Patent Application
20150023934
This patent disclosure contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.
Prostate disorders, such as prostatitis, benign prostate hyperplasia and prostate cancer are the most common male-related pathologies. Despite recent advances in basic and translational research, prostate cancer remains the second leading cause of cancer in men and a complete cure remains elusive. Complications in the clinic arise from prostate cancer phenotypic heterogeneity, imperfect early prognostic markers able to predict the evolution of the disease to aggressive forms, and the progression to castration-resistant forms.
The present invention relates generally to the finding that induced pluripotent stem cells (iPSCs) can be directly differentiated and that mouse and human fibroblasts can be transdifferentiated into prostate and urinary bladder epithelium.
An aspect of the invention is directed to a method for reprogramming embryonic fibroblast cells in culture to epithelial cells. In one embodiment, the method comprises: (a) isolating embryonic fibroblasts (EFs); (b) infecting EFs with a retrovirus comprising a reprogramming factor; and (c) incubating for at least 24 hours at about 37° C. In another embodiment, the method further comprises switching culture medium to a serum-free basal epithelial medium. In some embodiments, the basal epithelial medium contains EGF, FGF, or a combination of the listed growth factors. In one embodiment, the embryonic fibroblasts (EF) has a wild-type genotype, an Oct4-GFP knock-in genotype, or a Nkx3.1-lacZ knock-in genotype. In one embodiment, the embryonic fibroblasts (EF) have a GATA6CreERT2; R26R-CAG-YFP genotype. In one embodiment, the embryonic fibroblasts (EF) have a CK18CreERT2; R26R-Tomato genotype. In another embodiment, the retrovirus is a Rebna retrovirus. In one embodiment, the embryonic fibroblasts are mouse embryonic fibroblasts. In a further embodiment, the reprogramming factor is Oct4, Sox2, Klf4, c-Myc, or a combination of the listed reprogramming factors. In some embodiments, the epithelial cells are induced epithelial cells. In yet other embodiments, the induced epithelial cells express cytokeratin 5 (CK5), CK8, CK14, CK18, beta-catenin, E-cadherin, or a combination of such listed markers. In one embodiment, the induced epithelial cells express EpCAM, CD24, or a combination thereof. In some embodiments, the induced epithelial cells are stably maintained for at least 3 passages, at least 4 passages, at least 5 passages, at least 6 passages, at least 7 passages, at least 8 passages, at least 9 passages, at least 10 passages, at least 11 passages, at least 12 passages, at least 13 passages, at least 14 passages, or at least 15 passages. In further embodiments, the induced epithelial cells are further differentiated in prostate epithelia or bladder epithelia. In some embodiments, the retrovirus is a lentivirus. In another embodiment, the lentivirus is doxycycline regulated.
In one embodiment, the embryonic fibroblasts of (a) express CD140. In another embodiment, the embryonic fibroblasts of (a) do not express CD11, EpCAM, CD24, or a combination thereof.
An aspect of the invention is directed to a method for reconstituting induced epithelial cells into an organ tissue. In one embodiment, the method comprises: (a) isolating induced epithelial cells prepared according to the method described above; (b) transducing the induced epithelial cells with a retrovirus comprising a master regulatory gene; (c) recombining the induced epithelial cells with mesenchymal cells; and (d) performing a graft in an immunodeficient subject. In another embodiment, the master regulatory gene is a master regulatory gene for prostate development. In a further embodiment, the master regulatory gene for prostate development comprises NKX3.1, Androgen Receptor (AR), FOXA1, FOXA2, or a combination of the listed master regulatory genes. In some embodiments, the master regulatory gene is a master regulatory gene for bladder development. In other embodiments, the master regulatory gene for bladder development comprises KLF5, Pparγ, Grhl3, Ovol1, Foxa1, Elf3, Ehf, or a combination of the listed master regulatory genes. In further embodiments, the mesenchymal cells comprise urogenital mesenchyme. In one embodiment, the graft is a renal graft. In another embodiment, the organ tissue is prostate epithelial tissue. In a further embodiment, the organ tissue is bladder epithelial tissue. In some embodiments, the organ tissue expresses p63 and CK5 in the basal layer. In other embodiments, the prostate tissue expresses AR and CK8 in the luminal layer. In further embodiments, the prostate tissue expresses Probasin or PSA. In one embodiment, the bladder tissue expresses CK8 in the luminal layer and uroplakins. In yet other embodiments, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome. In some embodiments, the retrovirus is a lentivirus. In another embodiment, the lentivirus is doxycycline regulated.
An aspect of the invention is directed to an isolated population of induced epithelial cells obtained from the method described herein. In one embodiment, the cells express cytokeratin 5 (CK5), CK8, CK14, CK18, beta-catenin, E-cadherin, or a combination of the listed markers.
An aspect of the invention is directed to a method for transdifferentiation of embryonic fibroblast cells into an organ tissue, the method comprising: (a) isolating embryonic fibroblasts (EFs); (b) transducing EFs with a retrovirus comprising a reprogramming factor; (c) culturing the infected EFs in stem cell media for at least 24 hours at about 37° C. to generate induced pluripotent stem cells (iPSCs); (d) isolating iPSCs; (e) recombining the cells of (d) with mesenchymal cells; and (f) performing a graft of the recombined cells of (e) into an immunodeficient subject. In one embodiment, the stem cell media comprises LIF. In one embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In one embodiment, the mesenchymal cells comprise urogenital mesenchyme. In one embodiment, the mesenchymal cells comprise bladder mesenchyme. In one embodiment, the graft is a renal graft. In one embodiment, the organ tissue is prostate epithelial tissue. In one embodiment, the organ tissue is bladder epithelial tissue. In one embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In one embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In one embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In one embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome. In one embodiment, the retrovirus is a lentivirus. In one embodiment, the lentivirus is doxycycline regulated.
An aspect of the invention is directed to a method for differentiation of induced pluripotent stem cells (iPSCs) into an organ tissue, the method comprising: (a) isolating iPSCs; (b) recombining the cells of (a) with mesenchymal cells; and (c) performing a graft of the recombined cells of (b) into an immunodeficient subject. In one embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In one embodiment, the mesenchymal cells comprise urogenital mesenchyme. In one embodiment, the mesenchymal cells comprise bladder mesenchyme. In one embodiment, the graft is a renal graft. In one embodiment, the organ tissue is prostate epithelial tissue. In one embodiment, the organ tissue is bladder epithelial tissue. In one embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In one embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In one embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In one embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome.
An aspect of the invention is directed to a method for differentiation of induced pluripotent stem cells (iPSCs) into an organ tissue, the method comprising: (a) isolating iPSCs; (b) culturing iPSCs in endodermal differentiation media; (c) isolating iPSCs that express an endodermal marker; (d) recombining the cells of (c) with mesenchymal cells; and (e) performing a graft of the recombined cells of (d) into an immunodeficient subject. In one embodiment, the endodermal differentiation media contains Activin A, Noggin, and a GSK3β inhibitor. In another embodiment, the endodermal marker is GATA6. In one embodiment, the iPSCs are cultured in a three-dimensional culture. In one embodiment, the iPSCs are cultured in Matrigel. In another embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In another embodiment, the mesenchymal cells comprise urogenital mesenchyme. In another embodiment, the mesenchymal cells comprise bladder mesenchyme. In another embodiment, the graft is a renal graft. In another embodiment, the organ tissue is prostate epithelial tissue. In another embodiment, the organ tissue is bladder epithelial tissue. In another embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In another embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In another embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In another embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In another embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In another embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawings will be provided by the Office upon request and payment of the necessary fee.
Stem cell biologists have sought to generate desired cell types by activating lineage-specific differentiation pathways in the context of pluripotent embryonic stem cells (ESC) or induced pluripotent stem cells (iPSC). The directed differentiation of many epithelial cell types from ESC or iPSC can be challenging, perhaps since they typically reside in heterogeneous tissues containing multiple epithelial cell types within a stromal microenvironment. To overcome this challenge, the invention provides for the use of appropriate cell culture systems as well as tissue recombination methods in which mesenchymal cells are supplied to promote differentiation.
There has also been interest in transdifferentiation as another method for the generation of desired cell types [A1, A2], starting from the original demonstration that MyoD can be a master regulator that can reprogram fibroblasts into muscle cells [A3]. Furthermore, the generation of iPSC by Yamanaka and colleagues through ectopic expression of four “pluripotency factors” (OSKM: Oct4, Sox2, Klf4, c-Myc) [A4] has caused a resurgence of interest in molecular mechanisms of transdifferentiation. Several studies have now demonstrated that expression of lineage-specific master regulators can promote direct conversion or transdifferentiation from one mature differentiated cell type into a distinct differentiated cell type in the apparent absence of an intermediate pluripotent state. For example, fibroblasts can be directly converted to neurons or cardiomyocytes in culture by expression of lineage-specific MR genes [A5-A9], while induction of the pluripotency gene Oct4 combined with cytokine treatment can generate hematopoietic progenitors [A10].
An alternative approach for direct conversion, which has been termed “primed conversion” or “indirect lineage conversion” [A1, A2], has been to use transient expression of pluripotency factors to induce a plastic developmental state permissive for transdifferentiation into desired cell fates after exposure to appropriate external cues, such as specific cell culture conditions [A11, A12]. Neural progenitors generated by this methodology can be expanded in culture and generate different neuronal and glial types after multiple passages [A12, A13]. Thus, pluripotency factors can induce an epigenetically unstable state that is responsive to environmental signals and can be directed to lineage-specific progenitors and differentiated derivatives. The combination of this approach with the expression of lineage-specific master regulators can provide additional specificity or higher efficiency of direct conversion.
For direct conversion approaches, the generation of entire tissue, not just specific cell types, is desirable. This can be accomplished for epithelial tissues by combining epithelial progenitors generated by transdifferentiation with mesenchymal/stromal tissue that is specific for the tissue of interest, thereby recapitulating normal processes of organogenesis. In the case of the prostate, this approach can take advantage of a classic assay for prostate formation involving tissue recombination with rodent embryonic urogenital mesenchyme and renal grafting [A14, A15], which has been used for several studies of prostate differentiation and stem cell function [A16-A21]. This assay has been used for analyses of prostate stem/progenitor cells [A20-A23], and has also shown that human ESC can generate prostate epithelium in the context of teratomas following tissue recombination [A24]. Furthermore, embryonic urogenital mesenchyme is known to have potent reprogramming activity in tissue recombination assays, being capable of respecifying a range of epithelial cell types, such as bladder, vaginal, and mammary gland, to prostate epithelium [A15, A25-A27]. The contribution of organ-specific mesenchyme in enforcing correct lineage-specification and expansion of tissue progenitors has also been recognized for directed differentiation from pluripotent stem cells in culture [A28]. Direct conversion or differentiation to appropriate stem/progenitor cells, such as the prostate luminal stem cells that have previously been identified [A20], can enhance the production of desired cell types of interest.
Systems Analysis of Lineage Specific Master Regulators
The success and efficiency of direct conversion/transdifferentiation approaches depend upon the identification of suitable lineage-specific master regulator (MR) genes that can drive the direct conversion process. Candidate gene approaches to identify such MRs have been used, often by starting with a list of 10-20 transcription factors known to be important in the development and/or differentiation of the cell type of interest. This methodology relies upon the existence of a considerable body of literature on the cell type/tissue of interest, and is not feasible for cell types/tissues that are less well understood.
Candidate MRs for direct conversion can be systematically identified using a systems biology approach. Until recently, the molecular mechanisms underlying cell fate specification have been investigated without the benefit of comprehensive maps of the regulatory interactions that control lineage-specific differentiation. Recent work has led to the development of a large repertoire of computational methods for dissecting the molecular interactions that define the regulatory logic of cells and tissues. Methods for the dissection of cell type-specific regulatory networks and for identification of drivers of both physiological and pathological biological processes can be used. These include methods to infer transcriptional (ARACNe [A29, A30]) and post-translational (MINDy [A31]) interactions from large mRNA profile datasets. The resulting regulatory networks can then be interrogated to identify MR genes whose activity is both necessary and sufficient to implement a specific physiologic or pathologic cell state [A32, A33]. For example, this approach elucidated the synergistic role of the transcription factors C/EBPβ and Stat3 in reprogramming neural stem cells along a mesenchymal lineage [A32], and of the Huwe1-n-Myc-D113 cascade in brain morphogenesis in vivo [A34]. Without being bound by theory, the availability of an appropriate interactome and of signatures representing the gene expression differences of a progenitor state versus a fully differentiated tissue/cell type of interest can allow inference of MR genes governing transitions between these states that can be experimentally validated [A32, A33].
These computational/systems can be used for the identification of MRs of biological processes of interest. This methodology is unbiased, as it does not rely upon prior biological knowledge from functional studies using molecular genetic approaches. Many systems-based approaches have used expression profiling to identify differentially expressed genes, with the premise that highly differentially expressed genes can be enriched for master regulators. In contrast, the MARINa algorithm identifies candidate MRs on the basis of the differential expression of their inferred targets, and consequently can identify MRs that are not themselves differentially expressed, but display differential activity, for example, as a result of post-transcriptional regulation or post-translational modification such as phosphorylation.
Cancer Modeling by Gene Targeting and its Application to Human Prostate Cancer
Genetically-engineered mouse models of cancer have led to advances in understanding the biological and molecular mechanisms of cancer initiation and progression. Genetically-engineered mice can be intrinsically limited as models of human disease due to lack of conservation of tissue morphology, physiological states, and/or molecular pathways and regulatory genes. It is fundamentally important to generate appropriate human cancer models, but, the creation of precise genetically-engineered models can be hampered by technical difficulties with gene targeting in human cells.
Reagents, including zinc-finger nucleases and TALE nucleases (TALENs), can be used as gene targeting methods in experimental systems that have previously not been amenable to such approaches [A35]. TALENs correspond to fusions of sequence-specific TALE DNA-binding domains with the FokI restriction endonuclease [A36, A37], and can be engineered to bind and create a double-stranded break at a specific DNA sequence of interest in genomic DNA. TALENs have technical advantages since TALENs of any desired target specificity can be readily generated from standard starting reagents [A38]. Such TALENs can be used to mutate target genes by small insertions/deletions generated by TALEN-mediated double-strand DNA cleavage followed by non-homologous end-joining, or can be used as the basis for homologous recombination using an insertion vector as is the case for gene targeting in mouse ESCs. TALENs can be used for genetic engineering of human cells using approaches that have been well-developed over the past twenty years for manipulation of mouse ESC. The TALEN methodology is high-efficiency (often able to target both alleles in a single targeting experiment), non-cytotoxic, and has minimal off-target effects [A36, A37].
TALEN-mediated gene targeting can be utilized for the generation of genetically-engineered human models of cancer by mutation of tumor suppressor genes. In combination with direct conversion to generate tissues/cell types of interest, TALEN-mediated targeting can be used in fibroblasts or directly converted progeny cells to mutate target genes, followed by generation of human tissue that is cancer-prone or is undergoing cancer initiation. Since there are histological and physiological differences between the rodent and human prostate that limit the applicability of mouse models, these methods can be used for the generation of models of human prostate cancer. Genetically-engineered human models of prostate cancer based on gene targeting do not currently exist. An existing model that uses human prostate cells for oncogene overexpression in renal grafts [A39] uses primary normal prostate epithelial cells, which are difficult to obtain and cannot be propagated for use in gene targeting approaches.
The availability of genetically-engineered human models of prostate cancer can allow for the direct experimental analysis of prostate cancer initiation. The early events of human prostate cancer formation are poorly understood, due to the general lack of availability of human prostate tissue from men prior to clinical presentation of the disease [A40]. It is unclear when clinically-significant prostate cancer actually arises. Although prostate tissue from men in the twenties and thirties can contain localized areas of prostatic intraepithelial neoplasia (PIN) and latent adenocarcinoma, it is unknown whether this latent prostate cancer actually progresses to give rise to clinically aggressive disease in much older men (discussed in [A40]). Instead, this latent disease may be related to low-grade prostate cancer (histological Gleason grade 6 and 7 (3+4)) that is considered indolent and does not generally require treatment, whereas more aggressive prostate cancer (Gleason grade 7 (4+3) and above) can have an entirely different origin. There can be different origins of human prostate cancer that can be clinically distinct in terms of outcome, and it is unknown whether these differences are related to the mutational events that occur in prostate cancer initiation.
The invention provides for a direct conversion approach that can generate an entire tissue, not just a desired cell type of interest. In some embodiments, a computational systems biology approach can be used for the comprehensive identification of master regulator genes to optimize the direct conversion process. This approach can be combined with new gene targeting methods for the generation of novel genetically-engineered models of human cancer. Without being bound by theory, these approaches can be utilized for the analysis of human prostate cancer, but can also be used to model tumorigenesis in other tissues, as well as other diseases. For example, issues of primary clinical importance can be addressed, such as the molecular mechanisms that underlie the initiation and progression of human prostate cancer as the basis for aggressive versus indolent disease.
