Molecular Insights into the Function, Fate, and ... - Wiley Online Library

This material is protected by U.S. Copyright law. Unauthorized reproduction is prohibited. For reprints contact: [email protected]

Meeting Report Molecular Insights into the Function, Fate, and Prospects of Stem Cells Vinagolu K. Rajasekhar,a Mohan C. Vemurib a

Memorial Sloan-Kettering Cancer Center, New York, New York, USA; b Reprogenetics LLC, West Orange, New Jersey, USA

Key Words. Adult stem cells • Embryonic stem cells • Cancer stem cells • Differentiation Hematopoietic stem cells • Self-renewal • Stem cell niche • Stem cell markers

Abstract This article forms a review and an appraisal of the third annual meeting of the International Society for Stem Cell Research (http://www.isscr.org), held in San Francisco on June 23–25, 2005. The focus of the meeting was recent advances in stem cell biology. More than 2,000 scientists from around the world met to discuss stem cell research, clinical applications, and the ethical hurdles facing the field. Major topics highlighted during the meeting included the self-renewal and differentiation of embryonic stem cells as well as adult stem cells. Presentations included diverse topics such as cancer stem cells, tis-

sue-specific stem cells, technology development, and clinical aspects of stem cells. Given the excitement the field has generated, linking basic stem cell research and clinical applications was paramount for discussion at the meeting. With the current resources in molecular biology research, improvements in genetic engineering, postgenomic capabilities, and biotechnological advances, it appears timely that stem cell biology research is headed toward making a major therapeutic contribution to human health. Stem Cells 2005;23:1212–1220

Introduction

human system. However, the emerging differences between mouse and human embryonic stem (ES) cells in the requirements for growth and in the behavior observed when grown under certain conditions unveiled the necessity for directly studying human ES cells. In this context, the experiments to derive new human ES cell lines from specific patients using somatic cell nuclear transfer were certainly a translational highlight in stem cell research. For the convenience of the reader and to be consistent with the sessions, the meeting overview is summarized into five sections: (1) Embryonic Stem Cells and Their Maintenance; (2) Adult Stem Cells and Stem Cell Niches; (3) Cancer Stem Cells; (4) Stem Cell Regulatory Networks, Lineage Commitment, and Differentiation; and (5) Gene Therapy, Tissue Engineering, and Clinical Aspects. Although the scope of this overview does not permit coverage of each presentation from the meeting, we attempted to accommodate the exem-

Progress in stem cell research has long been hindered by two pervasive problems—the inability to produce various stem cell types in sufficient quantities and the lack of a methodology to derive differentiated cells from stem cells in a predictable manner. Various aspects of stem cell biology were discussed with these goals in mind. These studies have embraced contemporary technological advances to manipulate large-scale production and enrichment of the stem cells, analysis of their global expression at genomic and postgenomic/proteome levels, high-throughput whole-embryo in situ hybridization screens, short hairpin RNA (shRNA) screens, gene corrections using engineered zinc-finger nucleases, nuclear cloning, and genetic/epigenetic reprogramming. Almost three decades of experimental findings have been accumulated from mouse cell systems in attempts to extend the knowledge to the

Correspondence: Vinagolu K. Rajasekhar, Ph.D., Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021, USA. Telephone: 212-639-8510; Fax: 212-717-3627; e-mail: [email protected]; and Mohan C. Vemuri, Ph.D., Reprogenetics LLC, 101 Old Short Hills Rd., Suite 501, West Orange, New Jersey 07052, USA. Telephone: 973-322-6236; Fax: 973322-6235; e-mail: [email protected] Received July 26, 2005; accepted for publication July 28, 2005; first published online in Stem Cells EXPRESS August 2, 2005. ©AlphaMed Press 1066-5099/2005/$12.00/0 doi: 10.1634/stemcells.2005-0336

Stem Cells 2005;23:1212–1220 www.StemCells.com

Rajasekhar, Vemuri plary conceptual themes as best we could. The reader is directed to International Society for Stem Cell Research online support material for easy access to relevant abstracts (http://www.isscr.org).

