HK1113171B - Medium and culture of embryonic stem cells - Google Patents

Description

Culture medium and culture of embryonic stem cells

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. provisional patent application serial No. 60/608,040, filed on 8.9.2004.

Statement regarding federally sponsored research or development

The invention was made with U.S. government support awarded by the following agencies: NIH RR 17721. The united states government has certain rights in this invention.

Background

Stem cells are defined as cells that are capable of differentiating into many other differentiated cell types. Embryonic stem cells are stem cells from embryos that are capable of differentiating into most, if not all, differentiated cell types of the mature body. A stem cell is said to be pluripotent, meaning that it is capable of differentiating into many cell types. One type of pluripotent stem cell that has received much attention from research institutions is the human embryonic stem cell, herein abbreviated as human ES cell, which is an embryonic stem cell derived from a human embryonic source. Human embryonic stem cells are of great scientific interest because they are capable of unlimited proliferation in culture and thus are capable, at least in principle, of providing cells and tissues to replace defective or damaged human tissue. The existence of human embryonic stem cell cultures has the potential to provide unlimited quantities of human cells and tissues for use in various therapeutic protocols and research programs that contribute to human health. It is conceivable that in the future, human embryonic stem cells will be able to proliferate and be directed to differentiate into specific lineages to produce differentiated cells or tissues that can be transplanted into the human body for therapeutic purposes. Human embryonic stem cells and the differentiated cells that can be derived from them are also powerful scientific tools for studying human cells and the developmental system.

Basic techniques for the production and culture of human embryonic stem cells have been described. The prior art is certainly effective, but some of the current methods for culturing human embryonic stem cells have limitations and disadvantages. One limitation is particularly significant. Most existing human embryonic stem cell lines have been, to some extent or otherwise, exposed directly to mouse cells or to a medium in which mouse cells have been cultured. The fact that some human ES cells from existing cell lines were found to exhibit the sialic acid residue Neu5Gc, which is not normally produced by human cells, is of great interest. The initial technique for generating and culturing human embryonic stem cells used Mouse Embryonic Fibroblast (MEF) feeder cells as a feeder layer on which human embryonic stem cells could be cultured. The fibroblast feeder layer allows the stem cells to remain in an undifferentiated state through a number of mechanisms that are still not completely cleared. It was subsequently found that the same phenomenon can be obtained if the stem cells are exposed to "conditioned medium". Conditioned medium is stem cell medium that has been cultured with feeder cells (e.g., MEF). Whether feeder cells confer or remove certain factors from the medium, the result is conditioned medium that can be used to culture stem cells without differentiation. Any culture condition, i.e., growth of human ES cells directly on murine feeder cells or the use of conditioned medium, involves the problem that one or more agents, such as viruses, can be transferred from mouse cells to human ES cells. If one of the goals of culturing human embryonic stem cells is to produce tissue that can ultimately be implanted into the human body, it is highly desirable that the stem cells never be exposed to cells of other species or to the media that has been used to culture cells of other species. Therefore, it is of great interest in the continued development of techniques for long-term culturing of human embryonic stem cells to find culture conditions that will proliferate and culture human embryonic stem cells without a fibroblast feeder layer.

One characteristic trait of human embryonic stem cells in culture is that such cells tend to differentiate if conditions are not ideal enough. Human ES cells are readily induced to differentiate when it is highly desirable to maintain them in an undifferentiated state in culture. Most culture conditions will result in some level of unwanted differentiation, especially around the edges of growing ES cell colonies. Although ES cells may be cultured with some degree of unwanted differentiation, the objective is to define culture conditions that will maintain the culture as undifferentiated as possible, i.e., that will result in as few differentiated cells as possible. We believe we have used particularly stringent criteria to define conditions that will support the indefinite culture of undifferentiated ES cell cultures.

Some media formulations are capable of maintaining human ES cells in an undifferentiated state for a period of time, but will not be able to maintain this state when cultured for extended periods of time. Specifically, we define human ES cells from an initial seed culture to grow on a culture vessel until the cells are confluent in the same culture vessel as a "passage". We have found that some media formulations allow human ES cells to be cultured without severe differentiation for one or two passages, but that the cells differentiate rapidly during subsequent passages. We have now come to believe that in order to obtain a medium that does support the immortalization of human ES cells without differentiation and without feeder cells or conditioned medium, the medium must support the culture of human ES cells in a substantially homogeneous and undifferentiated state for at least 5 passages. It is also important that the culture maintain relative homogeneity and undifferentiated throughout the culture and retain all of the important characteristics of human ES cells.

