Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jul 1.
Published in final edited form as: Br J Haematol. 2014 Mar 27;166(2):268–278. doi: 10.1111/bjh.12870

Temporal Definition of Haematopoietic Stem Cell Niches in a Large Animal Model of In Utero Stem Cell Transplantation

Christine Jeanblanc 1, A Daisy Goodrich 1, Evan Colletti 1, Saloomeh Mokhtari 2, Christopher D Porada 2, Esmail D Zanjani 1,*, Graça Almeida-Porada 2,*
PMCID: PMC4079736  NIHMSID: NIHMS575200  PMID: 24673111

Abstract

The fetal sheep model has served as a biologically relevant and translational model to study in utero haematopoietic stem cell transplantation (IUHSCT), yet little is known about the ontogeny of the bone marrow (BM) niches in this model. Because the BMmicroenvironment plays a critical role in the outcome of haematopoietic engraftment, we have established the correlation between the fetal-sheep and fetal-human BM niche ontogeny, so that studies addressing the role of niche development at the time of IUHSCT could be accurately performed. Immunofluorescence confocal microscopic analysis of sheep fetal bone from gestational days (gd) 25-68 showed that the BM microenvironment commences development with formation of the vascular niche between 25-36 gd in sheep; correlating with the events at 10-11 gestational weeks (gw) in humans. Subsequently, between 45-51 gd in sheep (~14 gw in humans), the osteoblastic/endosteal niche started developing, the presence of CD34+CD45+ cells were promptly detected, and their number increased with gestational age. IUHSCT, performed in sheep at 45 and 65 gd, showed significant haematopoietic engraftment only at the later time point, indicating that a fully functional BM microenvironment improved engraftment. These studies show that sheep niche ontogeny closely parallels human, validating this model for investigating niche influence/manipulation in IUHSCT engraftment.

Keywords: In utero, Transplantation, Niche, Haematopoietic, Vascular, Osteoblast

INTRODUCTION

Advances in molecular biological techniques and obstetrical screening procedures, such as chorionic villous sampling and amniocentesis, now allow genetic disorders to be diagnosed much earlier in gestation, thus making it possible to begin exploring the feasibility of clinical correction before birth. Among the possible treatments that could be implemented prenatally, in utero haematopoietic stem cell transplantation (IUHSCT) has been touted as a promising approach for correcting an array of congenital haematological and immunological disorders that collectively account for 20% of infant death worldwide (Diukman and Golbus 1992, Merianos, et al 2008, Roybal, et al 2010, Slavin, et al 1992, Surbek, et al 2001, Westgren, et al 1996).

Several animal models have contributed to the understanding of fetal immunology and helped delineate the parameters governing successful IUHSCT (Flake 2004, Fujiki, et al 2003, Huang, et al 2002, Merianos, et al 2008, Merianos, et al 2009, Peranteau, et al 2007, Pixley, et al 1994, Schoeberlein, et al 2004, Shields, et al 1995, Sun, et al 2007a, Sun, et al 2007b, Tarantal, et al 2000, Xiao, et al 2003). Fetal sheep share many important physiological and developmental characteristics with humans, and have, therefore, been used extensively in the study of mammalian fetal physiology, and results obtained with this model have been directly applicable to the understanding of human fetal growth and development. Some specific characteristics that make sheep well-suited for developing/testing IUHSCT-based treatments and obtaining results of high clinical relevance are: 1) close size to humans; 2) immune development closely parallels that of human (Maddox, et al 1987a, Maddox, et al 1987b, Maddox, et al 1987c, Miyasaka and Trnka 1986, Osburn 1981, Sawyer, et al 1978); 3) sheep exhibit the same pattern of fetal to adult haemoglobin switching as humans, and also undergo naturally occurring changes in the primary sites of haematopoiesis from yolk sac to fetal liver and finally to the bone marrow (BM) near the end of gestation (Zanjani, et al 1996); 4) as a large, long-lived animal (lifespan 8-12 years), sheep allow critical questions of long-term efficacy and safety to be properly addressed; and 5) the long gestational period in sheep (145 days) provides sufficient temporal resolution to translate findings obtained in sheep into human parameters. It was with these advantages in mind that Flake et al (1996) used the fetal sheep model to delineate the conditions that enabled them to achieve the first clinical cure with IUHSCT, completely correcting a child with X-linked severe combined immunodeficiency.

Currently, 17 years after this first clinically curative IUHSCT, 50 human patients have now been treated with this procedure, for 14 different genetic disorders (Muench 2005, Muench and Barcena 2004). However, complete therapeutic success has only been seen in patients with primary immunodeficiencies, in whom donor haematopoietic stem cells (HSC) would be predicted to have an obvious proliferative/survival advantage over host cells (Flake, et al 1996, Merianos, et al 2008, Muench 2005). Thus, for IUHSCT to fulfill its promise of correcting a wider range of inherited disorders, a better understanding of the factors limiting engraftment in utero is needed to develop strategies to overcome these barriers and achieve therapeutic levels of engraftment (Flake and Zanjani 1999, Muench 2005, Muench and Barcena 2004, Roybal, et al 2010). Studies in mice have identified several properties of the developing fetus that may negatively impact upon its ability to serve as an amenable HSCT recipient. These include: competition from endogenous host cells (Flake, et al 1996), more significant fetal immune barriers than originally supposed (Peranteau, et al 2007), and even maternal immunity, in species that allow transplacental passage of maternal cells (Merianos, et al 2009, Nijagal, et al 2011). While these barriers could potentially be reduced/overcome through the use of autologous or maternal donor cells, one aspect of the developing fetus that has not yet been explored in detail in the context of IUHSCT is the degree of maturity and receptivity of nascent niches within the BM microenvironment, that are required for thr engraftment of donor cells (Flake, et al 1986, Flake and Zanjani 1999, Zanjani, et al 1992a, Zanjani, et al 1994, Zanjani, et al 1992b). Because the BM microenvironment plays a critical role in the outcome of haematopoietic engraftment following postnatal HSC transplantation, here we delineated fetal BM niche ontogeny in sheep and human, and show that sheep closely parallels human with respect to the key temporal events in these two species, including development of the vascular and osteoblastic/endosteal niches, and the first appearance of primitive haematopoietic cells within these niches during development. We then performed IUHSCT with allogeneic sheep and xenogeneic (human) HSC to define the temporal window during which HSC engraftment was optimal.

In conclusion, our studies demonstrate that the sheep is a valuable and clinically relevant model for delineating the role of niche development at the time of IUHSCT, and suggest that a better understanding of niche availability/receptivity can be used to develop approaches for maximizing HSC engraftment, thus enabling IUHSCT to deliver its promise of correction/cure in a wider range of genetic diseases.

