- Email: [email protected]
Adult CNS explants as a source of neural progenitors
Brain Research Protocols 14 (2005) 146 – 153 www.elsevier.com/locate/brainresprot
Protocol
Adult CNS explants as a source of neural progenitors Hiram Chipperfielda,*, Simon M. Coolb, Kuldip Bedic, Victor Nurcombeb a
Department of Molecular and Cellular Biology, Harvard University, 7 Divinity Avenue, Cambridge, MA 02138, USA b IMCB (Institute of Molecular and Cellular Biology), 61 Biopolis Drive, Proteos, 138673, Singapore c Department of Anatomy and Developmental Biology, University of Queensland, Street Lucia, QLD 4072, Australia Accepted 6 December 2004 Available online 5 February 2005
Abstract Adult neural progenitors have been isolated from diverse regions of the CNS using methods which primarily involve the enzymatic digestion of tissue pieces; however, interpretation of these experiments can be complicated by the loss of anatomical resolution during the isolation procedures. We have developed a novel, explant-based technique for the isolation of neural progenitors. Living CNS regions were sectioned using a vibratome and small, well-defined discs of tissue punched out. When cultured, explants from the cortex, hippocampus, cerebellum, spinal cord, hypothalamus, and caudate nucleus all robustly gave rise to proliferating progenitors. These progenitors were similar in behaviour and morphology to previously characterised multipotent hippocampal progenitor lines. Clones from all regions examined could proliferate from single cells and give rise to secondary neurospheres at a low but consistent frequency. Immunostaining demonstrated that clonal cortical progenitors were able to differentiate into both neurons and glial cells, indicating their multipotent characteristics. These results demonstrate it is possible to isolate anatomically resolved adult neural progenitors from small amounts of tissue throughout the CNS, thus, providing a tool for investigating the frequency and characteristics of progenitor cells from different regions. D 2005 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Regeneration Keywords: Adult neural progenitors; Explant; Clonal
1. Introduction Adult neural progenitors have now been isolated from a number of CNS regions, including the striatum [16], spinal cord [19] and substantia nigra [12]. Progenitors have also been isolated from the post-natal cerebellum [11]. They are best characterised from the hippocampus and the subventricular zone (SVZ) (reviewed in [7]). Amongst the defining characteristics of these progenitors include their potential to proliferate for extended periods, and the ability to differentiate into multiple cell types. Current methods for the isolation of adult neural progenitor cells involve the digestion of whole brain regions * Corresponding author. Fax: +1 617 496 8116. E-mail address: [email protected] (H. Chipperfield). 1385-299X/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainresprot.2004.12.003
with various proteolytic enzymes, followed by a density centrifugation step or, more recently, FACS [17]. These methods have provided a fruitful source of cells, enabling a full assessment of their potential through a variety of approaches, most notably transplantation [8,18]. A major disadvantage of this approach, however, is the quantity of tissue needed, and the subsequent requirement for difficult microdissection. The problem of anatomical resolution is evident in the contrast between the behaviours of adult stem cells derived from the SVZ and those from the dentate gyrus of the hippocampus [17]. Using precise microdissection techniques, it was found that cells derived from the SVZ readily displayed the hallmark characteristics of adult neural progenitor cells, while cells derived from the dentate gyrus of the hippocampus in contrast displayed only a limited
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
147
proliferative potential. These results contradict reports from other studies [8,18], which provided evidence for the continued proliferation of hippocampal progenitors. Among other factors, difficult dissections might have led to a contamination of immediately adjacent SVZ-derived cells into the hippocampal preparations, leading to the differences in observed results. We have previously used a digestion method to isolate adult neural progenitors from the hippocampus that were capable of proliferating from single cells [3] and subsequently differentiating into electrically-active neurons [9]. To improve the anatomical resolution of progenitor isolation, we have developed a novel, explant-based protocol for the isolation of adult neural progenitors. This technique facilitated the isolation of progenitors from small, welldefined explants from a range of CNS regions; they were able to proliferate extensively from clonal cells, and subsequently differentiate into both neurons and glia.
2. Materials and methods 2.1. Explant Isolation Male adult (over 4 months) dark Agouti rats were sacrificed via an overdose of ketamine/xylazine, their brains and/or spinal cords quickly dissected out and placed into ice cold HibernateA/2% B-27 (HibeA). Depending on the region of interest, the brain or spinal cord was then cut coronally into pieces approximately 0.5–1 cm in length. These tissue pieces were stored in ice cold HibeA for up to 2 h until they were attached to the stage of a MA752 Motorised Advance Vibroslice (Campden Instruments, USA) vibratome using a small amount of superglue. The tissue was then bathed in ice-cold HibeA and 100 Am thick coronal (sagittal, in the case of the spinal cord) sections cut. A straight-edged, flat razor blade with high-speed vibration and slow forward speed was found to produce the best sections. Once cut, the tissue sections were transferred to a 10 cm Petri dish containing a small amount (~500 Al) of HibeA. Under observation through a dissecting microscope, the end of a 1-ml Gilson pipette tip was used to punch out a small disc of tissue, which was then transferred into culture. The locations of representative punches are shown in Fig. 1. Correct punch location was verified by examining the location of the hole in the tissue section. It was possible to punch multiple explant discs from a single section using this technique. 2.2. Explant culture The explant discs were first placed into the centres of poly-d-lysine-coated (50 Ag/ml in H2O for 2 h at room temperature), 12 mm diameter glass coverslips in 24-well plates containing NeurobasalA (Invitrogen), 2% B-27 (Invitrogen) and 2 mM glutamine (NeuroA), supplemented
Fig. 1. The locations of the explants used to culture progenitor cells. Living brain tissue was sectioned and small discs of tissue punched out. A shows example punches in the head of the caudate nucleus used to culture caudate nucleus tissue. B shows example punches 300–500 Am lateral to the third ventricle used to culture hypothalamic tissue. C shows example punches of dorsal cortex gray matter used to culture cortex tissue and the CA1, CA2 and CA3 regions used to culture hippocampal tissue. No explants were taken from the dentate gyrus. Explants were also taken from dorsal grey matter of the cerebellum and the grey matter of the spinal cord between T2 and T8 (not shown).
