Cellular Oncology

, Volume 39, Issue 2, pp 149–159

Human IGF1 pro-forms induce breast cancer cell proliferation via the IGF1 receptor

  • Mauro De Santi
  • Giosuè Annibalini
  • Elena Barbieri
  • Anna Villarini
  • Luciana Vallorani
  • Serena Contarelli
  • Franco Berrino
  • Vilberto Stocchi
  • Giorgio Brandi
Original Paper

DOI: 10.1007/s13402-015-0263-3

Cite this article as:
De Santi, M., Annibalini, G., Barbieri, E. et al. Cell Oncol. (2016) 39: 149. doi:10.1007/s13402-015-0263-3

Abstract

Background

IGF1 is a key regulator of tissue growth and development and has been implicated in the initiation and progression of various cancers, including breast cancer. Through IGF1 mRNA splicing different precursor pro-peptides, i.e., the IGF1Ea, IGF1Eb and IGF1Ec pro-forms, are formed whose biological roles in the pathogenesis of breast cancer have not been established yet. The objective of this study was to assess the biological activity of the IGF1 pro-forms in human breast cancer-derived cells.

Methods

The different IGF1 pro-forms were generated through transient transfection of HEK293 cells with the respective vector constructs. The resulting conditioned media were applied in vitro to MCF7, T47D and ZR751 breast cancer-derived cell cultures. The recombinant human IGF1 pro-forms were also tested for their binding affinity to an anti-IGF1 specific antibody by immunoprecipitation. To determine whether the IGF1 pro-forms induce cell proliferation, mature IGF1 was neutralised in HEK293-derived conditioned media.

Results

We found that the IGF1 pro-forms were the only forms that were produced intra-cellularly, whereas both mature IGF1 and the IGF1 pro-forms were detected extra-cellularly. We also found that E peptides can impair the IGF1 pro-form binding affinity for the anti-IGF1 antibody and, thus, hamper an accurate measurement of the IGF1 pro-forms. Additionally, we found that the IGF1 antibody can completely inhibit IGF1-induced breast cancer cell proliferation and IGF1 receptor (IGF1R) phosphorylation, wheras the same antibody was found to only partially inhibit the biological activity of the pro-forms. Moreover, we found that the IGF1 pro-form activities can completely be inhibited by neutralising the IGF1R. Finally, we compared the bioactivity of the IGF1 pro-forms to that of mature IGF1, and found that the IGF1 pro-forms were less capable of phosphorylating the IGF1R in the breast cancer-derived cells tested.

Conclusions

Our data indicate that IGF1 pro-forms can induce breast cancer cell proliferation via the IGF1R, independent from the mature IGF1 form. These results underline the importance of an accurate assessment of the presence of IGF1 pro-forms within the breast cancer microenvironment.

Keywords

IGF1 pro-forms Breast cancer IGF1 receptor 

1 Introduction

Insulin-like growth factor-1 (IGF1) plays an important role in normal tissue growth and development. In addition, several studies have shown associations between circulating IGF1 levels and the risk to develop breast cancer [1, 2, 3]. Since the IGF1 receptor (IGF1R) is over-expressed in about 90 % of the breast cancer cases and since IGF1R levels are higher in breast cancer cells than in normal breast tissues [4], targeting the IGF1 system appears to be an attractive therapeutic option.

IGF1 is synthesized as a precursor protein requiring proteolysis at both the N- and C- termini to produce mature IGF1 [5, 6]. The full-length IGF1 precursor, pre-pro-IGF1, contains an N-terminal signal peptide, a 70 amino acid mature IGF1 peptide and a C-terminal E-peptide extension [7]. The signal peptide is cleaved off during translation in the endoplasmic reticulum, resulting in pro-IGF1. The E-peptide can subsequently be cleaved off from pro-IGF1 by proprotein convertases like furin, resulting in mature IGF1. The uncleaved pro-IGF1 has, however, also been detected in vitro in conditioned media and in vivo in sera [8, 9, 10, 11, 12, 13].

