1. Field of the Invention
The invention relates to compositions and methods whereby, e.g., human pancreatic acinar cells are cultured under conditions that support expansion and transdifferentiation into glandular epithelial cells and subsequently into insulin-producing cells.
2. Background Information
The potential benefits of taking insulin-producing cells from organ donors and transplanting them into insulin-dependent, Type I diabetic patients is clear. In the Edmonton clinical trials, many patients have lived free from the delivery of exogenous insulin for approximately 2 years after being transplanted with intact islets from organ donor sources. However, current technology requires two organ donor pancreata to generate a sufficient number of islets (about 1 million islets, comprised of about 1,000 cells each) to transplant into one diabetic patient for cellular therapy. Thus, there has been emphasis in the diabetes field to identify new sources of insulin-producing cells for transplantation. Many avenues are being explored, including expansion of islets after harvest and before transplantation and the generation of new islets from stem-like cells derived from the bone marrow, or from precursor cells located in the pancreas. The challenges presented by these approaches are related to maintenance of function of islets over long periods of culture, and of the relative rarity of stem-like cells that can be harnessed for insulin production from the bone marrow and pancreas. The ductular precursor stem-like cells derived from the pancreas are reported to be more efficient than bone-marrow derived cells at differentiation into insulin-producing cells, and this may reflect their site of origin (i.e., pancreas) where they are certainly exposed to many differentiation signals related to the pancreatic microenvironment. The most abundant cell type in the pancreas is the acinar cell, which comprises about 85% of the pancreas. The acinar cells serve to produce and secrete digestive enzymes and, like islet cells, arise during development from the ductular cell compartment.
There have been reports that acinar cells, when cultivated in vitro, especially under conditions of stress, can undergo a ‘transdifferentiation’ into a cell type that resembles ductular cells, as determined by expression of CK19, CK7, and carbonic anhydrase (all reputed by the authors to be markers of duct cells) (Kerr-Conte, 1996; WO 02/29010 A2), Hall et al., 1992). Furthermore, Bouwens et al. (1998) have shown in vivo, in a model of pancreatic duct ligation, that acinar cells in the ligated portion of the pancreas undergo transdifferentiation into cells with a ductular phenotype. Further work has suggested that insulin-producing cells can be produced upon further differentiation of the duct cells in the ligated portion of the pancreas. The acinar cells are also reported to be of limited survivability in primary culture, with some culture conditions leading to loss of at least 50% of cells within a week. While primary duct cells have been demonstrated in vitro to convert into insulin-producing cells under some culture conditions (e.g. Bonner Weir, 2000, U.S. Pat. No. 6,011,647), there are no reports of cells that arose from acinar cells in vitro differentiating further to produce islet-like cells.
Prior to the development of the present system, primary pancreatic acinar cells were expanded without differentiation into insulin-producing cells, either in serum-containing medium (undesirable both because of the risk and the uncertainty associated with the use of serum), or in complex serum free media formulations. Likewise, primary pancreatic acinar cells have been transdifferentiated into insulin-producing cells without expansion, producing cells with an insulin-producing phenotype in small numbers. Furthermore, it has not been previously possible to obtain insulin-producing cells in good numbers using acinar cells as starting material.
Thus, there is a need for a simple cell culture system and method for rapidly generating large numbers of cells that can further differentiate into, e.g., insulin-producing cells, through expansion and transdifferentiation of the abundant pancreatic acinar cells. Further, there is a need for a cell culture system and method for culturing and transforming such cells into insulin-producing cells. One cell culture system and related method disclosed herein allows a simple, one-step approach that generates expanded cultures that contain at least 80% intermediate progenitor cells that can give rise to insulin-producing cells. A second cell culture system and related method allows the further culturing of these intermediate progenitor cells or other glandular epithelial cells to obtain insulin-producing cells. Both IP cells and insulin-producing cells will be useful for cell-based therapies for the treatment of diseases such as diabetes.
The present invention provides compositions and methods whereby, e.g., acinar cells can be cultivated successfully in vitro, undergoing a 3-4 fold increase in cell number over time, and giving rise to a cell population that co-expresses acinar and ductal markers early during the culture (2-3 days ex vivo), then ultimately (e.g., about 7-8 days ex vivo) acquires a modified phenotype characterized by expression of some acinar-associated genes, as well as some liver-associated genes. The genes expressed by these modified cells at about 7-8 days ex vivo include, e.g., ductular cytokeratins (CK7, CK8, CK18 and CK19), hepatic nuclear factor 1 (HNF1), alpha-1 antitrypsin, pi-glutathione s transferase (pi-GST), liver-specific (basic helix-loop-helix (bHLH) transcription factor, Thy-1, CCAAT/enhancer-binding protein (C/EBP)-alpha and C/EBP-beta. These cells exhibit little if any expression of the pancreas-associated genes carbonic anhydrase, cystic fibrosis transmembrane conductance regulator (CFTR), elastase and amylase. By “little if any” expression of a gene is meant herein that gene expression is generally undetectable under conventional methods, such as the hybridization and immunocytochemical methods described herein, but expression may be detected by extraordinarily sensitive methods, such as PCR-based analysis. This type of modified cell is referred to herein as an intermediate progenitor (“IP”) cell. The expanded/transdifferentiated acinar cells (IP cells) can be produced using a general serum-containing media, or, in a preferred method, can be produced without serum on a surface comprising one or more extracellular matrix molecules (ECMs) in the presence of one or more soluble active factors. ECMs can be presented in 2 dimensional or 3 dimensional culture systems in the presence of soluble active factors.
The IP cells generated from these cultures are expected to be useful directly in certain medical applications. For example, there is evidence that such cells may under certain conditions become functioning insulin-producing cells when implanted in diabetic patients. The cells can also be used for drug discovery and toxicity studies.
In addition, according to a further aspect of the invention, the IP cells can be cultivated further, in a serum-free medium composed of any standard serum-free base medium (DMEM:HamsF12, for example) with BSA and combinations of factors, including ECMs, small molecules, and growth factors. After 5-10 days of culture, the IP cells undergo additional steps of differentiation, culminating in the formation of cell aggregates that express pro-insulin and C-peptide. Challenge of these cultures with a high-glucose medium causes release of insulin and C-peptide into the medium, indicating the production in these cultures of functional islet-like cells.
