CN112813019A

CN112813019A - Definitive endoderm

Info

Publication number: CN112813019A
Application number: CN202110089971.8A
Authority: CN
Inventors: 凯文·艾伦·德阿姆尔; 艾兰·D·奥戈尔尼克; 埃马纽埃尔·E·拜特格
Original assignee: Viahit
Current assignee: Viahit
Priority date: 2003-12-23
Filing date: 2004-12-23
Publication date: 2021-05-18
Also published as: CN109628371A; CN109628371B; ZA200605100B; CN1946838A

Abstract

Cell cultures comprising definitive endoderm cells and methods of making the same are disclosed. Also disclosed are cell populations of substantially purified definitive endoderm cells and methods of isolating, enriching and purifying definitive endoderm cells from other cell types.

Description

Definitive endoderm

The present application is a divisional application of a co-owned application having an application number of 201910016274.2, with the application date of 2004 being 12/23.

RELATED APPLICATIONS

This application is a non-provisional application entitled under the provisions of 35u.s.c. § 119(e) priority of us provisional patent application No. 60/532,004 filed 12/23/2003, entitled definitive endoderm, priority of us provisional patent application No. 60/586,566 filed 7/9/2004, entitled chemical factor cell surface receptor for separating definitive endoderm, also entitled "priority of us provisional patent application No. 60/586,566 filed 7/9/2004, and priority of us provisional patent application No. 60/587,942 filed 7/14/2004, entitled chemical factor cell surface receptor for separating definitive endoderm, under the provisions of 35u.s.c. § 119 (e). The disclosures of each of the priority applications listed above are incorporated herein by reference in their entirety.

Technical Field

The present invention relates to the fields of medicine and cell biology. In particular, the invention relates to compositions, including mammalian definitive endoderm cells and methods of making, isolating and using these cells.

Background

In 1994, human pluripotent stem cells, such as Embryonic Stem (ES) cells and Embryonic Germ (EG) cells, were first isolated in cultures without fibroblast feeder (Bongso et al, 1994), and then isolated in cultures with fibroblast feeder (Hogan, 1997). Later, Thomson, Reubinoff and Shamblott established continuous cultures of human ES and EG cells using mitotically inactivated mouse feeder layers (Reubinoff et al, 2000; Shamblott et al, 1998; Thomson et al, 1998).

Human ES and EG cells (hESCs) provide new opportunities to study early development in humans and to intervene in the treatment of some diseases such as diabetes and parkinson's disease. For example, the use of insulin-producing cells derived from hESCs would provide a tremendous improvement over current cell therapy approaches that utilize donor pancreatic cells. However, it is currently unknown how to generate insulin-producing beta cells from hESCs. Likewise, current use of diabetes is from the donor pancreasCell therapy of glandular islet cells is limited by the shortage of high quality islet cells required for transplantation. Cell therapy for a type I diabetic patient requires transplantation of approximately 8X 10⁸The pancreatic islet cells of (Shapiro et al, 2000; Shapiro et al, 2001 a; Shapiro et al, 2001b). Similarly, at least two healthy donor organs are required to obtain sufficient islet cells for successful transplantation. HESCs provide a source of starting material from which a substantial amount of high quality differentiated cells are developed for human cell therapy.

Two characteristics that make hESCs well suited for cell therapy applications are pluripotency and the ability to maintain these cells in long-term culture without the accumulation of genetic changes. Pluripotency refers to the ability of hESCs to differentiate into derivatives of all 3 primitive germ cell layers (endoderm, mesoderm, ectoderm), which then form all somatic cell types and extraembryonic tissues (e.g., placenta) as well as germ cells of the adult. While pluripotency confers particular usefulness to hESCs, this property also presents particular challenges for the study and manipulation of these cells and their derivatives. Since a large number of cell types may be produced in differentiated hESC cultures, the efficiency of production of most cell types is low. Furthermore, successful assessment of the production of any given cell type is critically dependent on the determination of appropriate markers. Achieving high efficiency of directed differentiation is important for therapeutic applications of hESCs.

In order to use hESCs as starting material for the generation of cells that can be used in cell therapy applications, it would be beneficial to overcome the aforementioned problems. For example, to achieve the level of cellular material required for islet cell transplantation therapy, it is advantageous to target hESCs to the islet/β cell line with high efficiency at the earliest stages of differentiation.

In addition to efficient directed differentiation of the differentiation process, the isolation and characterization of intermediate cell types differentiated along the differentiation pathway into islet/β cell lines, and the use of such cells as suitable lineage precursors for other steps in differentiation, are also beneficial.

Summary of The Invention

Some embodiments of the invention relate to cell cultures comprising definitive endoderm cells, wherein the definitive endoderm cells are pluripotent cells capable of differentiating into gut tube cells or organs derived from the gut tube. According to some embodiments, the definitive endoderm cell is a mammalian cell, and in preferred embodiments, the definitive endoderm cell is a human cell. In some embodiments of the invention, definitive endoderm cells express, or do not significantly express, a particular marker. In some embodiments, definitive endoderm cells express one or more markers selected from SOX17, CXCR4, MIXL1, GATA4, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1. In other embodiments, definitive endoderm cells do not significantly express one or more markers selected from OCT4, alpha-fetoprotein (AFP), Thrombomodulin (TM), SPARC, and SOX 7.

According to other embodiments of the invention, methods of producing definitive endoderm from pluripotent cells are described. In some embodiments, the pluripotent cells are derived from morula. In some embodiments, the pluripotent stem cells are stem cells. Stem cells for use in these methods may include, but are not limited to, embryonic stem cells. Embryonic stem cells may be derived from the inner cell mass of an embryo or the gonadal ridges of an embryo. Embryonic stem cells can originate from a variety of animal species including, but not limited to, various mammalian species, including humans. In a preferred embodiment, human embryonic stem cells are used to produce definitive endoderm.

In some embodiments of the invention, one or more growth factors are used in the differentiation process from pluripotent cells to definitive endoderm cells. These one or more growth factors used in the differentiation process may include growth factors from the TGF β superfamily. In these embodiments, the one or more growth factors include Nodal/activin and/or a subpopulation of BMPs of the TGF β superfamily of growth factors. In some embodiments, the one or more growth factors are selected from Nodal, activin a, activin B, BMP4, Wnt3a, or a combination of any of these growth factors.

Embodiments of the invention also relate to cell populations enriched in definitive endoderm cells. In certain embodiments, the definitive endoderm cells are isolated or substantially purified. In some embodiments, the isolated or substantially purified definitive endoderm cells express SOX17 and/or CXRC4 markers to a greater extent than OCT4, AFP, TM, SPARC, and/or SOX7 markers.

Methods of enriching a population of cells having definitive endoderm are also provided. In some embodiments, definitive endoderm cells can be isolated or substantially purified from a mixed population of cells by contacting the cells with an agent that binds to a molecule on the surface of the definitive endoderm cells but not on the surface of other cells in the mixed population of cells, and then isolating the cells bound to the agent. In certain embodiments, the molecule located on the surface of the definitive endoderm cell is CXCR 4.

Other embodiments of the invention also relate to CXCR4 antibodies, SDF-1 ligands or other ligands of CXCR4 for obtaining definitive endoderm cells in enriched, isolated or substantially purified form. For example, a CXCR4 antibody, an SDF-1 ligand, or another ligand of CXCR4 can be used as an agent in a method such as affinity separation or magnetic separation to enrich, isolate, or substantially purify definitive endoderm cells bound to the agent.

Other embodiments of the invention described herein relate to compositions, such as cell cultures, comprising pluripotent cells and definitive endoderm cells. In certain embodiments, the cell culture comprises both stem cells and definitive endoderm cells. The number of stem cells in these cultures can be greater than, equal to, or less than the number of definitive endoderm cells in the culture. In some embodiments, the stem cell is a human embryonic stem cell. In certain embodiments, hESCs are maintained on a trophoblast. In these embodiments, the trophoblast cells may be cells such as fibroblasts obtained from humans, mice, or other suitable organisms.

In some embodiments of the invention, the composition comprising definitive endoderm cells and hESCs further comprises one or more growth factors. These growth factors may include growth factors from the TGF β superfamily. In these embodiments, the one or more growth factors include Nodal/activin and/or a subpopulation of BMPs of the TGF β superfamily of growth factors. In some embodiments, the one or more growth factors are selected from Nodal, activin a, activin B, BMP4, Wnt3a, or a combination of any of these growth factors.

Other embodiments of the invention are described with reference to the following numbered paragraphs:

1. a cell culture comprising human cells, wherein at least about 10% of said human cells are definitive endoderm cells, said definitive endoderm cells being pluripotent cells capable of differentiating into gut tube cells or organs derived from the gut tube.

2. The cell culture of paragraph 1, wherein at least about 50% of the human cells are definitive endoderm cells.

3. The cell culture of paragraph 1, wherein at least about 80% of the human cells are definitive endoderm cells.

4. The cell culture of paragraph 1, wherein the definitive endoderm cells express a marker selected from the group consisting of SOX17 and CXCR 4.

5. The cell culture of paragraph 4, wherein in the definitive endoderm cells, expression of the marker selected from the group consisting of SOX17 and CXCR4 is greater than expression of a marker selected from the group consisting of OCT4, alpha-fetoprotein (AFP), Thrombomodulin (TM), SPARC, and SOX 7.

6. The cell culture of paragraph 4, wherein the definitive endoderm cells do not express a marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

7. The cell culture of paragraph 4, wherein the definitive endoderm cells express a marker selected from the group consisting of MIXL1, GATA4, and HNF3 b.

8. The cell culture of paragraph 4, wherein the definitive endoderm cells express a marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1.

9. The cell culture of paragraph 1, wherein the definitive endoderm cells express SOX17 and CXCR 4.

10. The cell culture of paragraph 9, wherein in said definitive endoderm cells, the expression of SOX17 and CXCR4 is greater than the expression of OCT4, AFP, TM, SPARC, and SOX 7.

11. The cell culture of paragraph 9, wherein the definitive endoderm cells do not express OCT4, AFP, TM, SPARC, and SOX 7.

12. The cell culture of paragraph 9, wherein the definitive endoderm cells express MIXL1, GATA4, and HNF3 b.

13. The cell culture of paragraph 9, wherein the definitive endoderm cells express a marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1.

14. The cell culture of paragraph 1, wherein in said cell culture there are at least about 2 definitive endoderm cells for each pluripotent cell.

15. The cell culture of paragraph 14, wherein the pluripotent cells comprise embryonic stem cells.

16. The cell culture of paragraph 15, wherein the embryonic stem cells are derived from a tissue selected from the group consisting of the morula, the Inner Cell Mass (ICM) of the embryo, and the gonadal ridges of the embryo.

17. The cell culture of paragraph 1 further comprising a culture medium comprising less than about 10% serum.

18. The cell culture of paragraph 1 further comprising a growth factor of the Nodal/activin subgroup of the TGF β superfamily.

19. The cell culture of paragraph 1, further comprising a growth factor selected from Nodal, activin A, activin B, and combinations thereof.

20. A cell population comprising cells, wherein at least about 90% of said cells are human definitive endoderm cells, said human definitive endoderm cells being pluripotent cells capable of differentiating into gut tube cells or organs derived from the gut tube.

21. The cell population of paragraph 20, wherein at least about 95% of said cells are human definitive endoderm cells.

22. The cell population of paragraph 20, wherein at least about 98% of said cells are human definitive endoderm cells.

23. The cell population of paragraph 20, wherein the human definitive endoderm cells express a marker selected from SOX17 and CXCR 4.

24. The cell population of paragraph 23, wherein in the human definitive endoderm cells, the expression of the marker selected from the group consisting of SOX17 and CXCR4 is greater than the expression of a marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

25. The cell population of paragraph 23, wherein the human definitive endoderm cells do not express a marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

26. The cell population of paragraph 23, wherein the human definitive endoderm cells express a marker selected from the group consisting of MIXL1, GATA4, and HNF3 b.

27. The population of cells of paragraph 23, wherein the definitive endoderm cells express a marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1.

28. The cell population of paragraph 20, wherein the human definitive endoderm cells express SOX17 and CXCR 4.

29. The cell population of paragraph 28, wherein in said human definitive endoderm cells the expression of SOX17 and CXCR4 is greater than the expression of OCT4, AFP, TM, SPARC, and SOX 7.

30. The cell population of paragraph 28, wherein the human definitive endoderm cells do not express OCT4, AFP, TM, SPARC, and SOX 7.

31. The population of cells of paragraph 28, wherein the human definitive endoderm cells express MIXL1, GATA4, and HNF3 b.

32. The cell population of paragraph 28, wherein the definitive endoderm cells express a marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1.

33. The cell population of paragraph 20, wherein there are at least about 2 definitive endoderm cells per pluripotent cell in said cell population.

34. The population of paragraph 33, wherein the pluripotent cells comprise embryonic stem cells.

35. The population of cells of paragraph 34, wherein the embryonic stem cells are derived from a tissue selected from the group consisting of the morula, the ICM of the embryo, and the gonadal ridges of the embryo.

36. A method of producing definitive endoderm cells, the method comprising the steps of:

obtaining a population of cells comprising human pluripotent cells;

providing said cell population with at least one growth factor of the TGF β superfamily in an amount sufficient to promote differentiation of said pluripotent cells into definitive endoderm cells, said definitive endoderm cells being pluripotent cells capable of differentiating into gut tube cells or organs derived from the gut tube; and

allowing sufficient time to form definitive endoderm cells, wherein said formation of definitive endoderm cells is determined by detecting the presence of definitive endoderm cells in said cell population.

37. The method of paragraph 36, wherein at least about 10% of said pluripotent cells differentiate into definitive endoderm cells.

38. The method of paragraph 36, wherein at least about 50% of said pluripotent cells differentiate into definitive endoderm cells.

39. The method of paragraph 36, wherein at least about 70% of said pluripotent cells differentiate into definitive endoderm cells.

40. The method of paragraph 36, wherein at least about 80% of said pluripotent cells differentiate into definitive endoderm cells.

41. The method of paragraph 36, wherein detecting the presence of definitive endoderm cells in the cell population comprises detecting in the cell population cells the expression of at least one marker selected from the group consisting of SOX17 and CXCR4, and the expression of at least one marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX7, wherein in the definitive endoderm cells the expression of the marker selected from the group consisting of SOX17 and CXCR4 is greater than the expression of the marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

42. The method of paragraph 36, wherein detecting the presence of definitive endoderm cells in said cell population comprises detecting in cells of said cell population expression of at least one marker selected from the group consisting of SOX17 and CXCR4, and expression of at least one marker selected from the group consisting of AFP, TM, and SOX7, wherein in said definitive endoderm cells expression of said marker selected from the group consisting of SOX17 and CXCR4 is greater than expression of said marker selected from the group consisting of AFP, TM, and SOX 7.

43. The method of paragraph 42 wherein the expression of at least one of said markers is determined by Q-PCR.

44. The method of paragraph 42 wherein the expression of at least one of said markers is determined by immunocytochemistry.

45. The method of paragraph 36, wherein detecting the presence of definitive endoderm cells in the cell population comprises detecting expression of at least one marker selected from the group consisting of VWF, CALCR, FOXQ1, CMKOR1, and CRIP1, and expression of at least one marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX7 in cells of the cell population, wherein expression of the marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, and CRIP1 is higher in the definitive endoderm cells than in the marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

46. The method of paragraph 36, wherein said at least one growth factor is a subset of Nodal/activin of the TGF β superfamily.

47. The method of paragraph 46, wherein the at least one growth factor is selected from Nodal, activin A, activin B, and combinations thereof.

48. The method of paragraph 47, wherein said at least one growth factor is Nodal.

49. The method of paragraph 47, wherein said at least one growth factor is activin A.

50. The method of paragraph 47, wherein said at least one growth factor is activin B.

51. The method of paragraph 36 wherein a plurality of growth factors of the TGF β superfamily are provided.

52. The method of paragraph 51, wherein the plurality of growth factors comprises Nodal, activin A and activin B.

53. The method of paragraph 36, wherein the concentration of the at least one growth factor is at least about 10 ng/ml.

54. The method of paragraph 36, wherein the concentration of the at least one growth factor is at least about 100 ng/ml.

55. The method of paragraph 36, wherein the concentration of the at least one growth factor is at least about 500 ng/ml.

56. The method of paragraph 36, wherein the concentration of the at least one growth factor is at least about 1000 ng/ml.

