Freshly dissociated fetal neural stem/progenitor cells do not turn into blood

Available online at www.sciencedirect.com R

Molecular and Cellular Neuroscience 22 (2003) 179 –187

www.elsevier.com/locate/ymcne

Freshly dissociated fetal neural stem/progenitor cells do not turn into blood Lorenzo Magrassi,a,* Sandra Castello,b Laura Ciardelli,c Marina Podesta,b Antonella Gasparoni,d Luciano Conti,e Stefano Pezzotta,a Francesco Frassoni,b,1 and Elena Cattaneoe,1 a

Neurochirurgia Dipartimento di Chirurgia, Universita` di Pavia, IRCCS Policlinico S. Matteo, Pavia 27100, Italy Centro Cellule Staminali e Terapia Cellulare, Divisione di Ematologia, Ospedale San Martino, 16100 Genova, Italy c Lab. Immunologia Neonatale, Lab. Sper. Di Ricerca Area trapiantologica, IRCCS Policlinico S. Matteo d Divisione di Neonatologia e Terapia Intensiva Neonatale, Spedali Civili Brescia, 25100, Italy e Dipartimento di Scienze Farmacologiche e Centro di Eccellenza sulle Malattie Neurodegenerative, Universita` di Milano, 20133 Milano, Italy b

Received 17 March 2002; revised 13 September 2002; accepted 19 September 2002

Abstract Earlier studies suggested that stem cells from one somatic tissue may generate differentiated elements of another, embryologically unrelated, tissue after an exchange in their positions through transplantation. Two reports indicated that murine and human neural stem cells of clonogenic origin after in vitro expansion in growth factor-supplemented media, may sustain hematopoiesis when injected into sublethally irradiated mice. Here we investigated if freshly dissociated fetal neural cells (fNC) share the reported hemopoietic potential of in vitro expanded neural cells. In order to minimize the risk of hemopoietic contamination, donor cells were taken from mouse E10.5 developing brains, before completion of blood vessel ingrowth into the brain; 106 fNC derived directly from fetal brains of transgenic mouse expressing an enhanced version of the green fluorescent protein were injected into the tail vein or directly into the bone marrow of sublethally irradiated (6 Gy) C57B16 mice. After transplantation, the presence of donor-derived cells was assessed at different survival times by FACS analysis, PCR, and clonogenic stem cell assays on peripheral blood and bone marrow. While bone marrow-derived cells were detected from 2 weeks onward after grafting, none of the mice grafted with neural embryonic cells demonstrated any sign of transdifferentiation into hemopoietic cells up to 16 months after transplantation. Our data indicate that ability to transdifferentiate from neural into the hematopoietic phenotype, if present, is acquired only after in vitro expansion of neural stem/progenitor cells and it is not present in vivo. © 2003 Elsevier Science (USA). All rights reserved.

Introduction Repopulation of the hematopoietic system in sublethally irradiated allogenic host by cells originating in the brain has been described in the past using cells dissociated from neurospheres of clonogenic origin (Bjornson et al., 1999; Shih et al., 2001). Do stem/progenitor cells freshly dissociated from the developing mice brain and transplanted into

* Corresponding author. Neurosurgery, Department of Surgery, University of Pavia. P.zle Golgi 2, 27100 Pavia, Italy. Fax: ⫹39-0382-422231/ 86. E-mail address: [email protected] (L. Magrassi). 1 Both authors should be considered last author.

sublethally irradiated allogenic hosts without in vitro expansion also display hemopoietic potential? In the absence of a clear in vivo assay to identify the properties of neural stem cells, most authors use the ability to grow, in vitro, neurospheres which contain cells capable to differentiate as glia and neurons, as an operative definition for “neural stem cells” (Rossi and Cattaneo, 2002). Neural cells freshly dissociated from the developing murine brain do not represent a pure preparation of stem cells according to the abovementioned definition. However, methods proposed for neural stem cells purification from murine fetal brain differ widely (Uchida et al., 2000; Rietze et al., 2001) and their general utility is still unclear and not validated by independent laboratories.

