NOD2-DEPENDENT PATHWAY OF CYTOPROTECTION OF STEM CELLS

Abstract

The present invention is directed to an agonist of NOD2 for use in therapy for increasing the autonomous capacity of survival of vertebrate adult stem cells, without loss of their capacity to multiply and differentiate, and preferably the capacity of survival of intestinal stem cells, especially in response to a stress. The invention also concerns the use of an agonist of Nod2 for increasing in vitro or ex vivo the autonomous capacity of survival, without loss of multiplication and differentiation capacity of mammalian adult stem cell. The invention also discloses different media and support for mammalian adult stem cells. The invention also concerns an in vitro screening process for identifying molecules capable increasing, in response to a stress, the autonomous capacity of survival, without loss of multiplication and differentiation capacity of mammalian adult stem cells.

Claims

1-16. (canceled)

17. A method of promoting survival of mammalian adult stem cells, comprising contacting mammalian adult stem cells with an agonist of NOD2, wherein the mammalian adult stem cells are not derived from human umbilical cord blood, wherein the rate of survival of the mammalian adult stem cells is increased, and wherein the capacity of the mammalian adult stem cells to multiply and differentiate is maintained.

18. The method of claim 17, wherein the mammalian adult stem cells are intestinal stem cells, mesenchymal stem cells, hematopoietic stem cells or skin stem cells.

19. The method of claim 17, wherein the mammalian adult stem cells are contacted while under stress conditions.

20. The method of claim 19, wherein the stress conditions comprise at least one of chemical stress, oxidative stress, irradiation, pharmacological stress, physical stress, physiopathological stress and infectious stress.

21. The method of claim 19, wherein the stress conditions are induced by chemotherapy, by radiotherapy, by infection, by tissue injury, or by tissue dissection.

22. The method of claim 19, wherein the stress conditions are induced by manipulation of the mammalian adult stem cells during harvest or implantation of the cells.

23. The method of claim 22, wherein the mammalian adult stem cells are intestinal stem cells and the stress conditions are induced by transplantation of the intestinal stem cells for treating intestinal atrophy, crypt dysfunction or inflammatory bowel diseases.

24. The method of claim 22, wherein the mammalian adult stem cells are skin stem cells and the stress conditions are induced by transplantation of the skin stem cells for accelerating repair of skin following an epithelial wound.

25. The method of claim 22, wherein the mammalian adult stem cells are hematopoietic stem cells and the stress conditions are induced by bone marrow transplantation.

26. The method of claim 17, further comprising transplanting the mammalian adult stem cells into a mammalian host.

27. The method of claim 17, wherein the mammalian adult stem cells are contacted ex vivo.

28. The method of claim 17, wherein the mammalian adult stem cells are contacted in vitro.

29. The method of claim 17, wherein the agonist of NOD2 is administered to a mammal to contact the mammalian adult stem cells, thereby increasing the in situ rate of survival of the mammalian adult stem cells and maintaining the capacity of the mammalian adult stem cells to multiply and differentiate.

30. The method of claim 17, wherein the mammalian adult stem cells are contacted during transplantation.

31. The method of claim 17, wherein the agonist of NOD2 is amyl dipeptide (N-acetylmuramyl-L-Alanyl-D-Isoglutamine).

32. The method of claim 17, wherein the agonist of NOD2 is a chemical derivative of Muramyl dipeptide (N-acetylmuramyl-L-Alanyl-D-Isoglutamine).

33. The method of claim 32, wherein the chemical derivative of Muramyl dipeptide (N-acetylmuramyl-L-Alanyl-D-Isoglutamine) is selected from N-Glycolyl-Muramyl dipeptide, a 6-O-acyl derivative of Muramyl dipeptide, muramyldipeptide with a C18 fatty acid chain, MurNAc-Ala-D-isoGln-Lys, MurNAc-Ala-D-isoGln-L-Lys(D-Asn) and murabutide (N-Acetyl-muramyl-L-Alanyl-D-Glutam in-n-butyl-ester).

Description

LEGEND OF FIGURES

[0106] FIG. 1: Growth curve of organoids.

[0107] FIG. 1A: FIG. 1A is the picture of an organoid, grown from intestinal crypt. This is a structure delineated by a definite monolayer of cells.

[0108] FIG. 1B: FIG. 1B represents difference between a viable and a dying organoid. On the left a living organoid is shown, i.e. a structure with a monolayer of cells surrounding a lumen, with new crypts (protusions). On the right a dead organoid is shown, that appears as a clump of dead cells. Bar 100 μm.

[0109] FIG. 1C: FIG. 1C represents a typical growth curve of the number of organoids upon stimulation by MDP, different other MAMPs, and without stimulation (control). The curves depicted are the results from one experiment, but are representative of the typical growth curve.

[0110] FIG. 2: Effects of MAMPs on organoids formation: MDP induced higher yield of organoids through Nod2 receptor cecognition.

[0111] FIG. 2 illustrates the fold change in number of organoids, formed from extracted crypts, depending on the MAMPs applied for stimulation. Crypts were stimulated with soluble sonicated peptidoglycan (PGN), muramyl-dipeptide (MDP), muramyl-tetrapeptide (Tetra-dap), Escherichia coli lipopolysaccharide (LPS) (each at 10 μg/ml), 10 ng/ml flagellin (Fla), 500 ng/ml synthetic lipoprotein (Pam3CSK), or 1 μM unmethylated CpG dinucleotides (CpG). The average number of non-stimulated organoids (control) is around 50, and the fold change value of the control is fixed to 1. The fold change upon the various MAMPs applied is thus by reference to non-stimulated organoids (Ctrl).

