Antimicrobial agents

Abstract

The application relates to antimicrobial agents against Gram-negative bacteria, in particular to fusion proteins composed of an enzyme having the activity of degrading the cell wall of Gram-negative bacteria and a peptide stretch fused to the enzyme at the N- or C-terminus, as well as pharmaceutical compositions comprising the same. Moreover, it relates to nucleic acid molecules encoding such a fusion protein, vectors comprising said nucleic acid molecules and host cells comprising either said nucleic acid molecules or said vectors. In addition, it relates to such a fusion protein for use as a medicament, in particular for the treatment or prevention of Gram-negative bacterial infections, as diagnostic means or as cosmetic substance. The application also relates to the treatment or prevention of Gram-negative bacterial contamination of foodstuff, of food processing equipment, of food processing plants, of surfaces coming into contact with foodstuff, of medical devices, of surfaces in hospitals and surgeries.

Claims

1. A fusion protein comprising an endolysin having the activity of degrading the cell wall of Gram-negative bacteria and a peptide segment fused to the endolysin at the N- or C-terminus or at both termini, wherein the peptide segment is a cathelicidine or a magainine.

2. The fusion protein according to claim 1, wherein the peptide segment consists of about 12 to about 100 amino acid residues.

3. The fusion protein according to claim 2, wherein the peptide segment consists of about 12 to 50 amino acid residues.

4. The fusion protein according to claim 2, wherein the peptide segment consists of about 12 to 30 amino acid residues.

5. The fusion protein according to claim 1, wherein said fusion protein comprises an additional amino acid residue on the N-terminus.

6. The fusion protein according to claim 1, wherein said fusion protein comprises a tag or additional protein on the C- and/or N-terminus.

7. The fusion protein according to claim 1, wherein the peptide segment is linked to the endolysin by one or more additional amino acid residues.

8. The fusion protein according to claim 1, wherein the endolysin comprises an amino acid sequence according to any of SEQ ID NO: 1, 2, 3, 4, 5, 18, 20, 21, 22, 23, 24, 25 or 34.

9. The fusion protein according to claim 1, wherein the peptide segment comprises an amino acid sequence according to any of SEQ ID NO: 9, 10, 11, 12, or 13.

10. The fusion protein according to claim 1, wherein the Gram-negative bacteria are selected from the group consisting of: Enterobacteriaceae, Pseudomonadaceae, Neisseria, Moraxella, Vibrio, Aeromonas, Brucella, Francisella, Bordetella, Legionella, Bartonella, Coxiella, Haemophilus, Pasteurella, Mannheimia, Actinobacillus, Gardnerella, Spirochaetaceae, Leptospiraceae, Campylobacter, Helicobacter, Spirillum, Streptobacillus, Bacteroidaceae and Acinetobacter.

11. The fusion protein according to claim 10, wherein the Gram-negative bacteria are selected from the group consisting of Escherichia, Salmonella, Shingella, Citrobacter, Edwardsiella, Enterobacter, Hafnia, Klebsiella, Moganella, Proteus, Providencia, Serratia, Yersinia, Pseudomonas, Burkholderia, Stenotrophomonas, Shewanella, Sphingomonas, Comamonas, Treponema, Borrelia, Bacteroides, Fusobacterium, Prevotella, Porphyromonas and A. baumanii.

12. An isolated nucleic acid molecule encoding a fusion protein according to claim 1.

13. A vector comprising the nucleic acid molecule according to claim 12.

14. A host cell comprising the nucleic acid molecule according to claim 12 or the vector according to claim 13.

15. The host cell according to claim 14, wherein the cell is a bacterial cell or a yeast cell.

16. A pharmaceutical composition comprising a fusion protein according to claim 1.

17. A fusion protein comprising an endolysin having the activity of degrading the cell wall of Gram-negative bacteria and a peptide segment fused to the endolysin at the N- or C-terminus or at both termini, wherein the peptide segment comprises SEQ ID NO: 33.

Description

EXAMPLE 1. CLONING, EXPRESSION AND PURIFICATION OF GP144 AND GP188 MODIFIED WITH AN AMPHIPATHIC PEPTIDE

(1) As a proof of principle, the potential of the LPS disrupting activity of amphipathic peptides to lead gp144 and gp188 through the outer membrane and the consequent antibacterial activity against Gram-negative bacteria is demonstrated. Gp144 and gp188 are modular endolysins originating from Pseudomonas aeruginosa phages φKZ and EL with an N-terminal peptidoglycan binding and C-terminal catalytic domain (Briers et al., 2007).

