Fibroblast Growth Factor 2 Promotes the Self-Renewal of Bipotent Glial Smooth Muscle Neural Crest Progenitors

Abstract

The neural crest (NC) is an attractive system for investigating the mechanisms underlying cell lineage diversification in higher vertebrates. The NC contains a mixed population of already defined precursors and multipotent cells that can give rise to a great variety of cell types, including glial cells and neurons of the peripheral nervous system, melanocytes, and smooth muscle cells (SMCs). Microenvironmental factors, such as the fibroblast growth factor 2 (FGF2), found along migratory paths and in target tissues, strongly influence the fate of multipotent NC precursors. We have previously demonstrated that the FGF2 promotes the differentiation of NC cells to glial phenotypes, while the epidermal growth factor induces NC differentiation to neurons and melanocytes. In the present study, we used mass cultures and single-cell culture assays to demonstrate that FGF2 influences NC cell differentiation and increases the proportion of multipotent progenitors. Furthermore, we demonstrate for the first time that avian tripotent glial, melanocyte and smooth muscle NC progenitors, as well as bipotent melanocyte and smooth muscle NC progenitors, are capable of self-renewal. FGF2 significantly stimulated the self-renewal of bipotent progenitor cells with glial cells and SMC potentials. These cells propagated for many generations and behaved as stem cells. These results suggest an important role of FGF2 in maintaining the stemness of avian NC cells.

Introduction

T he neural crest (NC) is a multipotent cell population in vertebrate embryos that originates during neurulation and gives rise to neural melanocytic derivatives (neurons and glial cells of the peripheral nervous system and skin melanocytes) and mesenchymal cell types, such as chondrocytes, osteocytes, myofibroblast/smooth muscle cells (SMCs), and adipocytes, all of which contribute to the head structure in amniotes [1,6,8,9]. Highly multipotent NC progenitors endowed with both mesenchymal and neural melanocytic potentials have been described in vitro in both the cephalic and the trunk NC [2 –4,35].

The self-renewal capacity of some NC cells has been demonstrated in vitro, indicating that they are true stem cells. Stemple and Anderson (1992) described rat trunk oligopotent NC progenitors that could be propagated in culture and were capable of giving rise to neurons, glial cells, and SMCs [34]. In addition, the self-renewal capacity of quail trunk glial melanocyte and glial smooth muscle bipotent NC progenitors has been demonstrated [35]. However, the self-renewal of NC cells and the factors supporting their propagation have not yet been fully described.

The fibroblast growth factor (FGF) signaling pathway controls multiple aspects of the nervous system development, including neural progenitor survival, proliferation, maintenance, and differentiation. It is also involved in tissue patterning and compartmentalization [19]. FGF2 has a role in both neurogenesis and gliogenesis in the central nervous system [29]. Furthermore, it regulates NC cell proliferation [11,12] and migration [17], in addition to skeletogenesis [20,31], chondrogenesis [13,14,23,28], and gliogenesis [12,14,27]. However, the involvement of FGF2 in NC multipotentiality is still unclear.

Therefore, in the present study, we investigated the role of FGF2 in the multipotentiality and self-renewal of quail trunk NC cells. FGF2 affected NC cell differentiation and increased the proportion of multipotent progenitors. Importantly, FGF2 greatly improved the self-renewal capacity of bipotent glial smooth muscle (GF) progenitors, such that these cells could propagate for many generations.

Materials and Methods

NC cell cultures

Neural tubes from quail embryos (20–25 somite stage) were dissected at the trunk level and plated in uncoated plastic culture dishes (Corning) in a basal medium: the α-minimum essential medium (α-MEM; Invitrogen), enriched with 10% fetal calf serum (FCS; Vitrocell), 2% chicken embryonic extract, penicillin (200 U/mL; Invitrogen), and streptomycin (10 μg/mL; Invitrogen). After 24 h, NC cells that migrated from the neural tubes were harvested for secondary plating as mass cultures (400 cells/well of a 96-well plate) on rat-tail collagen-coated surfaces (Sigma). Cultures were maintained for an additional 6 days in the basal medium with or without FGF2 (1, 10 or 20 ng/mL; Sigma). In some experiments, after a 6-day treatment with 10 ng/mL FGF2, cultures were maintained for an additional 4 days in a complex medium: the basal medium supplemented with hydrocortisone (0.1 μg/mL), transferrin (10 μg/mL), insulin (1 ng/mL), 3-3′-5 triiodothyronine (T₃) (0.4 ng/mL), glucagon (0.01 ng/mL), epidermal growth factor (EGF) (0.1 ng/mL), and FGF2 (0.2 ng/mL) (all from Sigma) [35].

Clonal cultures were performed by micromanipulation as described elsewhere [6,21]. Briefly, individual cells were obtained after 24 h from primary cultures and were seeded under microscopic control in 96-well culture plates coated with rat-tail collagen. Clonal cultures were maintained for 6 days in the basal medium alone with or without 10 ng/mL FGF2 supplementation, followed by an additional 4 days of culture in a complex medium. Cultures were incubated at 37°C in a humidified 5% CO₂ atmosphere. The medium was replaced every 3 days.

Subcloning procedure

After 6 days of culture on rat-tail collagen-coated wells, primary colonies (I) derived from founder NC cells were evaluated for their total number of cells under microscopic control. Some of the clones I containing 50–1,000 cells were detached using trypsin/EDTA (Invitrogen). For each colony, a 48 cell sample of the resulting suspension was used for subcloning. The cells remaining in the suspension were plated for overnight subculture, and then phenotypically analyzed with lineage-specific markers by immunofluorescence staining as described below. This method permitted to retrospectively identify the type of the primary colonies subjected to further cloning and therefore, the potentials of the initial founder cells. Such procedure was repeated at each subsequent subcloning and led us to follow the progeny of identified founder cells along successive rounds of subcloning (I to IV).

