WO 00/29588 PCT/US99/27403 LYSOZYME FUSION PROTEINS IN INFECTIONS Related Applications This application is a Continuation-In-Part of U.S. Application Serial No. 09/193,877 filed November 18, 1998, now pending. The U.S. Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others on reasonable terms as provided for by the terms of Grant Nos. R01-HL56285 and HL56285S awarded by the National Institutes of Health. Field of the Invention The invention relates to prophylactic and therapeutic uses of recombinant lysozyme and a lysozyme/surfactant protein-B fusion protein in bacterial infections. 5 Background of the Invention Bacterial infections remain a leading cause of worldwide morbidity and mortality. While antibiotics administered to treat bacterial infections are often safe and efficacious, there is widespread concern over bacterial strains WO 00/29588 PCT/US99/27403 -2 that have become resistant to classic antibiotic treatment regimens. In individuals infected with resistant strains, antibiotic administration results in incomplete and ineffective treatment, necessitating additional treatment along with propagation of the resistant strains. Preventative measures and 5 alternative treatments that do not rely on antibiotics are therefore desirable. In certain individuals such as those who are immunocompromised, who are in less than optimal health, who lack fully functional immune systems such as neonates or geriatric patients, or who suffer from a disease affecting the respiratory tract such as cystic fibrosis or the gastrointestinal tract such 10 as ulcerative colitis or sprue, bacterial infections may have severe consequences leading to serious illness or even death. For example, production of altered mucus in cystic fibrosis patients leads to dilation of the exocrine ducts, destruction of acinar tissue, and replacement of the destroyed tissue by fibrous connective tissue. Involvement of the lungs leads to 15 pneumonia and bronchiectasis. The paucity of systemic dissemination of infection in these patients, even in the presence of substantial bacterial colonization of the lungs, indicates that their systemic immunity is essentially intact, yet they are susceptible to pulmonary infections. These patients often die in their teens or early twenties from terminal lung infections in spite of 20 aggressive antibiotic therapy. The respiratory and gastrointestinal tracts are frequent sites of bacterial infections in normal individuals. The normal respiratory tract has natural clearance mechanisms that help to prevent bacterial colonization. These mechanisms include the presence of a mucus gel that acts as a barrier WO 00/29588 PCT/US99/27403 -3 to bacterial invasion, the propulsive forces of the cilia on the epithelial lining of the airways, and the secretion of antibacterial humoral factors such as the secretory immunoglobulins IgA and IgM, the proteins lactoferrin, betalysin and fibronectin, complement components and the enzyme lysozyme. Of these 5 antimicrobial factors, lysozyme is the best established antimicrobial substance in airway secretions. Human lysozyme is a naturally occurring enzyme that is known to exhibit bactericidal activity in vitro and thus would appear to be a promising way to prevent and/or treat bacterial infections. Lysozyme is a small (15 10 kilodaltons), basic protein that is produced in most tissues. It is secreted and is present in most body secretions such as mucus. In the lungs, immunohistochemical methods have localized lysozyme to the bronchial serous submucosal glands, alveolar macrophages and lamellar bodies in Type II alveolar epithelial cells. Approximately 80% of lysozyme secreted into the 15 airway comes from the mucosal layer of the upper airways. In vitro, lysozyme has been demonstrated to act independently to cause bacterial death. It is known that one way lysozyme kills bacteria is by hydrolyzing the glycosidic bond