with 10 ng/ml FGF-2 (Sigma). The amount of medium in each well was adjusted so that it just covered the disc of tissue, usually ~180 Al/well. The small amount of medium
148
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
seems to have provided good oxygen exchange, while simultaneously ensuring that the explants did not float freely about the well, so allowing attachments to form between the tissue and the coverslip. The medium was replaced every day for 2 weeks, after which the volume was increased to 500 Al/well; half changes of medium were then performed every 2–3 days. Once a significant number of cells and/or neurospheres were visible, they were mechanically dislodged from the coverslip and transferred to fresh poly-dlysine-coated coverslips. After this sub-plating, the cells were maintained as previously described [3] in NeuroA with 10 ng/ml of freshly thawed FGF-2. 2.3. Clonal analysis Neurospheres were mechanically dissociated by trituration 10–20 times with a 1-ml pipette into single cells and plated into poly-d-lysine-coated 96-well plates at a limiting dilutions corresponding to 1 viable cell per well. The culture medium used was NeuroA with 10 ng/ml FGF-2 and 0.5 Ag/ ml low molecular weight heparin. On the following day, the plates were carefully examined under phase-contrast microscopy, with only wells containing a single cell of healthy appearance (intact membrane under phase contrast) being considered for further analysis. Wells containing either multiple or no cells were excluded from further study. Half of the medium was then replaced every 2–3 days and the cells monitored continuously. A very small proportion of wells contained two colonies of proliferating cells; these were also excluded from further analysis, as it is possible they arose from two initial cells that were missed in the screening. 2.4. Immunostaining Immunostaining was performed as previously described [3]. Briefly, 1–5 103 cells/well were plated onto poly-dlysine coated coverslips in 24-well plates and incubated overnight, before being fixed. In some cases, the explant cultures were fixed directly. The coverslips were rinsed once in PBS, fixed for 30 min in 4% paraformaldehyde in PBS and rinsed 3 times in PBS. The cells were blocked in 5% foetal calf serum (FCS) and 0.5% Triton X-100 in PBS for 1 h, incubated with primary antibody diluted in blocking buffer for 2 h, washed 3 times in PBS, incubated with secondary antibody diluted in PBS for 1–2 h, washed 3 times in PBS and mounted in fluorescent mounting medium. In some cases a propidium iodide counter-stain was used to visualise all of the cells in which case the coverslips were incubated in 500 nM propidium iodide in PBS for 5 min after the final wash and then washed and additional 3 times in PBS. All incubations were performed at room temperature and the fluorescence was visualised under a BioRad 1024 confocal microscope. The following primary antibodies (dilution in brackets) were used, all were from Sigma, Australia unless noted:
mouse anti-MAP2 (1/50), mouse anti-nestin (1/200), mouse anti-GFAP (1/100), rabbit anti-GFAP (1/200) and mouse anti-GAP-43, Zymed, USA (1/50). Anti-mouse and antirabbit secondary antibodies labelled with FITC, Texas Red or Cy5 were used at a dilution of 1/200.
3. Results Living CNS tissue was cut into 100 Am thick sections and small pieces of tissue bpunched outQ using a standard Gilson 1 ml pipette tip. This process produced small discs of tissue approximately 800 Am in diameter and 100 Am thick. Neural progenitors from a variety of regions could be obtained at high efficiency when these discs were cultured in medium containing FGF-2 (Table 1). The most difficult technical aspect of the explant protocol was ensuring that the correct amount of medium was maintained in the wells. If there was too much medium, the disc of tissue would fail to attach, in which case, the cells would not migrate out onto the coverslip; if there was too little medium, the explant would dry out. Cells were seen to migrate out of the discs as soon as 3 days in vitro. These cells had a wide variety of phenotypes, including fibroblastic and stellate. Long neurites also extended out of some of the hypothalamic explants. Fig. 2A shows cells with a progenitor-cell phenotype after they had migrated out of a disc of cerebellum tissue. Within 1–2 weeks of culture, the disc of tissue would usually become dislodged from its original location, often leaving behind progenitor cells. After 3–8 weeks in culture, the majority of the progenitor cells formed lightly attached or free-floating neurospheres (Fig. 2B), while in a small number of cases fibroblastic-like cells predominated. A number of typical explant cultures were fixed after 2 or 6 weeks and immunostained with different neuronal markers. Low percentages of cells were positive for nestin (Fig. 2C), GAP-43 (Fig. 2D), MAP-2, neurofilament-200 and GFAP. A number of explants were fixed and immunostained after 2 and 6 weeks. In the 2-week
Table 1 Proportion of explants producing neural progenitor cells after 2 weeks Brain region
Number of explants with proliferating cells/total number of explants
Hippocampus Hypothalamus Cerebellum Cortex Spinal cord Caudate nucleus
9/21 9/12 9/12 10/12 9/12 12/12
Small tissue explants were punched out of living tissue sections and cultured as detailed in the text. After 2 weeks, it was observed that a proportion of the wells, as summarised in the table above, contained proliferating cells with a neural progenitor morphology. The above results are the totals from 4 independent animals.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
149
Fig. 2. Cells derived from cultured CNS tissue explants. A, cells with morphologies similar to progenitor cells can be seen next to a cerebellum explant after 2 weeks in culture. B, cells that have proliferated and migrated away from a cerebellum explant are beginning to form neurospheres after 6 weeks in culture (arrows point to forming neurospheres). This behaviour was typical, with cells from the majority of viable explants forming neurospheres within 3–8 weeks. C shows an example of a progenitor expressing nestin. A small number of cells also expressed GAP-43, a marker of axonal growth, shown in D. Scale bars; A and B are 200 Am, C is 25 Am and D is 100 Am.
explants, some MAP-2-positive cells were found on the coverslips although none were found in 6-week-old cultures; conversely no GFAP-positive cells were found at the earlier time point, while a low number were found in the 6-week-old explants (Table 2). Over the two time points, the percentage of nestin-positive cells dropped from approximately 12% to 3% (Table 2). There was no systematic difference in the proportions of different cell types between explants from different locations. After progenitor cells were re-plated into secondary wells they appeared indistinguishable in morphology and behaviour from hippocampal progenitor cells that we have previously isolated by conventional digestion techniques [3]. Three cell lines from each CNS region were expanded and maintained for further analysis. One of the cerebellar lines ceased to proliferate after approximately 1 month in culture, and eventually died. All other lines proliferated extensively, and could be frozen and thawed, passaged at least 15 times extensively and maintained for at least 2 months. Some lines were kept in continuous culture for over 6 months. No systematic differences in behaviour or morphology were observed between progenitors derived from different regions of the CNS, and there were no observable differences in their rates of proliferation in 10 ng/ml of FGF-2 (data not shown).
It proved predictably difficult to clone any of the progenitor lines from single cells. At very low densities, the progenitors would usually not survive for extended periods of time. This critical density effect was similar to that
Table 2 The proportion of nestin- and MAP-2-positive cells in explant cultures After 2 weeks in culture
After 6 weeks in culture
Nestin-positive 7/126 (6%) 18/211 (9%) 28/66 (42%) 4/66 (6%) Total 57/469 (12%) MAP-2-positive 5/42 (12%) 7/86 (8%)
Nestin-positive 25/1100 (2%) 40/1400 (3%) 57/2250 (3%)
Total 12/128 (9%)
Total 122/4750 (3%) GFAP-positive 8/1200 (0.7%) 12/2400 (0.5%) 0/320 (0%) Total 20/3920 (0.5%)
Cells that migrated out of explants were immunostained for various markers after 2 and 6 weeks in culture. The number of immunoreactive cells was counted in random microscope fields across the coverslip and compared to the number of total number of cells as visualised by propidium iodide staining. No GFAP-positive cells were observed at 2 weeks and no MAP-2positive cells were observed at 6 weeks. The majority of cells at 6 weeks could be nestin-negative progenitors. The explants used were from the cortex, spinal cord and hypothalamus and were from 3 independent experiments.