Immunofluorescence staining

Cell phenotypes were analyzed using lineage-specific markers. Melanocytes were recognized by staining with a melanoblast/melanocyte early marker (MelEM) monoclonal antibody (mAb) [24]. Glial cells were identified by staining with a Schwann cell myelin protein (SMP) mAb [7]. SMCs were identified by staining with an α-smooth muscle actin (α-SMA) mAb (Sigma). Neurons were identified by staining with a tyrosine hydroxylase (TH) mAb [10] and a βIII-Tubulin (βIII-Tub) mAb (Promega). Secondary antibodies were obtained from Southern Biotechnology Associates. Cell nuclei were stained with 4′, 6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma). Detailed procedures are described elsewhere [5,35]. Fluorescence labeling was observed using an epifluorescent microscope (Olympus IX71).

5-bromo-2′-deoxyuridine incorporation and detection

Cell proliferation was analyzed by 5-bromo-2′-deoxyuridine (BrdU; Invitrogen) incorporation with some modifications [5]. Briefly, NC cells were incubated with BrdU (2 h), fixed (4% paraformaldehyde; Sigma), permeabilized (0.25% Triton X-100; Sigma), washed (distilled water), and incubated with 2N HCl (15 min, 37°C). After washing in PBS, cultures were immunostained with an anti-BrdU mAb (Millipore) and a rabbit anti-IgG-FITC Ab (Southern Biotechnology). Cell nuclei were stained with DAPI and visualized as described above.

Cell death assay

Cell death was quantified by Sytox® green staining according to the manufacturer's instructions (Molecular Probes).

Statistical analysis

Significant differences were evaluated by one-way ANOVA with the Tukey's post-test or by the χ ² test. The level of significance was set at P<0.05.

Results

FGF2 affects NC cell differentiation, proliferation, and survival

FGF2 treatment during the 24 h of quail trunk NC primary culture (when NC cells migrated out of neural tube) was recently demonstrated to result in increased Schwann cell differentiation [12]. Therefore, we investigated the phenotype of the NC cell population during 6 days of secondary culture with FGF2 treatment (after the NC migration period). Fig. 1 shows representative pictures (Fig. 1A–D) and quantifications (Fig. 1E) of the cell types identified in the cultures. In the absence of FGF2 treatment, the cell population was composed of 62% glial cells, 18% melanocytes, 7% SMCs, and 1.5% neurons. Cells negative for these phenotypic markers (unmarked cells) corresponded to 11.7% of the cells (Fig. 1E). Treatment with 10 ng/mL FGF2 reduced the amount of glial cells (6-fold), melanocytes (3-fold), and SMCs (2.7-fold) and increased the number of unmarked cells (6-fold) compared to untreated controls (Fig. 1E). No differences in the number of neurons were observed. Similar results were obtained when FGF2 treatment was performed during both primary and secondary cultures (Supplementary Fig. S1; Supplementary Data are available online at www.liebertpub.com/scd). Although FGF2 treatment during only the initial period of culture stimulated glial cell differentiation of NC cells [12], extended and continuous treatment prevented differentiation to the main NC-derived cell types listed above.

FIG. 1.

Fibroblast growth factor 2 (FGF2) treatment influences NC cell differentiation (A–D, E), proliferation (G), and death (G) in mass cultures. FGF2 treatments were performed during the 6 days of secondary culture. Cell phenotypes were identified by the presence of specific cell markers: (A) SMP for glial cells, (B) MelEM for melanocytes/melanoblasts, (C) α-smooth muscle actin (αSMA) for smooth muscle cells (SMCs), and (D) βIII-Tub for neurons. (E) Quantification of each cell phenotype and the total number of cells. (F) Cell proliferation was assessed by 5-bromo-2′-deoxyuridine (BrdU) incorporation, and (G) cell death was assessed by SYTOX^® green staining. Data are represented as the proportion of the total number of cells stained by DAPI. Values were obtained from analysis of 60 random fields from 3 independent experiments. The results are expressed as the mean±S.E.M. Significant differences are indicated by *P<0.05 and ***P<0.001 versus the respective control (0 ng/mL FGF2), as determined by one-way ANOVA with Tukey post hoc test. Scale bar=20 μm.

The decrease in differentiated cell types was accompanied by a huge increase in the NC cell population (Fig. 1E). Therefore, we performed BrdU incorporation (Fig. 1F) and Sytox® green labeling (Fig. 1G) experiments to assess cell proliferation and death, respectively. FGF2 treatment, at all analyzed concentrations, resulted in increased proliferation of NC cells (1.2-fold at 10 ng/mL, maximal effect) (Fig. 1F). There were relatively few dead cells in our cultures (less than 2%); however, FGF2 treatment significantly decreased the proportion of Sytox® green-stained cells (9-fold at 10 ng/mL, maximal effect), indicating a positive effect on NC cell survival (Fig. 1G). Although, specific assays for cell survival would be needed, it is important to consider that even reduced values for overall cell death (when evaluated at specific time points) can contribute to much larger effects after 6 days of culture, specifically, in terms of the expansion of the NC cell population observed in our experiments.

NC cells differentiate after FGF2 removal

Next, we investigated if the unmarked cells reported in Fig. 1E were able to differentiate into the main NC-derived phenotypes after FGF2 was withdrawn. To do so, we followed the 6 days of FGF2 treatment with an additional 4-day culture period in a differentiation medium (complex medium) (Table 1). Consequently, the number of unmarked cells was significantly reduced (25.5-fold), whereas the proportions of glial cells and neurons were significantly increased (62-fold and 6.5-fold, respectively). No alterations were detected in the amount of melanocytes and SMCs. Therefore, FGF2 removal and replacement by the complex medium stimulates the expression of all of the main NC-derived phenotypes, especially glial cells and neurons, suggesting an effect of FGF2 on differentiation.

Table 1.