(2) To extend the 5′ end of the open reading frame encoding gp144 or gp188 with a gene fragment encoding the amphipathic α4 helix of T4 lysozyme (aa 143-155: Pro-Asn-Arg-Ala-Lys-Arg-Val-Ile-Thr-Thr-Phe-Arg-Thr according to SEQ ID NO: 92) a tail PCR with an extended 5′ primer and standard 3′ primer was applied. The PCR product was cloned in the pEXP5CT/TOPO® expression vector (Invitrogen, Carlsbad, Calif., USA).

(3) Expression of all constructs was performed in E. coli BL21 (DE3) pLysS cells. All proteins were purified by Ni.sup.2+ affinity chromatography using the C-terminal 6×His-tag. The yields for different purifications are shown in table 4. Remarkably, α4-KZ144 production was not toxic for the host, in contrast to KZ144, resulting in a significant higher yield.

(4) Purified stock solutions were ˜90% pure. All gp144 derivatives showed multimer formation which could be converted to monomers by addition of β-mercapto-ethanol, indicating that interdisulfide bonds cause multimerization.

(5) TABLE-US-00004 TABLE 4 Yields of recombinant purification of endolysins modified with an amphipathic peptide*. Endolysin Fusion gp144 gp188 α4 helix 179 mg 38 mg *The total yield of purified recombinant protein per liter E. colt expression culture is shown. This value was determined by spectrophotometric measurement of the protein concentration and the total volume of the purified stock solution. The purification of gp188 derivatives was performed under more stringent conditions (65 mM imidazole) compared to gp144 derivatives (50 mM imidazole) to ensure high purity.
Characterization of Gp144 and Gp188 Modified with an Amphipathic Peptide
1.A. Enzymatic Activity of Gp144 and Gp188 Modified with an Amphipathic Peptide

(6) To assess the influence of the modification on the enzymatic activity of gp144 or gp188, the specific activity of the variants was measured on chloroform-permeabilized Pseudomonas aeruginosa cells and compared to the corresponding unmodified endolysin. Different incremental amounts of all modified endolysins were tested to determine the corresponding saturation curve.

(7) The slope of the linear regression of the linear region of this curve is a measure for the specific activity and was expressed relatively to the slope of unmodified gp144 or gp188 (Table 5).

(8) TABLE-US-00005 TABLE 5 Enzymatic activity of gp144 or gp188 modified with an amphipathic peptide*. Endolysin Fusion gp144 gp188 α4 helix 23% 146% *The specific enzymatic activity of the different variants was determined and expressed relatively to the specific activity of the corresponding original endolysin (=100%), which was tested simultaneously. The buffer conditions of the assay were the optimal conditions of the corresponding endolysins (KH.sub.2P0.sub.4/K.sub.2HP0.sub.4 I = 120 mM pH 6.2 and I = 80 mM pH 7.3 for gp144 and gp188, respectively).
1.B. Antibacterial Activity of Gp144 and Gp188 Modified with an Amphipathic Peptide

(9) Exponential (˜10.sup.6/ml) P. aeruginosa PAO1 cells were incubated at room temperature with unmodified and modified gp144/gp188. After 1 hour, cell suspensions were diluted and plated. The residual colonies were counted after an overnight incubation (Table 6). Unmodified gp144 gp188 does not reduce cell numbers significantly compared to the negative control. This observation illustrates the efficacy of the outer membrane as a barrier. Fusion proteins with the amphipathic α4-helix inactivate exponential cells with 50±11 and 34±11% for α4-KZ144 and α4-EL188, respectively. When stationary cells with a 100-fold higher density are used, these values are similar (35±18 and 32±17%, respectively). Despite the rather high variability between different replicates, these values differ significantly from the untreated cells (α=0.05). In general, modified gp144 derivatives tend to have a higher antibacterial activity than gp188 derivatives.

(10) TABLE-US-00006 TABLE 6 Antibacterial effect of endolysins gp144 and gp188 and their derivatives*. Exponentially Endolysins growing cells gp144 gp188 Fusion % log % log unmodified  0 ± 15 0.00 ± 0.06 10 ± 13 0.05 ± 0.06 α4 helix 50 ± 11 0.31 ± 0.09 34 ± 11 0.19 ± 0.07 *Exponentially growing P. aeruginosa PAO1 cells were 100 x diluted and incubated (final density was ~10.sup.6/ml) with 10 μg undialyzed protein (final concentration 100 μg/ml, buffer: 20 mM NaH.sub.2P0.sub.4—NaOH pH 7.4; 0.5M NaCl; 0.5M imidazole) for 1 hour at room temperature. Aliquots are diluted and plated. The antibacterial activity is expressed as the relative inactivation (%) (=100 − (N.sub.i/No)*100 with N.sub.0 = number of untreated cells and N.sub.i = number of treated cells) and in logarithmic units (=log.sub.10N.sub.0/N.sub.i). All samples were replicated in six fold. Averages/standard deviations are represented. Statistical analysis was performed using a student's t-test.

