Novel actions of IGFBP-3 on intracellular signaling pathways of insulin-secreting cells

Abstract

Understanding mechanisms underlying apoptotic destruction of insulin-secreting cells is critical to validate therapeutic targets for type 1 diabetes mellitus. We recently reported insulin-like growth factor binding protein-3 (IGFBP-3) as a novel mediator of apoptosis in insulin-secreting cells. In light of emerging IGF-independent roles for IGFBP-3, we investigated the mechanisms underlying actions of the novel, recombinant human mutant G⁵⁶G⁸⁰G⁸¹-IGFBP-3, which lacks intrinsic IGF binding affinity. Using the rat insulinoma RINm5F cell line, we report the first studies in insulin-secreting cells that IGFBP-3 selectively suppresses multiple, key intracellular phosphorelays. By immunoblot, we demonstrate that G⁵⁶G⁸⁰G⁸¹-IGFBP-3 suppresses phosphorylation of c-raf-MEK-ERK pathway and p38 kinase in time-dependent and dose-dependent manners. SAPK/JNK signaling was unaffected. These data delineate several novel intracellular sites of action for IGFBP-3 in insulin-secreting cells.

1. Introduction

Apoptotic destruction of insulin-secreting cells is the final common pathway of diverse insults causing type 1 diabetes mellitus in vivo [1-3]. Insulin-like growth factors (IGFs) and their high affinity IGF binding proteins (IGFBPs) comprise a complex super-family of multi-functional proteins regulating cell survival, fate, apoptosis, and metabolism [4]. Dating from the discovery of IGFBPs in the early 1980s, the first revision of the classic somatomedin hypothesis held that IGFBPs functioned solely as passive regulators of IGF bioavailability to the type 1 IGF receptors (heterotetrameric tyrosine kinases on the cell surface) [5]. IGFBP-3, as the most abundant IGFBP species in human serum, represents the prime regulator of IGF half-life in the circulation. Recently, several independent in vitro studies support the emerging concept of IGFBP-3 as a significant regulator of cell survival acting via IGF-independent mechanisms [6,7]. Studies of several mutant IGFBP-3 molecules suggest that the IGF-independent action of IGFBP-3 to promote apoptosis resides in its variable mid-region [8,9], which contains a heparin-like domain.

Triple substitution of glycine at three critical residues in the IGF binding domain of recombinant human (rh) IGFBP-3 eliminates intrinsic IGF binding affinity, while preserving other properties of the native molecule [10]. Global transgenic over-expression of this mutant G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 in the mouse resulted in increased brain size and modest increases in circulating IGF-I and IGFBP-3 levels [11]. Several studies with native IGFBP-3 have reported specific IGFBP-3 binding partners in the plasma compartment [12,13], within extracellular matrix [14], and inside the cell [15,16]. Neither a classic IGFBP-3 receptor nor signal transduction pathways restricted to IGFBP-3 action have been identified to date. Recently, we first reported IGFBP-3 as a novel mediator of apoptosis in insulin-secreting cells [17]. To identify the IGF-independent mechanisms of IGFBP-3 in insulin-secreting cells, we studied the actions of G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 using the rat insulinoma cell model RIN m5F.

2. Materials and methods

2.1. Cell culture and reagents

Derived from Rattus norvegicus, clonal insulin-secreting RIN-m5F cells were purchased from American Type Tissue Collection (Manassas, VA, USA) [18]. Cells were grown at 37 °C under humidified atmosphere with 5% CO₂ and RPMI 1640 medium containing 2 mM l-glutamine, adjusted to 1.5 g/L sodium bicarbonate, 4.5 g/L glucose, 10 mM HEPES, 1 mM sodium pyruvate, and 10% fetal bovine serum. Growth media was changed every third day.

pCMV6 vector containing G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 was generously provided by Dr. Caroline Buckway of Oregen Health Sciences University (Portland, OR, USA). Monoclonal antibody to tubulin was purchased from Sigma (St. Louis, MO, USA). Polyclonal antibody for IGFBP-3 was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Polyclonal antibodies to phospho-ERK, total ERK, phospho-MEK1/2, MEK1/2, phospho-c-raf, c-raf, phospho-p38, total p38, phospho-SAPK/JNK, and total SAPK/JNK were purchased from Cell Signaling Technology, Inc. (Beverly, MA, USA). Secondary antibodies conjugated with horseradish peroxidase were purchased from Jackson Immunologicals (West Grove, PA, USA). ECL for immunoblot development and signal detection was purchased from Amersham Biosciences (Piscataway, NJ, USA). Fluorescein-conjugated goat anti-rabbit IgG and mounting medium for fluorescence with DAPI were purchased from Vector Laboratories, Inc. (Burlingame, CA, USA). Chamber slides were obtained from NalgeNunc International (Rochester, NY, USA). Nitrocellulose membrane (0.2 μm) and Bio-Rad Protein Assay were purchased from Bio-Rad (Hercules, CA, USA). Lipofectamine and plus reagents were purchased from Invitrogen. Cell Death ELISA to detect mono- and oligonucleosome release during apoptosis was purchased from Roche Applied Science (Nonnenwald, Penzberg, Germany).

