- Email: [email protected]
Stem cell potential: Can anything make anything?
Dispatch
R7
Stem cell potential: Can anything make anything? Sean J. Morrison
Recent results suggest that stem cells from one tissue can give rise to cells from developmentally unrelated tissues. These results strongly support the idea that certain progenitors retain much broader developmental potentials than expected, and other progenitors may be able to acquire broader potentials in culture. Address: Howard Hughes Medical Institute, Departments of Internal Medicine and Cell and Developmental Biology, 3215 CCGC, University of Michigan, Ann Arbor, Michigan 48109-0934, USA. Current Biology 2001, 11:R7–R9 0960-9822/01/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
It seems like everyone wants to be the first to show that an arm can give rise to a leg, and a nose to an ear. Stem cell potential has become a free-for-all in which the old rules that were understood to govern germ layers and lineage relationships in embryonic development are now being questioned. First, transplanted mixed populations of bone marrow progenitors were observed to give rise to unexpected derivatives, such as muscle and liver (see Table 1 for a summary). Then, progenitors cultured from the central nervous system (CNS) were observed to give rise to many non-neural tissues. And now, oligodendrocyte precursors, previously thought to be committed to the glial lineage, have been found to make neurons under certain culture conditions [1], and purified hematopoietic stem cells have been found to give rise to hepatocytes upon transplantation [2]. These new studies [1,2] are significant, because they have shown that prospectively identified, purified and wellcharacterized progenitor populations can differentiate into unexpected derivatives. Have we recently evolved to abandon classical lineage relationships? Or can stem cells retain potentials that they never exhibit during normal development? Or have we identified an intriguing phenomenon in which the developmental potential of cells can be reprogrammed under specific conditions in culture or after transplantation, without reflecting lineage relationships that exist in normal development? A ‘de-differentiation’ phenomenon was previously documented [3,4] by showing that primordial germ cells become totipotent when they are cultured in the growth factors bFGF and LIF. During normal development, primordial germ cells give rise only to germ cells, and when they are transplanted into blastocysts they do not detectably contribute to any somatic tissue; but Donovan and colleagues found that, when they are first cultured in bFGF
and LIF containing medium, primordial germ cells can give rise to all somatic tissues as well as the germline after transplantation into blastocysts [3,4]. The simplest interpretation of this result is that primordial germ cells can be reprogrammed in culture to exhibit a much broader developmental potential. As the gametes produced by primordial germ cells contribute to all tissues after fertilization, it was proposed that this broadening of developmental potential might represent a normal event in gametogenesis. A number of other groups, however, have recently discovered that the developmental potentials of other types of somatic cell can be broader than expected under a variety of conditions. This suggests that the ability of progenitor cells to exhibit an unexpectedly broad developmental potential in certain circumstances may be a general phenomenon. Bone marrow progenitors, which are best known for making blood cells, have recently been observed to contribute to a variety of tissues upon transplantation. Cells from mouse bone marrow have been observed to give rise to skeletal muscle [5,6], hepatocytes [2,7], and neurons and glia [8]. The bone marrow is known to contain at least two different types of stem cell, including hematopoietic stem cells and mesenchymal stem cells. It is also conceivable that stem cells from other tissues, such as liver, nervous system and muscle, may circulate at low levels and be found in the bone marrow. So which stem cells gave rise to the unexpected derivatives? The cells that gave rise to neurons and glia were derived from cultures of adherent bone marrow stroma, suggesting that they included mesenchymal stem cells [8]. The cells that gave rise to skeletal muscle [6] were isolated based on Hoechst dye exclusion, and so were enriched for hematopoietic stem cells; many different types of stem cell may exclude Hoechst, however, so this population may have contained other types of stem cell as well. Hematopoietic stem cells have now [2] been purified, using thirteen cell-surface markers, and shown to efficiently give rise to hepatocytes. Moreover, in this study hematopoietic stem cells were the only bone marrow cells that gave rise to detectable hepatocytes. This conclusively demonstrates that hematopoietic stem cells have at least one unexpected developmental potential, and supports the possibility that they may be able to give rise to other unexpected derivatives as well. In a reciprocal experiment, progenitors cultured from muscle were observed to give rise to blood cells upon
R8
Current Biology Vol 11 No 1
Table 1 Progenitors that have been found to exhibit unexpected developmental potentials.