The complexity of the IGF1 system is further enhanced by alternative splicing of the IGF1 mRNA, thereby producing multiple IGF1 isoforms (IGF1 pro-forms) that, while bearing the same mature IGF1 sequence, contain different N- and C-terminal extensions [5]. In humans, the alternative splicing that occurs at the 3′ end of the IGF1 mRNA gives rise to three possible IGF1 pro-forms with different C-terminal extensions, called the Ea, Eb and Ec peptides (Fig. 1a). Another level of complexity results from glycosylation of the IGF1Ea pro-form, as the human Ea-peptide of IGF1 contains an N-linked glycosylation site at Asn92 [6].
Fig. 1

Human IGF1 pro-forms. a Schematic presentation of mature IGF1 and IGF1 pro-forms. b Western blot analysis of HEK293 cells transfected with specific constructs. Cell lysates and supernatants (SN) were analysed 24 h post-transfection using an anti-IGF1 antibody. Images are representative of three replicates giving similar results

Recent studies in humans have shown that the IGF1 splice variants can be differentially transcribed in response to varying conditions and pathologies, such as skeletal muscle damage [14, 15], endometriosis [16], or prostate [17], cervix [18] and colorectal cancer [19]. Moreover, although it is generally assumed that IGF1 exerts its biological actions predominantly through the mature peptide, different biological activities have been reported for the different IGF1 pro-forms and/or for their E-peptides, either exogenously administrated or over-expressed in various in vitro model systems [6, 14, 17, 20, 21].

Even though circulating IGF1 levels are affected by physical activity and diet [22], the biological significance of the IGF1 pro-forms is currently unknown. Also, the physiological and molecular mechanisms that regulate their expression and their circulating levels are unclear [6]. Despite the fact that the regenerative properties of the IGF-1Ea pro-form in cardiac and skeletal muscles has extensively been documented [21, 29, 30], little is known about the role of the various IGF1 pro-forms in cancer.

Here we report the biological activity of IGF1 pro-forms on human breast cancer-derived cell lines. We analysed the intracellular and extracellular expression patterns of the IGF1 pro-forms in transfected HEK293 cells and conditioned media. MCF7, T47D and ZR751 breast cancer-derived cells were grown in conditioned media to assess whether the IGF1 pro-forms induce cell proliferation and/or IGF1R phosphorylation. We further evaluated the bioactivities of the IGF1 pro-forms compared to the mature IGF1 form, in terms of cellular proliferation and IGF1R, AKT or ERK1/2 phosphorylation.

2 Materials and methods

2.1 Cell cultures

The MCF7, T47D, ZR751 and HEK293 cell lines were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The cells were cultured in DMEM (MCF7 and HEK293) or RPMI-1640 (T47D and ZR751) media supplemented with 10 % fetal bovine serum (FBS), 10 mg/L insulin (MCF7 and T47D), 2 mmol/L L-glutamine, 1× MEM Non-essential Amino Acid Solution, 0.1 mg/ml streptomycin and 0.1 U/L penicillin (growth media). Cells were maintained in a humidified incubator (5 % CO2) at 37 °C during at maximum fifteen passages.

For the experiments, the breast cancer-derived cells were starved overnight in red phenol-free DMEM or RPMI-1640 media without FBS, after which the media were replaced by the same media with or without hormones. All cell culture materials were purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.2 MTS cell proliferation assay

Triplicate samples of 5 × 103 MCF7, T47D and ZR751 cells in 96-well plates were treated for 4 days with mature IGF1 or IGF1 pro-forms. Cell viabilities were evaluated using a CellTiter 96® Aqueous Non-Radioactive Cell Proliferation Assay (Promega, Madison, WI, USA) based on the ability of viable cells to convert soluble tetrazolium salt (MTS) into a formazan product, as reported before [23]. The results are expressed as the relative number of viable cells in treated samples relative to controls (untreated cells).

2.3 Plasmid constructs

Plasmid constructs containing sequences encoding human prepro-IGF1Ea, prepro-IGF1Eb and prepro-IGF1Ec were kindly provided by Dr. Joanne Tonkin and Dr. Tommaso Nastasi, European Molecular Biology Laboratory (EMBL), Monterotondo (Rome, Italy). Each plasmid contained DNA encoding the class 1 IGF1 48-amino acid signal peptide, the mature 70-amino acid IGF1 peptide, the first 16 amino acids (aa) of the COOH-terminal peptide, and C-terminal sequences encoding either the Ea (19 aa), the Eb (61 aa) or the Ec (24 aa) peptide.

