Portal de Periódicos da CAPES

Stem and progenitor cell dysfunction in human trisomies

2014; Springer Nature; Volume: 16; Issue: 1 Linguagem: Inglês

10.15252/embr.201439583

1469-3178

Binbin Liu, Sarah Filippi, Anindita Roy, Irene Roberts,

Genomic variations and chromosomal abnormalities

Review17 December 2014free access Stem and progenitor cell dysfunction in human trisomies Binbin Liu Binbin Liu Department of Paediatrics and Molecular Haematology Unit, Weatherall Institute of Molecular Medicine, Oxford, UK Search for more papers by this author Sarah Filippi Sarah Filippi Department of Statistics, University of Oxford, Oxford, UK Search for more papers by this author Anindita Roy Anindita Roy Centre for Haematology, Imperial College London, London, UK Search for more papers by this author Irene Roberts Corresponding Author Irene Roberts Department of Paediatrics and Molecular Haematology Unit, Weatherall Institute of Molecular Medicine, Oxford, UK Search for more papers by this author Binbin Liu Binbin Liu Department of Paediatrics and Molecular Haematology Unit, Weatherall Institute of Molecular Medicine, Oxford, UK Search for more papers by this author Sarah Filippi Sarah Filippi Department of Statistics, University of Oxford, Oxford, UK Search for more papers by this author Anindita Roy Anindita Roy Centre for Haematology, Imperial College London, London, UK Search for more papers by this author Irene Roberts Corresponding Author Irene Roberts Department of Paediatrics and Molecular Haematology Unit, Weatherall Institute of Molecular Medicine, Oxford, UK Search for more papers by this author Author Information Binbin Liu1, Sarah Filippi2, Anindita Roy3 and Irene Roberts 1 1Department of Paediatrics and Molecular Haematology Unit, Weatherall Institute of Molecular Medicine, Oxford, UK 2Department of Statistics, University of Oxford, Oxford, UK 3Centre for Haematology, Imperial College London, London, UK *Corresponding author. Tel: +44 1865 222316; Fax: +44 1865 234251; E-mail: [email protected] EMBO Reports (2015)16:44-62https://doi.org/10.15252/embr.201439583 See the Glossary for abbreviations used in this article. PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Trisomy 21, the commonest constitutional aneuploidy in humans, causes profound perturbation of stem and progenitor cell growth, which is both cell context dependent and developmental stage specific and mediated by complex genetic mechanisms beyond increased Hsa21 gene dosage. While proliferation of fetal hematopoietic and testicular stem/progenitors is increased and may underlie increased susceptibility to childhood leukemia and testicular cancer, fetal stem/progenitor proliferation in other tissues is markedly impaired leading to the characteristic craniofacial, neurocognitive and cardiac features in individuals with Down syndrome. After birth, trisomy 21-mediated premature aging of stem/progenitor cells may contribute to the progressive multi-system deterioration, including development of Alzheimer's disease. Glossary AICD APP intracellular domain ALL acute lymphoblastic leukemia AVSD atrioventricular septal defect BCP B-cell progenitor DS Down syndrome FDR false discovery rate GEDD gene expression dysregulation domain GEP gene expression profile GMP granulocyte–monocyte progenitor GSK3b glycogen synthase kinase 3b hESC human embryonic stem cell Hsa21 Homo sapiens chromosome 21 HSC hematopoietic stem cell HSPC hematopoietic stem and progenitor cell IGF insulin-like growth factor iPSC induced pluripotent stem cell MEP megakaryocyte–erythroid progenitor miR microRNA ML-DS myeloid leukemia of Down syndrome Mmu Mus musculus NFAT nuclear factor of activated T cells qPCR quantitative polymerase chain reaction Shh sonic hedgehog T21 trisomy 21 TAM transient abnormal myelopoiesis VEGF vascular endothelial growth factor Introduction Trisomy 21, trisomy 18 and trisomy 13 are the commonest constitutional trisomies in humans 1. In contrast to trisomy 18 and 13, where fewer than 10% of affected children survive beyond the first year of life 234, median life expectancy for individuals with trisomy 21 (Down syndrome; DS) is around 60 years 5. Most attention has focused on trisomy 21, not only because