379
Views
0
CrossRef citations to date
0
Altmetric
Meeting Report

38th Annual Meeting of the International Society of Experimental Hematology

&
Pages 645-647 | Published online: 10 Jan 2014

Abstract

The International Society of Experimental Hematology holds its annual meeting every northern summer. This year the meeting comprised of eight plenary sessions with distinguished invited speakers on in vivo imaging and tracking of hematopoietic stem cells (HSCs), HSC niches, epigenetic regulations of stem cells and regulation of stem cell fate, leukemogenesis, and mesenchymal stem cells. The small size of the meeting (300 attendees) permitted excellent discussion and face-to-face contacs between students, junior scientists and experts. Owing to the large number of keynote speakers, this report focuses on the most novel, unpublished data presented during the meeting.

In vivo imaging & hematopoietic stem cells niches

The meeting started with the Don Metcalf Lecture, this year given by Hiromistu Nakauchi (University of Tokyo, Japan). He showed that in bone marrow (BM), a rare astrocyte-like cell population expressing glial fibrillary acidic protein (GFAP) co-localizes with phenotypic lineage-negative CD41-CD48-CD150+ hematopoietic stem cells (HSCs) and expresses cytokines essential to HSCs, such as CXCL12, angiopoietin-1, KIT ligand and thrombopoietin, as well as expressing the three TGF-β isoforms and integrin αvβ8. Integrin αvβ8, which is uniquely expressed by these GFAP+ cells, may play a crucial role in maintaining HSC quiescence by specifically binding to the complex between latent pro-TGF-β and latent TGF-β binding protein (LTBP). Nakauchi also highlighted that the interaction of integrin αvβ8 with the latent TGF-β–LTBP complex enables the opening of latent TGF-β and its activation by proteases present in the BM microenvironment. Once activated, TGF-β maintains HSC quiescence by inhibiting lipid raft formation at the surface of HSCs in response to the binding of KIT ligand and thrombopoietin to their cognate receptors. Therefore, GFAP+ astrocyte-like niche cells could be key to the control of HSC quiescence in the BM.

David Scadden (Massachussetts General Hospital, MA, USA) showed that BM cells expressing the transcription factor osterix/SP7 are also contenders for this role. It is now well established that osteoblast-lineage cells are important contributors of HSC niches in the BM. However, whether these niche cells are mature osteoblasts or immature osteoblast precursors is unknown. Osterix is required for differentiation of mesenchymal stem cells to osteoblasts and is expressed for a short period at the beginning of the commitment of mesenchymal precursors to the osteoblastic lineage. Transgenic mice expressing Cre recombinase under the control of either the osterix promoter (osxCre) or the osteocalcin promoter (which is unique to mature osteoblasts [oscalCre]) were crossed with mice in which the Dicer-1 gene (which is required for the formation of microRNA [miR]) was floxed in order to delete the Dicer-1 gene specifically either in mature osteoblasts (in oscalCre × Dicerf/f mice) or in primitive osteoblastic progenitors (in osxCre ×Dicerf/f mice). Although both mouse strains had skeletal abnormalities, osxCre × Dicerf/f mice also had a marked hematopoietic defect similar to myelodysplastic syndrome with lymphopenia, thrombopenia and hyperlobulated neutrophils. oscalCre × Dicerf/f mice had no hematopoietic defect. The phenotype was present in mutant recipients transplanted with wild-type HSCs however absent in reciprocal chimeras. All osxCre × Dicerf/f mice eventually died of leukemia. Therefore, these experiments show that it is not the mature osteoblasts lining the bone surface that control HSCs, but rather their more primitive precursors. It also gives traction to the notion of leukemic niches by further demonstrating that deregulated niche cells can favor leukogenesis and that miR play an important role in this process.

In the final session on the stem cell microenvironment, Paul Frenette (Mount Sinai, NY, USA) presented results from an investigation that used transgenic mice in which green fluorescent protein (GFP) was under the control of the nestin promoter (nestin-GFP), to show that nestin, an intermediate filament protein mostly expressed by neurons, is expressed in the BM mostly by pericytes around blood vessels. These nestin-positive pericytes express most proteins known to control HSC fate, such as angiopoietin-1, KIT ligand, N-cadherin, osteopontin, VCAM-1 and Il-7; and Frenette’s results showed 50% of phenotypic lineage-negative CD41-CD48-CD150+ HSCs to be in direct contact with these nestin+ cells in the BM. Nestin-GFP+ cells sorted from the BM can be cloned when cultured as spheres and differentiated into osteoblasts, adipocytes and chondrocytes in vitro. When transplanted subcutaneously in phosphocalcic ceramic cubes, nestin-positive cells form ectopic bones containing hematopoietic BM. Nestin-GFP+ cells were resorted from these primary grafts and shown to re-form ectopic bones in secondary recipients. Furthermore, using mice containing a tamoxifen-inducible Cre recombinase under the control of the nestin promoter (nestinCreER) and a Cre-inducible diphtheria toxin receptor, in vivo ablation of these nestin-positive cells resulted in robust HSC mobilization from the BM into the blood. Similarly, ablation of these nestin-positive cells impaired homing of transplanted HSC into the BM. Therefore, nestin-positive BM cells present the hallmarks of self-renewing mesenchymal stem cells forming the HSC niches in the BM. It is clear from the following discussions that each of these three speakers’ groups will now assess whether GFAP, osterix and nestin are in fact all expressed in a unique cell population controlling HSC fate in vivo.