Progenitor cells Oligodendrocyte precursors Hematopoietic stem cells Primordial germ cells Bone marrow Hoechst-excluding bone marrow Bone marrow Bone marrow stroma Hoechst-excluding muscle Neurospheres Neurospheres Neurospheres
Unexpected derivative
Progenitors purified?*
Progenitors cultured?†
Occurs during normal development
Reference
Neurons Hepatocytes All tissue types Muscle Muscle Hepatocytes Neurons and glia Blood cells Blood cells Muscle Many somatic lineages
Yes Yes Uncertain No Enriched No No Enriched Yes Yes Yes
Yes No Yes No No No Yes Yes Yes Yes Yes
? ? ? ? ? ? ? ? ? ? ?
[1] [2] [3,4] [5] [6] [7] [8] [6,9] [10] [11] [12]
*This refers to the degree to which the tested progenitors were purified. In some cases, uncultured bone marrow was transplanted without identifying the progenitors that gave rise to the unexpected derivatives. In other cases, neurospheres were cultured to isolate multipotent progenitors. These were considered purified, though the relationship of neurospheres to normal CNS stem cells in vivo is
uncertain. In cases where progenitor populations were isolated based on Hoechst exclusion [5,8], these progenitors may have been purified, but it is also possible that a heterogeneous collection of different types of stem cell were included within the ‘hoechst side-population’ gates. †This refers to whether the progenitors were cultured before or during the assay of developmental potential.
transplantation into irradiated mice [6,9]. One possibility is that some sort of cultured muscle progenitor has hematopoietic potential. Alternatively, the small number of blood cells that contaminated the transplanted cells could have included some hematopoietic stem cells. Thus, additional work will be necessary to identify and characterize the muscle-derived progenitors with hematopoietic potential.
whether their developmental potential has been broadened in culture. When it becomes possible to prospectively identify and purify uncultured CNS stem cells, it will be important to test whether these cells still give rise to blood, muscle and other somatic lineages upon transplantation, or whether they must be cultured first in order to exhibit these potentials. For example, uncultured CNS stem cells could be either injected into irradiated mice, or cultured for variable periods of time first and then injected into irradiated mice to test whether time in culture affects their ability to give rise to blood cells.
Another series of experiments has demonstrated that stem cells cultured from the mouse or human CNS have the potential to give rise to non-neural derivatives. CNS stem cells are often isolated by culturing ‘neurospheres’, which are spheres of multipotent progenitors that grow out of mixed populations of CNS cells in bFGF-containing media. Such neurospheres are thought of as CNS stem cells, but because the neurosphere cells have only been studied in culture, it is not certain whether they have properties that are similar to normal CNS stem cells in vivo. Neurosphere cells have been observed to give rise to blood cells upon transplantation into irradiated mice [10], skeletal muscle upon co-culture with a myogenic cell line or upon transplantation into regenerating muscle in vivo [11], or to derivatives of all three germ layers upon injection into blastocysts or early chick embryos [12]. All of these papers [10–12] concluded that CNS stem cells retain a much broader developmental potential than previously thought. There is, however, concern that CNS progenitors can lose patterning information and acquire a broader developmental potential as a result of being cultured in high concentrations of mitogens such as bFGF [13]. Given the earlier work on primordial germ cells [3,4], this is a serious concern. As a result, it is not known whether neurosphere cells correspond to normal CNS stem cells, or
Oligodendrocyte precursor cells (OPCs) are progenitors from the optic nerve that have been extensively studied and thought to be committed to form glia. The recent demonstration [1] that OPCs can make neurons after being exposed in culture to a specific series of growth factors, including bFGF, is consistent with the idea that neural progenitors can acquire neuronal potential in culture. Indeed, the OPCs not only gave rise to neurons, but to neurospheres as well. This bolsters the concern that neurospheres can sometimes be produced in culture from cells that have properties that are different from multipotent CNS stem cells in vivo. The demonstration that OPCs can give rise to neurons is important, because OPCs can be purified from uncultured postnatal optic nerve by immunopanning. Thus, this is the first demonstration that a well-characterized, prospectively identified nervous system progenitor can acquire a broader developmental potential than previously thought in culture. This supports the idea that specific progenitors can exhibit unexpected developmental plasticity in culture, but cautions us that such plasticity may not be exhibited during normal development.