2.4 Cell transfection assays

HEK293 cells were cultured in DMEM without antibiotics at a density of 1x106/well in 6 well plates. After overnight incubation, the cells were transfected using a TransIT®-LT1 Transfection Reagent (Mirus Bio, Madison, WI, USA) according the manufacturer’s instructions. Briefly, 2.5 μg plasmid DNA was added to 250 μl growth medium without FBS and antibiotics and gently mixed, after which 7.5 μl TransIT®-LT1 Reagent was added. After a 30 min incubation at room temperature, the mixture was added drop-wise to the cells. After a 5 h incubation, the culture medium was replaced by red phenol-free DMEM without FBS. Next, the cells were incubated for another 24 h, after which supernatants were collected, clarified by 1000 rpm centrifugation for 5 min, and directly used or stored at −80 °C for further experiments. Transfected HEK293 cells were lysed for Western blotting or real-time PCR analyses. To increase IGF1 pro-form production, the furin convertase inhibitor chloromethylketone (CMK) (Enzo Life Sciences Inc., Farmingdale, NY, USA) was added at a 2.5 μmol/L final concentration during transfection. For E peptide cleavage, supernatants without CMK were treated with 10 nmol/L recombinant furin (R&D Systems Ltd., Minneapolis, MN, USA) overnight at room temperature while gently shaking [6].

2.5 ELISA assay

For the quantitative determination of human IGF1 concentrations in transfected HEK293 cell culture supernatants, a commercially available ELISA kit was used according to the manufacturer’s instructions (Quantikine® ELISA DG100, R&D Systems). Data were acquired in duplicate using a microplate reader (Multiskan EX, Thermo Fisher Scientific, Waltham, MA, USA) at 450 nm, after which the results were averaged.

2.6 Western blot analysis

MCF7, T47D, ZR751 and HEK293 cells were processed for Western blot analysis as previously reported [24]. Briefly, cells were lysed for 20 min on ice with 20 mmol/L HEPES (pH 7.9), 25 % v/v glycerol, 0.42 mol/L NaCl, 0.2 mmol/L EDTA, 1.5 mmol/L MgCl2, 0.5 % v/v Nonidet P-40, 1 mmol/L DTT, 1 mmol/L Naf, 1 mmol/L Na3VO4, and 1× complete protease inhibitor cocktail (Roche Diagnostics Ltd., Mannheim, Germany). The cell lysates were frozen and thawed twice and clarified by centrifugation at 12,000 rpm for 10 min at 4 °C. The proteins from the HEK293 cell supernatants were concentrated using an Amicon Ultra 3 K centrifugal filter unit (Merck Millipore, Billerica, MA, USA). Total cell lysates and concentrated supernatants were fractionated by SDS-PAGE and electroblotted onto nitrocellulose membranes (0.2 μm pore size) (Bio-Rad Laboratories Inc., Hercules, CA, USA). The resulting blots were probed with the following primary antibodies: anti-phospho-IGF1 Receptor β (3024), anti-IGF1 Receptor β (3027), anti-phospho-p44/42 (ERK1/2) (9101), anti-p44/42 (ERK1/2) (9102), anti-phospho-Akt (Ser473) (9271) and anti-Akt (9272), all purchased from Cell Signalling Technology (Beverly, MA, USA) and anti-IGF1 (I8773) purchased from Sigma-Aldrich. Protein bands were detected using a horseradish peroxidase-conjugated secondary antibody (Bio-Rad Laboratories Inc). The blots were treated with enhanced chemiluminescence reagents (ECL Kit, Amersham Bioscience, Arlington Heights, IL, USA), and the immunoreactive bands were detected and quantified using a Chemi-Doc System (Bio-Rad Laboratories Inc) equipped with Quantity One software.

2.7 RNA extraction, cDNA synthesis and qRT-PCR

Total RNA was extracted and purified using an Omega Bio-Tek E.Z.N.A.™ Total RNA kit (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. After digestion with DNase I (Qiagen, Hilden, Germany), cDNA was synthesized from 1 μg of total RNA using Omniscript RT (Qiagen) and random hexamers. Subsequently, quantitative real-time PCR was performed using an Applied Biosystems StepOnePlus™ Real Time PCR System in conjunction with a TaqMan® Universal PCR Master Mix No AmpErase® UNG and commercially available 6-carboxyfluorescein (FAM)-labeled TaqMan primers for human IGF1 (Hs01547656_m1) and GAPDH (Hs03929097_g1), respectively (Applied Biosystems, Foster City, CA, USA). mRNA expression data were generated using the 2–ΔCT method. The real-time PCR conditions were: 95 °C for 10 min followed by 40 cycles of two-steps at 95 °C for 15 s and 60 °C for 1 min. The specificity of the amplification products was confirmed by thermal denaturation plots and by separation in 4 % agarose gels.

