Selective modulation of type 1 insulin-like growth factor receptor signaling and functions by beta1 integrins

Abstract

We show here that beta1 integrins selectively modulate insulin-like growth factor type I receptor (IGF-IR) signaling in response to IGF stimulation. The beta1A integrin forms a complex with the IGF-IR and insulin receptor substrate-1 (IRS-1); this complex does not promote IGF-I mediated cell adhesion to laminin (LN), although it does support IGF-mediated cell proliferation. In contrast, beta1C, an integrin cytoplasmic variant, increases cell adhesion to LN in response to IGF-I and its down-regulation by a ribozyme prevents IGF-mediated adhesion to LN. Moreover, beta1C completely prevents IGF-mediated cell proliferation and tumor growth by inhibiting IGF-IR auto-phosphorylation in response to IGF-I stimulation. Evidence is provided that the beta1 cytodomain plays an important role in mediating beta1 integrin association with either IRS-1 or Grb2-associated binder1 (Gab1)/SH2-containing protein-tyrosine phosphate 2 (Shp2), downstream effectors of IGF-IR: specifically, beta1A associates with IRS-1 and beta1C with Gab1/Shp2. This study unravels a novel mechanism mediated by the integrin cytoplasmic domain that differentially regulates cell adhesion to LN and cell proliferation in response to IGF.

Figure 1.

IGF stimulates adhesion to LN-1 of β_1C-expressing cells. β_1C-CHO (clones 16.3 and 16.4) and β_1A-CHO (clones 10.18 and 10.2) clones were cultured in the absence (A) or in the presence (B) of tet for 48 h. Cells were labeled using ⁵¹Cr-sodium chromate for 1 h at 37°C. ⁵¹Cr-labeled cells were incubated in the presence or in the absence of purified rabbit Ab to IGF-II or ni-rIgG for 1 h on ice. Cells were plated on BSA, LN-1, or FN at 37°C for 2 h in the presence or in the absence of IGF-II. Attached cells were washed, lysed, and the amount of ⁵¹Cr associated with the attached cells was measured by liquid scintillation counting. Data are expressed as mean ± SEM. (C) Representative β_1C-CHO and β_1A-CHO clones were cultured in the absence or in the presence of tet for 48 h and analyzed by FACS^® using a β₁ integrin Ab TS2/16 specific to human β₁ or 7E2 specific to hamster β₁ or, as negative control, 12CA5.

Figure 2.

Down-regulation of β_1C expression in 267B1 cells inhibits IGF-I–stimulated cell adhesion to LN-1. (A) PrEC and PC3 cells were serum starved for 12 h, detached, and plated on LN-1 or FN at 37°C for 2 h in the presence or in the absence of IGF-I. Cell adhesion was analyzed by crystal violet staining. (B) pRNS-1-1 and 267B1 cells were lysed and immunoblotted with mAb to β_1C or normal IgM. (C) 267B1 cells were transiently transfected with pBJ1-RZ-β_1C or vector alone, lysed, and immunoblotted with mAb to β_1C, Ab to Akt, or Ab to IGF-IR. CHO-β_1C cell lysate was used as a positive control. (D) 267B1 cells were transiently transfected with pBJ1-RZ-β_1C or vector alone were detached and seeded on BSA or LN-1 or FN-coated plates at 37°C for 2 h in the presence or in the absence of IGF-I and stained with β-gal. Cell attachment was expressed as percentage of cells transfected with vector and attached to LN-1 in the absence of IGF-I, set at 100. Data are expressed as mean ± SEM. The experiments were repeated at least twice with similar results.

Figure 3.

β_1C enhances IGF-I– and II–mediated PC3 cell adhesion to LN-1. (A–D) β_1A- and β_1C-PC3 clones were cultured in the presence or in the absence of tet for 48 h. Cells were detached and plated on BSA or LN-1 or FN at 37°C for 2 h in the presence of IGF-II (A and B), of IGF-I (C and D), or of Ab to LN-1 (D) or ni-rIgG (D). In A, the differences in cell adhesion to LN-1 in the presence or in the absence of tet are statistically significant (*P ≤ 0.03). In C, the differences in cell adhesion to LN-1 between β_1A and β_1C expressing cells are statistically significant (*P ≤ 0.024). The experiments were repeated at least three times with similar results using two clones each of β_1A- and β_1C-PC3 cells. (E and F) β_1A- and β_1C-PC3 clones were cultured in the presence or in the absence of tet for 48 h. Cells were serum starved for 12 h, detached, and plated on LN-1 (E), FN or BSA (F) in serum-free medium at 37°C for 2 h in the presence or in the absence of IGF-I or GoH3 or rtIgG. Cell adhesion was analyzed by crystal violet staining. Data are expressed as mean ± SEM. (G) Surface expression of endogenous or exogenous β_1A or β_1C integrin or endogenous α₆ integrin was analyzed in PC3 cells by FACS^® using Ab to human β₁, TS2/16, chicken β₁, W1B10, α₆, GoH3; or, as negative controls, mIgG, rtIgG, or 12CA5.

