CD34+ hematopoietic progenitor cells are not a major reservoir of the human immunodeficiency virus.

Hematologic abnormalities occur in the majority of patients with acquired immunodeficiency syndrome (AIDS). Infection of the hematopoietic progenitor cells has been proposed as a potential explanation. In this study, different bone marrow cell populations, including the CD34+ hematopoietic progenitor cells, were purified by a fluorescence-activated cell sorter (FACS) and analyzed for the presence of human immunodeficiency virus-1 (HIV-1) proviral DNA using the polymerase chain reaction. A group of 14 patients with AIDS or AIDS-related complex (ARC) was studied (11 with peripheral blood cytopenias). The CD4+ helper cells in the bone marrow were found positive for HIV-1 DNA in all patients. In contrast, CD34+ progenitor cells were positive in only one patient. Two monocyte samples and two samples of CD4-/CD34- lymphocytes/blasts (mainly B and CD8 lymphocytes) were positive. Proviral DNA could not be detected in granulocytes. FACS analysis showed that the percentage of CD34+ hematopoietic progenitor cells was not altered in the bone marrow of AIDS patients in comparison with the HIV-1 seronegative controls. In contrast, the number of CD4+ lymphocytes was markedly reduced in the bone marrow of AIDS patients. These results show that the hematologic abnormalities in AIDS patients are neither explained by direct infection of the hematopoietic progenitor cells with HIV-1 nor by a depletion of progenitor cells.     

Reduced mobilization of CD34+ stem cells in advanced human immunodeficiency virus type 1 disease

Granulocyte colony-stimulating factor (r-met Hu G-CSF; filgrastim; 10 microgram/kg/day for 7 days) was used to mobilize CD34+stem cells into the peripheral blood of human immunodeficiency virus type 1 (HIV-1)-infected individuals and a group of HIV-1-uninfected donors as a measure of immunologic reserve in HIV-1-infected people. G-CSF mobilized CD34+ cells of HIV-1-infected individuals with cell counts >500 CD4+ cells/mm3, as well as in HIV-1-uninfected donors. In contrast, CD34 cell mobilization was significantly blunted in HIV-1-infected individuals with cell counts <500 CD4+ cells/mm3 (<200 cell days vs. >650 cell days, P<.0005, compared with the >500 CD4+ cell cohort). At least 1.75x10(7) CD34 cells were harvested by leukapheresis from patients in each study cohort. CD34+ cell viability and the ability to differentiate precursor cells into myeloid and erythroid progenitor cells were not affected by HIV-1 infection.     

In vitro expansion of human peripheral blood CD34+ cells

To elucidate the role of recombinant human colony-stimulating factors (CSFs) for expanding peripheral blood (PB) CD34+ cells, these cells were purified up to 94.5% +/- 1.3% and the effects of individual and combined CSFs on the proliferation and differentiation of these cells were studied in a 7-day suspension culture. The majority of CD34+ cells coexpressed CD38 (81.8% +/- 5.1%), but was negative for CD33 (88.5% +/- 3.4%). Among the individual CSFs examined, recombinant interleukin-3 (rIL-3) was identified as the most potent factor for expanding PB progenitor cells and increased nonerythroid progenitor cells 13- +/- 4- fold (P < .01). Recombinant granulocyte-macrophage colony-stimulating factor (rGM-CSF), recombinant granulocyte-CSF (rG-CSF), recombinant macrophage-CSF (rM-CSF), rIL-6, rIL-11, and recombinant stem cell factor (rSCF) did not alone expand nonerythroid progenitor cells. A combination of 5 CSFs, ie, rIL-3, rIL-6, rGM-CSF, rG-CSF, and rSCF, was identified as the most potent combination of those tested and increased nonerythroid progenitor cells 57- +/- 11-fold. After a 7-day suspension culture of CD34+ cells with these 5 CSFs, CD34+ cells expanded 14.5- fold, and CD34+/CD33- cells and CD34+/CD33+ cells were also expanded 2.9-fold and 307-fold, respectively. Most secondary colonies derived from expanded cells were small; however, the absolute number of large- sized colonies expanded 5.9- +/- 3.3-fold. Thus, the combination of CSFs can achieve a degree of amplification of PB CD34+ cells. The capability of in vitro expansion of PB CD34+ cells as an adjunct to PB stem cell transplantation is worthy of consideration.     