150
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
observed in our previous hippocampal progenitor lines [3]. The difficulty of purifying heparan sulfate from bulk cultures of progenitor cells, a technique that previously facilitated the cloning of hippocampal progenitors, precluded its widespread use in this study. After extensive testing it was found that NeuroA with at least 10 ng/ml of FGF-2 and 0.5 Ag/ml of low molecular weight heparin could facilitate the cloning of the progenitors from single cells; the progression of a single cell to a non-adherent neurosphere is depicted in Fig. 3. The general health of the cells before they were cloned was also important for their survival. Addition of FCS improved the initial viability of the clones, but tended to induce their differentiation rather than promote their proliferation. Neither conditioned medium from bulk cultures of progenitor cells, nor attempts at providing trophic support using NGF, BDNF or NT-3 had any noticeable effect on the survival of single cells. The initial week after plating the single cells proved critical during the cloning process. If a cell did not undergo 2–3 replications during this time, it was unlikely to survive. After this first week, the heparin was often omitted from the medium because it precipitated the detachment of the cells from the substrate; it simply proved easier to feed the cells when they were attached. The total number of clones surviving to form neurospheres for each brain region is given in Table 3. When some of the cloned neurospheres
Table 3 A low percentage of single cell clones generate neurospheres Brain region
Number of neurospheres generated/ total number of single cells observed (percentage)
Hippocampus Hypothalamus Cerebellum Cortex Spinal cord Caudate nucleus
1/46 21/75 1/31 6/112 5/59 10/110
(2%) (28%) (3%) (5%) (8%) (9%)
Progenitors from different CNS regions were disassociated and plated at clonal densities into 96-well plates. Wells containing a single viable cell were observed and a low proportion of these cells proliferated into secondary neurospheres as summarised above. The table above is the total from 24 independent experiments.
were re-cloned, they were all able to form secondary neurospheres. When FGF-2 was withdrawn from the medium and replaced with 10% FCS, the progenitors would begin to differentiate. After 7 days of differentiation, cells with morphologies corresponding to neurons, astroglial cells and possibly oligodendrocytes were visible. Cells from all regions of the CNS exhibited similar behaviour during their differentiation, but only the 3 cortical lines were subsequently examined in detail. When two independent lines of clonal cortical progenitors were differentiated and stained for lineage markers, cells positive for the neuronal markers
Fig. 3. Phase contrast micrographs of adult neural progenitors grown from single cells. Cells were seeded into 96-well plates at a clonal density. A shows a cell that has differentiated into what appears to be a neuron. B shows a cell that has undergone 2 divisions after 4 days in culture. C shows a 5 week old attached neurosphere, note the short radiating processes. D shows a large unattached 6 week old neurosphere. Scale bar is 100 Am in all panels.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
Fig. 4. Differentiated cortical progenitors express MAP-2 and GFAP. Progenitors were differentiated for 7 days in 10% FCS and then immunostained. A shows a phase-contrast micrograph of differentiated cells while B shows the same field immunostained for the neuronal marker MAP-2. C shows another coverslip that was immunostained for GFAP. The scale bar is 100 Am.
GAP-43 (12%) and MAP-2 (13%), and the glial marker GFAP (25%), were evident (Fig. 4).
4. Discussion This study demonstrates a robust technique for the isolation of neural progenitor cells from anatomicallydefined subregions of the CNS; it further demonstrates that cells cloned from wide areas of the brain respond to an established in vitro propagation method in essentially
151
identical ways. In comparison to other approaches, the explant isolation technique detailed here is technically simple to perform. Apart from the requirement for a vibratome, there is no need for specialised equipment and there are no steps that are particularly sensitive to variation. In some cases, it was difficult to ensure that the vibratome sections of the CNS tissue were of an even thickness, but this could be corrected in most cases by changing the orientation of the tissue on the cutting stage. The other step that required some care was the amount of medium covering the explants in the culture wells. It was sometimes difficult to find a balance between sufficient medium to avoid drying and too much medium where the explant remained detached from the bottom of the well. The first cells to migrate out of the explants displayed a number of distinct morphologies; possibly corresponding to neurons, glia and fibroblast cells in addition to progenitor cells. In a few cases, cells with a fibroblastic appearance proliferated relatively rapidly and overtook any progenitor cells within the well. However, in most cases, cells with a progenitor phenotype became the predominant cell type after 1–2 weeks in culture. This apparent selection for the progenitor cells could reflect the highly selective nature of the medium. Serum-free Neurobasal/B-27 was specifically formulated for the culture of neurons and does not promote the growth of glial cells [1]. The high concentration of FGF2 could have also preferentially selected for the proliferation of progenitor cells. At the first passage, the majority of nonprogenitor cells were not dislodged; the secondary cultures were therefore highly enriched for progenitors. Cells were isolated from at least 75% of the discs in all regions, except for the hippocampus where the yield was less than 50% (Table 1). Explants may have failed because there were no progenitors present within the tissue or because all of the progenitors died during the sectioning process. The latter case is pertinent to the hippocampal explants, as all 12 explants from one animal failed to produce cells, while 9 out of 11 explants from a different animal produced cells. By calculating the size of the explant and the density of cells in the region of interest it is possible to establish the frequency of progenitor cells in different CNS regions. This calculation would be assisted by reducing the size of each explant, perhaps through the use of a fine metal punch. This future refinement of the method will facilitate accurate calculations of progenitor densities, as well as reducing the possibility of cells arising from multiple progenitors within a single explant. It is of interest that after 2 weeks in culture a percentage of the adherent explant-derived cells were positive for the neural marker MAP-2 (Table 2), while no GFAP-positive cells were found. This somewhat surprising result could reflect the selective nature of the medium for neurons. The small number of MAP-2-positive cells were probably neurons that had become dislodged from the edges of the explant and settled onto the coverslip. These putative
152
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
neurons could have either died after 6 weeks of culture in response to the proliferative but not trophic conditions, or the large number of other types of cells arising at the later time period may have obscured them. After 2 weeks in culture, nestin-positive cells were found spread throughout the cultures and did not appear to form distinct colonies (Fig. 2 and Table 2). This even distribution indicates that it is unlikely the nestin-positive cells were descendants from a few founder cells; rather, the nestinpositive cells may have given rise to the other cells, or they could be the relatively uncommon daughter cells of another cell type. One study has shown that during development FGF-2-responsive, nestin-negative, neural progenitor cells give rise to large, morphologically distinct, nestin-positive daughter cells [4]. The majority of the nestin-positive cells found in the explant cultures of this study were relatively large and had longer, more extensive processes than their nestin-negative counterparts (Fig. 2C). It is therefore possible that the nestin-positive cells were the uncommon progeny of an FGF-2-responsive progenitor cell. The molecular identity of the nestin-negative progenitors is currently unknown although published transcriptional profiles of adult neural progenitors [5,15] provide a clear basis for the future screening of potential markers. Analysis of the cloning results presented here (Table 3) is somewhat limited as we did not determine the proportion of the single cells intrinsically lacking the potential to proliferate and the proportion that simply died. It is however clear that progenitors from all of the CNS regions examined could form neurospheres from a single cell. As can be seen in Fig. 3B, the first 2–3 divisions from a single cell could be observed, after which the individual cells became unclear (Fig. 3C). Although it is unclear if the first divisions occurred at the same time, they did appear to be symmetrical. The rate of proliferation at these first divisions was slower than if the cells were in bulk culture; a single cell replicated 2–3 times during its first week in culture compared to a doubling time of 22 h for bulk cultures of hippocampal progenitors (data not shown). This slower rate of proliferation could reflect the low-density culture conditions or may be a property of the progenitor cells that are capable of continued proliferation. The low percentage of single cells forming neurospheres was unremarkable (Table 3), as other groups have reported similar efficiencies [2,10] notwithstanding that differences in technique and the medium used make it difficult to compare results quantitatively. One obvious difference between the regions was the relatively high percentage of hypothalamic clones that formed secondary spheres (28% compared to 2–9%). A single 96-well plate contributed 19 (approximately 75% of the single cells in the plate) of the 21 clones that were observed from the hypothalamic lines. This could have been the result of chance, or the culture from which the cells were taken could have been particularly enriched for progenitor cells. Assessing if there are stable differences between cell lines in the proportion of cells that
can proliferate from a single cell would of necessity form the basis of another whole study, given the time-consuming nature of the assay. A previous study which found that explants from the SVZ but not the striatum or cortex could give rise to newly formed neurons and glial cells [13] apparently contradicts the evidence of cortical neural progenitors presented here. This discrepancy is most likely due to two key differences in the media that was used. Firstly, FGF-2 has been shown to be important in stimulating the proliferation of adult neural progenitors [14,16,18] but was absent in the previous study [13]; in this case, the progenitors may not have been sufficiently stimulated to proliferate. Secondly, FCS which induces progenitors to differentiate [14,16] was used in the previous study, which may caused the relatively rare progenitor cells present in the cortex to differentiate in the explant rather than migrate out of the tissue. A more recent study found that astrocyte-derived progenitors could be isolated from a number of CNS regions, notably the cerebellum, of post-natal mice but not adult animals [11]. Again, the absence of FGF-2 and the addition of FCS in their initial cultures may explain why no progenitors were observed from adult tissue. Several factors may be important for the success of this technique in cultivating adult neural progenitors. FGF-2 is a relatively unstable growth factor and incorrect storage can have a major effect on its potency [6]. We have previously shown that heparin, a co-factor for FGF-2, can greatly effect the proliferation and morphology of neural progenitors [3] and may have a confounding effect in some neural progenitor cultures [2]. The absence of any enzymatic dissociation may also reduce the exposure of progenitor cells to toxic metabolites released during digestion. The novel explant technique of isolating neural progenitors presented here has the advantage of being relatively simple to perform while being able to isolate progenitor cells from the CNS at a very high anatomical resolution. The finding that the isolated progenitors can proliferate for extended periods and are capable of producing secondary neurospheres from single cells (Fig. 3 and Table 3) is clear evidence that they possess at least one cardinal characteristic of progenitor cells. Cortical progenitors were also able to differentiate into both neurons and glial cells demonstrating a further characteristic of progenitor cells. The differentiation potential of the cortical and other lines remains to be fully assessed, but given their similarities in behaviour and proliferation to existing hippocampal lines it is expected that they will show all the characteristics of multipotent progenitors. This technique facilitates the isolation adult neural progenitors from small well-defined pieces of tissue from regions such as the cortex where progenitor cell isolation was not previously possible. The possible presence of progenitor cells spread throughout the CNS at relatively high proportions has important implications for neurological therapies seeking to utilise endogenous progenitor cells.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
References [1] G.J. Brewer, J.R. Torricelli, E.K. Evege, P.J. Price, Optimized survival of hippocampal neurons in B27-supplemented neurobasal, a new serum-free medium combination, J. Neurosci. Res. 35 (1993) 567 – 576. [2] B.J. Chiasson, V. Tropepe, C.M. Morshead, D. van der Kooy, Adult mammalian forebrain ependymal and subependymal cells demonstrate proliferative potential, but only subependymal cells have neural stem cell characteristics, J. Neurosci. 19 (1999) 4462 – 4471. [3] H. Chipperfield, K.S. Bedi, S.M. Cool, V. Nurcombe, Heparan sulfates isolated from adult neural progenitor cells can direct phenotypic maturation, Int. J. Dev. Biol. 46 (2002) 661 – 670. [4] F. Ciccolini, Identification of two distinct types of multipotent neural precursors that appear sequentially during CNS development, Mol. Cell. Neurosci. 17 (2001). [5] K.A. D’Amour, F.H. Gage, Genetic and functional differences between multipotent neural and pluripotent embryonic stem cells, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 11866 – 11872 (Electronic publication 2003 Aug. 15). [6] M.J. Davies, C.A. Mitchell, M.A. Maley, M.D. Grounds, A.R. Harvey, G.W. Plant, D.J. Wood, Y. Hong, T.V. Chirila, In vitro assessment of the biological activity of basic fibroblast growth factor released from various polymers and biomatrices, J. Biomater. Appl. 12 (1997) 31 – 56. [7] F.H. Gage, Neurogenesis in the adult brain, J. Neurosci. 22 (2002) 612 – 613. [8] D.G. Herrera, J.M. Garcia-Verdugo, A. Alvarez-Buylla, Adult-derived neural precursors transplanted into multiple regions in the adult brain, Ann. Neurol. 46 (1999) 867 – 877. [9] R.C. Hogg, H. Chipperfield, K.A. Whyte, M.R. Stafford, M.A. Hansen, S.M. Cool, V. Nurcombe, D.J. Adams, Functional maturation of isolated neural progenitor cells from the adult rat hippocampus, Eur. J. Neurosci. 19 (2004) 2410 – 2420.
153
[10] T.J. Kilpatrick, P.F. Bartlett, Cloning and growth of multipotential neural precursors: requirements for proliferation and differentiation, Neuron 10 (1993) 255 – 265. [11] E.D. Laywell, P. Rakic, V.G. Kukekov, E.C. Holland, D.A. Steindler, Identification of a multipotent astrocytic stem cell in the immature and adult mouse brain, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 13883 – 13888. [12] D.C. Lie, G. Dziewczapolski, A.R. Willhoite, B.K. Kaspar, C.W. Shults, F.H. Gage, The adult substantia nigra contains progenitor cells with neurogenic potential, J. Neurosci. 22 (2002) 6639 – 6649. [13] C. Lois, A. Alvarez-Buylla, Proliferating subventricular zone cells in the adult mammalian forebrain can differentiate into neurons and glia, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 2074 – 2077. [14] T.D. Palmer, J. Ray, F.H. Gage, FGF-2-responsive neuronal progenitors reside in proliferative and quiescent regions of the adult rodent brain, Mol. Cell. Neurosci. 6 (1995) 474 – 486. [15] M. Ramalho-Santos, S. Yoon, Y. Matsuzaki, R.C. Mulligan, D.A. Melton, bStemnessQ: transcriptional profiling of embryonic and adult stem cells, Science 298 (2002) 597 – 600. [16] B.A. Reynolds, S. Weiss, Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system, Science 255 (1992). [17] R.L. Rietze, H. Valcanis, G.F. Brooker, T. Thomas, A.K. Voss, P.F. Bartlett, Purification of a pluripotent neural stem cell from the adult mouse brain, Nature 412 (2001). [18] L.S. Shihabuddin, J. Ray, F.H. Gage, FGF-2 is sufficient to isolate progenitors found in the adult mammalian spinal cord, Exp. Neurol. 148 (1997) 577 – 586. [19] S. Weiss, C. Dunne, J. Hewson, C. Wohl, M. Wheatley, A.C. Peterson, B.A. Reynolds, Multipotent CNS stem cells are present in the adult mammalian spinal cord and ventricular neuroaxis, J. Neurosci. 16 (1996) 7599 – 7609.