[0112] The results presented in FIG. 2 are the average of 5 different independent experiments. The fold change is measured at day 4.

[0113] ***P<0.001, **P<0.01 Mann-Whitney.

[0114] As can be seen from this figure, PGN (peptidoglycan) and MDP induce the formation of higher amount of organoids.

[0115] FIG. 3: proliferation assays.

[0116] Cell proliferation was analyzed by cytometry after 4 days of culture.

[0117] FIG. 3A illustrates the rate of Ki-67 expression upon different stimulation conditions, Ki-67 being a marker for proliferative cells.

[0118] FIG. 3B illustrates the rate of EdU incorporation. After 4 days of culture, 10 μM EdU were added for 2 hours and then the organoids were recovered to monitor the EdU incorporation. Representation profiles of MDP-treated (black) and non-treated organoids (grey) are shown. No differences in the rate of proliferation between treated and un-treated organoids can be seen.

[0119] FIG. 4: Organoids from Nod1 and Nod2 KO mice.

[0120] FIG. 4 illustrates the fold change in number of organoids, formed from extracted crypts, from mice knocked out for Nod1 or Nod2 genes, upon different MAMPs stimulation conditions. By definition, the fold change of the untreated organoids is set to 1, and the fold changes upon the various stimulation conditions applied are by reference to the non-stimulated organoids.

[0121] FIG. 5: Nod2 gene expression

[0122] FIG. 5 illustrates the Nod2 gene expression in Lgr5-expressing stem cells, in Paneth cells and in the whole crypts. The level of expression of Nod2 gene in whole crypts is arbitrarily fixed to “1”.

[0123] As can be inferred from the results of FIG. 5, the level of nod2 mRNA is 5 times more in stem cells than in Paneth cells.

[0124] FIG. 6: Markers of stem cells and of Paneth cells.

[0125] The relative level of expression of different markers is illustrated in FIG. 6. As for FIG. 5, the level of expression for the different markers is arbitrarily fixed to “1” at the level of the whole crypts.

[0126] As can be seen from this figure, Igr5, ascl2 and olfm4 which are specific markers of intestinal stem cells are preferentially expressed in intestinal stem cells extracted from crypts, whereas CD24, lyzP and Defcr-rs1 are mainly expressed in Paneth cells, thus confirming the nature of these cells.

[0127] FIG. 7: Markers of stem cells and of Paneth cells in WT and Nod2 KO cells.

[0128] FIG. 7 illustrates results similar to those presented in FIG. 6, for crypts extracted from WT mice and from mice KO for Nod2 gene.

[0129] FIG. 8: Single cells stimulation.

[0130] FIG. 8 depicts the average number of organoids formed per well, depending on the cell types seeded in the well, without stimulation, or under stimulation with MDP. 500 cells were seeded per well in a 96-wells-plate.

[0131] Stem cells are either wild type or KO for the nod2 gene; they are seeded in all wells.

[0132] Paneth cells are also either wild type or KO for the nod2 gene; They are present in the 3.sup.rd to 6.sup.th series of experiments and absent in the 1.sup.st, 2.sup.nd and 7.sup.th series of experiments.

[0133] In the last (7.sup.th) series of experiments, Wnt is added.

[0134] FIG. 9: Inducing stress to stem cells.

[0135] FIG. 9 illustrates the fold change in number of organoids, formed from extracted crypts, upon stimulation with MDP or un-treated, with or without application of a specific stress, either Doxorubicin or H.sub.2O.sub.2. By definition, the fold change of the untreated organoids is set to 1 in the absence of any additional stress, and the fold changes upon the various stimulation or stress conditions applied are by reference to the non-stimulated non-stressed organoids.

[0136] Doxorubicin (1 μM) or H.sub.2O.sub.2 (200 μM) are added during the embedding of the crypts in the matrigel. Doxorubicin (DOX) belongs to the anthracycline class of chemotherapeutic agents and works by intercalating DNA. Doxorubicin has been shown to preferentially attack the cells at the bottom of the crypt.

[0137] As can be seen from FIG. 9, MDP is able to protect stem cells.

[0138] FIG. 10: LDH release.

[0139] FIG. 10 illustrates the percentage of LDH release from crypts treated by different MAMPs, by reference to untreated crypts (corresponding to the control, value set to 100).

[0140] As can be deduced from the results of FIG. 10, the stimulation with PGN or MDP results in a higher viability of the crypts, by reference to untreated crypts.

[0141] FIG. 11: Caspase-3 activation.

[0142] FIG. 11 depicts the percentage of caspase-3 positive crypts in different conditions. The results are obtained after 6 hours.

[0143] Line 1: WT, non-treated mice. Line 2: Nod2 KO mice, non-treated. Line 3: WT mice, treated with doxorubicin during 6 hours. Line 4: Nod2 KO mice, treated with doxorubicin during 6 hours. It can be observed that the absence of Nod2 induced more death in the crypts.

[0144] FIG. 12: Organoids from Doxorubicin-treated mice.

