Culture medium for epithelial stem cells and organoids comprising the stem cells
09752124 · 2017-09-05
Assignee
Inventors
- Toshiro Sato (Hilversum, NL)
- Johannes C. Clevers (Huis ter Heide, NL)
- Meritxell Huch Ortega (Utrecht, NL)
Cpc classification
C12N2501/119
CHEMISTRY; METALLURGY
C12N2533/90
CHEMISTRY; METALLURGY
C12N5/0677
CHEMISTRY; METALLURGY
C12N5/0671
CHEMISTRY; METALLURGY
C12N2501/117
CHEMISTRY; METALLURGY
C12N2501/13
CHEMISTRY; METALLURGY
C12N2501/335
CHEMISTRY; METALLURGY
C12N2501/155
CHEMISTRY; METALLURGY
International classification
Abstract
The invention relates to a method for culturing epithelial stem cells, isolated tissue fragments comprising the epithelial stem cells, or adenoma cells, and culturing the cells or fragments in the presence of a Bone Morphogenetic Protein (BMP) inhibitor, a mitogenic growth factor, and a Wnt agonist when culturing epithelial stem cells and isolated tissue fragments. The invention further relates to a cell culture medium comprising a BMP inhibitor, a mitogenic growth factor, and a Wnt agonist, to the use of the culture medium, and to crypt-villus organoids, gastric organoids, pancreatic organoids, liver organoids, colon organoids, Barrett's Esophagus organoids, adenocarcinoma organoids and colon carcinoma organoids that are formed in the culture medium.
Claims
1. A composition comprising a three-dimensional organoid obtained by in vitro expansion of one or more adult adenoma stem cells, wherein the organoid has a sealed central lumen lined by epithelial cells, wherein the organoid comprises adult adenoma stem cells which are capable of expansion for at least three months, and wherein non-epithelial cells are absent from said organoid and said composition.
2. The composition of claim 1, wherein the three-dimensional organoid comprises cells expressing Lgr5.
3. The composition of claim 1, wherein every epithelial cell within the three-dimensional organoid comprises nuclear beta-catenin.
4. The composition of claim 1, wherein the cells lining the central lumen are polarised.
5. The composition of claim 4, wherein the lumen contains apoptotic cell bodies.
6. The composition of claim 1, comprising differentiated cell types.
7. The composition of claim 1, wherein the organoid is derived from human epithelial stem cells.
8. The composition of claim 1, wherein the composition further comprises an exogenous extracellular matrix.
9. The composition of claim 8, wherein the composition further comprises a cell culture medium comprising a Bone Morphogenetic Protein (BMP) inhibitor; between 5 and 500 ngram/ml of a mitogenic growth factor; and a Wnt agonist.
10. The composition of claim 9, wherein the BMP inhibitor is selected from the group consisting of Noggin, DAN, Cerberus and Gremlin.
11. The composition of claim 9, wherein the Wnt agonist is selected from the group consisting of one or more of Wnt, R-spondin 1 through R-spondin 4, Norrin, and a GSK-inhibitor.
12. The composition of claim 9, wherein the BMP inhibitor is Noggin, the mitogenic growth factor is Epidermal Growth Factor, and the Wnt agonist comprises any one of R-spondin 1 through R-spondin 4.
13. The composition of claim 1, wherein the cells lining the lumen are randomly oriented towards either the periphery or the central lumen.
Description
DESCRIPTION OF FIGURES
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
(38)
(39)
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)
(60)
(61)
(62)
(63)
(64)
(65)
(66)
(67)
(68)
(69)
(70)
(71)
(72)
EXAMPLES
Example 1: Culturing of Small Intestine Crypts and Villi in Vitro
(73) Materials and Methods
(74) Mice: Outbred mice of 6-12 weeks of age were used. Generation and genotyping of the Lgr5-EGFP-Ires-CreERT2 allele.sup.1 has been previously described.sup.1. Rosa26-lacZ or YFP Cre reporter mice were obtained from Jackson Labs.
(75) Crypt isolation, cell dissociation and culture: Crypts were released from murine small intestine by incubation in 2 mM EDTA/PBS for 30 min at 4 C. Isolated crypts were counted and pelleted. 500 crypts were mixed with 50 l Matrigel (BD Bioscience) and plated in 24-well plates. After polymerization of Matrigel, 500 l of crypt culture medium (Advanced DMEM/F12 with growth factors (10-50 ng/ml EGF (Peprotech), 500 ng/ml R-spondin 1.sup.11 and 100 ng/ml Noggin (Peprotech)) was added. For sorting experiments, isolated crypts were incubated in culture medium for 45 min at 37 C., following by resuspension with a glass pipette. Dissociated cells were passed through 20-m cell strainer. GFP.sup.hi, GFP.sup.low or GFP.sup. cells were sorted by flow cytometry (MoFlo, Dako). Single viable epithelial cells were gated by forward scatter, side scatter and pulse-width parameter, and negative staining for propidium iodide. Sorted cells were collected in crypt culture medium and embedded in Matrigel including Jagged-1 peptide (Ana Spec, 1 M) at 1 cell/well (in 96-well plate, 5 l Matrigel). Crypt culture medium (250 l for 48-well plate, 100 l for 96-well plate) including Y-27632 (10 M) was overlaid. Growth factors were added every other day and the entire medium was changed every 4 days. For passage, organoids were removed from Matrigel and mechanically dissociated into single-crypt domains, and transferred to new Matrigel. Passage was performed every 1-2 weeks with 1:5 split ratio.
(76) Reagents; Murine recombinant EGF and Noggin were purchased from Peprotech. Human recombinant R-spondin 1.sup.11, Y-27632 (Sigma), 4-hydroxytamoxifen (Sigma) and Edu (Invitrogen) were used for culture experiments. The following antibodies were used for immunostaining: anti-lysozyme (Dako), anti-Synaptophysin (Dako), anti-BrdU (Roche), anti--catenin (BD Bioscience), anti-E-cadherin (BD Bioscience), anti-Smooth muscle actin (Sigma), anti-EphB2 and B3 (R&D), anti-villin, anti-Muc2, anti-chromogranin A (Santa Cruz), anti-caspase-3 (Cell Signaling).
(77) Crypt Isolation: Isolated small intestines were opened longitudinally, and washed with cold PBS. The tissue was chopped into around 5 mm pieces, and further washed with cold PBS. The tissue fragments were incubated in 2 mM EDTA with PBS for 30 min on ice. After removal of EDTA medium, the tissue fragments were vigorously suspended by 10 ml pipette with cold PBS. The supernatant was the villous fraction and was discarded; the sediment was resuspended with PBS. After further vigorous suspension and centrifugation, the supernatant was enriched for crypts. This fraction was passed through a 70-um cell strainer (BD bioscience) to remove residual villous material. Isolated crypts were centrifuged at 300 rpm for 3 min to separate crypts from single cells. The final fraction consisted of essentially pure crypts and was used for culture or single cell dissociation.
(78) Tamoxifen induction and X-gal staining: To activate CreERT2, crypts were incubated with low dose 4-hydroxytamoxifen (100 nM) for 12 hr and cultured in crypt culture medium. X-gal staining was performed as previously described.sup.1. No staining was seen without 4-hydroxytamoxifen treatment.
(79) Electron microscopy analysis: As described previously.sup.1 Matrigel including crypt organoids were fixed in Karnovsky's fixative (2% paraformaldehyde, 2.5% glutaraldehyde, 0.1 M Na-cacodylate, 2.5 mM CaCl.sub.2 and 5 mM MgCl.sub.2, pH 7.4) for 5 hr at room temperature. The samples were embedded in Epon resin and were examined with a Phillips CM10 microscope (Eindhoven, The Netherlands).
(80) Microarray analysis: Gene expression analysis of colonic crypts, small intestinal crypts and organoids. Freshly isolated small intestinal crypts from two mice were divided into two parts. RNA was directly isolated from one part (RNeasy Mini Kit, Qiagen), the other part was cultured for one week, followed by RNA isolation. We prepared labeled cRNA following the manufacturer's instruction (Agilent Technologies). Differentially labelled cRNA from small intestinal crypts and organoids were hybridized separately for the two mice on a 444k Agilent Whole Mouse Genome dual colour Microarrays (G4122F) in two dye swap experiments, resulting in four individual arrays. Additionally, isolated colonic crypts were hybridized against differentially labelled small intestinal crypts in two dye swap experiments, resulting in four individual arrays. Microarray signal and background information were retrieved using Feature Extraction (V.9.5.3, Agilent Technologies). All data analyses were performed using ArrayAssist (5.5.1, Stratagene Inc.) and Microsoft Excel (Microsoft Corporation). Raw signal intensities were corrected by subtracting local background. Negative values were changed into a positive value close to zero (standard deviation of the local background) in order to allow calculation of ratios between intensities for features only present in one channel (small intestinal crypts or organoids) or (small intestinal crypts or colonic crypts). Normalization was performed by applying a Lowess algorithm and individual features were filtered if both (small intestinal crypts or organoids) or (small intestinal crypts or colonic crypts) intensities were changed or if both intensities were less than two times the background signal. Furthermore, non-uniform features were filtered. Data are available at GEO (Gene Expression Omnibus, number GSE14594) upon publication. Unsupervised hierarchical clustering was performed on normalized intensities (processed signal in Feature Extraction) of small intestinal/colonic crypts and organoids using Cluster 3 (distance: city block, correlation: average linkage) and visualized with TreeView. Genes were considered significantly changed if they were consistently in all arrays more than 3-fold enriched in organoids or crypts.
(81) Imaging analysis: The images of crypt organoids were taken with either confocal microscopy (Leica, SP5), inverted microscope (Nikon DM-IL) or stereomicroscope (Leica, MZ16-FA). For immunohistochemistry, samples were fixed with 4% paraformaldehyde (PFA) for 1 hr at room temperature, and Paraffin sections were processed with standard technique.sup.1. Immunohistochemistry was performed as previously described.sup.1. For whole-mount immunostaining, crypts organoids were isolated from matrigel using with Dispase (Invitrogen), and fixed with 4% PFA, following by permiabilization with 0.1% Triton-X. EdU staining was performed following the manufacturer's protocol (Click-IT, Invitrogen). DNA was stained by DAPI or ToPro-3 (Molecular Probe). 3D images were acquired with confocal microscopy (Leica, SP5) and reconstructed with Volocity Software (Improvision).
(82) Results
(83) The intestinal epithelium is the most rapidly self-renewing tissue in adult mammals. We have recently demonstrated the presence of approximately six cycling Lgr5.sup.+ stem cells at the bottoms of small intestinal crypts.sup.1. We have now established long-term culture conditions under which single crypts undergo multiple crypt fission events, whilst simultaneously generating villus-like epithelial domains in which all differentiated cell types are present. Single sorted Lgr5.sup.+ stem cells can also initiate these crypt-villus organoids. Tracing experiments indicate that the Lgr5.sup.+ stem cell hierarchy is maintained in organoids. We conclude that intestinal crypt-villus units are self-organizing structures, which can be built from a single stem cell in the absence of a non-epithelial cellular niche.
(84) The self-renewing epithelium of the small intestine is ordered into crypts and villi. Cells are newly generated in the crypts and are lost by apoptosis at the tips of the villi, with a turn-over time of 5 days in the mouse. Self-renewing stem cells have long been known to reside near the crypt bottom and to produce the rapidly proliferating transit amplifying (TA) cells. The estimated number of stem cells is between 4 and 6 per crypt. Enterocytes, goblet cells and enteroendocrine cells develop from TA cells and continue their migration in coherent bands along the crypt-villus axis. The fourth major differentiated cell-type, the Paneth cell, resides at the crypt bottom. We have recently identified a gene, Lgr5, which is specifically expressed in cycling Crypt Base Columnar cells that are interspersed between the Paneth cells.sup.1. Using a mouse in which a GFP/tamoxifen-inducible Cre recombinase cassette was integrated into the Lgr5 locus, we showed by lineage tracing that the Lgr5.sup.+ cells constitute multipotent stem cells which generate all cell types of the epithelium.sup.1, even when assessed 14 months after Cre induction.sup.3.
(85) Although a variety of culture systems has been described.sup.4-7, no long-term culture system has been established which maintains basic crypt-villus physiology.
(86) Mouse crypt preparations were suspended in Matrigel. Crypt growth required EGF and R-spondin 1 (
(87) Culture of Lgr5-EGFP-ires-CreERT2 crypts revealed Lgr5-GFP.sup.+ stem cells intermingled with Paneth cells at the crypt base. Wnt activation, as evidenced by nuclear -catenin (
(88) We cultured crypts from Lgr5-EGFP-ires-CreERT2 mice crossed with the Cre-activatable Rosa26-LacZ reporter to allow lineage tracing. Directly after induction by low-dose tamoxifen, we noted single labeled cells (
(89) Recently, mammary gland epithelial structures were established from single stem cells in vitro.sup.21. When single Lgr5-GFP.sup.hi cells were sorted, these died immediately. The Rho kinase inhibitor Y-27632, significantly decreased this cell death. A Notch agonistic peptide.sup.24 was found to support maintenance of proliferative crypts.sup.23. Under these conditions, significant numbers of Lgr5-GFP.sup.hi cells survived and formed large crypt organoids. Organoids formed rarely when GFP.sup.low daughter cells were seeded (
(90) We sorted cells at 1 cell/well, visually verified the presence of single cells and followed the resulting growth. In each of four individual experiments, we identified and followed 100 single cells. On average, approximately 6% of the Lgr5-GFP.sup.hi cells grew out into organoids, while the remaining cells typically died within the first 12 hours, presumably due to physical and/or biological stress inherent to the isolation procedure. GFP.sup.low cells rarely grew out (
(91) The single stem cell-derived organoids appeared indistinguishable from those derived from whole crypts. Paneth cells and stem cells were located at crypt bottoms (
(92) It is well known that epithelial crypts are in intimate contact with subepithelial myofibroblasts.sup.26-28 and it is generally believed that the latter cells create a specialized cellular niche at crypt bottoms.sup.27, 29, 30. Such a niche would create a unique environment to anchor and support the intestinal stem cells. We now show that a self-renewing epithelium can be established by a limited set of growth signals that are uniformly presented. Despite this, the isolated stem cells autonomously generate asymmetry in a highly stereotypic fashion. This rapidly leads to the formation of crypt-like structures with de novo generated stem cells and Paneth cells located at their bottoms and filled with TA cells. These crypt-like structures feed into villus-like luminal domains consisting of postmitotic enterocytes, where apoptotic cells pinch off into the lumen, reminiscent of cell loss at villus tips. The paradoxical observation that single cells exposed to a uniform growth-promoting environment can generate asymmetric structures is particularly evident upon scrutiny of the Wnt pathway. While all cells are exposed to R-spondin 1, only cells in crypts display hallmarks of active Wnt signaling, i.e., nuclear -catenin and the expression of Wnt target genes. Apparently, differential responsiveness to Wnt signaling rather than differential exposure to extracellular Wnt signals lies at the heart of the formation of a crypt-villus axis.
