Acidic Peptide-Mediated Expression of the Antimicrobial Peptide Buforin II as Tandem Repeats inEscherichia coli

PROTEIN EXPRESSION AND PURIFICATION ARTICLE NO.

12, 53–60 (1998)

PT970814

Acidic Peptide-Mediated Expression of the Antimicrobial Peptide Buforin II as Tandem Repeats in Escherichia coli Jae H. Lee, Il Minn, Chan B. Park, and Sun C. Kim1 Department of Biological Sciences, Korea Advanced Institute of Science and Technology, 373-1 Kusong-dong, Yusong-gu, Taejon 305-701, Korea

Received July 1, 1997, and in revised form September 22, 1997

Antimicrobial peptides have received increasing attention as a new pharmaceutical substance, because of their broad spectrum of antimicrobial activities and the rapid development of multidrug-resistant pathogenic microorganisms. The main obstacle to the wide application of antimicrobial peptides has been the lack of a cost-effective, mass-production method. A novel mass-production method for an antimicrobial peptide of 21 amino acids, buforin II, which was isolated from the stomach of the amphibian Bufo bufo gargarizans, has been developed. This method is based on the neutralization of the positive charges of buforin II by fusing to an acidic peptide to avoid the lethal effect of the expressed antimicrobial peptide on the host cells. The fusion peptide was expressed in Escherichia coli as tandem repeats to incr ease the product yield. Multimers of the acidic peptide–buforin II fusion peptide were expressed at high levels without causing damage to the cells. The presence of cysteine residues in the acidic peptide was critical for the high level expression of the fusion peptide multimers. Multimers of this fusion peptide were expressed as inclusion bodies, and about 107 mg of pure buforin II was obtained from 1 L of E. coli culture by cleaving the multimers with CNBr. Recombinant buforin II had an antimicrobial activity identical to that of natural buforin II. These results may lead to a general, cost-effective solution to the mass production of antimicrobial peptides and other basic peptides which are lethal to the host strain. q 1998 Academic Press Key Words: antimicrobial peptide, expression, acidic peptide, tandem multimer, Escherichia coli.

Antimicrobial peptides, which play one of the key roles in primary host defense against infections of 1 To whom correspondence should be addressed. Fax: /82 42 869 2610. E-mail: [email protected].

pathogenic microorganisms, have received increasing attention as new antimicrobial substances in recent years. This is because of their broad spectrum of antimicrobial activities and the rapid development of microbial resistance to conventional antibiotics (1). Antimicrobial peptides are part of the innate immune system widely distributed in nature (2). Many different kinds of antimicrobial peptides have been identified from various sources such as amphibians, insects, mammalians, plants, invertebrates, and prokaryotes (3–8). Active antimicrobial peptides, derived by the processing of large precursors, share common structural features such as a high content of basic amino acid residues and a global distribution of hydrophobic and hydrophilic residues leading to amphipathic a-helical conformations under hydrophobic conditions or bsheet conformations (9). These peptides exhibit potent antimicrobial activities against a broad range of microorganisms, including bacteria, protozoa, fungi, and viruses (5). They have been shown to exert their activities directly through the lipid bilayer of the cell membranes by the formation of multimeric pores or by disrupting the cell membranes (10,11). The most intriguing feature of these molecules is their potential not to fall victim to the same microbial resistance problems as conventional antibiotics, because their mechanism of action is different (1). To date, researchers have been unable to induce bacterial resistance to antimicrobial peptides (1). Despite the activities of these peptides on prokaryotes, they do not induce lysis of erythrocytes or lymphocytes at comparable concentrations (12–14). Therefore, from a drug development point of view, these host-defense peptides may provide a powerful new class of antimicrobial pharmaceuticals. For pharmaceutical applications, a large quantity of antimicrobial peptides needs to be produced economically. Chemical synthesis is not economically practical for peptides more than 10 amino acids in length. Therefore, if developed successfully, a biological expression 53

1046-5928/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

pepa

AP: PEP

54

LEE ET AL.

system would be the most cost effective for mass production of antimicrobial peptides. Recently, several researchers have attempted to develop biological expression systems for the production of antimicrobial peptides. Cecropin A has been expressed in baculovirus expression systems (15,16) and in Escherichia coli (17), an insect defensin from Phormia terranovae in yeast (18), human defensin HNP-1 and CEME (19) in various bacterial expression systems (20), magainin in the erythrocytes of transgenic mice (21), and moricin from Bombyx mori in E. coli (22). However, each of the above systems has its own limitations and is far from satisfactory for mass production. The yields were rather low (15,17,22), proteolysis of the fusion protein occurred (20), or the product was toxic to the host (23). As a new approach, the concatemeric expression of the antimicrobial peptide magainin, from the skin of Xenopus laevis, was attempted (24). However, the expression level of repetitive antimicrobial peptide genes was extremely low. To overcome current biological expression problems, we explored a novel expression method by fusing an acidic peptide to a positively charged antimicrobial peptide to form a neutrally charged peptide, mimicking the natural precursor of an antimicrobial peptide. In this study, the effectiveness of our new approach was demonstrated by the highly increased expression of the antimicrobial peptide buforin II (25) fused to an acidic peptide as tandem repeats in E. coli. MATERIALS AND METHODS