The differentiation state of stem cell cultures can be very easily assessed by morphological features. Undifferentiated stem cells have a characteristic morphology, i.e., small and dense cells with clear cell boundaries, which can be readily observed by examining stem cell cultures under a microscope. In contrast, cells that have differentiated appear larger and more diffuse, with no apparent boundaries. Although some differentiated cells may appear at the edges of colonies of undifferentiated cells (which is often the case), the most desirable stem cell culture is one that proliferates in the culture vessel with only a minimal amount of cells around the seemingly differentiated culture. Differentiation and health of human ES cell cultures can be accurately determined visually from experience. The biochemical marker used to trace the undifferentiated state of ES cells is the transcription factor Oct4, which is considered to be the most reliable marker of the undifferentiated state of ES cells and one of the markers that is lost first when the undifferentiated cells begin to differentiate.

Summary of The Invention

The present invention can be summarized as a method for culturing human embryonic stem cells without the need for feeder cells or conditioned medium, comprising the step of culturing human embryonic stem cells in a medium comprising salts, vitamins, amino acids, glucose, fibroblast growth factor, gamma aminobutyric acid, 2-piperidinoic acid (pipecolic acid), lithium, and transforming growth factor beta in amounts sufficient to maintain the stem cells in an undifferentiated state in multiple subcultures.

The present invention also relates to an in vitro cell culture of human embryonic stem cells cultured in a medium containing high levels of fibroblast growth factor, gamma aminobutyric acid, 2-pipecolic acid, lithium and transforming growth factor beta, such that the stem cells can be cultured indefinitely in an undifferentiated state without the need for fibroblast feeder cells or conditioned medium.

It is an object of the present invention to define long-term culture conditions for human embryonic stem cells that avoid the use or exposure to animal cells and animal proteins, whether feeder cells or other animal cells and animal proteins used to condition the medium in which the stem cells are to be cultured.

It is another object of the present invention to define culture conditions for human embryonic stem cells that are as defined as possible while maintaining as far as possible the maximum proportion of cells in culture in an undifferentiated state.

Other objects, features and advantages of the present invention will become apparent from the following description.

Brief Description of Drawings

FIG. 1 presents data obtained from experimental work described in the following specification, showing that media composition reduces the proportion of differentiated cells in human ES cell cultures grown therein.

FIG. 2 is a graph showing data that media containing human matrix protein allows cultured human ES cells to exhibit no sialic acid residues of non-human origin.

Figure 3 is a graphical representation of data showing high levels of undifferentiated cells in human stem cell culture.

FIG. 4 is a graphical representation of data showing robust growth of stem cell cultures by the media described herein.

Detailed Description

We have identified a number of culture conditions and media that enable human embryonic stem cells to be cultured and robustly propagated in an undifferentiated state for an indefinite period of time while eliminating the need for feeder cells and conditioned medium at all. The development of these media and culture conditions enables the derivation and maintenance of human ES cell lines under defined and controlled conditions without the need for direct or indirect contact with any kind of animal cells. These media have been shown to support the proliferation of undifferentiated ES cells for many generations, at least 5 generations, which is strong evidence that these media will support this indefinite culture.

Defined humanized media used to culture and propagate human ES cells typically comprise salts, vitamins, glucose sources, minerals, and amino acids. To supplement the media and provide conditions to support cell growth, the original stem cell media contains serum from one source or another. It has also been previously reported that the addition of fibroblast growth factor and a serum replacement supplement allows the culture of human ES cells without the need for serum. Serum replacement is commercially available for this purpose or may be a mixture formulated with proteins such as serum albumin, vitamins, minerals, transferrin or transferrin substitutes, and insulin or insulin substitutes. Selenium may also be added to this serum replacement component. It is preferred here to replace serum of any origin from which human ES cells are cultured with defined replacement serum in order to avoid problems with changes in serum composition and to use a medium which is as defined as possible. We have defined a sufficient medium and all the components of the medium referred to herein are listed in Table 1 below, which lists all the components of our medium known as TeSR1 and the concentrations of the individual components. The TeSR1 culture medium is composed of DMEM/DF12 basal medium added with human serum albumin, vitamins, antioxidants, trace minerals, specific lipids and cloned growth factors.