(93) In summary, we conclude that a single Lgr5.sup.+ve intestinal stem cell can operate independently of positional cues from its environment and that it can generate a continuously expanding, self-organizing epithelial structure reminiscent of normal gut. The described culture system will simplify the study of stem cell-driven crypt-villus biology. Moreover, it may open up new avenues for regenerative medicine and gene therapy.
Example 2: Culturing of Colon Crypts and Villi in Vitro
(94) Material and Methods
(95) Wnt3a Conditioned Medium
(96) A Wnt3a ligand expressing cell line and the same cell line, without the Wnt3a ligand (control medium) are cultured for a period of 3-4 weeks. The cells will produce Wnt3a as soon as they stop grown at confluency. The medium will be harvested and tested in the TOPflash assay, a luciferase assay using a TCF responsive elements-luc construct (TOP) and the same construct, but with mutations in the TCF responsive elements (FOP). The ratio between TOP/FOP should be more than 20 for the medium to be used in cultures. The medium is diluted 25-50% when used in the cultures to regenerate tissue.
(97) Freshly isolated colon was opened and washed with PBS or DMEM, and cut into small pieces. The fragments were incubated with 2 mM EDTA/PBS for 1 hour at 4 C. under gentle rocking. Following removal of EDTA solution, the tissue fragments were vigorously suspended in 10 ml of cold PBS with a 10 ml pipette. The first supernatant containing debris was discarded and the sediment was suspended with 10-15 ml PBS. After further vigorous suspension of the tissue fragments the supernatant is enriched in colonic crypts. The fragments were pelleted and mixed with matrigel and cultured as small intestinal organoid culture system. The matrigel was incubated for 5-10 min at 37 C. After matrigel polymerization, 500 l of tissue culture media (50% Advanced-DMEM/F12/50% Wnt-3a conditioned medium-supplemented with 200 ng/ml N-Acetylcysteine, 50 ng/ml EGF, 1 g/ml R-spondin1, 100 ng/ml Noggin, 100 ng/ml BDNF (Peprotech) was added. The entire medium was changed every 2-3 days. For passage, the organoids were removed from the Matrigel using a 1000 l pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio at least once every two weeks. Under these conditions cultures have been maintained for at least for 3 months.
(98) Results
(99) Colonic organoids grow slower and less efficient as compared with small intestinal organoids. With the same growth factors condition as small intestine, less than 5% of colonic crypts isolated from distal colon grew and formed organoid structure (
(100) In conclusion, both small intestine derived and colon derived crypts can be maintained and propagated in vitro using the above described conditions, making this the first culture method ever described to result in the generation of intestinal epithelium in an artificial system.
Example 3: Culturing of Adenomas in Vitro
(101) Materials and Methods
(102) (See example 1)
(103) Results
(104) Adenomas have been historically difficult to culture in vitro. Since the above described conditions were used to successfully culture healthy crypts derived from small intestine as well as colon, it was determined whether similar conditions could sustain adenomas in vitro. After isolation of adenoma from APC/ mice using 2.5 mM EDTA, single adenomas were cultured under similar conditions as described above. Importantly, these conditions were adequate to maintain growth of the adenomas in vitro, however, R-spondin had become redundant. This can be easily explained by the fact that it no longer is necessary to induce the Wnt signaling pathway, since the absence of APC in these cells will automatically result in nuclear -Catenin. This makes R-spondin, a Wnt agonist, redundant in culturing adenomas in vitro.
(105) It was further tested whether single Lgr5+ sorted cells derived from the adenomas in Lgr5-EGFP-Ires-CreERT2/APCflox/flox mice were able to form similar adenoma organoids in vitro using the aforementioned culture conditions (without R-spondin). Indeed, this was the case and the organoids obtained were highly comparable in structure to those that were obtained using complete adenomas as starting material for the in vitro culture (data not shown).
Example 4: Testing the Effect of Other Wnt Agonists
(106) To determine whether other Wnt agonists have the same effect as R-spondin does, namely facilitate formation of crypt-villus organoids in vitro, soluble Wnt3a was added to Lgr5.sup.+ sorted single cells and the effect on crypt-villus formation in vitro was assessed.
(107) Materials and Methods
(108) Lgr5-GFP.sup.hi cells were sorted and cultured with or without Wnt3a (100 ng/ml) in addition to conventional single cell culture condition (EGF, noggin, R-spondin, Notch ligand and Y-27632, as described above for single cells). We seeded 100 cells/well and counted the number of organoids 14 days after seeding.
(109) Isolated crypts were incubated with 1 uM Newport Green-DCF (MolecularProbes) in PBS+0.1% Pluronic 127 (Sigma) for 3 min at room temperature, following by PBS wash. After this, crypts were embedded in Matrigel and cultured using the standard conditions as described above.
(110) Results
(111) The addition of Wnt3a in the absence of R-spondin did not have any effect on colony formation: little to no colonies were formed in the absence of R-spondin. However, in the presence of R-spondin, an increased efficiency in organoid formation was observed only in the presence of Wnt3a (
(112) Wnt3a is a soluble factor that, under physiological circumstances, is produced by Paneth cells. These cells are generally located adjacent to the stem cells (
Example 5: Culture Conditions Support the Growth of Stomach Organoids as Well
(113) The stomach consists of 3 topographic regions (fundus, corpus, and antrum) and 2 functional glandular areas (oxyntic and pyloric). The oxyntic gland area comprises 80% of the organ whereas the pyloric area comprises the 20% of the organ. The mammalian gastric epithelium is organized into gastric units consisting of a planar surface epithelium, a short pit and a long gland. The pit is lined by mucus secreting cells whereas the gland is composed of secreting cells separated in three regions: the isthmus, the neck and the base. The gastric epithelium is constantly renewed. Tracing studies performed in our laboratory have shown that LGR5 positive cells located at the gland base fulfil the definition of stemness (Barker et al. under preparation).
(114) So far, gastric monolayer cultures have not been able to recapitulate the features of the gastric unit, which is formed by several differentiated gastric cells. Furthermore, the 3-D culture method systems reported only reconstruct highly differentiated gastric surface mucous cells, without showing any endocrine cells. Moreover, these cultures had only been carried out over a period of 7 days, thus indicating a lack of self-renewing capacity (Ootani A, Toda S, Fujimoto K, Sugihara H., Am. J. Pathol. 2003 June; 162(6):1905-12). Here we have developed a method to isolate gastric units from the pyloric region of the murine stomach and have been able to develop a 3D-culture system that shows longer-lived maintenance.
(115) Materials and Methods
(116) Gastric Unit Isolation
(117) Isolated stomachs were opened longitudinally and washed in cold Advanced-DMEM/F12 (Invitrogen). Under the stereoscope, the pyloric region was excised and isolated from the body and forestomach and the pyloric mucosa was carefully separated from the muscle layer with tweezers. Then, the tissue was chopped into pieces of around 5 mm and further washed with cold isolation buffer (Na.sub.2HPO.sub.4 28 mM+KH.sub.2PO.sub.4 40 mM+NaCl 480 mM+KCl 8 mM+Sucrose 220 mM+D-Sorbitol 274 mM+DL-Dithiotreitol 2.6 mM). The tissue fragments were incubated in 5 mM EDTA with isolation buffer for 2 h at 4 C. under gentle rocking. Following removal of EDTA solution, the tissue fragments were vigorously suspended in 10 ml of cold isolation buffer with a 10 ml pipette. The first supernatant containing dead cells was discarded and the sediment was suspended with 10-15 ml cold isolation buffer. After further vigorous suspension of the tissue fragments the supernatant is enriched in gastric units. Every 10-20 suspensions the supernatant is replaced for fresh cold isolation buffer and is kept on ice and checked for the presence of gastric units. This procedure is repeated until the complete release of the gastric units, usually 4-5 times. Enriched gastric unit suspensions are centrifuged at 600 rpm for 2-3 min to separate the isolated gastric units from single cells and the sediment is used for culture.
(118) Gastric Culture
(119) Entire gastric units containing the gland, isthmus and pit regions were isolated from the pyloric region of murine stomach by incubating with 5 mM EDTA at 4 C. for 2 h as indicated in the previous section. Isolated gastric units were counted and pelleted. 100 gastric units were mixed with 25 l of Matrigel (BD Bioscience), seeded on 48-well tissue culture plates and incubated for 5-10 min at 37 C. until complete polymerization of the Matrigel. After polymerization, 250 l of tissue culture media (Advanced-DMEM/F12 supplemented with B27, N2, 200 ng/ml N-Acetylcysteine, 50 ng/ml EGF, 1 g/ml R-spondin1, 100 ng/ml Noggin, 100 ng/ml Wnt3A, 50 or 100 ng/ml KGF) was added. The entire medium was changed every 2 days. For passage, the organoids were removed from the Matrigel using a 1000 l pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio once or twice per week. Under these conditions cultures have been maintained for at least 1 month.
(120) Reagents
(121) Advanced DMEM/F12 and supplements N2 and B-27 Serum-Free Supplement were purchased from Invitrogen and N-Acetylcysteine from Sigma. Murine recombinant EGF, Noggin and human KGF were purchased from Peprotech, and Wnt3A recombinant protein from Stem Cell Research. From the mentioned growth factors, different concentrations have only been tested for R-Spondin1 and KGF. At 50 ng/ml R-Spondin 1 inhibits culture growth. KGF can be used either at 50 or 100 ng/ml but the budding efficiency is higher in the 100 ng/ml condition.
(122) Wnt3A conditioned media was prepared as previously described (Willert K., Brown J. D., Danenberg E., Duncan A. W., Weissman I. L., Reya T., Yates J. R. 3rd, Nusse R. Nature 2003 May 22; 423(6938):448-52).
(123) Immunohistochemistry and Imaging Analysis
(124) For X-gal staining, organoids were directly fixed in the matrigel with 0.25% glutaraldehyde (Sigma) in 100 mM MgCl.sub.2 in PBS, for 1-2 h at room temperature. After, cultures were washed 3-times with washing solution (0.01% Sodium Deoxycholate+0.02% NP40+5 mM MgCl.sub.2 in PBS) and incubated for 16 h at 37 C. with 1 mg/ml X-Gal (Invitrogen) in the presence of 0.21% K.sub.4Fe(CN).sub.6 and 0.16% K.sub.3Fe(CN).sub.6. After washing in PBS, cultures were post fixed with 2% PFA in PBS for 15 min at room temperature. All reagents were acquired from Sigma.
(125) For immunohistochemistry, organoids were isolated from the matrigel using trypsine (Tryple Select, Invitrogen), fixed with 4% PFA for 1 h at room temperature and embedded in paraffin. Paraffin sections were processed with standard techniques and immunohistochemistry was performed as previously described. The following antibodies were used anti-mouse Ki67 (clone MM1, Monosan) (1:200), anti-rabbit cleaved caspase-3 (Cell Signaling Technology) (1:400) and anti-human gastric mucin 5AC (Novocastra clone 45M1) (1:200). Citrate buffer antigen retrieval was performed in all cases. Sections were counterstained with Mayer's haematoxylin.
(126) The images of gastric organoids and isolated gastric glands were taken with either inverted microscope (Nikon DM-IL) or confocal microscopy (Leica SP5).
(127) Results
(128) So far, gastric cultures have been grown in monolayers. Monolayer cultures, however, lack the ability to recapitulate the features of the entire gastric unit, which is formed by several differentiated gastric cells (pit mucous cells, enteroendocrine cells and proliferating mucous-free cells). Recently our laboratory has demonstrated by in vivo lineage tracing, that the Lgr5 positive cells present at the bottom of the intestinal crypts are true intestinal stem cells (Barker N., van Es J. H., Kuipers J., Kujala P., van den Born M., Cozijnsen M., Haegebarth A., Korving J., Begthel H., Peters P. J., Clevers H. Nature 2007; 449:1003-7). As the intestinal epithelium, the gastric epithelium is constantly renewed. Lgr5 positive cells have been found at the bottom of the pyloric gastric gland units and, tracing studies have shown that these LGR5 positive cells fulfil the definition of stemness by showing self-renewal and multipotency capability (Barker et al. under preparation). Since we have been able to culture intestinal crypts from single Lgr5+ cells in 3-D structures, it was determined whether similar conditions could sustain the growth of pyloric gastric units in vitro.
(129) After isolation of gastric gland units using 5 mM EDTA, gastric glands (
(130) Organoids have been cultured for at least 1 month without losing the properties described. Weekly, organoids are passaged 1:4 by mechanical dissociation (
Example 6a
Pancreas Organoids can be Grown In Vitro
(131) Material and Methods
(132) Freshly isolated pancreas was cut into small pieces, and incubated in DMEM (Invitrogen) with digestive enzyme mixture (300 U/ml Collagenase typeXI (Sigma), 0.01 mg Dispase I (Roche) and 0.1 mg DNase) for 10 minutes in orbital shaker (80 rpm, 37 C.). After incubation, the tissue fragments were mildly dissociated by mechanical pipetting. Undigested fragments were settled down for 1 minute with normal gravity, and the supernatant was transferred to a new tube. The supernatant was passed through 70 um-cell strainer, and the residue was washed with DMEM. The fragments remaining on the cell strainer were harvested by rinsing the inverted cell strainer with DMEM, and pelletted. The fragments mostly consist of pancreatic acinar tissue and included pancreatic ducts. The pellet was mixed with matrigel and cultured as small intestinal organoid culture system (see materials and methods of example 1). The matrigel was incubated for 5-10 min at 37 C. After polymerization of matrigel, 500 l of tissue culture media (Advanced-DMEM/F12 supplemented with B27, N2, 200 ng/ml N-Acetylcysteine, 50 ng/ml EGF, 1 g/ml R-spondin 1, 100 ng/ml Noggin, 50 or 100 ng/ml KGF (Peprotech) was added. The growth factors were added every 2 days. The entire medium was changed every 4-6 days. For passage, the organoids were removed from the Matrigel using a 1000 l pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio once or twice per week. Under these conditions cultures have been maintained for at least for 2 months.