2.2. Transfection

Cells were seeded in 35 mm² cell culture dishes, then grown to 50–70% confluence. Cells were washed with phosphate-buffered saline, and 1.5 mL serum-free medium were added to each dish. pCMV vector with or without G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 cDNA (up to 4 μg for the dose experiments and 4 μg for the time course experiments), 4 μL lipofectamine reagent and 4 μL plus reagent in 400 μL serum-free medium were then added to each well. After incubation for 5 h, 1.5 mL medium containing 20% serum (to stimulate MAPK cascades) was added in each dish and incubated for the time indicated.

2.3. DNA fragmentation ELISA

Cells were plated at 70% confluence in 96-well plates and allowed to adhere overnight. G⁵⁶G⁸⁰G⁸¹-rhI-GFBP-3 and empty vector were transfected into cells as described above. To end transfection, the plate was centrifuged at 200g for 10 min to pellet the cells. Apoptosis in the floating and attached cells was quantified by cell death ELISA kit performed according to the manufacturer’s instructions. Briefly, cells resuspended in 200 μL 1X Lysis Buffer were incubated for 30 min at 25 °C. Lysates were centrifuged at 200g for 10 min, and 20 μL of cell lysates were transferred carefully into streptavidin-coated, 96-well plates. Immunoreagent 80 μL was added to each well, and plates were covered with adhesive foil under gentle shaking (300 rpm) for 2 h at 20 °C. Supernatant was thoroughly removed by gentle suction, then each well was rinsed thrice with 25 μL 1X incubation buffer. ABTS solution, 100 μL/well, was added to develop color detected at 405 nm (vs. 490 nm reference).

2.4. Cell growth rate quantitation

The cells were plated in the 6-well plate at about 50% confluence. After 24 h, the cells were transfected with or without 4 μg control vector or GGG-BP-3. After transfection for 24 or 48 h, the cells were trypsinized and counted on a hemacytometer.

2.5. Western blot

Transfection terminated when cells were washed with PBS. Two hundred microlitres of lysis buffer (containing 20 mM Tris, pH 7.5, 1% Nonidet P-40, 150 mM NaCl, 5 mM EDTA, 0.1 mM sodium ortho-vanadate, 1 mM PMSF, and 10 μL/mL aprotinin) were added to each 35 mm² cell culture dish. Whole cell lysates were collected using the cell scraper. Cell lysates were mixed and incubated on ice for 30 min. Lysates were cleared by centrifugation for 15 min at 13,000g. Protein concentrations were determined by Bio-Rad Protein Assay. Equal amounts of lysates were separated by 15% SDS–PAGE, and resolved proteins were transferred to nitrocellulose membranes. Membranes were blocked with 5% non-fat dried milk for 1 h at 25 °C, then incubated with primary antibodies for 3 h at 25 °C. After incubation with horseradish peroxidase-conjugated secondary antibodies, immunoreactive bands were detected by ECL, performed according to the manufacturer’s instructions. After detection of phosphoforms, we stripped membranes with 0.25 mM NaOH for 15 min at 25 °C, then re-probed with antibodies directed against the total specific protein.

2.6. Immunofluorescent microscopy

Cells plated in the chamber slides were grown to 50–70% confluence, then transfected with 0.25 μg DNA using Lipofectamine as described earlier. After two washes with PBS, cells were fixed with 4% paraformaldehyde (pH 7.5) for 20 min at 25 °C. Cells were rinsed twice with PBS and thrice washed with 1 mg/mL sodium borohydrate (5 min/wash) to reduce background. Cells were permeablized with 0.2% Triton X-100 for 5 min at 25 °C. After two washes with PBS, cells were blocked with 2% bovine serum for 1 h at 25 °C. Cells were then incubated with 10 μg/mL anti-IGFBP-3 antibody for 1 h at 25 °C. After two washes with PBS, cells were incubated for 1 h under darkness with fluorescein-conjugated goat anti-rabbit IgG, diluted 400-fold. After two washes in PBS, slides were covered under mounting medium containing DAPI. The 461 nm line from the Ar laser (Leica Inverted confocal microscope) induced fluorescence, signal captured by Himamatsu digital camera and processed with Microscope Control Module Software (QED Imaging, Inc., Pittsburgh, PA, USA).

