Cell cycle regulation by ADP and IGF-1 in cultured late developing glia progenitors of the avian retina

Abstract

In the avian retina, ADP induces the proliferation of late developing glia progenitors. Here, we show that in serum-containing retinal cell cultures, ADP-induced increase in [³H]-thymidine incorporation can be prevented by the IGF-1 receptor antagonists AG1024 and I-OMe-Tyrphostin AG 538, suggesting the participation of IGF-1 in ADP-mediated progenitor proliferation. In contrast, no increase in [³H]-thymidine incorporation is observed in retinal cultures treated only with IGF-1. Under serum starvation, while no increase in cell proliferation is detected in cultures treated only with ADP or IGF-1, a significant increase in [³H]-thymidine incorporation and number of PCNA expressing cells is observed in cultures treated concomitantly with ADP plus IGF-1, suggesting that both molecules are required to induce proliferation of retinal progenitors. In serum-starved cultures, although an increase in cell viability is detected by MTT assays in IGF-1-treated cultures, no significant increase in viability of [³H]-thymidine labeled progenitors is observed, suggesting that IGF-1 may contribute to survival of postmitotic cells in culture. While only ADP increases intracellular calcium, only IGF-1 induces the phosphorylation of Akt in the retinal cultures. IGF-1 through the PI3K/Akt pathway induces a significant increase in the transcription and expression of CDK1 with a decrease in phospho-histone H3 expression that is concomitant with an increase in the expression of cyclins D1 and E and CDK2. These findings suggest that IGF-1 stimulates CDK-1 mRNA and protein expression that enable progenitors to progress through the cell cycle. However, signaling of ADP in the presence IGF-I seems to be required for DNA synthesis.

Introduction

The vertebrate retina has been extensively used for studies approaching the neurochemical and morphological development of CNS and signaling by nucleotide receptors is involved in the regulation of its development. Nucleotide responses mediated by P2X7 receptors are associated with neuronal death [1, 2], while P2Y receptor-dependent responses are implicated in glial migration [3] and induction of cell proliferation [4–8]. At the very beginning of chick embryo retina development, while activation of UTP-sensitive P2Y2/4 receptors induces the proliferation of early developing progenitors of ganglion, amacrine, horizontal cells and photoreceptors [6], activation of ADP-sensitive P2Y receptors induces the proliferation of late developing glial / bipolar progenitors [9, 10]. In this tissue, ADP strongly stimulates the formation of phophoinositides, calcium mobilization, phosphorylation of ERK and CREB, signaling pathways that activation is required for nucleotide-dependent proliferation of retinal progenitors [4, 9, 11–13].

PI3K is typically activated upon growth factor stimulation of the cells and is a canonical signaling pathway for Insulin and IGF-1 in the retina [14], where it seems crucial in mediating cell survival. For instance, IGF-1 mediates the activation of PI3K/Akt and prevents the death of ganglion cells induced by hypoxia or optic nerve transection in the rat retina [15, 16]. Moreover, IGF-1/PI3K/Akt signaling also induces the proliferation of several developing cell types, including neural progenitor cells [17], oligodendrocyte progenitors [18] and murine retinal progenitors [19]. In the developing chick embryo retina, the presence of IGF-1 and specific receptors as well as the involvement of IGF-1 in cell survival and/or proliferation is well established [20–24].

Involvement of PI3K/Akt signaling in specific phases of the cell cycle such as G1/S or G2/M transitions was suggested. In synchronized cell lines, for example, Akt is active during mitosis and inhibition of PI3K/Akt promotes a delay in S phase exit and G2/M entry due a decrease in CDK1 activity [25, 26]. Similarly, in chick embryo retinal cultures, phospho-Akt is observed in dividing cells (from prophase to anaphase) and pharmacological inhibition of PI3K decreases CDK1 protein expression that leads to an accumulation of phospho-histone H3 positive cells arrested at G2/M transition of the cell cycle[27]. In addition, pharmacological blockade of PI3K also inhibits ADP-dependent increases in cyclin D1 expression and [³H]-thymidine incorporation, impairing G1/S transition of the cell cycle of the retinal progenitors in culture [28].

Although a growing body of evidences supports the involvement of PI3K/Akt pathway in cell proliferation, still limited information is available regarding the factors that stimulate PI3K/Akt pathway during progenitor’s cell cycle. In the present work, we aimed to investigate the contribution of ADP and IGF-1 signaling to the proliferation of late developing retinal glial progenitors in culture. Our data reveal that while ADP induces calcium mobilization, IGF-1, but not ADP, stimulates the PI3K/Akt pathway in the cultures. While IGF-1-stimulated PI3K/Akt signaling seems to be necessary for expression of cyclins and CDKs that enables progenitors to progress through cell cycle, both ADP and IGF-I signaling seems to be required for progenitors to accomplish DNA synthesis.

Experimental Procedures

Materials

Fertilized White Leghorn chicken eggs were obtained from a local hatchery and incubated at 38 °C in a humidified atmosphere up to the appropriate stage. [³H]-thymidine (5 Ci/mmol) was from PerkinElmer (São Paulo, SP, Brazil), ADP, LY294002 (PI3K inhibitor), I-OMe-Tyrphostin AG 538 (IOME-AG 538), U0126 (MEK inhibitor), insulin-like growth factor-1 (IGF-I), anti-α-tubulin (catalog # T5168) were from Sigma-Aldrich (St. Louis, MO, USA); Minimum Essential Medium (MEM), Fetal Bovine Serum were from Life Technologies (São Paulo, SP, Brazil). Antibodies against phospho-IGFR (catalog # 3024), phospho-Akt (catalog # 4060), Akt (catalog # 4691), phospho-ERK (catalog # 4370), ERK (catalog # 9102), phospho-4E-BP1 (Thr 36/46, catalog # 2855), PCNA (catalog # 2586), phospho-Histone H3 (Ser 10, catalog # 3377), CDK1 (catalog # 9112) were from Cell Signaling Technology (MA, USA). Antibodies against cyclin D1 (catalog sc-718), cyclin E (catalog sc-481) and AG-1024 were from Santa Cruz Biotechnology. All other reagents were of analytical grade.