(133) Results
(134) Pancreatic tissue formed simple cyst structure 3-4 days after culture in the presence of EGF. Noggin and R-spondin synergistically increased the size of cyst structure, but did not affect morphogenesis of organoids. KGF significantly induced budding formation as well as culture efficiency. Using the optimal combination of growth factors (EGF, Noggin, R-spondin-1 and KGF), more than 80% of pancreatic duct grew in the best combination of growth factors
(135) Once the pancreatic ducts had been taken in culture, the ducts quickly sealed both ends of the structure and form a simple structure. Approximately 20% of organoids started to form a budding structure 7 days after the start of the culture (
(136) Interestingly, after passage of the organoids, appr. 2-3 weeks after the start of the culture, pancreatic islet-like structure were observed (
Example 6b
Pancreas Organoids can be Grown In Vitro
(137) Material and Methods
(138) Freshly isolated pancreas was cut into small pieces, and incubated in DMEM (Invitrogen) with digestive enzyme mixture (300 U/ml Collagenase typeXI (Sigma), 0.01 mg/ml Dispase I (Roche) and 0.1 mg/ml DNase) for 10 minutes in orbital shaker (80 rpm, 37 C.). After incubation, the tissue fragments were mildly dissociated by mechanical pipetting. Undigested fragments were settled down for 1 minute with normal gravity. The undigested fragments were further digestive with the digestive enzyme mixture for 10 minutes. This digestion procedure was repeated until the undigested fragments were mostly consist of pancreas ducts. Pancreas duct structures were manually picked up from undigested fragments under the microscopy. The pancreas ducts were mixed with matrigel and cultured as small intestinal organoid culture system (see materials and methods of example 1). The matrigel was incubated for 5-10 min at 37 C. After polymerization of matrigel, 500 l of tissue culture media (Advanced-DMEM/F12 supplemented with 1 Glutamax, Penicilin/Streptomycin, 10 mM Hepes, B27, N2, 10 mM N-Acetylcysteine, 10 nM [Leu.sup.15]-Gastrin I, 100 nM Exendin4, 10 mM Nicotinamide, 50 ng/ml EGF, 1 g/ml R-spondin 1, 100 ng/ml Noggin, 50 or 100 ng/ml FGF7 (KGF) or FGF10 (Peprotech) was added. The culture medium was changed every 2 days. For passage, the organoids were removed from the Matrigel using a 1000 l pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio once or twice per week. Under these conditions cultures have been maintained for at least for 10 months.
(139) Results
(140) Pancreatic tissue formed simple cyst structure 3-4 days after culture in the presence of EGF. Noggin and R-spondin synergistically increased the size of cyst structure, but did not affect morphogenesis of organoids. FGF7 (KGF)/FGF10 significantly induced budding formation as well as culture efficiency. Using the optimal combination of growth factors (EGF, Noggin, R-spondin-1 and FGF7 (KGF)/FGF10), more than 80% of pancreatic duct grew in the best combination of growth factors
(141) Once the pancreatic ducts had been taken in culture, the ducts quickly sealed both ends of the structure and form a simple structure. Approximately 80% of organoids started to form a budding structure 7 days after the start of the culture (
(142) Interestingly, after passage of the organoids, appr. 2-3 weeks after the start of the culture, pancreatic islet-like structure were observed (
Example 7: Unimpeded Expansion of Adult Pancreatic Progenitors In Vitro by Driving a Wnt/Lgr5 Regenerative Response
(143) Materials and Methods
(144) Mice, Reagents and Tissues
(145) Pancreatic tissue was obtained from the following mice: Axin-LacZ knock in (Lustig et al., Mol. Cell. Biol. 2002), Lgr5-LacZ Knockin (Barker et al., 2007), Lgr5-GFP (Barker et al., 2007). Axin-LacZ mice were injected IP with 100 g of purified human R-spondin 1 (kindly provided by A. Abo, Nuvelo Inc, CA, USA) and sacrificed 48 hours later for LacZ expression analysis in the pancreas.
(146) Pancreatic duct ligation was performed as described in rats (Wang et al., 1995) with some minor modifications: The experimental procedure for PDL was the following: animals are anesthetized with a mixture of fluanisone:fentanyl:midazolam injected intraperitoneally at a dosage of 3.3 mg/Kg, 0.105 mg/Kg and 1.25 mg/Kg respectively. Animals are placed in supine position and the abdominal surface is shaved and cleaned with antiseptic solution (iodine solution). Following, a median incision at the upper anterior abdominal wall from the xiphisternum is performed and the pancreas is exposed. Under a dissecting microscope, the pancreatic splenic lobe is localized and the pancreatic duct is ligated with a 7-0 polypropylene suture monofilament at approximately 1 mm distal to the junction with the gastric lobe duct. Following surgery the analgesic buprenorphine is administered s.c. at a dose 0.01-0.05 mg/Kg. Following, the abdominal wall and skin was closed with 5-0 silk sutures.
(147) Freshly isolated pancreas was treated as described under example 6, resulting in pancreatic fragments that were cultured under conditions as described below. The main pancreatic duct and first branch of ducts are mechanically isolated. The fragments were cut into small pieces and incubated in DMEM (Invitrogen) with digestive enzyme mixture (300 U/ml Collagenase type XI (Sigma), 0.01 mg/ml Dispase I (Roche) and 0.1 mg/ml DNase) for 30 minutes in orbital shaker (80 rpm, 37 C.). After the digestion, most of acinar cells were released from the fragments. Undigested fragments mostly consist of pancreatic duct cells were settled down for 1 minute with normal gravity, and the supernatant was discarded. After three times washing with PBS, the undigested fragments were incubated with 2 mM EDTA/PBS for 30 min at room temperature. The fragments were vigorously pipetted and settled down for 1 minute with normal gravity. The supernatant enriched with duct cells were transferred to new tubes and washed with PBS for three times. The duct cells were pelleted and mixed with the Matrigel. The matrigel was incubated for 5-10 min at 37 C. After polymerization of matrigel, 500 l of Expansion medium (Advanced-DMEM/F12 supplemented with 1 Glutamax, Penicilin/Streptomycin, 10 mM Hepes, B27, N2, 1 mM N-Acetylcysteine, 10 nM [Leu.sup.15]-Gastrin I, 100 nM Exendin4, 10 mM Nicotinamide, 50 ng/ml EGF, 1 g/ml R-spondin1, 100 ng/ml Noggin, 50 or 100 ng/ml FGF7 (KGF) or FGF10 (Peprotech) was added. The entire medium was changed every 2 days. For passage, the organoids were removed from the Matrigel using a 1000 A pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio once per week. Under these conditions cultures have been maintained for at least for 2 months. For differentiation, expansion medium were changed into differentiation medium (Advanced-DMEM/F12 supplemented with Glutamax, Penicilin/Streptomycin, 10 mM Hepes, B27, N2, 200 ng/ml N-Acetylcysteine, 10 nM [Leu.sup.15]-Gastrin I, 100 nM Exendin4, 50 ng/ml EGF, 1 g/ml R-spondin 1, 100 ng/ml Noggin).
(148) FGF10 was obtained from Peprotech. BrdU was obtained from Sigma.
(149) Q-PCR
(150) RNA was isolated by RNA easy mini kit (Quiagen), and reverse transcribed using Moloney Murine Leukemia Virus reverse transcriptase (Promega). cDNA was amplified in a thermal cycler.
(151) Primers used are shown below.
(152) TABLE-US-00001 mmTBP(forward): TATTGTATCTACCGTGAATCTTGG (SEQIDNO:2) mmTBP(reverse): CAGTTGTCCGTGGCTCTC (SEQIDNO:3) Lgr5(forward) TCCAACCTCAGCGTCTTC (SEQIDNO:4) Lgr5(reverse) TGGGAATGTGTGTCAAAG(Tm= 57 C.) (SEQIDNO:5)
PCR
(153) All primers were designed to flank or span intron sequences in order to distinguish genomic DNA.
(154) TABLE-US-00002 Hprt (F)AAGTTTGTTGTTGGATATGC (SEQIDNO:6) (R)CATCTTAGGCTTTGTATTTGG (SEQIDNO:7) (Tm)57 C.,106bp Ngn3 (F)TCCTCGGAGCTTTTCTACGA (SEQIDNO:8) (R)TGTGTCTCTGGGGACACTTG (SEQIDNO:9) (Tm)60 C.,239bp/373bp(genomicband) Pax6 (F)AACAACCTGCCTATGCAACC (SEQIDNO:10) (R)ACTTGGACGGGAACTGACAC (SEQIDNO:11) TM60 C.,206bp Glucokinase (F)AAGATCATTGGCGGAAAG (SEQIDNO:12) (R)GAGTGCTCAGGATGTTAAG (SEQIDNO:13) (Tm)57 C.193bp ChromograninA (F)GCTGACAGCAGAGAAGCGGCT (SEQIDNO:14) (R)GACAGGCTCTCTAGCTCCTGG (SEQIDNO:15) (Tm)60 C.231bp Glut2(slc2a2) (F)AAGTTGGAAGAGGAAGTCAG (SEQIDNO:16) (R)AGACCTTCTGCTCAGTCG (SEQIDNO:17) (Tm)57 C.124bp Insulin (F)TTTGTCAAGCAGCACCTTTG (SEQIDNO:18) (R)TCTACAATGCCACGCTTCTG (SEQIDNO:19) (Tm)57 C.,214bp Somatostatin (F)GAGGCAAGGAAGATGCTGTC (SEQIDNO:20) (R)GGGCATCATTCTCTGTCTGG (SEQIDNO:21) (Tm)57 C.,214bp Glucagon (F)TTACTTTGTGGCTGGATTGCTT (SEQIDNO:22) (R)AGTGGCGTTTGTCTTCATTCA (SEQIDNO:23) (Tm)57 C.,149bp
Image Analysis
(155) The images of crypt organoids were taken by either confocal microscopy with a Leica SP5, an inverted microscope (Nikon DM-IL) or a stereomicroscope (Leica, MZ16-FA). For immunohistochemistry, samples were fixed with 4% paraformaldehyde (PFA) for 1 h at room temperature, and paraffin sections were processed with standard techniques (Barker et al., Nature 2007). Immunohistochemistry was performed as described previously (Barker et al., Nature 2007). For whole-mount immunostaining, pancreas organoids were isolated from Matrigel using Dispase (Invitrogen), and fixed with 4% PFA, followed by permeabilization with 0.1% Triton X-100. Following antibodies were used for immunohistochemistry; anti-BrdU (Amersham), anti-Ki67 (Dako), anti-Insulin (Sigma), anti-C-peptide (Cell signaling), anti-Ngn3 (Developmental hybridoma studies bank)
(156) DNA was stained with DAPI or ToPro-3 (Molecular Probes). Three-dimensional images were acquired with confocal microscopy. The staining with X-gal was performed as described under example 5 under immunohistochemistry and imaging analysis.
(157) FACS
(158) Pancreatic organoids were cultured in the presence or absence of R-Spondin (1 g/ml) were removed from matrigel mechanically or enzymatically (TrypLE). Isolated organoids were further digested by TrypLE for 10 min at 37 C. Dissociated cells were passed through 40 um cell strainer (BD bioscience) and stained with APC conjugated anti-EpCAM (eBioscience). LacZ was stained by FluoReporter kit(Invitrogen) following manufacturer's protocol. Single viable cells were gated with Pulse-width, Side scatter parameter and propidium iodide staining.
(159) In Vitro Expansion of Single Axin2-LacZ Positive Pancreatic Cells
(160) Pancreas was isolated from mice 7 days after PDL treatment, and pancreas ducts were isolated as described above. Isolated pancreas ducts were incubated with TrypLE Express (Invitrogen) for 20 min at 37 C, following by passing through 40 um cell strainer (BD bioscience). Cells were stained with EpCAM-APC and fluorescent substrate for LacZ (FluoroReporter kit) as described in Example 7. Cells were analyzed and single viable epithelial cells were sorted by flow cytometer (MoFlo; Dako Cytomation), and collected in the EM medium. Sorted cells were pelleted, mixed with Matrigel and cultured with EM medium including 50% Wnt conditioned medium and 10 mM Y-27632 for 4 days. Culture medium was changed into EM medium without Wnt and Y-27632 after 4 days.
(161) Results
(162) Single Wnt-dependent Lgr5.sup.+ stem cells derived from the small intestine can be cultured to form continuously expanding gut-like organoids (Sato et al., 2009) In healthy adult pancreas, the Wnt pathway is inactive and -consequently-Lgr5 is not expressed. Upon injury by partial duct ligation (PDL), we find that the Wnt pathway becomes robustly activated, while Lgr5 expression appears at the buds of regenerating ducts. Under conditions modified from the intestinal culture system, freshly isolated adult duct fragments initiate expression of Lgr5 and form budding cysts which expand 10-fold weekly for >30 weeks. Removal of growth stimuli converts these cysts into structures with immature islet morphology, expressing endocrine and -cell markers. Single Wnt-stimulated cells from injured pancreas can also initiated these long-term cultures. We conclude that the Hayflick limit does not apply to adult progenitor cells when cultured under optimized conditions. Thus, culture methods favoring expansion of organ-specific adult stem cells may represent an alternative to ES- or iPS-based tissue generation.
(163) While development of the exocrine and endocrine compartments of the embryonic pancreas are understood in great detail (Jensen, 2004), much less is known about the generation of islet cells in the postnatal pancreas (Bonner-Weir and Weir, 2005; Bouwens and Rooman, 2005). Genetic lineage tracing has provided proof that pre-existing cells, rather than stem/progenitor cells, generate new cells in adult mice both under normal physiological conditions and after partial pancreatectomy (Dor et al., 2004; Teta et al., 2007). The existence of multipotent progenitor cells in the ductal lining of the pancreas of adult mice has recently described, which can be activated in injured pancreas to increase the functional cell mass (Xu et al. 2008). Controlled injury was obtained by performing PDL on the pancreas of adult mice carrying a promoter reporter of Ngn3, which encodes a master switch for embryonic islet cell progenitors (Apelqvist et al., 1999; Gradwohl et al., 2000; Gu et al., 2002; Schwitzgebel et al., 2000) and which is silent in normal postnatal pancreas (Gu et al., 2002). Differentiation of these cell progenitors is Ngn3-dependent and gives rise to all islet cell types, including glucose-responsive cells (Xu et al., 2008). It is currently not known which signals drive the appearance of these progenitors upon injury. Such insights appear important as they may guide the design of in vitro approaches to progenitor expansion.
