CA2351737A1 - Efficient methods for producing antimicrobial cationic peptides in host cells - Google Patents

Abstract

Endogenously produced cationic antimicrobial peptides are ubiquitous components of host defenses in mammals, birds, amphibia, insects, and plants.
Cationic peptides are also effective when administered as therapeutic agents.
A practical drawback in cationic peptide therapy, however, is the cost of producing the agents. The methods described herein provide a means to efficiently produce cationic peptides from recombinant host cells. These recombinantly-produced cationic peptides can be rapidly purified from host cell proteins using anion exchange chromatography.

Description

EFFICIENT METHODS FOR PRODUCING ,aNTIMIGROBIAL CATIONIC
PEPTIDES IN HOST CELLS
TECHNICAL FIELD
The present invention relates generally to methods for obtaining recombinant peptides and proteins from host cells. In particular, the present invention relates to improved processes for producing and purifying cationic peptides from recombinant host cells in which the peptide is ea~pressed in high yield and is easily recovered.
BACKGROUND OF THE INVENTION
tp Antimicrobial peptides, particularly cationic peptides have received increasing attention as a new pharmaceutical substance, because of their broad spectrum of antimicrobial activities and the rapid development of mufti-drug-resistant pathogenic microorganisms. Endogenous peptide antibiotics are ubiquitous components of host defenses in mammals, birds, amphibia, insects, and plants. These endogenous 15 antimicrobial peptides are usually cationic amphipathic molecules that contain 10 to 45 amino acid residues and an excess of Lysine and arginine residues. (for a review, see Broekaert et al., Plant Phyriol. 108:1353, 1995; Ganz and Lehrer, Pharmacol.
Ther.
66:191, 1995; Martin et al., J. Lezrkoc. Biol. .18:128, I995; Hancock and Lehrer, TIBTECH 16:82, 1998). Examples of cationic peptides include rabbit defensin, crab 2t} tachyplesin, bovine bactenecin; silk-moth cecropin A, frog magainins, and bovine indolicidin. The main site of action of the peptides is the cytoplasmic membrane of bacteria and other microbes. Due to their arnphipathic nature, the peptides disrupt the membrane, causing a loss of potassium ions, membrane depolarization, and a decrease in cytoplasmic ATP.
2s . Since their de rrovo synthesis or release from storage sites can be induced rapidly, cationic peptides are particularly important in the initial phases of resistance to microbial invasion. Cationic peptides are also effective when administered as therapeutic agents. In the treatment of topical infection, for example, an a-helical magainin variant peptide has been shown to be effective , against 3n polymicrobic foot ulcer infections in diabetics; arid a protegrin-derived peptide was found useful for treatment of oral polymicrobic ulcers in cancer patients (Hancock and Lehrer; TIBTECH 16:82; 1998). Efficacy against systemic infection has been shown with an a-helical peptide used to treat PseZrdomonas aerrrginosa peritoneal infection; a j3-sheet protegrin against methicillin-resistant ~itaphyloeocczrs aZrrezrs and against vancomvcin-resistant Interoc~rrc'cvr.,~ .fcre'c:crli.,~. and extended-helix indolicidin against .~L~perRillrr.a fun~~al infections (Gou~.h e~l crL, Irtfi.~cu. Irnrnrrrr.
<.1:4922, 1996; Steinberg et al., .-Itrtinricr~oh: .-l~,~errts ('hemuther. -11:1736. 1997; and Ahmad ct crl., l3iochim.
Biophv.s~. ,~cta I?3~:109, 1995). Therefore. naturally-occurring cationic peptides, and s their synthetic variants, are valuable antimicrobiaf therapeutics.
A practical drawback in cationic peptide therapy is the lack of a cost effective, mass-production method of the agents. Typically, the isolation of cationic peptides from natural sources is not cost-effective, and does not apply to the production of engineered cationic peptide variants which rr~ay have increased efficacy.
While to chemical peptide synthesis can be used to manufacture either natural or engineered cationic peptides, this approach is very costly.
Therefore, alternate, more economical and efficient methods of synthesis are needed, such as irt viva synthesis in host cells using' recombinant DNA
methods.
Researchers have attempted various methods for recombinant production of cationic 15 peptides. For example, cationic peptides have been produced in bacteria, such as E
coli or Staphylococerra° atrrerra, yeast, insect cells, and transgenic mammals (Piers et al., Gene l3-1:7, 1993, Reichhart et al., luvertehrate l~eprou! Develop. 21:15, 1992, HeIlers et aL, E:rr. J. Biochem. 199:435, 1991, and Sharma el al., Proc. Nat'! Acac~
Sci. USA
91:9337, 1994).
2y Much attention has focused on production in E. coli, since those versed in the art are familiar with the fact that high productivity can be obtained in E. coli using the recombinant DNA technology. However, for small peptides it is often necessary to produce them as pan of a larger fusion protein. In this technique the gene for the peptide is joined to that of a larger carrier protein and the fusion expressed as a 25 single larger protein. Following synthesis the peptide must be cleaved from the fusion partner. There is an extensive body of literature on protein fusion, especially in the gene expression host E toll. For example, a number of recombinant proteins have been produced as fusion proteins in E. toll, such as, insulin A and B chain, calcitonin, Beta-globin, myoglobin, and a human growth hormone (Uhlen and Moks, "Gene 3o Fusions for Purposes of Expression, An Introduction" in Methods irt Ettzymology 18.5:129-143 Academic Press, Inc. 1990). Nevertheless, recombinant gene expression from a host cell presents a number of technical problems, particularly if it is desired to produce large quantities of a particular protein. For example, if the protein is a cationic peptide, such peptides are very susceptible to proteolytic degradation, possibly due to 35 their small size or lack of highly ordered tertiary structure. One approach to solving this problem is to produce recombinant cationic: proteins in protease-deficient E. toll - WO 00!31279 PCT/CA99I01107 host cell strains (see. for example. ~~'illiams et crl.. U.S. Patent No.
S.s89,364, and WO
96/04373). Yet there is no general wav to predict: which protease-deficient strains will be effective for a particular recombinant protein.
In principle the recombinant DNA technique is straight forward.
However, anv sequence that interferes with bacterial growth throuvh replication or production of products toxic to the bacteria, such as lytic cationic peptides, are problematic for cloning. Foreign peptide 'ene products that are unstable or toxic, like cationic peptides, can also be stabilized by expressing the peptides as part of a fusion protein comprising a host cell protein. For examlale, Callaway et al.et aL, Antimicrob.
to .4gerrts Ghemvther. 3":1614, 1993, expressed cecropin A in E. cull as a fusion peptide with a truncated portion of the L-ribulokinase f;ene product, Piers et al.et al., Genre I3-1:7, 1993, expressed fusion proteins in E toll that comprised ~lutathione-S-transferase and either defensin (HNP-1) or a synthetic cecropin-melittin hybrid, while Hara et aL, Biuchem. Biophy:s. Res. Cvmmun. 220:664, 1996, expressed silkworm 15 moricin in E. cull as a fusion protein with a ~3-galactosidase or a maltose-binding protein moiety.
One of the better options to avoid the toxic effects of a bacteriolytic peptide on the host bacterial cells in highly efficient production, and to avoid proteolytic degradation of the peptides, is to utilize the intrinsic bacterial host Zc} mechanism of driving heterologous proteins into inclusion bodies as a denatured insoluble form.
The approach outlined above suffers from the inherent limitation on overall productivity imposed by the use of a small single peptide {circa 10%) in the large fusion protein.
Accordingly, a need exists for a means to efficiently produce cationic peptides from recombinant host cells.
SUNINIARY OF THIr INVENTION
The present invention provides compositions and methods for expressing large quantities of a selected polypeptide. Within one aspect of the invention, large ;U quantities of a selected polypeptide can be expressed utilizing a multi-domain fusion protein expression cassette which comprises a promoter operably linked to a nucleic acid molecule which is expressed as an insoluble protein, wherein the nucleic acid molecule encodes a polypeptide comprising the structure {cationic peptide) -[(cleavage site)-{cationic peptide)]", wherein n is an integer having a value between 1 and 100.
a5 Within certain embodiments, a cleavage site may be inserted on either side of the structure. e.~., (clear°a~re site)-(cationic peptide) -[(cleava'=a site)-(cationTC peptide))n cleavage site, wherein n is an integer having a value between 1 and 100.
Within certain embodiments, utilizing the methods described herein, the unit: -(cleavage site)-{cationic peptide)- can be added to the above expression cassette in order to specifically add a defined number of cationic sequences to be expressed.
Within various embodiments, n is an integer having a value of 2, 3, ~, 10, or, 20 on the lower end, and 10, 15, 20, 30, 40, 50, 75, or 80 on the upper end (e.g., n may be an integer between about 2 and 30, 2 and 40, etc., 5 amd 30, 6, or, 7 and 40, etc., up to 10 or 20 to 40, 50, 70 or 80). As an example, within one embodiment n has a value of to between 5 and 40 or 10 and 40.
Within certain embodiments, the nucleic acid molecule may further comprise a carrier protein. Within various embodiments, to the extent that a carrier protein is to be expressed by the expression cassette, it can be located at either the N-terminus or the C-terminus of the fusion protein. A wide range of carrier proteins can be utilized, including for example, a cellulose binding domain (CBD), or, a fragment of CBD. Within various embodiments, the carrier protein can be greater than, equal to, or less than 100 amino acids in length.
Within further embodiments, the cleavage sites within the expression cassette can be cleaved by, for example, low pht, or, by a reagent such as cyanogen 2o bromide, 2-(2-nitrophenylsulphenyl)-3-methyl-3'-bromoindolenine, hydroxylamine, v iodosobenzoic acid, Factor Xa, thrombin, enterokinase, collagenase, Staphylococcus aurezrs V8 protease, endoproteinase Arg-C, or trylpsin.
Within another embodiment, tlhe expression cassette may more specifically be comprised of (a) . a carrier protein, (b) an anionic spacer peptide component having at least one peptide with the structure (cleavage site)-(anionic spacer peptide), and (c) a cationic peptide component having at least peptide with the structure (cleavage site)-(cationic peptide) wherein the cleavage site can be on either side of the anionic spacer peptide or cationic peptide, and elements (a), (b), and (e) can be in any order and or number. Within a further related err~bodiment, the expression cassette may 3o be comprised of (a) an anionic spacer peptide component having at least one peptide with the structure (cleavage site)-(anionic spacer peptide), and (b) a cationic peptide component having at least peptide with the strucaure (cleavage site)-(cationic peptide), wherein the cumulative charge of said anionic ;pacer peptide component reduces the cumulative charge of said cationic peptide component.
To the extent an anionic spacer is included, such a spacer may have, 0, 1, 2, or more cysteine residues. Within certain embodiments, there can be more, the WO 00131279 PCTlCA99/OI 107 S
same number, or fewer anionic spacers than cationic peptides in the fusion construct.
lh'ithin certain embodiments. the anionic spacer is smaller in size than the cationic peptide.
A wide ~rarietv of cleavage sites can be utilized, including for example, a methionine residue. In addition, a wide variety of promoters can be utilized, including for example the lcrcP promoter, ~crcl' promoter, trcP promoter, srpP promoter, promoter; T7 promoter, crraP promoter, upP protn~oter, and ~. promoter.
The present invention also provides methods for producing fusion proteins utilizing the above-described expression cassettes. Within one embodiment, lt> such methods generally comprise the step of culturing a recombinant host cell containing an expression cassette, under conditions and for a time sufficient to produce the fusion protein. Representative examples of suitable host cells include yeast, fungi, bacteria (e.g., E. culi), insect, and plant cells.
Once the fusion protein has been produced, it may be further purified and isolated. Further, the fusion protein may be cleaved into its respective components (e.g., utilizing low pH, or, a reagent such as cyanogen bromide, 2-(2 nitrophenylsulphenyl)-3-methyl-3'-bromoindolenine, hydroxylamine, v-iodosobenzoic acid, Factor Xa, thrombin, enterokinase, collagenase, Staphylococcus arrretrs protease, endoproteinase Arg-C, or trypsin).
2p Further, the fusion protein or cleaned cationic peptide may be purified utilizing a chromatographic method (e.g., an anion chromatography column or resin}.
Within certain embodiments, the column can be charged with a base, and washed with water prior to loading the column with said cationic peptide. Within various embodiments, the column can be equilibrated with water and up to about 8 M
urea.
Moreover, the cationic peptide is solubilized in a solution comprising up to about 8 M
urea. Within further embodiments, the cationic peptide is solubilized in a solution comprising a mild organic solvent, such as, far example, acetonitrile, or, an alcohol such as methanol or ethanol.
These and other aspects of the present invention will become evident upon reference to the following detailed description and attached drawings. In addition, various references are identified below and are incorporated by reference in their entirety.

BRIEF DESCRIPTION OF THE DR..~W1NGS
Figure 1 is the maps of: (.~) pfasmid pET21(+). {B) plasmid pET-CBD180, (C) a PCR fragment containing cbd180 and (D) plasmids pET21CBD180-B
and pET21 CBD 180-X.
Figure 2 shows maps of fusion poly cationic peptide ~=enes.
Figure 3 is an SDS-PAGE analysis. showing the expression of different CBD-poly-MBl-11 fusion proteins. Column ST: molecular weight markers: 14.4, 21.5, 31, 45, 66.2 and 97.4 liDa; Column l: whole cell', lysate of E. coli MC4100 (pGPl-2) cultivated at 30°C; Column 2: whole cell lysate of E. cvli MC4 i 00 (pGP 1-2, 1o pET21 CBD96-11 ) cultivated at 30°C; Column 3: whole cell lysate of induced E. cull MC4100 {pGP 1-2, pET21 CBD96-11 ) at 42°C; Column 4: whole cell lysate of E. toll MC4100 (pGP t -2, pET21 CBD96-2x I I ) cultivated at 30°C; Column S:
whole cell lysate of induced E. cull 1~'IC4100 (pGPI-2, pET'21CBD96-2x11) at 42°C;
Column 6:
whole cell lysate of E. cUli MC4100 (pGPl-2) cultivated at 30°C; Column 7: whole cell E 5 lysate of E. cull MC4100 (pGP I -2, pET21 CBD9fi-3x 11 ) cultivated at 30°C; Column 8:
whole cell lysate of induced 1:. cull MC4100 (p~GP l -2, pET2 i CBD96-3x 11 ) at 42°C;
Column 9: whole cell lysate of E. cull MC4100 (pGPI-2, pET21CBD96-4x11) cultivated at 30°C; Column 10: whole cell lysate of induced E. cull MC4100 (pGPl-2, pET21 CBD96-4x 11 ) at 42°C;
Figure 4 shows maps of cassettes used for construction of genes of mufti-domain proteins.
Figure 5 presents maps of plasmid pET21 CBD96 and one of its inserts.
Figure 6 shows maps of fusion mufti-domain protein genes.
Figure 7 is SDS-PAGE analyses showing the results of fermentation of 25 mufti-domain clones having five or more MB1-11 B7 copies. The upper panels represent the multidomain clones fused to CBD carrier. The lower panels show the mufti-domain clones carrier-free. The left panels show the whole cell lysates, where the right side panels show the inclusion bodies partitioning step. The major band in each lane represents the relevant multidomain protein and the "x" numbers appearing at 3o the bottom of each lane indicate the number of the MB1- peptide copies.
Numbers appearing along the left edge of the gels represents molecular weight standards (kD).
Figure 8 shows maps of portions of plasmids pET21-3s-5xi 1B7 and pET21-Ss-7x 11 B7.
Figure 9 is a chromatogram of the; Q-Sepharose chromatography step for cationic peptide purification, which monitors UV absorbance at 280 nm and conductivity.