EXAMPLE 2. CLONING, EXPRESSION AND PURIFICATION OF GP144 AND GP188 MODIFIED WITH A HYDROPHOBIC PEPTIDE

(11) As a proof of principle, the potential of the LPS disrupting activity of a hydrophobic pentapeptides to lead gp144 and gp188 through the outer membrane and the consequent antibacterial activity against Gram-negative bacteria is demonstrated. Gp144 and gp188 are modular endolysins originating from Pseudomonas aeruginosa phages φKZ and EL with an N-terminal peptidoglycan binding and C-terminal catalytic domain (Briers et al., 2007).

(12) To extend the 5′ end of the open reading frame encoding gp144 or gp188 with a gene fragment encoding 5 hydrophobic residues (Phe-Phe-Val-Ala-Pro) a tail PCR with an extended 5′ primer and standard 3′ primer was applied. The PCR product was cloned in the pEXP5CT/TOPO® expression vector (Invitrogen, Carlsbad, Calif., USA).

(13) Expression of all constructs was performed in E. coli BL21 (DE3) pLysS cells. All proteins were purified by Ni2+ affinity chromatography using the C-terminal 6×His-tag. The yields for different purifications are shown in table 7.

(14) Purified stock solutions were ˜90% pure. All gp144 derivatives showed multimer formation which could be converted to monomers by addition of β-mercapto-ethanol, indicating that interdisulfide bonds cause multimerization.

(15) TABLE-US-00007 TABLE 7 Yields of recombinant purification of endolysin derivatives*. Endolysin Fusion gp144 gp188 Phe-Phe-Val-Ala-Pro 25 mg 85 mg *The total yield of purified recombinant protein per liter E. colt expression culture is shown. This value was determined by spectrophotometric measurement of the protein concentration and the total volume of the purified stock solution. The purification of gp188 derivatives was performed under more stringent conditions (65 mM imidazole) compared to gp144 derivatives (50 mM imidazole) to ensure high purity.
Characterization of Gp144 and Gp188 Modified with a Hydrophobic Pentapeptide
2.A. Enzymatic Activity of Gp144 and Gp188 Modified with a Hydrophobic Pentapeptide

(16) To assess the influence of the modifications on the enzymatic activity of gp144 or gp188, the specific activity of the variants was measured on chloroform-permeabilized Pseudomonas aeruginosa cells and compared to the corresponding unmodified endolysin. Different incremental amounts of all modified endolysins were tested to determine the corresponding saturation curve. The slope of the linear regression of the linear region of this curve is a measure for the specific activity and was expressed relatively to the slope of unmodified gp144 or gp188 (Table 8).

(17) TABLE-US-00008 TABLE 8 Enzymatic activity of gp144 or gp188 modified with a hydrophobic peptide*. Endolysin Fusion gp144 gp188 Hydrophobic pentapeptide 150% 100% *The specific enzymatic activity of the different variants was determined and expressed relatively to the specific activity of the corresponding original endolysin (=100%), which was tested simultaneously. The buffer conditions of the assay were the optimal conditions of the corresponding endolysins (KH.sub.2P0.sub.4/K.sub.2HP0.sub.4 I = 120 mM pH 6.2 and I = 80 mM pH 7.3 for gp144 and gp188, respectively).
2.B. Antibacterial Activity of Gp144 and Gp188 Modified with a Hydrophobic Pentapeptide

(18) Exponential (˜10.sup.6/m1) P. aeruginosa PAO1 cells were incubated at room temperature with unmodified and modified gp144/gp188. After 1 hour, cell suspensions were diluted and plated. The residual colonies were counted after an overnight incubation (Table 9). Unmodified gp144 gp188 does not reduce cell numbers significantly compared to the negative control. This observation illustrates the efficacy of the outer membrane as a barrier. Incubation with the hydrophobic pentapeptide fusion proteins causes a significant reduction (α=0.05) of the bacterial cell number (83±7 and 69±21% for modified gp144 and gp188, respectively). In general, modified gp144 derivatives tend to have a higher antibacterial activity than gp188 derivatives.

