research paper
In vivo haematopoietic potential of human neural stem cells
G. Almeida-Porada,1 K. Crapnell,1 C. Porada,1 B. Benoit,2 H. Nakauchi,3 P. Quesenberry2 and E. D. Zanjani1 1
Department of Animal Biotechnology, School of
Veterinary Medicine and Department of Medicine, University of Nevada, Reno and VA Medical Center, Reno, NV, 2Department of Medicine, Roger Williams Medical Center, Providence, RI, USA, and 3Department of Immunology, University of Tsukuba, Ibaraki, Japan
Received 28 March 2005; accepted for publication 9 May 2005 Correspondence: Grac¸a Almeida-Porada, MD, PhD, Department of Animal Biotechnology,
Summary The fetal sheep model was used to compare the in vivo haematopoietic potential of human neural stem cells (NSC) versus bone marrow (BM)derived haematopoietic stem cells (HSC). To this end, sheep were transplanted with either 8 · 105 NSC (n ¼ 11) or HSC, CD34+Lin) (n ¼ 5), and subsequently analysed for haematopoietic chimaerism. While HSC-transplanted sheep displayed robust donor-derived haematopoiesis starting at less than 2 months post-transplant, NSC recipients exhibited haematopoietic engraftment at much later time points. Nevertheless, chimaerism persisted in both groups throughout the course of this study. Transplantation of secondary recipients with human CD45+/HLA-DR+ cells from the BM of NSC primary recipients at 14 and 16 months post-transplant demonstrated that long-term engrafting HSC were present in these animals. At 6 months post-transplant, both NSC- and HSC-transplanted sheep were mobilised with granulocyte colony-stimulating factor. In contrast to HSCtransplanted animals, levels of human blood cells in peripheral blood of NSCtransplanted sheep remained low throughout mobilisation. Our results show that, although human NSC were able to give rise to multilineage haematopoiesis in our model, the levels, timing of blood cell production and the ability to respond to cytokine mobilisation were different, suggesting that human NSCs latent haematopoietic potential is inherently different from that of true HSC.
School of Veterinary Medicine, University of Nevada, Reno, Mail Stop 202, Reno, NV 89557 0104, USA. E-mail:
[email protected]
Keywords: neural stem cells, haematopoietic stem cells, mobilisation, transplantation, differentiation.
Recent studies have shown that adult tissues contain stem cells with fairly pronounced multipotentiality that persist throughout life and participate in normal replacement and repair within their respective organs. For instance, the adult mammalian brain, which was once thought to be incapable of regeneration, was found not only to contain self-renewing multipotent stem cells, but also to produce new neurons throughout adult life (Gage et al, 1995; Eriksson et al, 1998; Clarke et al, 2000). Furthermore, several research groups provided evidence that stem cells isolated from the murine and human brain also had the ability to give rise to cells of other seemingly unrelated organs, such as blood, kidney myocytes and cardiomyocytes (Bjornson et al, 1999; Clarke et al, 2000; Galli et al, 2000; Tsai & McKay, 2000; Shih et al, 2001; Tropepe et al, 2001). But, while Bjornson et al (1999) reported that neural stem cells (NSCs) isolated from the embryonic and adult murine forebrain engrafted into the haematopoietic system of irradiated hosts to produce a wide
doi:10.1111/j.1365-2141.2005.05588.x
range of blood cells, and Shih et al (2001), demonstrated that human NSC were able to reconstitute long-term haematopoiesis following transplantation into severe-combined immunodeficient (SCID)-hu mice possessing a human bone marrow (HBM) microenvironment, other studies using mice suggested that blood production by embryonic neural cells was rare or because of contamination with other cells (Morshead et al, 2002; Magrassi et al, 2003; Yusta-Boyo et al, 2004). These studies raised questions as to whether the haematopoietic competence of NSC is a reliable property of these cells or rather, that the discrepancies found are a result of different methodological approaches. To date, the demonstration of in vivo differentiation of adult stem cells into a different unrelated phenotype has been shown primarily in animal studies in which a need for a specific cell/tissue type was created in the host either by an external stress such as radiation/chemicalinduced injury or an experimentally created shortage of a specific-cell type (Gussoni et al, 1999; Theise et al, 2000;
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
Neural Stem Cells Give Rise to Blood Krause et al, 2001; Lagasse et al, 2001; Wang et al, 2003; Almeida-Porada et al, 2004). While the mechanisms controlling changes in cell fate are at present unclear, lessons drawn from published studies suggest that the outcome is dependent on the animal model used, the source of cells, their phenotype and methods of culture (Wang et al, 2003; Almeida-Porada et al, 2004). Thus, if NSC indeed have the potential to differentiate into haematopoietic stem cells (HSC), the question remains whether this NSC population is able to fulfill the requisites imposed upon HSC, i.e. whether blood cells generated from NSC are able to function physiologically and respond to stimuli as do blood cells generated by BM-derived HSC. To this end, we used the fetal sheep model to perform a side-by-side comparison of the haematopoietic potential of human HSC versus human NSC. We have successfully used the fetal sheep transplantation model to identify and characterise novel human HSC phenotypes from various fetal and adult sources (Civin et al, 1996; Kawashima et al, 1996; Yin et al, 1997; Ziegler et al, 1999). Serial transplantation of these HSC populations into primary, secondary and, if needed, tertiary recipients, enabled us to readily distinguish between short- and long-term repopulating cells (Zanjani et al, 1995, 1996). Furthermore, human HSC engrafted within our model respond to human cytokine treatment in the same fashion as human recipients do, enabling us to use this model to study the physiology of mobilised HSC populations (Porada et al, 1999; Verfaillie et al, 2000). In the present studies, we used a serial transplantation approach to compare the haematopoietic engraftment potential and kinetics of human HSC versus human NSC. We showed that, although human fetal brainderived NSC harbour the ability to produce multiple haematopoietic lineages in primary and secondary sheep recipients, the difference in levels and timing of production of blood cells and the disparate ability to respond to cytokine mobilisation suggest the latent haematopoietic potential of human NSC is inherently different from that of human HSC.