,~ _ Fi~~ure 10 is a schematic drawing that illustrates the construction of plasmids pET~ 1 CBD-X and pET' l CBD-B.
Figure 11 is a ~__=raph showinV the results of reverse-phase analysis of the Q-Sepharose chromato'raphy leading peak. repre:>enting pure cationic peptide.
In this a study. a C8-column (4.6 x 10. Nova-Pak, Waters,) was equilibrated with 0.1°,%TFA in water at 1 mllmin flow rate. Then >0 yl of Q-Sepharose chromatography leading peak material, diluted with 50 yl equilibration solution, was loaded on the column.
Elution was performed with a 0-45% gradient of solution .B ( 0.1% TFA, 99.9%
Acetonitrile) at 1% increase B per min, then step to 100% B.
to DETAILED DESCRIPT10N OF THE INVENT11JN
1. Overview As discussed above., a successful approach to stabilizing foreign peptide gene products which are inherently unstable or toxic is to express them fused to a protein which displays stability in the relevant host cell. In the case of small cationic IS peptides, however, production of a fusion protein will lead to a small portion of the desired peptide and an apparent low yield. A major gain in productivity and therefore economics of the process can be made if the fraction of desired peptide in the fusion protein is substantially greater. A favored route for this concept concerns expression of a fusion protein containing multiple sequential copies (a concatomer or multi-domain 2t~ protein) of the peptide separated by linker sequences. The linkers are the points at which the concatomer (multi-domain protein) will be cleaved to give monomers of the desired peptide with most probably modified C-termini as a result of the cleavage process.
On the other hand, increasing the number of copies of a cationic peptide 25 per fusion protein will make it a more and more basic protein, which may effect the expression of the fusion protein and/or increase iia toxicity for the host cell.
An approach to overcome the high basicity of the recombinantly-produced multi-domain cationic protein, and also decrease its toxicity, is to include small acidic peptide sequences in the linker sequences that neutralize the positive ;o charge of the cationic peptide. To keep the economic concept of high ratio of the cationic peptide in the multi-domain protein it is important to engineer the acidic peptide to be as small as possible, preferably smaller than the cationic peptide. The natural phenomenon of a multipeptide precursor structure consisting of cationic peptide and anionic spacer has been described (Casteels-Josson et al (1993) EMBO J., vol. 12, ;s 1569-1578}. In this publication the authors describe the natural production of g _ apidaecin, an antibacterial cationic peptide. in insects such as the honeybee (,~pis ntellifer-cr). Apidaecin is generated as a single gene comprising multiple repeated precursor units, each consisting of an apidaecin peptide ;.,jene ( 18 amino acids).preceded by an acidic spacer re~~ion (6-8 amino acids). In .a further example. Lee ct ul., I'rn~eirt a f :rp. Punif' I?:53 ( 1998), e~cpressed in E. ~wli six copies of the cationic peptide buforin II per fusion protein, which also included as acidic peptide modified magainin intervening sequences that alternated with the cationic peptide sequences. The magainin intervening sequences were "modifie:d" in that the sequences included flanking cysteine residues. According to Lee et crl., the "presence of cysteine residues to in the acidic peptide was critical for the high ie:vel expression of the fusion peptide multimers."
In initial studies, the present inventors used carrier proteins of different sizes to express monomer and polymer forms of cationic peptides. The test carrier protein of these studies were CBD and a fragment of the same derived from 15 C'lustriditrm ceJluloW rcrrt~~ celhrlvse hirtcJirtg proneirr ~I. The chosen carrier protein fulfilled the requirements of high expression and accumulation in E. coli as insojuble forms. This approach was limited by a si'nificant decrease in expression when the number of cationic fused peptide genes exceeded three copies. There was essentially no expression from vectors containing more than four copies of a peptide gene.
A new 2o procedure was designed which allowed the multiplication of relevant cationic peptide genes using a specific anionic spacer sequence; that encoded a negatively charged peptide. In these studies, the anionic spacer peptide consisted of Il amino acids.
Various genes encoding cationic peptide-anionic spacer peptide mufti-domain proteins were constructed and fused to the carrier protein. A high level of expression was 25 achieved for all constructs harboring more than thirty copies of the relevant cationic peptide gene. In subsequent studies, polymers of cationic peptide genes with anionic spacers were liberated from the carrier and expressed directly. These constructs achieved high levels of expression and a high percentage of target cationic peptide in the carrier-free mufti-domain protein.
2. Definitions In the description that follows, a number of terms are used extensively.
The following definitions are provided to facilitate understanding of the invention.
A "structural gene" is a nucleotide sequence that is transcribed into ;5 messenger RNA (mRNA), which is then translated into a sequence of amino acids characteristic of a specific polypeptide.

WO 00/31279 PCT/CA99l01107 As used herein. "nucleic acid" or "nucleic acid molecule ~ refers. to any of deoxyribonucleic acid (DNA), ribonucleic acid (RNA), oli~lonucleotides.
fra;ments generated by the poivmerase chain reaction (PCR), and fragmems ~~enerated by any of ligation, scission, endonuclease action. and exonuclease action. Nucleic acids can be composed of monomers that are naturally-occurring nucleotides (such as deoxyribonucleotides and ribonucleotides), or analogs of naturally-occurring nucleotides (e.g., a.-enantiomeric forms of naturally-occurring nucleotides), or a combination of both. :an ''isolated nucleic acid molecule" is a nucleic acid molecule that is not integrated in the ~enomic DNA of an organism. For example, a DNA molecule that to encodes a cationic peptide that has been separated from the ~enomic DNA of a cell is an isolated DNA molecule. Another example of an isolated nucleic acid molecule is a chemically-synthesized nucleic acid molecule that is not integrated in the genome of an orsanism.
An "isolated polypeptide or protein" is a polypeptide that is essentially l5 free from contaminating cellular components, such as carbohydrate, lipid, nucleic acid {DNA or RNA) or other proteinaceous impurities associated with the polypeptide in nature. Preferably the isolated polypeptide is sufi:iciently pure for clinical injection at the desired dose. Whether a particular cationic polypeptide preparation contains an isolated cationic polypeptide can be determined utilizing methods such as Urea / acetic 2o acid polyacryiamide gel electrophoresis and Coomassie Brilliant Blue staining of the gel, reverse phase high pressure liquid chromatography, capillary electrophoresis, nucleic acid detection assays, and the Limulus Am~eb.ocyte Lysate test.
Utilizing such a method an isolated polypeptide preparation will be: at least about 95% pure polypeptide.
An "insoluble polypeptide" refers to a polypeptide that, when cells are 25 broken open and cellular debris precipitated by centrifugation (e.g., 10,000 ~ to 15,000 g), produces substantially no soluble component, a.s determined by SDS
polyacrylamide gel with Coomassie Blue staining.
A "promoter" is a nucleotide sequence that directs the transcription of a structural gene. Typically, a promoter is located in the 5' region of a gene, proximal to 3o the transcriptional start site of a structural gene. if a promoter is an inducible promoter, then the rate of transcription increases in response to an inducing agent. In contrast, the rate of transcription is not regulated by an inducing agent if the promoter is a constitutive promoter.
The term "expression" refers to the; biosynthesis of a gene product. For 35 example, in the case of a structural gene, expression involves transcription of the structural gene into mRNA and the translation of mRNA into one or more poiypeptides.

WO 00/31279 PC'd'/CA99/01107 A "cloning vector: ~ is a nucleic acid molecule. such as a plasmid.
cosmid. or bacteriophage. that has the capability of replicating autonomously in a host cell. Cloning vectors typically contain one or a small number of restriction endonuclease recognition sites at which foreign nucleotide sequences can be inserted in s a determinable fashion without loss of an essemial biological function of the vector, as well as nucleotide sequences encoding a marker gene that is suitable for use in the identification and selection of cells transformed with the cloning vector.
Warker genes typically include genes that provide antibiotic resistance.
An "expression vector" is a nucleiic acid molecule (plasmid, cosmid, or bacteriophage) encoding a ~,~ene that is expressed in a host cell. Typically, gene expression is placed urider the control of a promoter, and optionally, under the control of at least one regulatory element. Such a gene is said to be "operably linked to" the promoter. Similarly, a regulatory element and .a promoter are operably linked if the regulatory element modulates the activity of the promoter.
lg A "recombinant host" may be any prokaryotic or eukaryotic cell that contains either a cloning vector or expression vector. This term also includes those prokaryotic or eukaryotic cells that have been genetically engineered to contain the cloned genes) in the chromosome or genome of the host cell.
As used herein, "cationic peptide" refers to a peptide that possesses an 2o isoelectric point (pl) of 9 and above. A catianic; peptide is at least five amino acids in length, and has at least one basic amino acid {e.g~., arginine, lysine, histidine). Cationic peptides commonly do not have more than 50 amino acids, and typically contain 10 to 35 amino acid residues.
A "carrier protein' is an amino acid sequence that can be individually 25 expressed in host cells and, by recombinant fusion to a desired peptide or polypeptide can act as a carrier, enabling the expression of the desired peptide in host cells.
An "anionic spacer peptide domain" is a peptide sequence that is sufficiently anionic to decrease the positive charge of an associated cationic peptide.
That is, the combination of a cationic peptide and an anionic spacer peptide has a net charge that is essentially slightly positive, negative or neutral. The size of an anionic spacer domain is similar to but preferably smaller than the size of the cationic peptide domain.
As used herein, a ''fusion protein" is a hybrid protein expressed by a nucleic acid molecule comprising nucleotide sequences of at least two genes. A
"multi 35 domain protein" comprises a combination of preferably more than one "cationic peptide domain," and an equal, smaller or higher number of "anionic spacer peptide domains'' with suitable cieava~Je sites for separatin'T cationic peptide from the rest ~of the muiti-domain protein. The mufti-domain protein can be t:used to a carrier protein to achieve hi;~her expression andior stability. If stability and expression level of mufti-domain protein are satisfactory°, there is no need to use a carrier protein.
An "anionic spacer a peptide component ~ comprises at least one anionic spacer peptide with a cleavage site.
The "cumulative char~e'~ of a cationic peptide component refers to the total charge of all cationic peptides that comprise the cationic peptide component. Similarly, the "cumulative charge' of an anionic spacer peptide component refers to the total charge of al! anionic spacer peptides that comprise the aniionic spacer peptide component.
t« As used herein, "antimicrobial activity" refers to the ability to kill or to prevent the growth of a microbe, or to kill or to prevem the growth of microbe-infected cells. The term "microbe" includes bacteria, fungi, yeast, algae, protozoa, and viruses.
This term includes but will not be limited to all these irnerpretive descriptions of the biolo~icai activity of the cationic peptide.
t 3. Construction and Expression of Vectors Comprising Cationic Peptide Genes cr. Cationic Peptide Expressia~n Vectors The present invention contemplates the production of "cationic peptide,"
2~~ as that term is defined above. For example, suitable cationic peptides include but are not limited to, naturally occurrinU cationic peptides and analogs thereof, cecropins, normally made by lepidoptera (Steiner e~ al., Natrrre ?>2:246, 1981 } and diptera (Merrifield et aL, C'iha Found Sj~mp. 186:5, 1994), by porcine intestine (Lee et al., Prvc. Nat'l Acad Sci. LISA 8b:91~9, 1989), by blood cells of a marine protochordate 25 (Zhao et al., FEBS Lett. -/12:144, 1997), synthetic analogs of cecropin A, melittin, and cecropin~melittin chimeric peptides (Wade et a,C, Int. .L Pept. Protein Res. -/0:429, 1992), cecropin B analogs (Jaynes et aL, Plaut Sci. 89:43, 3993), chimeric cecropin A/B hybrids (During, Mol. Breed. 2:297, 1996), magainins (Zasloff, Prc~c.
Nat'I Acacl Sci USA 8-I:5449, 1987}, cathelin-associated antimicrobial peptides from leukocytes of ;o humans, cattle, pigs, mice, rabbits, and sheep (Zanetti et al., FEBS Lett.
.3--l:1, 1995), vertebrate defensins, such as human neutrophil defensins [HNP 1-4], paneth cell defensins of mouse and human small intestine (Oulette and Selsted, FASEB .l.
10:1280, 1996; Porter et al., lufect. Immma. 6:2396, 1997), vertebrate [3-defensins, .
such as HBD-1 of human epithelial cells (Zhao et al., FEBS Lout. 3b8:331, 1995), HBD-Z
of 35 inflamed human skin (Harder et al., Nature 38":861, 1997), bovine j3-defensins (Russell et al., lyect. Inamun. 6-/:1565, 1996), plant defensins, such as Rs-AFPI of WO 00/31279 PCTlCA991011t)7 radish seeds (Fehlbaum c't crl.. .l. liinl. ( ~lr~'m. 'i5>:33 l 59, 1994). c~-and (i-thionins (Stuart et crl.. ('meal ('hear. I J:288. 1942: Bohlrnann and Apel. .-lrnut.
IZc't~. I'l?vsiul.
Plant tLlol. I3iol. -1':?~7. 1991 ), ;_thionins (Broel::aert m ctl., I'lcuu I'l7voiol. IOa:I3s3.
1995), the anti-fungal drosomvcin (Fehlbaum et crl., .I. l3iul. ('hc'm.
'fJ:33159, 1994), s apidaecins, produced by honey bee,. bumble bee. cicada killer, hornet, yellow jacket, and wasp (Casteels c't crl., .l. l3iol. 1_.'lac,~m. ?fi~:26107, 1994;
L.evashina c't crl., Eur. .l.
Bivchem. ?33:694, 1995), cathelicidins, such as indoiicidin from bovine neutrophils (Falla et aL, .l. BiuL C'lrcm. 2" :1929$, 1996), bacteriocins, such as nisin (Delves-Broughton et aL, ~rrrunie wrrr Leerrwerrhvek .l. Micruhiol. b9:193, 1996), and the It) prote~rins and tachyplesins, which have antifunga~i, antibacterial and antiviral activities (Tamamura et ctl., Biochinr. Bivphy:s. .9cla ll6.i':209, 1993; Aumelas et aL, Errr. .I.
Biochem. ?3":575, 1996; Iwanga ct al., Ci~Sa Fartrru! Sy°mp. Ib'6:160, 1994).
Illustrative cationic peptides are listed in Table 1.

ILLUSTRA'i~IVE CATIONIC; PEPT)DES**
Group Name ~ Peptiule Sequence Reference*

Abaecins Abaecin YVPLPNVPQPGF~RPFPTFPGQGCasteels et al.
(199t)) PFNPKIKWPQGS.' Andropins Andropin VFIDILDKVENAIHNAAQVGIGSamakovlis et al. (1991) FAKPFEKLINPK

Apidaecins Apidaecin GNNRPVYIPQPF;PPHPRI Casteels er ai.
IA ( 1989) Apidaecin GNNRPVYIPQPF;PPHPRL Casteels er al.
IB (1989) Apidaecin GNNRPIYIPQPRPPHPRL Casteels et al.
II ( t 989) AS AS-~8 7.-I kDa - Galvez et al.
( 1989) BactenecinsBactenecin RLCRIVVIRVCR: Romeo et al. (1988) Bac Bac~ RFRPPiRRPPIRP~PFYPPFRPPIRPFrank et at. ( 1990) PIFPPIRPPFRPPLRFP

Bac7 RRIRPRPPRLPRPRPRPLPFPRPFrank et al. (1990) GPRPIPRPLPFPItPGPRPIPRPLP

FPRPGPRPIPRP

BactericidinsBactericidinWNPFKELERA(iQRVRDAVISADickinson et al.
B2 (1988) APAVATVGQA,AAIARG*

WO 00!31279 PCT/CA99f01107 Croup Name Pelnide ~~;4uence Rcfercnce*

B,ctcricidinWNPFILELERAGQRVRDAIISADickinson ~t al.
B-s ( 1988) GP AVATVGQ AA?,IARG

Bactcricidin~VNPFKELERAGQRb'RDAIISADickinson et crl.
B--t ( 1988) APAVATVGQAAAIARG*

BactcricidinWNPFI~ELERAGQ!RVRDAVISADickinson er al.
B- (1988) gp AAVATVGQAAALARGG*

BacteriocinsBacteriocin -1.8 kDa T,kada et al. ( 198-t) C;6o.;

Bactcriocin s kDa Nakamura et crl.
IY~2 (1983) Bombinins Bombinin GIGALSAKGALKGLAKGLAZHCsordas and Michl (1970) FAN*

gLp-I GIGASILSAGKSA.LKGLAKGLAGibson et al. (1991) EHFAN*

BLP-2 GIGSA1LSAGKSA.,LKGLAKGLAGibson ct al. ( 1991) EHFAN*

BornbolitinsBombolitin IKITT1W.AKLGK'\tI.AHV*Argiolas and Pisano B1 (1985) Bombolitin SK1TDILAKLGKVLAHV* Argiolas and Pis~uto BII ( 1985) BPTI Bovine RPDFCLEPPYTG;PCKARIIRYFCreighton and Charles Pancreatic YNAKAGLCQTFVYGGCRAKR ( 1987) Tn~psin InhibitorNNFKSAEDCMR'TCGGA

(BPTI) Brevinins Brevinin-IE FLPLLAGLAANFLPKIFCKITRSimmaco et al.
(1993}

KC

Brevinin-2E GIMDTLKNLAK'TAGKGALQSLSimmaco et al.
(i993) LNKASCKLSGQC

Cecropins Cecropin KWKLFKKIEKVGQNIRDGIIKAGudmundsson et A al. (1991) GPAVAWGQATQIAK*

Cecropin KWKVFKKIEKNIGRNIRNGIVKXanthopoulos et B al. (1988) AGPAIAVLGEAKAL*

Cecropin GWLKKLGKRIERIGQHTR.DATTryseiius et al.
C ( 1992) IQGLG1AQQAAP~IVAATARG*

Cecropin WNPFKELEKVCiQRVRDAVISAHultmark et al.
D (1982) GPAVATVAQA'1'ALAK*

Grout Dame Peptide Sequence Reference*

Cccropin S~'VLSIiT.~IthLE;'ISAKKRISEGILcc er aI. t 198'J) P;

.Af AIQGGPR

ChawbdtoxinsCharybdtoxinZFTNVSCTTSKECWSVCQRLHSchwcitz et al.
( 1989) NTSRGIS.C~tNKK~~RCYS

CaleoptericinsColeoptericin8.1 kDa Bulet et al. ( 1991) Crabrolins Crabroiin FLPLILRKIVTAL"' Argiolas and Pisano (198-t) cx-DefensinsCn~ptdin LRDLVCYCRSRGCKGRERMN Setsted et al.
1 ( 1992) GTCRKGNLLYTL,CCR

Crs'ptdin LRDLVCYCRTRCiCKRRERMNSelsicd et al.
2 ( 1992) GTCRKGHLMYTLCCR

MCP 1 WCACRRALCLPRERRAGFCR Selsted et al.
( 1 ')83) IRGRIHPLCCRR

MCP2 WCACRRALCLI'LERRAGFCRGanz et al. ( 1989) iRGRIHPLCCRR

GNCP-1 RRCiCTTRTCRFI'YRRLGTCIFYamashita and Saito QNRVYTFCC ( 1989) GNCP-2 RRCICTTRTCRFPYRRLGTCLFYamashita and Saito QNRWTFCC ( 1989) HNP-1 ~ ACYCRIPACIAGERRYGTCIYQLehrer et al. ( 1991) GRL W AFCC