CAGTTTCCGGTGGGCCGTGTGCATCGTCTGCTGCGTAAAATG 3*) and B (5* GGGGCATTTTACGCAGCAGACGATGCACACGGCCCACCGGAAACTGCAGGCCCGCACGGCTGCTACGGGTCATCA 3*) were synthesized, annealed, and then ligated into BbsI-digested pBBS1, resulting in pBBS1-B1. The DNA sequence was confirmed by the dideoxy chain-termination method (28) with Sequenase Version 2.0 (U.S. Biochemicals, Cleveland, OH). To construct the fusion of the buforin II gene and the modified magainin intervening sequence (MMIS) in which cysteine codons flank the magainin intervening sequence (MIS), DNA sequences coding for MIS and buforin II were amplified from pBSX3 kindly provided by Dr. Zasloff (12) with primers C (5* AAAGAAGACGGCCCCTGTGCGATGCAGAAGCAGTA 3*) and D (5* CGGGTCATCAGGGGGCATTCATCTAAATCTTC 3*) and from pBBS1-B1 with primers E (5* GAAGATTTAGATGAATGCCCCCTGATGACCCG 3*) and F (5* TGCATGCCTGCAGGTCGA 3*), respectively. Primers C and D were designed to introduce a cysteine codon on both ends of MIS, and primer D is complementary to primer E. The two PCR products were then purified from an agarose gel and assembled with primers C and F. The product was purified, BbsI digested, and cloned into BbsI-cut pBBS1, resulting in pBBS1-MB1 in which MMIS is followed by the buforin II gene. The schematic representation of the construction of MMIS–buforin II fusion gene is shown in Fig. 1.

Strains, Vectors, and Enzymes

Multimerization of the Peptide Genes Using a Gene Amplification Vector

E. coli strain XL1-Blue (Stratagene, La Jolla, CA) was used as a host for subcloning and E. coli strain BL21(DE3) (Novagen, Madison, WI) for gene expression. E. coli strains were grown in LB medium at 377C and 50 mg/ml ampicillin was added for plasmid-containing strains. pUC19 (New England Biolabs, Beverly, MA) and pBBS1 (24) were used as vectors for subcloning and amplification of peptide genes, respectively, and pET21c (Novagen) for expression. Restriction enzymes and modifying enzymes were purchased from New England Biolabs and used according to the recommendations of the supplier. A miniscale preparation of vector DNA was carried out using the alkaline lysis method (26) and large quantities of vector DNA were prepared by the PEG precipitation method (27). Other recombinant DNA techniques were exploited as described by Maniatis et al. (26) and Sambrook et al. (27).

The DNA fragments encoding buforin II or the acidic peptide–buforin II fusion were tandemly multimerized using the vector pBBS1 (24) as described in Fig. 2. Vectors pBBS1-B1 and pBBS1-MB1, which contain the buforin II gene and MMIS–buforin II fusion gene, respectively, produced monomeric DNA fragments with asymmetric cohesive ends of 5*-CCCC/5*-GGGG upon digestion with BbsI. Purified monomeric fragments were self-ligated for 2 h at 167C and tandemly multimerized by cloning into the BbsI-digested pBBS1. With XL1-Blue transformants, the number of monomers in the vector was determined by cleaving the vector with BamHI and XbaI, whose sites flank the multimer. The orientation of the individual monomers within the vector was determined by digesting the vectors with PstI which cuts a site in the vector and a site in each monomeric fragment.

Construction of Genes Coding for Buforin II and Acidic Peptide–Buforin II Fusions

Expression of the Multimeric Peptide Genes in E. coli

For the construction of the gene encoding buforin II, two complementary deoxyoligonucleotides (oligos) A (5* CCCCTGATGACCCGTAGCAGCCGTGCGGGCCTG-