Phenotypic Analysis of Neural Crest Cells After Fibroblast Growth Factor 2 Removal

	Percentage of cells
Cell phenotype	FGF2	Complex medium
Glial cells	1.3±0.6	80.7±8^a
Melanocytes	2.8±0.5	2.2±1
SMCs	0.4±0.2	3.7±2
Neurons	1.5±1.5	9.7±2^a
Unmarked	94.5±1.2	3.7±0.6^a

Mass cultures were maintained in the presence of 10 ng/mL FGF2 until day 6 of secondary culture (left column) and then in complex media until day 10 (right column). Cell phenotypes were identified by the presence of specific cell markers: SMP for glial cells, MelEM for melanocytes/melanoblasts, αSMA for SMCs, and βIII-Tub for neurons. Values were obtained from the analysis of 40 random fields of 3 independent experiments. The results are expressed as the mean±S.E.M. Significant differences are indicated by ^a P<0.001 as determined by t test versus FGF2.

FGF2, fibroblast growth factor 2; SMCs, smooth muscle cells; αSMA, α-smooth muscle actin.

FGF2 influences NC progenitors in single-cell cultures

To verify whether FGF2 affects the developmental potential of NC progenitors, cell-cloning experiments were performed and the progeny of individual NC cells were analyzed. Clonal cultures of NC cells were treated with 10 ng/mL FGF2 (6 days), followed by growth factor removal and replacement by the complex medium for an additional 3 days (Supplementary Fig. S2). Cloning efficiency, given by the total number of clonogenic cells in relation to the total number of plated cells was not significantly altered by FGF2 (Fig. 2F), indicating that FGF2 did not influence the survival of NC progenitors. In addition, NC colonies were heterogeneous with respect to their size (1 to 35,000 cells). Colonies were classified as small, medium, and large according to the total number of cells. Similar distribution of the colony size groups was detected for both conditions (Supplementary Table S1).

FIG. 2.

Influence of FGF2 treatment on trunk NC clonal cultures. Clones were maintained in the presence of 10 ng/mL FGF2 until day 6 of secondary culture and then in complex media until day 10. Control and FGF2-treated clonal cultures of NC cells were analyzed by immunofluorescence to detect glial cells, melanocytes, SMCs and neurons. (A–D) Representative epifluorescence micrographs of a typical multipotent glial–neuron–melanocyte–smooth muscle (GMNF) colony from an FGF2-treated culture showing (A) SMP⁺ glial cells, (B) βIII-Tub⁺ neurons, (C) MelEM⁺ melanocytic cells, and (D) αSMA⁺ SMCs. Scale bar=20 μm. (E) Scheme for the various types of progenitors found in both the control and the FGF2-treated conditions, classified according to the combination of cell types in their progeny (G, glial cells; N, neurons; M, melanocytes; and F, SMCs). The frequency of each progenitor type (percentage of the total clone number) is shown for both medium conditions. Multipotent clones (GMNF+GMF+GNF+GMN) are represented in gray. (F) Quantification of clones containing each cell phenotype expressed as mean percentages (data from 319 control and 357 FGF2-treated colonies). Cloning efficiency was calculated for each condition by dividing the total number of clones or the sum of multipotent clones by the total number of plated cells (n=718). Values were obtained from 15 independent experiments. Statistically significant differences are indicated by *P<0.5, **P<0.01, and *** P<0.001 as determined by χ² test.

Colonies were next analyzed for the presence of glial cells (G), neurons (N), melanocytes (M), and SMCs (F), which led us to identify twelve types of colonies: unipotent colonies containing only one phenotype, other than neurons (G, M, and F clones), bipotent colonies containing combinations of 2 phenotypes (GM, GF, GN, and MF, the only colony type that did not contain glial cells), and multipotent colonies with several combinations of cell phenotypes (GMF, GNF, GMN, and GMNF). A subset of unmarked colonies was found (U). Figure 2E shows the distribution of these progenitors, and Fig. 2A–D displays representative pictures from one highly multipotent GMNF colony.

FGF2 treatment significantly increased the overall frequency of colonies containing SMCs and neurons (Fig. 2F). Surprisingly, the proportions of glial cell- or melanocyte-containing colonies were not altered. Moreover, FGF2 significantly increased (2.6-fold) the percentage of multipotent progenitors (i.e., GMF+GNF+GMN+GMNF), especially the multipotent GMNF (Fig. 2A–D) and the tripotent GMF (Fig. 2E) cells. Furthermore, FGF2 treatment promoted a 2.2-fold increase in the cloning efficiency of multipotent progenitors (i.e., GMF+GNF+GMN+GMNF). Therefore, FGF2 seems to specifically stimulate the survival of NC multipotent progenitors instead of the entire NC cell population.

Multipotent colonies (i.e., GMF+GNF+GMN+GMNF) contained ≈50 to 35,000 cells. In the control condition, 38% of colonies were of large size (i.e., ≈1,000 to 15,000 cells, Table 2). FGF2 treatment significantly increased (1.5-fold) the proportion of large, multipotent colonies containing ≈1,000 to 35,000 cells. These results suggest that the mitogenic and/or survival effects of FGF2 could be specific to multipotent NC progenitors rather than the entire NC cell population.

Table 2.

Analysis of Multipotent Colony Size

	Percentage of colonies
Colony size (Cell/colony)	Control	FGF2
S (1–100)	5.9	3.57
M (101–1000)	55.9	39.3
L (>1000)	38.2	57.1^a

Multipotent colonies (GMNF+GMF+GNF+GMN) were classified as small (S), medium (M), and large (L) according to the total number of cells. The frequency of each colony type (percentage of the total multipotent colony number) is shown for both medium conditions (data from n=34 and n=84 for control and FGF2-treated cultures, respectively). Values were obtained from 15 independent experiments. Significant differences are indicated by ^a P<0.001 as determined by χ² test.

Propagation of NC precursors by serial subcloning

We next investigated the possible action of FGF2 in the self-renewal of NC progenitors by serially subcloning the NC clonal cultures (Materials and Methods, Supplementary Fig. S2 and Supplementary Table S2). Briefly, at day 6 of FGF2 treatment (or day 6 of control treatment), some colonies were harvested and replated under the same culture conditions for an additional 6 days, after which time the procedure was repeated. The remaining clonal colonies were maintained in the complex medium to permit cell differentiation and retrospectively identify the type of clone subject to further cloning.

