INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

CS-type antimicrobial peptides contain a single  helix and a pair of antiparallel  sheets stabilized by three or four intramolecular disulfide bridges (9). Insect defensins that contain three intramolecular disulfide bridges (7, 12), drosomycin (isolated from the fruit fly [Drosophila melanogaster]) (16), and plant defensin (3), which contains four intramolecular disulfide bridges, have been experimentally demonstrated to be members of a CS-type antimicrobial peptide by three-dimensional structural analyses. Some antimicrobial peptides that contain the consensus sequence of insect defensins (6) have been isolated from other arthropods and mollusks (4, 6, 10, 13), suggesting that they should also be CS-type antimicrobial peptides. Although these peptides form similar three-dimensional structures, the CS-type antimicrobial peptides exhibit diversified antimicrobial spectra. For instance, insect defensins are active against gram-positive bacteria but not against gram-negative bacteria or eukaryotic microbes (9), despite some exceptions (16). In contrast, drosomycin is active against fungi but not against bacteria (14). Plant defensins are categorized into four groups with different antimicrobial spectra, i.e., group I (active against gram-positive bacteria and fungi), group II (active against fungi but inactive against bacteria), group III (active against gram-positive and gram-negative bacteria but inactive against fungi), and group IV (active against gram-positive and gram-negative bacteria and fungi) (11). In addition, the antimicrobial activities of some CS-type antimicrobial peptides are inhibited by salts (5, 17), but those of others are not (15). However, the structure-activity relationship that can explain such diversified antimicrobial characteristics has not been well elucidated. Characterization of novel CS-type antimicrobial peptides has been providing novel antimicrobial substances with unique properties (15).

ASABF is a novel CS-type antimicrobial peptide that contains four intramolecular disulfide bridges isolated from the nematode Ascaris suum (14). In this study, a recombinant ASABF was produced with a yeast expression system, and the antimicrobial activity was characterized in detail.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of ASABF expression vector. The mature peptide region of ASABF was amplified by a high-fidelity PCR with the full-length cDNA clone of ASABF (14) as a template and the following set of primers: a sense primer (44-mer) that includes the region of ASABF from Ala¹⁹ to Cys²⁵ and the XhoI site (underlined) (5'-CCTTAGGGCCCCTCGAGAAAAGAGCAGTCGACTTTTCATCATGC-3') and an antisense primer (45-mer) that included the region from the terminal codon to Lys⁸³ and the NotI site (5'-GCCGAGCTCTGCAGCGGCCGCCTATCCACGTGAACTTCGCCCTTT-3'). This product was ligated to the pPIC9 vector by using the XhoI-NotI sites. By using this construct, the recombinant ASABF was expected to be produced as a secretory fusion peptide flanking the -factor secretory signal at the N terminus.

Expression of recombinant ASABF using a yeast expression system. After linearization by digestion with SacI, the ASABF expression vector was transformed into Picha pastoris GS115. The P. pastoris transformant was inoculated into 40 ml of BMGY medium (1% yeast extract, 2% peptone, 100 mM potassium phosphate [pH 6.0], 1.34% yeast nitrogen broth, 4 × 10⁵% biotin, 1% glycerol), and the culture was grown at 30°C until the culture reached an optical density at 600 nm of 6. The transformants were harvested by centrifugation and resuspended to an optical density at 600 nm of 1 in 400 ml of BMMY medium (1% yeast extract, 2% peptone, 100 mM potassium phosphate [pH 6.0], 1.34% yeast nitrogen broth, 4 × 10⁵% biotin, 1% methanol) to induce recombinant ASABF expression. The culture was grown for 6 days at 30°C in a shaking incubator by adding 100% methanol to a final concentration of 1% every 24 h to maintain the induction. After the incubation, the expression culture was centrifuged, and the supernatant was mixed with Hyflo Super-cell and filtered through a nitrocellulose membrane filter.

Purification of recombinant ASABF. (i) Step 1. Positive-ion-exchange chromatography. The supernatant was diluted fivefold to obtain a lower ionic strength and was applied to a SP-Sepharose FF column (Amersham Pharmacia Biotech) which had been equilibrated with 20 mM potassium phosphate (pH 6.0) at 4°C. Recombinant ASABF was eluted with 0.5 M NaCl. The A₂₈₀ was monitored, and the ASABF-containing fractions were collected and dialyzed against deionized water.