[0145] FIG. 12 illustrates the fold change in number of organoids, formed from extracted crypts from either WT or Nod2 KO mice, 72 hours after treatment with doxorubin, upon stimulation with MDP or un-treated.

[0146] FIG. 13: Proliferation index.

[0147] FIG. 13 illustrates the percentage of EdU-positive cells per crypt, in wild-type or Nod2 KO mice, after 2 hours of EdU (5-ethynyl-2-deoxyuridine) treatment.

[0148] In WT mice, the crypts are prone to regenerate.

[0149] By definition, the fold change of the untreated organoids, extracted from WT mice, is set to 1, and the fold changes upon the various stimulation conditions are by reference to the non-stimulated WT organoids.

[0150] As can be seen from this figure, upon stress (namely doxorubicin), the stem cells are actively reactivated by the MDP.

[0151] FIG.14: Stimulated and non-stimulated organoids present the same maximal size. On the fourth day of culture, MDP-treated or control organoids, were imaged. The area of the organoids was measured using Axiovision software, and the values were plotted. In the figure a box and whisker-plot shows the values obtained from one representative experiment.

EXPERIMENTAL SECTION

EXAMPLE 1

Material and Methods

[0152] Mice

[0153] B6.129P2-Lgr5.sup.tm1(cre/ESR1)Cle/mice (Lgr5-EGFP) (Barker et al., 2007) were purchased from The Jackson Laboratories. Card4/Nod1 (Nod1 KO) mice were generated by Millenium pharmaceuticals Boston. Card15/Nod2 deficient C57BL/6J (Nod2 KO) (Barreau et al., 2007) were provided by J.-P. Hugot (Hôpital Robert Debré, Paris, France). All mice were kept in specific pathogen-free conditions. When indicated the mice were injected intraperitoneally with 200 μg of EdU (Invitrogen) 2 hrs prior to sacrifice or with 10 mg/kg doxorubicin hydrochloride (Sigma) and sacrificed at indicated time points.

[0154] Crypts Isolation and Organoids Formation

[0155] Intestinal crypts were obtained following a protocol already described (Sato et al., 2009). After isolation small intestines were isolated, flushed with cold PBS followed by 0.3% bleach and again PBS, opened longitudinally and the villi were removed. The tissues were cutted in small pieces, washed in 5 mM PBS/EDTA for 5 min, and subsequently incubated in fresh 5 mM PBS/EDTA for 30 min on ice. After severals vigourous shakings the crypts, released in the supernatant, were passed through a cell strainer 70 μm (BD Falcon), spun down at 800 rpm, resuspended in DMEM and counted. The volume corresponding to 500 crypts was distributed in two tubes per condition, and spun down again. The pellet of crypts was or not incubated with the following microbe-associated molecular patterns (MAMPs) for 10 min at room temperature: 10 μg/ml Soluble sonicated peptidoglycan from E. coli K12 (PGN), 10 μg/ml Muramyl-dipeptide (MDP), 10 μg/ml MDP rhodamine, 10 μg/ml MDP control (MDP-ctrl), 10 μg/ml Lipopolysaccharide (LPS), 500 ng/ml Lipoprotein (Pam3), 10 ng/ml Flagellin (Fla) (all purchased from InvivoGen), 10 μg/ml MurNAc-Tetra(DAP) (Tetra-dap) (kindly provided by Dominique Mengin-Lecreulx). The crypts were embedded in 50 μl of growth factor reduced matrigel (BD Biosciences) in a 24-well plates, incubate at 37° C. 20 min and overload with 500 μl of crypts medium (CM) as described previously (Sato et al 2009). Briefly, Advanced DMEM/F12 was supplemented with 100 U/ml penicillin/streptomycin, 10 mM Hepes, 1× N2, 1× B27 (all from Gibco), 50 ng/ml EGF (Peprotech), 100 ng/ml Noggin (Peprotech), 500 ng/ml R-spondinl (R&D). The medium was exchanged every 4 days. To induce toxic stress to the organoids, 1 μM Doxorubicin hydrochloride (from Sigma) or 200 μM H.sub.2O.sub.2 were added to the medium upon the embedding of the crypts into the matrigel.

[0156] The number of organoids was evaluating with an inverted microscope equipped with thermostatic chamber (temperature and CO.sub.2).

[0157] Sorting and Culture of Single Cells

[0158] Isolated crypts from Lgr5-EGFP mice or Nod2KO mice were incubated in HBSS w/o Ca.sup.+2 and Mg.sup.2+ supplemented with 0.3 U/ml Dispase (BD Biosciences), 0.8 U/ml DNase (Sigma) and 10 μM Y-27632 (Sigma) for 30 min at 37° C.

[0159] Dissociated cells were washed with 1% PBS/BSA and stained with CD24-APC antibody (clone M1/69 BioLegend) and EpCam Pe-Cy7 (clone G8.8 BioLegend) for 20 min at 4° C., resuspended in crypts medium supplemented with 1 μM N-acetyl-L-cystein, 10 μM Jag-1 (Anaspec) and 10 μM Y-27632 (single cells medium-SCM) and filtered with a 20 μm mesh and analyzed with MoFlo legacy (Beckman Coulter).

[0160] Singlets of viable epithelial cells were gated by negative staining for propidium iodide and positive staining for EpCam. The stem cells and Paneth cells were isolated respectively as GFP.sup.hi+ population, and GFP.sup.−CD24.sup.hi+SSC.sup.hi population.