Figure 3 shows the progeny generated at successive rounds of subcloning. Subcloning of the primary progeny of the NC cells (cloning I, shown in Fig. 2) yielded the cells in cloning II. Cloning II was subcloned to yield cloning III, and after that, another subcloning round produced cloning IV. The cloning efficiency (Fig. 3A) and the number of cells per clone decreased as subcloning proceeded and were similar for the control and the FGF2-treated conditions. Although a higher cloning efficiency was recorded at cloning IV, this could be explained by the marked reduction in the number of plated cells rather than by increased cell viability. Cloning II yielded colonies ranging from ≈50 to 2,000 cells, and cloning III produced colonies with ≈5 to 1,000 cells, whereas the size of the colonies in cloning IV decreased to ≈1 to 100 cells.

FIG. 3.

FGF2 affects the propagation of trunk NC progenitors in subcloning experiments. (A) The types of colonies derived from NC cells through serial subcloning were analyzed and quantified (as the percentage of the total number of clones) during successive rounds II to IV of subcloning, as summarized in the scheme. The frequency of each progenitor type is indicated (percentage of total clones in each subcloning). Clonal efficiency was calculated as the percentage of total clones from plated cells in each subcloning round (n=1536 at subcloning II, n=3072 at cloning III, and n=96 at cloning IV for both control and FGF2-treated conditions). (B) Quantification of clones containing glial cells (top panel), SMCs (middle panel), and melanocytes (bottom panel) expressed as mean percentages during clonings I to IV. The number of colonies used for subcloning was 29 at cloning I, 16 at cloning II, and 2 at cloning III in control medium. In the FGF2-treated condition, the number of colonies used for subcloning was 27 at cloning I, 20 at cloning II, and 4 at cloning III. Values were obtained from 8 independent experiments. Statistically significant differences between 2 subsequent clonings were tested by χ ² tests and are indicated by *P<0.05, **P<0.01, and ***P<0.001.

Phenotypic analysis indicated that FGF2 significantly increased the overall frequency of colonies containing glial cells during the subcloning procedure (Fig. 3B top panel). In contrast, the overall frequency of melanocyte-containing colonies was progressively reduced during subsequent rounds (I to III) of subcloning until there were no more found at round IV (Fig. 3B, middle panel). FGF2 increased the percentage of SMC-containing colonies until round III of subcloning, with a significant reduction at round IV (Fig. 3B, bottom panel). In contrast, neuron-containing colonies were not observed during these procedures, possibly due to the very strict culture conditions required for differentiation to this cell type.

Six different types of precursors were obtained in cloning II, including the tripotent GMF, the bipotent MF and GF, and the restricted G, M, and F progenitors (Fig. 3A top panel). FGF2 treatment increased the number of G progenitors, accompanied by a reduction in M progenitors. Although not statistically significant, the growth factor reduced in 2.1-fold and 2.8-fold the proportions of GMF and GM progenitors, respectively, and improved in 1.6-fold the amount of GF cells (Fig. 3A top panel).

In the control condition, only restricted G-, M-, and F-producing cells were found at cloning III, and only F progenitors were found at cloning IV (Fig. 3A middle and bottom panels, respectively). In contrast, significant proportions of GF progenitors, in addition to restricted cells, were found after FGF2 treatment in these subcloning rounds (Fig. 3A middle and bottom panels). Moreover, a subset of U clones was observed in both the control and the FGF2-treated conditions. In cloning II, the proportion of these U clones, was increased in 2.2-fold after FGF2 treatment (Fig. 3A top panel).

Therefore, FGF2 stimulates the propagation of glial- and smooth muscle-endowed precursors, especially the bipotent GF cell type.

FGF2 stimulates the self-renewal of bipotent GF progenitors

The propagation of progenitors during the 4 successive rounds of subcloning prompted us to investigate if FGF2 promotes self-renewal. Thus, the progeny of each founder cell was analyzed and the self-renewal was evidenced by progeny that included the parental phenotype in addition to differentiated cells, for example, a colony containing G, M, and F was considered to derive from a GMF founder cell and the presence of the GMF colony in the subsequent subcloning procedure indicated the self-renewal of this progenitor. Such procedure was repeated at each subsequent subcloning and led us to follow the progeny of identified GF, GM, MF, and GMF founder cells along successive rounds of subcloning (I to IV). The number of self-renewing cell divisions was estimated by the quantification of progenitors identical to the founder in each round of the subcloning procedure.

Quantitative analysis of self-renewing GMF, GF, MF, and GM cells is shown in Fig. 4. These data are detailed in Supplementary Table S2. From the 7 GMF colonies that were subcloned, 2 produced parental-like colonies in the second generation, indicating that GMF founders were able to self-renew. Importantly, one of the GMF founders gave rise to 2 GMF progenitors (Supplementary Table S2, clone I.4), suggesting that (i) one symmetrical division of the GMF founder produced the 2 identical GMF progenitors or (ii) 2 asymmetrical divisions, one by the GMF founder and another by the GMF cell produced from the founder, occurred. Detailed analysis of GMF self-renewing cells revealed that they gave rise to bipotent MF progenitors in addition to restricted M and F progeny, indicating that they must have undergone at least one asymmetric cell division. Furthermore, subcloning of GMF progeny from subcloning II gave rise to only restricted G, F, and U progeny, suggesting the occurrence of differentiating divisions (Fig. 4A and Supplementary Table S2, clones II.5 and II.4, respectively).

FIG. 4.