epidermal growth factor or fibroblast growth factor 2 (Sigma Chemical, St Louis, MO, USA). Insulin (25 lg/ml), 100 lg/ml transferrin, 20 nmol/l progesterone, 60 lmol/l putrescine and 30 lmol/l selenium chloride were used in place of serum (Sigma Chemical). Passaging was carried out by gentle mechanical dissociation using a fire-polished Pasteur pipette, after which the mixture of intact spheres and single cells was re-seeded into fresh medium as above.
Phenotypic analysis In order to assess whether cultured neurosphere cells contained haematopoietic markers, after dissociation and prior to transplantation, neurospheres were analysed by flow cytometry, (50 000 events per sample were analysed) using monoclonal antibodies against CD3, CD7, CD14, CD19, CD33, CD34, CD45, CD90, CD117 and HLA-DR [Becton Dickinson Immunocytometry Systems (BDIS), San Jose, CA, USA] and Gly-A (Immunotech, Miami, FL, USA). As a control for their neural potential, immunocytochemistry for nestin (Chemicon, Temecula, CA, USA), a marker of precursor and putative neural stem cells, was also performed on the neurospheres prior to transplant.
Bone marrow HSC preparation Heparinised HBM was obtained from healthy donors after informed consent. Low density BM mononuclear cells (BMNC) were separated by Ficoll density gradient (1Æ077 g/ ml) (Sigma) and washed twice in Iscove’s modified Dulbecco’s medium, (Gibco Laboratories, Grand Island, NY, USA). BMNC were then enriched for the CD34+ fraction using magnetic cell sorting (Miltenyi Biotec, Inc., Auburn, CA, USA), followed by lineage depletion for CD3, CD4, CD8, CD14, CD15, CD19 and CD56 in a FACSVantage (BDIS).
Creation of human sheep chimaeras Materials and methods Isolation and culture of neurospheres Human fetal neurospheres were kindly provided by Drs Peter Quesenberry and H. Nakauchi from Roger Williams Medical Center, Providence, RI and University of Tsukuba, Ibaraki, Japan respectively (Hulspas & Quesenberry, 2000; Benoit et al, 2001; Engstrom et al, 2002). All the studies were performed with passage two or three human neurospheres. Briefly, to obtain human neurospheres, human fetal brain was harvested and mechanically dissociated using fired-polished Pasteur pipettes in 1:1 mixture of Dulbecco’s modified Eagle’s medium (DMEM) (Gibco BRL, Gaithersburg, MD, USA) and F12 nutrient (Gibco). Cells were plated onto poly-l-ornithine coated dishes and cultured for 10–14 d in DMEM/F12 (1:1) with glucose (0Æ6%), 2 mmol/l glutamine, 3 mmol/l sodium bicarbonate, 5 mmol/l HEPES buffer and 20 ng/ml of either
Neural stem cell studies were performed in 15 fetal sheep (11 primary recipients and four secondary recipients) following the transplantation procedure that has been described in detail previously (Flake et al, 1986). In short, cultured NSC were injected intraperitoneally in a 0Æ5-ml volume into 55–60-d-old fetal sheep (term, 145 d) at a dose of 8 · 105 NSC/fetus. The transplanted sheep were then analysed for donor (human) cell engraftment at intervals between 2 and 16 months posttransplant. At 14 and 16 months post-transplant, human cells were sorted from the BM of primary sheep recipients using a FACSVantage (BDIS) and were used to perform secondary transplants. At 14 months, 22 · 104 human cells were isolated and transplanted into two 55–60-d-old fetal sheep, and at 16 months 3Æ3 · 104 human cells were obtained and transplanted into two additional fetal sheep of the same gestational age. Secondary recipients were analysed for human haematopoietic activity at 2 months post-transplant. Age-matched
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
277
G. Almeida-Porada et al sheep that had been transplanted with 8 · 105 adult human BM CD34+Lin) cells as previously described (Zanjani et al, 1995, 1996; Kawashima et al, 1996) were also analysed at the same time points, to allow comparison for haematopoietic engraftment and cytokine mobilisation.
