Stem Cells http://www.peprotech.com/
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online August 11, 2005
Stem Cells Vol. 24 No. 2 February 2006, pp. 443 -453
doi:10.1634/stemcells.2004-0264; www.StemCells.com
© 2006 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2004-0264v1
24/2/443    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Biagiotti, T.
Right arrow Articles by Olivotto, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Biagiotti, T.
Right arrow Articles by Olivotto, M.

CANCER STEM CELLS

Cell Renewing in Neuroblastoma: Electrophysiological and Immunocytochemical Characterization of Stem Cells and Derivatives

Tiziana Biagiottia, Massimo D’Amicoa, Ilaria Marzia, Paola Di Gennarob, Annarosa Arcangelia, Enzo Wankec, Massimo Olivottoa

a Departments of Experimental Pathology and Oncology and
b Human Anatomy, Histology and Legal Medicine, University of Florence, Florence, Italy;
c Department of Biotechnology and Biosciences, University of Milan-Bicocca, Milan, Italy

Key Words. Neuroblastoma stem cells • Ion channels • Electrophysiological cluster of differentiation

Correspondence: Massimo Olivotto, M.D., Dipartimento di Patologia e Oncologia Sperimentali, Viale G.B. Morgagni 50, 50134 Firenze, Italy. Telephone: 39-055-4598203; Fax: 39-055-4598900; e-mail: olivotto{at}unifi.it


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We explored the stem cell compartment of the SH-SY5Y neuroblastoma (NB) clone and its development by a novel approach, integrating clonal and immunocytochemical investigations with patch-clamp measurements of ion currents simultaneously expressed on single cells. The currents selected were the triad IHERG, IKDR, INa, normally expressed at varying mutual ratios during development of neural crest stem cells, from which NB derives upon neoplastic transformation. These ratios could be used as electrophysiological clusters of differentiation (ECDs), identifying otherwise indistinguishable stages in maturation. Subcloning procedures allowed the isolation of highly clonogenic substrate-adherent (S-type) cells that proved to be p75- and nestinpositive and were characterized by a nude electrophysiological profile (ECDS0). These cells expressed negligible levels of the triad and manifested the capacity of generating the two following lineages: first, a terminally differentiating, smooth muscular lineage, positive for calponin and smooth muscle actin, whose electrophysiological profile is characterized by a progressive diminution of IHERG, the increase of IKDR and INa, and the acquisition of IKIR (ECDS2); second, a neuronal abortive pathway (NF-68 positive), characterized by a variable expression of IHERG and IKDR and a low expression of INa (ECDNS). This population manifested a vigorous amplification, monopolizing the stem cell compartment at the expense of the smooth muscular lineage to such an extent that neuronal-like (N-type) cells must be continuously removed if the latter are to develop.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Recent studies have proposed a new way of thinking about cancer, considering neoplasia as an ecosystem in which polyclonality generates a cell population more or less imperfectly organized in an abortive organogenesis [1, 2]. A crucial assumption in this new perspective is that a pool of cancer stem cells, endowed with unlimited capacity for self-renewal, will generate multiform progenies variously committed to terminal fates. It is further assumed that this committed compartment constitutes the highly mitotic bulk of tumors (which are targeted by conventional antiblastic treatments), whereas the staminal elements usually survive and prepare relapses [2]. For this reason, recent research is attempting to identify cancer stem cells and patterns of differentiation in the apparently anarchic manifestations of cancer to design radical therapies of malignancy [3, 4].

Within this framework, the neuroblastoma (NB) represents a first-choice clinical and experimental model. It is a tumor peculiar to childhood, characterized by an unpredictable clinical course and multiform phenotypes, as well as by a propensity to differentiate in various directions, either spontaneously or as a result of different treatments [5, 6]. Extensive studies by Biedler’s group identified in human NB three principal cell sub-populations: N-type cells, endowed with a neuronal-type immature phenotype; S-type cells, which are epithelial-like and strongly substrate-adherent cells; and intermediate cells (I-type), proposed as common ancestors of the other two subpopulations or as an intermediate stage of an interconversion between N and S [7–9]. Numerous phenotypes have been described in different NB lines and clones developed in vitro either spontaneously or upon different treatments, and this confirms the unique plasticity of this tumor under environmental and epigenetic influences [5, 10].

A major cause of this plasticity is the fact that NB originates from pluripotent staminal compartments of the neural crest (NC) [7, 11]. The NC is a transient embryonic structure whose cells migrate through embryonic tissues and come to a halt at elected sites where they differentiate into many various cell types [12–14]. At various steps of this process, several intermediate oligopotent precursors are produced, some of which behave as stem cells, in the sense that they are able to renew themselves and to generate a restricted progeny with an extended potential for proliferation [15]. This explains the striking heterogeneity displayed by neural crest cells (NCCs) when they are isolated in vitro and contributes to the phenotypic complexity of NC-derived tumors, especially of NB.

We designed an experimental protocol based on the use of the human NB adrenergic SH-SY5Y clone (hereinafter referred to as SY5Y) [7, 16] with the aim of providing an accurate map of its power of self-renewal and its potential for differentiation and, in the longer term, of contributing to a radical therapy of similar tumors. The protocol was suggested by our previous electrophysiological studies on NB [17] compared with NCC developing in vivo and in vitro [18]. We showed that the NC neuronal maturation is marked by an ordered expression of the following currents: IHERG, mediated by the human eag-related (HERG) potassium channels; IKDR, mediated by the delayed rectifier potassium (KDR) channels; INa, mediated by sodium tetrodotoxin–sensitive channels; and IKIR, mediated by the inward rectifier potassium (KIR) channels [17–21]. Only IHERG and IKDR were significantly expressed at the early stages of NC development; in later stages, IHERG disappeared altogether, whereas IKDR increased and IKIR and INa were newly expressed. These changes were accompanied by a substantial hyperpolarization of the resting potential (VREST) and acquisition of a mature neuronal phenotype. These results, in keeping with previous reports [22, 23], revealed that neuronal differentiation from immature precursors includes the expression, in defined sequences, of distinct sets of voltage-gated ion channels, which mark various developmental stages of NCC. Some evidence emerged that this is also true for the SY5Y clone. To be more precise, the N-type cells constituting the tumor bulk are apparently blocked at an immature stage of a neuronal commitment, with an electrophysiological profile dominated by IHERG and IKDR, a depolarized VREST, and the incapability of eliciting action potentials (APs). They become excitable, however, and express IKIR upon exposure to differentiating agents, such as retinoic acid and brain-derived neurotrophic factor (BDNF) [17, 18, 20, 21]. Moreover, we were able to recover S-type cells from the SY5Y population, which were endowed with a distinct electrophysiological profile, well polarized, and capable of eliciting AP [24].