[0161] Sorted cells were collected or in SCM for culturing experiments or in RNAprotect Cell Reagent (Qiagen) for the RNA extraction.

[0162] Stem cells and Paneth cells derived from Lgr5-EGFP or Nod2KO mice were cultured alone or in different combinations. 500 cells from each condition were re-suspended in 100 μl of single cells medium with or without MAMPs and, when indicated, with 100 ng/ml wnt3 (R&D), seeded in 96-wells round-bottom plates and incubated on ice for 15 min. The plate was spun at 300 g for 5 min and 10 μl of matrigel was added in each well. The number of organoids was counting after 1 week.

[0163] Real Time PCR

[0164] Total RNA was extracted with the RNeasy Micro Kit (Qiagen) and the cDNA was made with SuperScript II Reverse Transcriptase (Invitrogen) and oligo(dT)12-18 primer (Invitrogen) as recommended by the suppliers. The following primers (purchased from Invitrogen) were used:

TABLE-US-00001 Ascl2, (SEQ ID No: 3) 5′-AAGCACACCTTGACTGGTACG-3′ and (SEQ ID No 4) 5′-AAGTGGACGTTTGCACCTTCA-3′; b2m, (SEQ ID No 5) 5′-TCAGTCGTCAGCATGGCTCGC-3′ and (SEQ ID No 6) 5′-CTCCGGTGGGTGGCGTGAGTAT-3′; CD24 (SEQ ID No 7) 5′-CGAGCTTAGCAGATCTCCACT-3′ and (SEQ ID No 8) 5′-GGATTTGGGGAAGCAGAAAT-3′; defcr-rs1, (SEQ ID No 9) 5′-AAGAGACTAAAACTGAGGAGCAGC-3′ and (SEQ ID No 10) 5′-CGACAGCAGAGCGTGTA-3′; Lgr5, (SEQ ID No 11) 5′-GACAATGCTCTCACAGAC-3′ and (SEQ ID No 12) 5′-GGAGTGGATTCTATTATTATGG-3′; LyzP, (SEQ ID No 13) 5′-GAGACCGAAGCACCGACTATG-3′ and (SEQ ID No 14) 5′-CGGTTTTGACATTGTGTTCGC-3′; nod2, (SEQ ID No 15) 5′-GGAGTGGAACAGCTGCGACCG-3′ and (SEQ ID No 16) 5′-GCACACTCAACCAGCGTGCG-3′; OLfm4, (SEQ ID No 17) 5′-TGGGCAGAAGGTGGGACTGTGT-3′ and (SEQ ID No 18) 5′-CGGGAAAGGCGGTATCCGGC-3′.

[0165] Reactions were run on ABI 7900HT (Applied Biosystems) using Power SYBR Green mix (Applied Biosystems) according with the manufactures instructions. B2M RNA was used as an internal control, and ΔΔCT (cycle threshold) values were calculated.

[0166] Immunohistochemistry

[0167] The tissue was processed as already described (Escobar et al., 2011). The small intestines were flushed with PBS and everted on 3-mm OD wooden skewers before fixation in 4% paraformaldehyde (Electron Microscopy Sciences) for 2 h. Intestines were released from the skewers by a longitudinal incision and rolled up using wooden toothpicks, incubated for 2 h in 15% sucrose and overnight at 4° C. in 30% sucrose. Rolls were then embedded in OCT compound (VWR) frozen in isopentane, cooled with dry ice, and stored at −80° C. Frozen blocks were cut with a thickness of 7 μm using a CM 3050S cryostat (Leica), and sections were collected on Superfrost plus slides (VWR) and stored at −80° C. Alternatively after the fixation the tissues were embedded in paraffin using standard procedures.

[0168] Paneth cells were stained using a rabbit anti-Lysozyme (1:500, Thermo) and Alexa fluor 568 as secondary antibody. DNA was stained by DAPI (1 μg/ml, Molecular Probes). The sections were mounted with Prolong gold antifade (Invitrogen). Images were acquired with confocal microscopy (Leica, SP5).

[0169] For the caspase-3 staining, tissue sections were deparaffinized in xylene and then rehydrated in a series of alcohols and PBS. Endogenous peroxidase activity was removed by incubating the sections in 3% hydrogen peroxide for 10 min. Antigen retrieval was done by boiling the sections in 10 mM sodium citrate buffer for 20 min. Sections were permeabilized for 10 min with 0.5% triton X-100, blocked with ultra V block (Labvision) 5 min and incubated overnight with a rabbit cleaved Caspase-3 antibody (Cell Signaling Technology, 1:200). Dako EnVision-system HRP anti-rabbit (Dako) was then applied for 40 min and the reaction developed with DAB (Dako). Slides were counterstained with Mayer hematoxylin and mounted with Aquatex (Merck). The sections were imaging with Mirax scanner (Zeiss).

[0170] Cytotoxicity Assay

[0171] Cultures of organoids were analyzed for the release of lactate dehydrogenase (LDH) using Promega CytoTox 96 kit according to provider's instructions. The absorbance was mesaured with the Tecan Sunrise microplate reader and the ratio between the LDH released into the supernatant and the total LDH in cell lysate was measured.