Dispatch
In order to understand this phenomenon by which cells from one tissue can give rise to developmentally unrelated tissues it will be important to study further the identity of the progenitors involved, as well as the circumstances under which the phenomenon occurs. First, the progenitors that give rise to the unexpected derivatives must be purified and identified either as a particular normal stem cell population or as a novel progenitor type. Second, it must be determined whether freshly purified cells have the unexpected potential in their native state in vivo, or whether such cells only acquire that potential after being cultured. Third, if the uncultured stem cells exhibit the unexpected potential, do they give rise to the unexpected derivatives during normal development or is their unexpected potential never expressed during normal development? It seems likely that some of the potentials described above will be exhibited only after the progenitors have been cultured, while other surprising potentials may exist among normal uncultured stem cells. With so many cells exhibiting unexpected developmental potentials, some researchers have begun to question the classical germ layer origins of some tissues. Is it possible that CNS stem cells give rise to blood, muscle and liver during normal development? This question is testable by fate mapping using Cre recombinase, which can identify all of the progeny produced by a specific lineage of progenitors during normal development, as long as a promoter can be identified that is absolutely specific to that progenitor lineage and can be used to direct expression of the cre gene. Several studies have used this approach to examine the progeny produced by cells that express wnt-1, which include neural stem cells in the midbrain, dorsal neural tube and neural crest. Cells expressing wnt-1 were observed to give rise to many expected cells of the CNS and peripheral nervous system, as well as other neural crest derivatives, but they have not yet been observed to give rise to blood, skeletal muscle, liver, kidney, lung or other unexpected derivatives ([14,15] and S.J.M. and D.J. Anderson, unpublished data). This is not a perfect experiment, in that not all neural stem cells are marked by wnt-1 expression, and it is possible that the CNS stem cells with unusually broad developmental potentials derive from lineages that never expressed wnt-1. Nonetheless, the failure to observe any unusual derivatives from a wide cross-section of neural progenitors in vivo during normal development contrasts with the spate of papers reporting the broad potentials of a variety of cultured CNS stem cells after transplantation. This suggests that the potentials exhibited by cultured CNS stem cells may not be potentials that are exercised during normal development. The question of what cell types are produced by particular progenitors during normal development is distinct from that of what potentials a progenitor can exhibit in culture or after transplantation.
R9
Even if most of the surprising stem cell potentials arise as a result of ‘de-differentiation’ in culture, and do not reflect potentials that are normally exercised in vivo, it will still be scientifically and clinically important to understand the phenomenon. On the other hand, if normal somatic stem cells routinely retain much broader developmental potentials than they exhibit during normal development, this would pose the question as to whether differentiation is primarily environmentally regulated. That is, are the normal lineage relationships that exist between progenitors in different germ layers governed primarily by controlling the locations and interactions of such progenitors? Perhaps the only reason why neural stem cells do not seem to make liver or blood during normal development is that they are prevented from accessing hepatic or hematopoietic microenvironments. However these questions are resolved, we will need to redefine what we mean by developmental potential. Acknowledgements I am supported as an Assistant Investigator of the Howard Hughes Medical Institute. I am grateful for comments from Theodora Ross, Ben Barres, Eric Lagasse and Peter Donovan.