2.7. Statistic analysis

Data from replicate experiments are expressed as means ± SEM, analyzed by ANOVA and post-hoc Kruskal–Wallis test using Instat (version 2.00; Graph-Pad Software Inc., San Diago, CA, USA). Probability values (p) below 5% were considered significant.

3. Results

3.1. G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 induces apoptosis and inhibits growth in RIN cells

Confirming our prior report with the native protein, we observed that mutant G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection induced apoptosis within 24 h as detected by oligonucleosome generation (Fig. 1;*p < 0.05 vs. transfection with empty vector). Cell growth was also inhibited in the GGG-rhIGFBP-3 transfected cells compared with the empty vector transfected cells (Fig. 2, p < 0.05 vs. empty vector).

Fig. 1.

Nucleosome ELISA. Cells were transfected with G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 or empty vector for 24 or 48 h in the presence of serum (n = 8, *p < 0.05 vs. control by ANOVA).

Fig. 2.

Cell growth rate study. Cells were transfected with empty vector (EV) or GGG-BP-3 for 24 or 48 h. The cells were then trypsinized and counted (p < 0.05 vs. control by ANOVA). Data were from 3 independent experiments.

3.2. Dose-dependent and time-dependent G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection

Dramatic, dose-dependent G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 synthesis was identified in whole cell lysates (Fig. 3A–B). Of note, mutant G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 migrated in a manner identical to native IGFBP-3 as detected by Western immunoblot: 43 and 45 kDa glycosylated doublet bands, a minor fraction in the proteolyzed 29 kDa form. Time-dependent transfection, up to 3 days in the presence of serum, produced consistent increases (not reduction) in IGFBP-3 synthesis (Fig. 3C–D). Confocal microscopy of these transfected cells disclosed immunofluorescent IGFBP-3 localized primarily in the cytoplasm; control cells transfected with empty vector showed no appreciable signal (Fig. 4).

Fig. 3.

Dose- and time-dependent G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection. Western blot of whole cell lysates after 24-h transfection using various amount of DNA (A) with densitometric analysis (B) and time course (C,D). Densitometry was normalized with respect to the tubulin control without transfection (**p < 0.01 vs. control by ANOVA). Data were from 3 independent experiments.

Fig. 4.

Confocal microscopy. Cells transfected 24 h with empty vector (A) or G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 (B). DAPI staining of nuclear heterochromatin (upper left panels); IGFBP-3 immunostain (upper right panels); light microscopy (lower left panels); DAPI and IGFBP-3 stain overlay (lower right panels).

IGFBP-3 is a secreted protein. The level of BP-3 in the medium was also analyzed by Western blot. It shows that the level of BP-3 is higher in the GGG-rhIGFBP-3 transfected cell medium than the level in the control medium (Fig. 5A–D).

Fig. 5.

Cells were transfected with empty vector (EV) or GGG-BP-3 as indicated for 48 h. Equal amount of medium were analyzed by Western blot (*p < 0.05, **p < 0.01 vs. control by ANOVA). Data were from 4 independent experiments.

3.3. G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 inhibits phosphorylation of ERK, MEK, c-raf, p38 kinase, and Akt

We systematically examined the downstream kinases of the mitogen activated protein kinase (MAPK) cascades due to their central roles for signaling survival over apoptosis. Despite the presence of serum, G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection significantly blocked ERK phosphorylation in a dose-dependent (Fig. 6A–B) and time-dependent manner (Fig. 6C–D) by 72 h. The upstream kinases MEK1/2 and c-raf were also inhibited by G⁵⁶G⁸⁰G⁸¹-rhI-GFBP-3 in both time- and dose-dependent manners (Figs. 7, 8). In the presence of serum, G⁵⁶G⁸⁰G⁸¹-rhI-GFBP-3 significantly inhibited p38 phosphorylation in a dose-dependent (Fig. 9A–B) and time-dependent manner (Fig. 9C–D). No effect was observed on SAPK/JNK signaling (Fig. 10). Fig. 11 illustrates confirmation of the major actions by G⁵⁶G⁸⁰G⁸¹-rhI-GFBP-3 vs. mock transfections at the 48-h time point.