To avoid the need for a fibroblast feeder layer previously thought to be necessary to maintain human ES cells in an undifferentiated state, it is reported herein that the use of higher concentrations of FGF (10-1000ng/ml) in combination with GABA (gamma aminobutyric acid), 2-Pipecolic Acid (PA), lithium (LiCl) and transforming growth factor beta (TGF β) will enable the medium to support the growth of undifferentiated stem cells. It has been found that the combination of these supplements is sufficient to maintain a human ES cell culture in an undifferentiated state indefinitely without contact with feeder cells or conditioned medium. These additions proved to be sufficient. However, not all of them are required for every medium formulation. By selective deletion of these additives, one or more of these components can be deleted, resulting in a human ES cell culture that will still grow but with a loss of purity in the undifferentiated state of the culture. Such cultures may or may not remain stable for many generations. However, it is clear that this combination is sufficient for various media to support long-term culture and proliferation of undifferentiated human ES cells in the absence of feeder cells or conditioned media for an unlimited number of passages.

We initially conducted a subjective screening of these growth factors, and selected for receptors expressed by human ES cells, and identified some factors that had a positive effect on the proliferation of undifferentiated cells. Among these factors, bFGF, LiCl, gamma-aminobutyric acid (GABA), 2-pipecolic acid and TGF β are finally contained in TeSR 1. The proliferation rate and percentage of cells in TeSR1 that consistently expressed the human ES cell signature was higher for each of the four cell lines tested than for control cells cultured in fibroblast conditioned medium, and removal of any of these 5 factors reduced the effectiveness of the culture. Some of these data are shown in FIG. 1, which shows that cultures grown in media omitting any of these components show a lower proportion of undifferentiated cells than cultures containing all 4 media components. Note that Oct4, SSEA-1, SSEA-4, Tral-60, and Tral-80 are all cell surface markers or transcription factors used to track the differentiation status of stem cells (Oct 4). Figure 4 shows a similar experiment which demonstrates that the growth rate of the culture is highest in multiple passages when all these components are contained in the medium.

It is also beneficial for the culture conditions of human ES cells to contain a biological matrix in the culture vessel. One such material that has been used previously is Matrigel^TMThis is an artificial basement membrane from mouse cells, which is provided as a commercial product without mouse cells. Another human-derived material now known for similar purposes is fibronectin, a human glycoprotein, used in an insoluble form to form a fibrous matrix that also serves as a basement membrane for ES cell culture. To our knowledge, fibronectin matrices alone are not sufficient. However, it has now been found that human matrix materials can be made from a combination of the human matrix proteins collagen IV, fibronectin, laminin and vitronectin, which is sufficient to support human ES cells in a permanently undifferentiated state in TeSR1 medium.

The above listed media additions were obtained after systematically testing over 80 growth factors. While some of these supplements support growth of human ES cells in culture for at least several passages, many fail to maintain ES cells in an undifferentiated state during subsequent passages. We did not identify other combinations of these factors that produced the results of the media additions described in the examples below. This is not to say that no changes can be made to these components. For example, LiCl is used in this medium because it stimulates the wnt pathway. Wnt itself or other stimulators of the pathway such as activin could be used as an equivalent to LiCl in place of LiCl, although LiCl is probably the most economical agent for this purpose. Similarly, GABA is thought to react with the GABA receptor and several molecules have been identified in the scientific literature as agonists of this same receptor, which molecules can also serve as equivalents, replacing GABA in the culture medium. It is also believed that PA also reacts with GABA receptors. While both PA and GABA were found to be helpful in the media at the concentrations used herein, it is also envisioned that the concentration of one or the other of these components could be significantly increased while avoiding the need for the other component.

Higher concentrations (40-100ng/ml) of fibroblast growth factor appear to eliminate the need for feeder cells. Preferred FGFs are basic FGFs, also known as bFGF and FGF2, but other FGFs, including at least FGF4, FGF9, FGF17 and FGF18, are sufficient for this purpose. Other FGFs may be effective even at higher concentrations.