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

To construct expression vectors having the multimeric peptide genes under the control of the T7 promoter, the BamHI–HindIII fragments containing the multimeric peptide genes were cloned into the vector

pepa

AP: PEP

ACIDIC PEPTIDE-MEDIATED EXPRESSION OF THE ANTIMICROBIAL PEPTIDE

55

pET21c digested with BamHI and HindIII as shown in Fig. 2. E. coli host BL21(DE3) was transformed with the expression vectors and the desired clones were selected. Each transformant was inoculated into 3 ml LB medium supplemented to a final concentration of 50 mg/ml ampicillin and grown at 377C with shaking for 9 to 12 h. Each culture was then diluted 1:100 into fresh LB medium supplemented to a final concentration of 50 mg/ml ampicillin and grown at 377C. At OD600 Å 0.6, IPTG was added to a final concentration of 0.6 mM. The cells were harvested 3 h after induction and wholecell lysates from induced cultures (equivalent OD600) were analyzed by SDS–PAGE (29). Production and Purification of Recombinant Buforin II E. coli cells harboring the expression vector pET21cMB6, which has six copies of the MMIS–buforin II

FIG. 2. Schematic representation of the multimerization of target genes using the gene amplification vector pBBS1. The gene amplification cassette contains two inversely oriented BbsI sites and the same cleavage sequences. The monomeric peptide gene cloned into pBBS1 could be amplified by: (i) excision of the monomeric insert by digestion with BbsI; (ii) isolation of the fragments; (iii) self-ligation of the fragments; and (iv) cloning into the original pBBS1 vector digested with BbsI. B and MB stand for genes coding for buforin II and MMIS–buforin II, respectively, which were constructed as described in the legend to Fig. 1.

FIG. 1. Schematic representation of the construction of MMIS– buforin II fusion gene. The MMIS gene was fused to the buforin II gene by recombinant PCR, as described under Materials and Methods. Charged residues are indicated by / or 0 above the amino acid sequences. The cysteine residues are represented by the C-SH. The buforin II sequence is underlined in the fusion peptide.

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

fusion gene, were cultivated in a 30-L fermentor. At OD600 Å 0.6, IPTG was added to the cultures (20 L) to a final concentration of 0.6 mM. The cells were harvested by centrifugation at 6,000g for 10 min 3 h after induction. After the lysis of cells by sonication, the inclusion bodies were recovered by centrifugation at 10,000g for 30 min at 47C and washed with 50 mM Tris/HCl buffer (pH 8.0) containing 2 mM EDTA. The inclusion bodies were denatured and solubilized in 1 N HCl and 6 M guanidinium chloride and cleaved by incubating with 1 M CNBr at 307C for 20 h. After the cleavage reaction, the insoluble materials were removed by centrifugation at 10,000g for 30 min. The

pepa

AP: PEP

56

LEE ET AL.

supernatant containing the cleaved peptides was dialyzed against 50 mM glycine–NaOH buffer (pH 10.0) and applied to a QAE–Sephadex column equilibrated with the same buffer. The unbound fractions were collected and concentrated by lyophilization. The lyophilized samples were applied to a 3.9 1 300-mm Delta Pak C18 column (Waters Associates, Milford, MA) with a linear gradient of 0% buffer A to 50% buffer A at 1 ml/min for 1 h [buffer A; acetonitrile containing 0.1% (v/v) trifluoroacetic acid]. Each peak was collected and tested for antimicrobial activity after lyophilization. Electrophoretic analysis of recombinant buforin II on tricine–SDS–polyacrylamide gels was carried out by the method of Scha¨gger and von Ja¨gow (30). Total protein concentration was determined by BCA protein assay reagent (Pierce, Rockford, IL) using bovine serum albumin as a standard. The amount of fusion protein in the crude extracts and inclusion bodies and the amount of buforin II after QAE chromatography and HPLC was determined by quantifying the amount in each gel lane by densitometry at 600 nm (Bio/Profile Image Analysis Software Bio-1D, Vilber Lourmat, France). Characterization of Recombinant Buforin II The molecular weight and homogeneity of recombinant buforin II were analyzed by mass spectrometry on a matrix-assisted laser desorption/ionization (MALDI) mass spectrometer (Kartos Kompact MALDI, Manchester, England). Amino acid sequences of the recombinant buforin II were determined by automated Edman degradation performed on an Applied Biosystems (Foster City, CA) gas-phase sequencer (Model 447). Antimicrobial activity was examined during each purification step by the radial diffusion assay on Bacillus subtilis lawn as described by Lehrer et al. (31). A 20-ml B. subtilis culture in midlogarithmic phase was washed with cold 10 mM sodium phosphate buffer (NAPB), pH 7.4, and resuspended in 10 ml of cold NAPB. A volume containing 1 1 106 bacterial CFU was added to 6 ml of underlay agar [10 mM sodium phosphate, 1% (v/v) trypticase soy broth, and 1% (w/v) agarose, pH 6.5] and the mixture was poured into a petri dish. Samples were added directly to the 3-mm wells made on the solidified underlayer agar. After incubation for 3 h at 377C, the underlayer agar was covered with a nutrient-rich top agar overlay and incubated overnight at 377C. Antimicrobial activity was determined by observing the zone of suppression of bacterial growth around the 3-mm wells. Water and natural buforin II were used as negative and positive controls, respectively. The minimal inhibitory concentrations (MIC) of the peptide against several gram-positive and gram-negative bacteria and fungi were determined by incubating approximately 104 –105 CFU/ml of cells with serial dilutions of pep-