1044-7431/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1044-7431(02)00029-5

180

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

The preparations used by Bjornson et al. (1999) and Shih et al. (2001) in their experiments were also not pure. In those studies, description of partial repopulation of the hemopoietic system by transdifferentiation of neural stem cells were based on the injection of unpurified neural cells dissociated from neurospheres after expansion in vitro by growth factor addition (Bjornson et al., 1999; Shih et al., 2001). Neurospheres contain cells that are clearly not uniform in their differentiative stage (Mokry et al., 1996) and estimation of the number of bona fide stem cells contained in a preparation of cells dissociated from neurospheres varies as widely as for those contained in the fetal brain (Hulspas and Quesenberry, 2000). As for the implantation of cells derived from neurospheres, the E10.5 brain-derived cells we grafted into sublethally irradiated host are an heterogeneous population of proliferating nestin-positive stem/progenitor cells. E10.5 fetal mouse brain contains a population of “primitive” FGF2 responsive stem cells (Tropepe et al., 1999) that later will generate a progeny of “more restricted” multipotential stem cells responsive to both EGF and FGF2 (Ciccolini, 2001; Martens et al., 2000). This implies that stem cells with the capacity of generating neurospheres are present in vivo among all cycling neural precursors at the embryonic age selected for our studies. Moreover brain circulation at E10.5 is still developing and very limited, thus minimizing the risk for contamination by circulating hemopoietic cells. Finally, the very primitive character of the stem cells present in the brain of an E10.5 fetus should favor transdifferentiation. However, in the present study we did not observe intrinsic hematopoietic potential of any of the grafted cells despite their survival of the grafting procedure.

Results Fetal neural cells (fNC) isolated from the embryonic brain do not generate circulating blood cells at any survival times To test the potential of freshly dissociated fNC to contribute to the hemopoietic lineage, fNC derived from mice ubiquitously expressing enhanced green fluorescent protein (EGFP) were injected in the tail vein (106 cells) or directly into the marrow cavity of the tibia (105 cells) of sublethally irradiated C57B1/6 mice. We chose to restrict our analysis to cells derived from E10.5 mice embryos in order to minimize the risk of contamination by circulating hematopoietic stem cells and maximize the size of the donor brain. All transplantation experiments included separate control animals subjected to the same grafting procedures. Control mice received an injection of 106 bone marrow cells (BMC) derived from transgenic mice ubiquitously expressing EGFP in the tail vein or 105 BMC injected directly into the marrow. As expected, donor fNC from each preparation generated, in vitro, both neurons and glia (not shown).

Fig. 1. PCR amplification. for detection of EGFP-expressing donor cells. Sample were separated on 4% agarose gel and stained by ethidium bromide. M.W., pBR322 cut with HaeIII; Lane 1, positive control (peripheral blood from EGFP donor mouse); lanes 2– 6, amplifications of DNA extracted from peripheral blood collected at 1, 2, 9, 60, 485 days after BMC grafts; lanes 7–11, amplifications of DNA extracted from peripheral blood collected at 1, 2, 9, 60, 485 days after grafting of 106 freshly dissociated fNC.

A short-term analysis of the fNC-grafted mice revealed that a consistent fraction of transplanted cells survived the injection in the tail vein or in the bone marrow. PCR analysis using EGFP primers demonstrated that a specific band could be detected in all grafts by amplification of DNA derived from peripheral blood collected from mice at 1, 2, and 9 days postgrafting (Fig. 1). Fluorescence activated cell sorting (FACS) analysis on a fraction of peripheral blood collected from the same animals revealed the presence of circulating donor-derived fNC. In fact, a fluorescent signal above background was seen at 9 days after transplant, even if no mice grafted by fNC ever reached a percentage of EGFP-positive cells above 3% (Fig. 2). We observed FACS detectable circulating fluorescent cells in fNC-grafted mice also at 15 days after injection, although the number was steadily decreasing (Fig. 2). However, when we performed the same PCR analyses at later time points (i.e., 30, 60, 120, and 495 days after grafting), we were unable to obtain a positive amplification signal from any of the fNC-grafted mice (Fig. 1). This was consistent among animals that received the neural graft intravenously and those directly injected into the bone marrow by puncturing the tibial shaft (the follow-up of the latter group of animals was limited to 4 months). FACS analysis of a fraction of the same sample of blood used for DNA extraction confirmed the PCR data, demonstrating that 1 month postimplantation and later none of the fNC-grafted mice exhibited a fluorescent signal distinguishable from background, as defined by the values obtained in untreated mice (Fig. 2). Moreover, no fluorescent cells with distinct hematic morphologies were ever seen in blood smears of mice grafted with fNC at any survival times after grafting. As expected, control mice grafted intravenously or directly into the tibia with EGFP-positive BMC consistently tested positive. PCR analysis on peripheral blood samples performed using EGFP-specific primers showed a positive band of increasing intensity until about the second month after transplantation (Fig. 1). The same result was obtained by FACS analysis, with the percentage of circulating fluorescent cells also increasing up to the second month after grafting. Fig. 2 shows the results, up to 7 months after