Embryonic Stem Cells and Their Maintenance ES cells can self-renew indefinitely in culture and yet maintain their ability to differentiate into multiple lineages. Although the science of ES cells was the focus of this meeting, the ethics of human ES cell research received its share of the spotlight. Because human ES cells are derived from the inner cell mass of in vitro–fertilized blastocyst embryos, the controversy surrounding this issue warranted an entire public meeting convened solely to address the ethical issues of human ES cell research. It should be pointed out that extant human ES cell lines were derived, with the consent of donors, from spare in vitro–fertilized embryos at fertility clinics that would otherwise have been discarded. Although the isolation of human ES cells has unveiled new avenues for studies on human basic stem cell biology, early human embryonic development, and therapeutic potential in humans, the wealth of information obtained from research on mouse and other model systems, such as zebrafish and Drosophila, is nonetheless providing strong foundations for the future of stem cell research. James A. Thomson (University of Wisconsin-Madison Medical School, Madison, WI), who derived the first human ES cell lines in 1998, presented the latest developments from his laboratory in developing feeder-free culture of human ES cells. His findings underscored a critical difference in the self-renewal mechanisms between mouse and human ES cells. In mouse ES cells, the leukemia inhibitory factor maintains self-renewal by synergizing with bone morphogenetic protein (BMP)-4. In contrast, the BMPs induce differentiation of human ES cells, and the cells grown in unconditioned medium are exposed to activation of BMP signaling. The main finding is that basic fibroblast growth factor (bFGF) is the primary growth factor that maintains ES cells in the absence of feeders. The bFGF at 40 ng/ml synergizes with noggin-like BMP antagonists and maintains an undifferentiated self-renewal of human ES cells. Higher concentrations of bFGF, such as 100 ng/ml, suppressed the BMP signaling to the levels comparable with those of conditioned medium, and such high FGF concentrations maintained the undifferentiated proliferation. Thomson further reported simplified protocols for the culture of human ES cells in feeder-free conditions that take human ES cells a step closer to clinical utility since the feeders used were typically of mouse origin. Peter W. Andrews (University of Sheffield, Sheffield, U.K.) discussed the maintenance of the pluripotent state of human ES cells and emphasized that the selective pressures in cell culture acting on the mechanisms that control self-renewal versus differentiation may lead to the appearance of genetically variant lines better adapted to growth under prevailing culture conditions. His study showed how expression of key human ES cell markers

1213

SSEA3, SSEA4, TRA-1-60, GCTM2, THY-1, MHC, ALP, Oct4, and Nanog changes as the cells progress toward differentiation. Roger Pedersen (University of Cambridge, Cambridge, U.K.) discussed mechanisms of pluripotency and epigenesis in human ES cells. His studies showed that the transforming growth factor (TGF)-beta family members activin and nodal maintain expression of pluripotency markers in the absence of feeder layers, suggesting the role of TGF-beta signaling in human ES pluripotency mechanisms. Inhibition of activin/nodal receptor signaling by a chemical antagonist further confirms the essential role of such factors in the maintenance of human ES cell pluripotency. In addition, human ES cells demonstrated a substantial epigenetic stability, as revealed by monoallelic expression and maintenance of gametic methylation imprints. Laurie Boyer, associated with Rudolf Jaenisch (Massachusetts Institute of Technology, Cambridge, MA), and Richard Young (Whitehead Institute for Biomedical Research, Cambridge, MA) presented ways to identify regulatory networks that govern pluripotency and stem cell maintenance. To this goal, they generated the first genome-wide map of protein–DNA interactions in human cells by localizing the RNA polymerase II across the entire nonrepeat region of the human genome. Combined with the available expression data, transcription factors, such as Oct4 and Nanog, have been demonstrated to establish and maintain undifferentiated ES cells. Putative signaling networks that govern human ES cell pluripotency and self-renewal were postulated from these studies. On the other hand, Natalia Ivanova and Ihor Lemischka (Princeton University, Princeton, NJ) reported a functional genomic strategy to systematically identify regulators of self-renewal in mouse ES cells. Using Systems Biology approaches, they first analyzed genome-wide transcriptional changes that occur after treatment with retinoic acid (RA) using Affymetrix microarrays (Affymetrix, Santa Clara, CA, http://www.affymetrix.com). Using an shRNA loss-of-function strategy, they identified that the shRNAs for self-renewal transcription factors Oct4 and Nanog cause differentiation into trophectodermal and endomesodermal lineages, respectively. Furthermore, Sox2, Tbx3, Esrrb, Tcl1, Dppa4, Mm.5829, Mm.276044, and Mm.219358 were validated as additional factors involved in self-renewal. Dynamics of global gene expression were followed by the downregulation of individual genes, and three distinct clusters of perturbed gene expression were identified. The first cluster was found to be a function of downregulating Nanog, Oct 4, or Sox 2; the second was a function of downregulation of Esrrb, Tbx3, Tcl1, Mm. 5829, or Dppa4; and the third was a function of all or most of the shRNAs. Such studies would assist in predicting the architecture of transcriptional regulatory networks during ES cell maintenance. Shawn M. Burgess (Genome Technology Branch, NHGRI/ NIH, Bethesda, MD) analyzed the regulatory network of Oct4 in mouse ES cells and its homologue, POU2, in zebrafish embryos.