between C-1 of N-acetylmuramic acid and C-4 of N-acetylglucosamine in the bacterial polysaccharide cell wall. Lysozyme can 20 also kill bacteria by acting synergistically with other proteins such as complement or antibody to lyse bacterial cells. Lysozyme, produced by polymorphonuclear leukocytes such as neutrophils, inhibits chemotaxis of polymorphonuclear leukocytes and limits the production of oxygen free radicals following an infection. This limits the degree of inflammation, while at the WO 00/29588 PCT/US99/27403 -4 same time enhances phagocytosis by these cells. Lysozyme is also probably involved in the response of airway tissue to injury. While the antibacterial effects of lysozyme in vitro have been well documented, there has heretofore been no way to exploit these effects of 5 lysozyme for in vivo use. Previous reports furthermore implied that sustained lysozyme administration would be harmful. Pulmonary surfactant, a complex mixture of phospholipids and proteins, is synthesized and secreted by alveolar type II epithelial cells, a specialized exocrine cell. Surfactant protein-B (SP-B) is one of the protein 10 components of pulmonary surfactant. Normal respiratory function requires pulmonary surfactant for maintenance of alveolar patency. A method and composition for the prophylaxis and/or therapeutic treatment of bacterial infections, particularly respiratory bacterial infections as frequently occurs in patients with cystic fibrosis, is thus desirable. 1 5 Summary of the Invention This invention is directed to a composition for the prophylaxis or therapeutic treatment of a bacterial infection in a mammal. The composition is either a lysozyme/surfactant protein-B (SP-B) fusion protein SEQ ID NO:3 (rat lysozyme SEQ ID NO:1 fused with the carboxyl terminal SP-B propeptide 20 and the preceding ten amino acids of the mature SP-B peptide SEQ ID NO:4), or a lysozyme/SP-B fusion protein SEQ ID NO:6 (human lysozyme SEQ ID NO:5 fused with the carboxyl terminal SP-B propeptide and the preceding ten amino acids of the mature SP-B peptide SEQ ID NO:4), or recombinant lysozyme alone SEQ ID NO:1. The composition prevents or treats a respiratory WO 00/29588 PCT/US99/27403 -5 infection such as occurs frequently in individuals with cystic fibrosis, or may occur in individuals having other types of respiratory diseases, or a gastrointestinal infection. The invention's use of recombinant lysozyme reduces mortality and enhances in vivo clearance of bacteria such as 5 Pseudomonas aeruginosa, which is the major airway pathogen in patients with cystic fibrosis. Furthermore, the elevated lysozyme activity in bronchoalveolar lavage fluid resulting from administration of lysozyme is not associated with altered lung function or structure, or with chronic inflammatory disease. The invention is also directed to a method of preventing or 10 treating a bacterial infection in a mammal by administering a lysozyme/SP-B fusion protein SEQ ID NO:3 or SEQ ID NO:6 in a pharmaceutically acceptable composition in a dosing regimen sufficient to prevent or treat the infection. The route of administration may be parenteral, for example by inhalation, or nonparenteral. 15 The invention is still further directed to a fusion protein SEQ ID NO:3, comprising a rat lysozyme SEQ ID NO:1 and a carboxyl terminal SP-B propeptide with the ten terminal amino acids of the mature peptide SEQ ID NO:4, having antibacterial activity in a mammal. The invention is still further directed to a fusion protein SEQ ID 20 NO:6, comprising a human lysozyme SEQ ID NO:5 and a carboxyl terminal SP B propeptide with the ten terminal amino acids of the mature peptide SEQ ID NO:4, having antibacterial activity in a mammal. The invention is additionally directed to a method of treating a bacterial respiratory infection in an individual having cystic fibrosis by WO 00/29588 PCT/US99/27403 -6 administering a