References 1. Kondo T, Raff M: Oligodendrocyte precursor cells reprogrammed to become multipotential CNS stem cells. Science 2000, 289:1754-1757. 2. Lagasse E, Connors H, Al-Dhalimy M, Reitsma M, Dohse M, Osborne L, Wang X, Finegold M, Weissman IL, Grompe M: Purified hematopoietic stem cells can differentiate to hepatocytes in vivo. Nat Medicine 2000, 6:1229-1234. 3. Donovan PJ: Growth factor regulation of mouse primordial germ cell development. Curr Topics Dev Biol 1994, 29:189-225. 4. Matsui Y, Zsebo K, Hogan BLM: Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture. Cell 1992, 70:841-847. 5. Ferrari G, Cusella-De Angelis G, Coletta M, Paolucci E, Stornaiuolo A, Cossu G, Mavilio F: Muscle regeneration by bone marrow-derived myogenic progenitors. Science 1998, 279:1528-1530. 6. Gussoni E, Soneoka Y, Strickland CD, Buzney EA, Khan MK, Flint AF, Kunkel LM, Mulligan RC: Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 1999, 401:390-394. 7. Peterson BE, Bowen WC, Patrene KD, Mars WM, Sullivan AK, Murase N, Boggs SS, Greenberger JS, Goff JP: Bone marrow as a potential source of hepatic oval cells. Science 1999, 284:1168-1170. 8. Kopen GC, Prockop DJ, Phinney DG: Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains. Proc Natl Acad Sci USA 1999, 96:10711-10716. 9. Jackson KA, Mi T, Goodell MA: Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc Natl Acad Sci USA 1999, 96:14482-14486. 10. Bjornson CRR, Rietze RL, Reynolds BA, Magli MC, Vescovi AL: Turning brain into blood: a hematopoietic fated adopted by adult neural stem cells in vivo. Science 1999, 283:534-537. 11. Galli R, Borello U, Gritti A, Minasi MG, Bjornson C, Coletta M, Mora M, De Angelis MG, Fiocco R, Cossu G, Vescovi AL: Skeletal myogenic potential of human and mouse neural stem cells. Nat Neurosci 2000, 3:986-991. 12. Clarke DL, Johansson CB, Wilbertz J, Veress B, Nilsson E, Karlstrom H, Lendahl U, Frisen J: Generalized potential of adult neural stem cells. Science 2000, 288:1660-1663. 13. Gage FH: Stem cells of the central nervous system. Curr Opin Neurobiol 1998, 8:671-676. 14. Jiang X, Rowitch DH, Soriano P, McMahon AP, Sucov HM: Fate of the mammalian cardiac neural crest. Development 2000, 127:1607-1616. 15. Chai Y, Jiang X, Ito Y, Bringas P, Han J, Rowitch DH, Soriano P, McMahon AP, Sucov HM: Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 2000, 127:1671-1679.
R7
Stem cell potential: Can anything make anything? Sean J. Morrison
Recent results suggest that stem cells from one tissue can give rise to cells from developmentally unrelated tissues. These results strongly support the idea that certain progenitors retain much broader developmental potentials than expected, and other progenitors may be able to acquire broader potentials in culture. Address: Howard Hughes Medical Institute, Departments of Internal Medicine and Cell and Developmental Biology, 3215 CCGC, University of Michigan, Ann Arbor, Michigan 48109-0934, USA. Current Biology 2001, 11:R7–R9 0960-9822/01/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
It seems like everyone wants to be the first to show that an arm can give rise to a leg, and a nose to an ear. Stem cell potential has become a free-for-all in which the old rules that were understood to govern germ layers and lineage relationships in embryonic development are now being questioned. First, transplanted mixed populations of bone marrow progenitors were observed to give rise to unexpected derivatives, such as muscle and liver (see Table 1 for a summary). Then, progenitors cultured from the central nervous system (CNS) were observed to give rise to many non-neural tissues. And now, oligodendrocyte precursors, previously thought to be committed to the glial lineage, have been found to make neurons under certain culture conditions [1], and purified hematopoietic stem cells have been found to give rise to hepatocytes upon transplantation [2]. These new studies [1,2] are significant, because they have shown that prospectively identified, purified and wellcharacterized progenitor populations can differentiate into unexpected derivatives. Have we recently evolved to abandon classical lineage relationships? Or can stem cells retain potentials that they never exhibit during normal development? Or have we identified an intriguing phenomenon in which the developmental potential of cells can be reprogrammed under specific conditions in culture or after transplantation, without reflecting lineage relationships that exist in normal development? A ‘de-differentiation’ phenomenon was previously documented [3,4] by showing that primordial germ cells become totipotent when they are cultured in the growth factors bFGF and LIF. During normal development, primordial germ cells give rise only to germ cells, and when they are transplanted into blastocysts they do not detectably contribute to any somatic tissue; but Donovan and colleagues found that, when they are first cultured in bFGF