The invention is directed to methods for generating induced organ tissues. For example, the invention is directed to methods for the directed differentiation of mouse induced pluripotent stem cells (iPSC). The invention is also directed to transdifferentiation of mouse fibroblasts into prostate and urinary bladder epithelium, which have considerable clinical relevance for the patient-specific generation of normal and transformed prostate and bladder tissue. In one embodiment, the invention provides for methods of generating prostate tissue. In another embodiment, the invention provides for methods of generating bladder tissue. In some embodiments, the tissue is generated in vivo.
The invention encompasses methods for reprogramming fibroblast cells in culture, which are able to generate generic epithelial cells therefrom. These “primitive” epithelial cells can serve as the starting point for epithelial tissue formation in vivo upon transduction with specific tissue master regulatory genes together with grafting or co-culture of appropriate inductive mesenchyme or mesenchymal cells. Such tissues obtained by reprogramming include, but are not limited to prostate, urinary bladder, mammary gland, lung, as well as others.
Early stages of human prostate cancer are androgen-driven and thus respond to androgen-ablation therapy. However, in most cases a relapse occurs as a castration-resistant disease, which is progressive, metastatic and invariably lethal. These findings render mouse studies focused on generating new tissue engineering technologies to investigate the early events of prostate tumorigenesis highly relevant for human disease. Another leading cause of mortality in both men and women is urinary bladder cancer. In 90% of the cases, bladder cancer presents as urothelial cell carcinomas. In most cases, the treatment involves removal of the bladder wall followed by reconstructive surgeries, cystoplasty usually involving colon epithelium. These interventions leave the patient with highly debilitating long-term problems. Although a superior alternative, obtaining healthy functional autologous bladder urothelium has proved a challenging objective.
In one embodiment, the invention encompasses understanding the pathways involved in cellular identity and plasticity, as well as for developing patient-specific cell-based therapies for prostate and bladder disease. This approach can allow for the analysis of human prostate cancer initiation and early progression through the oncogenic transformation of prostate tissue generated by reprogramming. For example, such methods can allow for the analysis of the molecular basis for the differences between indolent and aggressive prostate cancer, which is likely to be established by early events in cancer initiation and progression [49]. This could lead to detection of new early prognostic biomarkers and would offer a new solution for drug screening. Generating bladder urothelium could have a more direct clinical applicability in regenerative medicine for patients with highly debilitating bladder exstrophy or cancer surgeries who need cystoplasty. More generally, the ability to generate patient-specific epithelial cell types from tissues that are otherwise difficult to access would represent a major advance in personalized and regenerative medicine.
Based on recent reprogramming studies [1, 2], the inherent plasticity of readily-accessible fibroblasts can be exploited to generate specific tissues (such as prostate and bladder epithelia) through a combination of reprogramming factors and tissue specific master regulator genes. As discussed in the Examples herein, mouse embryonic fibroblasts can be directly converted into epithelial cells in culture following expression of reprogramming factors, in the absence of an intermediate pluripotent stage. Moreover, these induced epithelial cells are amenable to further terminal differentiation into prostatic or bladder tissue in vivo in tissue recombination assays.
The invention encompasses methods directed to differentiation of mouse induced pluripotent stem cells (iPSC) into prostate and bladder epithelium by activation of master regulator genes of normal prostate and bladder epithelium, identified by bioinformatic analysis of regulatory genetic networks for mouse and human prostate or available from previous studies on urinary bladder development [3]. Expression of putative master regulator genes for prostate and bladder epithelium identified computationally or by a candidate gene approach can enhance prostate and bladder-specific differentiation of iPSC in tissue recombination experiments. In one embodiment, iPSC derived from various genetic backgrounds can be differentiated into mature epithelia through a temporal series of growth factors, genetic manipulations and in vivo recombination assays to mimic embryonic prostate and bladder development.
The invention further encompasses methods directed to conversion of mouse fibroblasts into prostate and bladder epithelium by transient expression of pluripotency factors (Oct4, Sox2, Klf4, c-Myc) to promote the directed transdifferentiation of mouse embryonic fibroblasts (MEFs) and human fibroblasts to “primitive” epithelial cells (iEpi) without undergoing an intermediate pluripotent state. Epithelial cells can be further directed toward prostate or bladder fate through expression of tissue specific master regulators and a pro-epithelial culture system. In one embodiment, MEFs derived from various genetic backgrounds and human fibroblasts can be briefly exposed to the pluripotency factors followed by transduction with prostate or bladder specific factors and cultured in epithelial conditions. In another embodiment, specific cell culture conditions (e.g., three-dimensional culture in Matrigel, co-culture with stromal cells) or tissue recombination assays can enhance the differentiation of desired epithelial cell.
The proposed studies aim at generating new ways to obtain complex tissues in vivo with a direct applicability in regenerative medicine. The resulting system would allow for functional studies to investigate the molecular nature of prostate tumorigenesis initiation in various oncogenic set-ups, and could lead to discovery of patient-specific early prognostic markers. Eventually, iPSC- and transdifferentiation-derived human bladder tissue could be considered for transplantation-based therapies in congenital defects (such as bladder exstrophy) or organ rehabilitation following cancer surgeries.
Direct Transdifferentiation in Regenerative Medicine and Disease Modeling
Stem cell biologists have sought to generate desired cell types by recapitulation of normal lineage-specific differentiation pathways from a pluripotent embryonic stem cell (ESC) or induced pluripotent stem cell (iPSC). To date, however, the directed differentiation of many epithelial cell types from ESC or iPSC has been relatively challenging, perhaps since they typically reside in a tissue containing multiple epithelial cell types within a stromal microenvironment. To overcome this challenge, the invention provides for the use of appropriate cell culture systems as well as tissue recombination methods in which mesenchymal cells are supplied to promote differentiation. Directed differentiation to appropriate adult stem/progenitor cells, such as the prostate luminal stem cells previously identified [4], can enhance the production of desired cell types of interest.
Previous studies have shown that human ESC can undergo complex differentiation along an endodermal lineage to generate prostate epithelium following recombination with rodent embryonic urogenital mesenchyme (UGM) and renal grafting [5, 6]. Similar to prostate, proper bladder development is dependent on proper stromal-epithelial crosstalk and paracrine signaling [7-10]. Tissue recombination techniques were employed to recapitulate bladder epithelium formation. Thus, embryonic bladder mesenchyme (EBLM) induces bladder morphogenesis when grafted together with mouse ESC [11] or bone marrow derived mesenchymal stem cells in tissue recombination models [12].
Prostate and bladder represent two functionally different types of epithelia. While prostate tissue is essentially a secretory glandular epthelium, the bladder is lined by urothelium, a permeability barrier epithelium, surrounded by lamina propria and a smooth muscle layer [13]. However, they appear similar from the point of view of tissue remodeling. Both prostate and urinary bladder are hindgut endodermal derivatives. The prostate develops from the pelvic (middle part) of the urogenital sinus (UGS), while urinary bladder forms from the cranial end of the UGS. Moreover, urogenital sinus mesenchyme (UGM) reprogrammed adult bladder epithelium to transdifferentiate into glandular epithelium in tissue recombination and renal grafting experiments [14]. Without being bound by theory, bladder and prostate can share a common stem cell/progenitor that is controlled by different inductive mesenchyme [11].
The efficiency of directed differentiation of pluripotent stem cells could be enhanced by the expression of lineage-specific master regulator genes that specify cell types of interest and can promote their differentiation. Without being bound by theory, such regulators can be determined by a candidate gene approach, or can be systematically identified using an unbiased reversed engineering approach. The candidate gene approach has been developed to generate and interrogate genome-wide regulatory networks, or interactomes, for cell types and tissues of interest [15-17]. The availability of such interactomes together with gene signatures of the tissue/cell types of interest allows the identification of master regulator genes that govern transitions to the differentiated cell type of interest [18, 19].
In one embodiment, lineage-specific master regulators can be used as an alternative approach to promote direct transdifferentiation from a distinct mature differentiated cell type in the absence of an intermediate pluripotent state. For instance, expression of four master regulator genes is sufficient to promote pancreatic beta-cell differentiation in vivo, albeit at low frequencies [20]; fibroblasts can be directly converted to neurons or cardiomyocytes in culture by expression of lineage-specific master regulator genes [21-23]; induction of the pluripotency gene Oct4 combined with cytokine treatment can generate hematopoietic progenitors [24]; and specific combinations of factors (Hnf4α, Foxa1, Foxa3, Gata4) can generate in vitro functional and proliferative hepatocyte-like cells from mouse fibroblasts [25, 26]. Moreover, the general reprogramming approach can be modified to serve as a platform for transdifferentiation [2]. Thus, transient expression of the four “pluripotency factors” (Oct4, Sox2, Klf4, c-Myc) in fibroblasts can lead to a plastic developmental state permissive for transdifferentiation into desired cell fates after exposure to appropriate external cues [27, 28]. Neural progenitors generated by this methodology can be expanded in culture and generate different neuronal and glial types after multiple passages [28]. Thus, pluripotency factors can induce an epigenetically unstable state that is responsive to environmental signals and can be directed to lineage-specific progenitors and differentiated derivatives. Directed transdifferentiation approaches can potentially overcome inherent limitations in the use of pluripotent cells for personalized treatments or regenerative medicine, such as low yields of differentiated cells, the need to generate patient-specific iPSC, or persistence of tumorigenic pluripotent cells.
Master Regulators of Direct Reprogramming to Prostate and Bladder Epithelium
As part of the candidate gene approach, an embodiment of the invention encompasses investigating whether genes with known biological function in regulating the developmental processes related to prostate and bladder are also appropriate master regulators of direct reprogramming.
The prostate is a secretory tissue of endodermal origin whose function is regulated by male sex hormones. Gene inactivation studies in the mouse, stem cell tracing mouse models combined with organ culture and tissue recombination assays, have highlighted the essential roles of androgenic signaling, epithelial-stromal interactions and specific stem cell populations in directing prostate development and regeneration[29]. The androgen receptor (AR) signaling axis plays a critical role in the development, function and homeostasis of the prostate[30, 31]. Mouse Nkx3.1 homeobox gene is the earliest known marker of prostate epithelium during embryogenesis and is subsequently expressed at all stages of prostate differentiation in vivo as well as in tissue recombinants. In the absence of Nkx3.1, the prostate ductal morphogenesis and secretory functions are disrupted [32]. Previous studies have placed the homeobox gene Nkx3.1, an important known regulator of prostate epithelial differentiation, at the center of prostate tissue homeostasis as a marker of a stem cell population active during prostate regeneration[29]. Based on genetic lineage-tracing analyses in mouse models, this work has shown that prostate stem cells reside among the Nkx3.1-positive luminal population, are castration resistant (Castration-resistant Nkx3.1-expressing cells, CARNs) and are able to regenerate prostatic glandular tissue after castration in an androgen-dependent manner [29]. Mouse Foxa1 expression marks the entire embryonic urogenital sinus epithelium (UGE), while Foxa2 is restricted to the basally located cells during prostate budding. Foxa1 plays a critical role in timing of prostate morphogenesis and cell differentiation. In Foxa1 deficient mice, the prostate has an abnormal ductal pattern composed of primitive epithelial cords surrounded by thick stromal layers [33]. Thus, the prostate epithelium development is blocked at a level similar to embryonic UGE and the primitive epithelial cells do not progress to differentiated and mature epithelial cells [33].
A recent study discussed the role for KLF5 in the formation and terminal differentiation of the urothelium [3]. When KLF5 is missing from the bladder epithelial cells, urothelial precursor cells remain in an undifferentiated state and the resulting urothelium fails to stratify and to express terminal differentiation markers (e.g. uroplakins). Moreover, the study uncovered and validated a plethora of transcriptional targets among the genes known to be coordinately expressed with KLF5 in the developing bladder: Pparγ, Grhl3, Ovol1, Foxa1, Elf3 and Ehf. Most importantly, Pparγ and Grhl3 participate in a KLF5-dependent gene network regulating maturation of the urothelium [3]. This study introduced order in the “black box” of the pathways involved in bladder development and opened the possibility that KLF5 could function as a master regulator of the reprogramming patterns in urothelium.
Without being bound by theory, focusing on a small number of core genes can significantly bias studies because other key players in determining epithelial tissue self-renewal and differentiation hierarchy would not be explored. An integrative systems biology approach can uncover whole gene pathways and networks, as well as new individual gene products which could be further validated experimentally. In one embodiment, the invention encompasses identifying and validating new master regulators (MRs) of epithelial reprogramming through unbiased genome-wide analysis of prostate and bladder urothelium.
Recent studies used powerful computational techniques of reverse-engineering designed to generate unbiased transcriptional and post-translational regulatory gene networks, or “interactomes” [17, 34]. These include an algorithm for the reconstruction of accurate cellular networks (ARACNe) [17], MARINa, for identification of most likely master regulators of specific expression signatures [18], MINDy, for the inference of post-transcriptional modulators of transcription factor activity [35], and master regulator analysis (MRA) [36]. These algorithms have accurately identified regulators of several human malignancies. Interrogation of a high-grade glioma interactome successfully identified two master regulator genes (C/EBPβ/δ and Stat3) that can reprogram neural stem cells along a mesenchymal lineage and that were validated both in vitro and in vivo [19]. In one embodiment, computational/systems biology approaches are used to construct genome-wide regulatory networks (interactomes) for mouse and human prostate tissue to allow identification of master regulator genes that govern prostate epithelial cell fates.
Methods for Isolating or Purifying Fibroblast Cells
The present invention provides methods for separating, enriching, isolating or purifying fibroblast cells from a tissue or mixed population of cells. The methods comprise obtaining a mixed population of cells, contacting the population of cells with an agent that binds to a mesenchymal marker, for example CD140a, and separating the subpopulation of cells that are bound by the agent from the subpopulation of cells that are not bound by the agent, wherein the subpopulation of cells that are bound by the agent is enriched for the mesenchymal marker (for example, CD140a-positive fibroblasts). The methods described herein may be performed using any mesenchymal marker known in the art, including, but not limited to N-cadherin (CD325), CD44, CD90, CD105, CD29, Sca-1, SSEA-4, vimentin, CD73, CD166, BMPR-1A, BMPR-1B, BMPR-II, CDCP1, fibronectin, CD49a, CD51, CD56, nestin, c-kit, STRO-1, and CD106.
The methods for separating, enriching, isolating or purifying fibroblast cells from a mixed population of cells according to the invention may be combined with other methods for separating, enriching, isolating or purifying fibroblast cells that are known in the art (for example, U.S. Pat. No. 4,777,145, U.S. Pat. No. 8,004,661, U.S. Pat. No. 5,367,474, U.S. Pat. No. 4,347,935) and are described in P. T. Sharpe, 1988, Laboratory Techniques in Biochemistry and Molecular Biology Volume 18: Methods of Cell Separation, Elsevier, Amsterdam; M. Zborowski and J. J. Chalmers, 2007, Laboratory Techniques in Biochemistry and Molecular Biology Volume 32: Magnetic Cell Separation, Elsevier, Amsterdam; and T. S. Hawley and R. G. Hawley, 2005, Methods in Molecular Biology Volume 263: Flow Cytometry Protocols, Humana Press Inc, Totowa, N.J. For example, the methods described herein may be performed in conjunction with techniques that use other markers. For example, additional selection steps maybe performed either before, after, or simultaneously with the mesenchymal marker selection step, in which a second agent, such as an antibody, that binds to a second marker is used, separating the subpopulation of cells that are bound by the agent from the subpopulation that are not bound by the agent, wherein the subpopulation of cells that are not bound by the agent is enriched. The second marker may be any marker known in the art that reduces the heterogeneity of the fibroblast population. For example, the second marker is the lineage surface antigens (Lin), Mac-1(CD11b), or epithelial cell adhesion molecule (EpCAM). In one embodiment, the second marker is a marker for blood cells (for example lineage surface antigens (Lin), Mac-1(CD11b), CD2, CD3, CD4, CD5, CD8, CD14, CD16, CD19, CD20, CD56, Ter119, B220, CD33, CD15, or CD45). In another embodiment, the second marker is a marker for endothelial cells (for example, CD34, CD146, CD202b, CD62e, CD54, VEGFR3, CD106, CD144, or CD309). In a further embodiment, the second marker is a marker for epithelial cells (for example, CD44R, CD66a, CD75, CD104, CD167, cytokeratin, EpCAM (CD326), CD138, or E-cadherin). In another embodiment, the second marker is a combination of any markers known in the art that reduce the heterogeneity of the fibroblast population (for example, Lin/Mac-1(CD11b)/EpCAM). The mixed population of cells can be any source of cells from which to obtain fibroblasts, including but not limited to an E13.5 mouse embryo, a P0 mouse, or a human foreskin. In one embodiment, mouse embryonic fibroblasts can be obtained from E13.5 mouse embryos. In another embodiment, mouse dermal fibroblasts can be obtained from P0 mice. In a further embodiment, BJ normal human foreskin fibroblasts can be obtained from human foreskins or from the American Type Culture Collection (for example cell line number CRL-2522).