lysozyme/SP-B fusion protein SEQ ID NO:3 or SEQ ID NO:6 to the individual. The fusion protein may be administered by aerosol installation and/or inhalation. The invention is also directed to a method of preventing or 5 treating a bacterial infection in a mammal by administering recombinant lysozyme SEQ ID NO:1 in a pharmaceutically acceptable composition at a dose sufficient to prevent or treat the infection, such as treating a bacterial infection in a patient with cystic fibrosis. The invention is additionally directed to a composition comprising 10 recombinant lysozyme SEQ ID NO:1 having antibacterial activity. These and other methods and compositions will be apparent in light of the following figures and detailed description. Brief Description of the Figures FIG. 1 is a histogram showing bacterial clearance from lungs of 15 transgenic (lysozyme/surfactant protein-B fusion protein) and control mice. FIG. 2A is a photograph showing expression of recombinant lysozyme SEQ ID NO:1 in a transgenic mouse line probed with mouse lysozyme complementary DNA (cDNA), and FIG 2B is a photograph showing expression of recombinant lysozyme SEQ ID NO:1 in the same transgenic 20 mouse line probed with rat cDNA. FIG. 3 is a photograph showing analysis of lysozyme protein expression. FIG. 4 is a histogram of lysozyme activity in bronchoalveolar lavage (BAL) fluid.
WO 00/29588 PCT/US99/27403 -7 FIG. 5A is a photograph showing lysozyme cellular localization in wild type mice, and FIG. 5B is a photograph showing lysozyme cellular localization in transgenic mice. FIG. 6A is a photograph showing lung structure in transgenic mice 5 that overexpress lysozyme, and FIG. 6B is a photograph showing lung structure in wild type mice. FIG. 7 is a histogram illustrating the cellular composition of BAL fluid. FIG 8 is a histogram showing clearance of Group B Streptococcus 10 (GBS) from the lungs at six hours post-infection. FIG 9 is a histogram showing clearance of Pseudomonas aeruginosa from the lungs at twenty-four hours post infection. Detailed Description of the Preferred Embodiment Preparation of Recombinant Lysozyme and SP-B and Fusion Protein 15 Rat lysozyme is a hydrophobic peptide of 148 amino acids SEQ ID NO:1. Human lysozyme SEQ ID NO:5 also has 148 amino acids and has 69% homology with rat lysozyme. Surfactant protein-B (SP-B) is a hydrophobic peptide of 79 amino 20 acids that avidly associates with surfactant phospholipids in the alveolar airspace. Human SP-B is synthesized by alveolar type II epithelial cells as a prepropeptide of 381 amino acids SEQ ID NO:2. Mature SP-B is generated by sequential cleavage of a 23 amino acid signal peptide, an amino terminal (N terminal) propeptide of 177 amino acids, and a carboxyl- terminal (C-terminal) 25 propeptide of 102 amino acids. The C-terminal propeptide has been shown to WO 00/29588 PCT/US99/27403 -8 function in maintenance of the size of lamellar bodies which store SP-B and in determining the intracellular surfactant pool size. Complementary DNA (cDNA) corresponding to rat lysozyme and human SP-B was synthesized. The protocol used for synthesis was that 5 described in Akinbi et al., J. Biol. Chem. 1997; 272: 9640-9647, which is expressly incorporated by reference herein in its entirety. Complementary DNA to rat lysozyme was generated as follows. Alveolar Type 11 epithelial cells were isolated from adult rat lung as described by Dobbs, et al., An improved method for isolating Type II cells in high yield and purity, Am. Rev. Respir. Dis. 10 134:140-145, 1986). Total RNA was isolated from Type II cells by the method of Chirgwin, et al., The isolation of biologically active ribonucleic acid from sources enriched in ribonuclease, Biochemistry 18:5294-5299, 1979) and polyA' RNA by the method of Aviv and Leder, Purification of biologically active globin messenger RNA by chromatography on oligothymidylic 15 acid-cellulose, Proc. Natl. Acad. Sci. USA 69:1408-1412, 1972. Single stranded cDNA, generated from isolated polyA' RNA with reverse transcriptase (Maniatis, et al. Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 1982), was used as a template for PCR amplification of the entire coding sequence of lysozyme using 20 oligonucleotide primers based on the published sequence of the rat enzyme by Yeh, et al. Evolution of rodent lysozymes: Isolation and sequence of the rat lysozyme genes, Mol. Physiol. Evol. 2:25-75, 1993. Isolation of the human SP-B cDNA has previously been described (Glasser, S.W., et al. cDNA and WO 00/29588 PCT/US99/27403 -9 deduced amino acid sequence of human pulmonary surfactant-associated proteolipid SPL (Phe)., Proc. Natl. Acad. Sci. USA 84:4007-4011, 1987). The synthesized cDNAs were either generated into a chimeric molecule consisting of the rat lysozyme protein and a carboxyl terminal (C 5 terminal) propeptide of a human surfactant protein B (SP-B) for evaluation, or recombinant rat lysozyme (SEQ ID NO:1) alone was generated. Specifically, this chimeric molecule was a fusion protein of residues 1-148 of rat lysozyme SEQ ID NO:1 and residues 270-381 of SEQ ID NO:2, shown in SEQ ID NO:4, forming a lysozyme/surfactant-B fusion protein SEQ ID NO:3. 10 Preparation of Transgenic Mice Overexpressing Lysozyme/SP-B Fusion Protein Three lines of transgenic mice were generated that expressed a cDNA construct comprising the entire coding sequence for rat lysozyme SEQ ID NO:1 and the entire C-terminal propeptide of SP-B along with the preceding ten amino acids from the C-terminal of the mature peptide SEQ ID NO:4. The 15 coding sequence for rat lysozyme SEQ ID NO:1 was cloned in frame with the coding sequence for the C-terminal propeptide and preceding 10 amino acids for human pulmonary surfactant protein B propeptide SEQ ID NO:4. FVB/N transgenic mice expressing a transgene construct encoding the fusion protein SEQ ID NO:3 under the control of the 3.7 kilobase (kb) human surfactant 20 protein C (SP-C) promoter were generated as described in Lin et al. (J. Biol. Chem. 1996; 271: 19689-19695) which is expressly incorporated by reference herein in its entirety. The expression of transgene RNA in these mice was restricted to the distal respiratory epithelium.
WO 00/29588 PCT/US99/27403 - 10 Expression of the chimeric protein SEQ ID NO:3 was confirmed by Western blot analysis, as is known to one skilled in the art, using antibody #R96189 generated and characterized as reported by Lin et al., Structural requirements for intracellular transport of pulmonary surfactant protein B 5 (SP-B) (Biochim. Biophys. Acta Mol. Cell. Res. 1312:177-185, 1996), that detects the C-terminal- propeptide of SP-B proprotein. A protein of approximately 29 kDa was detected in transgenic mice using this antibody, as would be predicted by the size of the construct of 15 kDa rat lysozyme and 14 kDa C-terminal propeptide and preceding 10 amino acids SEQ ID NO:3. The 10 transgene product was detected in both lung homogenates and in bronchoalveolar lavage (BAL) fluids, consistent with secretion of the chimeric protein SEQ ID NO:3 into the alveolar space. Constitutive expression of the chimeric protein SEQ ID NO:3 was not associated with altered lung structure, as assessed by light microscopy evaluation of lung tissue stained with 15 hematoxylin and eosin. Preparation of Transgenic Mice Overexpressing Lysozyme In order to assess the role of lysozyme in pulmonary host defense, transgenic mouse lines were generated in which rat lysozyme SEQ ID NO:1 was targeted to the distal airway epithelium under the direction of the 3.7 kb 20 human surfactant protein C (SP-C) promoter. Seven of twenty-one offspring from fertilized oocyte injections were positive for the transgene as assessed by PCR and confirmed by Southern blot analyses of tail DNA (now shown). Transgenic mice were indistinct from wild type