and LIF containing medium, primordial germ cells can give rise to all somatic tissues as well as the germline after transplantation into blastocysts [3,4]. The simplest interpretation of this result is that primordial germ cells can be reprogrammed in culture to exhibit a much broader developmental potential. As the gametes produced by primordial germ cells contribute to all tissues after fertilization, it was proposed that this broadening of developmental potential might represent a normal event in gametogenesis. A number of other groups, however, have recently discovered that the developmental potentials of other types of somatic cell can be broader than expected under a variety of conditions. This suggests that the ability of progenitor cells to exhibit an unexpectedly broad developmental potential in certain circumstances may be a general phenomenon. Bone marrow progenitors, which are best known for making blood cells, have recently been observed to contribute to a variety of tissues upon transplantation. Cells from mouse bone marrow have been observed to give rise to skeletal muscle [5,6], hepatocytes [2,7], and neurons and glia [8]. The bone marrow is known to contain at least two different types of stem cell, including hematopoietic stem cells and mesenchymal stem cells. It is also conceivable that stem cells from other tissues, such as liver, nervous system and muscle, may circulate at low levels and be found in the bone marrow. So which stem cells gave rise to the unexpected derivatives? The cells that gave rise to neurons and glia were derived from cultures of adherent bone marrow stroma, suggesting that they included mesenchymal stem cells [8]. The cells that gave rise to skeletal muscle [6] were isolated based on Hoechst dye exclusion, and so were enriched for hematopoietic stem cells; many different types of stem cell may exclude Hoechst, however, so this population may have contained other types of stem cell as well. Hematopoietic stem cells have now [2] been purified, using thirteen cell-surface markers, and shown to efficiently give rise to hepatocytes. Moreover, in this study hematopoietic stem cells were the only bone marrow cells that gave rise to detectable hepatocytes. This conclusively demonstrates that hematopoietic stem cells have at least one unexpected developmental potential, and supports the possibility that they may be able to give rise to other unexpected derivatives as well. In a reciprocal experiment, progenitors cultured from muscle were observed to give rise to blood cells upon
R8
Current Biology Vol 11 No 1
Table 1 Progenitors that have been found to exhibit unexpected developmental potentials.
Progenitor cells Oligodendrocyte precursors Hematopoietic stem cells Primordial germ cells Bone marrow Hoechst-excluding bone marrow Bone marrow Bone marrow stroma Hoechst-excluding muscle Neurospheres Neurospheres Neurospheres
Unexpected derivative
Progenitors purified?*
Progenitors cultured?†
Occurs during normal development
Reference
Neurons Hepatocytes All tissue types Muscle Muscle Hepatocytes Neurons and glia Blood cells Blood cells Muscle Many somatic lineages
Yes Yes Uncertain No Enriched No No Enriched Yes Yes Yes
Yes No Yes No No No Yes Yes Yes Yes Yes
? ? ? ? ? ? ? ? ? ? ?
[1] [2] [3,4] [5] [6] [7] [8] [6,9] [10] [11] [12]
*This refers to the degree to which the tested progenitors were purified. In some cases, uncultured bone marrow was transplanted without identifying the progenitors that gave rise to the unexpected derivatives. In other cases, neurospheres were cultured to isolate multipotent progenitors. These were considered purified, though the relationship of neurospheres to normal CNS stem cells in vivo is
uncertain. In cases where progenitor populations were isolated based on Hoechst exclusion [5,8], these progenitors may have been purified, but it is also possible that a heterogeneous collection of different types of stem cell were included within the ‘hoechst side-population’ gates. †This refers to whether the progenitors were cultured before or during the assay of developmental potential.
transplantation into irradiated mice [6,9]. One possibility is that some sort of cultured muscle progenitor has hematopoietic potential. Alternatively, the small number of blood cells that contaminated the transplanted cells could have included some hematopoietic stem cells. Thus, additional work will be necessary to identify and characterize the muscle-derived progenitors with hematopoietic potential.
whether their developmental potential has been broadened in culture. When it becomes possible to prospectively identify and purify uncultured CNS stem cells, it will be important to test whether these cells still give rise to blood, muscle and other somatic lineages upon transplantation, or whether they must be cultured first in order to exhibit these potentials. For example, uncultured CNS stem cells could be either injected into irradiated mice, or cultured for variable periods of time first and then injected into irradiated mice to test whether time in culture affects their ability to give rise to blood cells.