it is 20 times and 40 times more frequent than trisomy 18 and 13, respectively, but also because prolonged survival in DS suggests that most cells evolve epigenetic, transcriptional and/or translational regulatory mechanisms which allow them to adapt to the additional copy of chromosome 21 (Hsa21). To some extent, this may reflect the relatively low number of protein-encoding genes on Hsa21 (~240) (http://www.ensembl.org/biomart/martview/). However, the characteristic phenotypic variability between different individuals with DS points to considerable complexity. Understanding the genomic determinants of this complexity continues to reveal fascinating insights relevant not only to DS, but also to aneuploidy in general. Here, we review the impact of human trisomies on stem and progenitor cells. We will focus on trisomy 21, and particularly on hematopoiesis, where advances in techniques for characterization of highly purified primary cells and the development of induced pluripotent stem cell (iPSC) and animal models are beginning to answer some of the questions about the mechanisms by which trisomies cause human disease. Phenotypic variability in constitutional trisomy 21 (DS) DS is a multisystem disorder caused, in most cases, by meiotic non-disjunction of the maternal Hsa21, resulting in a third copy of the entire Hsa21 in all cells 67. The clinical and biological impact of trisomy 21 nevertheless varies widely, not only between individuals with DS, but also in different tissues, the cell types within these tissues and at different ages [reviewed in 8910]. Within this phenotypic variability, certain characteristics, such as the craniofacial abnormalities, hypotonia and cognitive impairment, are common to all individuals with DS (Table 1) and may therefore share temporal, biological or genetic mechanisms. Other features, such as cardiac defects or gastrointestinal anomalies, affect only a subset of patients and so may be more strongly influenced by inter-individual differences which interact with trisomy 21-driven changes, in heart and gut development, respectively, during embryogenesis. Many of these phenotypic abnormalities can be modeled using mouse segmental trisomies allowing the consequences of trisomy 21 to be investigated in an appropriate cellular context (Table 2). Table 1. Phenotypic characteristics of Down syndrome Characteristic Frequency (%) Reference Craniofacial ~100 10, 114, 194, 195 Epicanthal folds Upward slanting palpebral fissures Flat nasal bridge Small brachycephalic head Small ears Small mouth Other musculoskeletal abnormalities ~100 10, 196 Hypotonia Single transverse palmar crease Clinodactyly with wide spacing Cognitive impairment ~100 10, 17, 197 Reduced brain volume Learning and memory defects Dementia 40–50, increases with age 12, 13, 14, 15, 16, 17, 18, 198, 199 Visual 18–60, increase with age 10, 18 Hearing 18–80, increase with age 10, 18 Thyroid disease 1–54, increase with age 10, 18, 196, 200, 201 Cardiac defects 40–50 202, 203, 204 ASD (45%) VSD (35%) Isolated secundum (8%) Isolated PDA (7%) Isolated Fallot's (4%) Other Gastrointestinal defects 12 10 Benign hematological abnormalities ~100 11 Neonatal thrombocytopenia Neonatal polycythemia Neonatal neutrophilia, blast cells Macrocytosis Preleukemia and leukemia TAM and silent TAM 30 11, 53 Acute myeloid leukemia 1 53 Acute lymphoblastic leukemia 1 53 Non-hematologic cancers 50% of risk of individuals without DS 18, 53, 56, 58, 59 Table 2. Mouse models of Down syndrome Ts65Dn Ts1Cje Ts1Rhr Tc1 Ts1Yeh;Ts2Yeh;Ts3Yeh Dp(16)1Yeh/+; Dp(10)1Yeh/+; Dp(17)1Yeh/+ Ts1Yah Number of trisomic Hsa21 orthologs ~100 (also trisomic for ~60 genes on Mmu17 not syntenic for Hsa21) ~80 33 Transchromosomic (trisomic for 200 RefSeq Hsa21 genes*) ~175 ~12 Craniofacial Small mandible Brachycephaly Small mandible Large mandible Small mandible Normal appearance Not reported Differences in face, palate recapitulate human DS Differences in face, palate recapitulate human DS Abnormalities do not recapitulate human DS Learning and memory Impaired spatial learning and memory Impaired motor coordination Altered hippocampal dependent learning Impaired novel object recognition Defect in short-term memory and