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

181

Fig. 2. Multiple FACS analyses were performed at different time points after grafting to monitor the frequency of EGFP-positive cells in the peripheral blood of mice transplanted with fNC (open bars) or BMC (solid bars). Data from animals injected intravenously or directly into the bone marrow of the tibia were pooled. Values of fluorescent cells were at all survival times more than one order of magnitude greater in BMC-grafted than in fNC-grafted mice. Values below 0.5% should be considered background since similar values could be found also in untreated mice. Y-axis: EGFP-positive cells as percentage of total gated events by FACS analysis. X-axis: times elapsed in days from transplantation. Bar height represent the average of different animals; at least three animals were analyzed for each time point. Error bars refer to the standard deviation of the results.

transplantation, of multiple FACS analysis performed on peripheral blood samples of BMC grafted animals. The absence of donor-derived cells in the hematopoietic

system of fNC grafted mice was also confirmed at much longer survival times. Fig. 3 shows representative FACS profiles obtained from the peripheral blood of a mouse

Fig. 3. FACS profile of nucleated peripheral blood cells analyzed for the presence of EGFP-specific fluorescence. (A) Mice irradiated and grafted 16 months before with 1 ⫻ 106 freshly dissociated fNC. No fluorescent cells were detected above background (upper right quadrant). (B) Mice irradiated and grafted 16 months before with 1 ⫻ 106 BMC; 58.76% of the gated cells were EGFP-positive.

182

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

Fig. 4. fNC do not generate Sca1⫹ cells. Double-channel FACS analysis of purified Sca1⫹ cells grafted, respectively, with 1 ⫻ 106 freshly dissociated fNC (A) or with the same number of BMC (B). All mice were injected into the bone marrow cavity of the tibia and survived for 2 months. (A) Bone marrows of two mice grafted with fNC were pooled before purification of the Sca1⫹ cells. After two rounds of immunomagnetic separation, the resulting cells were stained by a phycoerythrinated anti-Sca1⫹ monoclonal antibody and subjected to FACS analysis (FL1, EGFP; FL2, phycoerythrin). In (A) the number of doubly fluorescent cells was very close to zero and indistinguishable from background also presents in bone marrows from C57B1/6 mice that did not received any transplant, while in BMC-grafted mice, 4.76% of the gated cells were both fluorescent and Sca1⫹.

intravenously injected with fNC (Fig. 3A) or BMC (Fig. 3B) 16 months before. While no fluorescent cells are detectable in the fNC-grafted animals, more than 58% of the cells were fluorescent in the animal injected with BMC. Survival of EGFP-expressing BMC for times comparable to the animal lifespan suggests that immunogenicity of EGFP, if present at all, does not significantly impair cell survival. Fetal neural cells do not generate hematopoietic colonyforming cells in vitro and in vivo To test if transplanted fNC adopted a hematopoietic identity, all cell populations contained in the bone marrow of mice subjected to intravenous fNC graft or that had received the fNC implant directly in the marrow were harvested at different survival times and tested for the presence of fluorescent cells by FACS analysis and in vitro clonogenic assay. FACS analysis of cells harvested 9 days after transplantation from the bone marrow of fNC-grafted mice revealed a percentage of EGFP-positive cells (approximately 3%) that was very similar to those seen in the circulating blood (Fig. 2). At later times the number quickly decreased and after 1 month it reached background values (not shown). On the contrary, FACS analysis of BMC of control mice after 1 to 16 months of survival resulted in a percentage of EGFP-positive cells more than one order of magnitude higher compared to fNC-transplanted mice (not shown). Similar results were obtained by performing clonogenic stem cell assays on BMC harvested from fNC- or BMC-grafted mice. Assays performed on host BMC sampled 1 to 2 months after fNC donor graft resulted in an average number of colonies of 2.7 ⫾ 2.1 every 105 cells