(164) To determine whether Wnt signaling plays a role in the induction of cell progenitors, the expression of the Axin2-LacZ allele was followed in the adult pancreas. The Axin2-LacZ allele has proven to represent a faithful, general reporter for Wnt signaling (Lustig et al., Mol. Cell. Biol. 2002). As expected, the reporter was inactive in adult pancreas (
(165) We have previously shown in the intestines that a certain population of Wnt responsive cells are stem cells (Barker et al., 2007). A marker for that population of cells was Lgr5. The Lgr5 gene is, like Axin2, a Wnt-responsive gene. Yet in the intestine and the skin it is only expressed in Wnt-stimulated stem cells but not in transit amplifying cells (Barker et al., 2007; Jaks et al., 2008)). It is therefore considered to be a genuine stem cell marker. We hypothesized that, similar to the Lgr5+ cells in the intestines, Lgr5+ cells in the pancreas may also be the origin of the cell progenitors as detected after injury. To test this hypothesis, we performed PDL in in the pancreas of Axin-LacZ and Lgr5-LacZ mice and determined Lgr5 mRNA expression and LacZ staining. Interestingly, Lgr5 became readily detectable by qPCR in a post-PDL time course (
(166) Given the appearance of the Wnt-dependent Lgr5 stem cell marker, we reasoned that adult pancreas progenitors may by expanded in the previously defined gut organoid culture conditions (Sato et al., 2009). Cultures of heterogeneous populations of pancreas cells have been previously established and typically include growth factors such as EGF (Githens et al. In Vitro Cell Dev. Biol. 1989), FGF10 (Miralles et al. Proc. Natl. Acad. Sci. U.S.A. 1999) and HGF (Lefebvre et al. Diabetes 1998, Suzuki et al., Diabetes 53, 2004) and serum supplements such as Gastrin (Rooman et al. Gastroenterology 2001), Nicotinamide (Rooman et al. Diabetologia 2000) and others. A number of such cultures resulted in the in vitro generation of cells with a cell-like phenotypes (Bonner-Weir et al., 2000; Seaberg et al., 2004; Suzuki et al., 2004) that under certain conditions were able to reverse hyperglycemia when transplanted in diabetic mice (Hao et al., 2006; Ramiya et al., 2000). Most of these approaches start with mixed cell populations that undergo senescence over time. It appears fair to say that no robust, long-term culture system exists today which maintains robust expansion of defined, non-transformed adult pancreas progenitors over long periods of time that maintain the capacity to differentiate along the endocrine lineage.
(167) We first attempted to grow purified duct fragments in Expansion Medium (EM). As shown in
(168) Since we demonstrated that Wnt signaling was activated upon PDL, the effect of addition of a Wnt agonist to freshly isolated pancreatic fragments in vitro on sustained growth was determined. When ducts were isolated from Axin2-LacZ mice, the entire budding cysts stained blue only in the presence of the Wnt agonist Rspondin1 (
(169) To test this notion directly we sorted Axin2-LacZ positive cells from mice 7 days post PDL and found that these cells efficiently initiated budding cysts that were indistinguishable from duct-initiated cysts (
(170) We then attempted to assess the potential of the budding cysts to generate endocrine lineage cells. To this end, we tested a number of changes to the EM to define a Differentiation Medium (DM). A series of factors was tested for their effect on the differentiation into the endocrine lineages. The removal of FGF10 seemed to be crucial to the induction of differentiation. Only in the absence of FGF10 did the islet like structures appear (
(171) When budding cysts after any period of culture were transferred from EM to DM, the cysts underwent a stereotypic involution process: Progressive inward folding of the wall lead to impaction of the cyst into a smaller compact body with morphological resemblance to an islet (
Example 8: In Vitro Expansion of Human Pancreas Fragments
(172) During embryonic pancreas development, neurogenin3+ or insulin expressing cells were seen in the pancreas ductal network, and it was suggested that pancreas duct cells give rise to endocrine progenitors and consequently mature endocrine cells. It has been shown that human pancreas duct cells differentiate into glucose responsive insulin producing cells in vitro (Bonner-Weir, S. et al. 2000 PNAS), and this finding made pancreas duct cells attractive source for beta cells replace therapy. However, it has been difficult to expand duct cells without losing endocrine differentiation capacity. In the previously reported culture system, human pancreas duct cells lost epithelial property or underwent senescence after 2 weeks up to 5 weeks (Trautmann B. et al., Pancreas vol. 8 248-254). Therefore, there is no robust culture system to expand human pancreas duct cells, which retain endocrine differentiation ability. Taking advantage of establishment of mouse pancreas organoid culture system, here, we attempted to establish human pancreas organoid culture system.
(173) Growth of Human Pancreatic Progenitors in Vitro
(174) Human pancreas was obtained from Leiden University Medical Center, The Netherlands. Importantly, under the same conditions as described for mouse pancreas fragments above (example 7), human freshly isolated pancreas fragments can also be grown in vitro (
(175) Under these expansion conditions, the culture efficiency of the pancreatic fragments was appr. 80%, meaning that 80% of the freshly isolated pancreatic fragments were efficiently expanded in vitro for a longer period of time. As compared with mouse pancreas, acinar tissue more easily forms cyst structures, however, these structures ceased to proliferate within 4 weeks. Pancreas duct cells from larger ductular network more efficiently produce cyst structures and eventually form organoids with bud. The pancreas organoids were splitted 1:5 ratio once per week and maintained in vitro at least 5 weeks without losing proliferation ability.
(176) In summary, we established human pancreas organoids culture system and succeeded in expansion of pancreas duct cells at least 3000 times from original volume. We are optimizing endocrine differentiation culture condition for human pancreas duct cells, and this in vitro approach, once optimized, might be important implications for making beta cell replacement therapy available to a larger number of people with type 1 and 2 diabetes mellitus.
REFERENCES
(177) Abe K., Watanabe S. (1995) Apoptosis of mouse pancreatic acinar cells after duct ligation. Arch. Histol. Cytol. 58:221-9 Apelqvist A., Li H., Sommer L., Beatus P., Anderson D. J., Honjo T., Hrabe de Angelis M., Lendahl U., Edlund H. (1999) Notch signaling controls pancreatic cell differentiation. Nature 400:877-81. Barker N., van Es J. H., Kuipers J., Kujala P., van den Born M., Cozijnsen M., Haegebarth A., Korving J., Begthel H., Peters P. J., Clevers H. (2007) Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449:1003-7. Bonner-Weir, S., and Weir, G. C. (2005). New sources of pancreatic beta-cells. Nat. Biotechnol. 23, 857-861. Bonner-Weir, S., Taneja, M., Weir, G. C., Tatarkiewicz, K., Song, K. H., Sharma, A., and O'Neil, J. J. (2000). In vitro cultivation of human islets from expanded ductal tissue. Proc. Natl. Acad. Sci. USA 97, 7999-8004. Bouwens, L., and Rooman, I. (2005). Regulation of pancreatic beta-cell mass. Physiol. Rev. 85, 1255-1270. Dor, Y., Brown, J., Martinez, O. I., and Melton, D. A. (2004). Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 429, 41-46. Githens S., Schexnayder J. A., Desai K., Patke C. L. (1989) Rat pancreatic interlobular duct epithelium: isolation and culture in collagen gel. In Vitro Cell Dev. Biol. 25:679-88. Gradwohl, G., Dierich, A., LeMeur, M., and Guillemot, F. (2000). neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc. Natl. Acad. Sci. U.S.A. 97, 1607-1611. Gu, G., Dubauskaite, J., and Melton, D. A. (2002). Direct evidence for the pancreatic lineage: NGN3+ cells are islet progenitors and are distinct from duct progenitors. Development 129, 2447-2457. Hao, E., Tyrberg, B., Itkin-Ansari, P., Lakey, J. R., Geron, I., Monosov, E. Z., Barcova, M., Mercola, M., and Levine, F. (2006). Beta-cell differentiation from nonendocrine epithelial cells of the adult human pancreas. Nat. Med. 12, 310-316. Jaks V., Barker N., Kasper M., van Es J. H., Snippert H. J., Clevers H., Toftgrd R. (2008) Lgr5 marks cycling, yet long-lived, hair follicle stem cells. Nat. Genet. 40:1291-9. Lefebvre V. H., Otonkoski T., Ustinov J., Huotari M. A., Pipeleers D. G., Bouwens L. (1998) Culture of adult human islet preparations with hepatocyte growth factor and 804G matrix is mitogenic for duct cells but not for beta-cells. Diabetes 47:134-7 Lustig B., Jerchow B., Sachs M., Weiler S., Pietsch T., Karsten U., van de Wetering M., Clevers H., Schlag P. M., Birchmeier W., Behrens J. (2002) Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors. Lustig B., Jerchow B., Sachs M., Weiler S., Pietsch T., Karsten U., van de Wetering M., Clevers H., Schlag P. M., Birchmeier W., Behrens J. Mol. Cell. Biol. 1184-93. Miralles F., Czernichow P., Ozaki K., Itoh N., Scharfmann R. (1999) Signaling through fibroblast growth factor receptor 2b plays a key role in the development of the exocrine pancreas. Proc. Natl. Acad. Sci. U.S.A. 96:6267-72. Ramiya, V. K., Maraist, M., Arfors, K. E., Schatz, D. A., Peck, A. B., and Cornelius, J. G. (2000). Reversal of insulin-dependent diabetes using islets generated in vitro from pancreatic stem cells. Nat. Med. 6, 278-282. Rooman I., Heremans Y., Heimberg H., Bouwens L. (2000) Modulation of rat pancreatic acinoductal transdifferentiation and expression of PDX-1 in vitro. Diabetologia 2000 July; 43(7):907-14. Rooman I., Lardon J., Flamez D., Schuit F., Bouwens L. (2001) Mitogenic effect of gastrin and expression of gastrin receptors in duct-like cells of rat pancreas. Gastroenterology 121:940-9. Sato T., Vries R. G., Snippert H. J., van de Wetering M., Barker N., Stange D. E., van Es J. H., Abo A., Kujala P., Peters P. J., Clevers H. (2009) Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459:262-5. Schwitzgebel, V. M., Scheel, D. W., Conners, J. R., Kalamaras, J., Lee, J. E., Anderson, D. J., Sussel, L., Johnson, J. D., and German, M. S. (2000). Expression of neurogenin3 reveals an islet cell precursor population in the pancreas. Development 127, 3533-3542. Seaberg, R. M., Smukler, S. R., Kieffer, T. J., Enikolopov, G., Asghar, Z., Wheeler, M. B., Korbutt, G., and van der Kooy, D. (2004). Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages. Nat. Biotechnol. 22, 1115-1124. Suzuki, A., Nakauchi, H., and Taniguchi, H. (2004). Prospective isolation of multipotent pancreatic progenitors using flow-cytometric cell sorting. Diabetes 53, 2143-2152. Teta, M., Rankin, M. M., Long, S. Y., Stein, G. M., and Kushner, J. A. (2007). Growth and regeneration of adult beta cells does not involve specialized progenitors. Dev. Cell. 12, 817-826. Trautmann B., Schlitt H. J., Hahn E. G., Lohr M. (1993) Isolation, culture, and characterization of human pancreatic duct cells. Pancreas 8:248-54. Wang, R. N., Kloppel, G., and Bouwens, L. (1995). Duct- to islet-cell differentiation and islet growth in the pancreas of duct-ligated adult rats. Diabetologia 38, 1405-1411. Xu X., D'Hoker J., Stang G., Bonne S., De Leu N., Xiao X., Van de Casteele M., Mellitzer G., Ling Z., Pipeleers D., Bouwens L., Scharfmann R., Gradwohl G., Heimberg H. (2008) Beta cells can be generated from endogenous progenitors in injured adult mouse pancreas. Cell 132(2):197-207.
Example 9: Culturing of Human Small Intestinal or Colon Crypts In Vitro
(178) As described in examples 1 and 2 it is now for the first time possible to generate long time culture conditions for mouse small intestine and colon epithelium. Crypt-villus organoids grow through the supplementation of a set of divined growth factors and an extracellular matrix. The organoids contain intestinal stem cells that actively divide and giving rise to all major differentiated cell lineages present in the intestine. In this example we show that these culture conditions are not unique to the mouse intestinal epithelium but can also be used to grow human intestinal epithelium.
(179) Material and Methods
(180) Mouse Colon Organoid Cultures
(181) Mouse organoid cultures were cultured as described in example 1. Inhibitor of Wnt production (IWP-2) was used to inhibit Wnt secretion (Chen et al., Nat. Chem. Biol. 2009 February; 5(2):100-7).
(182) Human Colon Organoid Cultures
(183) Human colon crypts were isolated from resected normal colonic specimen and cultured as organoid structures for 7 days using the established organoid culture system (Sato et al., 2009 Nature May 14; 459(7244):262-5). Since this protocol was optimized for mouse derived organoid cultures, we made a small change by the addition of Wnt3a conditioned medium, in order to ensure optimal growth of the human colon organoids. To obtain this conditioned medium, Wnt3a is expressed in a cell line by transfecting a suitable expression construct encoding the ligand. the cell line is cultured and the culture medium comprising the secreted ligand is harvested at suitable time intervals. For example, cells start the production of Wnt3a at the moment they reach confluency and stop growing. Culture medium from cells that were not transfected or infected with the empty expression construct was used as a negative control. The conditioned medium was harvested and tested, for example in an assay wherein luciferase expression in controlled by TCF responsive elements to quantitate the presence of a Wnt agonist such as Wnt3a (Korinek et al., 1997, Science 275:1784-1787).
(184) Results
(185) The proliferation of the intestinal epithelium is dependent on the Wnt signaling pathway. The exact location of the Wnt source is however unclear (Gregorieff and Clevers, 2005, Genes Dev. April 15; 19(8):877-90). Since the mouse intestinal organoids grew in a niche independent fashion (Sato et al., 2009 Nature May 14; 459(7244):262-5) we assumed that these organoids may produce their own Wnt ligands. To test this we inhibited Wnt secretion through incubation with a porcupine inhibitor. Porcupine is important for the Wnt secretion (schematic
(186) We next tried to culture human intestinal organoids. It turned out that the addition of Wnt3a to the medium was necessary because without, crypt organoids never formed budding structures and died within 5-10 days for the small intestine and in 3-4 days for the colon (
Example 10: Optimized Culture Conditions for the Growth of Stomach Organoids
(187) As described in example 5a culture medium has been identified which can be used to culture stomach epithelium for long periods. Here we describe the optimized conditions for these stomach organoid cultures.
(188) Materials and Methods
(189) Gastric Unit Isolation, Single Cell Dissociation and EGFP.sup.+ve Cell Sorting
(190) Gastric glands were isolated from mouse pylorus regions as previously described with some modifications (Bjerknes and Cheng, 2002, Am. J. Gastrointest. Liver Physiol. September; 283(3):G767-77). Briefly, under the microscope, the stomach was opened along the greater curvature, washed with saline solution and the pylorus isolated. The muscular layer of the stomach was removed and the remaining epithelia was divided into 5 mm pieces and incubated for 3-5 h in a buffered saline solution (Na2HPO4 28 mM, KH2PO4 40 mM, NaCl 480 mM, KCl 8 mM, Sucrose 220 mM, D-Sorbitol 274 mM, DL-Dithiotreitol 2.6 mM) containing 10 mM EDTA (for culturing or staining) or 5 mM EGTA (for RNA isolation) at 4 C. After removal of the chelating agent, the tissue fragments were vigorously suspended in the buffered solution using a 10 ml pipette. After suspension and centrifugation, the sediment was enriched in gastric glands. After gland isolation, cells were collected and resupended in calcium-free SMEM medium (Invitrogen), supplemented with 10 mg/ml Trypsine and 0.8 Units/l DNAse I (for microarray analysis) or resuspended in TrypleExpress (GIBCO) supplemented with 0.8 Units/ul DNAase (for culturing purposes). In both cases, after incubation at 37 C. for 20-25 minutes, cells were spun down, and filtered through a 40 M mesh. EGFPhi and EGFPlo cells were sorted by flow cytometry (MoFlo, Beckman Coulter). Single viable epithelial cells were gated by forward scatter and pulse-width parameter. Where stated, cells were either gated for negative staining of propidium iodide, collected in Trizol LS (Invitrogen) and RNA isolated according manufacturers' protocol or collected in gastric culture medium, embedded in Matrigel (BD Bioscience) and cultured according to the protocol detailed below.