Impaired in vitro growth of purified (CD34+) hematopoietic progenitors in human immunodeficiency virus-1 seropositive thrombocytopenic individuals.

In this report the role played by human immunodeficiency virus type-1 (HIV-1) in the pathogenesis of HIV-1-related thrombocytopenia was investigated. CD34+ hematopoietic stem/progenitor cells were purified from the bone marrow (BM) of HIV-1(+) thrombocytopenic patients, HIV-1(+) nonthrombocytopenic individuals, HIV-1(-) patients with immune thrombocytopenic purpura, and HIV-1(-) normal donors. CD34+ cells from HIV-1(+) thrombocytopenic individuals alone showed a reduced capacity to give rise to megakaryocytic colonies (CFU-Meg) and also a progressive and significant decline in cell number when placed in liquid culture containing recombinant human interleukin-3 (rIL-3). This decline involved not only megakaryocyte but also erythroid and granulocyte/macrophage progenitors. The defects in megakaryocyte colony formation and CD34+ cell growth did not result from a productive HIV-1 infection of CD34+ cells. Moreover, HIV-1 DNA was absent from CD34+ cells in 10 of 12 thrombocytopenic patients examined. On the other hand, the decreased survival/proliferation of CD34+ cells in liquid culture, within the HIV-1(+) thrombocytopenic patients, was correlated with the presence of HIV-1 p24 antigen in BM plasma. These results demonstrate an impairment of CD34+ cells in HIV-1(+) individuals presenting thrombocytopenia as the only hematologic manifestation. Furthermore, these findings suggest that increased viral replication in the BM microenvironment may cause this impairment and possibly contributes to HIV-induced thrombocytopenia.     

Ex vivo expansion of CD34+/CD41+ late progenitors from enriched peripheral blood CD34+ cells

 In our experience, patients with neuroblastoma who undergo transplantation with CD34+ cells following high-dose chemotherapy have prolonged delays in platelet recovery. In vitro expansion of megakaryocyte (MK) cells may provide a complementary transplant product able to enhance platelet production in the recipient. We investigated the ability of a combination of various hematopoietic growth factors to generate ex vivo MK progenitors. Immunoselected CD34+ cells from peripheral blood stems cells (PBSCs) were cultured in media with or without serum, supplemented by IL-3, IL-6, IL-11, SCF, TPO, Flt-3 ligand, and MIP-1α. In terms of MK phenotypes, we observed a maximal expansion of CD61+, CD41+, and CD42a of 69-, 60-, and 69-fold, respectively, i.e., 8–10 times greater than the expansion of total cell numbers. Whereas the absolute increment of CD34+ cells was slightly elevated (fourfold) we showed increases of 163-, 212-, and 128-fold for CD34+/CD61+, CD34+/CD41+, and CD34+/CD42a+ cells, respectively. We obtained only a modest expansion of CFU-MKs after only 4 days of culture (fourfold) and similar levels of CFU-MKs were observed after 7 days (fivefold). Morphology and immunohistochemistry CD41+ analyses confirmed expansion of a majority of CD41+ immature cells on days 4 and 7, while on day 10 mature cells began to appear. These results show that primarily MK progenitors are expanded after 4 days of culture, whereas MK precursor expansion occurs after 7 days. When we compared the two culture media (with and without serum) we observed that increases of all specific phenotypes of the MK lineage were more elevated in serum-free culture than in medium with serum. This difference was especially marked for CD34+/CD61+ and CD34+/CD41+ (163 vs 42 and 212 vs 36, respectively). We contaminated CD34+ cells with a neuroblastoma cell line and we observed no expansion of malignant cells in our culture conditions (RT-PCR for tyrosine hydroxylase positive at day 4 and negative at day 7). With our combination of hematopoietic growth factors we are able to sufficiently expand ex vivo MK late progenitor cells to be used as complementary transplant products in neuroblastoma patients who undergo transplantation with CD34+ cells. It is possible that these committed MK late progenitors could accelerate short-term platelet recovery in the recipient until more primitive progenitor cells have had time to engraft. Received: February 1, 1999 / Accepted: June 1, 1999     