motor co-ordination Recapitulates most of the behavioral features of Ts65Dn Impaired novel object recognition but improved hippocampal-dependent spatial learning Brain structure Reduced brain volume Reduced brain volume during embryogenesis Reduced brain volume at age 4 months Hydrocephalus (6.5%) Not reported Reduced cerebellar volume Reduced cerebellar volume Reduced cerebellar volume Reduced cerebellar volume Impaired neurogenesis: –Impaired neural precursor proliferation and differentiation –Abnormal cell cycle Impaired neurogenesis: –Impaired neural precursor proliferation and differentiation Cardiac defects Septal defects similar, but not identical, to human DS Not reported None Mainly VSD; also outflow tract defects and AVSD similar to human DS Cardiac defects include ASD VSD and AVSD (in several models: Dp(16)1Yeh/+,Dp(16)2Yeh/+Dp(16)4Yeh/+ andDp(16)1Yeh/+; Dp (16)2Yeh/+;Dp(16)3Yeh/+ Not reported Gastro-intestinal defects Not reported Not reported Not reported Not reported Not reported Not reported Thyroid disease Not reported Not reported Not reported Not reported Not reported Not reported Hematopoietic MPD in adults No leukemia Macrocytosis No MPD or leukemia Macrocytic anemia Co-operates with GATA1s and MPL to induce AMKL No MPD or leukemia even with GATA1s Macrocytosis Not reported Not reported Reduced adult HSC Impaired HSC self-renewal in adults Normal HSC numbers and function in adults Impaired fetal liver HSC and progenitor function Thrombocytosis, increased MKs and mild anemia in adults Increased MKs and erythrocytosis in older adults Increased GMP in adults Increased GMP in adults AMKL, acute megakaryocytic leukemia; ASD, atrioseptal defect; AVSD, atrioventricular septal defect; DS, Down syndrome; GMP, granulocyte–macrophage progenitor; HSC, hematopoietic stem cell; MPD, myeloproliferative disorder; VSD, ventriculoseptal defect. See text for details. The impact of age on the phenotypic expression of DS is increasingly recognized and, for many cells and tissues, is essential to consider in selecting the best experimental model to investigate the role of trisomy 21. Abnormalities of hematopoiesis begin in fetal life and have their maximal expression in the neonatal period when nearly all DS neonates have multiple hematologic defects, including 30% who develop a unique preleukemic syndrome confined to the first few months of life 11. By contrast, the effects of trisomy 21 on visual and hearing impairment, thyroid function and cognitive function increase with age with progressive pathological changes in the brain in almost all DS individuals and clinical evidence of dementia in ~50% 1213141516171819. These age-related differences in phenotypic expression in DS suggest that trisomy 21, through patterns of gene expression which may be established early in development, causes premature, or accelerated, aging of a range of cell types. Evidence in support of this is now emerging 20, as discussed below. Although several Hsa21 genes have been linked to the phenotypic expression of specific aspects of DS, such as leukemia and dementia, the mechanism(s) by which trisomy of individual genes or groups of genes contributes to the disorder remains unclear [reviewed in 8921222324]. Investigators have used three broad approaches to investigate this question: mouse models trisomic for one or more of the genes on Hsa21, genomic association studies and comparative studies between human cells trisomic or disomic for Hsa21. Mouse models of DS The phenotypic characteristics of the most well-established mouse models of DS, and the extent to which they recapitulate the human phenotype, are summarized in Table 2. These include the only transchromosomic mouse model (Tc1) in which most of Hsa21 is present 25 albeit with several deleted or rearranged genes 26. A number of more recent mouse mutants carrying genomic rearrangements of Hsa21 syntenic regions (on Mmu10, Mmu16 and Mmu17) that are trisomic for some, or most, of the ~250 mouse genes orthologous to Hsa21 genes have been described 2728293031. These interesting models, which may better mimic some aspects of human DS, have so far been used mainly to model the neurocognitive and cardiac defects in DS 2728293031. Details