HNP-2 CYCRIPACIAGERRYGTCIYQGLehrer et al. ( 1991 ) RL W AFCC

NP-1 WCACRRALCLPRERRAGFCR Ganz co al. (1989) IRGRIHPLCCRR

NP-2 WCACRRALCLPLERRAGFCR Ganz et al. ( 1989) IRGRIHPLCCRR

RatNP-1 VTCYCRRTRCGFRERLSGACGEisenhauer et al.
(1989) YRGRIYRLCCR

RatNP-2 VTCYCRSTRCGFRERLSGACGEisenhauer et al.
( 1989) YRGRIYRLCCR

~3-DefensinsBNBD-I DFASCHTNGGICLPNRCPGHMISelsted et at.
(1993) QIGICFRPRVKCCRSW

WO 00/31279 PCTlCA99/01107 Group N:~meReptile Srqucn ec Refcrencr*

BNBD-2 \'RNHVTCRINRGFC~'P1RCPGR

TRQIGTCFGPRIKCCRSW Sclsted et aJ.
( 1993) T.AP ~~VSCVRN1LGIC\~'PIRCPGSMKDiamond ct ul.
( 1991 ) Q1GTCVGRAVKC'CRKK

Defensins- Sapecin ATCDLLSGTGINI-ISACAAHCLHanzawa et al.
( 1990) insect LRGNRGGYCNGICAVCVCRN

Insect defensinGFGCPLDQMQCHRHCQTITGR Bulet et aJ. (1992) SGGYCSGPLKLTCTCYR

Defensins- Scorpion GFGCPLNQGACHRHCRSIRRR Cociancich et al. ( 1993) scorpion defensin GGYCAGFFKQTCTCYRN

DennaseptinsDenuaseptinALWK'I'MZ.KIhLG'~TMALHAGKMoreral. (1991) AALGAADTISQTQ

DiptericinsDiptericin 9 kDa Reicl:hardt et al. ( 1989) Drosocins Drosocin GKPRPYSPRPTSI~iPRPIRV Bulet et al. (1993) EsculentinsEsculentin GIFSKLGRKKIKrJL.LISGLKNVSimmaco et aJ.
(1993) GKEVGMD VVRTGIDIAGCKIK

GEC

lndolicidins)ndoticidinILPWKWPWWPWRR* Selsted et aL
(1992) LactoferricinsLactoferricinFKCRRWQWRMI~:Ki.GAPSITCBellamy et al.
B (1992b) VRRAF

LantibioticsNisin iTSISLCTPGCKTI.iALMGCNMHurst (1981) KTATCHCSIHVSI~

Pep 5 TAGPAIRASVKQCQKTLKATR Keletta et al.
( 1989) LFTVSCKGKNGC.'K

Subtilin MSKFDDFDLDVVKVSKQDSKI Banerjee and Hansen TPQWKSESLCTPGCVTGALQT (1988) CFLQTLTCNCKISK

Leukocins Leukocin KYYGNGVHCTKSGCSVNWGE Hastings et al.
(1991) A-vat Y87 AFSAGVHRLANCiGNGFW

Magainins Magainin GIGKFLHSAGKFGKAFVGEIM Zasloff (1987) i KS*

Magainin GIGKFLHSAKKFC rKAFVGEIMZasloff (1987) II

NS*

Grout Name r Seyucnne Rcf'erence*
Pclptide~

PGL: y GiVIASKAGAIAGKIAKVALKALKuchlcrcral. (1~J89) pGQ GV'L.SNVIGy'LKKI.GTGALNAVMoore er ad. ( 1989) LKQ

GW ASKIGQTLGKIAKVGLKELSures and Crippa ( i 98.1) ;~F

IQPK

MastoparansMastoparan INLKALAALAKK1L* Bcrnheimer and Rudy (1986) tin GIGAVLKVLTTGLPALISW1KRTosteson and Tosteson li M

Melittins e t KRQQ ( 198:1) ' Phormicins Phornticin ATCDLLSGTGINl-1SACAAHCLLambert ea al. ( A 1989) Phonnicin ATCDLLSGTGINIHSACAA.HCLLambert et al. ( B 1989) PoiyphcmusinsPolvphemusinRRWCFRVCYRGFCYRKCR* Mivata et al. ( 1 1989) PolvphemusinRRWCFRVCYKGFCYRKCR* Miyata et al. ( lI 1989) i rin I RGGRLCYCRRRIFCVCVGR Kokmakov et al.
Prote s (1993) n g Protegr Protegrin RGGRLCYCRRILFCICV Kokn~akov et al.
II ( 1993) Protegrin ~ RGGGLCYCRRRFCVCVGR Kokn~akov et al.
III (1993) i VTCDLLSFKGQ~~NDSACAANCFujiwart et al.
li ( 1990) R

Rovalisins n s oya LSLGKAGGHCE:KGVCICRKTS

FKDLWDKYF

xin IA GWLKKIGKKIEItVGQHTRDATOkada and Natori t (1985b) S

Sarcotaainso arco tQGLGIAQQAAPJVAATAR*

Sarcotoain GWLKKIGKKIE1~VGQHTRDATOkada and Natori IB (1985b) IQV1GVAQQAA1VVAATAR*

Seminal SeminalpiasminSDEKASPDKHHRFSLSRYAKLRedd<< and Bhargava plasmins ANRLANPKLLET'FLSKWIGDR( 1979) GNRSV

TachyplesinsTachvplesin KWCFRVCYRG:ICYRRCR* Nakamura et al.
I (1988) Tachvplesin RWCFRVCYRGfCYRKCR* Muta et al. ( 1990) II

Group Name !. Peptide Scyucnce Reference*

Thionins Thionin hSCCi~.DTL.4RNC5'NTCRFAGGBohlmann rr cyl.
( 19HR) BTH(> SRPVC AGACRCKIISGPKCPSD

YPK

Toxins Toxin 1 GGKPDLRPC1IPPCHYIPRPKPRSchmidt er al.
( 1992) Toxin 2 VKDGYIVDDVNC:TYFCGRNABontems et ai.
t f 991 ) YCNEECTKLKGE:SGYCQWASP

YGNACYCKi.PDt~VRTKGPGR

CH

*Argiolas and Pisano, .IBC' 259: i 0106 ( I 984); Argiolas and Pisano, .IBC' 260:1437 ( 1985); Banerjee and Hansen, .IBC' 263:9508 ( 1988); Bellamy et aL, .l. Appl.
Bacter.
73:472 (1992); Bernheimer and Rudy, BBA 864:123 (1986); Bohlmann et al., F~IBO.I.
7: I 559 ( 1988); Bontems et al., Science 254:1521 ( 1991 ); Bulet et aL, .IBC. 266:24520 (1991); Bulet et al., Errr-. .l. Bivcher~a. 209:977 (1992); Bulet et al., .IBC.' 268:14893 {1993); Casteels et al., Eh~IBO .l. 8:2387 {1989); Casteels et al.. Errr. .I.
Bivchem:
187:381 (1990); Cociancich et al., BBRC 194:1,7 (1993); Creighton and Charles, .!.
il~lvl. BrUI. 194:11 ( 198?); Csordas and Michl, Mvnatsh C:hemistrv 101:82 ( 1970);
Diamond et al., PNAS 88:3952 (1991); Dickionson et aJ., .IBC 263:19424 (1988);
Eisenhauer et aL, Iyect. and Imm. 57:2021 (198S~), Frank et al., .IBC 26518871 (1990);
Fujiwara et al., .IBC 265:11333 (1990); Galvez et al., Arrtimicrvbial Agents and Chemotherapy 33:437 (1989); Ganz et al., .l. I»tnnrrnol.. 143:1358 (1989);
Gibson et al., .IBC 266:23 I 03 ( 1991 ); Gudmundsson et al., .IB'C 266:1 I 510 ( 1991 );
Hanzawa et al., t5 FEBS Letters 269:413 (1990); Hastings et al:, .l. Bacteriology 173:7491 (1991);
Hultmark et al., Errr. .I. Bivchem. 127:207 (1982); Hurst, Adv. Appl. Micro.
27:85 ( 1981 ); Kaletta et al., Archives of Micrvhiolvgy 152:16 ( 1989); Kokryakov et al., FEBS
Letters 327:231 (1993); Kuchler et al., Errr. .l. Bivchem. 179:281 (1989);
Lambent et al., PNAS 86:262 {1989); Lee et aL, PNAS 86:9159 (1989); Lehrer et al., Cell 64:229 2t~ (1991); Miyata et al., .l. Bivchem. 106:663 ('1989); Moore et al., JBC
266:19851 ( I 991 ); Mor et al , Biochemistry 3 0:8824 ( 1991 ); Muta et al., .l.
Biochem. 108:261 (1990); Nakamura et al., .IBC 263:16709 (1988); Nakamura et al., Infection and Immunity 39:609 (1983); Okada and Natori, Binchem. .I. 229:453 (1985); Reddy and Bhargava, Nature 279:725 ( 1979); Reichhart et al., Eur. ~.I. Bivchem. 182:423 ( 1989);
25 Rameo et al., .IBC 263:9573 (1988); Samakavlis et al., EMBO .l. 10:163 (1991);
Schmidt et al., TC3XICtlr1 3O:I027 (1992); Schweitz et al., Bivchem. 28:9708 (1989);
Selsted et al., .IBC 258:14485 (1983); Selsted et crl., .IBC 267:4292 (1992);
Simmaco et crl:; l-'l~'Ij5 Lctl. 324:159 ( 1993): Sures and Crippa, I'.'V~.S 81:380 ( 1984), Takada m crl., Irrfi,°cr. crncl Imm. 44:.17~ ( 1984); Tosteson and Tosteson.
l3inphy~.,~iccrl .l. :~5:1 12 ( 1984);
Trvselius cu ul., I:ur. .l. l3inchenr. '_'04:395 ( 199?); Xanthopoulos e~I
crl., Eur. .l.
l3iochen?. 172:371 ( 1988); Yamashita and Saito, Irrfvcl: trod Imm. 57:2405 ( 1989);
a Zasloff, I'NAS 84:5449 ( 1987).
** S S Patent Nos. 4,822,608; 4,962,277; 1,980,163; 5,028,530;
also U

ee .
096,886; 5,166,321; 5,179,078; 5.202,420; 5,212,073; 5,242,902;

5,254,537;

, 5,300,629; 5,304,540; 5,324,716; 5,344,765; 5,422,424;
5,278,287; 5,424,395;

5,446,127; 5,459,235; 5,464,823; 5,466,671; 5,512,269; 5,516,682;
5,519,115;

5,519, I 5,547,939; 5,556,782; 5,610, f 39; 5,645,966; 5,567,681;
16; 5,585,353;

5,589,568; 5.594,103, x,610,139; 5,631,144; 5,635,479; 5,656,456;
5,707,855;

5,731,149; 5,714,467; 5,726,155; 5,747,449; 5,756,462; PCT Publication Nos. WO

89/00199; WO 90/11766; WO 90/1177.1; WO 9i1C10869; WO 91112815;
WO 91/17760;

WO 94/05251;
WO 94105156;
WO 94/07528;
'WO 95121601;
WO 97/00694;
WO

97i I 1713;
WO 97118826;
WO 97102287;
WO 98103192;
WO 98/07$33;
WO 98107745;

European Application Nos. EP
1'7785;
349451;
60?080;
665239;
and Japanese Patent/Patent Application Nos. 43411'79;
435883;
7196408;
798381;
and 8143 596.