Retinal cell monolayer cultures

Chick embryos at embryonic day 7 (E7) were killed instantaneously by decapitation and the eyes were removed and immediately transferred to 1 mL of Ca²⁺- and Mg²⁺-free balanced salt solution (CMF) where the retinas were dissected from other structures. Trypsin, at a final concentration of 0.1%, was added and the suspension incubated at 37 °C for 20–25 min. Trypsin solution was removed and retinas suspended in Minimum Essential Media (MEM) containing 5% fetal bovine serum (FBS), 2 mM glutamine, 100 U/ml penicillin and 100 µg/mL streptomycin. Tissues were mechanically dissociated by successive aspirations of the medium and cells counted in a Neubauer chamber. For immunofluorescence experiments, cells were seeded on coverslips at a density of 3 × 10⁶ cells/dish (3.1 × 10³ cells / mm²). For western blotting and qPCR experiments, 10⁷ cells were seeded on 35 mm plastic culture dishes (1.04 × 10⁴ cells / mm²). Cells were then incubated at 37 °C for the indicated periods of time, in humidified atmosphere of 95% air / 5% CO₂.

Serum starvation and culture treatment

One day after plating the cells (E7C1), cultures were rinsed 3 × with MEM to remove FBS and cells cultivated for an additional period of 24 h in MEM without FBS with 2 mM glutamine, 100 U/ml penicillin and 100 µg/mL streptomycin. In experiments in which signaling pathways were evaluated, serum starved cultures at E7C2 were incubated for 5 min with agonists (250 µM ADP and/or 100 ng/mL IGF-I) at 37 °C, in MEM buffered with 20 mM HEPES, pH 7.4. After incubation, cultures were processed for western blotting or immunofluorescence. For [³H]-thymidine incorporation, qPCR and western blotting for cell cycle regulatory proteins, serum-starved cultures at E7C2 were treated for additional 24 h with 250 µM ADP, 100 ng/mL IGF-I, 25 µM LY 294002, 1 µM or 10 µM AG-1024 in MEM without FBS containing glutamine and antibiotics.

Western blotting

After treatment of cultures, 60 µL of sample buffer without bromophenol blue was added to each dish, cell extracts were harvested, boiled and centrifuged at 27.000 × g for 10 min to remove non-soluble material. Protein content of 2 µL samples of extracts were estimated by the Bradford protein assay, using BSA solution containing 2 µL sample buffer as standard. Extract samples (50 µg / lane) were size fractionated on 9% or 12% SDS polyacrylamide gel, transferred to PVDF membranes (GE Healthcare), stained with Ponceau red and blocked in Tris-buffered saline (pH 7.6) with 0.1% Tween-20 and 5% non-fat milk or 5% BSA for cyclin D1, cyclin E and CDK2 detection. Membranes were incubated with diluted primary antibody (p-IGF-IR, 1:2000; p-Erk, 1:2000; p-Akt, 1:4000; p-4E-BP1, 1:4000; cyclin D1, 1:1000; cyclin E, 1:1000; CDK2, 1:1000; p27/kip1, 1:2000; CDK1, 1:2000; p-H3, 1:3000) overnight, at 4 °C. Blots were developed using the corresponding secondary antisera conjugated to horseradish peroxidase (Bio-Rad Labs Inc.) and enhanced chemiluminescence, according to the manufacturer’s protocol (ECL plus, GE Healthcare). In selected experiments, membranes were stripped and re-probed with anti-Erk (1:2000), anti-Akt (1:2000) or anti-α-tubulin (1:70,000), at 4 °C, followed by incubation with secondary antibody and detection as described above.

[³H]-thymidine incorporation

Treated cultures were incubated with [³H]-thymidine (0.5 µCi) for 60 min, at 37 °C. Cultures were then washed four times with 2 mL MEM buffered with 25 mM HEPES, pH 7.4 and cells dissolved with 0.2 mL of 0.4 N NaOH. After dilution of the samples with 3 mL H₂O, 0.6 mL of 50% trichloroacetic acid (TCA) was added and the mixtures incubated, at 4 °C, for at least 30 min. Samples were filtered through Whatmann GF/B glass fiber filters and washed three times with 5% TCA. Filters were dried and the radioactivity determined by scintillation spectroscopy.

MTT viability assay

Cell viability was determined by the MTT reduction method reproduced in chick retinal cultures by Anccasi et al. [2]. Briefly, cultures were incubated at 37° C with 0.3 mg/mL of MTT in MEM buffered with 25 mM HEPES, pH 7.4, without serum. After 20 min, medium was removed and formazan product was dissolved with a mixture of HCl/isopropanol and estimated by the absorbance at 550 nm after subtracting absorbance at 650 nm.

