Development of insect-resistant transgenic cauliflower plants expressing the trypsin inhibitor gene isolated from local sweet

Abstract Agrobacterium-mediated transformation was

used to introduce a trypsin inhibitor gene into Taiwan

cau-liflower (Brassica oleracea var. botrytis L.) cultivars. The

TI gene was isolated from a well-adapted Taiwan sweet

potato cultivar and was expected to be especially effective

in combating local pests. In vitro regeneration studies

in-dicated that 4-day-old cauliflower seedling hypocotyl

seg-ments, pretreated with 2,4-dichlorophenoxyacetic acid for

3 days and incubated on a silver-ion-containing shoot

in-duction medium, gave regeneration rates greater than 95%.

Optimum transformation conditions were determined.

G418 selection at 15 mg/l was initiated 1 week after

co-cultivation, and the dose was doubled 1 week later. Over

100 putative transgenic plants were produced. Transgenic

status was confirmed by in vitro TI activity, and Southern

and Western hybridization assays. The transgenic plants

demonstrated in planta resistance to local insects to which

the control plants were vulnerable.

Key words Cauliflower · Brassica oleracea var.

botrytis L. · Trypsin inhibitor · Insect resistance ·

Agrobacterium mediated transformation

Abbreviations BA N

-Benzyladenine · CI medium

Cal-lus-inducing medium · 2,4-D 2,4-Dichlorophenoxyacetic

acid · IAA Indole-3-acetic acid · NPTII Neomycin

phos-photransferase II · SI medium Short induction medium ·

TI Trypsin inhibitor

Introduction

Annual world production of cauliflower was over 12.7

mil-lion metric tons in 1996, with China leading with more than

30% of world production. One of the major problems in

cauliflower cultivation is insect pests. Massive quantities

of synthetic insecticides are used, giving rise to major

con-cerns about food safety and environmental pollution in

ad-dition to the high chemical and labor costs.

The development of genetic engineering protocols for

cauliflower with marker genes have been reported by

Da-vid and Tempe (1988), Srivastava et al. (1988), De Block

et al. (1989), and Eimert and Siegemund (1992) using

Ag-robacterium-mediated infection. The first case of

transfer-ring agronomically useful traits into cauliflower involved

insertion of the capsid gene and the antisense gene VI of

cauliflower mosaic virus via Agrobacterium (Passelegue

and Kerlan 1996). In this work, while the transcription of

the transgenes was detected in all transgenic plants,

trans-lation of the capsid protein was not detected.

Our work involves transferring the gene that encodes a

trypsin inhibitor (TI), a subgroup of protease inhibitor, into

Taiwan cauliflower cultivars to combat insect pests.

Pro-tease inhibitors are part of the natural plant defense system

against insect predation. They are present in the edible parts

of numerous crop species, e.g., TI makes up over 80% of

the soluble proteins in the storage roots of Taiwan sweet

potato cultivars (Lin 1996) and actinidin makes up to 60%

of the soluble proteins of the kiwi fruit tissue (Praekelt et

al. 1988). Moreover, they are capable of controlling a wide

spectrum of insect pests (Hilder et al. 1990). Since the gene

we used was isolated from the storage root of a

well-adapted local sweet potato cultivar (Yeh et al. 1997), we

expected that it would be especially effective in

combat-ing local insect pests. The sweet potato, of New World

or-igin, is well-established in China (over 86% of world

pro-duction) and is spreading as a wild plant in Taiwan.

L.-C. Ding · C.-y. Hu · K.-W. Yeh · P.-J. Wang

Development of insect-resistant transgenic cauliflower plants expressing

the trypsin inhibitor gene isolated from local sweet potato

Communicated by G. Phillips L.-C. Ding · P.-J. Wang

Graduate Institute of Agricultural Biotechnology, National Chung Hsing University,