(ii) Step 2. Positive-ion-exchange high-pressure liquid chromatography. The dialyzed fractions were applied to a HiTrap SP column (Amersham Pharmacia Biotech) connected to a Pharmacia fast-protein liquid chromatography system. The recombinant ASABF was eluted with a linear concentration gradient (0 to 300 mM) of NaCl containing 50 mM potassium phosphate buffer (pH 6.0). The fractions containing ASABF were collected.

(iii) Step 3. Reversed-phase high-pressure liquid chromatography. The final purification was carried out with an Asahipak C4P-90 2F column (Asahi Chemical Industry Co., Ltd.). A linear gradient elution was used (20 to 40% acetonitrile in water containing 1% trifluoroacetic acid). The purified recombinant ASABF was lyophilized and stored at 4°C.

Microorganisms. (i) Bacterial strains. Escherichia coli JM109 was purchased from Takara. Staphylococcus aureus ATCC 6538P was a gift from Masanori Yamamoto. Bacillus subtilis IFO3134 was purchased from the Institute for Fermentation (IFO), Osaka, Japan.

(ii) Yeast strains. Candida krusei MAFF114085, Debaryomyces hansenii MAFF113836, Kloeckera apiculata MAFF114302, Kluyveromyces thermotolerans MAFF113848, Picha anomala MAFF113717, Saccharomyces cerevisiae MAFF113011, Sporobolomyces sp. strain MAFF425173, Torulaspora delbrueckii MAFF113811, and Zygosaccharomyces rouxii MAFF113405 were obtained from the National Institute of Agrobiological Resources, Tsukuba, Japan.

Microbicidal assay. Each microbial strain in the logarithmic phase of growth was suspended in 10 µl of 10 mM Tris-HCl (pH 7.5) containing a series of purified recombinant ASABF for a threefold increase in concentration. The optical density of the microbial suspension was adjusted to 0.02 at 650 nm. After 2 h of incubation, the test suspension (5 µl) was diluted 1,000 times. The diluted sample (200 µl) was inoculated onto an optimum medium. The numbers of colonies were counted, and the minimum bactericidal concentration (MBC) was determined.

Hemolytic assay. Human type A erythrocytes were used for the hemolytic assay. Hemolysis was estimated as the leakage of hemoglobin. The erythrocytes were washed in 10 mM Tris-HCl (pH 7.6) containing 154 mM NaCl or 308 mM sucrose (308 mosM) and were resuspended in the same buffer containing recombinant ASABF. After 0.5 h of incubation, the test suspension was centrifuged to remove the intact erythrocytes. The supernatant was diluted, and the A₅₄₀ was measured.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Expression of recombinant ASABF. To characterize the antimicrobial activity in detail, recombinant ASABF was expressed by using the yeast P. pastoris (see Materials and Methods). The recombinant ASABF was designed as a fusion peptide with an -factor secretory signal at the N terminus with a downstream-flanking mature region of ASABF. The recombinant ASABF was collected as a secretory peptide from the culture supernatant. The purified peptide was analyzed by N-terminal sequencing and mass spectrometry (Fig. 1). The N-terminal sequence of the recombinant ASABF was AVDFSSCARM, which is completely identical to that of natural ASABF (14), suggesting that the -factor secretory signal was cleaved as expected. The recombinant ASABF was estimated to be a mixture of two peptides whose molecular masses were 7,415.8 and 7,358.4 Da, respectively, on the basis of mass spectrometry. These peptides are thought to be the processed peptides in which a four-residue peptide (calculated mass, 7,412.4 Da) and a five-residue peptide (calculated mass, 7,355.5 Da) at the C terminus were eliminated. It is noteworthy that a four-residue peptide at the C terminus of natural ASABF was also removed by processing (14). Because the differently processed recombinant ASABF was only slightly separated, the mixture of peptides was directly subjected to characterization of its antimicrobial properties.