Figure 4.

β_1A and β_1B differentially affect IGF-mediated cell proliferation and tumor growth. (A) β_1A- and β_1C-PC3 clones were cultured in the absence of tet for 48 h. Cells were plated on a 96-well plate in serum-free medium with or without IGF-I or IGF-II at 37°C. After 72 h of incubation, cell proliferation was analyzed using SRB. As control, sulforhodamine B (SRB) incorporation of cells attached to FN-coated substrates was measured in parallel (black bars). (B and C) PC3-β_1A (B) and PC3-β_1C (C) clones were injected subcutaneously in athymic male Balb/c mice. Mice were given water supplemented with either 5% sucrose to induce β_1A or β_1C expression, or 5% sucrose plus 100 μg/ml tet. The graphs show kinetics of tumor growth. Tumor growth is expressed as tumor volume in cubic millimeters. Data are the mean ± SEM of 10 animals per group. *P ≤ 0.001 at 10, 12, 14, and either 16 or 17 d from injection. These experiments were repeated three times with similar results.

Figure 5.

IGF-IR and β_1C act synergistically to support cell adhesion to LN-1. (A) R− and R+ cells were transiently transfected with human β_1A or β_1C. Surface expression of β_1A or β_1C was analyzed by FACS^® using TS2/16 or 12CA5 as a negative control. Thick line, TS2/16; thin line, 12CA5. (B) R+ cells transiently transfected with β_1A or β_1C were detached and seeded on BSA or LN-1–coated plates at 37°C for 2 h in the presence or in the absence of IGF-II and stained with β-gal. (C) R− and R+ cells (10⁶) transiently transfected with β_1A or β_1C were incubated with or without P4C10 (a-β₁) or 1C10 (neg-cont) on ice for 1 h. Cells were plated on BSA or LN-1 or FN at 37°C for 2 h and stained with β-gal. Attachment of cells transfected with β₁-integrin cDNA was expressed as percentage of cells transfected with pBJ (B) or pBJ-β_1A (C) that were attached to LN-1, set at 100. The experiments were repeated at least twice with similar results. Data are expressed as mean ± SEM.

Figure 6.

β_1C inhibits IGF-IR tyrosine phosphorylation and β_1A association with IGF-IR. β_1A- and β_1C-PC3 clones were cultured in the presence or in the absence of tet for 48 h, stimulated with IGF-I for 10 min at 37°C, lysed. Exogenous β₁ was immunoprecipitated using CSAT (A) and endogenous β₁ using K-20 (C). In A, immunoblotting (IB) was performed using CSAT or Ab to IGF-IR-β. (B) Cells were incubated with or without IGF-I for 0, 5, 10, and 20 min in suspension at 37°C. Immunoprecipitation was performed using Ab to IGF-IR-β. Immunoprecipitates were separated on SDS-PAGE and immunoblotted with PY20 or Ab to IGF-IR-β. In C, the endogenous β₁ immunoprecipitates were reprecipitated and immunoblotted with Ab to IGF-IR-β. (D) CHO clones expressing the full-length β_1A integrin or β₁COM or β_1CΔ802 truncated mutant were cultured for 48 h. Cells were lysed and proteins were immunoprecipitated with K-20. The immunocomplexes were dissociated and proteins were reimmunoprecipitated with Ab to IGF-IR or ni-rIgG. The immunoprecipitates were separated using SDS-PAGE and immunoblotted with Ab to IGF-IR. (E) Schematic representation of β_1C cytoplasmic domain deletion mutants. The experiments were repeated at least twice with similar results.

Figure 7.