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

tides in a 96-well microtiter plate (Nunc F96 microtiter plates, Denmark) as described by Moore et al. (32). RESULTS

Construction and Expression of the Buforin II Gene as Tandem Repeats The 75-nt synthetic oligos, which were synthesized based on the codon usage of E. coli, were hybridized to form a 71-bp DNA fragment with asymmetric cohesive ends, 5*-CCCC/5*-GGGG. The hybrid DNA sequence codes for buforin II and contains two methionine codons flanking the buforin II gene as shown in Fig. 1. The methionine residues were introduced for the cleavage of tandem multimers of buforin II with CNBr in order to produce active buforin II. The hybridized DNA fragment was ligated into the BbsI site of gene amplification vector pBBS1 (24), resulting in pBBS1-B1. DNA fragments encoding buforin II were isolated from pBBS1-B1 by BbsI digestion and then recloned into the BbsI site of pBBS1 after self-ligation, for the construction of tandem multimers as shown in Fig. 2. The clones containing 1, 2, 4, and 6 copies of buforin II were selected (Fig. 3A, lanes 2-5) and named pBBS1-B1, 2, 4, and 6, respectively. The BamHI–HindIII fragments of these clones were cloned into pET21c for the expression of the multimeric buforin II gene (Fig. 2), and the expression level was examined by SDS–PAGE. Both monomers and multimers were poorly expressed in E. coli, as shown in Fig. 3B (lanes 4–6). To find the reason for the decrease in the expression levels, gel retardation and in vitro translation experiments were performed (data not shown). The interaction of buforin II with DNA was confirmed by a gel retardation experiment in which various amounts of buforin II were added to a given amount of DNA, and the retardation of the DNA band was observed. The in vitro coupled transcription–translation experiment, which was performed with the E. coli T7 S30 Extract System for Circular DNA (Promega, Madison, WI), showed that the expression level decreased as the size of the multimers was increased. It seems that the strong positive charge of buforin II caused the decrease in the expression of repetitive buforin II, possibly by interfering with either transcription or translation through interaction with DNAs or RNAs. Therefore, an acidic peptide fusion to buforin II was explored to minimize the interaction between buforin II and nucleic acids by neutralizing its positive charge. Expression of the Fusion Peptide Gene of Buforin II and MMIS as Tandem Repeats To examine the effect of an acidic peptide on the expression of positively charged buforin II, a gene encoding an acidic peptide, the intervening segment of

pepa

AP: PEP

ACIDIC PEPTIDE-MEDIATED EXPRESSION OF THE ANTIMICROBIAL PEPTIDE

FIG. 3. Electrophoretic analysis and expression of multimeric peptide genes. (A) Electrophoretic analysis of the multimeric peptide genes. The number of peptide genes cloned in the gene amplification vector pBBS1 was determined by cleaving the vectors with BamHI / XbaI, whose sites flank the multimer. The digests were electrophoresed on a 2% agarose gel in TBE buffer for 2 h at 10 V/cm and DNA bands were stained with ethidium bromide. Lane M, size markers; lane 1, the pBBS1 vector digested with BamHI / XbaI; lanes 2–5, BamHI / XbaI-digested pBBS1-B1, -B2, -B4, and -B6, which contain 1, 2, 4, or 6 copies of the buforin II monomer, respectively; lanes 6– 9, BamHI / XbaI-digested pBBS1-MB1, -MB2, -MB4, and -MB6, which contain 1, 2, 4, or 6 copies of the MMIS–buforin II fusion monomer, respectively. A detailed explanation under Results. (B) Expression of multimers of buforin II and MMIS–buforin II fusions. Lane M, molecular weight markers; lanes 1 and 2, BL21(DE3) and BL21(DE3) containing pET21c, respectively; lanes 3–6, BL21(DE3) harboring pET21c-B1, -B2, -B4, and -B6, respectively; and lanes 7– 10, BL21(DE3) harboring pET21c-MB1, -MB2, -MB4, and -MB6, respectively. Total cell proteins were analyzed by SDS–PAGE. Samples were applied to the gel after boiling for 5 min. Proteins were stained with Coomassie brilliant blue.

the magainin precursor, which is similar to buforin II in number of the opposite charge and peptide length (12,33), was selected and fused to buforin II. The DNA sequence encoding the MIS, DAEAVGPEAFADEDLDE, was amplified by PCR from the cDNA sequence

