Treatment of topical and systemic bacterial infections

Abstract

Bacteriophage covalently attached to a carrier particle with an average diameter of from 0.1 microns to 15 microns, are used in topical treatment of bacterial infection. Bacteriophage covalently attached to a carrier particle of average diameter 7 microns or less are used in systemic treatment of bacterial infection. A plurality of bacteriophages lytic against different bacterial strains gives wide antibacterial activity. A combination therapy comprises administration of antibiotic and bacteriophage covalently attached to a carrier particle.

Claims

1. A method of treating an acne infection of intact skin in a subject in need thereof, comprising administering a topical formulation to the subject, wherein the topical formulation comprises a bacteriophage covalently attached to a carrier particle.

2. The method of claim 1, where in the carrier particle is approximately spherical and has an average diameter of up to 20 microns.

3. The method of claim 1, wherein the treatment is free of salicylic acid and free of benzoyl peroxide.

4. The method of claim 1, wherein the topical formulation comprises bacteriophage active against 3 or more strains of bacteria, wherein the bacteriophage is covalently attached to carrier particles.

5. The method of claim 1, wherein the topical formulation is in the form of a gel, cream, or lotion.

6. The method of claim 5, wherein the topical formulation comprises one or more or all of: a gel-forming agent, a cream-forming agent, a wax, an oil, a surfactant, and a binder.

7. The method of claim 6, wherein the topical formulation comprises carrier particles with an average diameter of up to 20 microns.

8. The method of claim 7, wherein the topical formulation comprises carrier particles with an average diameter of up to 10 microns.

9. The method of claim 1, wherein the topical formulation comprises bacteriophages lytic for P. acnes covalently attached to the carrier particles with an average diameter of up to 20 microns, wherein the topical formulation is in the form of a cream or a gel.

10. The method of claim 9, wherein the topical formulation comprises bacteriophages lytic for at least 3 different strains of P. acnes, wherein the bacteriophages are covalently attached to the carrier particles.

11. The method of claim 1, wherein the topical formulation comprises bacteriophages lytic for S. aureus covalently attached to carrier particles with an average diameter of up to 20 microns, wherein the formulation is in the form of a cream or a gel.

12. The method of claim 11, wherein the topical formulation comprises bacteriophages lytic for at least 3 different strains of S. aureus covalently attached to the carrier particles.

Description

(1) FIG. 1 shows the survival of phage in phage-containing formulations of the invention in storage;

(2) FIG. 2 shows the effect of high dose microspheres treated with purified bacteriophage on weight gain;

(3) FIG. 3 shows the temperature time course of two animals following a subcutaneous (s.c.) dose of 100 μl 1×10.sup.8 cfu/ml E15 and an intravenous (i.v.) tail injection of 5 μm bacteriophage-treated microspheres;

(4) FIG. 4 shows 24 hour bacterial counts following a dose of 100 μl 1×10.sup.8 cfu/ml E15 suspended in HM;

(5) FIG. 5 shows bacterial and bacteriophage time course in one animal (9676) following a s.c. dose of 100 μl 1×10.sup.8 cfu/ml E15 and an i.v. tail injection of 5 μm bacteriophage-treated microspheres;

(6) FIG. 6 shows the time course of bacteria in total blood. Test animals (open square, shaded triangle) were inoculated with 50 μl of 5×10.sup.7 cfu/ml E15 suspended in 5% hog gastric mucin intraperitoneally (i.p.) with a secondary injection of bacteriophage-treated microspheres intravenously The control animal (shaded diamond) was inoculated with 50 μl of 5×10.sup.7 cfu/ml E15 suspended in 5% hog gastric mucin i.p only;

(7) FIG. 7 shows the time course following cfu/total blood in test animals inoculated with 50 μl of 5×10.sup.7 cfu/ml E15 suspended in 5% hog gastric mucin i.p with a secondary injection of bacteriophage-treated microspheres i.v.;

(8) FIG. 8 shows the effect of ampicillin on S. aureus 8588;

(9) FIG. 9 shows the effect of Phage K on S. aureus 8588;

(10) FIG. 10 shows the effect of inhibitory ampicillin concentration on phage K killing of S. aureus 8588;

(11) FIG. 11 shows the results of testing an in vivo response to administration of phage on particles;

(12) FIG. 12 shows the effects of normal and heat-inactivated serum on free bacteriophage K;

(13) FIG. 13 shows the effects of normal and heat-inactivated serum on immobilized bacteriophage K;

(14) FIG. 14 shows sutures removed from rat after two weeks;

(15) FIG. 15 shows a photomicrograph of rat liver 14 days after administration of microspheres of the invention; and

(16) FIG. 16 shows the results of a trial of phage-containing cream on human skin.