1214

Gain or loss-of-function for Oct4 and POU2 levels were induced in each of the species and then followed with transcriptional profiling using microarrays. This allowed for a meta-analysis of gene-expression profiles across the species, with the data facilitating the identification of several common genes in a “core pathway” encoding signaling molecules and/or transcription factors directly responsive to Oct4/POU2 levels. These pathways seem to be conserved over 300 million years of divergent evolution. On the other hand, Shin-Ichi Nishikawa (RIKEN Center for Developmental Biology, Kobe, Japan) presented modeling of in vitro ES cell differentiation using microarray analysis and developed a database based on transcription of Affymetrix oligonucleotide arrays. His studies aided the definition of growth factor requirements for distinct stages of development, to develop novel surface markers to identify intermediates, and to discover novel genes related to embryogenesis. In addition, he presented a reanalysis of data sets that had previously been used to argue for the presence of stem-specific genes and was able to show that previous conclusions were based on inappropriate analyses and that in fact the data sets argued for the absence of such genes. Epigenetic variability exists between and within human ES cell lines. Lorraine Young (University of Nottingham, Nottingham, U.K.) analyzed a variety of human ES cell lines (BG01, NCL1, HES-2, and H7) for genome-wide methylation profiles of gene-rich CpG islands. Pairwise comparisons revealed significant epigenetic distances among them. Stochastic and unpredictable epigenetic changes were also observed and were associated with the passage numbers. On the other hand, epigenetic mechanisms, such as histone acetylation, have been reported to control the expression of Oct4 and Nanog. Rong Lu and Ihor Lemischka (Princeton University) investigated stem cell epigenetics by analyzing genome-wide histone acetylation and methylation patterns in mouse ES cells using the ChIP-on-Chip methodology. They observed a dynamic change of histone acetylation and methylation patterns during the induction of ES cell differentiation. They found that a global nonspecific erasure of the histone acetylation pattern takes place before a specific new pattern is built. Furthermore, they showed that single gene knock-downs of Oct4 and Nanog induce rapid genome-wide changes in histone acetylation patterns. Such studies contribute to our understanding of genomic reprogramming mechanisms. Stuart Orkin (Dana Farber Cancer Institute, Harvard Medical School, Boston) described protein networks controlling ES cells. It was shown that Nanog interacts with several different nuclear proteins. Oct4 is not a consistent component of Nanogassociated proteins. This network of proteins might define a pluripotent state of ES cells. In this context, Anthony Whetton (University of Manchester, Manchester, U.K.) applied state-ofthe-art advances in mass spectrometry to identify post-translational modifications, such as phosphorylation events, as regulatory processes in stem cells.

Third Annual Meeting of the ISSCR

Adult Stem Cells and Stem Cell Niches Stem cells found in adult tissues are referred to as adult stem cells. They exist in a dormant state in a so-called stem cell niche, maintaining cellular homeostasis or repairing damaged tissue upon stimulation. Adult stem cells have the capacity to self-renew and generate functionally differentiated cells that replenish lost cells in the organ throughout the lifespan of the organism. How a new stimulus activates the exit of adult stem cells from their specialized microenvironment/niche is yet to be understood. Once activated, the adult stem cell changes into a committed transit-amplifying cell and undergoes a limited number of divisions before differentiating. The stem cell niche is a dynamic multicellular structural unit consisting of a stem cell and its surrounding cells. Physiologically, the stem cell niche is thought to balance the dual processes of self-renewal and lineage commitment of adult stem cells by enabling the niche-specific spatiotemporal interactions. Although the molecular details are obscure, adult stem cells maintain regenerative tissues such as skin epidermis, gastrointestinal mucosa, hematopoietic system, and bladder epithelial system. Elaine Fuchs (Rockefeller University, Howard Hughes Medical Institute, New York) presented the molecular steps involved in skin stem cell activation and commitment. Her group developed a novel approach to fluorescently tag and isolate multipotent epithelial stem cells from both quiescent and activated states in mouse skin. Using embryonic skin stem cells, she illuminated the work of her laboratory associate, Terry Lechler, as the first example of asymmetric cell divisions in a mammalian tissue in which the Par3, Inscuteable, and Lgn/Pins complex is involved. Moreover, these divisions are dependent on beta-1 Integrin and alphaCatenin. This work demonstrates the distinct role of the basement membrane and the integrins that provide a natural mechanism for asymmetrically partitioning cell fate determinants. In addition, Fuchs presented their work on the complex signaling mechanisms in adult hair follicle stem cells, which involve multiple members of the LEF1/TCF proteins, Wnts, BMPs, and TGF-beta families. Gregor Adams and David Scadden (Harvard Medical School) proposed that calcium-sensing receptors may have an important role in engraftment of stem cells or recruitment of hematopoietic stem cells (HSCs) into the bone marrow (BM) niche. They showed that normal primitive HSCs use stromalderived factor domains for localization in BM, suggesting specialized microvasculature in BM as the actual niche. Their studies showed that GSα−/− cells do not translocate and engraft in BM—this is a deficiency of these stem cells and a failure in their ability to engraft. Therefore, GSα may be a pharmacological target for altering stem cell function. Thus, they used pharmacologic manipulation to target the stem cell niche rather than the stem cells themselves, providing in vivo therapeutic effects on HSCs that can alter the outcome of cytotoxic chemotherapy and stem cell transplantation. Such studies on stem cell niches may offer new opportunities for stem cell–based therapies.