of these, and of elegant refinements to narrow down the Hsa21 regions linked to defined phenotypes, are described in several reviews 3233343536373839404142 and are only briefly discussed here in relation to their insight into the effects of trisomy 21 on stem/progenitor cells. The impact of trisomy 21 on stem cell function There is increasing recognition that trisomy 21 impacts on stem cell function in a number of ways (Fig 1). In hematopoiesis, for example, trisomy 21 affects the self-renewal, proliferation and differentiation of hematopoietic stem and progenitor cells (HSPC) either directly or via the hematopoietic microenvironment 43444546474849505152. Studies in other tissue types, where stem and progenitor cells are often more difficult to identify and isolate, suggest that trisomy 21 also causes many of the defects in craniofacial, brain and cardiac development through perturbations of stem/progenitor cell growth and differentiation and altered interactions with microenvironmental and temporal cues. These alterations in stem/progenitor proliferation may underlie the increased susceptibility of some cell types, such as HSPC and primordial germ cells to malignant transformation 5354555657585960 and of HSPC to premature aging in DS 20, as discussed in detail below. Figure 1. Impact of trisomy 21 on stem and progenitor cell functionStudies in human cells and in animal models of Down syndrome (DS) show that trisomy 21 can affect the self-renewal, proliferation and differentiation of stem and progenitor cells either directly or via the supportive microenvironment. In fetal life in DS, proliferation of hematopoietic and testicular stem/progenitor cells is increased and susceptibility to malignant transformation (leukemia and testicular cancer) is increased in childhood. By contrast, proliferation of stem/progenitor cells of other lineages is impaired and is responsible for many of the developmental defects affecting the brain, craniofacial structures and heart in DS. After birth, trisomy 21 has been shown to cause premature aging of stem and progenitor cells both of hematopoietic and non-hematopoietic lineages, an effect which is likely to contribute to the phenotypic abnormalities in adults with DS, including Alzheimer's disease, bone marrow failure and impaired immunity. Download figure Download PowerPoint Hematopoiesis and leukemia The link between childhood leukemia and DS provides strong evidence for a particular susceptibility of hematopoietic cells early in life to perturbation of the normal mechanisms which control their growth and differentiation. Leukemias in DS have several unique features which hint at the ways in which trisomy 21 alters the behavior of HSPC [reviewed in 216162]. First, the frequency of both myeloid leukemias and lymphoid leukemias is increased, by 150-fold and ~30-fold, respectively 5359, indicating that trisomy 21 affects both myeloid and lymphoid progenitors. Second, these leukemias have a distinct temporal pattern of onset. Myeloid leukemia of DS (ML-DS) originates in fetal liver HSPC and presents either as a neonatal preleukemic syndrome known as transient abnormal myelopoiesis (TAM) or as full-blown ML-DS in children under the age of 5 years 216162. The peak age at presentation for acute lymphoblastic leukemia in DS (DS-ALL) is 1–4 years and, in contrast to ALL in individuals without DS, never presents in neonates or infants 53. Third, leukemias in DS have distinct biologic and molecular features. Leukemic cells in ML-DS and TAM harbor N-terminal truncating mutations in the key hematopoietic transcription factor GATA1, which result in exclusive production of a short GATA1 protein (Gata1s) with altered functional properties together with loss of expression of full-length Gata1 since the GATA1 gene is on the X chromosome 6364656667. Such mutations are not leukemogenic in the absence of trisomy 21 68. In DS-ALL, which in contrast to ALL in children without DS is always a B-precursor disease 69, ~60% of cases have aberrant expression of the CRLF2 receptor and around half of these have RAS mutations or mutations activating JAK-STAT growth-promoting signaling pathways 70717273747576. Impact of trisomy 21 on fetal, neonatal and adult