PerspectiveSTEM CELLS

Setting Standards for Human Embryonic Stem Cells

See allHide authors and affiliations

Science  09 May 2003:
Vol. 300, Issue 5621, pp. 913-916
DOI: 10.1126/science.1082940

Human embryonic stem cells [HN1] (HESCs) are important both for their potential in regenerative medicine and as a window on early human embryology [HN2]. A group of molecular biologists who study embryogenesis in human and animal systems gathered to review and assess molecular and cellular standards for HESCs—functional assays, and important future research avenues (1). This Perspective reports the proceedings of this workshop.

Our goal was to initiate a scientific discussion that may help to establish standards for evaluating HESCs, in particular, to enable the comparison and classification of existing and future cell lines. This approach also should allow a direct comparison between HESCs and their counterparts in mice and other animals that are more amenable to experimentation. Molecular markers have enabled many new insights in the field of modern (nonhuman) embryology, and in a similar way HESCs may shed light on human embryology.

Basic Characteristics of HESCs

Embryonic stem cells are endowed with at least two remarkable properties. First, stem cells can proliferate in an undifferentiated but pluripotent state, and therefore can self-renew. Second, they have the ability to differentiate into many specialized cell types. HESCs are derived from the early human blastocyst (5 days postfertilization) from a region of the embryo called the inner cell mass (ICM). As with mouse embryonic stem cells [HN3] (MESCs) (25), HESCs exhibit the following basic characteristics: (i) The cells are karyotypically normal, (ii) they survive and proliferate in vitro indefinitely under well-defined tissue culture conditions, (iii) most of the cells recover after freezing and thawing, and (iv) they differentiate into a variety of cell types in vitro and in vivo. These characteristics are well documented, but several important questions await resolution.

Reproducibility of HESC Derivation

Derivation of new HESCs in academic environments is limited by the moratorium on federal funding in a number of countries, including the United States. The premoratorium experience, combined with the experience of a few academic laboratories using private funds, as well as information available in the private sector and in countries where these experiments can be performed, suggests that scientists can derive HESCs with good success rates.

Effective Culture Conditions

Current protocols for dissociation of human embryos to generate stem cells are adapted from the mouse. Briefly, the ICM of a human blastocyst is removed by immunosurgery, dissociated in Ca2+-Mg2+-free medium, and plated over mouse embryonic fibroblasts or human feeder cells (6). The mouse cells are irradiated to suppress their proliferation. Current published culture conditions allow the maintenance of HESCs for many passages (up to 80) in the undifferentiated state (7, 8) [HN4].

Culture conditions enabling the successful generation of HESCs are similar to those for MESCs except regarding growth factor requirements. MESCs require the cytokine leukemia inhibitory factor (LIF) [HN5] to maintain their undifferentiated pluripotent state, whereas HESCs are not responsive to LIF for the maintenance of pluripotency (7, 9). An as yet unidentified signal in medium conditioned with mouse embryonic fibroblasts maintains HESCs, in a pluripotent state (10). In addition, HESCs, like other human cells in culture (but not MESCs) require fibroblast growth factor to grow (7) [HN6]. The longer cell cycle of HESCs in culture (11) may affect their responses in different ways from MESCs.

How Is HESC Self-Renewal Regulated?

LIF, which mediates its effect via the membrane protein gp130 and activation of the Stat3 signaling pathway, is sufficient to maintain the undifferentiated state of MESCs (9). In HESCs, LIF is not required to maintain the undifferentiated state, but whether or not the downstream signaling pathway (gp130, JAK/STAT) participates has not yet been studied. The extent to which signaling pathways are shared between HESCs and MESCs will determine the extent to which MESC experimental protocols must be adjusted for HESCs.

Are All HESC Lines the Same?

A related question is: How many states of “stemness” are there in the embryo? Progress in understanding these issues requires comparison of molecular markers of the undifferentiated pluripotent state as well as cellular characteristics and potency among different HESC lines [HN7]. These studies are well advanced for MESCs but have not yet been initiated for HESCs (9). Thus, investigators currently use molecular markers originally characterized in MESCs to test the state of an HESC line. These markers are defined as factors expressed consistently, and enriched, in embryonic stem cells. We propose two classes of markers (see the table): Class I markers are cloned, well defined, and unambiguous. Class II markers need more extensive characterization.