Taichung, Taiwan C.-y. Hu (

½

Center for Applied Science, William Paterson University, Wayne, NJ 07470, USA

Fax no.: +1-973-7202338

e-mail: huc@nebula.wilpaterson.edu K.-W. Yeh

Department of Botany, National Taiwan University, Taipei, Taiwan

Materials and methods

Plant materials and in vitro conditions

Three key Taiwanese cauliflower (Brassica oleracea var. botrytis L.) cultivars, Known You Early no. 2, Snow Lady, and Beauty Lady, were used. Seeds were disinfected in 70% ethanol for 10 s, followed by 0.5% sodium hypochlorite for 50 min (with sonication during the initial 20 min). B5 medium (Gamborg et al. 1968) with 3% sucrose was used as the basal medium, and for solid medium, 0.8% Bacto-agar was used. The incubation conditions for germination and in vi-tro culturing, unless stated otherwise, were 25°C and 16-h photope-riod of approximately 28µE m–2s–1. Four-day-old germinated seed-lings were cut into explant sections: 5 mm for hypocotyl and coty-ledon petiole, 2 mm2_{for cotyledon.}

Bacterium and plasmid

Agrobacterium tumefaciens LB4404::pBI121/TI was used to

trans-form seedling explants. Plasmid pBI121/TI (Fig. 1) was construct-ed by inserting the TI structural gene (Yeh et al. 1997) between the 35S promotor and the GUS structural gene of pBI121. The plasmid was introduced into LB4404 via microprojectile bombardment. A single colony of bacterium was transferred into liquid bacterial me-dium (YEP meme-dium containing 100 mg/l kanamycin) and shaken at 240 rpm under 28°C for 2 days. The suspensions were then mixed with sterile glycerol (1:1 vol/vol) and stored at –80°C until used. Eighteen hours before inoculation, the bacterium suspension was mixed with the bacterial medium (1:25 vol/vol) and shake-incubat-ed in the above conditions. Acetosyringone was addshake-incubat-ed to the suspen-sion after a 10-h incubation at a concentration of 50µM. By the end

of 18 h of incubation the suspension reached an OD600reading of

0.8 with approximately 3×108cells/ml. The bacterial medium in the suspension was then centrifuged and replaced with an equal volume of inoculation medium [liquid CI medium (see below)+10 mM D-glucose] before use.

Stage 1: developing an efficient in vitro regeneration system Initially three explant types were tested for their regeneration capac-ities in basal medium with 0.2 mg/l indole-3-acetic acid (IAA) and 0.5, 1.0, 2.0, or 4.0 mg/l N6-benzyladenine (BA). Hypocotyl explants were used in all the subsequent experiments including the testing of desirable BA concentration from a gradient ranging from 0.5 to 5.0 mg/l at 0.5 mg/l increments with a fixed amount of 0.2 mg/l IAA. The ability to pretreat explants on a callus-inducing (CI) medium [basal medium with 1 mg/l 2,4-dichlorophenoxyacetic acid (2,4-D) and 0.5 mg/l kinetin] and supplementing the shoot induction medi-um with silver ion (14.7, 29.4, 58.8, 86.2, and 117.6µM as silver thiosulfate; prepared by mixing 1:3 molar concentration solutions of silver nitrate and sodium thiosulfate and added to culture medium via filter sterilization) to increase shoot-bud regeneration percentag-es was also tpercentag-ested. Basal medium, supplemented with 0.2 mg/l IAA, 1.0 or 5.0 mg/l BA and 29.4µMsilver ion, was used as the standard shoot induction (SI) medium in later experiments.

Stage 2: defining the Agrobacterium transformation parameters Since neomycin phosphotransferase II (NPTII) was the selectable marker gene in our system, explants were first tested for their

re-sponses in SI medium containing various concentrations of the se-lective agents: kanamycin (50, 25, 12.5, and 6.25 mg/l) or G418 (30, 15, 7.5, and 3.25 mg/l). In transformation experiments, the 150 mg/l carbenicillin-containing SI medium with no selective agent was des-ignated as SI-1 medium; supplemented with 25 mg/l kanamycin or 15 mg/l G418 it was designated SI-2 medium; with 50 mg/l kanam-ycin or 30 mg/l G418 as SI-3 medium.

