TEXT

Top Abstract Text References

High levels of enzyme inhibitors found in the seeds of many plant species serve as storage or reserve proteins, as regulators of endogenous enzymes, and as defensive agents against attacks by animal predators and insect or microbial pests (15). Among these inhibitors, some were found to have activity against both trypsin and -amylase (3, 16, 17).

The most extensively studied enzyme inhibitor is trypsin inhibitor (TI). Direct evidence of TI involvement in plant defense is that the expression of the cowpea (Vigna unguiculata) TI gene in tobacco increased host resistance against herbivorous insects (7). Antifungal activities have also been reported for TI proteins from several crops, including TIs from barley (18) and trypsin and chymotrypsin inhibitors from cabbage (14) as well as the 22-kDa TI from corn (8) and the 24-kDa cysteine protease inhibitor from pearl millet (9). However, most were described to be active only against a very limited group of fungi. In a recent report from this laboratory (4), a TI was described with efficacy against Aspergillus flavus, the major fungus causing extensive crop loss in corn, cotton, peanut, and tree nuts due to contamination with aflatoxins. This 14-kDa corn TI is present at high levels in corn genotypes resistant to A. flavus infection but at low or undetectable levels in susceptible genotypes (4). The same TI has also been reported to be a specific inhibitor of activated Hageman factor (factor XIIa) of the intrinsic blood clotting process (6), as well as an inhibitor of -amylases from certain insects (1, 3). Purification of the 14-kDa TI from corn requires large quantities of resistant corn kernels, which are usually in short supply. This has hampered efforts to test its efficacy against other important pathogens and to investigate its mechanism of inhibition. Therefore, the objectives of the present study were to overexpress this protein in Escherichia coli to obtain large quantities and to use the purified active recombinant TI to test for inhibition of various plant-pathogenic fungi.

Overexpression of the TI gene in E. coli and purification technique. The complete coding region of mature corn 14-kDa TI cDNA (GenBank accession no. X54064) (19) was amplified from plasmid pT7-7 with Taq polymerase by using the primer pair 2041 (5' GAGCTCTTACTTGGAGGGCATCGTTCCGC) and 2164 (5' CATATGAGCGCCGGGACCTCCTGC) with mismatches (underlined) to introduce an NdeI (5' end) or SacI (3' end) restriction site. After cloning into TA cloning vector pCRII (Invitrogen, Carlsbad, Calif.) and complete sequence analysis, the amplified 0.4-kb PCR product was subcloned into the unique NdeI and SacI sites of an E. coli overexpression vector, pET-28b (Novagen, Madison, Wis.). Positive clones were identified by using PCR according to the manufacturer's instructions. The correct in-frame fusion of the construct was verified by DNA sequencing of positive transformants before it was transformed into an E. coli BL21 (DE3) expression host. TI expression was induced by adding isopropyl--D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM as previously described (5). The overexpressed TI was predicted to be 16.5 kDa, containing a vector His tag and a thrombin cleavage site at the N terminus (MGSSHHHHHHSSGLVPRGSHM) followed by the complete mature TI (127 amino acid residues) (19).

Trypsin activity assay. Trypsin-inhibiting activity of the purified recombinant TI was assayed as previously described with bovine pancreatic trypsin and, as a substrate, -N-benzyl-DL-arginine-p-nitroanilide HCl (BAPNA) (4). All assays were conducted three times.

Antifungal assays. The activity of this recombinant TI against the growth of nine plant-pathogenic fungi (see Fig. 2) was assayed in 10% potato dextrose broth (PDB) (Difco) by using microtiter plates as described previously (4). Conidia (7 days old) from V8 juice agar medium (5% [vol/vol] V8 juice, 2% [wt/vol] agar [pH 5.2]) maintained at 25°C were harvested in sterile water and diluted to 10⁶/ml. For Cercospora kikuchii and Fusarium species, this test was done with macroconidia. Conidia were allowed to germinate and grow in the presence of TI at 50, 100, 200, and 300 µg/ml at 25°C for 12 h. Negative controls were 10 mM phosphate buffer (pH 7.0) or heat-inactivated TI at a concentration of 100 µg/ml. The hyphal length of control or TI-treated fungi was measured with an ocular micrometer after 12 h of incubation at 25°C. For each treatment, the hyphal lengths were measured for at least 40 randomly selected hyphae, and the mean hyphal length was used for comparison. The hyphal length in the control containing heat-inactivated TI was similar to that in the phosphate buffer control. Conidium germination was based on counts of at least 100 conidia per replicate. For C. kikuchii, Fusarium graminearum, and Fusarium moniliforme, which produce multicelled conidia, a conidium was considered germinated if a hypha was visible for at least one of the cells. The bioassay was performed three times for A. flavus and Aspergillus parasiticus and twice for all other fungi, with three replicates per treatment. The data presented are means for all experiments.