Mobilization of CD34+ progenitor cells by granulocyte colony- stimulating factor in human immunodeficiency virus type 1-infected adults

We conducted a clinical trial to determine the feasibility of growth factor mobilization of CD34+ progenitor cells in human immunodeficiency virus type 1 (HIV-1)-infected individuals. Eight asymptomatic, HIV-1- infected adults (median CD4+ T-cell count, 415 cells/microL), received 480 micrograms/d of granulocyte colony-stimulating factor (G-CSF) for 6 days without evidence of viral activation. Despite concerns that HIV-1 might inhibit hematopoiesis, CD34+ cells were successfully mobilized to the periphery of all donors, independent of the baseline CD4+ T-cell count, and the status of antiretroviral therapy. Leukapheresis was performed on day 6, and yielded a median of 194 x 10(6) CD34+ cells per leukapheresis (n = 7). CD34-enriched cells from the leukapheresis were predominantly myeloid-committed, but between 0.2% and 1.7% were primitive CD34+/CD38- progenitors. A median of 21.7% of the mobilized CD34+ cells were dimly positive for CD4. Consequently, CD34(+)-enriched cells were purified on the cell sorter (mean purity, 97.7% +/- 2.4%; n = 7), and examined for HIV-1 DNA. Purified CD34+ cells from two of seven donors were polymerase chain reaction (PCR)-positive for HIV-1, but only from one of three samples from each donor. We conclude that G- CSF can safely mobilize CD34+ progenitor cells in HIV-1-infected subjects, and that these cells are suitable for consideration in gene- transfer strategies.     

Evidence for a human immunodeficiency virus type 1-mediated suppression of uninfected hematopoietic (CD34+) cells in AIDS patients.

Hematopoietic progenitor (CD34+) cells were purified from the bone marrow of 6 human immunodeficiency virus (HIV) type 1-seropositive cytopenic patients and 10 healthy donors. HIV-1-seropositive patients showed a reduced number of granulocyte/macrophage, erythroid, and megakaryocyte progenitors and also a progressive and significant decline of numbers of CD34+ cells in liquid culture, which did not result from a productive or latent HIV-1 infection of CD34+ cells. However, all HIV-1-seropositive patients showed signs of active viral replication at the bone marrow level. Moreover, virus isolates from 3 HIV-1-seropositive patients showed a dose-dependent inhibition on growth of normal CD34+ cells. This suppressive activity was almost completely reversed by incubating the virus isolates with an anti-gp120 polyclonal antibody before adding to normal CD34+ cells.     

Interferon-gamma enhances factor-dependent myeloid proliferation of human CD34+ hematopoietic progenitor cells.

Numerous studies have shown that interferon-gamma (IFN gamma) inhibits the proliferative effects of colony-stimulating factors (CSFs) on human bone marrow cells. In the present study we investigated the effects of IFN gamma and other described inhibitory factors on the proliferation of highly purified CD34+ human hematopoietic progenitor cells (HPC) in response to recombinant CSFs. While transforming growth factor-beta (TGF beta) and IFN alpha were highly inhibitory, IFN gamma strongly potentiated interleukin-3 (IL-3) and, to a lesser extent, granulocyte-macrophage-CSF (GM-CSF) induced growth of CD34+ HPC. IFN gamma had no significant proliferative effect per se, and did not affect granulocyte-CSF (G-CSF)-dependent cell proliferation. Within 10 days the number of viable cells generated in the presence of IL-3 + IFN gamma was two times higher than in the presence of IL-3 alone. Limiting dilution analysis showed that IFN gamma acts directly on its target cell to increase the frequency of IL-3-responding cells without affecting the average size of the IL-3-dependent clones. Enhanced frequency of IL-3- and GM-CSF-responding cells was also observed in colony assays where the addition of IFN gamma increased by twofold to threefold the number of granulocyte colony-forming units (CFU-G), macrophage CFUs (CFU-M), granulocyte-macrophage CFUs (CFU-GM), and mixed erythroid (E-MIX). In contrast, IFN gamma did not affect the generation of erythroid burst-forming units (BFU-e) in such cultures. In longer-term culture, the combination of IFN gamma and IL-3 did not alter the lineage distribution of the cells when compared with IL-3 alone. However, after 15 days, when mature cells were present in the cultures, IFN gamma displayed cell concentration-related growth-inhibitory effects. Thus, IFN gamma appears to stimulate the early stage of myelopoiesis by enhancing the frequency of growth factor-responding cells but, unlike tumor necrosis factor alpha (TNF alpha), does not alter cell differentiation.     