In this paper, we report that after SY5Y subcloning, it is possible to recover—and to recover constantly—highly clonogenic S-type cells that give rise both to N-committed cells and to a smooth muscular lineage committed to a terminal differentiation. The identification and development of various progenitors was made possible by a new experimental approach based on the association of classic immunocytochemical procedures with a qualitative and quantitative analysis of the currents expressed in each single cell by means of the patch-clamp technique.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell Culture and Cloning
SH-SY5Y cells were maintained in RPMI 1640 medium (Euroclone, Milan, Italy, http://www.euroclone.net) supplemented with 10% fetal calf serum (Hyclone, Logan, UT, http://hyclone.com) in air containing 5% CO2, as previously described [15]. Subclones were obtained by limiting dilution in 96 multi-wells. Single N- or S-type cells were observed after approximately 4 days and followed in their clonal expansion. At approximately 60 days in culture, S subclones produced N-type cells that were removed by vigorously rinsing the plates with phosphate-buffered saline (PBS), discarding the supernatant, and maintaining the adherent cells [7, 24]. To evaluate the S0 renewal capacity, a second round of cloning of S0 subclones was performed, diluting cells at clonal density and seeding them on CELLocate coverslips (Eppendorf, Hamburg, Germany, http://www.eppendorf.com), characterized by a microgrid, permitting the identification of single cells with precise coordinates.

The clonal analysis was performed by successive limiting of dilutions of S colonies expanded to not more than 20 cells [25]. At each passage, cells of the colonies were analyzed by the patch-clamp technique.

Contrast-phase images of cell populations were acquired using a Nikon TMS microscope with x 20 or x 40 objective using a digital camera.

Clonogenic Efficiency
N-type and S-type clones, obtained by the limiting dilution of SY5Y parental clone, were plated in 96-mm dishes at a dilution of 3 cells/ml and allowed to expand to form colonies of approximately 100 cells. Clones of this size were counted and the clonogenic efficiency estimated by the ratio of the number of colonies versus the number of cells seeded. When these colonies attained the number of approximately 100 cells, they were processed to start a new dilution procedure for two successive passages.

Immunofluorescence
Cells were seeded onto 13-mm glass coverslips and cultured as above in 24 multiwells (Corning, Cambridge, MA, http://www.corning.com). Cell cultures were fixed in 3.7% formaldehyde in PBS for 20 minutes at 4°C and then incubated with Hoechst at 37°C for 30 minutes to stain cell nuclei. Then cell cultures were washed with PBST (PBS + 0.1% Triton X-100) and incubated with blocking buffer (3% bovine serum albumin [BSA] in PBST) for 45 minutes. This was followed by incubation with a primary antibody for 16 hours at 4°C. Each cell population was tested for the following list of antibodies: mouse anti–smooth muscle actin (SMA) (1:800, Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com); mouse anti–Glial fibrillary acidic protein (GFAP) (1:2,000; Sigma); goat anti–neural cell adhesion molecule–L1 (1:1,000; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com); mouse anti-calponin (1:3,000, Sigma); mouse antinestin (1:800, BD Transduction Laboratories, Franklin Lakes, NJ, http://www.bdbiosciences.com); mouse anti-NF68 (1:1,500, Sigma); mouse anti-NF160 (1: 2,000, Sigma); mouse anti-NF200 (1:2,500, Sigma); rabbit anti-p75 (1:800, Chemicon, Temecula, CA, http://www.chemicon.com); goat anti–S-100{alpha} (1:1,000; Santa Cruz Bio-technology); and mouse anti-vimentin (1:3,000, Sigma). After washing with PBST, the cell cultures were incubated with anti-mouse {alpha}Cy3-conjugated (Chemicon), anti-rabbit fluoresce-in-conjugated (Chemicon), or anti-goat fluorescein-conjugated (Calbiochem, San Diego, http://www.emdbiosciences.com) secondary antibody (1:800 dilution in PBST/3% BSA) for 1 hour at room temperature. Cells fixed onto coverslips were mounted on a glass slide with n-propylgallate. Preparations were observed on a Nikon TMS microscope equipped with fluorescence at x 40 objective. Cell images were acquired under appropriately filtered light using a digital camera. Except in a few cases specified in the legends, populations negative to the above antibodies are not illustrated in the figures.

Patch-Clamp Recordings
Cells, seeded on 35-mm Petri dishes (Corning), were patched at room temperature after 2 days of culture, and traces were recorded with the patch-clamp amplifier MultiClamp 700A (Axon Instruments, Foster City, CA, http://www.moleculardevices.com) using the whole-cell configuration [26] or perforated patch to record the VREST and AP [27]. Measurements of the membrane potential (Vm) and currents were performed in current clamp (I = 0 mode) or voltage clamp, respectively. The pipettes used (borosilicate glass, Harvard Apparatus, Kent, UK, http://www.harvardapparatus.com) had resistances ranging between 3 and 5 M{Omega}. Gigaseal resistances were in the range 1 to 10 G{Omega}. The cell capacitance was obtained directly by reading the cell capacitance compensation. Input resistances of the cells were in the range 2 to 6 G{Omega}. Whole-cell currents were filtered at 1 to 3 KHz. For data acquisition and analysis, the pClamp and Axoscope software (Axon Instruments) and Origin (OriginLab, Northampton, MA, http://www.originlab.com) were routinely used. Extracellular solutions were delivered with hypodermic needles inserted into a capillary with a small hole (inner diameter, 0.4 mm), positioned near the cell under study. The extra-cellular solution with low potassium (low Ko solution [K]o = 5 mM) contained (in mM) NaCl 130, KCl 5, CaCl2 2, MgCl2 2, HEPES-NaOH 10, and glucose 5, pH 7.4. The extracellular solution with high potassium (high Ko solution [K]0 = 40 mM) contained (in mM) NaCl 95, KCl 40, CaCl2 2, MgCl2 2, HEPES-NaOH 10, and glucose 5, pH 7.4. The standard pipette solution at [Ca2+] = 10–7 M contained (in mM) K+ Aspartate 130, NaCl 10, MgCl2 2, CaCl2 2, EGTA-KOH 10, and HEPES-KOH 10, pH 7.4. The pipette solution at low [Ca2+]i, necessary for recording EAG currents, contained (in mM) K+ Aspartate 130, NaCl 10, MgCl22, EGTA-KOH 10, HEPES-KOH 10, pH 7.4. For current-clamp experiments, the pipette solution contained (in mM) K+ Aspartate 140, NaCl 10, MgCl22, HEPES-KOH 10, and Amphotericin B 150 µg/ml, pH 7.3.

For a precise measurement of the current gating parameters, pipette, cell capacitance, and series resistance (up to 70%–80%) were carefully compensated before each voltage clamp protocol was run. The protocol used to measure the tail IHERG maximal current (IMAX) started from a holding of 0 mV and tested the current at –120 mV, after preconditioning from 0 to –70 mV for 15 seconds. The IKIR and IHERG currents were elicited at various Vm (from 0 to –140 mV), starting with a holding potential of 0 or –70 mV, according to Faravelli et al. [28] (currents were obtained at [K]o = 40 mM). For the activation of KDR currents, cells were preconditioned at –80 mV, and test potential ranged from –10 to +70 mV, with step increments of 10 mV; for the inactivation, cells were preconditioned from –80 to 0 mV, with step increments of 10 mV, and test potential was at +60 mV. For activation of tetrodotoxin (TTX)-sensitive INa, cells were preconditioned at –90 mV and currents were recorded from –40 to 30 mV with step increments of 10 mV. This protocol also reveals IKDR, without overlapping of the two currents, which activate quite distantly on the time scale. For the activation of EAG currents, cells were preconditioned at –60 and –120 mV, and currents were recorded at +60 mV.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The Sequential Morphological Changes of SY5Y Subclones in Culture
The SY5Y bulk cells are characterized by small and dense cell bodies emitting subtle neurites (Fig. 1AGo). Limiting dilution experiments of this population invariably produced two types of subclones: N-type (hence referred to as NN), fast-growing and morphologically identical to the parental cells (Fig. 1BGo); and S-type (hence referred to as S subclones), slow-growing and composed of multiform, strongly substrate-adherent cells (Fig. 1CGo).