Rajasekhar, Vemuri Margaret Goodell (Baylor College of Medicine, Houston) reported interesting data on molecular regulation of HSC activation. She focused on the very early events involved in activation of HSCs from quiescence. By enriching mouse HSCs through isolation of side populations, she identified rare cDNA sequences in HSCs by suppressive subtractive hybridization followed by concatenation cDNA sequencing method and finally mapping to the mouse genome. The sequences were classified into two groups: one containing multiple intronic sequences and the other containing only exons. The former group is enriched for genes related to DNA binding, ribonucleoproteins, splicing components, etc., compared with the latter group. It was not immediately clear if the first group represented novel splice variants, noncoding RNAs, or unique properties of stem cells due to improper splicing. But levels of the unspliced gene products were found to decrease after activation of HSCs. Microarray analysis of gene expression profiles after HSC activation over different time periods validated the increase in expression of genes related to splicing, regulators of cell cycle, etc. Thus, the quiescent HSCs harbor substantial intronic RNA, and the activation of HSCs leads to induction of mRNA splicing. These data imply that an additional tier of control at the post-transcriptional level of gene expression exists in the maintenance of quiescence in HSCs. Sean J. Morrison (University of Michigan, Ann Arbor, MI) presented on a new family of markers for isolating HSCs. His laboratory demonstrated that the SLAM family of cell-surface receptors is differentially expressed among early hematopoietic progenitors in a way that correlates with primitiveness and that the simple combination of CD150 + CD48 −CD41− markers can highly purify HSCs. This is the first known family of receptors whose combinatorial expression distinguishes stem cells from progenitors. By simplifying the purification of HSCs, these new markers make it possible to identify the localization of HSCs in hematopoietic tissues, revealing endothelial niches for stem cells. Emmanuelle Passegue from Irving Weissman’s group (Stanford University School of Medicine, Stanford, CA) reported cell-cycle regulation defining cell fate decisions in HSCs. As HSCs exit quiescence to enter cell-cycle, they lose efficient in vivo engraftment activity and regulate specific cell-cycle proteins that are associated with either developmental outcome (self-renewal vs. differentiation) or developmental fate (myeloid vs. lymphoid). Stewart Fabb (Peter MacCallum Cancer Center, Melbourne, Australia) in association with Louise Purton (Massachusetts General Hospital, Boston) has identified a novel role for the homeobox (Hox) gene Hoxa1 in the self-renewal of blood-forming adult stem cells. Although Hoxa1 has been shown to enhance HSC self-renewal, Hoxa1-T (the alternatively spliced transcript lacking the homeobox domain) is proposed to promote differentiation of HSCs. The opposing effects of these two transcripts suggest a novel mechanism whereby Hoxa1 can self-regulate its function in hematopoiesis, with Hoxa1-T perhaps serving to restrict the selfrenewal, and hence the potentially oncogenic function of Hoxa1.