Granulomonocyte-associated lysosomal protein expression during in vitro expansion and differentiation of CD34+ hematopoietic progenitor cells

Using an in vitro expansion and differentiation system for human CD34+ cord blood (CB) progenitor cells, we analyzed the induction and expression kinetics of the granulomonocyte associated lysosomal proteins myeloperoxidase (MPO), lysozyme (LZ), lactoferrin (LF), and macrosialin (CD68). Freshly isolated CD34+ CB cells were negative for LZ and LF, and only small proportions expressed MPO (4% +/- 2%) or CD68 (3% +/- 1%). Culturing of CD34+ cells for 14 days with interleukin (IL)- 1, IL-3, IL-6, stem cell factor, granulocyte-macrophage colony- stimulating factor (GM-CSF), and G-CSF resulted in on average a 1,750- fold amplification of cell number, of which 83% +/- 7% were MPO+. Without addition of GM-CSF and G-CSF, lower increases in total cell numbers (mean, 211-fold) and lower proportions of MPO+ cells (54% +/- 11%) were observed. The proportion of MPO+ cells slightly exceeded but clearly correlated with the proportion of cells positive for the granulomonocyte associated surface molecules CD11b (Mac-1), CD15 (LeX), CD64 (Fc gamma RI) CD66, or CD89 (Fc alpha R). At day 14 MPO+ and LZ+ cells were virtually identical. However, at earlier time points during culture (days 4 and 7), single MPO+ or LZ+ cell populations were also observed, which only later acquired LZ and MPO, respectively. Maturation of cells into the neutrophilic pathway was indicated by the acquisition of MPO, followed by LZ. In contrast, maturation of cells into the monocytic pathway was indicated by the acquisition of LZ followed by MPO and CD14. CD68 was found to be expressed at day 4 by the majority of cells and was not restricted to the granulomonocytic cells, as cells with megakariocytic (CD41+) or erythroid (CD71hi) features were CD68+. LF expression was observed only in GM- plus G-CSF- supplemented cultures, in which only 26% +/- 5% of cells expressed LF by day 14.     

Human CD34+AC133+VEGFR-2+ cells are not endothelial progenitor cells but distinct, primitive hematopoietic progenitors

OBJECTIVE: Endothelial progenitor cells (EPCs) are used for angiogenic therapies or as biomarkers to assess cardiovascular disease risk. However, there is no uniform definition of an EPC, which confounds EPC studies. EPCs are widely described as cells that coexpress the cell-surface antigens CD34, AC133, and vascular endothelial growth factor receptor-2 (VEGFR-2). These antigens are also expressed on primitive hematopoietic progenitor cells (HPCs). Remarkably, despite their original identification, CD34+AC133+VEGFR-2+ cells have never been isolated and simultaneously plated in hematopoietic and endothelial cell (EC) clonogenic assays to assess the identity of their clonal progeny, which are presumably the cellular participants in vascular regeneration. METHODS: CD34+AC133+VEGFR-2+ cells were isolated from human umbilical cord blood (CB) or granulocyte colony-stimulating factor-mobilized peripheral blood and assayed for either EPCs or HPCs. RESULTS: CD34+AC133+VEGFR-2+ cells did not form EPCs and were devoid of vessel forming activity. However, CD34+AC133+VEGFR-2+ cells formed HPCs and expressed the hematopoietic lineage-specific antigen, CD45. We next tested whether EPCs could be separated from HPCs by immunoselection for CD34 and CD45. CD34+CD45+ cells formed HPCs but not EPCs, while CD34+CD45- cells formed EPCs but not HPCs. CONCLUSIONS: Therefore, CD34+AC133+VEGFR-2+ cells are HPCs that do not yield EC progeny, and the biological mechanism for their correlation with cardiovascular disease needs to be reexamined.     