za Nucleic acid molecules encoding cationic peptides can be isolated from natural sources or can also be obtained by automated synthesis of nucleic acid molecules or by using the polymerise chain reaction (PCR) with oligonucleotide primers having nucleotide sequences that are based upon known nucleotide sequences 25 of cationic peptides. In the latter approach, a cationic. peptide gene is synthesized using mutually priming long oligonucieotides (see, for example, Ausubel et al.
(eds.), Short Protocols in Molecular Biology, 3'° Edition, pages 8-8 to 8-9. (John Wiley ~ Sons 1995), "Ausutiel (1995)"). Established techniques using the polymerise chain reaction provide the ability to synthesize DNA molecules of at least two kilobases in length (Adang et ;o aL, Plaut Molec. Biol. 21:1131, 1993; Bambot Eat al., PCR Methods and Applications '266, 1993; Dillon e1 al., "Use of the Polym~erase Chain Reaction for the Rapid Construction of Synthetic Genes," in Methods in Molecular Biology, L'vl. l~:
PCR
Protocols: Current A~IeIhods,alrd Applicaliorrs, White (ed.), pages 263-268, (Humana Press, Inc. 1993); Holowachuk et al., PCR Methods Appl. -1:299, 1995).
As noted above, analogs of natural cationic peptides can also be recombinantly produced by the presently described methods. The presence of a codon WO OOI3t279 t'CT/CA99/01107 may have an adverse effect on expression and therevfore~a DNA sequence encoding the desired cationic peptide is optimized for a particular host system, in this case L:. cull.
amino acid sequences of novel cationic peptides are disclosed, for example, by Falla cu crl., WO97/08199, and by Fraser el crl:, WO 98/077=15.
One type of cationic peptide analos: is a peptide that has one or more - conservative amino acid substitutions, compared 'with the amino acid sequence of a naturally occurring cationic 'peptide. For example, a cationic peptide analog can be devised that contains one or more amino acid substitutions of a known cationic peptide sequence, in which an alkyl amino acid is substituted for an alkyl amino acid in the 1o natural amino acid sequence, an aromatic amino acid is substituted for an aromatic amino acid in the natural amino acid sequence., a sulfur-containing amino acid is substituted for a sulfur-containin; amino acid in the natural amino acid sequence, a hydroxy-containing amino acid is substituted, for a hydroxy-containing amino acid in the natural amino acid sequence, an acidic amino acid is substituted for an acidic amino acid in the natural amino acid sequence, a basic amino acid is substituted for a basic amino acid in the natural amino acid sequence, or a dibasic monocarboxylic amino acid is substituted for a dibasic monocarboxylic amino acid in the natural amino acid sequence.
Among the common amino acids, for example, a "conservative amino 2e~ acid substitution" is illustrated by a substitution among amino acids within each of the following groups: ( 1 ) glycine, alanine, valine, leucine, and isoleucine, (2) phenylalanine, tyrosine, and tryptophan, (3) serine and threonine, (4) aspartate and ~iutamate, (5) glutamine and asparagine, and (6) lysine, arginine and histidine.
Nucleotide sequences encoding such "conservative amino acid" analogs 25 can be obtained, for example, by oligonucleotide-directed mutagenesis, linker-scanning mutagenesis, mutagenesis using the polymerase chain reaction, and the like (see Ausubel (1995) at pages 8-10 to 8-22; and McP'herson (ed.), Directed Mutagenesis: A
Practical Approach (IRL Press 1991)). The antimicrobial activity of such analogs can be determined using a standard method, such as the assays described herein.
.;0 Alternatively, a cationic peptide analog can be identified by the ability to specifically bind anti-cationic peptide antibodies. Typically, cationic peptide analogs should exhibit at least 50%, and preferably, greater than 70, 80 or 90%, of the activity of the corresponding naturally occurring cationic peptide.
Although one objective in constructing a cationic peptide variant may be .;5 to improve its activity, it may also be desirable to alter the amino acid sequence of a naturally occurring cationic peptide to enhance its production in a recombinant host WO 00!31279 PCT/CA99/01107 cell. For example. a nucleotide sequence encoding a radish cationic Ypeptide may include a colon that is commonly found in radish, but is rare for 1.:. cwli:
The presence of a rare colon may have an adverse effect on protein levels when the radish cationic peptide is expressed in recombinant l: cvli. Methods for alterin~T nucleotide sequences to allee~iate the colon usage problem are well known to those of skill in the art (see, for example, Kane, (~'rrrr-. ()pin. Ijlr~l~chllvl. 6:494., i 995, Makrides, A.licrahivl. Rev.
60:512, 1996, and Brown (Ed.), h~uJecrrlcrr Bioly.> LahFcrx (BIOS Scientific Publishers, Ltd. 1991), which provides a colon usage table on pages 245-253).
The present invention contemplates the use of "anionic spacer peptide' as to that term is defined above. :~s described below, an illustrative anionic spacer peptide has the amino acid sequence of HEAEPEAEPIM where the methionine residue can be used as a cleavage site. Similar naturally occurrinf; examples of anionic spacer peptides include EAEPEAEP, EAKPEAEP, EAEPKAEP, EAESEAEP, EAELEAEP, EPEAEP
and EAEP (Casteels-Josson. W crl. EMBO J., 12:1.569-1578, 1993). Additional anionic 15 spacer peptides are suitable for use in producing, cationic peptides such as doubles or other combinations of those illustrated above. When designing an anionic spacer peptide for expression of a particular cationic peptide in the mufti-domain.
protein concept, the following criteria should be borne in mind: the negative charge of the anionic spacer peptide should substantially reduce the positive charge of the cationic 2o peptide in the mufti-domain fusion proteins, a cleavage point must be present at which the mufti-domain protein will be cleaved to give monomers of the desired peptide, and the anionic spacer peptide is preferably smaller than the cationic peptide.
Such fusion proteins can be designed with alternating units of cationic peptide and anionic spacer peptide. Such a~ configuration, however, is not: required. Any sequence of cationic 25 peptide and anionic spacer peptide is acceptable., as long as the cumulative charge of the concatomer in the multidomain protein will not effect its expression in host cells.
In the examples described herein, a cellulose binding domain (CBD) carrier protein was used to illustrate methods for producing cationic peptides.
Additional suitable examples of carrier proteins include, but are not limited to, 3U glutathione-S-transferase (GST), S>aphylc~coccu~, arrrerrs protein A, two synthetic IgG-binding domains (ZZ} of protein A, outer membrane protein F, ~3-galactosidase (lac2), and various products of bacteriophage ~. and bacteriophage T7. From the teachings provided herein, it is apparent that many other proteins may be used as carriers. As shown by the use of the CBD fragment, an entire carrier protein need not be used, as 35 long as it is highly expressed in the host cell. For the sake of simplicity and economics, suitable carrier proteins should be as small as possible, around 100 amino acid residues or preferably less. In certain cases. it is desirable to use a carrier protein that either lacks cysteine residues or that contains no more than one cysteine residue. It is also desirable to avoid methionine residues except in the cleavage site if CNBr reagent is to be used to release the linked peptide..
To facilitate isolation of the cationic peptide sequence, amino acids susceptible to cleavage can be used to bridge the carrier protein, a cationic peptide moiety, and an anionic spacer peptide moiety in the mufti-domain .protein. The determination and desigTn of the amino acid sequence of the cleavage site is hijhly dependent on the strategy of cleavage and the amino acid sequence of the cationic io peptide, anionic spacer peptide and carrier protein. The removal of the cationic peptide can be accomplished through any known chemical or enzymatic cleavages specific for peptide bonds. Chemical cleavages include (R. A. :lue & R. F. Doolittle, Biochemistry, (1985) 24: 162-170; R.L. Lundblad, Chemical Reagents for Protein Modification (CRC
Press, Boca Raton, FL; 1991 ), Chapter 5.), but: are not limited to those treated by 15 cyanogen bromide cleavages at methionine (Met~~), N-chlorosuccinimide or o-iodosobenzoic acid at tryptophan (Trp~~), hydrox.ylamine at asparaginyl-glycine bonds (Asn~.Gly), or low pH at aspartyl-proline bonds (Asp~.Pro). Alternatively, there are a vast number of proteases described in the literature but the majority have little specificity for a cleavage site. Enzymatic cleavages which can be performed include, 1o but are not limited to those catalyzed by Factor Xa, Factor XTIa, thrombin, enterokinase, colla~enase, Staphylvcvccrrs arrrerr;~ V8 protease (endoproteinase GIu-C), endoproteinase Arg-C, endoproteinase Lys-C, chymotrypsin ortrypsin.
To express a cationic peptide gene;, a nucleic acid molecule encoding the peptide must be operably linked to regulatory sequences that control transcription and 25 translation (expression) in an expression vector and then introduced imo a host cell. In addition, expression vectors can include a marker gene which is suitable for selection of cells that carry the expression vector.
The expression vectors of the present invention comprise nucleic acid molecules encoding mufti-domain fusion proteins with more than one copy of a 3tf cationic peptide ~ene. As can be shown, mufti-domain fusion proteins having a carrier protein domain, an anionic spacer peptide component, and a cationic peptide component may include from two to more than 30 copies, of a cationic peptide gene.
Mufti-domain fusion proteins that lack an anionic spacer peptide component, but contain a carrier protein domain and a cationic. peptide component include two to four ;s copies of a cationic peptide gene. Moreaver, mmlti-domain fusion proteins that lack a carrier protein domain, but include both anionic spacer peptide and cationic peptide components may include from five to more than 20 copies of a cationic peptide ~,.:ene.
Preferably, cationic peptides are produced in prokaryotic host cells.
Suitable promoters that can be used to express polypeptides in a prokaryotic host are well-known to those of skill in the art and for example include T4, T3, SP6 and T7 promoters recognized by specific phase RNA polymerases, the irrt , PR and PL
promoters of bacteriophage lambda; the trp, recA., heat shock, IcrcLIiW , rat, Ipp-IacSpr, phvA, IacP, tacP, lrcP, .mpP, araP, and IcreZ promoters of E. toll, promoters of B.
srrblilis, the promoters of the bacteriophages of Bacillus, Streptunryces promoters, the to bla promoter of the ctrl promoter. Prokaryotic promoters have been reviewed by Glick, .l. Inch Microhivl. 1:277, 1987, Watson el aL, Mulecrrlar Biology of the GerTe, -!th Ed (Benjamin Cummins 1987); and by Ausubel et al. (1995).
Preferred prokaryotic hosts include E. cull; Bacillus srrbtilrrs, and Staphylucoccnrs aureus. Suitable strains of E. toll include BL21(DE3), t5 BL2-1(DE3)pLysS, BL2I(DE3)pLysE, DH1, DH:41, DHS, DHSI, DHSIF', DHSIMCR, DH10B, DHl0Blp3, DHIIS, C600, HBlOI, JIv1101, JMI05, JM109, JM110, K38, RRI, Y1088, Y1089, CSH18, ERi451, and ER1647 (see, for example, Brown (Ed.), Molecular Biology Labfax (Academic Press 1991)). Suitable strains of Bacillus srrhtihrs include BR151, YB886, MIl I9, MI120, and B170 (see, for example, Hardy, 20 "Bacillus Cloning Methods," in DIVA (_'.lorring: A Practical .Approach, Glover (Ed.) {IRL Press 1985)). An illustrative strain of Staphylococcus arrreus is RN4220 {Kreiswirth et al., Nature 30:709, 1983). The; present invention does not require the use of bacterial strains that are protease deficient.
An expression vector can be introduced into host cells using a variety of 25 standard techniques including calcium phosphate transfection, microprojectile-mediated delivery, electroporation, and the like. Methods for introducing expression vectors into bacterial cells are provided by Ausubel (1995). Methods for expressing proteins in prokaryotic hosts are well-known to those of skill in the art (see, for example, Williams et al., "Expression of foreign proteins in E. toll using plasmid vectors and purification 30 of specific polycional antibodies," in DNA Cloning 2: Expression Systems, 2nd Edition, Glover et al. (eds.), page 15 {Oxford University Press 1995); and Georgiou, "Expression of Proteins in Bacteria," in Protein Errgineerirrg: Principles and Practice, Cleland et al. {eds.), page 101 {John Wiley & Sons, Inc. 1996)).
Cationic peptides can also be expressed in recombinant yeast cells.
35 Promoters for expression in yeast include promoters from GAL! (galactose), PGK
(phosphoglycerate kinase), ADH (alcohol deh:ydrogenase), AOXI (alcohol oxidase), NIS-l (histidinol dehydroUenase), and the like. Many yeast clonin' vectors have been designed and are readily available. These vector's include YIp-based vectors, such as YIpS, YRp vectors, such as 1'Rp 17, YEp vectors such as YEp 13 and YCp vectors, such as YCpl9. One skilled in the art will appreciate that there are a wide variety of suitable vectors for expression in yeast cells.
The baculovirus system prow ides an efficient means to express cationic peptide ~,enes in insect cells. Suitable expression vectors are based upon the .9rrtotrcrpha ccrlifornica multiple nuclear polyhedrosis virus (AcMNPV), and contain well-known promoters such as Drosophila heat shock protein (hsp} 70 promoter, tn .arrtographa californiccr nuclear polyhedrosis virus immediate-early gene promoter (ie-I) and the delayed early 39K promoter, baculovirus p10 promoter, and the Drosophila metallothionein promoter. Suitable insect host cells include cell lines derived from 1PLB=Sf 21, a Spodoptera-frrrgiperda pupal ovarian cell line, such as S~ (ATCC
CRL
1711), Sf2lAE, and Sf21 (Invitrogen Corporation; San Diego, CA), as well as is Drosophila Schneider-2 cells. Established techniques for producing recombinant proteins in baculovirus systems are provided by Bailey et aL, "Manipulation of Baculovirus Vectors," in Methods in Molecular Biology, Yf~lume .'": Gene Transfer at:d ~,XprLSSIOJrPIYJtUCOjS, IVIurray (ed.), pages 147-158 (The Humana Press, lnc.
1991), by Patel et aL, "The baculovirus expression syste;m," in DNA Gloniog 2:
Expression zo .fy'stems, 2nd Edition, Glover et al. (eds.), pages 205-244 (Oxford University Press 1995), by Ausubel (1995} at pages 16-37 to 16-57, by Richardson (ed.);
Bacrrlovirus Expressiorr Protocols (The Humana Press, Inc. 1995), and by Lucknow, "Insect Cell Expression Technology," in Protein Errgineerin~;: Principles and Practice, Cleland et al. (eds.), pages 183-218 {John Wiley & Sons, Inc. 1996). Established methods for 25 isolating recombinant proteins from a baculovinis system are described by Richardson (ed.), Baccrlovirrrs ~rpres.siorr Protocols (The Humana Press, Inc. 1995).
The recombinant host cells are cultured according to means known in the art to achieve optimal cell growth. In the case of recombinant bacterial hosts, preferably E. coli, the bacteria are introduced into a suitable culture medium containing 3t~ nutrient materials that meet growth requirements. After inoculation, the bacteria continue to divide and grow until reaching a concentration, saturation density. For example, shake flask fermentation may require 1.5-17 hours at 30°C to reach this point.
Then the bacteria are diluted 1:3 in fresh medium and allowed to grow to mid or late exponential phase, at which time synthesis of the cationic peptide is induced.
There are 35 several methods of inducing the bacteria to synthesize the relevant recombinant proteins. Suitable induction conditions will vary .with the strain of ~ toll and the plasmid it contains. For example, in the case of 'temperature-dependent induction, the induction is obtained by raising the temperature to 42°C and maintaining it from about 1 to about ~ hours at a preferred pH range of 6.5-7.2. When th,e expression of the desired ~'ene reaches optimum levels the bacteria are harvested and the cells are either frozen or continue through the recovery process.
b. Illustrative Vectors Having; a Nucleutirle .feyuence Eneocling a Carrier Protein-Cationic F'epti~le As described in detail in the examples, plasmids were constructed that contained illustrative carrier protein genes. Briefly, plasmid vector pET2la(+), a T7 1b expression plasmid {Novagen Corporation, USA), was used as the core piasmid for initial studies (Figure lA). Plasmid pET-CBD180 (see Shpigel et al., Biotech.
Bioeng.
65:17-23, 1999) was used as the source for the gene encoding the cellulose binding domain (CBD) carrier protein (Figure I B). A PCR. reaction was designed to amplify a fragment containing the CBD180 gene from opET-CBD180 as a 646 by fragment 15 (Figure 1 C). A BamHI restriction site (GGATCC) was incorporated at the 3'-end of the CBDI80 PCR fragment. The BgIII or Xhal sites of pET-CBD180 and BamHI were used to cleave the PCR fragment and two fragments were separately ligated into pETZI a(+) resulting in two plasmids pET21 CBD-B and pET21 CBD-X, respectively (Figure I D). Plasmid pET 21 CBD-X contains l~acU from pET21 a(+), which improves 2o the regulation of the T7 expression system. Both plasmids contain a stop codon downstream of BamHI to allow expression of CBD 180 protein. A T7 expression system was prepared in E. cvli MC4100 based on ,pGPI-2, which carries the T7 RNA
polymerase gene under a ~,R promoter controlled by cI857 thermo-sensitive repressor.
CBD180 protein was expressed at high levels in both systems. Plasmid pET2ICBD-X
25 was used for subsequent studies.
Indolicidin is a natural 13-amino acid antimicrobial cationic peptide present in the cytoplasmic granules of bovine ne;utrophiIs and has a unique composition consisting of 39% tryptophan and 23% proline. Initial studies used two cationic peptides derived from modifications of indolicidin, MBI-11 peptide (I L K K W
_P W
3o W P W R R K) and MBI-1 IB7 peptide {I L R W P W W P W R R K), as described by FaiIa et al., WO 97/08199, and by Fraser et a~C, WO 9?/07745. A gene encoding the indolicidin-type cationic peptide MBI-11 was synthesized with BamHI and HindIII
cloning sites, fused to CBD180 carrier protein and expressed. The level of expression was high and equal to that of CBD180 alone. Next, a tandem of two MBI-11 genes 35 (2x11) was fused to CBD180, and again high expression was achieved. In order to increase the ratio of cationic peptide to carrier protein, the 177 amino acids of CBD180 were truncated to 96 amino acids. and this version of the carrier protein.
desi~~nated CBD96, was used as a new carrier protein. The E)NA fra~~ment carryin~,~ CBD96 was prepared by PCR: usinU pET-CBD 180 as a template, and cloned into pET21 a(+) resulting in plasmid pET? 1 CBD96. Both sin~,.:le and double copies of the i~lBt-1 I gene a were fused to CBD96 and expressed at hiVh levels. Then poly genes containing up to ten MBI-1 I units were prepared. However, expression was only achieved with a fusion protein containin' four MBI-11 genes in tandem. A dramatic decrease in expression was encountered when the number of genes exceeded three (Figures 2 and 3).
c, pllustratine Vectors Hawing a Nucleotirle Serleeenc:e E'ncorling a pi Carrier Protein-Cationic Peptide l~tJz Anionic Spacers In another approach, vectors were .constructed comprising mufti-domain fusion proteins with small anionic peptide spacers between the cationic peptide domains. This method of construction of mufti-domain ,genes allows the polymerization of any cationic peptide gene without changing its amino acid sequence.
15 In initial studies, the MB1-1 1B7 cationic peptide vows used.
Three distinct DNA cassettes specifying NLBI-11B7 cationic peptide _~enes and a negatively charged peptide spacer .were synthesized: I 1 B7-poly, anionic spacer, and 2x11B7-last (Figure 4), and cloned into appropriate plasmid vectors.
Cassettes of IlB7-poly and spacer were linked together resulting in the 11B7poly-2e spacer cassette. The anionic spacer peptide and cationic peptide genes were separated by codons for Met to create sites for cleavage by cyanogen bromide (CNBr). Two codons, specifying Ala and a stop codon, were linked to the last ZxIlB7 gene.
The 2xI1B7-last cassette was then cloned downstream of the gene encoding CBD96 in pET21CBD96 (Figure 5) resulting in plasrnid pET21CBD96-2x1 iB7. This plasmid 25 was later used in the construction of several fused mufti-domain genes. The 1 lB7poly-spacer cassette was used in a serial cloning procedure which allowed polymerization of 11B7 genes into mufti-domain fusion CBD!~6-spacer-poiyllB7 proteins in the pET21CBD96 expression system (Figures 6). All's mufti-domain constructs containing n copies (where f~ = 3 to 30) of MBI-11B7 genes and (rr-2) spacers were expressed at high levels. Examples of expression are shown in Figures 7. In order to accelerate the serial cloning procedure a polymerization cassette containing five 1 I B7 domains and five anionic spacer domains was prepared and used for construction of multidomain genes containing more than fifteen 11B7 domains ( i.e., 20 copies, 25, 30, etc.). This cassette has an anionic spacer domain at the end followed by a stop codon. Use of this ;5 cassette allowed construction of CBD96-based rnulti-domain systems containing equal numbers of I IB7 and spacer domains.

~l. llleistrcttire t%cturc Hc~rin~l a Nucleuti~le .SeeJt~ence Eneurling ct Cationic Pepticle W itlr Anionic .Spacers, But I rtcking ~t Carrier Protein One series of the mufti-domain proteins comprises jr times MBI-1187 peptides and n-2 anionic spacer peptides. When n=~, the molecular weight of the mufti-domain protein equals 13.46 l:Da, whiclh should be sufficiently large for expression in E. coli. DNA fragments containing mufti-domain genes of approximately this size were excised from relevant plasmids using restriction endonucleases Ndel and HindIII and fused into plasmid containing specifically designed leader 11B7 domain. In to E. coli, the first methionine in all proteins is translated as formy!-methionine which cannot be cleaved by CNBr. Accordingly, the carrier-free mufti-domain proteins were modified in such a way that the first domain begins with M-T-M amino acids, allowing CNBr to cleave the first peptide at the second met.hionine and release authentic peptide.
The relevant portions of plasmids pET2I -3 S-Sx 11 B7 and pET21-5 S-7x I 1 B7 are shown 15 in Figures 8. All of the carrier-free mufti-domain constructs containing from 5 to 14 copies of M1B1-11B7 genes were expressed at high levels as shown in Figure 7.
In the same way, constructs were prepared containing an equal number of 11 B7 and anionic spacer domains with a spacer sequence at the end. They were also expressed at high levels. The theoretical yield of the N1BI-11B7 peptide, within experimentally obtained 2o mufti-domain proteins, can be seen in Table 2.
The invention also provides .an additional example of another antimicrobial cationic peptide (IvIBI-26), twice the size of the peptide described above (MBI-1 IB7), consisting of 26 amino acids, where seven of them are basic amino acids.
This peptide was artificially designed by a fusion between selected sequences of the 25 natural antimicrobial cationic peptides cecropin and melittin. In the present invention, the last amino acid serine at the carboxy end ws~s replaced with a methionine residue, which was used for release of the peptide lfrom the mufti-domain protein. The production of this peptide was obtained by recombinant synthesis in host cells, using the mufti-domain protein method, as described above for NIBI-I1B7 peptide.
Details 3o are provided in Example 8.
TABLE 2, SUMMARY OF SUCCESSFULLY EXPRESSED CONSTRUCTS* AND THEIR THEORETICAL
MBI-I I B7CATIONIC PEPTU~E RATIO >r1 TI-IE MULTI-DOMAIN PROTEINS, WITH AND WITHOUT CARRIER PROTEIN

Construct l.lulti-domain'%~ Cationic Peptide per Protein i1-las~~i~lulti-domain Protein ( pad ( DaIDa) With Carrier Protein pET2ICBD-1187 21.2:19 8.9 pET21 CBD-2x 11 B7 23.142 16.~

pET21CBD96-11B7 12.697 15.0 pET21 CBD96-2s 11 B7 14.590 26.1 pET21CBD96-1S-3aI lB7 17.718 32.3 pET21 CBD96-2S-~~ 11 20_84 '6.6 pET21 CBD96-3S- 11 B7 23.973 39.8 pET2 I CBD96-45-6x I 27.101 -12.2 pET21 CBD96-SS-7x 11 30.228 '1'12 pET21 CBD96-6S-8x 1 33.356 45.8 pET21 CBD96-7S-9x 11 36.184 47.1 8S-IO~c11B7 39.612 48.2 pET2lCBD96- .

pET21CBD96-9S-I1~11B7 42,739 ~ 49.1 pET21CBD96-IOS-12a11B7 45.867 49.9 pET21CBD96-11S-13xI1B7 48.995 X0.6 pET21CBD96-12S-14sI1B7 52.122 51.3 pET21CBD96-13S-lSxIlB7 -250 aI.B

pET21CBD96-185-20~11B7 70.888 53.9 23S-25aI1B7 86.527 awl pET21CBD96- .

pET21 CBD96-28S-30s 102. I 62 56. I

With egual spacers number-pET21CBD96-SS-~~11B7 26.282 36.3 pET21CBD96-lOS-IOx11B7 41,921 '155 135-15x11B7 57.559 a8.9 pET21CBD96- .

