groups of a-amylases, some remarkable examples of ®ne
speci®city e xist. A mong several interesting speci®city d iffer-
ences in the cereal-type inhibitors is the inability of WRP26
to inhibit Z SA, while WRP25, 98% i dentical in sequen ce to
WRP26, is an effective inhibitor [46].
Given the remarkable structural and functional variety
naturally found among a-amylaseinhibitors (Tables 1±3),
screening for inhibitorswith desirable characteristics is a
viable option. An attractive alternative, however, would be
the r ational redesign of known i nhibitors in order t o confer
upon them the required speci®city pro®le. Although
conceivably more rapid than the screening approach,
inhibitor redesign clearly requires a full understanding of
amylase±inhibitor interaction structural bases. With the
availability of an ever-increasing number of crystal struc-
tures a number of r ecent studies h ave addressed e xperimen-
tally observed issues of amylase-inhibitor speci®city,
generally through sequence analysis and modelling, some-
times supported by mutagenesis studies [46,62,82,133,134].
Two independent studies [62,133] address the speci®city
of a-AI1 for PPA, not inhibiting ZSA and the opposite
speci®city of a-AI2 for Z SA over PPA [56]. The reliance on
simple counting of hydrogen b onds weakens the conclusions
of Le Berre-Anton et al . [135] regarding the a-AI1/a-AI2
comparison but they show that the bulkier, single chain
a-AIL is sterically impeded from binding either ZSA or
PPA. In another study, analysis of other factors, including
electrostatics and hydrophobic interactions, fails to lead to a
simple explanation of a-AI1/a-AI2 speci®city and the
authors concluded that s peci®city was conferred by multiple
factors [133].
In studies aimed at explaining the ability of BASI to
inhibit AMY2 but not AMY1, previous indications of
the importance of electrostatic interactions [136] have
been e xamined through site-directed mu tagenesis [ 134].
Mutations of AMY2 residues known to make electro-
static interactions at the interface with this inhibitor,
Arg128 and Asp142, were made, weakening the e nzyme±
inhibitor interactionand reducing t he effect of charge
screening o n the interaction. Remarkably, the introduc-
tion of just two AMY2 residues, Arg128 and Pro129,
into the A MY1 e nvironment w as enough to enhance
BASI sensitivity at least 100-fold. As well as the
electrostatic characteristics of the Arg, the conformational
properties of t he pro line, which forms a cis peptide in
AMY2, are implicated in the AMY1/AMY2 spe ci®city.
The lysine, which replaces Pro129 in AMY1 is unlikely to
form a cis peptide bond, with consequent changes in the
conformation of neighbouring Arg128 and other residues
at the interface.
The varied inhibition speci®city pro®les of a family of
cereal a-amylaseinhibitors have been addressed by sequence
analysis and model building [46]. In the absence of crystal
structures for any of the analysed inhibitors in complex with
a-amylase, the complex of T MA with the more distantly
related RBI [88] was used as the basis for model construc-
406 O. L. Franco et al. (Eur. J. Biochem. 269) Ó FEBS 2002
[...]... cleavage of inhibitors by insect digestive proteinases [138] may complicate pest control by a-amylaseinhibitorsand could explain the low in vivo ef®ciency of some inhibitors against insect pests [16] The a-amylase diversity found in a single insect [14] indicates that unless an a-amylase inhibitor has reasonably broad speci®city, being capable of inhibiting all the a-amylases produced by the insect, ... TRANSGENIC PLANTS EXPRESSING a-AMYLASEINHIBITORS The transgenic plant approach provides an attractive alternative to the use of chemical pesticides and insecticides and could contribute to the production of crop varieties that are inherently