littermates with respect to body weight, lung weight, longevity and reproductive capability. Two WO 00/29588 PCT/US99/27403 -11 transgenic lines (3.5 and 2.6) had increased levels of lysozyme protein in bronchoalveolar lavage fluid. Efficacy of Lysozyme/SP-B Fusion Protein Five-week old transgenic mice (n = 56) carrying the fusion protein 5 SEQ ID NO:3 (treated) and their wild type litter mates (control) were infected with 106 strain III group B Streptococci (GBS) via intratracheal administration. Aliquots of GBS were grown in Todd Hewitt broth at 37 0 C overnight and bacteria were pelleted by centrifugation. The bacterial pellet was suspended in sterile phosphate buffered saline (PBS) at a concentration of 1 0 7 /ml. One 10 hundred microliters of bacterial suspension (106 bacteria) were drawn into tuberculin syringes fitted with 27 gauge needles in preparation for the intratracheal injection. All injections were carried out in a sterile environment. Five to six week old mice maintained in clean rooms were used for bacterial clearance studies. Mice were anesthetized with a mixture of 15 nitrous oxide and oxygen. The trachea was exposed through a midline incision and dissection through the thyroid gland. One hundred microliters of bacterial suspension was instilled into the trachea with a 27 gauge needle just below the cricoid cartilage. The two halves of the thyroid muscles were apposed at the midline and the skin incision was closed by approximating the two edges 20 with surgical glue. Animals were housed for either 6 or 24 hours prior to sacrifice. After either 6 or 24 hours post infection, lungs from treated and control mice were harvested, weighed, homogenized in PBS and plated on BAP to evaluate formation of GBS colonies. The cultured plates were incubated at WO 00/29588 PCT/US99/27403 -12 37 0 C for approximately 16-18 hours (overnight). Colony forming units per gram of lung tissue (CFU/g) were assessed by manual counting of bacterial colonies. As shown in FIG. 1, transgenic mice harboring the chimeric 5 protein SEQ ID NO:3 had significantly fewer CFU/g of lung tissue at both 6 hours and 24 hours following infection with GBS. The number of CFU in most transgenic mice was even less than the number of bacteria that had been administered. As shown in FIG. 1, transgenic mice had significantly enhanced clearance of GBS from the lungs at both the 6 h and 24 h post infection time 10 points. At 6 h post-infection, BAP inoculated with lung tissue from transgenic (treated) mice had 1.99 ± 1.4 x 104 CFU/g of tissue, while BAP inoculated with lung tissue from wild type (control) mice had 25.49 ± 12.43 x 104 CFU/g (p<0.006). At 24 h post-infection, BAP inoculated with lung tissue from transgenic (treated) mice had 9.9 ± 6.43 x 104 CFU/g tissue, while BAP 15 inoculated with lung tissue from wild type (control) mice had 67.29 ± 34.2 x 104 CFU/g tissue (p < 0.04). These results suggested that the expression of lysozyme/surfactant protein-B fusion protein SEQ ID NO:3 in the airway of the transgenic mice facilitated bacterial clearance from the airway. The results 20 show that in mice carrying the lysozyme/surfactant protein-B fusion protein SEQ ID NO:3, bacterial proliferation was inhibited at 6 hours post infection, while bacterial clearance was enhanced at 24 hours post infection. Transgenic mice which expressed a lysozyme/surfactant protein B fusion protein SEQ ID NO:3 in the distal airway were twelve-fold more efficient in clearing bacteria WO 00/29588 PCT/US99/27403 - 13 in the airway than their wild type littermates. This lysozyme-produced efficacy is particularly striking, since wild type mice are inherently very efficient in clearing bacteria in the airway. Two other lines of transgenic mice expressed the 5 lysozyme/surfactant protein-B chimeric protein SEQ ID NO:3 although at lower levels. Bacterial clearance in these lines was correspondingly