EXAMPLES

Example 1—Bacteriophage Formulations for Acne Treatment

(17) Propionibacterium acnes bacteriophages (FP pal) were immobilised onto nylon beads (average diameter 10 microns) and mixed into Formulations A, B and C as set out below. Each Formulation was then tested for survival of bacteriophage at room temperature.

(18) Formulation A—Aqueous Cream

(19) Anhydrous Lanolin 1.0% w/w White Soft Paraffin BP 14.5% w/w Light Liquid Paraffin PhEur 12.6% w/w Water [to 100%]
Formulation B—Face Wash (Commercially Available Under the Trade Mark “Clearasil”) Product contents: Aqua, Sodium Gluconate, Propylene Glycol, Octyldodecanol, Steareth-2, Cyclopentasiloxane, Steareth-21, Salicylic Acid, Cetyl Alcohol, Behenyl Alcohol, Cyclohexasiloxane, Polyacrylamide, C13-14 Isoparaffin, Xanthan Gum, Phenoxyethanol, Magnesium Aluminum Silicate, Laureth-7, Menthol, Methylparaben, Butylparaben, Ethylparaben, Isobutylparaben, Propylparaben, CI 77891.
Formulation C—Gel (Commercially Available Under the Trade Mark “Dr Spot”) Product contents: 2% Salicylic acid, Witch Hazel, Lactic acid.
Resistance to Salicylic Acid

(20) Formulation A was supplemented with salicylic acid at 0.5, 1.0, 1.5 and 2% w/w, labelled as Formulations A1, A2, A3 and A4. Infectivity of phage in these formulations was compared with infectivity of phage in Formulation A (containing no salicylic acid). After 2 weeks in storage no loss in infectivity was observed in any of A1-A4 compared to the control Formulation A, indicating no adverse effect of the salicylic acid even at 2% on phage survival.

(21) Survival of Phage-Containing Formulations in Storage

(22) Each of formulations A, B and C were supplemented with Propionibacterium acnes bacteriophages covalently immobilised onto nylon 12 particles and the infectivity of the phage compared with the same titre of free (i.e. non-immobilised) phage in the same base formulations. The results over a 6 week period are shown in FIG. 1:

(23) Hence, phage survival was significantly enhanced by immobilisation onto the nylon particles in each of Formulations A, B and C, in all cases by at least one order of magnitude and in 2 out of 3 by several orders of magnitude.

Example 2—Formulation for Topical Use (S. aureus)

(24) Base Cream Preparation:

(25) 120 g of Emulsifying Ointment BP was heated to 60 degrees C. and mixed with 270 ml water also heated to the same temperature. The mixture was carefully stirred as it cooled, producing a smooth, white cream formulation. The cream was cooled to room temperature and divided into 5 equal portions.

(26) Bacteriophage-Particle Production:

(27) Nylon 12 particles of average diameter 3 microns were treated by corona discharge (75 kV field) and rapidly added to a bacteriophage suspension at 1×10.sup.9 pfu/ml. Particles were washed 3 times to remove non-bound bacteriophages. Using this method, 5 separate 2 ml preparations were made utilizing one of each of 5 different strains of bacteriophage specific for S. aureus from stored phage stock.

(28) Formulation:

(29) To each of the 5 separate portions of cream base was added a separate bacteriophage-particle preparation by admixture and agitation until the suspension had been fully incorporated into the base. The 5 separate bases were then combined and thoroughly mixed to form a single cream base containing 5 different immobilised phage types.

Example 3—Formulation for Topical Use (P. acnes)

(30) A formulation was prepared as follows. Oil phase: Stearic acid 4%, stearyl alcohol 5%, lanolin 7%, isopropyl myristate 8%. Aqueous phase: Methyl cellulose 1%, in purified water

(31) The two phases were prepared separately by weight and heated to 70° C. The water phase was then mixed with the oil phase by trituration till the cream congealed and cooled.

(32) Nylon 6,6 particles of average diameter 10 microns were treated by passing through a corona discharge at 70 kv and rapidly added to a mixed bacteriophage preparation containing 5 different bacteriophages against P acnes at a final concentration of 1×10.sup.9 pfu/ml. The particles were washed 3 times to remove non-bound bacteriophages and added to the cream base to give a final bacteriophage concentration of 1×10.sup.5 pfu/ml.