2.8 Immunoprecipitation assay

To prepare magnetic beads for immunoprecipitation, Dynabeads® Protein G (Life Technologies, Monza, Italy) were washed twice with PBS/0.1 % Tween-20 and incubated with 5 μg of IGF1 antibody (Sigma) for 1 h at room temperature with end-over-end rotation. The bead-antibody complexes were washed with PBS/0.1 % Tween-20 after which the IGF1 monoclonal antibody was covalently bound to the beads using BS3 as cross-linkers according to the manufacturer’s instructions (Thermo Scientific, Milano, Italy). Subsequently, the beads were washed three times with PBS/0.1 % Tween-20 to remove non-covalently bound antibodies and incubated with 1 ml of tissue culture supernatant for 1 h at room temperature with end-over-end rotation. Finally, the beads were washed three times with washing buffer and the bound proteins were eluted by heating the beads for 10 min at 70 °C in 20 μl elution buffer and 10 μl SDS-PAGE sample buffer.

2.9 IGF1 and IGF1R neutralisation

To neutralise IGF1 activity, culture media containing IGF1 or IGF1 pro-forms were incubated with 3 μg/ml anti-IGF1 antibody (Sigma) for 1 h at 37 °C. Next, MCF7 and ZR751 cells were cultured in IGF1-neutralised media to evaluate their effects on cell proliferation and IGF1R phosphorylation. To neutralise IGF1R activity, cells were pre-incubated with 5 μg/ml of anti-IGF1R antibody (R&D Systems) for 1 h at 37 °C and treated with IGF1 or IGF1 pro-forms to evaluate their effects on cell proliferation and IGF1R phosphorylation.

2.10 Statistical analyses

Statistical analyses were performed using one-way or two-way ANOVA as appropriate, followed by Bonferroni’s multiple comparison post hoc tests (GraphPad Software, Inc., La Jolla, CA, USA).

3 Results

3.1 Expression of IGF1 and its pro-forms in HEK293 transfected cells

IGF1 pro-forms were generated through transient transfection of HEK293 cells with specific plasmid vector constructs for each pro-form. The resulting cell lysates and supernatants were analysed by Western blotting using an antibody directed against the mature region of IGF1. The amounts of IGF1 and its pro-forms in the supernatants were quantified by ELISA and concentrated using filter columns, after which 50 ng was loaded on gel. By doing so, we found that the IGF1 pro-forms were the only forms produced intra-cellularly by the transfected HEK293 cells, whereas both mature IGF1 and the IGF1 pro-forms were detected extra-cellularly (Fig. 1b). Notably, both glycosylated and non-glycosylated IGF1Ea were detected in the cell lysates, whereas only the glycosylated IGF1Ea pro-form (gly-IGF1Ea) was secreted. Moreover, we found that both IGF1Eb and IGF1Ec showed additional higher molecular weight bands, suggesting that also these pro-forms are subject to extensive post-translational modification.

3.2 E peptides impair an accurate quantification of the IGF1 pool

As previously reported [13], the quantity of the gly-IGF1Ea pro-form may be underestimated in non-denaturating conditions such as ELISA assays, suggesting that the E peptide could impair the binding affinity between IGF1 and anti-IGF1 antibodies. In order to evaluate whether the ELISA assay provides an accurate measure of the IGF pool, HEK293 cells were transfected with IGF1Ea, IGF1Eb and IGF1Ec expression vectors with or without the furin inhibitor CMK. In doing so, the same IGF1 mRNA expression efficiencies were obtained in CMK treated and untreated cells (Supplementary Fig. S1). Next, the supernatants were analysed by both ELISA and Western blotting after filter column concentration for IGF1 quantification. As shown in Fig. 2a, Western blot analysis of conditioned media from the CMK treated or untreated HEK293 cells did not show any significant variation in the total IGF1 pool (i.e., mature IGF1 and pro-forms). IGF1 quantification by ELISA of gly-IGF1Ea enriched medium did not show any change after CMK treatment, whereas a significant reduction of IGF1 was observed, after CMK treatment, in IGF1Eb and IGF1Ec enriched media (Fig. 2b). Therefore, we conclude that the E peptides in the IGF1 pro-forms may impair their affinity to the anti-IGF1 antibody under non-denaturating conditions and, hence, hamper the accuracy of the ELISA assay.
Fig. 2

