Cell replacement therapy
The application of ES cells that has received the most attention in recent years is as a novel source of cells for cell replacement therapy for the treatment of a wide range of debilitating diseases. Type I diabetes, cardiovascular disease, Parkinson's disease, blood cell diseases, and certain types of liver disease are considered candidates for cell replacement therapy. As indicated in this review, transplantation of specific ES-cell-derived cells into pre-clinical models of human disease is already underway. The success to date in delivering appropriate cell populations and demonstrating their functional capacity in vivo is limited. For this application to advance to the next step, several outstanding issues need to be addressed.
The first relates to the type and number of cells to be transplanted. One of the most important questions is at what stage of maturation should cells be transplanted. The answer to this will depend to a large extent on the lineage under investigation. For example, when considering hepatocyte replacement therapy, one might isolate and transplant relatively mature cells that are known to retain substantial proliferative and regenerative potential (Overturf et al. 1997
). Similarly, when replacing pancreatic 
-cell function, it may be essential to develop methods to induce end-stage maturation of the cells prior to transplantation. In contrast, sustained replacement of the hematopoietic system will require transplantation of the most immature cells of the system, the HSC. For cardiomyocyte replacement, different developmental stages will need to be tested. The number of cells to be transplanted will also depend on the lineage and stage of development. To address these issues, it will be important to first define the mechanisms controlling development of the desired lineage. With this information, optimal conditions for the generation of the appropriate cell populations can be developed.
A second concern when considering cell replacement therapy with ES-cell-derived populations is the danger associated with contamination of the graft with residual undifferentiated ES cells. Transplantation of ES cells can result in the development of teratomas (Evans and Kaufman 1983; Thomson et al. 1998). Techniques that allow for the selection of the lineage to be transplanted represent the first step in reducing the number of ES cells in the graft (Klug et al. 1996; Li et al. 1998). A second approach is the use of strategies to eliminate undifferentiated ES cells from the graft (Billon et al. 2002). With a combination of positive and negative selection, it should be possible to reproducibly generate grafts free of undifferentiated ES cells.
A third obstacle to be overcome is donor/recipient compatibility and graft rejection. One solution to this problem would be to bank large numbers of ES cell lines that would include a significant portion of the histocompatibility types in the population. A second strategy is to generate individualized ES cell lines through SCNT (Hochedlinger and Jaenisch 2003). With this approach, the cells used for transplantation would be genetically identical to those of the patient. Both mouse and human ES cells have been generated using SCNT (Munsie et al. 2000; Wakayama et al. 2001; Rideout et al. 2002; Hwang et al. 2004). An ethical and practical limitation of SCNT is the requirement of oocytes. The development of oocytes from differentiated ES cells may provide one solution to this problem (Hubner et al. 2003).
Drug discovery
In addition to developmental biology and cell-based therapy, the ES cell model has widespread applications in the areas of drug discovery and drug development. With respect to drug development, cell types such as cardiomyocytes and hepatocytes generated from hES cells could provide ideal populations for predictive toxicology. These human cells could reveal the toxicity of certain drugs that might not be detected using conventional assays that rely on animal models. Other advantages of such ES-cell-based assays include (1) the requirement of significantly less test compound than in vivo assays, enabling earlier use in the drug discovery phase, resulting in earlier prioritization between drug leads; and (2) the availability of cell lines with different genotypes that could be used to identify clinically important pharmacogenomic specific drug responses and/or toxicities.
ES cells offer many different strategies to develop drugs that can be used for regenerative medicine, that is, the therapeutic regrowth and/or repair of damaged cells. A unique strength of the ES cell system is the ability to engineer the ES cells to enable one to easily quantify the effect of drugs. This strategy has been discussed previously in this review, and could be used for drug discovery applications by introducing reporter molecules, such as GFP, into genes indicative of the development of specific progenitor populations or into genes associated with cell maturation and function. These customized assays could be used to identify those compounds that induce the growth and/or maturation of the cell types of interest. This capability offers a unique opportunity to develop clinically relevant commercial-grade pharmaceutical screening assays for certain human cell types that are not possible to produce by any other approach. An example of this approach would be the development of an ES cell line engineered to express an appropriate reporter in genes associated with pancreatic development or -cell maturation. Screens with such ES cell lines could lead to the identification of molecules that promote the growth of specific pancreatic progenitors or those that enhance the maturation of the insulin-producing -cells.
In addition to genetically modified cells, hES cell lines with genotypes characteristic of various genetic diseases could not only provide novel insights into the mechanisms of the disease process, but also offer powerful screening systems for developing drugs for treating those diseases. Given the successes in lineage-specific differentiation of ES cells in recent years, these approaches are already under preliminary evaluation. As the conditions for the differentiation of hES cells are optimized, ES-cell-based screens could be incorporated into drug discovery platforms of many biotechnology and pharmaceutical companies.
Acknowledgments
I thank Jim Palis, Ralph Snodgrass, Hans Snoeck, and members of my lab for many helpful discussions and critical reading of the manuscript and Dr. Steve Kattman for assistance with generating the figures.
Footnotes
Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/gad.1303605.
1Correspondence.
E-MAIL gordon.keller@mssm.edu
; FAX (212) 803-6740.
Answers to all your Biology Questions