Example 4—Treatment of Systemic Infection

(33) Five micron diameter microspheres were prepared by either chemical or corona discharge methods with the addition of the purified bound bacteriophage K. An intravenous (i.v.) injection of 5 μm microspheres suspended in PBS was given to a group of rats at a dose of 20 mg/kg body weight at day 7. There was no change in the weight gain profile, which indicated no significant adverse effect of the microspheres or the immobilised bacteriophages on the animals in the short term—as shown in FIG. 2.

(34) Prior to infection, a temperature transponder was implanted into 2 rats. Body temperature was recorded at intervals of 30 minutes pre-, during and post-infection phase. Animals were handled and weighed over a period of 7 days during which they became accustomed to the restraint process by which blood sampling was be carried out. The experiment used a dose of 100 μl 1×10.sup.8 cfu/ml EMRSA 15 suspended in 5% hogs gastric mucin using the subcutaneous (s.c.) route as the choice of administration in anaesthetized animals. Injections were given at 09:30 hours, the first sample time point. An i.v. tail injection of 5 μm bacteriophage-treated microspheres was carried out after removal of the tail tip. Numbers of bacteriophages on microspheres were calculated from plaque assays giving a value of 1.1×10.sup.3 pfu/ml, 100 ul inoculum contained 10.sup.2 pfu/ml.

(35) Recovery was supervised continuously with blood sampling hourly for 27 hours, then at 6 hourly intervals for the following 3 days. Animals were sacrificed at 14 days post infection during which a final blood sample was taken by cardiac puncture, organs removed, accompanied by blood swabs onto plates and broths to determine the extent of the infection. As microspheres were introduced, organs and blood would also be analysed for the presence of bacteriophage.

(36) Results

(37) Animals recovered fully following a s.c. dose of 100 μl 1×10.sup.8 cfu/ml E15 and an i.v. tail injection of 5 μm bacteriophage-treated microspheres. The animals showed no clinical signs of distress, and continued in a good general state of health.

(38) There was a steep rise in temperature which followed shortly after inoculations indicating a fever response in the animal associated with the infection—see FIG. 3. Temperature was maintained at approximately 38.5° C. for 24 hours post inoculation. The bacteriophage bound microspheres at this stage have no direct effect on the temperature when a comparison is made with s.c. injections of bacteria alone.

(39) Bacterial counts were performed on blood samples—see FIG. 4.

(40) A similar pattern was seen in both animals as the graph indicates that over time, bacterial numbers increased signifying the colonization and influx of bacteria in the animal's body. A correlation was seen between the temperature and the bacterial numbers; a temperature increase coincided with an increase in bacteria in the blood.

(41) Additional universal culture tubes were also filled with L-broth to grow up any bacteria present in the blood. Results reinforced the bacterial counts giving a positive result for each time point. No bacteria were found in the organs following sacrifice 1 week post inoculation indicating the infection had ceased to be systemic.

(42) Blood sampling results demonstrated a relationship between rising bacterial numbers with a raised temperature. However the longevity of this elevated temperature was shown to be affected by the blood sampling procedure itself. No bacteria were found in the blood at sacrifice, 1 week after sampling, which implied that the initial systemic infection had been eliminated.

(43) Bacterial counts in organs also gave negative results, which confirms the elimination of the model systemic infection—see FIG. 5; this was in contrast to models where no bacteriophages were introduced when substantial amounts of bacteria were found on sacrifice.

(44) Analysis of the plaques produced on the agar overlay using an inverted microscope did not show the presence of any nylon beads. It was therefore concluded that the bacteriophage isolated were free and not the initial immobilised dose.

(45) We compared bacterial count between treated and control animals—see FIG. 6.

(46) We measured bacterial numbers over 11 days—see FIG. 7. Test animals (C4-C10) showed an initial rise in bacterial numbers in the first two days; however, these figures rapidly declined and were depleted in the subsequent days. This trend was similar in all animals, unlike the continuous rate of bacterial growth seen in the control animals.

(47) We measured bacteriophage numbers in various organs—see Table 1. The results showed that bacterial numbers were significantly lower in all organs than those of bacteriophage. High numbers of free bacteriophage must result from bacterial infection within the animal's body. Their presence in all organs demonstrated the ability of the microspheres or free bacteriophage to travel throughout the body.