Mature IGF1 and IGF1 pro-form affinities to anti-IGF1 antibody. Quantification of mature IGF1 and IGF1 pro-forms from a representative set (n = 3) of transfected HEK293 cells using a Western blot analysis and b ELISA. The furin convertase inhibitor chloromethylketone (CMK) was used to increase the IGF1 pro-form production. c Representative Western blot after immunoprecipitation of mature IGF1 and HEK293 supernatants containing IGF1 pro-forms using Dynabeads-anti-IGF1 complexes. Arrows indicate IGF1 pro-forms

To further confirm this notion, conditioned media obtained from HEK293 cells transfected with the IGF1 pro-form vectors were immunoprecipitated with Dynabeads coupled to an anti-IGF1 antibody. The proteins bound to the bead-antibody complex were subsequently recovered and analysed by Western blotting. As shown in Fig. 2c, the gly-IGF1Ea pro-form was, at least partially, recognized by the anti-IGF1 antibody, whereas the IGF1Eb and IGF1Ec pro-forms were only weakly recognized by the anti-IGF1 antibody in the immunoprecipitates. These results confirm that E peptides, in the IGF1 pro-forms, hamper an accurate measurement of the IGF1 pool.

3.3 Biological activity of IGF1Ea, IGF1Eb and IGF1Ec enriched media

Next, the activity of each IGF1 pool in the MCF7 and ZR751 human breast cancer-derived cells was evaluated in terms of cell proliferation and IGF1R phosphorylation. Cell proliferation was evaluated using a MTS cell proliferation assay, an indirect assay that evaluates the cellular metabolic activity. First, MCF7 cells were grown in the presence of increasing concentrations of mature IGF1. After 4 days of culture, the cellular proliferation was evaluated by both the MTS assay and by cell counting. As shown in supplementary Fig. 2S, both methods yielded similar results. Subsequently, the cells were cultured in IGF1 pro-form-enriched media, previously normalised to 10 ng/ml using an ELISA assay. Since it was not possible to accurately quantify the total IGF1 pool (i.e., mature IGF1 and pro-forms) in IGF1 pro-form-enriched media using an ELISA assay (see above), we were unable to directly compare the effects between each IGF1 pool. We found, however, that the IGF1 pro-form-enriched media significantly induced both MCF7 and ZR751 cell proliferation compared to the control (unstimulated) cells (Fig. 3a and b). It was not possible to evaluate the proliferation response in T47D cells due to their poor growth in serum free medium (not shown). Supernatants of un-transfected HEK293 cells or HEK293 cells transfected with an empty vector did not affect cell proliferation (not shown). Next, the anti-IGF1 antibody was used to neutralize the activity of mature IGF1. As shown in Fig. 3a and b, we found that the anti-IGF1 antibody completely inhibited the IGF1-induced cell proliferation, whereas the same antibody only partially inhibited MCF7 (Fig. 3a) and ZR751 (Fig. 3b) cell proliferation induced by HEK293 supernatants containing the IGF1 pro-forms. Moreover, we found that the anti-IGF1 antibody markedly inhibited the IGF1R phosphorylation induced by mature IGF1, but not the phosphorylation induced by the IGF1 pro-forms (Fig. 3c and d). These results suggest that the IGF1 pro-forms can induce breast cancer cell proliferation and IGF1R phosphorylation. The activity of the IGF1 pro-forms is IGF1R dependent. In fact, by inhibiting IGF1R activation with an anti-IGF1R antibody, neither cell proliferation nor IGF1R phosphorylation, induced by either mature IGF1 or the IGF1 pro-forms, were detected (Fig. 3).
Fig. 3