(19) TABLE-US-00009 TABLE 9 Antibacterial effect of endolysins gp144 and gp188 and their derivatives*. Exponentially Endolysins growing cells gp144 gp188 Fusion % log % log unmodified  0 ± 15 0.00 ± 0.06 10 ± 13 0.05 ± 0.06 Hydrophobic 83 ± 7 0.9 ± 0.2 69 ± 21 0.7 ± 0.3 pentapeptide *Exponentially growing P. aeruginosa PAO1 cells were 100 x diluted and incubated (final density was ~10.sup.6/ml) with 10 μg undialyzed protein (final concentration 100 μg/ml, buffer: 20 mM NaH.sub.2P0.sub.4—NaOH pH 7.4; 0.5M NaCl; 0.5M imidazole) for 1 hour at room temperature. Aliquots are diluted and plated. The antibacterial activity is expressed as the relative inactivation (%) (=100 − (N.sub.i/No)*100 with N.sub.0 = number of untreated cells and N.sub.i = number of treated cells) and in logarithmic units (=log.sub.10N.sub.0/N.sub.i). All samples were replicated in six fold. Averages/standard deviations are represented. Statistical analysis was performed using a student's t-test.

EXAMPLE 3: CLONING, EXPRESSION AND PURIFICATION OF KZ144 AND STM0016 MODIFIED WITH VARIOUS PEPTIDE STRETCHES ON THE N-TERMINUS OF THE ENDOLYSIN

(20) KZ144 according to SEQ ID NO: 25 is a modular endolysin originating from Pseudomonas aeruginosa phage φKZ with an N-terminal peptidoglycan binding and C-terminal catalytic domain (Briers et al., 2007). The endolysin KZ144 is encoded by the nucleic acid molecule according to SEQ ID NO: 64. The nucleic acid molecule according to SEQ ID NO: 64 was synthetically produced with a BamH I (5″-GGA TCC-3′) restriction site at the 5″-end of the nucleic acid molecule and an Xho I (5′-CTC GAG-3′) restriction site at the 3′-end of the nucleic acid molecule.

(21) STM0016 is a hypothetical protein with homology to the E. coli phage N4 endolysin N4-gp61.

(22) The endolysin STM0016 is encoded by the nucleic acid molecule according to SEQ ID NO: 65. The nucleic acid molecule according to SEQ ID NO: 65 was synthetically produced with a BamH I (5′-GGA TCC-3′) restriction site at the 5′-end of the nucleic acid molecule and an Xho I (5″-CTC GAG-3′) restriction site at the 3′-end of the nucleic acid molecule.

(23) N4-gp61 is an E. coli N4 phage endolysin. The endolysin is encoded by the nucleic acid according to SEQ ID NO: 91. The nucleic acid molecule according to SEQ ID NO: 91 was synthetically produced with a BamH I (5″-GGA TCC-3′) restriction site at the 5″-end of the nucleic acid molecule and an Xho I (5′-CTC GAG-3′) restriction site at the 3′-end of the nucleic acid molecule.

(24) The following peptide stretches in table 10 were used for production of fusion proteins with the endolysin KZ144 or STM0016:

(25) TABLE-US-00010 TABLE 10 Nucleic acid molecule encoding Peptide stretch the peptide stretch Pseudin 1 SEQ ID NO: 66 (SEQ ID NO: 29) Ranalexin SEQ ID NO: 67 (SEQ ID NO: 30) Sushi 1 SEQ ID NO: 68 (SEQ ID NO: 32) WLBU2-Variant SEQ ID NO: 69 (SEQ ID NO: 33) Melittin SEQ ID NO: 70 (SEQ ID NO: 31) SMAP-29 SEQ ID NO: 71 (SEQ ID NO: 11) Pleurocidin SEQ ID NO: 72 (SEQ ID NO: 6) Cecropin A (A. SEQ ID NO: 73 aegypti) (SEQ ID NO: 14) Cecropin A (A. SEQ ID NO: 74 melanogaster) (SEQ ID NO: 15) Buforin II SEQ ID NO: 75 (SEQ ID NO: 8) Sarcotoxin IA SEQ ID NO: 76 (SEQ ID NO: 16)

(26) The nucleic acid molecules encoding the respective peptide stretches were synthetically produced with a Nde I (5′-CAT ATG-3′) restriction site at the 5′-end of the nucleic acid molecule and a BamH I (5′-GGA TCC-3′) restriction site at the 3′-end of the nucleic acid molecule, except the nucleic acid molecule encoding the Sushi 1 peptide, which was produced with a Nco I restriction site plus two additional nucleotides (5′-CCA TGG GC-3′) at the 5′-end of the nucleic acid molecule.

