Mass production method of antimicrobial peptide and DNA construct and expression system thereof

[0002] The present invention relates to the recombinant DNA technology. The present invention also relates to the mass-production of antimicrobial materials from microorganisms and a DNA construct and vector system. Biologically active peptide (antimicrobial peptide hereinafter) has little chance to develop resistance since the antimicrobial peptides show activity by a mechanism that is totally different from that of conventional antibiotics which have a serious problem of developing resistance. Therefore, the antimicrobial peptides have a high industrial applicability in the fields of pharmaceutics and the food industry.

[0003] The main obstacle in the industrial use of the antimicrobial peptide, however, is the difficulty in economical mass-production of the antimicrobial peptides. For instance, the production of the antimicrobial peptides by chemical synthesis is not economical. Also, there have been attempts to produce antimicrobial peptides by genetic engineering using microorganisms, in this case, however, the expression levels of the antimicrobial peptides are very low.

[0004] U.S. Pat. No. 5,206,154 provides a DNA construct which comprises a polypeptide gene which is capable of suppressing the bactericidal effect of cecropin, and a cecropin gene fused to the polypeptide gene. An example of such polypeptide disclosed in the patent is the araB gene.

[0005] U.S. Pat. No. 5,593,866 provides a method for a microbial production of a cationic antimicrobial peptide, wherein the cationic peptides is expressed as a fusion to an anionic peptide to avoid degradation by a bacterial protease.

[0006] The present invention provides a DNA construct to mass-produce a antimicrobial peptides. The present invention also provides a DNA construct that can produce and recover antimicrobial peptides effectively from microorganisms.

[0007] Also, the present invention provides gene multimers that can increase the efficiency of expression, separation and purification of desired peptides and the construction method of such construct.

[0008] Further, the present invention provides an expression vector to mass-produce antimicrobial peptides from microorganisms.

[0009] Further, the present invention provides a method to mass-produce antimicrobial peptides form microorganisms.

[0010] FIGS. 1A, 1B, 1C, 1D, 1E and 1F, herein referred to collectively as FIG. 1, are a nucleotide sequence coding for an antimicrobial peptide of the present invention.

[0011] FIGS. 2A, 2B, 2C and 2D, herein referred to collectively as FIG. 2, are a nucleotide sequence coding for a fusion partner.

[0012] FIGS. 3A and 3B, herein referred to collectively as FIG. 3, are a scheme of a fusion method between the fusion partner and the MSI-344 gene by generating a sequence encoding producing CNBr cleavage site.

[0013] FIGS. 4A and 4B, herein referred to collectively as FIG. 4, are a scheme of a fusion method between the fusion partner and the MSI-344 gene by generating a sequence encoding producing hydroxylamine cleavage site.

[0014] FIG. 5 is a scheme of the construction of the transcriptionally fused multimer.

[0015] FIGS. 6A and 6B, herein referred to collectively as FIG. 6, are a scheme of the construction of the pGNX2 vector.

[0016] FIG. 7 is a scheme of the construction of the pT7K2.1 vector.

[0017] FIG. 8 is a scheme of the construction of the pGNX3 vector.

[0018] FIG. 9 is the pGNX4 vector.

[0019] FIG. 10 is a scheme of the construction of the pGNX5 vector.

[0020] FIG. 11 is a SDS-PAGE electrophoretic analysis of the lysates of the transformants expressing MSI-344 by an induction with lactose or IPTG.

[0021] FIG. 12 is a SDS-PAGE electrophoretic analysis of MSI-344 expression with various vectors.

[0022] FIG. 13ais a SDS-PAGE electrophoretic analysis of the lysates of the transformants expressing various antimicrobial peptides by induction with lactose.

[0023] FIG. 13bis a SDS-PAGE electrophoretic analysis of the lysates of the transformants expressing various antimicrobial peptides by an induction with lactose.

[0024] FIGS. 14a, 14b, 14cand 14dare SDS-PAGE electrophoretic analyses of the lysates of the transformants expressing various antimicrobial peptides by an induction with lactose.

[0025] FIG. 15 is a SDS-PAGE electrophoretic analysis of the lysates of the transformants expressing the monomer, dimer and tetramer of the fusion genes.

[0026] The present invention relates to a DNA construct for mass-producing antimicrobial peptides effectively in E. colior other prokaryotes.