(191) Gastric Culture
(192) For culturing, isolated gastric glands were counted and a total of 100 glands mixed with 50 ul of Matrigel (BD Bioscience) and plated in 24-well plates. After polymerization of Matrigel, gastric culture medium (Advanced DMEM/F12 supplemented with B27, N2 and nAcetylcistein (Invitrogen) containing growth factors (50 ng/m EGF (Peprotech), 1 ug/ml R-spondin 1, 100 ng/ml Noggin (Peprotech), 100 ng/ml FGF10 (Preprotech) and Wnt3A conditioned media) was overlaid. For the single cell culture, a total of 100 sorted EGFP.sub.hi cells/well were collected in gastric culture medium and embedded in Matrigel (BD Bioscience). After polymerization of Matrigel, gastric culture media was overlaid. For the first 2 days after seeding, the media was also supplemented with 10 M ROCK inhibitor Y-27632 (Sigma Aldrich), to avoid anoikis. Growth factors were added every second day and the entire medium was changed every 4 days. For passage, gastric organoids were removed from Matrigel, mechanically dissociated and transferred to fresh Matrigel. Passage was performed every 1-2 weeks with a 1:5-1:8 split ratio. To confirm the Wnt3A requirement, mouse Wnt3A recombinant protein (Stem cell technologies) was supplemented instead of the Wnt3A conditioned media. For the in-vitro tracing experiments, 2-week old gastric organoids were incubated with 100 nM of 4-hydroxytamoxifen in gastric culture medium for 20 h to activate Lgr5-CreERT2. YFP was subsequently visualized and recorded in live organoids using confocal microscopy (Leica, SP5).
(193) Wnt3a Conditioned Media
(194) The Wnt3a media was prepared following protocol described elsewhere (Willert et al., 2003, Nature, May 22; 423(6938):448-52). The TOP/FOP assay was used to test the Wnt activity of the Wnt3a conditioned media and the Control conditioned media, as described by van de Wetering and colleagues (van de Wetering et al., 2001 Cancer Res. January 1; 61(1):278-84). A TOP/FOP ratio 50 was considered high Wnt media and diluted 1:1 with the gastric organoid culture media. A 1:10 dilution of this high Wnt3a media (TOP/FOP ratio 5) was considered low Wnt media and used for differentiation purposes.
(195) Gastric Organoid Immunohistochemistry
(196) For immunohistochemistry gastric organoids were washed once with PBS and immediately fixed with Paraformaldehyde 4% for 15-20 min at RT. When stated gastric organoids were embedded in paraffin and processed using standard techniques. For whole mount staining, samples were permeabilized with PBS 0.5% Triton-X100-1% BSA and incubated o/n with the primary antibodies. Following several washes in PBS 0.3% Triton X100, samples were incubated with the secondary antibody. EdU staining was performed following manufacturer's instructions (Click-IT; Invitrogen). Nuclei were stained with TOPRO3 iodine or Hoescht33342. The images of gastric glands and gastric organoids were acquired using confocal microscopy (Leica, SP5). Three-dimensional reconstruction was performed using Volocity Software (Improvision).
(197) RT-PCR
(198) RNA was extracted from gastric cell cultures or freshly isolated tissue using the RNeasy Mini RNA Extraction Kit (Qiagen) and reverse-transcribed using Moloney Murine Leukemia Virus reverse transcriptase (Promega). cDNA was amplified in a thermal cycler (GeneAmp PCR System 9700; Applied Biosystems, London, UK) as previously described (Huch et al., 2009). Primers used are shown below (Gene symbol followed by Forward (5-3) and Reverse (5-3) primers).
(199) TABLE-US-00003 Lgr5: GGAAATGCTTTGACACACATTC, (SEQIDNO:24) GGAAGTCATCAAGGTTATTATAA (SEQIDNO:25) Gif: TGAATCCTCGGCCTTCTATG, (SEQIDNO:26) CAGTTAAAGTTGGTGGCACTTC (SEQIDNO:27) Pgc: CCAACCTGTGGGTGTCTTCT, (SEQIDNO:28) TTAGGGACCTGGATGCTTTG (SEQIDNO:29) Muc6: TGCATGCTCAATGGTATGGT, (SEQIDNO:30) TGTGGGCTCTGGAGAAGAGT (SEQIDNO:31) Muc5ac: CCATGAAGTGGGAGTGTGTG, (SEQIDNO:32) TTGGGATAGCATCCTTCCAG (SEQIDNO:33) Ghrl: GCCCAGCAGAGAAAGGAATCCA, (SEQIDNO:34) GCGCCTCTTTGACCTCTTCC (SEQIDNO:35) Gast: GCCAACTATTCCCCAGCTCT, (SEQIDNO:36) GGCTCTGGAAGAGTGTTGCT (SEQIDNO:37) Stt: GAGGCAAGGAAGATGCTGTC, (SEQIDNO:38) GGGCATCATTCTCTGTCTGG (SEQIDNO:39) Muc2: GAACGGGGCCATGGTCAGCA, (SEQIDNO:40) CATAATTGGTCTTGCATGCC (SEQIDNO:41) Cdx2: CTTGCTGCAGACGCTCAAC, (SEQIDNO:42) TCTGTGTACACCACCCGGTA (SEQIDNO:43) Hprt: AAGCTTGCTGGTGAAAAGGA, (SEQIDNO:44) TTGCGCTCATCTTAGGCTTT (SEQIDNO:45)
Results
(200) To determine optimal growth of gastic units in vitro we isolated gastric gland units which were suspended in Matrigel and cultured under different conditions. Gastric culture growth conditions were similar to those of the small intestine cultures (including EGF, Noggin and R-spondin 1), except for a strict dependence on Wnt3A in the form of conditioned media. This requirement was confirmed using purified Wnt3a protein (
(201) To determine whether gastric Lgr5.sup.+ve cells (
Example 11
(202) To address the need for improved culture media and methods for human epithelial stem cells, the inventors investigated signaling pathways that are known to be subverted in certain cancers, e.g., colorectal cancer. It was hypothesized that these pathways, which affect cell fate in cancer, may also play a role in determining cell fate under in vitro cell culture conditions.
(203) In a first screening experiment, a series of vitamins, hormones and growth factors were tested in combination with standard stem cell culture media. Gastrin and nicotinamide were identified as resulting in significantly improved culture conditions. Incorporating these factors into the standard culture conditions, a second screening experiment was performed, in which certain small molecule inhibitors related to relevant signaling pathways, such as ERK, p38, JNK, PTEN, ROCK, and Hedgehog, were tested. Whilst there is reasonable basis for choosing to test these compounds, as described above, it is to be emphasized that in the present state of the art, there would be no reasonable way to predict what the outcome of each of these additional compounds would be on the culture medium properties.
(204) TABLE-US-00004 TABLE 1 List of reagents used for optimization of human intestinal organoids culture Description Source Concentration Activity* First screening (WENR**) Hormones, vitamins etc Hydrocortison Sigma 500 nM 0 Gastrin*** Sigma 1 uM 1+ Exendin4 GLP1 analog Sigma 100 nM 0 Nicotinamide Vitamin B derivative Sigma 10 mM 3+ L-Ascorbic acid Vitamin C Sigma 10 uM 0 anti-oxidant mixture Sigma 1x 0 Lipid mixture Sigma 1x 0 PGE2 Sigma 10 uM 1+ (Cystic) Cholera Toxin Sigma 100 nM 1+ (Cystic) Growth factors BDNF Peprotech 100 ng/ml 0 GDNF Peprotech 100 ng/ml 0 FGF2 Peprotech 100 ng/ml 0 FGF10 Peprotech 100 ng/ml 0 Follistatin Peprotech 100 ng/ml 0 Cyr61 Peprotech 1 ug/ml 0 LIF Millipore 1000 U/ml 0 Second screening (WENR + gastrin + Nicotinamide) Small molecule inhibitors PD98059 ERK inhibitor Sigma 10 uM 1 SB203580 p38 inhibitor Sigma 1-10 uM 2+ SB202190 p38 inhibitor Sigma 1-10 uM 2+ SP600125 JNK inhibitor Sigma 10 uM 0 PS48 PDK1 activator Sigma 5 uM 0 Y27632 ROCK inihibitor Sigma 10 uM 1+ cystic Cyclopamine Hedgehog inhibitor Sigma 100 nM 1 5 Azacytidin DNA methylase inhibitor Stemolecule 1 Dorsomorphin BMP inhibitor Stemolecule 0 A83-01 ALK4,5,7 inhibitor Tocris 50 n-1 uM 3+ VO-OHpic trihydrate PTEN inhibitor Sigma 500 nM 3 Pifithrin- p53 inhibitor Sigma 0 BIX01294 G9a HMTase inhibitor Stemolecule 1 *Activity scale (plating efficiency was compared with control after 4 days culture): 0 = no change; 1+ = <50% increase; 2+ = 50-100% increase; 3+ = >100% increase; 1 = 0-50%; 2 = 50-100% decrease; 3 = >100% decrease. **WENR comprises EGF + Noggin + R-spondin + Wnt-3a ***Highlighted in bold are the compounds which showed the greatest improvement to the culture medium.
(205) In summary, the inventors have established long-term culture conditions under which single crypts or stem cells derived from murine small intestine (SI) expand over long periods of time. Growing crypts undergo multiple crypt fission events, whilst simultaneously generating villus-like epithelial domains in which all differentiated cell types are present. The inventors have now adapted the culture conditions to grow similar epithelial organoids from mouse colon and human SI and colon. Based on the murine small intestinal culture system, the inventors optimized the murine and human colon culture system. They found that addition of Wnt3A to the growth factor cocktail allowed mouse colon crypts to expand indefinitely. Further addition of nicotinamide, a small molecule Alk inhibitor and a p38 inhibitor was preferable for long-term human SI and colon culture. The culture system also allowed growth of murine Apc.sup.min adenomas, human colorectal cancer and human esophageal metaplastic Barrett's epithelium. The culture technology should be widely applicable as a research tool for infectious, inflammatory and neoplastic pathologies of the human gastrointestinal tract. Moreover, regenerative applications may become feasible with ex vivo expanded intestinal epithelia. Self-renewal of the small intestinal and colonic epithelium is driven by the proliferation of stem cells and their progenitors located in crypts. Although multiple culture systems have been described (Evans G S et al. J. Cell Sci. 1992; 101 (Pt 1):219-31; Fukamachi H., J. Cell Sci. 1992; 103 (Pt 2):511-9; Perreault N & Jean-Francois B., Exp. Cell. Res. 1996; 224:354-64; Whitehead R. H. et al., Gastroenterology 1999; 117:858-65), only recently have long-term culture systems become available that maintain basic crypt physiology. Two different protocols were published which allow long-term expansion of murine small intestinal epithelium. Kuo and colleagues demonstrated long-term growth of small fragments containing epithelial as well as stromal elements in a growth factor-independent fashion (Ootani A et al. Nat. Med. 2009; 15:701-6). The inventors designed a culture system for single stem cells by combining previously defined insights in the growth requirements of intestinal epithelium. Wnt signaling is a pivotal requirement for crypt proliferation (Korinek V. et al. Nat. Genet. 1998; 19:379-83; Pinto D. et al., Genes Dev. 2003; 17:1709-13; Kuhnert F. et al., Proc. Natl. Acad. Sci. U.S.A. 2004; 101:266-71) and the Wnt agonist R-spondin1 induces dramatic crypt hyperplasia in vivo (Kim K. A. et al., Science 2005; 309:1256-9). Second, EGF signaling is associated with intestinal proliferation (Dignass AU & Sturm A., Eur. J. Gastroenterol. Hepatol 2001; 13:763-70). Third, transgenic expression of Noggin induces expansion of crypt numbers (Haramis A. P. et al. Science 2004; 303:1684-6). Fourth, isolated intestinal cells undergo anoikis outside the normal tissue context (Hofmann C. et al., Gastroenterology 2007; 132:587-600). Since laminin (1 and 2) is enriched at the crypt base (Sasaki T. et al., Exp. Cell Res. 2002; 275:185-), the inventors explored laminin-rich Matrigel to support intestinal epithelial growth. Matrigel-based cultures have successfully been used for growth of mammary epithelium (Stingl J. et al., Breast Cancer Res. Treat. 2001; 67:93-109). Under this culture condition (R-spondin1, EGF, and Noggin in Matrigel), the inventors obtained ever-expanding small intestinal organoids, which displayed all hallmarks of the small intestinal epithelium in terms of architecture, cell type composition and self-renewal dynamics.
(206) Despite extensive efforts, long-term adult human intestinal epithelial cell culture has remained difficult. There have been some long-term culture models, but these techniques and cell lines have not gained wide acceptance, possibly as a result of inherent technical difficulties in extracting and maintaining viable cells (Rogler G. et al., Scandinavian Journal of Gastroenterology, 2001; 36:389-98; Buset M. et al., In vitro cellular & developmental biology: journal of the Tissue Culture Association 1987; 23:403-12; Whitehead R. H. et al., In vitro cellular & developmental biology: journal of the Tissue Culture Association 1987; 23:436-42; Deveney C. W. et al., The Journal of surgical research 1996; 64:161-9; Pang G. et al., Gastroenterology 1996; 111:8-18; Latella G. et al., International Journal of Colorectal Disease 1996; 11:76-83; Panja A. Laboratory investigation; a journal of technical methods and pathology 2000; 80:1473-5; Grossmann J. et al., European Journal of Cell Biology 2003; 82:262-70). Encouraged by the establishment of murine small intestinal culture, the inventors aimed to adapt the culture condition to mouse and human colonic epithelium. The inventors now report the establishment of long-term culture protocols for murine and human colonic epithelium, which can be adapted to primary colonic adenoma/adenocarcinoma and Barrett's esophagus.