Fixation and Immunofluorescence

For the detection of p-Akt in retinal progenitors, cultures at E7C2 starved for 24 h were treated with IGF-I for 5 min, fixed with cold 4% paraformaldehyde in 0.16 M sodium phosphate buffer, pH 7.4, for 30 min and extensively washed with phosphate buffered saline (PBS). Coverslips were first permeabilized with PBS-1%Triton X-100, incubated with 10 mM citrate buffer, pH 6.0, at 100 °C for 10 min and washed with pure water and PBS. Non-specific binding sites were blocked with 1% bovine serum albumin (BSA) for 1 h and the sections incubated overnight at 4 °C with primary antisera against phospho-Akt (1:300) and PCNA (1:1000). Cells were washed with PBS and incubated with fluorescent secondary Alexa-conjugated antisera for 2 h at room temperature. Cell nuclei were stained with DAPI or Hoechst. Control sections were incubated in the absence of the primary antibody and no immunoreactivity was detected under this condition. Imaging was performed with a Leica SP5 II confocal microscope. For the detection of the increase in PCNA protein expression, coverslips were incubated for 24 h with IGF-1 and/or ADP. Counting of PCNA⁺ and Hoechst positive nuclei was performed with the ImageJ software in 10 fields photographed in each culture. At least 3 experiments in duplicate were performed.

Real-time quantitative PCR (pPCR)

Total RNA was extracted from cultures using Trizol reagent, according to the manufacturer’s protocol. Extracted RNA samples were treated for 10 min at 37 °C with RNase-free DNase I (0.5 U / μg of RNA). First-strand synthesis of cDNA was performed in a solution containing 1 μg of RNA, reverse transcriptase, random hexamers and DTT for 1 h, at 37 °C, using the First-Strand cDNA synthesis kit (GE Healthcare Life Sciences), following the instructions of the manufacturer. The quantitative real time PCR amplification was performed using GoTaq qPCR Master Mix (Promega) and carried out on a Real-Time PCR System (StepOne; Applied Biosystems) under the following conditions: an initial holding stage at 95 °C for 2 min that was followed by 40 cycles of denaturation at 95 °C for 15 s, annealing and extension at 60 °C for 1 min. Melt curve was performed (slow ramp of 0.3 °C/20 s to 95 °C). Specific primers for CDK1 (GenBank sequence no 396252) that amplified a fragment of 122 bp were as follows: forward primer = 5´-GACGAGTTCTGCACAGAGAC-3´; reverse primer = 5´-ATGAGTGTATACCCGCACCG-3`. Primers 5’-AAGCCGGGGAAGGTGGTG-3’ (base position 42) and 5’-GGGTGGGCATCAGGTGGT-3’ (base position 276), amplifying a 235-bp fragment of DNA for the ribosomal protein L27 (GenBank sequence no. X56852) were used in all PCR experiments as an internal control. For both control and treated samples, the L27 Ct values were subtracted from the CDK1 Ct values to obtain the ΔCt value. ΔΔCt values were obtained according to ΔCt (treated samples)—ΔCt (control samples). The relative quantity was calculated according to 2-ΔΔCt.

Calcium Imaging

The Fluo-3 AM indicator was used, folowing previous published protocol [3]. In brief, cultures were incubated for 1 h in complete MEM medium containing 5 µM Fluo-3 AM, 0.2% (v/v) Pluronic F-127, 0.5% (v/v) DMSO and 2.5 mM Probenecid. Cells were washed with HBSS with Probenecid and incubated for an additional 15 min to allow complete de-esterification of the AM ester. Live calcium imaging was carried out on a Leica TCS SP5 II confocal microscope, using a 488 laser line for excitation, images sizes of 256 × 256 pixels and acquisition rates of 1 frame/sec. Emitted fluorescence was recorded at wavelengths between 530—565 nm. Data were expressed as the ratio F/F_o with F and F_o representing Fluo-3 fluorescence emission during stimulation and the mean fluorescence emission before stimulation of the cells, respectively. Data was also expressed as the ratio F_max/F₀ where F_max represents the maximal Fluo-3 AM fluorescence emission during stimulation of the cells.

Statistical analysis

The statistical analysis was performed by ANOVA and the Bonferroni’s multiple comparison test.

Results

The concomitant effect of nucleotides and growth factors in proliferative events in the developing retina was demonstrated [7] and IGF-1 present in early developing retinas affects retinal cell proliferation, differentiation and survival [20, 22, 23]. To investigate a possible contribution of IGF-1 to the ADP-induced increase in the proliferation of retinal progenitors, we initially evaluated the effect of AG-1024 and I-OMe-Tyrphostin AG 538 (IOME AG 538), specific inhibitors of IGF-1 receptors, on the ADP-induced increase in [³H]-thymidine incorporation in retina cultures (Fig. 1). Although no decrease in [³H]-thymidine incorporation was observed when cultures were stimulated with 250 µM ADP in the presence of 1 µM AG-1024 (in cpm/culture: Basal = 1153 ± 172; ADP = 2096 ± 273; ADP + 1 µM AG-1024 = 2138 ± 213), ADP-induced increase in [³H]-thymidine incorporation was completely prevented when cultures were stimulated with the nucleotide in the presence of 10 µM AG-1024 (915 ± 123 cpm/culture) (Fig. 1a). A small but not significant decrease in [³H]-thymidine incorporation was observed when cultures were incubated only with 10 µM AG-1024 (758 ± 118 cpm/culture). A decrease in [³H]-thymidine incorporation was also detected with 10 µM IOME AG 538 (Fig. 1b).

Fig. 1.