57. The method of paragraph 36, wherein the concentration of the at least one growth factor is at least about 5000 ng/ml.

58. The method of paragraph 36, wherein the cell population is grown in a culture medium comprising less than about 10% serum.

59. The method of paragraph 36, wherein said pluripotent cells comprise stem cells.

60. The method of paragraph 59, wherein said pluripotent cells comprise embryonic stem cells.

61. The method of paragraph 60, wherein said embryonic stem cell is derived from a tissue selected from the group consisting of the morula, the ICM of the embryo, and the gonadal ridge of the embryo.

62. A definitive endoderm cell produced by the method of paragraph 36.

63. A method of producing a cell population enriched for definitive endoderm cells comprising the steps of:

differentiating cells in a population of pluripotent human cells to produce definitive endoderm cells, said definitive endoderm cells being pluripotent cells capable of differentiating into gut tube cells or organs derived from the gut tube;

providing an agent to said cell population, said agent binding to a marker expressed in said definitive endoderm cells and not substantially expressed in other cell types in said cell population; and

separating said definitive endoderm cells bound to said agent from said other cell types located in said cell population, thereby producing a cell population enriched for definitive endoderm cells.

64. The method of paragraph 63, wherein the step of differentiating further comprises,

obtaining a population of cells comprising pluripotent human cells,

providing said cell population with said at least one TGF β superfamily growth factor in an amount sufficient to promote differentiation of said pluripotent cells into definitive endoderm cells, said definitive endoderm cells being pluripotent cells that can differentiate into gut tube cells or organs derived from the gut tube and

allowing sufficient time for the formation of definitive endoderm cells, wherein said formation of definitive endoderm cells is determined by detecting the presence of definitive endoderm cells in said cell population.

65. The method of paragraph 63, wherein the detecting comprises,

detecting expression of at least one marker selected from the group consisting of SOX17 and CXCR4, and expression of at least one marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX7, in cells of said population of cells, wherein expression of a marker selected from the group consisting of SOX17 and CXCR4 is greater in said definitive endoderm cells than in the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

66. The method of paragraph 63, wherein detecting comprises detecting in cells of said cell population the expression of at least one marker selected from the group consisting of SOX17 and CXCR4, and the expression of at least one marker selected from the group consisting of AFP, TM and SOX7, wherein in said definitive endoderm cells the expression of a marker selected from the group consisting of SOX17 and CXCR4 is greater than the expression of a marker selected from the group consisting of AFP, TM and SOX 7.

67. The method of paragraph 63, wherein detecting comprises detecting expression of at least one marker selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR 1and CRIP1, and expression of at least one marker selected from the group consisting of OCT4, AFP, TM, SPARC and SOX7 in cells of the population of cells, wherein expression of the markers selected from the group consisting of FGF17, VWF, CALCR, FOXQ1, CMKOR 1and CRIP1 is higher than expression of the markers selected from the group consisting of OCT4, AFP, TM, SPARC and SOX7 in the definitive endoderm cells.

68. The method of paragraph 63, wherein at least about 95% of the cells are definitive endoderm cells.

69. The method of paragraph 63, wherein at least about 98% of the cells are definitive endoderm cells.

70. The method of paragraph 63, wherein said marker is CXCR 4.

71. The method of paragraph 63 wherein the agent is an antibody.

72. The method of paragraph 71 wherein said antibody has affinity for CXCR 4.

73. The cell population enriched for definitive endoderm cells produced by the method of paragraph 63.

74. The cell culture of any of

paragraphs

4 or 9, wherein the definitive endoderm cells do not significantly express a marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

75. The cell population of any of paragraphs 23 or 28, wherein said definitive endoderm cells do not significantly express a marker selected from the group consisting of OCT4, AFP, TM, SPARC, and SOX 7.

It will be appreciated that the methods and compositions described above relate to in vitro cell culture. However, the in vitro differentiated cell compositions described above may be used in vivo.

Other embodiments of the invention may also be found in U.S. provisional patent application No. 60/532,004, entitled definitive endoderm; U.S. provisional patent application No. 60/586,566, entitled "chemokine cell surface receptors for isolating definitive endoderm", filed 7/9/2004; and U.S. provisional patent application No. 60/587,942, entitled "cell surface receptors for isolating definitive endoderm", filed on 7/14/2004, the disclosures of which are incorporated herein by reference in their entirety.

Brief Description of Drawings

FIG. 1 is a schematic representation of the proposed differentiation pathway to generate beta-cells from hESCs. The first step in this pathway is the conversion of ES cells to the definitive endoderm lineage, which represents one of the earliest known steps to further differentiate ES cells into pancreatic endoderm, endocrine gland endoderm or islet/β -cells. Some of the factors that may be used to mediate this transformation are members of the TGF β family, including, but not limited to, activins, nodal and BMPs. Representative markers identified for definitive endoderm target cells were SOX17, GATA4, HNF3b, MIX1, and CXCR 4.

FIG. 2 is a diagram of human SOX17 cDNA showing the positions of conserved motifs and highlighting the regions of the immunization method for GENOVAC.

FIG. 3 is a correlation tree showing that SOX17 has the strongest correlation with SOX7 and weaker correlation with SOX 18. The SOX17 protein in homologous species is much more relevant than other members of the SOX group subfamily F in the same species.

FIG. 4 shows protein hybridization (Western blot) using rat anti-SOX 17 antibody as a probe. This hybridization demonstrates the specificity of the antibody for the human SOX17 protein that is overexpressed in fibroblasts (lane 1), and the lack of immunoreactivity for EGFP (lane 2) or the most relevant SOX family member SOX7 (lane 3).

FIGS. 5A-B are diagrams showing SOX17⁺Micrographs of cell clusters showing high amounts of co-labeled AFP⁺A cell (A). This is in contrast to other SOX17⁺Small amount of AFP was observed in the cell clusters (B)⁺Cellular or not observed AFP⁺The cells contrast sharply.

FIGS. 6A-C are micrographs showing body wall endoderm and SOX 17. Panel A shows immunocytochemistry for human Thrombomodulin (TM), which is localized on the cell surface of parietal endoderm cells in randomly differentiated hES cell cultures. Panel B shows the same region as Panel A for the dual TM and SOX17 logo. Panel C is a phase difference image of the same area of the nucleus labeled with DAPI. Note that the DAPI-labeled nucleus is fully associated with the SOX17 marker.

FIGS. 7A-B are histograms of anti-SOX 17 positive cells showing SOX17 gene expression and SOX17 specific antibodies by quantitative PCR (Q-PCR). Panel a shows that activin a increases SOX17 gene expression, while Retinoic Acid (RA) strongly inhibits SOX17 expression relative to undifferentiated control medium (SR 20). Panel B shows that the same pattern and degree of similarity are reflected in SOX17⁺In cell number, the Q-PCR test indicating the expression of the SOX17 gene was very sensitive to changes at the single cell level.

Figure 8A is a histogram showing that cultures of hESCs differentiated in the presence of activin a maintained low levels of AFP gene expression, whereas in 10% Fetal Bovine Serum (FBS), cells randomly differentiated, showing severe upregulation of AFP. The difference in expression level was about 7-fold.

FIGS. 8B-C are two photomicrograph images showing that activin A also inhibits AFP expression at a single cell level, relative to AFP observed in conditions of activin A treatment (bottom) with 10% FBS only (top)⁺Clusters of cells are rare and small.

FIGS. 9A-B are control charts showing the quantification of AFP using a flow cytometer⁺The number of cells. The graph shows the magnitude of the change in AFP gene expression (FIG. 8A) versus AFP in the presence (right panel) or absence (left panel) of activin A⁺The cell numbers were very consistent, further demonstrating the utility of the Q-PCR analysis to show changes that occurred at the single cell level.

FIGS. 10A-F are photomicrographs showing that exposure of hESCs to nodal, activin A and activin B (NAA) produced a significant increase in cell number over a 5 day period (A-C). By comparing SOX17⁺The relative amount of cells to the total amount of cells present in this region, as indicated by the DAPI chromogenic nucleus (D-F), shows that about 30-50% of all cells were immunoreactive to SOX17 after 5 days of NAA treatment.

FIG. 11 is a histogram showing that activin A (0, 10, 30, or 100ng/mL) dose-dependently increased SOX17 gene expression of differentiated hESCs. After the adhesion culture is treated for 3 days, the suspension culture is continued for 3-5 days, and the expression is obviously increased.

FIGS. 12A-C are histograms demonstrating the effect of activin A on MIXL1 (panel A), GATA4 (panel B), and HNF3B (panel C) expression. Dose-dependent increases in activin were also observed in the other three definitive endoderm markers MIXL1, GATA4, and HNF3 b. The magnitude of expression of dose-dependent increases in activin was very similar to that of SOX17, strongly indicating that activin A is specific for co-expression of all four genes (SOX 17)⁺,MIXL1⁺,GATA4⁺andHNF3b⁺) The cell population of (1).

FIGS. 13A-C are histograms demonstrating the effect of activin A on AFP (Panel A), SOX7 (Panel B), and SPARC (Panel C) expression. Activin a dose-dependently reduced the expression of the visceral endoderm marker AFP. The original endoderm (SOX7) and body wall endoderm (SPARC) markers remained unchanged or showed inhibition only at some points, indicating that activin a was not specific for these extra-embryonic endoderm cell types. This further supports that the increased expression of SOX17, MIXL1, GATA4 and HNF3b is due to an increase in the number of definitive endoderm cells caused by activin a.

FIGS. 14A-B are histograms showing the effect of activin A on expression of ZIC1 (panel A) and Brachyury (panel B). The sustained expression of the neural marker ZIC1 showed a dose-independent effect of activin a on neural differentiation. From the reduction in brachyury expression, it can be seen that 100ng/mL activin a treatment significantly inhibited mesodermal differentiation. This may be the result of an increase in definitive endoderm specificity from mesendoderm precursors. Low levels of activin A treatment (10 and 30ng/mL) maintained brachyury expression at the late differentiation time point compared to the untreated control blank culture.

FIGS. 15A-B are photomicrographs of activin treatment causing decreased differentiation of parietal endoderm. The finding of TM in serum-only differentiated cultures (A)^hiBody wall endoderm regions, however rarely differentiate into TM when activin (B) is included⁺The magnitude of cellular and overall immunoreactivity is low.

FIGS. 16A-D are photomicrographs showing marker expression resulting from activin A and activin B treatment. hESCs were treated with activin a and activin B for 4 consecutive days, tri-labeled with SOX17, AFP, and TM antibodies. Panel A-SOX 17; panel B-AFP; panel C-TM; and Panel D-Phase/DAPI. Note that a large number of SOX17 positive cells (A) were visible when there was no AFP (B) and TM (C) immunoreactivity at all.

Figure 17 is a micrograph showing the appearance of definitive endoderm and visceral endoderm from hESCs in vitro. Visceral endodermis area with AFP^hi/SOX17^lo/-Identified, however, definitive endoderm showed the completely opposite characteristics, SOX17^hi/AFP^lo/-. This domain was chosen because the two regions are close to each other, however, there are times when the AFP is completely separated^hiSOX17 was observed in the cell area^hi/AFP^lo/-Region, indicating partial definitive endoderm cell derivationOriginated in visceral endodermal cells.

Figure 18 is a diagram depicting TGF β family ligands and receptors. Factors that activate AR-Smads and BR-Smads favor the production of definitive endoderm from human embryonic stem cells (see, J Cell Physiol.187: 265-76).

FIG. 19 is a histogram showing the expression of SOX17 induced by treatment with single or multiple TGF- β factors over time.

FIG. 20 is a histogram showing that treatment with various TGF- β factors causes SOX17⁺The number of cells increased with time.

FIG. 21 is a histogram showing the expression of SOX17 induced over time by treatment with various TGF- β factors.

FIG. 22 is a histogram showing the dose-dependent increase of SOX17 for activin A⁺The number of cells.

Figure 23 is a histogram showing Wnt3a addition to activin a and activin B treated cultures increased SOX17 expression above the levels induced by activin a and activin B alone.

FIGS. 24A-C are histograms showing increased differentiation into definitive endoderm under low FBS conditions. Treatment of hESCs with activin a and B in media including 2% FBS (2AA) resulted in 2-3 fold higher SOX17 expression levels than treatment under the same conditions in 10% FBS (10AA) (panel a). The definitive endoderm marker MIXL1 (panel B) was also affected in the same way, and 2% FBS inhibited AFP (visceral endoderm) (panel C) more than under 10% FBS conditions.

FIGS. 25A-D are micrographs showing SOX17 in culture⁺The cells divide. SOX17 immunoreactive cells were at the hESC clonal differentiation edge (C, D), labeled with Proliferating Cell Nuclear Antigen (PCNA) (panel B), but not co-labeled with OCT4 (panel C). Furthermore, the nuclei were labeled with DAPI, at SOX17⁺Cell (arrow) and OCT4⁺Mitotic pictures are clearly visible in undifferentiated hESCs (arrows) (D).

Figure 26 is a histogram showing the relative expression levels of CXCR4 in differentiating hESCs in various media.

Figures 27A-D are histograms showing how a series of definitive endoderm markers have very similar CXCR4 expression patterns by the same treatment as figure 26.

Fig. 28A-E are histograms showing how mesodermal (BRACHYURY, MOX1), ectodermal (SOX1, ZIC1) and visceral endoderm (SOX7) markers have opposite CXCR4 expression by the same treatment as fig. 26.

FIGS. 29A-F are micrographs showing the relative differences of SOX17 immunoreactive cells under the three media conditions of FIGS. 26-28.

FIGS. 30A-C are flow cytometer spot plots demonstrating CXCR4 with increasing concentration of activin A added to differentiation media⁺The number of cells increases.

FIGS. 31A-D are histograms showing the isolated CXCR4 after high dose activin A treatment (A100-CX +)⁺The cells had an abundance of definitive endoderm markers compared to their parental cell population (a 100).

FIG. 32 is a histogram showing CXCR4 isolated using Fluorescence Activated Cell Sorter (FACS)⁺And CXCR4^-Gene expression in cell and maternal cell populations. This demonstrates CXCR4⁺The cells comprise essentially all of the CXCR4 gene expression, CXCR4, present on each maternal cell population^-Little or no CXCR4 gene expression is included.

FIGS. 33A-D are histograms demonstrating high dose activin A treated CXCR4⁺The expression of mesoderm (BRACHYURY, MOX1), ectoderm (SOX1, ZIC1) and visceral endoderm (SOX7) genes of the cells is deleted, and the expression of the non-definitive endoderm markers is inhibited.

FIGS. 34A-M are histograms showing marker gene expression patterns that can be used to identify definitive endoderm cells. Expression analysis of definitive endoderm markers FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 are shown in panels G-L, respectively. The aforementioned lineage marker genes SOX17, SOX7, SOX17/SOX7, TM, ZIC 1and MOX1 are shown in panels A-F, respectively. Panel M shows expression analysis of CXCR 4. For panels a-M, panel column for marker hESC shows purified human embryonic stem cell gene expression; 2NF shows cells treated with 2% FBS without activin; 0.1A100 shows cells treated with 0.1% FBS, 100ng/ml activin A; 1A100 shows cells treated with 1% FBS, 100ng/ml activin A; 2A100 shows cells treated with 2% FBS, 100ng/ml activin A.

Detailed Description

A key stage of early human development, the term gastrulation, occurs 2-3 weeks after fertilization. Gastrulation is extremely important because at this stage the three primary embryos are first specialized and ordered (Lu et al, 2001; Schoenwolf and Smith, 2000). The ectoderm is responsible for the formation of the outer layers of the body and the entire nervous system, however, the heart, blood, bone, skeletal muscle and other connective tissues are derived from the mesoderm. Definitive endoderm is defined as the germ layer responsible for the formation of the entire intestinal tract, including the esophagus, stomach, small and large intestine, as well as organs derived from the gut, such as the lung, liver, thymus, parathyroid and thyroid glands, gallbladder and pancreas (Grapin-Botton and Melton, 2000; Kimelman and Griffin, 2000; Tremblay et al, 2000; Wells and Melton, 1999; Wells and Melton, 2000). There is a very significant difference between definitive endoderm and the fully isolated cell line known as primitive endoderm. The primitive endoderm forms primarily the extraembryonic tissue, primarily the body wall and the visceral inner embryonic layer portion of the placental yolk sac, as well as the extracellular matrix material of Reichert's membrane.