human hematopoiesis In contrast to most other tissues, hematopoietic tissues contain a well-characterized hierarchy of stem and progenitor cells, which can be readily isolated for molecular and functional studies. Characterization of the hematologic abnormalities in human DS therefore offers one of the best ways to understand how trisomy 21 perturbs cell biology and how cells adapt to aneuploidy. Recent studies in primary human fetal liver and neonatal cells 1145, supported by data from human iPSC and hESC 4647, demonstrate that trisomy 21 causes major disturbance throughout the entire hematopoietic hierarchy from HSC through to progenitors and mature cells (Fig 2). In particular, in fetal liver, trisomy 21 alters the balance of HSPC differentiation, promoting expansion and proliferation of megakaryocyte–erythroid progenitors (MEP) and megakaryocytes during the second trimester at the expense of both granulocyte–monocyte progenitors (GMP) and B-cell progenitors (BCP) 45. There is also a 3.5-fold expansion in HSC numbers, and in vitro purified trisomy 21 fetal liver HSCs have erythroid–megakaryocyte-biased gene expression together with reduced expression of lymphoid genes. Consistent with this, fetal liver HSC function is also markedly abnormal in DS. In particular, fetal liver HSCs generate more megakaryocyte and erythroid cells while their B-cell potential is severely impaired 45. Since GATA1 mutations were not detectable in these cells, these data indicate that trisomy 21 itself perturbs fetal liver hematopoiesis. Figure 2. Perturbation of human fetal hematopoiesis by trisomy 21Comparison of the frequency and function of disomic and trisomy 21 second-trimester human fetal hematopoietic stem cells and progenitor cells (HSPC) has shown a consistent pattern of abnormalities in the trisomic populations. Trisomy 21 alters the balance of HSPC differentiation, promoting expansion and proliferation of megakaryocyte–erythroid progenitors (MEP) and megakaryocytes during the second trimester at the expense of both granulocyte–monocyte progenitors (GMP) and B-cell progenitors (BCP), which are both significantly reduced. Download figure Download PowerPoint The effects of trisomy 21 on primary human fetal liver HSPC raise many questions. First, since these studies were confined to second-trimester fetal liver, it is not clear whether the effects are confined to this gestation. Interestingly, Chou et al 47 found that trisomy 21 iPSC differentiated under conditions designed to model yolk sac hematopoiesis showed enhanced erythroid, but not megakaryocyte, differentiation in vitro, suggesting the effects of trisomy 21 may be developmental stage specific. More recently, our group studied hematopoiesis in neonates with DS. In the presence of GATA1 mutations, DS neonates developed the preleukemic condition, TAM. However, even in the absence of GATA1 mutations, DS neonates had trilineage perturbation of hematopoiesis with increased erythroid and myeloid cells and abnormal platelet development consistent with the effects of trisomy 21 on HSPC function persisting after birth 11. In contrast, the few studies in adults with DS suggest that trisomy 21 causes a different profile of hematologic abnormalities later in life. Adults with DS have a high prevalence of red cell macrocytosis and quantitative and qualitative B- and T-lymphocyte abnormalities, while some have unexplained thrombocytopenia and neutropenia 777879, myelodysplasia or bone marrow failure 80. This suggests that in adults, trisomy 21 may induce HSC aging, consistent with recent studies in Ts65Dn mice implicating increased expression USP16 as a possible mechanism for these effects 20. Second, the mechanisms linking trisomy 21-mediated perturbation of fetal liver hematopoiesis and the high frequency of GATA1 mutations in DS neonates are still unclear. Trisomy 21-mediated proliferation of fetal liver megakaryocyte/erythroid-biased HSPC may simply provide a permissive cellular environment for expansion of preleukemic mutant GATA1 clones. Alternatively, changes to pathways regulating fetal HSPC growth and differentiation in DS may be directly responsible for the increased frequency of GATA1 mutations. Similarly, the link