FGF2 promotes the self-renewal of GF progenitors. GMF, GF, MF, and GM primary progeny (I) were serially subcloned in (A) control and (B) FGF2-supplemented medium (10 ng/mL), and their progeny were analyzed after each round of subcloning (I to IV). The different types of progeny (and mean percentage of total clones) are indicated as well as the fold increase in the number of GMF, GF, and MF cells produced by individual founders during self-renewal (bold arrows). (A) The subcloned founders in the control condition were as follows: 7 GMF, 6 GF, 3 GM, and 2 MF clones at cloning I, and 2 GMF and 2 GF clones at cloning II. (B) In FGF2-treated cultures, the subcloned founders were as follows: twelve GMF, ten GF, one GM, and one MF clones at cloning I; 1 GMF and 19 GF clones at cloning II; and 4 GF clones at subcloning III (for details, see Supplementary Table S2).

In the first round of the control cloning experiments, 6 GF colonies were generated from subcloning GF founders (Fig. 4A and Supplementary Table S2, clones I.8 to I.13). In 2 cases, all progenitors found in the progeny from cloning II were GF, indicating at least 3 symmetrically proliferative divisions between cloning I and II (Supplementary Table S2, clones I.9 and I.12). In 3 cases, in addition to the GF cells, GF self-renewing founders also produced restricted G, F, k, and U progeny, indicating that they had divided asymmetrically at least one time (Fig. 4A and Supplementary Table S2, clones I.10, I.11, and I.13). In another case, only restricted G and F cells were found in the progeny from cloning II, showing that GF cells can also undergo divisions that lead only to restricted differentiated phenotypes (Supplementary Table S2, clone I.8). Additional subcloning of 2 GF colonies derived from progeny II gave rise only to differentiated G and F cells (Fig. 4A and Supplementary Table S2, clones II.9 to II.12).

In the control condition, 2 subcloned MF founder cells produced 4 MF progenitors in the progeny from cloning II, in addition to restricted M and F cells, indicating that at least one asymmetrical division had occurred (Fig. 4A and Supplementary Table S2, clone I.14). In another case, only F cells were observed (Fig. 4A and Supplementary Table S2, clone I.15). In contrast, 3 GM colonies were subcloned, producing only fate-committed G and M phenotypes, indicating that they were not able to self-renew in this condition (Fig. 4A and Supplementary Table S2, clones I.16 to I.18).

In FGF2-treated cultures, twelve GMF colonies were subcloned and only 2 GMF founders were able to self-renew in the second generation (cloning II). Bipotent GF progenitors and restricted cells (F and U) were concomitantly detected, indicating that GMF cells had divided asymmetrically at least one time between cloning I and II (Fig. 4B and Supplementary Table S3, clones I.5 and I.11). The remaining 10 GMF founders were not able to self-renew, but produced GF and MF cells in addition to restricted G, M, F, and U cells, indicating the occurrence of differentiating divisions (Fig. 4B and Supplementary Table S3, I.1 to I.4, I.6 to I.10 and I.12). Further, subcloning of GMF progeny II produced no parental-like progeny in round III of subcloning, but instead produced GF precursors (Fig. 4B and Supplementary Table S3, clone II.5). Importantly, GF cells originating from the GMF founders in cloning II and III could self-renew until round IV of subcloning by both symmetrical (Supplementary Table S3, clones III.5 and III16) and asymmetrical divisions (Supplementary Table S3, clone III.7). GF progenitors were greatly amplified in the GMF progeny (Fig. 4B and Supplementary Table S3, clones II.1 to II.12), indicating that GMF, GF, or both founders had self-renewed between cloning I and II, before restricting their lineage to GF cells. Subcloning of GMF primary progeny at earlier cell cycles (i.e., at earlier culture time) would be needed to ascertain the possibility that GMF founders self-renew before becoming restricted to GF bipotent progenitors.

In contrast to the GMF progenitors, GF cells displayed high self-renewal ability after exposure to FGF2. Ten GF colonies were subcloned (Fig. 4B and Supplementary Table S3, I.13 to I.22), and in 7 cases, GF progenitors were able to self-renew, yielding restricted G, F, and U cells and indicating that at least one asymmetrical division occurred between cloning I and II (Fig. 4B and Supplementary Table S3, clones I.13, I.15 to I.20). In 3 cases, GF founders produced only restricted F cells (Fig. 4B and Supplementary Table S3, clones I.14, I.21-I.22). Seven GF clones from the progeny of cloning II were subcloned, and in 4 cases, they gave rise to more GF progenitors (Fig. 4B and Supplementary Table S3, clones II.13, II.16-II.18). As restricted F and U cells were also produced, the self-renewing GF progeny must have undergone at least one asymmetrical division between cloning II and III. The 3 remaining GF progenitors produced only restricted G and F cells (Fig. 4B and Supplementary Table S3, clones II.15, II.19-II.20). One GF colony from the progeny of cloning III was subcloned once more yielding only GF cells, indicating symmetrical proliferative divisions even in round IV of cloning (Fig. 4B and Supplementary Table S3, clone III.16). Taken together, these results clearly indicate that FGF2 stimulates the propagation of GF progenitors. These cells were able to self-renew by both symmetrical and asymmetrical divisions and maintained their differentiation capacity, exhibiting true stem cell properties.

In the FGF2-treated condition, one MF (Fig. 4B and Supplementary Table S3, clone I.23) and one GM colony (Fig. 4B and Supplementary Table S3, clones I.24) were subcloned to produce the progeny of cloning II. In the first case, the MF progenitor was able to self-renew, producing another MF progenitor in addition to restricted M and F cells and indicating asymmetric division. In contrast, the GM progenitor underwent only differentiating division, producing melanocytes.

Finally, by assessing the number of self-renewing GF progenitors that also produced differentiated progeny, we could estimate the proportion of GF stem cells in each round of the subcloning procedure. In control cultures, the self-renewing GF stem cells were found only in the cloning I as 4.1% of the population. In FGF2-treated cultures, however, approximately 6% of the total population was GF stem cells. These were observed in all 4 rounds of successive subcloning. Although the proportion of GF stem cells was not significantly increased by the treatment, FGF2 maintained the proliferative capacity of GF cells for longer time periods, thus possibly contributing to amplification of the NC cell population.