In gastrulation, the process of definitive endoderm formation begins with a cell migration event in which mesendoderm cells (a cellular component capable of forming mesoderm or endoderm) pass through a structure known as the primitive line. Definitive endoderm is derived from cells that pass through the anterior part of the original line and the node (a specific structure located in the foremost part of the original line). When migration occurs, definitive endoderm forms the foremost part of the gut tube first until it ends when the posterior end of the gut tube is formed.

In vivo analysis of shape-setting endosymbiont formation, such as Conlon et al, 1994; feldman et al, 1998; zhou et al, 1993; aoki et al, 2002; dougan et al, 2003; tremblay et al, 2000; vincent et al, 2003; alexander et al, 1999; alexander and Stainier, 1999; kikuchi et al, 2001; studies of zebrafish and xenopus performed by Hudson et al, 1997 and mouse Kanai-Azuma et al, 2002, laid the foundation for how to use human embryonic stem cells in culture dishes to complete the development of specific germ layer cell types. Two aspects associated with the in vitro ESC culture constitute a major bottleneck in restarting development in the culture dish. First, no ordered germ layer or organ structure is produced. In differentiating hESC culture systems, most germ layer and organ specific genetic markers are expressed in heterologous forms. Thus, due to the lack of organ-specific delimitation, it is difficult to estimate the formation of specific tissues or cell types. Almost all genes expressed in cell types or tissue types of one particular germ layer are also expressed in cell types or tissue types of other germ layers. Without specific boundaries, it is difficult to assign specificity of gene expression to 1-3 gene samples. Thus, a considerable number of genes must be carefully examined, and some should also be expressed on specific cell types of organs or tissues not of interest. Second, the timing of the gene expression pattern is critical to the activity of developing along a particular channel.

As for more complex events, it should be noted that stem cell differentiation in vitro is rather asynchronous and may be much more pronounced than in vivo. Thus, one group of cells may be expressing genes associated with gastrulation, while another group may be beginning final differentiation. Moreover, treatment of hESC monolayers or Embryoid Bodies (EBs) with or without exogenous factors involvement may result in significant differences in overall gene expression patterns or differentiation status. Thus, in order to effectively follow a particular differentiation pathway, the use of exogenous factors must be time-controlled according to the gene expression pattern of the heterogeneous cell mixture. It is also beneficial to consider the morphological relationship of these cells in the culture vessel. The ability to uniformly affect hESCs in the formation of so-called embryoid bodies is far from optimal compared to the ability to grow or differentiate into monolayers and/or hESC clones in a culture medium container.

As an effective way to address the aforementioned heterogeneity and asynchrony, some embodiments of the invention contemplate combining methods of differentiating cells with methods of enriching, isolating, and/or purifying intermediate cell types in the differentiation channels.

Embodiments of the present invention relate to novel defined methods of generating definitive endoderm cells in culture by differentiating pluripotent cells, such as stem cells, into pluripotent definitive endoderm cells. As used herein, "pluripotent" or "multipotent cell" refers to a cell type that can give rise to a limited number of other specific cell types. As described above, definitive endoderm cells do not differentiate into tissues derived from ectoderm or mesoderm, but, however, differentiate into intestinal tracts and organs derived from intestinal tracts. In some preferred embodiments, the definitive endoderm cells are derived from hESCs. These methods provide a basis for efficient generation of human endoderm derived tissues such as pancreas, liver, lung, stomach, intestine and thyroid. For example, the first step in the generation of definitive endoderm differentiates stem cells into functional insulin-producing beta cells. In order to obtain useful amounts of insulin-producing beta cells, it is desirable that each differentiation step be a highly efficient differentiation step before reaching the islets/beta cells. Since the differentiation of stem cells into definitive endoderm cells may represent the initial step in the generation of functional islet/β cells (as shown in figure 1), this step of differentiation is particularly desirable for high efficiency.

In view of the need to differentiate pluripotent cells into definitive endoderm cells, aspects of the invention relate to in vitro methodologies that convert about 50-80% of pluripotent cells into definitive endoderm cells. Typically, these methods involve the use of cultures and growth factors in defined and temporally specified ways. Further enrichment of the definitive endoderm cell population can be obtained by isolating and/or purifying definitive endoderm cells from other cells in the cell population using an agent that specifically binds to definitive endoderm cells. Thus, the present invention relates to definitive endoderm cells and methods of making the cells for isolation and/or purification.

In order to determine the number of definitive endoderm cells within a cell culture or population of cells, methods are needed to distinguish such cells from other cells in the culture or population of cells. Thus, embodiments of the invention relate to cellular markers whose presence, absence, and/or relative expression levels are specific for definitive endoderm and methods of detecting and determining expression of these markers. As used herein, "expression" refers to the production of a material or substance, or the level or amount of production of a material or substance. Thus, determining the expression of a specific marker refers to detecting the relative or absolute amount of the marker expressed, or simply detecting the presence or absence of the marker. "marker" as used herein refers to any molecule that can be observed or detected. For example, a marker can include, but is not limited to, a nucleic acid, such as a transcript of a specific gene, a polypeptide product of a gene, a non-gene polypeptide product, a glycoprotein, a sugar, a glycolipid, a lipid, a lipoprotein, or a small molecule.

In some embodiments of the invention, the presence or absence and/or level of expression of a marker is determined by quantitative PCR (Q-PCR). For example, the amount of transcripts produced by some genetic markers, such as SOX17, CXCR4, OCT4, AFP, TM, SPARC, SOX7, MIXL1, GATA4, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, CRIP1, and others described herein, is determined by quantitative Q-PCR. In other embodiments, Q-PCR and immunohistochemical techniques are used to identify and determine the amount or relative proportion of these markers.

By using methods of expression of one or more suitable markers as identified above, it is possible to identify definitive endoderm cells within a cell culture or population of cells and to determine the proportion of definitive endoderm cells. For example, in some embodiments of the invention, definitive endoderm cells or cell populations are produced that express the SOX17 and/or CXCR4 gene at levels that are about 2 orders of magnitude greater than non-definitive endoderm cell types or cell populations. In other embodiments, definitive endoderm cells or cell populations are produced that express SOX17 and/or CXCR4 genes at levels that are more than 2 orders of magnitude greater than non-definitive endoderm cell types or cell populations. In still other embodiments, the definitive endoderm cells or cell populations produced express one or more markers selected from the group consisting of SOX17, CXCR4, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 on the order of about 2 or more than 2 higher than the expression of the non-definitive endoderm cell type or cell population.

The present invention relates to cell cultures comprising cell populations having significant amounts of definitive endoderm and comprising enriched definitive endoderm cells. Thus, some embodiments relate to a cell culture comprising definitive endoderm cells, wherein at least about 50-80% of the cells in the culture are definitive endoderm cells. A preferred embodiment relates to a cell culture comprising human cells, wherein at least about 50-80% of the human cells in the culture are definitive endoderm cells. Since the effectiveness of the differentiation process can be adjusted by varying several parameters, including but not limited to cell growth conditions, growth factor concentration, and time of culture step, the differentiation process described herein can convert about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, or greater than 95% of the pluripotent cells to definitive endoderm. In other embodiments of the invention, a population of pluripotent cells, such as a population of stem cells, is converted into substantially pure definitive endoderm cells.

The compositions and methods of the present invention have several useful features. For example, cell cultures and cell populations comprising definitive endoderm and methods of making such cell cultures and cell populations can be used to model early stages of human development. In addition, the compositions and methods of the present invention are also useful for the interventional treatment of diseases such as diabetes. For example, because definitive endoderm is only a limited number of tissue sources, it can be used to develop pure tissue or cell types.

Preparation of definitive endoderm from pluripotent cells

Definitive endoderm cell cultures and compositions comprising definitive endoderm cells described herein can be prepared from pluripotent cells such as embryonic stem cells. As used herein, "embryo" refers to a range of developmental stages of an organism, starting from a single fertilized egg to the end of a multicellular structure that no longer contains pluripotent or totipotent cells, except for the developing gamete cell. The term "embryo" refers to embryos derived from somatic cell nuclear transfer, except embryos derived from partner fusion. Preferred methods of deriving definitive endoderm cells utilize human embryonic stem cells as the starting material for preparing definitive endoderm. The embryonic stem cells used in the method may be cells derived from the morula, the inner cell mass of the embryo, or obtained from the gonadal ridges of the embryo. Using prior art methods, human stem cells can be maintained in a pluripotent state in culture without substantial differentiation. Such methods, for example, as disclosed in U.S. Pat. Nos. 5,670,5,690,9265,843,6,200,806, and 6,251,671, are incorporated herein by reference in their entirety.

In some embodiments of the methods of the invention, hESCs are maintained in the trophoblast. In these embodiments, any trophoblast capable of maintaining hESCs in a pluripotent state may be used in the methods of the invention. One commonly used feeder layer for culturing human embryonic stem cells is a layer of mouse fibroblasts. Recently, human fibroblast feeder layers have also been developed for culturing hESCs (see the method disclosed in U.S. patent application No. 2002/0072117, incorporated herein by reference in its entirety). Other embodiments of the methods of the invention allow for the maintenance of pluripotent hESCs without the use of trophoblasts. These methods are described in U.S. patent application No. 2003/0175956, the disclosure of which is incorporated herein by reference in its entirety.

The human embryonic stem cells of the present invention may be maintained in culture with or without serum. In some embodiments, a serum replacement is used. In other embodiments, serum-free media technology, such as that described in U.S. patent application No. 2003/0190748, the disclosure of which is incorporated herein by reference in its entirety.

Stem cells that remain pluripotent in culture are routinely passaged until differentiation into definitive endoderm. In some embodiments, differentiation to definitive endoderm is accomplished by adding a TGF β superfamily growth factor to the stem cell culture medium in an amount sufficient to promote differentiation to definitive endoderm. The TGF-beta superfamily growth factors used to prepare the definitive endoderm are selected from Nodal/activin or a subgroup of BMPs. In some embodiments of the differentiation methods of the present invention, the growth factor is selected from Nodal, activin A, activin B, and BMP 4. In addition, the growth factor Wnt3a and other Wnt family members can be used to prepare definitive endoderm cells. In some embodiments of the invention, any of the growth factors mentioned above may be used.

In some embodiments of the differentiation methods of the present invention, the above-described growth factors are added to the cells at a concentration in the culture medium sufficient to promote differentiation of at least a portion of the stem cells into shaped embryos. In some embodiments of the invention, the concentration of the above growth factor in the medium is at least about 5ng/ml, at least about 10ng/ml, at least about 25ng/ml, at least about 50ng/ml, at least about 75ng/ml, at least about 100ng/ml, at least about 200ng/ml, at least about 300ng/ml, at least about 400ng/ml, at least about 500ng/ml, at least about 1000ng/ml, at least about 2000ng/ml, at least about 3000ng/ml, at least about 4000ng/ml, at least about 5000ng/ml, or greater than 5000 ng/ml.

In some embodiments of the invention, the growth factors described above need to be removed from the cell culture after addition to the culture medium. For example, the removal time of the growth factor is about 1 day, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 8 days, about 9 days, or about 10 days after the addition. In a preferred embodiment, the removal time of the growth factor is about 4 days after the addition.

Culture of definitive endoderm cells can be in media containing little or no serum. In some embodiments of the invention, the concentration of serum ranges from about 0.05% v/v to about 20% v/v. For example, in some embodiments, the concentration of serum in the culture medium can be less than about 0.05% (v/v), less than about 0.1% (v/v), less than about 0.2% (v/v), less than about 0.3% (v/v), less than about 0.4% (v/v), less than about 0.5% (v/v), less than about 0.6% (v/v), less than about 0.7% (v/v), less than about 0.8% (v/v), less than about 0.9% (v/v), less than about 1% (v/v), less than about 2% (v/v), less than about 3% (v/v), less than about 4% (v/v), less than about 5% (v/v), less than about 6% (v/v), less than about 7% (v/v), less than about 8% (v/v), (v/v), Less than about 9% (v/v), less than about 10% (v/v), less than about 15% (v/v), or less than about 20% (v/v). In some embodiments, the definitive endoderm cells are grown under serum-free conditions. In other embodiments, definitive endoderm cells are grown in the presence of serum replacement. In other embodiments, definitive endoderm cells are grown in the presence of B27. In these embodiments, the concentration of the B27 additive ranges from about 0.2% to about 20% v/v.

Culture of hescs to definitive endoderm can be monitored by determining the expression of marker signatures of definitive endoderm. In some embodiments, the expression of some markers may be determined by the presence or absence of the marker. Alternatively, expression of some markers can be determined by measuring the level of the marker in a cell culture or cell population. In these embodiments, the expression of the marker may be determined qualitatively or quantitatively. The method of quantifying the expression marker produced by the marker gene may use quantitative PCR (Q-PCR). Methods for performing Q-PCR are known in the art. Other methods of the prior art can also be used to quantify the expression of the marker genes. For example, expression of a marker gene product can be detected by using an antibody directed against the specific marker gene product of interest. In some embodiments of the invention, the expression of marker signature genes with definitive endoderm and the expression of marker signature genes lacking significantly expressed hESCs and other cell types are determined.

As further described in the examples below, one reliable marker of definitive endoderm is the SOX17 gene. Thus, definitive endoderm cells produced by the methods of the invention express the SOX17 marker gene, thereby producing the SOX17 gene product. Other markers of definitive endoderm are MIXL1, GATA4, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1. In some embodiments of the invention, definitive endoderm cells express higher levels of the SOX17 marker gene than the SOX7 marker gene, which have the original and visceral endoderm characteristics (see table 1). Furthermore, in some embodiments, SOX17 marker gene expression is higher than OCT4 marker gene expression, which is characteristic of hESCs. In other embodiments of the invention, definitive endoderm cells express higher levels of the SOX17 marker gene than the levels of the AFP, SPARC, or Thrombomodulin (TM) marker genes. In some embodiments of the invention, definitive endoderm cells expressing SOX17 according to the methods described herein do not express significant levels or amounts of PDX1(PDX 1-negative).

Other markers of definitive endoderm are the CXCR4 gene. The CXCR4 gene encodes a cell surface chemokine receptor, the ligand of which is the chemoattractant SDF-1. The major role of CXCR4 receptor-containing cells in adults is believed to be the migration of hematopoietic cells to the bone marrow, the transport of lymphocytes, the differentiation of various B cells and the macrophage cell line [ Kim, c., and Broxmeyer, h.j. leukocyte biol.65,6-15(1999) ]. The CXCR4 receptor also acts as a co-receptor for HIV-1entry into T cells [ Feng, Y., et al. science,272,872-877(1996) ]. In a series of studies carried out [ McGrath, K.E.et al.Dev.Biology213,442-456(1999) ], the expression of the chemokine receptor CXCR4 and its unique ligand SDF-1 in adult mice and their early development [ Kim, C., andBroxmye, H., J.Leucocyte biol.65,6-15(1999) ] was described. The developmental interaction of CXCR4/SDF1 has been shown to result in late embryonic death in transgenic mice when either gene is disrupted [ Nagasawa et al Nature,382,635-638(1996) ], Ma, Q., et al Immunity,10,463-471(1999) ]. McGrath et al, using ribonuclease protection and in situ hybridization methods, demonstrated that CXCR4 is the most abundant chemokine receptor messenger RNA early in the formation of gastrulation (E7.5). At the gastrula stage, CXCR4/SDF-1 signaling appears to induce predominantly primitive line cell migration and is expressed on the definitive endoderm, mesoderm and ectoderm present at this time. In E7.2-7.8 mouse embryos, CXCR4 and alpha fetoprotein repel each other, showing a lack of expression in the visceral endoderm [ McGrath, K.E.et al. Dev. biology213,442-456(1999) ].

In some embodiments of the invention, definitive endoderm cells produced according to the methods of the invention express a CXCR4 marker gene. In other embodiments, definitive endoderm cells prepared according to the methods of the invention express the CXCR4 marker gene as well as other definitive endoderm markers including, but not limited to, SOX17, MIXL1, GATA4, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP 1. In some embodiments of the invention, definitive endoderm cells express CXCR4 marker genes at higher levels than SOX7 marker genes. Furthermore, in some embodiments, CXCR4 marker gene expression is higher than the level of OCT4 marker gene expression. In other embodiments of the invention, definitive endoderm cells express CXCR4 marker genes at levels greater than the levels of AFP, SPARC, or Thrombomodulin (TM) marker genes. In some embodiments of the invention, definitive endoderm cells expressing CXCR4 prepared according to the methods described herein do not express significant levels or amounts of PDX1(PDX 1-negative).