The previous observation that human Embryonic Stem (ES) cell cultures can only be maintained in an undifferentiated state in the presence of fibroblast feeder cells or when cultured in conditioned medium has led to the speculation that fibroblasts release into the medium certain factors that inhibit ES cell differentiation. However, regardless of the effect on the medium mediated by fibroblast feeder cells, it will now be clear that the medium described below will replace this effect. The three media described below are defined, do not contain animal cells, and are capable of long-term culture of undifferentiated human ES cells in an undifferentiated state. An example of a medium is provided in which the proteins in the medium are all human proteins, with "humanized" media and culture conditions, thereby avoiding any possible problems involving animal-derived subcellular products.

Examples

Unless otherwise indicated, the composition of TeSR1 medium used for all cultures described herein is listed in table 1 below. Our preliminary experiments suggest that undifferentiated human ES cells were at pH 7.2, osmolality (osmomolity) 350mOsmol and 10% CO₂/5％O₂The proliferation is optimum under the atmosphere. These conditions were used for all subsequent cultures described herein.

Cells of human ES cell lines H1, H7, H9 and H14 proliferated robustly in TeSR1 for 11, 7, 25 and 17 passages (2-6 months), respectively. The karyotype was confirmed to be normal after passage 7 for the cell line H14, and after passage 8 and 21 for H9. It was confirmed that H1 and H9 formed teratomas after 11 and 20 passages.

Previous ES cell cultures are known to be less than optimal due to the presence of Neu5Gc, a sialic acid that is not made by humans. Since the human matrix components collagen, fibronectin, laminin and vitronectin exclude the final animal product from the TeSR1 culture conditions of ES cells, we examined whether the existing human ES cell line eliminated Neu5Gc when cultured in this medium. We confirmed the presence of Neu5Gc on human ES cells cultured in fibroblast conditioned medium, reduced but detectable Neu5Gc on cells cultured in TeSR1 on Matrigel, and no Neu5Gc on ES cells cultured in TeSR1 using the four human matrix components. These data are shown in figure 2. Thus, human ES cells cultured on TeSR1 and human protein matrices do not display non-human sialic acid residues found in cells cultured on murine feeder cells.

To test the conditions of the ES cell colonies and to maintain the suitability for culture for long-term maintenance of human ES cell cultures, TeSR1 medium was compared with the best previous culture conditions (conditioned medium was used as our best knowledge). The TeSR1 medium was found to be able to maintain human ES cells in an undifferentiated state in which more than 90% of the cells continued to be Oct4 positive even after prolonged culture. The results of this measurement are shown in FIG. 3. This is the first indication that medium without any feeder cells and without conditioned medium can maintain undifferentiated growth levels of human ES cells to the point that more than 90% of the cells in culture remain undifferentiated at all stages.

Growth curves and FACS analysis of H1 cells cultured for 3 passages were tested in TeSR1 medium and TeSR medium omitting the following components: TGF beta, PA, LiCl, GABA and bFGF. At the beginning of the culture, 3X 10 cells were grown on day 0 of passage 1⁵Cell plating of each cell line. The number of cells from three duplicate wells was calculated to assess adherence (days 2-3) and final cell number at passage (days 6-7). The initial plank density and sampling time were repeated, possibly for 5 generations. Cells were analyzed by FACS on day 6 of passage 3 for cell surface markers SSEA1, SSEA4, Tra 1-60 and Tra 1-81 and the transcription factor Oct 4. The data is shown in figure 1. These data show that human ES cells can be cultured in media lacking each of these components, but at the expense of some unwanted differentiation of the cells upon culture, and that the highest level of undifferentiated culture can only be obtained with all of these components. Similar results were obtained with other cell lines.

Pluripotency of human ES cell lines maintained in TeSR1 medium was tested. Cells of the emerging cell lines WA01 and WA09, cultured in TeSR1 medium on Matrigel matrix for 11 and 20 passages, respectively, were injected into SCID-beige mice, respectively. Teratomas displaying complex differentiation appeared in mice 6-8 after inoculation.