The agent used can be any agent that binds to the mesenchymal marker (for example, CD140a), or the markers known in the art that reduce the heterogeneity of the fibroblast population (for example, Lin/Mac-1(CD11b)/EpCAM). The term “Agent” includes, but is not limited to small molecule drugs, peptides, proteins, peptidomimetic molecules, and antibodies. It also includes any molecule that binds to the mesenchymal marker, or to markers known in the art that reduce the heterogeneity of the fibroblast population, that is labeled with a detectable moiety, such as a histological stain, an enzyme substrate, a fluorescent moiety, a magnetic moiety or a radio-labeled moiety. Such “labeled” agents are particularly useful for embodiments involving isolation or purification of CD 140 positive cells, or detection of CD 140-positive cells, or isolation or purification of Lin/Mac-1(CD11b)/EpCAM negative cells. In some embodiments, the agent is an antibody that binds to CD140, Lin, Mac-1(CD11b), or EpCAM.
There are many cell separation techniques known in the art (U.S. Pat. No. 4,777,145, U.S. Pat. No. 8,004,661, U.S. Pat. No. 5,367,474, U.S. Pat. No. 4,347,935), and any such technique may be used. For example magnetic cell separation techniques can be used if the agent is labeled with an iron-containing moiety. Cells may also be passed over a solid support that has been conjugated to an agent that binds to a marker, such that the marker positive cells will be selectively retained on the solid support. Cells may also be separated by density gradient methods, particularly if the agent selected significantly increases the density of the marker positive cells to which it binds. For example, the agent can be a fluorescently labeled antibody against the marker, and the marker positive cells are separated from the other cells using fluorescence activated cell sorting (FACS).
DNA Manipulation for Reprogramming Factors and Master Regulatory Genes
One skilled in the art understands that polypeptides (for example Oct4, Sox2, Klf4, c-Myc, NKX3.1, Androgen receptor (AR), FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, Ehf, and the like) can be obtained in several ways, which include but are not limited to, expressing a nucleotide sequence encoding the protein of interest by genetic engineering methods.
The invention provides for a nucleic acid encoding a reprogramming factor molecule, such as an Oct4 molecule, a Sox2 molecule, a Klf4 molecule, a c-Myc molecule, or a combination thereof. The invention further provides for a nucleic acid encoding a master regulatory molecule, such as a NKX3.1 molecule, an AR molecule, a FOXA1 molecule, a FOXA2 molecule, a KLF5 molecule, a Pparγ molecule, a Grhl3 molecule, a Elf3 molecule, a Ehf molecule, or a combination thereof. In one embodiment, the molecule (such as an Oct4 molecule, a Sox2 molecule, a Klf4 molecule, a c-Myc molecule, a NKX3.1 molecule, an AR molecule, a FOXA1 molecule, a FOXA2 molecule, a KLF5 molecule, a Pparγ molecule, a Grhl3 molecule, a Elf3 molecule, or a Ehf molecule) comprises an expression cassette, for example to achieve overexpression in a cell. The nucleic acids of the invention can be an RNA, cDNA, cDNA-like, or a DNA nucleic acid molecule of interest in an expressible format, such as an expression cassette, which can be expressed from the natural promoter or a derivative thereof or an entirely heterologous promoter. The nucleic acid of interest can encode a protein (for example, Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf), and may or may not include introns. The nucleic acid of interest can encode only a single protein (for example, Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf), or can encode for more than one protein of interest (for example, combinations of Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf).
For example, the polypeptide sequence of human OCT4 (isoform 1) is depicted in SEQ ID NO: 1. OCT 4 is also known as POU5F1 (POU class 5 homeobox 1). The nucleotide sequence of human OCT4 (isoform 1) is shown in SEQ ID NO: 2. Sequence information related to OCT4 (isoform 1) is accessible in public databases by GenBank Accession numbers NP—002692.2 (protein) and NM—002701.4 (nucleic acid).
Sequence information related to OCT4 (isoform 2) is accessible in public databases by GenBank Accession numbers NP—976034.4 (protein) and NM—203289.4 (nucleic acid).
Sequence information related to OCT4 (transcript variant 3) is accessible in public databases by GenBank Accession numbers NP—001167002.1 (protein) and NM—001173531.1 (nucleic acid).
SEQ ID NO: 1 is the human wild type amino acid sequence corresponding to OCT4 isoform 1 (residues 1-360):
SEQ ID NO: 2 is the human wild type nucleotide sequence corresponding to OCT4 (isoform 1) (nucleotides 1-1411), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human SOX2 is depicted in SEQ ID NO: 3. The nucleotide sequence of human SOX2 is shown in SEQ ID NO: 4. Sequence information related to SOX2 is accessible in public databases by GenBank Accession numbers NP—003097.1 (protein) and NM—003106.3 (nucleic acid).
SEQ ID NO: 3 is the human wild type amino acid sequence corresponding to SOX2 (residues 1-317):
SEQ ID NO: 4 is the human wild type nucleotide sequence corresponding to SOX2 (nucleotides 1-2520), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human KLF4 is depicted in SEQ ID NO: 5. The nucleotide sequence of human KLF4 is shown in SEQ ID NO: 6. Sequence information related to KLF4 is accessible in public databases by GenBank Accession numbers NP—004226.3 (protein) and NM—004235.4 (nucleic acid).
SEQ ID NO: 5 is the human wild type amino acid sequence corresponding to KLF4 (residues 1-479):
SEQ ID NO: 6 is the human wild type nucleotide sequence corresponding to KLF4 (nucleotides 1-2949), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human c-MYC is depicted in SEQ ID NO: 7. c-MYC is also known as MYC. The nucleotide sequence of human c-MYC is shown in SEQ ID NO: 8. Sequence information related to c-MYC is accessible in public databases by GenBank Accession numbers NP—002458.2 (protein) and NM—002467.4 (nucleic acid).
SEQ ID NO: 7 is the human wild type amino acid sequence corresponding to c-MYC (residues 1-454):
SEQ ID NO: 8 is the human wild type nucleotide sequence corresponding to c-MYC (nucleotides 1-2379), wherein the underscored bolded “CTG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human NKX3.1 (isoform 1) is depicted in SEQ ID NO: 9. The nucleotide sequence of human NKX3.1 (isoform 1) is shown in SEQ ID NO: 10. Sequence information related to NKX3.1 (isoform 1) is accessible in public databases by GenBank Accession numbers NP—006158.2 (protein) and NM—006167.3 (nucleic acid).
Sequence information related to NKX3.1 (isoform 2) is accessible in public databases by GenBank Accession numbers NP—1243268.1 (protein) and NM—1256339.1 (nucleic acid).
SEQ ID NO: 9 is the human wild type amino acid sequence corresponding to NKX3.1 (isoform 1) (residues 1-234):
SEQ ID NO: 10 is the human wild type nucleotide sequence corresponding to NKX3.1 (isoform 1) (nucleotides 1-3281), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human AR (Androgen Receptor) (isoform 1) is depicted in SEQ ID NO: 11. The nucleotide sequence of human AR (isoform 1) is shown in SEQ ID NO: 12. Sequence information related to AR (isoform 1) is accessible in public databases by GenBank Accession numbers NP—000035.2 (protein) and NM—000044.3 (nucleic acid).
Sequence information related to AR (isoform 2) is accessible in public databases by GenBank Accession numbers NP—1011645.1 (protein) and NM—10111645.2 (nucleic acid).
SEQ ID NO: 11 is the human wild type amino acid sequence corresponding to AR (isoform 1) (residues 1-920):
SEQ ID NO: 12 is the human wild type nucleotide sequence corresponding to AR (isoform 1) (nucleotides 1-10661), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human FOXA1 is depicted in SEQ ID NO: 13. The nucleotide sequence of human FOXA1 is shown in SEQ ID NO: 14. Sequence information related to FOXA1 is accessible in public databases by GenBank Accession numbers NP—004487.2 (protein) and NM—004496.3 (nucleic acid).
SEQ ID NO: 13 is the human wild type amino acid sequence corresponding to FOXA1 (residues 1-472):
SEQ ID NO: 14 is the human wild type nucleotide sequence corresponding to FOXA1 (nucleotides 1-3396), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human FOXA2 (isoform 1) is depicted in SEQ ID NO: 15. The nucleotide sequence of human FOXA2 (isoform 1) is shown in SEQ ID NO: 16. Sequence information related to FOXA2 (isoform 1) is accessible in public databases by GenBank Accession numbers NP—068556.2 (protein) and NM—021784.4 (nucleic acid).
Sequence information related to FOXA2 (isoform 2) is accessible in public databases by GenBank Accession numbers NP—710141.1 (protein) and NM—153675.2 (nucleic acid).
SEQ ID NO: 15 is the human wild type amino acid sequence corresponding to FOXA2 (isoform 1) (residues 1-463):
SEQ ID NO: 16 is the human wild type nucleotide sequence corresponding to FOXA2 (isoform 1) (nucleotides 1-2428), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human KLF5 is depicted in SEQ ID NO: 17. The nucleotide sequence of human KLF5 is shown in SEQ ID NO: 18. Sequence information related to KLF5 is accessible in public databases by GenBank Accession numbers NP—001721.2 (protein) and NM—001730.3 (nucleic acid).
SEQ ID NO: 17 is the human wild type amino acid sequence corresponding to KLF5 (residues 1-457):
SEQ ID NO: 18 is the human wild type nucleotide sequence corresponding to KLF5 (nucleotides 1-3350), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human PPARγ (isoform 1, variant 1) is depicted in SEQ ID NO: 19. PPARγ is also known as PPARG. The nucleotide sequence of human PPARγ (isoform 1, variant 1) is shown in SEQ ID NO: 20. Sequence information related to PPARγ (isoform 1, variant 1) is accessible in public databases by GenBank Accession numbers NP—619726.2 (protein) and NM—138712.3 (nucleic acid).
Sequence information related to PPARγ (isoform 1, variant 3) is accessible in public databases by GenBank Accession numbers NP—619725.2 (protein) and NM—138711.3 (nucleic acid).
Sequence information related to PPARγ (isoform 1, variant 4) is accessible in public databases by GenBank Accession numbers NP—005028.4 (protein) and NM—005037.5 (nucleic acid).
Sequence information related to PPARγ (isoform 2, variant 2) is accessible in public databases by GenBank Accession numbers NP—056953.2 (protein) and NM—015869.4 (nucleic acid).
SEQ ID NO: 19 is the human wild type amino acid sequence corresponding to PPARγ (isoform 1, variant 1) (residues 1-477):
SEQ ID NO: 20 is the human wild type nucleotide sequence corresponding to PPARγ (isoform 1, variant 1) (nucleotides 1-1892), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human GRHL3 (isoform 1) is depicted in SEQ ID NO: 21. The nucleotide sequence of human GRHL3 (isoform 1) is shown in SEQ ID NO: 22. Sequence information related to GRHL3 (isoform 1) is accessible in public databases by GenBank Accession numbers NP—067003.2 (protein) and NM—021180.3 (nucleic acid).
Sequence information related to GRHL3 (isoform 2) is accessible in public databases by GenBank Accession numbers NP—937816.1 (protein) and NM—198173.2 (nucleic acid).
Sequence information related to GRHL3 (isoform 3) is accessible in public databases by GenBank Accession numbers NP—937817.3 (protein) and NM—198174.2 (nucleic acid).
Sequence information related to GRHL3 (isoform 4) is accessible in public databases by GenBank Accession numbers NP—1181939.1 (protein) and NM—1195010.1 (nucleic acid).
SEQ ID NO: 21 is the human wild type amino acid sequence corresponding to GRHL3 (isoform 1) (residues 1-607):
SEQ ID NO: 22 is the human wild type nucleotide sequence corresponding to GRHL3 (isoform 1) (nucleotides 1-2710), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human ELF3 (transcript variant 1) is depicted in SEQ ID NO: 23. The nucleotide sequence of human ELF3 (transcript variant 1) is shown in SEQ ID NO: 24. Sequence information related to ELF3 (transcript variant 1) is accessible in public databases by GenBank Accession numbers NP—004424.3 (protein) and NM—004433.4 (nucleic acid).
Sequence information related to ELF3 (transcript variant 2) is accessible in public databases by GenBank Accession numbers NP—1107781.1 (protein) and NM—1114309.1 (nucleic acid).
SEQ ID NO: 23 is the human wild type amino acid sequence corresponding to ELF3 (transcript variant 1) (residues 1-371):
SEQ ID NO: 24 is the human wild type nucleotide sequence corresponding to ELF3 (transcript variant 1) (nucleotides 1-3149), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
For example, the polypeptide sequence of human EHF (isoform 1) is depicted in SEQ ID NO: 25. The nucleotide sequence of human EHF (isoform 1) is shown in SEQ ID NO: 26. Sequence information related to EHF (isoform 1) is accessible in public databases by GenBank Accession numbers NP—1193545.1 (protein) and NM—1206616.1 (nucleic acid).
Sequence information related to EHF (isoform 2) is accessible in public databases by GenBank Accession numbers NP—036285.2 (protein) and NM—012153.5 (nucleic acid).
Sequence information related to EHF (isoform 3) is accessible in public databases by GenBank Accession numbers NP—1193544.1 (protein) and NM—1206615.1 (nucleic acid).
SEQ ID NO: 25 is the human wild type amino acid sequence corresponding to EHF (isoform 1) (residues 1-322):
SEQ ID NO: 26 is the human wild type nucleotide sequence corresponding to EHF (isoform 1) (nucleotides 1-5467), wherein the underscored bolded “ATG” denotes the beginning of the open reading frame:
A reprogramming factor molecule or a master regulatory molecule can also encompass ortholog genes, which are genes conserved among different biological species such as humans, dogs, cats, mice, and rats, that encode proteins (for example, homologs (including splice variants), mutants, and derivatives) having biologically equivalent functions as the human-derived protein. Orthologs of a reprogramming factor molecule or a master regulatory molecule include any mammalian ortholog inclusive of the ortholog in humans and other primates, experimental mammals (such as mice, rats, hamsters and guinea pigs), mammals of commercial significance (such as horses, cows, camels, pigs and sheep), and also companion mammals (such as domestic animals, e.g., rabbits, ferrets, dogs, and cats).
In one embodiment of the present invention, the gene encoding a protein of interest (for example for example, Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, Ehf, and the like), can be cloned from either a genomic library or a cDNA according to standard protocols familiar to one skilled in the art (J. Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, Plainview, N.Y.; F. M. Ausubel et al., 1989, Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y.). A cDNA, for example, encoding Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf, can be obtained by isolating total mRNA from a suitable cell line. Double stranded cDNAs can be prepared from the total mRNA using methods known in the art, and subsequently can be inserted into a suitable plasmid or vector. Genes can also be cloned using PCR techniques well established in the art. In one embodiment, a gene encoding Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf, can be cloned via PCR in accordance with the nucleotide sequence information provided by Genbank. In a further embodiment, a DNA vector containing Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf, can act as a template in PCR reactions wherein oligonucleotide primers designed to amplify a region of interest can be used, so as to obtain an isolated DNA fragment encompassing that region.
An expression vector of the current invention can include nucleotide sequences that encode either an Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf protein linked to at least one sequence in a manner allowing expression of the nucleotide sequence in a host cell. Regulatory sequences are well known to those skilled in the art, and can be selected to direct the expression of a protein of interest (such as Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf) in an appropriate host cell as described in Goeddel, Gene Expression Technology: Methods in Enzymology 185, Academic Press, San Diego, Calif. (1990). Non-limiting examples of regulatory sequences include: polyadenylation signals, promoters (such as CMV, ASV, SV40, or other viral promoters such as those derived from bovine papilloma, polyoma, and Adenovirus 2 viruses (Fiers, et al., 1973, Nature 273:113; Hager G L et al., Curr Opin Genet Dev, 2002, 12(2):137-41) enhancers, and other expression control elements.
One skilled in the art also understands that enhancer regions, which are those sequences found upstream or downstream of the promoter region in non-coding DNA regions, are also important in optimizing expression. If needed, origins of replication from viral sources can be employed, such as if a prokaryotic host is utilized for introduction of plasmid DNA. However, in eukaryotic organisms, chromosome integration is a common mechanism for DNA replication.
In one embodiment of the present invention, the gene encoding a protein of interest (such as Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf) is controlled by an inducible promoter. For example, transcription of the gene encoding a protein of interest is reversibly controlled by the presence of an antibiotic, such as doxycycline. Inducible expression systems are well known in the art, and include but are not limited to, the Tet-On system, or the Tet-Off system (U.S. Pat. No. 5,464,758; U.S. Pat. No. 5,814,618; Bujard H. & Gossen M., 1992, PNAS 89(12):5547-51)
It is understood by those skilled in the art that for stable amplification and expression of a desired protein, a vector harboring DNA encoding a protein of interest (for example, Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf) is stably integrated into the genome of eukaryotic cells (for example, mammalian cells, such as mouse embryonic fibroblasts, mouse dermal fibroblasts, or BJ normal human foreskin fibroblasts), resulting in the stable expression of transfected genes. The expression vector and method of introduction of the exogenous nucleic acid to the cell can be factors that contribute to a successful integration event. For example, an exogenous nucleic acid can be integrated into the genome of eukaryotic cells (such as a mammalian cell) for stable expression by using a retrovirus to introduce the exogenous nucleic acid into the cell. In another example, an exogenous nucleic acid sequence can be introduced into a cell by homologous recombination as disclosed in U.S. Pat. No. 5,641,670, the contents of which are herein incorporated by reference.