Discussion

FGFs regulate NC cell proliferation, survival, and differentiation [12,15,22]. However, the involvement of FGF2 in NC multipotentiality and self-renewal is unknown. Therefore, in this study, we used mass and serial clonal cultures to show that FGF2 increases the proportion of NC multipotent progenitors and stimulates the self-renewal and propagation capacities of bipotent GF stem cells. These data have significant implications in the development of the NC and its derived structures.

FGF2 regulates in vitro proliferation, survival, and differentiation of NC cells

FGF signaling interacts with several other signal transduction pathways. Integration of FGF signaling and Notch activation regulates gliogenesis and chondrogenesis [11,14,23] and inhibits neuronal differentiation and expansion in the NC [26]. Garcez et al. (2009) showed that FGF2 treatment in 24-h primary cultures of quail trunk NC stimulates Schwann cell differentiation, whereas EGF treatment stimulates melanocytic and neuronal differentiation [12]. In the present article, however, a 6-day treatment with FGF2 in quail NC secondary cultures resulted in reduced proportions of the main NC-derived phenotypes (i.e., glial cells, neurons, melanocytes, and SMCs). The majority of the NC cell population corresponded to unmarked cells, suggesting that FGF2 could prevent NC cell differentiation. In fact, experiments in which FGF2 was removed and replaced by a differentiating medium resulted in increased amounts of glial cells and neurons accompanied by unchanged numbers of melanocytes and SMCs, as well as a reduction in the amount of unmarked cells (Table 1).

FGF2 also promoted huge increases in the amount of NC cells (Fig. 1) due to enhanced cell proliferation (Fig. 1F) and/or reduced cell death (Fig. 1G). These data agree with previous studies in which FGF2 was reported to regulate NC cell proliferation and differentiation [12,15,22,27].

FGF2 stimulates multipotent NC progenitors

Several types of multipotent progenitors and more restricted cells were previously identified by in vitro clonal assays in both the cephalic and trunk NC [3,4,35]. In fact, the vast majority of the cephalic quail NC cells are highly multipotent precursors. Among them, the hexapotent progenitor able to produce glial cells, neurons, melanocytes, SMCs, chondrocytes, and osteocytes (GNMFCO) has been identified as the most highly multipotent NC cell [3]. More restricted potentiality has been observed at the trunk level, wherein the most highly multipotent NC cell found was the tetrapotent glial, neuron, melanocyte, and smooth muscle (GNMF) progenitor [35].

The fate of NC cells is most strongly influenced by environmental cues [6,9], although recent data have suggested that trunk NC cells are predetermined [16]. At the trunk level, multiple growth factors and substrate molecules influence the differentiation of NC progenitors. Fibronectin promotes the survival of multipotent and oligopotent NC progenitors endowed with smooth muscle potential [5]. BMP2/4 induces neurogenesis, whereas neuregulin 1 and the Notch ligand Delta 1 stimulate glial differentiation in mammalian oligopotent NC cells [21,32,33]. Smooth muscle differentiation of NC progenitors is stimulated by transforming growth factor β [30,32], and glial and melanocyte differentiation are stimulated by endothelin-3 [33,34].

Although FGF2 has been implicated in NC cell proliferation and differentiation of NC cells [11,20,25], its effects on NC multipotentiality have not been demonstrated thus far. Our results in which the production of glial cells and neurons was greatly stimulated at the expense of melanocytes after substitution of FGF2 with the complex medium suggest that FGF2 can influence specific NC progenitors. This issue was addressed by single-cell culture assays. Although FGF2 did not affect the cloning efficiency of the entire population of NC progenitors, it significantly enriched the clonogenic multipotent progenitors (GMF+GNF+GMN+GMNF) (Fig. 2E). Multipotent clones were also larger in FGF2-treated cultures than in control cultures, although no differences in clone size were recorded when the clones were analyzed collectively (Supplementary Table S1). Therefore, FGF2 could affect the proliferation and/or survival of specific subsets of NC progenitors, especially the multipotent progenitors, instead of the entire NC cell population. In addition, phenotypic analysis of colonies revealed that FGF2 increased the frequency of multipotent GMF and GMNF progenitors (Fig. 2E). In this line, the U clones, found in a minor proportion, could correspond to multipotent NC cells that remained undifferentiated even after removal of FGF2 and exposure to the differentiated culture conditions. Increased proportions of progenitors endowed with neuronal or smooth muscle potentials were also promoted by FGF2 (Fig. 2A). Interestingly, no alterations were recorded in the proportion of glial progenitors in the first round of the cloning assay, although significant increases in this cell phenotype were found in mass cultures. Differences in cell density between mass and single-cell cultures could explain these discrepancies.

FGF2 stimulates the self-renewal and propagation of bipotent glial smooth muscle (GF) progenitors that display true stem cell properties

Although the multipotentiality of NC cells has been extensively studied, their self-renewal capacity is not well understood. Stemple and Anderson (1992) showed that oligopotent progenitors from rat trunk NC could propagate in culture and were able to generate autonomic neurons, glial cells, and SMCs [34]. Furthermore, bipotent GM and GF progenitors from quail NC are also endowed with self-renewal capacity [35].

Although FGF2 has been implicated in the propagation and expansion of neural stem cells in the central nervous system, its involvement in the self-renewal of NC cells is unknown. This issue was addressed in the present study with sequential cloning experiments. The propagation of clonogenic NC cells progressively decreased with successive rounds of subcloning for both the control and the FGF2-treated conditions, as demonstrated by reduced cloning efficiency and colony size (Fig. 3). Importantly, FGF2 increased the propagation of progenitors endowed with glial and smooth muscle potentials, at the expense of the melanocytic fate.