It is understood that expression of CXCR4 in endodermal cells does not exclude expression of SOX 17. Accordingly, in some embodiments of the invention, definitive endoderm cells express higher levels of both the SOX17 and CXCR4 marker genes than the SOX7 marker gene. Furthermore, in some embodiments, the expression of both SOX17 and CXCR4 marker genes is higher than the expression of OCT4 marker gene. In other embodiments of the invention, definitive endoderm cells express higher levels of SOX17 and CXCR4 marker genes than do AFP, SPARC, or Thrombomodulin (TM) marker genes. In some embodiments of the invention, definitive endoderm cells expressing SOX17/CXCR4 prepared according to the methods described herein do not express significant levels or amounts of PDX1(PDX 1-negative).

It will be appreciated that depending on the differentiation conditions, different levels of SOX17 and/or CXCR4 marker expression are induced in definitive endoderm cells. Thus, in some embodiments of the invention, the expression of a SOX17 marker and/or CXCR4 marker in a definitive endoderm cell or cell population is at least about 2-fold to at least about 10,000-fold greater than the expression of a SOX17 marker and/or CXCR4 marker in a non-definitive endoderm cell or cell population, such as a pluripotent stem cell. In other embodiments of the invention, the expression of a SOX17 marker and/or a CXCR4 marker in a definitive endoderm cell or cell population is at least about 4 fold, at least about 6 fold, at least about 8 fold, at least about 10 fold, at least about 15 fold, at least about 20 fold, at least about 40 fold, at least about 80 fold, at least about 100 fold, at least about 150 fold, at least about 200 fold, at least about 500 fold, at least about 750 fold, at least about 1000 fold, at least about 2500 fold, at least about 5000 fold, at least about 7500 fold, or at least about 10,000 fold greater than the expression of a SOX17 marker and/or a CXCR4 marker in a non-definitive endoderm cell or cell population, such as a pluripotent stem cell. In some embodiments, the expression of a SOX17 marker and/or CXCR4 marker in a definitive endoderm cell or cell population is, without limitation, higher than the expression of a SOX17 marker and/or CXCR4 marker in a non-definitive endoderm cell or cell population, such as a pluripotent stem cell.

It will be appreciated that in some embodiments of the invention, expression of the GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 markers is increased in a definitive endoderm cell or population of cells as compared to expression of the GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 markers in a non-definitive endoderm cell or population of cells.

It will also be appreciated that the expression levels of the SOX17 marker in definitive endoderm cells differ from the OCT4, SPARC, AFP, TM and/or SOX7 marker expression levels. Similarly, the expression level of the CXCR4 marker in definitive endoderm cells differed from the OCT4, SPARC, AFP, TM, and/or SOX7 marker expression levels. Similarly, in some embodiments of the invention, the expression of the SOX17 marker or CXCR4 marker is at least about 2-fold to at least about 10,000-fold greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers. In other embodiments of the invention, the expression of the SOX17 marker or CXCR4 marker is at least about 4 fold, at least about 6 fold, at least about 8 fold, at least about 10 fold, at least about 15 fold, at least about 20 fold, at least about 40 fold, at least about 80 fold, at least about 100 fold, at least about 150 fold, at least about 200 fold, at least about 500 fold, at least about 750 fold, at least about 1000 fold, at least about 2500 fold, at least about 5000 fold, at least about 7500 fold, or at least about 10,000 fold greater than the expression of the OCT4, SPARC, AFP, TM, and/or SOX7 marker. In some embodiments, OCT4, SPARC, AFP, TM, and/or SOX7 markers are not significantly expressed in definitive endoderm cells.

It will be appreciated that in some embodiments of the invention, expression of a marker selected from GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR 1and CRIP1 is increased in definitive endoderm cells as compared to OCT4, SPARC, AFP, TM and/or SOX 7.

Compositions comprising definitive endoderm

Aspects of the invention relate to compositions, such as cell populations and cell cultures, comprising pluripotent cells, such as stem cells, and definitive endoderm cells. For example, using the methods described herein, compositions comprising mixtures of hESCs and definitive endoderm cells can be prepared. In some embodiments, the prepared composition comprises at least about 5 definitive endoderm cells per 95 pluripotent cells. In other embodiments, at least about 95 definitive endoderm cells are included per 5 pluripotent cells included. In addition, compositions include other ratios of definitive endoderm cells to pluripotent cells. For example, in a composition, at least about 1 definitive endoderm cell per 1,000,000 pluripotent cells, at least about 1 definitive endoderm cell per 100,000 pluripotent cells, at least about 1 definitive endoderm cell per 10,000 pluripotent cells, at least about 1 definitive endoderm cell per 1,000 pluripotent cells, at least about 1 definitive endoderm cell per 500 pluripotent cells, at least about 1 definitive endoderm cell per 100 pluripotent cells, at least about 1 definitive endoderm cell per 10 pluripotent cells, at least about 1 definitive endoderm cell per 5 pluripotent cells, at least about 1 definitive endoderm cell per 2 pluripotent cells, a composition of matter comprising a mixture of at least about 1 definitive endoderm cell per 1,000,000 pluripotent cells, a composition of matter comprising at least about 1 definitive endoderm cell per 1, a composition of matter comprising at least about 1 definitive endoderm cell per 100 pluripotent cells, a composition of matter comprising at least about 1 definitive endoderm cell per 2 pluripotent cells, at least about 2 definitive endoderm cells per 1 pluripotent cell included, at least about 5 definitive endoderm cells per 1 pluripotent cell included, at least about 10 definitive endoderm cells per 1 pluripotent cell included, at least about 20 definitive endoderm cells per 1 pluripotent cell included, at least about 50 definitive endoderm cells per 1 pluripotent cell included, at least about 100 definitive endoderm cells per 1 pluripotent cell included, at least about 1,000 definitive endoderm cells per 1 pluripotent cell included, at least about 10,000 definitive endoderm cells per 1 pluripotent cell included, at least about 100,000 definitive endoderm cells per 1 pluripotent cell included, at least about 1,000 definitive endoderm cells per 1 pluripotent cell included, and at least about 1,000,000 definitive endoderm cells per 1 pluripotent cell included. In some embodiments of the invention, the pluripotent cells are human pluripotent stem cells. In some embodiments, the stem cell is derived from the morula, the inner cell mass of an embryo, or the gonadal ridges of an embryo. In some other embodiments, the pluripotent cells are derived from gonadal or reproductive tissue of a multicellular structure that has undergone embryonic stage development.

Some aspects of the invention relate to a cell culture or cell population comprising at least about 5% definitive endoderm cells to at least about 95% definitive endoderm cells. In some embodiments, the cell culture or population of cells comprises mammalian cells. In a preferred embodiment, the cell culture or population of cells comprises human cells. For example, some embodiments relate to cell cultures, including human cells, wherein at least about 5% to at least about 95% of the human cells are definitive endoderm cells. Other embodiments of the invention relate to cell cultures comprising human cells wherein at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%, at least about 40%, at least about 45%, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, or greater than 90% of the human cells are definitive endoderm cells.

Other embodiments of the invention relate to compositions, such as cell cultures or cell populations, comprising human cells, such as human definitive endoderm cells, wherein the expression of SOX17 or CXCR4 marker is greater than the expression of OCT4, SPARC, Alpha Fetoprotein (AFP), Thrombomodulin (TM), and/or SOX7 marker in at least about 5% of the human cells. In other embodiments, wherein the expression of SOX17 or CXCR4 markers is greater than the expression of each of SOX17 or SOX 4 markers in at least about 10% human cells, at least about 15% human cells, at least about 20% human cells, at least about 25% human cells, at least about 30% human cells, at least about 35% human cells, at least about 40% human cells, at least about 45% human cells, at least about 50% human cells, at least about 55% human cells, at least about 60% human cells, at least about 65% human cells, at least about 70% human cells, at least about 75% human cells, at least about 80% human cells, at least about 85% human cells, at least about 90% human cells, at least about 95% human cells, or greater than 95% human cells than the expression of SOX 4, SPARC, AFP, TM, and/or SOX7 markers.

It is to be understood that some embodiments of the invention relate to compositions, such as cell cultures or cell populations, comprising human cells, such as human definitive endoderm cells, wherein the expression of one or more markers selected from GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 is greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers in at least about 5% to greater than at least about 95% of the human cells.

Other embodiments of the invention relate to compositions, such as cell cultures or cell populations, comprising human cells, such as human definitive endoderm cells, wherein the expression of both the SOX17 and CXCR4 markers is greater than OCT4, SPARC, AFP, TM, and/or SOX7 marker expression in at least about 5% of the human cells. In other embodiments, the expression of both the SOX17 and CXCR4 markers is greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers in at least about 10% human cells, at least about 15% human cells, at least about 20% human cells, at least about 25% human cells, at least about 30% human cells, at least about 40% human cells, at least about 50% human cells, at least about 55% human cells, at least about 60% human cells, at least about 65% human cells, at least about 70% human cells, at least about 75% human cells, at least about 80% human cells, at least about 85% human cells, at least about 90% human cells, at least about 95% human cells, or greater than 95% human cells.

It is to be understood that some embodiments of the invention relate to compositions, such as cell cultures or cell populations, comprising human cells, such as human definitive endoderm cells, in at least about 5% to greater than at least about 95% of the human cells, wherein the expression of the markers GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 is greater than the expression of the OCT4, SPARC, AFP, TM, and/or SOX7 markers.

Other embodiments of the invention relate to compositions, such as cell cultures or cell populations, including mammalian endoderm cells, such as human definitive endoderm cells, wherein the expression of SOX17 or CXCR4 marker is greater than the expression of OCT4, SPARC, Alpha Fetoprotein (AFP), Thrombomodulin (TM), and/or SOX7 marker in at least about 5% of the endoderm cells. In other embodiments, wherein in at least about 10% endoderm cells, at least about 15% endoderm cells, at least about 20% endoderm cells, at least about 25% endoderm cells, at least about 30% endoderm cells, at least about 35% endoderm cells, at least about 40% endoderm cells, at least about 45% endoderm cells, at least about 50% endoderm cells, at least about 55% endoderm cells, at least about 60% endoderm cells, at least about 65% endoderm cells, at least about 70% endoderm cells, at least about 75% endoderm cells, at least about 80% endoderm cells, at least about 85% endoderm cells, at least about 90% endoderm cells, at least about 95% endoderm cells, or greater than 95% endoderm cells, the expression of SOX17 or CXCR4 markers were all greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers.

It is to be understood that some embodiments of the present invention relate to compositions, such as cell cultures or cell populations, comprising mammalian endoderm cells, wherein expression of one or more markers selected from GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 is greater than expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers in at least about 5% to greater than at least about 95% of the endoderm cells.

Still further embodiments of the invention relate to compositions such as cell cultures or cell populations comprising mammalian endoderm cells such as human definitive endoderm cells wherein the expression of both SOX17 and CXCR4 markers is greater than the expression of OCT4, SPARC, Alpha Fetoprotein (AFP), Thrombomodulin (TM), and/or SOX7 markers in at least about 5% of said endoderm cells. In other embodiments, wherein in at least about 10% endoderm cells, at least about 15% endoderm cells, at least about 20% endoderm cells, at least about 25% endoderm cells, at least about 30% endoderm cells, at least about 35% endoderm cells, at least about 40% endoderm cells, at least about 45% endoderm cells, at least about 50% endoderm cells, at least about 55% endoderm cells, at least about 60% endoderm cells, at least about 65% endoderm cells, at least about 70% endoderm cells, at least about 75% endoderm cells, at least about 80% endoderm cells, at least about 85% endoderm cells, at least about 90% endoderm cells, at least about 95% endoderm cells, or greater than 95% endoderm cells, the expression of both SOX17 and CXCR4 markers was greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers.

It is to be understood that some embodiments of the invention relate to compositions, such as cell cultures or cell populations, including mammalian endoderm cells, wherein the expression of markers of GATA4, MIXL1, HNF3b, GSC, FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 is greater than the expression of OCT4, SPARC, AFP, TM, and/or SOX7 markers in at least about 5% to greater than at least about 95% of the endoderm cells.

Using the methods described herein, compositions comprising definitive endoderm cells substantially free of other cell types can be prepared. With respect to cells in a cell culture or cell population, the term "substantially free" refers to the absence of the particular cell in the cell culture or cell population, or less than about 5% of the total number of cells in the cell culture or cell population. In some embodiments of the invention, a population or culture of definitive endoderm cells prepared using the methods described herein is substantially free of cells that specifically and significantly express OCT4, SOX7, AFP, SPARC, TM, ZIC1, or BRACH marker genes.

In one embodiment of the invention, definitive endoderm cells are described as high in SOX17, high in MIXL1, low in AFP, low in SPARC, low in thrombomodulin, low in SOX7, high in CXCR4, based on the expression of the marker genes.

Enrichment, isolation and/or purification of definitive endoderm

With respect to other aspects of the invention, definitive endoderm cells can be enriched, isolated, and/or purified using affinity tags specific for these cells. Examples of affinity labels specific for definitive endoderm cells are antibodies, ligands, or other binding reagents specific for marker molecules, such as polypeptides, that are present on the surface of the cells definitive endoderm cells but are substantially absent from other cell types found in cell culture media prepared according to the methods of the invention. In some embodiments, antibodies that bind CXCR4 are used for affinity labeling for enrichment, isolation, and/or purification of definitive endoderm. In other embodiments, the chemokine SDF-1 or other SDF-1-based molecules can also be used for affinity labeling. These molecules include, but are not limited to, SDF-1 fragments, SDF-1 fusions or SDF-1 mimetics.

Methods for preparing antibodies and for cell isolation are known in the art and can be performed using the antibodies and cells of the invention. In one embodiment, antibodies that bind CXCR4 are bound to magnetic beads and then bound to definitive endoderm cells in cell culture media, which after enzyme treatment reduces cell-to-cell and substrate adhesion. The cell/antibody/magnetic bead complexes are exposed to a moving magnetic field to separate the magnetic bead-bound definitive endoderm cells from the unbound cells. Once definitive endoderm cells are physically separated from other cells in culture, the bound antibodies are destroyed and the cells are replaced in appropriate tissue culture media.

Other methods of enrichment, isolation and/or purification of cultures or populations of definitive endoderm cells are contemplated by embodiments of the invention. For example, in some embodiments, CXCR4 antibodies are incubated in cell culture comprising definitive endoderm, which when treated reduce cell-to-cell and substrate adhesion. The cells were then washed, centrifuged and resuspended. The cell suspension is then incubated with a second antibody, such as a FITC conjugated antibody capable of binding to the first antibody. The cells were then washed, centrifuged and resuspended in buffer. The cell suspension was then analyzed and sorted using a Fluorescence Activated Cell Sorter (FACS). Converting CXCR4^-Separating negative cells, collecting CXCR4^-Positive cells, thereby isolating cells of this type. If desired, the isolated cell compositions can be further sorted several times using the same or different markers specific for definitive endoderm using alternative affinity methods.

In other embodiments of the invention, definitive endoderm is enriched, isolated and/or purified using a ligand or other molecule that binds to CXCR 4. In some embodiments, the molecule is SDF-1 or a fragment, fusion, or mimetic thereof.

In preferred embodiments, definitive endoderm cells are enriched, isolated, and/or purified from other non-definitive endoderm cells after a stem cell culture is induced to differentiate towards a definitive endoderm. It is to be understood that the above enrichment, isolation and/or purification methods may utilize such cultures at any stage of differentiation.

In addition to the methods described above, definitive endoderm cells can also be isolated by other cell separation techniques. In addition, definitive endoderm cells can also be enriched for or isolated by a series of subculture under growth conditions which favor selective survival or selective expansion of said definitive endoderm cells.