Mobilisation of human cells with granulocyte colonystimulating factor
Assessment of human donor cell engraftment The presence of donor cells in haematopoietic tissues of the recipients (blood, marrow, liver, spleen and thymus) was determined at intervals post-transplantation using flow cytometric analysis and karyotyping of haematopoietic progenitors grown in methylcellulose assays. Flow cytometric analysis of the cell populations was performed on a FACScan (BDIS). Monoclonal antibodies to various cluster designations directly conjugated with fluorescein isothiocyanate or phycoerythrin were used according to the manufacturer’s recommendation. These included: CD1a, CD3, CD7, CD20, CD33, CD34, CD45, CD86, HLA-DR (BDIS) and glycophorin A (Coulter; Immunotech). Karyotyping of haematopoietic colonies was performed as previously described (Zanjani et al, 1992). In short, single colonies were identified by type, removed from plates individually, placed in buffered normal saline colcemid solution, mixed, and incubated at 37C for 30 min. The mixture was then centrifuged at 200 g for 8 min, the supernatant was removed, and the cell pellet was slowly resuspended in a solution of prewarmed (37C) potassium chloride. The mixture was incubated at 37C for 10 min, spun at 200 g for 8 min and the cell pellet resuspended. A fixative solution of 33% glacial acetic acid in methanol was slowly added to the cell suspension and the cells processed for karyotyping according to the standard methods.
Detection of human cells by polymerase chain reaction To evaluate various organs for the presence of human cells, single cell suspensions were prepared by passing each of the tissues through a glass homogeniser. Total genomic DNA was then extracted from 107 cells from each tissue using a commercially available kit according to the manufacturer’s recommendations (MasterPureTM Genomic DNA Purification Kit; Epicentre Technologies, Madison, WI, USA). A segment from the human HLA-DQa gene was then amplified by polymerase chain reaction (PCR) using the primers A (5¢gtgctgcaggtgtaaacttgtaccag-3¢) and B (5¢-cacggatccggtagcagcgg tagaagttg-3¢) by performing 2 min of denaturation at 98C and 5 min of annealing at 58C followed by 40 cycles with the following parameters: 95C for 45 s, 58C for 1 min, 69C for 1 min). After a 7-min extension at 69C, one tenth of each of the reaction products was run on a 1% agarose gel and subsequently transferred to Gene Screen Plus membrane (NEN, Boston, MA, USA) under denaturing conditions. An oligonucleotide probe (5¢-tggacctggagaggaaggagact-3¢) was then end-labelled with 32P-ATP and hybridised to the resultant blot for 2 h at 42C in Rapid-Hyb hybridisation solution 278
(Amersham, Buckinghamshire, UK). After four washings at 42C under conditions of increasing stringency, the blots were exposed to Bio-Max MS film (Kodak, Rochester, NY, USA) for 2–12 h at )70C with one Bio-Max intensifying screen.
At 6 months post-transplant, primary NSC and HSC recipients were administered subcutaneous injections of human recombinant granulocyte colony-stimulating factor (G-CSF) (Neupogen; Amgen, Inc., Thousand Oaks, CA, USA) once a day in the morning for three consecutive days. Animals were individually weighed prior to injection, and a dose of 5 lg/ kg was used. This dose and schedule was chosen as we had previously demonstrated that it efficiently mobilised human cells present in human sheep chimaeras, and that mobilising for more than 3 d did not result in further mobilisation of human cells (Porada et al, 1999)
Results In vitro haematopoietic potential of cultured neurospheres Neurospheres were harvested after two to three passages and confirmed, prior to transplant, to be positive for the expression of the intermediate filament protein, nestin, to verify that culture had not altered their NSC phenotype. In order to ensure that the neurospheres did not contain detectable haematopoietic contaminants, flow cytometric analysis was performed prior to transplantation for a panel of haematopoietic markers including stem cell- and lineage differentiation-specific markers. Flow cytometric analysis demonstrated NSC to be positive for CD90, but negative for all of the other markers tested including CD3, CD7, CD14, CD19, CD33, CD34, CD38, CD45, CD117, Gly-A and HLA-DR, demonstrating that these NSC were not contaminated with detectable levels of cells committed to the haematopoietic lineage.