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

57

encoding prepromagainin (12) as shown in Fig. 1. The synthetic oligo primers used for the amplification were designed to have a cysteine codon on the 5* end. The cysteine residues flanking MIS were introduced to enhance the interaction between MIS and buforin II in tandem multimers by forming aggregates caused by cysteine residues. Recombinant PCR was used to construct a fusion of the MMIS and buforin II gene as shown in Fig. 1. The fusion was cloned into the BbsI site of pBBS1, resulting in pBBS1-MB1. The 134-bp DNA fragments with asymmetric cohesive ends, 5*CCCC/5*-GGGG, which encode the fused peptide, were isolated from pBBS1-MB1 after BbsI digestion and tandemly multimerized into the BbsI-digested pBBS1 (Fig. 2). The clones, pBBS1-MB1, 2, 4, and 6, each containing 1, 2, 4, or 6 copies of the monomer, respectively, were selected (Fig. 3A, lanes 6–9). The tandem multimers of the fusion gene were expressed in E. coli using the pET21c vector. Expression of the tandem multimers was substantially improved (Fig. 3B, lanes 8–10), whereas expression of fusion monomer was very low. The in vitro translation experiment was also performed with multimers of the fusion gene containing cysteine codons in which high level expression was observed. It seemed that there was a close relationship between the in vitro translation and in vivo expression in our experimental system. The monomer bands of buforin II and fusion peptide were not shown on the gel because they ran out of the gel (Fig. 3B, lanes 3 and 7). With hexamers, more than 50% of the total proteins, which were expressed as inclusion bodies, were the fusion proteins. In addition, the expression level did not decrease as the size of the multimers was increased. Purification and Characterization of Recombinant Buforin II The fusion protein was expressed as inclusion bodies in E. coli BL21(DE3). The inclusion bodies were solubilized and cleaved by adding 1 M CNBr in 1 N HCl and 6 M guanidinium chloride. After CNBr cleavage, samples were subjected to QAE–Sephadex anion-exchange chromatography and reverse-phase HPLC, producing homogeneously pure buforin II as shown in Fig. 4A. About 107 mg of pure buforin II was obtained after QAE–Sephadex anion-exchange chromatography from 1 L of E. coli culture with over 90% purity which was estimated based on the results of MALDI-MS (Table 1). The molecular mass was 2512 Da, which was confirmed by MALDI-MS as shown in Fig. 4B (inset). The amino acid sequence of recombinant buforin II was identical to that of natural buforin II, except for an additional homoserine at the C-terminus. The purified recombinant buforin II was tested for its antimicrobial activity by determining MIC against selected microorganisms. Recombinant buforin II had an antimicrobial

pepa

AP: PEP

58

LEE ET AL. TABLE 1

Purification of Recombinant Buforin II from E. coli

Purification step Crude extracts d Inclusion bodies QAE chromatography (after CNBr cleavage) HPLC

Total protein (mg) a

Protein of interest (mg) b

Yield (%) c

1290.0 998.1

670.8 665.7

100 99.2

124.9 107.0

119.0 107.0

17.7 15.6

a

Total protein concentration was determined by BCA protein assay (Pierce, Rockford, IL) using bovine serum albumin as a standard. b The amount of fusion protein in the crude extracts and inclusion bodies and the amount of buforin II after QAE chromatography and HPLC were determined by quantifying the amount in each gel lane by densitometry at 600 nm (Bio/Profile Image Analysis Software Bio1D, Vilber Lourmat, France). c The purification fold and yield are calculated based on the amount of protein of interest. d The starting material was crude extracts from the lysis of 1 L of induced E. coli culture as described under Materials and Methods.

FIG. 4. Purification of recombinant buforin II from multimers expressed by BL21(DE3) containing pET21c-MB6. (A) SDS–PAGE analysis of purified recombinant buforin II. Lanes M1 and M2, molecular weight markers. Lanes 1 and 2, total cell proteins before and after induction, respectively; lanes 3 and 4, inclusion bodies isolated from total cell proteins and solubilized inclusion bodies cleaved by CNBr, respectively; lanes 5 and 6, unbound fractions on QAE–Sephadex chromatography and recombinant buforin II purified by HPLC, respectively; lane 7, the natural buforin II. (B) Purification of recombinant buforin II by reverse-phase HPLC of unbound fractions from QAE–Sephadex chromatography and determination of purity by MALDI-MS (inset).

activity identical to that of natural buforin II (Table 2). The additional homoserine residue derived from a methionine residue after CNBr cleavage does not interfere with the antimicrobial activity of recombinant buforin II. DISCUSSION