lower, although still significantly greater than that in wild type control mice. Lysozyme enzyme activity was increased 40% (relative to wild type littermates) in bronchoalveolar lavage fluid of the transgenic line expressing the highest levels 10 of the lysozyme/SP-B fusion protein SEQ ID NO:3. Since bacterial clearance was enhanced twelve-fold in this line, the antibacterial effect may have been conferred by the SP-B C-terminal propeptide with the preceding amino acids from the mature peptide SEQ ID NO:4 alone, or the result of the combined action of SP-B and lysozyme SEQ ID NO:3 components; alternatively the effect 15 may be due to increased lysozyme SEQ ID NO:1 activity. Analysis of Lysozyme Transgene Expression RNA With reference to FIGS. 2A and 2B, expression of the lysozyme transgene was assessed by Northern blot analysis of 2 pg of total cellular RNA isolated from lung tissue of five-week old transgenic mice (line 20 3.5). These are shown in FIG. 2A lanes 4 and 5, and in FIG. 2B lanes 3 and 4. Control wild type littermates are shown in FIG. 2A lanes 1-3 and in FIG. 2B lanes 1 and 2. Both transgenic and control samples were probed with mouse complementary DNA (cDNA) (FIG. 2A) and rat cDNA (FIG. 2B). In FIG. 2A, endogenous mouse lysozyme was the larger of the 2 transcripts. The mouse WO 00/29588 PCT/US99/27403 -14 cDNA probe, because of cross hybridization between rat lysozyme and mouse cDNA, detected rat lysozyme. A cDNA probe specific for rat lysozyme SEQ ID NO:1 detected a -1 kb transcript, the predicted size of the lysozyme transgene (FIG. 2B). 5 When the same RNA samples were probed with a mouse lysozyme cDNA (FIG. 2A), both the larger endogenous mouse lysozyme mRNA and rat lysozyme mRNA were detected. The mouse lysozyme mRNA was used as an internal control for normalization of samples. Protein Because rat and mouse lysozyme have very similar 10 molecular weights, it was not possible to resolve the proteins by SDS-PAGE. Total levels of lysozyme (rat and mouse) were estimated by Western blot analysis of equal amounts of protein from lung homogenates or BAL fluid from five-week old transgenic mice and wild type littermate controls. FIG. 3 shows a Western blot analysis of 1 plg of total lung protein from transgenic mouse line 15 3.5 (lanes 4-6) and control wild type littermates (lanes 1-3). Proteins were fractionated by SDS-PAGE under non-reducing conditions, blotted onto a nitrocellulose membrane and incubated with anti-human lysozyme antibody, which detected both mouse and rat lysozyme (molecular weight of about 15 kD). 20 As shown in FIG. 3, mice from transgenic line 3.5 (lanes 4-6) had a four-fold increase in the level of lysozyme protein in both lung homogenate and BAL fluid over wild type littermates (lanes 1-3). Mice from transgenic line 2.6 had a two-fold increase in lysozyme protein compared to control wild type littermates (not shown).
WO 00/29588 PCT/US99/27403 -15 Lysozyme Enzyme Activity With reference to FIG. 4, lysozyme enzyme activity was assessed in BAL fluid from five-week old transgenic mice from line 3.5 (number of animals (n)=4) and line 2.6 (n=4), and control wild type littermates (n= 7) by a turbidimetric assay. Ten nanograms of BAL fluid 5 protein was incubated with Micrococcus lysodeikticus suspended at an optical density (O.D.) of 1 at 450 nm. Following one hour of incubation at 370C, the change in O.D. of the suspension was determined. Purified chicken egg white lysozyme was used to generate a standard curve. As shown in FIG. 4, mice from transgenic line 3.5 had a 17.5-fold 10 increase in lysozyme activity compared to wild type mice (550 units/ng BAL protein versus 31 units/ng BAL protein in wild type mice). The data are mean + standard error of the mean (SEM), with p= <0.0001 for wild type versus transgenic line 3.5, p= <0.0001 for wild type versus 2.6, and p= <0.0008 for transgenic line 3.5 versus transgenic line 2.6. As