Evaluation of the differentiative potential of the neurospheres in vivo To evaluate the in vivo differentiative potential of the cultured human neurospheres, 11 55–60-d-old primary fetal sheep were transplanted with NSC at a cell dose of 8 · 105 NSC/fetus. Two of the transplanted lambs were lost to sepsis and as a result, no data was obtained from them. Recipients were evaluated for the presence/engraftment of human cells beginning at 2 months post-transplant and at intervals thereafter until 16 months posttransplant. At 2 months post-transplant, human cells were detected only by PCR in the spinal cord of two, and in haematopoietic organs of four of seven transplanted sheep (Fig 1A and B). At this time point and in these same two animals we were unable to detect human cells in other organs, such as
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
9
7 8
2 3
1
B
6
7
8
9
17
5
14 15 16
4
13
3
12
2
11
1
10
Lane#
6
A
4 5
Neural Stem Cells Give Rise to Blood
Fig 1. (A) Detection of human cells in central nervous system (CNS) of neural stem cell (NSC) primary sheep recipients. At 2 months posttransplant DNA was extracted from several organs of sheep transplanted with NSC (n ¼ 7). Polymerase chain reaction (PCR) analysis using primers specific for human HLA-DQa demonstrated the presence of human cells in spinal cord of two primary sheep recipients (lanes 3 and 6). Lane 1, non-transplanted sheep; lane 2, brain sheep B; lane 3, spinal cord sheep B; lane 4, muscle sheep B; lane 5, brain sheep C; lane 6, spinal cord sheep C B; lane 7, lung sheep C; lane 8, muscle sheep C; lane 9, human positive control. (B) Detection of human cells in haematopoietic organs of NSC primary sheep recipients. At 2 months post-transplant human cells were detected in haematopoietic organs of four transplanted NSC animals by human HLA-DQa-specific PCR. Lanes: 1, reagent control; 2, non-transplanted sheep; 3, peripheral blood (PB) sheep A; 4, PB sheep F; 5, BM sheep F; 6, PB sheep D; 7, BM sheep D; 8, liver sheep D; 9, spleen sheep D; 10, thymus sheep D; 11, PB sheep E; 12, BM sheep E; 13, liver sheep E; 14, spleen sheep E; 15, thymus sheep E; 16, blank; 17, positive control.
% Human CD45+ cells
3 2.5
BM PB
2 1·5 1 0·5 0 3
6 8 Months post-transplant
14
16
Fig 2. Levels of human cell engraftment in neural stem cell (NSC) primary recipients post-transplant. Flow cytometric analysis was performed at regular intervals in bone marrow and peripheral blood of NSC primary recipients using human-specific antibodies. The level of human cells, as assessed by their expression of CD45 cells are presented at the various times post-transplant. Although there was a decrease in the levels of human cells with increasing time post-transplant, at the time point of 16 months post-transplant the sheep still maintained their chimaeric status.
brain, lung and muscle (Fig 1A). While at 2 months posttransplant we were unable to identify haematopoietic cells by flow cytometry (probably the number of human cells present were below our detection threshold) in either BM or peripheral
blood (PB) of these NSC primary recipients, at 3 months, human cells were detectable by this method in the BM and PB of the remaining transplanted animals at the levels of 1Æ7 ± 0Æ71% and 0Æ8 ± 0Æ02% respectively. At 6 months post-transplant, the levels of human cells as assessed by CD45 expression was essentially maintained from the 3-month time point with 1Æ6 ± 0Æ95% and 1 ± 0Æ4% human cells present in BM and PB respectively. Although there was a decrease in the levels of human cells with increasing time post-transplant, at 16 months post-transplant the sheep still maintained their chimaeric status (Fig 2). Phenotypic analysis of circulating human cells in these primary recipients showed the presence of human cells displaying T (0Æ12–0Æ34%), B (0Æ01–0Æ03%), and red blood cell markers (0Æ5–0Æ72%), albeit at low frequency. At 8 months post-transplant, BM cells harvested from the primary recipients were cultured in methylcellulose in the presence of human growth factors and analysed by karyotyping. A total of 53 individual colonies were harvested and cultured for karyotyping. Eight of 53 colonies evaluated were of human origin. Further confirmation of the human origin of these colonies was obtained by performing PCR on the harvested colonies using human HLA-DQa-specific primers (data not shown).
Long-term engrafting HSC were present in BM of NSC primary recipients The ability to engraft secondary recipients upon serial transplantation is a critical functional characteristic of HSC and has been an essential tool to distinguish between short- and longterm engrafting cells (Civin et al, 1996; Verfaillie et al, 2000). Thus, in order to critically assess whether the haematopoietic cells that were generated in vivo following transplantation of NSC had long-term engrafting capability, human cells were sorted from BM of primary transplanted animals at 14 and 16 months post-transplant based on their positivity for CD45/ HLA-DR. We obtained 22 · 104 and 3Æ3 · 104 human cells at 14 and 16 months post-transplant respectively. Secondary recipients were transplanted at cell doses of 1Æ5 · 104 (n ¼ 2) and 11 · 104 cells/fetus (n ¼ 2). At 2 months post-transplant, BM and PB from secondary recipients were analysed for the presence of human cells. Human haematopoietic cells were detected by flow cytometry in PB and BM of all four animals, with levels of human cells ranging from 0Æ35% to 3Æ29% in BM and 3Æ5–5Æ39% in PB. The engraftment in secondary recipients was also multilineage (Table I), and of particular note was the presence of human CD34+ cells (0Æ12–0Æ48%) in these chimaeric secondary animals.