[0172] Proliferation Assay

[0173] The supernatant was removed and the organoids were dissociated using dispase (BD Bioscience). The cells were fixed with 2% PFA, permeabilized with 0.1% triton X-100 and stained with 1:100 dilution of rabbit polyclonal anti Ki-67 antibody (Abcam) for 1 h. Alexa Fluor 647 was used as a secondary antibody. Data were recorded using FACSCalibur flow cytometer (BD) and analyzed with Flowjo software.

[0174] The proliferation of organoids was also evaluated by flow cytometry using the Click-iT® EdU Alexa Fluor® 647 Flow Cytometry Assay Kit (Invitrogen) according to provider's instructions. Briefly, EdU (5-ethynyl-2-deoxyuridine) was added at 10 μM for 2 h. Then the culture supernatant was removed, and the organoids were dissociated by dispase (BD Bioscience). The cells were fixed and permeabilized, and the click-iT reactions were carried out. After the washes, the cells were analyzed with a FACSCalibur flow cytometer (BD) and the data analyzed by Flowjo software.

[0175] For the analysis of proliferation in the tissue, 10 mg/kg EdU were injected IP and the animals were sacrificed after 2 h. The Click-iT® EdU Alexa Fluor® 488 Imaging Kit was used on paraffin slides that were processed according to the manufacturer's instructions.

[0176] Scoring of Proliferative Index and Regenerative Crypts.

[0177] The proliferation index was calculated as a percentage of EdU positive cells over the total number of cells in each crypt. Data from five mice each group were obtained and at least 50 crypts per section were examined for all histological parameters. A regenerative crypt was defined as a crypt containing more than 10 EdU positive cells per crypt. Relative crypt size was determined as crypt width×crypt height.

[0178] Statistics. The descriptive statistical analysis was performed on Graphpad Prism version 5 (Graphpad software Inc., San Diego, Calif.). Results are expressed as mean±SD. Statistical comparisons were performed using the Mann-Whitney test U-test. A P-value <0.05 was considered as significant.

EXAMPLE 2

Results

[0179] Summary:

[0180] The present inventors have discovered a NOD2-dependant pathway of cytoprotection of intestinal stem cells, in the presence of muramyl-dipeptide, the best known NOD2-agonist. This is a totally novel function of the NOD signaling pathway that illuminates the pathogenesis of IBDs (inflammatory bowel diseases) like Crohn disease, and offers a novel approach to maintain stem cells alive in conditions of manipulation.

[0181] The intestinal mucosa surface is continuously exposed to a complex and dynamic community of microorganisms, the microbiota. These microbes establish symbiotic relationships with their hosts, and this interaction plays a crucial role in the maintenance of intestinal epithelial homeostasis. The bacteria present in the lumen release microbial components, namely MAMPs for microbe-associated molecular patterns, which access the intestinal crypts, a sensitive region responsible for epithelial regeneration where intestinal stem cells are located. Using a newly developed in vitro method described by Clevers group (Sato et al., 2009), the inventors cultured intestinal crypts organoids testing several MAMPs. Starting from the first day of culture, they found a higher number of organoids upon stimulation with bacterial peptidoglycan (PGN), especially with MDP, a PGN subunit present in Gram positive and Gram negative bacteria. On the contrary the inventors did not observe any differences with other PGN motifs such as Tetra-dap. This result is due to the fact that Tetradap and MDP are respectively recognized by two different murine receptors, Nod1 and Nod2 (Magalhaes et al., 2005; Girardin et al., 2003). Only Nod2 seems to be involved in the observed phenotype. Indeed comparing the effect of PGN or MDP on crypts extracted from Nod1 or Nod2 deficient mice, the inventors observed the same effect seen in the wt crypts in the Nod1 KO mice. As expected, organoids produced from Nod2 KO mice were not affected by the presence of MDP.

[0182] Using Lgr5-EGFP-ires-creERT2 mice, the inventors were able to sort intestinal stem cell (SC) and cultivate them to produce organoids. When the Paneth cells were added to SC the rate of organoids formation increased. Adding MDP to the medium of single SC or into a co-culture of SC and Paneth cells the inventors always observed an increase in numbers of organoids. It is known from the literature that Nod2 is expressed at the bottom of the intestinal crypts, at the level of Paneth cells. However, the higher survival of organoids was not associated to the expression of Nod2 in Paneth cells. In fact co-culturing Paneth cells isolated from Nod2 KO mice with wt stem cells, the increase in the number of organoids due to the presence of MDP was still present.

[0183] Analyzing the gene expression in those cells, the inventors found that not only the Paneth cells express the nod2 gene but also the Igr5 positive stem cells. Moreover nod2 gene is expressed 5 times more in SC compared to Paneth cells. This explains the direct effect of MDP on stem cells.

[0184] Organoids produced upon stimulation with MDP had the same rate of proliferation of the non-stimulated ones. Therefore MDP has a cyto-protective effect on isolated intestinal crypts. In fact, crypts stimulated with MDP release less LDH compared to untreated organoids.

[0185] MDP shed by the luminal microbiota can thus interact with the Nod2 receptor expressed by Paneth cell and stem cells. The Paneth cells could be induced to release factors like EGF, TGF-a, Wnt3 required to support the stem cells. On the other hand MDP interacting directly with stem cells increases their survival.