Protocol
Adult CNS explants as a source of neural progenitors Hiram Chipperfielda,*, Simon M. Coolb, Kuldip Bedic, Victor Nurcombeb a
Department of Molecular and Cellular Biology, Harvard University, 7 Divinity Avenue, Cambridge, MA 02138, USA b IMCB (Institute of Molecular and Cellular Biology), 61 Biopolis Drive, Proteos, 138673, Singapore c Department of Anatomy and Developmental Biology, University of Queensland, Street Lucia, QLD 4072, Australia Accepted 6 December 2004 Available online 5 February 2005
Abstract Adult neural progenitors have been isolated from diverse regions of the CNS using methods which primarily involve the enzymatic digestion of tissue pieces; however, interpretation of these experiments can be complicated by the loss of anatomical resolution during the isolation procedures. We have developed a novel, explant-based technique for the isolation of neural progenitors. Living CNS regions were sectioned using a vibratome and small, well-defined discs of tissue punched out. When cultured, explants from the cortex, hippocampus, cerebellum, spinal cord, hypothalamus, and caudate nucleus all robustly gave rise to proliferating progenitors. These progenitors were similar in behaviour and morphology to previously characterised multipotent hippocampal progenitor lines. Clones from all regions examined could proliferate from single cells and give rise to secondary neurospheres at a low but consistent frequency. Immunostaining demonstrated that clonal cortical progenitors were able to differentiate into both neurons and glial cells, indicating their multipotent characteristics. These results demonstrate it is possible to isolate anatomically resolved adult neural progenitors from small amounts of tissue throughout the CNS, thus, providing a tool for investigating the frequency and characteristics of progenitor cells from different regions. D 2005 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Regeneration Keywords: Adult neural progenitors; Explant; Clonal
1. Introduction Adult neural progenitors have now been isolated from a number of CNS regions, including the striatum [16], spinal cord [19] and substantia nigra [12]. Progenitors have also been isolated from the post-natal cerebellum [11]. They are best characterised from the hippocampus and the subventricular zone (SVZ) (reviewed in [7]). Amongst the defining characteristics of these progenitors include their potential to proliferate for extended periods, and the ability to differentiate into multiple cell types. Current methods for the isolation of adult neural progenitor cells involve the digestion of whole brain regions * Corresponding author. Fax: +1 617 496 8116. E-mail address: [email protected] (H. Chipperfield). 1385-299X/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainresprot.2004.12.003
with various proteolytic enzymes, followed by a density centrifugation step or, more recently, FACS [17]. These methods have provided a fruitful source of cells, enabling a full assessment of their potential through a variety of approaches, most notably transplantation [8,18]. A major disadvantage of this approach, however, is the quantity of tissue needed, and the subsequent requirement for difficult microdissection. The problem of anatomical resolution is evident in the contrast between the behaviours of adult stem cells derived from the SVZ and those from the dentate gyrus of the hippocampus [17]. Using precise microdissection techniques, it was found that cells derived from the SVZ readily displayed the hallmark characteristics of adult neural progenitor cells, while cells derived from the dentate gyrus of the hippocampus in contrast displayed only a limited
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
147
proliferative potential. These results contradict reports from other studies [8,18], which provided evidence for the continued proliferation of hippocampal progenitors. Among other factors, difficult dissections might have led to a contamination of immediately adjacent SVZ-derived cells into the hippocampal preparations, leading to the differences in observed results. We have previously used a digestion method to isolate adult neural progenitors from the hippocampus that were capable of proliferating from single cells [3] and subsequently differentiating into electrically-active neurons [9]. To improve the anatomical resolution of progenitor isolation, we have developed a novel, explant-based protocol for the isolation of adult neural progenitors. This technique facilitated the isolation of progenitors from small, welldefined explants from a range of CNS regions; they were able to proliferate extensively from clonal cells, and subsequently differentiate into both neurons and glia.
2. Materials and methods 2.1. Explant Isolation Male adult (over 4 months) dark Agouti rats were sacrificed via an overdose of ketamine/xylazine, their brains and/or spinal cords quickly dissected out and placed into ice cold HibernateA/2% B-27 (HibeA). Depending on the region of interest, the brain or spinal cord was then cut coronally into pieces approximately 0.5–1 cm in length. These tissue pieces were stored in ice cold HibeA for up to 2 h until they were attached to the stage of a MA752 Motorised Advance Vibroslice (Campden Instruments, USA) vibratome using a small amount of superglue. The tissue was then bathed in ice-cold HibeA and 100 Am thick coronal (sagittal, in the case of the spinal cord) sections cut. A straight-edged, flat razor blade with high-speed vibration and slow forward speed was found to produce the best sections. Once cut, the tissue sections were transferred to a 10 cm Petri dish containing a small amount (~500 Al) of HibeA. Under observation through a dissecting microscope, the end of a 1-ml Gilson pipette tip was used to punch out a small disc of tissue, which was then transferred into culture. The locations of representative punches are shown in Fig. 1. Correct punch location was verified by examining the location of the hole in the tissue section. It was possible to punch multiple explant discs from a single section using this technique. 2.2. Explant culture The explant discs were first placed into the centres of poly-d-lysine-coated (50 Ag/ml in H2O for 2 h at room temperature), 12 mm diameter glass coverslips in 24-well plates containing NeurobasalA (Invitrogen), 2% B-27 (Invitrogen) and 2 mM glutamine (NeuroA), supplemented
Fig. 1. The locations of the explants used to culture progenitor cells. Living brain tissue was sectioned and small discs of tissue punched out. A shows example punches in the head of the caudate nucleus used to culture caudate nucleus tissue. B shows example punches 300–500 Am lateral to the third ventricle used to culture hypothalamic tissue. C shows example punches of dorsal cortex gray matter used to culture cortex tissue and the CA1, CA2 and CA3 regions used to culture hippocampal tissue. No explants were taken from the dentate gyrus. Explants were also taken from dorsal grey matter of the cerebellum and the grey matter of the spinal cord between T2 and T8 (not shown).