Figure 1
View larger version (55K):
[in this window]
[in a new window]
  		Figure 1. Phase-contrast images of SY5Y clone and subclones and clonogenic efficiency of N and S clonogenic cells. (A): SY5Y cells, displaying the typical N-type morphology. (B): N-type subclone isolated from the SY5Y parental clone (NN cells). (C): S-type subclone isolated from the SY5Y parental clone and observed within 20 days after cloning (S0 cells). (D): N-type cells generated in the above S-type subclone, starting from approximately day 60 (NS cells); white arrows indicate five S-type cells closed to the NS cell island. (E): Substrate-adherent cells appearing in culture from day 60 in the same S-type subclone (S1 cells). (F): Substrate-adherent cells generated by the same S-type subclone from approximately day 120 (S2 cells). Bars = 100 µm. (G): The clonogenic efficiency (calculated as reported in Materials and Methods) is reported as a function of the dilution passages. Closed squares represent means ± standard errors of mean (n = 3) of the clonogenic efficiency of N cells; open squares represent the clonogenic efficiency of S cells undergoing the fibromuscular S1-S2 differentiation in culture.

 
The spontaneous development of S and NN subpopulations was followed up in long-term cultures, with no manipulation, except splitting at confluence. Whereas the NN subclones did not change substantially their morphology throughout the entire time course, the S subclones displayed a fairly constant pattern of changes that can be summarized as follows: soon after isolation, and up to day 60, cells appeared scattered and rather monomorphic, with rare small spikes protruding from the plasma membrane (Fig. 1CGo); because this represents the first progeny of single cells seeded, we called it S0. After day 60, cells gradually changed their shape, spreading out on the culture surface and becoming pleiomorphic (Fig. 1EGo); this latter aspect was assumed to identify an S1 subpopulation and remained substantially unchanged up to 120 to 140 days. Between 60 and 120 days, most S subclones produced N-type cells that gave rise to a fast-growing subpopulation (hence referred to as NS sub-clones; Fig. 1DGo), morphologically identical to the parental SY5Y and to NN cells. These NS cells need to be continuously removed to allow S1 cells to expand, giving rise to an S2 subpopulation (Fig. 1FGo), composed of larger cells. We do not have evidence to decide whether this inhibitory effect of N cell population on the S population is attributable to a direct cell-cell interaction or whether it is due to a prohibitive competition for nutrients or growing factors within the culture. We did not observe any evidence of apoptosis of S1 that appeared rarely dispersed but viable among the rapidly dividing NS cells.

The above S0-S1-S2 sequence was obtained with each of 12 tested S subclones, suggesting that a self-renewing, at least bipotential, compartment persists in the SY5Y bulk.

This issue was further explored by calculating the clonogenic efficiency of S0 and N cells (see Materials and Methods), obtaining the same value (approximately 55%) for both cell types at the first passage (Fig. 1GGo), which in turn enabled us to calculate the percentage of N and S0 clonogenic cells contained in the parental SY5Y clone. In this event, although the total clonogenic efficiency of this population was 30%, only 10% were S-type subclones. This leads us to estimate the percentages of clonogenic N and S0 cells in the SY5Y population to be 27% and 3%, respectively. Interestingly, the N clonogenic efficiency remained substantially constant over at least two successive passages, whereas for the S0 clonogenic cells we observed that in some cases (displaying signs of S1 differentiation), the efficiency dropped to 15% (Fig. 1GGo), whereas in others (with no sign of differentiation), it remained constant. Taken as a whole, these results suggest the following conclusions: the parental SY5Y population contains a large pool (approximately 27% of the entire population) of N clonogenic cells, endowed with an unlimited cell renewal; the S0 cells represent a minimal clonogenic pool (3% of the entire population), whose elements can either self-renew or undergo the S1-S2 differentiation. This finding is in keeping with the view that N cells represent the malignant abortive stage of neuronal components of the bulk tumor, whereas the S compartment is made of clonogenic cells susceptible to spontaneous terminal differentiation.

Following Up SY5Y Subclone Development with Patch-Clamp: Use of Ion Currents as Electrophysiological Clusters of Differentiation
Figure 2Go shows the typical currents recorded in the SY5Y clone and subclones by means of the patch-clamp protocols (in Figs. 2A7–2C7Go). In SY5Y cells, IHERG (Fig. 2A1Go) showed the same biophysical properties previously described by Arcangeli et al. [16], including its complete abolition by specific inhibitors (WAY123, 398 and E4031); the same holds true for IKDR (Fig. 2B1Go) (see details in Figs. 3A–3CGo) [16]. Figure 2C1Go shows the TTX-sensitive sodium currents (INa), similar to those recorded in S-type cells by D’Amico et al. [24]. Taken as a whole, the average SY5Y electrophysiological profile was characterized by the expression of substantial IHERG, a marked IKDR, and a small INa. At a lower scale, however, both NS and NN cells (Figs. 2A5–2C5 and 2A6–2C6Go, respectively) displayed a profile identical with that of SY5Y.


Figure 2
View larger version (25K):
[in this window]
[in a new window]
  		Figure 2. The electrophysiological clusters of differentiation (ECDs) identifying various developmental stages of SY5Y cells and subclones. The currents illustrated in (A–C) were measured in the same single cell by the whole cell configuration using the protocols reported in A7-C7. Currents are expressed in pA/pF (current density) and were specifically abolished by the following inhibitors, indicated in parentheses, with the respective concentrations: IHERG (2 µM WAY 123,398); IKDR (20 mM TEA); INa (2 µM tetrodotoxin).