initially plated, while from the BMC-grafted animals, 4 ⫾ 2.82 colonies were grown. No fluorescent colonies were detected among those grown from fNC-grafted animals, while 20% of those from BMC-transplanted animals expressed EGFP. Mice assayed by the same test 4 to 5 months after fNC transplantation showed no fluorescent colonies with an average growth of 16.5(⫾ 4.59) colonies every 105 cells plated. At the same survival time, BMC-grafted mice had 30% of fluorescent colonies with an average growth of 20.5(⫾ 0.7) colonies every 105 cells plated. We also pooled the cells derived from the bone marrow of two mice 2 months after intratibial grafting with fNC and isolated all Sca1⫹ cells by repeated positive immunomagnetic selection. This experiment was performed to increase the chance of finding donor-derived cells given that the pool of Sca1⫹ cells is greatly enriched in hematopoietic stem cells and early precursors cells already committed to a specific lineage. FACS analysis of these pools did not reveal any fluorescent cells in animals transplanted with fNC (Fig. 4A), while after the same survival time 4.76% of the cells isolated from one mouse grafted with BMC were fluorescent (Fig. 4B). Direct analysis of bone marrow content by fluorescence microscopy of bones of mice grafted with fNC confirmed the lack of any support to hemopoiesis by grafted fNC. No fluorescent cells were visible in the bone marrow 1 month after fNC transplantation independently from the site of the transplant (Fig. 5A). The same was true for longer survivals up to 16 months. On the contrary, EGFP-positive cells were aboundant in the marrow of BMC-transplanted animals at all survival times (Figs. 5C and 5E).

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

183

Fig. 5. (A) One hundred-micrometer vibratome section of a tibial diaphysis 1 month after injection of 105 freshly dissociated fNC into the bone marrow cavity. Only autofluorescent cells are visible (arrowheads); no EGFP-positive cells are present. The section was illuminated with blue light at 485 nm (EGFP fluorescent emission reaches its maximum at this value). Scale bar, 250 ␮m. (B) Same field and magnification as in (A) but illuminated with green light at 560 nm. At this wavelength, EGFP does not fluoresce and only autofluorescent cells are visible; arrowheads indicate the same cells that were indicated in (A). Note that all cells visible in (A) are also visible in (B), confirming that no EGFP-positive donor-derived cells were present in the marrow 1 month after grafting. (C) Tibial epiphysis of a control mice grafted in situ with 105 BMC. Both autofluorescent (arrowheads) and EGFP fluorescent cells (arrows) are visible. EGFP-positive cells are well integrated into both the marrow hematopoietic and stromal component. Excitation, 485 nm; scale bar, 100 ␮m. (D) Same field and magnification as in (C) but illuminated with green light (560 nm). Only autofluorescent cells are visible; arrowheads indicate the same cells that were indicated in (A). EGFP-positive cells indicated by arrows in (C) are, as expected, not visible at this excitation wavelength. (E) 100 One hundred-micrometer vibratome section of a tibial diaphysis from a control mouse that received 106 BMC 1 month before. EGFP-positive cells were readily visible in the bone marrow (arrows). Scale bar, 50 ␮m. (F) Same field and magnification as for (E), but shown under green light to identify autofluorescent cells (arrowheads). Note that all EGFP-positive cells are now invisible.

Fetal neural cells do not generate bone marrow stromal cells In order to test if engrafted fNC were able to generate bone marrow stromal cells, we performed analysis of serial

vibratome sections by fluorescence microscopy of femuri and tibiae of mice receving fNC grafts directly into the marrow space of the tibia. We were unable to detect any evidence of EGFP-positive cells at any of the times considered (Fig. 5A). On the contrary EGFP- positive stromal cells

184

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

and hematopoietic cells were readly visible at the same survival times after intratibial grafting of bone marrowderived cells (Fig. 5C).

Discussion We transplanted freshly isolated fetal neural cells derived from transgenic mice expressing EGFP into sublethally irradiated C57B1/6. Analysis of grafted animals did not reveal development of chimerism in the hematopoietic compartment even after very long survival times (16 months) after grafting. According to Bjornson and colleagues (1999), stable taking of CNS donor-derived peripheral blood cells is best assessed 20 to 22 weeks after grafting of putative neural stem cells expanded in vitro. In our experiments we could not detect donor-derived hematopoietic cells in animals grafted by fNC either at those time points or at much longer survival times, which are comparable with the normal lifespan of the animal. The same negative results were obtained by injecting the fNC in the tail vein or by transplanting them directly into the bone marrow cavity of sublethally irradiated mice. Instead, and as expected, stable chimerism was easily obtained after infusion of bone marrow-derived cells from the same strain of transgenic mice either into the tail vein or directly into the bone marrow of the host. At E10.5, the developmental stage of donor embryos in our experiments, the endothelial cells, pericytes and other mesenchymal cells move out of the capillaries contained in the perineural mesenchymal tissue and start to invade the developing CNS (Herken et al., 1989). While blood flow is present in the capillary invading the brain at E10 (Bauer et al., 1993), they are restricted to the more superficial layers of the CNS and perfusion is minimal compared to later stages. The risk of contamination of brain-derived cells by circulating hematopoietic stem cells is variable and increasing with the development of cerebral perfusion. Our negative results indicate that, as expected, contamination of the fNC preparation by circulating hematopoietic stem cells is indeed very rare if the cells are collected at this stage. Problems in the interpretation of experimental data due to contamination by circulating stem cells have already occurred in the past. The existence of a resident population of hemopoietic stem cells in the adult mouse brain has been described (Bartlett, 1982). However, these cells are not stably present in the brain but they derive from contamination by circulating stem cells (Benca et al., 1986; Hoogerbrugge et al., 1985; Stedra et al., 1988). The recently described hematopoietic potential of muscle cells (Jackson et al., 1999) has also been traced to contamination by cells of hematopoietic origin in the donor cell preparation (McKinney-Freeman et al., 2002). Earlier work (Bjornson et al., 1999) on transdifferentiation of neural cells into hemopoietic cells was based on cells derived from older embryos (E14); however, at that stage