Another series of experiments has demonstrated that stem cells cultured from the mouse or human CNS have the potential to give rise to non-neural derivatives. CNS stem cells are often isolated by culturing ‘neurospheres’, which are spheres of multipotent progenitors that grow out of mixed populations of CNS cells in bFGF-containing media. Such neurospheres are thought of as CNS stem cells, but because the neurosphere cells have only been studied in culture, it is not certain whether they have properties that are similar to normal CNS stem cells in vivo. Neurosphere cells have been observed to give rise to blood cells upon transplantation into irradiated mice [10], skeletal muscle upon co-culture with a myogenic cell line or upon transplantation into regenerating muscle in vivo [11], or to derivatives of all three germ layers upon injection into blastocysts or early chick embryos [12]. All of these papers [10–12] concluded that CNS stem cells retain a much broader developmental potential than previously thought. There is, however, concern that CNS progenitors can lose patterning information and acquire a broader developmental potential as a result of being cultured in high concentrations of mitogens such as bFGF [13]. Given the earlier work on primordial germ cells [3,4], this is a serious concern. As a result, it is not known whether neurosphere cells correspond to normal CNS stem cells, or
Oligodendrocyte precursor cells (OPCs) are progenitors from the optic nerve that have been extensively studied and thought to be committed to form glia. The recent demonstration [1] that OPCs can make neurons after being exposed in culture to a specific series of growth factors, including bFGF, is consistent with the idea that neural progenitors can acquire neuronal potential in culture. Indeed, the OPCs not only gave rise to neurons, but to neurospheres as well. This bolsters the concern that neurospheres can sometimes be produced in culture from cells that have properties that are different from multipotent CNS stem cells in vivo. The demonstration that OPCs can give rise to neurons is important, because OPCs can be purified from uncultured postnatal optic nerve by immunopanning. Thus, this is the first demonstration that a well-characterized, prospectively identified nervous system progenitor can acquire a broader developmental potential than previously thought in culture. This supports the idea that specific progenitors can exhibit unexpected developmental plasticity in culture, but cautions us that such plasticity may not be exhibited during normal development.
Dispatch
In order to understand this phenomenon by which cells from one tissue can give rise to developmentally unrelated tissues it will be important to study further the identity of the progenitors involved, as well as the circumstances under which the phenomenon occurs. First, the progenitors that give rise to the unexpected derivatives must be purified and identified either as a particular normal stem cell population or as a novel progenitor type. Second, it must be determined whether freshly purified cells have the unexpected potential in their native state in vivo, or whether such cells only acquire that potential after being cultured. Third, if the uncultured stem cells exhibit the unexpected potential, do they give rise to the unexpected derivatives during normal development or is their unexpected potential never expressed during normal development? It seems likely that some of the potentials described above will be exhibited only after the progenitors have been cultured, while other surprising potentials may exist among normal uncultured stem cells. With so many cells exhibiting unexpected developmental potentials, some researchers have begun to question the classical germ layer origins of some tissues. Is it possible that CNS stem cells give rise to blood, muscle and liver during normal development? This question is testable by fate mapping using Cre recombinase, which can identify all of the progeny produced by a specific lineage of progenitors during normal development, as long as a promoter can be identified that is absolutely specific to that progenitor lineage and can be used to direct expression of the cre gene. Several studies have used this approach to examine the progeny produced by cells that express wnt-1, which include neural stem cells in the midbrain, dorsal neural tube and neural crest. Cells expressing wnt-1 were observed to give rise to many expected cells of the CNS and peripheral nervous system, as well as other neural crest derivatives, but they have not yet been observed to give rise to blood, skeletal muscle, liver, kidney, lung or other unexpected derivatives ([14,15] and S.J.M. and D.J. Anderson, unpublished data). This is not a perfect experiment, in that not all neural stem cells are marked by wnt-1 expression, and it is possible that the CNS stem cells with unusually broad developmental potentials derive from lineages that never expressed wnt-1. Nonetheless, the failure to observe any unusual derivatives from a wide cross-section of neural progenitors in vivo during normal development contrasts with the spate of papers reporting the broad potentials of a variety of cultured CNS stem cells after transplantation. This suggests that the potentials exhibited by cultured CNS stem cells may not be potentials that are exercised during normal development. The question of what cell types are produced by particular progenitors during normal development is distinct from that of what potentials a progenitor can exhibit in culture or after transplantation.