Using the methods described herein, definitive endoderm cells can be enriched, isolated, and/or purified in vitro from a pluripotent cell culture or cell population, such as a stem cell culture or cell population, that has undergone at least some differentiation. In some embodiments, the cells are randomly differentiated. However, in some preferred embodiments, the cells differentiate predominantly into definitive endoderm. Some preferred enrichment, isolation and/or purification methods involve the preparation of definitive endoderm in vitro from human embryonic stem cells. Using the methods described herein, the population of cells or cell culture enriched in definitive endoderm is at least about 2-fold to about 1000-fold as compared to an untreated population of cells or cell culture. In some embodiments, the enriched definitive endoderm is at least about 5-fold to about 500-fold compared to an untreated cell population or cell culture. In other embodiments, the enriched definitive endoderm is at least about 10-fold to about 200-fold compared to an untreated cell population or cell culture. In other embodiments, the enriched definitive endoderm is at least about 20-fold to about 100-fold compared to an untreated cell population or cell culture. In other embodiments, the enriched definitive endoderm is at least about 40-fold to about 80-fold compared to an untreated cell population or cell culture. In certain embodiments, the enriched definitive endoderm is at least about 2-fold to about 20-fold compared to an untreated cell population or cell culture.

Having generally described the invention, a further understanding can be obtained by reference to certain specific examples which are provided herein for purposes of illustration only and are not intended to be limiting.

Examples

The following examples describe the use of pluripotent human cells. Methods for preparing pluripotent human cells are well known to those of ordinary skill in the art and numerous scientific publications, including U.S. Pat. Nos. 5,453,357, 5,670,372, 5,690,926, 6,090,622, 6,200,806, and 6,251,671, and U.S. patent application publication No. 2004/0229350, all of which are incorporated herein by reference in their entirety, are described.

Example 1

Human ES cells

To study endodermal development, we utilized human embryonic stem cells, which are pluripotent stem cells that appear to divide indefinitely in culture while maintaining a normal karyotype. ES cells are derived from the inner cell mass of a 5-day-old embryo and isolated using immunological or mechanical methods. In particular, the human embryonic stem cell line hESCyt-25 is from an extra frozen embryo of the in vitro fertilization cycle with patient consent. After thawing, the hatched blastocysts were seeded on Mouse Embryonic Fibroblasts (MEFs) using ES medium (DMEM, 20% FBS, non-essential amino acids, β -mercaptoethanol, ITS modulators). Approximately two weeks later, embryos were attached to culture dishes and the regions of undifferentiated hESCs were transferred to MEFs. The transfer is done by mechanical cleavage, simple digestion with neutral protease, mechanical removal of cell clusters, washing, reseeding. Since derivatization, hESCyt-25100 serial passages. We used the hESCyt-25 human embryonic stem cell line as the starting material to prepare definitive endoderm.

It will be appreciated by those of ordinary skill in the art that stem cells or other pluripotent cells may also be used as starting materials for the differentiation methods described herein. For example, embryonic gonadal ridge cells may be isolated according to methods known in the art for use as pluripotent cell starting material.

Example 2

Characterization of hESCyt-25

Human embryonic stem cell lines maintain normal morphological, karyotypic, growth and self-renewal properties for 18 months in culture. This cell line showed strong immunoreactivity to OCT4, SSEA-4 and TRA-1-60 antigens, which are characteristic of undifferentiated hESCs, and showed the same basic phosphate activity and morphology as other established hESC lines. Furthermore, Embryoid Bodies (EBs) are also readily formed when the human stem cell line hESCyt-25 is cultured in suspension. Because of its pluripotent nature, hESCyT-25 differentiates into different cell types representing the three major germ layers. The production of ectoderm was confirmed by Q-PCR detection of ZIC 1and Immunocytochemistry (ICC) detection of nestin and more mature neural markers. Immunocytochemical staining of beta-III microtubules can be observed in elongated cell clusters, characteristic of early neurons. Previously, we treated EBs in a suspension containing retinoic acid to induce differentiation of pluripotent stem cells into an extra-embryonic Visceral Endoderm (VE). After 54 hours of treatment, the treated cells expressed high levels of Alpha Fetoprotein (AFP), two VE markers of SOX 7. Immunocytochemical staining showed sporadic sheets of AFP expressed by cells differentiated as a monolayer. As described below, the hESCyT-25 cell line also formed definitive endoderm, confirmed by real-time quantitative polymerase chain reaction (Q-PCR) and immunocytochemistry detecting SOX17, without AFP expression. To confirm differentiation into mesoderm, differentiating EBs were analyzed at several time points to detect short tail (Brachyury) gene expression. During the course of the experiment, the expression of the brachyury gene was progressively increased. As previously described, hESCyT-25 is pluripotent and is capable of forming cells representing three germ layers.

Example 3

Preparation of SOX17 antibody

The major bottleneck in identifying definitive endoderm in hESC culture is the lack of adequate tools. We thus prepared antibodies against SOX17 protein.

The marker SOX17 was expressed throughout the definitive endoderm as it formed during gastrulation, and its expression remained in the gut tube (despite the difference in expression level along the a-P axis) until the organ began to form. SOX17 was also expressed on a series of extraembryonic endoderm cells. This protein is not expressed in mesoderm or ectoderm. When combinations with other markers exclude extraembryonic lineages, SOX17 has now been found to be a suitable marker for the definitive endoderm lineage.

As detailed herein, in order to produce SOX17 positive definitive endoderm cells, SOX17 antibody was used to specifically test the effect of various treatments and differentiation methods. Other antibodies reactive with AFP, SPARC, and thrombomodulin are also used to exclude the production of visceral and parietal endoderm (extraembryonic endoderm).

For the preparation of antibodies against SOX17, a part of the human SOX17 cDNA (SEQ ID NO:1) corresponding to the carboxy-terminal amino acid 172-414(SEQ NO:2) of the SOX17 protein (FIG. 2) was used for the genetic immunization of rats according to the method developed by the antibody preparation company GENOVAC (Freiberg, Germany). Methods of genetic immunization can be found in U.S. Pat. Nos. 5,830,876, 5,817,637, 6,165,993, and 6,261,281, and International patent application publication Nos. WO 00/29442 and WO99/13915, the disclosures of which are incorporated herein by reference.

Other methods of genetic immunization can also be found in the non-patent literature. For example, monoclonal antibodies prepared according to genetic immunization as described by Barry et al, Biotechniques 16:616-620,1994, the disclosure of which is incorporated herein by reference in its entirety. Specific methods for preparing antibodies against specific proteins based on genetic immunization, e.g., Costaglia et al (1998) genetic immunization against the human thyrotropin receptor causes thyroiditis and preparation of monoclonal antibodies recognizing the autoreceptor, J.Immunol.160: 1458-1465; DNA-based immunization delivered by the Kilparick et al (1998) Gene gun mediated the rapid preparation of murine monoclonal antibodies to the Flt-3 receptor, Hybridoma 17: 569-576; schmolke et al, (1998), hepatitis G virion J, Virol.72:4541-4545 was recognized by E2-specific monoclonal antibodies generated by DNA immunization in human serum; krasemann et al, (1999) use of non-traditional nucleic acid immunization strategies to generate monoclonal antibodies against proteins, J.73: 119-129; and Ulivieri et al (1996) DNA immunization to generate monoclonal antibodies to a defined portion of the H.pylori vacuolating toxin, 51:191-194, the disclosure of which is incorporated herein by reference in its entirety.

As shown in the relationship tree of FIG. 3, in the Sox family, SOX7 and SOX18 are the closest to SOX 17. We used human SOX7 polypeptide as a negative control to confirm that SOX17 antibody is specific for SOX17 and does not react with the most closely related family members. In particular, to demonstrate that antibodies generated by genetic immunization are specific to SOX17, SOX7 and other proteins were expressed on human fibroblasts, and then the total reactivity with SOX17 antibodies was analyzed by Western blot and ICC. For example, expression vectors for SOX17, SOX7, and EGFP were prepared by the following method, transfected into human fibroblasts, and analyzed by Western blot. The expression vectors used to prepare SOX17, SOX7, and EGFP were pCMV6(OriGene Technologies, Inc., Rockville, Md.), pCMV-SPORT6(Invitrogen, Carlsbad, Calif.), and pEGFP-N1(Clonetech, Palo Alto, Calif.), respectively. To prepare the protein, telomerase immortalized MDX human fibroblasts (Invitrogen, Carlsbad, CA) were transiently transfected with supercoiled DNA using Lipofectamine 2000. 36 hours after transfection, total cell lysates were collected IN 50mM TRIS-HCl (pH 8), 150mM NaCl, 0.1% SDS, 0.5% deoxycholic acid, and some protease inhibitors (Roche Diagnostics Corporation, Indianapolis, IN). Western blot analysis of 100. mu.g of cellular proteins, separated by SDS-PAGE on NuPAGE (4-12% gradient polyacrylamide, Invitrogen, Carlsbad, Calif.), transferred by electroblotting onto PDVF membranes (Hercules, Calif.), probed with murine SOX17 antiserum diluted to 1/1000 in 10mM TRIS-HCl (pH 8), 150mM NaCl, 10% BSA, 0.05% Tween-20(Sigma, St. Louis, Mo.), and then treated with anti-rat IgG binding alkaline phosphatase (Jackson ImmunoResearce Laboratories, West Grove, Pa.), with results shown by Vector Black alkaline phosphatase staining (Vector Laboratories, Burlingame, Calif.). The protein size standards used were a wide range of color markers (Sigma, st.

In FIG. 4, Western blot was performed using an SOX17 antibody as a probe on protein extracts from human fibroblasts transiently transfected with SOX17, SOX7, or EGFP cDNA. The protein extracted from hSOX 17-transfected cells alone produced a band of approximately 51Kda that best matched the predicted 46Kda molecular weight of human SOX17 protein. The SOX17 antibody was not reactive against extracts from human SOX7 or EGFP transfected cells. In addition, the SOX17 antibody clearly marked the human fibroblast nuclei transfected with hSOX17 expression construct, but not cells transfected with EGFP alone. Likewise, the SOX17 antibody showed specificity of detection by ICC.

Example 4

Confirmation of SOX17 antibody as definitive endoderm marker

Partially differentiated hESCs were co-labeled with SOX17 and AFP antibodies based on the specificity of the SOX17 antibody for the human SOX17 protein and further labeling definitive endoderm. SOX17, SOX7, and AFP were each shown to be expressed in the visceral endoderm, and SOX7 is a closely related member of the SOX gene family F subgroup (fig. 3). However, the expression of AFP and SOX7 in definitive endoderm cells is not at a level detectable by ICC, and therefore, it can be used as a negative marker for bonifide definitive endoderm cells. The SOX17 antibody marker cell population was shown to be present as a dispersed cell population, or mixed with AFP positive cells. In particular, FIG. 5A shows a small number of SOX17 cells co-marked with AFP; however, some areas were also found, among which SOX17⁺With little or no AFP in the cellular field⁺Cells (fig. 5B). Similarly, because it has also been reported that the parietal endoderm expresses SOX17, an antibody co-labeled with SOX17 can be used together with the parietal markers SPARC and/or Thrombomodulin (TM) to identify SOX17 of the parietal endoderm⁺A cell. As shown in FIGS. 6A-C, parietal endoderm cells, common to thrombomodulin and SOX17, were produced by stochastic differentiation of hES cells.

According to the cell marker test, the characteristics of the definitive endoderm cells can be expressed by a marker expression profile SOX17^hi/AFP^lo/[TM^loOr SPARC^lo]And (4) establishing. In other words, the expression of the SOX17 marker was higher than the expression of the AFP marker, which is a characteristic of visceral endoderm, and the TM or SPARC marker, which is a characteristic of parietal endoderm. Thus, those cells that are positive for SOX17, negative for AFP and negative for TM or SPARC are definitive endoderm.

SOX17^hi/AFP^lo/TM^lo/SPARC^loThe specificity of the marker expression profile can be used as further evidence for definitive endoderm prediction, quantitatively comparing SOX17 and AFP gene expression to the relative number of antibody-labeled cells. As shown in fig. 7A, hESCs treated with retinoic acid (visceral endoderm inducer) or activin a (definitive endoderm inducer) resulted in a 10-fold difference in SOX17 mRNA expression levels. This result reflects a 10-fold difference in the number of cells labeled with SOX17 antibody (fig. 7B). Furthermore, as shown in fig. 8A, activin a treatment of hESCs inhibited 6.8-fold expression of the AFP gene compared to no treatment. This change is visually reflected by the dramatic decrease in the number of AFP-labeled cells in these cultures, as shown in FIGS. 8B-C. For further quantification, a reduction of approximately 7-fold in AFP gene expression was demonstrated as a result of approximately 7-fold reduction in the number of AFP antibody-labeled cells as determined by flow cytometry (FIGS. 9A-B). This result is very important, indicating that the quantitative changes in gene expression observed in Q-PCR reflect typical changes in cell type specificity observed by antibody staining.

hESCs cultured in the presence of Nodal family members (Nodal, activin A and activin B-NAA) resulted in a significant increase in cells marked with the SOX17 antibody over time. After 5 consecutive days of activin treatment, more than 50% of the cells were labeled with SOX17 (FIGS. 10A-F). After 5 days of activin treatment, few cells were marked with AFP.

In conclusion, the 242 amino acid antibody at the carbon end of the anti-human SOX17 protein was generated, and the human SOX17 protein was identified in Western blots but no SOX7 was recognized, which is the close relative of the recent SOX family. The SOX17 antibody identified a subpopulation of cells in differentiated hESC cultures, which was predominantly SOX17⁺/AFP^lo/-(more than 95% of marker cells) and a small amount of (C)<5%) SOX17, AFP co-tagged cells (visceral endoderm). Treatment of hESC cultures with activin produced a significant increase in SOX17 gene expression and SOX17 labeled cells, and significantly inhibited the expression of AFP mRNA and the number of cells labeled with AFP antibodies.

Example 5

Q-PCR Gene expression analysis

In the experiments described below, real-time quantitative RT-PCR (Q-PCR) was the primary method used to screen various treatments for hESC differentiation. In particular, gene expression was measured in real time by Q-PCR at multiple time points and multiple marker genes were analyzed. Marker gene signatures of both desired and undesired cell types are evaluated to gain a better understanding of the overall dynamics of the cell population. The strengths of Q-PCR analysis include its extreme sensitivity and the relative ease with which genomic sequences can be exploited to develop the necessary markers. Furthermore, the extremely high sensitivity of Q-PCR allows the detection of gene expression in relatively small numbers of cells in a larger population. Furthermore, the ability to detect very low levels of gene expression provides an indication of "propensity to differentiate" within a population. Before significant differentiation of these cell phenotypes, their propensity to particular differentiation pathways could not be recognized by immunocytochemistry techniques. Thus, Q-PCR provides an assay that is at least complementary to, and potentially superior to, immunocytoactivity techniques for screening for success in differentiation processes. In addition, Q-PCR provides a mechanism to evaluate the success of differentiation methods by quantitative analysis on a half-high-throughput scale.

The method used here is relatively quantitative, using SYBR Green chemistry and a two-step RT-PCR procedure on a Rotor Gene 3000instrument (Corbett research). This method allows the storage of cDNA samples for analysis of future additional marker genes, thus avoiding variability in reverse transcription efficiency between samples.

If possible, the designed primers are located at exon-exon boundaries or span at least 800bp of introns, as this empirically eliminates amplification of contaminating genomic DNA. When a marker gene containing no intron or no pseudogene is used, DNase I treatment of the RNA sample is performed.

We generally use Q-PCR to determine gene expression of multiple markers for both target and non-target cell types to provide a broad expression profile describing gene expression in a cell sample. Markers associated with the early stage of hESC differentiation (specifically, ectodermal, mesodermal, definitive endoderm and extraembryonic endoderm) and available validation primers are provided in table 1 below. Human specificity of these primer sets has also been demonstrated. This fact is important because hESCs are typically grown on mouse trophoblasts. Most commonly, 3 samples are taken for each condition and analyzed twice independently to assess biological variability associated with each quantitative assay.

To prepare PCR templates, total RNA was isolated using RNeasy (Qiagen) and quantified using RiboGreen (molecular probes). Reverse transcription of 350-500ng total RNA was accomplished using the iScript reverse transcription kit (BioRad) which contains a mixture of oligo dT and random primers. Each 20. mu.L reaction was then diluted to a total volume of 100. mu.L, and 3. mu.L was used for each 10. mu. L Q-PCR reaction containing 400nM forward and reverse primers and 5. mu.L of 2X SYBR Green master mix (Qiagen). Two-step cycling parameters were selected, denaturation at 85-94 ℃ for 5 seconds (specific selection based on melting temperature of amplicons of each primer set), and annealing/extension at 60 ℃ for 45 seconds. Fluorescence data was collected during the last 15 seconds of each extension phase. Three points of the 10-fold dilution series were used to generate a standard curve for each round, and the cycle threshold (Ct's) was converted to a quantitative value based on the standard curve. Values for each sample were calibrated for housekeeping gene characterization, and then the mean and standard deviation of the three samples were calculated. At the end of the PCR cycle, the specificity of the reaction was determined by melting curve analysis. The single specific product is shown at the appropriate T for the PCR amplicon_mAt a single peak. In addition, reactions without reverse transcriptase were used as negative controls and were not amplified.