Without Carrier Protein- -pET21-3s-SS 11 B7-F 13,692 69.7 pET21-4s-6~IIB7-F ~ 16,820 68.1 pET21-SS-7~ 11 B7-F 19,947 67.0 pET2 I -6s-8x 11 B7-F 23, 075 66.2 pET21-7s-9x 11 B 7-F 2 6, 203 65 . 6 pET21-8s-1 Ox 11 B7-F 29,3 30 65.1 Construct ~tulti-domain '%~ Cationic Peptide Protein Mass per (Da) Multi-domain Protein (Da/Da) pET21-9s-1 1 a 1 t 3?.458 Hd.7 p ET21- I Os-121118 pET21-l ls-l3sl 1B7-F 38,713 6~t.i pET2l-12s-1~x11B7-F ~I.8~1 pET21-19s-21 s I 1 63.73 ~ 61.9 With eyual .spacers - -numher p ET21-bs-6x 11 B 7-F / 9.129 ~ 'w pET2 I - l l s-11 a '4.767 H~''1 pET21-16s-16x1 1 B7-F s0.'lO~

* Examples of the expression can be seen in Figure 7.
4. Purification and Assay of (:ationic Peptides Produced by Recombinant Host Cells g General techniques for recovering protein produced by a recombinant host cell are provided, for example, by Grissha~mmer et al., "Purification of over-produced proteins from E. eoli cells," in DNA ~C.IUTI!)ig 2: Expression Systems, 2nd Edition, Glover et al. (eds.), pages 59-92 (Oxford University Press 1995), Georgiou, "Expression of Proteins in Bacteria," in ProteirT .E:ugineering: Pritrciples crud Practice, to Cleland et aJ. (eds.), page 101 (John Wiley & Sons, Inc. 1996), Richardson (ed.), BaCtlIGVTrtlS Expressiot? Protocols (The Humana Press, Inc. 1995), and by Etcheverry, "Expression of Engineered Proteins in Mammalian Cell Culture," in Protein Engineering:
Pritrciples afrd Practice, Cleland et al. (eds.); pages 163 (Wiley-Liss, Inc.
1996).
Variations in cationic peptide isolation and purification can be devised by Chase of skill 15 in the art, including, for example, affinity chromatography, size exclusion chromatography, ion exchange chromatography, HPLC and the like (see, for example, Selsted, "HPLC Methods far Purification of Antimicrobial Peptides," in Afrtihacterial Peptide Protocols, Shafer (ed.) (Humana Press,, Inc. 1997)). Particular purification methods are described below:
The present invention provides a novel, scaieable, cost-effective purification process far recombinant production of cationic peptides in host cells. The multi-domain fused polypeptide forms an insoluble complex in E. coli called the inclusion body. After the bacteria are mechanically disrupted, these inclusion bodies can be separated from the soluble components of the cell, according to means known in the an such as filtration or precipitation. Host impurities can be removed usin~~ solvents such as detergent (Triton X-100), enzyme (lvsozvrne. DNAse) and salt.
Cationic peptides can be released from anionic spacer peptides and carrier protein (such as truncated CBD) using standard techniques. if methionine s residues have been included at desired cleavage points, for example, chemical cleavage with cyanogen bromide (CNBr) reagent in an acidic environment can be used. The reaction can be performed in 70% formic acid or 70% formic acid and 0.1 N HCl or 70% TFA {trifluoroacetic acid). At the end of the reaction, which can last 4-15 hours, the reaction mixture is diluted in water, preferably I S times the volume of the reaction W mixture, and then dried. At this stage, the carboxyl terminus of the cationic peptide is present as homoserine lactone.
The isoelectric point of polvcationic peptides is very high ( 10.5-12.5) which enables the development of a very unique; purification process using an anion exchange chromatography column under unusual conditions. Almost any anion 15 exchange resin coupled to weak or strong canon ligand, particularly those used for industrial purpose to purify proteins, peptides, carbohydrates and nucleic acids, can be used in the following purification process for cationic peptides. This procedure requires the,use of only one chromatography step to obtain 95% purification.
The advantages of this chromatography are that it is short, fast, does nit require high 2ct pressure equipment and can be performed without organic solvents. The preferred procedure relies on dissolving the dried cleavage materials in 7- 8 M urea (alternatively in 50% ethanol or water) and loading them onto an anion exchange column. At this stage, the pH of the loading sample is acidic {pH 2-3). The column is previously washed with two column volumes of 0.5-1 M NaOH and one short wash in water to a 25 conductivity of less than 10 mS, preferably less than lmS, detected at the exit of the column. If the dried cleavage materials have been dissolved in 8 M urea, one column wash with 8 M urea before loading is preferred. 'The cationic peptide, in contrast to the impurities, passes through the column whereas the impurities are bound to the resin and thus separated (Figure 9). At this stage, the carboxy terminus of the cationic peptide has 3c} been converted and appears as homoserine. In addition, the pH of the cationic peptide sample has changed from acidic to basic (above pH I 1).
If the dried cleavage materials loaded on the anion exchange column are in the presence of 7- 8 M urea, the flow through purified peptide will be in urea solution, which can be separated and further purified by high-throughput reverse phase 35 chromatography using the perfusive supports foros 20 or 50 R-2 resin (PerSeptive 3(3 -Biosvstems Inc.}. For mass production. Poros ~0 is preferred due to better~t~ow and the fact that the use of high pressure equipment is avoided..
.W other more common. but more e:~pensive procedure can be performed according to means known in the art. such as reverse phase chromatography where the dried cleavage materials may be dissolved in water or 0.1% TFA and loaded onto a C8 or C 18 column using the RP-HPLC technique. However, this method requires high pressure equipment and organic solvents and results in a cationic peptide with a C-terminal homoserine lactone.
In the studies described above, thf: recombinant cationic peptide MBI
11B7 was obtained from a mufti-domain construct. As a result, CNBr cleavage causes the formation of a homoserine lactone residue at the carboxy end which may be easily converted to homoserine by raising the pH. This c:arboxy terminus is different from the bactericidal amidated chemical synthetic I~~IBI~~11B7CN. Hence, the antimicrobial activity was compared between chemically andl recombinantly synthesized cationic peptide.
' There are various in vitro methods for determining the activity of a cationic peptide, including an agarose dilution MIC assay, a broth dilution, time-kill assay, or equivalent methods (see, for example., Shafer (ed.), ~ntibarterial Peptidx Prvtoculs (Humana Press, Ine. 1997)). Antibiotic activity is typically measured as 2o inhibition of growth or killing of a microorganism or a microorganism-infected cell.
For example, a cationic peptide is ~E'irst dissolved in Mueller Hinton broth supplemented with calcium and magnesium, and then this solution is mixed with molten agarose. Other broth and agars may be used as long as the peptide can freely diffuse through the medium. The agarose is poured into petri dishes or wells and allowed to solidify, and a.test strain is applied to the agarose plate. The test strain is chosen, in part, based on the intended application of the peptide. Plates are incubated overnight and inspected visually for bacterial growth. A minimum inhibitory concentration (MIC) of a cationic peptide is the lowest concentration of peptide that completely inhibits growth of the organism. Peptides that exhibit acceptable activity 3o against the test strain, or group of strains, typically having an MIC of less than or equal to 16 l.y/ml, can be subjected to further testing.
Alternatively, time kill curves can be used to determine the differences in colony counts over a set time period, typically 24 hours. Briefly, a suspension of organisms of known concentration is prepared and a cationic peptide is added.
Aliquots of the suspension are removed at sca times, diluted, plated on medium, WO 00/31279 PCTlCA99/01107 incubated, and counted. ~1IC is measured as the lowest concentration of peptide that completelv_ inhibits growth of the organism.
Cationic peptides may also be tested for their toxicity to normal mammalian cells. An exemplary assay is a red blood cell (RBC) (erythrocyte) hemoiysis assay. Briefly, in this assay, red blood cells are isolated from whole blood, typically by centrifugation, and washed free of plasma components. A 5% (v/v) suspension of erythrocytes in isotonic saline is incubated with different concentrations of cationic peptide. After incubation for approxinnately one hour at 37°C, the cells are centrifuged, and the absorbance of the supernatant: at 540 nm is determined. A
relative to measure of lysis is determined by comparison to absorbance after complete lysis of erythrocytes using NHaCI or equivalent (establishing a 100% value). A peptide with less than 10% lysis at 100 glml is suitable. Preferably, the cationic peptide induces less than 5% lysis at 100 glml. Cationic peptides that: are not lytic, or are only moderately lytic, are desirable and suitable for further screening. In vitro toxicity may also be t, assessed by measurement of toxicity towards cultured mammalian cells.
Additional in vitro assays may be carried out to assess the therapeutic potential of a cationic peptide. Such assays include peptide solubility in formulations, pharmacology in blood or plasma, serum protein binding, analysis of secondary structure, for example by circular dichroism, liposome permeabilization, and bacterial 2o inner membrane perrneabilization.
In the present case, the antimicrobial activities of MBI-11B7CN, MBI-11B7HSL (homoserine lactone form) and MBI-11B7HS (homoserine form) were tested against various gram-negative and gram-positive; strains, including antibiotic resistant strains. The assay was performed as described in "Methods for Dilution Antimicrobial z5 Susceptibility Tests for Bacteria That Grow Aerobically-Fourth Edition;
Approved Standard" NCCLS document M7-A4 (ISBN 1-:56238-309-4) Vol. 17, No. 2 {1977).
Determination of .the minimum inhibitory concentration (MIC) of the peptides, demonstrated that MBI-11B?HSL and MBI-11B7HS peptides maintain similar bactericidal activity to the amidated MBI-11B7CN peptide. See Table 3 in Example 30 1~.
Cationic peptides can also be testE:d ira vivo for efficacy, toxicity arid the like. The antibiotic activity of selected peptides may be assessed in vivo for their ability to ameliorate microbial infections using a. variety of animal models.
A cationic peptide is considered to be therapeutically u;sefui if inhibition of microorganism 35 growth, compared to inhibition with vehicle alone, is statistically significant. This measurement can be made directly from culture; isolated from body fluids or sites, or - - WO 00/31279 PC'f/CA99/01107 indirectly, by assessing survival rates of infected animals. For assessment of antibacterial activity, several animal models one available, such as acute infection models includinv 'those in which (a) norm2~1 mice receive a lethal dose of microorganisms, (b) neutropenic mice receive a lethal dose of microorganisms, or (c) rabbits receive an inoculum in the heart, and chronic infection models. The model selected will depend in part on the intended ciinic.al indication of the cationic peptide.
As an illustration, in a normal mouse model, mice are inoculated ip or iv with a lethal dose of bacteria. Typically, the dose is such that 90-100% of animals die within two days. The choice of a microorganism strain for this assay depends, in part, W upon the intended application of the cationic peptide. Briefly, shortly before or after inoculation (generally within b0 minutes), cationic peptide is injected in a suitable formulation buffer. Multiple injections of canionic peptide may be administered.
Animals are observed for up to eight days post-infection and the survival of animals is recorded. Successful treatment either rescues animals from death or delays death to a IS statistically significant level, as compared with non-treatment control animals.
lu. aiv~, toxicity of a peptide can be measured by administration of a range of daces to animals, typically mice, by a route defined in part by the intended clinical use. The survival of the animals is recorded and LDS°, LDP,.,°~, and maximum tolerated dose (MTDj can be calculated to enable comparison of cationic peptides.
2p Low immunoaenicity of the cationic peptide is also a preferred characteristic for in vivo use. To measure immunogenicity, peptides are injected into normal animals, generally rabbits. At various times after a single or multiple injections, serum is obtained and tested far antibody reactivity to the peptide analogue.
Antibodies to peptides may be identified by ELISA, immunoprecipitation assays, 25 Western blots, and other methods (see, generally, Harlow and Lane, Antibodies: A
Labvrato~y Mafrual, (Cold Spring Harbor Laborz~tory Press, I 988)).
Expression vectors comprising the mufti-domain fusion proteins described herein can be used to produce mufti-domain fusion protein representing more than 25% of the total protein of a recombinant host cell. Since the mufti-domain fusion ;o proteins comprise multiple copies of a cationic peptide gene, the cationic peptide component of a fusion protein can be practically attained as more than SO% of the fusion protein.
The present invention, thus generally described, will be understood more 35 readily by reference to the following examples, which are provided by way of illustration and are not intended to be limiting of the present invention.

EXA~'~1PLE 1 Co~srRnr~r~oN or Pt.~~MiDS uET2ICBl~-X (B) .~.NO PET21CBD96 _; Plasmid vector pET21 a{+} (Nova~en Corporation, USA), a T7 expression plasmid, was used as the core plasmid for all expression systems (Figures I A and 10). The cellulose binding domain (CBD) from ~'lmn'idium cellulovor-an~ was selected as a carrier protein for expression of antibacterial cationic peptides. Plasmid pET-CBD180 (Shpigel et al., .nrpna) was used as the starting material (Figures 1B and lc~ 10). Restriction enzymes except I~ :~pl and N.sil (Promega Corporation, USA), T4 DNA
ligase and Tad polymerase were purchased from Pharmacia Biotech. The relevant part of CBD, including the T7 promoter of pET-CBD 180, was amplified by PCR using pmol each of each of the primers GCGT CCGG C;GTA GAGG ATCG (SEQ ID NO:1) and CCGG GATC CHAT GTTG CAGA AGT A.G (SEQ ID N0:2), 2 U of Tay DNA
15 polymerase, corresponding reaction buffer (tOx PCR reaction buffer: X00 mM
KCh 15 mM MgCI,, 100 mM Tris-HCI, pH 9), 0.2 mM dNTPs (dATP, dGTP, dTTP and dCTP, Pharmacia Biotech) and 20 ng of heat-denatured pET-CBD180. PCR was performed in MJ-Research PTC-100 Thermo-cycler in 50 ~I reaction volume and 30 cycles of 94°C, 30 sec.; SS°C, 30 sec. and 72°C, 30 sec.
A BamHI restriction site 2c~ (GGATCC) was incorporated at the 3'-end of the chd gene to allow it to be cloned into pET2la(+). The Bglll (AGATCT) or Xbal (TC~CAGA) sites already present on pET-CBD180 were used to cut the 5'-end of the PCR i:ragment. One ~g of PCR product was digested in a 100 ~1 reaction containing 1.5 x t)PA (Pharmacia Biotech assay buffer One-Phor-All is supplied at 1 Ox concentration: 100 mM Tris-acetate, pH 7.5;
100 mM
25 magnesium acetate and 500 mM potassium acetate) and 10 U of BanrHI and 10 U
of HindIII. Plasmid pET21 a(+) was digested in the same way, in 2 x 50 ~tl reactions each containing 0.25 ~g of plasmid DNA, l.Sx OPA and 2 U of BamHI and HiudIIl each.
Reactions were stopped by phenol/CHC1; extraction and ethanol precipitation.
The resultant DNA pellets of digested pET21 a(+) and relevant cbd and cbd96 inserts were 3o dissolved in 8 lxl of water and mixed, then 2 pl of 10 mM ATP, 2 ~cl of l Ox OPA and 2 U of T4 DNA ligase were added and reactions were incubated at 10°C for 1 hour. Then 2 pl of each ligation mixture were used to electroporate 40 ~l of E. coli XLl Blue (Promega Corporation) using a sterile Gene Pulser cuvette (0.2 cm electrode gap) and Gene Pulser electroporator apparatus (Bio-Rad Laboratories) set to 2.5 kV, 200 ohms 35 and 250 ~F. After an electroporation pulse, 1', ml of TB media (Maniatis, et al., Mclecarlar Glor~irtg: A Laboratory Manual. 2'~ Ed., (Cold Spring Harbor Laboratory Press 1989) was added to the cell suspension and bacteria were incubated for 1 hour at 37°C with ri~_=orous shaking. Then 10, ~0 and l00 pl of cell suspension were plated on i\-lacKonkey aVar (BBL, Becton Dickinson and Company, USA) plates with 100 yglml of Ampicillin and incubated overnight at 37°C. The next day, several colonies were s transferred to 2 ml of TB and cultivated at 37°C with vigorous shaking overnight. Then plasmid DNA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones contained plasmids pET2ICBD-B or pET2ICBD-X, respectively (Figure 10). Plasmid pET2ICBD-X contains lcrcCa, which improves the regulation of the T7 expression sy:>tem. Both plasmids contain a stop to codon downstream of BamHI to allow expression of CBD180 protein. A T7 expression system was prepared in E. cull MC4100F (Strain MC4100F was prepared by mating E. coli XL I Blue and E. coli MC4100; ATCC Number 35695) based on pET variants and pGPI-2, which carries the T7 RNA polymerase gene under a ~,R promoter controlled by cI857 thermo-sensitive repressor ('Tabor and Richardson, Biochemistry t5 82:1074, 1985). CBD180 protein was expressed at high levels in both systems.
Plasmid pET2ICBD-X was used for subsequent v~~ork.
Plasmid pET21 CBD96 (Figure 5) was prepared using the same PCR
conditions and cloning procedures. In this experiment the carrier protein CBD180 was truncated to about 96 amino acids. Therefore a. pair of PCR primers GCGT CCGG
2U CGTA GAGG ATCG (SEQ ID N0:3) and ATAT GGAT CCAG ATAT GTAT CATA.
GGTT GATG TTGG GC (SEQ ID N0:4) was used to prepare the relevant DNA
fragment encoding chd9b (Figure 5), which wa~~ then cloned into pET21 a(+).
Then again a T7 expression system was prepared in E. c«li MC4100F based on plasmids pET21CBD96 and pGPI-Z and protein CBL>96 was expressed at high levels.
25 pET2I CBD96 was used for most of the subsequent work.