A gene that encodes a selectable marker (for example, resistance to antibiotics or drugs, such as ampicillin, G418, and hygromycin) can be introduced into host cells along with the gene of interest in order to identify and select clones that stably express a gene encoding a protein of interest. The gene encoding a selectable marker can be introduced into a host cell on the same plasmid as the gene of interest or can be introduces on a separate plasmid. Cells containing the gene of interest can be identified by drug selection wherein cells that have incorporated the selectable marker gene will survive in the presence of the drug. Cells that have not incorporated the gene for the selectable marker die. Surviving cells can then be screened for the production of the desired protein (for example, Oct4, Sox2, Klf4, c-Myc, NKX3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Elf3, or Ehf)
Introduction of Reprogramming Factors into Fibroblasts
A eukaryotic expression vector can be introduced into cells in order to produce proteins (for example, Oct4, Sox2, Klf4, or c-Myc) encoded by nucleotide sequences of the vector. Cells (such as embryonic fibroblasts, mouse dermal fibroblasts, or BJ normal human foreskin fibroblasts) can harbor an expression vector (for example, one that contains a gene encoding Oct4, Sox2, Klf4, or c-Myc) via introducing the expression vector into an appropriate host cell via methods known in the art.
An exogenous nucleic acid can be introduced into a cell via a variety of techniques known in the art. For example, a retrovirus can be used to introduce a nucleotide sequence into cells (such as embryonic fibroblasts, mouse dermal fibroblasts, or BJ normal human foreskin fibroblasts). In one embodiment, the retrovirus is a Rebna retrovirus. Other viral vectors known in the art can be used to introduce a nucleotide sequence, including, but not limited to a lentivirus, a adenovirus, or a adeno-associated virus.
In one embodiment, a retrovirus can be used to introduce a nucleotide sequence into embryonic fibroblasts, dermal fibroblasts, or human foreskin fibroblasts, in order to produce proteins encoded by said nucleotide sequences (for example, Oct4, Sox2, Klf4, and c-Myc). For example, the Rebna retrovirus is used to introduce DNA into an embryonic fibroblast, or a dermal fibroblast, to confer high-level stable expression of reprogramming factors (for example, Oct4, Sox2, Klf4, and c-Myc). In other embodiments, lentivirus is used to introduce DNA into embryonic fibroblasts, dermal fibroblasts, or human foreskin fibroblasts, to confer high-level stable expression of reprogramming factors (for example, Oct4, Sox2, Klf4, and c-Myc). In further embodiments, lentivirus is used to introduce DNA into embryonic fibroblasts, dermal fibroblasts, or human foreskin fibroblasts to confer transient doxycycline-inducible expression of reprogramming factors (for example, Oct4, Sox2, Klf4, and c-Myc). The nucleic acid of interest can encode only a single protein (for example, Oct4, Sox2, Klf4, or c-Myc), or can encode for more than one proteins of interest (for example, combinations of Oct4, Sox2, Klf4, c-Myc). In one embodiment, doxycycline-inducible expression of reprogramming factors (for example, Oct4, Sox2, Klf4, and/or c-Myc) is used. Reprogramming factors include, but are not limited to, Oct4, Sox2, Klf4, c-Myc, nanog, Lin28, Esrrb, or Nr5a2.
A eukaryotic expression vector can be used to transfect cells in order to produce proteins (for example, Oct4, Sox2, Klf4, or c-Myc) encoded by nucleotide sequences of the vector. Mammalian cells (such as mouse embryonic fibroblasts, mouse dermal fibroblasts, or BJ normal human foreskin fibroblasts) can harbor an expression vector (for example, one that encodes a gene encoding Oct4, Sox2, Klf4, or c-Myc) via introducing the expression vector into an appropriate host cell via methods known in the art.
An exogenous nucleic acid can be introduced into a cell via a variety of techniques known in the art, such as lipofection, microinjection, calcium phosphate or calcium chloride precipitation, DEAE-dextrin-mediated transfection, or electroporation. Other methods used to transfect cells can also include calcium phosphate precipitation, modified calcium phosphate precipitation, polybrene precipitation, microinjection liposome fusion, and receptor-mediated gene delivery.
Cells to be genetically engineered can be primary and secondary cells, which can be obtained from various tissues and include cell types which can be maintained and propagated in culture. Vertebrate tissue can be obtained by methods known to one skilled in the art, such as dissection of an E13.5 mouse embryo. In one embodiment, tissue can be obtained from an E12.5, E13, E13.5, E14, or E14.5 mouse embryo. In another embodiment, dissection of a E13.5 mouse embryo can be used to obtain a source of embryonic fibroblast cells. In further embodiments, tissue can be obtained from a P0, P1, P2, or P3 mouse. For example, dissection of a P0 mouse can be used to obtain a source of mouse dermal fibroblasts. In another embodiment, human foreskins can be used to obtain a source of BJ normal human foreskin fibroblasts.
In certain embodiments, embryonic fibroblast cells or mouse dermal fibroblasts can be acquired from a mouse which has been genetically engineered. For example, embryonic fibroblasts or mouse dermal fibroblasts may be derived from mice with an Oct4-GFP knock-in genotype. In another embodiment, embryonic fibroblasts or mouse dermal fibroblasts may be derived from mice with a Nkx3.1-lacZ knock-in genotype. In further embodiments, embryonic fibroblasts or mouse dermal fibroblasts may be derived from mice with a doxycycline-regulated transgene encoding a protein, or proteins of interest (for example, Oct4, Sox2, Klf4, c-Myc, or a combination thereof). Embryonic fibroblasts or mouse dermal fibroblasts may also be derived from mice with other genetically engineered genomes including, but not limited to, Nanog-CreERT2;R26R-Tomato mice, CK5-CreERT2; R26R-YFP mice, CK8-CreERT2; R26R-YFP mice, or CK18-CreERT2; R26R-YFP mice. In other embodiments, embryonic fibroblast cells or mouse dermal fibroblast cells can be acquired from a mouse which has a wild-type genome. In some embodiments, embryonic fibroblasts or mouse dermal fibroblasts may be derived from mice with a GATA6CreERT2; R26R-CAG-YFP genotype. In some embodiments, embryonic fibroblasts or mouse dermal fibroblasts may be derived from mice with a CK18CreERT2; R26R-Tomato genotype.
Cell Culturing of Eukaryotic Cells
Various culturing parameters can be used with respect to the host cell being cultured. Appropriate culture conditions for mammalian cells are well known in the art or can be determined by the skilled artisan (see, for example, Animal Cell Culture: A Practical Approach 2nd Ed., Rickwood, D. and Hames, B. D., eds. (Oxford University Press: New York, 1992)), and vary according to the particular cell selected. Commercially available medium can be utilized. Non-limiting examples of medium include, for example, Dulbecco's Modified Eagle Medium (DMEM, Life Technologies), Minimal Essential Medium (MEM, Sigma, St. Louis, Mo.); HyClone cell culture medium (HyClone, Logan, Utah); and serum-free basal epithelial medium (CellnTech).
The media described above can be supplemented as necessary with supplementary components or ingredients, including optional components, in appropriate concentrations or amounts, as necessary or desired. Cell medium solutions provide at least one component from one or more of the following categories: (1) an energy source, usually in the form of a carbohydrate such as glucose; (2) all essential amino acids, and usually the basic set of twenty amino acids plus cysteine; (3) vitamins and/or other organic compounds required at low concentrations; (4) free fatty acids or lipids, for example linoleic acid; and (5) trace elements, where trace elements are defined as inorganic compounds or naturally occurring elements that are typically required at very low concentrations, usually in the micromolar range.
The medium also can be supplemented electively with one or more components from any of the following categories: (1) salts, for example, magnesium, calcium, and phosphate; (2) hormones and other growth factors such as, serum, insulin, transferrin, epidermal growth factor and fibroblast growth factor; (3) protein and tissue hydrolysates, for example peptone or peptone mixtures which can be obtained from purified gelatin, plant material, or animal byproducts; (4) nucleosides and bases such as, adenosine, thymidine, and hypoxanthine; (5) buffers, such as HEPES; (6) antibiotics, such as gentamycin or ampicillin; (7) cell protective agents, for example, pluronic polyol; and (8) galactose.
The mammalian cell culture that can be used with the present invention is prepared in a medium suitable for the particular cell being cultured. In one embodiment, the culture medium can be one of the aforementioned (for example, DMEM) that is supplemented with serum from a mammalian source (for example, fetal bovine serum (FBS)). For example, DMEM supplemented with FBS can be used to sustain the growth of embryonic fibroblasts, dermal fibroblasts or human foreskin fibroblasts. In another embodiment, the medium can be serum-free basal epithelial medium. For example, serum-free basal epithelial medium can used to sustain the growth of epithelial cells obtained from the reprogramming of fibroblast cells. In further embodiments, serum-free basal epithelial medium contains epidermal growth factor (EGF), fibroblast growth factor (FGF), or a combination thereof.
In one embodiment, fibroblasts cultured in an acceptable medium (such as DMEM supplemented with FBS), can be transduced with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof). In one embodiment, following transduction with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof), fibroblasts are incubated for at least 24 hours at about 37° C. In another embodiment, cells are incubated for at least 48, 72, or 96 hours, following transduction. Cells are incubated at about 35° C., about 36° C., about 37° C., about 38° C., or about 39° C.
In one embodiment, following transduction of fibroblasts with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof), the medium used to sustain the growth of fibroblasts is switched to serum-free basal epithelial medium. In a further embodiments, the serum-free basal epithelial medium contains EGF, FGF or a combination thereof. In another embodiment, following transduction with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof), fibroblasts are reprogrammed to epithelial cells. For example, the epithelial cells are induced epithelial cells.
Cells maintained in culture can be passaged by their transfer from a previous culture to a culture with fresh medium. In one embodiment, induced epithelial cells are stably maintained in cell culture for at least 3 passages, at least 4 passages, at least 5 passages, at least 6 passages, at least 7 passages, at least 8 passages, at least 9 passages, at least 10 passages, at least 11 passages, at least 12 passages, at least 13 passages, at least 14 passages, at least 15 passages, at least 20 passages, at least 25 passages, or at least 30 passages.
The cells suitable for culturing according to the methods of the present invention can harbor introduced expression vectors (constructs), such as plasmids and the like. The expression vector constructs can be introduced via transformation, microinjection, transfection, lipofection, electroporation, or infection. The expression vectors can contain coding sequences, or portions thereof, encoding the proteins for expression and production. Expression vectors containing sequences encoding the produced proteins and polypeptides, as well as the appropriate transcriptional and translational control elements, can be generated using methods well known to and practiced by those skilled in the art. These methods include synthetic techniques, in vitro recombinant DNA techniques, and in vivo genetic recombination which are described in J. Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, Plainview, N.Y. and in F. M. Ausubel et al., 1989, Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y.
In one embodiment, induced epithelial cells can express a variety of markers that distinguish them from fibroblasts. These markers include, but are not limited to cytokeratin 5 (CK5), CK8, CK14, CK18, beta-catenin, E-cadherin, Epithelial Membrane Antigen (EMA/Muc1), or EpCAM or a combination thereof. Expression of markers can be evaluated by a variety of methods known in the art. The presence of markers can be determined at the DNA, RNA or polypeptide level.
In one embodiment, the method can comprise detecting the presence of a marker gene (such as, CK5, CK8, CK14, CK18, beta-catenin or E-cadherin) polypeptide expression. Polypeptide expression includes the presence of a marker gene polypeptide sequence, or the presence of an elevated quantity of marker gene polypeptide as compared to non-epithelial cells. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, polypeptide expression maybe evaluated by methods including, but not limited to, immunostaining, FACS analysis, or Western blot. These methods are well known in the art (for example, U.S. Pat. No. 8,004,661, U.S. Pat. No. 5,367,474, U.S. Pat. No. 4,347,935) and are described in T. S. Hawley & R. G. Hawley, 2005, Methods in Molecular Biology Volume 263: Flow Cytometry Protocols, Humana Press Inc; I. B. Buchwalow & W. BoEcker, 2010, Immunohistochemistry: Basics & Methods, Springer, Medford, Mass.; O. J. Bjerrum & N. H. H. Heegaard, 2009, Western Blotting: Immunoblotting, John Wiley & Sons, Chichester, UK.
In another embodiment, the method can comprise detecting the presence of marker gene (CK5, CK8, CK14, CK18, beta-catenin or E-cadherin) RNA expression, for example in reconstituted induced epithelial cells. RNA expression includes the presence of an RNA sequence, the presence of an RNA splicing or processing, or the presence of a quantity of RNA. These can be detected by various techniques known in the art, including by sequencing all or part of the marker gene RNA, or by selective hybridization or selective amplification of all or part of the RNA.
In one embodiment, following transduction of fibroblasts with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof), the medium used to sustain the growth of fibroblasts is switched to stem cell media. In a further embodiments, stem cell media is mouse embryonic stem cell media. In further embodiments, the stem cell media contains LIF, In another embodiment, following transduction with DNA vectors harboring genes that encode a protein of interest (such as Oct4, Sox2, Klf4 or c-Myc, or a combination thereof), fibroblasts are reprogrammed to induced pluripotent stem cells (iPSCs).
Cells maintained in culture can be passaged by their transfer from a previous culture to a culture with fresh medium. In one embodiment, iPSCs are stably maintained in cell culture for at least 3 passages, at least 4 passages, at least 5 passages, at least 6 passages, at least 7 passages, at least 8 passages, at least 9 passages, at least 10 passages, at least 11 passages, at least 12 passages, at least 13 passages, at least 14 passages, at least 15 passages, at least 20 passages, at least 25 passages, or at least 30 passages.
Methods for Reconstituting Induced Epithelial Cells into an Organ Tissue
A eukaryotic expression vector can be introduced into cells in order to produce proteins (for example, Nkx3.1, Androgen receptor (AR), FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, Ehf) encoded by nucleotide sequences of the vector. Cells (such as induced epithelial cells) can harbor an expression vector (for example, one that contains a gene encoding Nkx3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, or Ehf) via introducing the expression vector into an appropriate host cell via methods known in the art.
An exogenous nucleic acid can be introduced into a cell via a variety of techniques known in the art. For example, a retrovirus can be used to introduce a nucleotide sequence into cells (such as induced epithelial cells). In one embodiment, the retrovirus is a Rebna retrovirus. In another embodiment, the retrovirus is a lentivirus. In yet another embodiment, the retrovirus is a LZRS retrovirus. Other viral vectors known in the art can be used to introduce a nucleotide sequence, including, but not limited to a lentivirus, a adenovirus, or a adeno-associated virus.
In one embodiment, a retrovirus can be used to introduce a nucleotide sequence into induced epithelial cells to produce proteins encoded by said nucleotide sequences (for example, Nkx3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, or Ehf). For example, the LZRS retrovirus, or a lentivirus, is used to introduce DNA into an induced epithelial cells to confer high-level stable expression of master regulatory genes (for example, Nkx3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, or Ehf). The nucleic acid of interest can encode only a single protein (for example, Nkx3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, or Ehf), or can encode for more than one protein of interest (for example, combinations of Nkx3.1, AR, FOXA1, FOXA2, KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, or Ehf).
In one embodiment, induced epithelial cells can be transduced with DNA vectors harboring genes that encode a master regulatory gene. For example, a master regulatory gene can be a master regulatory gene for prostate development, such as Nkx3.1, AR, FOXA1, FOXA2, or a combination thereof. In another embodiment, a master regulatory gene can be a master regulatory gene for bladder development, such as KLF5, Pparγ, Grhl3, Ovo1, Foxa1, Elf3, Ehf, or a combination thereof. Master regulatory genes include, but are not limited to, XBP1, FOXA1, ACAD8, NKX3.1, MAP2K1, CREB3L4, HIPK2, YWHAQ, RIPK2, CREB3, FOXM1, TRIP13, CENPF, MEF2C, and ZNF423.
An exogenous nucleic acid can be introduced into a cell via a variety of techniques known in the art, such as lipofection, microinjection, calcium phosphate or calcium chloride precipitation, DEAE-dextrin-mediated transfection, or electroporation. Other methods used to transfect cells can also include calcium phosphate precipitation, modified calcium phosphate precipitation, polybrene precipitation, microinjection liposome fusion, and receptor-mediated gene delivery.
Cells to be genetically engineered can be primary and secondary cells, which can be obtained from various tissues and include cell types which can be maintained and propagated in culture. In one embodiment, cells are induced epithelial cells which can be obtained by the methods described by this invention.