1215

Michael Cooke (Genomics Institute of the Novartis Research Foundation, San Diego) reported c-myb and p300 as key regulators of mouse HSC renewal and differentiation. He used mice with a mutation in the transcription factor c-Myb in the transactivation domain, which partially disrupts the association of c-myb with the transactivator p300. Mutant animals display several alterations in hematopoiesis, including specific blocks in T- and B-cell development, a lack of eosinophils, reduced red blood cell levels, and greatly elevated levels of megakaryocytes and platelets, highlighting c-myb as a pleiotropic regulator of HSC differentiation. Surprisingly, in contrast to c-myb–null animals, which die of reduced numbers of HSCs, the number and the rate of proliferation of HSCs in these c-myb mutants was increased several-fold. It was suggested that c-myb controls HSC proliferation and that c-myb–p300 interactions are required for differentiation of HSCs and their progeny. It opens up the possibility that molecules affecting the c-myb–p300 interaction may be of clinical significance for HSC expansion. Andreas Trumpp and associates (Swiss Institute for Experimental Cancer Research and University of Lausanne, Epalinges, Switzerland) reported that the transcription factor and oncoprotein c-Myc controls the balance between HSC self-renewal and differentiation. This was reported to be achieved at the interface between the niche and the non-niche microenvironment of HSCs apparently by controlling the expression of N-cadherin and various integrin receptors. Such observations have also wide implications for tumorigenesis, especially in the context of stem-like cancer cells. David Traver (University of California–San Diego, San Diego) used the zebrafish model system and demonstrated that the earliest blood-forming cells are committed to the erythroid lineage, whereas multipotent HSCs seem to arise later in the aorta-gonadmesonephros region. Both David Traver and Leonard Zon in association with Caroline Burns identified the important function of Notch activation of Runx1-dependent regulation of HSC expansion and further developmental specification of HSC fate. Kateri Ann Moore (Princeton University) and Yoav Soen and colleagues (Stanford University) have taken genomic and also proteomic approaches to elucidate the role of Wnt and Wnt/Notch signaling pathways in hematopoietic and neural stem cell niches, respectively. Masatake Osawa from Shin-Ichi Nishikawa’s group (Riken Center for Developmental Biology, Kobe, Japan) reported the vital role of Notch signaling in the maintenance of melanoblasts (Mbs), the precursors for pigment cells in the embryonic epidermis and melanocyte stem cells (MSCs) in the hair follicle. Both the Mbs and MSCs have upregulated expression of Notch genes. Inhibition of Notch signaling resulted in apoptotic elimination of Mbs, and this effect could be rescued with the forced expression of Hes1, a direct downstream target of Notch. Transgenic mice expressing a dominant-negative form of Hes1 (Hes1dn) under the control of melanocyte-specific promoter (Dct-Hes1dn) were born with severe defects in hair pigmentation. Extensive hair graying occurred in these pigmented hair areas in the second hair cycle,

1216

pointing also to the maintenance defect in MSCs of the transgenic mice. They have also confirmed this activation of Notch signaling by in situ immunohistochemical analysis by including human melanoma tissue arrays, suggesting the role for Notch signaling in melanoma progression. Such studies have potential for the therapeutic improvements against melanoma progression. Arturo Alvarez-Buylla (University of California–San Francisco, San Francisco) presented new evidence showing that adult neural stem cells in the subventricular zone (SVZ) not only generate neurons but also oligodendrocytes. His previous work had identified GFAP-expressing astrocytes (also called type B cells) in the SVZ that function as stem cells and generate large numbers of new neurons through transit-amplifying C cells. These new neurons migrate to the olfactory bulb, where they complete their differentiation into local interneurons. He reviewed recent evidence from his laboratory demonstrating that adult SVZ stem cells are derived from radial glia. Radial glia not only give rise to adult stem cells but also to other brain cell types, including neurons, astrocytes, oligodendrocytes, and ependymal cells. This raised the question of whether adult neural stem cells generate cells other than neurons. In the adult brain, oligodendrocytes are continuously produced from NG2 + progenitor cells. Using specific retroviral labeling of GFAP-expressing cells, it was shown that NG2 progenitors and new oligodendroglial cells are also derived from SVZ astrocytes. The type B cells give rise to a subpopulation of transit-amplifying cells (type C cells) that express markers of the oligodendrocyte lineage. Isolated type B cells in culture generate oligodendrocytes, and a small subpopulation also yield oligodendrocytes and neurons, which suggests that some but not all type B cells in the adult SVZ are multipotential. Thus, the SVZ primary progenitors are more than mere neuronal precursors. Merritt Taylor and Sean Morrison used Cre-loxP–based targeted deletion of RBP-J gene in neural crest progenitors of the peripheral nervous system and found that Notch/RBP-J signaling was crucial for gliogenesis in vivo by promoting the neural crest stem cell maintenance through late fetal development as well as by facilitating the glial differentiation. Neural stem cells are propagated in vitro as neurospheres. However, neurospheres are comprised of not only stem cells but also progenitors and fully differentiated cells, and this has prevented characterization of the true stem cells. Austin Smith (University of Edinburgh, Edinburgh, Scotland) presented conditions that facilitate derivation of homogenous neural stem cell lineages from mouse ES cells without accompanying differentiation. Under these conditions, Austin Smith’s group identified EGF and FGF-2 as the sole extrinsic factors required to maintain the symmetrical self-renewal of neural stem cells. Even after prolonged maintenance, they remained competent to differentiate into neurons and astrocytes in vitro as well as after transplantation into the brain. These neural stem cells have morphological and molecular similarities to radial glia, which