Epigenetic controls of hematopoiesis

Two plenary sessions were dedicated to discussing epigenetic controls of hematopoiesis and there was a particular emphasis on miR. miR is synthesized from large mRNA – a typical hairpin loop that is processed by enzymes, such as Drosha and Dicer, to form single-stranded 23-nucleotide long miR, which regulate protein expression by either blocking translation or destabilizing partially complementary target mRNA. Various examples of how miR controls normal hematopoiesis or contributes to leukemogenesis were provided by the speakers. For instance, Antonio Sorrentino (Instituto Superiore di Sanità, Rome, Italy) showed that of the 706 known human miRs (the list is growing and predicted to exceed 1000), 49 are expressed by human CD34+ cells and their progenies, and they target expression of mRNA critical for HSCs, such as c-kit (miR221/22), CXCR4 (miR146a), or for lineage commitment. For instance, the AML1 gene product binds to the promoters and represses expression of miR17 and miR106a. Reciprocally, both miR17 and miR106a inhibit AML1 expression. Thus AML1, miR17 and miR106a form a self-inhibitory loop in CD34+ cells. This loop is broken when macrophage colony-stimulating factor binds to its receptor c-fms and initiates macrophage differentiation by repressing miR17 and miR106a expression, which in turn releases expression of AML1.

Florian Kuchenbauer (Terri Fox Lab, Vancouver, Canada) used deep sequencing to identify new miR and SNIPs in these miR genes. In his model, mouse BM cells were first retrovirally transduced with a NUP98–HoxD13 fusion protein, which immortalizes cells but does not give rise to leukemia once transplanted. Additional overexpression of the homeobox gene Meis1 led to leukemia. In a quest to determine whether miR could be responsible for this transition from immortalization to malignancy, he identified 55 novel miRs, two of which were overexpressed. Most interestingly, deep sequencing of the transcriptome revealed that miR*, the RNA product transcribed from the complementary strand of miR genes, is also expressed; sometimes in equal amount to their complementary miR, or sometimes in much larger amounts. This means that the number of miRs could potentially be the double what we know, and also that miR–miR* duplexes may exist and might have completely different functions and targets. Fernando Camargo (Whitehead Institute for Biomedical Research, MA, USA) demonstrated that when a single miR is overexpressed, most generally the target mRNA are destabilized. However, blockage of translation was much rarer.

Leukemogenesis & leukemia stem cells

The concept of leukemia stem cells or leukemia-initiating cells has gained traction during the last 10 years. This concept is based on the notion that as with normal stem cells, a pool of malignant cells has the ability to indefinitely self-renew and re-initiate tumors in distant locations. As normal adult stem cells (particularly HSC) self-renew extremely slowly in the BM, it is commonly thought that such leukemia stem cells may also self-renew extremely slowly, rendering them more resistant to cytotoxic therapy, while their actively dividing progenies are more sensitive. In his ‘Till and McCullock’ Lecture, Scott Armstrong (Children’s Hospital, MA, USA) showed that the genetic events causing leukemia need not occur in a HSC. For instance, chromosomal rearrangements involving the mixed-lineage leukemia-1 (MLL1) gene occurs in 70% of childhood acute leukemia. Overexpression of a MLL–AF4 fusion protein (corresponding to translocation t(4;11)) in committed granulocyte–macrophage progenitors (GMPs) gives rise to leukemia initiating cells than can propagate leukemia when transplanted into serial hosts. Therefore, these leukemic GMPs (L-GMPs) do self-renew and are leukemia-initiating cells. Expression microarray analysis revealed that genes, normally restricted to HSC, were re-activated in these L-GMP, particularly homeobox genes, such as HoxA5, HoxA9 and Meis1. Overexpression of HoxA9–Meis1 fusion protein in mouse HSC also caused leukemia, yet leukemia was not initiated when transduced into GMP. GMP–HoxA9–Meis1 progressively differentiated and lost their self-renewal potential in vivo. As β-catenin, a critical mediator of Wnt signaling, is expressed in normal in HSC but not in GMP, β-catenin modulation was tested in this model. Indeed, a constitutively active mutant of β-catenin rendered GMP–HoxA9–Meis1 leukemogenic in mice, whereas mice transplanted with HSC–HoxA9–Meis1 lacking the β-catenin gene survived much longer, thus confirming that Wnt signaling is necessary to the self-renewal of leukemia initiating cells transformed with the HoxA9Meis1 fusion gene. Cox1 was also identified as one of the most upregulated genes in leukemic GMP–MLL–AF4 compared with normal HSCs. Treatment of recipient mice with indomethacin for 5 days resulted in inhibition of β-catenin in GMP–MLL–AF4 leukemic cells, increased differentiation and lengthened animal survival. Therefore cox inhibitors may have therapeutic value in the treatment of certain acute leukemias. Leighton Grimes (Cincinnati Children’s Hospital Medical Center, OH, USA) demonstrated that miR21 and miR196b are overexpressed in these MLL-AF4 leukemic cells, and that infusion of antagomirs targeting these two miR significantly prolonged survival of recipient mice. Michael Heuser (Terry Fox Lab) showed that retroviral transduction of the meningioma-1 (MN1) gene, a very poor prognosis marker in acute myeloid leukemia, into common myeloid progenitors (CMP) also generated leukemia-initiating cells. Although transduction of MN1 into GMP never caused leukemia, co-transduction with Meis1 into GMP caused leukemia. Meis1 could work by further opening the chromatin in GMP, enabling MN1 to target more genes leading to leukemia. From these different mouse models, it is clear that leukemia-initiating cells need not arise from self-renewing noncommitted HSCs.

The next annual meeting will be held in Melbourne, Australia, in September 2010.

Acknowledgements

Jean-Pierre Levesque is a Senior Research Fellow of the Cancer Council of Queensland, Ingrid G Winkler is a Research Fellow of the National Health and Medical Research Council of Australia. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.