 

Figure 3
View larger version (33K):
[in this window]
[in a new window]
  		Figure 3. Biophysical and pharmacological characterization of the K+ currents observed in the cells belonging to the various clones and excitable properties of some S1 and S2 clone cells. (A): Clones N, NN, and NS. IKDR currents elicited by the activation protocol shown in Figure 2B7 in control (A1) and during the perfusion of solutions containing either 200 nM BDS I (A2) or 5 mM TEA (A3). (B): Clones N, NN, and NS. Same as (A) except the inactivation protocol is shown in the inset to panel (B3). (C): The activation/inactivation voltage-dependent curves obtained from recordings were similar to those shown in (A) and (B). Data derive from experiments done in the following clones: N (open squares), S1 (open circles), and S2 (open triangles). The Boltzmann curves had the following values of V1/2 and slope (mV): N (V1/2 18.02 ± 0.7, slope 9.76 ± 0.5, n = 21), S1 (V1/2 25.2 ± 2.4, slope 16.7 ± 1.1, n = 19), and S2 (V1/2 33.8 ± 3.0, slope 19.6 ± 1.5, n = 6). (D): Clone S1. Traces of IKDR currents elicited by the activation protocol shown in Figure 2B7Go in control (D1) and during the perfusion of a solution containing 5 mM TEA (D2). (E): Clone S1. Traces of IKDR currents (EAG currents) elicited by the activation protocol shown in the inset. The line and symbol traces were elicited from holding potentials of –60 and –120 mV, respectively. (F): Clones S1 and S2. Excitable responses, recorded in current clamp, elicited with the protocol shown in the inset from a holding potential of –70 mV. [K]o = 4 mM. (G): Clone S2. Traces elicited by the protocols shown in the inset. Holding potentials and 100 µM Ba2+ were 0 mV and absent in G1, 0 mV and present in G2, –70 mV and present in G3. [K]0 = 40 mM.

 
With regard to the S subclones, the following information was obtained: S0 cells turned out to be practically devoid of currents, deserving the definition of "nude cells" (Figs. 2A2–2C2Go); as compared with SY5Y, S1 cells expressed much higher levels of IHERG and significantly higher INa but only negligible amounts of IKDR (Figs. 2A3–2C3Go; see details in Figs. 3D, 3EGo); S2 cells expressed low IHERG, substantial IKDR, and higher INa (Figs. 2A4–2C4Go; see details in Figs. 3F, 3GGo).

The simultaneous quantitative measurements of a cluster of currents in the same cell, as reported in Figure 2Go, emerged as a powerful tool to identify different cell types and successive stages in differentiation. We are indeed led to believe that each estimate of the cluster can be used to identify a specific cell phenotype, in the same way that clusters of differentiation (CD) antigens are widely used in cell biology and immunology [29, 30]. By analogy, we here propose to use the term electrophysiological cluster of differentiation# (ECD#) to describe the quantitative expression of a group of ion channels on the cell under study, where # is a signal identifying a determined stage of differentiation along a specific pathway.

Biophysical Characterization of IKDR Properties Observed in Different Clones and Properties of S1 and S2 Cells
Before looking closely at the changes observed in the various clones and subclones during development time (Fig. 4Go), it is important to further introduce and clarify in detail some pharmacological and biophysical properties of the currents not described formerly in column B of Figure 2Go under the general name IKDR.


Figure 4
View larger version (28K):
[in this window]
[in a new window]
  		Figure 4. Time course of ion current expression in cell populations derived from a single S0 cell. Cells derived from a single S0 cell were expanded in culture as reported in Materials and Methods, removing NS cells generated in the interval indicated by the horizontal bar. At indicated times, cells were replated in dishes to be used for patching within 24 hours. Currents were recorded in the whole cell configuration, using the protocols reported in Figures 2A7–2C7Go, and are expressed in pA/pF. Each symbol represents the mean ± standard error of mean of seven cells. Closed circles = IHERG; open circles = IKDR; open squares = INa. The dashed area represents the percentage of excitable cells, responding to injected depolarizing currents as reported by D’Amico et al. [23]. (B): The normalized electrophysiological clusters of differentiation distribution in various subclones obtained from single S0 cells. Cells, taken from various S subclones at different stages of their spontaneous development, were patched in whole cell configuration. The triad of currents (IKDR, INa, IHERG) was simultaneously recorded in each cell, and IKDR and INa were expressed as a function of IHERG. The patch clamp protocols were the same as in Figures 2A7–2C7Go. Symbols represent the mean ± standard error of mean of current values (expressed in pA/pF) recorded in at least 15 cells taken from 10 different subclones. Closed circles = IKDR; open circles = INa.

 
As shown in Figure 3CGo, IKDR recorded in the parental N, NS, and NN clones resulted in very similar voltage-dependent activation (open squares at right) and inactivation (open squares at left) curves [16]. Experimental data were fitted to Boltzmann curves, and the following V1/2 and slope (mV) were obtained: activation V1/2 18.02 ± 0.7, slope 9.76 ± 0.5 (n = 21); inactivation V1/2 –8.0 ± 1.2, slope 9.6 ± 0.6 (n = 16). The current recordings at various membrane potentials and in the presence of the blood depressing substance I (BDSI) (200 nM, blocker of some inactivating K+ currents) or tetraethylammonium (TEA) (5 mM) are shown in panels A1, A2, A3 and B1, B2, B3 for the activation and inactivation, respectively. Whereas BDS I was ineffective, we found a complete reversible block by TEA, suggesting that other types of K+ currents (TEA-insensitive) are not present. By contrast, the IKDR current density recorded in the S1 clone turned out to be smaller than in the N cells but in general formed by a mixture of two components. The first, illustrated in panel D1, is sensitive to TEA (panel D2) but almost free from the considerably slow inactivation observed in the various N clones (panel A). The voltage-dependent curve of S1 cells, shown in panel C (open circles), is characterized by a small, but not significant, right shift of the V1/2 midpoint and a less-pronounced slope (activation V1/2 25.2 ± 2.4, slope 16.7 ± 1.1, n = 19). The second component observed in the S1 clone is shown in panel E (protocol in the inset) and has the typical properties of the EAG current [31]. Indeed, the currents tested at +60 mV, when preconditioned by highly negative holding potentials (–120 mV), had a sigmoidal time course (open circles line) as opposed to the exponential time course observed when elicited from –60 mV. The exemplary cell shown in Figure 3EGo was chosen because it was almost devoid of the TEA-sensitive current shown in Figure 3DGo1.

An interesting property of some of the S1 and S2 cells (see also Fig. 4Go) is the capability of becoming excitable. This property is shown in Figure 3FGo by using current-clamp recordings. This capability can be explained by the combination of a relatively high density of Na+ voltage-dependent current and the low density of IKDR currents. This combination helps the regenerative development of a long AP, because IKDR currents are unable to repolarize the membrane potential and IHERG currents take over this function, as in the cardiac AP [24].

In the clone S2, we were able to observe the appearance of the inward rectifying K+ current (IKIR) [32]. Given that the IHERG current is also present, Figure 3GGo shows the IKIR biophysical and pharmacological characterization using the protocol shown in the inset. When elicited from a holding potential of 0 mV, we recorded the traces shown in panel G1; after the perfusion of 100 µM Ba2+ (a specific inward rectifier blocker), the resulting current biophysically resembled a typical HERG current (G2), which disappeared completely when elicited from a holding potential of –70 mV (G3). Indeed, at this potential, all HERG channels are completely closed and the test levels do not show appreciable outward currents in an inward rectifying channel. In the cell shown in Figure 3Go, the two currents coexisted, IKIR being much larger than IHERG and completely concealing it. The noninactivating IKDR currents (not shown) were small and characterized by a right shift of the voltage-dependent activation (Fig. 3CGo, open triangles; see other details in the legend).

These supplementary results, taken in their entirety, would in principle increase the variety of the ECDs identifying the NB subpopulations, but we decided, for reasons of simplicity, not to introduce novel variables in our cluster characterization.