CONSTR.UCT1ON AND EXPRESSION OF' CBD - MBI-I 1 FUSIONS

Sequences encoding all cationic peptides were designed from modified indolicidin, a natural anti-microbial peptide. Plasmids pET2ICBD-X and pET21 CBD96 (0.25 ~g each) were digested with 2 U of BamHI and 2 U of HindIII
in l.Sx OPA in 50 p.l reactions at 37°C for I hour. In the same way, a frajment encoding ',s MBI-11 was digested (Example 4) using about 1 lzj of DNA and 25U of BamHI
and HindIII each in a 100 yI reaction. Both reactions were stopped by phenol/CHCI, (Si;;ma-Aldrich Canada Ltd.> ewraction and ethanol precipitation. The resultant DNAs of each vector and MIBi-11 insert were dissolved vin 8 yl of water and mixed.
then 2 l.tl of 10 mM ATP, 2 yl of IOx OPA and ? U of T:~ DNA lipase were added and ligation reactions were incubated at 10''C for 1 hour. Thf:n 2 l.rl of each ii~ation mixture was used to electroporate 40 yl of 1:. coli ?Q.l Blue using sterile Gene Pulser cuvettes (0.2 cm electrode dap) and Gene Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 250 ~tF. After an electroporation pulse, 1 ml of TB media was added to the cell suspension and bacteria were incubated for I hour at 37°C with rigorous shaking. Then 10, 50 and 100 l.tl of cell suspension were plated on MacKonkey agar plates with .100 p io glml of Ampicillin (Sigma-Aldrich Canada Ltd.} and incubated overnight at 37°C. The next day, several colonies were transferred to 2 rnl of TB and cultivated at 37°C with vigorous shaking overnight. Then plasmid DNA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones contained MBI-11 fused to CBD180 or CBD96. Expression strains of E. cull I5 harboring plasmids pGPI-2 and pET2ICBD-11 or pET21CBD96-I 1 respectively were prepared by electroporation. Final strains were incubated overnight in 2 ml TB
at 30°C
with rigorous shaking and the next day 1 ml of cell suspension was diluted with the equal volume of fresh TB and cultivation temperature was increased to 42°C for a minimum of 2 hours. Samples of preinduced and induced cells were analyzed by SDS-2c~ PAGE. The level of expression of the fusion protein caring MBI-l f or 2x MBI- I 1 gene was high and equal to expression of CBD180 or C:BD96 alone.

This experiment was designed to test how many peptide genes in tandem can be fused to a carrier protein and expressed. It was necessary to create two DIVA
fragments encoding MBI-11, one for polymerization by DNA cloning and another one 3o as the last gene in the tandem. Therefore, the original DNA fragment encoding MBI-11 peptide with COOH end was modified in order to create the last gene in tandem (Example 4} and a new gene was designed for a specifrc cloning procedure, which allowed construction of multiple tandem peptide genes fused to CBD 180 or carrier proteins genes {Example 4). The clonin;; procedure resulted in addition of an 35 extra isoleucine to the MBI-I1 tandem sequences. Therefore in order to produce identical peptide molecules. an isoleucine codon was also added to the last gene sequence. C:~'Br wil!'be used to cleave the peptide from fusion proteins, which means that peptide molecules would have a homoserine lactone on the end. Therefore the last peptide gene was also modified to have a methionine followed by two tyrosines at the end for CNBr cleava~_>e in order to produce equivalent peptide products.
_; CBD180 and CBD96 fused peptide polygenes of up to 10 units in tandem were prepared. However good expression was only achieved with a fusion containing two and three MBi-1 l domains and practically stopped when the number of peptide genes exceeded four. DNA synthesis and construction of plasmids containing 1~-~IBI-11 polymers is described in Example 4.
EXAMPLE ~4 SYNTHESIS-AND MOUffICA'I'ION OF DNA FRAGMENTS ENCODING CAT10NIC PEPTIDES
The desired sequences were conventionally synthesized by the phosphoramidite method of oligonucleotide synthesis using the Applied Biosystems Mode! 391 DNA Symhesizer with connected chemicals and protocols. Desired oligonucleotides were used as templates in the PC:R reaction to produce double stranded DNA suitable for DNA cloning.
A: SYNTHESIS OF THE MBI-1 I DNA DOMAlIV
2p An oligonucleotide TTTA ACGG GGAT CCGT CTCA TATG ATCC
TGAA AAAA TGG (SEQ ID NO:S) CCGT GGTG GCCG TGGC GTCG TAAA
TAAG CTTG ATAT CTTG GTAC CTGC G (SEQ ID N0:6) was synthesized and used as a template for PCR using primers TTTA ACGG GGAT CCG TCTC ATAT G
(SEQ ID N0:7) and TAAG CTTG ATAT CT'L~G GTAC CTGC G (SEQ ID N0:8).
The PCR was performed in MJ-Research PTC-1.00 Thermo-cycler in a SO pl reaction volume with .30 cycles of 94°C, 30 sec.; 50°C, 30 sec. and 72°C, 30 sec., 2 U of Taq DNA polymerase, corresponding reaction buffer (lOx PCR reaction buffer: 500 mM
KC1, 15 mM MgCI.,, 100 mM Tris-HCl pH 9), 0.2 mM dNTPs (dATP, dGTP, dTTP
and dCTP), 25 pmol of each primer and SO pmol of template oligonucleotide resulting 3o in an 88 by dsDNA MBI-11 fragment. DNA was used for the cloning procedure described in Example 2.

WO 00/31279 PCTICA9910i107 13. :'IIODIFIG;4770N l)F lLll3/-I1 I)llM.Al:''.4.5 7h'E I.4ST DOMAIN IN
TAnrDEM
PCR was used to modify the original DNA fray=ment encoding NfBI-1 l for use as the last gene in the tandem polvpeptide Y~ene. The original oligonucleotide (A) was used as a template. The sense primer TTTA ACGG GGAT CCGT CTCA
s TATG (SEQ 1D N0:9) was identical to that~used in the synthesis PCR reaction, but a new antisense primer CGCG AAGC TTAA TAA'T ACAT AATT TTAC GACG CCAC
GGCC ACCA CGGC (SEQ ID NO:10) was desi;;ned to modify the end of the MBI-I l gene {for explanation, see Example 3). The PCR was performed in MJ-Research PTC-100 Thermo-cycler in a 50 l.Ei reaction volume with 30 cycles of 94°C
30 sec., 51°C, 30 its sec. and 72°C, 30 sec., 2 U of Tad DNA polymerase, corresponding reaction buffer (lOx PCR reaction buffer: 500 mM KCI, 15 mM M~CI~, 100 mM Tris-HC1 pH 9), 0.2 mM dNTPs (dATP, dGTP, dTTP and dCTP), 2_'~ pmol of each primer and 50 pmol of the template oligonucleotide. The PCR product was then cloned as a BamHI-Hir~ctIII
fragment into pBCKS(+). (Stratagene, USA;/ resulting in plasmid pBCKS-11.
Modification was verified by DNA sequencing.
C: SYNTHESIS OFMI3J-11 FRAGMENT DESIGNATED FOR THE POLYMERIZATION
CLONING PROCEDURE
An oligonucleotide CGCC A~GGG TTTT GCCA GTCA CGAC GGAT
CCGT CTCA TATG ATCC TGAA AAAA 'TGGC CGTG GTGG CCGT GGCG
2a TCGT AAAA TTAA TTGA ATTC GTCA TALC TGTT TCCT GTGT GA (SEQ ID
NO:11) was synthesized and used as a template for PCR using primers CGCC AGGG
TTTT CCCA GTCA CGAC (SEQ ID N0:12) and TCAC ACAG GAAA CAGC
TATG AC (SEQ ID N0:13). The PCR was performed in MJ-Research PTC-100 Thermo-cycler in a 50 ~l reaction volume with 30 cycles of 94°C, 30 sec., 51°C, 30 z5 sec. and 72°C, 30 sec., 2U of Taq DNA polymerase, corresponding reaction buffer (lOx PCR reaction buffer: 500 mM KCI, 15 mM Me;CI~, 100 mM Tris-HC1 pH 9), 0.2 mM
dNTPs (dATP, dGTP, dTTP and dCTP), 25 pmol of each primer and 50 pmol of template oligonucleotide resulting in 114 by dsDNA MBI-I 1-BE fragment. This fragment was cloned as a BamHI - EcoRI insert into vector pBCKS(+) resulting in ~o pBCKS-1IBE.
D. POLYMERIZATION CLONING PROCEDURE;
The copy of MBI-11 designed for the polymerization cloning procedure was cloned into pET21 CBD96-11 resulting in pET21 CBD96-2x 11. pBCKS-11 BE was digested with 2 U of BcrnaHl and I-:spI in 2x OP'A in 50 Etl reactions at 37°C for 1 hour and pET2 iCBD96-1 1 was digested with 2 U of BcrmHI and ~'Vcle~l in 2x OPA in a ~0 ~ti reaction at 37°C for 1 hour. Reactions were stopped~by phenol/CHCI;
extraction and ethanol precipitation. The resulting DNA pellets were dissolved in 8 ~tl of water each and mixed, then 2 Ltl of 10 m~~9 ATP. '- l.tl of l Ox OPA and 2 U of T4 DNA
li~,;ase were added and reactions were incubated at 10°C for 1 hour. Then 2 i.tl of the ligation mixture was used to eiectroporate 40 ~tl of E. cull XLI Blue using Gene Pulser cuvettes (0.2 cm electrode dap) and Gene Puiser (Bio-Rad lLaboratories) set to 2.5 kV, 200 ohms and 250 ~tF. After an electroporation pulse, 1 mll of TB media was added to the cell suspension and bacteria were incubated 1 hour at 37°C with rigorous shaking. Then 10, w 50 and 100 Etl of cell suspension mere plated on MacKonkey agar plates with 100 ~t g/ml of Ampicillin and incubated overnight at 37°C. The next day, several colonies were transferred to 2 ml of TB and cultivated at 37°C with vigorous shaking overnight.
Then plasmid DNA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones contained pET21CBD96-2x11.
The is ligation of compatible I~a~I and Nc~eI cohesive ends resulted to elimination of both restriction sites. At the same time, the insertion of the mbi-I lbe cassette introduced a new Nc~eI site, which allowed repetition of the. cloning procedure and insertion of another mbi-1 Ibe. This procedure could be repeated theoretically without limitation. In this particular case the serial cloning was repented nine times and constructs up to .
20 pET21CBD96-10x1 1 were prepared.

$YN'rI-IES1S OF DNA CASSETTES FOR CONSTRUC'flON OF FUSED AND UNFUSED MULTI
DOMAIN EXPRESSION SYSTEMS
A. SYNTHESIS OFMBI 2X11 B7 LAST CASSETTE
An oligonucleotide CGCC AGGCi TTTT CCCA GTCA CGAC GGAT
CCGT CTCA TATG ATTC TGCG TTGG CCG'T GGTG GCCG TGGC GTCG CAAA
ATGA TTCT GCGT TGGC CGTG GTGG I~CGT GGCG TCGC AAAA TGGC
3o GGCC TAAG CTTC GATC CTCT ACGC CGGA CGC (SEQ ID N0:14) was synthesized and used as a template for PCR using primers CGCC AGGG TTTT CCCA
GTCA CGAC (SEQ ID N0:15) and GCGT C~CGG CGTA GAGG ATCG (SEQ ID
N0:16). The PCR was performed in MJ-Research PTC-100 Thermo-cycler in a 50 pl reaction volume with ~0 cycles of 94°C, 30 sec.; 55°C, 30 sec.
and 72°C, 30 sec. 2 U of 7W p DNA polymerase, correspondin~~ reaction buffer ( 10~: PCR reaction buffer:
>00 mM KC1, I S mM M~CI,_, 100 mM Tris-HC1 pH 9), 4.2 ml~I dNTPs (dATP, dGTP, dTTP and dCTP), 2~ pmol of each primer and 50 pmol of template oli~~onucleotide resulting in 151 by dsDN.A MBI-11 fraUment. The PCR product was purified by a phenollCHCI; extraction and ethanol precipitation. The resulting DNA was dissolved in 100 p.i Ix OPA , 20U of f3amHI and 20U of Niraca'III and the reaction was incubated at 37°C for 2 hours. The vector pBCKS(+) (0.25 ~y;) was di'ested in the same way. Both reactions were stopped by phenol/CHCI; extraction and ethanol precipitation.
The resultant DNAs of each vector and MBI-l 1 insert were dissolved in 8 yl of water and jct mixed, then 2 p.l of 10 mM ATP, 2 yl of lOx OPA and 2 U of T4 DNA ligase were added and ligation reactions were incubated at 10°C for 1 hour. Then 2 pl of each li~ation mixture was used to electroporate 40 yl. of E. cull XL1 Blue using a sterile Gene Pulser cuvette (0.2 cm electrode gap) and Gene Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 250 ~F. After an elc~ctroporation pulse, 1 ml of TB media was added to the cell suspension and bacteria were incubated 1 hour at 37°C with rigorous shaking. Then 10, 50 and 100 p.l of cell suspension were plated on MacKonkey agar plates with 100 yg/ml of Ampicillin and incubated overnight at 37°C.
The next day, several colonies were transferred to 2 ml of TB and cultivated at 37°C
with vigorous shaking overnight. Then plasmid DNA was isolated and analyzed, 2U including DNA sequencing by methods known to those skilled in the art. The resulting plasmid was pBCKS-2x11B7. The insert was later recloned into pBCKS-V resulting in pBCKS-V-2x 11 B7.
B: SYNTHESIS O~' MBI-11 B7 POL Y CASSETTE
An oligonucleotide CGCC AGG(.i TTTT CCCA GTCA CGAC GGAT
2s CCGT CTCA TATG ATTC TGCG TTGG CCG'T GGTG GCCG TGGC GTCG CAAA
ATGC ATAA GCTT CGAT CCTC TACG t:CGG ACGC (SEQ ID N0:17) was synthesized and used as a template for PCR using primers CGCC AGGG TTTT CCCA
GTCA CGAC {SEQ ID N0:18) and GCGT CCGG CGTA GAGE ATCG (SEQ ID
N0:19). The PCR was performed in MJ-Research PTC-100 Thermo-cycler in a 50 ~tl 3U reaction volume with 30 cycles of 94°C, 30 sec.; 55°C, 30 sec. and 72°C, 30 sec., 2 U
of Tay DNA paIymerase, corresponding reaction buffer (lOx PCR reaction buffer:
500 mM KC1, 15 mM MgCI~, 100 mM Tris-HCl pH 9), 0.2 mM dNTPs (dATP, dGTP, dTTP and dCTP), 2~ pmoi-of each primer and 50 pmol of template oligonucleotide resulting in a 112 by dsDNA MBI-11 fragment. The resulting DNA fragment was cloned into pTZIgR (Pharmacia Biotech) as a l3cunHl-Hinc~ltl fra~~ment as described in para~=raph (A) resulting in plasmid pTZIBR-1 IB7Isoly.
C. .f YNTHESIS OF ANIONIC .S P.4 CER CA S.SETTE, An oligonucleotide CGCC aGGG TTTT CCCA GTCA CGAC GGAT
CCGT CTAT GCAT GAAG CGGA ACCG GAAG CGGA ACCG ATTA ATTA
AGCT TCGA TCCT CTAC GCCG GACG C (SEQ ID N0:20) was synthesized and used as a template for PCR using primers CGCC AGGG TTTT CCCA GTCA CGAC
(SEQ ID N0:2 i ) and GCGT CCGG CGTA GAGG~ ATCG (SEQ ID N0:22). The PCR
was performed in MJ-Research PTC-100 Thermo-cycler in a 50 yl reaction volume with 30 cycles of 94°C, 30 sec.; 55°C, 30 sec. wind 72°C, 30 sec., 2 U of Taq DNA
polymerase, corresponding reaction buffer ( 1 Ox PCR reaction buffer: 500 mM
KCI, mM MgCI~, 100 mM Tris-HCl pH 9), 0.2 m.M dNTPs (dATP, dGTP, dTTP and dCTP), 25 pmol of each primer and 50 pmol of teciiplate oligonucleotide resulting in a 97 by dsDNA MBI-I 1 fragment. The resulting DNA fragment was cloned into pBCKS-15 V as a BcrmHt-Hinc~itl fragment as described in paragraph (A), resulting in plasmid pBCKS-V-S.
D. .SYNTHESIS OF MBI-I I B7 FIRST CASSETT E
An oligonucieotide CGCC AGGG TTTT CCCA GTCA CGAC GGAT
CCGT CTCA TATG ACTA TGAT TCTG CGT'f GGCC GTGG TGGC CGTG GCGT
2o CGCA AAAT GCAT AAGC TTCG ATCC 'TCTA CGCC GGAC GC (SEQ ID
N0:23) was synthesized and used as a template for PCR using primers CGCC AGGG
TT TT CCCA GTCA CGAC (SEQ ID N0:24) and GCGT CCGG CGTA GAGG
ATCG (SEQ ID N0:25). The PCR was performed in MJ-Research PTC-100 Thermo-cycler in a 50 p.i reaction volume with 30 cycles of 94°C, 30 sec.;
55°C, 30 sec. and 72°
C, 30 sec, 2 U of Taq DNA polymerase, corresponding reaction buffer (lOx PCR
reaction buffer: 500 mM KCI, 15 mM MgCi~, 100 mM Tris-HCI, pH 9), 0.2 mM
dNTPs (dATP, dGTP, dTTP and dCTP), 25 pmol of each primer and 50 pmol of template oli~onucleotide resulting in a 114 by dsDNA MBI-11 fragment. The resulting DNA fragment was Cloned into pBCKS-V-S as a BamHI- Nsit fragment basically as 3o described in paragraph (A), resulting in plasmid pBCKS-V-11B7S-F. The only exception was that 2x OPA was used in the restriction enzyme digest reaction.

