Guided antimicrobial peptides for targeting h. pylori

Abstract

The present disclosure pertains to a treatment strategy to combat select bacteria in the gut, such as H. pylori. The strategy uses a probiotic-based system to express and deliver a guided antimicrobial peptide to the gut. The guided antimicrobial peptide is expressed from a hybrid gene that codes for an antimicrobial peptide fused to a guide peptide, the latter binding to a protein of the target bacterium. This technology can eliminate the target bacterium selectively and specifically from the gut microbiota. The specificity of the targeting, being at the strain, species or genus level, depends on the sequence of the guide peptide used to provide the targeting. The treatment can be administered orally, such as by using an ingestible probiotic.

Description

Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION GUIDED ANTIMICROBIAL PEPTIDES FOR TARGETING H. PYLORI BACKGROUND

[0001] This application claims priority to U.S. Provisional Patent Application Serial No.63 / 652,842, entitled “Guided Antimicrobial Peptides for Targeting H. Pylori,” filed on May 29, 2024, the entire contents of which are hereby incorporated by reference.

[0002] The present disclosure relates to a means of eliminating a specific gut bacterial species, such as Helicobacter pylori, without altering the microbiome.

[0003] The microbiota of the gut affects human health in many ways. The gut microbiome contains 100+ trillion bacteria and is largely involved in mediating the host’s immune response while also performing other essential functions including the extraction of nutrients and energy from food. The bacterial makeup of the gut predisposes humans to health issues ranging from obesity to cancer to psychological disorders. Disruption to the microbiome (dysbiosis) results in an imbalance in the types and number of bacteria that comprise a person’s normal, protective microflora. There are a number of factors that lead to dysbiosis including ingestion of pathogenic bacteria and antibiotic-mediated or immunosuppressive mediated depletion of the microbiome. Dysbiosis has been linked to numerous human diseases including both intestinal as well as extra-intestinal disorders. The literature indicates dysbiosis in the pathogenesis of IBS, inflammatory bowel disease, and colorectal cancer as well as allergies, cardiovascular disease, and mental illness. Additionally, gut microbiota have been implicated as precursor for autoimmune diseases given that severity and / or incidence of disease has been shown to be reduced in germ-free animal models.

[0004] In other cases, changes to gut bacteria result from ingestion of a dangerous pathogen that can produce an intestinal disease. There are few, if any, reported means to effectively knock out a specific bacterial species which is causing problems in the gut, either as an active pathogen or as a player in the microbiome that predisposes humans to various disorders.

[0005] Helicobacter pylori is a gut bacterium that is the primary cause of peptic ulcers and gastric cancer. Gastric cancer causes the third most fatalities worldwide among cancers and is especially common in the Far East (Bahkti et al., 2020). Only 1 in 5 patients 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION survive gastric cancer 5 years after diagnosis. H. pylori is recognized by the International Agency for Research on Cancer as a Group 1 carcinogen. It is estimated that 4.4 billion people are infected with H. pylori, with developing countries having the highest infection rates (70% prevalence in Africa) (Hooi et al., 2017). In the United States, H. pylori occurs twice as frequently in the non-white population as in the white population (Everhart et al., 2000) and is associated with lower socio-economic status worldwide.

[0006] No commercial vaccine exists against H. pylori. Though some progress has been seen in lowered H. pylori prevalence in some countries using antibiotic treatment, large increases in antibiotic resistance rates are now being seen in H. pylori isolates. The prevalence of clarithromycin-resistance in H. pylori rose from 11% to 60% in just 4 years (2005-2009) in Korea, with similar increases recorded in China and Japan (Thung et al., 2016). Though the standard treatment is in fact a triple antibiotic therapy, antibiotic resistance rates continue to rise. Thus, it is difficult to see a path forward with H. pylori treatment via antibiotics. Other bacteria offer similar challenges. SUMMARY

[0007] The present disclosure pertains to a treatment strategy to combat select bacteria in the gut, such as H. pylori. The strategy uses a probiotic-based system for the expression and delivery of guided antimicrobial peptides to the gut. The guided antimicrobial peptides are expressed from a hybrid gene in the probiotic bacterium’s DNA, and can be the sequence coding for an antimicrobial peptide fused to the sequence coding for a guide peptide, with the latter peptide responsible for binding to a protein of the target bacterium. The fusing can occur with or without a linker sequence, that is, independent of the presence of a linker sequence. This technology can eliminate the target bacterium selectively and specifically from the gut microbiota. The specificity of the targeting, being at the strain, species or genus level, depends on the guide protein used to provide the targeting. The treatment can be administered orally, such as by using an ingestible probiotic.

[0008] Preferred embodiments described herein relate to a method for the control of a target bacterium such as H. pylori which does not involve antibiotics. For delivery of the active protein, this method uses engineered probiotic bacteria. Preferred embodiments utilize lactic acid bacteria, including Lactococcus and Lactobacillus species, such as 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION Lactococcus lactis and Lactobacillus acidophilus, which are food grade bacterium that are safe for human consumption or have been granted GRAS status (Generally Regarded As Safe) by the FDA and are in widespread commercial use for processing dairy food products. Probiotics constitute a well-established technology which is inexpensive, highly scalable, and very successful commercially. These commercial traits make this technology especially amenable to large-scale application, particularly in developing countries.

[0009] The probiotic bacterium can be formulated as a recognizable food product that is commonly found in the probiotics market, such as dried yoghurt pellets, which can be stored without refrigeration for months. In this format, the product may be taken by travelers to foreign countries or by long-term expatriates or soldiers with food, perhaps twice per week, as a preventative (“prophylactic”) to disease. The treatment could also serve as a therapy, being eaten after the patient is sick.

[0010] The present technology is important and advantageous because it utilizes guided antimicrobial peptides that eliminate only the target bacterium while leaving all the other members of the microbial community undisturbed. The use of probiotic bacteria that are ingested and remain active in the digestive system in order to secrete the guided recombinant antimicrobial peptide directly in the gut of the patient is also significantly different from previous technologies. BRIEF DESCRIPTION OF DRAWINGS

[0011] FIG. 1 shows the pE-SUMOstar vector carrying AMP for expression in E. coli BL21 cells. SUMO protease site is between SUMO and A12C-AMP.

[0012] FIG. 2 shows expression of SUMO / AMP in E. coli and cleavage of AMP free of SUMO fusion partner.

[0013] FIG. 3 shows log values for minimum inhibitory concentrations (MIC) in ^M for non-targeted and targeted analogues of eurocin and plectasin against Bacillus subtilis, Enterococcus faecalis, Staphylococcus aureus and Staphylococcus epidermidis.

[0014] FIG.4 shows the cell-kinetic profile for B. subtilis, S. epidermidis, S. aureus and E. faecalis (clockwise), created by plotting log CFU / ml of the bacteria grown in the presence of each peptide.

[0015] FIG.5 shows biofilm inhibition activity evaluated by plotting the absorbance 3 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION of crystal violet (540 nm) against the concentration of 4 AMPs on the 4 bacteria - B. subtilis, S. epidermidis, S. aureus and E. faecalis.

[0016] FIG. 6A shows all 18 non-redundant sequences selected by the four rounds of biopanning, analyzed to show consensus between them.

[0017] FIG. 6B shows the phage ELISA results quantifying the binding of phage clones of FIG.6A to whole H. pylori cells expressing VacA (WT) or engineered to prevent VacA expression (Del).

[0018] FIG. 7 shows flow cytometry of GFP protein with different peptide guides (gGFP) binding to H. pylori cells engineered to prevent VacA virulence factor expression (Deletion) or expressing VacA (WT).

[0019] FIG. 8 shows cytotoxicity of various gAMPs against human gastric AGS cells.

[0020] FIG.9 shows a quantitative summary for the top gAMP, P5-pexiganan, with all parameters from the previous figures included for direct comparison in terms of the concentration of gAMP required to be toxic to mammalian cells (Hemolysis, AGS gastric cells), off-target bacteria (P. aeruginosa, A. baumanii, A. faecalis), and the very low concentration actually needed to kill the target bacterium, H. pylori.