IGF1 pro-forms induce cell proliferation via the IGF1R. a MCF7 and b ZR751 cells were cultured 4 days with mature IGF1 (10 ng/ml) or HEK293 supernatants containing IGF1 pro-forms (means ± SEM; n = 3). Cell proliferation was evaluated by MTS assay. Data are expressed as relative proliferation vs. unstimulated cells. *** Significantly different, P < 0.001; ns: not significantly different; 1-way ANOVA followed by Bonferroni’s multiple comparison test. Representative Western blot (n = 3) of phospho-IGF-R levels in c MCF7 and d ZR751 cells stimulated for 10 min with mature IGF1 or HEK293 supernatants containing IGF1 pro-forms. IGF1R was used as a loading control. Densitometry values for specific proteins relative to unstimulated cells (set as one-fold) are included below the lanes. An anti-IGF1 antibody was used to neutralise the biological activity of IGF1. An anti-IGF1R antibody was used to inhibit IGF1R phosphorylation/activation

3.4 Biological activity of mature IGF1 versus the IGF1 pro-forms

To evaluate the activity of the IGF1 pro-forms compared to mature IGF1, supernatants containing different ratios of mature IGF1 and pro-forms were produced. Recombinant furine was used to induce E peptide cleavage and to increase the amount of mature IGF1. The furin convertase inhibitor CMK was used to inhibit E peptide cleavage and to increase the IGF1 pro-form amounts during transfection. The supernatants were concentrated using filter columns and analysed by Western blotting using an anti-IGF1 antibody. We found that CMK markedly increased the IGF1 pro-form amounts, while in the furine-treated supernatants the IGF1 pro-forms were not detectable (Fig. 4a, b and c).
Fig. 4

IGF1 pro-form production and E peptides cleavage. Representative Western blots are shown for supernatants of HEK293 cells transfected with specific constructs for the a IGF1Ea, b IGF1Eb and c IGF1Ec pro-forms. Mature IGF1 and IGF1 pro-forms were detected using an anti-IGF1 antibody. The furin convertase inhibitor CMK was used to increase IGF1 pro-form production. Recombinant furin was used to induce E peptides cleavage

Next, MCF7 cells were cultured in two-fold diluted conditioned media (from 1:4 to 1:32) containing different ratios of mature IGF1 and the IGF1 pro-forms, after which cellular proliferation and IGF1R, AKT and ERK1/2 phosphorylation were evaluated at the indicated time points. No significant differences in MCF7 cell proliferation were detected (supplementary Fig. S3a-b-c), but by increasing the amount of the gly-IGF1Ea pro-form in the cell culture medium, IGF1R phosphorylation was found to be markedly reduced (Fig. 5a), suggesting a minor binding affinity of gly-IGF1Ea for the IGF1R. While increasing the amount of gly-IGF1Ea also reduced AKT phosphorylation in MCF7 cells, no effect on ERK1/2 phosphorylation was observed (Fig. 5a). Decreased levels of IGF1R phosphorylation were also observed in MCF7 cells cultured with higher amounts of IGF1Ec and, partially, IGF1Eb, whereas no differences in AKT and ERK1/2 phosphorylation were observed (Fig. 5b and c).
Fig. 5

Phosphorylation of IGF1R, AKT and ERK1/2 in MCF7 cells. Representative Western blot (n = 3) showing phospho-IGFR, phospho-AKT and phospho-ERK1/2 levels in MCF7 cells stimulated for the indicated times with HEK293 supernatants containing mature IGF1 and the a IGF1Ea, b IGF1Eb and c IGF1Ec pro-forms. IGF1R, AKT and ERK1/2 were used as loading controls. Densitometry values for specific proteins relative to unstimulated cells (set as one-fold) are included below the lanes

The activity of the IGF1 pro-forms compared to mature IGF1 was also evaluated in T47D and ZR751 breast cancer-derived cells (Fig. 6). In conformity with the results obtained with MCF7 cells, we found that the glycosylated IGF1Ea and IGF1Ec pro-forms were less potent in phosphorylating IGF1R in both the T47D (Fig. 6a) and ZR751 (Fig. 6b) cells. Additionally, we found that furin and its inhibitor CMK did not alter the phosphorylation status of IGF1R, AKT and ERK1/2 when induced by mature IGF1 (supplementary Fig. S3d).
Fig. 6