Molecular markers of HESCs.

Listed are 13 molecular markers for identifying undifferentiated pluripotent HESCs. These markers are expressed (enriched) in undifferentiated HESCs and are turned off after differentiation. There are two groups: class I, cloned markers that are well defined and unambiguous, and class II, cell surface antigens detected by antibodies but requiring more extensive characterization.

View this table:

We note that not all mouse markers are expressed in human cells. For example, SSEA3 and SSEA4 are expressed in HESCs but not in MESCs, whereas SSEA1 is expressed in MESCs but not in HESCs (12) [HN8]. We anticipate that this list will grow rapidly as expression-profiling data from both MESCs and HESCs are further analyzed. The process could be expedited by making raw data from embryonic stem cell profiling available to the scientific community (13, 14). Such state-specific expression fingerprints should help standardize our definitions of HESCs.

How Can HESCs Be Genetically Altered?

A number of reports have documented successful transfection and subsequent expression of genes in HESCs (1517). Human retroviral vectors (such as lentiviruses) and biochemical strategies have shown promise for transfecting HESCs. If HESCs are going to be used for therapy, the possibility of random integration of transfected genes cloned into viral vectors should be addressed with care. Loss-of-function experiments by site-specific recombination (knockouts or knockins) routinely performed in MESCs have recently been reported in HESCs, and provide the opportunity to genetically alter HESCs (18).

How Robust Are HESCs?

We do not have a clear understanding of the molecular properties that underlie stem cell biology in any system, including mouse, and significant differences between MESCs and HESCs are already apparent. For instance, MESCs grow much more rapidly in culture than HESCs; the ligand involved in MESC self-renewal has been identified; MESCs can be derived from genetically manipulated backgrounds; and MESCs are much more amenable to a variety of in vivo assays free of ethical and sociopolitical issues. HESCs, on the other hand, are difficult to obtain and slow to grow; the molecular mechanisms underlying their self-renewal remain unknown. HESCs are also genetically difficult to manipulate and are very difficult to assay in vivo. Thus, basic questions about MESCs and HESCs should be addressed in parallel.

Additionally, because of the differences between MESCs and HESCs, the study of HESCs, rather than stem cells from model systems, is necessary if HESCs are to be developed for clinical applications.

How Well Do HESCs Differentiate?

In vitro, HESCs differentiate into a variety of cell types (but not all types) that derive from the three embryonic germ layers. Cultured HESCs form neurons and skin cells (indicating ectodermal differentiation) (1922); blood, muscle, cartilage, endothelial cells, and cardiac cells (indicating mesodermal differentiation) (2325); and pancreatic cells (indicating endodermal differentiation) (26).

Additionally, HESCs form embryoid bodies containing all three germ layers (27). Thus, HESCs exhibit an in vitro differentiation potential parallel to that of MESCs, but it is unclear whether this occurs during normal embryogenesis, or whether differentiation can be achieved through other pathways.

Developmental decisions follow a hierarchy in time. Before discrete organs and specialized cell types are made, the embryo (and, by inference, HESCs) generates three embryonic germ layers: ectoderm, mesoderm, and endoderm [HN9]. In seeking the molecular pathways underlying organ formation, it is possible to identify activities that are necessary but not sufficient for the formation of a particular cell type. For example, a certain treatment might push HESCs to generate endoderm without giving rise to more specific endodermal derivatives, such as pancreas. Thus, it is important to follow differentiation in time to account for the intermediate steps a cell undergoes before it commits to its final fate. Understanding the normal progression of development should help us to design the sequence of treatments that can drive differentiation of embryonic stem cells. Although many markers of germ layers exist in model embryos, there is currently no comprehensive survey of these markers in human embryos.

An embryonic stem cell is defined as a cell that contributes to all embryonic germ layers and the germ line in vivo (9). The contribution to the germ layer of marked MESCs transplanted into mouse embryos at different stages of embryogenesis (most often at the blastocyst stage) is estimated by following the marked cells. A rigorous test of HESC potency in vivo is required.

For ethical reasons, the implantation of HESCs should not be performed with human embryos as hosts. The only published in vivo assay performed so far is the subcutaneous injection of HESCs into immunocompromised mice (such as NOD/SCID mice) (7, 8). This assay revealed that HESCs behave like MESCs. The injected mice formed teratomas (embryonic-like tumors) that contain derivatives of the three embryonic germ layers. This in vivo assay is useful to assess HESC potency, but additional assays are needed to evaluate the contribution of HESCs to extraembryonic lineages and germ cells.