Third Annual Meeting of the ISSCR are endogenous neuronal precursor cells that self-renew without undergoing differentiation. Analogous neuronal stem cells can be established from human ES cells as well as from fetal brains of mouse and human. On the other hand, Jianxue Li et al. (Harvard Medical School) identified the restoration of tissue plasminogen activator, an agonist of protein kinase C, as one of the molecular clues by which neural stem cells rescue cerebellar Purkinje neurons from impending degeneration in homozygous nervous mutant mice. Eftekhar Eftekharpour (Toronto Western Hospital, Toronto) reported that when adult-derived neural stem cells from mice transplanted into dysmyelinated spinal cords of adult Shiverer mice, more than 50% of the surviving cells differentiated into the oligodendrocyte lineage and successfully remyelinated the axons. Kiminobu Sugaya (University of Central Florida, Orlando, FL) provided the first evidence that those ACT-N cells (adult BMderived cells) migrate out and differentiate into cells expressing neuronal markers and induced an improvement in cognitive function when implanted into the ventricle of an adult rat brain. The mechanism of the pluripotency of the ACT-N cells may involve epigenetic modifications of the adherent stem cells from BM. The results are significant because they suggest a possibility of autologous neuroreplacement therapy using the patient’s cells. Thomas Payne of Johnny Huard’s group (Children’s Hospital of Pittsburgh, Pittsburgh) reported the isolation of a population of skeletal muscle–derived stem cells that are more effective than satellite cells in cardiac regeneration and repair upon transplantation into the murine model of acute myocardial infarction. Their present molecular genetic approaches are aimed at identifying and characterizing the factors responsible in this specific process. Ting Xie and Rongwen Xi (Stowers Institute for Medical Research, Kansas City, MO) showed that ATP-dependent chromatin-remodeling factors control stem cell self-renewal in a cell type–specific manner. Although two SWI/SNF2-like chromatin remodeling factors, Imitation Switch and Domino, are expressed in both germ-line and somatic stem cells in the Drosophila ovary, they are required for controlling self-renewal of germ-line and somatic stem cells, respectively. As a technical advance in the identification and characterization of adult stem cells from a variety of tissues, David Breault (Harvard Medical School) exploited the expression of mouse telomerase (mTert) (known to be expressed in ES cells, HSCs, and their progenitors as well as in the proliferating cells of regenerative tissues) and generated mTert-GFP transgenic mice. These mice expressed GFP in BM, including the HSC compartment based on cell-surface marker analysis and serial transplantation potential, in germ cells of the testis and in the crypts of the small intestine. The GFP-expressing intestinal epithelial cells were isolated by fluorescence-activated cell sorting method and shown to be negative for CD45. This model may enable the identification, isolation, and characterization of adult stem cells from a variety of tissues.

Rajasekhar, Vemuri

1217

Cancer Stem Cells

the normal stem cell compartment. They also demonstrated that LSCs still require interaction with microenvironmental niches because interference with LSC homing via blocking antibodies to CD44 can lead to their eradication. These studies also demonstrate a potential strategy to kill LSCs that does not involve attacking cell cycling, as LSCs are normally found in a quiescent state. Peter Lansdorp (BC Cancer Agency, Vancouver, British Columbia, Canada) discussed the molecular mechanisms of telomere attrition in relation to DNA replication and repair in normal and malignant cells. Typical telomere losses occur with every round of telomere replication, whereas sporadic telomere losses can result from either failure to properly repair (oxidative) damage to telomeric DNA or from failure to properly process higher-order structures of G-rich DNA. Sporadic losses of telomeric DNA typically involve homologous recombination reactions. Differences in telomere maintenance and repair between different stem cell types were highlighted. For example, ES cells appear quite resistant to DNA damage and maintain the length of telomere repeats upon serial passage, whereas HSCs are quite sensitive to DNA damage and unable to maintain telomere length. Levels of telomerase in human HSCs are under extremely tight control, as illustrated by marrow failure in patients with (mild) telomerase deficiencies. Differences between normal and malignant cells in telomere-erosion pathways and DNA repair and telomere maintenance provide novel targets for the prevention and therapy of disease. Maartin van Lohuizen (The Netherlands Cancer Institute, Amsterdam) reported that polycomb repressors regulate stem cell fate. Overexpression of a repressive polycomb group (PcG) of gene, BMI-1, results in proliferation of granule precursor cells by repressing Ink4a/Arf locus and activation of sonic hedgehog (Shh) signaling, culminating in cancers such as primary medulloblastomas. His work further showed that PcG silencing is essential to maintain stem cell fate in the nervous system and in hematopoietic cells. Carla Bender Kim and Tyler Jacks and associates (Massachusetts Institute of Technology) reported the successful isolation of a regional pulmonary stem cell population (bronchioalveolar stem cells [BASCs]) from the bronchioalveolar duct junction of mouse lungs. The BASCs exhibited self-renewal, were multipotential in clonal assays, and expanded in response to oncogenic K-ras in culture and in vivo. Understanding how BASCs are regulated during normal lung homeostasis and in lung cancer may offer insights into rational therapeutic interventions to combat lung cancer. Owen Witte (University of California–Los Angeles, Los Angeles) reported the cell-surface marker-dependent isolation of stem/progenitor cell subpopulations from the proximal regions of the murine prostatic tubules, where slow-cycling stem-like cells have also been identified. He also demonstrated that cellautonomous activation of PTEN-Akt signaling pathway in these stem cell fractions induces prostate cancer. These studies not only suggest that prostate stem/progenitor cells are efficient targets for prostate cancer but also open up opportunities to find new potent