circulation into the brain is already well developed and it is impossible to collect fNC without contamination by circulating hematopoietic cells. E10.5 fetal mouse brain contains stem cells with the capacity of generating neurospheres indistinguishable from those generated by cells at later stages as E14 (Tropepe et al., 1999). Moreover the very primitive character of the stem cells present in the brain of an E10.5 fetus should favor transdifferentiation into extraneural phenotypes. Our experiments, as the original observations on transdifferentiation of neural stem cells cultured in vitro, were performed using not histocompatibility matching animals (Bjornson et al., 1999; Shih et al., 2001). However, stable chimerism in the hematolimphopoietic system without signs of rejection or graft versus host disease can be routinely obtained and maintained in mice as confirmed by the result of our control group of animals. In control mice injected with BMC the level of chimerism was variable but never less than 15% of the nucleated blood cells originated from the graft. Moreover after 16 months of survival many animals engrafted with BMC had values of donor-derived nucleated blood cells above 50%. Histocompatibility should thus not be at the basis of our inability of observing transdifferentiation of fNC into hematopoietic cells. For the same reasons immunogenicity of EGFP, if present at all, does not seem to significantly impair cell survival under our experimental conditions. PCR analysis of circulating blood cells containing the EGFP transgene at various times after grafting showed that a positive signal was obtained from control mice at all survival times. On the contrary, after the 9th day of survival from the transplant we were unable to obtain any positive signal by PCR amplification of DNA extracted from the blood of mice grafted with fNC. Due to the sensitivity of the PCR protocol adopted, it is unlikely that latent microchimerism was induced by fNC transplantation in the blood of host mice. On the contrary, a low but consistent level of fluorescent cells in the blood of grafted mice 1 and 9 days after grafting was revealed by FACS analysis. PCR analysis of blood samples confirmed the presence of host-derived cells in the circulating blood during the first 2 weeks after transplantation. The results of PCR analysis indicate that at least some of the fluorescent cells revealed by FACS analysis are not macrophages loaded by EGFP due to the engulfment of the grafted fNC but donor cells surviving the transplant. Our ability to amplify EGFP DNA up to 9 days from the transplant suggests that grafted fNC survived and some of them could be found in peripheral blood several days after the transplant. The alternative explanation, i.e., that fNC died quickly after transplant releasing their DNA into serum, is unlikely since free DNA in mice serum is short-lived (minutes) and is degraded or rapidly taken up by the liver nonparenchymal cells (Kawabata et al., 1995; Kobayashi et al., 2001). Furthermore, the detection of minimal residual disease of hematological malignancies in animal models

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

indicate that wherever a positive PCR signal is demonstrated the expected neoplastic cells are present in the blood and massive cell death induced by experimental therapies does not interfere with the signal (Gobbi et al., 1997; Hosler et al., 2000; Pugatsch et al., 1993). Our results are consistent with the idea that repopulation of the bone marrow of irradiated mice by fNC is not possible without previous in vitro manipulation. It is of note that, while this article was in preparation, Morshead and collegues published evidence indicating that in their hands, even putative neural stem cells extensively cultivated in vitro do not turn into hematopoietic precursors with any appreciable frequence (Morshead et al., 2002). The above-mentioned study, together with our data obtained with freshly dissociated fNC, indicate that transdifferentiation into hematopoietic cells is not one of the potential differentiative pathways of neural stem cells. Morshead et al. suggested that genetic instability of putative neural stem cells after long-term in vitro expansion may explain the results of Bjornson et al. and Shih et al. that described in vivo transdifferentiation of neural stem cells into hematopoietic cells. They suggested that hematopoietic reconstitution could result from the presence of rare genetic mutations or misregulated transcriptional activity in the donor “cell line.” Support for this idea may come from the observation that immortalization of proliferating neural cells broadens their developmental potential with respect to their normal nonimmortalized counterpart (Gao and Hatten, 1994; Shihabuddin et al., 1995). Finally, recent studies indicate that transdifferentiation of in vitro expanded neural stem cells may be explained by fusion of donor cells with host cells (Terada et al., 2002; Ying et al., 2002). In our study, the absence of EGFPpositive cells after the first month posttransplant of fNC suggests that fusion is not a frequent event in vivo, or at least it does not occur with an appreciable frequency in the host hematopoietic system. Taken together the present evidence indicates that transdifferentiation of brain cells into blood, if it occurs, requires more than heterotopic transplantation from the central nervous system to the blood or even the bone marrow.