with 10 ng/ml FGF-2 (Sigma). The amount of medium in each well was adjusted so that it just covered the disc of tissue, usually ~180 Al/well. The small amount of medium
148
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
seems to have provided good oxygen exchange, while simultaneously ensuring that the explants did not float freely about the well, so allowing attachments to form between the tissue and the coverslip. The medium was replaced every day for 2 weeks, after which the volume was increased to 500 Al/well; half changes of medium were then performed every 2–3 days. Once a significant number of cells and/or neurospheres were visible, they were mechanically dislodged from the coverslip and transferred to fresh poly-dlysine-coated coverslips. After this sub-plating, the cells were maintained as previously described [3] in NeuroA with 10 ng/ml of freshly thawed FGF-2. 2.3. Clonal analysis Neurospheres were mechanically dissociated by trituration 10–20 times with a 1-ml pipette into single cells and plated into poly-d-lysine-coated 96-well plates at a limiting dilutions corresponding to 1 viable cell per well. The culture medium used was NeuroA with 10 ng/ml FGF-2 and 0.5 Ag/ ml low molecular weight heparin. On the following day, the plates were carefully examined under phase-contrast microscopy, with only wells containing a single cell of healthy appearance (intact membrane under phase contrast) being considered for further analysis. Wells containing either multiple or no cells were excluded from further study. Half of the medium was then replaced every 2–3 days and the cells monitored continuously. A very small proportion of wells contained two colonies of proliferating cells; these were also excluded from further analysis, as it is possible they arose from two initial cells that were missed in the screening. 2.4. Immunostaining Immunostaining was performed as previously described [3]. Briefly, 1–5 103 cells/well were plated onto poly-dlysine coated coverslips in 24-well plates and incubated overnight, before being fixed. In some cases, the explant cultures were fixed directly. The coverslips were rinsed once in PBS, fixed for 30 min in 4% paraformaldehyde in PBS and rinsed 3 times in PBS. The cells were blocked in 5% foetal calf serum (FCS) and 0.5% Triton X-100 in PBS for 1 h, incubated with primary antibody diluted in blocking buffer for 2 h, washed 3 times in PBS, incubated with secondary antibody diluted in PBS for 1–2 h, washed 3 times in PBS and mounted in fluorescent mounting medium. In some cases a propidium iodide counter-stain was used to visualise all of the cells in which case the coverslips were incubated in 500 nM propidium iodide in PBS for 5 min after the final wash and then washed and additional 3 times in PBS. All incubations were performed at room temperature and the fluorescence was visualised under a BioRad 1024 confocal microscope. The following primary antibodies (dilution in brackets) were used, all were from Sigma, Australia unless noted:
mouse anti-MAP2 (1/50), mouse anti-nestin (1/200), mouse anti-GFAP (1/100), rabbit anti-GFAP (1/200) and mouse anti-GAP-43, Zymed, USA (1/50). Anti-mouse and antirabbit secondary antibodies labelled with FITC, Texas Red or Cy5 were used at a dilution of 1/200.
3. Results Living CNS tissue was cut into 100 Am thick sections and small pieces of tissue bpunched outQ using a standard Gilson 1 ml pipette tip. This process produced small discs of tissue approximately 800 Am in diameter and 100 Am thick. Neural progenitors from a variety of regions could be obtained at high efficiency when these discs were cultured in medium containing FGF-2 (Table 1). The most difficult technical aspect of the explant protocol was ensuring that the correct amount of medium was maintained in the wells. If there was too much medium, the disc of tissue would fail to attach, in which case, the cells would not migrate out onto the coverslip; if there was too little medium, the explant would dry out. Cells were seen to migrate out of the discs as soon as 3 days in vitro. These cells had a wide variety of phenotypes, including fibroblastic and stellate. Long neurites also extended out of some of the hypothalamic explants. Fig. 2A shows cells with a progenitor-cell phenotype after they had migrated out of a disc of cerebellum tissue. Within 1–2 weeks of culture, the disc of tissue would usually become dislodged from its original location, often leaving behind progenitor cells. After 3–8 weeks in culture, the majority of the progenitor cells formed lightly attached or free-floating neurospheres (Fig. 2B), while in a small number of cases fibroblastic-like cells predominated. A number of typical explant cultures were fixed after 2 or 6 weeks and immunostained with different neuronal markers. Low percentages of cells were positive for nestin (Fig. 2C), GAP-43 (Fig. 2D), MAP-2, neurofilament-200 and GFAP. A number of explants were fixed and immunostained after 2 and 6 weeks. In the 2-week
Table 1 Proportion of explants producing neural progenitor cells after 2 weeks Brain region
Number of explants with proliferating cells/total number of explants
Hippocampus Hypothalamus Cerebellum Cortex Spinal cord Caudate nucleus
9/21 9/12 9/12 10/12 9/12 12/12
Small tissue explants were punched out of living tissue sections and cultured as detailed in the text. After 2 weeks, it was observed that a proportion of the wells, as summarised in the table above, contained proliferating cells with a neural progenitor morphology. The above results are the totals from 4 independent animals.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
149
Fig. 2. Cells derived from cultured CNS tissue explants. A, cells with morphologies similar to progenitor cells can be seen next to a cerebellum explant after 2 weeks in culture. B, cells that have proliferated and migrated away from a cerebellum explant are beginning to form neurospheres after 6 weeks in culture (arrows point to forming neurospheres). This behaviour was typical, with cells from the majority of viable explants forming neurospheres within 3–8 weeks. C shows an example of a progenitor expressing nestin. A small number of cells also expressed GAP-43, a marker of axonal growth, shown in D. Scale bars; A and B are 200 Am, C is 25 Am and D is 100 Am.
explants, some MAP-2-positive cells were found on the coverslips although none were found in 6-week-old cultures; conversely no GFAP-positive cells were found at the earlier time point, while a low number were found in the 6-week-old explants (Table 2). Over the two time points, the percentage of nestin-positive cells dropped from approximately 12% to 3% (Table 2). There was no systematic difference in the proportions of different cell types between explants from different locations. After progenitor cells were re-plated into secondary wells they appeared indistinguishable in morphology and behaviour from hippocampal progenitor cells that we have previously isolated by conventional digestion techniques [3]. Three cell lines from each CNS region were expanded and maintained for further analysis. One of the cerebellar lines ceased to proliferate after approximately 1 month in culture, and eventually died. All other lines proliferated extensively, and could be frozen and thawed, passaged at least 15 times extensively and maintained for at least 2 months. Some lines were kept in continuous culture for over 6 months. No systematic differences in behaviour or morphology were observed between progenitors derived from different regions of the CNS, and there were no observable differences in their rates of proliferation in 10 ng/ml of FGF-2 (data not shown).
It proved predictably difficult to clone any of the progenitor lines from single cells. At very low densities, the progenitors would usually not survive for extended periods of time. This critical density effect was similar to that
Table 2 The proportion of nestin- and MAP-2-positive cells in explant cultures After 2 weeks in culture
After 6 weeks in culture
Nestin-positive 7/126 (6%) 18/211 (9%) 28/66 (42%) 4/66 (6%) Total 57/469 (12%) MAP-2-positive 5/42 (12%) 7/86 (8%)
Nestin-positive 25/1100 (2%) 40/1400 (3%) 57/2250 (3%)
Total 12/128 (9%)
Total 122/4750 (3%) GFAP-positive 8/1200 (0.7%) 12/2400 (0.5%) 0/320 (0%) Total 20/3920 (0.5%)
Cells that migrated out of explants were immunostained for various markers after 2 and 6 weeks in culture. The number of immunoreactive cells was counted in random microscope fields across the coverslip and compared to the number of total number of cells as visualised by propidium iodide staining. No GFAP-positive cells were observed at 2 weeks and no MAP-2positive cells were observed at 6 weeks. The majority of cells at 6 weeks could be nestin-negative progenitors. The explants used were from the cortex, spinal cord and hypothalamus and were from 3 independent experiments.