Effect of the IGF-1 receptor inhibitors on ADP-induced cell proliferation (a, b) and survival of progenitors (c). Cultures at E7C1 were treated with 250 µM ADP in the absence or presence of 1 or 10 µM AG 1024 or 0.1 or 10 µM IOME AG 538. After 24 h, cultures were processed for [³H]-thymidine incorporation and the radioactivity determined by scintillation spectroscopy. For determination of progenitor’s survival, cultures were incubated with 0.5 µCi of [³H]-thymidine for 2 h to label progenitors. Culture medium was removed, cells washed and incubated for an additional 24 h period in the presence of 5 or 10 µM AG-1024 and/or 250 µM ADP. H₂O₂ (500 µM) was used as a positive control. Data represent the mean ± S.E.M. of at least 3 experiments performed in duplicate. *p < 0.05 and **p < 0.01, relative to control cultures; ^###p < 0.001 relative to ADP-treated cultures

The effect of AG-1024 on the survival of retinal progenitors in culture is shown in Fig. 1c. Retinal cultures were incubated with 0.5 µCi of [³H]-thymidine for 2 h to label cultured progenitors. Culture medium containing non-incorporated [³H]-thymidine was removed, cells washed and incubated for an additional 24 h period in the presence of 5 or 10 µM AG-1024 and/or 250 µM ADP. H₂O₂ (500 µM) was used as a positive control. After incubation, incorporated [³H]-thymidine was estimated. While incubation of the cultures with H₂O₂ induced a significant decrease of ~ 50% in the levels of pre-incorporated [³H]-thymidine, no significant decrease in the radioactivity present in the cultures was observed with 5 µM or 10 µM AG-1024, either in the presence or absence of ADP.

The effect of ADP and IGF-1 on the incorporation of [³H]-thymidine in retinal cultures is shown in Fig. 2. Retinal cultures obtained from 7-day-old chick embryos were cultured for 48 h (E7C2) and then incubated with 250 µM ADP and/or 100 ng/mL IGF-1 for an additional period of 24 h. ADP-treated cultures showed a significant increase of ~ 240% of control levels in [³H]-thymidine incorporation, consistent with data previously described. No increase in [³H]-thymidine incorporation was observed in cultures treated only with IGF-I. Moreover, [³H]-thymidine incorporation in cultures treated with both agonists was similar to the incorporation observed in the cultures treated only with ADP (~ 271% of control levels).

Fig. 2.

Effect of ADP and IGF-1 treatment on proliferation of cultured retinal progenitors. Retinal cultures were seeded and maintained in MEM containing 5% FBS during the whole experiment. At E7C2 cells were treated with 250 µM ADP and/or 100 ng/mL IGF-1 and after 24 h they were processed for [³H]-thymidine incorporation. Data represent the mean ± S.E.M. of at least 3 experiments performed in duplicate. **p < 0.01; ***p < 0.001 relative to control

Both fetal and adult bovine sera contain significant amounts of IGF-1 and IGF-2 [29]. To eliminate the effect of IGF-1 coming from FBS, serum-starved cultures were used in all next experiments. Primary cultures were prepared in the presence of 2% FBS to allow proper initial establishment. After 24 h, they were rinsed with MEM without FBS and cultured for the next period in medium without FBS. In contrast to the responses observed in cultures containing FBS, cultures free of FBS for 24 h followed by a 24 h incubation with agonists in the absence of serum, showed no increase in [³H]-thymidine incorporation when incubated with 250 µM ADP (Fig. 3a). No increase in [³H]-thymidine incorporation was either detected in the cultures incubated only with IGF-I. However, treatment of the cultures with both ADP and IGF-1 was able to induce a significant increase in the incorporation of [³H]-thymidine that was ~ 195% of the level obtained in control cultures.

Fig. 3.

Effect of ADP and IGF-1 treatment on proliferation of retinal progenitors in serum-starved cultures. a Retinal cultures were seeded in MEM containing 2% FBS and after 24 h (E7C1), they were rinsed and maintained in MEM without FBS until the end of the experiment. At E7C2, cells were treated with 250 µM ADP and/or 100 ng/mL IGF-1 and after 24 h they were processed for [³H]-thymidine incorporation. b Cultures in coverslips were treated as for [³H]-thymidine incorporation, fixed and processed for PCNA labeling. Hoechst was used to label nuclei. Data represent the mean ± S.E.M. of at least 3 experiments performed in duplicate. ⁺p < 0.05 versus ADP; *p < 0.05 versus control; ^##p < 0.01 and **p < 0.01 relative to control or IGF-1-treated cultures

The effect of ADP and IGF-1 on the proliferation of retinal progenitors in culture was also evaluated by immunocytochemistry against Proliferating Cell Nuclear Antigen (PCNA), a protein that is expressed by proliferative progenitors and that increases during S phase of the cell cycle [30, 31]. Retinal cultures were prepared over coverslips and incubated with agonists in serum-deprived medium. Similar to the [³H]-thymidine experiments, no increase in the number of PCNA⁺ nuclei was observed in cultures treated only with ADP or IGF-1. However, an increase corresponding to ~ 156% of the control cultures in the expression of PCNA was observed in the cultures treated with ADP + IGF-1 (Fig. 3b).

The effect of ADP and IGF-1 on the survival of cultured retinal neurons and progenitors was evaluated by MTT assays or by measuring the amount of pre-incorporated [³H]-thymidine. Retinal cultures at E7C2 were treated for 24 h with 250 µM ADP and/or 100 ng/mL IGF-1 and processed for MTT reduction assays as described in Methods. Alternatively, cultures were firstly incubated with 0.5 µCi of [³H]-thymidine for 2 h to label proliferating progenitors. Then, culture medium was removed, cells washed and cells incubated with agonists for 24 h in fresh medium. [³H]-thymidine incorporation was measured as described in methods (Fig. 4b). In MTT experiments (Fig. 4a), significant increases of ~ 58% and ~ 50% in cell viability relative to control cultures was observed in cultures treated with IGF-1 or IGF-1 plus ADP, respectively. On the other hand, [ ³H]-thymidine experiments (Fig. 4b) revealed that, while 500 µM H₂O₂ decreased the radioactivity present in the cultures by ~ 48%, very small and not significant increases in pre-incorporated [³H]-thymidine levels were observed in the cultures treated with ADP or IGF-1 plus ADP.