185

crossed together and the homozygotes further crossed together for at least five generations. The progeny obtained was homozygous for the transgene viable and fertile. All cells in the tissues of those mice can be easily detected when transplanted into nonfluorescent hosts by their fluorescence. Host mice C57B16J mice (5 to 8 weeks old) were subjected to whole-body irradiation (LINAC source) with a dose of 6Gy 24 h before grafting. Isolation of cells for transplantation E10.5 donor embryos were collected from two to three timed pregnant mothers. After the initial separation of the head in a different dish, brains were dissected free of the developing skull and meninges. For each transplantation session, an average of 18 brains was pooled. The duration of the dissection was kept below 150 min, and the dissected tissue maintained at 4°C in serum-free glucose-supplemented Earle medium. Tissue dissociation started immediately after the last brain was dissected. Cells were dissociated to a single-cell suspension by repeated pipetting through sterile plastic pipette tips of decreasing diameter. After centrifugation and washing (three times), cells were resuspended to a final concentration of 106 in 300 ␮l when injected into the tail vein, or 105 in 30 ␮l when injected into bone marrow. A small aliquot of the dissociated cells was also plated in vitro to check viability and capacity of generating neurons and glia. In all cases, neurospheres were grown and cells obtained were able to produce glia and neurons as assessed by immunohystochemistry (data not shown). Control BMC were obtained from the femurs of the pregnant mothers by washing the bone central canal with serum-free glucose-supplemented Earle medium. After washing three times and centrifugation, BMC were resuspended to a final concentration of 106 cells in 300 ␮l when injected into the tail vein, or 105 cells in 30 ␮l when injected into the tibia. Cell grafting and sample collection

Experimental methods Donor mice fNC were obtained from E10.5 embryos of timed pregnant transgenic mice homozygotes for the enhanced green fluorescent protein under the control of the chicken ␤-actin promoter. To obtain mice homozygotes for the transgene, hemizygotes transgenic mice carrying an “enhanced” version of the green fluorescent protein (EGFP) under the control of the chicken ␤-actin promoter and cytomegalovirus enhancer (Okabe et al., 1997) were crossed to 129sv mice. Brown mice carrying one copy of the transgene were

Twenty-seven C57B16J mice were grafted with freshly dissociated fNC; 23 animals received a single injection of 106 cells into the tail vein, while 4 were injected with 105 fNC into the marrow cavity of the left tibia by puncturing the tibial shaft. Fifteen control animals were injected with BMC; 10 mice were grafted by tail vein injection of 106 cells, while five received 105 BMC directly into the bone marrow cavity of the left tibia. Injection of the dissociated fNC and BMC into the tibial shaft was performed through a 27-gauge needle percutaneously implanted into the tibia by gentle but firm pressure. The correct position for the injection into the marrow cavity

186

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187

was reached by advancing the needle into the bone until a sudden drop into the injection pressure was experienced with free flow of the 30 ␮l of the cell suspension out of the needle into bone central canal. Peripheral blood was collected by puncture of the periorbital plexus at regular intervals up to 16 months after grafting. Blood smears were immediately obtained and inspected under a fluorescence microscope for the presence of EGFP-positive cells.

Sca1⫹ cell purification Cells were isolated by flushing the bone marrow content of long bones using a 26-gauge needle attached to a syringe. Bone marrow cells from two mice were collected, pooled, and subjected to two rounds of magnetic cell sorting using the Sca-1 MultiSort kit by Miltenyi Biotec. The efficency of sorting was then checked by FACS analysis using anti Sca1 phycoerythrinated monoclonal antibody.