The need for large quantities of antimicrobial peptides has prompted investigations into developing biologically efficient production methods. Not a single method has been developed so far without intrinsic limitations such as poor expression, instability of the expressed proteins or toxic effects to the host. Therefore, we attempted to develop a new method for the mass production of antimicrobial peptides, thereby accelerating their practical applications. The expression of bu-

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

forin II as a concatemeric multimer was explored first, because the recombinant concatemer could avoid a suicide situation by masking the intrinsic antimicrobial activity and the susceptibility of the peptides to proteolytic degradation (34). In addition, the multimers could be converted quantitatively into monomeric peptides, resulting in a high yield compared with monomeric ex-

TABLE 2

Comparison of the Antimicrobial Activities of Recombinant Buforin II, Natural Buforin II, and Magainin II Minimal inhibitory concentration (mg/ml) a

Microorganism Gram positive Bacillus subtilis Staphylococcus aureus Streptococcus mutans Streptococcus pneumoniae Gram negative Escherichia coli Salmonella typhimurium Serratia sp. Pseudomonas putida Fungi Candida albicans Cryptococcus neoformans Saccharomyces cerevisiae a

Recombinant buforin II Buforin II Magainin II

2 4 2 4

2 4 2 4

50 50 100 50

4 1 4 2

4 1 4 2

100 25 50 50

1 1 1

1 1 1

25 12 25

Each MIC was determined from two independent experiments performed in duplicate.

pepa

AP: PEP

ACIDIC PEPTIDE-MEDIATED EXPRESSION OF THE ANTIMICROBIAL PEPTIDE

pression (24). However, our first attempt to produce buforin II as a polypeptide was unsuccessful. The multimers were very poorly expressed even with a strong T7 promoter. A similar result was also observed with the multimeric expression of the magainin gene (24). In vitro translation and gel retardation experiments showed that the low expression level of multimeric buforin genes was due to the interactions of multimeric peptides with DNAs and RNAs in E. coli. These interactions seemed to be ascribed to the basic nature of antimicrobial peptides, which might inhibit transcription and translation by interacting with DNAs or RNAs (35). Since buforin II is a highly positively charged peptide, it has an increased chance of interacting with nucleic acids in the host when expressed in multimeric forms. Therefore, the neutralization of its basic charge was needed to reduce this interaction and increase the expression of the multimeric peptides. Many of the naturally occurring basic peptides and proteins, including antimicrobial peptides (12,36–39) and the major basic protein of the human eosinophil granule (40), are synthesized as precursors in which acidic proparts electrically neutralize the basic peptides. One of the proposed roles of these acidic peptides is that they mask the toxicity of the mature basic peptides in a charge-dependent manner by counterbalancing the overall positive charge and protect the host cell from the cytolytic effects of high concentrations of peptides while they are processed through the endoplasmic reticulum to their sequestered sites in the granules (23). The proteolysis of defensin was overcome by the addition of a prepropiece of defensin between the fusion partner and defensin (20). This might be resulted from the complete protection of the fusion protein from proteolytic degradation, presumably due to secondary structure formed between the anionic prepropiece and the cationic defensin sequence (41). We wanted to mimic a neutral precursor present in nature for the efficient expression of antimicrobial peptide buforin II as tandem multimers. We looked for an acidic peptide whose length and opposite charge would complement those of buforin II. MIS was an ideal peptide considering its length, charge balance, and distribution. As a first step, to examine the effectiveness of MIS on neutralization of the positive charge of buforin II, we attempted to express the magainin precursor itself which has five MIS, as an acidic peptide bridge between magainins. However, the expression of the magainin precursor gene was unsuccessful. Therefore, we modified the MIS such that one cysteine residue was located on both the N- and C-terminal ends of MIS, naming it MMIS. With the fusion of MMIS and buforin II, the expression level substantially increased except the monomer expression, and there was no decrease in the expression even with increasing the size of the multimers. Our results indicate that the sulfur hy-

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

59

droxyl groups on an acidic peptide, along with adequate distribution of the negative charge, are important in neutralizing the positive charge of an antimicrobial peptide for the efficient expression in multimeric forms. The effect of sulfhydryl groups of the acidic peptide on monomeric expression of the fusion peptide was negligible. It seems that the aggregate formed by the monomers was not stable enough for efficient neutralization of the basic peptide by the acidic peptide because there are fewer cysteine residues present on the monomers compared to the multimers. The multimeric MMIS– buforin II fusion protein was produced as inclusion bodies. This has an advantage over soluble forms, because it can be easily purified and buforin II need not be refolded to be active. The purified inclusion bodies were easily cleaved with CNBr under denaturing conditions, producing a mixture of buforin II and MMIS, from which buforin II could be easily purified to over 90% homogeneity by passing it through an anion-exchange column in a single step. These results clearly open a door to the mass production of antimicrobial peptides, which has been a bottleneck to the wide application of antimicrobial peptides up to now. ACKNOWLEDGMENTS This work was partially supported by the grants from Korea Science and Engineering Foundation (KOSEF, 951-0502-046-2), KOSEF through Research Center for New Bio-Materials in Agriculture (RCNBMA) at Seoul National University, and Special Grants Research Program/High-Technology Development Project for Agriculture, Forestry, and Fisheries (SGRP/HTDP) in Korea.