(27) Fusion proteins are constructed by linking at least two nucleic acid sequences using standard cloning techniques as described e.g. by Sambrook et al. 2001, Molecular Cloning: A Laboratory Manual. Therefore the nucleic acid molecules encoding the peptide stretches were cleaved in a digest with the respective restriction enzymes Nde I and BamH I and in case of the nucleic acid molecule encoding the peptide stretch Sushi 1 the digest was performed with the restriction enzymes Nco I and BamH I. Subsequently the cleaved nucleic acids encoding the peptide stretches were ligated into the pET21 b expression vector (Novagen, Darmstadt, Germany), which was also cleaved in a digest with the respective restriction enzymes Nde land BamH I before. The cleaved nucleic acid molecule encoding the peptide stretch Sushi I was ligated into a modified pET32 b expression vector (unmodified vector obtainable from Novagen, Darmstadt, Germany), which was also cleaved in a digest with the respective restriction enzymes Nco I and BamH I before. The modification of the pET32b expression vector refers to the deletion of the sequence encoding a S-tag and the central His-tag.

(28) Afterwards, the nucleic acid molecule encoding the endolysin KZ144 was cleaved in a digest with the restriction enzyme BamH I and Xho I, so that the endolysin could be ligated into the pET21b expression vector (Novagen, Darmstadt, Germany) and the modified pET32 b expression vector, respectively, which were also cleaved in a digest with the respective restriction enzymes BamH I and Xho I before. The nucleic acid molecule encoding the endolysin STM0016 and the nucleic acid molecule encoding the endolysin N4gp61 were cleaved in a digest with the restriction enzyme BamH I and Xho I, so that the respective endolysin could be ligated into the pET21b expression vector (Novagen, Darmstadt, Germany).

(29) Thus, the nucleic acid molecule encoding the peptide stretch is ligated into the respective vector at the 5′-end of the nucleic acid molecule encoding the endolysin KZ144 or STM0016. Moreover, the nucleic acid molecule encoding the endolysin KZ144 or STM0016 is ligated into the respective plasmid, so that a nucleic acid molecule encoding a His-tag consisting of six histidine residues is associated at the 3′-end of the nucleic acid molecule encoding the endolysin.

(30) As some fusion proteins may either be toxic upon expression in bacteria, or not homogenous due to protein degradation, the strategy might be to express these fusion proteins fused or linked to other additional proteins. Example for these other additional protein is thioredoxin, which was shown to mediate expression of toxic antimicrobial peptides in E. coli (TrxA mediating fusion expression of antimicrobial peptide CM4 from multiple joined genes in Escherichia coli. Zhou L, Zhao Z, Li B, Cai Y, Zhang S. Protein Expr Purif. 2009 April; 64(2):225-230). In the case of the fusion protein consisting of the N-terminal Sushi 1 peptide and the endolysin KZ144, the Sushi 1 peptide is ligated into the modified pET32 b expression vector, so that an additional thioredoxin is associated at the 5′-end of the Sushi 1 peptide. The thioredoxin could be removed from the expressed fusion protein by the use of enterokinase, therefore between the nucleic acid molecule encoding the Sushi peptide and the one encoding the thioredoxin is an enterokinase restriction site introduced.

(31) The sequence of the endolysin-peptide-fusions was controlled via DNA-sequencing and correct clones were transformed into E. coli BL21(DE3) (Novagen, Darmstadt, Germany) for protein expression.

(32) Recombinant expression of the fusion proteins according to SEQ ID NO: 77 to 90 is performed in E. coli BL21 (DE3) pLysS and E. coli BL21 (DE3) cells (Novagen, Darmstadt, Germany). The cells were growing until an optical density of OD600 nm of 0.5-0.8 was reached. Then the expression of the fusion protein was induced with 1 mM IPTG (isopropylthiogalactoside) and the expression was performed at 37° C. for a period of 4 hours.

(33) E. coli BL21 cells were harvested by centrifugation for 20 min at 6000 g and disrupted via sonication on ice. Soluble and insoluble fraction of the E. coli crude extract were separated by centrifugation (Sorvall, SS34, 30 min, 15 000 rpm). All proteins were purified by Ni.sup.2+ affinity chromatography (Akta FPLC, GE Healthcare) using the C-terminal 6×His-tag, encoded by the pET21b or pET32b vectors.