[0186] Detailed Results:

[0187] Recently a culture system for intestinal crypts has been described (Sato et al., 2009). Embedding isolated crypts in matrigel and adding growth factors, leads to three-dimensional structures, called organoids. Organoids recapitulate the crypt-villus architecture, with an internal lumen, the different epithelial lineages and stem cells located in surface protrusions, corresponding to novel crypts.

[0188] Organoids stimulation. To evaluate the effect of bacterial products on intestinal crypts the inventors grew organoids in the presence of different purified or synthetic microbe-associated molecular patterns (MAMPs). By definition, MAMPs are common to all the bacteria, pathogenic and not pathogenic, and can be released from the microbiota normally present in the gut (Garret et al., 2010). To allow MAMPs to be entrapped in the lumen of organoids they added them prior to embedding of the crypts into the matrigel. To check that the MAMPs localized inside the organoid lumen, they used, for instance, MDP-rhodamine in order to monitor the location of this molecule upon stimulation. They observed that the red signal of the MDP localized inside the organoid lumen.

[0189] To monitor if pattern recognition receptors (PRRs), upon recognition of their ligands, affected the occurrence and development of organoids, they assayed different MAMPs. Nod1 and Nod2 receptors were stimulated using soluble sonicated peptidoglycan (PGN); moreover they used Tetra-dap and MDP to discriminate between the two Nod proteins, respectively Nod1 and Nod2. Synthetic lipoprotein (Pam3), lipopolysaccharide (LPS), and flagellin (Fla) and unmethylated CpG were also assayed to respectively stimulate TLR2, TLR4, TLR5 or TLR9.

[0190] Upon MAMP-stimulation they monitored the number of formed organoids. They considered as viable organoid a structure delineated by a definite monolayer of cells (FIG. 1A). FIG. 1B are pictures showing a viable organoid on the left and a dying organoid on the right. Only the viable organoids are counted. A typical growth curve of the number of viable organoids upon stimulation is shown in FIG. 1C. Starting from the first day they observed a higher number of organoids in the conditions stimulated with PGN and with MDP, compared to the others, and this difference was maintained, and even increased over the time of the experiments.

[0191] They elected to monitor the number of organoids at day 4, the time at which the medium was normally exchanged for long culturing. Upon stimulation with PGN they observed around 5 times more organoids compared with untreated cultures (FIG. 2). To distinguish between the role of Nod1 and Nod2, they compared the effect of their dedicated agonists and found the same effect seen with PGN using MDP, the Nod2 agonist, but not with Tetra-dap, the Nod1 agonist.

[0192] The average area of the MDP-treated organoids was smaller than untreated organoids meaning that in presence of MDP more crypts could survive and generate new organoids. Indeed, following the crypt purification process, a very heterogeneous population of explanted crypts with different degrees of viability was obtained, thus introducing an asynchronous pattern of initiation of growth. In presence of MDP more stem cells could survive giving new organoids from these immature structures or also from isolated single stem cells. Moreover, no difference was observed in the maximum size of organoids compared to controls, indicating that stimulation with MAMPs did not affect their growth rate. This point is illustrated in FIG. 14. To confirm this observation, after 4 days of culture, treated or not treated organoids were tested for the expression of Ki-67, a marker for proliferative cells and for EdU (5-ethynyl-2′-deoxyuridine) incorporation. As shown in FIG. 3A (Ki-67) and FIG. 3B (EdU) they did not observe any variation in the rate of proliferation between organoids stimulated with PGN or MDP, compared with untreated controls.

[0193] To further investigate the effect of PGN, they generated organoids from Nod1 KO and Nod2 KO mice. After stimulation with PGN and MDP, the number of organoids from Nod1 KO mice, was 3 to 5 times higher, compared with untreated ones (FIG. 4). To confirm that the effect on the numbers of organoids was MDP-dependent, the inactive form of MDP, the D-D isomer (MDP-ctrl) was tested. No difference was observed with the control samples. They applied the same stimulus to organoids from Nod2 KO mice and did not observe any difference.

[0194] It has been shown that in the gut the Nod2 receptor is present at the level of the crypts in mice (Kobayashi et al., 2005). In human intestinal tissues it was demonstrated that Nod2 is expressed in Paneth cells (Ogura et al., 2003; Lala et al., 2003). Moreover an important role was shown for Paneth cells as main constituent of the niche of Lgr5 stem cells in intestinal crypts (Sato et al., 2010). To further investigate the expression of the nod2 gene in intestinal crypts, stem cell and Paneth cells were sorted, starting from Lgr5-EGFP knock-in mice. Epithelial cells were selected using an anti-Epcam antibody. Stem cells were sorted as GFP “highly positive”, CD24 “medium” positive cells, while Paneth cells were isolated as high-SSC and high-CD24 population.

[0195] Purity after sorting. To confirm the purity of the two cell populations, they performed real time-PCR checking for expression of the respective markers of stem cells and Paneth cells. B2M was used as an internal control gene and ΔΔCt values were calculated to obtain relative expression compared to whole crypt expression. For the stem cells they found high expression of Lgr5 (Barker et al., 2007), Ascl2 (Van der Flier et al, 2009a) and Olfm4 (Van der Flier et al, 2009b), and low expression of CD24. Paneth cells highly expressed CD24 (Sato et al 2009), LyzP and defcr-rs1 (Garcia et al., 2009) (FIG. 6). Then they analyzed stem cells and Paneth cells for expression of the nod2 gene. Paneth cells expressed Nod2, as previously reported, but in a lower level compared to stem cells, that showed to have 5 times more Nod2 gene than Paneth cells (see FIG. 5).

