TREATMENT OF INFLAMMATORY BOWEL DISEASE

Abstract

A delivery system comprising hydrogen peroxide or a hydrogen peroxide generator system, for use in a method for the treatment or prevention of inflammatory bowel disease in a mammal, is described. The method comprises a step of orally administering to the mammal the delivery system, wherein the delivery system is formulated for oral delivery and release of the hydrogen peroxide or hydrogen peroxide generator system in-situ at a target location in the mammalian gastrointestinal tract. Also described is an oral dosage delivery system comprising hydrogen peroxide or a hydrogen peroxide generator system, in which the delivery system is configured for oral administration to a mammal and release of the hydrogen peroxide or hydrogen peroxide generator system at a target location within the mammalian gastrointestinal tract thereby effecting a localised increase in hydrogen peroxide levels at the target location in the mammalian gut or airways.

Claims

1. A method for the treatment or prevention of inflammatory bowel disease in a mammal, in which the method comprises a step of orally administering to the mammal a delivery system comprising hydrogen peroxide or a hydrogen peroxide generator system, wherein the delivery system is formulated for oral delivery and release of the hydrogen peroxide or hydrogen peroxide generator system in-situ at a target location in the mammalian gastrointestinal tract.

2. A method according to claim 1, in which the delivery system is formulated for release of the hydrogen peroxide or the hydrogen peroxide generator system in-situ in the colon.

3. A method according to claim 1, in which the hydrogen peroxide generator system comprises an enzyme and substrate configured to react in-situ at a target location in the mammalian gastrointestinal tract to produce hydrogen peroxide.

4. A method according to claim 1, in which the hydrogen peroxide generator system comprises an enzyme and substrate configured to react in-situ at a target location in the mammalian gastrointestinal tract to produce hydrogen peroxide, in which the substrate is glucose and the enzyme is glucose oxidase.

5. A method according to claim 1, in which the hydrogen peroxide generator system is substantially free of water.

6. A method according to claim 3, in which the hydrogen peroxide generator system is configured to keep the enzyme and substrate separate prior to release of the enzyme and substrate at the target location in the gastrointestinal tract.

7. A method according to claim 3 in which the enzyme is provided as a first particulate composition and the substrate is provided as a second particulate composition, optionally in which the first particulate composition and second particulate composition are each, independently, selected from microparticles, microspheres, nanoparticles, and granulated particles.

8. (canceled)

9. A method according to claim 7, in which the first and second particulate compositions are contained within a container configured to release the compositions at a target location within the gastrointestinal tract, in which the container is optionally an enteric coating.

10. (canceled)

11. (canceled)

12. A method according to claim 1, in which the delivery system comprises hydrogen peroxide.

13. A method according to claim 1, in which the delivery system comprises a polymeric matrix configured for slow release of hydrogen peroxide in-situ at the target location in the gastrointestinal tract.

14. A method according to claim 1, in which the hydrogen peroxide generator system comprises an active agent capable of localised stimulation of the mammalian epithelial tissue to produce hydrogen peroxide in-situ at the target location in the gastrointestinal tract, in which the active agent is optionally an agonist that enhances NOX1 or DUOX2 expression and/or activity.

15-17. (canceled)

18. A method according to claim 1, in which the hydrogen peroxide generator system is a bacterium that produces hydrogen peroxide.

19. A method according to claim 1, in which the hydrogen peroxide generator system is a bacterium that produces hydrogen peroxide, and in which the bacterium is genetically engineered to overexpress hydrogen peroxide.

20. A method according to claim 1, in which the delivery system is configured to release or generate 5 nmol to 20 mol hydrogen peroxide in-situ at the target location in the gastrointestinal tract.

21. A method according to claim 1, in which the inflammatory bowel disease is ulcerative colitis or Crohn's disease, optionally in which the delivery system is administered during a remission phase of the disease to prevent acute recurrence of the symptoms of the disease.

22-43. (canceled)

Description

BRIEF DESCRIPTION OF THE FIGURES

[0079] FIG. 1. Lactobacilli generate H.sub.2O.sub.2H.sub.2O.sub.2 release by L. rhamnosus, L. murinus and L. reuteri. Catalase (CAT; 25 U/ml) was used for H.sub.2O.sub.2 decomposition; n=3.