Thus, in a first aspect, the present invention provides a cell culture system comprising a superior cell attachment surface that also stimulates cellular expansion, and a simple culture medium including effective amounts of one or more soluble active factors, or serum (e.g. fetal bovine serum), added to a base medium composition. The cell culture system will be particularly useful for primary culture of mammalian epithelial cells, particularly human epithelial cells. In a preferred embodiment the cell culture system is used for the expansion and transdifferentiation of primary acinar cells, especially human pancreatic acinar cells.
The cell attachment surface for this cell culture system is any surface to which the cells can attach and expand, including both 2 dimensional (e.g. plates, flasks, roller bottles, petri dishes, wells etc.) and 3 dimensional (e.g. scaffold) environments. Preferably the surface comprises at least one type of ECM, or a peptide fragment thereof. Cells may, in some circumstances, detach from these surfaces and form self-supporting aggregates. Suitable fragments include peptides consisting of a sequence of three of more amino acid residues that are identical to any portion of the amino acid sequence of the ECM. Such fragments can be easily made and tested by means known to those of skill in the art. Most preferably the surface is a layer of collagen I. Many other surfaces known in the art are also suitable, such as Collagen VI, Collagen IV, Vitronectin, or Fibronectin. Collagen I is preferred due to ease and cost.
The base medium to which the soluble active factors are added may be any cell culture medium appropriate for growth and differentiation of epithelial cells. These include, but are not limited to, DMEM, Hams F12, MEM, M-199 and RPMI. The general requirements for such culture media and many suitable examples are known to those of skill in the art. To this base medium is added either serum (such as fetal bovine serum), or a stabilizing protein such as bovine serum albumin (BSA) along with effective amounts of soluble active factors. The medium is preferably serum-free.
Soluble active factors for the expansion and transdifferentiation of primary pancreatic acinar cells into IP cells include growth factors such as HGF receptor activators and EGF receptor activators. Preferred soluble active factors include one or more of EGF and Transforming Growth Factor-α, IGF1, HGF, betacellulin, prolactin and gastrin 1. HGF, EGF and/or Transforming Growth Factor-α are particularly preferred. Also preferred is the combination of IGF1 and betacellulin.
In one particularly preferred embodiment, the base medium contains a 1:1 mixture of DMEM and Hams F12. The base medium is completed with the addition of glutamine to a final concentration of ˜4 mM, insulin (˜0.1-10 μg/ml, preferably ˜0.01 mg/ml), transferrin (˜0.5-10 μg/ml, preferably ˜0.0055 mg/ml), selenium (˜0.25-5.0 ng/ml, preferably ˜0.0067 μg/ml of sodium selenite), and Epidermal Growth Factor (EGF) (˜1-20 ng/ml, preferably ˜10 ng/ml); this medium is hereafter referred to as pancreatic cell medium, or PCM. To this base medium formulation, up to ˜20% Fetal Bovine Serum (or other serum), preferably between ˜10-˜15% fetal bovine serum, most preferably about 10% or up to about 15% fetal bovine serum) may be added, or, to create a serum-free culture environment, the following components are added in place of serum: heat-inactivated bovine serum albumin (0.1-2%), Hepatocyte growth factor (HGF) (1-20 ng/ml), and/or Transforming Growth Factor Alpha (TGFα) (1-10 ng/ml). In addition, the medium may contain Betacellulin (0.5-20 ng/ml), Gastrin 1 (0.05-10 ng/ml), Prolactin (1.0-10 ng/ml), and/or IGF-1 (5-100 ng/ml). In particular formulations, greater or lesser amounts of these components may be added in order to achieve a formulation that is effective in supporting the expansion and transdifferentiation of the cells. Persons of skill in the art will appreciate that determining effective amounts of the components will require no more than routine experimentation.
By the use of this attachment surface and medium, the expansion and transdifferentiation of primary pancreatic cells with the desired phenotype is simplified greatly.
In a particularly preferred embodiment, the cell culture system is a combination of collagen I coated tissue culture surface (presented in a 2 dimensional or 3 dimensional form) and a serum-free medium containing BSA, insulin, transferrin, selenium, Hepatocyte growth factor (HGF), Epidermal Growth Factor (EGF) and Transforming Growth Factor Alpha (TGFA).
The cell culture system enables superior attachment in vitro of primary pancreatic epithelial cells for adherent culture compared to prior methods, while creating a cellular environment that promotes expansion of the epithelial component of primary pancreatic cultures with concomitant transdifferentiation of the acinar cells present in the starting material into IP cells, while minimizing emergence of undesired fibroblasts. Advantages of this culture system are ease of construction, few components needed, and that all components are readily available and easily used in the required manner.
The components of this aspect of the invention may be conveniently packaged in the form of a kit. The kit may include, for example, 1) a cell culture medium such as DMEM: 2) a serum-free medium supplement containing BSA, insulin, transferrin, selenium, HGF, EGF and TGFA, in suitable amounts to yield the concentrations noted above in the completed medium; and 3) at least one collagen I coated substrate, such as a vessel for tissue culture (e.g., dish(es) with at least one collagen-1 coated tissue culture surface), or collagen-1 coated inserts for use in culture dishes or other laboratory ware. The kit may also optionally include a tissue culture dish or other cell culture accessories and additional reagents that may be required to carry out epithelial cell culture and differentiation.
Culture systems consisting of scaffolds, collagen coated flasks or other vessels and serum-free base medium may be packaged along with the soluble active factors as a separate vial that would be added to the culture medium just prior to use. The active factor combination can be added to a variety of base media to accomplish the same end, e.g., growth and differentiation of primary pancreatic acinar cells in vitro. Such culture systems should also be useful for other cell types, particularly glandular epithelial cells derived from other organs and tissues, including those from liver, pancreas, intestine, prostate, and breast.
The collagen I surface provides superior cell attachment (thereby increasing the number of cells that adhere during initial culture and thus enhancing culture efficiency), while the collagen I and the combination of soluble active factors (e.g., HGF, TGFA and EGF) promote continued proliferation of cells over time, leading to an increase in cell number above what has been previously reported for primary pancreatic acinar cells. Furthermore, the expansion of the acinar cells is accompanied by a transdifferentiation in the majority of cells to an IP phenotype, which is potentially a therapeutically useful cell phenotype for the treatment of diseases such as diabetes. This likely occurs due to convergence of the intracellular signaling pathways associated with collagen I, HGF, TGFA and EGF, creating a synergistic response.