The CI-medium-pretreated hypocotyl explants were inoculated with 10 ml bacteria-containing inoculation medium. Bacterial inoc-ulation concentrations tested were 1, 10 and 100×dilutions. Dishes were incubated in darkness for 1 h at 10, 15, 20, 25, or 30°C. After blotting dry with sterile filter papers the explants were incubated on solid CI medium for 72 h, 25°C, dark cocultivation. The cocultivat-ed explants were washcocultivat-ed with 1.5% D-mannitol with 1500 mg/l car-Fig. 1 Construct of pBI121/TI

vector used in cauliflower transformation (TI 0.66 kb, pBI121 13 kb, pBI121/TI 13.66 kb)

Fig. 2A, B Effects of timing and concentration of antibiotic selec-tive regimes on resistant cauliflower plant regeneration following cocultivation with pBI121/TI vector containing Agrobacterium. A Flow chart of antibiotic selection regimes [1–4 week after cocul-tivation; selective media: N no selection, using SI-1 medium (SI dium+150 mg/l carbenicillin), L low-level selection, using SI-2 me-dium (SI-1+25 mg/l kanamycin or 15 mg/l G418), H high-level se-lection, using SI-3 medium (SI-1+50 mg/l kanamycin or 30 mg/l G418]. B Percentages of explants of Known You Early no. 2, Snow Lady, and Beauty Lady producing putative transgenic plants under different G418 selection regimes (vertical bar standard error, ND not determined)

benicillin, blotted dry and placed in solid SI-3 medium to induce re-generation.

Timing for antibiotic selection was determined by transferring explants into SI-1, SI-2, and SI-3 media in various combination re-gimes (Fig. 2A) during the 4-week periods following cocultivation.

Stage 3: confirmation of the TI transgenic status of the resultant plants

Some of the putative transgenic plants derived from the selection re-gime “e” (Fig. 2A) were used for in vitro TI activity, Southern blot-ting and immunoblot (Western blotblot-ting) assays.

Protein extraction

Soluble proteins were extracted from 0.3 g of fresh leaf of each of ten putative transgenic and two control plants by grinding in liquid nitrogen with 3× vol of extraction buffer containing 30 mMTris-HCl (pH 7), 1% PVPP, and 1% vitamin C. The soluble-protein-contain-ing supernatant obtained after centrifugation at 12 000 rpm for 40 min was stored at –70°C for in vitro TI activity and immunoblot assays. The protein concentration of the supernatant was determined with Bio-Rad Protein Assay Kit II following its microassay proce-dure.

In vitro TI activity assay

The TI activity assay was based on the Geiger and Fritz (1984) cell-free trypsin assay method with Bz-L-Arg-4NA [Na

-benzoxyl-1-arginine-4-nitroanilide hydrochloride (Merck)] as substrate. Trypsin hydrolyzes the substrate and forms 4-nitroaniline which absorbs light at OD405. TI activities were measured by the reduction in the

above OD405readings due to the presence of TI over the TI-free

con-trol.

Genomic DNA isolation

DNA was isolated from fresh leaf tissues as described by Dellapor-ta et al. (1984). Primers used in PCR were the flanking sequences of the TI cDNA with 5′CATGAAAGCCCTCACACTG3′at the 5′end and 5′CATTACACATCGGTAGGTTTG3′at the 3′end. PCR prod-ucts and biotinylated probe of TI cDNA [prepared according to the Feinberg and Vogelstein (1983) protocol] were used in Southern hy-bridization using the NEBlot Phototope Kit and its procedure from New England Biolabs (Beverly, Mass.).

Immunoblotting

Appropriate quantities of protein were fractionated by SDS-PAGE following the Laemmli (1970) procedure, and using Mini Trans-Blot Cell (Bio-Rad, Hercules, Calif.) transferred to nitrocellulose. The TI polypeptide was detected using a rabbit anti-TI serum and a goat-rabbit IgG coupled to horseradish peroxidase as secondary anti-body.

In planta bioassays

Small-scale insect-feeding trials were carried out. Leaves of two in vitro cloned plants of the transgenic Snow Lady plant no. 1296 1-1 were used to feed the first-instar larvae of the common Brassica man-dible Lepidopteran pests, Spodoptera litura and Plutella xylostella, in closed containers. The remaining confirmed transgenic plants were cloned in vitro and used in an open infestation test carried out in a greenhouse with opened windows.

Fig. 3A–C Effects of 3 days 2,4-D preincubation and addi-tion of ethylene inhibitor (29.4µMsilver thiosulfate) in regeneration medium on adven-titious bud formation of cauli-flower hypocotyl explants

(ver-tical bar standard error, some

of which were smaller than the symbols)

Results and discussion

Stage 1: developing an efficient

in vitro regeneration system

Among the three explant types tested, the hypocotyl

ex-plants of all the cultivars expressed the highest

adventi-tious bud regeneration capacity (data not shown) after 3

weeks incubation, primarily at the cut ends adjacent to the

apical meristems. The regeneration capacity was weaker

on petiole explants and few, if any, buds appeared on the

cotyledon explants. The optimum BA concentration for

shoot bud regeneration for Known You Early no. 2 and

Snow Lady was 5.0 mg/l and for Beauty Lady 1.0 mg/l. At

these BA concentrations, the percentage of explants

regen-erating for Beauty Lady, Snow Lady and Known You Early

no. 2 were 74, 68, and 37, respectively (Fig. 3, Controls).