[0027] One of the essential conditions for mass production of the antimicrobial peptides from microorganisms is to efficiently neutralize the toxicity of the antimicrobial peptides against the microorganisms. To this end, the present invention provides a DNA construct in which a whole gene, partial or derivatives of the purF gene (glutamine pyrophosphoribosyl pyrophosphate amidotransferase; Genbank No.: X12423) (Tso et al., J. Biol. Chem., 257: 3525, 1982, Makaroff et al., J. Bio. Chem., 258: 10586, 1983) is fused as a fusion partner to the gene coding for antimicrobial peptides.

[0028] The derivatives of purF gene used as a fusion partner in the DNA construct according to the present invention allows mass-production of the antimicrobial peptides as a fused polypeptide with purF derivatives in Escherichia coliwithout killing the host cells. Therefore, it is possible to mass-produce the desired antimicrobial peptides from the host microorganisms using a strong expression system since they are not lethal to the host cell. In the case of using a fusion partner according to the present invention to express peptides, it is possible to cleave and separate the antimicrobial peptides from the fusion protein by using a protease or other chemicals. To achieve this, for instance it is possible to insert a DNA sequence between the fusion partner and antimicrobial peptide genes encoding the cleavage site for proteases such as Factor Xa or enterokinase or chemicals such as CNBr or hydroxylamine.

[0029] For instance, to provide a CNBr cleavage site, restriction enzyme site containing Met codon (ATG) with correct leading frame such as Afl III, Bsm I, BspH I, BspLU11 I, Nco I, Nde I, Nsi I, Ppu10 I, Sph I, Sty I, or their isoschizomers could be inserted into the 3′ end of the fusion partner. It is possible to make in-frame fusion of the fusion partner and the gene coding for antimicrobial peptide by inserting the restriction enzyme site into the 5 end of the gene coding for antimicrobial peptide that produces a compatible end to the enzyme site of the fusion partner.

[0030] It is also possible to insert a DNA sequence coding for Asn-Gly between the fusion partner and antimicrobial peptide genes. For instance, two genes can be fused by the following method. After inserting a restriction enzyme or isoschizomer site containing an Asn codon with correct reading frame at the 3′ end of the fusion partner, the fusion partner is cleaved by the enzyme. At the 5′ end of the gene coding for antimicrobial peptide, a restriction enzyme site containing a Gly codon with correct reading frame that produces a compatible or blunt end with the corresponding site of the fusion partner is inserted and cleaved with the corresponding enzyme. The two cleaved DNA fragments may be connected to produce the fused gene. The genetic construct according to the present invention may be inserted into the host cell by cloning into any kind of expression vector, that is conventionally used in this field such as plasmid, virus or other vehicles that can be used to insert or incorporate the structural genes.

[0031] The present invention relates to a multimer that can increase the expression level by increasing the copy number of the gene of the required product and which can be separated and purified conveniently and the preparation method thereof.

[0032] The multimer according to the present invention is constructed by the following units.

[0033] 1) A first restriction enzyme site that can generate an initiation codon Met, 2) a structural gene, 3) a ribosome binding site (RBS), and 4) a second restriction enzyme site generating a cohesive end which can be in-frame fused to the cohesive end generated by the first restriction enzyme and which can generate the initiation codon. Here, the stop codon and the RBS of the structural gene may overlap by ca. 2 bp or may be separated as far as 500 bp. The distance between the RBS and the second restriction enzyme site that can generate the initiation codon may be ca. 5 to 30 bp. The 3′ and 5′ ends of the multimer may be cleaved by the first or second restriction enzyme, respectively.

[0034] The multimer according to the present invention may be prepared by a variety of techniques known in the field of genetic engineering. One of the examples of such preparation method is given below.