(48) TABLE-US-00001 TABLE 1 cfu/organ pfu/organ animal no. Organ 9676 9677 9676 9677 Heart 23 0 263 90 Lung 0 0 56 127 Spleen 47 22 141 60 Liver 859 0 7 23 Kidney 18 0 6 52 (cfu = colony forming units of bacteria; pfu = plaque forming units of bacteriophages)

(49) In the experiments of this example of the invention a relatively low dose of immobilised bacteriophages were used (100) compared with “free” bacteriophages used in experiments reported in the literature (where 10.sup.9 bacteriophages are often used). The model infection was induced by around 10.sup.6 bacteria so that bacteria multiplied before bacteriophage numbers rose to a level at which the bacteria were largely eliminated. This is why bacterial numbers continued to increase until 12 to 24 hours after inoculation. The kinetics are complex; bacteria multiply logarithmically but are phagocytosed by the immune system, as well as being infected and destroyed by the bacteriophages. “Free” bacteriophages will also be removed by the immune system (as per the prior art) and immobilised bacteriophages on microbeads as per the invention are now shown to be removed from the circulatory system into the spleen and liver where they remain active (see other data and results herein).

(50) The results showed that: Nylon microspheres at 5 μm diameter did not cause any obvious adverse effects to the animal Immobilised bacteriophages in this dose did not cause any measurable adverse effects. Immobilised bacteriophages infected bacteria in viva Low doses of immobilised bacteriophages eliminated infecting bacteria in the model. Free bacteriophages resulted from initial infections with immobilised bacteriophages and the immobilised bacteriophages that do not cause an initial infection remained active. Both corona treated and chemically treated microbeads were equally active.

Example 5—Formulation Comprising Bacteriophages and Antibiotics

(51) We tested (1) ampicillin, (2) nylon 12 particles onto which bacteriophage K was covalently bound and (3) a formulation in which both were co-administered for effectiveness against S. aureus 8588.

(52) FIG. 8 shows that bacteriophage K was unaffected by the presence of ampicillin at close to MIC concentrations.

(53) The data showed that ampicillin was able to kill the 8588 strain of S. aureus. The MIC of ampicillin against this strain of bacteria was found to be 0.02 mg/ml (20 μg/ml) which showed this to be a resistant strain (“MRSA”) (MIC breakpoint 0.125 μg/ml). The absorbance readings of treatments above the MIC were similar to the negative control of Mueller-Hinton broth alone indicating no bacterial growth.

(54) FIG. 9 shows that phage K killed the bacteria 8588 at every concentration. The absorbance readings of the phage K treated wells were similar to the negative control Mueller-Hinton broth alone.

(55) The data in FIG. 10 showed no interaction between ampicillin at 0.01 mg/ml (below MIC) and bacteriophage lysis of S. aureus 8588. The ampicillin value (amp) is from an experiment analogous to that shown in FIG. 8. At 0.01 mg/ml ampicillin (below MIC) the ampicillin has some effect on bacterial growth but is not totally effective as resistant bacteria survive. Addition of phage results in total kill, showing that the phage kills resistant bacteria that would have survived the sub-MIC dose of antibiotic. This showed that resistant bacterial strains that emerge (and would survive if treated only with antibiotic) can be killed by the combination therapy. Hence, co-administration of antibiotic and immobilised phage was effective in killing bacteria in this model.

Example 6—In Vivo Response to Administration of Phage on Particles

(56) P388.D1 cells (mouse lymphoid macrophage cells) were stimulated with (i) 5 micron nylon beads, (ii) 100 microlitres of free phage, (iii) 5 micron beads with phage covalently attached and (iv) a control. Samples were periodically taken and IL-1alpha levels measured using ELISA—see FIG. 11.

(57) A small increase in IL-1alpha was seen after 3 hours stimulation but no effect after 24 hr stimulation. These results indicated the covalently immobilised phage did not induce an immune response in this model.

Example 7—Survival of Immobilised Bacteriophage in Serum

(58) Previous studies (Donlan R. M. (2006) Controlling clinically relevant biofilms using bacteriophage Biofilm Perspectives No. 2006:02. www.BiofilmsOnline.com and Sokoloff A., Bock I., Zhang G., Sebestyen M., and Wolff J. (2000) The interaction of peptides with the innate immune system studied with the use of T7 phage peptide display. Molecular Therapy 2, 131-139) have shown phages to be inactivated after 3 minutes following contact with serum incubated at 37° C.