In one embodiment, following transduction of induced epithelial cells with DNA vectors harboring genes that encode a master regulatory gene, cells are recombined with mesenchymal cells and a graft is performed in a subject. Tissue recombination assays are well known to one in the art (A14-A21). In one example, the mesenchymal cells comprise urogenital mesenchyme. In another example, the mesenchymal cells comprise embryonic bladder mesenchyme. Various routes of administration and various sites of graft can be utilized, such as, a renal graft, in order to introduced the transduced recombined cells into a site of preference. Once implanted into a subject (such as, a mouse, rat, or human), the transduced recombined cells can reconstitute into an organ tissue (such as, prostate epithelial tissue, or bladder epithelial tissue). In one example the graft is a renal graft. Administration of the recombined cells is not restricted to a single route, but may encompass administration by multiple routes. Exemplary administrations include a renal graft. Other modes of administration by multiple routes will be apparent to the skilled artisan.
In some embodiments, the cells used for administration will generally be subject-specific genetically engineered cells. In another embodiment, cells obtained from a different species or another individual of the same species can be used. Thus, using such cells may require administering an immunosuppressant to prevent rejection of the administered cells. Such methods have also been described in United States Patent Application Publication 2004/0057937 and PCT application publication WO 2001/32840, and are hereby incorporated by reference.
In one embodiment, cells may be introduced into an immunodeficient subject. For example, the cells may be introduced into an immunodeficient mouse such as an athymic nude mouse, a BALB/c nude mouse, a CD-1 nude mouse, a Fox Chase SCID beige mouse, a Fox Chase SCID mouse, a NIH-III nude mouse, a NOD SCID mouse, a NU/NU nude mouse, a SCID hairless congenic mouse, or a SCID hairless outbred mouse.
In one embodiment, induced epithelial cells are reconstituted into an organ tissue. For example, induced epithelial cells can be reconstituted into prostate epithelial tissue. In another example, induced epithelial cells can be reconstituted into bladder epithelial tissue. In one embodiment, reconstituted organ tissue can express a variety of markers that distinguish them as, for example, prostate epithelial tissue, or bladder epithelial tissue. These markers include, but are not limited to p63, CK5, AR, CK8, NKX3.1, PSA, Probasin, uroplakins or a combination thereof.
Expression of markers can be evaluated by a variety of methods known in the art. The presence of markers can be determined at the DNA, RNA or polypeptide level. In one embodiment, the method can comprise detecting the presence of a marker gene polypeptide expression. Polypeptide expression includes the presence of a marker gene polypeptide sequence, or the presence of an elevated quantity of marker gene polypeptide as compared to non-epithelial cells. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, polypeptide expression maybe evaluated by methods including, but not limited to, immunostaining, FACS analysis, or Western blot. These methods are well known in the art (for example, U.S. Pat. No. 8,004,661, U.S. Pat. No. 5,367,474, U.S. Pat. No. 4,347,935) and are described in T. S. Hawley & R. G. Hawley, 2005, Methods in Molecular Biology Volume 263: Flow Cytometry Protocols, Humana Press Inc; I. B. Buchwalow & W. BoEcker, 2010, Immunohistochemistry: Basics & Methods, Springer, Medford, Mass.; O. J. Bjerrum & N. H. H. Heegaard, 2009, Western Blotting: Immunoblotting, John Wiley & Sons, Chichester, UK.
In another embodiment, the method can comprise detecting the presence of marker gene (such as, p63, CK5, AR, CK8, Probasin, or a combination thereof) RNA expression, for example in reconstituted organ tissue. RNA expression includes the presence of an RNA sequence, the presence of an RNA splicing or processing, or the presence of a quantity of RNA. These can be detected by various techniques known in the art, including by sequencing all or part of the marker gene RNA, or by selective hybridization or selective amplification of all or part of the RNA.
In another embodiment, reconstituted organ tissue can express markers that reveal reconstituted organ tissue architecture and are localized to specific areas. For example, the method can comprise detecting the presence of a marker gene (for example, p63, CK5, or a combination thereof) in the basal layer of prostate epithelial tissue, or bladder epithelial tissue. In another example, the method can comprise detecting the presence of a marker gene (for example, AR, CK8, or a combination thereof) in the luminal layer of prostate epithelial tissue. In a further example, the method can comprise detecting the presence of a marker gene (for example, CK8) in the luminal layer of bladder epithelial tissue. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, marker gene expression can be evaluated by immunostaining. Other markers that known in the art that reveal reconstituted organ tissue architecture can also be used.
In one embodiment, reconstituted organ tissue can express markers that reveal reconstituted organ tissue functionality. For example, the method can comprise detecting the presence of a marker gene (for example, Probasin) in prostate epithelial tissue. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, marker gene expression can be evaluated by immunostaining.
In one embodiment, reconstituted organ tissue can display characteristic tissue architecture. For example, reconstituted bladder epithelium can stain positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome. The method can comprise detecting other characteristic tissue architecture in reconstituted organ tissue using various techniques known in the art, including staining of tissue with various stains including, but not limited to, Gomori's trichrome, haematoxylin and eosin, periodic acid-Schiff, Masson's trichrome, Silver staining, or Sudan staining.
Methods for Reconstituting Induced Pluripotent Stem Cells (iPSCs) into an Organ Tissue
In one embodiment, following the reprogramming of fibroblasts into iPSCs, iPSCs are recombined with mesenchymal cells and a graft is performed in a subject. Tissue recombination assays are well known to one in the art (A14-A21). In one example, the mesenchymal cells comprise urogenital mesenchyme. In another example, the mesenchymal cells comprise embryonic bladder mesenchyme. Various routes of administration and various sites of graft can be utilized, such as, a renal graft, in order to introduced the transduced recombined cells into a site of preference. Once implanted into a subject (such as, a mouse, rat, or human), the iPSCs can reconstitute into an organ tissue (such as, prostate epithelial tissue, or bladder epithelial tissue). In one example the graft is a renal graft. Administration of the recombined cells is not restricted to a single route, but may encompass administration by multiple routes. Exemplary administrations include a renal graft. Other modes of administration by multiple routes will be apparent to the skilled artisan.
In another embodiment, following the reprogramming of fibroblasts into iPSCs, the medium used to sustain the growth of iPSCs is switched to endodermal differentiation media. In one embodiment, the endodermal differentiation media contains Activin A, Noggin, and a GSK3β inhibitor. In one embodiment, iPSCs expressing endodermal markers are isolated. For example, endodermal markers include, but are not limited to GATA6. In one embodiment, the iPSCs express GATA6. The methods for separating, enriching, isolating or purifying iPSCs expressing endodermal markers according to the invention may be combined with other methods for separating, enriching, isolating or purifying cells that are known in the art. The presence of markers can be determined at the DNA, RNA or polypeptide level. In one embodiment, following the isolation of iPSCs expressing endodermal markers (e.g. GATA6), the iPSCs are recombined with mesenchymal cells and a graft is performed in a subject. In one embodiment, the iPSCs are cultured in a three-dimensional culture. In one embodiment, the iPSCs are cultured in Matrigel.
In some embodiments, the cells used for administration will generally be subject-specific genetically engineered cells. In another embodiment, cells obtained from a different species or another individual of the same species can be used. Thus, using such cells may require administering an immunosuppressant to prevent rejection of the administered cells. Such methods have also been described in United States Patent Application Publication 2004/0057937 and PCT application publication WO 2001/32840, and are hereby incorporated by reference.
In one embodiment, cells may be introduced into an immunodeficient subject. For example, the cells may be introduced into an immunodeficient mouse such as an athymic nude mouse, a BALB/c nude mouse, a CD-1 nude mouse, a Fox Chase SCID beige mouse, a Fox Chase SCID mouse, a NIH-III nude mouse, a NOD SCID mouse, a NU/NU nude mouse, a SCID hairless congenic mouse, or a SCID hairless outbred mouse.
In one embodiment, iPSCs are reconstituted into an organ tissue. For example, iPSCs can be reconstituted into prostate epithelial tissue. In another example, iPSCs can be reconstituted into bladder epithelial tissue. In one embodiment, reconstituted organ tissue can express a variety of markers that distinguish them as, for example, prostate epithelial tissue, or bladder epithelial tissue. These markers include, but are not limited to p63, CK5, AR, CK8, NKX3.1, PSA, Probasin, uroplakins or a combination thereof
In one embodiment, iPSCs expressing an endodermal marker are reconstituted into an organ tissue. For example, iPSCs expressing an endodermal marker can be reconstituted into prostate epithelial tissue. In another example, iPSCs expressing an endodermal marker can be reconstituted into bladder epithelial tissue. In one embodiment, reconstituted organ tissue can express a variety of markers that distinguish them as, for example, prostate epithelial tissue, or bladder epithelial tissue. These markers include, but are not limited to p63, CK5, AR, CK8, NKX3.1, PSA, Probasin, uroplakins or a combination thereof.
Expression of markers can be evaluated by a variety of methods known in the art. The presence of markers can be determined at the DNA, RNA or polypeptide level. In one embodiment, the method can comprise detecting the presence of a marker gene polypeptide expression. Polypeptide expression includes the presence of a marker gene polypeptide sequence, or the presence of an elevated quantity of marker gene polypeptide as compared to non-epithelial cells. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, polypeptide expression maybe evaluated by methods including, but not limited to, immunostaining, FACS analysis, or Western blot. These methods are well known in the art (for example, U.S. Pat. No. 8,004,661, U.S. Pat. No. 5,367,474, U.S. Pat. No. 4,347,935) and are described in T. S. Hawley & R. G. Hawley, 2005, Methods in Molecular Biology Volume 263: Flow Cytometry Protocols, Humana Press Inc; I. B. Buchwalow & W. BoEcker, 2010, Immunohistochemistry: Basics & Methods, Springer, Medford, Mass.; O. J. Bjerrum & N. H. H. Heegaard, 2009, Western Blotting: Immunoblotting, John Wiley & Sons, Chichester, UK.
In another embodiment, the method can comprise detecting the presence of marker gene (such as, p63, CK5, AR, CK8, Probasin, or a combination thereof) RNA expression, for example in reconstituted organ tissue. RNA expression includes the presence of an RNA sequence, the presence of an RNA splicing or processing, or the presence of a quantity of RNA. These can be detected by various techniques known in the art, including by sequencing all or part of the marker gene RNA, or by selective hybridization or selective amplification of all or part of the RNA.
In another embodiment, reconstituted organ tissue can express markers that reveal reconstituted organ tissue architecture and are localized to specific areas. For example, the method can comprise detecting the presence of a marker gene (for example, p63, CK5, or a combination thereof) in the basal layer of prostate epithelial tissue, or bladder epithelial tissue. In another example, the method can comprise detecting the presence of a marker gene (for example, AR, CK8, or a combination thereof) in the luminal layer of prostate epithelial tissue. In a further example, the method can comprise detecting the presence of a marker gene (for example, CK8) in the luminal layer of bladder epithelial tissue. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, marker gene expression can be evaluated by immunostaining. Other markers that known in the art that reveal reconstituted organ tissue architecture can also be used.
In one embodiment, reconstituted organ tissue can express markers that reveal reconstituted organ tissue functionality. For example, the method can comprise detecting the presence of a marker gene (for example, Probasin) in prostate epithelial tissue. These can be detected by various techniques known in the art, including by sequencing and/or binding to specific ligands (such as antibodies). For example, marker gene expression can be evaluated by immunostaining.
In one embodiment, reconstituted organ tissue can display characteristic tissue architecture. For example, reconstituted bladder epithelium can stain positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome. The method can comprise detecting other characteristic tissue architecture in reconstituted organ tissue using various techniques known in the art, including staining of tissue with various stains including, but not limited to, Gomori's trichrome, haematoxylin and eosin, periodic acid-Schiff, Masson's trichrome, Silver staining, or Sudan staining.
An aspect of the invention is directed to a method for transdifferentiation of embryonic fibroblast cells into an organ tissue, the method comprising: (a) isolating embryonic fibroblasts (EFs); (b) transducing EFs with a retrovirus comprising a reprogramming factor; (c) culturing the infected EFs in stem cell media for at least 24 hours at about 37° C. to generate induced pluripotent stem cells (iPSCs); (d) isolating iPSCs; (e) recombining the cells of (d) with mesenchymal cells; and (f) performing a graft of the recombined cells of (e) into an immunodeficient subject. In one embodiment, the stem cell media comprises LIF. In one embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In one embodiment, the mesenchymal cells comprise urogenital mesenchyme. In one embodiment, the mesenchymal cells comprise bladder mesenchyme. In one embodiment, the graft is a renal graft. In one embodiment, the organ tissue is prostate epithelial tissue. In one embodiment, the organ tissue is bladder epithelial tissue. In one embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In one embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In one embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In one embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome. In one embodiment, the retrovirus is a lentivirus. In one embodiment, the lentivirus is doxycycline regulated.
An aspect of the invention is directed to a method for differentiation of induced pluripotent stem cells (iPSCs) into an organ tissue, the method comprising: (a) isolating iPSCs; (b) recombining the cells of (a) with mesenchymal cells; and (c) performing a graft of the recombined cells of (b) into an immunodeficient subject. In one embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In one embodiment, the mesenchymal cells comprise urogenital mesenchyme. In one embodiment, the mesenchymal cells comprise bladder mesenchyme. In one embodiment, the graft is a renal graft. In one embodiment, the organ tissue is prostate epithelial tissue. In one embodiment, the organ tissue is bladder epithelial tissue. In one embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In one embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In one embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In one embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In one embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome.
An aspect of the invention is directed to a method for differentiation of induced pluripotent stem cells (iPSCs) into an organ tissue, the method comprising: (a) isolating iPSCs; (b) culturing iPSCs in endodermal differentiation media; (c) isolating iPSCs that express an endodermal marker; (d) recombining the cells of (c) with mesenchymal cells; and (e) performing a graft of the recombined cells of (d) into an immunodeficient subject. In one embodiment, the endodermal differentiation media contains Activin A, Noggin, and a GSK3β inhibitor. In another embodiment, the endodermal marker is GATA6. In one embodiment, the iPSCs are cultured in a three-dimensional culture. In one embodiment, the iPSCs are cultured in Matrigel. In another embodiment, the graft is maintained in the subject for about 6 to 8 weeks. In another embodiment, the mesenchymal cells comprise urogenital mesenchyme. In another embodiment, the mesenchymal cells comprise bladder mesenchyme. In another embodiment, the graft is a renal graft. In another embodiment, the organ tissue is prostate epithelial tissue. In another embodiment, the organ tissue is bladder epithelial tissue. In another embodiment, the prostate tissue expresses p63, CK5, or a combination thereof, in the basal layer. In another embodiment, the bladder tissue expresses p63, CK5, or a combination thereof, in the basal layer. In another embodiment, the prostate tissue expresses AR, CK8, or a combination thereof, in the luminal layer. In another embodiment, the prostate tissue expresses Probasin, PSA, or a combination thereof. In another embodiment, the bladder tissue expresses CK8, uroplakins, or a combination thereof. In another embodiment, the bladder tissue stains positive for the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium with Gomori's trichrome.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Exemplary methods and materials are described below, although methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present invention.
All publications and other references mentioned herein are incorporated by reference in their entirety, as if each individual publication or reference were specifically and individually indicated to be incorporated by reference. Publications and references cited herein are not admitted to be prior art.
Examples are provided below to facilitate a more complete understanding of the invention. The following examples illustrate the exemplary modes of making and practicing the invention. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only, since alternative methods can be utilized to obtain similar results.
Interactomes have been generated for mouse and human prostate tissue, using an established algorithm for reverse engineering, such as ARACNe [15-17]. The mouse prostate interactome was constructed using a large collection of gene expression profiles from drug-induced perturbation of several transgenic models, with phenotypes ranging from normal tissue to advanced prostate cancer. The human prostate cancer interactome was constructed from a large published dataset comprised of prostate cancer specimens and adjacent normal tissue [37]. These interactomes, which are being validated using cell culture assays, have been interrogated to identify master regulator genes for prostate cancer initiation, using the MARINa algorithm [18, 19] (
Expression of reprogramming factors have been used in fibroblasts to generate cells with epithelial morphologies in culture. Mouse embryonic fibroblasts (MEFs) of distinct genotypes (wild-type, Oct4-GFP knock-in, and Nkx3.1-lacZ knock-in) have been derived from E13.5 mouse embryos after the head and pelvis were removed to exclude neural and prostate progenitors. These MEFs were used after sorting for the mesenchymal marker CD140 or sorting against Lin/Mac-1(CD11b)/EpCAM markers to exclude blood, endothelial, and epithelial contaminants, thereby reducing the heterogeneity of the primary fibroblast population (
The “primitive” epithelial cells were further stably transduced with Nkx3.1 and AR-known master regulators of prostate development followed by tissue recombination assays with rat UGM in renal grafts (
Without being bound by theory, these studies can identify master regulator genes for the normal prostate epithelium by regulatory network analysis using existing or newly generated interactomes for mouse and human prostate and bladder tissue. Together with master regulators identified by the candidate gene approach, these genes can be used in gain- or loss-of-function experiments to promote prostate differentiation by mouse iPSC using an in vivo tissue recombination/renal grafting system.