[0196] Sorting from Nod2 KO mice. Nod2 KO mice were generated by the disruption of the Nod2 gene with a EGFP cassette (Barreau et al., 2007). Cells that normally express Nod2 thus became fluorescent. They decided to evaluate the GFP signal at the level of intestinal crypts and verified that the signal was generated by the stem cells and not by the Paneth cells. They stained the Paneth cells with an anti-lysozyme antibody, and observed that the green cells, were interspersed between Paneth cells, exactly in the position of the Lgr5 stem cells. Starting from this observation they decided to extract crypts from Nod2 KO mice and to sort the GFP positive cells to verify that those cells correspond to Lgr5-expressing stem cells. Then they analyzed by RT-PCR the gene expression pattern of the sorted cells, checking the respective markers of stem cells and Paneth cells. They confirmed that GFP positive cells were expressing the markers of stem cells at the level observed in Lgr5-EGFP mice, and the same for the Paneth cells markers (FIG. 7).

[0197] Single cells stimulation. Using the same sorting strategy used to isolate cells for gene analysis, they recovered stem cells and Paneth cells from Lgr5-EGFP and Nod2 KO mice. When GFP positive cells were sorted from Nod2 KO mice and cultivated in appropriate medium, they observed the formation of organoids, meaning that the cells had properties of stem cells (FIG. 8). In this condition the average number of organoids was about 1.5 fold lower compared with wt stem cells. No differences were found when MDP was added to stem cells originating from Nod2 KO mice. On the contrary, upon MDP stimulation of wt stem cells, 2.5 more organoids were observed, compared to the unstimulated condition (p<0.0001). Co-culturing Paneth cells, derived from both wt and Nod2 KO mice, with wt stem cells, or adding the wnt ligand to wt stem cells alone, the average of organoids per well was around 3, a value similar to the one obtained in MDP-treated wt stem cells. Adding MDP to a co-culture associating wt stem cells and wt Paneth cells they observed again an increase in the average of organoids, around 6 organoids per well. The same result was obtained by co-culturing wt stem cells and Nod2 KO Paneth cells, meaning that the increase in the number of organoids upon stimulation was dependent on Nod2 expressed in stem cells. Moreover upon MDP-stimulation, no differences were observed in co-cultures associating Nod2 KO stem cells with wt or Nod2 KO Paneth cells.

[0198] Inducing stress. To test if MDP had a protective effect on stem cells present in the crypt, they decided to induce a stress targeted to intestinal stem cells. As a drug they used 1 μM doxorubicin hydrochloride (dox), known to be toxic for the stem cell (Dekaney et al., 2009), or 200 μM H.sub.2O.sub.2 to induce an oxidative stress.

[0199] After 4 days of treatment they counted 50% less organoids. When the MDP was added to the crypts in presence of doxorubicin, they were able to restore the normal growth of the organoids maintaining the same ratio between stimulated and non-stimulated organoids of 5 times. In H.sub.2O.sub.2 treated samples, the increase in the number of organoids upon MDP-stimulation was still present but in a lower ratio (FIG. 9).

[0200] Cytoprotection. Finally to assess if the differences in the number of organoids upon stimulation was dependent from a better survival of the crypts in the presence of MDP, they tested organoid culture for the release of lactate dehydrogenase (LDH). After 4 days of culture in presence of PGN or MDP the organoids produced 50% less LDH to non-stimulated crypts (FIG. 10).

[0201] Doxorubicin in vivo. To evaluate if the cytoprotective effect induced by MDP in a Nod2-dependent pathway on stem cells, that they observed in vitro, was also observed in vivo, they treated wt and Nod2 KO mice with doxorubicin. Mice were injected intraperitoneally with 20 mg/kg doxorubicin. Due to the fluorescent properties of doxorubicin, they could observe in tissue sections that red signal of doxorubicin was colocalizing with the GFP signal of the stem cells (in a Lgr5-EGFP mouse). The percentage of crypts presenting at least one caspase3-positve cell was counted. As shown in FIG. 11 upon treatment with doxorubicin wt mice presented 30% of the crypts with apoptotic cells, compare to 10% in the non-treated. In the Nod2 KO mice the effect of doxorubicin caused an increase in the percentage of apoptotic crypts, going from 14% in the non-treated to 54% in the doxorubicin injected mice. To evaluate if upon doxorubicin treatment the proliferation index was affected, the number of EdU positive cells per crypts was counted and the percentage versus the total number of cells in the crypts was calculated. Both wt and Nod2 KO mice showed a decrease in the rate of proliferation starting from 6h upon the injection of doxorubicin until 24 h. The Nod2 KO mice showed a low rate of proliferation in the crypts also in the following days, instead at 48 h the percentage of proliferative cells in the crypts of wt mice increased again until 72 h reaching almost the same rate observed in the non-treated mice (FIG. 13).

[0202] In order to evaluate the capacity of stem cells to survive and repair the damaged tissue upon doxorubicin-treatment, the inventors used a modified clonogenic microcolony assay (Tustison et al). At 72 h wt mice presented higher percentage of regenerative/surviving crypts than in Nod2 KO mice and this was associated with a bigger size of the crypts (Table 1).