MATERIALS AND METHODS

HSC preparation and in utero HSC transplantation

All studies involving animals were approved by the University of Nevada, Reno Institutional Animal Care and Use Committee and were in accordance with Public Health Service guidelines. Random-bred, time-dated, pregnant Dorset-merino sheep (term: 145 days) were used in all studies.

Allogeneic transplantation I; with T cell-depleted, BM mononuclear cells

Sheep that had been typed at the β-globin locus as homozygous AA or BB (Zanjani, et al 1979) were used to facilitate detection of donor (sheep) haematopoietic chimerism. Each homozygous type BB fetus was transplanted intraperitoneally at either gestational day (gd) 45 or 65 (10 animals/group) with 5 × 105 T cell-depleted, BM mononuclear cells obtained from 1-year-old type AA donors of the opposite sex (Flake, et al 1986).

Allogeneic transplantation II; with BM-derived CD34+ cells

In a second allogeneic transplantation study, sheep BM-derived CD34+ cells were isolated from the mononuclear fraction by magnetic cell sorting (Miltenyi Biotec, Inc., Auburn, CA) using a monoclonal antibody to sheep CD34(Porada, et al 2008). Sheep CD34+ cells were stimulated for 24 h in serum-free QBSF-60 media (Quality Biologicals, Gaithersburg, MD), supplemented with the following human cytokine cocktail (all from Peprotech, Rocky Hill, NJ): stem cell factor (SCF) at 50 ng/ml, Fms-like tyrosine kinase-3 (FLT3-L) at 50 ng/ml, and thrombopoietin (TPO) at 20 ng/ml. The CD34+ cells were then transduced, using standard procedures, with a human immunodeficiency virus (HIV)-based vector encoding Enhanced Green Fluorescent Protein (eGFP; Capital Biosciences, Rockville, MD), and extensively washed, prior to transplantation into eight 65-gd recipient sheep fetuses, at a dose of 1.4 × 106 cells/fetus.

Xenogeneic transplantation with human BM CD34+ cells

For studies involving xenogeneic transplantation, each fetus received 4 × 104-5 × 105 CD34+ human BM cells at either 45 gd or 65 gd (10 animals/group). All transplants were performed using methods previously described in detail (Flake, et al 1986, Zanjani et al 1992). Human cells were obtained from normal, healthy volunteers. All studies involving the use of human cell/tissues were approved by the University of Nevada, Reno Institutional Review Board.

Monitoring of donor cell engraftment

Peripheral blood, BM cells and fetal tissue were obtained from the recipient fetuses at 10, 60 and 130 days (1 month after birth) post-transplantation. In allogeneic transplantation studies I, donor cell activity was monitored by haemoglobin analyses of the recipients’ peripheral blood and BM as previously described (Flake, et al 1986, Zanjani, et al 1979), while in allogeneic transplantation studies II, donor cell activity was determined by immunofluorescence staining and confocal imaging to detect the eGFP-expressing cells.

Donor cell engraftment in human xenograft recipients was determined by performing flow cytometry on the recipients’ peripheral blood and BM at early time points post-transplantation (10 or 60 days after transplant) following euthanasia of the recipient fetus, using human-specific antibodies to various haematopoietic cell surface markers as described (Zanjani et al 1992). To assess long-term engraftment, other cohorts of animals were euthanized at 130 days or 9 months post-transplant, and their femurs were collected for immunofluorescence confocal analyses, as detailed below, to quantitate the levels of human cell engraftment.

Fetal bone preparation for immunohistochemisty/immunofluorescence and confocal imaging

For examination of haematopoietic niche development, long bones were obtained from fetal sheep at 25, 36, 40, 46, 51, 65, and 68 gd (5 fetuses/gd). For analyses of human BM niche development, 10-, 11-, 14-, 15- 18- and 20 gestational weeks (gw) human fetal bone samples were purchased from Advanced Bioresources (ABR, Portola Valley, CA) (5 samples /age). Fetal bones were rinsed, fixed in paraformaldehyde and embedded in paraffin. Paraffin-embedded tissue was cut into 5 μm sections, adhered to glass slides, dried and baked. After deparaffinization with xylene (2 × 1 min) and rehydration with 100%, 90%, 75% ethanol for 1 min each, tissue sections were rinsed in deionized water (diH2O), washed in phosphate-buffered saline (PBS) (Gibco-Invitrogen Corp., Camarillo, CA), and blocked with 10% Normal Goat Serum (NGS) (Gibco-Invitrogen) in PBS for 30 min. After 3 rinses in 2% NGS in PBS, the sections were dual-labelled with primary antibodies to various cell types present within the vascular and osteoblastic/endosteal niches. To characterize the vascular niche in sheep, we used rabbit anti-sheep CD31 (PECAM1) (Abcam, Cambridge, MA) and mouse anti-sheep CD34(Porada, et al 2008) while for human, we used rabbit anti-human CD146, rabbit anti-human CD31, and mouse anti human CD34, all from Abcam. For osteoblastic/endosteal niche characterization in sheep and in human, rabbit anti-human/mouse/sheep N-Cadherin (AnaSpec, Fremont, CA) was used to identify preosteoblasts, and anti-human/mouse/sheep Osteopontin (Anaspec) was used to identify mature osteoblasts.

For HSC detection in sheep tissue, the following combinations of primary antibodies were employed: mouse anti-sheep CD34 (Porada, et al 2008) and rabbit anti-human/mouse/rat CD45 (Abcam); mouse anti-sheep CD34 (Porada, et al 2008) and rabbit anti-mouse/human/rat/horse/cow aldehyde dehydrogenase 1/2 (ALDH1/2; Santa Cruz Biotechnology, Santa Cruz, CA). For HSC detection in human tissue, mouse anti-human CD34 (Invitrogen) and rabbit anti-human CD45 (Enzo Life Sciences, Farmingdale, NY) were used. For T cell detection, rabbit anti-sheep CD8 (Abcam) and rabbit anti-sheep CD3 (Abcam) antibodies were used. For detection of eGFP-tagged CD34+ cells, rabbit anti-eGFP (Abcam) and mouse anti-sheep CD34 were used. For human HSC detection in sheep, the mouse anti-human nuclei antibody (Millipore, Billerica, MA) and the rabbit anti-human CD45 (Enzo Life Sciences) were used. Tissue sections were incubated in the above primary antibodies overnight at 4°C at a concentration suggested by the manufacturer, washed with PBS/2% NGS, and incubated for 1 h at 4°C with two of the following AlexaFluor®-conjugated secondary antibodies (Invitrogen) at a concentration of 1:400; AlexaFluor anti-mouse 488 (green), AlexaFluor goat anti-rabbit 488 (green), AlexaFluor goat anti-mouse 532 (yellow), AlexaFluor goat anti-rabbit 647 (red) and Alexa Fluor goat anti-mouse 647 (red). After 3 rinses in PBS/2% NGS, the sections were counterstained with DAPI (4,6 diamidino-2-phenylindole) (Biogenex, San Ramon, CA) for nuclear detection, air-dried and coverslipped with Cytoseal 60 (Thermo Scientific, Portsmouth, NH). Confocal images were subsequently acquired with an Olympus Fluoview FV1000 confocal microscope (Olympus, Melville, NY), with an Olympus UPlanFLN- 40x/1.30 oil, Plan ApoN 60x/1.42 ∞/0.17/FN 26.5 numeric aperture objective lenses, an Olympus 1×81camera, and Olympus FV10 MCPSU. Confocal images were taken as z-stacks with 5-10 slices per image before being projected as 2D images and saved. In each experiment, a slide stained with the two secondary antibodies alone served as a negative control for all slides processed at the same time. Adobe Photoshop CS5 was then used to stack the respective colour channels, assemble multi-panel figures, and perform minimal global processing, such as brightness, contrast adjustment and colour balance.