Experimental Design:
To identify master regulators of prostate and bladder epithelium, expression signatures can first be generated for adult and embryonic mouse prostate epithelium and bladder urothelium as well as mammary epithelium as control comparisons. These signatures can be produced by gene expression profiling of six biological replicate samples using standard protocols and hybridization to Illumina BeadArrays. Alternatively, transcriptomes can be generated in a more comprehensive way through RNA-seq. These expression signatures can be used to interrogate the mouse prostate and bladder interactomes using the MARINa and MINDy algorithms to identify master regulator (MR) genes and their modulators, as previously reported [18, 19]. The algorithms infer direct and indirect interactions among specific gene products, mRNA and DNA sequences from statistically significant co-regulation data. The power of this approach lies in its basis on genome-wide gene expression profiles data gathered from biological samples and consideration for all genes equally. Thus it is unbiased, unlike other approaches relying on a priori knowledge and probabilistic assumptions about how genes interact. Without being bound by theory, additional putative master regulators can be inferred by a candidate gene approach (e.g., Nkx3.1, FoxA1, androgen receptor, KLF5, Pparγ and Grhl3), based upon biological and biochemical identification of key transcription factors for prostate and bladder development (e.g., [40]).
In the next step, validation of the identified candidate MRs can be performed. The ability of each candidate to affect the propensity for epithelial differentiation of induced pluripotent stem cell (iPSCs) can be tested. To determine whether these master regulators can enhance the differentiation of mouse iPSC, lentiviral infection can be used to overexpress positive master regulators or knock-down negative regulators, as appropriate. Synergistic master regulators can be identified using the approach described in [18, 19], and experimentally tested. To assess the ability of these iPSCs to differentiate into mature prostate epithelium in vivo, a tissue recombination system can be employed in which these cells can be combined with dissociated rat embryonic urogenital mesenchyme, followed by renal grafting into immunodeficient nude mice. This basic strategy was successfully used previously to explore prostate differentiation and stem cell function ([4, 41-43]). As positive controls, mouse ESC can be used as well as human ESC, since human ESC have been shown to generate prostate epithelial cells under similar conditions [5]. For induction of bladder urothelium, embryonic bladder mesenchyme can be used in a similar experimental setting. Immunostaining for specific tissue markers can be performed to confirm the prostatic (mouse Nkx3.1, mouse AR, prostate secretions) or urothelial (uroplakins) phenotype. Epithelial tissue architecture can be confirmed with immunostaining for basal (p63, CK5) and luminal (CK8) markers. Gomori's trichrome staining can be used to demonstrate the presence of the sub-epithelial connective tissue layer (lamina propria) surrounding the urothelium. SMA immunolocalization can be performed to visualize the outer smooth muscle layer. Prostate epithelium and bladder urothelium can be used as controls for both tissue recombination experiments and immunostainings. In addition, the transcriptional profile of the induced tissues can be compared with normal mouse tissues through DNA microarray analysis.
Without being bound by theory, the interactome analysis can highlight known regulators of tissue development, such as AR or KLF5 pathways, as well as new, context-specific gene regulatory networks. For example, new master regulatory genes involved in early stages of tissue commitment and differentiation can be uncovered and validated. Prostate and bladder epithelia can be generated in vivo in renal grafts. Uncontrolled cell proliferation determined by the positive master regulators in different cell compartments resulting in an unbalanced basal:luminal cell ratio and improper epithelial-mesenchymal interactions can result. For instance, overexpression of KLF5 in stratified epithelium determines proliferation of the basal compartment [3]. If this event would occur in the urothelium, a lentiviral tet-on/tet-off system can be used to transduce the tissue master regulators and downregulate them in vivo in renal grafts.
These studies can employ expression of pluripotency factors to promote the reprogramming of mouse embryonic fibroblasts (MEFs) to normal prostate epithelial cells without undergoing an intermediate pluripotent state followed by expression of tissue specific master regulators. One approach relies on retroviral expression of Oct4, Sox2, Klf4, and c-Myc in MEFs, while a second approach uses transient doxycycline-inducible expression of pluripotency factors in MEFs. In both cases, reprogrammed cells with epithelial characteristics can be isolated by flow cytometry and used for tissue recombination and renal grafting to assess prostate and bladder differentiation. In addition, these studies can seek to optimize reprogramming conditions in the absence of c-Myc to reduce oncogenic transformation of the resulting epithelial cells.
Experimental Design:
In initial studies, a system can be used in which the expression of reprogramming factors is regulated by administration of doxycycline, which allows temporal control over their expression and avoid issues associated with their continuous expression. In one approach, mouse embryonic fibroblasts (MEFs) can be derived, as well as dermal fibroblasts and keratinocytes, from mice carrying a doxycycline-regulated single-copy transgene expressing Oct4, Sox2, Klf4, and c-Myc as a polycistronic transcript [44]. In a second approach, doxycycline-regulated lentiviruses can be used for each of the reprogramming factors, which can allow their use of desired combinations of interest (for example, Oct4, Sox2, and Klf4, without c-Myc). Without being bound by theory, additional 1-factor and 2-factor combinations can allow systematic investigation of the mechanisms by which the epithelial switch is activated.
Following these initial studies, the functional properties of the reprogrammed epithelial cells can be examined. In particular, it can be determined whether they display characteristic features of epithelial growth using in vitro assays, such as growth in three-dimensional culture in Matrigel, in the presence or absence of stromal cells. Their growth can also be examined in anchorage-independent conditions promoting the growth of spheres or organoids, as have been previously described for prostate epithelial cells [45, 46]. Finally, gene expression profiling of these reprogrammed epithelial cells can be performed to determine their similarity to immature epithelial cell types (e.g. primitive urogenital epithelium). The gene signatures of the reprogrammed epithelial cells can also be compared under a variety of culture conditions and ascertain their similarity to signatures of mature epithelium from mouse prostate, bladder, and breast, using Principal Components Analysis (PCA) and Gene Set Enrichment Analysis (GSEA) [36, 47], which have previously been used in other studies [48].
To determine whether the master regulators can enhance the differentiation of reprogrammed epithelial cells in culture, lentiviral infection can be used to overexpress positive master regulators or knock-down negative regulators. The resulting reprogrammed cells can be assayed for their morphological features and marker expression, and cells with promising phenotypes can be analyzed by expression profiling for comparison to the gene signatures of normal prostate and bladder epithelium. To assess prostate and bladder differentiation, flow cytometry can be used to isolate EpCAM+/CD24+ reprogrammed epithelial cells that have been maintained in prostate basal medium, followed by lentiviral infection with master regulators, tissue recombination, and renal grafting. Renal grafts can be harvested at various time points post-implantation and the epithelial cells can be dissociated and FACS sorted. Expression profiles of epithelial cells can be generated in order to identify new factors involved in terminal differentiation of prostate and bladder tissue.
Without being bound by theory, reprogrammed epithelial cells can display properties of a “primitive” epithelial cell. Although it may be found that specific culture conditions do not promote their terminal differentiation or formation of organoid structures, tissue recombination assays provide an in vivo microenvironment that is more conducive to cellular differentiation.
Expression of reprogramming factors have been used in fibroblasts to generate cells with epithelial morphologies in culture. For this purpose, mouse embryonic fibroblasts (MEFs) of distinct genotypes (wild-type, Oct4-GFP knock-in, and Nkx3.1-lacZ knock-in) were derived from E13.5 mouse embryos after the head and pelvis were removed to exclude neural and prostate progenitors. These MEFs were used after sorting for the mesenchymal marker CD140 or sorting against Lin/Mac-1(CD11b)/EpCAM markers to exclude blood, endothelial, and epithelial contaminants, thereby reducing the heterogeneity of the primary fibroblast population (
These induced epithelial cells were further stably transduced with viruses expressing Nkx3.1 and AR or NKX3.1, AR and FOXA1, which are known master regulatory genes for prostate development, followed by tissue recombination assays with rat urogenital mesenchyme (UGM) in renal grafts in immunodeficient male mice (
A goal of stem cell biology is the creation of desired cell types and tissues, which can be achieved by directed differentiation from pluripotent cells, or alternatively by direct lineage conversion in which transdifferentiation of cell types occurs. While these approaches are utilized for applications in regenerative medicine, they can also be used as the basis for genetically-engineered models of human disease, including cancer. Without being bound by theory, direct lineage conversion can be used in combination with gene targeting methods for the creation of genetically-engineered human models of cancer. In this application, direct conversion and tissue recombination can be used to generate mouse and human prostate tissue, and this reprogramming methodology can be applied to generate human tumor tissue for modeling of prostate cancer. Mouse and human fibroblasts can be directly converted to prostate tissue using a three-step process involving transient induction of pluripotency factors, expression of master regulators of prostate epithelium, and tissue recombination with urogenital mesenchyme followed by renal grafting. This direct conversion approach can be used to analyze the molecular mechanisms of reprogramming to prostate tissue as well as to generate genetically-engineered human models of prostate cancer.
Without being bound by theory, the mechanisms of direct conversion and the generation of human models of prostate cancer can be investigated. For example, the direct conversion of mouse and human fibroblasts into prostate epithelium can be investigated by systems analyses to identify optimal master regulators of prostate epithelial differentiation and by molecular analyses of reprogrammed prostate tissue. Mechanisms of direct conversion to prostate epithelium can be analyzed by investigating the multiple steps of cellular reprogramming. These studies can determine whether there is a transient intermediate pluripotent state, identify the cell(s) of origin for reprogrammed prostate epithelium, and analyze the reprogramming activity of urogenital mesenchyme. Modeling of human prostate cancer initiation by gene targeting and direct conversion can be investigated using Transcription Activator-Like Effector nucleases (TALENs) for the specific alteration of tumor suppressor genes that are mutated in human prostate cancer, followed by generation of reprogrammed human prostate tissue. In combination, these studies can provide the basis for an innovative approach for human cancer modeling, which can yield insights into the molecular mechanisms of human prostate cancer initiation.
Without being bound by theory, the proposed studies can yield insights into the basis for direct lineage conversion and cellular reprogramming, which have multiple applications in regenerative medicine and disease modeling. For example, this can also provide the basis for an approach for generating genetically-engineered human models of prostate cancer, which can have important implications for understanding the molecular mechanisms of prostate cancer initiation and progression.
Mouse as well as human fibroblasts can be directly converted into epithelial cells in culture following transient expression of the four “pluripotency factors” (Oct4, Sox2, Klf4, c-Myc). Following expression of prostate regulatory genes such as androgen receptor (AR), FoxA1, and Nkx3.1 in these induced epithelial cells, and recombination with embryonic urogenital mesenchyme, the resulting renal grafts can generate histologically normal prostate tissue with appropriate expression of tissue-specific markers. TALENs have also been used for gene targeting in prostate epithelial cell lines. Computational/systems biology approaches have been used to construct genome-wide regulatory networks (interactomes) for mouse and human prostate tissue, which can allow identification of master regulator (MR) genes that govern prostate epithelial cell fates, and thereby promote optimization of the reprogramming process.
Based on these findings, and without being bound by theory, this direct conversion/transdifferentiation approach can be used successfully to generate normal human prostate tissue, and in combination with gene targeting approaches, can be used to generate genetically-engineered human models of prostate cancer. This experimental methodology can be validated and the mechanistic basis for the direct conversion process can be investigated. For example, the direct conversion of mouse and human fibroblasts into prostate epithelium can be investigated by the identification of master regulators (MRs) of prostate epithelial differentiation, and molecular analyses of the reprogrammed prostate tissue. These studies can employ systems analyses of mouse and human prostate gene regulatory networks to identify candidate MRs, followed by functional assessment of their ability to promote direct conversion. These studies can provide a comprehensive analysis of MR combinations for optimization of reprogramming to prostate epithelium.
A general strategy for reprogramming to generate mouse and human prostate tissue has been developed (
Generation of Induced Epithelial Cells by Transient Expression of Pluripotency Factors:
Expression of pluripotency factors in fibroblasts can induce the formation of cells with epithelial morphologies in culture, termed induced epithelial cells (iEpt) cells. Mouse embryonic fibroblasts (MEFs), generated from E13.5 limb buds of wild-type mice to exclude neural and prostate progenitors, as well as dermal fibroblasts (MDFs) from P0 mice, were used. These MEFs and MDFs were then flow-sorted for the mesenchymal marker CD140a and against Lin/Mac-1(CD11b)/EpCAM markers to exclude blood, endothelial, and epithelial contaminants, thereby reducing the heterogeneity of the fibroblast population (
The system for the expression of reprogramming factors was changed to one that is regulated by administration of doxycycline, which allows temporal control over their expression and avoids issues associated with their continuous expression. In this approach, MEFs and MDFs were derived using the same strategy as above from mice carrying a doxycycline-regulated single-copy transgene expressing Oct4, Sox2, Klf4, and c-Myc as a polycistronic transcript [A44]. These fibroblast cultures were treated with doxycycline for 5-9 days to induce pluripotency factor expression, followed by 10 days in the absence of doxycycline to select for OSKM-independent iEpt cells. Under these conditions, approximately 10% of cells were EpCAM+CD24+ and displayed a stable epithelial morphology. The transient expression of OSKM can induce iEpt cells to form in basal epithelial medium.
Production of Mouse Prostate Tissue from Reprogrammed Fibroblasts by Tissue Recombination:
iEpt cells were investigated for their ability to be further reprogrammed to generate prostate tissue. The expression of putative master regulators (MRs) of prostate differentiation was combined with a tissue recombination assay. A candidate gene approach was used to select putative prostate epithelial MRs based upon biological and biochemical identification of key transcription factors for prostate development (e.g., [A45]). Androgen receptor (AR) was selected due to its central roles in prostate specification, organogenesis, and adult homeostasis and regeneration [A40, A46]. FoxA1 was selected because it is known to be critical for prostate development and functions as a pioneer factor in opening chromatin for AR binding [A45, A47-A50]. Nkx3.1 was selected due to its role in prostate development and luminal epithelial differentiation, and its participation in many AR transcriptional complexes [A16, A45, A51, A52].
Using retroviruses that constitutively express AR, FoxA1, and Nkx3.1 [A19, A53], the ability of iEpt cells to form prostate tissue following recombination with urogenital mesenchyme was investigated. Urogenital mesenchyme from E18.5 rat embryos and renal grafting in immunodeficient NCR nude mice (Taconic), using between 50,000 and 250,000 iEpt cells together with 250,000 mesenchymal cells, was used. To determine the contribution of each MR to prostate tissue formation, iEpt cells that received different combinations and proportions of these factors were used. iEpt cells were generated using the constitutively-expressed OSKM factors with retroviruses expressing AR, FoxA1, or Nkx3.1 individually, or in combination. The resulting renal grafts were harvested after 6-8 weeks, and analyzed by hematoxylin-eosin staining and immunostaining for specific markers. As positive controls, adult mouse prostate epithelial cells in tissue recombinations performed in parallel were used. As negative controls, renal grafts were generated from iEpt cells in the absence of urogenital mesenchyme, which never formed prostate tissue, with or without prostate MR expression (n=0/11); instead, 9 of these grafts only formed teratomas, while the remaining 2 grafts formed teratomas with areas of endoderm differentiation, but no prostate formation. As another negative control, 17 grafts were generated from iEpt cells that were not infected by retroviruses expressing candidate MRs. Of these, 6 grafts formed teratomas, while an additional 11 grafts formed teratomas with areas of endodermal epithelial differentiation, characterized by formation of large ducts as well as tubular and glandular structures, but not prostate differentiation.
Overall, 13% (n=6/47) of the successful tissue grafts formed tissue structures that histologically resembled prostate tissue, as shown by hematoxylin-eosin staining of paraffin sections (
To confirm that the successful grafts reconstituted prostate tissue, immunostaining for specific markers of basal and luminal epithelial cells was performed. These marker analyses revealed a proper tissue architecture containing a basal epithelial layer expressing p63 and CK5, as well as a luminal epithelial layer expressing CK8, CK18, and AR (
Production of Human Prostate Tissue from Reprogrammed Fibroblasts by Tissue Recombination:
The ability of fibroblasts to generate human prostate tissue was investigated using a similar direct conversion approach. For this purpose, lentiviruses expressing doxycycline-inducible human OSKM was used together with the reverse tetracycline transactivator rtTA (Stemgent) to infect BJ normal human foreskin fibroblasts. Doxycycline was added at 2 days post-infection, and cells were cultured for 8 days in basal epithelial media, which resulted in approximately 15% frequency of conversion into iEpt cells. These human iEpt cells resembled the mouse iEpt cells in their expression of CK5, CK8, CK18, and beta-catenin (
The resulting grafts were analyzed by H&E staining and immunostaining for specific epithelial markers, which showed their strong similarity to normal human prostate tissue (
The direct conversion process can be investigated using the optimization of direct conversion to prostate tissue using systems approaches to identify candidate master regulators for prostate epithelium. The mechanisms of direct conversion can be investigated, including analyses of potential intermediate pluripotent states, lineage-tracing of iEpt cells to identify potential progenitor cells, and molecular analyses of the reprogramming activity of urogenital mesenchyme. Direct conversion can be combined with gene targeting to establish genetically-engineered models of human prostate cancer.