(207) Results
(208) Establishment of a Mouse Colon Culture System
(209) In an attempt to establish a mouse colon culture system, the inventors explored our small intestinal culture condition (here termed ENR: EGF+Noggin+R-spondin). In our experience, initial growth of colon epithelium is often observed under the ENR culture condition, but is invariably abortive. Organoid formation was studied using epithelium isolated from the distal part of the mouse colon. Under ENR conditions, the plating efficiency of single distal colonic crypts was much lower than that of small intestine (1-3% vs >90%) and these organoids could not be passaged. Recently, the inventors have shown that Paneth cells produce several Wnt ligands (Gregorieff A. et al. Gastroenterology 2005; 129:626-38), and that the production of Wnt by these Paneth cells is essential to maintain intestinal stem cells (Sato T. et al., Nature 469:415-8). To determine the Wnt signaling status in colon organoids, the inventors cultured colon crypts from Axin2-lacZ mice, (a faithful Wnt reporter) (Lustig B. et al., Mol. Cell. Biol. 2002; 22:1184-93) or Lgr5-GFP knock-in mice (Lgr5 being a Wnt-dependent stem cell marker)(Barker N. et al. Nature 2007; 449:1003-7).
(210) Freshly isolated colon crypts readily expressed Axin2-LacZ or Lgr5-GFP at their bottoms, but they lost expression of the Wnt reporters shortly after initiation of culture (
(211) Mouse small intestinal epithelium grown under ENR conditions generates all differentiated epithelial cell types concomitant with stem cell self-renewal. The inventors have shown previously that the addition of Wnt3A to these cultures interferes with intestinal differentiation and yields organoids that largely consist of undifferentiated progenitors (Sato T. et al., Nature 469:415-8). This is not unexpected given the central role of Wnt signaling in the maintenance of the undifferentiated crypt progenitor state (van de Wetering M. et al. Cell 2002; 111:241-50). Consistent with this observation, colonic organoids in WENR condition failed to differentiate properly. Upon withdrawal of Wnt-3A, the inventors observed differentiation along all epithelial lineages (
(212) Establishment of Human Colon Culture System
(213) Encouraged by the success of the improved mouse colon crypt culture, the inventors applied the culture condition to human colon crypts. Although these crypts initially survived, most subsequently disintegrated within 7 days. To increase the plating efficiency of human colon crypts, the inventors screened candidate growth factors, hormones and vitamins (list in
(214) The inventors assumed that growth arrest occurred because of inadequate culture conditions rather than a cell-intrinsic property of senescence/replicative aging. The inventors therefore extended our attempts to optimized the culture condition. The inventors screened various small molecule modulators of MAP kinases, of signaling molecules mutated in colon cancer, and of histone modifiers (
(215) These results implied that Alk receptor and p38 signaling negatively regulate long-term maintenance of human intestinal epithelial cells. The inventors refer to the optimized culture condition as the HISC (Human intestinal stem cell culture) condition.
(216) Human Intestinal Organoids Mimic in Vivo Differentiation
(217) Under the HISC condition, the inventors failed to observe differentiated cells. As was seen in the mouse colon organoids, withdrawal of Wnt was required for mature enterocyte differentiation in human organoids (
(218) The same differentiation inhibitory effects of Wnt, Nicotinamide and SB202190 were observed in human small intestinal organoids. Lysozyme+Paneth cells were observed in small intestinal organoids, but not in colonic organoids (
(219) The inventors further examined the response of human intestinal organoids to Notch-inhibition. The inventors have previously shown that Notch inhibition with either -secretase inhibitors (dibenzazepine; DBZ) or by conditional targeting of the Notch pathway transcription factor CSL depleted intestinal stem cells, terminated intestinal epithelial proliferation and induced goblet cell hyperplasia in vivo (van Es J. H. et al., Nature 2005; 435:959-63). Indeed, upon treatment with DBZ, the intestinal organoids ceased their proliferation and most cells converted into goblet cells within 3 days (
(220) Establishment of APC-Deficient Adenoma and Colon Adenocarcinoma
(221) Recently, the inventors reported efficient mouse intestinal adenoma formation from Lgr5 stem cells in Lgr5-GFP-ires-CreERT2APC.sup.flox/flow mice upon Tamoxifen-induced Cre activation_ENREF_32 (Barker N. et al., Genes Dev. 2008; 22:1856-64). The inventors isolated the intestinal adenomas 10 days after induction and optimized the culture condition. The adenomas efficiently formed cystic organoid structure without budding. Since APC loss constitutively activates the Wnt pathway, the inventors expected that R-spondin1 would become dispensable for adenoma organoid growth. This was indeed observed. Furthermore, Noggin, which is essential for long-term culture of normal small intestine, was dispensable in adenoma organoids. Interestingly, the inventors observed a loss of Lgr5-GFP but not Axin2-LacZ in adenomatous organoids 7 days after withdrawal of Noggin (
(222) Many colorectal cancer cell lines have been isolated over the past four decades. Typically, such cell lines emerge as rare, clonal outgrowths after primary cultures of colon tumors enter tissue-culture crisis. Currently, no robust culture system exists which allows the consistent culture of primary human colon cancer samples without culture crisis and the consequent clonal outgrowth of culture-adapted cells. As an obvious next step, the inventors applied intestinal adenoma culture conditions to human colorectal cancer samples. As expected, colon cancer cells required neither R-spondin nor Noggin. EGF was dispensable in most colon cancer organoids, while some colon cancer organoids decelerated their proliferation after withdrawal of EGF. Distinct from mouse intestinal adenoma, colorectal cancer organoids in the culture condition grew as irregular compact structures rather than as simple cystic structures (
(223) The inventors examined the proliferation/differentiation status of adenoma and colon cancer organoids. As expected, most of cells were Ki67+. Consistent with the strong inhibitory effect of Wnt on enterocyte differentiation (
(224) Culturing Human Metaplastic Barrett's Epithelium
(225) Barrett's Esophagus is marked by the presence of columnar epithelium in the lower esophagus, replacing the normal squamous cell epithelium as a result of metaplasia (Odze R. D. Nat. Rev. Gastroenterol. Hepatol. 2009; 6:478-90). The histological hallmark of Barrett's Esophagus is the presence of intestinal goblet cells in the esophagus. Exploiting the similarity between Barrett and intestinal epithelium, the inventors subjected small Barrett's epithelium (BE) biopsies to the human colon culture condition. Under these culture conditions, normal esophageal squamous cells transiently proliferated for 1 week, but the organoids could not be passaged. Barrett's Esophagus epithelium could be maintained for up to 1 month under HISC conditions (
(226) Discussion
(227) The protocols developed here allow robust and long-term culture of primary human epithelial cells isolated from small intestine, colon, adeno(carcino)mas and Barrett's Esophagus (table 2).
(228) TABLE-US-00005 TABLE 2 List of components of the organoid culture systems Reagent name Supplier Cat No. Solvent Stock solution Final conc. Matrigel, GFR, phenol BD bioscience 356231 free Advanced DMEM/F12 Invitrogen 12634-028 GlutaMAX-I Invitrogen 35050-079 200 mM 2 mM HEPES 1M Invitrogen 15630-056 10 mM Penicillin/Streptomycin Invitrogen 15140-122 10000/10000 U/ml 100/100 U/ml N2 supplement Invitrogen 17502-048 100x 1x B27 supplement Invitrogen 17504-044 50x 1x N-Acetylcysteine Sigma-Aldrich A9165-5G DW 500 mM = 81.5 mg/ml 1 mM EDTA Sigma-Aldrich 431788-25g DW 500 mM = 2 mM 14.6 g/100 ml Mouse recombinant Peprotech 250-38 PBS/BSA 100 mg/ml 100 ng/ml noggin 100 ug mouse recombinant Invitrogen PMG8043 PBS/BSA 500 mg/ml 50 ng/ml EGF human recombinant R- Nuvelo PBS/BSA 1 mg/ml 1 mg/ml spondin human recombinant Peprotech 100-26 PBS/BSA 100 mg/ml 100 ng/ml FGF10 mouse recombinant Millipore GF-160 PBS 10 mg/ml 100 ng/ml Wnt-3A Y-27632 Sigma-Aldrich Y0503 PBS 10 mM = 1 g/338 ml 10 mM A-83-01 Tocris 2939 DMSO 500 mM 500 nM SB202190 Sigma-Aldrich S7067 DMSO 30 mM 10 mM Nicotinamide Sigma-Aldrich DW 1M 10 mM [Leu15]-Gastrin I Sigma-Aldrich G9145 PBS/BSA 100 mM 10 nM DNase Sigma-Aldrich DN25-1g PBS 200000 U/ml 2000 U/ml TrypLE express Invitrogen 12605-036 Collagenase type XI Sigma-Aldrich C9407 Dispase Invitrogen 17105-041 70 um Cell strainer BD falcon 352350 All stock solutions and aliquoted Matrigel are stored in 20 C.
(229) In contrast to murine small intestine, murine colonic epithelial cells require Wnt ligand in the culture medium. The inventors have previously reported that CD24.sup.hi Paneth cells produce Wnt-3/11, which are essential for stem cell maintenance in small intestine (Sato T., et al., Nature 2011; 469:415-8). Wnt-6 and -9b mRNA are expresses at the bottom of colon crypts (Gregorieff A., et al. Gastroenterology 2005; 129:626-38.). It remains undetermined whether this local Wnt production by colon crypt base cells is sufficient to activate canonical Wnt signal in vivo or there is another source of Wnt ligand in colon mucosa. The difference between human and mouse intestinal organoid culture conditions was unexpectedly large. A83-01 inhibits ALK4/5/7, receptors that are detected in both murine and human crypts by microarray. The inventors are currently investigating the mechanism by which ALK signal regulates human organoid growth. The inventors have not observed cellular transformation in long-term cultures and no chromosomal changes become obvious under the optimized culture conditions. Furthermore, the organoids can undergo a considerably higher number of cell division than reported for other adult human epithelial culture system (Dey D. et al., PloS one, 2009; 4:e5329; Garraway I. P. et al., The Prostate 2010; 70:491-501). It is generally believed that somatic cells are inherently limited in their proliferative capacity, a phenomenon called replicative aging (Walen K. H., In vitro cellular & developmental biology, Animal 2004; 40:150-8). Most normal human cells are believed to count the number of times they have divided, eventually undergoing a growth arrest termed cellular senescence. This process may be triggered by the shortening of telomeres, and the consequent activation of DNA damage signals (M1), or telomere attrition (M2). In the absence of the two small molecule kinase inhibitors, human intestinal organoids underwent growth arrest after 10-20 population doublings. By contrast, the replicative capacity in the optimized culture condition was extended at least up to 100 population doublings upon addition of the inhibitors, which exceeded the Hayflick limit (Hayflick L. The Journal of Investigative Dermatology 1979; 73:8-14). This result clearly indicates that the senescent phenotype seen in the first culture system reflects inadequate growth conditions, rather than inherent replicative aging.
(230) The culture techniques can be used to study basic aspects of stem cell biology and the control of differentiation, exemplified by depletion of stem cells and goblet cell differentiation upon Notch inhibitor treatment. Moreover, the organoid culture platform may be used for pharmacological, toxicological or microbiological studies on pathologies of the intestinal tract, as the organoids represent more closely the intestinal epithelium than often-used colon cancer cell lines such as CaCo2 or DLD1. Lastly, since small biopsies taken from adult donors can be expanded without any apparent limit or genetic harm, the technology may serve to generate transplantable epithelium for regenerative purposes.
Example 12: Culturing Mouse Pancreatic Organoids
(231) The use of a TGF-beta inhibitor was also tested in a culture medium for mouse pancreatic organoids. The expansion medium that was used was DMEM/F12 media (supplemented with P/S, Glutamax, 10 mM Hepes, B27, N2 and N-Acetylcysteine), EGF (50 ng/ml), R-spondin (10%), Noggin (100 ng/ml), FGF10 (100 ng/ml), A8301 (TGF-beta inhibitor, 500 nM) and Gastrin (10 M). This differs slightly from that of the HISC culture used in Example 16 in that there is no Wnt agonist (other than Rspondin) or Nicotinamide and FGF10 is added. However, these culture media share a number of key components (ENR+gastrin+TGF-beta inhibitor), the addition of the TGF-beta inhibitor being advantageous in both cases. Pancreas organoids grown in these conditions could be expanded for >3 months and passaged at least 5 times.
(232) Microarray experiments were carried out for the pancreas organoids grown in the above-described expansion medium and the results were compared to the adult pancreas, adult liver and newborn liver (see
(233)
(234) The expression levels of genes important for the generation of insulin-producing cells are low. However, it is clear that in the expansion medium, proliferation and expression patterns of the pancreatic organoids closely resemble those seen in early progenitor endocrine cells.
(235) The pancreas is mainly formed by three different cell types: acinar cells, ductal cells and endocrine cells. In a total RNA sample of adult pancreas, 90% of the RNA comes from acinar cells, so the expression levels of endocrine markers are very diluted in a total pancreas sample. Therefore, further experiments are planned for each specific cell type. For example, the inventors plan to carry out a microarray comparison between pancreas organoids, enriched acinar cell preparation, enriched ductal cell preparation and enriched endocrine cell preparation, to have a better estimation of the mRNA levels of the important genes in our pancreas organoids compared with the levels present in insulin producing cells. For example, in an enriched endocrine cell sample, 75-85% of the cells present would be insulin-secreting cells).
Example 13: The Effect of Noggin on the Expansion Medium
(236) To investigate the role of the BMP inhibitor, Noggin, in the expansion medium, the inventors compared mRNA levels of early endocrine markers and ductal markers in pancreatic organoids that have always been cultured in EGFRA medium so have never been cultured in the presence of Noggin with the level of expression of the same markers in organoids that have always been cultured in EGFRAN medium (i.e., always in the presence of Noggin). The inventors also compared mRNA levels of these markers in pancreatic organoids from which Noggin was added or removed from the cultures respectively. Specifically, one sample of pancreatic organoids was cultured in EGFRA medium and then Noggin was added and the organoids were cultured for a further 2 or 4 days. Another sample of pancreatic organoids was cultured in EGFRAN medium and then Noggin was removed and the organoids were cultured for a further 2 or 4 days. The gene expression was compared and the results are shown in
(237) The effect of the presence or absence of Noggin, or its addition or withdrawal to EGFRA medium on Lgr5 gene expression was assessed using pancreatic organoids obtained from pancreatic ducts. The results in
(238) This is surprising, because in the literature it is described that BMP activity is useful for culture of pancreatic cells. This conclusion is based on the observations that BMP signaling is required for the differentiation into both the ductal (see keratin7 and 19 expression) and endocrine cells. Thus, the skilled person would expect the inclusion of a BMP inhibitor, such as Noggin, to be disadvantageous. However, the inventors surprisingly found that the use of a BMP inhibitor was advantageous because it resulted in more proliferative organoids and higher expression of Lgr5.
Example 14: Transplantation of Human Pancreatic Organoids Under the Kidney Capsule in Mice
(239) Pancreatic organoids, that had been expanded using the protocol described in example 11 (see
(240) Just before transplantation, organoids were treated with cell recovery solution (BD#354253, BD Biosciences) to get rid of matrigel residues. Organoids were washed several times with PBS and pelleted.