PCR amplification One-twentieth of the DNA extracted from 200 ␮l of heparinated peripheral blood collected by puncturing the orbital venous plexus was utilized for PCR analysis using EGFP-specific primers. Genomic DNA was extracted by a silica gel membrane affinity method according to the manifacturer’s instructions (DNeasy System Qiagen) and amplified using 5⬘ and 3⬘ green PCR primers (BD Clontech) specific for many derivatives of the green fluorescent protein of Aequorea victoria. PCR amplification was performed in 50 ␮l of a solution containing 50 mM KCl, 10 mM Tris–HCl (pH 9.0 at 25°C), 0.1% Triton X-100, 2.5 mM MgCl, deoxynucleotide triphosphates 0.2 mM each, 1 ␮l of each primer solution as provided by the producer, and 1 unit of TAQ DNA polymerase (Promega). The reactions were subjected to 35 amplification cycles on a Thermal cycler 480 (Perkin Elmer). Each cycle consisted of the following steps: 94°C for 40 s, 58°C for 45 s, and 72°C for 1 min.

Acknowledgments This work was supported by the following grants: Ministero della Sanita`-Progetto Alzheimer (300RFA00/01-02) to L.M. and E.C., Consiglio Nazionale delle Ricerche (P.S. basi biologiche delle malattie degenerative del sistema nervoso centrale) to L.M., Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica (COFIN 2001) to S.P, and Associazione Italiana Ricerca Emato-Oncologica (AIREO) to F.F. We are indebted to Dr. Masaru Okabe (Genome Information Research Center Osaka University) for the gift of EGFP mice and to Dr. Ferdinando Rossi (University of Turin) for help with fluorescence microphotograpy. We also thank Mrs. Irene Marini and Mrs. Pina Dellisanti for invaluable technical assistance.

References Flow cytometry Blood was subjected to red cell lysis by treatment with ammonium chloride, washed once, and then resuspended in phosphate-buffered saline. The relative percentage of peripheral blood cells or BMC of donor origin was directly distinguished by their intrinsic EGFP fluorescence using one-color flow citometry (FACS Calibur, Becton Dickinson, Mountain View, CA, USA). Sca1⫹ cells were labeled with a phycoerythrinated monoclonal antibody against Sca1⫹ (Miltenyi Biotec) and the cells were subjected to two-color fluorescence analysis using phycoerithrin and EGFP fluorescence as parameters. Stem cell colony assay Standard methylcellulose-based semisolid assays using recombinant human IL3 (10 ng/ml, Genzyme, Cambridge, MA, USA) ⫹ GM-CSF (10 ng/ml, Genzyme) ⫹ SCF (10 ng/ml, Genzyme) ⫹ erythropoietin (Epo, 5 U/ml, Genzyme) were used to evaluate myeloid (CFU-GM), erythroid (BFUE), and mixed (CFU-GEMM) progenitors. After 14 days of incubation at 37°C in a humified 5% CO2 atmosphere, colonies were scored using an inverted microscope applying standard criteria for their identification.

Bartlett, P.F., 1982. Pluripotential hematopoietic stem cells in adult mouse brain. Proc Natl Acad Sci USA 79, 2722–2725. Bauer, H.C., Bauer, H., Lametschwandtner, A., Amberger, A., Ruiz, P., Steiner, M., 1993. Neovascularization and the appearance of morphological characteristics of the blood-brain barrier in the embryonic mouse central nervous system. Brain Res. Dev. Brain Res. 75, 269 – 278. Benca, R.M., Wemhoff, G., Quintans, J., 1986. Functional studies of pluripotential hematopoietic stem cells in mouse brain. J. Neuroimmunol. 10, 341–352. 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 stemcells in vivo. Science 283, 534 –537. Ciccolini, F., 2001. Identification of two distinct types of multipotent neural precursors that appear sequentially during CNS development. Mol. Cell. Neurosci. 17, 895–907. Gao, W.Q., Hatten, M.E., 1994. Immortalizing oncogenes subvert the establishment of granule cell identity in developing cerebellum. Development 120, 1059 –1070. Gobbi, A., Di Berardino, C., Scanziani, E., Garofalo, A., Rivolta, A., Fontana, G., Rambaldi, A., Giavazzi, R., Biondi, A., 1997. A human acute lymphoblastic leukemia line with the T(4;11) translocation as a model of minimal residual disease in SCID mice. Leuk. Res. 21, 1107–1114. Herken, R., Gotz, W., Wattjes, K.H., 1989. Initial development of capillaries in the neuroepithelium of the mouse. J. Anat. 164, 85–92. Hoogerbrugge, P.M., Wagemaker, G., van Bekkum, D.W., 1985. Failure to demonstrate pluripotential hematopoietic stem cells in mouse brains. Proc. Natl. Acad. Sci. USA 82, 4268 – 4269.