Optimization of Direct Conversion into Prostate Epithelium:
Using candidate MRs identified by systems analyses, functional validation assays can be performed to identify successful reprogramming MR combinations for optimization of the direct conversion process. The quality of the reprogrammed mouse and human prostate tissue can be assessed using histopathological and molecular analyses. The efficiency of the reprogramming process can be assessed to determine the number of iEpt cells necessary for successful graft formation.
Experimental Design:
To determine whether candidate MRs can improve the reprogramming of iEpt cells in culture, lentiviral infection can be used to overexpress positive MRs or knock-down negative MRs in mouse and human iEpt cells, followed by tissue recombination and renal grafting. These experiments can be performed using synergistic combinations of candidate MRs identified bioinformatically, as well as using combinations of candidate MRs together with AR, Nkx3.1 and FoxA1, or individually as a control. If new MR combinations that appear to greatly enhance the efficiency or quality of direct conversion are identified, limiting dilution analyses can be performed as well as detailed marker studies of the reprogrammed prostate tissue.
For reprogrammed prostate tissues, H&E staining and immunostaining for specific markers can be performed (
To assess the efficiency of direct conversion, limiting dilution analyses can be performed to determine the number of iEpt cells required for successful formation of prostate grafts. The number of urogenital mesenchyme cells remains constant at 250,000/graft, while the number of iEpt cells can be varied from 100 to 50,000. The results can then be analyzed by the extreme limiting dilution algorithm (ELDA) [A59], which has been used previously for analyses of graft formation by isolated prostate basal cells [A21]. In each experiment, the number of iEpt cells co-expressing prostate lineage master regulators can be determined retrospectively by immunostaining to adjust the cell numbers for the starting iEpt population.
Without being bound by theory, molecular analyses to investigate the similarity of reprogrammed prostate tissue to native mouse and human prostate tissue can be performed. Control mouse and human tissue grafts produced by tissue recombination of normal mouse and human prostate tissue with rat urogenital mesenchyme can also be analyzed. For example, expression profiles from at least six independent reprogrammed prostate grafts can be generated, as well as control grafts by RNA-sequencing. RNA-seq can then be performed using 30 million single-end reads generated on a high-throughput sequencing platform, such as the Illumina HiSeq 2000 platform. Expression profiles of normal adult mouse prostate tissue can be obtained by RNA-seq, while expression profiles of normal human prostate tissue can be obtained from publically available datasets [A57] and by RNA-seq analysis. The resulting expression profiles can be analyzed by Principal Components Analysis (PCA) and unsupervised hierarchical clustering to determine the overall similarity of these expression profiles [A21, A60]. Gene expression signatures of the reprogrammed tissue grafts versus normal control grafts can be generated to investigate their similarity to native mouse and human prostate tissue using Gene Set Enrichment Analysis (GSEA) [A21, A60].
Normal adult human prostate tissue can be obtained from primary cystectomy samples in which normal prostate tissue is surgically excised in conjunction with the removal of bladder tumors. The normal histology of the prostate tissue can be verified by pathological analysis.
In one embodiment, it is conceivable that these analyses can identify putative MR combinations that can promote direct conversion of fibroblasts to prostate tissue in the absence of transient expression of pluripotency factors. The properties of efficient reprogramming combinations can be investigated using alternative methods for direct conversion.
An interactome for human prostate tissue has been generated, using the ARACNe algorithm for reverse engineering [A29, A30, A56]. This human prostate interactome was constructed from a large published dataset comprised of prostate cancer specimens and adjacent normal tissue [A57], and was validated by computational analysis of published genome-wide chromatin immunoprecipitation (ChIP) data for transcription factors such as c-Myc, AR, and BCL6, showing consistently high statistical significance.
To identify master regulators (MRs) for normal prostate epithelium, the human prostate interactome was used for analysis using the MARINa algorithm [A32, A33]. Published gene expression profiles were used for mouse prostate tissue during organogenesis as well as adulthood [A58] to generate gene signatures for normal prostate tissue. Cross-species interrogation of the human prostate interactome using signatures for normal prostate differentiation during organogenesis (comparing embryonic to adult prostate) consistently identified both FoxA1 and Nkx3.1 among the top candidate MRs (
Successful reprogramming mouse and human fibroblasts into prostate tissue has been shown. A candidate gene approach has been used to identify putative master regulators (MRs) that promote direct conversion to prostate epithelium. A systems approach for the unbiased identification of such master regulators and their potential synergistic interactions can be used, and functional validation of the top candidate master regulators can be performed in the direct conversion assay. The direct conversion process can then be optimized by performing detailed histological and molecular analyses of the quality and efficiency of reprogramming by these MRs.
Experimental Design:
Published array data has been used for the identification of candidate MRs using the MARINa algorithm to interrogate the human prostate interactome, and has identified FOXA1 and NKX3.1, among others, as candidate MRs for prostate epithelium (
To identify additional candidate MRs of prostate epithelium, gene expression profiling of adult mouse prostate tissue can be performed, as well as from embryonic (18.5 dpc) and neonatal (postnatal day 4 and day 12) prostate, with at least six samples for each time point. These tissues can be dissociated and used in flow cytometry using EpCAM antibodies to purify epithelial cells, followed by RNA-seq analysis. The resulting expression profiles can be used to generate signatures corresponding to embryonic, neonatal, and adult prostate epithelium. These expression signatures can be used to interrogate the human prostate interactome using the MARINa algorithm to identify candidate MR genes [A32, A33]; in parallel, similar analyses can be performed using a recently constructed mouse prostate interactome. Without being bound by theory, this approach can be used to identify potential synergistic pairs of candidate MRs [A32, A33].
Without being bound by theory, new candidate master regulators of prostate epithelium can be identified by these systems analyses. These candidate MRs can function synergistically with other prostate reprogramming factors to induce direct conversion to prostate epithelium. These system analyses can also identify negative MRs whose expression needs to be down-regulated to facilitate direct conversion; such reprogramming inhibitors are difficult to identify with candidate gene approaches. In one embodiment, candidate MRs can require co-expression in combination with several other reprogramming factors to induce prostate reprogramming.
Without being bound by theory, the mechanisms of direct conversion to prostate epithelium can be analyzed by investigation of the steps of cellular reprogramming involved in the multi-step conversion process. For example, these studies can use lineage-tracing to identify the induced epithelial cell type(s) that are most amenable for reprogramming by prostate MRs, can examine whether successful reprogramming requires traversal through a transient pluripotent state, and can address the role of embryonic urogenital mesenchyme in promoting prostate transdifferentiation.
To understand the cellular and molecular mechanisms of direct conversion, the key features of the reprogramming process can be investigated. These studies can examine whether direct conversion proceeds through a pluripotent state, identify the cell type that gives rise to the prostate epithelial cells, and analyze the secreted factor(s) in the urogenital mesenchyme that is involved in prostate specification. These studies can provide important mechanistic insights into the reprogramming process.
Analysis of Traversal of the Pluripotent State:
Previous analyses of direct conversion protocols have concluded that the reprogramming process does not traverse a pluripotent state during the transdifferentiation process [A61-A63]. These analyses have not addressed the possibility that this pluripotent state may be extremely transient, and can only occur in a small percentage of the cell population that gives rise to the reprogrammed cells/tissue. Sporadic and transient expression of pluripotency markers in a small population of cells can be detected using a sensitive reporter. A mouse reagent that allows detection of Nanog expression, even if it occurs very transiently in a limited cell population has been developed.
Experimental Design:
Whether fibroblasts traverse the pluripotent state during generation of iEpt cells in culture can be investigated. MEFs from a mouse line carrying an IRES-GFP knock-in within the 3′ untranslated region of Oct4 [A64] can be generated. These Oct4-GFP MEFs can be used to determine whether rare GFP-positive cells can be identified during the formation of iEpt cells in basal medium. As a positive control, parallel cultures in mESC/LIF medium to generate iPSC colonies (GFP-positive) can be performed.
An inducible Nanog-CreERT2 transgene can be used in combination with the fluorescent Cre-reporter R26R-Tomato to perform lineage-marking of cells that express Nanog during direct conversion. MEFs containing the Nanog-CreERT2 transgene can only express the Tomato reporter if the Nanog promoter is activated by 4-hydroxy-tamoxifen (4-OHT), but continue to express Tomato even if Nanog is no longer expressed. (It is essential to use an inducible Cre driver under the control of the Nanog promoter, since a constitutively active Cre would promote Cre-reporter expression in pluripotent epiblast cells and thus all of the cells of the resulting mouse.) Two independent BAC (bacterial artificial chromosome) transgenic mouse lines that express CreERT2 under the control of the endogenous Nanog promoter (
MEFs from Nanog-CreERT2; R26R-Tomato/+ mouse embryos can be generated, using the protocols that have been followed previously for MEF isolation and culture. The resulting MEFs can be utilized for the direct conversion protocol using doxycycline-inducible lentiviruses expressing human OSKM and rtTA for transient expression of pluripotency factors as described previously, but also cultured in the presence of 4-OHT. As a positive control, parallel reprogramming experiments can be performed using cell culture conditions that promote iPSC formation. Finally, if such traversal is observed, the contribution of Tomato-positive cells to the formation of reprogrammed prostate tissue can be investigated.
Without being bound by theory, Nanog-CreERT2 MEFs represent a sensitive reagent, since transient Nanog expression can be detected no matter when it occurs in the culture due to the indelible lineage-mark, and the level of Cre expression only needs to be sufficient to induce a single recombination event at the ROSA26 locus. Upon detection of Tomato expression in our cultures, the time point at which Cre-mediated recombination occurs can be identified, and the expression of Nanog and other pluripotency markers can be examined by quantitative RT-PCR and RNA-seq approaches. If reprogramming to prostate epithelium traverses a transient pluripotent state, as detected using the Nanog-CreERT2 mice, other direct conversion processes that have been reported in the literature can be investigated to determine whether a similar transient pluripotent state may occur.
Lineage-Tracing of the Cell of Origin for Converted Prostate Epithelium:
To determine whether the formation of reprogrammed prostate tissue in renal grafts recapitulates processes of normal organogenesis, or whether instead it mimics features of adult tissue homeostasis and/or regeneration, the cell type that gives rise to reprogrammed prostate epithelium can be investigated. During organogenesis, the basal epithelium contains progenitors for both basal and luminal cell types, whereas the luminal epithelium appears to be unipotent [A65]. In the adult prostate, bipotential progenitors exist in the basal epithelium during homeostasis and regeneration, but are relatively rare [A21], while luminal stem/progenitors have been identified during regeneration [A20]. Lineage-tracing of the iEpt cells in culture can be performed to determine which cell type(s) within this heterogeneous cell population can generate prostate epithelium in renal grafts. Specifically, inducible Cre drivers can be used to mark iEpt cells expressing basal or luminal markers to determine whether either or both cell populations can generate reprogrammed prostate epithelium in tissue recombinants. These studies can also be relevant for understanding the cell of origin for the human prostate tumors.
Experimental Design:
Lineage-tracing can be performed using inducible Cre drivers that mark basal or luminal subpopulations of the iEpt cells, which display heterogeneous marker phenotypes in culture (
Without being bound by theory, if the reprogrammed prostate epithelium is derived from basal iEpt cells, lineage-tracing using the CK5-CreERT2 transgenic line would reveal extensive contribution of YFP-positive cells to the renal grafts. If luminal iEpt cells give rise to reprogrammed prostate tissue, lineage-tracing using the CK8-CreERT2 and CK18-CreERT2 mice would generate extensive YFP-positive contribution in the grafts. An interaction between basal and luminal iEpt cells can be necessary for generation of reprogrammed prostate tissue, which in this case would not be clonally derived. This interpretation would be suggested if flow-sorted basal and luminal iEpt cells are unable to form prostate tissue as purified populations, but can do so if mixed together prior to tissue recombination with urogenital mesenchyme. It may be the case that reprogrammed prostate tissue is generated from “intermediate” cells that co-express basal and luminal markers (such as CK5+CK8+ cells), which would be suggested if both purified populations of basal (CK5+) and luminal (CK8+) iEpt cells are able to generate prostate tissue. Further flow-sorting studies using cell-surface markers can be performed, such as the basal cell marker CD49f, in combination with CK8-CreERT2 lineage-tracing to isolate intermediate cells co-expressing basal and luminal markers. The ability of iEpt population(s) that generate reprogrammed prostate tissue to display stem cell properties, can be determined using assays that have been previously employed to identify stem cell populations in the adult prostate epithelium [A20, A21].
Systems Analysis of Embryonic Urogenital Mesenchyme:
Without being bound by theory, to identify the critical factor(s) responsible for the reprogramming properties of embryonic urogenital mesenchyme, a candidate pathway approach can be pursued, in combination with an unbiased systems analysis. For example, specific signaling pathways known to be active in embryonic urogenital mesenchyme can be tested for their necessity for reprogramming. Gene signatures of urogenital mesenchyme can be generated to interrogate the prostate interactomes.
Experimental Design:
In a candidate pathway approach, signaling pathways that have been implicated in prostate specification can be focused on, these include the canonical Wnt, FGF, and BMP pathways [A66]. To test whether these pathways are critical for prostate tissue reprogramming, lentiviral infection can be used to express secreted inhibitors of these pathways in mouse urogenital mesenchyme or to knock-down candidate signaling factors. For example, to test the role of canonical Wnt signaling, lentiviral overexpression of Dkk1 can be used to inhibit Wnt signaling, and as a control for its effects, the sensitive TCF/LefH2B-GFP transgenic reporter for canonical Wnt signaling activity [A67] can be used to monitor the consequences of Dkk1 overexpression. Similar approaches have been used to investigate the role of canonical Wnt signaling in early stages of prostate organogenesis [A51].
In the systems approach, differentially expressed genes as well as candidate master regulators can be identified. For this purpose, RNA-seq analyses can be performed to generate expression profiles of mouse embryonic urogenital mesenchyme as well as the neighboring bladder mesenchyme, which lacks reprogramming activity. Differentially expressed genes between urogenital mesenchyme and bladder mesenchyme can be identified, and gene ontology-biological process (GO-BP) analyses can be performed to identify differentially active signaling pathways. Expression signatures can be generated for urogenital mesenchyme to interrogate the mouse prostate interactome (which is based upon samples containing stromal tissue) for the identification of candidate MRs and synergistic MRs. These analyses can provide insights into signaling pathways and candidate ligands that can correspond to the reprogramming activity of the urogenital mesenchyme. Such candidate ligands can then be further investigated by lentiviral knock-down in the urogenital mesenchyme to determine whether their loss-of-function reduces or eliminates reprogramming activity.
For both approaches, if a candidate signaling ligand/pathway is identified as being critical for reprogramming activity using loss-of-function approaches, gain-of-function approaches to validate this finding can be used. Lentiviral infection can be performed to overexpress candidate ligands in rodent stromal cell lines that are derived from urogenital mesenchyme, but lack reprogramming activity, such as UGSM-2 [A68]. The resulting stromal cells can be investigated for its ability to support growth of normal prostate epithelium in tissue recombinants, as well as its ability to participate in direct conversion to prostate tissue.
Without being bound by theory, among the signaling pathways that have been investigated in prostate formation, there is evidence supporting a central role for canonical Wnt signaling [A51, A69-A71], and the candidate pathway approach can initially focus on canonical Wnt signaling. The reprogramming activity of urogenital mesenchyme can be at least partially unrelated to its inductive activity during prostate formation, and all candidate signaling pathways identified by systems analysis can be analyzed. In some embodiments, there can be cooperative effects and/or functional redundancy of multiple signaling factors that correspond to the reprogramming activity, analyses of synergistic MRs and GO biological processes can provide insights into the activities and identities of such cooperative signaling factors.
An objective in stem cell biology is the development of therapies based on the generation of clinically relevant human cell types and tissues. In the context of disease, such approaches can also be harnessed for the creation of genetically engineered models of human cancer. Without being bound by theory, direct conversion/transdifferentiation methodologies can be employed to generate desired cell types and tissues from fibroblasts in culture, followed by their oncogenic transformation. In combination with gene targeting technologies, such approaches can be used to create precise genetically-engineered models of human cancer.
Despite the widespread use of mouse models of cancer, such models can be limited by their inability to fully recapitulate the physiological processes underlying human cancer, and can be limited for applications such as preclinical testing of candidate therapeutics. For example, analogous mouse and human tissues can have important anatomical and/or physiological differences, such as the strictly ductal histology of the mouse prostate gland versus the ductal-acinar structure of the human prostate. Consequently, it is essential to develop model systems using human tissue that can accurately recapitulate cancer, yet are amenable to gene targeting approaches and other genetic manipulations.
Without being bound by theory, cellular reprogramming methods can be used to develop a new generation of models of human cancer, using prostate cancer as a model system. For example, the direct conversion of mouse and human fibroblasts into prostate epithelium together with tissue recombination approaches can be used to generate histologically normal prostate tissue in renal grafts. In combination with gene targeting of tumor suppressors using Transcription Activator-Like Effector nucleases (TALENs), this approach can generate oncogenically transformed prostate tissue, which can have considerable clinical relevance for the generation of prostate cancer models.