E. C~U:~'.S7RUC770NOFPL.9SMID PI3Ch:f-V
Plasmid pBCKS-V was prepared from pBCKS(+). The ~~oal was to eliminate all t :,pl restriction sites from the ori~~inal plasmid and use the resulting plasmid for cloning of some of DNA cassettes.
.About t ~y of pBCKS(+) was diested with t :,pl {Promega) in 50 yi reaction usin' 1 x OPA. The reaction was stopiped by phenol/CHC1; extraction and ethanol precipitation. The resulting DNA was dissolved in 50 ~tl of 1 x OPA, 0.2 mM
dNTPs and 1 U of Klenow polymerase. The reaction was incubated at 30°C, for 30 min. and then stopped by phenol/CHCI; extraction and ethanol precipitation..
DNA was is then dissolved in 50 ~tl of lx OPA, 0.5 mM ATP and 15U of T4 DNA ligase and the reaction was incubated at 10°C and after 4 hours stopped by incubation at 65°C for 30 min. Then 20 U of I',;~I was added to the reaction to digest any remaining pBCKS(+) molecules and after 3 hours incubation at 37°C, 2 ul of the ligation mixture were used to electroporate 40 tzl of E. cull MC4100F using a sterile Gene Pulser cuvette (0.2 cm electrode gap) and Gene Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 250 yP. After an electroporation pulse, i ml of TB media was added to the cell suspension arid bacteria were incubated 1 hour at 3?°C with rigorous shaking. Then 10, 50 and 100 l.il of cell suspension were plated on rrlacKonkey agar plates with 2S lzg/ml of Chloramphenicol and incubated overnight at 37°C. The next day, several colonies 2o were transferred to 2 ml of TB and cultivated at 37°C with vigorous shaking overnight.
Then plasmid DNA was isolated and analyzed by V.spI restriction analysis. All plasmids lacked I~,pI sites and their size corresponded with the calculated size of pBCKS-V.

CONSTRUCTION OF FUSED MULTI-DUN.IAIN EXPRESSION SYSTEMS
A. ColVSrRUCTroN of PET21 CBD96-2x11 Bar Plasmids pET21 CBD96 (0.25 ug) and pBCKS-2x I I B7 (2:5 fig) were digested with BcrmHI and HindIII in I.Sx OPA i:n a SO p.l reaction at 37°C for 1 hour 3o using 2 U of each restriction enzyme and 20U of each enzyme respectively.
Both reactions were stopped by phenoi/CHCI; extraction and ethanol precipitation.
The resulting DNAs were dissolved in 8 ~l of water and rriixed, then 2 ~l of 10 mM
ATP, 2 l.tl of lOx OPA and 2 U of T4 DNA iigase were: added and the ligation reaction was incubated at 10°C for 1 hour. Then 2 ~1 of the ligation mixture was used to 4z electroporate 40 yl of 1.:. cull XL 1 Btue usin4 a sterile Gene Pulser cuvette (0.2 cm electrode dap) and Gene Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 250 yF. After an electroporation pulse, 1 ml of TB media was added to the cell suspension and bacteria were incubated 1 hour at i7°C with rigorous shaking. Then 10, 50 and 100 l.il of cell suspension were plated om MacKonkey agar plates with .100 p g/rnl of Ampicillin and incubated overnight at 37°C. The next day several colonies were transferred to 2 ml of TB and cultivated at 37°C with vigorous shaking overnight.
Then plasmid DNA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones pET21CBD96-2x11B7 contained io tandem MBI-11 genes fused to cbdy6.
THE USE OF.SERIAL CLONING PROCEDURE FOR CONSTRUCTION OF FUSED
MULTI-DOMAIN PL.9 SMIDS
The idea of the serial cloning procedure is that the insertion of the BcrmHI- MB1-11B7-P-6'.~pl cassette into the BcrmHI - Ndel sites of pET21CBD96 '?x11B7 and subsequent mufti-domain clones al~Nays eliminates the original Ndel site by Ndell Y Cpl ligation and a new IUdeI site is introduced with each insertion, which together with BcrnrHI is used for the next cycle of cloniing:
Plasmid pET21CBD96-2xlIB7 (0.25 l.tg) was digested with 2 U of BamHI and NdeI in 2x OPA in 50 ~1 reaction at 37°C for I hour. Plasmid pBCKS-V
11B7S (2.5 fig) was digested in a 100 ~I reaction with 20 U of BamHI and V~pI
in 2x OPA -at 37°C for 1 hour. Both reactions were stopped by phenollCHCI, extraction and ethanol precipitation. The resulting DNAs were dissolved in 8 p.l of water and mixed, then 2 ~l of 10 mM ATP, 2 ~t of l Ox OPA and :z U of T4 DNA ligase were added and the ligation reaction was incubated at 10°C for 1 hour. Then 2 p.l of the ligation mixture were used to electroporate 40 ~1 of E. cull XLI Blue using a sterile Gene Pulser cuvette (0.2 cm electrode gap) and Gene: Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 25.0 ~tF. After an electroporation pulse, 1 ml of TB
media was added to the cell suspension and bacteria were incubated 1 hour at 37°C
with rigorous shaking. Then 10, 50 and 100 ~1 of cell suspension were plated on MacKonkey agar 3o plates with 100 p.g/ml of Ampicillin and incubated overnight at 37°C. The next day, several colonies were transferred to 2 ml of TEt and cultivated at 37°C
with vigorous shaking overnight. Then plasmid DNA was isolated and analyzed, including DNA
sequencing by methods known to those skilled in the art. Positive clones pET21CBD96-Is-3xI1B7 contained three MH~I-11 units with one spacer fused to cbd96. 'This was the f rst cycle of the serial cloning. In the next cycle pET21 CBD96-WO 00131279 PCT/CA99/Of I07 ls-3x1 iB7 and pBCKS-V-1 IBIS were used and clonin' was repeated resulting in pET21 CBD96-2s-~x 11 B7. Then pET21 CBD96-2s-~x 1 I B7 and pBCKS-V-1 IBIS
were used far the next cloning resulting in pET21 CBD96-3s-5x 1 1 B7 and so on.
In order to accelerate the serial cloning procedure plasmid pBCKS-V
5 5x I -I B7S was prepared and each cloning cycle would add five 1 I B7S
domains. First the i 1B7S insert of pBCKS-V-1 IB7S was reclonf:d into pTZl8R, resulting in pTZl8R
1 I B7S. Then this plasmid was used as the donor of the 11 B7S domain for the serial cloning into pBCKS-V-l IBIS using the BcrmHI-.NclellL:vpl strategy. The serial cloning procedure was repeated four times resulting in pBCKS-V-5S-5xI1B7S. The 5S
tc~ 5x11B7 cassette was tl;en used for construction of CBD96-fused systems containing more than fifteen 11 B7 domains and also CBI~96-fused multidomain systems with equal numbers of I 1B7 and anionic spacer domains {Table 2).
The cassette 5 S-5x 11 B'7 of pBCKS-V-5 S-5x 1 I B~ with anionic spacer domain at the end was cloned into pET21 CBD!a6 using BamHI and Kpul restriction 15 enzymes resulting in pET21CBD96-5S-5x11B7. In the second cloning cycle the same cassette was ligated as BamHI-Y:spI fragment of pBCKS-V-Sxl IBIS into BcrnrHI-Ndel.
sites of pET2 l CBD96-5S-5x l t B7 resulting in pET21 CBD96-l OS-l Ox 11 B7.
This can be repeated several times to receive constructs with 15, 20, 25 etc. 1 I B7 domains and equal numbers of anionic spacer domains. Conditions for restriction enzymes, ligation, 2et electroporation and analysis of recombinant plasrnids are described above.

CONSTRUCTION OF UNFUSED MULTI-DOMAIN EXPRESSION SYSTEMS

In E. coli, the first amino acid in all proteins is f methionine. However, this amino acid is not cleaved by CNBr, which means that one peptide domain released from a mufti-domain protein would start with f methionine. The solution was to create a modified MBI-I 1 cassette encoding f methionine and methionine in tandem at the 3o beginning of the peptide, so the second one would be cleaved by CNBr. The result was the synthesis of the special first domain in mufti-domain genes, cassette MBI-11B7F, encoding MTM amino acids at the beginning. This domain was fused to the spacer domain in pBCKS-V-S resulting in plasmid pBC;KS-V-11B7S-F.
Plasmid pBCKS-V-11B7S-F and the relevant pET21CBD96-multi 35 domain-11B7 plasmids were used for construction of unfused mufti-domain MBI-11B7~
genes. Mufti-domain genes were liberated from cbct96 by Nc~I - XhoI digestion and cloned into the b apl - .Ylx~l sites of pBCKS-V-11 B7S-F downstream of the 11 insert. This created a line of unfused mufti-domain 11 B7 genes in plasmid pBCKS-V.
These genes were then recloned as Nclel - .k7~ol fragments into pET21 a(+) resulting in a series of pET plasmids capable of expression of mufti-domain proteins using the T7 promoter system.
Plasmid pBCKS-V-11B7S-F (0.25 l.y) was digested with 2 U of NdeI
and Xhol in 2x OPA in several 50 yl reactions at: 37°C for 1 hour.
Relevant plasmids pET2lCBD96-multidomain-11B7 (2.5 ~~g) were digested in 100 ~l reactions with of Ndel and Xhol in 2x OPA at 37°C for 1 hour. All reactions were stopped by to phenol/CHCI; extraction and ethanol precipitation. The resultant vector and insert DNAs were dissolved in 8 yl of water and mixed, then 2 pl of lO mM ATP, 2 ~tl of l Ox OPA and 2 U of T4 DNA lipase were added and ligation reactions were incubated at 10°C for 1 hour. Then 2 yl of each ligation mixture was used to electroporate 40 y~l of E. cola XL.1 Blue using a sterile Gene Pulser cuvette (0.2 cm electrode gap) and Gene Pulser electroporator apparatus set to 2.5 kV, 200 ohms and 250 pF. After an electroporation pulse, 1 ml of TB media was added to the cell suspension and bacteria were incubated 1 hour at 37°C with rigorous sharking. Then 10, 50 and 100 p.l of cell suspension were plated on MacKonkey agar plates with 100 ~giml of Ampicillin and incubated overnight at 37°C. The next day several colonies were transferred to 2 ml of 2o TB and cultivated at 37°C with rigorous shaking overnight. Then plasmid D.NA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones contained pET21-multidomain-I1B7 plasmids containing 5, 6, 7, 8, 9, 10, 1 I, 12, 13, 14, 15, 16 and 21 MBI-11137 domains.
In the same way, constructs were prepared containing equal numbers of I1B7 and anionic spacer domains. By way of illustration: pET2ICBD96-SS-Sx11B7 was digested with BarrrHl and XhoI (or HiradIII) and fragment SS-Sxl 1B7 was ligated into BarnHl-Xhol (or HiradIII) of pBCKS-V-I1B7S-F resulting in pBCKS-V-6S
6x11B7. The BamHI-6S-bxl IB7 XhoI cassette; of pBCKS-V-6S-6xI lB7 was then recloned into BamHI-XhoI of pET21 a(+) resulting in pET21-6S-6x I 1 B7. All cloning 3o procedures and clone analysis are described above.

CONSTRUCTION OF FUSED MULTIDOMAIhT MBI26 EXPRESSION SYSTEMS
;g In our previous work we solved all major problems connected to the ' construction of multidomain cationic peptide expression systems. This example demonstrates we were able to simplify the process, especially the need for synthesis of multiple specific DNA cassettes; only one mbi26 cassette was prepared and used at the first and last position as well as for the serial cloning procedure. Plasmids pET21CBD96-ls-26 and pET21CBD96-2s-2x26 were prepared. We tested expression 5 of a combination of mbi26 and mbil 1B7 domains. We performed two cloning cycles, inserting mbi26S cassettes into pET21CBD96-IS-3xI1B7, resulting in pET21CBD96-26S-3x11B7 and pET21CBD96-2x26S-~xlIB7. Both constructs expressed the combined mbi26-i 1B7 multidomain proteins at good levels.
A. .f YNTHESIS OF UNIVERSAL MBI26 DOMAln!
An oligonucleotide CGCC AGGG TTTT CCCA GTCA CGAC GGAT
CCGT CTCA TATG ACCA TGAA ATGG A.AAT CTTT CATC AAAA AACT
GACC TCTG CTGC TAAA AAAG TTGT TACO ACCG CTAA ACCG CTGA TCTC
TATG CATG CTTA AGCT TCGA TCCT CTAC: GCCG GACG C (SEQ ID N0: 26) was synthesized and used as a template for PCR using primers CGCC AGGG TTTT
Is CCCA GTCA CGAC (SEQ ID N0:18) and GCGT CCGG CGTA GAGG ATCG (SEQ
ID N0:19). PC:R was performed in an MJ-Researc;h PTC-100 Thermo-cycler in a 50 ~I
reaction volume with 30 cycles of 94°C, 30 sec.; 55°C, 30 sec.
and 72°C, 30 sec., 2 U
of Taq DNA polymerase, corresponding reaction buffer ( l Ox PCR reaction buffer:
500 mM KCI, 15 mM MgCI=, 100 mM Tris-HC1 pH 9), 0.2 mM dNTPs (dATP, dGTP, 2o dTTP and dCTP), 25 pmol of each primer and 50 pmol of template oligonucleotide resulting in a I 12 by dsDNA MBI26 fragment. The resulting DNA fragment was cloned into pTZ 18R as a BamHI-Hinc~lII fragment as described in Example 2, paragraph (A} resulting in piasmid pTZ 18R-26GT. After verification of DNA
sequence, the BamHI-HindIII mbi26 fragmem was recloned into pBCKS(~-) resulting in 25 pBCKS-26GT.
B. CONSTRUCTION OFMBI26 FUSEDMULTLDOMAIN.S'YSTEM
The first step in construction was a direct fusion of the mbi26 cassette to cbd96 in pET21 CBD96. Plasmids pET21 CBD96 (0.25 fig) and pBCKS-26GT (2.5 p.g) were digested with BamHI and HindIII in 1.5x OPA (50 ~l reaction volume) at 37°C
3o for 1 hour using 2 U of each restriction enzyme .and 20U of each enzyme respectively.
Both reactions were stopped by phenol/CHCI~ extraction and ethanol precipitation.
Each resulting DNA was dissolved in 8 lzl of water; the two were mixed together with 2 ~1 of 10 mM ATP, 2 ul of l Ox OPA and 2 'U of T4 DNA lipase and the ligation reaction was incubated at 10°C for 1 hour. 2 ~,l of the ligation mixture was used to 46 _ electroporate ~0 yl of 1. cull XL t Blue using a :>terile Gene Pulses cuvette (0.2 cm electrode gap) and Gene Pulses eiectroporator apparatus set to 2.5 kV, 200 ohms and 250 ltF. After an electroporation pulse, 1 ml of TB media was added to the cell suspension and bacteria were incubated i hour at 37°C with rigorous shaking. Then I0, 50 and 100 ~1 of cell suspension were plated on MacKonkey agar plates with 100 a glml of Ampicillin and incubated overnight at 3'7°C. The next day several colonies were transferred to 2 ml of TB and cultivated at 3',1°C with rigorous shaking overnight.
Then plasmid DNA was isolated and analyzed, including DNA sequencing by methods known to those skilled in the art. Positive clones pET21 CBD96-26 contained the MBI
tt 26 gene fused to cbd9h.
The second step was preparation of a cassette for the serial cloning procedure. The mbi26 fragment of pTZ 18R-26G'T was cloned into pBCKS-V-S as a BamHI-NsiI fragment basically as described in lExample 5 {A), resulting in plasmid . pBCKS-V-26S with an mbi26 domain fused to the anionic spacer-encoding sequence.
15 The only exception was that 2x OPA was used in the restriction enzyme digest reaction.
The insert was then cloned into pTZl8R resulting in pTZl8R-26S. This allowed the cloning of the BamHI-26S-L :~pl insert into the l3arreHI-Ndel sites of pBCKS-V-26S, resulting in pBCKS-V-2S-2x26.
The third step was the actual serial cloning procedure (for details see 2o Example 6B). Briefly, pBCKS-V-26S was digested with BamHI and Vspl resulting in fragment BamHI-26S-V.spl, which was ligated; into plasmid pET21CBD96-26GT
digested with BamHI and NdeI. Positive clones pET21CBD96-Is-2x26 contained two MBI-26 units with one spacer fused to cbd9b. 'This was the first cycle of the serial cloning. In the next cycle pET21CBD96-ls-2x26 and pBCKS-V-26S could be used to 2S prepare pET21 CBD96-2s-3x26 and so on.
C. CONSTRUCTIONAND EXPRESSION OF COA~BINED MBI26 - MBIll B7 MULTIDOMAIN lIENES
Plasmid pET21CBD96-1S-3x11H~7 was used as a vector for serial cloning of the mbi26S domain of pBCKS-V-265. Briefly, pBCKS-V-26S was digested 3o with BamHI and Vspl restriction endonucleases resulting in fragment BamHI-26S-Vspl, which was ligated into plasmid pET21 CBD96- I S-3x 11. B7 digested with BamHI
and Ndel. Positive clones pET21CBD96-2S-26-3x11B7 contained an MBI-26 unit with one spacer fused to three 11B7 units with one spacer. This was the first cycle of the serial cloning. In the next cycle pET21 CBD96-2S-26-:3x11 B7 and pBCKS-V-26S were used 3~ to prepare pET21 CBD96-3 S-2x26-3x 11 B7.