L. Magrassi et al. / Molecular and Cellular Neuroscience 22 (2003) 179 –187 Hosler, G.A., Bash, R., Scheuermann, R.H., 2000. Kinetics of early therapeutic response as measured by quantitative PCR predict survival in a murine xenograft model of human T cell acute lymphoblastic leukemia. Leukemia 14, 1215–1224. Hulspas, R., Quesenberry, P.J., 2000. Characterization of neurosphere cell phenotypes by flow cytometry. Cytometry 40, 245–250. Jackson, K.A., Mi, T., Goodell, M.A., 1999. Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc. Natl. Acad. Sci. USA 96, 14482–14486. Kawabata, K., Takakura, Y., Hashida, M., 1995. The fate of plasmid DNA after intravenous injection in mice: Involvement of scavenger receptors in its hepatic uptake. Pharm. Res. 12, 825– 830. Kobayashi, N., Kuramoto, T., Yamaoka, K., Hashida, M., Takakura, Y., 2001. Hepatic uptake and gene expression mechanisms following intravenous administration of plasmid DNA by conventional and hydrodynamics-based procedures. J. Pharmacol. Exp. Ther. 297, 853– 860. Martens, D.J., Tropepe, V., van Der Kooy, D., 2000. Separate proliferation kinetics of fibroblast growth factor-responsive and epidermal growth factor-responsive neural stem cells within the embryonic forebrain germinal zone. J. Neurosci. 20, 1085–1095. McKinney-Freeman, S.L., Jackson, K.A., Camargo, F.D., Ferrari, G., Mavilio, F., Goodell, M.A., 2002. Muscle-derived hematopoietic stem cells are hematopoietic in origin. Proc. Natl. Acad. Sci. USA 99, 1341–1346. Mokry, J., Subrtova, D., Nemecek, S., 1996. Differentiation of epidermal growth factor-responsive neural precursor cells within neurospheres. Acta Medica (Hradec Kralove) 39, 7–20. Morshead, C.M., Benveniste, P., Iscove, N.N., van Der Kooy, D., 2002. Hematopoietic competence is a rare property of neural stem cells that may depend on genetic and epigenetic alterations. Nat. Med. 8, 268–273. Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., Nishimune, Y., 1997. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett. 407, 313–319.

187

Pugatsch, T., Weiss, L., Slavin, S., 1993. Minimal residual disease in murine B-cell leukemia (BCL1) detected by PCR. Leuk. Res. 17, 999 –1002. 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. Rossi, F., Cattaneo, E., 2002. Neural stem cell therapy for neurological diseases: dreams and reality. Nat. Rev. Neurosci. 3, 404 – 409. Shih, C.C., Weng, Y., Mamelak, A., LeBon, T., Hu, M.C., Forman, S.J., 2001. Identification of a candidate human neurohematopoietic stemcell population. Blood 98, 2412–2422. Shihabuddin, L.S., Hertz, J.A., Holets, V.R., Whittemore, S.R., 1995. The adult CNS retains the potential to direct region-specific differentiation of a transplanted neuronal precursor cell line. J. Neurosci. 15, 6666 – 6678. Stedra, J., Lodin, Z., Rossmann, P., Hartman, J., Sterzl, J., 1988. Failure to detect the presence of pluripotential haemopoietic stem cells in the mouse brain. J. Neuroimmunol. 18, 217–222. Terada, N., Hamazaki, T., Oka, M., Hoki, M., Mastalerz, D.M., Nakano, Y., Meyer, E.M., Morel, L., Petersen, B.E., Scott, E.W., 2002. Bone marrow cells adopt the phenotype of other cells by spontaneous cell fusion. Nature 416 (6880), 542–545. Tropepe, V., Sibilia, M., Ciruna, B.G., Rossant, J., Wagner, E.F., van der Kooy, D., 1999. Distinct neural stem cells proliferate in response to EGF and FGF in the developing mouse telencephalon. Dev. Biol. 208, 166 –188. Uchida, N., Buck, D.W., He, D., Reitsma, M.J., Masek, M., Phan, T.V., Tsukamoto, A.S., Gage, F.H., Weissman, I.L., 2000. Direct isolation of human central nervous system stem cells. Proc. Natl. Acad. Sci. USA 97, 14720 –14725. Ying, Q.L., Nichols, J., Evans, E.P., Smith, A.G., 2002. Changing potency by spontaneous fusion. Nature 416 (6880), 545–548.