Haematoxylin and Eosin staining

Paraffin-embedded sheep fetal bones processed as above were cut into 5 μm sections, adhered to glass slides, dried and baked. After deparaffinization with xylene (2 × 1 min) and rehydration with 100%, 90%, 75% ethanol for 1 min each, the sections were rinsed in diH2O. The sections were stained with haematoxylin (Sigma, St. Louis, MO) for 5 min for nuclear staining, and rinsed in tap water. The sections were then placed briefly in Scott’s tap water (Fisher Scientific, Portsmouth, NH), rinsed in tap water, and placed in 1% acid alcohol for a few seconds to lighten the staining, especially outside the nucleus. After rinsing in tap water, the sections were incubated for 5 min in eosin (Sigma) to stain the cytoplasm and cartilage, and again rinsed with tap water. The sections were then allowed to dry in the dark and coverslipped with Cytoseal 60 (Thermo Scientific).

RESULTS

Vascular niche development begins in fetal sheep BM between 25-36 gd; similar changes occur in the human fetus at 10-11 gestational weeks

Confocal microscopy analyses of fetal sheep long bones with antibodies against the endothelial markers CD31 and CD34, demonstrated the presence, at 25 gd, of unevenly distributed endothelial cells within the mesenchymal structure (Figure 1A). In contrast, by 36 gd, the vascular niche was already much more mature in appearance, with abundant erythrocytes circulating in the sinusoid vessels; Figure 1B shows an elongated vessel densely populated with biconcave-shaped red blood cells and some CD8+ lymphocytes. Their detection confirms the successful trafficking of cells in circulation, an indication of the presence of an established vascular network by 36 gd in the sheep BM, and that of immune cells at a very early time point of gestation. Further confirmation of well-formed vasculature at this relatively early stage of marrow development came from analyses of haematoxylin and eosin (H&E)-stained 40-gd fetal sheep bone samples. They revealed the presence of clearly defined sinusoid vessels (Figures 1C, 1D) and elongated arterioles (Figure 1E), the key features of the nascent vasculature. CD34+ALDH+ HSCs within the chondrocytic matrix of the medullary cavity were observed as early as 46 gd in sheep (Figure 1F), suggesting that the niche had sufficiently matured at this stage to support primitive HSCs, at least to some degree. By 51 gd, the vascular microenvironment had considerably expanded within the medullary cavity, as evidenced by the presence of elongated sinusoid vessels, delineated by CD34+ endothelium, containing abundant erythrocytes (Figure 1G). By 65 gd, an even higher proportion of red blood cells were seen within well-defined sinusoids of the developing medullary cavity, suggesting that the vascular niche had attained relative maturity (Figure 1H).

Figure 1. Assessment of vascular niche development in fetal sheep bone marrow (BM).

Figure 1

Open in a new tab

Sheep bone samples, obtained at varying stages of fetal development (25-65 gestational days (gd); term 145 days), were stained and analysed by confocal microscopy to delineate the ontogeny of the vascular niche. (A) [40x] Representative image depicting the nascent vasculature in the 25-gd sheep fetal BM, showing a few CD31+(red)/CD34+(green) endothelial cells associated with a nascent sinusoid vessel. Insets show individual red (CD31) and green (CD34) channels, confirming dual-positivity and thus establishing the endothelial identity of the cells. (B) [40x] Representative section of 36-gd sheep fetal bone, demonstrating CD8+ lymphocytes (red) and biconcave-shaped erythrocytes populating an elongated arteriole of the early BM vascular network (nuclei are blue, and the transmitted display (TD) channel has been overlaid for visualization of morphology). (C-E) Haematoxylin and eosin stained 40-gd fetal sheep bone sections, showing the presence of sinusoid vessels (C [40x]; D [zoom]) and elongated arterioles (E [zoom]). (F) [40x] Confocal image of 46-gd sheep fetal bone revealing the presence of a few CD34+(green) ALDH+(red) HSCs within the chondrocytic matrix (nuclei are stained blue with DAPI). (G) [40x] Representative 51-gd sheep fetal bone, exhibiting numerous red (autofluorescing) erythrocytes populating a vessel delineated by CD34+ endothelium (green). (H) [40x] Image from 65-gd sheep fetal bone depicting the mature vasculature with abundant red fluorescing erythrocytes (red channel) within well-formed sinusoidal vessels.

Although the ontogeny of the BM haematopoietic niches has been fairly well characterized in humans (Charbord, et al 1996), we performed staining of human fetal bones in parallel to the sheep samples to determine whether sheep recapitulate the key temporal events in human marrow ontogeny, and thereby validate the ability of the fetal sheep to serve as an accurate and predictive model for assessing the ontogeny of BM niches, and the role they play in engraftment following IUHSCT.

In similarity to what we had observed in sheep, in the 10-gw human fetus, only a few unevenly distributed CD31+CD34+ endothelial cells could be detected in the mesenchymal matrix (Figure 2A). The medullary cavity in the 10- to 11-gw human fetus merely consisted of a cartilaginous matrix in which CD146+ vascular endothelium formation (Figure 2B) was clearly observed. CD34+CD45+ HSCs were also detected within the BM chondrocytic matrix of the human fetus as early as 10 gw (Figure 2C), suggesting that the BM vascular microenvironment had matured sufficiently at this stage to allow and support primitive HSC trafficking, at least to some degree. By 11 gw, endothelial cells appeared to be adopting a circular pattern, suggesting sinusoidal vessel lumen formation within the mesenchymal matrix (Figure 2D). By that age, well-formed elongated arterioles, delineated by CD146+ vascular endothelium, could also be observed (Figure 2E). By 14 gw in the human fetus, the vascular microenvironment had also considerably expanded within the medullary cavity, as indicated by the presence of abundant red fluorescing erythrocytes circulating within the elongated sinusoid vessels (Figures 2F, 2G). In the human fetus, an even higher proportion of erythrocytes was visible within the medulla by 20 gw, suggesting that the vascular microenvironment had reached relative maturity by that gestational age (Figure 2H).