The cell culture system of the present invention has unexpected advantages over systems previously in use. Collagen I, IV, VI, Vitronectin and Fibronectin were expected to enhance cell attachment. However, other extracellular matrix molecules that yielded equivalent attachment of cells during the initial 18 hours of culture did not promote consistent growth of the cells over time in the serum-free medium containing HGF/EGF/TGFA. The most efficient and cost-effective method of achieving cell expansion AND differentiation into IP cells is to utilize a collagen-I surface and a medium containing reduced serum (preferably less than 20%, more preferably less than 15%, 10%, or 5%, most preferably 2%).
Another aspect of the invention is a method for culturing mammalian epithelial cells comprising adding said cells to the cell culture system described above, and maintaining them at suitable temperature and atmospheric conditions. By “mammalian epithelial cell” is meant any cell of a tissue or organ with an epithelial cell phenotype, defined by the presence of expression of cytokeratins and often through the presence of markers that suggest a tissue-specific function (i.e., epithelial cells of the skin make keratin, epithelial cells of the intestine make mucin, epithelial cells of the prostate make PSA). In a preferred embodiment, the cells are primary pancreatic cells, particularly human pancreatic cells. Suitable temperature for mammalian cells is usually in the range of about 37° C., but may be varied somewhat according to cell type. The atmosphere can be ordinary air, or other specialized mixtures of gasses suitable for maintaining cells, as will be familiar to persons of skill in the art. Expansion of pancreatic acinar cells can be maximized by decreasing the oxygen tension in the culture atmosphere to less than 21%, while transdifferentiation to IP cells can be enhanced by increasing oxygen tension to greater than 5%. A preferred range of oxygen tension is between about 5% and about 21%.
In a second aspect, the invention also provides methods and compositions for transforming glandular epithelial cells that have acquired expression of markers characteristic of an intermediate progenitor (IP) phenotype as described above into insulin-producing cells. By “glandular epithelial cell” is meant an epithelial cell that is a component of a gland. Glands are tissues that have a specific function related to secretion of key molecules—most organs in the body have glandular function (liver, intestine, pancreas, prostate, breast, pituitary, adrenal, kidney) whereby they produce and release hormones, digestive enzymes, or other life-essential fluids. Glandular epithelial cells from endoderm-derived organs (e.g., liver, intestine, pancreas) share many characteristics, including the ability to express many of the same genes. Particularly preferred are glandular epithelial cells from pancreas, for example acinar cells. As used herein, the terms “express” and “expression” generally refer to nucleic acids (e.g., mRNAs) or to protein gene products that are detectable by standard immunocytochemical methods.
In this aspect, the invention provides a second cell culture system comprising a cell attachment surface and a culture medium that supports and promotes the transformation of glandular epithelial cells into insulin-producing cells. The cell attachment surface is similar to and may be identical to the attachment surface for expanding primary pancreatic acinar cells. It may be presented in the form of a flat surface coated on a vessel or in the form of a scaffold or other surface adapted for cell culture. It can be comprised of, or coated with, any composition that is capable of maintaining cells or supporting cell growth. In a preferred embodiment, it comprises at least one ECM, such as Collagen I, Collagen VI, Collagen IV, Vitronectin or Fibronectin. In a particularly preferred embodiment, the cell attachment surface is Collagen-I.
In this aspect, the invention provides a further culture medium comprising at least one differentiation promoting factor (“DPF”) that promotes the transformation of glandular epithelial cells into insulin producing cells. The DPFs for the transformation of glandular epithelial cells into insulin producing cells can be one or more of Activin A, acidic FGF, basic FGF, C-Natriuretic Peptide (CNP), Calcitonin Gene Related Peptide, Cholera Toxin B Subunit, Dexamethasone, Gastrin-Releasing Peptide, Glucagon-like Peptide-1 (GLP-1), Glucose, IGF1, IGF2, Insulin, Laminin, LIF, Met-Enkephalin, PDGFAA+PDGFBB, Prolactin, Sonic Hedgehog, Substance P, TGF-alpha, Trolox (alpha-tocopherol derivative), or VEGF. Preferred concentrations in culture medium of each of these 23 DPFs are listed in Table 1. Although in some cases one DPF is sufficient, preferably two or more factors are used. As many as all 23 of the factors may be used.
In a preferred embodiment of this aspect of the invention, the culture medium comprises at least one (or as many as all 10) of the following differentiation promoting DPFs:.C-Natriuretic Peptide (CNP), Calcitonin Gene Related Peptide, Cholera Toxin B Subunit, Dexamethasone, Gastrin-Releasing Peptide, Laminin, Met-Enkephalin, PDGFAA+PDGFBB, Sonic Hedgehog, and Substance P.
In a preferred embodiment, the culture medium that promotes the transformation of glandular epithelial cells into insulin producing cells consists of a 1:1 mixture of DMEM and Hams F12 plus the components listed in Table 2. This medium is sometimes referred to herein as “Media or Medium G9.”
The components of this aspect of the invention may also be conveniently packaged in the form of a kit. The kit may include, for example, 1) a cell culture medium such as DMEM, Hams F12, or a combination thereof, 2) a serum-free medium supplement containing: BSA and the DPFs Activin A, acidic FGF, basic FGF, C-Natriuretic Peptide (CNP), Calcitonin Gene Related Peptide, Cholera Toxin B Subunit, Dexamethasone, Gastrin-Releasing Peptide, Glucagon-like Peptide-1 (GLP-1), Glucose, IGF1, IGF2, Insulin, Laminin, LIF, Met-Enkephalin, PDGFAA+PDGFBB, Prolactin, Sonic Hedgehog, Substance P, TGF-alpha, Trolox (alpha-tocopherol derivative), or VEGF, or two or more of these components in combination, in suitable amounts to yield the concentrations noted in Table 1 in the completed medium; and 3) tissue culture dish(es) with at least one collagen-1 coated tissue culture surface (or collagen-1 coated inserts for use in culture dishes or other laboratory ware). The kit may also optionally include a tissue culture dish and/or other cell culture accessories and additional reagents that may be required to carry out epithelial cell culture and differentiation. In other embodiments, the kit may contain any of the media or media components discussed herein.