Comparison of the haematopoietic activity between sheep transplanted with BM HSC and NSC In the previous experiments, we had successfully demonstrated that NSC were able to generate low levels of haematopoietic cells in our model, and that these cells possessed long-term
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
279
G. Almeida-Porada et al
Phenotype
% Human cells
CD7 CD20 GLY A CD34
6Æ7–14Æ6 0Æ08–0Æ21 0Æ05–2Æ7 0Æ12–0Æ48
A
6
% Human cells
Table I. Multilineage engraftment at 2 months post-transplant in secondary sheep recipients transplanted with human bone marrow cells harvested from primary neural stem cell recipients.
5 4 3 2 1 0
Discussion Studies performed by Bjornson et al (1999) and Shih et al (2001) demonstrated that murine and human NSC have the
280
6
101 102 103 CD7 FITC
Anti-HLA-DR PE 100 101 102 103 104
104
100 101 102 103 104 CD45 FITC
100
100
CD34 PE
104
100 101 102 103 104
100
101 102 103 CD7 FITC
Anti-HLA-DR PE
CD3 PE 100 101 102 103 104
CD3 PE
100 101 102 103 104
100
8 months
101 102 103 CD45 FITC
104
101 102 103 GLY A FITC 100 101 102 103 104
3
B
CD34 PE 100 101 102 103 104
2
100
104
101 102 103 GLY A FITC
104
Fig 3. (A) Comparison of the haematopoietic activity between sheep transplanted with bone marrow (BM) haematopoietic stem cell (HSC) and neural stem cell (NSC). Levels of human cells as assessed by flow cytometry using an antibody specific for human CD45 in primary sheep recipients transplanted with NSC (black squares) or adult BM CD34+ cells (black triangles) are compared at time points from 2 to 8 months post-transplant. Primary animals transplanted with human HSC (triangles) exhibited significantly higher levels of human blood cells than NSC transplanted sheep (squares) at all time points post-transplant. This is especially evident early post-transplant at the time point of 2 months, when NSC animals had undetectable levels of human cells by flow cytometry, whereas HSC-transplanted animals already showed 2Æ2% human cells by the same method. (B) Flow cytometric analysis of peripheral blood (PB) at 8 months post-transplant. Multilineage engraftment in the PB of animals injected with either CD34+Lin) cells (top row) or NSCs (bottom row) can be seen. Animals transplanted with human HSC exhibit significantly higher levels of different human blood cell phenotypes than NSC transplanted sheep.
A
B Human WBC X10E4
14
Sheep WBC X10E6
reconstitution ability. However, we wished to further characterise this population to determine whether they were functionally indistinguishable from HSC. To accomplish this, we compared the haematopoietic activity in NSC-transplanted animals with sheep that were transplanted with comparable numbers of human BM HSC. Figure 3A compares the levels of human haematopoietic cells in primary sheep recipients transplanted with NSC or adult BM CD34+Lin) cells (n ¼ 5). Figure 3B shows the multilineage engraftment in the PB at 8 months post-transplant in animals injected with either CD34+Lin) cells or NSCs. Clearly, primary animals transplanted with human BM CD34+Lin) cells exhibited a significantly higher level of human chimaerism than NSC-transplanted sheep at all time points post-transplant. This was especially evident at the early time points, when sheep transplanted with NSC had undetectable levels of human blood cells by flow cytometric analysis, while sheep made chimaeric with adult BM cells already displayed significant levels of chimaerism (2Æ2 ± 0Æ3%). The difference was also apparent at late time points post-transplant, when sheep transplanted with human CD34+Lin) cells had almost 13 times more human blood cells than their NSC counterparts (Fig 3A). Another significant difference between NSC and HSC transplanted sheep was the levels of CD34+ cells present in the marrow and circulation of the primary recipients. While HSC transplanted sheep had levels of human CD34+ cells ranging from 0Æ1% to 0Æ4% starting at early time points post-transplant, human CD34+ cells were not detectable by flow cytometry in NSC transplanted sheep until 6 months post-transplant and even then were only detected after G-CSF mobilisation. To further evaluate the haematopoietic activity in NSC transplanted sheep, primary sheep recipients that had received NSC were mobilised at 6 months posttransplant with human G-CSF and their pattern of mobilisation compared with those that had received human BM CD34+ cells. As observed in Fig 4A, total sheep white blood cells increased to similar levels in both groups. However, unlike HSC-transplanted sheep, human cell activity, as assessed by CD45 positivity, remained low in NSC transplanted sheep throughout mobilisation (Fig 4B), except for the appearance of 0Æ12% CD34+ cells in PB on day 3 (data not shown).