View larger version (8K): [in this window] [in a new window]   FIG. 1.   Amino acid sequence of recombinant ASABF. The recombinant ASABF was analyzed by N-terminal sequencing and mass spectrometry. The sequence obtained by N-terminal sequencing is underlined. Numbers indicate molecular masses (in Daltons). The experimentally determined molecular mass is followed by its theoretical value in parentheses. The filled triangle indicates the cleavage site of processing in native ASABF.

Time dependence of bactericidal activity. S. aureus ATCC 6538P, which was the bacterium the most sensitive to natural ASABF (14), was treated with 0.1 µg of recombinant ASABF per ml (13 mM) in 10 mM Tris-HCl buffer (pH 7.6) for various times, and the viability was tested (Fig. 2). When the viability of S. aureus was tested in the presence of 100 µg chloramphenicol per ml or in buffer with no antibiotics, no significant decrease in viability was observed. In contrast, 93% of the bacteria were killed within 1 min of contact with ASABF, suggesting that ASABF is bactericidal.

View larger version (19K): [in this window] [in a new window]   FIG. 2.   Rate of killing of S. aureus. Bacteria were incubated for various times, and their viabilities were tested. Most of the bacteria (93%) were killed within 1 min of contact with ASABF.

Inhibition by salts. The antimicrobial activities of some known CS-type antimicrobial peptides, such as insect defensins or plant defensins, are inhibited by salts (see the introduction). The bactericidal activity of recombinant ASABF (10 ng/ml [0.13 nM]) against S. aureus ATCC 6538P was tested in 10 mM Tris-HCl buffer (pH 7.6) with or without 150 mM NaCl (Fig. 3A). The bactericidal activity decreased 1/400 in 150 mM NaCl, suggesting that NaCl inhibits the activity of ASABF. Next, the bactericidal activity of recombinant ASABF in the presence of various concentrations of NaCl was explored (Fig. 3B). The inhibitory effect of NaCl was concentration dependent and was saturated at about 40 to 80 mM. The bactericidal activity was more strongly inhibited by CaCl₂ (Fig. 3B). In addition, 0 to 330 mM sucrose slightly affected the bactericidal activity of recombinant ASABF, but the influence of sucrose was far different from the influence (inhibition) of NaCl (Fig. 3C), suggesting that the inhibition by salts should be attributed to an electrostatic interaction.

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Inhibition of bactericidal activity of ASABF by salts. S. aureus ATCC 6538P was treated with 10 ng of ASABF per ml in 10 mM Tris-HCl (pH 7.6) for 2 h. (A) Dose-effect relationship in the presence or absence of 150 mM NaCl. The bactericidal activity decreased 1/400 in 150 mM NaCl. (B) Influence of NaCl and CaCl2. CaCl2 more strongly inhibited bactericidal activity than NaCl did. (C) Influence of osmotic pressure. Sucrose did not strongly affect the activity of ASABF. Osm, osmolar.

Antimicrobial spectrum. The antimicrobial spectrum of recombinant ASABF was tested under an optimum condition, i.e., in 10 mM Tris-HCl buffer (Table 1). Due to the good reproducibility, we estimated the activity intensity as the MBC. All (seven of seven) gram-positive bacteria tested were sensitive. The MBC for gram-positive bacteria was estimated to be 0.03 to 1 µg/ml (4 to 130 nM) except only for that for L. mesenteroides (10 µg/ml [130 nM]). Although some (8 of 14) gram-negative bacteria were also sensitive, a higher concentration was required (i.e., MBC, >0.5 µg/ml [70 nM]). Interestingly, recombinant ASABF was also active against some (three of nine) yeasts (i.e., MBC, >3 µg/ml [400 nM]).

Hemolytic activity. Recombinant ASABF is effective not only against bacteria but also against some yeasts, suggesting that the toxicity of ASABF is not limited to prokaryotes. A CS-type antimicrobial peptide, an insect defensin, is reported to disrupt the permeability barrier of the cytoplasmic membranes of gram-positive bacteria (5). To test the toxic activity of ASABF against the cytoplasmic membranes of higher animals, the hemolytic activity against human red blood cells was explored (Fig. 4). No hemolytic activity was detected in 150 mM NaCl, although amphotericin B caused severe hemolysis (19). The substitution of 300 mM sucrose for 150 mM NaCl caused a slight hemolysis (4.2%) in the presence of 100 µg of recombinant ASABF per ml (13 µM) (note that 0.8% hemolysis was observed even for the ASABF-free control). No prominent increase in hemolysis was observed at a higher concentration (4.5% at 1,000 µg of ASABF per ml). These results suggest that ASABF is much less harmful to the cytoplasmic membranes of higher animals, and its limited toxicity is inhibited by salts.