[0021] FIG. 10 shows a vector for transformation of Lactococcus lactis in accordance with preferred embodiments described herein.

[0022] FIG. 11 shows the viability of E. coli in the presence of different antibiotic dilutions and supernatants of broth cultures of Lactococcus lactis secreting antimicrobial peptide with or without a guide peptide.

[0023] FIG. 12 shows an exemplary vector for Lactococcus lactis secretion of AMPs and gAMPs.

[0024] FIG. 13 shows results of qPCR amplifying the vacA gene, showing elimination of H. pylori expressing VacA by co-culturing in vitro with L. lactis expressing gAMPs targeting VacA or unguided AMPs.

[0025] FIG. 14 shows therapeutic clearance of H. pylori from mouse gastric lumen using probiotics expressing gAMPs or AMPs. 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0026] FIG. 15 shows prophylactic protection from H. pylori infection of mouse gastric lumen using probiotics expressing gAMPs or AMPs. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0027] The present disclosure relates to a means for targeting and eliminating a target bacterium using a probiotic that expresses and secretes a protein that kills the disruptive bacterium without harming other bacteria.

[0028] In preferred embodiments, the present technology pertains to a probiotic bacterium that has been transformed to include a DNA construct for a guided antimicrobial peptide. In preferred embodiments, the probiotic bacterium is a bacterium that is safe for human consumption, such as Lactococcus lactis. The sequence coding for the guided antimicrobial peptide includes the sequence coding for a targeting (guide) peptide fused to the sequence coding for an antimicrobial peptide and expressed by the probiotic bacterium as a hybrid protein. The guide peptide is specific for the target bacterium and limits the action of the antimicrobial peptide to that particular bacterium.

[0029] Accordingly, preferred embodiments described herein relate to a probiotic for the prevention or treatment of a condition caused by a target bacterium living in the gastrointestinal tract of a subject, comprising a probiotic bacterium. The probiotic bacterium is preferably a lactic acid bacterium, such as a Lactococcus bacterium, and preferably Lactococcus lactis. The probiotic bacterium has been transformed to comprise a DNA construct expressing a guided antimicrobial peptide, wherein the sequence coding for the guided antimicrobial peptide comprises the sequence coding for an antimicrobial peptide fused to the sequence coding for a guide peptide that binds to a protein of the target bacterium. The guided antimicrobial peptide kills the target bacterium in the gastrointestinal tract of the subject. The guided antimicrobial peptide also minimally disrupts other bacteria found in the gastrointestinal tract of the subject when compared to unguided antimicrobial peptides, antibiotics, or other broad spectrum treatments.

[0030] As used herein, “minimally disrupts” means the guided antimicrobial peptide does not cause a disruption that would cause a health effect, as opposed to a technical change in bacterial abundance. “Minimally disrupts” also means the guided antimicrobial peptide does not significantly disrupt other non-target bacteria, where the disruption would cause a 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION health effect.

[0031] Preferred embodiments relate to a probiotic system which delivers antimicrobial peptides (AMPs) to the gut. Antimicrobial peptides are natural products produced by plants, animals and fungi to protect against bacterial infection (Ngyuen et al., 2011). However, an AMP by itself has broad spectrum activity, similar to an antibiotic. The broad activity of antibiotics has been well-documented to lead to microbiota dysbiosis. Many publications have demonstrated connections between antibiotic-induced dysbiosis and rheumatoid arthritis, inflammatory bowel disease, diabetes, obesity and other disorders (for a review, see Keeney et al., 2014). This is one of the consequences of the overuse of antibiotics and nonselective AMPs share the same weakness. Exemplary AMPs used in preferred embodiments described herein include plectasin, eurocin, pexiganan, laterosporulin, alyteserin, and cathelin-related anti-microbial peptide (CRAMP).

[0032] To solve this problem of dysbiosis, the preferred embodiments described herein include a guide peptide fused to an AMP, produced from a corresponding guide- AMP hybrid gene of the probiotic bacterium. This enables the resulting guided AMP (gAMP) to bind specifically to the targeted bacterium such as H. pylori, leaving the commensal bacteria of the gut largely undisturbed. In this way, a probiotic expressing gAMP will multiply in the stomach and selectively kill the target pathogen, H. pylori without the health issues associated with antibiotics and other broad-spectrum treatments. Other targeted bacterium can be treated similarly, and H. pylori is used herein as an example.

[0033] In preferred embodiments, the guide peptide described herein binds to the VacA protein that is at least sometimes displayed on the surface of H. pylori. In this way, these gAMPs will be localized to the surface of the H. pylori via binding to VacA and the AMP portion can then act to destabilize the bacterial membrane and specifically kill the H. pylori cell.

[0034] In additional preferred embodiments, the guide peptides used in the gAMPs. have one of the following sequences: WNSWPFPMASFE (SEQ ID NO:1) TQQGINGVLVSV (SEQ ID NO:2) FHEGTYFKVWAG (SEQ ID NO:3) 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0035] Exemplary DNA sequences that can be used to express the guide peptides of SEQ ID NOs: 1-3 are shown below: TGGAATTCTTGGCCGTTTCCTATGGCTTCGTTTGAG (SEQ ID NO:4) ACGCAGCAGGGTATTAATGGGGTGCTTGTTTCTGTC (SEQ ID NO:5) TTTCATGAGGGTACTTATTTTAAGGTTTGGGCGGGG (SEQ ID NO:6)

[0036] In additional preferred embodiments, the gAMPs also include a linker peptide between the guide peptide and the antimicrobial peptide. Examples of the linker include linkers having the peptide sequence VGCGSGG (SEQ ID NO:11) or the peptide sequence VGCGSGGHGSGGGS (SEQ ID NO:12).

[0037] The probiotic gAMPs described in preferred embodiments are distinct from similar technologies. They possess a selectivity not found in antibiotics and unguided AMPs. The use of probiotics makes it possible to produce probiotic gAMPs much more cheaply than gAMP proteins purified from a heterologous expression system or synthesized chemically. This combination of selectivity and low-cost scalability is essential for any replacement for cheap and abundant antibiotics to be successful commercially and therefore reach the intended patients.

[0038] Preferred embodiments disclosed herein relate to an edible Lactococcus lactis probiotic bacterium, wherein the probiotic bacterium has been transformed to comprise a DNA construct expressing a guided antimicrobial peptide, wherein the sequence coding for the guided antimicrobial peptide comprises the sequence coding for an antimicrobial peptide fused to the sequence coding for a guide peptide that binds to the VacA peptide of H. pylori, produced from the corresponding hybrid gene of the L. lactis bacterium, wherein the antimicrobial peptide is plectasin, eurocin, pexiganan, laterosporulin, alyteserin, or cathelin-related anti-microbial peptide, and wherein the guided antimicrobial peptide kills H. pylori in the gastrointestinal tract of the patient without causing a significant disruptive effect on other bacterial species. In other words, the probiotic bacterium expressing the guided antimicrobial peptide will not disrupt the taxonomic balance of the stomach microbiota and will not cause long-term damage.

[0039] Additional preferred embodiments relate to a method for treating a disease or condition associated with H. pylori by administering an edible probiotic to a subject, 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION where the edible probiotic is ingested and remains active in the subject’s gut long enough to secrete a guided antimicrobial peptide that kills H. pylori.

[0040] In another aspect of the present invention there is provided a probiotic composition including a therapeutically effective amount of a transformed probiotic L. lactis bacterium expressing a guided antimicrobial peptide and an acceptable excipient, adjuvant, carrier, buffer or stabiliser. A “therapeutically effective amount” is to be understood as an amount of an exemplary probiotic that is sufficient to show inhibitory effects on H. pylori. The actual amount, rate and time-course of administration will depend on the nature and severity of the condition or disease being treated. Prescription of treatment is within the responsibility of general practitioners and other medical doctors. The acceptable excipient, adjuvant, carrier, buffer or stabiliser should be non-toxic and should not interfere with the efficacy of the secreted antimicrobial protein. The precise nature of the carrier or other material will depend on the route of administration, which is preferably oral. The amount that is “effective” will vary from subject to subject, depending on the age and general condition of the individual, the particular concentration and composition being administered, and the like. Thus, it is not always possible to specify an exact effective amount. However, an appropriate effective amount in any individual case may be determined by one of ordinary skill in the art using routine experimentation. Furthermore, the effective amount is the concentration that is within a range sufficient to permit ready application of the formulation so as to deliver an amount of the drug that is within a therapeutically effective range.