between impaired B-cell development in DS fetal liver HSPC and the increase in B-ALL 698182 and of immune deficiency in children with DS 83 is unclear, although delayed expression of the normal fetal B-cell development program might increase the likelihood of acquiring leukemogenic mutations in lymphoid genes in early childhood. Third, given that alterations in the microenvironment can promote myeloproliferative disorders and leukemia in mouse models 84858687, the DS fetal liver microenvironment may support, or even drive, the abnormal growth and differentiation of DS fetal liver HSPC. The natural history of TAM, which resolves within a few weeks of life in most cases and is characterized by infiltration of the liver by mutant GATA1 blast cells 8889, also suggests that factors produced in the fetal liver microenvironment may be necessary to maintain these cells. In support of this, in vitro survival of TAM blast cells and in vivo survival of leukemia cells in a mouse model of DS-like acute myeloid leukemia has been shown to be dependent on insulin-like growth factors 90. Finally, the molecular basis for perturbation of fetal liver HSPC growth and differentiation by trisomy 21 remains to be explained. Even using highly purified fetal liver HSPC, we found no significant increase in expression of selected trisomic genes on Hsa21 (RUNX1, ERG, DYRK1A) known to influence HSPC behavior and development of leukemia through gene dosage 5191. This does not exclude a role for trisomy 21 dose-related changes in these genes given limitations in the sensitivity of the methodology 92 and the confounding influence of interindividual variation 93 as discussed below, especially since even small changes in expression of the Hsa21 genes are associated with DS-like defects in mouse models 94 and multiple genes may be involved 95. Animal models of leukemia and abnormal hematopoiesis in DS Although ML-DS and TAM provide a natural human model to interrogate the impact of trisomy 21 on HSPC and the mechanisms which contribute to the development of leukemia in DS, mechanistic experiments to identify the exact role of specific genes are often difficult in human cells. Initial attempts to model ML-DS and TAM in mouse models were disappointing as no spontaneous leukemias developed (Table 2). However, this is consistent with human DS where trisomy 21 dysregulates HSPC proliferation and differentiation but is insufficient to promote leukemia without additional, acquired mutations. All of the DS mouse models have abnormal hematopoiesis, typically affecting erythroid and megakaryocyte development 484950, although the defects do not accurately recapitulate those seen in human fetal liver 45. Nevertheless, by co-expressing additional oncogenes, DS mouse models provide potential insight into genes and pathways, which may contribute to perturbation of HSPC development by trisomy 21, including ERG, DYRK1A, HMGN1 and miR125b 519192969798. The myeloproliferative disorder in adult Ts65Dn mice 48, for example, is clearly linked to gene dosage of ERG since reducing the number of copies of ERG from 3 to 2 in this model corrects the hematologic abnormalities 91. Since neonatal TsDn mice are not affected, Birger et al 97 used a different approach to modeling TAM. Building on data showing potent effects of ERG overexpression on megakaryocyte proliferation and leukemia in disomic mice 100, they recently created a double transgenic mouse model of TAM/ML-DS on a non-trisomic background in which overexpression of ERG promoted fetal liver MEP expansion similar to that seen in human fetal liver, and this synergized in vivo with expression of GATA1s to cause a TAM-like disease and subsequent progression to megakaryocyte–erythroid leukemia 97. Nevertheless, ERG has not yet been shown to be significantly overexpressed in trisomy 21-containing human hematopoietic cells, including leukemias 45100 and hESC/iPSC 4647. Malinge et al 51 recently used Ts1Rhr mice, trisomic for 33 Hsa21 orthologs (Table 2), to create a trisomy 21-dependent ML-DS model by crossing them with GATA1s knock in mice and over-expressing a transforming MPL allele (MPLW515L), which has been reported in ML-DS 101102. In this model, they showed that