Human prostate cancer initiation can be modeled by gene targeting and direct conversion using TALENs for the specific alteration of tumor suppressor genes that are mutated in human prostate cancer, followed by the generation of prostate tissue using the direct conversion methodology. Histopathological and molecular analysis of the resulting transformed prostate tissue can allow functional analysis of the roles of these tumor suppressors in human prostate cancer initiation and progression.
Without being bound by theory, these studies can provide the basis for an approach to human cancer modeling, which can lead to new insights into the molecular basis of human cancer initiation and progression as well as improved pre-clinical studies of candidate therapeutics.
TALEN-Mediated Gene Targeting in Human Fibroblasts and Prostate Epithelial Cells:
To demonstrate the feasibility of gene targeting in combination with direct conversion, TALENs have been used for gene targeting in the RWPE-1 human prostate epithelial cell line as well as in BJ foreskin fibroblasts. AAVS1, which encodes the PPR1R12C gene has been targeted and is a well-characterized locus used previously for gene targeting in human embryonic stem cells [A37]. Using published TALEN pairs and a GFP-expressing puromycin-resistance donor cassette [A37], AAVS1 was successfully targeted in both cell lines. To eliminate non-specific targeting, the cells were selected in puromycin followed by clonal growth by limiting dilution. Analysis of the AAVS1 locus showed proper targeting and integration of the donor GFP cassette (
To generate genetically-engineered models of human prostate cancer initiation and early progression, gene targeting using TALE nucleases can be performed in human fibroblasts followed by direct conversion into prostate tissue. Straightforward targeting mediated by non-homologous end joining to generate loss-of-function alleles, or a two-step homologous recombination approach to create specific point mutations, can be used. These studies can permit the analysis of early events in cancer initiation in human prostate, which has previously been inaccessible to molecular genetic analysis.
Experimental design: Gene targeting of PTEN and TP53 in human fibroblasts can be performed. These tumor suppressors have been selected since their loss-of-function can yield prostate cancer phenotypes. Notably, in mouse models, loss of PTEN function results in high-grade PIN and eventually adenocarcinoma [A72-A75], while TP53 loss does not have a cancer phenotype, but deletion of both genes results in aggressive adenocarcinoma [A76]. To introduce deletions at the start codon of these two genes, published TALENs (Addgene) that cleave near the N-terminus of the protein coding sequence [A38] can be used. Targeting of PTEN and TP53 in human BJ fibroblasts can be performed, followed by the direct conversion protocol to form prostate tissue in renal grafts using immunodeficient NCR nude mice. These studies can be performed using targeting of PTEN or TP53 individually, or can use sequential targeting of both tumor suppressors. The resulting tissue grafts can be analyzed histologically for a PIN and/or adenocarcinoma phenotype. Basal (p63, CK5, CK14) and luminal (CK8, CK18) markers can be analyzed to ascertain whether the PIN/tumor lesions have a strong luminal phenotype that is typical of human prostate adenocarcinoma. The expression of alpha-methylacyl-CoA racemase (AMACR), which is up-regulated in human prostate cancer [A77], can be assessed. If robust tumor formation is observed, these tumors can then be propagated by renal or orthotopic grafting in immunodeficient mice.
The creation of a specific point mutation in TP53 can be performed, using an approach similar to that employed for genetic-engineering in mouse ES cells. TALENs can mediate gene targeting in human cells by homologous recombination with insertion vectors, analogous to conventional approaches in mouse ES cells, including two-step procedures that can introduce point mutations followed by Cre-loxP recombination to remove inserted drug-selection cassettes [A37]. These studies can use a two-step targeting approach to introduce a specific missense mutation, R273H, into the TP53 coding region in fibroblast cells that are either wild-type or contain a homozygous PTEN null mutation, followed by phenotypic analysis of reprogrammed prostate tissue. The TP53 residue 8273 is a mutational hotspot in human cancer, including prostate cancer [A78]. Studies in genetically engineered mice show that the corresponding Tp53R27OH mutation has a prostate cancer phenotype distinct from that of Tp53 null mutants, suggesting a potential role for TP53 in prostate cancer initiation rather than in advanced disease [A79].
The creation of mutations in genes that have recently been identified in whole-genome and exome sequencing projects as mutated in human prostate cancer can be performed. Although human prostate cancer displays a relatively low mutation rate in general, particularly for many known tumor suppressor genes, a significant number of genes have been found to be mutated that have not been functionally characterized to any significant degree, including genes such as SPOP, MED12, and HOXB13 [A57, A78, A80-A83]. To address the functional significance of these genes in human prostate cancer progression, these genes can be mutated either individually or in combination with PTEN or other tumor suppressors in human fibroblasts to investigate the phenotype of the resulting reprogrammed prostate tissue. TALENs can be created to mutate the desired target sites using currently available reagents (Addgene) [A38], and use non-homologous end joining to mutate genes to create simple loss-of-function alleles (e.g., for SPOP mutations) or homologous recombination to create specific point mutations (e.g., for the HOXB13 G48E allele).
Without being bound by theory, these studies can provide the foundation for new genetically-engineered models of human prostate cancer. Studies of the cell of origin of reprogrammed prostate tissue can be relevant for understanding the cell of origin for prostate cancer, which can originate either from luminal or basal cells in mouse models [A21, A84]. In some embodiments, there may be intrinsic variability in the extent of reprogramming that can complicate the interpretation of tumor phenotype. Continued development of the TALEN technology can undoubtedly lead to its application for chromosomal engineering, as is now commonly performed using Cre-loxP technology [A85], and allow for the recapitulation of the extensive genomic rearrangements that typically take place in prostate cancer, such as the frequent TMPRSS2-ERG gene fusion. In other embodiments, targeting of certain tumor suppressor genes may affect the efficiency and possibly the outcome of direct conversion, since reduced function of the p53-p21 pathway greatly increases efficiency of fibroblast reprogramming to iPSC [A86-A89]. The generation of human prostate tumor models using TALEN-mediated gene targeting, allows for future studies that can extend the applicability of this approach. Chromosomal engineering approaches can be used to generate the TMPRSS2-ERG fusion and other genomic rearrangements in reprogrammed prostate tumors. The molecular mechanisms of castration-resistance in this system can also be investigated, including the possibility of endogenous androgen biosynthesis by reprogrammed tumors.
Without being bound by theory, the direct conversion/transdifferentiation to prostate epithelium can provide the basis for many future studies of reprogramming. In particular, the approaches developed herein can be generally applicable for reprogramming to other tissues of interest, and for creating genetically-engineered models for a range of human cancers. The systems analyses coupled with mechanistic and functional studies can yield insights into normal processes of prostate organogenesis and stem cell biology. The use of xenograft-based genetically-engineered models of human cancer permits the extension to analyses of candidate therapeutics and drug response.
Doxycycline-inducible lentiviral pluripotency factors, OSKM, were used to reprogram mouse embryonic fibroblasts (MEFs) to induced epithelial (iEpt) cells in culture. This allows precise timing of expression of the pluripotency factors, OSKM. Lentiviruses were produced in 293FT packaging cells using established protocols. Lentiviruses were pooled and filtered prior to infection. 2 days after infection, MEFs were treated with Dox for 7-9 days to induce the pluripotency factors in 10% FBS/DMEM or 10% KSR/DMEM, no LIF was added to the media. After 7-9 days, Dox was withdrawn from the media and cells were infected with lentiviruses expressing human NKX3.1 (pLOC NKX3.1 iresGFP), human AR (pLentiV6.2 HA-AR), and human FOXA1 (pSIN-EF2 Foxa1-puro) (NAF cocktail) and cultured in prostate basal media (Cnt-12, Cnt-Prime media, CellnTEC) for 7 days. To avoid confusion with host derived cells, prior to tissue recombination, an additional infection with pLOC RFP lentiviruses was performed to color-mark the iEpt-NAF cells.
In the next step, the iEpt-NAF cells were recombined with rat embryonic urogenital sinus mesenchyme (UGM) and grafted under the renal capsule of athymic nude mice. The tissue recombinants were harvested after 6-8 weeks and analyzed by hematoxylin-eosin staining and immunostaining for prostate tissue specific markers. Similar to our experimental set-up, this combination of transient expression of lentiviral pluripotency factors and lentiviral transduced master regulators of prostate development were able to reprogram MEFs to iEpt cells which were able to grow into prostate tissue under the inductive force of UGM (
KLF5 has been used as a master regulator of bladder development [B1] to re-specify iEpt cells towards bladder epithelia in tissue recombination experiments with rat embryonic bladder mesenchyme. When KLF5 is missing from the bladder epithelial cells, urothelial precursor cells remain in an undifferentiated state and the resulting urothelium fails to stratify and to express terminal differentiation markers (e.g. uroplakins). Similar to the reprogramming to prostate tissue experiments, we have used KLF5 expressing lentiviruses to infect iEpt cells. iEpt-KLF5 cells were further recombined with rat embryonic bladder mesenchyme and grafted under the renal capsule. In this set-up, 4/4 renal grafts grew (
The same doxycycline-inducible pluripotency factors, OSKM, were used to reprogram MEFs from CK18CreERT2/Rosa26-Tomato to induced pluripotent cells (iPS) cells in culture. Cells of the above genotypes were infected with OSKM and rtTA lentiviruses and cultured in mouse embryonic stem cell media in the presence of LIF. According to iPS published protocols, Dox was added to the media for 11 days to induce the pluripotency factors, followed by Dox-free media for another 5-7 days when iPS colonies were picked and moved on a mitomycin-treated fibroblast feeder layer. 1 μM 4-hydroxy Tamoxifen (4-OHT, (Z)-4-Hydroxytamoxifen, H7904, Sigma) was also added to the media after the OSKM infection until the iPS colonies picking to lineage-trace cells which expressed CK18 or Gata6. In accord with previous literature, upon OSKM activation, a proportion of the MEFs undergo a transition to an CK18+ epithelial phenotype and express Tomato in the presence of 4-OHT (FIG. 14A,B). Some of these Tomato-positive cells developed into iPS colonies after 11 days of Dox induction (FIG. 14C,D). A single Tomato-positive iPS colony was picked from the plate at Day 12 and recombined undissociated with rat UGM in collagen. The resulting cell recombinant was grafted under the renal capsule of an athymic nude mouse. The renal graft was harvested at 8 weeks post-grafting and analyzed by gross microscopy (
A similar strategy can be employed to generate bladder tissue from a single iPS colony after recombination with rat embryonic bladder mesenchyme.
Using the same Dox-inducible reprogramming protocol, iPS cells were generated from Gata6CreERT2/Rosa26-caggEYFP MEFs. Passaged 2 iPS colonies (FIG. 15A,B) (4 independent colonies) were replated on 0.1% gelatin coated plates and the mES media was changed to endodermal differentiation media containing Activin A (50 ng/ml; RnD Systems, Minneapolis, USA), Noggin (200 ng/ml; RnD Systems) and a GSK3β inhibitor (1 μM of 6-bromo indirubin-3-oxine, BIO; Merck KGaA, Darmstadt, Germany) in 25% F-12/75% IMDM/2 mM Glutamax/0.55 mM beta-mercaptoethanol/N2 supplement [2]. 4-OHT was added to the differentiation media to mark endodermal differentiated cells. Numerous YFP+ colonies were observed at 4-6 days of culturing in this media indicating that these cells express or passed through a GATA6-positive state (FIG. 15C,D). The YFP+ cells were sorted after 6 days of differentiation and analyzed for expression of endodermal markers by RT-PCR. As expected, these cells expressed GATA6 and SOX7 mRNA at high levels compared with MEFs. For the differentiation towards prostate and bladder lineages, YFP+ endodermal cells were plated in 3D-culture conditions in matrigel with (for prostate) or without (for bladder) dihydrotestosterone propionate (DHT, Sigma). In these culture conditions, spherical growth of some of the YFP+ cells was observed (FIG. 15E,F). These endodermal 3D-structures can be grafted under the renal capsule of nude mice after recombination with rat embryonic UGM or bladder mesenchyme.
As an alternative to continuous activation of the pluripotency factors, our reprogramming protocols were switched to a lentiviral OSKM cocktail. Specifically, doxycycline-inducible lentiviral vectors expressing the pluripotency factors, Oct4, Sox2, KLF4 and cMyc together with the vector expressing the reverse tetracycline transactivator (rtTA) were acquired from Addgene (FU-tet-o-hOct4, cat.no 19778; FU-tet-o-hSox2, cat.no 19779; FU-tet-o-hKLF4, cat.no 19777; FU-tet-o-hc-myc, cat.no 19775; FUdeltaGW-rtTA, cat.no 19780). Lentiviruses were produced in 293FT packaging cells using established protocols for second generation lentiviral system based on the packaging plasmids pMD2.G (VSV-G envelope expressing plasmid, cat. no 12259) and psPAX2 (Addgene cat. no 12260). Briefly, 293FT cells were transfected with the packaging plasmids and the OSKM and rtTA encoding plasmids using Lipofectamine 2000 (Invitrogen, cat.no 11668-019). Each lentivirus was produced separately. Lentiviruses were collected at 48 hrs and 72 hrs post-transfection, pooled and filtered prior to infection. Thus, mouse embryonic fibroblasts derived from WT 129Sv mice, Oct4-GFP knock-in, Nkx3.1 Lacz+/−, CK18CreERT2/Rosa26-Tomato, Gata6CreERT2/Rosa26-caggEYFP mice were infected twice at 6 hours interval with a pool of lentiviruses encoding OSKM and rtTA. 48 hours after the last infection, MEFs cultured in 10% FBS/DMEM or 10% KSR/DMEM (FBS from Gemini, KSR and DMEM from Invitrogen) were treated with doxycycline (Dox) for 7-9 days to induce the pluripotency factors OSKM.
For generation of prostate tissue: After 7-9 days, Dox was withdrawn from the media and induced epithelial cells (iEpt) cells were infected twice at 6 hrs interval with lentiviruses expressing human NKX3.1 (pLOC NKX3.1 iresGFP; human AR (pLentiV6.2 HA-AR), and human FOXA1 (pSIN-EF2 Foxa1-puro) (NAF cocktail). The lentiviruses were produced in 293FT cells using the same packaging plasmid system as above. After the last NAF infection, the cell media was switched to prostate basal epithelial media (Cnt-12, CellnTEC) or generic basal epithelial media (Cnt-Prime media, CellnTEC) for 7 days. In some experiments, to avoid confusion with host-derived cells, prior to tissue recombination, an additional infection with pLOC RFP lentiviruses (derived from the pLOC RFP ires GFP vector obtained from the Califano Lab by removing the ires GFP cassette) was performed to color-mark the iEpt-NAF cells.
For generation of bladder tissue: After 7-9 days, Dox was withdrawn from the media and induced epithelial cells (iEpt) cells were infected twice at 6 hrs interval with lentiviruses expressing human KLF5 (pSIN-EF2 KLF5-puro). The KLF5 lentiviruses were produced in 293FT cells using the same packaging plasmid system as above. After the last KLF5 infection, the cell media was switched to generic basal epithelial media (Cnt-Prime media, CellnTEC) for 7 days. In some experiments, to avoid confusion with host-derived cells, prior to tissue recombination, an additional infection with pLOC RFP lentiviruses was performed to color-mark the iEpt-KLF5 cells.
In the next step, the iEpt-NAF and iEpt-KLF5 cells were recombined with rat embryonic urogenital sinus mesenchyme (UGM) and rat embryonic bladder mesenchyme, respectively in collagen. The recombined cells in collagen were grafted under the renal capsule of athymic nude mice. The tissue recombinants were harvested after 6-8 weeks and analyzed by hematoxylin-eosin staining and immunostaining for epithelial (CK5, CK8, CK18); endodermal (Foxa1, KLF5); prostate tissue specific (AR, Probasin) or bladder specific markers (Uroplakin III). The cultured origin of the tissues in the grafts was verified by GFP (for Nkx3.1 ires GFP) and RFP (for pLOC RFP) immunostaining.
Two further new approaches to generate prostate and bladder epithelial tissues in vivo are described. In the first instance, prostate tissue was generated from CK18CREert2/R26r-Tomato iPS after recombination with rat embryonic UGM. In the second instance, endodermal differentiation experiments with Gata6CreERT2/R26r-caggYFP iPS were performed. The endodermal cells can be recombined with tissue specific mesenchyme and renal grafted.
This application is a continuation-in-part of International Application No. PCT/US2013/028265, filed on Feb. 28, 2013, which claims priority to U.S. Application Ser. No. 61/604,455, filed on Feb. 28, 2012, the contents of each of which are hereby incorporated by reference in their entireties.
The invention was made with government support under Grant No. R01 DK076602 awarded by the National Institute of Diabetes and Digestive and Kidney Diseases, and under Grant No. P01 CA154293 awarded by the National Cancer Institute. The Government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61604455 | Feb 2012 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US2013/028265 | Feb 2013 | US |
| Child | 14471836 | US |