[0203] After 72 h the intestinal crypts were extracted from the doxorubicin treated mice, and tested for the capacity to form organoids and to respond to MDP. As shown in FIG. 12 upon MDP stimulation the number of organoids was increasing more than 10 times compare to non-stimulated samples.

TABLE-US-00002 TABLE 1 Comparison of intestinal regeneration in wild-type and Nod2 KO mice after 72 hours upon doxorubicin injection. wt 72 h dox Nod2 KO 72 h dox Percentage of regenerative crypts 59.9 ± 6.4  33.9 ± 9.4 ** Relative size of crypts 3231 ± 105 2538 ± 180**

REFERENCES

[0204] Barker, N., van Es, J. H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth, A., et al. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature, 449(7165), 1003-1007.

[0205] Barreau, F., Meinzer, U., Chareyre, F., Berrebi, D., Niwa-Kawakita, M., Dussaillant, M., Foligne, B., et al. (2007). CARD15/NOD2 Is Required for Peyer's Patches Homeostasis in Mice. PLoS ONE, 2(6), e523.

[0206] Dekaney, C. M., Gulati, A. S., Garrison, A. P., Helmrath, M. A., & Henning, S. J. (2009). Regeneration of intestinal stem/progenitor cells following doxorubicin treatment of mice. AJP: Gastrointestinal and Liver Physiology, 297(3), G461-G470.

[0207] Escobar, M., Nicolas, P., Sangar, F., Laurent-Chabalier, S., Clair, P., Joubert, D., Jay, P., et al. (2011). Intestinal epithelial stem cells do not protect their genome by asymmetric chromosome segregation. Nature Communications, 2, 258.

[0208] Garcia, M. I., Ghiani, M., Lefort, A., Libert, F., Strollo, S., & Vassart, G. (2009). LGRS deficiency deregulates Wnt signaling and leads to precocious Paneth cell differentiation in the fetal intestine. Developmental Biology, 331(1), 58-67.

[0209] Garrett, W. S., Gordon, J. I., & Glimcher, L. H. (2010). Homeostasis and Inflammation in the Intestine. Cell, 140(6), 859-870.

[0210] Girardin, S. E. et at (2001). CARD4/Nod1 mediates NF-kappaB and JNK activation by invasive Shigella flexneri. EMBO Rep.,2:736-742.

[0211] Girardin, S. E. (2003). Peptidoglycan Molecular Requirements Allowing Detection by Nod1 and Nod2. Journal of Biological Chemistry, 278(43), 41702-41708.

[0212] Kim et al, (2010). Implication of NOD1 and NOD2 for the differentiation of multipotent mesenchymal stem cells derived from human umbilical cord blood. PLoS One. 5(10):e15369 Kobayashi, K. S. (2005). Nod2-Dependent Regulation of Innate and Adaptive Immunity in the Intestinal Tract. Science, 307(5710), 731-734.

[0213] Lala, S., Ogura, Y., Osborne, C., Hor, S. Y., Bromfield, A., Davies, S., Ogunbiyi, O., et al. (2003). Crohn's disease and the NOD2 gene: a role for paneth cells. Gastroenterology, 125(1), 47-57.

[0214] Macho Fernandez E. et al., (2011). Anti-inflammatory capacity of selected lactobacilli in experimental colitis is driven by NOD2-mediated recognition of a specific peptidoglycan derived muropeptide, Gut, 60: 1050-1059

[0215] Magalhaes, J. G., Philpott, D. J., Nahori, M.-A., Jéhanno, M., Fritz, J., Bourhis, L. L., Viala, J., et al. (2005). Murine Nod1 but not its human orthologue mediates innate immune detection of tracheal cytotoxin. EMBO reports, 6(12), 1201-1207

[0216] Ogura, Y., Lala, S., Xin, W., Smith, E., Dowds, T. A., Chen, F. F., Zimmermann, E., et al. (2003). Expression of NOD2 in Paneth cells: a possible link to Crohn's ileitis. Gut, 52(11), 1591-1597.

[0217] Rubino S. J. et al (2013). Identification of a synthetic muramyl peptide derivative with enhanced Nod2 stimulatory capacity. Innate Immun. 19(5):493-503.

[0218] Sato, T., Vries, R. G., Snippert, H. J., van de Wetering, M., Barker, N., Stange, D. E., van Es, J. H., et al. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature, 459(7244), 262-265.

[0219] Stelzner M, Chen DC (2006). To make a new intestinal mucosa. Rejuvenation Res.; 9(1):20-5.

[0220] Tustison, K. S., Yu J., Cohn S. M. (2001). Assessment of intestinal stem cell survival using the microcolony formation assay. Methods Mol. Med. 50, 267-273.

[0221] Tower J (2012). Stress and stem cells. WIREs Dev Biol.doi:10.1002/wdev.56 van der Flier, L. G., van Gijn, M. E., Hatzis, P., Kujala, P., Haegebarth, A., Stange, D. E., Begthel, H., et al. (2009a). Transcription Factor Achaete Scute-Like 2 Controls Intestinal Stem Cell Fate. Cell, 136(5), 903-912.

[0222] van der Flier, L. G., Haegebarth, A., Stange, D. E., van de Wetering, M., & Clevers, H. (2009b). OLFM4 Is a Robust Marker for Stem Cells in Human Intestine and Marks a Subset of Colorectal Cancer Cells. YGAST, 137(1), 15-17.