150
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
observed in our previous hippocampal progenitor lines [3]. The difficulty of purifying heparan sulfate from bulk cultures of progenitor cells, a technique that previously facilitated the cloning of hippocampal progenitors, precluded its widespread use in this study. After extensive testing it was found that NeuroA with at least 10 ng/ml of FGF-2 and 0.5 Ag/ml of low molecular weight heparin could facilitate the cloning of the progenitors from single cells; the progression of a single cell to a non-adherent neurosphere is depicted in Fig. 3. The general health of the cells before they were cloned was also important for their survival. Addition of FCS improved the initial viability of the clones, but tended to induce their differentiation rather than promote their proliferation. Neither conditioned medium from bulk cultures of progenitor cells, nor attempts at providing trophic support using NGF, BDNF or NT-3 had any noticeable effect on the survival of single cells. The initial week after plating the single cells proved critical during the cloning process. If a cell did not undergo 2–3 replications during this time, it was unlikely to survive. After this first week, the heparin was often omitted from the medium because it precipitated the detachment of the cells from the substrate; it simply proved easier to feed the cells when they were attached. The total number of clones surviving to form neurospheres for each brain region is given in Table 3. When some of the cloned neurospheres
Table 3 A low percentage of single cell clones generate neurospheres Brain region
Number of neurospheres generated/ total number of single cells observed (percentage)
Hippocampus Hypothalamus Cerebellum Cortex Spinal cord Caudate nucleus
1/46 21/75 1/31 6/112 5/59 10/110
(2%) (28%) (3%) (5%) (8%) (9%)
Progenitors from different CNS regions were disassociated and plated at clonal densities into 96-well plates. Wells containing a single viable cell were observed and a low proportion of these cells proliferated into secondary neurospheres as summarised above. The table above is the total from 24 independent experiments.
were re-cloned, they were all able to form secondary neurospheres. When FGF-2 was withdrawn from the medium and replaced with 10% FCS, the progenitors would begin to differentiate. After 7 days of differentiation, cells with morphologies corresponding to neurons, astroglial cells and possibly oligodendrocytes were visible. Cells from all regions of the CNS exhibited similar behaviour during their differentiation, but only the 3 cortical lines were subsequently examined in detail. When two independent lines of clonal cortical progenitors were differentiated and stained for lineage markers, cells positive for the neuronal markers
Fig. 3. Phase contrast micrographs of adult neural progenitors grown from single cells. Cells were seeded into 96-well plates at a clonal density. A shows a cell that has differentiated into what appears to be a neuron. B shows a cell that has undergone 2 divisions after 4 days in culture. C shows a 5 week old attached neurosphere, note the short radiating processes. D shows a large unattached 6 week old neurosphere. Scale bar is 100 Am in all panels.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
Fig. 4. Differentiated cortical progenitors express MAP-2 and GFAP. Progenitors were differentiated for 7 days in 10% FCS and then immunostained. A shows a phase-contrast micrograph of differentiated cells while B shows the same field immunostained for the neuronal marker MAP-2. C shows another coverslip that was immunostained for GFAP. The scale bar is 100 Am.
GAP-43 (12%) and MAP-2 (13%), and the glial marker GFAP (25%), were evident (Fig. 4).
4. Discussion This study demonstrates a robust technique for the isolation of neural progenitor cells from anatomicallydefined subregions of the CNS; it further demonstrates that cells cloned from wide areas of the brain respond to an established in vitro propagation method in essentially
151
identical ways. In comparison to other approaches, the explant isolation technique detailed here is technically simple to perform. Apart from the requirement for a vibratome, there is no need for specialised equipment and there are no steps that are particularly sensitive to variation. In some cases, it was difficult to ensure that the vibratome sections of the CNS tissue were of an even thickness, but this could be corrected in most cases by changing the orientation of the tissue on the cutting stage. The other step that required some care was the amount of medium covering the explants in the culture wells. It was sometimes difficult to find a balance between sufficient medium to avoid drying and too much medium where the explant remained detached from the bottom of the well. The first cells to migrate out of the explants displayed a number of distinct morphologies; possibly corresponding to neurons, glia and fibroblast cells in addition to progenitor cells. In a few cases, cells with a fibroblastic appearance proliferated relatively rapidly and overtook any progenitor cells within the well. However, in most cases, cells with a progenitor phenotype became the predominant cell type after 1–2 weeks in culture. This apparent selection for the progenitor cells could reflect the highly selective nature of the medium. Serum-free Neurobasal/B-27 was specifically formulated for the culture of neurons and does not promote the growth of glial cells [1]. The high concentration of FGF2 could have also preferentially selected for the proliferation of progenitor cells. At the first passage, the majority of nonprogenitor cells were not dislodged; the secondary cultures were therefore highly enriched for progenitors. Cells were isolated from at least 75% of the discs in all regions, except for the hippocampus where the yield was less than 50% (Table 1). Explants may have failed because there were no progenitors present within the tissue or because all of the progenitors died during the sectioning process. The latter case is pertinent to the hippocampal explants, as all 12 explants from one animal failed to produce cells, while 9 out of 11 explants from a different animal produced cells. By calculating the size of the explant and the density of cells in the region of interest it is possible to establish the frequency of progenitor cells in different CNS regions. This calculation would be assisted by reducing the size of each explant, perhaps through the use of a fine metal punch. This future refinement of the method will facilitate accurate calculations of progenitor densities, as well as reducing the possibility of cells arising from multiple progenitors within a single explant. It is of interest that after 2 weeks in culture a percentage of the adherent explant-derived cells were positive for the neural marker MAP-2 (Table 2), while no GFAP-positive cells were found. This somewhat surprising result could reflect the selective nature of the medium for neurons. The small number of MAP-2-positive cells were probably neurons that had become dislodged from the edges of the explant and settled onto the coverslip. These putative
152
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
neurons could have either died after 6 weeks of culture in response to the proliferative but not trophic conditions, or the large number of other types of cells arising at the later time period may have obscured them. After 2 weeks in culture, nestin-positive cells were found spread throughout the cultures and did not appear to form distinct colonies (Fig. 2 and Table 2). This even distribution indicates that it is unlikely the nestin-positive cells were descendants from a few founder cells; rather, the nestinpositive cells may have given rise to the other cells, or they could be the relatively uncommon daughter cells of another cell type. One study has shown that during development FGF-2-responsive, nestin-negative, neural progenitor cells give rise to large, morphologically distinct, nestin-positive daughter cells [4]. The majority of the nestin-positive cells found in the explant cultures of this study were relatively large and had longer, more extensive processes than their nestin-negative counterparts (Fig. 2C). It is therefore possible that the nestin-positive cells were the uncommon progeny of an FGF-2-responsive progenitor cell. The molecular identity of the nestin-negative progenitors is currently unknown although published transcriptional profiles of adult neural progenitors [5,15] provide a clear basis for the future screening of potential markers. Analysis of the cloning results presented here (Table 3) is