Phosphorylation of IGF1R, AKT and ERK1/2 in a T47D and b ZR751 cells. Representative Western blot (n = 3) showing phospho-IGF1R, phospho-AKT and phospho-ERK1/2 levels in cells stimulated for 60 min with HEK293 supernatants containing mature IGF1 and the IGF1Ea, IGF1Eb and IGF1Ec pro-forms. IGF1R, AKT and ERK1/2 were used as loading controls. Densitometry values for specific proteins relative to unstimulated cells (set as one-fold) are included below the lanes

Fig. 7

Schematic presentation of IGF1 pro-form biological activity in breast cancer cells. a IGF1 induces IGF1R phosphorylation that is completely inhibited by neutralising IGF1 or IGF1R. b E peptide decreases IGF1R phosphorylation induced by IGF1. IGF1 neutralisation is ineffective in inhibiting IGF1R phosphorylation that, on the other hand, is completely inhibited by IGF1R neutralisation

4 Discussion

The IGF1 pathway plays a well-documented role in the development and/or progression of breast carcinomas [2]. IGF1 mRNA splicing events generate different precursor IGF1 polypeptides, namely the IGF1Ea, IGF1Eb and IGF1Ec pro-forms in humans, that share the mature peptide, but differ by the structure of their extension peptides, or E-peptides, in the C-terminus [5, 6]. The IGF1 pro-forms also undergo posttranslational modifications, such as glycosylation and proteolytic processing by proprotein convertases such as furin [6]. Convertase-mediated cleavage generally occurs intra-cellularly [25], but it has also been reported that there are potential proprotein convertases that may process pro-IGF1 extra-cellularly, resulting in the secretion of unprocessed IGF1 pro-forms [10, 13]. Our data confirm this latter notion by revealing that the IGF1 pro-forms are the predominant forms inside the transfected HEK293 cells, and that they are also abundantly secreted in the cell culture media. Our results also showed that the non-glycosylated IGF1Ea form was detectable in the cell lysates only, whereas only the glycosylated form was secreted. The Ea-peptide of human IGF1 is a unique E peptide that contains a N-linked glycosylation site, and it has been hypothesized that its glycosylation may play a role in IGF1 biological activity modulation, such as bioavailability [26]. Interestingly, our data revealed that both IGF1Eb and IGF1Ec are subject to posttranslational modifications. These modifications still require detailed characterization.

As reported by Durzyńska et al. [13], ELISA measurements are more sensitive to mature IGF1 than to the IGF1 pro-forms, suggesting that the presence of the E-peptide may impair the ability of the anti-IGF1 antibody to recognize the native protein. Our results support this hypothesis, showing that the anti-IGF1 antibody has a higher affinity for mature IGF1 compared to the IGF1 pro-forms, especially IGF1Eb and IGF1Ec. Moreover, according to the literature [13], the ELISA quantification appears to be impaired in supernatants with large amounts of pro-forms after CMK treatment during the transfections, even though our Western blotting results did not show a decrease in total IGF1. Thus, it is difficult to compare the bioactivities of the different pro-forms, as the E-peptide in pro-IGF1 hampers the ability to accurately measure and subsequently normalize the IGF1 content under non-denaturing conditions.

Since it was unclear whether pro-IGF1 is bioactive or simply an inactive precursor or source for mature IGF1 [7], we cultured MCF7 and ZR751 cells in IGF1-neutralised conditioned media. As expected, the anti-IGF1 antibody was able to completely neutralise the activity of mature IGF1 in terms of the induction of cell proliferation and IGF1R phosphorylation. On the contrary, the anti-IGF1 antibody was found to be ineffective in inhibiting the proliferation and IGF1R phosphorylation in cells cultured in conditioned media containing the IGF1 pro-forms (Fig. 7). These results suggest that the IGF1 pro-forms are able to induce breast cancer cell proliferation. In vitro studies have suggested that the E-peptides of the human IGF1 precursors may act as independent growth factors, inducing mitosis independently from IGF1R [6]. In contrast, we found that by neutralising the IGF1R, the induction of cell proliferation by mature IGF1 or the IGF1 pro-forms was completely inhibited, suggesting that IGF1 pro-forms induce cell proliferation via IGF1R activation (Fig. 7).