Culture systems consisting of scaffolds, collagen coated flasks or other vessels and serum-free base medium may be packaged along with the DPF(s) as a separate vial that would be added to the culture medium just prior to use. The DPF combination can be added to a variety of base media to accomplish the same end, e.g., growth and differentiation of primary pancreatic acinar cells in vitro. Such culture systems may also be useful for other cell types, particularly other epithelial cells derived from glandular tissues, including those from liver, pancreas, intestine, prostate, and breast.
The invention also provides a method for converting glandular epithelial cells into insulin-producing cells comprising culturing the glandular epithelial cells in the cell culture system described above. The method may further comprise removing the culture medium from the cell culture, re-feeding the cell culture with a serum-free medium with glucose, and measuring proinsulin production C-peptide production, or insulin release.
Furthermore, the invention provides an isolated population of insulin-producing cells containing cytoplasmic granules with immunodetectable proinsulin, insulin, and/or c-peptide that is derived from a population of cells of which a subset of cells expressed at least one marker associated with IP cells (e.g., expressed some acinar-associated genes, as well as some liver-associated genes, including, e.g., ductular cytokeratins (CK7, CK8, CK18 and CK19), HNF1, alpha-1 antitrypsin, pi-glutathione s transferase (pi-GST), liver-specific bHLH transcription factor, Thy-1, C/EBP-alpha and C/EBP-beta, and expressed little if any of the pancreas-associated genes carbonic anhydrase, cystic fibrosis transmembrane conductance regulator (CFTR), elastase and amylase).
By an “isolated” cell or population of cells is meant herein that the cell or cell population is removed from its original environment (e.g., the natural environment if it is naturally occurring), and isolated or separated from at least one other component with which it is naturally associated. For example, a naturally-occurring cell present in its natural living host is not isolated, but the same cell, separated from some or all of the coexisting materials in the natural system, is isolated. Such cell or cell populations could be part of a cell culture or cell population, and still be isolated in that such culture or population is not part of its natural environment.
In one preferred embodiment, the insulin-producing cells are derived from glandular epithelial cells obtained from mammalian pancreas, such as primary acinar cells.
The data disclosed in the examples below are generated from freshly isolated human pancreatic cells. The expansion of primary human pancreatic cells in these conditions produces cultures with a mixed epithelial IP phenotype, suitable for in vitro studies of IP cells for a variety of purposes, and suitable for transplantation in vivo for cell therapy for the treatment of diseases such as diabetes. The IP cells generated by these methods may also be useful in the study of pancreatic cell biology, as normal controls in the study of pancreatic epithelial cancers, and to test the effects of drugs/compounds on normal pancreatic epithelial cells (ductal or acinar). Furthermore, the cells may be further cultured to yield insulin-producing cells as demonstrated below.
FIGS. 1A-D show microscopic images after treatment of starting material with antibodies to amylase (
FIGS. 4A-B shows the effect of different culture surfaces on total cell number (
FIGS. 5A-B show a comparison of cell phenotype after expansion in serum-containing (5A) and serum-free (5B) medium containing all soluble active factors.
In describing preferred embodiments of the present invention, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. It is to be understood that each specific element includes all technical equivalents, which operate in a similar manner to accomplish a similar purpose. Each reference cited here is incorporated by reference as if each were individually incorporated by reference.
The following abbreviations are used:
BSA: bovine serum albumin
BMP Bone Morphogenetic Protein
bHLH: basic helix loop helix
DMEM: Dulbecco's Modified Eagle's Medium
TGFβ1: Transforming Growth Factor β1
ECM: extracellular matrix molecules; naturally occurring proteins produced by cells of a tissue that provide structural support as well as a source of cellular signals related to adhesion. Examples are collagen, vitronectin, fibronectin, laminin.
EGF: Epidermal Growth Factor
Hams F12: Ham's Nutrient Mixture F12
HGF: Hepatocyte growth factor
HNF-1: Hepatic nuclear factor 1
IGF1: Insulin-like growth factor 1
IGF-II: Insulin-like growth factor 2
IP cells: Intermediate progenitor cells derived from an epithelial cell, such as, e.g., a pancreatic acinar cell or a liver cell, wherein the derived cells express some acinar-associated genes, as well as some liver-associated genes, including, e.g., cytokeratins (CK7, CK8, CK18 and CK19), HNF1, alpha-1 antitrypsin, pi-glutathione s transferase (pi-GST), liver-specific bHLH transcription factor, Thy-1, C/EBP-alpha and C/EBP-beta, and express little if any of the pancreas-associated genes carbonic anhydrase, cystic fibrosis transmembrane conductance regulator (CFTR), elastase and amylase).
PDGF-A: Platelet derived growth factor alpha
PDGF-B: Platelet derived growth factor beta
TGFA, TGF-α Transforming Growth Factor α
As used herein, the term “culture system” is intended to mean a system for growing and/or differentiating cells in culture, which comprises a cell attachment surface, preferably one that also stimulates cellular expansion, and a culture medium, which includes effective amounts of one or more factors, or serum (e.g. fetal bovine serum), added to a base medium composition.
When referring to active soluble factors and DPFs herein, “effective amount” means an amount that either alone or in combination with other included factors is effective in promoting either expansion and differentiation into IP cells, or into insulin-producing cells, as applicable.
I. Expansion and Transdifferentiation of Primary Acinar Cells into Glandular Epithelial Cells (Culture Phase I)
Materials and Methods:
Starting Material: Primary human pancreatic acinar cells are collected as waste from standard COBE gradient preparation of islet cells for transplantation (Lake et al., 1989). After density gradient centrifugation, the islets are present as a layer between 1.063 density and 1.10 density, and the remaining cells are collected as the pellet that sediments to the bottom of the gradient based on density. Approximately 48 hours after collection of the cells at the transplant center are received by the inventors in non-tissue-culture treated polystyrene flasks and are suspended in RPMI+10% fetal calf serum at a density of approximately 2.0 million cells/ml. Cell number and viability is assessed by trypan blue exclusion and enumeration on a hemacytometer by light microscopic observation.