(59) An experiment was designed to elucidate whether bacteriophage covalently attached to sutures showed the same effect.

(60) Free phage and phage on sutures were combined with serum (“normal” and heat inactivated), incubated at 37° C. and tested at regular intervals (see results below) for retention of PFU activity.

(61) Results

(62) Referring to FIGS. 12 and 13, it was found that after 10 minutes in the presence of serum the phage titre of free phage was thereafter rapidly reduced. With heat treated serum (in which complement has been inactivated) no loss occurred. Separately, the immobilized phage was resistant to (normal) serum—the phage titre remained stable throughout the experiment. This indicated that the anti-bacteriophage activity of untreated serum is immunologically based.

Example 8

(63) Analysis of Production of IgM and IgG Antibodies Against Immobilised Phage

(64) We investigated the antibody response in rats to administration of phage covalently immobilised onto 5 micron nylon particles. Blood samples were taken by cardiac puncture and serum from these samples was subjected to an ELISA test to determine whether any antibody response could be detected.

(65) Measurement of IgG & IgM by ELISA

(66) Enzyme-linked immunosorbent assay (ELISA) was used to determine whether blood serum contained levels of antibodies to bacteriophage 9563 in experimental rats.

(67) The experiment was performed by the following method: blood was collected from rats at sacrifice and centrifuged at 13,000 rpm for 10 minutes and the resulting serum collected and stored at −20° C. Fresh double-strength coating buffer (100 μl) was added into each well of a 96 well microtitre plate (Greiner Bio-One, Germany) with purified bacteriophage (100 μl) in order to immobilize them to the microplate. PBS (100 μl) was added to control wells and the plate left overnight at 4° C. or for 2 hr at 37° C. The remaining liquid was tipped out and plate was washed thoroughly 3 times with PBS-Tween (Fisher Scientific, Leicestershire, UK). Plates were tapped repeatedly onto paper towels until no liquid remained in the wells. PBS-BSA (10 mg/ml) (200 μl) was added to each well for 0.5 hr at 37° C. to block non-specific binding sites. The remaining liquid was tipped out and the plate was washed thoroughly 3 times with PBS-Tween before 100 μl of serum was added and incubated for 2 hr at room temperature in a sealed box containing wet tissue paper to create a humid atmosphere. Following incubation the remaining liquid was tipped out and the plate washed 3 times with PBS-Tween.

(68) Following washing, 100 μl of a 1 in 1000 dilution of a second antibody of either HRP-mouse anti-Rat IgM, or IgG (Invitrogen, Paisley, UK), was added and the plate incubated at room temperature for 1 hr, or at 37° C. for 30-40 min. The remaining liquid was tipped out and plate was washed 3 times with PBS-Tween. Tetramethyl benzidine (TMB) (150 μl) (Sigma, Aldrich, UK) in acetate citrate buffer (5.2.1.) was added to each well and the plate was incubated at room temperature for 30-40 minutes in the dark, until the reaction mixture turned blue. The reaction was stopped by adding 4M H.sub.2SO.sub.4 (50 μl) causing it to turn yellow in colour. Absorbance was read at 450 nm using a plate reader (Labsystems iEMS Reader MF, Finland). Results were compared to control blood samples in which rats were not challenged with phage.

(69) Results

(70) The tables below show the results separately for IgM and IgG. In both cases, there was no statistically significant difference in antibody production—hence the immobilised phage stimulated no production of anti-phage antibodies in the rats tested.

(71) TABLE-US-00002 TABLE 2 Statistical analysis of IgM + E15 alone vs IgM E15 + beads & phage IgM Group IgM + E15 alone IgM E15 + beads & phage Mean 0.37278200 0.26216200 SD 0.23527212 0.24058500 SEM 0.06074700 0.04811700 N 15 25

(72) TABLE-US-00003 TABLE 3 Statistical analysis of IgG + E15 alone vs IgG E15 + beads & phage IgG Group IgG + E15 alone IgG E15 + beads & phage Mean 0.37535700 0.31158300 SD 0.18974907 0.16150500 SEM 0.04899300 0.03230100 N 15 25

(73) The results showed that the immobilised bacteriophages were poorly immunogenic: there was no immune IgG or IgM response. Repeated subsequent experiments by the inventors, using phage covalently bound to nylon beads, have all consistently shown no detectable antibody response to immobilised phage in rat models.

Example 9—Phage Activity after In Vivo Exposure

(74) Sutures to which phage K were covalently immobilised were prepared and used as per Example 6 in WO2012/175749. At day 14, sutures were removed, washed and their activity tested.