A significant breakthrough was achieved when the

ex-plants were (a) pretreated with CI medium for 3 days,

fol-lowed by (b) supplementing the regeneration medium with

an ethylene inhibitor, 29.4

µ

silver ion. The regeneration

percentages of all three cultivars, even the weakly

regen-erating Known You Early no. 2, exceeded 95% (Fig. 3).

All silver ion concentrations tested showed a similar

de-gree of stimulation (data not shown). This work is the first

successful case of applying an ethylene inhibitor to in

vi-tro cauliflower regeneration. The stimulating effects of the

silver ion and other ethylene inhibitors on in vitro

regen-eration of Brassica have been documented in recent years

(Chi et al. 1990; Palmer 1992; Burnett et al. 1994). Added

proof of ethylene regulating in vitro shoot regeneration was

provided by constructing transgenic Brassica plants with

an antisense ACC oxidase gene (Pua and Lee 1995).

cauliflower plants and the non-resistant control (CK) plants (Sample

1296 Snow Lady, 1298 Beauty Lady, 1-1, 1-2, etc. putative

transgen-ic plants). Values are the average of three readings ±SD. The lower the OD value the stronger the suppression of trypsin by TI; thus, the lower the OD value, the higher the TI level. Values followed by dif-ferent letters are significantly difdif-ferent at the 5% level

Sample OD405 Sample OD405 1296 1-1 1.256b± 0.061 1298 2-2 1.101d± 0.037 1-2 1.120c± 0.022 3-1 1.175c± 0.007 1-3 1.265b± 0.041 3-2 1.127b± 0.051 2-1 0.943d_{± 0.010 3-3 1.184}bc_{± 0.009} CK 1.387a_{± 0.016 5-1 1.074}d_{± 0.014} 5-3 0.831e± 0.031 CK 1.358a± 0.003 LSD = 0.0643 LSD = 0.0649 Fig. 4 PCR and Southern blot

analysis of the selected G418-resistant regenerant (R0)

cauli-flower plants (1296 Snow Lady, 1298 Beauty Lady, C. K. control)

Fig. 5 SDS-PAGE and West-ern blot analysis of TI protein from the selected G418-resist-ant regenerG418-resist-ant (R0) cauliflower

plants (1296 Snow Lady, 1298 Beauty Lady, C. K. control)

Stage 2: defining the Agrobacterium transformation

parameters

The optimum inoculation conditions determined (data not

shown) were 100

×

dilution of bacterium suspension

inoc-ulated at 20–25°C. All three cauliflower cultivars had

sim-ilar antibiotic sensitivities (data not shown). In media

con-taining the higher two concentrations of either antibiotic,

the regenerated buds, when produced, were chlorotic and

the explants gradually turned white or brown. At least 2%

of the explants in the lower two antibiotic concentrations

regenerated green buds. Therefore, the higher

concentra-tions of 25–50 mg/l kanamycin or 15–30 mg/l G418 were

used to exert selective pressures in subsequent

transforma-tion experiments.

Among the various antibiotic selection regimes (Fig. 2),

the regime “f” produced the highest percentage of

antibio-tic-resistant buds. Regime “e” had the next highest

percent-ages. The in vitro quantitative TI protein activity analysis

performed later indicated (data not shown) that the average

TI content of the putative transgenic plants from regime “f”

was considerably lower than that from regimes “d” and “e”

which were comparable to each other. This suggested that

many plantlets from regime “f” might be escapes or

chime-ras. Thus, this regime could not be relied on. There was no

green-bud regeneration when antibiotic selective pressure

in-festation test carried out in a greenhouse with open windows (left TI transgenic cauliflower plants in vitro cloned from the R0plants, right control plants).

The insects were identified as

Pieris conidia (A) and Plutella xylostella (B)

began right after cocultivation in regimes “a”, “b”, and “c”.