(34) As described above, some of the fusion proteins were expressed using a modified pET32b vector (S-tag and central His-tag deleted), which fuses thioredoxin on the N-terminus of the proteins of interest. The vector also contains an enterokinase cleavage site right before the protein of interest. This site allows the proteolytic cleavage between thioredoxin and the protein of interest, which can purified via the remaining C-terminal His-tag. For antimicrobial function of the fusion protein Sushi 1-KZ144 it may be necessary to remove the thioredoxin by proteolytic cleavage. Therefore the fusion protein was cleaved with 2-4 units/mg recombinant enterokinase (Novagen, Darmstadt, Germany) to remove the thioredoxin following the protocol provided by the manufacturer. After enterokinase cleavage the fusion protein was purified via His-tag purification as described below.

(35) The Ni.sup.2+ affinity chromatography is performed in 4 subsequent steps, all at room temperature: 1. Equilibration of the Histrap FF 5 ml column (GE Healthcare) with up to 10 column volumes of Washing Buffer (20 mM imidazole, 1 M NaCl and 20 mM Hepes on pH 7.4) at a flow rate of 3-5 ml/min. 2. Loading of the total lysate (with wanted fusion protein) on the Histrap FF 5 ml column at a flow rate of 3-5 ml/min. 3. Washing of the column with up to 10 column volumes of Washing Buffer to remove unbound sample followed by a second washing step with 10% Elution buffer (500 mM imidazole, 0.5 M NaCl and 20 mM Hepes on pH 7.4) at a flow rate of 3-5 ml/min. 4. Elution of bounded fusion proteins from the column with a linear gradient of 4 column volumes of Elution Buffer (500 mM imidazole, 0.5 M NaCl and 20 mM Hepes on pH 7.4) to 100% at a flow rate of 3-5 ml/min.

(36) Purified stock solutions of fusion proteins in Elution Buffer (20 mM Hepes pH 7.4; 0.5 M NaCl; 500 mM imidazole) were at least 90% pure as determined visually on SDS-PAGE gels (data not shown).

EXAMPLE 4: ANTIMICROBIAL ACTIVITY OF THE ENDOLYSIN KZ144 MODIFIED WITH VARIOUS PEPTIDE STRETCHES ON THE N-TERMINUS

(37) The fusion protein consisting of KZ144 and the peptide stretch α4 helix was constructed as described in example 1. The other fusion proteins consisting of KZ144 and the respective peptide stretches were constructed as described in example 3.

(38) E. coli DSMZ 11753, Acinetobacter baumannii DSMZ 30007 and Pseudomonas aeruginosa PAO1p cells (Burn wound isolate, Queen Astrid Hospital, Brussels; Pirnay J P et al. (2003), J Clin Microbiol., 41(3):1192-1202) were used as test strains. Overnight cultures were diluted 10-fold in fresh LB medium and grown to OD.sub.600=0.6. The culture was spun down and diluted 10-fold in dilution buffer (10 mM HEPES, 0.5 mM EDTA; pH 7.4). Bacteria were incubated at room temperature with each 10 μg undialyzed fusion protein at a final concentration of 100 μg/ml in buffer (20 mM NaH.sub.2PO.sub.4—NaOH pH 7.4; 0.5 M NaCl; 0.5 M imidazole). After 1 hour cell dilution series were made in PBS and plated on LB. Additionally, a negative control was plated using buffer (20 mM NaH.sub.2PO.sub.4—NaOH pH 7.4; 0.5 M NaCl; 0.5 M imidazole). The residual colonies were counted after an overnight incubation at 37° C. Based on the counted cell numbers the antibacterial activity as logarithmic units (=log.sub.10N.sub.0/N.sub.i with N.sub.0=number of untreated cells and N.sub.i=number of treated cells) was calculated (Table 11). All samples were replicated at least in four fold.

(39) The antimicrobial activity of these fusion proteins is given in the following table.