12 10 8 6 4 2 0 Transplanted NSC
Transplanted HSC
450 400 350 300 250 200 150 100 50 0
DAY 0 DAY 2 DAY 3
Transplanted
Transplanted
NSC
HSC
Fig 4. Mobilisation of human cells using human granulocyte colonystimulating factor (G-CSF). Primary sheep recipients that had received neural stem cell (NSC) were mobilised at 6 months post-transplant with human G-CSF (5 lg/kg) and their pattern of mobilisation compared with those that had received human BM CD34+ cells. As in (A), total sheep white blood cells increased upon mobilisation to similar levels in both groups. However unlike haematopoietic stem celltransplanted sheep, human cell activity, as assessed by CD45 positivity, remained low in NSC transplanted sheep throughout mobilisation (B).
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
Neural Stem Cells Give Rise to Blood ability to produce haematopoietic cells in vivo. However, the studies by Bjornson et al (1999) also suggested that the transplanted NSC may function differently than HSC as a haematopoietic graft, in that a significant delay was observed before the NSC reconstituted haematopoiesis. While it was hypothesised that this delay may have been caused by the additional time required for the NSC to differentiate to HSC before haematopoiesis could be re-established, this was never experimentally addressed. In the present study, we employed a fetal sheep transplantation model to examine the ability of human fetal brain-derived NSC to contribute to haematopoiesis in vivo and compared the haematopoietic engraftment potential obtained with NSC to those attained with sheep transplanted at the same age in utero with human BM-derived HSC. Although this study did not employ clonally derived NSC, our cultured neurospheres were extensively characterised prior to transplantation by phenotypic analysis to rule out the possibility of any haematopoietic cell contamination within the neurospheres. The cultured neurospheres were negative for all haematopoietic markers tested, with the exception of CD90, a marker whose expression is clearly not limited to HSC (Woodbury et al, 2000). Indeed, this finding is in agreement with recent studies showing that a percentage of multipotent murine neurospheres, but not those that are most enriched for NSC activity, express CD90 (Rietze et al, 2001). The fact that CD90-positive NSC are not the population most enriched for neurogenic potential may be a possible explanation of why we were unable to find human-derived neural cells in transplanted animals. Another possible explanation is that either the cell dose employed and/or the administration route was not appropriate for these cells to engraft the brain and give rise to neural cells. Although BM-derived mesenchymal stem cells (MSC) have been reported to express several neural markers (Tondreau et al, 2004; Wislet-Gendebien et al, 2005), the fact that the cells utilised in the present study were derived from brain, coupled with the unique culture and passaging conditions used, and the inherent structure of the neurospheres from which our cells were harvested, preclude the possibility that the cells used in this study were MSCs. In order to examine whether there were basic differences in the in vivo haematopoietic potential of NSC versus human HSC, fetal sheep were transplanted with either 8 · 105 NSC (n ¼ 11) or comparable numbers of BM-HSC (n ¼ 5) and subsequently analysed for haematopoietic chimaerism. While HSC-transplanted sheep displayed robust long-term donor haematopoiesis starting at 2-months post-transplant, NSC recipients not only showed a delay in the ability to produce blood when compared with human HSC, but also clearly exhibited a significantly lower level of human chimaerism than HSC-transplanted sheep at all time points post-transplant. Another significant difference between NSC- and HSC-transplanted sheep was the levels of CD34+ cells present in the marrow and circulation of the primary recipients, while HSCtransplanted sheep had levels of human CD34+ cells ranging from 0Æ1% to 0Æ4% starting at early time points post-
transplant, human CD34+ cells were not detectable by flow cytometry in NSC-transplanted sheep until 6 months posttransplant and were only found after mobilisation with G-CSF. It is curious to note that secondary sheep recipients transplanted with BM cells from the NSC primary animals demonstrated higher levels of human haematopoietic engraftment, suggesting that NSC had generated long-term engrafting haematopoietic cells within the BM of the primary recipients that more efficiently generated blood cells upon re-transplantation. But, although we successfully demonstrated that the population of NSC used was indeed able to generate haematopoietic cells in our model, we also showed that there were differences in their differentiation/commitment processes. Recipients that had received NSC were mobilised at 6 months post-transplant with human G-CSF and their pattern of mobilisation compared with that of sheep that had received human BM CD34+ cells. Unlike HSC-transplanted sheep, human cell activity as assessed by CD45 positivity remained low in NSC-transplanted sheep throughout mobilisation (Fig 4B). Although our studies contradict recent studies (Morshead et al, 2002; Magrassi et al, 2003; Yusta-Boyo et al, 2004), it is most likely that the population of NSC used in our studies differed in their characteristics from those used in the other studies, as our NSC were of fetal origin, not embryonic, were isolated from a different brain region, and consisted of a more differentiated population, as they expressed CD90. In conclusion, the use of this unique model system has enabled us to demonstrate that a human non-clonally derived NSC population had the ability to give rise to both a range of differentiated haematopoietic cells types and HSC that were capable of engraftment upon serial transplantation in vivo, in a competitive and physiologically normal haematopoietic microenvironment. However, this model has also revealed that these cells seem to differ from human HSC regarding their haematopoietic potential, as evidenced by lower levels of haematopoietic activity in primary recipients, late detection of human CD34+ cells in BM, and inefficient mobilisation of the human component of their haematopoietic system. Although this study does not demonstrate that true plasticity exists, it shows beyond a doubt that a human fetal brain NSC population is able to give rise, to some extent, to cells with a haematopoietic phenotype, and may explain the controversial results obtained by the different groups working in this area.