Oligodeoxynucleotides antisense to c-abl specifically inhibit entry into S-phase of CD34+ hematopoietic cells and their differentiation to granulocyte-macrophage progenitors

A number of experimental observations suggest that the proto-oncogene c- abl participates in the regulation of hematopoietic cell growth. We used an antisense strategy to study the relationship between c-abl expression and hematopoietic cell proliferation and differentiation. Purified normal human bone marrow-derived CD34+ cells were obtained by immunomagnetic selection and incubated with 18-base-unmodified antisense oligodeoxynucleotides complementary to the first six codons of the two alternative first exons of c-abl, la and lb. At the end of incubation, an aliquot of cells was assayed for clonogenic growth and the remainder was used for flow cytometric analyses. Cell kinetics were evaluated by means of both single parameter DNA and bivariate DNA/bromodeoxyuridine (BrdU) flow cytometry. Apoptosis was routinely studied by DNA flow cytometric analysis and, in some cases, also through DNA agarose gel electrophoresis for detection of oligonucleosomal DNA fragments. Expression of differentiation markers was studied by flow cytometry. Exposure to antisense oligonucleotides specifically inhibited the accumulation of c-abl mRNA in CD34+ cells. Preincubation with the c-abl antisense oligomers reduced the proportion of cells in S-phase from 19% +/- 5% (mean +/- SD) to 7% +/- 4% (P < .05), and BrdU labeling from 13% +/- 6% to 6% +/- 3% (P < .05). Flow cytometry and DNA agarose gel electrophoresis showed that treated CD34+ cells accumulated in the G0/G1 region of the DNA histogram with no evidence of either differentiation or apoptosis. By contrast, both growth factor deprivation and exposure of CD34+ cells to the tyrosine kinase inhibitor tyrphostin AG82 clearly induced apoptosis. When cells were preincubated with antisense oligonucleotides and then plated for evaluation of colony formation, this resulted in a significant inhibition of colony forming unit granulocyte-macrophage growth (from 44 +/- 15 to 22 +/- 9; P < .01) but had no effect on burst-forming unit erythroid growth (24 +/- 11 v 21 +/- 11; P < .05). These results suggest that c-abl expression is critical for entry of human CD34+ hematopoietic cells into S-phase and for their differentiation to granulocyte-macrophage progenitors. They also indicate that other tyrosine kinases besides p145c-alb are active in the prevention of apoptosis, so that inhibition of c-abl RNA accumulation arrests CD34+ cells in G0/G1 without activating programmed death.     

Myeloperoxidase expression in CD34+ normal human hematopoietic cells

Bone marrow (BM), adult peripheral blood (aPB), and umbilical cord blood (CB) samples contain small proportions of CD34+ cells that include virtually all hematopoietic progenitor cells. Myeloperoxidase (MPO) is considered to be selectively expressed in cells committed to granulomonocytic differentiation. Using flow cytometry and an antibody against MPO, we studied at which stage of normal hematopoietic differentiation CD34+ cells being to express MPO. We consistently observed a characteristic MPO/CD34 staining pattern and found that 35% +/- 9% of CD34+ BM cells express MPO. The MPO+ CD34+ subset and the CD33+ CD34+ subset were of similar size and overlapped considerably. MPO+ CD34+ cells expressed high levels of HLA-D molecules, were weakly CD71/transferrin receptor positive to negative, were CD45RA+ and lacked the CD45RO isoform of the leukocyte common antigen. Additionally, MPO+ CD34+ cells were on average larger in size than MPO- CD34+ cells. Virtually identical phenotypic features have previously been described for in vitro colony-forming granulomonocytic progenitor cells. In vitro clonogenic assays performed with MPO-enriched and MPO-depleted fractions of CD34+ BM cells performed by us also suggest, but do not formally prove, that at least a portion of MPO+ CD34+ cells have in vitro cluster (10 to 50 cells/colony) or colony-forming unit granulocyte-macrophage (> or = 50 cells/colony) forming capacity. CD34+ cells from CB and aPB resembled CD34+ BM cells in that considerable proportions of them coexpressed CD33. However, in contrast to BM, CD34+ cells from CB and aPB samples lacked significant MPO expression and, in line with this, the majority of them (CB, 59% +/- 7%; aPB, 66% +/- 5%) coexpressed CD45RO.     