[0041] The L. lactis bacteria useful in the disclosed probiotic composition may be provided as a live culture, as a dormant material or a combination thereof. Those skilled in the art will appreciate that the L. lactis bacteria may be rendered dormant by, for example, a lyophilization process, as is well known to those skilled in the art.

[0042] An example of an appropriate lyophilization process may begin with a media carrying appropriate L. lactis bacteria to which an appropriate protectant may be added for cell protection prior to lyophilization. Examples of appropriate protectants include, but are not limited to, distilled water, polyethylene glycol, sucrose, trehalose, skim milk, xylose, hemicellulose, pectin, amylose, amylopectin, xylan, arabinogalactan, starch (e.g., potato starch or rice starch) and polyvinylpyrrolidone. Gasses useful for the lyophilization process 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION include but are not limited to nitrogen and carbon dioxide.

[0043] In one aspect, the L. lactis bacteria in the disclosed probiotic composition may be provided as a dispersion in a solution or media. In another aspect, the L. lactis bacteria in the disclosed probiotic may be provided as a semi-solid or cake. In another aspect, the L. lactis bacteria in the disclosed probiotic may be provided in powdered form.

[0044] Quantities of appropriate L. lactis bacteria may be generated using a fermentation process. For example, a sterile, anaerobic fermentor may be charged with media, such as glucose, polysaccharides, oligosaccharides, mono- and disaccharides, yeast extract, protein / nitrogen sources, macronutrients and trace nutrients (vitamins and minerals), and cultures of the desired L. lactis bacteria may be added to the media. During fermentation, concentration (colony forming units per gram), purity, safety and lack of contaminants may be monitored to ensure a quality end result. After fermentation, the L. lactis bacteria cells may be separated from the media using various well known techniques, such as filtering, centrifuging and the like. The separated cells may be dried by, for example, lyophilization, spray drying, heat drying or combinations thereof, with protective solutions / media added as needed.

[0045] The probiotic compositions may be prepared in various forms, such as capsules, suppositories, tablets, food / drink and the like. The probiotic compositions may include various pharmaceutically acceptable excipients, such as microcrystalline cellulose, mannitol, glucose, defatted milk powder, polyvinylpyrrolidone, starch and combinations thereof.

[0046] The probiotic composition may be prepared as a capsule. The capsule (i.e., the carrier) may be a hollow, generally cylindrical capsule formed from various substances, such as gelatin, cellulose, carbohydrate or the like. The capsule may receive the probiotic bacteria therein. Optionally, and in addition to the appropriate probiotic bacteria, the capsule may include but is not limited to coloring, flavoring, rice or other starch, glycerin, caramel color and / or titanium dioxide.

[0047] The probiotic composition may be prepared as a suppository. The suppository may include but is not limited to the appropriate probiotic bacteria and one or more carriers, such as polyethylene glycol, acacia, acetylated monoglycerides, carnuba wax, 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION cellulose acetate phthalate, corn starch, dibutyl phthalate, docusate sodium, gelatin, glycerin, iron oxides, kaolin, lactose, magnesium stearate, methyl paraben, pharmaceutical glaze, povidone, propyl paraben, sodium benzoate, sorbitan monoleate, sucrose talc, titanium dioxide, white wax and coloring agents.

[0048] The probiotic composition may be prepared as a tablet. The tablet may include the appropriate probiotic bacteria and one or more tableting agents (i.e., carriers), such as dibasic calcium phosphate, stearic acid, croscarmellose, silica, cellulose and cellulose coating. The tablets may be formed using a direct compression process, though those skilled in the art will appreciate that various techniques may be used to form the tablets. A capsule may also be used to contain the composition.

[0049] The probiotic composition may be formed as food or drink or, alternatively, as an additive to food or drink, wherein an appropriate quantity of probiotic bacteria is added to the food or drink to render the food or drink the carrier.

[0050] The concentration of probiotic bacteria in the probiotic composition may vary depending upon the desired result, the type of bacteria used, the form and method of administration, among other things. For example, a probiotic composition may be prepared having a count of probiotic bacteria in the preparation of no less than about 1×106colony forming units (CFUs) per gram, based upon the total weight of the preparation.

[0051] When lactic acid bacteria are used as gut expression vehicles, various dairy products, such as youghurt, youghurt pellets, or other milk products may be used as the physical carrier for oral administration, with or without the above mentioned adjuvants or carriers.

[0052] In another aspect, there is provided the use in the manufacture of a medicament of a therapeutically effective amount of a probiotic as defined above for administration to a subject.

[0053] The probiotic in its final form is expected to have a very low production cost and be highly scalable. In addition, it should have a long shelf life and not require refrigeration. A physician's prescription may not be required. Thus, the market is expected to be unusually wide. The probiotic is expected to provide sophisticated control at a very low price. 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0054] The probiotic compositions described herein can be used to prevent or treat H. pylori infections, or diseases or disorders caused by H. pylori, in humans and animals. The probiotic compositions may be administered as a prophylactic, prior to an exposure or challenge with H. pylori. The probiotic compositions may be administered therapeutically, after an infection with H. pylori has occurred. The probiotic compositions may be incorporated into animal feed or animal drinking water.

[0055] Helicobacter pylori is the primary cause of 78% of gastric cancer cases, providing an opportunity to prevent cancer by controlling a single bacterial pathogen within the complex gastric microbiota. There is a great need for alternatives to antibiotics in the control of H. pylori and the prevention of gastric cancer. The high prevalence of H. pylori in the human population, the induction of microbial dysbiosis by antibiotics, and increasing antibiotic resistance call for a more sustainable approach. By selectively eliminating the pathogen and retaining the commensal community, H. pylori control may be achieved without adverse health outcomes. Antibiotics are typically used as a therapeutic post- infection, but a more targeted, less disruptive approach can be used as a long-term prophylactic against H. pylori or, by extension, against other gastrointestinal pathogens. Furthermore, the modular nature of the guided AMP (gAMP) technology allows for the substitution of different guides for different pathogens and the use of a cocktail of gAMPs to avoid the development of pathogen resistance. Example 1

[0056] Engineered proteins that specifically kill certain pathogenic bacteria without harming unrelated commensal bacteria have been developed. The specificity of killing is due to a targeting (guide) peptide attached to an antimicrobial peptide as expressed from a hybrid gene. In the present example the skin pathogen, Staphylococcus aureus, was targeted using purified guided antimicrobial protein produced from an E. coli expression system. However, the targeting system can be modified to specifically kill any bacterium.

[0057] In this example, two commonly used antimicrobial peptides (AMPs), plectasin and eurocin, were genetically fused to the targeting peptide A12C, which selectively binds to Staphylococcus species. It should be noted that A12C peptide was developed using a generic biopanning technique; in theory, any bacterium can be targeted 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION using this method for producing guide proteins. A12C was developed by another laboratory to serve as a guide protein for vesicles, which also illustrates that peptides developed for other purposes can be repurposed to serve as guide proteins for antimicrobial peptides. The targeting peptide did not decrease activity against the targeted Staphylococcus aureus and Staphylococcus epidermidis, but drastically decreased activity against the non-targeted species, Enterococcus faecalis and Bacillus subtilis. This effect was equally evident across two different AMPs, two different species of Staphylococcus, two different negative control bacteria, and against biofilm and planktonic forms of the bacteria.