predicted from Western 15 blot analyses, mice from transgenic line 2.6 had increased lysozyme enzyme activity (205 units/ng BAL protein) relative to wild type controls (p = 0.02), but lower activity relative to transgenic line 3.5. Spatial Expression of Lysozyme With reference to FIGS. 5A and 5B, cellular localization of 20 lysozyme protein in transgenic and control mice was obtained. Paraffin embedded lung sections from five-week old transgenic mice from line 3.5 (FIG 5B) and wild type control littermates (FIG. 5A) were immunostained with hematoxylin/eosin using an anti-human lysozyme antibody which detects both rat and mouse lysozyme and were photographed at a magnification of 80 x.
WO 00/29588 PCT/US99/27403 -16 The overall architecture of lung sections of transgenic mice were indistinct from wild type littermates. There was no evidence of pulmonary edema or vascular congestion, and inflammatory cells were not detected in lung sections from uninfected transgenic mice. 5 Lysozyme was detected in Type II cells in wild type and transgenic mice, with more intense staining in Type II cells from transgenic mice. Transgenic mice, in addition, have expression in bronchiolar epithelial cells. In wild type mice, endogenous expression of epithelial lysozyme was restricted to Type II cells, whereas in transgenic mice, lysozyme expression 10 was equally prominent in non-ciliated bronchiolar cells and Type II cells. Effect of Lysozyme Overexpression on Lung Structure With reference to FIGS. 6A and 6B, gross histologic features of paraffin-embedded, hematoxylin/eosin-stained sections of lungs from uninfected five-week old transgenic mice (FIG. 6A) and control wild type 15 littermates (FIG. 6B) were compared at a magnification of 40 x. Total and differential cell counts (Cytospin® preparations stained with Diff-Quick®) in BAL fluids were performed. As shown in FIG. 7, there were no significant differences in either the total cell count or in the distribution of cell types recovered from BAL fluid 20 in transgenic mice compared to wild type control littermates, with p = 0.98 for total cell counts, p= 0.74 for macrophages, and p=0.95 for lymphocytes. OData are mean ± SEM. Data are mean +_ SEM.
WO 00/29588 PCT/US99/27403 - 17 Effect of Lysozyme Overexpression on Bacterial Pathogen Clearance from the Airway Clearance of Group B Streptococcus (GBS). To determine if 5 increased lysozyme levels in the airway enhanced clearance of bacteria from the lungs, bacterial counts in quantitative cultures of lung homogenates from transgenic mice and wild type littermate controls were compared following an intratracheal injection of 106 colony forming units (CFU) of GBS. All mice survived until sacrifice at six hours post infection. 10 The results, shown in FIG. 8, are expressed as CFU/g lung tissue ± SEM, with p=0.0172 in wild type (n=20) versus transgenic line 3.5 (n=19), and p=0.19 in wild type versus transgenic line 2.6 (n=10). Mice from transgenic line 3.5 had a three-fold enhancement of GBS clearance (2.1 ± 0.1 x 106 CFU/g lung tissue versus 6.8 ± 0.5 x 106 CFU/g lung tissue in 15 wild type mice, p=0.0172). The incidence of systemic dissemination of infection, as assessed by growth of GBS on plates inoculated with splenic homogenates, was less in transgenic mice (27% versus 60%, p=0.04). Clearance of GBS from the lungs was enhanced less than two-fold in mice from transgenic line 2.6 (4.2 ± 0.8 x 106 CFU/g lung tissue versus 7.1 ± 0.6 20 x 106 in wild type mice, p=0.19). Clearance of Pseudomonas aeruginosa. Transgenic and wild type mice received intratracheal injections with 107 CFU of Pseudomonas aeruginosa. All transgenic mice survived until sacrifice at 24 hours post infection. In contrast, 20% of infected wild type mice died. 25 FIG. 9 shows CFU/g lung tissue ± SEM. Bacterial clearance was enhanced approximately thirty-fold in mice from transgenic line 3.5 (n= 10) WO 