The above results indicated that the buds of cauliflower,

even the NPTII-gene-containing ones, were more sensitive

to the selective agents at the early stages of development.

They became progressively more tolerant to the selective

antibiotics, even without the resistant gene. Thus, the lack

of selective agent during the first week after cocultivation,

followed by progressively increasing the selective pressure,

i.e. regime “e”, provided the optimum conditions for

select-ing NPTII-containselect-ing transgenic buds.

Data from the same experiment also indicated that G418

selection resulted in a higher number of regenerated buds

than that of kanamycin. In regime “e”, for example, the

percentages of explants regenerating buds from Known

You Early no. 2, Snow Lady, and Beauty Lady were 8.7,

10.3, and 33.3, respectively, from G418 selection (see

Fig. 2B), while from kanamycin selection the percentages

were 0, 2.0, and 0, respectively (data not shown). Thus,

G418 selection provided a higher probability of obtaining

putative transgenic plants.

Stage 3: confirmation of the TI transgenic status

of the resultant plants

Over 100 G418/kanamycin-resistant plantlets were

ob-tained from stage 2 experiments. The functional integrity

of the TI gene products in these plants was demonstrated

by an in vitro TI activity assay (Table 1). All the tested

pu-tative transgenic plants showed significantly higher TI

pro-tein content than control plants.

The genomic DNAs from all tested plants produced

0.66-kb DNA segments following PCR, while that of the

control plant did not (Fig. 4). Southern hybridization

(Fig. 4) with TI cDNA probe confirmed that these 0.66-kb

DNA segments were TI gene sequences. We recognize that

the Southern procedure we performed could not preclude

the possibility of DNA contamination from Agrobacterium

which might still be present on the tested regenerant (R

)

plants. Nevertheless, the combination of results from the

in vitro TI activity assay, immunoblotting, and in planta

bioassay indicated that functional TI protein indeed had

been translated in the tissues of these R

plants. A

prog-eny test is currently underway which will verify

inheri-tance of the TI transgene.

The TI polypeptide was detected as a 24-kD band

(Fig. 5) which was present in all six transformant plants

tested, while the control plant did not give such a band.

The expression of the TI genes at both transcriptional and

translational levels was, thus, verified in the transgenic

plants. The in vitro TI activity assays performed above

(Ta-ble 1) indicated that those translational products were

functionally active in vitro as well.

A high degree of insect protection was evident in our

bioassays (Table 2, Fig. 6). These results demonstrated

that the TI proteins produced in the transgenic cauliflower

plants were functionally active in planta. High expression

of protease inhibitor genes in transgenic plants have

pre-viously been reported to have enhanced insect resistance

potentials (Hilder et al. 1990; Thomas et al. 1995; Duan et

al. 1996; Xu et al. 1996). Our insect bioassays, although

at small scales initially, also provided evidence that the

in-troduction of a protease inhibitor gene into transgenic crop

plants can be efficacious.

Our protocol developed for the effective genetic

engi-neering of Taiwan cauliflower can be used as a practical

starting point for the transformation of other cauliflower

cultivars. Analysis of the resultant R

and R

generation

plants as well as a larger scale field test of the clonal R

plants are currently in progress.

Acknowledgements This joint project was supported by (Taiwan) National Science Council (NSC84-2331-B002-050-B14 and NSC85-2621-B005-012-B14), A. R. T. and the Center for Research of William Paterson University. C. Y. H.’s trip to Taiwan was sup-ported by Known You Seed Company. We are grateful to Dr. Ste-phen Vail, Ms. Donna Potacco and Dr. Kevin Martus for preparing computer graphics.

References

Burnett L, Arnoldo M, Huang B (1994) Enhancement of shoot re-generation from cotyledon explants of Brassica rapa ssp. oleife-ra through pretreatment with auxin and cytokinin and use of eth-ylene inhibitors. Plant Cell Tissue Organ Cult 37:253–256 Chi GL, Barfield DG, Sim GE, Pua EC (1990) Effect of AgNO3and

aminoethoxyvinylglycine on in vitro shoot and root organogen-esis from seedling explants of recalcitrant Brassica genotypes. Plant Cell Rep 9:195–198