[0058] Methods:

[0059] Reagents. The pE-SUMOstar vector (LifeSensors) was grown in 10-^ and BL21 E. coli (New England Biolabs) and AMP was released from expressed fusion / AMP using Ulp1 protease produced in house. The AMPs plectasin (GFGCNGPWDEDDMQCHNHCKSIKGYKGGYCAKGGFVCKCY (SEQ ID NO:7); MW 4408) and eurocin (GFGCPGDAYQCSEHCRALGGGRTGGYCAGPWYLGHPTCTCSF (SEQ ID NO:8); MW 4345) were expressed from pE58 SUMOstar as were A12C-plectasin (MW 6137) and A12C-eurocin (MW 6074), both of which had the A12C targeting peptide (underlined) plus a short linker (GVHMVAGPGREPTGGGHM) (SEQ ID NO:9) genetically fused to the N- terminus of the respective AMP sequences. As a control, plectasin and eurocin were also conjugated with the AgrD1 bacterial pheromone sequence (YSTCYFIM)(SEQ ID NO:10) (Mao et al.2013) at the N- terminus. Synthetic A12C peptide (Biosynthesis) was used as a “target peptide only” control. FIG. 1 shows the pE-SUMOstar vector carrying AMP for expression in E. coli BL21 cells. SUMO protease site is between SUMO and A12C-AMP.

[0060] Expression, Purification and Analysis of Fusion Proteins. The DNA sequences for the AMPs were synthesized (Integrated DNA Technologies) and ligated into the pE66 SUMOstar vector and cloned into E. coli 10-beta cells. Plasmid from these were used to transform E. coli BL21 cells for protein expression. Transformed cultures were grown out and induced with IPTG according to standard procedures. The resulting bacterial pellets were resuspended in PBS / 25 mM imidazole / 0.1 mg / ml lysozyme and frozen overnight. The cells were then thawed, sonicated, and ultracentrifuged at 80,000 x g for 1 h at 4˚C and the 6his / SUMO / AMP fusion protein in the supernatant was purified by nickel 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION column chromatography. The AMP was separated from SUMO by proteolysis using Ulp1 (1U per 100 ^g fusion protein) at 4˚C overnight and the cleavage was evaluated by SDS- PAGE. Yields were calculated from the SDS-PAGE data, using NIH ImageJ to measure band density and the marker lane bands for mass reference. Mass spectrometry was used to ensure the proper cleavage of the AMP from the SUMO carrier protein. In-gel tryptic digest (Thermo Fisher) was performed on the AMP excised from the SDS-PAGE gel. The digest was examined by LC-ESI-MS (Synapt G2-S, Waters) at the Baylor University Mass Spectrometry Center. The analysis of the MS data was done by MassLynx (v4.1) The spectra of each protein, both non-targeted and targeted, were peak centered and MaxEnt3 processed and then matched against hypothetical peaks from peptides generated by simulated trypsin digestion of the respective proteins.

[0061] Hemolytic Activity Assay. Guided AMPs, non-guided AMPs and synthetic A12C peptide were assessed for human hemolytic activity via exposure to washed human erythrocytes. Whole blood cells were collected a healthy volunteer using standard procedures (Evans et al.2013) and cells were diluted in phosphate buffered saline to 5x108 cells / ml. To initiate hemolysis, 190 ^l of the cells was added to 20 ^l of a 2-fold serially diluted peptide / test reagent in phosphate buffered saline. Wells without peptide were used as negative controls, while wells containing 1% 85 Triton X-100 were used as positive controls.

[0062] In Vitro Bactericidal Activity Assay. The Ulp-1 protease-cleaved proteins were tested for antimicrobial assays against four strains of bacteria: Staphylococcus aureus, Staphylococcus epidermidis, Enterococcus faecalis and Bacillus subtilis. These four species were selected because they are gram positive and the AMPs plectasin and eurocin are specifically active against gram positive bacteria (Mygind et al.2005, Oeemig et al.2012). The component controls were free SUMO protein and synthetically produced A12C peptide. Vancomycin was used as the positive control. The standard protocol for a microtiter plate assay with serial dilution was used in which serial 2-fold dilutions of test peptide were made across a 96-well plate containing uniform bacterial inoculum across the peptide dilutions. After bacterial growth in the presence of peptide, cell viability was assayed with resazurin. Experiments with all peptides against all bacterial species were performed with >5 replicates each. 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0063] In Vitro cell kinetics study. Ulp-1 protease-cleaved peptides were assayed to determine their dynamic action against the bacteria in a growing culture. The bacteria were grown at 37°C with shaking and diluted to ~1x108CFU / ml. To these cultures were added plectasin or eurocin, at 3x the respective minimum inhibitory concentrations, or the A12C-targeted versions at these same respective concentrations. The vancomycin control concentration was the mean of the molar concentration of plectasin and eurocin used. Growth was then monitored from 2-10 h after addition of the peptides, diluting 10 ^l of culture in medium and plating onto Mueller-Hinton agar plates. The number of colonies was recorded the next day.

[0064] In Vitro biofilm inhibition assay. In addition to planktonic cultures, biofilm cultures were used to assay inhibition by the peptides, using standard procedures (O’Toole 2011). Briefly, overnight cultures were diluted 1:100 and added to serially diluted peptides. Biofilms were allowed to grow for 24-36 h of unshaken culture. The liquid was removed and the biofilms were washed, dried and fixed with methanol and then stained with Crystal Violet, which was later dissolved with 30% acetic acid and the resulting solution measured for absorbance at 540 nm to quantify the amount of biofilm formed. All assays were run in triplicate or greater.

[0065] Results:

[0066] Protein Expression and Purification. AMP / SUMO fusion proteins, with or without the A12C targeting domain, were highly expressed in E. coli BL21 cells. These were successfully cleaved with SUMO protease (Ulp-1) into their component AMP and SUMO carrier protein and were clearly visualized with SDS-PAGE as 4-6 kDa free AMP and ~17 kDa SUMO / AMP fusion proteins. FIG.2 shows expression of SUMO / AMP in E. coli and cleavage of AMP free of SUMO fusion partner, where Lane 1: free SUMO control and Lanes 2-9: Intact fusion proteins (even lanes) and cleaved products (odd lanes) in the following order: SUMO / plectasin, SUMO / A12C-plectasin, SUMO / eurocin, SUMO / A12C- eurocin. Arrows: free AMP The average yields (n>=3) of the proteins plectasin, A12C- plectasin, eurocin and A12C-eurocin were 15-26 mg (3-4 ^moles) per L of culture. For peptide confirmation, peptides were extracted from the SDS-PAGE gel bands, digested by trypsin and analyzed by mass spectrometry. Peptide identities were confirmed using the MassLynx (v4.1) application (Waters). 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0067] Hemolytic Activity Assay. In concordance with previously published individual studies on plectasin and eurocin (Mygind et al.2005, Oeemig et al.2012, Yacoby et al. 2006), both guided and un-guided fusion peptides, along with the free A12C peptide control, displayed no hemolytic effect on human erythrocytes in comparison to a 20% Triton-X positive control (data not shown).

[0068] In Vitro Bactericidal Activity Assay. Differential toxicity against off target bacteria was observed with the A12C targeting peptide added to the AMPs. A12C-AMPs retained their toxicity against both of the targeted staphylococci bacterial species but showed a dramatic decrease in toxicity against the off target bacterial species relative to unmodified AMPs. FIG. 3 shows log values for minimum inhibitory concentrations (MIC) in ^M for non-targeted and targeted analogues of eurocin and plectasin against Bacillus subtilis, Enterococcus faecalis, Staphylococcus aureus and Staphylococcus epidermidis. The boxed regions represent 50% of the values while the bars represent 95%. Unmodified plectasin and eurocin had the expected mean MIC values of 3-6 ^M, which are typical values for AMPs with sequential tri-disulfide bonds produced in E. coli expression systems (Li et al. 2010, Parachin et al. 2012, Li et al. 2017). In contrast, the addition of the A12C guide peptide rendered these AMPs essentially noninhibitory to the off target bacteria, with MIC values >70 ^M. In all cases, the MIC values for A12C / AMP versus AMP were significantly different for both of the off target bacteria, E. faecalis and B. subtilis (p<0.001; ANOVA 2- 139 tailed test). Negative controls (SUMO alone and A12C alone) showed no antimicrobial activity (data not shown) and these were run for all experiments.