Figure 2. Assessment of vascular niche development in human fetal BM.

Figure 2

Open in a new tab

Human fetal bone samples were stained for confocal microscopy and processed as described in the methods. The 2-or 3-colour channel composite photographs with or without transmitted display (TD) used DAPI (blue) for cell nuclei staining. (A) [40x] Confocal image acquired at 40x magnification depicting 10-gestational week (gw) human fetal bone. A few isolated CD31+(red)/CD34+(green) endothelial cells can be visualized unorganized within the chondrocytic matrix. Inset: 60x magnification of dual positive CD31+ /CD34+ endothelial cells. (B) [40x] Confocal image of 10-gw human bone showing the linear assembly of CD146+ endothelial cells, possibly the first stage in capillary lumen formation within the bone marrow (BM) cavity. (C) [60x] Image of 10 gw human fetal bone, showing scarce CD34+(green)/CD45+(red) HSCs present within the BM mesenchymal matrix. Inset: close up on the HSC shown in the main picture. (D) [40x] Confocal image of 11-gw human fetal bone (~36-40 gd in sheep) depicting CD31+(red)/CD34+(green) endothelial cells depicting the formation of several capillary lumen within the chondrocytic matrix, the precursor of bone. Inset: 60x magnification of CD31+/CD34+ endothelial cells without TD. (E) [40x] Image of 11-gw human fetal bone revealing the presence of sinusoid vessels lined with CD146+ vascular endothelium (red) within the mesenchymal matrix. (F) [40x] Confocal image from 14-gw human fetal bone showing the BM densely populated with red (autofluorescing) erythrocytes within elongated arterioles. (G) [40x] Image of 14gw human fetal bone section revealing the presence of red (autofluorescing) erythrocytes within an elongated arteriole lined with elongated CD146+ (green) endothelium, attesting to established vasculature. Inset: Confocal image acquired at 60x magnification showing autofluorescing erythrocytes (red) within an elongated arteriole delineated by CD146+ vascular endothelial cells. (H) [40x] Image of 20-gw human fetal bone stained for CD146 revealing the presence of abundant autofluorescing RBCs (red channel) within vessels delineated with circularly arranged CD146+ (green) vascular endothelium.

Collectively, these results demonstrate that early marrow development in fetal sheep closely parallels that of human, both as shown here, and as has previously been described (Charbord, et al 1996), with the vascular niche being the first component of the BM microenvironment to arise during fetal development. Formation of this niche commences between 25-36 gd in sheep and ~10-11 gw in the human fetus, and reaches relative maturity by 65 gd in sheep and ~18 gw in the human fetus.

The endosteal niche develops between 46-51 gd in sheep (~12-14 gw in human), maturing by 68 gd (~20 gw in human)

As described above, confocal imaging analyses of fetal sheep bone revealed that, by 36 gd, the vascular niche had already reached a sufficient state of maturity to permit cell trafficking within the nascent BM. At that gestational age, however, the medullary cavity only consisted of a mesenchymal matrix, completely devoid of calcified structures (Figures 1A-E, 2A, 2B, 2D, 2E). Even as late as 46 gd, when the vascular network appeared to be relatively well established, no calcified features were detectable, and the only N-Cadherin+ cells that could be identified within the mesenchymal structure appeared, morphologically, to be relatively immature preosteoblasts (Figure 3A). In contrast, at 51 gd, there was clear evidence of calcified bone in the form of trabeculae, consisting of well-organized, cuboidal-shaped osteoblasts expressing high levels of N-Cadherin+ (Figure 3F). The formation of calcified bone was confirmed by H&E-staining of 51-gd sheep bone, which clearly showed bone trabeculae formation adjacent to areas of endochondral ossification (Figures 3B-E). In 65-gd fetal sheep (Figure 3G), endochondral ossification appeared to be complete, as revealed by the presence of a well-established network of bone trabeculae and mature osteoblasts expressing high levels of N-Cadherin (Figure 3G), which had replaced the chondrocytic matrix seen 3 weeks earlier. The cell adhesion molecule NCadherin, which is abundantly expressed by osteoblasts, has a high affinity for the hydroxyapatite crystals of the cartilaginous matrix. As a result, N-Cadherin+ osteoblasts bind to the hydroxyapatite-rich matrix during endochondral ossification, producing the red-stained trabecular bone seen in Figure 3G. By 68 gd, the mineralized trabecular bone was well defined, and had become lined with N-Cadherin+ osteoblasts (Figure 3I). Moreover, at this same time point, CD34+ CD45+ HSCs were observed along the endosteal surface of the bone trabeculae (Figures 3H), attesting to the presence of CD34+ haematopoietic cells adjacent to N-Cadherin+ osteoblasts, documenting the presence of an established osteoblastic/endosteal niche.

Figure 3. Assessment of endosteal niche development and emergence/presence of HSC in the BM of fetal sheep.

Figure 3

Open in a new tab

Sheep fetal bone samples were immunostained for confocal microscopy as described in the methods section. The 2- or 3-colour channel composite photographs with or without transmitted display (TD) used DAPI (blue) for cell nuclei staining. (A) [40x] Composite image of 46-gd sheep fetal bone representing N-Cadherin+ preosteoblasts (red) within the bone marrow (BM) cavity which, at that gestational stage, merely consists of a chondrocytic matrix devoid of any mineralized structure. Inset: 10x magnification of the outlined section showing an isolated N-Cadherin+ preosteoblast. (B) [40x] 51-gd sheep bone representing trabecular bone formation during the process of endochondral ossification, as indicated by the black vertical bar (Magnification: 10x). (C) Representative image of endochondral ossification within 51-gd sheep femoral bone. Close up/zoom of “B” showing the product of endochondral ossification: calcified trabecular bone adjacent to the perichondrium or future bone cortex. It is indicated by the vertical bar (Magnification 20x). While area “1” represents the zone of chondrocyte proliferation with chondrocytes organized in isogenous groups separated by longitudinal and transversal septae, area “2” corresponds to the future bone cortex or perichondrium. The endosteum “3” lines the perichondrium and houses the germinal/stem cell zone where HSCs will home. Finally “4” represents the zone of cartilage mineralization characterized by elongated bone trabeculae within the BM cavity. (D) [40x] 51-gd sheep bone of the same animal depicting clearly delineated trabecular bone structures within the medullary cavity. The trabecular bone area is indicated by the horizontal black bar. (Magnification: 40x). (E) Close-up/zoom of the dark pink trabecular bone area shown in “D”. (F) [40x] Image of 51-gd sheep fetal bone revealing cuboidal NCadherin+ osteoblasts (red) organized into trabecular bone lined with CD34+ (green) cells. The presence of trabeculae at that fetal age suggests bone development in the sheep fetus in the 46- to 51-gd developmental window. (G) [40x] Three-colour plus TD composite confocal image of 65-gd sheep fetal bone featuring well-defined, elongated, finger-like N-Cadherin+ (red) calcified bone trabeculae. Inset: 2x magnification without TD of the outlined section representing mature N-Cadherin+ trabecular bone (red) lined with CD34+ endothelial cells (green). (H) [40x] Confocal image of 65-gd fetal sheep bone depicting CD34+(yellow) /CD45+(red) HSCs lining the endosteal area. (I) [40x] Composite confocal image of 68-gd sheep fetal bone featuring a section of trabecular bone lined with N-Cadherin+ cells (red). Trabecular bone is already better defined at 68 gd than at 65 gd, as revealed by the presence of wide trabecular bone lined with NCadherin+ osteoblasts. Inset: 2x magnification of the top right area showing N-Cadherin+ osteoblasts lining the trabecular bone structure.