(241) Transplantation of these organoids under the renal capsule of immunodeficient recipients was carried out using an NIH recommended procedure for islet transplantation under the kidney capsule (Purified Human Pancreatic Islets, In Vivo Islets Function, Document No. 3104, A04, Effective Date 7 Jul. 2008, DAIT, NIAID, NIH). A week before the transplantation, hyperglycemia was chemically induced in the recipient mice (NOD/SCID/IL2RgammaKO a.k.a. NSG) with a high dose 130 mg/kg streptozotocin injection. Blood glucose levels were monitored and mice having a blood glucose above 18 mmol/l were considered hyperglycaemic.
(242) For transplantation, the hyperglycemic recipient was anesthetized and a small incision was made in the left flank to expose the left kidney. Approximately 2.5-3.0 mm.sup.3 of organoids were collected in a siliconized PESO transplantation tube and transplanted under the kidney capsule using a Hamilton syringe. After cauterizing the damaged capsule the kidney was placed back into the abdominal cavity. The peritoneum and the skin were then closed with 5-0 silk sutures.
(243) One mouse was sacrificed three hours post-transplantation and the graft was analyzed for mature beta cell and progenitor markers. In this mouse, no insulin-producing cells could be seen in the murine peri-renal capsule (
(244) A further mouse was allowed to recover in the cage with a heat pad, under close supervision. Bodyweights and blood glucose levels of the transplanted mouse were monitored for 1 month. After one month the mouse was sacrificed and the graft was analyzed for mature beta cell and progenitor markers.
(245) 1 month after transplantation, a number of insulin-producing cells could be identified. These insulin-producing cells are all the stained cells in
(246) The finding that the insulin producing cells are present 1 month after transplantation but are not present 3 hours after transplantation demonstrates that the insulin producing cells largely or only arise after transplantation.
(247) These results show that cells taken from pancreatic organoids of the present invention, cultured with the media and methods of the present invention, can be transplanted into mice and can promote the growth of insulin-producing cells in the pancreas. Excitingly, human pancreatic organoids could be transplanted. This opens a number of exciting possibilities for using transplanted organoid cells to promote insulin production, e.g., for treatment of diabetes.
Example 15: An Expansion Medium for Liver Organoid Growth and Expansion
(248) After isolation, biliary ducts (see
(249) We analyzed the expression of the Wnt target genes Axin2 and Lgr5 in the cultures. Cultures of both Axin2LacZ and Lgr5-LacZ livers revealed the presence of Axin2- and Lgr5-positive cells in the liver organoids 1 month after seeding, thus confirming that the Wnt signaling is active and required for culture growth (
(250) When single Lgr5 cells from a Lgr5LacZ or LgrSGFP mouse were sorted, single colonies grew into organoids. These cultures also express markers of cholangiocyte and hepatocyte lineages and have been maintained and regularly split into 1:6-1:8 for more than 4 months (see
(251) Having established that the liver organoids are derived from Lgr5+ve cells we set out to determine their individual gene signature as compared to the adult liver signature. RNA was isolated from adult liver and from liver organoids grown in ER or ENRW media supplemented with FGF10, Nicotinamide and Hepatocyte Growth Factor. The genetic signature of the adult liver and the 2 liver culture conditions was subsequently derived via comparative gene expression profiling in respect to the expression of a Universal RNA reference. The use of the same reference RNA for the hybridization to all the samples allowed us to compare the 3 independent samples among them (adult liver, ER and ENRW). The heat map analysis revealed that the expression profile of both culture conditions highly resemble the adult liver tissue expression profile, whereas they do not share the same profile when compared to muscle or adipose tissue profile (see
(252) Also, in both culture conditions as well as in the adult liver, non-specific adult liver genes as AFP, and non-liver transcription factors as Pdx1 or NeuroD can be detected.
(253) It is remarkable that, in both culture conditions but not in the adult liver, the stem cell marker Lgr5 was one of the most highly enriched genes in the liver culture signature. Also, cell markers of progenitor populations in small intestine and stomach as Cd44 and Sox9 (Barker & Huch et al., Cell stem cell 2010) were highly expressed in both culture conditions but not in adult liver, indicating again the self-renewal capacity of the liver cultures as well as the quiescent status of the normal adult liver.
(254) Additionally, apart from Lgr5, multiple Wnt target genes were also highly upregulated in the liver cultures compared to the adult liver including MMP7, Sp5 and Tnfrs19, among others, providing strong evidence of the requirement of an active and robust canonical Wnt signaling activity to maintain the self renewing capacity of the cultures.
Example 16: An Improved Differentiation Medium
(255) Under ER or ENRW conditions the liver cultures self-renew, and can be maintained and expanded in a weekly basis, for up to 1 year (
(256) The combination of EGF (50 ng/ml) and R-spondin 1 (1 ug/ml) supplemented with FGF10 (100 ng/ml), HGF (25-50 ng/ml) and Nicotinamide (1-10 mM), were preferable for the long term maintenance of the cultures. Under these conditions, we obtained long-lived cell cultures that express biliary duct and some hepatoblast or immature-hepatocyte markers (Glu1, Albumine). However, the number of cells positive for these hepatocyte markers was very low. Under these culture conditions, no mature hepatocyte markers (e.g., p450 Cytochromes) were detected. These results suggest that the culture conditions described here facilitate the expansion of liver progenitors able to generate hepatocyte-like cells, albeit at lower numbers, but not fully mature hepatocytes (
(257) To enhance the hepatocytic nature of the cultures and obtain mature hepatocytes in vitro, we first determined whether the three supplemental factors (FGF10, HGF and Nicotinamide) added to EGF and R-spondin 1 were exerting either a positive or negative effect on the hepatocyte expression, as well as on the self-renewal of the culture. We generated liver organoid cultures and cultured them either with EGF or EGF and R-spondin 1 plus FGF10 or HGF or Nicotinamide or the combination of these, and we split the cultures once a week for a total period of 10 weeks. At each time-point we also analyzed the expression of several mature hepatocyte markers (FAH, CYP3A11) and hepatoblast markers (albumin) (
(258) In agreement with the data in
(259) The first approach involved testing a series of compounds in addition to the EGF+FGF10 or HGF condition. A complete list of the compounds analyzed is shown in Table 3.
(260) TABLE-US-00006 TABLE 3 Result Compounds Signal Concentration Alb CYP3AII Exendin4 Glucagon like Sigma 0.1-1 uM peptide 2 analog E7144 Retinoic Acid RAR-RXR Sigma 25 nM receptor ligand Retinoic Acid + Exendin 4 Sonic Hedgehog Invitrogen 500-100 ng/ml C25II BMP4 BMP signaling Peprotech 20 ng/ml 120-05 DAPT Gamma-secretase Sigma 10 nM inhibitor D5942 A8301 Alk5/4/7 inhibitor Tocris 50 nM Bioscience 2939 DAPT + A8301 +++ +++ FGF4 FGFR1,2 ligand Peprotech 50 ng/ml FGF1 FGFR1,2,3,4 Peprotech 100 ng/ml ligand 450-33A Dexamethasone Sigma 10 M-1 mM D4902 25MG Oncostatin M R&D 10-1000 ng/ml (OSM) systems 495- MO-025 FGF4 + OSM + Dexa IGF peprotech 100 ng/ml Valproic acid histone deacetylase Stemgent 250 M inhibitor and 04-0007 regulator of ERK, PKC Wnt/-catenin pathways Sodium Butyrate histone deacetylase Stemgent 250 M inhibitor 04-0005 BIX01294 G9a HMTase Stemgent 1 M inhibitor 04-0002 RG 108 DNA Stemgent 1 M methyltransferase 04-0001 inhibitor TSA 100 nM + Hydrocortisone glucocorticoid Sigma 5 nM H6909 Oncostatin M R&D 10-1000 ng/ml (OSM) systems 495- MO-025 ARA Sigma 500 nM A 0937 R 59022 Diacylglycerol Sigma 500 nM-50 nM + + kinase inhibitor D 5919 Arterenol bitrartre: ---- andrenoreceptor sigma 500 nM-50 nM-5 nM agonist A 0937 LIF 10.sup.3 PD 035901 MEK1 inhibitor Axon 500 nM Medchem cat n 1386 CHIR99021 GSK3 inhibitor Axon 3 uM Medchem cat n 1408 DMSO 1% L-Ascobic acid Sigma 1 mM 077K13021 VEGF Peprotech Matrigel 50% Matrigel 20% VEGF + DEXA
(261) The second approach took into account knowledge from published developmental studies regarding the expression of the transcription factors essential to achieve biliary and hepatocyte differentiation in vivo. A comparative analysis of the expression of transcription factors in the organoids under E or ER or ENRW conditions supplemented with FGF10, HGF and Nicotinamide is shown in
(262) Notch and TGF-beta signaling pathways have been implicated in biliary cell fate in vivo. In fact, deletion of Rbpj (essential to achieve active Notch signaling) results in abnormal tubulogenesis (Zong Y., Development 2009) and the addition of TGF-beta to liver explants facilitates the biliary differentiation in vitro (Clotman F., Genes and Development 2005). Since both Notch and TGF-beta signaling pathways were highly upregulated in the liver cultures (
(263) The combination of EGF+FGF10 with DAPT and A8301 resulted in surprisingly large enhancement of expression of the hepatocyte markers analyzed (CYP3A11, TAT, Albumin) (
Example 17: Human Liver Organoids
(264) Using these expansion conditions (ERFHNic and ENRWFHNic) we have also been able to expand human biliary-derived cultures (
(265) Material and Methods (For Examples 15-17)
(266) Liver Culture-Biliary Duct Isolation
(267) Isolated adult liver tissue was washed in cold Advanced-DMEM/F12 (Invitrogen) and then, the tissue was chopped into pieces of around 5 mm animals and further washed with cold dissociation buffer (collagenase, dispase, FBS in DMEM media). The tissue fragments were incubated with the dissociation buffer for 2 h at 37 C. Then, the tissue fragments were vigorously suspended in 10 ml of cold isolation buffer with a 10 ml pipette. The first supernatant containing death cells was discarded and the sediment was suspended with 10-15 ml of dissociation buffer. After further vigorous suspension of the tissue fragments the supernatant is enriched in biliary ducts. This procedure is repeated until enough biliary ducts are obtained.
(268) Isolated biliary ducts are pelleted and mixed with 50 l of Matrigel (BD Bioscience), seeded on 24-well tissue culture plates and incubated for 5-10 min at 37 C. until complete polymerization of the Matrigel. After polymerization, 500 l of tissue culture media are overloaded.
(269) Media Composition:
(270) Advanced-DMEM/F12 supplemented with B27, N2, 200 ng/ml N-Acetylcysteine, 50 ng/ml EGF, 1 g/ml R-spondin1, gastrin: 10 nM, FGF10 100 ng/ml, Nicotinamide 10 mM and HGF: 50 ng/ml and 50% Wnt conditioned media.
(271) The entire medium was changed every 2 days. After 1 week, Wnt conditioned media is withdrawal and the formed organoids removed from the Matrigel using a 1000 l pipette and were dissociated mechanically into small fragments and transferred to fresh Matrigel. Passage was performed in 1:4 split ratio once or twice per week. Under these conditions cultures have been maintained for at least 6 month.
(272) Reagents
(273) Human Hepatocyte Growth Factor (HGF) was purchased from Peprotech, EGF invitrogen, R-Spondin Nuvelo, Noggin peprotech, FGF10 Peprotech, gastrin Sigma Aldrich, nicotinamide Sigma.
(274) Microarray
(275) For the expression analysis of Lgr5-derived liver cultures, RNA was isolated using a Qiagen RNAase kit, from adult liver or from liver cultures cultured in media without Wntcm and Noggin (ER) or with Wntcm and Noggin (ENRW). 150 ng of total RNA was labelled with low RNA Input Linear Amp kit (Agilent Technologies, Palo Alto, Calif.). Universal mouse Reference RNA (Agilent) was differentially labelled and hybridized to either adult liver tissue or ER or ENRW treated cultures. A 444K Agilent Whole Mouse Genome dual colour Microarrays (G4122F) was used. Labelling, hybridization, and washing were performed according to Agilent guidelines.
Example 18: Lgr5 Expression is Upregulated Following Liver Injury
(276) In the liver, Wnt signaling is active in central vein areas. We have recently observed that Wnt signaling plays a key role in liver metabolism (Boj et al. personal communication). In the liver duct cells, Wnt signaling is activated following liver injury (Hu et al. 2007, Gastroenterology 133(5): 1579-91). Similarly, using the Axin2-LacZ allele, which represents a faithful, general reporter for Wnt signaling we also have observed upregulation of Wnt signaling in the whole liver parenquima after injection of the Wnt agonist Rspo1 (see
(277) The Wnt target gene Lgr5 marks stem cells in several actively self-renewing tissues, but has not previously been reported to be expressed upon injury. Our previously described Lgr5-LacZ knockin mice (Barker et al, 2007, Nature 449 (7165): 1003-7) show that Lgr5 is essentially undetectable in healthy liver although a residual mRNA expression is detected by qPCR. Following injection of CCl4 on Lgr5-LacZ knockin mice (see Barker et al, 2007, supra for LacZ mice and Furuyama K. et al., Nat. Genetics 43, 34-41, 2001 for description of CCl4 method), we observed a clear expression of the reporter in newly formed bud structures in the liver (see
(278) The appearance of Lgr5 expression at sites of active regeneration, suggested that Lgr5 might herald de novo activation by Wnt of regenerative stem cells/progenitors upon injury. Indeed, we found that the novo appearing Lgr5 cells do not express markers of mature liver cells (K19 or FAH) or stellate cells (SMA) but instead, they are positive for the recently described liver progenitor marker Sox9 (
Example 19: Long-Term Expansion of Liver Organoid Cultures
(279) In example 15, it was found that the combination of EGF (50 ng/ml) and R-spondin 1 (1 ug/ml) supplemented with FGF10 (100 ng/ml), HGF (25-50 ng/ml) and Nicotinamide (1-10 mM), were preferable for the long term maintenance of the cultures. We now also have evidence that the three supplemental factors (FGF10, HGF and Nicotinamide) added to EGF and R-spondin1 are all necessary for the expansion of the cultures for longer than 3 months. To assess that, we isolated biliary ducts from the liver parenquima, as shown in
Example 20: Markers Expressed in Liver Organoids Under Differentiation Conditions
(280) Using the differentiation protocol described in example 16, we were able to detect a hepatoblast marker (albumin) and a hepatocyte surface marker in the liver organoids. To quantify the number of these hepatocyte-like cells, we performed flow cytometry analysis of the cultures using a hepatocyte surface marker. We observed that, whereas in the expansion culture condition almost no hepatocyte surface marker-positive cells were detected, after differentiation, up to 35% of the cells were positive for this hepatocyte surface marker (see
(281) We then analyzed the gene expression profile of the mouse liver organoids under these differentiation conditions (
Example 21: Transplantation of Liver Organoids into Mice
(282) Cells were taken from the organoids that had been grown using ERFHNic expansion conditions and EAFD differentiation conditions and were transplanted into immunodeficient strain of mice deficient in the tyrosine catabolic enzyme fumarylacetoacetate hydrolase (FAH), a mouse model for Tyrosinemia type I human disease (Azuma et al. 2007, Nature Biotech. 25(8), 903-910). The transplantation schedule is shown in
Example 22
(283) Organoids from mouse liver cultured in accordance with a method of the invention were analyzed using microarray analysis to determine which genes are expressed and which genes are not expressed.