Integration of adeno-associated virus vectors in CD34+ human hematopoietic progenitor cells after transduction

Gene transfer vectors based on adeno-associated virus (AAV) appear promising because of their high transduction frequencies regardless of cell cycle status and ability to integrate into chromosomal DNA. We tested AAV-mediated gene transfer into a panel of human bone marrow or umbilical cord-derived CD34+ hematopoietic progenitor cells, using vectors encoding several transgenes under the control of viral and cellular promoters. Gene transfer was evaluated by (1) chromosomal integration of vector sequences and (2) analysis of transgene expression. Southern hybridization and fluorescence in situ hybridization analysis of transduced CD34 genomic DNA showed the presence of integrated vector sequences in chromosomal DNA in a portion of transduced cells and showed that integrated vector sequences were replicated along with cellular DNA during mitosis. Transgene expression in transduced CD34 cells in suspension cultures and in myeloid colonies differentiating in vitro from transduced CD34 cells approximated that predicted by the multiplicity of transduction. This was true in CD34 cells from different donors, regardless of the transgene or selective pressure. Comparisons of CD34 cell transduction either before or after cytokine stimulation showed similar gene transfer frequencies. Our findings suggest that AAV transduction of CD34+ hematopoietic progenitor cells is efficient, can lead to stable integration in a population of transduced cells, and may therefore provide the basis for safe and efficient ex vivo gene therapy of the hematopoietic system.     

CD34+ endothelial cell lines derived from murine yolk sac induce the proliferation and differentiation of yolk sac CD34+ hematopoietic progenitors

Embryonic hematopoiesis is initiated in part in the blood islands of the yolk sac. Previous confocal microscopic analysis has shown that the CD34 antigen, a mucin-like cell surface glycoprotein that is expressed by hematopoietic progenitors and all endothelial cells of the adult and embryo, is also found on a subset of luminal hematopoietic-like cells in the yolk sac blood islands as well as on the vascular endothelium lining these early hematopoietic locations. We show here that, as in all other hematopoietic sites thus far examined, immunoaffinity- purified CD34+ nonadherent cells from murine yolk sacs contain the vast majority of erythroid and myeloid progenitor cell colony forming activity. To examine the developmental interactions between these CD34+ hematopoietic progenitor cells of the yolk sac and the CD34+ yolk sac endothelium, we have immunaffinity-purified adherent endothelial cells from day 10.5 yolk sacs using CD34 antiserum and produced cell lines by transformation with a retrovirus expressing the polyoma middle T antigen. Analysis of these cell lines for CD34, von Willebrand's factor, FLK 1 and FLT 1 expression, and capillary growth in Matrigel indicates that they appear to be endothelial cells, consistent with their original phenotype in vivo. Coculture of yolk sac CD34+ hematopoietic cells on these endothelial cell lines results in up to a 60-fold increase in total hematopoietic cell number after approximately 8 days. Analysis of these expanded hematopoietic cells showed that the majority were of the monocyte/macrophage lineage. In addition, examination of the cultures showed the rapid formation of numerous cobblestone areas, a previously described morphologic entity thought to be representative of early pluripotential stem cells. Scrutiny of the ability of these endothelial cell lines to expand committed progenitor cells showed up to a sixfold increase in erythroid and myeloid colony- forming cells after 3 to 6 days in culture, consistent with the notion that these embryonic endothelial cells mediate the expansion of these precursor cells. Polymerase chain reaction analyses showed that most of the cell lines produce FLK-2/FLT-3 ligand, stem cell factor, macrophage colony-stimulating factor, leukemia-inhibitory factor, and interleukin- 6 (IL-6), whereas there is a generally low or not measurable production of granulocyte colony-stimulating factor, granulocyte-macrophage colony- stimulating factor, IL-1, IL-3, transforming growth factor beta-1, erythropoietin, or thrombopoietin. The output of mature hematopoietic cells from these cocultures can be modified to include an erythroid population by the addition of exogenous erythropoietin. These data suggest that endothelial cell lines derived form the yolk sac provide an appropriate hematopoietic environment for the expansion and differentiation of yolk sac progenitor cells into at least the myeloid and erythroid lineages.     