[0035] After cleaving the units of a gene given above by the first and second restriction enzymes, the cleaved units is connected to produce a mixture containing multimers that include each unit with the same direction and multimers that have more than one unit with reverse direction. Since the multimers that contain more than one unit with reverse direction will have the first or second restriction enzyme site regenerated at the connection site, the multimer mixture may be cleaved simultaneously by the first and second restriction enzymes and separated by agarose gel electrophoresis, for instance, to separate the multimers those have units with the same direction only. The multimer according to the present invention is a transcriptionally fused multimer. This means that the repeated genes are transcribed into a single mRNA, but the gene expression product is not connected. In other words, the multimer is translated into many copies of a single product, In the case of the conventional translationally fused multimer, the desired product is present as a concatemer in a single polynucleotide, and an additional cleavage process is necessary to obtain the desired active product in case that the expression product is a fusion protein, it requires a greater amount of reagent to cleave only with lower efficiency when compared to the transcriptionally fused multimer. Compared to the translationally fused multimer, the expressed multimer of the present invention does not require additional cleavage processes or in the case it requires cleavage processes such as fused proteins, the amount of the reagent for the cleavage may be reduced since the number of peptide bonds to be cleaved per mole of the fused peptide is relatively smaller than the translationally fused multimer.

[0036] The multimer of the present invention may increase the gene expression in the host cell, have advantages in cleaving and purifying the desired product, and express in the host more efficiently when compared to the monomer. The multimer and the preparation method thereof are not limited in preparing peptides or fusion peptides. It can be widely applicable in expressing the unfused or fused gene coding for enzymes, hormones and antimicrobial polypeptides in microorganism.

[0037] Therefore, it is desirable to produce the DNA construct of the present invention in the form of transcriptionally fused multimer. In the case of preparing the DNA construct of the present invention in the form of transcriptionally fused multimer, it is advantageous to cleave and purify the products, and the multimer may be expressed in the host more efficiently than the monomer.

[0038] The present invention also relates to the expression vector that may induce the expression of foreign genes by lactose which is more economical than IPTG.

[0039] The expression vector according to the present invention is composed of high copy number replication origin, strong promoter and structural gene, and does not include lacl^qgene.

[0040] The replication origin may be colE1 or p15A in the present invention. Examples of the strong promoters include tac, trc, trp, T7Φ10, P_L, other inducible or constitutive promoters in the microorganisms. Additionally, a selection marker gene that may be used to select the transformants of the vector may be included. These marker genes include antibiotic resistant genes against antibiotics such as ampicillin, kanamycin, tetracyline and chloramphenicol, or the genes that complement the auxotrophy of the host. Gene expression using the expression vector according to the present invention can be induced efficiently by adding lactose instead of IPTG preferably by adding IPTG and lactose simultaneously.

[0041] As an example, after transforming the plasmid containing the structural gene into the host cells, transformants are primary-cultured for 5 to 18 hours at 30-37° C. in a culture medium that include 50-300 μg/ml kanamycin. Afterwards, they are diluted to 1% (v/v) in a fresh media and cultured at 30-37° C. To induce the expression, 0.01 mM-10 mM IPTG is added when the OD₆₀₀reaches 0.2-2 in case of IPTG induction, or 0.2-2% lactose is added when the OD₆₀₀reaches 0.2-2, or at the time of inoculation in the case of lactose induction. IPTG and lactose can be used simultaneously with a significantly reduced amount of IPTG. Additionally, it is desirable to include a transcriptional terminator in the expression vector according to the present invention.

[0042] It is possible to obtain the expression product as an inclusion bodies using the expression vector of the present invention. This property is useful in producing a product lethal to the host.

[0043] A vector containing a structural gene of the present invention may be transformed into microorganisms by using conventional methods used in the fields of the present invention. For instance, the transformation may be achieved by CaCl₂method or by physical methods such as electrophoration or microinjection into prokaryotic cells such as E. coli. There is no specific limitation for the host. For instance E. colistrain may be selected form BL21(DE3), BLR(DE3), B834(DE3), AD494(DE3), JM109(DE3), HMS174(DE3), UT400(DE3) and UT5600(DE3). Culture medium could be selected from LB, M9, M9CA, and R according to the characteristics of the host or transformants cells. Growth factors may be added to the media depending on the host requirements.

[0044] LB medium (bacto-tryptone 10 g/l, yeast extract 5 g/l, NaCl 10 g/l)

[0045] M9 medium (Na₂PO₄7H₂O 12.8 g/l, KH₂PO₄3.0 g/l, NaCl 0.5 g/l, NH₄Cl 1 g/l, glucose 4 g/l, MgSO₄2 mM, CaCl₂0.1 mM)

[0046] M9CA medium (M9 medium+0.2% casamino acid)

[0047] R medium (Reisenberg medium; KH₂PO4 13.3 g/l, (NH₄)₂PO₄4.0 g/l, citric acid 0.17 g/l, MgSO₄7H₂O 0.22 g/l, glucose 20 g/l, trace element solution 10 ml/l)