In similarity to what we had observed in sheep, performing identical staining on 11 gw human fetal bone samples (equivalent to 36 gd of development in sheep), the vascular niche had also matured sufficiently to permit trafficking of haematopoietic cells within the forming BM, despite the absence of any calcified structures (Figure 4A). However, images of 14 gw human bone (~51 gd in sheep) revealed elongated trabecular bone structures within the medullary cavity that contained N-cadherin positive cells (Figure 4B, red stain), just as we had observed in sheep at this same stage of development. Figure 4C further confirmed the presence of calcified structures around that gestational period via the transmitted display of 15-gw human bone sections, which revealed newly formed bone matrix (colourless) adjacent to a large vascular cavity. By 18 gw in human (equivalent to 65 gd in sheep), endochondral ossification appeared to be complete (Figure 4D), and well-defined, mineralized trabecular bone was present at 20 gw, and was lined with osteopontin+ osteoblasts (Figure 4F). In addition, at this same time point, CD34+CD45+ HSCs were readily observed along the endosteal surface (Figures 4E, 4G), suggesting the osteoblastic/endosteal niche had reached a sufficient state of maturity to support/retain these haematopoietic cells.

Figure 4. Assessment of endosteal niche development and emergence/presence of HSC in the fetal human BM.

Figure 4

Open in a new tab

Human fetal bone samples at various gestational ages were stained for confocal microscopy as described in the methods section. (A) [40x] Confocal image of 11-gw human fetal bone (~40 gd in sheep) showing CD31+(red)/CD34+(green) endothelial cells lining large vascular cavities within the chondrocytic matrix, which is devoid of any calcified structure at that gestational age. (B) [40x] Confocal image of human fetal bone at 14 gw, showing the bone matrix with well-defined bone trabeculae (blue), suggesting that bone development probably takes place in the human fetus between 11-14 gw. In this confocal image, trabecular bone appears as elongated, deep blue structures, due to the non-specific binding of DAPI to the mineralized bone matrix. CD34+ cells are labelled green and N-cadherin+ cells are red. (C) [60x] Confocal image of 15-gw human fetal bone depicting a small cluster of HSCs lined up within the bone matrix in proximity to a large vascular cavity as shown by the transmitted display (TD). The image confirms the presence of a well-established vascular network at 15 gw, enabling HSC trafficking within the BM. Inset: Composite image acquired at 60x without TD showing an enlarged view of the square box highlighted in the main image, showing CD34+ (green) CD45+ (red) dual-positive HSCs along with single positive cells. (D) [40x] Confocal image of the human fetus at 18 gw, showing a clearly established N-Cadherin+ bone matrix (red). (E) [60x] Confocal image of human fetal bone acquired at 18 gw (~65 gd in sheep). HSCs are present within the BM as revealed by this composite of green (CD34), red (CD45) blue (nuclei) and TD channels. The bottom left inset is a magnification of the section highlighted in the main image, showing dual positive HSCs in close proximity to CD45+ haematopoietic cells. (F) [40x] Confocal image of human fetal bone acquired at 20gw, depicting how bone structures are already better developed than 2 weeks earlier, with Osteonpontin+ cells (green) prominently lining the well-established trabecular bone and trabecular bone pockets. Also at 20 gw in the human fetus (G), [60x] dual positive CD34+ (green)/CD45+ (red) HSCs are detected in higher numbers within the BM along with single positive CD34+ cells and single positive CD45+ haematopoietic cells. The bottom right inset shown in “G” is a magnification of the area delineated in the main image. The latter focuses on 6 dual positive HSCs.

Collectively, these results support the conclusion that the osteoblastic/endosteal niche commences development between 46-51 gd in fetal sheep (12-14 gw in human), but only reaches relative maturity by 65-68 gd in fetal sheep (18-20 gw in the human fetus).

Donor haematopoietic engraftment is enhanced at later gestational ages when both the vascular and endosteal niches are present

We next wished to investigate whether engraftment of allogeneic and/or xenogeneic cells was affected by gestational age. Fetal sheep were transplanted with sheep (allogeneic) or human (xenogeneic) HSCs at either 45 gd (only vascular niche present) or at 65 gd (both vascular and endosteal niche present) and analysed for donor cell engraftment/chimerism. As shown by the results presented in Figure 5, regardless of the source of HSCs used and the period of time post-transplant, no donor cell activity was ever detected in animals that were transplanted in utero at 45 gd (Figure 5A, 5B). In contrast, the fetuses that were transplanted with donor HSCs (sheep or human) at 65 gd exhibited significant donor cell activity that persisted after birth (Figure 5A, 5B). Donor HSC engraftment ranged from 3-14% after allogeneic (sheep) HSC transplantation (Figure 5A), and from 1-3% after xenogeneic (human) HSC transplantation (Figure 5B). These results were confirmed by confocal imaging analyses on the bones of recipients from two separate transplantation studies. Sheep fetuses transplanted at 65 gd with eGFP-tagged sheep CD34+ cells were analysed by immunofluorescence and confocal imaging for the presence of eGFP+ donor cells in the BM at 63 days post-transplant, as described in the Materials and Methods section. Figure 5C shows a representative image of these analyses, and demonstrates the robust engraftment of the eGFP-tagged CD34+ cells observed following IUHSCT at 65 gd.

Figure 5. Comparative assessment of HSC engraftment in the 45- and 65-gd sheep BM after in-utero transplantation of sheep or human cells.