Phenotypic Evaluation of Starting Material. A preparation of starting material was formalin fixed and paraffin-embedded as a cell pellet about 24 hours after initial harvest of the pancreatic cells. Paraffin sections were prepared, placed on slides, and subjected to immunocytochemical analysis with antibodies to insulin (Biogenex, San Ramon, Calif.), CK19 (Biogenex), and Amylase (Biogenex). A minimum of (3) sections per sample were assessed with each marker. All antibody staining was carried out according to the manufacturer's suggestion with pre-diluted commercial antibodies. For CK19, a 3 minute treatment with pepsin enzyme (Biogenex) preceded the blocking step for the purpose of antigen retrieval. Briefly, the sections were rehydrated through graded ethanols, followed by a 15 minute incubation in phosphate buffered saline (PBS) without calcium and magnesium. Protein Blocker (Biogenex) was added for 30 minutes prior to addition of primary antibodies. After (3) 5-minute washes, biotinylated secondary antibody (Biogenex) was added at a 1:100 dilution and sections were incubated for 30 minutes at room temperature. After (3) 5-minute washes, Alexa488 or Alexa-596-conjugated StreptAvidin (Molecular Probes, Eugene, Oreg.) was added for fluorescent visualization. For each slide, a minimum of (3) 200× images were captured using a Nikon fluorescent microscope fitted with a SPOT camera (Diagnostic Systems, Inc., Webster, Tex.). The images were assessed quantitatively using image analysis software (MetaMorph/Universal Imaging Corporation, Downington, Pa.) to determine relative fraction of insulin-positive, CK19+, and amylase+ cells. Insulin+ cells are the beta cells of the islets, CK19+ cells are the primary ductal cells, and amylase+ cells are the acinar cells (see Example 1).
A. Serum-Free Medium
Freshly isolated primary human pancreatic cells were collected as a pellet from a COBE cell separator, fixed in formalin, paraffin-embedded, sectioned, and analyzed with antibodies to amylase, CK19, and Insulin. Images (
Primary human pancreatic cells were then seeded at 104 or 105 cells/cm2 onto tissue culture treated polystyrene in either DMEM commercial medium plus 10% fetal bovine serum or in PCM plus 10% fetal bovine serum. Replicate cultures were harvested at 3 day intervals via trypsinization and live cells (as determined by trypan blue exclusion) and enumerated on a hemocytometer. The results (shown in
The cell expansion experiment was repeated essentially as above, except that the base medium was supplemented with only three of the soluble active factors: TGF, HGF, and EGF.
B. ECM Surfaces
The attachment of primary human pancreatic cells was evaluated by counting the number of attached cells vs. the number of cells initially seeded on a panel of ECM surfaces comprised of Collagen I (1 μg/cm2), Fibronectin (3 μg/cm2), Laminin (2 μg/cm2), Vitronectin (1 μg/cm2), Matrigel (1 μg/cm2), Human ECM (1 μg/cm2), or Poly-D-Lysine (3 μg/cm2). In one condition, a mixture of Collagen IV, Laminin, and Fibronectin was utilized. ECMs were placed into solution at the above concentrations and allowed to coat tissue culture-treated polystyrene surfaces according to manufacturer's suggestions of 1 hour at room temp. Excess ECM solution was then removed and surfaces were rinsed twice in water. Just before seeding cells, the water was aspirated, then cells were seeded onto the ECM surface at a density of 1×105 cells/cm2 in growth medium (PCM) composed of DMEM:HamsF12 mixture (1:1) with 4 mM glutamine, 1×ITS supplement (GIBCO 51500-056), 10% Fetal Calf Serum (Inactivated, Qualified, GIBC 26140-079), and 10 ng/ml Epidermal Growth Factor (EGF) (BD 4001). Cells were seeded onto tissue-culture polystyrene surface as a control. After 18 hours, unattached cells were washed away and remaining attached cells were re-fed with PCM and allowed to grow for 7 days prior to evaluation. Cultures were fixed in 10% formalin and subjected to immunocytochemistry with antibodies for CK19 and Amylase as described previously to determine phenotypic composition. Cells were counterstained with DAPI fluorescent blue nuclear stain to visualize individual cell nuclei for cell counting. The metabolic activity of cells subjected to the various conditions was determined by an MTS assay. Viable cells were measured using the MTS assay (Promega CellTiter 96 Aqueous One Solution Cell Proliferation Assay), a colorimetric method for determining the number of viable cells in proliferation or cytotoxicity. The results of this analysis are shown in
Primary pancreatic cells, composed of >90% non-islet/non-duct cells, were plated onto various coated surfaces at a density of 28,900 cells/well (105 cells/cm2). Unattached cells were washed off after 18 hours, and cultures were re-fed and allowed to grow for 8 days. Cultures were fixed in formalin (10%) and subjected to phenotypic analysis with antibodies to CK19 and Amylase. The results are shown in
Tissue culture-treated polystyrene culture surfaces were coated with Collagen I as described above. Tissue culture medium (PCM) was prepared as described above. In some cases, serum was replaced with Fraction V BSA (99% pure, heat inactivated, Sigma), along with combinations of soluble growth factors, including IGF1, IGF2, betacellulin, HGF, EGF, and TGF-alpha. Optimal seeding density is between 104 and 105 cells/cm2, as demonstrated in Example 3. Cells were seeded onto collagen-coated flasks (150 cm2) at 1.5×106 cells/flask in PCM. After an ˜18 hr attachment period, unattached cells were washed away with gentle aspiration/rinse, followed by re-feeding with fresh medium. Cultures were monitored over time by metabolic assay (MTT) and by trypsinization and cell counting, to establish cell number (see example 3). Cell phenotype at the end of the culture period was assessed as follows: small-scale cultures were set up simultaneously in 96-well plates. At the end of the culture phase, monolayer cells were fixed in 10% formalin for a minimum of 1 hour. After formalin was removed and monolayers were rinsed, cultures were subjected to immunocytochemistry as described in previous section for CK19, amylase, insulin, and vimentin (a marker of fibroblasts). The relative fraction of CK19+ cells was determined by quantitative image analysis as described above (see Example 4). After formalin was removed and monolayers were rinsed, cultures were subjected to immunocytochemistry as described in previous section for CK19 and vimentin (a marker of fibroblasts). Cells were also stained with amylase antibodies, but did not produce positive results due to release of digestive enzymes, such as amylase, by the cells over time in culture. The relative fraction of CK19+ cells was determined by quantitative image analysis as described above (see example 4). Acquisition of ductal markers by acinar cells was verified by demonstrating concomitant expression of CK19 and amylase in cell subpopulations during days 2-3 of culture (see example 5). For these experiments, CK19 primary antibodies were reacted with formalin-fixed cell cultures, followed by visualization with Alexa488-conjugated Goat anti-mouse IgG (Molecular Probes). Then, cells were subjected to a blocking step (Protein Blocker, BioGenex), followed by application of the second primary antibody (anti-amylase). Visualization of the amylase was accomplished by application of Alexa594-conjugated Goat Anti-Mouse IgG. Images were collected as described above. At the end of a 7-day culture period in the conditions described herein, between 65-90% of the cells in the culture express CK19, while less than 20% express vimentin (see example 6). Variations in the relative proportion of CK19+ cells probably reflect heterogeneity due to age, gender, and other unique characteristics of individual patients.