[0048] Trace element solution (ferric citrate 7.3 g/l, CoCl₂6H₂O 0.5 g/l, MnCl₂4H₂O 3.2 g/l, CuCl₂2H₂O 0.3 g/l, H₃BO₃0.7 g/l, NaMoO₄2H₂O 1.68 g/l, Thiamin HCl 0.5 g/l, EDTA 1 g/l)

[0049] The invention will be further illustrated in detail by the following examples. It will be apparent to those having conventional knowledge in this field that these examples are given only to explain the present invention more clearly, but the invention is not limited to the examples given below.

[0050] Two different MSI-344 genes were synthesized by the PCR method to express MSI-344 gene efficiently in E. coliand to ease the gene manipulation (FIG. 1). Template for PCR was pNH18a-MBP-MSI-78 described in Korean patent application 97-29426. Sequence (a) (SEQ ID NO. 55) was synthesized using primers No. 1 (SEQ ID NO. 1) and No. 2 (SEQ ID NO. 2) in Table 1 which was designed to separate MSI-344 by CNBr cleavage from the fusion peptide, and Sequence (b) (SEQ ID NO. 57) was synthesized using primers No. 3 (SEQ ID NO. 3) and No. 4 (SEQ ID NO. 4) in Table 1 which was designed to be cleaved by hydroxylamine. To subclone MSI-344 gene with correct reading frame into the expression vector, Nde1 (Sequence (a)) and SmaI (Sequence (b)) sites were inserted in front of MSI-344 gene and stop codons TAA and TGA were inserted behind the MSI-344 gene. Also to construct the transcriptional multimer, a ribosome binding site that overlaps 1 base pair with the stop codon and Ase I site were inserted. These two MSI-344 genes were cloned into pCR2.1 vector (Invitrogen, USA) to prepare vector pCRMSI containing sequence (a) and vector pCRMSI' containing sequence (b).

[0051] The antimicrobial peptide genes in FIG. 1 were prepared by annealing chemically synthesized oligonucleotides (Table 1) or by performing PCR after annealing. In the case of Apidaecin I (SEQ ID NO. 41), Indolicidin (SEQ ID NO. 51), and Tachyplesin I (SEQ ID NO. 61), DNA sequence was based on the amino acid sequence of a peptide (Maloy and Kark, Peptide Science, 37: 105, 1995) and the gene was chemically synthesized by using codons that can maximize the expression level in E. coli. In the case of Bombinin (SEQ ID NO. 43), CPF1 (SEQ ID NO. 45), Drosocin (SEQ ID NO. 47), Melittin (SEQ ID NO. 53), HNP-I (SEQ ID NO. 49), PGQ (SEQ ID NO. 59), and XPF (SEQ ID NO. 63), the N- and C-terminal oligonucleotides which were designed to anneal to each other by 8-10 bp overlaps, were synthesized and the peptide gene was synthesized by PCR after annealing two oligonucleotides. The characteristics of each antimicrobial peptide are listed in Table 2.

[0052] To use as a fusion partner, purF derivatives shown in FIG. 2 were obtained from the chromosomes of E. coliand Bacillus subtilisusing PCR. The fusion partner F was prepared by CNBr cleavage, and F′, F5 and BF by for hydroxylamine cleavage. F3 and F4 were prepared as two different forms; one for CNBr cleavage (F3(CB), F4(CB)), and another for hydroxylamine cleavage (F3(HA), F4(HA), F4a(HA)). Fusion partners F, F′, F3(HA), F3(CB), F4(HA), F4a(HA), F4a(CB), F5, BF are indicated in sequences No. 1-9, respectively.

[0053] 1) purF derivative F (SEQ ID NO. 73)

[0054] The derivative is a coding for 61 amino acid from the N-terminus of the E. colipurF protein (FIG. 2). Nde I site including start codon Met was inserted at the 5′ end, and Nde I site including Met codon that encodes cleavage site for CNBr was inserted at the 3′ end.

[0055] 2) purF derivative F′ (SEQ ID NO. 75)

[0056] To remove the internal hydroxylamine cleavage site, the 49^thglycine residue (GGG) was substituted with alanine (GCG, see FIG. 2) by site-directed mutagenesis using primer #36 in Table 1, and Ssp I site containing AAT coding for asparagine was added after alanine codon (number 57) by PCR to form a hydroxylamine cleavage site.