Figure 5

Open in a new tab

Allogeneic and xenogeneic cell transplantation studies used random-bred, time-dated, pregnant Dorset-Merino sheep (term ~145 days). (A) For allogeneic haematopoietic stem cell (HSC) transplantation studies, each type BB sheep fetus (typed at the β-globin locus of the haemoglobin gene) received an intraperitoneal injection of 5 × 105 T cell-depleted mononuclear (MN) cells obtained from the bone marrow (BM) of 1-year-old type AA donors of the opposite sex, at either 45 or 65 gd (n=10/group). BM cells obtained from recipients at 10 (n=4), 60 (n=3) and 130 (n=3) days post-transplantation were enumerated by isoelectric focusing haemoglobin analyses of the recipients. (B) For xenogeneic HSC transplantations, each sheep fetus received 4 × 105 CD34+ human BM cells at either 45 or 65 gd (n=10/group). BM cells were harvested from the sheep recipients at 10 (n=4), 60 (n=3), and 130 (n=3) days post-transplantation and assessed by flow cytometry for the presence of human CD45. (C) [40x] Confocal image of a 128-gd sheep fetus that was injected at 65 gd with 1.4 × 106 eGFP-tagged allogeneic sheep BM CD34+ cells, as described in the Materials and Methods section. The transplanted eGFP+ CD34+ cells were visualized by confocal microscopy as described in the methods section. They were detected in proximity to both niches of the BM: the endosteal region (“trabecular bone”) and the vascularized area (“sinusoid vessels”) toward the centre of the medullary cavity. HSCs are believed to transit between the endosteal and vascular microenvironments where they are kept quiescent (endosteum) or induced to expand and differentiate (vasculature). (D-G) Confocal images of a 6-month-old lamb that was injected at 65 gd with 4.5 × 104 human cord blood CD34+ cells. Bone samples were collected 9 months after injection and analysed for the presence of the transplanted cells as described in the methods section. The 2- or 3-colour channel composite photomicrographs shown with or without transmitted display (TD) used DAPI (blue) for cell nuclei staining. (D, E) [60x] Confocal images of the green (CD34), red (CD45), blue (DAPI) and TD channels demonstrating the successful engraftment of the transplanted human CD34+ (green)/ CD45+ (red) HSC along the endosteum, within approximately 10 μm of the endosteal surface. Insets: 2- and 3- channel images showing the individual and merged staining of the cells highlighted in the white box (2x magnification). (F) [40x] Composite confocal image of the green (CD34), red (CD45), blue (nuclei) and TD channels showing HSCs in close proximity to a large sinusoid vessel, confirming engraftment within the vascular niche. Inset: 3- colour image without TD acquired at 60x magnification representing a close-up of the cells outlined in the main image. (G) [60x] Composite confocal image of the green (CD34), red (CD45), blue (nuclei) and TD channels depicting several human HSCs engrafted within the medullary cavity. Inset is a close-up/zoom of the section outlined in the main image. It shows the dual positive CD34+ CD45+ human cells along with 3 endogenous CD45+ haematopoietic cells.

Sheep fetuses, transplanted at 65 gd with human CD34+ cells, were also evaluated for donor cell engraftment at 9 months post-transplant. Figure 5D is a representative image showing evidence of successful long-term (9 months post-transplant) engraftment of the human CD34+ cells following in utero transplantation into sheep fetuses at 65 gd. Importantly, these data demonstrate that the cells engrafted in the 3 expected regions of the BM, namely the osteoblastic/endosteal niche (Figure 5E), the vascular niche (Figure 5F), and the medullary cavity (Figure 5G) when transplanted at 65 gd, a time during gestation when our prior analyses demonstrated both BM microenvironments are fully developed. In contrast, no significant donor cell engraftment was documented following IUHSCT with allogeneic or xenogeneic HSC at the earlier time point of 45 gd when maturity of the HSC niches were lacking.

DISCUSSION

Using the fetal sheep model, we performed studies to delineate the temporal window during which the various components of the BM microenvironment develop, with the goal of establishing a correlation between the fetal-sheep and fetal-human BM niche ontogeny, so that studies addressing the role of niche development at the time of IUHSCT could be accurately addressed using this animal model.

Our results show that, in both ovine and human, the vascular and osteoblastic/endosteal niches reach maturity at different time points during gestation, with the vascular niche developing between 25-36 gd in sheep (equivalent to ~10-11 gw in human), and the osteoblastic/endosteal niche starting development around 46-50 gd in sheep, reaching maturity by 65-68 gd (~18-20 gw in human). Our results showing that, in sheep, vasculogenesis occurs before osteogenesis in the nascent BM are in agreement with earlier studies on the ontogeny of the human BM microenvironment (Bollerot, et al 2005, Dzierzak and Enver 2008, Tavian, et al 1999), showing that the vasculature starts developing in the human embryo between 9.0-10.5 gw, while osteogenesis does not occur until about 10.5-15 gw. Our analysis of BM microenvironment development in sheep revealed that vasculogenesis takes place within the sheep BM between 25-36 gd (~10-11 gw in the human fetus), at a time when neither osteoblasts nor mineralized structures are present in the BM. Indeed, osteogenesis was not observed to commence until ~46 gd in the sheep (~14-15 gw in the human fetus), thus establishing that critical events in BM ontogeny occur at parallel time points in both species (Charbord, et al 1996). These findings thus further validate the fetal sheep as a model for not only haematopoiesis, but also BM niche development. In addition, the early establishment of a network of vessels within the BM has been shown to be necessary for cell trafficking to the medullary cavity, as it permits the migration of the cells required for BM haematopoiesis and for tissue formation and differentiation, including osteogenesis. It has previously been reported that the vascular networks are the first differentiated tissues to develop during early vertebrate ontogeny (Tavian and Peault 2005), in order to provide the erythrocytes that are essential for the oxygenation of the developing tissues, including bone. In addition, previous studies on endochondral ossification have further shown that blood vessels first have to invade the cartilaginous matrix in order to initiate the processes of ossification and trabecular bone formation (Chen and Weiss 1975, Kronenberg 2003).

It is important to note that the first concomitant appearance of osteoblasts and sinusoid vessels we observed during early osteogenesis does not imply that the BM niches have reached maturity and are receptive to HSC homing after IUHSCT. It has been postulated that BM niche maturity is not established until the haematopoietic marrow invades the medullary space (Kronenberg 2003), where they then interact with the stromal cells of both niches and set the stage for permanent haematopoiesis. However, the demonstration of CD34+ALDH+ HSC within the trabecular matrix starting at ~46 gd in sheep BM probably does not indicate complete maturity of the BM microenvironment in this model at this time point. BM niche maturity is probably characterized by the simultaneous presence of the 3 independent factors including vascular niche maturity, endosteal niche maturity and the presence of HSC populations within both niches. This was confirmed by results obtained from transplanting allogeneic (sheep) or xenogeneic (human) HSC at either 45 gd, a specific time in gestation when our data indicated only the vascular niche to be present, and at 65 gd, when both niches were shown to be receptive to HSC homing. We found that, at 45 gd, no HSC engraftment occurred after transplantation. However, when transplantation was performed at later time points, consistent engraftment of either allogeneic or xenogeneic donor cells was found. By performing cell transplantation at a gestational age when only one niche, the vascular niche, was present, we were able to show that successful donor cell engraftment after IUHSCT was only achieved when both niches were concomitantly established. Nevertheless, other factors may also contribute to the lack of engraftment following IUHSCT at 45 gd, such as the inability of adult donor haematopoietic cells to efficiently migrate from the peritoneal cavity through the lymphatics into circulation at earlier time points of gestation and/or entrapment in other organs possessing haematopoietic function at the time of transplant. Nevertheless, as we analysed donor cell engraftment at a time when haematopoiesis had already been established within the BM, donor cells transplanted at earlier time points in gestation would have migrated, along with the endogenous haematopoietic cells, and contributed to the establishment of haematopoiesis in the newly formed marrow.