Time Course of ECDs
The time course of ECDs in various populations generated in culture from one single S0 cell is shown in Figure 4AGo. Cells, appearing in culture within 0 to 60 days, maintained the nude ECDS0 profile; cells patched from 60 to 80 days progressively increased the expression of IHERG and INa while maintaining low IKDR; this produced the typical S1 profile (ECDS1 = high IHERG, negligible IKDR, relatively high INa), already evidenced in Figures 2A3–2C3. In the interval from 80 to 120 days, the IHERG and INa increase was followed by a marked decrease of both these currents, with no substantial change of IKDR. There-after, a progressive loss of IHERG occurred, with a substantial recovery of INa, a progressive increase of IKDR, and the expression of IKIR; these features characterized the S2 electrophysiological profile (ECDS2 = low IHERG, high IKDR, high INa). After 120 days, a sensible diminution of IKDR and INa took place, accompanying the exhaustion of the culture. Figure 4AGo also shows the time window within which S1 and S2 populations manifest their ability to elicit AP upon injection of depolarizing currents. A significant percentage (approximately 25%) of these types of cells was found only after the beginning of S1 cell generation, increasing up to 52% when the S2 cells prevailed in culture.

The generation of NS cells (see bar in Fig. 4AGo) started simultaneously with that of S1 cells, when the only cells present in culture were S0; when patched at days 20 to 30 after removal from the original S0 culture, NS cells displayed an electrophysiological profile (ECDNS) qualitatively identical to that of SY5Y (Figs. 2A1–2C1, 2A5–2C5Go). In general, although NS are morphologically indistinguishable from SY5Y cells (Figs. 1A, 1CGo), they appeared electrophysiologically immature compared with the bulk tumor cells.

The S0-S1-S2 developmental sequence was repeated with high consistency in various tested clones, although with shifts in the time scale from one experiment to another. We then normalized the current variations as a function of their ECDs, choosing IHERG as the independent variable. This choice was suggested by the fact that this current dominates the VREST of neuroblastoma cells throughout their fates; moreover, despite its huge variations in density at various stages of S cell development, IHERG maintained substantially unchanged in its biophysical parameters (activation and inactivation kinetics, time constants), which were similar to those previously described [24].

The normalized data (obtained from 10 separate experiments carried out as in Fig. 4AGo) are reported in Figure 4BGo, from which it will be seen that cells appear distributed as a function of ECDs, irrespective of their generation time. This plot allows the identification of three areas in which the following ECD kinetics are represented: the initial stages of the electrophysiological maturation of S0 cells starting from the "nude" ECDS0; the progressive achievement of ECDS1 by S1-committed cells; and the maturation of S2 cells up to their distinctive ECDS2. The limits of the areas can then be used, approximately, to assign single cells to one or the other of the S cell compartments as follows: S0 compartment = IHERG < 60 pA/pF, IKDR and INa< 15 pA/pF; S1 compartment = IHERG > 60 pA/pF, IKDR ≤ 15 pA/pF, INa > 20 pA/pF; S2 compartment = IHERG < 60 pA/pF, IKDR and INa > 15 pA/pF.

To sum up, the sequential expression of ECDs allows a step-by-step study of the fluctuations of the cells through the S0-S1-S2 and neuronal differentiation pathways originating from the S0 compartment.

S0 Cell Staminal Features and NC Differentiation Markers Expressed Along with the Electrophysiological Maturation of SY5Y Cells
The biological and oncological implications of the electrophysiological studies described above were explored by defining the correlations between tumor ECDs and immunocytochemical markers of normal NC development (Figs. 5Go–8GoGoGo). At first, we analyzed the immunocytochemical and self-renewal features of the S0 cells, identified above as the first cells appearing in culture and characterized by the nude profile ECDS0. These cells were 100% positive to vimentin and nestin (Figs. 5A, 5BGo) and 60% positive to p75 (Figs. 5C, 5DGo). These cells are therefore characterized by a common marker of substrate-adherent cells, vimentin [7]; a typical marker of NC immaturity, nestin, that persists in aggressive malignancy [33, 34]; and a sign of high rank of NCC staminality, p75 [25, 35], something that was never expressed elsewhere in any other cell type explored in this work.


Figure 5
View larger version (52K):
[in this window]
[in a new window]
  		Figure 5. Neural crest stem cell markers and self-renewal in S0 cells. (A, B): Cells were examined at approximately 30 days after subcloning from SY5Y clone. One hundred percent of cells resulted positive to vimentin and nestin. Bars = 100 µm. (C, D): Cells were stained with anti-p75 at day 30 after cloning. The percentage of p75-positive cells in this population was 40%. Bars = 50 µm. Cell nuclei were stained with Hoechst 33258. (E–H) Phase-contrast images of the cell population generated by a single S0 cell. (E): The cell founder identified by the coordinates on a CELLocate coverslip. (F–H): The cells generated by the cell founder at days 2, 6, and 12; the asterisk indicates the original position of the cell founder at the beginning of the experiment. Bars = 100 µm.

 

Figure 6
View larger version (19K):
[in this window]
[in a new window]
  		Figure 6. The immunocytochemical profile of S1 cells. Cells were examined at approximately 90 days after subcloning from SY5Y cells. One hundred percent of cells were positive to vimentin (A), nestin (B), and smooth muscle actin (C), but only some were positive to calponin (D). Cell nuclei were stained with Hoechst 33258. Bars = 100 µm.

 

Figure 7
View larger version (29K):
[in this window]
[in a new window]
  		Figure 7. Smooth muscle markers of S2 cells. Cells were examined at approximately 150 days after subcloning from SY5Y. S2 cells resulted in 100% positive to vimentin (A), negative to nestin (B), positive to smooth muscle actin (C), and positive to calponin (D). Cell nuclei were stained with Hoechst 33258, except for (A). Bars = 100 µm.

 

Figure 8
View larger version (20K):
[in this window]
[in a new window]
  		Figure 8. (A–C): The immunocytochemical profile of bulk SY5Y cells. Cells were harvested 2 days after seeding in six multiwells. Fifty percent of cell population turned out weakly positive to vimentin (A), whereas 100% of cells were strongly positive to nestin (B) and to NF-68 (C). (D–F): The immunocytochemical profile of NS cells. N-type cells were collected from one S0 progeny at approximately 55 days after subcloning; they resulted weakly positive to vimentin (90%) (D) and clearly positive to nestin (E) and NF-68 (F). The white arrows indicate S-type cells interspersed among the N-type. (G–I): The immunocytochemical profile of NN cells. SY5Y cells were subcloned and N-type single cells were isolated and followed up for 3 months. Immunocytochemistry was performed at approximately day 60 after cloning. Fifty percent of cells were weakly positive to vimentin (G), and 100% were positive to nestin (H) and NF-68 (I). Cell nuclei were stained with Hoechst 33258. Bars = 100 µm.