Fig. 4.

Effect of ADP and IGF-1 on the survival of retinal cells in culture. a Serum-deprived retinal cultures at E7C2 were treated with 250 µM ADP and/or 100 ng/mL IGF-1. After 24 h, cultures were processed for MTT viability assays as described in methods section. b Cultures at E7C2 were incubated with 0.5 µCi of [³H]-thymidine for 2 h to label progenitors. Culture medium was removed, cells washed and incubated for an additional 24 h period with 250 µM ADP and/or 100 ng/mL IGF-1. H₂O₂ (500 µM) was used as a control. Data represent the mean ± S.E.M. of at least 3 experiments performed in duplicate. **p < 0.01 and ***p < 0.001 compared to control cultures

To identify potential signaling pathways activated selectively by ADP and/or IGF-I, cultures maintained in the absence of serum for 24 h were stimulated for 5 min with 250 µM ADP and/or 100 ng/mL IGF-1. The phosphorylation of IGF-1 receptor, ERKs, Akt and the eIF4E-binding protein 4E-BP1 was evaluated by western blotting (Fig. 5). Only IGF-I, in the presence or not of ADP, was able to induce IGF-IR phosphorylation. No effect of ADP on the phosphorylation of IGF-1 receptors was noticed (Fig. 5a).

Fig. 5.

Signaling pathways activated by ADP and IGF-1 in retinal cells. Serum-starved retinal cultures at E7C2 were treated with 250 µM ADP, 100 ng/mL IGF-1 or both for 5 min at 37ºC. a Representative blots showing the detection of phospho-IGF-IR, phospho-ERK, phospho-Akt and phospho-4E-BP1. Blots detecting p-ERK b and pAkt c were quantified by densitometry and expressed as the mean ± S.E.M. in the graphs. Protein gel loading was assessed with anti-ERK or anti-Akt antiserum. Blots are representative of at least 3 experiments performed in duplicate. *p < 0.05 and ***p < 0.001 versus control cultures

In contrast, both ADP and IGF-I were able to stimulate the phosphorylation of ERK, promoting increases of 188% and 245% of the control levels, respectively. Noteworthy, when cells were incubated with both agonists concomitantly, levels of phospho-ERK increased to ~ 310% of the control condition (Fig. 5b). IGF-I, no matter if in the presence of ADP or not, was also able to increase the phosphorylation of both Akt and 4E-BP1, a well-established target of the PI3K/Akt/mTOR signaling pathway. No change of Akt and 4E-BP1 phosphorylation levels could be detected when retinal cultures were incubated only with ADP (Figs. 5a, c).

Akt and the downstream mTOR target eIF4E-binding protein (4E-BP1) are phosphorylated in mitotic retinal progenitors in culture [27]. In order to verify if IGF-I induces the activation of Akt in proliferating retinal progenitors, immunostaining against phosphorylated Akt (pAkt) and PCNA was performed in the retinal cultures (Fig. 6). Treatment with 100 ng/mL IGF-I for 5 min promoted a consistent increase in phospho-Akt labeling that could be noticed over the majority of the cultured cells. Furthermore, several pAkt and PCNA double-labeled cells could be distinguished in IGF-1-stimulated cultures, although no increase in PCNA labeling was noticed in these IGF-1-treated cultures (Fig. 3b).

Fig. 6.

IGF-1 induces the phosphorylation of Akt in cultured retinal progenitors. Serum-starved retinal cultures at E7C2 were incubated with 100 ng/mL IGF-1 for 5 min at 37 ºC, fixed with 4% paraformaldehyde and then labeled with anti-phospho-Akt (red) and anti-PCNA (green). Arrow points to Akt activation in proliferating, double labeled cells. DAPI (blue) was used for nuclear staining. Scale bar = 10 µm

Besides ERK and Akt phosphorylation, nucleotide-dependent proliferation of chick retinal progenitors requires intracellular Ca²⁺ mobilization [4, 11]. The ability of ADP and IGF-1 to increase intracellular calcium content in the cultures is shown in Figs. 7a-c. After 48 h of serum starvation, cultures were loaded with the calcium indicator Fluo-3 AM as described in the “Materials and Methods” section and stimulated with 100 ng/mL IGF-1 or 250 µM ADP. Several glial progenitors consistently responded to ADP. Peak values of the transient increases in [Ca²⁺]_I induced by ADP represented 290% of the resting levels of [Ca²⁺]_i (n = 40 cells). However, no glial progenitors responding to IGF-1 (100 ng/mL) with increases in [Ca²⁺]_i were observed in the cultures (n = 56 cells).

Fig. 7.

Effect of ADP and IGF-1 on the intracellular calcium content of glial progenitors in cultures at E7C3. After serum starvation, retinal cultures were loaded with the calcium indicator Fluo3-AM as described in Methods. (a, b) Representative traces of [Ca²]_I transients after ADP or IGF-1 stimulation (arrows). (c) Peak intensity of fluorescence relative to the mean basal fluorescence (F_max/F₀) in glial cells after the stimulation of the cultures with agonists. Data in c represents the mean ± S.E.M. of the ratio F_max/F₀ calculated from 40 and 56 glial cells recorded in 3 separate experiments. ***p < 0.001 versus control cultures

Expression of CDK1, an enzyme that triggers the entry of progenitors into mitosis, is decreased when retinal cultures are incubated with the PI3K/Akt signaling inhibitor LY 294002 [27]. Since IGF-1 stimulated the phosphorylation of Akt, the effect of IGF-I and/or ADP in the regulation of CDK1 expression in developing retinal progenitors was investigated. Cultures were treated for 24 h with 250 µM ADP, 100 ng/mL IGF-I or both and CDK1 mRNA and protein levels determined as described in Materials and Methods. As can be observed in Fig. 8a, while no increase in the levels of CDK1 mRNA could be detected in cultures incubated only with ADP, cells treated with IGF-I showed a ~ 2.6-fold increase in the content of CDK1 mRNA, as compared to control non-treated cultures. Levels of CDK1 mRNA in cultures treated with both agonists were also significantly increased by ~ twofold, an increase that was not significantly different from the increase obtained by incubating the cultures only with IGF-I.

Fig. 8.

IGF-I increases CDK1 mRNA and protein levels and decreases phospho-histone H3 content in retinal cultures. Cultures were treated for 24 h with 250 µM ADP and/or 100 ng/mL IGF-1 and processed for detection of CDK1 mRNA (a), CDK1protein (b) and p-H3 (c) levels. (A) Data represent the mean ± S.E.M. of at least four independent experiments. **p < 0.01; ***p < 0.001 relative to control. (b, c) Blots are representative of at least 3 independent experiments. Protein gel loading was assessed with anti-ERK antiserum

In addition to the increase in the content of CDK1 mRNA, CDK1 protein levels were clearly up-regulated when the cultures were treated with IGF-1 (Fig. 8b), no matter if in the presence or absence of ADP (in arbitrary units: control, 90.068 ± 7.629; IGF-1, 164.922 ± 8.119; IGF-1 + ADP, 136.080 ± 18.510). Similar to CDK1 mRNA, no increase in CDK1 protein content was observed in cultures treated only with ADP (103.440 ± 4.117).

Levels of CDK1 are correlated with the progression of retinal progenitors through G2/M transition. Since cultures treated with IGF-I showed a significant increase in CDK1 expression, the effect of IGF-1 and/or ADP on the levels of phospho-Histone H3 (p-H3), an unambiguous marker of progenitors at G2 and M phases was investigated [32]. As can be noted in Fig. 8c, treatment with IGF-I, the absence or presence of ADP, promoted a substantial decrease in p-H3 expression that corresponded to ~ 14% and 15% of control levels, respectively (in arbitrary units: control, 137.115 ± 21.033; IGF-1, 19.158 ± 6.247; IGF-1 + ADP, 20.544 ± 6461). On the other hand, treatment of the cells only with ADP did not affect p-H3 expression that remained highly expressed as in control cultures (161.898 ± 20.086).

IGF-1 induced increase in CDK1 protein expression was blocked by 25 µM of the PI3K inhibitor LY 294002 (Fig. 9a). While treatment of the cultures with 100 ng/mL IGF-I increased CDK1 expression to 227% of the control levels, levels of CDK1 were similar to the levels observed in control cultures when the PI3K inhibitor was present. LY294002 also reduced CDK1 content by ~ 45% in control, non-treated cultures.

Fig. 9.

PI3K inhibition decreases IGF-I regulation of CDK1 expression and phosphorylation of Histone H3. Retinal cultures at E7C2 were treated with 100 ng/mL IGF-1 and/or 25 µM LY 294002 for 24 h. Protein extracts were processed for detection of CDK1 (a) or p-H3 (b) Protein gel loading was assessed with anti-ERK antiserum. Representative blots from at least 3 independent experiments are shown. Blots were quantified by densitometry and expressed the mean ± S.E.M. in the graphs. ^*p < 0.05 and ^**p < 0.01 as compared to control; ⁺p < 0.05 and ⁺⁺p < 0.01 as compared to IGF-1- treated cultures

The decrease in phospho-H3 expression mediated by IGF-1 was also affected by the addition of the PI3K inhibitor to the cultures (Fig. 9b). While IGF-1 decreased p-H3 content to ~ 25% of the control levels, the addition of LY 294002 to the IGF-1-treated cultures increased p-H3 content to ~ 55% of the control level. No apparent effect of LY294002 was noticed on p-H3 expression when cultures were incubated only with this inhibitor.

Since p-H3 is largely not detected following mitosis [32], the decrease in p-H3 content mediated by IGF-1 could be related to the passage of progenitors thought mitosis and entry in G1 phase. The effect of 250 µM ADP and/or 100 ng/mL IGF-1 on the expression of cyclin D1, cyclin E, CDK2 and p27/kip1is shown in Fig. 10. No increase in the expression of all these proteins was observed in cultures treated with ADP. Although not statistically significant, treatment of the cultures with IGF-1 induced an increase in cyclin D1 expression of approximately 180% of the levels observed in control cultures. Combined treatment of the cultures with ADP plus IGF-1 promoted a higher increase in cyclin D1 expression that reached ~ 230% of the control levels of non-treated cultures (Fig. 10a).

Fig. 10.

Modulation of the expression of cell cycle regulatory proteins induced by IGF-I and ADP in retinal cell cultures. Cultures at E7C2 were treated with 250 µM ADP and/or 100 ng/mL IGF-1 for 24 h and processed for detection of cyclin D1 (a), cyclin E (b), CDK2 (c) and P27/Kip1 (d) by western blotting. Tubulin was used as loading control. Quantification of blots is shown at right and data represent the mean ± S.E.M. of experiments performed in duplicate. *p < 0.05 and ***p < 0.001, versus control; ^##p < 0.01, relative to ADP-treated cultures. Blots are representative from at least 3 independent experiments

Treatment of the cells with IGF-I promoted an increase of ~ 216% of the control levels in the expression of cyclin E (Fig. 10b). Moreover, treatment of the cells with ADP + IGF-1 induced an increase of ~ 304% of control levels, an increase that was significantly higher than levels observed in cultures treated only with IGF-I.

Treatment of the cultures with Igf-1 also increased consistently the expression of CDK2, no matter if in the presence or absence of ADP (Fig. 10c). In contrast, no effect of IGF-1 was noticed in the expression of the CDK inhibitory protein P27/kip1 (Fig. 10d).

Discussion

In the chick embryo retina, ADP-sensitive nucleotide receptors induce the proliferation of late developing glia/bipolar progenitors through the sequential activation of phospholipase C, PKC and MAP kinases [12, 33] and activation of the PI3K/Akt pathway[28]. In the present work, we show that ADP-induced proliferation of retinal progenitors requires the activation of tyrosine kinase receptors. We investigated the participation of IGF-1 in this process as the presence of IGF-1 in the developing retina is well established [20–22, 24] and IGF-1 signaling increases the proliferation of developing progenitors [23]. The present work shows that incubation of the cultures with tyrphostin AG1024 or IOME AG 538, two antagonists of IGF-1 receptors, completely prevents the increase of [³H]-thymidine incorporation induced by ADP in chick embryo retinal cultures, suggesting that activation of IGF-1 receptors is required for ADP-dependent proliferation of retinal progenitors in culture. Activation of IGF-1 receptors could be due to their transactivation mediated by P2Y receptors or to their activation induced by IGF-1 present in the culture medium. Since no effect of ADP on [³H]-thymidine incorporation was observed in serum-deprived cultures, we favor the second possibility, although we cannot exclude completely the occurrence of transactivation mechanisms between P2Y and IGF-1 receptors. In good agreement with the second idea is the observation that fetal bovine serum contains significant amounts of IGF-1 and IGF-2 [29] and that developing retinas in culture release IGFs to the medium [23].

Besides proliferation, IGF-1 signaling is also involved in the survival of retinal cells [15, 16]. Thus, the decrease in [³H]-thymidine incorporation observed with the IGF-1 antagonists could be due to an increase in cell death induced by the inhibitors that would result in a smaller population of retinal progenitors incorporating [³H]-thymidine. This possibility however, can be ruled out since no decrease in the amounts of [³H]-thymidine that was incorporated prior to the treatment with AG 1024 or IOME- AG 538 was observed, suggesting that these compounds do not decrease the proliferation of retinal progenitors by decreasing their survival.

As mentioned above, ADP does not induce any increase in [³H]-thymidine incorporation in serum-starved cultures. Under this condition, ADP does not induce the phosphorylation of Akt, an enzyme that together with ERK is required for nucleotide-induced proliferation of retinal glial progenitors in culture [28]. In contrast to ADP, IGF-1 significantly stimulates the phosphorylation of Akt, 4E-BP1 and ERK in the serum-starved cultures. Moreover, IGF-1 induces Akt phosphorylation in many cultured cells, including PCNA⁺ positive progenitors. Thus, our findings strongly suggest that ADP-induced proliferation of retinal glia progenitors in culture requires the signaling of additional factors such as IGF-1 that stimulate PI3K/Akt pathway in progenitors.

Although levels of distinct cell cycle proteins including cyclins D1, E and CDK1 and 2 are elevated by IGF-1 in our retinal cultures, no increase in [³H]-thymidine incorporation is observed when cultures are incubated only with IGF-1, no matter if cultivated in the presence or absence of serum. No increase in PCNA expression was also noticed in cultures treated only with IGF-1. Previous studies have demonstrated that exogenous ADP is capable of inducing an increase in the intracellular calcium in retinal progenitors in culture [13]. Also, intracellular calcium mobilization and/or calcium capacity entry is required for DNA synthesis in S phase of the cell cycle in chick embryo retinas in culture [4, 34]. In the present study, calcium experiments showed that only ADP, but not IGF-1, is capable of inducing increases in [Ca²⁺]_I in the serum-free cultures, a condition where incubation of the cultures with ADP or IGF-1 separately does not induce any increase in [³H]-thymidine incorporation in the cells. Since only IGF-1, but not ADP, stimulates PI3K/Akt in the serum-deprived cultures, these findings suggest that the increase in [³H]-thymidine incorporation and PCNA expression in late developing retinal progenitors in culture occurs only when both signaling modules are stimulated. In this scenario, ADP-induced calcium mobilization and IGF-1-stimulated PI3K/Akt pathway would cooperate, together with other signaling modules, to induce DNA synthesis in cultured retinal progenitors. In good agreement with this idea is the present observations showing that incubation of serum-starved cultures with IGF-1 plus ADP indeed results in increases in [³H]-thymidine incorporation in the cultured cells.

Serum deprivation induces a decrease in viability of cultured retinal cells such as R28 neuroretinal precursors [35] and amacrine cells [36] that can be prevented by the addition of IGF-1 to the culture medium. IGF-1 also increases the survival of other retinal cell types, including photoreceptors [14] and ganglion cells [16]. Here, MTT viability assays show that IGF-1 increases significantly the viability of retinal cells in the cultures deprived of serum, no matter if in the absence or presence of ADP. In contrast, no effect of this factor is observed in [³H]-thymidine incorporation or PCNA labeling when the cultures are incubated only with IGF-1 and increase in DNA synthesis is observed only when cultures are incubated with IGF-1 plus ADP. These distinct responses of IGF-1 suggest that IGF-1 signaling is not directly related to DNA synthesis in retinal progenitors. In addition, since a large proportion of retina cells at the beginning of the cultures are postmitotic, it is possible that the IGF-1-mediated increase in cell viability observed in the MTT assays largely reflects an increase in the viability of postmitotic cells rather than an increase in the viability of proliferating progenitors. Accordingly, no relevant increase in the pre-incorporated [³H]-thymidine added prior to treatment is observed in cultures treated only with IGF-1. However, since IGF-1 increases the expression of cyclins D1 and E besides CDKs 1 and 2, we cannot completely exclude a beneficial influence of IGF-1 on the survival of cycling progenitors. To investigate the expression of proteins associated with cell death in retinal cultures under IGF-1 and/or ADP treatments may contribute elucidate to this point.

CDK1 is an enzyme that when associated with cyclin B1 controls several events in late G2 and M phases of the cell cycle that ultimately will lead progenitors to divide in mitosis. Several studies using different cell types suggest that Akt is a direct regulator of progenitor’s passage through G2/M transition of the cell cycle. Katayama et al. [37] showed that activation of Akt phosphorylates and inactivates Wee1Hu, preventing the inhibitory phosphorylation of CDK1 in HEK293T cells. In starfish oocytes, activation of Akt phosphorylates Myt and induces its down-regulation, favoring the meiotic G2/M transition of these cells [38]. In retinal progenitors in culture, phosphorylated Akt is observed over centrosomes of p-H3⁺ dividing cells and inhibition of PI3K/Akt pathway decreases CDK1 expression and impairs the progression of proliferating progenitors through the G2/M transition [27]. Our present results show that IGF-1 is able to increase CDK1 expression in a PI3K-dependent manner in retinal cell cultures. Moreover, content of mRNA for CDK1 also increases with the treatment with IGF-1, no matter if in the presence or not of ADP that does not modify CDK1 levels in the cultures. In good agreement with the present findings are previous results showing that IGF-I controls CDK1 mRNA as well as protein expression and activation in other types of cells such as fibroblasts, epithelial and tumor cells and oligodendrocyte progenitors [39–42].

Histone H3 phosphorylation on serine 10 (p-H3) begins late in G2 phase, increases during early mitosis and decreases after anaphase [32, 43, 44]. The present results show that IGF-1 induces a significant decrease in p-H3 levels, in the presence or not of ADP that does not affect the expression of p-H3 in the cultures. Although we cannot exclude the possibility that IGF-1 affects directly the phosphorylation of histone H3, one hypothesis to explain the decrease in the phosphorylation of this protein is that IGF-1 stimulates the passage of progenitors through mitosis. Alternatively, IGF-1 can induce the death of proliferating cells, decreasing the levels of p-H3. In good agreement with the first hypothesis is the observation that IGF-1 stimulates the expression of CDK1, enzyme that together with cyclin B1 triggers the entry of progenitors into mitosis [45]. Moreover, IGF-1 treatment increases the expression of cyclin D1, cyclin E and CDK2, proteins that increase and accumulate during G1 phase of the cell cycle [46]. Thus, under IGF-1 treatment, cultured progenitors most probably complete mitosis and progress through G1 phase of the cell cycle.

Isis Moraes Ornelas

- I.M. Ornelas is graduated with a PhD degree in Neuroscience from Universidade Federal Fluminense (UFF), in Rio de Janeiro, Brazil. She did postdoctoral trainning in Rutgers University, in New Jersey, USA, and DO’R Institute, in Rio de Janeiro. She is passionate about understanding the factors that regulate the development of the nervous system, and she has experience studying signaling pathways in retinal and oligodendrocyte progenitors and iPS-derived neural progenitor cells. Isis is currently a Professor at the Universidade Federal do Espírito Santo (UFES), Vitória-Brazil, where she teaches Human Physiology and is now dedicated to stablish her own research lab.

Authors’ contribution

Conceived and designed the experiments: IMO, GRF, ALMV; Performed experiments: IMO, TMS, GRF, MRP. Analyzed the data: IMO, GRF, MRP, TMS, ALMV; Manuscript writing: ALMV. All authors contributed to the article and approved the submitted version.

Meet the Author Guidelines

Isis Ornelas is graduated with a PhD degree in Neuroscience from Universidade Federal Fluminense (UFF), in Rio de Janeiro, Brazil. She did postdoctoral trainning in Rutgers University, in New Jersey, USA, and DO’R Institute, in Rio de Janeiro.

She is passionate about understanding the factors that regulate the development of the nervous system, and she has experience studying signaling pathways in retinal and oligodendrocyte progenitors and iPS-derived neural progenitor cells. Isis is currently a Professor at the Universidade Federal do Espírito Santo (UFES), Vitória-Brazil, where she teaches Human Physiology and is now dedicated to stablish her own research lab.

Current Address: I.M.O.: Department of Physiological Sciences, Universidade Federal do Espírito Santo, Vitória, Espírito Santo, 29,047–105, Brazil.

Funding

This work was supported by grants from Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ- E26-111.327/2013), Coordenação de Aperfeiçoamento de Pessoal do Ensino Superior (CAPES-AUXPE-PNPD2808/2010), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq-303018/2016–0) and Pró-reitoria de Pesquisa, Pós-graduação e Inovação (Proppi-UFF) to A.L.M.V.. During this work, I.M.O. and T.M.S. were recipients of post-doctoral fellowships from PNPD-CAPES.

Declarations

Ethical Approval

Experiments were carried out according to the rules issued by the National Council for Control of Animal Experimentation (CONCEA) and approved by the Ethics Committee for Animal Research of the Federal Fluminense University (CEUA, protocol number 806141221).

Competing interests

Authors declare no competing interest.