An emerging concept in cancer is that tumors are fueled by rare cells that have the characteristics of stem cells. This led to the socalled cancer stem cell hypothesis, which suggests that the primary target of transformation is either a somatic cell with stem cell properties or a dedifferentiated cell that regains the ability to self-renew by extensive proliferation. Classically, dedifferentiation is a process of lineage reversal from a terminally differentiated state, and it is suggested to play a vital role in tissue regeneration. Only recently have high-throughput cellular screens begun to identify the small molecules that render the dedifferentiation of lineage-committed somatic cells as reported by Shuibing Chen (The Scripps Research Institute, La Jolla, CA). Such approaches may lead to therapeutics for in vivo tissue regeneration or other proliferative and/or degenerative disease prevention. Discussing malignant hematopoiesis, Irving Weissman (Stanford University School of Medicine) envisioned that a typical HSC (HSC markers C-Kit+, Thy1.1low, Lin−, Sca 1+) potential varies between young and old animals. The cell-intrinsic functional and molecular properties of long-term repopulating HSCs (LT-HSCs) differ with age. Specifically, the proportion of lymphoid and myeloid cells derived from young HSCs is approximately equal, as mirrored by the expression of myeloid and lymphoid transcription factors in these LT-HSCs. However, there are changes in the aged HSC population in which lymphoid potential and lymphoid-specific transcripts are both markedly diminished while myeloid transcripts and myeloid differentiation potential are dominant; among the myeloid transcripts are those that are involved as translocation partners in human myelogenous leukemias. In mice, conditional knockout of JunB, a negative regulator of LT-HSCs, resulted in a transplantable chronic myeloid leukemia. This is consistent with previous studies showing that myelogenous leukemias begin with initiating events at the level of LT-HSCs, the only cells in the myeloid lineage that self-renew, but that progression from the chronic to the acute or myeloid blast crisis stages involve several events that can give rise to a subclone of progenitors, for example, at the stage of the granulocyte/monocyte progenitors, that have gained unregulated self-renewal and are the leukemia stem cells. Previously, Weissman, Miyamoto, and Akashi showed that LT-HSCs from a Japanese population of aml-1/eto acute myelogenous leukemia (AML) patients have only normal differentiation outcomes but that the leukemia stem cell in that disease was a downstream progenitor. John Dick (Princess Margaret Hospital, Toronto) focused on the function of stem cells in leukemogenesis and delineated that leukemia stem cells are not homogeneous but comprised of distinct hierarchically arranged leukemic stem cells (LSCs) with long-term and short-term potential to sustain the AML clone. These studies demonstrate that LSCs possess both extensive selfrenewal potential as well as the ability to regulate self-renewal potential to create stem cell hierarchies in a manner that resembles

1218

therapeutic targets in Akt signaling axis, wherein additional intermediates were recently identified. There was also a report from Dong Fang in Meenhard Herlyn’s group (The Wistar Institute, Philadelphia) that showed the isolation of multipotent stem-like cells from fresh metastatic melanomas that persisted after a serial cloning in vitro using a growth medium suitable for human ES cells and after transplantation in vivo. Dan Duda, from the group of Rakesh K. Jain (Massachusetts General Hospital), evaluated the phenotype and circulation kinetics of precursor cells as potential surrogate markers for the efficacy of antiangiogenic agents in cancer patients.