[0057] 3) purF derivative F3

[0058] The 49^thglycine residue was substituted with alanine as in F′. Asparagine at the 58th residue was substituted with alanine and alanine-asparagine was added after the 59th histidine (F3(HA)) (SEQ ID NO. 77). In case of F3 for CNBr cleavage (F3(CB)) (SEQ ID NO. 79), a DNA sequence that codes for Met and includes BspLU11I site was added after histidine at the 59th residue.

[0059] 4) PurF derivative F4

[0060] This derivative is composed of 159 amino acid residues from the N-terminus of the purF protein. There exists two hydroxylamine sites in wild-type purF protein. To remove these sites, the 102nd asparagine codon (AAC) was substituted with leucine codon (CTC, underlined in Table 2) by site-directed mutagenesis with primer #37 (Table 1) to form F4(HA) (SEQ ID NO. 81). F4a(HA) (SEQ ID NO. 83) was prepared by double substitution of the 49th glycine with alanine and the 102^ndasparagine with leucine. In the case of F4(HA) and F4a(HA) for hydroxylamine cleavage, the SspI site including asparagine codon was added at the 3′ end. In the case of F4a(CB) (SEQ ID NO. 85) for CNBr cleavage, BspLU11 I site including Met codon was added at the 3′ end.

[0061] 5) purF derivative F5 (SEQ ID NO. 87)

[0062] This derivative composed of a sequence from the 60^thmethionine to the 148^thaspartic acid of the purF protein, and Ssp I site was added at the 3′ end.

[0063] 6) purF derivative BF (SEQ ID NO. 89)

[0064] BF is a purF derivative of B. subtilisand includes 43 amino acid residues and Ssp I site coding for Asn at the 3′ end.

[0065] Among the peptide genes prepared in Example 1, the genes encoding peptide that contains glycine at the first amino acid were fused to fusion partners for the hydroxylamine cleavage, F4a(HA), F5 and BF. Other peptides (HNP-I, Indolicidin, Tachyplesin) were fused to the fusion partners for the CNBr cleavage, F, F3(CB) and F4a(CB) (Table 3).

[0066] A method of fusion between the fusion partner and the gene coding for an antimicrobial peptide while producing the CNBr cleavage site (Met) or hydroxylamine cleavage site (Asn-Gly) is shown in FIGS. 3 and 4, respectively. In the case of fusion with fusion partner F for CNBr cleavage, the fusion partner and the MSI-344 gene were fused using the Nde I site to produce DNA construct FM (FIG. 3a). In case of fusion with F3(CB) or F4(CB), the peptide genes are chemically synthesized and fused to 3′ end BspLU11 I site of the fusion partner by complementary 5′ Nco I site for HNP-I, and 5′ BspLu11 I site for indolicidin and tachyplesin, respectively.

[0067] The fusion with the fusion partner for hydroxylamine cleavage (F′, F3(HA), F4(HA), F4a(HA), F5, BF) was carried out by cleaving the fusion partner with Ssp I and MSI-344 by Sma I, and connecting these DNA fragments to generate Asn-Gly site for the hydroxylamine cleavage. In the case of the genes for Apidaecin I, Bombinin, CPF1, Drosocin, Melittin, PGQ and XPF, it was not necessary to digest with restriction enzyme before the fusion with the fusion partner cleaved with Ssp I, since they have 5′ blunt ends.

[0068] A monomeric unit that can produce multimers was constructed consisting of Nde I site coding for Met, structural gene, RBS (SEQ ID NO. 91), and Ase I site that connects with Nde I to generate Met. As structural genes, F4a(HA)-MS 1344 fusion gene ( F4Ma ) and F5-MSI344 fusion gene ( F5M ) were used. The monomeric units were digested with Nde I and Ase I, and the isolated monomeric units were reconnected. Obtained DNA fragments were digested again with Nde I and Ase I, and the multimers were separated by agarose gel electrophoreses. By using this method, monomer (F4Ma), dimer (F4MaX2) and tetramer (F4MaX4) of F4Ma and monomer (F5M), dimer (Fm5MX2) and tetramer (F5MX4) of F5M were obtained.