In conclusion, this study demonstrates that that critical events of fetal BM niche development occur at parallel time points in both sheep and human, confirming the reliability of the sheep as an experimental model for IUHSCT. The results also suggest that the ideal time to perform IUHSCT would probably be at or later than 55-65 gd, to ensure that both the vascular and osteoblastic/endosteal niche are fully functional.

Acknowledgements

This work was supported by grants from the National Institutes of Health to E.D.Z (HL52955) and G.A-P. (R01HL097623).

Footnotes

Authorship Contribution: C.J., E.C., and S.M. performed experiments; C.J., E.C., and C.D.P. made figures; C.J and A.D.G. analysed results and wrote the paper; E.D.Z., C.D.P., and G.A.P. designed research, evaluated results, and wrote paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

References

  1. Bollerot K, Pouget C, Jaffredo T. The embryonic origins of hematopoietic stem cells: a tale of hemangioblast and hemogenic endothelium. Apmis. 2005;113:790–803. doi: 10.1111/j.1600-0463.2005.apm_317.x. [DOI] [PubMed] [Google Scholar]
  2. Charbord P, Tavian M, Humeau L, Peault B. Early ontogeny of the human marrow from long bones: an immunohistochemical study of hematopoiesis and its microenvironment. Blood. 1996;87:4109–4119. [PubMed] [Google Scholar]
  3. Chen LT, Weiss L. The development of vertebral bone marrow of human fetuses. Blood. 1975;46:389–408. [PubMed] [Google Scholar]
  4. Diukman R, Golbus MS. In utero stem cell therapy. J Reprod Med. 1992;37:515–520. [PubMed] [Google Scholar]
  5. Dzierzak E, Enver T. Stem cell researchers find their niche. Development. 2008;135:1569–1573. doi: 10.1242/dev.019943. [DOI] [PubMed] [Google Scholar]
  6. Flake AW. In utero stem cell transplantation. Best Pract Res Clin Obstet Gynaecol. 2004;18:941–958. doi: 10.1016/j.bpobgyn.2004.06.006. [DOI] [PubMed] [Google Scholar]
  7. Flake AW, Zanjani ED. In utero hematopoietic stem cell transplantation: ontogenic opportunities and biologic barriers. Blood. 1999;94:2179–2191. [PubMed] [Google Scholar]
  8. Flake AW, Harrison MR, Adzick NS, Zanjani ED. Transplantation of fetal hematopoietic stem cells in utero: the creation of hematopoietic chimeras. Science. 1986;233:776–778. doi: 10.1126/science.2874611. [DOI] [PubMed] [Google Scholar]
  9. Flake AW, Roncarolo MG, Puck JM, Almeida-Porada G, Evans MI, Johnson MP, Abella EM, Harrison DD, Zanjani ED. Treatment of X-linked severe combined immunodeficiency by in utero transplantation of paternal bone marrow. N Engl J Med. 1996;335:1806–1810. doi: 10.1056/NEJM199612123352404. [DOI] [PubMed] [Google Scholar]
  10. Fujiki Y, Fukawa K, Kameyama K, Kudo O, Onodera M, Nakamura Y, Yagami K, Shiina Y, Hamada H, Shibuya A, Nakauchi H. Successful multilineage engraftment of human cord blood cells in pigs after in utero transplantation. Transplantation. 2003;75:916–922. doi: 10.1097/01.TP.0000057243.12110.7C. [DOI] [PubMed] [Google Scholar]
  11. Huang S, Zeng F, Gong Z, Lu J, Huang W, Hu W, Yan J, Fang Y. [A study of the engraftment, expansion and differentiation of human hematopoietic stem cells in goats] Zhonghua Yi Xue Za Zhi. 2002;82:86–89. [PubMed] [Google Scholar]
  12. Kronenberg HM. Developmental regulation of the growth plate. Nature. 2003;423:332–336. doi: 10.1038/nature01657. [DOI] [PubMed] [Google Scholar]
  13. Maddox JF, Mackay CR, Brandon MR. Ontogeny of ovine lymphocytes. I. An immunohistological study on the development of T lymphocytes in the sheep embryo and fetal thymus. Immunology. 1987a;62:97–105. [PMC free article] [PubMed] [Google Scholar]
  14. Maddox JF, Mackay CR, Brandon MR. Ontogeny of ovine lymphocytes. II. An immunohistological study on the development of T lymphocytes in the sheep fetal spleen. Immunology. 1987b;62:107–112. [PMC free article] [PubMed] [Google Scholar]
  15. Maddox JF, Mackay CR, Brandon MR. Ontogeny of ovine lymphocytes. III. An immunohistological study on the development of T lymphocytes in sheep fetal lymph nodes. Immunology. 1987c;62:113–118. [PMC free article] [PubMed] [Google Scholar]
  16. Merianos D, Heaton T, Flake AW. In utero hematopoietic stem cell transplantation: progress toward clinical application. Biol Blood Marrow Transplant. 2008;14:729–740. doi: 10.1016/j.bbmt.2008.02.012. [DOI] [PubMed] [Google Scholar]
  17. Merianos DJ, Tiblad E, Santore MT, Todorow CA, Laje P, Endo M, Zoltick PW, Flake AW. Maternal alloantibodies induce a postnatal immune response that limits engraftment following in utero hematopoietic cell transplantation in mice. J Clin Invest. 2009;119:2590–2600. doi: 10.1172/JCI38979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Miyasaka M, Trnka Z. Lymphocyte migration and differentiation in a large-animal model: the sheep. Immunol Rev. 1986;91:87–114. doi: 10.1111/j.1600-065x.1986.tb01485.x. [DOI] [PubMed] [Google Scholar]
  19. Muench MO. In utero transplantation: baby steps towards an effective therapy. Bone Marrow Transplant. 2005;35:537–547. doi: 10.1038/sj.bmt.1704811. [DOI] [PubMed] [Google Scholar]
  20. Muench MO, Barcena A. Stem cell transplantation in the fetus. Cancer Control. 2004;11:105–118. doi: 10.1177/107327480401100217. [DOI] [PubMed] [Google Scholar]
  21. Nijagal A, Wegorzewska M, Jarvis E, Le T, Tang Q, MacKenzie TC. Maternal T cells limit engraftment after in utero hematopoietic cell transplantation in mice. J Clin Invest. 2011;121:582–592. doi: 10.1172/JCI44907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Osburn BI. The ontogeny of the ruminant immune system and its significance in the understanding of maternal-fetal-neonatal relationships. Adv Exp Med Biol. 1981;137:91–103. [PubMed] [Google Scholar]