Primary pancreatic cells were seeded at (3) densities on tissue-culture treated polystyrene dishes (60 mm) and fed with PCM. Light microcopic observations were made daily. At the 24-hour timepoint, dishes were sacrificed and stained with trypan blue to assess viability. The results are shown in Table 3.
Cells were grown on a Collagen I surface, at 37° C. in 21% oxygen, in PCM medium or in base medium with 2% BSA, 2 ng/ml TGF-α, 10 ng/ml EGF, and 10 ng/ml HGF. After 7 days, cultures were fixed in 10% formalin and subjected to immunocytochemical analysis with fluorescent detection, followed by automated image collection and analysis. The results are shown in
Primary pancreatic acinar cells were cultured for several days in a 1:1 ratio of DMEM and HamsF12, with 10% fetal bovine serum, 0.01 mg/ml insulin, 0.0055 mg/ml transferrin, 0.0067 μg/ml sodium selenite, 10 ng/ml EGF, 4 mmol/liter glutamine and antibiotics. After 2 days of culture (4 days ex vivo), expression of amylase by the acinar cells is still strong (
After 7 days of growth in PCM/Collagen I surface, cells were fixed, stained with antibodies to CK19, and counterstained with nuclear DAPI. Total cell number was evaluated by automated image analysis (
Two independent IP cell cultures were subjected to Clontech 8K Atlas Gene Array analysis. IP cells were obtained by culturing primary acinar cells in a cell culture system comprising PCM and a Collagen I surface. Monolayer cultures were rinsed 2× with PBS, then detached from the flasks with 0.25% trypsin. Cells were pelleted by centrifugation at 1,200 RPM for 3 minutes in a swinging bucket centrifuge. Cell pellets were resuspended and washed 2× in PBS before a final centrifugation at 1,200 RPM for 3 minutes as described above. The supernatant was discarded and gently aspirated to remove as much liquid as possible from the cell pellet, which was then quick-frozen in a dry-ice/ethanol bath and stored at −80° C. until transfer to BD Clontech where gene expression analysis was performed, using conventional techniques.
Labelled P-33 cDNA probes were prepared from the 30 μg of total RNA from each sample by first enriching for poly A+RNA using a streptavidin-magnetic bead separation method that is part of the Atlas Pure Total RNA Labeling system. The labeled probes from each sample were hybridized with the plastic human 8 K gene arrays for about 16 hours, the arrays were washed and imaged according to the Atlas array protocols. The Atlas image 2.7 software was used to align array images with the array grid template and to exclude false background signals or false signals due to strong signal bleedover. The transcript signals were then extracted from these aligned arrays using the Atlas Image 2.7 software and further statistical analysis of the changes in gene expression were performed.
In general, mRNA transcription was assayed, by hybridization to suitable oligonucleotide probes. In a few cases, e.g., for CK19 and amylase, the protein expression product was measured, using conventional methods of immunohistochemistry. A summary of the expression by these cell populations of a selected set of genes is presented in Table 4. Table 4 contains a list of genes expressed in IP cells and a comparison of expression patterns in primary acinar cells and primary ductal cells. Gene products identified as “+” were expressed; those identified as “++” were strongly expressed. Gene products designated) are found in regenerating pancreas.
II. Transformation of Glandular Epithelial Cells into Insulin Producing Cells—Generating Insulin-Producing Cells by Differentiation of IP Cultures (Culture Phase II)
IP cultures can be utilized to generate insulin-producing cells by placing the cells in a second phase of culture that includes a surface, such as Collagen I, that promotes attachment of the IP cells combined with a defined medium formula that lacks serum but contains combinations of the following differentiation promoting factors: Activin A, acidic FGF, basic FGF, C-Natriuretic Peptide (CNP), Calcitonin Gene Related Peptide, Cholera Toxin B Subunit, Dexamethasone, Gastrin-Releasing Peptide, Glucagon-like Peptide-1 (GLP-1), Glucose, IGF1, IGF2, Insulin, Laminin, LIF, Met-Enkephalin, PDGFAA+PDGFBB, Prolactin, Sonic Hedgehog, Substance P, TGF-alpha, Trolox (alpha-tocopherol derivative), and VEGF. In the following examples, the base medium is composed of a 1:1 mixture of HamsF12 and DMEM with antibiotics and 0.2% Bovine Serum Albumin (Fraction V, heat inactivated 99% pure). In one example (Combination 1), the base medium contained Cholera Toxin B, Dexamethasone, GRP, GLP-1, Glucose, IGF-1, IGF-2, Insulin, Prolactin, Sonic Hedgehog, Trolox, aFGF, and bFGF. In another example (Combination 2), the base medium contained Activin A, CGRP-alpha, CNP, Glucose, GLP-1, IGF-2, Insulin, LIF, Met-Enkephalin, Prolactin, Sonic Hedgehog, aFGF, and vEGF. In a third example (Combination 3), the base medium contains Activin A, CGRP-alpha, Cholera Toxin B, Dexamethasone, Glucose, GLP-1, Insulin, LIF, Laminin, Met-Enkephalin, PDGFAA/BB, Sonic Hedgehog, Substance P, TGF-alpha, aFGF, and VEGF. The concentrations of these media supplements are listed in Table 1.