(40) TABLE-US-00011 TABLE 11 Antimicrobial activity of KZ144 modified with various peptide stretches against gram-negative bacteria Activity against Peptide stretch Activity against Activity against Acinetobacter (N-terminal unless Pseudomonas E. coli baumannii Fusion protein Enzyme part otherwise indicated) aeruginosa DSMZ 11753 DSMZ 30007 SEQ ID NO: 77 KZ144 Pseudin 1 + n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 29) SEQ ID NO: 78 KZ144 Ranalexin + n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 30) SEQ ID NO: 79 KZ144 Sushi 1 + n.d. ++ (SEQ ID NO: 25) (SEQ ID NO: 32) SEQ ID NO: 80 KZ144 WLBU2-Variant n.d. + n.d. (SEQ ID NO: 25) (SEQ ID NO: 33) SEQ ID NO: 81 KZ144 Melittin + n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 31) SEQ ID NO: 82 KZ144 SMAP-29 +++ +++ n.d. (SEQ ID NO: 25) (SEQ ID NO: 11) SEQ ID NO: 83 KZ144 Cecropin A (A. ++ + ++ (SEQ ID NO: 25) aegypti) (SEQ ID NO: 14) SEQ ID NO: 84 KZ144 Pleurocidin + n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 6) SEQ ID NO: 85 KZ144 Cecropin A (A. + n.d. n.d. (SEQ ID NO: 25) melanogaster) (SEQ ID NO: 15) SEQ ID NO: 86 KZ144 Buforin II + n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 8) SEQ ID NO: 87 KZ144 Sarcotoxin IA ++ ++ ++ (SEQ ID NO: 25) (SEQ ID NO: 16) SEQ ID NO: 93 KZ144 α4 helix ± n.d. n.d. (SEQ ID NO: 25) (SEQ ID NO: 92) Abreviations: ± <1 log; +: 1 log; ++: 2-3 log; +++: 4 or more logs; n.d. means that this strain was not tested with the respective fusion protein.

EXAMPLE 5: ANTIMICROBIAL ACTIVITY OF THE ENDOLYSIN STM0016 MODIFIED WITH VARIOUS PEPTIDE STRETCHES ON THE N-TERMINUS

(41) The fusion proteins consisting of STM0016 and the peptide stretch Sarcotoxin IA or SMAP-29 was constructed as described in example 3.

(42) E. coli DSMZ 11753, Salmonella typhimujrium DSMZ 17058 and Pseudomonas aeruginosa PAO1p cells (Burn wound isolate, Queen Astrid Hospital, Brussels; Pirnay J P et al. (2003), J Clin Microbiol., 41(3):1192-1202) were used as test strains. The antimicrobial activity of the fusion proteins consisting of the endolysin STM0016 and the peptide Sarcotoxin IA or SMAP-29 was examined as described in example 4. The antimicrobial activity of these fusion proteins is given in the following table.

(43) TABLE-US-00012 TABLE 12 Activity against Peptide stretch Activity against Activity against Salmonella (N-terminal unless Pseudomonas E. coli typhimurium Fusion protein Enzyme part otherwise indicated) aeruginosa DSMZ 11753 DSMZ 17058 SEQ ID NO: 88 STM0016 Sarcotoxin IA + n.d. + (SEQ ID NO: 22) (SEQ ID NO: 16) SEQ ID NO: 89 STM0016 SMAP-29 + + + (SEQ ID NO: 22) (SEQ ID NO: 11) Abreviations: +: 1 log; n.d. means that this strain was not tested with the respective fusion protein.

EXAMPLE 6: ANTIMICROBIAL ACTIVITY OF THE ENDOLYSIN N4GP61 MODIFIED WITH A PEPTIDE STRETCH ON THE N-TERMINUS

(44) The fusion protein consisting of N4gp61 and the peptide stretch SMAP-29 was constructed as described in example 3.

(45) E. coli DSMZ 11753, Salmonella typhimujrium DSMZ 17058 and Pseudomonas aeruginosa PAO1p cells (Burn wound isolate, Queen Astrid Hospital, Brussels; Pirnay J P et al. (2003), J Clin Microbiol., 41(3):1192-1202) were used as test strains. The antimicrobial activity of the fusion protein consisting of the endolysin N4gp61 and the peptide SMAP-29 was examined as described in example 4. The antimicrobial activity of this fusion protein is given in the following table.

(46) TABLE-US-00013 TABLE 13 Activity against Peptide stretch Activity against Activity against Salmonella (N-terminal unless Pseudomonas E. coli typhimurium Fusion protein Enzyme part otherwise indicated) aeruginosa DSMZ 11753 DSMZ 17058 SEQ ID NO: 90 N4-gp61 SMAP-29 + + + (SEQ ID NO: 23) (SEQ ID NO: 11) Abreviations: +: 1 log; n.d. means that this strain was not tested with the respective fusion protein.