  23. Peranteau WH, Endo M, Adibe OO, Flake AW. Evidence for an immune barrier after in utero hematopoietic-cell transplantation. Blood. 2007;109:1331–1333. doi: 10.1182/blood-2006-04-018606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pixley JS, Tavassoli M, Zanjani ED, Shaft DM, Futamachi KJ, Sauter T, Tavassoli A, MacKintosh FR. Transplantation in utero of fetal human hematopoietic stem cells into mice results in hematopoietic chimerism. Pathobiology. 1994;62:238–244. doi: 10.1159/000163916. [DOI] [PubMed] [Google Scholar]
  25. Porada CD, Harrison-Findik DD, Sanada C, Valiente V, Thain D, Simmons PJ, Almeida-Porada G, Zanjani ED. Development and characterization of a novel CD34 monoclonal antibody that identifies sheep hematopoietic stem/progenitor cells. Exp Hematol. 2008;36:1739–1749. doi: 10.1016/j.exphem.2008.09.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Roybal JL, Santore MT, Flake AW. Stem cell and genetic therapies for the fetus. Semin Fetal Neonatal Med. 2010;15:46–51. doi: 10.1016/j.siny.2009.05.005. [DOI] [PubMed] [Google Scholar]
  27. Sawyer M, Moe J, Osburn BI. Ontogeny of immunity and leukocytes in the ovine fetus and elevation of immunoglobulins related to congenital infection. Am J Vet Res. 1978;39:643–648. [PubMed] [Google Scholar]
  28. Schoeberlein A, Schatt S, Troeger C, Surbek D, Holzgreve W, Hahn S. Engraftment kinetics of human cord blood and murine fetal liver stem cells following in utero transplantation into immunodeficient mice. Stem Cells Dev. 2004;13:677–684. doi: 10.1089/scd.2004.13.677. [DOI] [PubMed] [Google Scholar]
  29. Shields LE, Bryant EM, Easterling TR, Andrews RG. Fetal liver cell transplantation for the creation of lymphohematopoietic chimerism in fetal baboons. Am J Obstet Gynecol. 1995;173:1157–1160. doi: 10.1016/0002-9378(95)91344-0. [DOI] [PubMed] [Google Scholar]
  30. Slavin S, Naparstek E, Ziegler M, Lewin A. Clinical application of intrauterine bone marrow transplantation for treatment of genetic diseases--feasibility studies. Bone Marrow Transplant. 1992;9(Suppl 1):189–190. [PubMed] [Google Scholar]
  31. Sun Y, Xiao D, Pan XH, Zhang RS, Cui GH, Chen XG. Generation of human/rat xenograft animal model for the study of human donor stem cell behaviors in vivo. World J Gastroenterol. 2007a;13:2707–2716. doi: 10.3748/wjg.v13.i19.2707. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sun Y, Xiao D, Zhang RS, Cui GH, Wang XH, Chen XG. Formation of human hepatocyte-like cells with different cellular phenotypes by human umbilical cord blood-derived cells in the human-rat chimeras. Biochem Biophys Res Commun. 2007b;357:1160–1165. doi: 10.1016/j.bbrc.2007.04.087. [DOI] [PubMed] [Google Scholar]
  33. Surbek DV, Holzgreve W, Nicolaides KH. Haematopoietic stem cell transplantation and gene therapy in the fetus: ready for clinical use? Hum Reprod Update. 2001;7:85–91. doi: 10.1093/humupd/7.1.085. [DOI] [PubMed] [Google Scholar]
  34. Tarantal AF, Goldstein O, Barley F, Cowan MJ. Transplantation of human peripheral blood stem cells into fetal rhesus monkeys (Macaca mulatta) Transplantation. 2000;69:1818–1823. doi: 10.1097/00007890-200005150-00015. [DOI] [PubMed] [Google Scholar]
  35. Tavian M, Peault B. The changing cellular environments of hematopoiesis in human development in utero. Exp Hematol. 2005;33:1062–1069. doi: 10.1016/j.exphem.2005.06.025. [DOI] [PubMed] [Google Scholar]
  36. Tavian M, Cortes F, Charbord P, Labastie MC, Peault B. Emergence of the haematopoietic system in the human embryo and foetus. Haematologica. 1999;84:1–3. Suppl EHA-4. [PubMed] [Google Scholar]
  37. Westgren M, Ringden O, Eik-Nes S, Ek S, Anvret M, Brubakk AM, Bui TH, Giambona A, Kiserud T, Kjaeldgaard A, Maggio A, Markling L, Seiger A, Orlandi F. Lack of evidence of permanent engraftment after in utero fetal stem cell transplantation in congenital hemoglobinopathies. Transplantation. 1996;61:1176–1179. doi: 10.1097/00007890-199604270-00010. [DOI] [PubMed] [Google Scholar]
  38. Xiao YP, Chen MJ, Sheng M, Gong ZJ, Wang S, Huang SZ. [Analysis of human cells in transplanted goats using fluorescence in situ hybridization] Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2003;25:129–133. [PubMed] [Google Scholar]
  39. Zanjani ED, McGlave PB, Bhakthavathsalan A, Stamatoyannopoulos G. Sheep fetal haematopoietic cells produce adult haemoglobin when transplanted in the adult animal. Nature. 1979;280:495–496. doi: 10.1038/280495a0. [DOI] [PubMed] [Google Scholar]
  40. Zanjani ED, Ascensao JL, Flake AW, Harrison MR, Tavassoli M. The fetus as an optimal donor and recipient of hemopoietic stem cells. Bone Marrow Transplant. 1992a;10(Suppl 1):107–114. [PubMed] [Google Scholar]
  41. Zanjani ED, Pallavicini MG, Ascensao JL, Flake AW, Langlois RG, Reitsma M, MacKintosh FR, Stutes D, Harrison MR, Tavassoli M. Engraftment and long-term expression of human fetal hemopoietic stem cells in sheep following transplantation in utero. J Clin Invest. 1992b;89:1178–1188. doi: 10.1172/JCI115701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zanjani ED, Flake AW, Rice H, Hedrick M, Tavassoli M. Long-term repopulating ability of xenogeneic transplanted human fetal liver hematopoietic stem cells in sheep. J Clin Invest. 1994;93:1051–1055. doi: 10.1172/JCI117054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zanjani ED, Almeida-Porada G, Flake AW. The human/sheep xenograft model: a large animal model of human hematopoiesis. Int J Hematol. 1996;63:179–192. doi: 10.1016/0925-5710(96)00445-8. [DOI] [PubMed] [Google Scholar]

RESOURCES