WO 00!31279 PCT/CA99/01107 T7 expression systems were prepared in l:. cull MC4100F based on plasmids pET21 CBD96-2S-26-3x I 1 B7 or pET21 ~~BD96-3S-2x26-3x l i B7 and pGP
I-2. Proteins CBD96-2S-26-3xi IB7 and CBD9b-pS-2x26-3x1187 were expressed at ;ood levels after temperature induction.
EXAMPLE ~~
PRODUCTION IN SHAKE FLASK FERMENTATION O!F MULTI-DOMAW CATIONIC PEPTmE
FUSED TO TRUNCATED CBD OR UNFUSED SYSTEMS
to Each of the different pET2ICBDS~6-(rr-2)S-rrx11B7, pET21CBD96-rrS-nx 11 B7, pET21-(rr-2)S-rrx 11 B7 and pET21-nS-rrx:l 1 B7F constructs (where rr = number of copies, S represents the anionic spacer, and I1B7 or 11B7F represents the cationic MBI-1187 peptide) were expressed in E. call strain MC4100F.
All fermentations are done in TB broth, which is prepared as follows: 12 15 g of Trypticase Peptone (BBL), 24 g of yeast Extract (BBL) and 4 ml of glycerol (Fisher] is added to 900 ml of Milli-Q water. Ttie material is allowed to dissolve and 100 mI of 0.17 M KH~PO, (BDH), 0.72 M K~JHPO~ (Fisher) is added . The broth is autoclaved at 121°C for 20 minutes. The resulting pH is 7.4.
A one liter Erlenmeyer flask with 170 ml medium, containing 100 ~tg/ml 2o ampicillin (Sigma-Aldrich Corp.) and 30 pglml kanamycin A (Sigma-Aldrich Corp.), was inoculated with the relevant 0.5 rril frozen stock and shaken at 300 rpm in a shaking incubator (model 4628, Lab Line Instrument Inc.), at 30°C for 16 hr.
The culture was then transferred to a 2.0 L flask with 330 ml fresh TB
medium (no antibiotics), preincubated at 30°C. After dilution, protein expression was 25 induced by raising the culture temperature to 42"C and shaking at 300 rpm for another to 7 hours. The pH was kept between 6.7 and 7.1 using 30% ammonium hydroxide.
Bacteria were fed at least twice during inductic>n with 0.5 g glucose per flask. Cells were harvested by centrifugation (Sorvall~ RC-SB) at' 1 S,OOOxg for I S
minutes and cell pellets were stored at -70°C prior to cell lysis.

CRUDE FRACTIONATION AND INCLt.ISION BODIES ISOLATION
The bacteria produce the multi-domain proteins as insoluble inclusion bodies. To release and isolate the inclusion bodies, the harvested cells were suspended in 200 ml buffer (50' mM Tris-HCI, 10 mM EDTA, pH 8.0) and lysed-by sonication (Vibra-CellT'~, Sonic and material lne.) five rimes for =~5 seconds, , on ice, then centrifuged (Sorvall0 RC-SB) at '' 1,875xg for 15 min at 4°C. The pellet was homogenized (PolyScience, Niles, IL USA) in 160 ml of lysis buffer (20 mM Tris-HCI, 5 100 Lylml lysozyme, pH 8.0) and incubated at room temperature for 45 min.
Next Triton X-100 was added (1% v/v), and the mixture was homogenized thoroughly and centrifuged at 21,875xg for 15 min at 4°C. The inclusion bodies pellet was resuspended in 200 ml of O. I M NaCI, homogenized, and precipitated by centrifugation as described above, then resuspended in 200 ml water and precipitated again by centrifugation. At to this stage, the inclusion bodies contained greater than 70% fusion protein.

RELEASING OF CATIONIC PEPT)DE F3Y CHEMICAL CLEAVAGE
15 The isolated inclusion bodies were dissolved in 70% formic acid (I00 mg wet weight IB per ml), then CNBr was added to a final concentration of 0.1 to 0.15 M. The cleavage reaction which allowed the release of cationic peptide from the fusion protein and spacer was performed under nitrogen and with stirring, in the dark, for 4 hr.
Next the reaction mixture was diluted with 15 volumes of Milli-Q water and dried in a 2o rotovap machine (Rotovapore, R-124VP, BUCHI Switzerland). The dried pellet was then dissolved in 10 ml of 7-8. M urea and insoluble materials were separated by centrifugation at 21,875xg for 15 min.
At this stage, the soluble, materials, at acidic pH (2-J.~) and low conductivity (1-5 mS), contain the homoserine; lactone form of the cationic peptide.
25 This material was further purified using a chromatography procedure.
EXAMPLE, 12 FREE CATIONIC PEPT)Dl: PURIFICATION
3o The purification of the homoserine lactone form of MBI-I IB7 peptide was performed on a BioSys''~ 2000 chromatography work station (Beckman Instruments, Inc.), using Fast Flow Q-Sepha~rose anion exchange resin (Pharmacia Biotech AB) packed in an XK column (1.6 x 11 cm). The column was equilibrated with 2 column volumes (CV) of 1 M NaOH at a flow rate of 9 ml/min, followed by a water 35 wash. Conductivity, pH and absorbency at 280 nm were monitored. When the conductivity dropped below s mS, the dried cieava~~e materials, dissolved in 7-urea, were loaded onto the column and washed with 4 M urea. The unbound pure cationic peptide flowed through the column and was monitored as the leading peak.
When the absorbance dropped to baseline, the bound material (i.v., impurities) was washed off the column with I M NaOH and appeared as the second peak (Figure 9}.
The flow-through peak was collected and pooled and the pH was adjusted to 7.0-7.5 with 0.2 N HCI. The sample was analyzed for purity by reverse phase HPLC {Figure 11 ), using a C8 column (4.6x i 0, Nova-Pak, Waters) and by acid-urea gel electrophoresis (West and Bonner, Biochemisxry 19:3238, 1980). The identity to of the MBI-11B7 peptide was confirmed by mass spectrometry to show that the flow -through peak represents the homoserine form of the MBI-11 B7 peptide.
EXAMPLE i 3 UREA SEPARATION AND FURTI-LER PURIFICATION
I~
The separation of the urea from the purified peptide utilized a high-throughput reverse phase chromatography technique by using the BioGADT~' (PerSeptive Biosystems lnc.) perfusion chromatography workstation and Poros~ R-Il 20 column, 4.6 x 100 mm (PerSeptive Biosystems lnc.). About 10 mg of the peptide 20 were applied on the column at 5 ml/min, followed by equilibration of the column with 0.1% TFA. The peptide was eluted from the column by a gradient of increasing acetonitrile from 0 to 50% for 10 minutes at a flow rate of 5 ml/min. The peak of the further purified arid urea free peptide was collectf;d and lyophilized.

BACTERICIDAL. ACTIVITY OF MBI-11B7CN PEPTLDE AND ITS
HOMOSERINEI HOMOSERINE L.ACTONE 1SOFORMS
A comparison of anti-microbial activity between chemically and 3o recombinantly synthesized cationic peptide was carried out.
The antimicrobial activities of the chemically synthesized MBI-11B7CN
peptide and recombinant DNA synthesized MBI.-11B7HSL (homoserine lactone form}
and MBI-11B7HS (homoserine form} peptides were tested against various gram-negative anal positive strains of bacteria, inch.~ding antibiotic resistant strains. The 35 Agarose Dilz~tion Assay was performed as described in "Methods for Dilution :~ntimicrobial Susceptibility Tests for Bacteria That Grow Aerobically-Fourth Edition;
.approved Standard" NCCLS document lVl7-A4 ( I;SBN I-56238-309-4) Vol. 17, No ( 1977).
The agarose dilution assay measures antimicrobia! activity of peptides and peptide analogues, which is expressed as thc~ minimum inhibitory concentration (MIC) of the peptides.
In order to mimic ire vivv conditions, calcium and magnesium supplemented Mueller 1-iinton broth is used in combination with a low EEU
agarose as the bacterial growth medium. Agarose, rather than agar, is used as the charged groups in agar prevent peptide diffusion through the media. The medium is autoclaved and then cooled to 50°C - 55°C in a water bath before aseptic addition of anti-microbial solutions. The same volume of different concentrations of peptide solution are added to the cooled molten agarose which is then poured to a depth of 3 - 4 mm.
The bacterial inocufum is adjusted t:o a 0.5 McFarland turbidity standard (PML Microbiological) and then diluted 1:10 before application on to the agarose plate.
The final inoculum applied to the agarose is approximately 10' CFU in a 5 - 8 mm diameter spot. The agarose plates are incubated at 35°C - 37°C
for 16 to 20 hours.
The MIC is recorded as the lowest concentration of peptide - tha't completely inhibits growth of the, organism as determined by visual inspection.
2o Representative MICs for the cationic peptides against various bacterial strains are shown in Table 3.

MINIMUM INHIBITORY CONCENTRATION (MIC) VALUES FOR MBI-1 1B7CN
(CARBOXY-AMIDATED), MBI-1 1B7HSL (HOMOSERINE LACTONE FORM) AND MBI
I 1B7HS (HOMOSERINE FORM) PEPTIDES, AGATI'JST VARIOUS GRAM-NEGATIVE AND
GRAM-POSITIVE BACTERIA STRAINS
MIC
m ~ Organism # ~ Source ~ 11B7CN ( 11B7HSL I 11B7HS
92A1 ~ 203B1 ~ 203B1 calcvaceticz~s AC2 ATCC 2 ~ 4 ~ 2 cloacae ECL7 ATCC >64 >64 >64 E. cull EC~5 ..~TCC 8 8 32 K. pnetrntvoiaeKP1 .~TCC 8 8 32 I'. GlC'Yll~lltl).SaPA4 ATCC >64 >64 >64 S. ntaltphiliaSMA2 ATCC 32 32 64 S. ntarcescettsSMS3 ATCC >64 >64 >64 E. faecalisEFS 1 ATCC 2 t 2 E. faecalirEFSB ATCC 16 16 32 S. aurears SA 14 Bayer 4 1 2 v: arrrerrsSA93 Bayer 1 1 1 5: epidermidisSE10 Chow 2 4 8 Although the foregoing refers to particular preferred ernbodirnents, it will be understood that the present invention is not so limited. It will occur to those of ordinary skill in the art that various modifications may be made to the disclosed embodiments and that such modifications are intended to be within the scope of the present invention, which is defined by the following claims.

Claims (28)

We claim:

1. A multi-domain fusion protein expression cassette, comprising a promoter operably linked to a nucleic acid molecule which is expressed as an insoluble protein, wherein said nucleic acid molecule encodes a polypeptide comprising the structure (cationic peptide)-[(cleavage site)-(cationic peptide)]n, wherein n is an integer having a value between 1 and 100 and said cationic peptide has antimicrobial activity:

2. The expression cassette according to claim 1 wherein said nucleic acid molecule also encodes a carrier protein.

3. The expression cassette according to claim 1 wherein n has a value of between 5 and 40.

4. The expression cassette according, to claim 1 wherein said cleavage site can be cleaved by low pH or by a reagent selected from the group consisting of cyanogen bromide, 2-(2-nitrophenylsulphenyl)-3-methyl-3'-bramoindolenine, hydroxylamine, o-iodosobenzoic acid; Factor Xa, thrombin, enterokinase, collagenase, Staphylycoccus aureus V8 protease, endoproteinase Arg-C, and trypsin.

5. The expression cassette according to claim 1 wherein said nucleic acid molecule encodes a fusion protein comprising (a) a carrier protein, (b) an anionic spacer peptide component having at least one peptide with the structure (cleavage site)-(anionic spacer peptide), and (c) a cationic peptide component having at least peptide with the structure (cleavage site)-(cationic peptide) wherein the cleavage site can be on either side of the anionic spacer peptide or cationic peptide, and elements (a), (b), and (c) can be in any order and or number.

6. The multi-domain fusion protein expression cassette of claim 5 wherein said carrier protein is located at the N-terminus of said fusion protein.

7. The expression cassette according to claim 5 wherein said anionic spacer lacks a cysteine residue.

8. The expression cassette of claim 1 wherein said fusion protein comprises from 2 to 40 cationic peptides.

9, The expression cassette according to claim 8 wherein said fusion protein comprises from 3 to 15 cationic peptides.

10. The expression cassette according to claim 5 wherein the number of anionic spacer peptides greater than or the same as the number of cationic peptides.

11. The expression cassette according to claim 5 wherein the number of anionic spacer peptides is less than the number of cationic peptides.

12. The expression cassette according to claim 2 wherein said carrier protein is less than 100 amino acid residues in length.

13. The expression cassette according to claim 2 wherein said carrier protein is a truncated cellulose binding domain of less than 100 amino acids.
.

14. The expression cassette according to claim 1 wherein said nucleic acid molecule encodes a fusion protein comprising (a) an anionic spacer peptide component having at least one peptide with the structure (cleavage site)-(anionic spacer peptide), and (b) a cationic peptide component having at least peptide with the structure (cleavage site)-(cationic peptide), wherein the cumulative charge of said anionic spacer peptide component reduces the cumulative charge of said cationic peptide component.

15. The expression cassette according to claim 1 wherein said promoter is selected from the group consisting of IacP promoter-, tacP promoter, trcP
promoter, SrpP
promoter, SP6 promoter, T7 promoter, araP promoter, trpP promoter, and .lambda., promoter.

16. A recombinant host cell comprising the expression cassette according to claim 1.

17. The recombinant host cell of claim 16 wherein said host cell is a yeast, fungi, bacterial or plant cell.

18. The recombinant host cell of claim 17 wherein said bacterial host cell is Escherichia call.

19. A polypeptide encoded by the expression cassette according to claim 1.

20. A method of producing fusion proteins that contain a cationic peptide, comprising: (a) culturing the recombinant host cell of claim 15 under conditions and for a time sufficient to produce said fusion protein.

21. The method according to claim 19, further comprising the step of isolating said fusion protein.

22. The method according to claim 20, further comprising the step of cleaving said polypeptide by treating said fusion protein with low pH or with a reagent selected from the group consisting of cyanogen bromide, 2-(2-nitrophenylsulphenyl)-3-rnethyl-3'-bromoindolenine, hydroxylamine, o-iodosobenzoic acid, Factor Xa, thrombin.
enterokinase, collagenase, Staphylococcus aureus V8 protease, endoproteinase Arg-C, and trypsin.

23. The method according to claim 21, further comprising the step of purifying said cationic peptides by applying said cleaved polypeptide to an anion exchange chromatography resin.

24. The method according to claim 23 wherein said anion exchange column is charged with a base, and washed with water prior to loading the column with said cationic peptide.

25. The method according to claim 23 wherein said column is equilibrated with water and up to about 8 M urea.

26. The method according to claim 23 wherein said cationic peptide is in a solubilized in a solution comprising up to about 8 M
urea.

27. The method according to claim 23 wherein said cationic peptide is solubilized in a solution comprising a mild organic solvent.

28. The method according to claim 27 wherein said mild organic solvent is methanol, ethanol, or acetonitrile.