The first step in establishing Q-PCR methodology is to identify the appropriate Housekeeping Genes (HGs) in the assay system. Since HG is used for inter-sample RNA input, RNA integrity and RT efficiency calibration, for calibration to be meaningful, HG constant expression levels over time in all sample types is valuable. We determined the expression levels of Cyclophilin G, hypoxanthine phosphoribosyltransferase 1(HPRT), beta-2-microglobulin (microroglobulin), hydroxymethybian synthetase (HMBS), TATA-binding protein (TBP) and glucoronidase beta (GUS) in differentiating hESCs. Our results show that β -2-microglobulin expression levels are increased during differentiation, and therefore we exclude this gene from being used for calibration. Other gene expression levels were consistent with time and throughout the treatment. We generally used Cyclophilin G and GUS simultaneously to calculate calibration coefficients for all samples. The simultaneous use of multiple HGs reduces variations inherent in the calibration process, increasing the reliability of the relative gene expression values.

After obtaining the genes for calibration, Q-PCR was used to examine samples treated with different assays to determine the relative gene expression levels of a number of marker genes. The marker genes selected for use are groups of genes that are enriched in a representative specific population of early germ layers, particularly those differentially expressed in definitive endoderm and extraembryonic endoderm. These genes and their associated enrichment characteristics are highlighted in table 1.

TABLE 1

Because many genes are expressed in more than one germ layer, it is useful to quantitatively compare the expression levels of many genes in the same assay. SOX17 is expressed in definitive endoderm and to a lesser extent in visceral and parietal endoderm. SOX7 and AFP were expressed in the endoderm of the inner organ at early developmental time points. SPARC and TM are expressed in the parietal endoderm and Brachyury is expressed in the early mesoderm.

Definitive endoderm cells are predicted to express SOX17 mRNA at high levels, AFP and SOX7 (visceral endoderm), SPARC (parietal endoderm) and Brachyury (mesoderm) at low levels. In addition, the present invention uses ZIC1 to further rule out induction of early ectoderm. Finally, GATA4 and HNF3b were expressed simultaneously in definitive and extraembryonic endoderm and, therefore, correlated with SOX17 expression in definitive endoderm (table 1). Representative experiments are shown in FIGS. 11-14, demonstrating how the marker genes described in Table 1 correlate with each other with each sample, thus emphasizing the specific differentiation patterns of definitive endoderm, extraembryonic endoderm, mesoderm, and neural cell types.

The data clearly show that increased doses of activin resulted in increased expression of SOX17 gene. Further, the SOX17 expression represented primarily definitive endoderm, but not the opposite extraembryonic endoderm. The observation concluded that SOX17 gene expression was inversely correlated with AFP, SOX7 and SPARC.

Example 6

Directed differentiation of human ES cells into definitive endoderm

If a human ES cell culture is cultured without actively maintaining its undifferentiated state, the culture will differentiate randomly. The differential differentiation results in the formation of extraembryonic endoderm cells, which include both the body wall and the visceral endoderm (AFP, SPARC and SOX7 expression), as well as ectodermal, mesodermal derivatives marked by expression of ZIC1, Nestin (ectodermal) and Brachyury (mesodermal). In ES cell cultures, definitive endoderm cell appearance has not been detected or determined by traditional methods due to the lack of specific antibody markers. Likewise, the production of early definitive endoderm from cultures of ES cells has not been carefully investigated due to lack of means. The vast majority of confirmations have focused on ectoderm and extraembryonic endoderm due to the lack of good, available antibody reagents for definitive endoderm cells. In summary, in random differentiated ES cell cultures, the cells were cultured with SOX17^hiDefinitive endoderm cells are produced in significantly greater numbers of extra-embryonic and neuroectodermal cell types than definitive endoderm cells.

When undifferentiated hESC clones were expanded in fibroblast feeder, the edges of the clones exhibited different morphological characteristics than the cells inside the clones. Many limbal cells could be distinguished by their large heterogeneous cell size and high level expression of OCT 4. It has been described that when ES cells begin to differentiate, their level of expression of OCT4 is altered up or down relative to the level of undifferentiated ES cells. The initial differentiation state, which shows a departure from the pluripotent state, was altered up or down relative to the OCT4 threshold level for undifferentiated cells.

When undifferentiated clones were detected immunocytochemically with SOX17, small clusters of 10-15 cells of SOX17 positive cells were occasionally randomly detected around and at the junctions of undifferentiated ESC clones. As described above, when the clone volume expanded so as to become crowded, the outer edge of the clone distributed pockets appeared to be the first cells to differentiate from the classical ESC morphology. The older, less voluminous undifferentiated clones (<1 mm; 4-5 day size) within and at the margins of the clones were free of SOX17 positive cells, whereas older, more voluminous clones (1-2mm diameter, >5 day size) within the perimeter and at the margins of some clones contained sporadic, free, flaked SOX17 positive, AFP negative cells, differentiated from the classical hESC morphology as described above. Given that this was the first development of an effective SOX17 antibody, definitive endoderm cells generated in this early "undifferentiated" ESC culture have not been previously demonstrated.

Based on the negative correlation of SOX17 and SPARC gene expression levels as determined by Q-PCR, the vast majority of SOX 17-positive, AFP-negative cells will be negative for antibody co-labeled body wall markers. As shown in FIGS. 15A-B, specific confirmation was obtained in the TM-expressing parietal endoderm cells. Contact with Nodal factor activins a and B resulted in a dramatic decrease in TM expression intensity and the number of TM positive cells. SOX17 positive cells that were also negative for AFP and TM were observed on activin-treated medium by triple labeling with SOX17, AFP and TM antibodies (FIGS. 16A-D). These were the first cells to confirm SOX17 positive definitive endoderm cells on cultures of differentiated ESCs (FIGS. 16A-D and 17).

Using the SOX17 antibody and Q-PCR tools described above, we have developed a series of tools that can effectively program ESCs into SOX17^hi/AFP^lo/SPARC/TM^loMethods of definitive endoderm cells. We applied a series of differentiation methods aimed at increasing the number and proliferative capacity of these cells, using Q-PCR to detect SOX17 gene expression at the population level and SOX17 protein antibody labeling on individual cells.

We first analyzed and described the effects of TGF-beta family growth factors, such as Nodal/activin/BMP, on the creation of definitive endoderm cells from embryonic stem cells in vitro cell culture. In a typical experiment, we added activin A, activin B, BMP, or a combination thereof, to an undifferentiated human stem cell line hESCyt-25 culture to begin the differentiation process.

As shown in FIG. 19, addition of 100ng/ml concentration of activin A for differentiation for 4 days induced 19-fold SOX17 gene expression compared to undifferentiated hESCs. Addition of activin a with activin a second member of the activin family activin B induced 37-fold SOX17 gene expression compared to undifferentiated hESCs by 4 days of combined activin treatment. Addition of BMP4, a third member of the Nodal/activin and BMP subgroups of the TGF β family, along with activin a and activin B, induced 57-fold SOX17 gene expression compared to undifferentiated hESCs (fig. 19). When SOX17 was induced with activin and BMP, differentiation for 4 days resulted in 5-,10-, and 15-fold induction compared to the no factor culture control. SOX17 induced more than 70-fold higher than hESCs by 5 days of triple treatment with activin A, B and BMP. These data show that higher doses of the Nodal/activin TGF β family member, longer treatment times, resulted in increased SOX17 expression.

Nodal and related activin A, B and BMP molecules promote the expression of SOX17, the formation of definitive endoderm in vivo and in vitro. In addition, the addition of BMP caused an increase in SOX17 induction, possibly through further induction of Nodal co-receptor Cripto.

We have demonstrated that the combined use of activin A, B and BMP4 results in an SOX 17-induced increase and thus formation of a definitive endoderm. In combination with activins A and B, chronic addition of BMP4(>4 days) induced an increase in SOX17 in the conductor walls and in the visceral and definitive endoderm. Thus, in some embodiments of the invention, it is important that BMP4 be removed within 4 days of the addition treatment.

To determine the effect of TGF β treatment at the single cell level, the effect of TGF β addition over a time course was tested using SOX17 antibody labeling. As shown in FIGS. 10A-F, the relative number of SOX 17-labeled cells dramatically increased over time. Relative quantification (FIG. 20) showed a more than 20-fold increase in SOX 17-labeled cells. The results indicate that both cell number and gene expression levels increase with increased time of TGF β factor exposure. As shown in fig. 21, SOX17 induced levels of 168-fold higher than undifferentiated hESCs after 4 days of exposure to Nodal, activin a, activin B, and BMP 4. Figure 22 shows that the number of SOX17 positive cells is also dose dependent. Activin A at doses of 100ng/mL or higher was able to strongly induce gene expression of SOX17 and an increase in cell number.

In addition to TGF β family members, Wnt family molecules may play a role in definitive endoderm specificity and/or maintenance. Sample SOX17 gene expression using activin + Wnt3a increased compared to activin alone, indicating that Wnt molecules were also beneficial for differentiation of hESCs into definitive endoderm (figure 23).

All the above experiments were performed in tissue culture medium containing 10% serum and added factors. Surprisingly, we found that serum concentrations in the presence of added activin had an effect on SOX17 expression levels, as shown in FIGS. 24A-C. When serum levels were reduced from 10% to 2%, the expression of SOX17 increased 3-fold in the presence of activin a and B.

Finally, we demonstrated that activin induces SOX17⁺Cells divide in culture as shown in FIGS. 25A-D. The arrows show that cells labeled with SOX17/PCNA/DAPI are in mitosis, as evidenced by the PCNA/DAPI-labeled mitotic plate pattern differing in mitotic characteristics in time.

Example 7

Expression of chemokine receptor 4(CXCR4) is associated with definitive endoderm markers and with mesoderm, ectoderm or mesoderm Visceral endoderm markers not related

As described above, ESCs can be induced to differentiate into definitive endoderm layers by using cytokines from the TGF β family and the more specific activin/nodal subfamily. Furthermore, we have shown that the proportion of Fetal Bovine Serum (FBS) in differentiation medium affects the efficiency of differentiation of definitive endoderm from ESCs. This effect is that higher levels of FBS, at a given concentration of activin a in the medium, will inhibit its maximum differentiation to definitive endoderm. In the absence of exogenous activin a, the differentiation efficiency of ESCs into definitive endoderm lineage was very low, and FBS concentration had a weak effect on the differentiation process of ESCs.

In these experiments, hESCs were grown for 6 days in RPMI medium (Invitrogen, Carlsbad, CA; cat #61870-036) supplemented with 0.5%, 2.0% or 10% FBS, with or without 100ng/mL activin A. In addition, a gradient of 0.5% to 2.0% FBS was also used in combination with 100ng/mL activin A on the first three days of differentiation. After 6 days, replicate samples were collected from each culture condition for relative gene expression analysis by real-time quantitative PCR. The remaining cells were mixed and the SOX17 protein was detected by immunofluorescence.

The expression level of CXCR4 differed greatly among the 7 culture conditions used (fig. 26). Generally, CXCR4 expression was high in activin a treated media (a100) and low in media without exogenous activin a (nf). Furthermore, CXCR4 expression was highest in a100 treated medium when FBS concentration was lowest. Under 10% FBS conditions, CXCR4 levels were significantly reduced so that relative expression was more consistent with conditions without activin a (nf).

As described above, the expression of the SOX17, GSC, MIXL1, and HNF3 β genes is consistent with the characteristics of definitive endoderm cells. The relative expression of these 4 genes under 7 differentiation conditions mapped the expression of CXCR4 (fig. 27A-D). This also confirms that CXCR4 is also a definitive endoderm marker.

Ectodermal and mesodermal lineages can be distinguished from definitive endoderm by various markers of their expression. Early mesoderm expresses Brachyury and MOX1 genes, whereas primary neuroectoderm expresses SOX 1and ZIC 1. FIGS. 28A-D demonstrate that the cultures without exogenous activin A favoured mesodermal and ectodermal gene expression, and that 10% FBS conditions also increased the level of mesodermal and ectodermal marker expression in the activin-treated cultures. These expression patterns are in contrast to the CXCR4 pattern, indicating that CXCR4 is not highly expressed in germ or ectoderm derived from ESCs during this developmental time.

Differentiation to extra-embryonic lineages also occurs early in mammalian development. Differentiation of visceral endoderm is of specific relevance here, with the same expression of many genes common to definitive endoderm, including SOX 17. To distinguish definitive endoderm from extra-embryonic visceral endoderm, two different markers should be detected. SOX7 represents a marker expressed on the visceral endoderm, but not the definitive endoderm lineage. Thus, culture conditions showing strong SOX17 gene expression under conditions without SOX7 expression might include definitive endoderm, not visceral endoderm. As shown in fig. 28E, SOX7 was highly expressed in the absence of activin a medium, and SOX7 was expressed in increased amounts when FBS included 10% even in the presence of activin a. This pattern is in contrast to CXCR4 expression patterns, indicating that CXCR4 is not highly expressed in visceral endoderm.

The immunoreactivity of SOX17 was also examined under each of the above differentiation conditions (SOX 17)⁺) Relative number of cells. SOX17 when hESCs differentiated at high dose activin A and low FBS concentrations (0.5% -2.0%)⁺Cells are distributed throughout the culture. SOX17 when high dose activin A was used and FBS concentration was 10% (v/v)⁺The frequency of cell appearance was reduced, often in isolated clusters, rather than being evenly distributed in the culture (fig. 29A, C, B and E). When used without exogenous activin A, SOX17 was found⁺The cells were further decreased. Under these conditions, SOX17⁺Cells also appeared in clusters, but these clusters were smaller and higher activin a, less with low FBS treatment (fig. 29C and F). These results indicate that the CXCR4 expression pattern is consistent not only with definitive endoderm gene expression under various conditions, but also with the number of definitive endoderm cells.

Example 8

Differentiation conditions enriched for definitive endoderm increase the proportion of CXCR4 positive cells

The dose of activin a also affected the efficiency with which definitive endoderm was derived from ESCs. This example increased activin a dosing increased CXCR4⁺Proportion of cells in culture.

hESCs were differentiated in RPMI medium supplemented with 0.5% -2% FBS (gradually increasing from 0.5% to 1.0% to 2.0% on the first 3 days of differentiation) and 0, 10 or 100ng/mL activin a. After 7 days of differentiation, the cells were maintained in the absence of Ca²⁺/Mg² ⁺And in PBS containing 2% FBS and 2mM (EDTA) for 5 minutes at room temperature. Cells were filtered through a 35 μm nylon filter, counted and pelleted. The pellet was resuspended in 50% human serum/50% normal donkey serum and incubated on ice for 2 minutes to block non-specific antibody binding sites. To each 50. mu.L (containing about 10)⁵Individual cells) suspension was added 1 μ L of mouse anti-CXCR 4 antibody (Abcam, cat # ab10403-100) and labeled on ice for 45 minutes. Cells were washed and pelleted with 5mL PBS containing 2% human serum (buffer). After another 5mL buffer wash, the cells were washed with another 50. mu.L buffer/10⁵Cell concentration suspension. Secondary antibody (conjugated FITC donkey anti-mouse antibody; Jackson Immuno Research, cat # 715-. Cells were then washed at 5X10⁶The cells/mL were suspended in The buffer, analyzed and sorted by The flow cytometer instrument operator (The Scripps Research Institute) using a FACS Vantage (Beckton Dickenson). Cells were collected directly in RLT lysis buffer (Qiagen) for subsequent total RNA isolation and gene expression analysis by real-time quantitative PCR.

As the dose of activin A in differentiation media (FIGS. 30A-C) was increased, flow cytometry-determined CXCR4 was observed⁺Cells also proliferated (FIGS. 30A-C). CXCR4⁺Cells fell into the R4 gate, which used only the secondary antibody as a control, with 0.2% of this control event in the R4 gate. CXCR4 when dose of activin A is increased⁺The dramatic increase in cell number correlated with a significant increase in definitive endoderm gene expression (FIGS. 31A-D).