Analysis of HESC Differentiation

Improved in vitro and in vivo assays are essential for understanding the biology of HESC differentiation, a prerequisite for their use in regenerative medicine. Additional in vitro assays can be derived from those already in use for fish, frog, chick, and mouse embryonic tissues.

In vivo assays are more difficult to design. Drawing on lessons from model systems, there are two simple assays. First, marked HESCs could be transplanted into defined tissue environments in discrete regions of nonhuman adults or fetuses to test for their ability to be incorporated into these tissues. Similar experiments with mouse stem cells differentiated in vitro into motor neurons and transplanted back into a chick fetus have shown that these mouse neurons contribute to the chick spinal cord (28). Human fetal and adult neuronal and hematopoietic stem cells have been transplanted into mouse embryos, and their contribution to a variety of organs has been reported (29, 30).

The first assay would test local contributions to specific fetal environments. In the second assay, marked HESCs could be transplanted at an earlier time point into nonhuman blastocysts to test for global incorporation into host tissues. HESCs transplanted into embryos of model systems, including mice and chickens, could be evaluated for their contributions to tissues and organs in these hosts. The assay should be performed transiently, that is, embryos would be removed from the host during different stages of gestation. We recognize that the second assay will require endorsement by the appropriate scientific institutional review board.

Infrastructure Requirements

Because the source of biological material (the human embryo) will always be limiting, the establishment of infrastructure is essential for effective progress. We suggest the creation of a HESC repository and registry, similar to the proposed Stem Cell Bank in the United Kingdom [HN10]. The repository would collect all HESC lines (including those generated after 9 August 2001), test them for the criteria described above, and make them available to academic institutions. The quality controls and criteria for submission could be similar to those established for centers such as the American Type Culture Collection [HN11]. This repository would take away the burden of obtaining the cell lines from third parties and would assign the task to a centralized facility, which would take responsibility for providing the cells to interested laboratories in a timely fashion. This could provide a solution to the dilemma faced by U.S. researchers.

The registry would include a Web-based database available to university and private-sector researchers where all the data about HESCs and human embryos would be collected. The registry would collect and distribute information pertinent to a number of areas, including results of microarray analysis (raw data) or other high-throughput methodologies, growth and culture conditions of the cell lines, differentiation potential of the cell lines, and number of passages the cell lines can sustain.

Two aspects of this endeavor will require special attention. First, quality control of the deposited information (for example, raw microarray data) must be stringent, defined and imposed by a committee, which we suggest be composed of scientists with expertise in molecular embryology, high-throughput data analysis, and bioinformatics. Second, the maintenance and upgrade of the information will require a committed, long-term effort.

This document provides a starting point, which we anticipate will be refined and strengthened as our knowledge of HESCs and human embryology expands.

HyperNotes Related Resources on the World Wide Web

General Hypernotes

This issue also has an Enhanced Policy Forum by E. Zerhouni titled “Stem cell programs.”

Dictionaries and Glossaries

The xrefer Web site provides searchable scientific dictionaries and other references.

The On-line Medical Dictionary is available on CancerWEB.

A Glossary of Stem Cell-Related Terms is provided by the International Society for Stem Cell Research.

A glossary is included in Stem Cells: A Primer.

Web Collections, References, and Resource Lists

The Virtual Library of Developmental Biology is maintained by the Society for Developmental Biology.

The Google Web Directory provides links to Internet resources on stem cells.

The Yahoo Directory provides links to Internet resources and news on stem cell research.

Genomics: A Global Resource from the Pharmaceutical Research and Manufacturers of America provides links to stem cell resources.

The National Institutes of Health (NIH) News & Events Web site provides a resource page of stem cell information.

The National Stem Cell Resource Web site of the American Type Culture Collection provides links to Internet resources.

The Stem Cell Network at the University of Ottawa provides stem cell news, an overview of stem cell research, and summaries of research themes.

Online Texts and Lecture Notes

Stem Cells: A Primer is provided by NIH; an NIH backgrounder on stem cells is also available.

The University of Wisconsin Communications Office offers a presentation on embryonic stem cells with links to other Internet resources.