[0080] FIG. 2. Downregulation of virulence genes located in the LEE pathogenicity island (ler, escN) in C. rodentium infected, genetically modified mice (cyba .sup.f/f Vilcre) with highly increased abundance of indigenous lactobacilli (e.g. L. reuteri, L. murinus). qPCR analysis of ler and escN expression in C. rodentium isolated from the lumen (I) or mucus (J) of infected mice (6 dpi); n=5.

[0081] FIG. 3. Downregulation of LEE island genes after exposure of C. rodentium to H.sub.2O.sub.2 generated by engineered Cos cells overexpressing constitutively active NADPH oxidase NOX4. qPCR analysis of ler and escN expression in C. rodentium after exposure to Cos-p22 (negative control) and Cos-NOX4 cells (4 h, DMEM); n=4. Cells and bacteria were separated by a filter.

[0082] FIG. 4. Downregulation of ler in C. rodentium exposed to H.sub.2O.sub.2C. rodentium in DMEM was treated with a single bolus of 1 mM H.sub.2O.sub.2 and analysed by qPCR for ler expression 3 h later.

[0083] FIG. 5. H.sub.2O.sub.2 generation of Lactobacillus johnsonii wildtype (NCC533) and isogenic ROS source deletion mutant (NCC9359). H.sub.2O.sub.2 release by L. johnsonii WT NCC533 and isogenic NADH-dependent flavin reductase deletion mutant NCC9359; n=3.

[0084] FIG. 6. H.sub.2O.sub.2-producing L. johnsonii NCC533, but not the mutant strain (NCC9359) downregulates the C. rodentium ler gene. qPCR analysis of ler expression in C. rodentium cultured in NCC533- or NCC9359-derived culture supernatant for 3 h; n=5.

[0085] FIG. 7. Association of wildtype C57Bl/6 mice with NCC533 (H.sub.2O.sub.2 producing), but not with NCC9359 downregulates C. rodentium virulence genes. qPCR analysis of ler and escN expression in C. rodentium isolated from the lumen of infected WT3 mice at 6 dpi. Mice were pretreated and treated with NCC533 or NCC9359 during infection; n=4-5.

[0086] FIG. 8. H.sub.2O.sub.2 production by L. johnsonii is required to prevent host epithelial tissue damage during C. rodentium infection. C57Bl/6 mice were treated using daily oral gavage with NCC533 or NCC9359 prior and during infection with C. rodentium. Immunohistochemistry images (H&E staining) of representative colon sections, 6 dpi, scale bar 100 m.

[0087] FIG. 9. Downregulation of LEE island genes in the human pathogen EPEC by exposure to H.sub.2O.sub.2 generated by Cos-NOX4 cells. Exposure of EPEC to H.sub.2O.sub.2 released from Cos-NOX4 cells (but not from Cos-p22 cells; description of cells in Cell Signal 18:69-82, 2006; J Biol Chem 283:35273-82, 2008; Mol Cell Biol 30:961-75, 2010) for 3 h followed by qPCR analysis for ler and escN expression in EPEC. Cells and bacteria were separated by a filter.

[0088] FIG. 10. H.sub.2O.sub.2 generated by colon cells, human biopsies and in the rabbit loop model during pathogen exposure downregulates bacterial phosphotyrosine signalling, which acts as virulence modifier. Anti-phosphotyrosine immunoblots of lysates derived from extracellular bacteria after co-culture (3 h) with HCT-8 colon cells (left panel) or human biopsies (middle panel), or at indicated time points after injection into rabbit ileal loops (right panel).

[0089] FIG. 11. Cos-NOX4 cell-produced H.sub.2O.sub.2 leads to a marked decrease in bacterial phosphotyrosine content. Anti-phosphotyrosine immunoblot of lysates derived from L. monocytogenes after exposure to Cos-NOX4 cells or Cos-p22 cells seeded into the lower chamber of a Boyden chamber.

[0090] FIG. 12. Addition of H.sub.2O.sub.2 decreases bacterial phosphotyrosine content when iron is present (iron and 5% oxygen are the normal conditions in the gut). Anti-phosphotyrosine immunoblots of bacteria cultured in 5% or 21% O.sub.2 (left panel) or in low or high iron conditions at 5% O.sub.2 (right panel) prior to 0.7 mM H.sub.2O.sub.2 addition (3 h).