Example 9

Enriching and separating CXCR4 positive cells for definitive endoderm gene expression, and removing expressed mesoderm, ectoderm and viscera Endoderm marker cells

Collection of CXCR4 identified in example 8 above⁺And CXCR4^-Cells were analyzed for relative gene expression, while the gene expression of the maternal cell population was determined.

CXCR4 when dose of activin A is increased⁺The relative levels of gene expression sharply increased (fig. 32). This is in conjunction with a dose-dependent increase in CXCR4 by activin a⁺Cell correlation was good (FIGS. 30A-C). It is also clear that CXCR4 was isolated from each cell population⁺The cells occupy almost all of the CXCR4 gene expressing cells in the cell population. This confirms that FACS method collects these cellsEfficiency. Gene expression analysis shows that CXCR4⁺Cells include not only the majority of CXCR4 gene expression, but also gene expression of other definitive endoderm markers. As shown in FIGS. 31A-D, the maternal A100 cell population from SOX17, GSC, HNF3B, and MIXL1 was further enriched for CXCR4⁺A cell. In addition, CXCR4^-Some include very little expression of these definitive endoderm marker genes. Also, CXCR4⁺And CXCR4^-The cell populations showed the opposite pattern of mesodermal, ectodermal and extraembryonic endodermal marker gene expression. FIGS. 33A-D show the removal of CXCR4 relative to A100 maternal cell population⁺Cells were expressed with Brachyury, MOX1, ZIC1, and SOX7 genes. The A100 maternal cell population expressed these markers already at low levels relative to conditions with low or no activin A. These results indicate that CXCR4 isolated from hESCs in the presence of high doses of activin A under differentiation conditions⁺The cells are obtained as highly enriched substantially pure definitive endoderm cells.

Example 10

Quantification of definitive endoderm cells in a cell population using CXCR4

To quantify the proportion of definitive endoderm cells in cell cultures or cell populations, cells expressing CXCR4 and other definitive endoderm markers were analyzed by FACS according to the methods described previously or in U.S. provisional patent application No. 60/532,004, entitled "definitive endoderm", filed 12/23/2003, the disclosure of which is incorporated herein by reference in its entirety.

Using methods such as those described in the examples above, hESCs were differentiated to produce definitive endoderm. In particular, the productivity and purity of the differentiated cell culture are increased, and the serum concentration in the culture medium is strictly controlled as follows: 0.2% FBS on day one, 1.0% FBS on day two, and 2.0% FBS on days 3-6. The differentiated cultures were sorted by FACS using three cell surface antigenic determinants, E-Cadherin (Cadherin), CXCR4 and thrombomodulin. The cell populations were then sorted by Q-PCR analysis to determine the relative expression levels of markers for definitive endoderm, extraembryonic endoderm and other cell types. Sorting cells from CXCR4 obtained from optimally differentiated cultures yielded isolation of definitive endoderm cells > 98% pure.

Table 2 shows the results of definitive endoderm culture marker assays differentiated from hESCs using the methods described herein

TABLE 2

Composition of definitive endoderm cultures

In particular, table 2 shows CXCR4 and SOX17 positive cells (endoderm) comprise 70% -80% of cells in cell culture. In these cells expressing SOX17, less than 2% expressed TM (parietal endoderm) and less than 1% expressed AFP (visceral endoderm). When TM-positive and AFP-positive cells (combination of body wall and visceral endoderm; total 3%) were subtracted from the ratio SOX17/CXCR 4-positive cells, about 67% -77% of the cell cultures were found to be definitive endoderm. Approximately 10% of the cells were E-cadherin (ECAD) positive, which is a marker of hESCs, and approximately 10-20% of the cells were of other cell types.

We also found that the purity of definitive endoderm in differentiated cell cultures can be increased prior to FACS isolation compared to the aforementioned low serum method in which FBS concentrations were maintained at < 0.5% throughout the 5-6 day differentiation procedure. However, maintaining cell culture concentrations at ≦ 0.5% throughout the 5-6 day differentiation procedure also resulted in a reduction in the total number of definitive endoderm cells produced.

Definitive endoderm cells prepared according to the methods of the invention remain expanded in culture in the presence of activin for more than 50 days without significant differentiation. In these cases, the expression of SOX17, CXCR4, MIXL1, GATA4, HNF3 β was maintained during the culture. In addition, no TM, SPARC, OCT4, AFP, SOX7, ZIC1, and BRACH were detected in these cultures. It is possible to maintain definitive endoderm cells expanded in culture in the presence of activin for substantially more than 50 days without significant differentiation.

Example 11

Other markers of definitive endoderm cells

In the following experiments, RNA was isolated from purified populations of definitive endoderm and human embryonic stem cells. RNA from each purified cell population was then analyzed on a gene chip. Q-PCR was used to further examine the potential of genes expressed by definitive endoderm, but not embryonic stem cells, as markers for definitive endoderm.

Human embryonic stem cells (hESCs) were maintained in DMEM/F12 medium supplemented with 20% KnockOut serum replacement, 4ng/mL recombinant human basic fibroblast growth factor (bFGF), 0.1mM 2-mercaptoethanol, L-glutamic acid, non-essential amino acids, and penicillin/streptomycin. hESCs were differentiated to definitive endoderm by culturing for 5 days in RPMI medium supplemented with 100ng/mL recombinant human activin A, Fetal Bovine Serum (FBS) and penicillin/streptomycin. The daily concentration changes of FBS were 0.1% (day one), 0.2% (day two) and 2% (days 3-5).

Gene expression analysis was performed to obtain a pure population of cells hESCs and definitive endoderm, which were separated by Fluorescence Activated Cell Sorting (FACS). Use of SSEA4 antigen (R)&D Systems, cat # FAB1435P) immunopurification of hESCs using CXCR4 (R)&D Systems, cat # FAB170P) purified definitive endoderm. Cell dissociation was blocked for non-specific binding using trypsin/EDTA (Invitrogen, cat #25300-054), washed with Phosphate Buffered Saline (PBS) containing 2% human serum, resuspended in 100% human serum and placed on ice for 10 min. Add 200. mu.L of phycoerythrin-binding antibody to 800. mu.L of human serum at 5X10⁶In cells, staining was performed for 30 min on ice. Cells were washed twice with 8mL PBS buffer and resuspended in 1mL PBS. FACS isolation was performed using FACS Vantage (BD Biosciences) with core equipment of the Scripps institute. Cells were collected directly in RLT lysis buffer and RNA was isolated as RNeasy according to the instructions (Qiagen).

Purified RNA was sampled twice (Durham, NC) and expression profiling data generated using a U133Plus2.0 high density oligonucleotide array of the Affymetrix platform. The data presented is a set of comparisons that identify genes differentially expressed by two cell populations of hESCs versus definitive endoderm. Genes with strongly elevated changes in expression levels compared to those found by hESCs were selected as new candidate markers with a high degree of definitive endoderm characteristics. Selected genes were assayed using Q-PCR according to the described method to verify gene expression changes found on the gene chip and to examine the expression pattern of these genes over the time course of hESC differentiation.

FIGS. 34A-M show the results of some marker gene expression. Cell cultures were analyzed on

days

1,3 and 5 after addition of 100ng/ml activin A, showing the results of definitive endoderm cells expressing CXCR4 and Human Embryonic Stem Cells (HESC) at the end of 5 days of differentiation procedures (CXDE). Comparison of fig. 34C and G-M shows that the six marker genes FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1, show nearly identical expression patterns to each other, which are also identical to CXCR4 and SOX17/SOX7 expression patterns. As described above, SOX17 was expressed in both definitive endoderm and in embryonic endoderm expressing SOX 7. Since SOX7 is not expressed in definitive endoderm, the SOX17/SOX7 ratio reliably estimates the contribution of SOX17 expression in definitive endoderm indicated by the whole population. The similarity of panels G-L and M to panel C shows that FGF17, VWF, CALCR, FOXQ1, CMKOR1, and CRIP1 are likely markers of definitive endoderm and indicate their insignificant expression in extraembryonic endoderm cells.

It is understood that the Q-PCR results described herein may be further corroborated with ICC.

The methods, compositions and apparatus described herein are representative of preferred embodiments, are exemplary, and are not to be construed as limiting the scope of the invention. Variations and other uses will occur to those skilled in the art which are encompassed within the spirit of the invention and are defined by the scope of the disclosure. It is therefore evident that those skilled in the art may now make alterations and modifications to the invention disclosed herein without departing from the scope and spirit of the invention.

As set forth in the claims and the disclosure below, the phrase "consisting essentially of … …" is meant to include any of the elements listed after the phrase and is limited to those other elements that do not interfere with or contribute to the activity or effect described in the disclosure. Thus, the phrase "consisting essentially of … …" indicates that the listed ingredients are required or necessary, but that other ingredients are optional, with or without the inclusion of an optional ingredient, depending on whether the activity or effect of the listed ingredients is affected.

Reference to the literature

A number of documents and patent references are cited in the present application, each of which is incorporated herein by reference in its entirety.

Some references are already cited in the text in their entirety. Some other references cited in the text, which are cited by author and year only, are incorporated in their entirety as follows:

alexander, j., Rothenberg, m., Henry, g.l., and Stainier, D.Y (1999) Casanova systems an early and essential role in end role formation in zebrafish Dev Biol 215, 343-.

Alexander, J. and Stainier, D.Y. (1999). A molecular pathway leading to endoderm formation in zebrafish. Curr Biol 9, 1147-.

Aoki, t.o., Mathieu, j., Saint-etianne, l., Rebagliati, m.r., peyreeras, n, and Rosa, F.M (2002), Regulation of nodal and sensory formation by tar-a, a tgfbeA-related type I receptor (tamam-a, a TGF beA-related type I receptor modulating nodal signaling system and mesogenic formation), Dev Biol 241, 273-.

Beck, s., Le Good, j.a., Guzman, m., Ben Haim, n., Roy, k., Beermann, f., and Constam, D.B (2002). Extra-embroynic proteins regulating Nodal signaling during gastrulation. Nat Cell Biol 4, 981-.

Beddington, R.S., Rashbass, P. and Wilson, V. (1992) Brachyury-a gene expression and early organogenesis (Brachyury-a gene affecting mouse gastrulation and early organogenesis) Dev Suppl,157 + 165.

Bongso, A., Fong, C.Y., Ng, S.C., and Ratnam, S. (1994). Isolation and culture of inner cell mass cells from human blastula cells, Hum Reprod 9,2110-2117).

Chang, H., Brown, C.W., and Matzuk, M.M (2002). Genetic analysis of the mammalian transforming growth factor-beta superfamily, Endocr Rev 23, 787-.

Conlon, F.L., Lyons, K.M., Takaesu, N., Barth, K.S., Kisprt, A., Herrmann, B, and Robertson, E.J, (1994). A primary requirement for nodal in the formation and main of the primary strand in the motor (primary requirement for formation and maintenance of mouse origin), Development 120, 1919-.

Dougan, S.T., Warga, R.M., Kane, D.A., Schier, A.F., and Talbot, W.S (2003.) The role of The zebrafish nodal-related genes and loops in determining mesendoderm (The role of zebrafish nodal-related genes and loops in determining mesendoderm), Development 130,1837-1851.

Feldman, B., Gates, M.A., Egan, E.S., Dougan, S.T., Rennebeck, G., Sirotkin, H.I., Schier, A.F., and Talbot, W.S (1998) Zebraf orange definition and gem-layer formation require nodal-related signals for development and germ layer formation Nature 395,181-185.

Feng, Y., Broder, C.C., Kennedy, P.E., and Berger, E.A. (1996). HIV-1entry factor: functional cDNA cloning of a seven transmembrane G-protein coupled receptor (HIV-1 entry cofactor). Science 272,872-877.

Futaki, S.A., Hayashi, Y.A., Yamashita, M.A., Yagi, K.A., Bono, H.A., Hayashizaki, Y.A., Okazaki, Y.A., and Sekiguchi, K.2003. Molecular weights of connective production of basic membrane components: Gene expression profiles of engelbretth-hole-switch fabrics and F9 expression of systemic cancer cells (Molecular basis of constituent products of basement membrane components: engelbretth-hole-switch tumors and F9 embryonic tumor cell Gene expression profiles). J.l Chem.

Grapin-Botton, A. and Melton, D.A. (2000). endo derm degradation: from patterning to organogenesis. Trends Genet 16, 124. 130.

Harris, T.M. and Childs, G. (2002). Global gene expression patterns of F9 embedded cardiovascular cells inter partial end-meter (whole gene expression pattern during differentiation of F9 embryonic tumor cells into parietal endoderm) Funct Integr Genomics 2, 105. entoderm 119.

Hog B.L. (1996) Bone morphogenetic proteins in depletion, Curr Opin Genet Dev 6, 432-.

Hog, B.L, (1997). Pluripotent embryo cells and methods of making same (U.S. A., Vanderbilt University).

Howe, c.c., overturn, g.c., Sawicki, j., Solter, d., Stein, p. and Strickland, s. (1988) Expression of SPARC/osteonectin transcription in mice embryos and gonads.

Hudson, C., Clements, D., Friday, R.V., Stott, D., and Woodland, H.R (1997). Xsox17alpha and-beta medium endoderm formation in Xenopus (Xsox 17. alpha. and. beta. mediate formation of African toad endoderm). Cell 91, 397-.

Imada, M., Imada, S., Iwasaki, H., Kume, A., Yamaguchi, H., and Moore, E.E, (1987) Fetomodulin: marker surface protein of fetal development simulated by periodic AMP. Dev Biol 122,483 491.

Kanai-Azuma, M.A., Kanai, Y.A., Gad, J.M., Tajima, Y.A., Taya, C.A., Kurohmaru, M.A., Sanai, Y.A., Yonekawa, H.A., Yazaki, K.A., Tam, P.P., and Hayashi, Y.A. (2002) a deletion of a defined gut end-effector in Sox17-null mutant microorganism (deletion of the definitive gut of Sox 17) Development 129,2367-2379.

Katoh, M. (2002). Expression of human SOX7 in normal tissues and tumors (Expression of human SOX7 in normal tissues and tumors). Int J Mol Med 9, 363-.

Kikuchi, Y, Agathon, A, Alexander, J, Thisse, C, Waldron, S, Yelon, D, Thisse, B, and Stainier, D.Y, (2001) casanova encodes a novel Sox-related protein processing and subsequent for early end expression in zebrafish (casanova encodes a new Sox-related protein sufficiently necessary for early endoderm formation of zebrafish early embryo layer), Genes Dev 15,1493-1505.

Kim, C.H. and Broxmeyer, H.E. (1999). Chemokines: signal lamps for influencing of T and B cells for influencing and influencing functions.J.Leukoc Biol 65,6-15.

Kimelman, D. and Griffin, K.J. (2000). Vertebrate mesendoderm induction and patterning. Curr Opin Genet Dev 10, 350-.

Kubo A, Shinozaki K, Shannon JM, Kouskoff V, Kennedy M, Woo S, Fehling HJ, Keller G. (2004) Development of a defined end-effector from engineered stem cells in culture (Development of definitive embryonic stem cell definitive endoderm) development.131,1651-62.

Kumar, A., Novoselov, V., Celeste, A.J., Wolfman, N.M., ten Dijke, P., and Kuehn, M.R, (2001). Nodal signaling uses and transforming growth factor-beta receptor-modulated Smads (Nodal signaling uses activin and transforming growth factor-beta receptor modulated ds). J Biol Chem 276,656 661.

Labosky, P.A., Barlow, D.P., and Hogan, B.L (1994 a.) Embryonic germcell lines and the derivative from mouse primordial cells Ciba Foundation Symp 182,157 one 168; discission 168-.

Labosky, P.A., Bar, D.P., and Hogan, B.L (1994b), Mouse embryo culture (EG) cell lines: transmission through the culture and differentiation in the methylation imprints of insulin-like growth factor 2receptor (Igf2r) gene compounded with insulin-like growth factor (ES) cell lines (Mouse Embryonic Germ (EG) cell lines: germline transmission and difference in methylation blots of insulin-like growth factor 2receptor (Igf2r) gene compared to Embryonic Stem (ES) cell lines) — Development 120,3197 3204.

Wickert, H., Kutsch, S., Kanzler, B., Tamai, Y., Taketo, M.M., and Kemler, R. (2002). Formation of multiple hearts in the embryonic endoderm mice.Dev Cell 3, 171-.