J. Kimball maintains Kimball's Biology Pages, an online biology textbook and glossary. A presentation on stem cells is included.

Stem Cell Research is a student project prepared for a course on biotechnology and its social impact at Princeton University. Stem cell basics and a stem cell tutorial are included.

An education module on human embryonic and fetal stem cell research is provided by the University of Pittsburgh Health Sciences' Research & Practice Fundamentals Web site.

E. Lutz, Department of Bioscience, University of Strathclyde, Glasgow, provides lecture notes on stem and germ cell technology for a course on the applications of molecular biotechnology.

General Reports and Articles

Free access to the contents of the journal Stem Cells is available until 1 July 2003.

The April 2003 issue of Molecular Pathology had a review article by S. Preston et al. titled “The new stem cell biology: Something for everyone.”

Stem Cells: Scientific Progress and Future Research Directions is a June 2001 report (available in PDF format) issued by the U.S. Department of Health and Human Services.

Stem Cells and the Future of Regenerative Medicine is a 2002 report available from the National Academies Press.

The United Kingdom Parliament makes available a February 2002 report on stem cell research.

The AAAS Center for Science, Technology, and Congress offers a policy brief on stem cell research. Stem Cell Research and Applications: Monitoring the Frontiers of Biomedical Research is a November 1999 AAAS/Institute for Civil Society report available in PDF format. The report (with terms linked to a glossary) is also available in HTML format from the Counterbalance Meta-Library.

The October 2001 issue of Differentiation was a special double issue devoted to invited reviews and original research on stem cells. A review article by A.-K. Hadjantonakis and V. Papaioannou titled “The stem cells of early embryos” was included.

The 25 February 2000 issue of Science was a special issue on stem cell research and ethics. Included were a review article by F. Watt and B. Hogan titled “Out of Eden: Stem cells and their niches,” a review article by I. Weissman titled “Translating stem and progenitor cell biology to the clinic: Barriers and opportunities,” and a review article by D. van der Kooy and S. Weiss titled “Why stem cells?” The 9 August 2002 issue had a News Focus article by C. Holden and G. Vogel titled “‘Show us the cells,’ U.S. researchers say.”

Focus on Stem Cells is 20 June 2002 Web feature from Nature. A 1 November 2001 Insight feature on stem cells is available to registered users.

Numbered Hypernotes

1. Human embryonic stem cells. Stem Cell Research and Applications: Monitoring the Frontiers of Biomedical Research provides an introduction to human embryonic stem cells. The June 2001 report Stem Cells: Scientific Progress and Future Research Directions has a chapter on the human embryonic stem cell.

2. Early human embryology. Kimball's Biology Pages provides an introduction to embryonic development. The University of Pennsylvania Health System offers a basic overview of embryology. The Division of Anatomy, University of California, San Diego, School of Medicine, makes available D. Rapoport's lecture notes on early embryogenesis. UNSW Embryology is a resource provided by M. Hill, Department of Anatomy, University of New South Wales; sections on human embryology and molecular development are included. D. Linden, Biology Department, Occidental College, Los Angeles, offers lecture notes for a course on developmental cell biology.

3. Mouse embryonic stem cells. W. Richardson, Department of Biology, University College London, makes available lecture notes by H. Smith on mouse embryogenesis and genetics for a course on the biology of development. The June 2001 report Stem Cells: Scientific Progress and Future Research Directions includes a section on mouse embryonic stem cells. The November 2001 issue of Stem Cells had a review by J. Rossant titled “Stem cells from the mammalian blastocyst.” Nature makes available (in PDF format) classic papers on mouse embryonic stem cells from 1981 (“Establishment in culture of pluripotential cells from mouse embryos” by M. Evans and M. Kaufman) and 1984 (“Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines” by A. Bradley, M. Evans, M. Kaufman, and E. Robertson).

4. Culturing embryonic stem cells. The University of Wisconsin's embryonic stem cells Web site offers an illustration and a movie on stem cell culturing, as well as an illustrated 5 November 1998 article by T. Devitt titled “Wisconsin scientists culture elusive embryonic stem cells.” Kimball's Biology Pages offers a presentation on culturing human embryonic stem cells. The September 2002 issue of Nature Biotechnology had a News & Views article by R. Pedersen titled “Feeding hungry stem cells” about the research by M. Richards et al. (“Human feeders support prolonged undifferentiated growth of human inner cell masses and embryonic stem cells”) reported in that issue. The 6 November 1998 issue of Science had a report by J. Thomson et al. titled “Embryonic stem cell lines derived from human blastocysts” (7) and a related Perspective by J. Gearhart titled “New potential for human embryonic stem cells.” Access Excellence offers a 5 November 1998 article by S. Henahan titled “Stem cell breakthrough.”