[0091] FIG. 13. L. monocytogenes adherence and invasion of colon cells is reduced after prior exposure of bacteria to H.sub.2O.sub.2. Bacteria were pretreated with 0.7 mM H.sub.2O.sub.2 for 3 h (bacteria remained fully viable after treatment). Adherence and invasion to HCT-8 cells was analysed as described in Cell Host Microbe 12:47-59, 2012. Error bars represent SEM and asterisks indicate significance (** p<0.01 and * p<0.05).

[0092] FIG. 14. Reduced biofilm formation by S. pneumoniae after prior exposure to H.sub.2O.sub.2. Bacteria were pretreated with 0.7 mM H.sub.2O.sub.2 for 3 h (bacteria remained fully viable after treatment) before biofilm formation was assessed. 3D measurement of S. pneumoniae biofilm in the presence (red) or absence (black) of H.sub.2O.sub.2 is shown (mean fluorescent values for images in a 13.5 m thick stack). Representative confocal image of S. pneumoniae biofilm (green) used for quantification. Scale bar 10 m.

[0093] FIG. 15: H.sub.2O.sub.2 accelerates recovery in murine colitis. Mice were treated with 110.sup.9 bacteria by daily oral gavage 5 days before, during colitis (3% DSS in water) and until the end of the experiment (day 16). On day 7 of DSS treatment, DSS was replaced with water to enable the healing process. Lactobacillus johnsonii wildtype (NCC533, H.sub.2O.sub.2 production) enhances tissue restitution and reduces inflammation, while an isogenic L. johnsonii mutant (NCC9360, nfr, deletion of the H.sub.2O.sub.2-generating enzyme) is not effective. Daily administration of low nanomolar H.sub.2O.sub.2 accelerates recovery after insult (adisease index, NCC533 triangle, cbody weight), downregulates inflammation (b, colon length as indicator of inflammation), and supports rapid tissue restitution (dH&E staining of colon day 11). One-way Anova with Tukey post-hoc test; **p=0.01, ***p=0.001.

[0094] FIG. 16: Deficiency of H.sub.2O.sub.2 generating enzymes (NOX) in the colon epithelium of mice leads to reduced or even abolished mucus when compared to normal wildtype mice (white line). In wildtype mice (and humans) the mucus layer provides an impenetrable barrier and separates the intestinal bacteria (microbiota) from the host epithelium. Loss of this protective mucus layer in H.sub.2O.sub.2-deficient mice leads to bacterial colonization in crypts (arrow). Adding H.sub.2O.sub.2 can restore mucus secretion (Birchenough G. et al., Science 352:1535-1542, 2017).

DETAILED DESCRIPTION OF THE INVENTION

Mice:

[0095] Cyba.sup.flox/flox mice were bred to B6.SJL-Tg (Vil-cre) 997Gum/J mice (Jackson Laboratories) to generate mice with a targeted deletion of p22.sup.phox in the epithelium of the small intestine and colon (p22IEC). Mice were infected by oral gavage with 0.3 ml of an overnight culture of Luria Bertani broth containing 510.sup.9 Citrobacter rodentium and analyzed at day 6 post infection. L. johnsonii NCC533/NCC9359 pretreatment and treatment: L. johnsonii NCC533 and NCC9359 strains were administered at a concentration of 10.sup.9 CFU to WT mice via oral gavage starting 3 days prior C. rodentium inoculation and daily during C. rodentium infection. For histopathology Carnoy-fixed distal colons were embedded in paraffin and 5 m sections were stained with hematoxylin and eosin (H&E).

Cell Experiments:

[0096] Cos-NOX4 and Cos-p22 cells in DMEM, 5% FBS (Cell Signal 18:69-82, 2006; J Biol Chem 283:35273-82, 2008; Mol Cell Biol 30:961-75, 2010) were seeded into the lower chamber of a Boyden chamber (pore width 0.4 m) 48 h before the assay. Lactobacilli grown in DMEM were suspended in media onto the upper chamber. Co-culture was initiated for 3-4 h in DMEM, 5% FBS media followed by harvesting of bacteria on the filter for RNA isolation.

Analysis of C. rodentium Virulence-Associated Genes:

[0097] Luminal content and adherent, mucus associated scrapings were obtained from infected mice (6 dpi) and used for RNA isolation. Total RNA was isolated using RNeasy Mini Kit and reverse transcribed using the High Capacity cDNA Reverse Transcription Kit. Quantitative real-time PCR for ler and escN expression was performed using a SYBR Green Master mix and normalized to the expression of gfp expressed in C. rodentium. Relative expression was determined as fold expression in comparison to LB grown C. rodentium.