DYRK1A was able to act as a megakaryoblastic tumor-promoting gene and they found increased expression of DYRK1A in human ML-DS samples 51. Although this identifies a possible role for increased expression of DYRK1A in the transformation of TAM to ML-DS, this model does not fully recapitulate the human disease. For reasons that are still not clear, this model can only be produced in adult, and not fetal, hematopoietic cells, and indeed, DYRK1A expression does not appear to be significantly increased in human fetal HSPC 45, perhaps indicating altered mechanisms of DYRK1A regulation in fetal cells compared to postnatal or leukemic cells. Furthermore, MPLW515L is able to induce a fatal, rapid onset myeloproliferative disorder even in the absence of Gata1s and a trisomic background 103 highlighting the importance of the cellular context in understanding the contribution of individual genes. The Ts1Rhr mouse model has also proved useful for investigating the role of Hsa21 orthologs in B-cell development and B-ALL. Lane et al 98 recently showed that, as in human fetal liver 45, B progenitors were reduced in bone marrow from young Ts1Rhr mice but were more clonogenic than wild-type progenitors and could be replated indefinitely in vitro. Furthermore, Ts1Rhr B progenitors were transformed into B-ALL in vivo by CRLF2 with activated JAK2, a known oncogenic stimulus in DS-ALL. Lane et al then identified differential expression of PRC2 targets and sites of H3 K27 trimethylation as a specific 'signature' common to DS-ALL and Ts1Rhr B cells and, through a series of elegant experiments, showed that overexpression of HMGN1, an Hsa21 ortholog trisomic in Ts1Rhr mice which encodes a nucleosome remodeling protein, is responsible both for this gene expression signature and for the proliferative and leukemia-promoting effects on Ts1Rhr B cells. These data provide compelling evidence in support of a role for HMGN1 in the perturbation of B-cell development by trisomy 21 and the increased susceptibility of children with DS to B-ALL. USP16 and defects in HSC self-renewal and stem cell aging By com-paring hematopoiesis in Ts65Dn, Ts1Cje and wild-type mice, Adorno et al 20 identified a role for the mouse homolog of the Hsa21 gene USP16 in HSC self-renewal. HSC frequency was reduced by greater than threefold in Ts65Dn mice, which are trisomic for USP16, compared to Ts1Cje mice and wild-type mice, which have only 2 copies of USP16. HSC function was also impaired in the USP16 trisomic mice with reduced clonogenicity and multilineage engraftment following secondary transplantation. These features were associated with a 1.5-fold increase in USP16 gene expression and were reversed by short interfering RNAs. Interestingly, similar defects were seen in Ts65Dn neural progenitors and fibroblasts consistent with previously reported defective proliferation of primary human DS fibroblasts 104105. They also went on to demonstrate a link between trisomy for USP16 and reduced activity of the PRC1 complex and its target CDK2NA, which regulate senescence and self-renewal of several somatic stem cell types [106107; reviewed in 108]. The reduction in HSC frequency and clonogenicity contrasts with the increase in HSC frequency and clonogenicity in human DS fetal liver 45. However, although this may reflect species-specific differences in hematopoiesis and/or the role played by other genes/pathways in the senescence of Ts65Dn mouse HSC, another important issue is age. The impaired HSC self-renewal reported by Adorno et al in adult Ts65Dn mice 20 is compatible with the increasing recognition of the occurrence of hematologic abnormalities, including myelodysplasia and bone marrow failure, in older adults with DS 80. Non-hematologic cancers It is likely that several mechanisms contribute to the 50% reduction in the frequency of solid tumors with DS, including the effects of trisomy 21 on stem and progenitor proliferation, tumor-associated angiogenesis and tumor suppression [reviewed in 22]. It is notable that the only malignancy, apart from leukemia, to be increased in DS is testicular germ cell tumors. These tumors are derived from primordial germ cells and, in DS, are believed to arise in utero through a pre-