5. LIF (leukemia inhibitory factor). Swiss-Prot has entries for mouse LIF and human LIF. GeneCards has an entry for LIF with links to other resources. Online Mendelian Inheritance in Man has an entry for LIF. The Ovarian Kaleidoscope Database (OKDB) (provided by the Hsueh Lab in the Division of Reproductive Biology, Stanford University Medical Center) includes an entry for LIF. The January 2003 issue of Stem Cells had a review article by D. Metcalf titled “The unsolved enigmas of leukemia inhibitory factor.”

6. Fibroblast growth factor. The Medical Biochemistry Page, maintained by M. King, Indiana University School of Medicine, offers a presentation on fibroblast growth factors. M. Hill's UNSW Embryology includes a section on fibroblast growth factor. H. Ibelgaufts' Cytokines Online Pathfinder Encyclopaedia has an entry for the fibroblast growth factor family. The Howard Hughes Medical Institute provides a 10 October 2000 article titled “Study reveals how growth factors affect human stem cells” about the research by D. Melton and N. Benvenisty reported in the 10 October 2000 issue of the Proceedings of the National Academy of Sciences (“Effects of eight growth factors on the differentiation of cells derived from human embryonic stem cells” by M. Schuldiner, O. Yanuka, J. Itskovitz-Eldor, D. A. Melton, and N. Benvenisty) (19).

7. “Stemness” and molecular markers. The June 2001 report Stem Cells: Scientific Progress and Future Research Directions includes a section on stem cell markers. The January 2003 issue of Stem Cells had an article by R. Hawley and A. Sobieski titled “Stem cell molecular blueprint: ‘Life, the universe, and everything’.” The 18 October 2002 issue of Science had an article by M. Ramalho-Santos, S. Yoon, Y. Matsuzaki, R. Mulligan, and D. Melton titled “‘Stemness’: Transcriptional profiling of embryonic and adult stem cells” (13) and an article by N. Ivanova et al. titled “A stem cell molecular signature” (14). The Howard Hughes Medical Institute provides a 12 September 2002 article titled “Gene profiling reveals the essence of ‘stemness’” about the research by Melton and colleagues. Nature offers a 13 September 2002 Science Update by K. Powell titled “Stem cells fingerprinted: Gene screen hints at what makes stem cells special.” A mini-review on stem call markers is provided by R&D Systems. Chemicon International offers a presentation on stem cell marker antibodies.

8. SSEAs. The July 2002 issue of Stem Cells had an article by J. K. Henderson et al. titled “Preimplantation human embryos and embryonic stem cells show comparable expression of stage-specific embryonic antigens” (12).

9. The three embryonic germ layers. Germ layers, ectoderm, mesoderm, and endoderm are defined in xrefer's Dictionary of Biology. Developmental Biology Online offers a presentation on the derivatives of the three germ layer cells. Embryo Images offers an presentation on the development of germ cell layers.

10. The UK Stem Cell Bank. The Medical Research Council's stem cells resource page includes an announcement titled “UK stem cell bank launched.” The UK National Institute for Biological Standards and Control provides information about the UK Stem Cell Bank. The 13 September 2002 issue of Science had a News of the Week article by G. Vogel titled “Pioneering stem cell bank will soon be open for deposits.”

11. The American Type Culture Collection is a nonprofit bioresource center that provides biological products, technical services, and educational programs to private industry, government, and academic organizations around the world.

12. A. H. Brivanlou is at the Rockefeller University. F. H. Gage is at the Salk Institute for Biological Studies. R. Jaenisch is at the Whitehead Institute for Biomedical Research. T. Jessell is at the Columbia University College of Physicians and Surgeons. D. Melton is in the Department of Molecular and Cellular Biology, Harvard University. J. Rossant is at Mount Sinai Hospital, Toronto.

References and Notes

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
View Abstract

Stay Connected to Science

Navigate This Article