Despite the vast amount of evidence regarding the biological activity of E peptides, little is known about the biological activities of the IGF1 pro-forms [6, 13, 21, 27]. Here, we evaluated the biological activities of the IGF1 pro-forms compared to those of mature IGF1. To this end, we generated a set of conditioned media containing different ratios of mature IGF1 and IGF1 pro-forms by using the proprotein convertase furin to induce pro-form cleavage and to increase the mature IGF1 amounts and, on the other hand, by using the convertase inhibitor CMK during transfection to inhibit pro-form cleavage and to increase the IGF1 pro-form amounts. We found that after culturing MCF7, T47D and ZR751 cells with increasing amounts of the IGF1 proform, phosphorylation of the IGF1R markedly decreased (Fig. 7). This result correlates with recent data showing that glycosylated pro-IGF1Ea is less efficient in receptor activation than pro-IGF1 and mature IGF1 [13]. Despite the finding that the pro-forms decreased the activation of the IGF1R, no significant differences were observed in cellular proliferation and ERK1/2 phosphorylation compared to mature IGF1. Interestingly, AKT phosphorylation in MCF7 cells seems to be affected by gly-IGF1Ea. It has previously been suggested that IGF1Ea may activate alternative IGF1R downstream pathways [6], as the canonical PI3K/AKT/mTOR signaling pathway was not induced in transgenic mice over-expressing IGF1Ea [28, 29]. The effect of gly-IGF1Ea on AKT phosphorylation in breast cancer cells requires, however, independent confirmation.

In conclusion, we found that IGF1 pro-forms can induce breast cancer cell proliferation via IGF1R phosphorylation. There are other data supporting a role of the IGF1 pro-forms in cancer, such as prostate [17], cervical [18] and colorectal cancer [19]. As yet, however, the biological activity of IGF1 variants in breast cancer development has not been established, and no analytical methods are available to correctly detect and quantify the IGF1 pro-forms. In fact, the available methods rely on the use of antibodies that primarily recognise the mature IGF1 peptide, thereby underestimating the pro-forms. The low specificity of anti-IGF1 antibodies to the pro-forms could also have implications for the design of breast cancer therapies, since current targeted strategies include anti-IGF1 antibodies to neutralise the IGF system [31]. The low affinities of pro-IGF1s for the IGF1R, together with the poor prognosis associated with high IGF1R expression, make the search for regulatory mechanism(s) and potentially specific bioactivities of the various IGF1 peptides an area of particular interest. Further studies will focus on the identification of the pro-IGF1s as candidate prognostic factors.

Acknowledgments

The authors would like to thank Dr. Joanne Tonkin and Dr. Tommaso Nastasi for kindly providing the IGF1 pro-form specific constructs for the cell transfections.

Compliance with ethical standards

Funding

This study was supported by the RF-2009-1,532,789 Ministry of Health Project-Italy.

Conflict of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Supplementary material

13402_2015_263_MOESM1_ESM.pdf (169 kb)
Supplementary Fig. S1(PDF 169 kb)
13402_2015_263_MOESM2_ESM.pdf (174 kb)
Supplementary Fig. S2(PDF 173 kb)
13402_2015_263_MOESM3_ESM.pdf (369 kb)
Supplementary Fig. S3(PDF 369 kb)

Funding information

Funder NameGrant NumberFunding Note
Ministry of Health - Italy
  • RF-2009-1532789

Copyright information

© International Society for Cellular Oncology 2015

Authors and Affiliations

  • Mauro De Santi
    • 1
  • Giosuè Annibalini
    • 2
  • Elena Barbieri
    • 2
  • Anna Villarini
    • 3
  • Luciana Vallorani
    • 2
  • Serena Contarelli
    • 2
  • Franco Berrino
    • 3
  • Vilberto Stocchi
    • 2
  • Giorgio Brandi
    • 1
  1. 1.Department of Biomolecular Sciences, Hygiene UnitUniversity of Urbino Carlo BoUrbinoItaly
  2. 2.Department of Biomolecular Sciences, Exercise and Health Sciences UnitUniversity of Urbino Carlo BoUrbinoItaly
  3. 3.Epidemiology & Prevention Unit, Department of Preventive & Predictive MedicineFondazione IRCCS Istituto Nazionale dei TumoriMilanItaly

Personalised recommendations