View larger version (17K): [in this window] [in a new window]   FIG. 4.   Hemolytic activity of ASABF. The hyposmotic hemolysis in 10 mM Tris-HCl (pH 7.6) was defined as 100% hemolysis. The hemolysis was measured in 10 mM Tris-HCl (pH 7.6) containing isotoinc NaCl (stippled column) or sucrose (filled column). Amphotericin B was used as a positive control. Slight hemolysis (4.2% at 100 µg/ml) was observed only under low-ionic strength conditions.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, ASABF, a CS-type antimicrobial peptide isolated from the nematode A. suum, was produced as a recombinant peptide with a yeast expression system, and its antimicrobial activity was explored in detail.

A four-residue peptide at the C terminus of natural ASABF was removed by processing (14). The removed peptide contains a dibasic site thought to be a recognition site for the processing of enzymes. Although the recombinant ASABF was also processed in a similar way, we observed recombinant peptides from which not only the four-residue peptide but also the five-residue peptide had been removed. Many proteases which recognize dibasic sites have been reported, and their cleavage sites are diversified (18). The recombinant ASABF could be processed by multiple proteases which recognized the dibasic site and cleaved it at different positions.

In this study, we tested the bactericidal activity of ASABF against 31 species of various microbes. The results suggest that ASABF is effective against gram-positive and gram-negative bacteria and yeasts. To the best of our knowledge, only a group IV plant defensin, So-D2, and an insect defensin, Phormia defensin, have also been reported to be CS-type antimicrobial peptides active against both prokaryotic and eukaryotic microbes. Further analysis of ASABF will provide a new key to revealing the structure-activity relationships of the CS-type antimicrobial peptides. Since recombinant ASABF could be obtained with a yeast expression system, the use of mutational analysis should be possible (8).

The hemolytic activity of ASABF is limited and is observed only under lower-ionic-strength conditions, suggesting that ASABF is much less harmful to the cytoplasmic membranes of higher vertebrates. On the other hand, ASABF was active against some yeasts, suggesting that the effective site of ASABF is not specific in prokaryotes. It remains to be elucidated whether the selective toxicity of ASABF is attributed to resistance factors in higher animals or sensitivity factors that may be common in both prokaryotic and eukaryotic microbes.

ASABF killed S. aureus within 1 min after exposure. Similar results have been reported for some membrane-disrupting antimicrobial peptides such as insect defensins (5) and cecropins (1). It is possible to argue that ASABF could also kill microbes by disrupting the cytoplasmic membrane. The exact bactericidal mechanism of ASABF remains to be elucidated.

The bactericidal activity of ASABF was inhibited by salts. The salt inhibition was also reported in other known CS-type antimicrobial peptides such as insect defensins, plant defensins, and drosomycin (see introduction). The antimicrobial activities of cationic lantibiotics (e.g., nisin) are also inhibited by salts (2). The inhibition by salts seems to be a hallmark of antimicrobial peptides as a class. The calcium ion was more inhibitory than the sodium ion. In addition, osmotic pressure did not strongly affect the bactericidal activity of ASABF. These results suggest that inhibition by salts is attributed to the inhibition of the electrostatic interaction between microbial target molecules presumably charged negative and ASABF charged positive at neutral pH. Although the hemolytic activity of ASABF was faint, the inhibition by salts was also observed, suggesting that the nature of the toxicity of ASABF may be a membrane-disrupting activity.

In conclusion, ASABF is a good candidate as a clinically applicable antimicrobial agent because of its wide antimicrobial spectrum and confirmed activity against some pathogens, especially S. aureus. The inhibition by salts both of the microbicidal activity and of the toxicity toward human erythrocytes should be considered for effective clinical applications.