[0069] In Vitro cell kinetics study. Growth kinetics over an 8 to 10 hour period more conclusively demonstrated the loss of antimicrobial activity of the A12C / AMP against the off target bacterial species. For these bacteria, A12C / AMP treatment resulted in bacterial growth that lagged only slightly behind buffer control treated cultures. FIG. 4 shows the cell-kinetic profile for B. subtilis, S. epidermidis, S. aureus and E. faecalis (clockwise), created by plotting log CFU / ml of the bacteria grown in the presence of each peptide for 8- 10 hours collected in 2-3 hour intervals. Unmodified AMPs were bactericidal similar to the vancomycin control. In contrast, all peptides - both guided and unguided - demonstrated a strong bactericidal effect against the target bacteria S. epidermidis and S. aureus, similar to the vancomycin positive control. The relatively flatter growth curve for the B. subtilis 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION control cultures reflects its growth kinetics, which is far slower than that of other bacteria.

[0070] In Vitro biofilm inhibition assay. Growing bacterial cultures with the peptides demonstrated the preferential inhibition of bacterial biofilm of the Staphylococcus strains by the targeted AMPs over the non-Staphylococcus bacteria. FIG. 5 shows biofilm inhibition activity evaluated by plotting the absorbance of crystal violet (540 nm) against the concentration of 4 AMPs on the 4 bacteria - B. subtilis, S. epidermidis, S. aureus and E. faecalis (clockwise). (* = p<0.1, ** = p<0.05, n>=3). The absorption reading (hence, the quantity of biofilm formed) decreased with the increase in peptide concentration for all the 4 bacteria when treated with unguided peptides but the guided peptides did not have similar effects on B. subtilis and E. faecalis with significant (p <0.10 or p<0.05) difference in the absorbance values between targeted and non-targeted AMPs at concentrations beyond 6.25 ^M.

[0071] This example demonstrates successful targeting of the AMPs plectasin and eurocin against two staphylococcal bacteria. Importantly, this was achieved by essentially eliminating the activity against the two off target bacteria tested. This is the expected outcome for an antimicrobial therapy that preserves the commensal members of the microbiome while killing the pathogenic target bacteria. This is also the outcome that was achieved against S. aureus by Mao et al. (2013) with the use of a bacterial pheromone peptide for targeting of plectasin. Other than a lower MIC for the unmodified plectasin itself, the same drastic degree of reduction in the activity against the off target bacteria, E. faecalis and B. subtilis was seen, as was reported by Mao et al. (2013). Thus, it is demonstrated that a biopanning-derived ligand works as efficiently as a pheromone-derived ligand, which is the class of targeting peptide used in all targeted AMPs to date. It should be noted that the pheromone-derived ligand was more specific than A12C, with activity against S. aureus but not S. epidermis, while A12C / plectasin was highly active against both species.

[0072] Four main sources of ligands exist for use as guide peptides for AMPs. First, bacterial pheromones are species-specific peptide signals which trigger the development of competence, virulence, or other capabilities, and pheromone peptides have been determined for many pathogenic bacteria (Monnet et al. 2016). Second, biopanning is a means of screening random libraries of peptides for the ability to bind to a target sequence, such as a receptor on a bacterial cell. Usually, a bacteriophage is used to display the members of the 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION peptide library (Wu et al.2016). Third, bacteriophage receptor binding proteins can be used as a resource for the development of targeting peptides for AMPs. The receptor binding proteins of phages against many pathogenic bacteria have already been characterized (Dowah and Clokie 2018, Nobrega et al.2018). In addition, screens for new phages against lesser studied bacterial pathogens can be carried out (Weber-D^browska et al.2016). Fourth, virulence factors of the targeted bacterial pathogen can be targeted by using targeting (guide) peptides consisting of the sequence of the host receptor that is bound by the bacterial virulence factor. In this way, the host receptor sequence is used as a guide peptide to direct an AMP back to the bacterial pathogen. EXAMPLE 2

[0073] Potential guide peptide sequences were selected from a random library of 1 billion unique 12-mer peptides rather than being taken from any published work. In this process, strongly binding peptides were selected by exposing the library to purified recombinant VacA protein and retaining the peptides binding to the VacA. VacA protein bound to a strep tag was produced from H. pylori 60190 harboring this mutant vacA gene. VacA protein was purified by affinity chromatography on an affinity column. The candidate peptides were in the form of a phage display and this phage display library is sold by New England Biolabs, Inc., specifically as a tool for selecting peptides that bind to a protein of interest. This tool is derived from methods leading to a Nobel Prize awarded in 2018 on phage display or “biopanning.” The technique works because the phages displaying peptides that allow binding to the target protein can be washed off the target protein even in tiny quantities and then amplified by being allowed to replicate in E. coli, their natural host. This new population of phages is then allowed to bind fresh VacA protein but in the presence of a detergent that allows only the strongest binding peptides of the first round to bind in the second round. This process is repeated for a third or fourth round, after which the identity of the best binding peptides is revealed by sequencing the phage DNA.

[0074] In this example, purified VacA was bound to polystyrene 96-well plates and then blocked with bovine serum albumin. The commercial 12-mer peptide library (Ph.D.- 12 from New England Biolabs, Inc., USA), containing 1 billion unique peptide sequences displayed on M13 phage was added to allow binding to the immobilized VacA. Following washings to remove unbound phages, the target bound phages were eluted and used to infect 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION the E. coli host strain supplied by the kit. Phage particles were recovered as a pellet using 20% polyethylene glycol-8000 / 2.5 M NaCl precipitation. The process was repeated using increasing concentrations of detergent in the buffer to increase selection stringency with each round of panning. The first round involved panning against purified VacA; the second against whole cell H. pylori, the third against purified VacA, and the fourth against whole cell H. pylori.

[0075] The next steps quantitate the effectiveness of binding peptides and choose the most effective ones to serve as guide peptides for gAMP. First, the best binding phages are quantified by allowing them to bind to VacA or H. pylori adhering to a well in a microtiter plate and quantitatively measuring binding using an antibody when binds the phage. In essence, this is an ELISA-type assay to measure binding. In this example, thirty- six (36) phage plaques were selected from the fourth round of panning and assayed by phage ELISA (enzyme-linked immunosorbent assay). The guide peptide of each phage was sequenced and after eliminating identical sequences, 18 unique phage clones were analyzed. Whole cells of H. pylori fixed with 4% paraformaldehyde were bound to a polystyrene 96- well plate, blocked with bovine serum albumin, and then provided with one phage clone per well (in three replicates). After washing the wells, HRP-conjugated anti-M13 monoclonal antibody was added and allowed to bind. The HRP substrate, 3,3', 5, 5' tetramethylbenzidine was added and allowed to react. The colorimetric product was read at 450 nm on a plate reader to quantitatively measure strength of binding of each phage clone to the whole H. pylori cells. The binding to both wild type H. pylori and mutant H. pylori with the vacA gene deleted was quantified. As shown in FIG. 6B, several phage clones showed clear differential binding of the VacA on the wild type (WT) containing wells over the wells containing the VacA-deleted cells (P11, P12, P16, P18, P21, P4, P5, P8). Some degree of consensus was noted among the guide candidates (FIG 6A). These guide sequences advanced to testing as guided GFP constructs.

[0076] Second, the peptide sequences of the best binding phages are then used to construct guided green fluorescent protein. In order to narrow down the guide candidate field, the sequences were tested for their ability to bind a GFP protein to whole H. pylori cells harboring or lacking VacA protein. These gGFPs are then measured for their ability to bind H. pylori using fluorescently activated cell sorting (FACS). Each guide sequence 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION was fused to a 12 bp linker at the N-terminus of the GFP ORF of expression vector u- mspGFP (UC Berkeley QB3 MacroLab), which is a ligation independent cloning vector based on pET28. Guided GFP proteins, with a C-terminal 6his tag, were generated by expression in E. coli BL21 cells and purification via nickel affinity chromatography.