00/29588 PCT/US99/27403 -18 (1.06 ± 0.05 x 106 CFU/g lung tissue, compared to 3.24 ± 0.41 x 10' in wild type littermate controls (n=10, p=O.03). Bacterial clearance was enhanced approximately six-fold in mice from transgenic line 2.6 (n = 10) (6.52 ± 0.71 x 106 CFU/g lung tissue, p= 0.05). Systemic bacterial dissemination 5 was not detected in surviving mice at 24 hours post-infection. The lysozyme/SP-B fusion protein SEQ ID NO:3 or SEQ ID NO:6 or recombinant lysozyme SEQ ID NO:1 of the invention may be used to treat and/or reduce bacterial colonization of the airway. The latter use would be extremely beneficial in treating individuals with cystic fibrosis, since chronic 10 bacterial colonization of the major airways, particularly by Pseudomonas aeruginosa, with consequent debilitating exacerbations is the major cause of the morbidity and mortality suffered by cystic fibrosis patients. Cystic fibrosis is a systemic disease in which mucus secretion is altered so that a viscid mucus is produced. Production of altered mucus leads 15 to dilation of the exocrine ducts, destruction of acinar tissue, and replacement of the destroyed tissue by fibrous connective tissue. Involvement of the lungs leads to pneumonia and bronchiectasis. These patients often succumb at a young age to terminal lung infections with Pseudomonas aeruginosa in spite of aggressive antibiotic therapy. Therefore, the bactericidal activity of 20 lysozyme or the lysozyme/surfactant protein-B fusion protein in vivo offers a potential therapeutic strategy for suppressing bacterial colonization of the airways in cystic fibrosis patients without compromising whatever degree of respiratory function the patient exhibits.
WO 00/29588 PCT/US99/27403 -19 It will be appreciated that prophylaxis or therapeutic treatment by the method and composition of the invention may range from total prevention, reduction of the bacterial load, amelioration of the severity of, or elimination of a bacterial infection. The lysozyme/SP-B fusion protein SEQ ID NO:3 or 5 SEQ ID NO:6 or recombinant lysozyme SEQ ID NO:1 may be used to protect against all bacterial strains which colonize the respiratory tract such as, for example, Staphylococcus aureus, Streptococcus species, Streptococcus pneumoniae, Neisseria meningitidis, Neisseria gonorrhoeae, Klebsiellae species, Proteus species, Pseudomonas cepacia, Haemophilus influenzae, Bordetella 10 pertussis, Mycoplasma pneumoniae, Legionella pneumophila. Additionally, the invention may be used to combat gastrointestinal infections. The lysozyme/surfactant protein-B fusion protein SEQ ID NO:3 or SEQ ID NO:6 may be administered by an enteral route to target the gastrointestinal tract for treating or preventing infections with bacterial strains 15 that colonize the gastrointestinal tract such as, for example, Salmonellae species, Shigellae species, Escherichia coli, and Vibrio species, Yersinia enterocolitica, Campy/obacter fetus, ssp. jejuni, and Helicobacter pylori. The lysozyme/SP-B may be formulated for oral administered as a solid or liquid in the form of a capsule, tablet, syrup, and so on. 20 Other variations or embodiments of the invention will also be apparent to one of ordinary skill in the art from the above description. For example, other fusion proteins besides surfactant protein-B, such as members of the structurally related saposin protein family such as saposin A, saposin B, saposin C, saposin D, NK lysin, pore forming peptide of amoebapore A etc., WO 00/29588 PCT/US99/27403 - 20 could be generated. Various modes of administration besides inhalation could be used, such as injection, etc. The fusion protein SEQ ID NO:3 or SEQ ID NO:6 or recombinant lysozyme SEQ ID NO:1 may be administered either alone or in combination with antibiotic therapy. Thus, the forgoing embodiments are 5 not to be construed as limiting the scope of this invention. What is claimed is: