Differential modulation of Sonic-hedgehog-induced cerebellar granule cell precursor proliferation by the IGF signaling network

Abstract

The molecular mechanisms regulating organ growth and size remain unclear. Sonic hedgehog (SHH) signaling is a major player in the regulation of cerebellar development: SHH is secreted by Purkinje neurons and acts on the proliferation of granule cell precursors (GCPs) in the external germinal layer. These then become postmitotic and form the internal granular layer but do so in the presence of SHH ligand, begging the question of how the proliferative response to SHH signaling is downregulated in differentiating GCPs. Here, we have determined the precise cellular localization of the expression of insulin-like growth factor (IGF) network components in the developing mouse cerebellum and show that this network modulates the proliferative effects of SHH signaling on GCPs. IGF1 and IGF2 are potent mitogens for GCPs and both synergize with SHH in inducing GCP proliferation. Whereas the proliferative activity of IGF1 or IGF2 on GCPs does not require intact SHH signaling, aspects of SHH activity on GCP proliferation require signaling through the IGF receptor 1. Moreover, we find that 3 of the IGF-binding proteins, IGFBP2, IGFBP3 and IGFBP5, inhibit IGF1/2-induced cell proliferation, whereas IGFBP5 also inhibits SHH-induced GCPs proliferation. This novel function of IGFBP5 that we have uncovered demonstrates the exquisite regulation of SHH signaling by different components of the IGF network.

Fig. 1

Expression analysis of members of the IGF signaling network in the developing embryonic and postnatal cerebellum. Sagittal sections of mouse embryonic (E17.5) and postnatal (P0, P1, P2 or P5) cerebellum were hybridized with RNA probes corresponding to mouse Shh, Gli1 and members of the IGF signaling network. Shh is expressed in the PNs (a) and Gli1 expression in the EGL and in cells in the PL (b). Igfr1 expression is ubiquitous in the cerebellum, with stronger expression in EGL cells (c, arrow). Igf1 expression in the PNs is apparent as early as E17.5 (d, arrow), becoming stronger during the early postnatal stages (e, f). Igf2 is detected in meninges (g–i, arrows), choroid plexus (arrowhead in g) and in blood vessels (red arrows in g–i) in the cerebellar tissue. Igfbp2 is expressed in the PL, choroids plexus and meninges (j). Igfbp3 is expressed in the PL (k, l), while Igfbp4 expression is observed in the meninges (arrow) and choroid plexus (arrowhead in m) (m, n). Igfbp6 is expressed in a subset of PNs (o); note that proliferating phosphoshistone H3 positive cells are detected in the EGL of P15 cerebellum (arrow in o). Igfbp5 expression is detected at all stages in the EGL and cells within the PL (arrows in p, r). M = Meninges; ChP = choroid plexus.

Fig. 2

IGF1 and IGF2 act as mitogens on GCP proliferation. The effects of IGFs on GCP proliferation were tested using 250-μm postnatal cerebellar explants (a), GCP aggregates (b) or isolated, purified GCPs (c). After 48 h in culture, IGF1 (20 ng/ml) and IGF2 (40 ng/ml) induce an increase in cell proliferation in the outer side of the explants containing the GCPs (arrows in a), in the GCP aggregates (b) or in isolated cells (c) as visualized by BrdU incorporation. b, c All nuclei are stained with Hoechst in blue and BrdU is in green. The quantification of the IGF1 and IGF2 proliferative effect on GCPs in isolated cultures shows a dose effect of these growth factors (d).

Fig. 3

SHH acts synergistically with IGF1 and IGF2. Purified GCPs were either not treated (control) or treated with SHH or SHH + IGF1 (n = 4 for each condition). GCPs were also treated with SHH or SHH + IGF2 (n = 5 for each condition). Cell proliferation is significantly higher when GCPs are treated with both SHH + IGF1 (p < 0.001) and SHH + IGF2 (p < 0.001) than when they are treated with SHH alone. The concentrations of growth factors were as follows: IGF1: 20 ng/ml; IGF2: 40 ng/ml; SHH was used at 400 or 600 ng/ml. Similar results were obtained with these 2 concentrations of SHH in all experiments.

Fig. 4

Cyclopamine does not affect IGF1- or IGF2-induced proliferation of GCPs. Purified GCPs were treated with SHH (400 ng/ml; n = 3), IGF1 (20 ng/ml) or IGF2 (40 ng/ml) alone or in combination with cyclopamine (5 μM). All growth factors alone enhance cell proliferation. Addition of cyclopamine significantly decreases cell proliferation for cells treated with SHH (p < 0.0001; n = 4), but it does not induce any significant decrease in the number of BrdU-positive cells when GCPs are treated with either IGF1 (p = 0.19; n = 4) or IGF2 (p = 0.15; n = 4). ∗ p < 0.05.

Fig. 5

IGF signaling is required for SHH-induced proliferation of GCPs. Purified GCPs were treated with SHH (200 ng/ml, n = 7 independent tests) or IGF1 (20 ng/ml, n = 6) alone or in combination with the monoclonal antibody A12, an anti-IGFR1 blocking antibody (20 μg/ml). Blockade of IGFR1 induces a decrease in proliferation on GCPs following treatment with IGF1 (p < 0.0001) or SHH (200 ng/ml; p < 0.0001).

Fig. 6

SHH-induced proliferation is specifically inhibited by IGFBP5 but not by IGFBP2 or IGFBP3. Purified GCPs were either left untreated (control, n = 14 independent tests) or treated with the following binary combinations of factors: SHH or SHH + IGFBP2 (n = 5 for each condition); SHH or SHH + IGFBP3 (n = 5 for each condition); SHH or SHH + IGFBP5 (n = 9 for each condition), IGF1 or IGF1 + IGFBP2 (n = 5 for each condition), IGF1 or IGF1 + IGFBP3 (n = 4 for each condition), IGF1 or IGF1 + IGFBP5 (n = 5 for each condition). IGFBPs were used at 400 ng/ml, IGF1 at 20 ng/ml and IGF2 at 40 ng/ml. Similar results were obtained with SHH concentrations of 400 or 600 ng/ml. IGF1-induced proliferation on GCPs is reduced when cells are co-treated with IGFBP2 (p < 0.0003), IGFBP3 (p < 0.0001) or IGFBP5 (p < 0.0001). However, only IGFBP5 causes a significant decrease in SHH-induced GCP proliferation (p < 0.0001), while IGFBP2 (p = 0.4) or IGFBP3 (p = 0.8) have no significant effect on SHH activity. ∗ = Significant differences.