REFERENCES 1. Kelley, K. J. (1996) Using host defenses to fight infectious diseases. Nature Biotechnol. 14, 587–590. 2. Gabay, J. E. (1994) Ubiquitous natural antibiotics. Science 264, 373–374. 3. Barra, D., and Simmaco, M. (1995) Amphibian skin: A promising resource for antimicrobial peptides. Trends Biotechnol. 13, 205– 209. 4. Hoffmann, J. A. (1995) Innate immunity of insects. Curr. Opin. Immunol. 7,4–10. 5. Lehrer, R. I., Lichtenstein, A. K., and Ganz, T. (1993) Defensins: Antimicrobial and cytotoxic peptides of mammalian cells. Annu. Rev. Immunol. 11, 105–128. 6. Broekaert, W. F., Terras, F. R., Cammue, B. P., and Osborn, R. W. (1995) Plant defensins: Novel antimicrobial peptides as components of the host defense system. Plant Physiol. 108, 1353–1358. 7. Nakamura, T., Furunaka, H., Miyata, T., Tokunaga, F., Muta, T., and Iwanaga, S. (1988) Tachyplesin, a class of antimicrobial peptide from the hemocytes of the Horseshoe Crab (Tachypleus tridentatus). J. Biol. Chem. 263, 16709–16713. 8. Jack, R. W., Tagg, J. R., and Ray, B. (1995) Bacteriocins of grampositive bacteria. Microbiol. Rev. 59, 171–200. 9. Saberwal, G., and Nagaraj, R. (1994) Cell-lytic and antibacterial peptides that act by perturbing the barrier function of membranes: facets of their conformational features, structure–func-

pepa

AP: PEP

60

LEE ET AL. tion correlations and membrane-perturbing abilities. Biochim. Biophys. Acta 1197, 109–131.

10. Wimley, W. C., Selsted, M. E., and White, S. H. (1994) Interactions between human defensins and lipid bilayers: evidence for formation of multimeric pores. Protein Sci. 3, 1362–1373.

25.

11. Matsuzaki, K., Murase, O., Fujii, N., and Miyajima, K. (1995) Translocation of a channel-forming antimicrobial peptide, magainin 2, across lipid bilayers by forming a pore. Biochemistry 34, 6521–6526.

26.

27.

12. Zasloff, M. (1987) Magainin, a class of antimicrobial peptides from Xenopus skin: Isolation, characterization of two active forms, and partial cDNA sequence of a precursor. Proc. Natl. Acad. Sci. USA 84, 5449–5453.

28.

13. Chen, H. C., Brown, J. H., Morell, J. L., and Huang, C. M. (1988) Synthetic magainin analogues with improved antimicrobial activity. FEBS Lett. 236, 462–466.

29.

14. Zasloff, M. (1992) Antibiotic peptides as mediator of innate immunity. Curr. Opin. Immunol. 4, 3–7. 15. Andersons, D., Engstrom, A., Josephson, S., Hansson, L., and Steiner, H. (1991) Biologically active and amidated cecropin produced in a baculovirus expression system from a fusion construct containing the antibody-binding part of protein A. Biochem. J. 280, 219–224.

30.

31.

32.

16. Hellers, M., Gunne, H., and Steiner, H. (1991) Expression and post-translational processing of preprocecropin A using a baculovirus vector. Eur. J. Biochem.199, 435–439. 17. Callaway, J. E., Lai, J., Haselbeck, B., Baltaian, M., Bonnesen, S. P., Weickmann, J., Wilcox, G., and Lei, S. P. (1993) Modification of the C terminus of cecropin is essential for broad-spectrum antimicrobial activity. Antimicrob. Agents Chemother. 37, 1614– 1619.

33.

18. Reichhart, J. M., Petit, I., Legrain, M., Dimarcq, J. L., Keppi, E., Lecocq, J. P., Hoffmann, J. A., and Achstetter, T. (1992) Expression and secretion in yeast of active insect defensin, an inducible antimicrobial peptide from the fleshfly Phormia terranovae. Invertebrate Reprod. Develop. 21, 15–24.

34.

35.

19. Wade, D., Andrew, D., Mitchell, S. A., Silveira, A. M. V., Boman, A., Boman, H. G., and Merrifield, R. B. (1992) Antimicrobial peptides designed as analogs or hybrids of cecropins and melittin. Int. J. Peptide Protein Res. 40, 429–436.