tolerant/resistant to their major target insect pests Besides the bene®t on agricultural crop production, the use of genes that encode insecticidal proteins in Plant a-amylase. .. cloning of bruchid (Zabrotes subfasciatus) a-amylase cDNA and interactions of the expressed enzyme with bean amylase inhibitorsInsect Biochem Mol Biol 27, 271±281 18 Titarenko, E & Chrispeels, M.J (2000) cDNA cloning, biochemical characterization and inhibition by plantinhibitors of the a-amylases of the Western corn rootworm, Diabrotica virgifera virgifera Insect Biochem Mol Biol 30, 979±990 19 Buonocore,... Speci®c inhibition of insect a-amylases: yellow meal worm a-amylase in complex with the Ê Amaranth a-amylase inhibitor at 2.0 A resolution Structure 7, 1079±1088 Buonocore, V., Petrucci, T & Silano, V (1977) Wheat protein inhibitors of a-amylase Phytochemistry 16, 811±820 Lyons, A., Richardson, M., Tatham, A.S & Shewry, P.R (1987) Characterization of homologous inhibitors of trypsin anda-amylase Biochim... speci®city and stability Plant Mol Biol 25, 141±157 MacGregor, E.A., Janececk, S & Svensson, B (2001) Relationship of sequence and structure to speci®city in the a-amylase family of enzymes Biochem Biophys Acta 1546, 1±20 Â Machius, M., Vertesy, L., Huber, R & Wiegand, G (1996) Carbohydrate and protein-based inhibitors of porcine pancreatic a-amylase: structure analysis and comparison of their binding... structural variety of the a-amylaseinhibitors so far characterized is striking, encompassing proteins with mainly alpha, beta and mixed folds The continuing discovery of new a-amylaseinhibitors suggests that the list of a-amylaseinhibitors is far from complete As has been emphasized, consideration of a-amylase inhibitor speci®city is of prime importance The inhibition spectra of inhibitors so far characterized... evolution in plant chitinases: molecular targets of selection in plant- pathogen co-evolution Proc Natl Acad Sci USA 97, 5322±5327 Â 8 Sales, M.P., Gerhardt, I.R., Grossi-de-Sa, M.F & Xavier-Filho, J (2000) Do legumes storage proteins play a role in defending seeds against bruchids? Plant Physiol 124, 515±522 9 Konarev, A.V (1996) Interaction of insect digestive enzymes withplant protein inhibitorsand host-parasite... inhibitor and design of a routine procedure Clin Chem 23, 560±566 45 Silano, V (1987) a-Amylaseinhibitors In Enzymes andTheir Role in Cereal Technology (Kruger, J & Lineback, D., eds), pp 141±199 American Association of Cereal Chemists, St Paul, MD, USA 46 Franco, O.L., Rigden, D.J., Melo, F.R., Bloch Jr, C., Silva, C.P Â & Grossi de Sa, M.F (2000) Activity of wheat a-amylaseinhibitors towards bruchid a-amylases. .. Further characterization studies of the alpha-amylase protein inhibitor of gel eletrophoretic mobility 0 19 from the wheat kernel Biochim Biophys Acta 420, 288±297 49 Feng, G.H., Richardson, M., Chen, M.S., Kramer, K.J., Morgan, T.D & Reeck, G.R (1996) a-Amylaseinhibitors from wheat: a sequences and patterns of inhibition of insectand human a-amylasesInsect Biochem Mol Biol 26, 419±426 50 Abe, J.I.,... Salcedo, G (1994) Rye inhibitors of animal a-amylases shown dierent speci®cities, aggregative properties and IgE-binding capacities than their homologues from wheat and barley Eur J Biochem 224, 525±531 53 Yamagata, H., Kunimatsu, K., Kamasaka, H., Kuramoto, T & Iwasaki, T (1998) Rice bifunctional a-amylase/ subtilisin inhibitor: characterization, localization, and changes in developing and germinating seeds . REVIEW ARTICLE
Plant a-amylase inhibitors and their interaction with insect a-amylases
Structure, function and potential for crop protection
Octa
Â
vio. Amaranthus a-amylase inhibitor; a-AI1 and
a-AI2, a-amylase inhibitors 1 and 2 from the common bean; AMY1
and AMY2, a-amylases from barley seeds; BASI, barley a-amylase
subtilisin