Fig. 6.

Western blot for phospho-ERK in whole cell lysates. Dose-range study of 24-h G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection (A) with corresponding densitometry normalized to “0” control (B). Time course (C) with corresponding densitometry normalized to “0” control (D) (**p < 0.01 vs. control by ANOVA). Data were from 3 independent experiments.

Fig. 7.

Western blot for phospho-MEK1/2 in whole cell lysates. Dose-range study of 24-h G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection (A) with corresponding densitometry normalized to “0” control (B). Time course (C) with corresponding densitometry normalized to “0” control (D) (**p < 0.01 vs. control by ANOVA). Data were from 3 independent experiments.

Fig. 8.

Western blot for phospho-c-raf in whole cell lysates. Dose-range study of 24-h G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection (A) with corresponding densitometry normalized to “0” control (B). Time course (C) with corresponding densitometry normalized to “0” control (D) (**p < 0.01 vs. control by ANOVA). Data were from 3 independent experiments.

Fig. 9.

Western blot for phospho-p38 in whole cell lysates. Dose-range study of 24-h G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection (A) with corresponding densitometry normalized to “0” control (B). Time course (C) with corresponding densitometry normalized to “0” control (D) (*p < 0.05, **p < 0.01 vs. control by ANOVA). Data were from 3 independent experiments.

Fig. 10.

Western blot for phospho-SAPK/JNK in whole cell lysates. Dose-range study of 24-h G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection (A) and time course study (B). A representative blot is from 3 independent experiments.

Fig. 11.

Western blot of whole cell lysates. ERK, c-raf, MEK and p38 phosphorylation were not inhibited by transfection of empty vector (EV), but by G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection.

4. Discussion

We have presented the first evidence that IGFBP-3 inhibited specific MAP kinase pathway which may be the potential mechanism underlying IGFBP-3 apoptosis actions. Moreover, we have confirmed that these effects can be independent of this multifunctional protein’s intrinsic IGF binding affinity. In the presence of serum, G⁵⁶G⁸⁰G⁸¹-rhIGFBP-3 transfection significantly inhibited the phospho-c-raf-MEK-ERK pathway as well as the phospho-p38 in this insulin-secreting line. IGFBP-3 has been shown to block proliferation and to promote apoptosis by inhibiting mitogens acting via cell surface tyrosine kinases, including receptors for vascular endothelial growth factor and the type 1 IGF receptors[17,19-21].

Several novel binding partners for IGFBP-3 have been proposed to explain its IGF-independent actions [20], yet no classic receptor for any IGFBP has been accepted globally. Extracellular signal kinase 1/2 (ERK1/2) at the terminus of the MAPK cascade plays a critical supporting role for cell growth and differentiation [22-25]. Upregulation of ERK1/2 supports β-cell replication in mice induced by neuropeptide Y [26], which indicates ERK1/2 is an important regulation factor in pancreatic β-cell growth. Interrupting this pathway by BP-3 may lead to cell death. However, the relationship between this MAP kinase pathway and BP-3 apoptosis effect need to be evaluated.

p38, another MAP kinase, is activated by diverse cellular stresses, including osmotic shock, inflammatory cytokines, lipopolysaccharides, ultraviolet irradiation, and growth factors [27-31]. The inhibitory action of IGFBP-3 on this distal phosphorelay represents a novel, specific mechanism for its actions. The lack of effect on SAPK/JNK phosphorylation suggests that the effects of IGFBP-3 on MAP kinase regulation are kinase-specific or relay-specific.

Other data obtained from insulin-secreting cell models provide conflicting evidence for the complex roles of such phosphorelays in apoptosis. ERK activation appears necessary for interleukin-1β (IL-1β)-induced apoptosis in human β-cells [32]. The p38 inhibitor, SB203580, counteracts apoptosis induced by the combination of IL-1β and interferon-γ in RINm5F cells [33]. These observations suggest that the balance from multiple signaling transduction pathways determines cell fate, say, survival vs. apoptosis. Our use of transformed cells limits extrapolation to the situation in vivo with human islets. In conclusion, we have presented the first evidence that IGFBP-3 selectively inhibits serum-induced phosphorylation of major MAP kinases in insulin-secreting cells. Further investigations of this novel pathway underlying destruction of insulin-secreting cells are needed and this mechanism may be warranted to validate IGFBP-3 blockade as a novel therapeutic approach to preserve residual β-cell mass in the face of imminent clinical diabetes.