36.

20. Piers, K. L., Brown, M. H., and Hancock, R. E. W. (1993) Recombinant DNA procedures for producing small antimicrobial cationic peptides in bacteria. Gene 134, 7–13.

37.

21. Sharma, A., Khoury-Christianson, A. M., White, S. P., Dhanjal, N. K., Huang, W., Paulhiac, C., Friedman, E. J., Manjula, B. N., and Kumar, R. (1994) High-efficiency synthesis of human a-endorphin and magainin in the erythrocytes of transgenic mice: A production system for therapeutic peptides. Proc. Natl. Acad. Sci. USA 91, 9337–9341.

38.

39.

22. Hara, S., and Yamakawa, M. (1996) Production in Escherichia coli of moricin, a novel type antibacterial peptide from the silkworm, Bombyx mori. Biochem. Biophys. Res. Commun. 220, 664– 669.

40.

23. Chan, R. Y. K., Palfree, R. G. E., Congote, L. F., and Solomon, S. (1994) Development of a novel type of cloning vector for suicide selection of recombinants. DNA Cell Biol. 13, 311–319.

41.

24. Lee, J. H., Skowron, P. M., Rutkowska, S. M., Hong, S. S., and Kim, S. C. (1996) Sequential amplification of cloned DNA as tan-

AID

PEP 0814

/

6q17h$$201

01-16-98 12:01:52

dem multimers using class-IIS restriction enzymes. Genetic Anal. Biomol. Eng. 13, 139–145. Park, C. B., Kim, M. S., and Kim, S. C. (1996) A novel antimicrobial peptide from Bufo bufo gargarizans. Biochem. Biophys. Res. Commun. 218, 408–413. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) ‘‘Molecular Cloning: A Laboratory Manual,’’ Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) ‘‘Molecular Cloning: A Laboratory Manual,’’ 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. Laemmli, U. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Scha¨gger, H., and von Ja¨gow, G. (1987) Tricine-sodium dodecyl sulfate–polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166, 368–379. Lehrer, R. I., Rosenman, M., Harwig, S. S. S. L., Jackson, R., and Eisenhauer, P. (1991) Ultrasensitive assays for endogenous antimicrobial polypeptides. J. Immunol. Methods 137, 167–173. Moore, K. S., Bevins, C. L., Brasseur, M. M., Tomassini, N., Turner, K., Eck, H., and Zasloff, M. (1991) Antimicrobial peptides in the stomach of Xenopus laevis. J. Biol. Chem. 266, 19851–19857. Terry, A. S., Poulter, L., Williams, D. H., Nutkins, J. C., Giovannini, M. G., Moore, C. H., and Gibson, B. W. (1988) The cDNA sequence coding for prepro-PGS (prepro-magainins) and aspects of the processing of this prepro-polypeptide. J. Biol. Chem. 263, 5745–5751. Shen, S. H. (1984) Multiple joined genes prevent product degradation in Escherichia coli. Proc. Natl. Acad. Sci. USA 81, 4627– 4631. Miller, K. W., Evans, R. J., Eisenberg, S. P., and Thompson, R. C. (1989) Secretory leukocyte protease inhibitor binding to mRNA and DNA as a possible cause of toxicity to Escherichia coli. J. Bacteriol. 171, 2166–2172. Hunt, L. T., and Barker, W. C. (1988) Relationship of promagainin to three other prohormones from the skin of Xenopus laevis: a different perspective. FEBS Lett. 233, 282–288. Nutkins, J. C., and Williams, D. H. (1989) Identification of highly acidic peptides from processing of the skin prepropeptides of Xenopus laevis. Eur. J. Biochem. 181, 97–102. Michaelson, D., Rayner, J., Couto, M., and Ganz, T. (1992) Cationic defensins from charge-neutralized propeptides: a mechanism for avoiding leukocyte autotoxicity? J. Leukocyte Biol. 51, 634–639. Liu, L., and Ganz, T. (1995) The pro region of human neutrophil defensin contains a motif that is essential for normal subcellular sorting. Blood 85, 1095–1103. Barker, R. L., Gundel, R. H., Gleich, G. J., Checkel, J. L., Loegering, D. A., Pease, L. R., and Hamann, K. J. (1991) Acidic polyamino acids inhibit human eosinophil granule major basic protein toxicity. J. Clin. Invest. 88, 798–805. Saito, Y., Ishii, Y., Koyama, S., Tsuji, K., Yamada, H., Terai, T., Kobayashi, M., Ono, T., Niwa, M., Ueda, I., and Kikuchi, H. (1987) Bacterial synthesis of recombinant a-human atrial natriuretic polypeptide. J. Biochem. 102, 111–122.

pepa

AP: PEP