R9
Even if most of the surprising stem cell potentials arise as a result of ‘de-differentiation’ in culture, and do not reflect potentials that are normally exercised in vivo, it will still be scientifically and clinically important to understand the phenomenon. On the other hand, if normal somatic stem cells routinely retain much broader developmental potentials than they exhibit during normal development, this would pose the question as to whether differentiation is primarily environmentally regulated. That is, are the normal lineage relationships that exist between progenitors in different germ layers governed primarily by controlling the locations and interactions of such progenitors? Perhaps the only reason why neural stem cells do not seem to make liver or blood during normal development is that they are prevented from accessing hepatic or hematopoietic microenvironments. However these questions are resolved, we will need to redefine what we mean by developmental potential. Acknowledgements I am supported as an Assistant Investigator of the Howard Hughes Medical Institute. I am grateful for comments from Theodora Ross, Ben Barres, Eric Lagasse and Peter Donovan.
References 1. Kondo T, Raff M: Oligodendrocyte precursor cells reprogrammed to become multipotential CNS stem cells. Science 2000, 289:1754-1757. 2. Lagasse E, Connors H, Al-Dhalimy M, Reitsma M, Dohse M, Osborne L, Wang X, Finegold M, Weissman IL, Grompe M: Purified hematopoietic stem cells can differentiate to hepatocytes in vivo. Nat Medicine 2000, 6:1229-1234. 3. Donovan PJ: Growth factor regulation of mouse primordial germ cell development. Curr Topics Dev Biol 1994, 29:189-225. 4. Matsui Y, Zsebo K, Hogan BLM: Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture. Cell 1992, 70:841-847. 5. Ferrari G, Cusella-De Angelis G, Coletta M, Paolucci E, Stornaiuolo A, Cossu G, Mavilio F: Muscle regeneration by bone marrow-derived myogenic progenitors. Science 1998, 279:1528-1530. 6. Gussoni E, Soneoka Y, Strickland CD, Buzney EA, Khan MK, Flint AF, Kunkel LM, Mulligan RC: Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 1999, 401:390-394. 7. Peterson BE, Bowen WC, Patrene KD, Mars WM, Sullivan AK, Murase N, Boggs SS, Greenberger JS, Goff JP: Bone marrow as a potential source of hepatic oval cells. Science 1999, 284:1168-1170. 8. Kopen GC, Prockop DJ, Phinney DG: Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains. Proc Natl Acad Sci USA 1999, 96:10711-10716. 9. Jackson KA, Mi T, Goodell MA: Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc Natl Acad Sci USA 1999, 96:14482-14486. 10. Bjornson CRR, Rietze RL, Reynolds BA, Magli MC, Vescovi AL: Turning brain into blood: a hematopoietic fated adopted by adult neural stem cells in vivo. Science 1999, 283:534-537. 11. Galli R, Borello U, Gritti A, Minasi MG, Bjornson C, Coletta M, Mora M, De Angelis MG, Fiocco R, Cossu G, Vescovi AL: Skeletal myogenic potential of human and mouse neural stem cells. Nat Neurosci 2000, 3:986-991. 12. Clarke DL, Johansson CB, Wilbertz J, Veress B, Nilsson E, Karlstrom H, Lendahl U, Frisen J: Generalized potential of adult neural stem cells. Science 2000, 288:1660-1663. 13. Gage FH: Stem cells of the central nervous system. Curr Opin Neurobiol 1998, 8:671-676. 14. Jiang X, Rowitch DH, Soriano P, McMahon AP, Sucov HM: Fate of the mammalian cardiac neural crest. Development 2000, 127:1607-1616. 15. Chai Y, Jiang X, Ito Y, Bringas P, Han J, Rowitch DH, Soriano P, McMahon AP, Sucov HM: Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 2000, 127:1671-1679.