[0069] To express foreign gene in E. coli, two expression vectors pGNX2 and pT7K2.1 were constructed by using T7Φ10 promoter, high copy number replication origin (colEI of pUC family), and kanamycin resistance gene. To construct pGNX2, bla gene in commercially available pUC19 (ampicillin resistance gene: Amp^R) was substituted with kanamycin resistance gene (Kan^R). To this end, pUC19 was digested with Ssp I and Dra I to separate 1748 bp DNA fragment having 1748 bp, and Kan^Rgene was amplified by PCR by using Tn5 of E. colias a template and primers #39 and #40 (Table 1). The PCR product was digested with BamH I and Hind III, filled-in by Klenow treatment, and cloned int pUC19 digested with Ssp I and Dra I, resulting in pUCK2. After this vector was digested with Nde I and filled in by Klenow treatment, it was religated to contruct pUCK2ΔNdeI. The final plasmid pGNX2 was constructed by cloning the fragment containing T7Φ10 promoter and RBS from pT7-7 (USB, USA) that was digested with BamH I, filled-in by Klenow treatment, and then digested with Ase I, into the pUCK2ΔNdeI vector that was digested with Hind III, filld-in by Klenow treatment, and then digested with Ase I. T7Φ10 promoter and kanamycin resistant gene (Kan^R) are oriented to the same direction in pGNX2 (FIG. 6).

[0070] To construct the plasmid pT7K2.1, the bla gene was removed from pT7-7 by digestion with SspI and Bgl I, and the following treatment with T4 DNA polymerase to make blunt ends. Kan^Rgene was prepared as in pGNX2 and the two DNA fragments were ligated to construct pT7K2. Final plasmid pT7K2.1 was constructed by removing Ase I site from this vector (FIG. 7). E. coliHMS174 (DE3) transformed with pGNX2 was deposited to Korean Collection of Type Cultures (KCTC) in Korea Research Institute of Bioscience and Biotechnology located at Yusong-gu Eun-dong, Taejon, Korea on May 29, 1998 and the number KCTC0486BP was given. To construct pGNX3, pGNX2F4M was partially digested with BspH I, and the fragment that has a cut in a single BspH I site was separated and further digested with BamH I. To prepare fragment containing T7 and rrnBT1T2 terminators, 132 bp fragment from pET11 a digested with BamH I and EcoR V and a 488 bp fragment from ptrc99a digested with BamH I and EcoR V were ligated. These fused fragments were cleaved by BamH I and BspH I, and cloned into the vector prepared as above to construct pGNX3F4M (FIG. 8).

[0071] To prepare pGNX4, a 3052 bp fragment was isolated from pETACc digested with Xba I and AlwN I, and a 2405 bp vector fragment from the pGNX3F4M digested with Xba I and AlwN I resulting in pGNX4F4M (FIG. 9).

[0072] To construct pGNX5, pGNX3F4M was partially digested with Ase I, then digested with Xba I, and treated with Klenow fragment. A fragment obtained from PCR-TrpPO digested with EcoR I and Nde I and then treated with Klenow fragment was cloned with the above vector fragment to construct pGNX5F4M (FIG. 10).

[0073] DNA constructs obtained by fusing the MSI-344 to fusion partners, F3, F4, F4a, F5 and BF, were cloned into pGNX2 digested with Nde I and BamH I and pT7K2.1 digested with Nde I and BamH I, respectively. In case F (entire purF) was used as the fusion partner, it was cloned into pET24a (Novagen, USA) digested with Nde I and Xho I. In case of a multimer, it was cloned into the Nde I site of pGNX2 and pT7K2.1. The genes coding for Apidaecin I, Indolicidin, Tachyplesin I, Bombinin, CPF1, Drosocin, Melittin, HNP-I, PGQ and XPF were fused to the fusion partner F4 and cloned into pGNX2 digested with Nde I and EcoRI. When F3 was used, BamH I and EcoR I sites of pRSETc were used for cloning (Table 3).