Purification and characterization of overexpressed TI. SDS-PAGE analysis of each fraction during purification showed that the overexpressed TI comprised 30 to 35% of total cell protein when the E. coli cells were induced and that it was not readily dissolvable in 6 M urea (Fig. 1). Overexpressed TI that remained insoluble in 6 M urea in the absence of -mercaptoethanol prevented the use of traditional nickel ion affinity chromatography to purify this His-tagged recombinant TI; -mercaptoethanol reacts with nickel ion affinity column material and forms precipitates. However, by taking advantage of this characteristic, the overexpressed TI in the urea-insoluble fraction was further enriched through repeated washings with 50 mM Tris-HCl (pH 9.0) containing 6 M urea, and the need for further purification was eliminated after refolding of the TI. The overexpressed TI accounted for more than 90% of the total protein in the -mercaptoethanol-soluble fraction and for more than 96% of the total protein in the final refolded active TI fraction (Fig. 1). The yield of purified active TI was about 70 mg per liter of culture. It is believed that the second protein, having an apparent molecular mass of 33 kDa in the purified TI fraction (Fig. 1), is a dimer of the overexpressed TI based on the following observations (data not shown): (i) this 33-kDa protein cross-reacted with the antibody raised against purified native 14-kDa corn trypsin inhibitor, (ii) the resolving of the excised 33-kDa protein from a native gel on SDS-PAGE led to the appearance of the 16-kDa protein (the expected size of overexpressed TI), and (iii) the longer the denaturing time, the higher the intensity of the 16-kDa protein and the lower the intensity of the 33-kDa protein. However, we were not able to completely convert the 33-kDa protein to a 16-kDa protein by extending the denaturing time. This eliminated a possible concern that any future antifungal activity observed could be due to the presence of an antifungal compound other than TI in the extract.

View larger version (132K): [in this window] [in a new window]   FIG. 1.   SDS-PAGE analysis of TI overexpressed in E. coli at different stages of purification and refolding. Lane 1, molecular mass standards (5 µg); lane 2, total protein extract from noninduced E. coli cells (10 µg); lane 3, total protein extract from induced E. coli cells (10 µg); lane 4, water-soluble-fraction proteins (10 µg); lane 5, urea-soluble-fraction proteins (5 µg); lane 6, -mercaptoethanol-soluble-fraction proteins (5 µg); lane 7, purified and refolded active TI (2.5 µg, indicated by an arrow). The purity of this TI is 96% as quantified by gel densitometry. Molecular mass standards included were (from top to bottom) albumin (66 kDa), ovalbumin (45 kDa), glyceraldehyde-3-phosphate dehydrogenase (36 kDA), carbonic anhydrase (29 kDa), trypsinogen (24 kDa), TI (20 kDa), -lactalbumin (14.2 kDa), and aprotinin (6.5 kDa). The asterisk indicates the possible dimer of this TI.

Inhibition of fungal growth. The refolded active TI inhibited both conidium germination and hyphal growth of all nine fungi tested (Fig. 2). For most fungi, conidium germination decreased dramatically with increasing TI concentrations (Fig. 2). For six of the nine fungal species tested, conidium germination was reduced 50% at TI concentrations less than 115 µg/ml (Fig. 2, top panels). However, for F. graminearum, F. moniliforme, and C. kikuchii even the presence of 300 µg of TI per ml did not reduce conidium germination by 50% (Fig. 2). It might be significant that, of the nine species tested, the latter fungi are the only ones that produce large multicelled conidia. These conidia may have sufficient endogenous nutritional supplies to facilitate germination under adverse conditions.

View larger version (35K): [in this window] [in a new window]   FIG. 2.   Conidium germination and hyphal growth in the presence of TI overexpressed in E. coli. The percentage conidium germination and hyphal growth were determined after 12 h of incubation at 25°C in 10% PDB medium with different concentrations of TI. The data are means ± standard errors of three (for A. flavus and A. parasiticus) or two (all other fungi) repeated experiments. Symbols: , A. flavus AF13; , A. niger; , A. parasiticus; , F. graminearum; , F. moniliforme; , C. kikuchii; , P. chrysogenum; , R. stolonifer; , T. viride.

Simultaneous inhibition of hyphal growth of A. flavus and F. moniliforme. It was found that hyphal growth of A. flavus and F. moniliforme grown together was inhibited by TI as effectively as when they were grown separately (Fig. 3). More than 50% reduction of hyphal growth was obtained in the presence of 100 µg of TI per ml for F. moniliforme when microconidia were used. However, the calculated IC₅₀ of TI for F. moniliforme hyphal growth was 124 µg/ml when macroconidia were used (Fig. 2). This supported our observation that macroconidia are less sensitive to TI inhibition than microconidia.

View larger version (186K): [in this window] [in a new window]   FIG. 3.   Simultaneous inhibition of A. flavus and F. moniliforme hyphal growth by TI protein overexpressed in E. coli at 25°C. (A and B) A. flavus treated with 0 and 100 µg of TI per ml, respectively, for 12 h; (C and D) A. flavus and F. moniliforme grown together and treated with 0 and 100 µg of TI per ml, respectively, for 12 h; (E and F) F. moniliforme treated with 0 and 100 µg of TI per ml, respectively, for 12 h. Bar = 25 µm.

Conclusions. The two objectives of this study were to overexpress the 14-kDa TI in E. coli to obtain quantities suitable for testing protein bioactivity against several phytopathogenic fungi and to perform the actual antifungal bioassays. The original clone, pT7-7, was able to express TI in E. coli; however, the level was so low that it could be detected only by Western blot analysis. After the coding region of TI from pT7-7 was subcloned into vector pET-28b, the level of recombinant TI in E. coli total cell protein extract increased dramatically compared to that of clone pT7-7. Not only were suitable quantities obtained through overexpression in E. coli but also TI was purified by a novel procedure based on TI's insolubility in urea in the absence of -mercaptoethanol. The possibility exists that proteins overexpressed either in E. coli or in yeast with a high number of disulfide bonds also could be purified by this method.