Lu, C.C., Brennan, J. and Robertson, E.J (2001). From fertilization to transformation: axis formation in the mouse embryo Dev 11, 384-year 392.

Ma, Q., Jones, D., and Springer, T.A (1999). The chemokine receptor CXCR4 is required for The permission of B lines and granules precursors with The bone marrow microenvironment (chemokine receptor CXCR4 is required to retain B lineage and granulocytic precursors in The bone marrow microenvironment). Immunity 10,463-471.

McGrath KE, Koniski AD, Maltby KM, McGann JK, Palis J. (1999) immunological expression and function of the chemokine SDF-1and its receptor, CXCR4 (chemokine SDF-1and its receptor, CXCR4 Embryonic expression and function). Dev biol.213,442-56.

Miyazono, K., Kusanagi, K., and Inoue, H. (2001) Divergene and conversion of TGF-beta/BMP signaling (irradiation and aggregation of TGF-beta/BMP signaling) J Cell Physiol 187, 265-.

Nagasawa, t.s., Hirota, s.a., Tachibana, k.a., Takakura, n.a., Nishikawa, s.a., Kitamura, y.a., Yoshida, n.a., Kikutani, h. and Kishimoto, t.1996. Defects of B-cell lymphopoiesis and bone-marrow myogenesis in mice lacking the CXC chemokine PBSF/SDF-1 (lacking CXC chemokine PBSF/SDF-1 mouse B-cell lymphopoiesis and myelopoiesis 635) Nature 382, 638.

Niwa, H. (2001) Molecular mechanism to main stem Cell connecting of ES cells (Molecular mechanism for maintaining ES Cell stem Cell regeneration), Cell Structure Funct 26,137-148.

Ogura, H., Aruga, J. and Mikoshiba, K. (2001). Behavior antigens of Zic 1and Zic2 mutant mice: abnormalities as models for human neurological disorders (Behavioral abnormalities in Zic 1and Zic2 mutant mice: model of underlying human neurological disease). Behav Genet 31, 317-.

Reubinoff, B.E., Pera, M.F., Fong, C.Y., Trousnon, A. and Bongso, A. (2000). Embrnic stem cell lines from human blasts: physiological differentiation in vitro [ Nat Biotechnol 18, 399-.

Rodaway, a. and parent, R. (2001). mesendoderm.an extrinsic germ layer (mesendoderm.an ancient germ layer)? Cell 105,169-172.

Rodaway, A., Takeda, H., Koshida, S., Broadbent, J., Price, B., Smith, J.C., Patient, R. and Holder, N. (1999) the Induction of the sensory under the zebrafish germ ring by yolk-derived TGF-beta family signals and differentiation of media and end by FGF. Development 126,3067 + 3078.

Rohr, K.B., Schulte-Merker, S. and Tautz, D. (1999). Zebraphis zic1expression in brains and transmitters is infected by BMP and hedgehog signaling (BMP and hedgehog signaling system affects the expression of Zebrafish zic1 in brain and body segments). Mech Dev 85, 147-.

Schier, A.F, (2003). Nodal signaling in verterblate reduction (Nodal signaling system in vertebrate development). Annu Rev cells Dev Biol 19, 589-.

Schoenwolf, G.C. and Smith, J.L. (2000). stimulation and early media patterning in vertebrates (patterned development of vertebrate gastral and early germ layers). Methods Mol Biol 135, 113-.

Shamblott, M.J., Axelman, J., Wang, S., Bugg, E.M., Littlefield, J.W., Donovan, P.J., Blumenhal, P.D., Huggins, G.R., and Gearhart, J.D, (1998) Derivation of pluripotent stem cells from cultured human primordial germ cells, Proc Natl Acad Sci U S A95, 13726-13731.

Shapiro, A.M., Lakey, J.R., Ryan, E.A., Korbutt, G.S., Toth, E., Warnock, G.L., Kneteman, N.M., and Rajotte, R.V (2000). Islet transfer in a derived patient with type 1diabetes using a glucose-interactive-free immunological region registration (7 cases of type I diabetic patients who do not use glucocorticoid immunosuppressive therapy), N Engl J Med 343, 230-.

Shapiro, A.M., Ryan, E.A., and Lakey, J.R, (2001 a.) Pancreatic islet transplantation in the treatment of diabetes mellitus, Best practice Res Clin endothelial Metab 15, 241-.

Shariro, j., Ryan, e., warneck, g.l., kneeman, n.m., Lakey, j., Korbutt, g.s., and Rajotte, R.V. (2001 b.) Could power islets be transferred in type 1diabetes insulin absorption stroke be in transplantation (insulin independent less in type I diabetes mellitus). Bmj 322,861.

Shiozawa, m., Hiraoka, y, Komatsu, n, Ogawa, m., Sakai, y, and Aiso, s. (1996) Cloning and characterization of Xenopus laevis xSox7 cDNA (african claw frog xSox7 cDNA clone and characterization) biochem biophysis Acta 1309,73-76.

Smith, J. (1997). Brachyury and the T-box genes (Brachyury and T-box genes). Curr Opin Genet Dev 7, 474-.

Smith, J.C., arms, N.A., Conlon, F.L., Tada, M., Umbrauer, M.and Weston, K.M (1997). Upstream and downstream from Brachyury, a gene required for a verterburiate function formation (Upstream and downstream of a vertebrate mesoderm-forming essential gene Brachyury). Cold Spring Harb Symp Quant Biol 62, 337-.

Takash, w., canizes, j., Bonneaud, n., Poulat, f., Mattei, m.g., Jay, p., and Berta, P. (2001) SOX7 transcription factor, sequence, chromosomal localization, expression, transcription and interference with Wnt signaling (SOX7 transcription factor: sequencing, chromosomal localization, expression, transduction, and interference with Wnt signaling), Nucleic Acids Res 29, 4274-.

Taniguchi, K., Hiraoka, Y., Ogawa, M., Sakai, Y., Kido, S and Aiso, S. (1999) Isolation and characterization of a motor SRY-related cDNA, mSOx7 (mouse SRY-related cDNA, Isolation and characterization of mSOx 7) Biochim Biophys Acta 1445,225-231.

Technau, U. (2001), Brachyury, the blastopore and the evolution of the mesoderm evolution, Bioessays 23, 788-.

Thomson, J.A., Itskovitz-Eldor, J.Shapiro, S.S., Waknitz, M.A., Swergiel, J.J., Marshall, V.S., and Jones, J.M (1998) Embroken stem cells lines derived from human blastocyst Embryonic stem cells (human blastocyst stem cell line) Science 282,1145 and 1147.

Tremblay, K.D., Hoodeless, P.A., Bikoff, E.K., and Robertson, E.J. (2000). Formation of the defined end-effector in mouse a Smad2-dependent processes. (mouse definitive endoderm forms Smad2-dependent processes.) Development 127, 3079-.

Vandesompele, j., De Preter, k., Pattyn, f., Poppe, b., Van Roy, n., De Paepe, a. and Speleman, F. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometrical averaging of multiple internal control genes Accurate calibration of multiple internal control genes.

Varlet, i.e., Collignon, j. and Robertson, E.J (1997). nodal expression in the private end-compressor is required for specification of the antioxidant axing mouse, Development 1044, where nodal expression in primitive endoderm is essential for specialization of mouse primitive gut embryonic axis.

Vincent, S.D., Dunn, N.R., Hayashi, S., Norris, D.P., and Robertson, E.J (2003). Cell failure determinations with the use of the mouse organ r modified by modified Nodal signals in mouse formations. Genes Dev 17,1646- "1662.

Weiler-Guittler, H., Aird, W.C., Rayburn, H., Husain, M., and Rosenberg, R.D, (1996) Development regulated gene expression of thrombomodulin in fertilization mice embryos (expression of developmental regulatory genes of thrombomodulin in mouse embryos), Development 122,2271-2281.

Weiler-Guettler, H., Yu, K., Soff, G., Gudas, L.J., and Rosenberg, R.D (1992). Thrombobodulin gene regulation by cAMP and lubricating acid in F9embryo cardiovascular cells (F9 genes regulated by cAMP and retinoic acid). Procedings Of The National Academy Of Sciences Of The United States Of America 89, 2155-.

Wells, J.M. and Melton, D.A, (1999) Vertebrate endogerm degradation (Vertebrate endoderm development), Annu Rev Cell Dev Biol 15,393 410.

Wells, J.M. and Melton, D.A. (2000). Early mouse endogenerm is patterned by soluble factors from adjacent germ layers. Development 127, 1563-.

Willison, K. (1990) The mouse Brachyury gene and mesoderm formation Trends Genet 6,104-105.

Zhao, G.Q, (2003). sequences of knocking out BMP signaling in the mouse (results of knocking out mouse BMP signaling system). Genesis 35,43-56.

Zhou, X., Sasaki, H., Lowe, L., Hogan, B.L., and Kuehn, M.R (1993) Nodal is a novel TGF-beta-like gene expressed in mouse node during gastrulation Nature 361, 543-.

Sequence listing

<110> ViaCyte, Inc.

D'Amour, Kevin A.

Agulnick, Alan D.

Baetge, Emmanuel E.

<120> definitive endoderm

<130> 96183-000242US

<140> US 14/072,642

<141> 2013-11-05

<150> US 12/710,300

<151> 2010-02-22

<150> US 10/584,338

<151> 2007-01-09

<150> WO PCT/US2004/043696

<151> 2004-12-23

<150> US 60/587,942

<151> 2004-07-14

<150> US 60/586,566

<151> 2004-07-09

<150> US 60/532,004

<151> 2003-12-23

<160> 2

<170> PatentIn version 3.5

<210> 1

<211> 1245

<212> DNA

<213> Intelligent (homo sapiens)

<400> 1

atgagcagcc cggatgcggg atacgccagt gacgaccaga gccagaccca gagcgcgctg 60

cccgcggtga tggccgggct gggcccctgc ccctgggccg agtcgctgag ccccatcggg 120

gacatgaagg tgaagggcga ggcgccggcg aacagcggag caccggccgg ggccgcgggc 180

cgagccaagg gcgagtcccg tatccggcgg ccgatgaacg ctttcatggt gtgggctaag 240

gacgagcgca agcggctggc gcagcagaat ccagacctgc acaacgccga gttgagcaag 300

atgctgggca agtcgtggaa ggcgctgacg ctggcggaga agcggccctt cgtggaggag 360

gcagagcggc tgcgcgtgca gcacatgcag gaccacccca actacaagta ccggccgcgg 420

cggcgcaagc aggtgaagcg gctgaagcgg gtggagggcg gcttcctgca cggcctggct 480

gagccgcagg cggccgcgct gggccccgag ggcggccgcg tggccatgga cggcctgggc 540

ctccagttcc ccgagcaggg cttccccgcc ggcccgccgc tgctgcctcc gcacatgggc 600

ggccactacc gcgactgcca gagtctgggc gcgcctccgc tcgacggcta cccgttgccc 660

acgcccgaca cgtccccgct ggacggcgtg gaccccgacc cggctttctt cgccgccccg 720

atgcccgggg actgcccggc ggccggcacc tacagctacg cgcaggtctc ggactacgct 780

ggccccccgg agcctcccgc cggtcccatg cacccccgac tcggcccaga gcccgcgggt 840

ccctcgattc cgggcctcct ggcgccaccc agcgcccttc acgtgtacta cggcgcgatg 900

ggctcgcccg gggcgggcgg cgggcgcggc ttccagatgc agccgcaaca ccagcaccag 960

caccagcacc agcaccaccc cccgggcccc ggacagccgt cgccccctcc ggaggcactg 1020

ccctgccggg acggcacgga ccccagtcag cccgccgagc tcctcgggga ggtggaccgc 1080

acggaatttg aacagtatct gcacttcgtg tgcaagcctg agatgggcct cccctaccag 1140

gggcatgact ccggtgtgaa tctccccgac agccacgggg ccatttcctc ggtggtgtcc 1200

gacgccagct ccgcggtata ttactgcaac tatcctgacg tgtga 1245

<210> 2

<211> 414

<212> PRT

<213> Intelligent (homo sapiens)

<400> 2

Met Ser Ser Pro Asp Ala Gly Tyr Ala Ser Asp Asp Gln Ser Gln Thr

1 5 10 15

Gln Ser Ala Leu Pro Ala Val Met Ala Gly Leu Gly Pro Cys Pro Trp

20 25 30

Ala Glu Ser Leu Ser Pro Ile Gly Asp Met Lys Val Lys Gly Glu Ala

35 40 45

Pro Ala Asn Ser Gly Ala Pro Ala Gly Ala Ala Gly Arg Ala Lys Gly

50 55 60

Glu Ser Arg Ile Arg Arg Pro Met Asn Ala Phe Met Val Trp Ala Lys

65 70 75 80

Asp Glu Arg Lys Arg Leu Ala Gln Gln Asn Pro Asp Leu His Asn Ala

85 90 95

Glu Leu Ser Lys Met Leu Gly Lys Ser Trp Lys Ala Leu Thr Leu Ala

100 105 110

Glu Lys Arg Pro Phe Val Glu Glu Ala Glu Arg Leu Arg Val Gln His

115 120 125

Met Gln Asp His Pro Asn Tyr Lys Tyr Arg Pro Arg Arg Arg Lys Gln

130 135 140

Val Lys Arg Leu Lys Arg Val Glu Gly Gly Phe Leu His Gly Leu Ala

145 150 155 160

Glu Pro Gln Ala Ala Ala Leu Gly Pro Glu Gly Gly Arg Val Ala Met

165 170 175

Asp Gly Leu Gly Leu Gln Phe Pro Glu Gln Gly Phe Pro Ala Gly Pro

180 185 190

Pro Leu Leu Pro Pro His Met Gly Gly His Tyr Arg Asp Cys Gln Ser

195 200 205

Leu Gly Ala Pro Pro Leu Asp Gly Tyr Pro Leu Pro Thr Pro Asp Thr

210 215 220

Ser Pro Leu Asp Gly Val Asp Pro Asp Pro Ala Phe Phe Ala Ala Pro

225 230 235 240

Met Pro Gly Asp Cys Pro Ala Ala Gly Thr Tyr Ser Tyr Ala Gln Val

245 250 255

Ser Asp Tyr Ala Gly Pro Pro Glu Pro Pro Ala Gly Pro Met His Pro

260 265 270

Arg Leu Gly Pro Glu Pro Ala Gly Pro Ser Ile Pro Gly Leu Leu Ala

275 280 285

Pro Pro Ser Ala Leu His Val Tyr Tyr Gly Ala Met Gly Ser Pro Gly

290 295 300

Ala Gly Gly Gly Arg Gly Phe Gln Met Gln Pro Gln His Gln His Gln

305 310 315 320

His Gln His Gln His His Pro Pro Gly Pro Gly Gln Pro Ser Pro Pro

325 330 335

Pro Glu Ala Leu Pro Cys Arg Asp Gly Thr Asp Pro Ser Gln Pro Ala

340 345 350

Glu Leu Leu Gly Glu Val Asp Arg Thr Glu Phe Glu Gln Tyr Leu His

355 360 365

Phe Val Cys Lys Pro Glu Met Gly Leu Pro Tyr Gln Gly His Asp Ser

370 375 380

Gly Val Asn Leu Pro Asp Ser His Gly Ala Ile Ser Ser Val Val Ser

385 390 395 400

Asp Ala Ser Ser Ala Val Tyr Tyr Cys Asn Tyr Pro Asp Val

405 410

Claims

1. An in vitro cell culture comprising definitive endoderm cells, at least 100ng/ml activin A, Wnt3a, and a culture medium comprising no more than about 10% (v/v) serum, wherein the definitive endoderm cells are not obtained from a human embryo after 14 days of fertilization, and wherein the in vitro cell culture does not comprise human germ cells or a fertilized egg.

2. The in vitro cell culture of claim 1, wherein said in vitro cell culture further comprises a subgroup of BMPs.

3. The in vitro cell culture of claim 2, wherein said subgroup of BMPs is BMP 4.

4. The in vitro cell culture of claim 1, wherein said in vitro cell culture further comprises activin B.

5. The in vitro cell culture of claim 1, wherein said in vitro cell culture further comprises Nodal.

6. The in vitro cell culture of claim 1, wherein said medium comprises about 10% (v/v) serum.

7. The in vitro cell culture of claim 1, wherein said medium comprises about 2% (v/v) serum.

8. The in vitro cell culture of claim 1, wherein said medium comprises about 0.5% (v/v) serum.