[0074] The plasmids 2,3,4,5,6 and 7 in Table 3 were transformed into E.coliHMS174(DE3) by using the CaCl₂method. R medium supplemented with casamino acid was used as a culture medium, and the peptide expression was induced when OD₆₀₀was between 0.2 and 0.4 by adding 2% lactose and 2 mM IPTG, respectively. The expression level was quantified by scanning the results from SDS-PAGE by a densitometer and as the percent of fusion peptide in total cell proteins (FIG. 11). In FIG. 11, M represents molecular weight standard marker, and lanes 1 through 6 represent the expression from the transformants with plasmids 2,3,4,5,6, and 7 in Table 3 by lactose induction, and lanes 7 through 12 represent the expression from the transformants with plasmids 2,3,4,5,6, and 7 in Table 3 by IPTG induction. Lanes 13 and 14 represent the expression from the transformant with plasmid 43 (E. colipurF; EF) by lactose and IPTG induction, respectively. As in the same manner, MSI-344 was expressed using E.coliHMS174(DE3) transformed with plasmids 44, 45 and 46 in Table 3 and by lactose induction (FIG. 12). It can be seen that the expression level is higher with the plasmid having transcriptional terminator. With the HMS174 (DE3) transformed with plasmid 4 in Table 3, the expression of fusion peptide was induced by lactose and cells were harvested 9 hours after induction. The cells were sonicated and precipitates were obtained by centrifugation. After dissolving the precipitates by placing for 2 hours at room temperature in solution containing 9 M urea, 20 mM potassium phosphate (pH 8.5), the sample was loaded onto SP-sepharose FF column (Pharmacia, Sweden), and the fusion peptide F4Ma was eluted using 0.3˜1.0 M NaCl. Purified F4Ma was reacted in 0.5˜2 M hydroxylamine and 0.4 M potassium carbonate (pH 7.5-9.5) buffer to cleave MSI-344 from the fusion partner. After desalting, the reaction mixture was loaded onto SP sepharose FF column (Pharmacia, Sweden) again to elute MSI-344 with 0.4˜1 M NaCl. Purified MSI-344 was identified by HPLC, MALDI-MS and amino acid sequencing.

[0075] Other plasmids in Table 3 were transformed into E. coliHMS1 74 (DE3) by CaCl₂method. R medium supplemented with casamino acid was used as a culture medium, and the peptide expression was induced by adding 2% lactose when OD₆₀₀was between 0.4 and 0.6. The expression level was quantified by scanning the results from SDS-PAGE by a densitometer and as the percent of fusion peptide in total cell proteins. The results of the expression of each antimicrobial peptide are shown in FIGS. 13aand 13band Table 3. In FIG. 13a, lanes 1 through 6 represent the results from the transformants with plasmids 10,12,15,20,21 and 23 in Table 3. In FIG. 13b, lanes 1 through 9 represent the results from the transformants with plasmids 11,13,14,16,22,17,18,24 and 19 in Table 3. FIGS. 14a-14drepresent the expression results of plasmids 25-42 in Table 3. Buforin IIbx2 and Buforin IIx4 are dimer and tetramer of Buforin IIb, respectively, and constructed as described in Example 4. The corresponding plasmids, systems and expression results were indicated in parenthesis below:

[0076] FIG. 14a: 1(25)

[0077] FIG. 14b: 1(26), 2(31), 3(36)

[0078] FIG. 14c: 1(27), 3(32), 3(37), 4(28), 5(33), 6(38), 7(29), 8(34), 9(39), 10(30), 11(35), 12(40)

[0079] FIG. 14d: 1(41), 2(42), 3(43)

[0080] The constructs prepared in Example 4, such as monomer (F4Ma), dimer (F4MaX2) and tetramer (F4MaX4) of F4Ma and monomer (F5M), dimer (Fm5MX2) and tetramer (F5MX4) of F5M were transformed into E. coliHMS174 (DE3) after cloning them into Nde I site of pGNX2 and at Nde I site of pT7K2.1. Fusion protein was expressed following the method in Example 6, and the expression level was quantified by scanning the results from SDS-PAGE by a densitometer and as the percent of fusion peptide in total cell proteins. In FIG. 15, lanes 1-6 in pT7K2.1 represent F4Ma, F4MaX2, F4MaX4, F5M, F5MX2, and F5MX4, respectively. Lanes 1-4 in pGNX2 represent F4Ma, F4MaX2, F5M and F5MX2, respectively. As can be seen from FIG. 15, the expression level increased from 30% to 40% when the expression of tetramer was compared with that of the monomer. In the case of F5M, the expression level increased from 20% to 25% when the expression of tetramer was compared with that of monomer.

[0081] According to the present invention, antimicrobial peptides can be efficiently mass-produced from microorganisms more economically and can be separated and purified easily.