Example 23
(284) Organoids from human liver cultured using the EM1, EM2 and DM media of the invention and human liver were analyzed using oligonucleotide microarray analysis to determine which genes are expressed and which genes are not expressed. A significantly different gene expression profile was noticeable between the genes expressed in expansion media, the genes expressed in differentiation medium and the genes expressed in adult liver. The trend for hepatocyte gene expression is roughly the same as for in the mouse but the differentiation of the organoids was less than in the mouse liver organoids. This may be due to use of the human cell used.
(285) As often happens in an analysis using an oligonucleotide microarray, Lgr5 and Tnfrsf19 were not detected. However, they were found to be present in organoids cultured in the expansion medium.
(286) Materials & Methods (for Examples 18 to 23)
(287) Animal Treatment
(288) Two-Eight month old Lgr5LacZ or Axin2-LacZ or WT littermates BL6/Balbc F1 mice received an intraperitoneal injection of 0.8 ml/kg of CCL4 dissolved in corn oil (n=) or corn oil alone (n=). Mice were sacrificed 2 or 5 or 9 or 13 days later and the liver was isolated and further processed for RNA or bgalactosidase staining.
(289) -Galactosidase (lacZ) Staining
(290) Liver tissues were isolated and immediately incubated for 2 hours in a 20-fold volume of ice-cold fixative (1% Formaldehyde; 0.2% Gluteraldehyde; 0.02% NP40 in PBS0) at 4 C. on a rolling platform. The fixative was removed and the tissues washed twice in washing buffer (PBS0; 2 mM MgCl.sub.2; 0.02% NP40; 0.1% NaDeoxycholate) for 20 minutes at room temperature on a rolling platform. The -galactosidase substrate (5 mM K.sub.3FE(CN).sub.6; 5 mM K.sub.4Fe(CN).sub.6.3H.sub.2O; 2 mM MgCl.sub.2; 0.02% NP40; 0.1% NaDeoxycholate; 1 mg/ml X-gal in PBS0) was then added and the tissues incubated in the dark at 37 C. for 2 h and overnight at room temperature. The substrate was removed and the tissues washed twice PBS0 for 20 minutes at room temperature on a rolling platform. The tissues were then fixed overnight in a 20-fold volume of 4% Paraformaldehyde (PFA) in PBS0 at 4 C. in the dark on a rolling platform. The PFA was removed and the tissues washed twice in PBS0 for 20 minutes at room temperature on a rolling platform.
(291) The stained tissues were transferred to tissue cassettes and paraffin blocks prepared using standard methods. Tissue sections (4 M) were prepared and counterstained with neutral red.
(292) R-spondin1 Treatment
(293) Axin2-lacZ mice aged 6-8 weeks were injected IP with 100 g of purified human R-spondin1 and sacrificed 48 hours later for LacZ expression analysis in the liver.
(294) RT-PCR
(295) RNA was extracted from gastric cell cultures or freshly isolated tissue using the RNeasy Mini RNA Extraction Kit (Qiagen) and reverse-transcribed using Moloney Murine Leukemia Virus reverse transcriptase (Promega). cDNA was amplified in a thermal cycler (GeneAmp PCR System 9700; Applied Biosystems, London, UK) as previously described (Huch et al., 2009). Primers used are shown in Table 4 below.
(296) TABLE-US-00007 TABLE4 PrimersforRT-PCR PCR Gene product Genename Symbol Sequence bp) cytochromeP450,family3, CYP3A11 TGGTCAAACGCCTCTCCTTGCT 100 w G (SEQIDNO:46) subfamilya,polypeptide11 ACTGGGCCAAAATCCCGCCG v (SEQIDNO:47) Glucose-6-phoshatase G6P GAATTACCAAGACTCCAGG 581 w (SEQIDNO:48) TGAGACAATACTTCCGGAGG v (SEQIDNO:49) Keratin19 Krt19 GTCCTACAGATTGACAATGC 549 w (SEQIDNO:50) CACGCTCTGGATCTGTGACA v (SEQIDNO:51) Albumin Alb GCGCAGATGACAGGGCGGAA 358 W (SEQIDNO:52) GTGCCGTAGCATGCGGGAGG v (SEQIDNO:53) t-box3 Tbx3 AGCGATCACGCAACGTGGCA 441 w (SEQIDNO:54) GGCTTCGCTGGGACACAGATCT v TT (SEQIDNO:55) Prospero-relatedhomeobox Prox1 TTCAACAGATGCATTACC 270 w (SEQIDNO:56) protein1 TCTTTGCCCGCGATGATG v (SEQIDNO:57) Fumarylacetoacetate- Fah ACGACTGGAGCGCCGAGAC 183 w (SEQIDNO:58) hydrolase AGGGCTGGCTGTGGCAGAGA v (SEQIDNO:59) Tyrosineaminotransferase Tat TTTGGCAGTGGCTGAAAGGCA 258 w (SEQIDNO:60) GGGCCCAGGATCCGCTGACT v (SEQIDNO:61) Trytophane2,3-dioxygenase Tdo2 ACTCCCCGTAGAAGGCAGCGA 583 w (SEQIDNO:62) TCTTTCCAGCCATGCCTCCACT v (SEQIDNO:63) Leucine-richrepeat- Lgr5 GGAAATGCTTTGACACACATTC 413 w (SEQIDNO:64) containingG-protein GGAAGTCATCAAGGTTATTATA v A (SEQIDNO:65) coupledreceptor5 Transthyretin TTR ATGGTCAAAGTCTGGATGC 220 w (SEQIDNO:66) AATTCATGGAACGGGGAAAT v (SEQIDNO:67) Glucokinase Gck AAGATCATTGGCGGAAAG 193 w (SEQIDNO:68) GAGTGCTCAGGATGTTAAG v (SEQIDNO:69) hypoxanthine Hprt AAGCTTGCTGGTGAAAAGGA 186 w (SEQIDNO:70) phosphoribosyltransferase TTGCGCTCATCTTAGGCTTT v (SEQIDNO:71)
Immunohistochemistry
(297) Immunostaining procedure used here was previously described in Huch et al. 2009. Briefly, five-micrometer sections were deparaffinized, rehydrated, and tissue sections were permeabilized using PBS-T (PBS; Tween 20 0.1%). When required, sections were treated with 10 mMcitrate buffer (pH 6.0) for antigen retrieval, blocked using Universal blocking buffer (BioGenex)) and incubated with the primary antibody. Then, sections were washed twice with PBS and incubated with peroxidase conjugated secondary antibodies. DAB+ (DAKO) was used as a chromogen substrate. Sections were counterstained with Mayer's hematoxylin and visualized on a Leica DMR microscope. The primary antibodies used were rabbit anti-Sox9 (1:600; 1 h at RT, Millipore), mouse anti-SMA (1:1000, overnight at 4 C., Sigma), rabbit anti-FAH (1:5000; overnight 37 C., gift from M. Grompe), rabbit anti-K19 (1:500; overnight 4 C., gift from M. Grompe). The peroxidase conjugated secondary antibodies used were Mouse or Rabbit Brightvision (Immunologic).
(298) Immunofluorescence
(299) For whole mount staining, organoids or isolated biliary ducts were fixed with acetone (organoids) or PFA4% (biliary ducts) for 30 min, washed once with PBS, permeabilized with PBS 0.3% Triton-X100 for 5 min, blocked using Universal blocking solution (Power block HK085-5KE BioGenex) and incubated overnight with the primary antibodies diluted in PBS1% FBS. Following several washes in PBS, samples were incubated with the secondary antibody. Nuclei were stained with Hoescht33342. Images were acquired using confocal microscopy (Leica, SP5). Three-dimensional reconstruction was performed using Volocity Software (Improvision). The primary antibodies used were rabbit anti-K19 (1:500; gift from M. Grompe), rat anti-hepatocyte surface marker (1:50, gift M. Grompe), goat anti-albumin (1:50, santa Cruz). The secondary antibodies used were all raised in donkey and conjugated to different Alexa fluorofores (donkey anti-goat 568, donkey anti rat-488, donkey anti rabbit-647, Molecular probes).
(300) Flow Cytometry
(301) Dissociated cells were resuspended at 110.sup.4 cells per milliliter in 1 ml of DMEM+2% FBS prior to the addition of MIC1-1C3 hybridoma supernatant at a 1:20 dilution or OC2-2F8 hybridoma supernatant at a 1:50 dilution, and incubated for 30 min at 4 C. After a wash with cold Dulbecco's Phosphate Buffered Saline (DPBS), cells were resuspended in DMEM+2% FBS containing a 1:200 dilution of APC-conjugated goat anti-rat secondary antibody adsorbed against mouse serum proteins (Jackson Immunoresearch). Propidium iodide staining was used to label dead cells for exclusion. Cells were analyzed and sorted with a Cytopeia in FluxV-GS (Becton-Dickenson).
(302) Transplantation Assay
(303) The injection of sorted cell populations to the spleen and the withdrawal of NTBC to induce hepatocyte selection were performed as described previously (Overturf et al. 1996). Drug withdrawal was done in periods of 3 wk, followed by readministration until normal weight was restored in the recipient animals.
REFERENCES
(304) 1. Barker, N. et al. Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003-7 (2007). 2. Bjerknes, M. & Cheng, H. Intestinal epithelial stem cells and progenitors. Methods Enzymol. 419, 337-83 (2006). 3. Barker, N., van de Wetering, M. & Clevers, H. The intestinal stem cell. Genes Dev. 22, 1856-64 (2008). 4. Evans, G. S., Flint, N., Somers, A. S., Eyden, B. & Potten, C. S. The development of a method for the preparation of rat intestinal epithelial cell primary cultures. J. Cell. Sci. 101 (Pt 1), 219-31 (1992). 5. Whitehead, R. H., Demmler, K., Rockman, S. P. & Watson, N. K. Clonogenic growth of epithelial cells from normal colonic mucosa from both mice and humans. Gastroenterology 117, 858-65 (1999). 6. Fukamachi, H. Proliferation and differentiation of fetal rat intestinal epithelial cells in primary serum-free culture. J. Cell. Sci. 103 (Pt 2), 511-9 (1992). 7. Perreault, N. & Jean-Francois, B. Use of the dissociating enzyme thermolysin to generate viable human normal intestinal epithelial cell cultures. Exp. Cell. Res. 224, 354-64 (1996). 8. Korinek, V. et al. Depletion of epithelial stem-cell compartments in the small intestine of mice lacking Tcf-4. Nat. Genet. 19, 379-83 (1998). 9. Pinto, D., Gregorieff, A., Begthel, H. & Clevers, H. Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev. 17, 1709-13 (2003). 10. Kuhnert, F. et al. Essential requirement for Wnt signaling in proliferation of adult small intestine and colon revealed by adenoviral expression of Dickkopf-1. Proc. Natl. Acad. Sci. U.S.A. 101, 266-71 (2004). 11. Kim, K. A. et al. Mitogenic influence of human R-spondin1 on the intestinal epithelium. Science 309, 1256-9 (2005). 12. Dignass, A. U. & Sturm, A. Peptide growth factors in the intestine. Eur. J. Gastroenterol. Hepatol. 13, 763-70 (2001). 13. Haramis, A. P. et al. De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine. Science 303, 1684-6 (2004). 14. Hofmann, C. et al. Cell-cell contacts prevent anoikis in primary human colonic epithelial cells. Gastroenterology 132, 587-600 (2007). 15. Sasaki, T., Giltay, R., Talts, U., Timpl, R. & Talts, J. F. Expression and distribution of laminin alpha1 and alpha2 chains in embryonic and adult mouse tissues: an immunochemical approach. Exp. Cell Res. 275, 185-99 (2002). 16. Stingl, J., Eaves, C. J., Zandieh, I. & Emerman, J. T. Characterization of bipotent mammary epithelial progenitor cells in normal adult human breast tissue. Breast Cancer Res. Treat. 67, 93-109 (2001). 17. St Clair, W. H. & Osborne, J. W. Crypt fission and crypt number in the small and large bowel of postnatal rats. Cell Tissue Kinet. 18, 255-62 (1985). 18. Batlle, E. et al. Beta-catenin and TCF mediate cell positioning in the intestinal epithelium by controlling the expression of EphB/ephrinB. Cell 111, 251-63 (2002). 19. Srinivas, S. et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev. Biol. 1, 4 (2001). 20. Soriano, P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70-1 (1999). 21. Stingl, J. et al. Purification and unique properties of mammary epithelial stem cells. Nature 439, 993-7 (2006). 22. Watanabe, K. et al. A ROCK inhibitor permits survival of dissociated human embryonic stem cells. Nat. Biotechnol. 25, 681-6 (2007). 23. van Es, J. H. et al. Notch/gamma-secretase inhibition turns proliferative cells in intestinal crypts and adenomas into goblet cells. Nature 435, 959-63 (2005). 24. Li, L. et al. The human homolog of rat Jagged1 expressed by marrow stroma inhibits differentiation of 32D cells through interaction with Notch1. Immunity 8, 43-55 (1998). 25. Cheng, H. & Leblond, C. P. Origin, differentiation and renewal of the four main epithelial cell types in the mouse small intestine. I. Columnar cell. Am. J. Anat. 141, 461-79 (1974). 26. Powell, D. W. et al. Myofibroblasts. II. Intestinal subepithelial myofibroblasts. Am. J. Physiol. 277, C183-201 (1999). 27. Yen, T. H. & Wright, N. A. The gastrointestinal tract stem cell niche. Stem Cell Rev. 2, 203-12 (2006). 28. Kedinger, M. et al. Intestinal epithelial-mesenchymal cell interactions. Ann. N.Y. Acad. Sci. 859, 1-17 (1998). 29. Spradling, A., Drummond-Barbosa, D. & Kai, T. Stem cells find their niche. Nature 414, 98-104 (2001). 30. Li, L. & Xie, T. Stem cell niche: structure and function. Annu. Rev. Cell Dev. Biol. 21, 605-31 (2005). 31. Binnerts M E et al, PNAS 104:14700-5 (2007). 32. Sawada et al. Int. J. Exp. Pathol. 72:407-21 (1991).