(75) Results

(76) As shown in FIG. 14, the sutures removed after 2 weeks retained phage activity as measured by conventional plaque formation assay. The immobilised phage had not been inactivated by antibodies.

Example 10—In Vivo Retention of Immobilised Phage

(77) Liver samples from rats treated as per example 4 were analyzed 14 days after administration of 5 micron microspheres.

(78) Results

(79) As seen in FIG. 15, microspheres (subsequently confirmed to carry active phage) were detectable (see arrows) in the liver 14 days after systemic administration.

Example 11—Treatment of P. acnes on Human Skin

(80) In order to assess the effectiveness of bacteriophage immobilized onto particles on treatment of infection on human skin a short study was undertaken to determine if cream containing bacteriophages would reduce the bacterial load on skin.

(81) Human skin of a volunteer was swabbed with alcohol and inoculated with 1×10.sup.4 cfu/mL of P. acnes (ATCC 6919). This was allowed to air dry and then treated with: 1. E45 cream, 2. E45 cream with 1×10.sup.5 pfu/g bacteriophage, or 3. E45 cream with 1×10.sup.5 pfu/g bacteriophages immobilised on approx 10 micron diameter, nylon 12 beads.

(82) At different time points swabs were taken to monitor the quantity of bacteria on the skin.

(83) The results are shown in FIG. 16. As seen, in this study immobilised bacteriophages in a cream base were more effective than free phages and also more effective than the other control (cream alone).

Example 12—Pig Skin as a Model for Skin Infections

(84) We developed the protocol below to use pig skin as a model for human skin infections and treatment thereof according to the invention.

(85) Immobilisation of Bacteriophages

(86) Previously isolated bacteriophages are immobilised onto nylon beads that can be incorporated into acne treatment products; use 10 micron nylon beads, cosmetic grade—as previously described in earlier published work by the applicant e.g. in WO 2007/072049.

(87) System Protocol

(88) The nylon beads act as a model system for immobilisation of bacteriophage to nanoparticles that are to be incorporated into acne treatments and creams.

(89) Each treated material is tested for inhibition of bacterial growth. The effects are compared to material exposed to non-immobilised bacteriophage and material alone. The experiments incorporate multiple tests and continual testing until the inhibitory effect is no longer observed.

(90) Upon completion, the efficacy of immobilised bacteriophage for controlling bacterial growth and the shelf life of immobilised bacteriophage are determined on each material.

(91) Treatment of Skin Infection Using Immobilised Bacteriophages

(92) Pig skin is prepared and inoculated as follows.

(93) Fresh pig skin is handled aseptically. The skin is swabbed with 70% alcohol to remove contaminating bacteria.

(94) Once dry the skin is contaminated with the intended bacterial strain to a concentration of 1×10.sup.4 cfu/cm.sup.2. Using a sterile swab the bacterial solution is smeared into the skin and allowed to air dry in the laminar flow hood for 15 minutes. The desired bacteriophage treatment is applied—using the beads onto which bacteriophage have been immobilised as described above, and the skin then incubated at 37° C. for 16 hours.

(95) The pig skin is assayed for contaminating bacteria on selective media. Bacterial counts are made following inoculation in order to determine if the treatment was successful.

Example 13—Acne Treatment Efficacy on Model Pig Skin

(96) The pig skin model described above was used to demonstrate the efficacy of administration of immobilised bacteriophage for treating acne.

(97) The host bacterium used to inoculate the pig skin was Staphylococcus aureus 8588 and the bacteriophage used to infect this bacterium was Phage K.

(98) Phage K were immobilised onto 10 micron diameter nylon beads; 1×10.sup.7 pfu/mL bacteriophage were immobilised onto 1 g of 10 micron beads.

(99) Pig skin was inoculated with S. aureus as described above. The pig skin (4 cm×4 cm) was treated with E45 base cream (1 mL) and approximately 1×10.sup.4 pfu beads, also as described above.

(100) The results of this test are set out in Table 4 and show that the number of bacteria on the skin test samples was successfully reduced by application of the cream containing immobilised bacteriophages.

(101) TABLE-US-00004 TABLE 4 Number of Bacteria (cfu) Test No. Control Free Immobilised 1 120000 150000 60000 2 280000 170000 30000 3 220000 180000 20000 Mean 206667 166667 36667

(102) The invention thus provides compositions and methods for treatment of topical and systemic bacterial infections.