Even in the absence of FGF2, GMF, GF, and MF, precursors could propagate to generate identical cells and restricted progeny, indicating that they display true stem cell properties (Fig. 4). This was the first evidence for avian GMF and MF self-renewal. Whereas all 3 precursors were able to undergo asymmetrical divisions, the GF cells could only undergo symmetrical proliferative divisions (Supplementary Table S2). We could not detect the self-renewal of GM progenitors, in agreement with our previous data in which self-renewal required endothelin-3 [35]. In the absence of FGF2, GMF, GF, and MF, progenitors displayed very limited propagation. Although FGF2-treatment did not affect the propagation of GMF and MF cells, it greatly improved the self-renewal of GF cells. The GF progeny generated by asymmetrical divisions of GMF founders displayed similar properties. Importantly, even at round IV of the subcloning procedure, and in spite of the greatly reduced number of cells per clone, the GF precursor could undergo both symmetrically proliferative and asymmetrical divisions.

Quantification of self-renewing GF cells led us to estimate the proportion of stem cells in each round of the subcloning assay (Fig. 5). In the control condition, 4% of GF cells that could not propagate thereafter were found in cloning I. After FGF2 treatment, however, a constant proportion of approximately 6% GF stem cells was recorded until the last evaluated round of the subcloning assay. These data indicate that FGF2 stimulates the propagation of GF stem cells at a constant rate over the subcloning procedure. FGF2 did not seem to increase the cell division capacity or alter the ratio between symmetrically proliferative and asymmetrical divisions of GF stem cells. It could, however, promote their stemness. Therefore, by promoting the self-renewal of GF stem cells, FGF2 can permit their propagation for longer time periods, increasing the entire NC cell population.

FIG. 5.

FGF2 enhances the propagation of GF stem cells. The mean percentage (±SEM) of self-renewing GF stem cells present in each colony subjected to subcloning was estimated from the proportion of parental-like multipotent subclones that developed in control and FGF2-supplemented media (for details, see Supplementary Table S2).

Methodological limitations impaired our ability to detect the self-renewal capacity of the highly multipotent GNMF progenitors, which are near the top of the hierarchical lineage tree. However, as FGF2 increased the proportion of this progenitor in cloning I, it might also regulate GNMF self-renewal.

The mechanisms regulating NC stem cell maintenance and proliferation are only partially understood. NC progenitors are induced at the neural plate border, and subsequently in the dorsal neural tube, as a consequence of complex signaling events involving the BMP, Wnt, and FGF pathways during gastrulation [6]. The role of these signaling molecules in NC induction has been derived from studies in Xenopus laevis at cephalic level and thus, the signal-promoting trunk NC specification was not fully elucidated. In that sense, FGF signaling has been recently demonstrated in chicken embryos to control the initiation of NC specification and together with retinoic acid signaling, the timing of NC epithelial–mesenchymal transition and emigration from the dorsal neural tube [18]. These events are necessary for the distribution of NC progenitors along the body axis that will ultimately differentiate into a diverse array of cell type distributes through the vertebrate body plan, including neurons and glia of the peripheral nervous system, SMCs, and melanocytes [6]. Furthermore, the transcription factor Stat3 has recently been identified as an essential FGF-transducing signal for cell cycle progression and NC specification in Xenopus [25]. Therefore, Stat3 has been suggested to promote the proliferation and maintenance of NC cells in an undifferentiated state [25], and it is a candidate for mediating the effects of FGF2 on NC multipotency and self-renewal. However, the FGF2 signaling components and the specific target genes involved in NC cell self-renewal remain to be identified.

In summary, our in vitro studies of cloning and subcloning procedures provided new evidence for the self-renewal capacity of trunk NC cells. We showed that FGF2 promoted the propagation of GF stem cells by means of both symmetrical and asymmetrical divisions, similar to the previous reported action of endothelin-3 on GM cells [35]. The data from cloning, along with the mass culture data, suggest that FGF2 can drive multipotent NC progenitors to glial, smooth muscle, and possibly neuronal fates at the expense of the melanocytic fate.

Footnotes

Acknowledgments

This work was supported by Ministério da Ciência e Tecnologia/Conselho Nacional de Desenvolvimento Científico e Tecnológico (MCT/CNPq/Brazil), CNPq/PIBIC/PIBIT (Brazil), MCT/INFRA (Brazil), PRONEX/CNPq, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, Brazil), Instituto Nacional de Neurociência Translacional (INNT), and Fundação de Amparo à Pesquisa do Estado de Santa Catarina (FAPESC, SC, Brazil). Authors thanks Ricardo Castilho Garcez for the careful reading in the manuscript and Tyson do Brasil for providing chicken eggs.

Author Disclosure Statement

No competing financial interest exists.

References

Achilleos

, Trainor

. 2012. Neural crest stem cells: discovery, properties and potential for therapy. Cell Res, 22:288–304.

Baroffio

, Dupin

, Le Douarin

. 1988. Clone-forming ability and differentiation potential of migratory neural crest cells. Proc Natl Acad Sci U S A, 85:5325–5329.

Calloni

, Le Douarin

, Dupin

. 2009. High frequency of cephalic neural crest cells shows coexistence of neurogenic, melanogenic, and osteogenic differentiation capacities. Proc Natil Acad Sci U S A, 106:8947–8952.

Calloni

, Glavieux-Pardanaud

, Le Douarin

, Dupin

. 2007. Sonic Hedgehog promotes the development of multipotent neural crest progenitors endowed with both mesenchymal and neural potentials. Proc Natl Acad Sci U S A, 104:19879–19884.

Costa-Silva

, da Costa

, Melo

, Neves

, Alvarez-Silva

, Calloni

, Trentin

. 2009. Fibronectin promotes differentiation of neural crest progenitors endowed with smooth muscle cell potential. Exp Cell Res, 315:955–967.

Le Douarin

, Kalcheim

. 1999. The Neural Crest. Cambridge University Press: Cambridge.

Dulac

, Cameron-Curry

, Ziller

, Le Douarin

. 1988. A surface protein expressed by avian myelinating and nonmyelinating Schwann cells but not by satellite or enteric glial cells. Neuron, 1:211–220.

Dupin

, Calloni

, Real

, Goncalves-Trentin

, Le Douarin

. 2007. Neural crest progenitors and stem cells. C R Biol, 330:521–529.

Dupin

, Calloni

, Le Douarin

. 2010. The cephalic neural crest of amniote vertebrates is composed of a large majority of precursors endowed with neural, melanocytic, chondrogenic and osteogenic potentialities. Cell Cycle, 9:238–249.

10.

Fauquet

, Ziller

. 1989. A monoclonal antibody directed against quail tyrosine hydroxylase: description and use in immunocytochemical studies on differentiating neural crest cells. J Histochem Cytochem, 37:1197–1205.

11.

Fujita

, Yasui

, Shinohara

, Ito

. 2011. Interaction between NF-κB signaling and Notch signaling in gliogenesis of mouse mesencephalic neural crest cells. Mech Dev, 128:496–509.

12.

Garcez

, Teixeira

, Schmitt Sdos

, Alvarez-Silva

, Trentin

. 2009. Epidermal growth factor (EGF) promotes the in vitro differentiation of neural crest cells to neurons and melanocytes. Cell Mol Neurobiol, 29:1087–1091.

13.

Ido

, Ito

. 2006. Expression of chondrogenic potential of mouse trunk neural crest cells by FGF2 treatment. Dev Dyn, 235:361–367.

14.

Ijuin

, Nakanishi

, Ito

. 2008. Different downstream pathways for Notch signaling are required for gliogenic and chondrogenic specification of mouse mesencephalic neural crest cells. Mech Dev, 125:462–474.

15.

Kalcheim

. 1989. Basic fibroblast growth factor stimulates survival of nonneuronal cells developing from trunk neural crest. Dev Biol, 134:1–10.

16.

Krispin

, Nitzan

, Kalcheim

. 2010. The dorsal neural tube: a dynamic setting for cell fate decisions. Dev Neurobiol, 70:796–812.

17.

Kubota

, Ito

. 2000. Chemotactic migration of mesencephalic neural crest cells in the mouse. Dev Dyn, 217:170–179.

18.

Martínez-Morales

, Diez Del Corral

, Olivera-Martínez

, Quiroga

, Das

, Barbas

, Storey

, Morales

. 2011. FGF and retinoic acid activity gradients control the timing of neural crest cell emigration in the trunk. J Cell Biol, 194:489–503.

19.

Mason

. 2007. Initiation to end point: the multiple roles of fibroblast growth factors in neural development. Nat Rev Neurosci, 8:583–596.

20.

Moore

, Ferretti

, Copp

, Thorogood

. 2002. Blocking endogenous FGF-2 activity prevents cranial osteogenesis. Dev Biol, 243:99–114.

21.

Morrison

, Csete

, Groves

, Melega

, Wold

, Anderson

. 2000. Culture in reduced levels of oxygen promotes clonogenic sympathoadrenal differentiation by isolated neural crest stem cells. J Neurosci, 20:7370–7376.

22.

Murphy

, Reid

, Ford

, Furness

, Bartlett

. 1994. FGF2 regulates proliferation of neural crest cells, with subsequent neuronal differentiation regulated by LIF or related factors. Development, 120:3519–3528.

23.

Nakanishi

, Chan

, Ito

. 2007. Notch signaling is required for the chondrogenic specification of mouse mesencephalic neural crest cells. Mech Dev, 124:190–203.

24.

Nataf

, Mercier

, Ziller

, Le Douarin

. 1993. Novel markers of melanocyte differentiation in the avian embryo. Exp Cell Res, 207:171–182.

25.

Nichane

, Ren

, Bellefroid

. 2010. Self-regulation of Stat3 activity coordinates cell-cycle progression and neural crest specification. EMBO J, 29:55–67.

26.

Nikopoulos

, Duarte

, Kubu

, Bellum

, Friesel

, Maciag

, Prudovsky

, Verdi

. 2007. Soluble Jagged1 attenuates lateral inhibition, allowing for the clonal expansion of neural crest stem cells. Stem Cells, 25:3133–3142.

27.

Ota

, Ito

. 2006. BMP and FGF-2 regulate neurogenin-2 expression and the differentiation of sensory neurons and glia. Dev Dyn, 235:646–655.

28.

Petiot

, Ferretti

, Copp

, Chan

C-TJ

. 2002. Induction of chondrogenesis in neural crest cells by mutant fibroblast growth factor receptors. Dev Dyn, 224:210–221.

29.

Qian

, Davis

, Goderie

, Temple

. 1997. FGF2 concentration regulates the generation of neurons and glia from multipotent cortical stem cells. Neuron, 18:81–93.

30.

Real

, Glavieux-Pardanaud

, Vaigot

, Le-Douarin

, Dupin

. 2005. The instability of the neural crest phenotypes: schwann cells can differentiate into myofibroblasts. Int J Dev Biol, 49:151–159.

31.

Sarkar

, Petiot

, Copp

, Ferretti

, Thorogood

. 2001. FGF2 promotes skeletogenic differentiation of cranial neural crest cells. Development, 128:2143–2152.

32.

Shah

, Groves

, Anderson

. 1996. Alternative neural crest cell fates are instructively promoted by TGFbeta superfamily members. Cell, 85:331–343.

33.

Shah

, Marchionni

, Isaacs

, Stroobant

, Anderson

. 1994. Glial growth factor restricts mammalian neural crest stem cells to a glial fate. Cell, 77:349–360.

34.

Stemple

, Anderson

. 1992. Isolation of a stem cell for neurons and glia from the mammalian neural crest. Cell, 71:973–985.

35.

Trentin

, Glavieux-Pardanaud

, Le Douarin

, Dupin

. 2004. Self-renewal capacity is a widespread property of various types of neural crest precursor cells. Proc Natl Acad Sci U S A, 101:4495–4500.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

0.03 MB

0.02 MB

0.03 MB