David C, Tempe J (1988) Genetic transformation of cauliflower (Brassica oleracea L. var. botrytis) by Agrobacterium

rhizo-genes. Plant Cell Rep 7:88–91

De Block M, De Brouwer D, Tenning P (1989) Transformation of

Brassica napus and Brassica oleracea using Agrobacterium tu-mefaciens and the expression of the bar and neo genes in the

transgenic plants. Plant Physiol 91:694–701

Dellaporta SL, Wood J, Hicks JB (1984) Maize DNA miniprep. In: Malmberg R, Messing J, Sussex I (eds) Molecular biology of plants. Cold Spring Harbor Laboratory Press, Cold Spring Har-bor, NY, pp 36–37

Duan X, Li X, Xue Q, Abo-El-Saad M, Xu D, Wu R (1996) Trans-genic rice plants harboring an introduced potato proteinase in-hibitor II gene are insect resistant. Nature Biotechnol 14:484–498 Eimert K, Siegemund F (1992) Transformation of cauliflower

(Bras-sica oleracea L. var. botrytis) – an experimental survey. Plant

Mol Biol 19:485–490

Feinberg AP, Vogelstein B (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6–13

Table 2 In planta feeding bioassays using two clonal plantlets of a TI transgenic cauliflower Snow Lady plant 1296 1-1 and non-trans-genic control (CK) plants. The experiment was carried out once with-out replication

Insect species Host Larvae Larvae Larvae plant survived reaching reaching

(%) pupa stage adult stage

(%) (%)

Spodoptera litura 1-1 56.7 22.7 13.2

CK 86.7 63.6 49.1

Plutella xylostella 1-1 60.0 36.0 21.6

Gamborg OL, Miller RA, Ojima K (1968) Nutrient requirements of suspension cultures of soybean root cells. Exp Cell Res 50: 151–158

Geiger R, Fritz H (1984) Two proteinases and their inhibitors – tryp-sin. In: Bergmeyer J, Grabl M (eds) Methods of enzymatic anal-ysis, vol V. Enzymes 3: peptidases, proteinases and their inhibi-tors. Chemiew, Weinheim, pp 119–129

Hilder VA, Gatehouse AMR, Boulter D (1990) Genetic engineering of crops for insect resistance using genes of plant origin. In: Ly-cett GW, Grierson D (eds) Genetic engineering of crop plants. Butterworths, London, pp 51–65

Laemmli UK (1970) Cleavage of structural proteins during the as-sembly of the head of bacteriophage T4. Nature 227:680–685 Lin MZ (1996) Study on gene expression and insect-resistance of

sweet potato sporamin in transgenic tobacco. MS thesis, Depart-ment of Botany, National Taiwan University

Palmer CE (1992) Enhanced shoot regeneration from Brassica

cam-pestris by silver nitrate. Plant Cell Rep 11:541–545

Passelegue E, Kerlan C (1996) Transformation of cauliflower

(Bras-sica oleracea var. botrytis) by transfer of cauliflower mosaic

vi-rus genes through combined cocultivation with virulent and avir-ulent strains of Agrobacterium. Plant Sci 113:79–89

Praekelt UM, McKee A, Smith H (1988) Molecular analysis of ac-tinidin, the cysteine protease of Actinidia chinensis. Plant Mol Biol 10:193–202

Pua EC, Lee JEE (1995) Enhanced de novo shoot morphogenesis in vitro by expression of antisense 1-aminocyclopropane-1-carbox-ylate oxidase gene in transgenic mustard plants. Planta 196:69–76 Srivastava V, Reddy AS, Guha-Mukherjee S (1988) Transformation and regeneration of Brassica oleracea mediated by an oncogen-ic Agrobacterium tumefaciens. Plant Cell Rep 7:504–507 Thomas JC, Adams DG, Keppenne VD, Wasmann CC, Brown JK,

Kanost MR, Bohnert HJ (1995) Proteinase inhibitors of

Mandu-ca sexta expressed in transgenic cotton. Plant Cell Rep 14:

758–762

Xu D, Xue Q, McElroy D, Mawal Y, Hilder VA, Wu R (1996) Con-stitutive expression of a cowpea trysin inhibitor gene, CpTi, in transgenic rice plants confers resistance to two major rice insect pests. Mol Breed 2:167–173

Yeh KW, Chen JC, Lin MI, Chen YM, Lin CY (1997) Functional ac-tivity of sporamin from sweet potato (Ipomoea batatas Lam.): a tuber storage protein wtih trypsin inhibitory activity. Plant Mol Biol 33:565–570