AD cells were placed into culture by either: 1) trypsinizing the cells from the surface on which they were generated, and redistribution onto a fresh attachment-promoting surface at a density of ˜5×104 cells/cm2, or 2) removing the medium, washing 2× in PBS to remove traces of old medium, and cultures re-fed with the new medium (described above) containing differentiation promoting factors. Cells are cultured for a period of 4-10 days at 37° C. and 21% oxygen. On Day 5, half of the medium is removed and replaced with an equal volume of fresh medium containing differentiation promoting factors.
Phenotypic Analysis of IP cells After Differentiation Culture.
Morphological assessment of IP cells cultured in differentiation conditions described above was captured by light microscopy (see Example 8, below). The cellular phenotype of the cells comprising these cultures was assessed by immunocytochemistry as described above using monoclonal antibodies to vimentin, pro-insulin, C-peptide, MUC-1, and CK19 (See Example 10, below). Briefly, cultures were fixed with 10% formalin for 1 hour at room temperature, then washed with PBS and subjected to immunocytochemical protocol. (See Example 9, below).
Functional Analysis of IP cells After Differentiation Culture.
The ability of the aggregated cell clusters to release insulin and C-peptide was assessed by subjecting the cultured cells to a glucose challenge as follows. Cells that had been cultured in differentiation medium for 7-10 days were washed 3× in PBS, then re-fed with either 1) base medium (described above) with 5 mM Glucose, or 2) base medium with 22 mM glucose. After 18 hours, the cell-conditioned medium was collected and subjected to ELISA analysis for insulin and C-peptide release (Diagnostic Systems Laboratories (DSL)). ELISAs were conducted using the standard range assay procedure according to manufacturer's specifications. Plates were incubated on a shaker during the assay and results were read in a Tecan spectrophotometric plate reader. Total ng of insulin or C-peptide per well were calculated for each media condition, for both 5 mM glucose media and 22 mM glucose media (See Example 10).
Pancreatic acinar cells were cultured in Base Medium+ITS+Serum (10%) for 1 week, then trypsinized (treated with 0.25% Trypsin without EDTA for 10 minutes at 37° C.) and transferred to a fresh collagen-1 coated surface and placed in a medium containing all 23 DFPs listed. Over a period of 3-5 days, the cells readily formed three-dimensional pod-like structures, clearly observable by light microscopy (
Pod-like structures, generated the same manner as described in the previous example, were fixed in 10% formalin and subjected to immunocytochemical analysis first with CK19 monoclonal antibodies, then with C-peptide monoclonal antibodies, as described above.
Cells cultured in base medium (negative control), or in Combinations 1, 2 and 3 of the differentiation promoting media, were evaluated for their ability to release insulin and C-peptide into the culture medium. In addition, we assessed whether increasing concentrations of glucose led to the release of a greater quantity of insulin and C-peptide, indicating an islet-like functionality. First, the cells were cultured for 1 week in base medium+EGF (10 ng/ml)+ITS+10% fetal bovine serum (PCM). Then, cells were either subjected to a wash and medium change (non-subcultured), or to a wash, trypsinization/detachment, reseeding, and medium change. Replicate cultures were re-fed with either base medium (serum-free), fresh PCM, or one of the three combinations of differentiation promoting media (all serum-free). After 10 days, differentiation media were removed, cultures were washed 3× with PBS, then re-fed with serum-free base medium containing either 5 mM glucose or 22 mM glucose (final concentration). After 18 hours, the conditioned media were collected and subjected to ELISA analysis with antibodies to either Insulin or C-Peptide (DSL laboratories).
Both the quantity of insulin and the quantity of DNA were measured in IP cells subjected to differentiation culture with or without enzymatic detachment and subculturing. Cultures were carried out precisely as described in the previous paragraph. DNA was measured utilizing a standard Picogreen assay (Molecular Probes), while insulin was measured by ELISA assay. Total ng of Insulin was divided by total μg of DNA in the sample, thus providing the insulin:DNA ratio value, in order to calculate a ratio of the quantity of insulin present vs. the number of cells present (reflected by DNA content). The results are shown in
Insulin-producing cells obtained by the preceding method were subjected to gene expression analysis as described above. Table 5 contains a list of the highest expressed genes, their position on the Clontech atlas 8K gene array, and relative expression of these genes (after normalization). Table 5 is attached hereto as Appendix 1
Primary human pancreatic cells were seeded at 0.5×105 cells/cm2 in PCM on a collagen-1 surface and grown for 7 days. Insulin was measured at Days 1, 7, and 10 as follows: Growth medium was removed, wells were washed 3× in phosphate buffered saline. After a pre-incubation for 1 hour at 37 C in base medium without insulin, with 5 mM glucose, media was removed and replaced with either 1) base medium (without insulin) with 5 mM glucose, or 2) base medium (without insulin) with 22 mM glucose. Insulin was measured in cell-conditioned media after 18 hours at 37° C. After 7 days of culture, PCM medium was replaced with either 1) fresh PCM, 2) serum-free base medium, 3) serum free base medium with all 23 differentiating factors, 4) serum-free combination 1, or 5) serum-free combination 2. The results are shown in
Human pancreatic acinar cells were cultured on a collagen I surface in PCM from Day 1 to Day 7, thus generating a culture of IP cells at Day 7. On Day 7, the EP cells were washed and the PCM medium was replaced with the G09 differentiation medium containing the 30 factors listed in Table 2. At each time point (Days 1, 7, 10 and 14), insulin release was measured by washing the cultures three times with PBS, then challenging the cultures with a 1:1 mixture of DMEM and HAMs F12 containing either 5 mM or 22 mM glucose. After 18 hours of exposure to the glucose, supernatants were collected and insulin measured by ELISA. The results are shown in
III. Expression Studies at Several Time Points of Primary Human Acinar Cells that are Expanded, Allowed to Differentiate into IP Cells and then Allowed to Differentiate Further into Insulin-Producing Cells
Three independent samples of primary human pancreatic acinar cells were seeded and expanded described above. From Day 0 to Day 8, cells were on collagen I surface, seeded at 104 cells/cm2, in PCM. On Day 8, the medium was changed from PCM to the medium with the active factors shown in Table 2. Cells were fed twice with G09 (50% of medium replaced) between days 8 and 16. The cells remained on the surface throughout the culture process. Cultures were harvested at 3 days after the initial plating (actively trans-differentiating acinar cells), 8 days after plating (IP cells) and 16 days after plating (putative insulin producing cells) and subjected to gene expression analysis, as described in Example 7. mRNA expression data were obtained with 12K microarrays from Clonetech.