somewhat limited as we did not determine the proportion of the single cells intrinsically lacking the potential to proliferate and the proportion that simply died. It is however clear that progenitors from all of the CNS regions examined could form neurospheres from a single cell. As can be seen in Fig. 3B, the first 2–3 divisions from a single cell could be observed, after which the individual cells became unclear (Fig. 3C). Although it is unclear if the first divisions occurred at the same time, they did appear to be symmetrical. The rate of proliferation at these first divisions was slower than if the cells were in bulk culture; a single cell replicated 2–3 times during its first week in culture compared to a doubling time of 22 h for bulk cultures of hippocampal progenitors (data not shown). This slower rate of proliferation could reflect the low-density culture conditions or may be a property of the progenitor cells that are capable of continued proliferation. The low percentage of single cells forming neurospheres was unremarkable (Table 3), as other groups have reported similar efficiencies [2,10] notwithstanding that differences in technique and the medium used make it difficult to compare results quantitatively. One obvious difference between the regions was the relatively high percentage of hypothalamic clones that formed secondary spheres (28% compared to 2–9%). A single 96-well plate contributed 19 (approximately 75% of the single cells in the plate) of the 21 clones that were observed from the hypothalamic lines. This could have been the result of chance, or the culture from which the cells were taken could have been particularly enriched for progenitor cells. Assessing if there are stable differences between cell lines in the proportion of cells that
can proliferate from a single cell would of necessity form the basis of another whole study, given the time-consuming nature of the assay. A previous study which found that explants from the SVZ but not the striatum or cortex could give rise to newly formed neurons and glial cells [13] apparently contradicts the evidence of cortical neural progenitors presented here. This discrepancy is most likely due to two key differences in the media that was used. Firstly, FGF-2 has been shown to be important in stimulating the proliferation of adult neural progenitors [14,16,18] but was absent in the previous study [13]; in this case, the progenitors may not have been sufficiently stimulated to proliferate. Secondly, FCS which induces progenitors to differentiate [14,16] was used in the previous study, which may caused the relatively rare progenitor cells present in the cortex to differentiate in the explant rather than migrate out of the tissue. A more recent study found that astrocyte-derived progenitors could be isolated from a number of CNS regions, notably the cerebellum, of post-natal mice but not adult animals [11]. Again, the absence of FGF-2 and the addition of FCS in their initial cultures may explain why no progenitors were observed from adult tissue. Several factors may be important for the success of this technique in cultivating adult neural progenitors. FGF-2 is a relatively unstable growth factor and incorrect storage can have a major effect on its potency [6]. We have previously shown that heparin, a co-factor for FGF-2, can greatly effect the proliferation and morphology of neural progenitors [3] and may have a confounding effect in some neural progenitor cultures [2]. The absence of any enzymatic dissociation may also reduce the exposure of progenitor cells to toxic metabolites released during digestion. The novel explant technique of isolating neural progenitors presented here has the advantage of being relatively simple to perform while being able to isolate progenitor cells from the CNS at a very high anatomical resolution. The finding that the isolated progenitors can proliferate for extended periods and are capable of producing secondary neurospheres from single cells (Fig. 3 and Table 3) is clear evidence that they possess at least one cardinal characteristic of progenitor cells. Cortical progenitors were also able to differentiate into both neurons and glial cells demonstrating a further characteristic of progenitor cells. The differentiation potential of the cortical and other lines remains to be fully assessed, but given their similarities in behaviour and proliferation to existing hippocampal lines it is expected that they will show all the characteristics of multipotent progenitors. This technique facilitates the isolation adult neural progenitors from small well-defined pieces of tissue from regions such as the cortex where progenitor cell isolation was not previously possible. The possible presence of progenitor cells spread throughout the CNS at relatively high proportions has important implications for neurological therapies seeking to utilise endogenous progenitor cells.
H. Chipperfield et al. / Brain Research Protocols 14 (2005) 146–153
References [1] G.J. Brewer, J.R. Torricelli, E.K. Evege, P.J. Price, Optimized survival of hippocampal neurons in B27-supplemented neurobasal, a new serum-free medium combination, J. Neurosci. Res. 35 (1993) 567 – 576. [2] B.J. Chiasson, V. Tropepe, C.M. Morshead, D. van der Kooy, Adult mammalian forebrain ependymal and subependymal cells demonstrate proliferative potential, but only subependymal cells have neural stem cell characteristics, J. Neurosci. 19 (1999) 4462 – 4471. [3] H. Chipperfield, K.S. Bedi, S.M. Cool, V. Nurcombe, Heparan sulfates isolated from adult neural progenitor cells can direct phenotypic maturation, Int. J. Dev. Biol. 46 (2002) 661 – 670. [4] F. Ciccolini, Identification of two distinct types of multipotent neural precursors that appear sequentially during CNS development, Mol. Cell. Neurosci. 17 (2001). [5] K.A. D’Amour, F.H. Gage, Genetic and functional differences between multipotent neural and pluripotent embryonic stem cells, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 11866 – 11872 (Electronic publication 2003 Aug. 15). [6] M.J. Davies, C.A. Mitchell, M.A. Maley, M.D. Grounds, A.R. Harvey, G.W. Plant, D.J. Wood, Y. Hong, T.V. Chirila, In vitro assessment of the biological activity of basic fibroblast growth factor released from various polymers and biomatrices, J. Biomater. Appl. 12 (1997) 31 – 56. [7] F.H. Gage, Neurogenesis in the adult brain, J. Neurosci. 22 (2002) 612 – 613. [8] D.G. Herrera, J.M. Garcia-Verdugo, A. Alvarez-Buylla, Adult-derived neural precursors transplanted into multiple regions in the adult brain, Ann. Neurol. 46 (1999) 867 – 877. [9] R.C. Hogg, H. Chipperfield, K.A. Whyte, M.R. Stafford, M.A. Hansen, S.M. Cool, V. Nurcombe, D.J. Adams, Functional maturation of isolated neural progenitor cells from the adult rat hippocampus, Eur. J. Neurosci. 19 (2004) 2410 – 2420.
153
[10] T.J. Kilpatrick, P.F. Bartlett, Cloning and growth of multipotential neural precursors: requirements for proliferation and differentiation, Neuron 10 (1993) 255 – 265. [11] E.D. Laywell, P. Rakic, V.G. Kukekov, E.C. Holland, D.A. Steindler, Identification of a multipotent astrocytic stem cell in the immature and adult mouse brain, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 13883 – 13888. [12] D.C. Lie, G. Dziewczapolski, A.R. Willhoite, B.K. Kaspar, C.W. Shults, F.H. Gage, The adult substantia nigra contains progenitor cells with neurogenic potential, J. Neurosci. 22 (2002) 6639 – 6649. [13] C. Lois, A. Alvarez-Buylla, Proliferating subventricular zone cells in the adult mammalian forebrain can differentiate into neurons and glia, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 2074 – 2077. [14] T.D. Palmer, J. Ray, F.H. Gage, FGF-2-responsive neuronal progenitors reside in proliferative and quiescent regions of the adult rodent brain, Mol. Cell. Neurosci. 6 (1995) 474 – 486. [15] M. Ramalho-Santos, S. Yoon, Y. Matsuzaki, R.C. Mulligan, D.A. Melton, bStemnessQ: transcriptional profiling of embryonic and adult stem cells, Science 298 (2002) 597 – 600. [16] B.A. Reynolds, S. Weiss, Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system, Science 255 (1992). [17] R.L. Rietze, H. Valcanis, G.F. Brooker, T. Thomas, A.K. Voss, P.F. Bartlett, Purification of a pluripotent neural stem cell from the adult mouse brain, Nature 412 (2001). [18] L.S. Shihabuddin, J. Ray, F.H. Gage, FGF-2 is sufficient to isolate progenitors found in the adult mammalian spinal cord, Exp. Neurol. 148 (1997) 577 – 586. [19] S. Weiss, C. Dunne, J. Hewson, C. Wohl, M. Wheatley, A.C. Peterson, B.A. Reynolds, Multipotent CNS stem cells are present in the adult mammalian spinal cord and ventricular neuroaxis, J. Neurosci. 16 (1996) 7599 – 7609.