Stem Cell Regulatory Networks, Lineage Commitment, and Differentiation By virtue of their ability to recapitulate embryonic germ layer differentiation, ES cells offer an ideal system to study the process of early embryogenesis in vitro. Signaling cascades and genes involved in differentiation can easily be identified through functional genomic and proteomic approaches, thereby permitting the delineation of stem cell regulatory networks underlying lineage commitment and differentiation. Aspects of ES cell self-renewal and terminal fate commitment to produce virtually all somatic lineages have been intensively investigated. Martin Pera (Monash University, Victoria, Australia) identified the presence of subpopulations of cells in human ES cell cultures that could be fractionated by monoclonal antibodies against cell-surface markers. Transcriptional profiling of these subpopulations reveals that under conditions that promote stem cell renewal, a hierarchical pattern of differentiation is evident, even in cells expressing relatively high levels of stem cell markers. It is hypothesized that paracrine interactions among ES cells and their descendants at the early stages of differentiation probably dictates the stem cell fate. Jenna Galloway from Leonard Zon’s group (Harvard Medical School) investigated vertebrate embryonic hematopoiesis using zebrafish as a model system. They unveiled a precise transcriptional hierarchy that governs the differentiation of HSCs into the mature blood lineages. First, blood cells were characterized by the expression of Scl (stem cell leukemia gene encoded basic helix-loop-helix protein) and GATA-2 transcription factors, followed by erythroid lineage marking GATA-1. Using a high-throughput whole-embryo in situ hybridization with hematopoietic cDNA library, different subsets of genes were found to be expressed at different stages of maturation from hematopoietic progenitors to terminally differentiated erythrocytes. Using antisense morpholinos to the above transcription factors, the authors also dissected the regulation of the gene subsets and found that some of the erythroid genes were regulated in a GATA-independent manner. A dual role for GATA-1 was also uncovered in promoting erythroid differentiation and suppressing myeloid formation. In this zebrafish model system,

Third Annual Meeting of the ISSCR the progenitors of blood and kidney arise by a spatiotemporal regulation of a common multipotent mesodermal cell pool. Alan Davidson (Massachusetts General Hospital) in association with Leonard Zon used a zebrafish mutant for the caudal-related homeobox gene cdx4 and antisense morpholinos to the related gene cdx1a to demonstrate that the cdx-hox pathway plays a central role in the formation of blood and kidney cell fate. The cdx genes suppressed the production of RA, which in turn was found to positively regulate renal fate and negatively modulate blood cell formation. From these observations, a model was proposed in which the cdx-hox pathway acts to restrict RA signaling along the embryonic axis and thereby control the induction of kidney and blood progenitors. This is an attractive model because it links together a number of key developmental pathways to create a molecular framework for understanding how and where these tissues are induced during embryogenesis. Such knowledge will be instrumental for targeted differentiation–based approaches in which ES cells are coaxed to differentiate into specific cell lineages. Furthermore, Leonard Zon along with George Daley and associates (Children’s Hospital of Boston) extended this framework with mouse hematopoietic progenitors that are successfully generated from ES cell–derived embryoid bodies (EBs). They reported a significant increase in primitive erythroid colonies by inhibiting RA formation in 2- to 3-day-old mouse EBs, analogous to their observations in zebrafish embryos. George Daley has also used murine ES cells to study the regulation of fate-switching to trophoblast, which is the membrane of the blastocyst. ES cells are normally excluded from the trophoectoderm, but they can be adapted to this lineage by insertion of H-Ras, an oncogene. The H-Ras activation led to trophoblast differentiation that could then be reverted after withdrawal of H-Ras induction. It was also shown that activation of homeobox proteins cdx4 and hoxb4 promotes definite HSC production from ES cells. Yuan Wang from George Daley’s group in further collaboration with Leonard Zon and associates showed the intrinsic requirement of cdx4-hox pathway during murine embryonic hematopoiesis by either conditionally inducing or abrogating expression of cdx4. All of these data have potential in furthering our understanding of developmental hematopoiesis. These studies may yield insights relevant for gene therapeutic approaches and for the molecular pathogenesis of leukemias. Koichi Akashi (Dana-Farber Cancer Institute) reported data on how the multipotent early lymphoid progenitors manifest lineage commitment and developmental plasticity. Isolation of stem and progenitor cell subsets by cell-surface phenotype has identified the branch points at which major hematopoietic lineage decisions take place. Importantly, the author demonstrated that even at these branch points, plasticity for differentiation into all lineages could be achieved by genetically manipulating the levels of essential transcription factors such as C/EBP, PU.1, GATA-1, and GATA-2.

Rajasekhar, Vemuri A dosage-sensitive role for Sox2 was also identified as a key regulator of progenitor identity by Larysa Pevny (University of North Carolina at Chapel Hill, Chapel Hill, NC) through the analysis of an allelic series of Sox2 mutations in the mouse. Alterations in the levels of Sox2 regulate the choice between maintenance of neural progenitor cell identity and differentiation such that wild-type levels of Sox2 maintain progenitor developmental potency and the ability to proliferate and differentiate and decreased Sox2 levels (