[0077] The gGFP proteins were allowed to bind either wild type of VacA deleted H. pylori cells and then the binding was assayed by flow cytometry. As seen in FIG. 7, only three guide sequences gave differential binding for wild type H. pylori cells displaying VacA ("WT") compared to H. pylori cells lacking VacA ("Deletion"). These were sequences identified as P5 (WNSWPFPMASFE (SEQ ID NO:1)), P8 (TQQGINGVLVSV (SEQ ID NO:2)), and P18 (FHEGTYFKVWA (SEQ ID NO:3)). The top 3 guide peptides advanced to gAMP testing. EXAMPLE 3

[0078] The selectivity of killing of guided AMPs (gAMPs) due to these phage display derived sequences was then examined. Specifically, the increased toxicity against H. pylori of gAMPs utilizing the P5, P8, or P18 guide sequence was measured as well as the relative reduction in activity against off-target bacterial species, human cells, and H. pylori missing an intact vacA gene.

[0079] The protein expression system used was generated in-house and was based on the use of the onconase RNase carrier protein which confers insolubility at neutral pH to proteins it is fused to (Pane et al., 2016). In this way, pexiganan, an AMP highly toxic to both H. pylori and E. coli, could be produced in an E. coli expression system since the AMP and gAMP peptides would be insoluble and inactive in the E. coli cell. After the E. coli cell was harvested and lysed, the fusion protein (onconase:AMP) was solubilized at pH 3 while all other proteins dropped out of solution and were eliminated by centrifugation, resulting in a highly pure final product. The onconase carrier itself was eliminated by treatment at pH 2 / 60˚ C to cleave an Asp / Pro bond between carrier and gAMP. In this custom system, the guide sequence was separated from the pexiganan AMP by a linker peptide. The short linker version comprised the peptide sequence VGCGSGG (SEQ ID NO:11). The long linker version comprised the peptide sequence VGCGSGGHGSGGGS (SEQ ID NO:12). Thus, the sequence compositions of each linker were very similar with the linker differing primarily in length. The guide peptide was added via a ligation independent cloning method 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION directly upstream of the linker. The final protein product produced in E. coli was the following insoluble fusion protein: onconase / AspPro cleavage site / guide / linker / pexiganan.

[0080] The toxicity against bacteria was measured by minimum inhibitory concentration (MIC). The standard protocol for an MIC microtiter plate assay with serial dilution was used in which serial two-fold dilutions of test peptide were made across a 96- well plate containing uniform bacterial inoculum across the peptide dilutions. After bacterial growth in the presence of peptide, cell viability was assayed with resazurin. Experiments with all peptides against all bacterial species were performed with 6 replicates each.

[0081] The results of the MIC assays are shown below in Table 1. Table 1 shows the minimum inhibitory concentration (MIC) in µM needed to completely inhibit growth of target (H. pylori 60190) and Table 2 shows the MIC values against the off-target bacteria. Table 1. Minimum Inhibitory Concentration (MIC) of guided pexiganan against H. pylori Bacterium Tested Pexiganan P5-Pexiganan P8-Pexiganan P18- Pexiganan (µM) (µM) (µM) (µM) Short Long Short Long Short Long H. pylori 60190 2 0.25 0.25 0.5 0.5 1 0.25 H. pylori ^vacA21 1 0.5 0.5 1 2 Table 2. MICs for guided pexiganan against off-target gram-negative bacteria Bacterium Pexiganan P5-Pexiganan P8-Pexiganan P18- Pexiganan Tested (µM) (µM) (µM) (µM) Short Long Short Long Short Long P. aeruginosa 16 16 >64 64 64 >64 >64 A. faecalis 8 16 8 8 8 16 8 A. baumanii 4 8 8 4 4 4 4

[0082] The lower the MIC value, the more toxic the AMP is against that bacterium. Six replicates were run for each cell of data displayed but no error bars or significant digits are shown since the test is run with discrete dilutions of the peptide and all replicates showed 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION identical results except for H. pylori 60190 / P5-Pex Long. The internal negative control was H. pylori ^VacA. The off-target bacteria used were Pseudomonas aeruginosa, Enterococcus faecalis, and Acinetobacter baumanii. Bacterial densities were 10^5 (Log 5) except for the noted high titer tests which investigated overloading with bacterial culture.

[0083] As seen in Table 1, the targeted H. pylori 60190 human virulent strain was strongly inhibited by pexiganan AMP, with an impressive MIC value of 2. P18 / short linker was twice as toxic while the P8 guide AMPs were four times as toxic. The P5 guide conferred 8x the toxicity of unguided pexiganan, as did P18-long linker. Thus, the guides provided a substantial increase in toxicity against H. pylori. Furthermore, the guides provided a shielding from AMP toxicity for off-target bacteria in some cases, with values mostly at 8 to greater than 64 µM. Compared to the MIC value of 0.25 against the targeted H. pylori, this indicates a differential toxicity of 32- to greater than 256-fold between targeted and off-target bacteria. In summary, the MIC assays reveal both increased toxicity against the targeted pathogen and decreased toxicity against off-targets compared to the standard, unguarded AMP. This will translate into selective killing of H. pylori, giving the commensal bacteria the advantage for outcompeting the H. pylori in the gastric microbiota.

[0084] The cytotoxicity against human gastric AGS cells was next tested. Results are shown in FIG.8. The cytotoxicity assay was run as follows.200 µL of AGS cells at a concentration of roughly 1.2x106cells / ml were added to wells of a 96-well, cell culture treated polystyrene plate. Cells were allowed to grow 48 h, with media changed after 24 h. After 48 h, 180 µl of media with 20 µl of 10x concentrated AMP were added to wells. AMP treatment was removed after 20 hours and media was replaced with media supplemented with 0.03 mg / ml resazurin to measure cellular respiration and returned to incubator. After 4 hours of respiration, supernatant was transferred to a black, clear-bottomed 96-well plate and fluorescence read (Excitation: 530 / Emission: 590). All experiments were performed twice with three replicates each (n=6).

[0085] Untreated control wells represented 100% respiration. A decrease in respiration of 30% was considered to be cytotoxic. Cytotoxicity was calculated as a reduction of respiration, using the formula: Toxicity = (RFUUntreated – RFUAMP Treated ) / RFUUntreated 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION

[0086] FIG. 8 shows cytotoxicity of the tested gAMPs against human gastric AGS cells. The dotted line indicates 30% toxicity which is generally regarded the limit for drug toxicity in cytotoxicity assays. Guides P8 and P18 showed some toxicity at high concentrations (16 – 32 µM), but Guide P5 showed consistently low cytotoxicity even at the highest concentrations. This assay clearly identified the strongest lead candidate gAMP in the combination of the P5 guide and the long linker. With this combination, very acceptable results were seen even at 32 µM gAMP. Combined with the results seen in Table 1 (0.25 µM MIC against H. pylori), this results in an exceptionally wide 128-fold therapeutic window. It should be noted that gAMPs using guide P8 with the short linker and guide P18 with either linker showed lack of cytotoxicity at 16 or 32 µM which is also acceptable, resulting in a wide therapeutic window.

[0087] FIG. 9 provides a quantitative summary of all the parameters discussed above for the top gAMP, P5-pexiganan with the short linker. It is evident that a concentration effective against H. pylori (0.25 µM) does not affect human cells, with concentrations of 32 to >64 µM required for cytotoxicity. Thus, a very wide therapeutic window exists. Similarly, 0.25 µM P5-pexiganan has little effect on the off-targets tested, which require 8 to 16 µM of gAMP for toxicity, meaning that a wide differential in toxicity exists between target and off-target, which may allow the preservation of the gastric microbiota at concentrations used to treat H. pylori. Similar trends were seen for P8 and P18 pexiganan gAMPs. EXAMPLE 4

[0088] An exemplary probiotic bacterium, Lactococcus lactis, has been shown to survive well in the stomach of mice. A vector has also been developed that greatly facilitates Lactococcus lactis engineering. To create this vector (shown in FIG. 10), the original Lactococcus lactis vector, pT1NX, was modified by the addition of an E. coli origin of replication and a kanamycin resistance cassette, both from the SUMO-based E. coli expression vector, pE-SUMOstar. In FIG. 10, the kanamycin resistance block represents both the kanR cassette and the E. coli origin of replication. This binary vector (pT1bin1) can be grown in E. coli to facilitate the addition of AMP or guide sequence inserts by recombinant DNA techniques. Generous quantities of plasmid can be produced via standard plasmid preparation techniques in order to ease the transformation of Lactococcus lactis. 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION This latter transformation is difficult to achieve with ligation products, but is easier with DNA from plasmid preparations.