EXAMPLE 7: ANTIMICROBIAL ACTIVITY OF THE ENDOLYSIN GP188 MODIFIED WITH A PEPTIDE STRETCH ON THE N-TERMINUS

(47) The fusion proteins consisting of the endolysin gp188 and the peptide stretches α4 helix, SMAP-29 or Sarcotoxin IA were constructed as described in example 1. E. coli DSMZ 11753, Acinetobacter baumannii DSMZ 30007 and Pseudomonas aeruginosa PAO1p cells (Burn wound isolate, Queen Astrid Hospital, Brussels; Pirnay J P et al. (2003), J Clin Microbiol., 41(3):1192-1202) were used as test strains. The antimicrobial activity of the fusion proteins consisting of the endolysin gp188 and the respective peptide stretches was examined as described in example 4. The antimicrobial activity of these fusion proteins is given in the following table.

(48) TABLE-US-00014 TABLE 14 Activity against Peptide stretch Activity against Activity against Acinetobacter (N-terminal unless Pseudomonas E. coli baumannii Fusion protein Enzyme part otherwise indicated) aeruginosa DSMZ 11753 DSMZ 30007 SEQ ID NO: 94 gp188 α4 helix ± n.d. n.d. (SEQ ID NO: 2) (SEQ ID NO: 92) SEQ ID NO: 95 gp188 SMAP-29 ++ ++ ++ (SEQ ID NO: 2) (SEQ ID NO: 11) SEQ ID NO: 96 gp188 Sarcotoxin IA + + + (SEQ ID NO: 2) (SEQ ID NO: 16) Abreviations: ± <1 log; +: 1 log; ++: 2-3 log; n.d. means that this strain was not tested with the respective fusion protein.

EXAMPLE 8: ANTIMICROBIAL ACTIVITY OF THE SALMONELLA ENDOLYSIN MODIFIED WITH THE PEPTIDE STRETCH SMAP-29 ON THE N-TERMINUS

(49) The fusion proteins consisting of the Salmonella endolysin having an amino acid sequence according to SEQ ID NO: 3 and the peptide stretch SMAP-29 were constructed analogous to example 3. E. coli DSMZ 11753 and Salmonella typhimurium DSMZ 17058 were used as test strains. The antimicrobial activity of the fusion protein was examined as described in example 4. The antimicrobial activity of this fusion protein is given in the following table.

(50) TABLE-US-00015 TABLE 15 Activity against Peptide stretch Activity against Salmonella (N-terminal unless E. coli typhimurium Fusion protein Enzyme part otherwise indicated) DSMZ 11753 DSMZ 17058 SEQ ID NO: 97 Salmonella SMAP-29 + + endolysin (SEQ ID NO: 11) (SEQ ID NO: 3) Abreviations: +: 1 log;

EXAMPLE 9: ANTIMICROBIAL ACTIVITY OF THE ACINETOBACTER BAUMANNII ENDOLYSIN MODIFIED WITH VARIOUS PEPTIDE STRETCHES ON THE N-TERMINUS

(51) The fusion proteins consisting of the Acinetobacter baumannii endolysin having an amino acid sequence according to SEQ ID NO: 5 and the peptide stretches SMAP-29, Pseudin 1 and Sushi 1 were constructed analogous to example 3. Acinetobacter baumannii DSMZ 30007 and Pseudomonas aeruginosa PAO1p cells (Burn wound isolate, Queen Astrid Hospital, Brussels; Pirnay J P et al. (2003), J Clin Microbiol., 41(3):1192-1202) were used as test strains. The antimicrobial activity of the fusion proteins was examined as described in example 4. The antimicrobial activity of these fusion proteins is given in the following table.

(52) TABLE-US-00016 TABLE 16 Activity against Peptide stretch Activity against Acinetobacter (N-terminal unless Pseudomonas baumannii Fusion protein Enzyme part otherwise indicated) aeruginosa DSMZ 30007 SEQ ID NO: 98 Acinetobacter Pseudin 1 ± n.d. baumannii (SEQ ID NO: 29) endolysin (SEQ ID NO: 5) SEQ ID NO: 99 Acinetobacter SMAP-29 ++ ++ baumannii (SEQ ID NO: 11) endolysin (SEQ ID NO: 5) SEQ ID NO: 100 Acinetobacter Sushi 1 + + baumannii (SEQ ID NO: 32) endolysin (SEQ ID NO: 5) Abreviations: ± <1 log; +: 1 log; ++: 2-3 log; n.d. means that this strain was not tested with the respective fusion protein.

(53) The fusion proteins in Table 11 to 16 without any tag and linker were also tested with the activity assays described above. They all showed antimicrobial activity against the used bacterial strains (data not shown).