Briefly, growth medium was removed from the culture flasks and cells were lysed in trizol LS (Invitrogen) chaotrope/phenol reagent for about 2 minutes by pipetting the lysis solution over the cell layer. Three ml of RNAse free water was added per 9 ml of lysis solution in an Oak Ridge Cetrifuge tube. 2.4 ml chloroform was then added and the solution vigorously vortexed for 1 minute. The aqueous and organic phases were then separated by cetrifugation at 4° C. and the upper aqueous phase containing RNA was removed to a clean PET tube. The RNA was precipitated by isopropanol precipitation, washed with 70% ethanol and redissolved in 200 μl of RNAse free water. A chaotrope lysis reagent was immediately added to the RNA and it was further purified using a Qiagen spin column method with a DNAse digestion step. The purified RNA was finally eluted in 80 μl RNAse free water and stored at −80° C.
Labelled P-33 cDNA probes were prepared from the 30 μg of total RNA from each sample by first enriching for poly A+RNA using a streptavidin-magnetic bead separation method that is part of the Atlas Pure Total RNA Labeeling system. The labeled probes from each sample were hybridized with the plastic human 12 K gene arrays for about 16 hours, the arrays were washed and imaged according to the Atlas array protocols. The Atlas image 2.7 software was used to alighn array images with the array grid template and to exclude false background signals or false signals due to strong signal bleedover. The transcript signals were then extracted from these aligned arrays using the Atlas Image 2.7 software and further statistical analysis of the changes in gene expression were performed.
The raw expression data were analyzed as follows: (1) We filtered out genes that were not expressed at any of the 3 conditions/time points; (2) We normalized all of the microarrays against each other to remove differences from array-to-array and the effects of variability in sample processing, hybridization, etc.; (3) We identified genes which showed a statistically significant difference among the conditions/time points; and (4) We clustered the genes based on their temporal patterns in a way that is consistent with the design of the study and the changes in phenotype.
Table 6 shows expression data for the genes that were identified by the above analysis. This Table is attached hereto as Appendix 2. These identified genes were expressed at high levels at both Day 3 and Day 8, or their expression increased substantially between Day 3 and Day 8. The Table also shows the expression levels of these genes at Day 16, and the mean expression for all three condition/time points. The Table also shows the ratios of expression at various times: “I to A” is the ratio of expression of putative insulin-producing cells (Day 16) to acinar (Day 8) cells; “Int to A” is the ratio of IP cells (Day 8) cells to acinar cells (Day 3).
The data shown in Table 6 were further analyzed by clustering them into one of 17 “classes,” whose features are summarized on the Table. A graphical representation of the characteristics of these 17 classes in presented in
The data from the Day 8 time points in Table 6 were also grouped with regard to whether the genes expressed at Day 8 in these cells belong to the classes of genes expressed normally in (1) liver and pancreas; (2) pancreas-associated genes; (3) liver-associated genes; or (4) progenitor-associated genes. The results are shown in Table 7.
As can be seen, at Day 8 IP cells no longer expressed genes consistent with pancreatic acinar cells, nor did they express a complement of genes specific for pancreatic ductular cells. The IP cells expressed low levels of some markers associated with pancreatic islets, including insulin, somatostatin and pancreatic polypeptide, suggesting that at least some cells in the population are competent to express endocrine genes of the pancreatic islets.
Surprisingly, the IP cells also expressed several liver-specific transcription factors (e.g., C/EBP alpha, C-EBP-beta) and other markers of mature and developing liver, including low levels of Thy-1, a marker associated with hepatic “oval” stem cells. This suggests that the differentiating cells were not moving simply from pancreatic acinar to pancreatic ductal, but had developed into a cell with both hepatic and pancreatic characteristics, while not fitting into any single gene expression profile of one of these cell types. The cells generated in this example resemble the cells that emerge from the pancreas of rodents that are fed a copper-deficient diet. (See, e.g. Rao et al., 1988). The pancreas of such animals goes through an acute phase of pancreatitis followed by “hepatization” of liver (which means cells that begin to express hepatic genes rather than pancreatic genes). Liver-like cells have also been reported in human fetal pancreas (Tsanadis et al., 1995) Isolated cells generated by the methods of the present invention (e.g., by propagating primary acinar cells or other types of endodermal cells or progenitor cells by the methods of the invention) are to be distinguished from naturally occurring cells that may have some of the characteristics of IP cells, such as oval cells or cells isolated from the pancreas of a rodent on a copper-deficient diet.
Cells having the characteristics of these IP cells may be useful for, e.g., therapeutic approaches in the treatment of diabetes. Furthermore, although the cells in this example were derived from pancreas, other epithelial tissues, or perhaps even any endoderm-derived tissue, may provide additional sources of cells that can be differentiated into cells having a similar phenotype. Suitable tissue types include, e.g., liver or intestine. These IP cells express genes associated with pancreas, liver, intestine and neuronal tissues. For example, they express mucin, CK19 and CK7, which are common markers associated with duct cells in the pancreas, liver and intestine. Thus, the gene expression pattern seen in these IP cells may serve as a predictive measure for cells derived from each of these tissues for the purpose of generating insulin-producing cells. Furthermore, IP cells may, under appropriate conditions, give rise, not only to pancreatic islet cells, but also to hepatocytes or any endoderm-derived tissue.
The disclosures of the following references, cited above in part, relate to the present invention:
The embodiments illustrated and discussed in the present specification are intended only to teach those skilled in the art the best way known to the inventors to make and use the invention, and should not be considered as limiting the scope of the present invention. The exemplified embodiments of the invention may be modified or varied, and elements added or omitted, without departing from the invention, as appreciated by those skilled in the art in light of the above teachings. It is therefore to be understood that, within the scope of the claims and their equivalents, the invention may be practiced otherwise than as specifically described.
The entire disclosure of all applications, patents and publications, cited above and in the figures are hereby incorporated in their entirety by reference.
adenomatous polyposis coli like
ef7
This application claims the benefit of provisional application 60/384,000, filed May 28, 2002, which disclosure is incorporated by reference in its entirety herein.
Number | Date | Country | |
---|---|---|---|
60384000 | May 2002 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 10443732 | May 2003 | US |
Child | 11410954 | Apr 2006 | US |