In Vitro Analysis of Lactobacilli:

[0098] To obtain a cell free supernatant (CFS) the culture of lactobacilli (24 h) was centrifuged at 10000 g for 30 min (4 C.); the supernatant was collected and passed through a sterile 0.22 m filter unit Millex GS. An exponential culture of C. rodentium (10.sup.4 CFU, 500 l) was incubated with CFS (500 l) at 37 C. for 4 h. In other experiments lactobacilli were grown o/n in MRS to a total density of 10.sup.9 CFU/ml and resuspended in PBS. After 30 min incubation, the lactobacilli cultures were centrifuged at 14000 rpm (5 min) and 10 ul of the supernatant was used to measure H.sub.2O.sub.2 by Amplex Red assay.

Co-Culture of Intestinal Bacteria with Cells in Normal or Iron-Modified Conditions

[0099] Bacteria in RPMI 1640 containing 3% FBS (OD.sub.600=0.4; 10.sup.7 bacteria, 1 ml) were incubated with HCT-8 cells at 37 C. in microaerophilic conditions with MOI 50. DPI pretreatment of HCT-8 cells was for 20 min (25 M). Non-adherent bacteria were removed by centrifugation of media (3300 g, 5 min), and were used for immunoblotting. Viable counts were performed for inocula to ensure that comparable numbers of live bacteria were present for each bacterial strain. Exposure of bacteria to H.sub.2O.sub.2 released by Cos cells stable expressing the NOX4-p22.sup.phox complex or as negative control Cos cells expressing only p22.sup.phox was performed using Boyden chambers. Cells in DMEM, 10% FBS medium were seeded into the bottom chamber 24 h before start of the experiment and then moved to microaerophilic conditions for 3 h. Bacteria grown in microaerophilic conditions were resuspended in 5% O.sub.2 conditioned DMEM, 3-10% FBS medium, placed on top of the filter (3 m pore size) and incubated for 3 h. Bacteria were harvested from the filter for analysis.

Quantification of Protein Phosphotyrosine Levels in Modified Iron Conditions

[0100] C. jejuni 81-176, L. monocytogenes EGDe and K. pneumoniae were grown microaerophilic to mid-log phase in minimal essential media. Bacteria were collected by centrifugation (5000 rpm, 10 min) and diluted to OD.sub.600=0.2 in MEM. Iron (II) sulfate (40 M) was added as indicated and cultures were grown microaerophilic until OD.sub.600=0.5 was reached. To C. jejuni 81-176 cultures 5 mM H.sub.2O.sub.2 was added and the growth continued microaerophilic for 8 h. L. monocytogenes and K. pneumoniae cultures were exposed to 0.7 mM H.sub.2O.sub.2 for 3 h in microaerophilic conditions. Bacteria were then diluted to OD.sub.600=0.2, collected by centrifugation, washed twice with 25 mM Tris-HCl, resuspended in 30 l Laemmli buffer and heated for 5 min at 95 C. Boiled samples were loaded on 10% SDS-PAGE gels, separated by electrophoresis and immunoblotted. The control samples followed an identical protocol without supplemental iron or added H.sub.2O.sub.2.

Various Methods

[0101] Invasion and adhesion assays were performed as previously reported (Cell Host Microbe 12:47-59, 2012). Bacterial cell viability was assessed using the XTT (2,3-Bis-(2-Methoxy-4-Nitro-5-Sulfophenyl)-2H-Tetrazolium-5-Carboxanilide) assay. For in vitro experiments bacteria were exposed at time zero once to 0.7 mM H.sub.2O.sub.2, except for C. jejuni (5 mM), and analyzed after indicated time periods (up to 3 h, except C. jejuni up to 6 h).

Biofilm Assay

[0102] Biofilm was analyzed using SYTO9. Cultured bacteria were grown in microaerophilic conditions overnight. The bacterial culture was then adjusted to OD.sub.600=0.1 and 1 ml of this solution was added to an optical grade plastic bottom -plate 24 well. Bacteria were incubated for an additional 72 h in microaerophilic conditions. Bacteria were washed gently with autoclaved water to remove planktonic cells and fixed with 3% paraformaldehyde in PBS for 1 h. Images were then analyzed for total intensity of SYTO9 fluorescence over the imaged volumes using Fiji. Briefly, after staining with SYTO9, 50 independent 13.2 m stacks of 60 images at a 0.22 m Z resolution were imaged for all conditions. The mean fluorescence of the different slices in the stacks was measured and a standard deviation was computed from the aggregated data per slice. The mean fluorescence with the corresponding standard deviations per slice was then plotted as a function of depth with the start of the stack set to zero.