[0089] It has been demonstrated in vitro that engineered Lactococcus lactis secreting antimicrobial peptide kills other bacteria in vitro. This is reported in FIG. 11 as the survival of E. coli in the presence of broth culture of Lactococcus lactis secreting antimicrobial peptide with or without a guide peptide. FIG.11 shows the viability of E. coli in the presence of different antibiotic dilutions and supernatants. It should be noted that the legend is in reverse order of the lines, top to bottom, with the upper line in the graph being the buffer control and the lower line being vancomycin. To obtain the results shown in FIG. 11, cultures of Lactococcus lactis containing either the empty pT1bin1 vector, pT1bin1 harboring the antimicrobial peptide laterosporulin, or pT1bin1 harboring laterosporulin genetically fused to the guide peptide from multimerin were centrifuged to remove bacterial cells and the resulting supernatants were added to separate starter cultures of E. coli to check for inhibition of E. coli growth. The starter culture used supplying all replicates consisted of 500 µl of overnight culture of E. coli diluted in 50 ml of LB broth. Three replicates of each treatment were conducted and each point in the graph represents an average with correspoding error bars. To run the treatments and replicates, a 96-well microtiter plate was used. For each well, 100 µl of diluted Lactococcus lactis supernatant was added to 100 µl of E. coli starter culture. As seen in the x-axis of FIG.11, the dilutions used ranged from no dilution (100 µl of 100% supernatant added to the 100 µl of E. coli) down to 1 / 200 dilution of supernatant (100 µl of 0.5% supernatant added). Antibiotic positive controls were diluted similarly, with the starting concentrations (undiluted) stated in the legend. The y-axis of FIG. 11 represents the inhibition of E. coli viability by these supernatant and antibiotic dilutions. E. coli viability was measured by plating onto LB agar plates the cultures in each well after 4 hours of exposure to supernatant or antibiotic. The resulting colonies appearing on the plates were recorded, with the undiluted buffer control treatment being set to 100% and all other treatments being converted to a fraction of this value, as plotted on the y-axis.

[0090] Looking at FIG. 11, it can be seen that the buffer control did not inhibit E. coli. However, Lactococcus lactis broth culture (with cells removed) did inhibit E. coli even with no recombinant antimicrobial peptide present (empty vector control). This is 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION considered the baseline for examining the effect of the secreted recombinant proteins. The expression of laterosporulin by Lactococcus lactis resulted in a significant decrease in viability of E. coli compared to this baseline. However, there was no significant difference seen between the empty vector baseline and the multimerin-guided (targeted) lactosporulin. This means that the guide peptide completely abolished antimicrobial activity of laterosporulin against the nontarget bacterium E. coli. This is in agreement with results shown in Example 1 with Staphylococcus. This data supports the ability of these extracts to kill different target bacterium, such as Helicobacter pylori.

[0091] In this example, multimerin 1 (MM1) was used as the guide peptide in various gAMPs, as MM1 specifically binds to H. pylori cells. When fused to green fluorescent protein (GFP), the MM1-guided GFP bound strongly to H. pylori but showed no affinity to a panel of six off-target gastrointestinal bacteria. Guided AMPs (gAMPs) expressed by the probiotic were generated. Results show that an antimicrobial peptide (AMP) fused to the multimerin-derived guide peptide specific for Helicobacter pylori, expressed from a hybrid gene and secreted from the probiotic Lactococcus lactis, can specifically kill H. pylori when the probiotic is co-cultivated with H. pylori in vitro.

[0092] In the co-cultures, different dilutions of L. lactis were used but each well had 10 µl of H. pylori culture (~3000 CFUs). The L. lactis secreted AMP, gAMP or contained an empty expression vector. Alyteserin and CRAMP were the AMPs tested. These were constructed either genetically fused to the multimerin-derived guide peptide (guide AMP or gAMP) or not (AMP). The amount of H. pylori present in the co-culture at any given time point was measured by qPCR, using primers specific for the vacA gene itself, which codes for the receptor protein to which the gAMP binds. The entire experiment was run in triplicate.

[0093] FIG. 12 shows the vector for Lactococcus lactis secretion of AMPs and gAMPs. The ORFs of the AMPs, codon-optimized for Lactococcus lactis, were cloned into the modified pT1NX-kanR (pTKR) vector for L. lactis expression / secretion in between the restriction enzyme sites BamHI and SpeI by replacing the spaX protein of the original plasmid. The P1 promoter upstream of the BamHI cut-site controls the downstream expression as a constitutive promoter which is upregulated by low pH. The usp45 gene immediately upstream of BamHI site codes for an endogenous signal peptide of L. lactis 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION that allows secretion of the resulting fusion peptide. After ligation of the AMP / tAMP into pTKR vector, it was transformed into E. coli (10^, NEB) and plated onto kanamycin selective plate. The pT1NX plasmid (LMBP 3498) has erythromycin resistance but was modified to create pTKR as shown in FIG. 12, which also has kanamycin resistance for cloning into electrocompetent E. coli (10^, NEB) for plasmid propagation. Extracted plasmid from the E. coli was then electroporated into electrocompetent L. lactis MG1363 (LMBP 3019) and plated on erythromycin selective GM17 plates (30oC, microaerobic, overnight). After screening for the presence of the AMP / gAMP ORFs with PCR, selected colonies were propagated in liquid cultures of M17 broth with glucose (0.5% w / v) in the presence of erythromycin (5 µg / ml).

[0094] The following AMPs and guided AMPs (gAMPs) were cloned into the secretion vector pTKR. The multimerin1 (MM1) guide peptide sequence MQKMTDQVNYQAMKLTLLQK (SEQ ID NO:13) is underlined and the serine / glycine linker sequence is in bold.

[0095] L. lactis AMP / gAMP clones were propagated from glycerol stocks and grown in GM17 broth overnight with erythromycin (5 µg / ml) with no shaking. H. pylori stocks were first propagated on Blood-TS agar overnight with microaerobic condition and 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION >5% CO2environment. Then colonies from the plate were transferred to a TS broth with newborn calf serum (5%) and grown overnight under microaerobic condition and >5% CO2environment. The L. lactis cultures were serially diluted in a 96-well culture plate with TSB broth to make up a volume of 100 µL. To each well, 10 µL of the overnight H. pylori culture was added and each well volume was brought up to 200 µL with more TS broth. The plate was left to grow overnight in a microaerobic environment with >5% CO2. After 24 h, well contents from the culture plate were transferred to a 96-well PCR plate. That PCR plate was sealed and heated for 15 min at 100oC and chilled at 4oC for 5 min. Then the plate was centrifuged at 2000 g for 2 min and the supernatant was used as the template for qPCR. The qPCR was done using primers for VacA gene to quantify H. pylori (forward: 5^- ATGGAAATACAACAAACACAC-3^ (SEQ ID NO:20), reverse: 5^- CTGCTTGAATGCGCCAAAC-3^ (SEQ ID NO:21) and primers for acma gene for quantifying L. lactis. Standard curves for H. pylori and L. lactis were constructed by determining CTvalues for different dilutions of the overnight cultures of the respective bacteria (1 / 10, 1 / 100, 1 / 1000, 1 / 10000) in the qPCR plates, the CFUs for the dilutions were determined by plating on their respective agar plates.