Acknowledgements This work was supported by the Grants HL 70566, HL73737, HL52955 and NAG9-1340 from National Institutes of Health and NASA.
References Almeida-Porada, G., Porada, C.D., Chamberlain, J., Torabi, A. & Zanjani, E.D. (2004) Formation of human hepatocytes by human hematopoietic stem cells in sheep. Blood, 104, 2582–2590.
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
281
G. Almeida-Porada et al Benoit, B.O., Savarese, T., Joly, M., Engstrom, C.M., Pang, L., Reilly, J., Recht, L.D., Ross, A.H. & Quesenberry, P.J. (2001) Neurotrophin channeling of neural progenitor cell differentiation. Journal of Neurobiology, 46, 265–280. Bjornson, C.R., Rietze, R.L., Reynolds, B.A., Magli, M.C. & Vescovi, A.L. (1999) Turning brain into blood: a hematopoietic fate adopted by adult neural stem cells in vivo. Science, 283, 534–537. Civin, C.I., Almeida-Porada, G., Lee, M.-J., Olweus, J., Terstappen, L.W.M.M. & Zanjani, E.D. (1996) Sustained, retransplantable, multilineage engraftment of highly purified adult human bone marrow stem cells in vivo. Blood, 88, 4102–4109. Clarke, D.L., Johansson, C.B., Wilbertz, J., Veress, B., Nilsson, E., Karlstrom, H., Lendahl, U. & Frisen, J. (2000) Generalized potential of adult neural stem cells. Science, 288, 1660–1663. Engstrom, C.M., Demers, D., Dooner, M., McAuliffe, C., Benoit, B.O., Stencel, K., Joly, M., Hulspas, R., Reilly, J.L., Savarese, T., Recht, L.D., Ross, A.H. & Quesenberry, P.J. (2002) A method for clonal analysis of epidermal growth factor-responsive neural progenitors. Journal of Neuroscience Methods, 117, 111–121. Eriksson, P.S., Perfilieva, E., Bjork-Eriksson, T., Alborn, A.M., Nordborg, C., Peterson, D.A. & Gage, F.H. (1998) Neurogenesis in the adult human hippocampus. Nature Medicine, 4, 1313– 1317. Flake, A.W., Harrison, M.R., Adzick, N.S. & Zanjani, E.D. (1986) Transplantation of fetal hematopoietic stem cells in utero: the creation of hematopoietic chimeras. Science, 233, 776–778. Gage, F.H., Ray, J. & Fisher, L.J. (1995) Isolation, characterization, and use of stem cells from the CNS. Annual Review of Neuroscience, 18, 59–192. Galli, R., Borello, U., Gritti, A., Minasi, M.G., Bjornson, C., Coletta, M., Mora, M., De Angelis, M.G., Fiocco, R., Cossu, G. & Vescovi, A.L. (2000) Skeletal myogenic potential of human and mouse neural stem cells. Nature Neuroscience, 3, 986–991. Gussoni, E., Soneoka, Y., Strickland, C.D., Buzney, E.A., Khan, M.K., Flint, A.F., Kunkel, L.M. & Mulligan, R.C. (1999) Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature, 401, 390–394. Hulspas, R. & Quesenberry, P.J. (2000) Characterization of neurosphere cell phenotypes by flow cytometry. Cytometry, 40, 245– 250. Kawashima, I., Zanjani, E.D., Almeida-Porada, G., Flake, A.W., Zeng, H.-Q. & Ogawa, M. (1996) CD34-positive human marrow cells that express low levels of kit protein are enriched for long-term marrow engrafting cells. Blood, 87, 4136–4142. Krause, D.S., Theise, N.D., Collector, M.I., Henegariu, O., Hwang, S., Gardner, R., Neutzel, S. & Sharkis, S.J. (2001) Multi-organ, multilineage engraftment by a single bone marrow-derived stem cell. Cell, 105, 369–377. Lagasse, E., Connors, H., Al-Dhalimy, M., Reitsma, M., Dohse, M., Osborne, L., Wang, X., Finegold, M., Weissman, I.L. & Grompe, M. (2001) Liver repopulation and correction of metabolic liver disease by transplanted adult mouse pancreatic cells. American Journal of Pathology, 158, 571–579. Magrassi, L., Castello, S., Ciardelli, L., Podesta, M., Gasparoni, A., Conti, L., Pezzotta, S., Frassoni, F. & Cattaneo, E. (2003) Freshly dissociated fetal neural stem/progenitor cells do not turn into blood. Molecular and Cellular Neurosciences, 22, 179–187. Morshead, C.M., Benveniste, P., Iscove, N.N. & van der Kooy, D. (2002) Hematopoietic competence is a rare property of neural stem
282