TABLE 1 complete formulation of TeSR1 Medium

Inorganic salt mM amino acid mM

Calcium chloride (anhydrous) 0.8232L-alanine 0.1392

HEPES 11.76L-arginine hydrochloride 0.5488

Magnesium chloride (anhydrous) 0.2352L-asparagine-H2O 0.1392

Magnesium sulfate (MgSO4) 0.319088L-aspartic acid 0.1392

Potassium chloride (KCl) 3.26144L-cysteine-HCl-H2O 0.0784

Sodium bicarbonate (NaHCO3) 11.2112L-cystine 2HCl 0.0784

Sodium chloride (NaCl) 94.55824L-glutamic acid 0.1392

Disodium hydrogen phosphate (anhydrous) 0.392L-Glutamine 2.96

Sodium dihydrogen phosphate (NaH2PO4-H2O) 0.355152 Glycine 0.296

L-histidine-HCl-H2O 0.1176

Trace mineral L-isoleucine 0.326144

Ferric nitrate (Fe (NO3)3-9H2O) 0.00009408L-leucine 0.353584

Ferric sulfate (FeSO4-7H2O) 0.001176 hydrochloric acid L-lysine 0.391216

Copper sulfate (CuSO4-5H2O) 4.0768E-06L-methionine 0.090944

Zinc sulfate (ZnSO4-7H2O) 0.001176L-phenylalanine 0.16856

Ammonium metavanadate NH4VO30.000056L-proline 0.2176

Manganese sulfate MnSO 4H 2O 1.00592E-05L-serine 0.296

Ammonium molybdate 1.00404E-05L-threonine 0.352016

NiSO 46H 2O 4.94861E-06L-Tryptophan 0.0346528

Sodium metasilicate Na2SiO 39H₂O0.004926108L-tyrosine 2Na 2H2O 0.167776

SnCl25.32544E-06L-valine 0.354368

CdCl2 6.21931E-05

CrCl39.41176E-06 vitamin

AgNO35.00293E-06 ascorbic acid 0.375

AlCl 36H 2O 2.4855E-05 biotin 1.12112E-05

Ba (C2H3O2) 24.99217E-05 choline hydrochloride 0.0502544

CoCl 26H 2O 5.0021E-05D-calcium pantothenate 0.0036064

GeO22.5337E-05 Folic acid 0.004704

KBr 5.04202E-06 i-inositol 0.05488

KI 5.12048E-06 nicotinamide 0.012936

NaF 0.000500119 pyridoxine hydrochloride 0.0076048

RbCl 5.00414E-05 Riboflavin 0.0004704

ZrOCl 28H 2O 9.03834E-05 thiamine hydrochloride 0.02460217

Vitamin B120.000392

Growth factor

GABA 0.979 energy substance

2-Piperidinic acid 0.000984D-glucose 13.72784

bFGF 5.80E-06 sodium pyruvate 0.392

LiCl 0.979

TGF-beta 12.35E-08 protein

Human insulin 0.0034438

Lipid human holotransferrin (Holo-Transferrin) 0.14

Linoleic acid 0.0070976 human serum albumin 199.7

Lipoic acid 0.00039984

Arachidonic acid 0.001312 other Components

Cholesterol 0.0113798 glutathione (reduced form) 0.00592996

DL-alpha tocopherol-acetate 0.02962 hypoxanthine Na 0.01176

Linolenic acid 0.007184 phenol Red 0.0159936

Myristic acid 0.008758 putrescine-2 HCl 0.000394352

Oleic acid 0.00708 thymidine 0.001176

Palmitoleic acid 0.0078622-mercaptoethanol 0.1

Stearic acid 0.00703 selenium 0.000177304

Pluronic F-68 0.238

Tween 80 0.3358

Claims (4)

1. A cell culture medium comprising sufficient amounts of albumin, salts, minerals, vitamins, amino acids, glucose, transferrin, insulin, fibroblast growth factor, transforming growth factor beta, gamma aminobutyric acid, 2-pipecolic acid, and lithium, whereby human embryonic stem cells cultured in said medium maintain an undifferentiated state through multiple culture passages.

2. The culture medium of claim 1, being feeder-free and never exposed to feeder cells.

3. The culture medium of claim 1, wherein the lithium is LiCl.

4. The culture medium of claim 1, comprising fibroblast growth factor at a concentration of at least 40 ng/ml.