[0096] FIG. 13 shows the results of qPCR detecting the vacA gene of H. pylori co- cultured with L. lactis expressing gAMPs or AMPs. L. lactis expressing AMPs with or without guide peptides knocked down the H. pylori culture to below the baseline of detection for this experiment (CTvalue of 40). Plain AMPs are represented with open symbols while gAMPs are represented with solid gray symbols. Alyteserin was not very effective unless fused to the guide peptide. The control experiment (solid line), with L. lactis carrying the empty vector, showed that the L. lactis probiotic, by itself, had little to no influence on the growth of H. pylori over 24 hours. Error bars represent 95% confidence limits.

[0097] Results showed that L. lactis expressing two different AMPs was able to knock down, to baseline levels, a vigorous H. pylori culture in vitro. The multimerin guide peptide sequence was shown to not interfere with AMP toxicity in CRAMP, with targeted and untargeted CRAMP equally toxic to H. pylori. In all cases, the gAMP (“MM1” prefix) was more toxic (lower on y-axis) than the corresponding AMP. In the case of the alyteserin AMP / gAMP pair, the guide peptide appeared to be a requirement for high toxicity to H. 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION pylori.

[0098] These precision antimicrobial probiotics, evaluated in co-culture in vitro with test bacteria, retained full toxicity against H. pylori but were 40 to 80-fold less toxic than unguided AMP probiotics against off-target bacteria. This was true with three different AMPs: alyteserin, CRAMP, and laterosporulin.

[0099] The same probiotics successfully served as therapy or prophylactic treatment against H. pylori infection in mice. In the therapeutic experiment, healthy mice were assayed on Day 0 for initial H. pylori titer by flushing the stomach with PBS by oral gavage, followed by analysis of the fluid by qPCR calibrated to CFU plating. Results are shown in FIG. 14. In FIG.14, the x-axis legend is as follows: (1) buffer control, (2) probiotic / no AMP control, (3) tetracycline / amoxicillin control, (4) alyteserin AMP probiotic, (5) guided alyteserin AMP probiotic, (6) CRAMP AMP probiotic, (7) guided CRAMP AMP probiotic, (8) laterosporulin AMP probiotic, (9) guided laterosporulin AMP probiotic. Since mice do not harbor H. pylori natively, none was detected. On Day 1, 2, and 3, mice were inoculated with H. pylori by oral gavage to produce a strong infection. Though there was some variation, by Day 5, mouse cohort groups were well infected as shown by qPCR of the gastric flush samples. After the flush was collected, mice were immediately inoculated with the appropriate probiotic treatment or with antibiotics or PBS buffer. Five days later (Day 10), mice receiving the probiotic expressing no AMP or the buffer control showed strong H. pylori growth relative to their starting titers on Day 5. However, mice receiving the antibiotic treatment or any of the probiotics expressing either AMP or gAMP showed a steep decline in H. pylori titer, with the AMP or gAMP probiotics producing complete clearance. As shown in FIG. 14, probiotics expressing three different AMPs, either unguided (treatments 4, 6, 8) or guided (5, 7, 9), all eliminated H. pylori from stomach by Day 10, bettering antibiotic treatment and empty vector L. lactis probiotic.

[0100] These results were in agreement with the prophylactic experiment. Mouse samples were collected by gastric flushes on Days 0, 5, 8 and 10 and these samples were analyzed for H. pylori titer as before. The probiotic treatments or PBS control were inoculated immediately following the sampling on Day 0. The H. pylori challenge was inoculated on Days 3, 4, and 5 and resulted in a strong infection for the PBS and empty probiotic mice five days later (Day 10). Results are shown in FIG. 15. In FIG. 15, the x- 27 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION axis legend is as follows: (1) buffer control, (2) probiotic / no AMP control, (3) alyteserin AMP probiotic, (4) guided alyteserin AMP probiotic, (5) CRAMP AMP probiotic, (6) guided CRAMP AMP probiotic, (7) laterosporulin AMP probiotic, (8) guided laterosporulin AMP probiotic. There was no antibiotic control in this experiment. As shown in FIG.15, probiotics expressing three different AMPs either unguided (treatments 4, 6, 8) or guided (5, 7, 9), all protected mice from H. pylori challenge infection. All probiotics expressing either AMP or gAMP gave strong protection against infection. REFERENCES Choudhury A, Ortiz PS, Young M, Mahmud MT, Stoffel RT, Greathouse KL, Kearney CM. 2023. Control of Helicobacter pylori with engineered probiotics secreting selective guided antimicrobial peptides. Microbiol Spectr. Sep 15:e0201423. doi: 10.1128 / spectrum.02014-23. Epub ahead of print. PMID: 37712669. Pane K, Durante L, Pizzo E, Varcamonti M, Zanfardino A, Sgambati V, Di Maro A, Carpentieri A, Izzo V, Di Donato A, Cafaro V, Notomista E.2016. Rational Design of a Carrier Protein for the Production of Recombinant Toxic Peptides in Escherichia coli. PLoS One 11:e0146552. US Patent Application Publication No. US / 20210052679 and all references cited therein. 45120619

Claims

Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION WHAT IS CLAIMED IS:

1. A probiotic for the prevention or treatment of a condition caused by H. pylori in the gastrointestinal tract of a subject, comprising: a probiotic Lactococcus bacterium, wherein the probiotic Lactococcus bacterium comprises a DNA construct expressing a guided antimicrobial peptide in the gastrointestinal tract of the subject, wherein the guided antimicrobial peptide comprises an antimicrobial peptide fused to a guide peptide that binds to H. pylori, wherein the guide peptide has a sequence comprising SEQ ID NO:1, SEQ ID NO:2, or SEQ ID NO:3, and wherein the guided antimicrobial peptide kills H. pylori in the gastrointestinal tract of the subject.

2. The probiotic of claim 1, wherein the probiotic Lactococcus bacterium is Lactococcus lactis.

3. The probiotic of claim 1, wherein the antimicrobial peptide is plectasin, eurocin, pexiganan, laterosporulin, alyteserin, or cathelin-related anti-microbial peptide.

4. The probiotic of claim 1, wherein the guided antimicrobial peptide further comprises a linker peptide between the guide peptide and the antimicrobial peptide.

5. The probiotic of claim 4, wherein the linker peptide has a sequence comprising SEQ ID NO:11 or SEQ ID NO:

12.

6. A probiotic composition for the prevention or treatment of a condition caused by H. pylori in the gastrointestinal tract of a subject, comprising: the probiotic of claim 1; and an acceptable excipient or carrier.

7. The probiotic composition of claim 6, wherein the probiotic Lactococcus bacterium is edible, and wherein the acceptable excipient or carrier is edible.

8. A method for preventing or treating a condition in a patient caused by H. pylori in 45120619Attorney Docket No.: 208614.00463 PCT PATENT APPLICATION the gastrointestinal tract of the subject, comprising: administering a probiotic composition to the subject, wherein the probiotic composition comprises a probiotic Lactococcus bacterium and an acceptable excipient or carrier, wherein the probiotic Lactococcus bacterium comprises a DNA construct expressing a guided antimicrobial peptide in the gastrointestinal tract of the subject, wherein the guided antimicrobial peptide comprises an antimicrobial peptide fused to a guide peptide that binds to H. pylori, and wherein the guide peptide has a sequence comprising SEQ ID NO:1, SEQ ID NO:2, or SEQ ID NO:3; and allowing the guided antimicrobial peptide to bind and kill H. pylori in the gastrointestinal tract of the subject.

9. The method of claim 8, wherein the Lactococcus bacterium comprises Lactococcus lactis.

10. The method of claim 8, wherein the antimicrobial peptide is plectasin, eurocin, pexiganan, laterosporulin, alyteserin, or cathelin-related anti-microbial peptide.

11. The method of claim 8, wherein the guided antimicrobial peptide further comprises a linker peptide between the guide peptide and the antimicrobial peptide.

12. The method of claim 8, wherein the linker peptide has a sequence comprising SEQ ID NO:11 or SEQ ID NO:

12.

13. The method of claim 8, wherein the subject is an animal.

14. The method of claim 8, wherein the subject is a human.

15. The method of claim 8, wherein the probiotic bacterium is edible, and wherein the acceptable excipient or carrier is edible 16. The method of claim 8, wherein the probiotic composition is administered orally. 45120619

Images