 
The staminal features of S0 cells were further investigated by analyzing their capacity for self-renewal as follows: having obtained the first S0 population by an SY5Y-limiting dilution, we performed a second round of cloning. After 24 hours, we limited our follow up to single, isolated cells, identifying them by the coordinates on a CELLocate coverslip (see an example in Fig. 5EGo). The progenies of these cells were followed up for 12 days, producing in 58% of cases colonies composed of approximately 20 cells. All of these cells maintained the S0 morphology (Figs. 5F–5HGo) and the ECDS0 nude profile. Taken as a whole, these results indicate that ECDS0 identifies a self-renewal compartment (S0 cells) within the SY5Y population.

The staminal nature of S0 cells was further supported by their ability to generate at least two types of NC derivatives (neuronal and smooth muscular) [15, 25, 35]. Indeed, the S1 cells were all positive to vimentin and nestin (Figs. 6A, 6BGo) but also positive to classic markers of the smooth muscle lineage, such as SMA and calponin [36] (Figs. 6C, 6DGo). This smooth muscular commitment progressed with the onset of S2 cells that turned out to be strongly positive to SMA and calponin, whereas they lost the nestin expression (Fig. 7Go). Recalling the ECDS1-ECDS2 development illustrated in Figures 2Go–4GoGo, one should conclude that this S1-S2 progression coincides with the concomitant IKDR increase and IHERG decline as well as with a frequent excitability. Moreover, as shown in Figure 3GGo, the S2 cells start expressing the typical muscle current IKIR [32].

A completely different immunocytochemical profile was displayed by the bulk N-type SY5Y cells (Figs. 8A–8CGo), which were generally negative to vimentin and positive to NF-68 and nestin. This profile argues for a neuronal commitment (NF-68 positivity) of cells that are scarcely able to adhere to the substrate (vimentin negative) [7] but maintain clear signs of immaturity and malignancy (nestin positivity) [33, 34]. A similar degree of neuronal commitment was displayed in the NS and NN phenotype (Fig. 8D–8IGo), positive to NF-68 and nestin.

Clonal Analysis of S0 and S1 Cells
To assess whether the above S0-S1-S2 progression was initiated by environmental factors or whether it was a cell-autonomous process, we performed a clonal analysis of S0 and S1 cells. In these experiments, we started from an S colony morphologically displaying an S0 phenotype, isolated from the SY5Y parental clone. This colony was chosen at the stage when it consisted of not more than 20 cells, some of which were used for the electrophysiological analysis. Only after all the patched cells displayed the nude ECDS0 profile were the remaining sister cells diluted and seeded as single cells. These founder cells generated colonies of 20 to 30 cells within approximately 23 days, endowed with the enlarged body and the ECDS1 profile typical of the S1 cells. In some cases, moreover, these cells were endowed with the capability of responding with APs to injected depolarizing currents. Single cells from this S1 population were further processed with limiting dilutions to obtain single founder S1 cells. The latter were allowed to develop spontaneously for 33 days, by which time they produced 20 to 30 cell colonies and were then tested for their electrophysiological profile. All of the cells of these new colonies displayed large dimensions, the ECDS2 profile and the IKIR, typical of S2 cells. Taken as a whole, this clonal analysis confirmed the results obtained in the long-term cultures and lends support to the conclusion that the S0-S1-S2 differentiation pathway is both spontaneous and independent of cell interactions, even though it is strongly accelerated. It is interesting to record here that in these experiments we did not observe any substantial production of NS cells, a fact that suggests that the generation of the latter is initiated by progenitors only after the S0 colonies expand beyond the 20 cell number.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this work we investigated the developmental repertoire of NB cells with the principal aim of identifying the cell compartment responsible for tumor renewal. The relevance of this task for modern oncology can hardly be overstated, especially in the case of NB, which is a product of altered progenies of the staminal NC compartment [8, 11, 15, 37].

A major technical novelty of this research is the use of patch clamp in an attempt to characterize the differentiative steps of NB elements emerging throughout the experimental time. The chosen method proved able to characterize single cells at various stages of differentiation by means of the simultaneous expression of grouped ion channels, for which we have proposed (see above) the denomination of ECD. This staging method has the following advantages: the high reliability afforded by the patch-clamp measurements; the simultaneous measurements of the currents cluster in the same cell; and the quantitative estimate of each component in the cluster, providing a new staging parameter based on the ratios among the expressed currents. This method proved to be decisive in disentangling the N and S cell maturation steps described in the present study and summarized in the model illustrated in Figure 9Go (see Figure legend and text below for explanation).


Figure 9
View larger version (53K):
[in this window]
[in a new window]
  		Figure 9. Model for SY5Y cell renewal and commitment. Cell populations are classified according to their electrophysiological cluster of differentiation and immunocytochemical profile. The parental SY5Y clone contains a minute pool of S0 cells, endowed with self-renewal and committed to both the smooth muscle and neuronal differentiation lineage. The bulk tumor also includes neuronally committed highly clonogenic cells (NN cells). S0 cells can, in turn, generate neuronal progenitors endowed with high clonogenic potential (NS cells), whose progeny becomes identical to SY5Y bulk cells. The S0 commitment to the smooth muscle lineage is allowed to proceed only if the NS progeny is removed. This removal, while permitting the cell progressive amplification and differentiation along this smooth muscle lineage (S1-S2 cells), leads to the culture exhaustion. This implies that N cell accumulation causes the block of the S0-S1-S2 lineage and the maintenance of the S0 self-renewal. Without N cell removal, the final fate of the overall population repeats the configuration of the SY5Y parental clone, in which rare S0 cells are renewed within the overwhelming N-type bulk. A hypothetical interconversion N-S0 is represented by the dashed arrows.

 
The choice of the cluster IHERG-IKDR-INa was suggested by our previous studies on NC electrophysiological maturation in vivo and in vitro, which was paced by an initial increase of IHERG, followed by a progressive loss of this current, substituted by IKDR [17, 19]. We here show that this cluster exhibits striking variations during the spontaneous development of NB in culture, and we present the remarkable finding that it begins with the "nude" profile displayed by the S0 cells (Fig. 2Go). This tiny cell population constitutes a self-renewing pool, occupying a high position in the hierarchical scale of NB staminal compartment.

Our claim is supported by three main facts. First, S0 cells are nestin and p75 positive. Nestin is considered a marker of neural progenitors in both the central and the peripheral nervous system [33, 38, 39]; p75, identified as the low-affinity nerve growth factor receptor [40–42], is regarded as a reliable marker of neural crest stem cells [33, 35]. We were able to confirm a previous report that the SY5Y bulk cells are p75 negative [43]; more importantly, we also showed that this negativity is shared by all the other cell populations individuated in this work, being lost before any restriction in the differentiation of NB cells occurs. Second, S0 cells maintained unvaried morphological and electrophysiological features throughout hundreds of passages in vitro of the SY5Y clone for at least four to five doublings (Fig. 5E–5HGo) and over a period of about 60 days (Fig. 4AGo), which implies that they are endowed with a high capacity for self-renewal. Third, S0 cells possess a high clonogenic power, generating a vast cell population that spontaneously differentiates into the two lineages so far identified (S1-S2 smooth muscle cells and NS cells). No GFAP-positive population emerged in this work, but it has to be said that we did not explore a glial-oriented pathway from S0 by adding specific growth and differentiation factors. The same consideration applies to other lineages, which still need to be explored to evaluate more exactly the differentiative potency of S0 cells. These factors were deliberately omitted in our experimental protocols because there was a danger of masking the natural fate of the tumor by forcing a commitment of the cells in predetermined directions.