Human mesenchymal stem cells promote human osteoclast differentiation from CD34+ bone marrow hematopoietic progenitors.

Interactions between osteoclast progenitors and stromal cells derived from mesenchymal stem cells (MSCs) within the bone marrow are important for osteoclast differentiation. In vitro models of osteoclastogenesis are well established in animal species; however, such assays do not necessarily reflect human osteoclastogenesis. We sought to establish a reproducible coculture model of human osteoclastogenesis using highly purified human marrow-derived MSCs (hMSCs) and CD34+ hematopoietic stem cells (HSCs). After 3 weeks, coculture of hMSCs and HSCs resulted in an increase in hematopoietic cell number with formation of multinucleated osteoclast-like cells (Ocls). Coculture of hMSCs with HSCs, transduced with a retroviral vector that expresses enhanced green fluorescent protein, produced enhanced green fluorescent protein+ Ocls, further demonstrating that Ocls arise from HSCs. These Ocls express calcitonin and vitronectin receptors and tartrate-resistant acid phosphatase and possess the ability to resorb bone. Ocl formation in this assay is cell contact dependent and is independent of added exogenous factors. Conditioned medium from the coculture contained high levels of interleukin (IL)-6, IL-11, leukemia inhibitory factor (LIF), and macrophage-colony stimulating factor. IL-6 and LIF were present at low levels in cultures of hMSCs but undetectable in cultures of HSCs alone. These data suggest that coculture with HSCs induce hMSCs to secrete cytokines involved in Ocl formation. Addition of neutralizing anti-IL-6, IL-11, LIF, or macrophage-colony stimulating factor antibodies to the coculture inhibited Ocl formation. hMSCs seem to support Ocl formation as undifferentiated progenitor cells, because treatment of hMSCs with dexamethasone, ascorbic acid, and beta-glycerophosphate (to induce osteogenic differentiation) actually inhibited osteoclastogenesis in this coculture model. In conclusion, we have developed a simple and reproducible assay using culture-expanded hMSCs and purified HSCs with which to study the mechanisms of human osteoclastogenesis.     

Aorta-associated CD34+ hematopoietic cells in the early human embryo

Hematopoiesis is established from circulating blood stem cells that seed the embryonic rudiments of blood-forming tissues, a basic notion in developmental hematology. However, the assumption that these stem cells originate from the extraembryonic mesoderm, where primitive hematopoiesis is initiated by intrinsic precursors, has been reconsidered after analysis of blood cell development in avian embryo chimeras: yolk-sac-derived stem cells do not contribute significantly to the definitive blood system, whose first forerunners develop independently along the ventral aspect of the embryonic aorta. Recently, the homologous intraembryonic tissues of the mouse have been submitted to sensitive in vivo and in vitro assays, which showed that they also harbor multipotential hematopoietic stem cells. We have now identified a dense population of hematogenous cells, marked by the surface expression of the CD34 glycoprotein, associated with the ventral endothelium of the aorta in the 5-week human embryo. Therefore, we extend to the human species the growing evidence that intraembryonic hematopoietic cells developing independently of the yolk sac might be the real stem of the whole blood system.