cells that may depend on genetic and epigenetic alterations. Nature Medicine, 8, 268–273. Porada, C., Almeida-Porada, G. & Zanjani, E.D. (1999) Differing engraftment potential of human hematopoietic stem cells from bone marrow and mobilized peripheral blood of human/sheep chimeras. Blood, 94, 664a. Rietze, R.L., Valcanis, H., Brooker, G.F., Thomas, T., Voss, A.K. & Bartlett, P.F. (2001) Purification of a pluripotent neural stem cell from the adult mouse brain. Nature, 412, 736–739. Shih, C.C., Weng, Y., Mamelak, A., LeBon, T., Hu, M.C. & Forman, S.J. (2001) Identification of a candidate human neuro-hematopoietic stem-cell population. Blood, 98, 2412–2422. Theise, N.D., Badve, S., Saxena, R., Henegariu, O., Sell, S., Crawford, J.M. & Krause, D.S. (2000) Derivation of hepatocytes from bone marrow cells of mice after radiation-induced myeloablation. Hepatology, 31, 235–240. Tondreau, T., Lagneaux, L., Dejeneffe, M., Massy, M., Mortier, C., Delforge, A. & Bron, D. (2004) Bone marrow-derived mesenchymal stem cells already express specific neural proteins before any differentiation. Differentiation, 72, 319–326. Tropepe, V., Hitoshi, S., Sirard, C., Mak, T.W., Rossant, J. & van der Kooy, D. (2001) Direct neural fate specification from embryonic stem cells: a primitive mammalian neural stem cell stage acquired through a default mechanism. Neuron, 30, 65–78. Tsai, R.Y. & McKay, R.D. (2000) Cell contact regulates fate choice by cortical stem cells. Journal of Neuroscience, 20, 3725–3735. Verfaillie, C.M., Almeida-Porada, G., Wissink, S. & Zanjani, E.D. (2000) Kinetics of engraftment of CD34()) and CD34(+) cells from mobilized blood differs from that of CD34()) and CD34(+) cells from bone marrow. Experimental Hematology, 28, 1071–1079. Wang, X., Willenbring, H., Akkari, Y., Torimaru, Y., Foster, M., AlDhalimy, M., Lagasse, E., Finegold, M., Olson, S. & Grompe, M. (2003) Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature, 422, 897–901. Wislet-Gendebien, S., Hans, G., Leprince, P., Rigo, J.M., Moonen, G. & Rogister, B. (2005) Plasticity of cultured mesenchymal stem cells: switch from nestin-positive to excitable neuron-like phenotype. Stem Cells, 23, 392–402. Woodbury, D., Schwarz, E.J., Prockop, D.J. & Black, I.B. (2000) Adult rat and human bone marrow stromal cells differentiate into neurons. Journal of Neuroscience Research, 61, 364–370. Yin, A.H., Miragi, A.S., Zanjani, E.D., Almeida-Porada, G., Ogawa, M., Leary, A.G., Olweus, J., Kearney, J. & Buck, D.W. (1997) AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood, 90, 5002–5012. Yusta-Boyo, M.J., Gonzalez, M.A., Pavon, N., Martin, A.B., De La Fuente, R., Garcia-Castro, J., De Pablo, F., Moratalla, R., Bernad, A. & Vicario-Abejon, C. (2004) Absence of hematopoiesis from transplanted olfactory bulb neural stem cells. European Journal of Neuroscience, 19, 505–512. Zanjani, E.D., Pallavicini, M.G., Ascensao, J.L., Flake, A.W., Langlois, R.G., Reitsma, M., MacKintosh, F.R., Stutes, D., Harrison, M.R. & Tavassoli, M. (1992) Engraftment and long-term expression of human fetal hematopoietic stem cells in sheep following transplantation in utero. Journal of Clinical Investigation, 89, 1178– 1788. Zanjani, E.D., Srour, E.F. & Hoffman, R. (1995) Retention of longterm repopulating ability of xenogeneic transplanted purified adult
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
Neural Stem Cells Give Rise to Blood human bone marrow hematopoietic stem cells in sheep. Journal of Laboratory and Clinical Medicine, 126, 24–28. Zanjani, E.D., Almeida-Porada, G. & Flake, A.W. (1996) The human/ sheep xenograft model: a large animal model of human hematopoiesis. International Journal of Hematology, 63, 179–192.
Ziegler, B.L., Valtieri, M., Almeida-Porada, G., De Maria, R., Muller, R., Masella, B., Gabbianelli, M., Casella, I., Pelosi, E., Bock, T., Zanjani, E.D. & Peschle, C. (1999) KDR receptor: a key marker defining hematopoietic stem cells. Science, 285, 1553– 1558.
ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 276–283
283