The S0-S1-S2 progression occurred with a remarkably constant sequence in each subclone examined; its smooth muscular commitment was attested by its positivity to SMA and calponin (Figs. 6Go, 7Go) as well as by its frequent (50%) excitability and IKIR expression (Figs. 3F, 3GGo). The clonal analysis experiments showed that this progression was spontaneous and independent of the environment. It must be observed, however, that this pathway leads to the exhaustion of the culture by a progressively restricting differentiation. This fact, together with the loss of nestin positivity, indicates that S0 cells are fully committed to a nonmalignant phenotype [34]. Another intriguing feature of this pathway is that it can be accomplished only if NS cells, concomitantly generated in culture, are continuously removed. If they are not removed, NS cells produce an overwhelming population committed to the neuronal lineage (clearly positive to NF-68; Fig. 8FGo), with an ECDNS qualitatively identical to that of N cells of the SY5Y clone (Fig. 2Go). Like the latter, NS cells are only able to exploit a defective neuronal differentiation, whereas they are blocked at immature, nestin-positive malignant stages. However, it is not yet possible to state with certainty whether this neuronal-committed lineage originates directly from S0 cells or from some of their highly clonogenic derivatives. It should be noted, however, that the distinction between the neuronal and smooth muscle lineages is clear-cut because their morphological, electrophysiological, and immunocytochemical profiles diverge drastically. The bipotency of S0 raises the question of the place of these cells in the NC staminal hierarchy [15], but a definitive answer to this question could be given only after a study taking into account various growth and differentiation factors and lies beyond the limits of the present work.

The problem of the renewal of N cells within the SY5Y clone was further addressed by analyzing the electrophysiological and immunocytochemical maturation of NN and NS subclones (Figs. 2Go, 8Go). Such renewal started from NF-68–positive cells, which displayed a rather high IKDR/IHERG ratio and again produced a defective neuronally differentiated progeny with a deficient INa expression that accounts for their low excitability [20, 21]. This NN expansion implies the existence of N-committed, highly clonogenic cells and again raises the question of whether these cells derive from the S0 compartment.

The N generation from S0 cells might well represent an example of the S-N interconversion, which has been supported by reliable experiments from Biedler’s group [8, 44]. We cannot exclude the inverse N-S interconversion (also proposed by these authors), which would further complicate the population network described in this paper (Fig. 9Go). This possibility, however, would provide a clue to understanding a major issue that emerges from this study, namely the necessity of an overwhelming N cell population to maintain the S0 compartment (see above). In fact, the accumulation of N cells depends on the coexistence of two factors: the maintenance of an indefinite replenishment of the N bulk and the blockade of the S0-S1-S2 differentiative pathway. This last blockade emerges as a decisive factor in determining the recruitment of the entire staminal pool into a fate exhausting the self-renewal. This behavior is reminiscent of the fate of hemopoietic stem cells once they have been exported out of their physiological niche [45]. In summary, N cells seem necessary to maintain the conditions required for S0 renewal, but they also seem to monopolize the S0 reservoir at the expense of the S1-S2 direction.

As an alternative to the hypothesis of N-S interconversion, it might be possible to explain the necessity of N cells in maintaining the S0 compartment by drawing inferences from the fact that S0 phenotype is characterized by the expression of p75. The latter binds various neurotrophins (BDNF, NT-3/4/5, NGF) produced by neurons and NB cells [6], and these interact with the p75 protein to bring about effects varying from death signals to the modulation of mitogenic stimuli elicited by TrkA receptors [46–48]. It is conceivable that in the SY5Y bulk the dominant effect of this interaction is the protection of S0 cells from mitogenic stimuli that would recruit them in the S1-S2 pathway to exhaustion. On this hypothesis, the ordinary maintenance of the tumor and its amplification would be attributable to N clonogenic progenitors produced by S0 cells at a pace compatible with their self-renewal. If these assumptions prove to be correct, they would entail important consequences for the clinical treatment of cancer in that the elimination of the N cells by available therapies, both chemotherapies and radiotherapies, would be seen to have two distinct functions: eliminating the bulk of the tumor and helping to induce S0 compartment exhaustion in the S1-S2 pathway.

Finally, our study revealed the advantages of using the ECD description of complex cell populations based on an integrated analysis of electrophysiological and immunocytochemical studies. This procedure makes it possible to retrace the natural history of the individual cells that converge in such populations at any given time, and this provides an overall insight into the distinct stages of their maturation, which is an important parameter in developmental biology and a crucial oncological index.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We thank Dr. Lukas Sommer for his scientific suggestions and discussion, Marco Cutrì for his skilled technical support, and Professor Patrick Boyde for help in the revision of the manuscript. This work was supported by grants from the Associazi-one Italiana per la Ricerca sul Cancro (AIRC), Ministero dell’Istruzione, dell’Università e della Ricerca Scientifica e Tecnologica (MIUR, COFIN 2002–03), Associazione Italiana contro le Leucemie (AIL) Firenze, and Ente Cassa di Risparmio di Firenze (CARIFI).

DISCLOSURES
The authors indicate no potential conflicts of interest.


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Cheng KC, Loeb LA. Genomic instability and tumor progression: Mechanistic considerations. Adv Cancer Res 1993;60:121–156.[Medline]

  2. Reya T, Morrison SJ, Clarke MF et al. Stem cells, cancer, and cancer stem cells. Nature 2001;415:105–111.

  3. Pardal R, Clarke MF, Morrison SJ. Applying the principles of stem-cell biology to cancer. Nat Rev Cancer 2003;3:895–902.[CrossRef][Medline]

  4. Jones RJ, Matsui WH, Smith BD. Cancer stem cells: Are we missing the target? J Natl Cancer Inst 2004;96:583–585.[Free Full Text]

  5. Triche TJ. Neuroblastoma and other childhood neural tumors: A review. Pediatr Pathol 1990;10:175–193.[Medline]

  6. McConville CM, Forsyth J. Neuroblastoma: A developmental perspective. Cancer Lett 2003;197:3–9.[CrossRef][Medline]

  7. Ciccarone V, Spengler BA, Meyers MB et al. Phenotypic diversification in human neuroblastoma cells: Expression of distinct neural crest lineages. Cancer Res 1989;49:219–225.[Abstract/Free Full Text]

  8. Ross RA, Biedler JL, Spengler BA. A role for distinct cell types in determining malignancy in human neuroblastoma cell lines and tumors. Cancer Lett 2003;197:35–39.[CrossRef][Medline]

  9. Walton JD, Kattan DR, Thomas SK et al. Characteristics of stem cells from human neuroblastoma cell lines and in tumors. Neoplasia 2004;6: 838–845.[CrossRef][Medline]

  10. van Noesel MM, Versteeg R. Pediatric neuroblastomas: Genetic and epigenetic "danse macabre." Gene 2004;325:1–15.[CrossRef][Medline]