Ex Vivo Biopsy Analysis

[0103] Polarized In Vitro Organ Culture (pIVOC): pIVOC experiments of colon biopsies were performed as described (Cell Host Microbe 12:47-59, 2012). Bacterial infections were conducted using microaerophilic preconditioned media or buffers. L. monocytogenes was added at a final OD.sub.600=0.2 to the apical side of biopsies and H.sub.2O.sub.2 release was measured. Controls were either non-infected biopsies or biopsies pretreated with 20 M DPI for 20 min before addition of C. jejuni or L. monocytogenes to the top chamber.

Ligated Rabbit Ileal Loops

[0104] Female rabbits (Chinchilla breed) were starved for 2 h prior to infection. Anesthesia was performed iv with ketamine (35 mg/kg) and xylazine (5 mg/kg). The incision line was injected subcutaneously with 2 ml xylazine 1%. After laparotomy four ileal loops (5 cm in length) were isolated and ligated. The loops were injected with either PBS, C. jejuni 81-176 or L. monocytogenes EDGe using 1 ml of culture at OD.sub.600=0.3 in PBS (pH7.4). For each microorganism 3 rabbits were used. After closure of the abdomen rabbits were placed in cages for 180-360 minutes (C. jejuni) or 90-180 minutes (L. monocytogenes). Rabbits were sacrificed by intravenous injection of sodium pentobarbital (120 mg/kg). Fluid accumulated in each loop was collected separately and spun at 1000 rpm for 5 min to remove debris, followed by centrifugation at 5000 rpm to collect bacteria.

In-Vivo Murine Colitis Model (Administration of H.sub.2O.sub.2-Producing Bacteria)

[0105] Wildtype mice were treated with 110.sup.9 bacteria by daily oral gavage 5 days before, during colitis (3% DSS in water) and until the end of the experiment (day 16). On day 7 of DSS treatment, DSS was replaced with water to enable the healing process. Lactobacillus johnsonii wildtype (NCC533, H.sub.2O.sub.2 production) enhances tissue restitution and reduces inflammation, while an isogenic L. johnsonii mutant (NCC9360, nfr, deletion of the H.sub.2O.sub.2-generating enzyme) is not effective. Daily administration of low nanomolar H.sub.2O.sub.2 accelerates recovery after insult (adisease index, NCC533 triangle, cbody weight), downregulates inflammation (b, colon length as indicator of inflammation), and supports rapid tissue restitution (dH&E staining of colon day 11). One-way Anova with Tukey post-hoc test; **p=0.01, ***p=0.001.

In-Vivo Murine Colitis Model (Administration of H.sub.2O.sub.2-Producing Capsules)

[0106] The source for nanomolar to low micromolar release of H.sub.2O.sub.2 in the intestine is controlled release capsules containing the enzyme glucose oxidase and its substrate glucose, albeit other combinations or sources can be used. Capsules containing a source for generating H.sub.2O.sub.2 in the intestine or control capsules (e.g. empty capsules, heat-inactivated enzyme, absence of substrate) are given once or multiple times daily by oral gavage at the onset of acute disease (day 0), at the height of disease (time varies dependent on the model) or when the healing process begins (time varies). Mice are monitored daily for weight and the extent of disease is scored according to an animal welfare and colitis adjusted scale. Animals are sacrificed at various time points to assess inflammation and tissue injury/tissue restitution. These parameters are determined by procedures including colon length measurement, immunohistochemistry with stains (e.g. H&E, Mason's), immunofluorescence using antibodies (e.g. anti-mucin 2) or staining with proliferation markers (e.g. BrdU). Markers of inflammation also include recruitment of immune cells, chemokine and cytokine levels in the mucosa and blood, measurement of mucus density, rheology and chemical composition, as well as microbiome and metabolome composition and quantitative parameters (e.g. cytokines, chemokines). Pharmacokinetic in vitro, ex vivo and in vivo studies are performed according to reported best practice, and safety and efficacy studies (up to 10 weeks) with and without prior insult to determine putative physiological changes due to prolonged H.sub.2O.sub.2 delivery and the ability to resist an acute colitis insult.