  11. Nakagawara A, Ohira M. Comprehensive genomics linking between neural development and cancer: Neuroblastoma as a model. Cancer Lett 2004;204:213–224.[CrossRef][Medline]

  12. Le Dourain NM. The Neural Crest. Cambridge: Cambridge University Press, 1982.

  13. Le Dourain NM, Kalcheim C. The Neural Crest, 2nd ed. New York: Cambridge University Press, 1999.

  14. Etchevers HC, Vincent C, Le Dourain NM et al. Development 2001;128: 1059–1068.[Abstract]

  15. Trentin A, Galvieux-Pardanaud C, Le Dourain NM et al. Self-renewal capacity is a widespread property of various types of neural crest precursor cells. Proc Natl Acad Sci U S A 2004;101:4495–4500.[Abstract/Free Full Text]

  16. Arcangeli A, Bianchi L, Becchetti A et al. A novel inward-rectifying K+ current with a cell-cycle dependence governs the resting potential of mammalian neuroblastoma cells. J Physiol 1995;489:455–471.[Medline]

  17. Arcangeli A, Rosati B, Cherubini A et al. Long-term exposure to retinoic acid induces the expression of IRK1 channels in HERG channel-endowed neuroblastoma cells. Biochem Biophys Res Commun 1998;244: 706–711.[CrossRef][Medline]

  18. Arcangeli A, Rosati B, Crociani O et al. Modulation of HERG current and herg gene expression during retinoic acid treatment of human neuroblastoma cells: Potentiating effects of Bdnf. J Neurobiol 1999;40:214–225.[CrossRef][Medline]

  19. Arcangeli A, Rosati B, Cherubini A et al. HERG- and IRK-like inward rectifier currents are sequentially expressed during neuronal development of neural crest cells and their derivatives. Eur J Neurosci 1997;9:2596–2604.[CrossRef][Medline]

  20. Toselli M, Tosetti P, Taglietti V. Functional changes in sodium conductance in the human neuroblastoma cell line SH-SY5Y during in vitro differentiation. J Neurophysiol 1996;76:3920–3927.[Abstract/Free Full Text]

  21. Tosetti P, Taglietti V, Toselli M. Functional changes in potassium conductances of the human neuroblastoma cell line SH-SY5Y during in vitro differentiation. J Neurophysiol 1998;79:648–658.[Abstract/Free Full Text]

  22. Moody WJ, Simoncini L, Coombs JL et al. Development of ion channels in early embryos. J Neurobiol 1991;22:674–684.[CrossRef][Medline]

  23. Moody WJ. Critical periods of early development created by the coordinate modulation of ion channel properties. Perspect Dev Neurobiol 1995;2:309–315.[Medline]

  24. D’Amico M, Biagiotti T, Fontana L et al. A HERG current sustains a cardiac-type action potential in neuroblastoma S cells. Biochem Biophys Res Commun 2003;302:101–108.[CrossRef][Medline]

  25. Shah NM, Groves AK, Anderson DJ. Alternative neural crest cell fates are instructively promoted by TGFbeta superfamily members. Cell 1996; 85:331–343.[CrossRef][Medline]

  26. Hamil OP, Marty A, Neher E et al. Improved patch-clamp techniques for high resolution current recording from cells and cell-free membrane patches. Pflugers Arch 1981;391:85–100.[CrossRef][Medline]

  27. Rae J, Cooper K, Gates E et al. Low access resistance perforated patch recordings using amphotericin B. J Neurosci Methods 1991;37:15–26.[CrossRef][Medline]

  28. Faravelli L, Arcangeli A, Olivotto M et al. A HERG-like K+ channel in rat F-11 DRG cell line: Pharmacological identification and biophysical characterization. J Physiol 1996;496:13–23.[Medline]

  29. Kenny AJ, O’Hare MJ, Gusterson BA. Cell-surface peptidases as modulators of growth and differentiation. Lancet 1989;2:785–787.[Medline]

  30. Lunney JK, Walker K, Goldman T et al. Overview of the First International Workshop to Define Swine Leukocyte Cluster of Differentiation (CD) Antigens. Vet Immunol Immunopathol 1994;43:193–206.[CrossRef][Medline]

  31. Meyer R, Heinemann SH. Characterization of an eag-like potassium channel in human neuroblastoma cells. J Physiol 1998;508:49–56.[Abstract/Free Full Text]

  32. Hille B. Ionic Channels of Excitable Membranes, 2nd ed. Sunderland. MA: Sinauer Associates, 1992.

  33. Stemple DL, Anderson DJ. Isolation of a stem cell for neurons and glia from the mammalian neural crest. Cell 1992;71:973–985.[CrossRef][Medline]

  34. Thomas SK, Messam CA, Spengler BA et al. Nestin is a potential mediator of malignancy in human neuroblastoma cells. J Biol Chem 2004;279:27994–27999.[Abstract/Free Full Text]

  35. Hagedorn L, Suter U, Sommer L. P0 and PMP22 mark a multipotent neural crest-derived cell type that displays community effects in response to TGF-beta family factors. Development 1999;126:3781–3794.[Abstract]

  36. Sugimoto T, Mine H, Horii Y et al. Neuroblastoma cell lines showing smooth muscle cell phenotypes. Diagn Mol Pathol 2000;9:221–228.[CrossRef][Medline]

  37. Nakagawara A. Neural crest development and neuroblastoma: The genetic and biological link. Prog Brain Res 2004;146:233–242.[Medline]

  38. Okano H. Stem cell biology of the central nervous system. J Neurosci Res 2002;69:698–707.[CrossRef][Medline]

  39. Taupin P, Gage FH. Adult neurogenesis and neural stem cells of the central nervous system in mammals. J Neurosci Res 2002;69:745–749.[CrossRef][Medline]

  40. Chao MV, Bothwell MA, Ross AH et al. Gene transfer and molecular cloning of the human NGF receptor. Science 1986;232:518–521.[Abstract/Free Full Text]

  41. Radeke MJ, Misko TP, Hsu C et al. Gene transfer and molecular cloning of the rat nerve growth factor receptor. Nature 1987;325:593–597.[CrossRef][Medline]

  42. Bothwell M. p75NTR: A receptor after all. Science 1996;272:506–507.[Medline]

  43. Eggert A, Sieverts H, Ikegaki N et al. p75 mediated apoptosis in neuroblastoma cells is inhibited by expression of TrkA. Med Pediatr Oncol 2000;35:573–576.[CrossRef][Medline]

  44. Rettig WJ, Spengler BA, Chesa PG et al. Coordinate changes in neuronal phenotype and surface antigen expression in human neuroblastoma cells variants. Cancer Res 1987;47:1383–1389.[Abstract/Free Full Text]

  45. Cipolleschi MG, Dello Sbarba P, Olivotto M. The role of hypoxia in the maintenance of hematopoietic stem cells. Blood 1993;82:2031–2037.[Abstract/Free Full Text]

  46. Sommer L, Rao M. Neural stem cells and regulation of cell number. Prog Neurobiol 2002;66:1–18.[CrossRef][Medline]