β_1A associates with and supports tyrosine phosphorylation of IRS-1, whereas β_1C associates with and supports tyrosine phosphorylation of Gab1 and Shp2. (A) CHO clones expressing either β_1C or β_1A integrin were lysed and β₁ integrin was immunoprecipitated using K-20. (A) The β₁ immunoprecipitates were reprecipitated with Ab to IGF-IR or ni-rIgG and immunoblotted with polyclonal Ab to IGF-IR (top). As control, proteins from CHO cell lysates were immunoprecipitated with Ab to β₁ integrin or ni-mIgG and immunoblotted with Ab to β₁ integrin (Ab13, bottom). (B) CHO clones expressing either β_1C or β_1A integrin were incubated with or without IGF-I for 10 min, lysed, and proteins were immunoprecipitated with Ab to β₁ integrins (K-20) and immunoblotted with Ab to IRS-1 (top), Gab1 (middle), or β₁ integrin (bottom). (C–E) CHO clones expressing either β_1C or β_1A integrin were incubated with or without IGF-I for 10 min, lysed, and proteins were immunoprecipitated with Ab to IRS-1 (C) or Gab1 (D and E). Immunoprecipitates were separated using SDS-PAGE and immunoblotted with Ab to p-Tyr (PY20) (C and D), IRS-1 (C), Gab1 (D and E), Shp2 (E), or phospho-Shp2 (E). (F) β_1A- and β_1C-CHO clones were transiently transfected with wt-Shp2 or Shp2 C/S or vector alone. Cells were then cultured in the absence of tet for 48 h, plated on LN-1 and incubated for 2 h at 37°C in the presence or in the absence of IGF-I. After incubation, cells were fixed and stained with β-gal. Attachment of cells transfected with wt-Shp2 or Shp2 C/S cDNA was expressed as percentage of the number of attached cells transfected with vector alone, set at 100 in the absence of IGF-I. Data are expressed as mean ± SEM. (G) β_1A- and β_1C-CHO clones were transiently transfected with either HA-tagged wt-Shp2 or FLAG-tagged Shp2 C/S or vector alone. ERK1 was used as loading control. Cells were lysed after 48 h and immunoblotted with Ab to HA for wt-Shp2 or Ab to FLAG for Shp2 C/S. The experiments were repeated at least three times with similar results. White lines indicate that intervening lanes have been spliced out.

Figure 7.

β_1A associates with and supports tyrosine phosphorylation of IRS-1, whereas β_1C associates with and supports tyrosine phosphorylation of Gab1 and Shp2. (A) CHO clones expressing either β_1C or β_1A integrin were lysed and β₁ integrin was immunoprecipitated using K-20. (A) The β₁ immunoprecipitates were reprecipitated with Ab to IGF-IR or ni-rIgG and immunoblotted with polyclonal Ab to IGF-IR (top). As control, proteins from CHO cell lysates were immunoprecipitated with Ab to β₁ integrin or ni-mIgG and immunoblotted with Ab to β₁ integrin (Ab13, bottom). (B) CHO clones expressing either β_1C or β_1A integrin were incubated with or without IGF-I for 10 min, lysed, and proteins were immunoprecipitated with Ab to β₁ integrins (K-20) and immunoblotted with Ab to IRS-1 (top), Gab1 (middle), or β₁ integrin (bottom). (C–E) CHO clones expressing either β_1C or β_1A integrin were incubated with or without IGF-I for 10 min, lysed, and proteins were immunoprecipitated with Ab to IRS-1 (C) or Gab1 (D and E). Immunoprecipitates were separated using SDS-PAGE and immunoblotted with Ab to p-Tyr (PY20) (C and D), IRS-1 (C), Gab1 (D and E), Shp2 (E), or phospho-Shp2 (E). (F) β_1A- and β_1C-CHO clones were transiently transfected with wt-Shp2 or Shp2 C/S or vector alone. Cells were then cultured in the absence of tet for 48 h, plated on LN-1 and incubated for 2 h at 37°C in the presence or in the absence of IGF-I. After incubation, cells were fixed and stained with β-gal. Attachment of cells transfected with wt-Shp2 or Shp2 C/S cDNA was expressed as percentage of the number of attached cells transfected with vector alone, set at 100 in the absence of IGF-I. Data are expressed as mean ± SEM. (G) β_1A- and β_1C-CHO clones were transiently transfected with either HA-tagged wt-Shp2 or FLAG-tagged Shp2 C/S or vector alone. ERK1 was used as loading control. Cells were lysed after 48 h and immunoblotted with Ab to HA for wt-Shp2 or Ab to FLAG for Shp2 C/S. The experiments were repeated at least three times with similar results. White lines indicate that intervening lanes have been spliced out.

Figure 8.

PI 3-kinase mediates IGF-stimulated adhesion to LN-1. (A) β_1C-CHO clones were transiently transfected with vector alone (pcDNA3), wt-p85, or DNp85. Cells were cultured in the absence of tet for 48 h. 1.5 × 10⁵ cells were plated per well coated with LN-1 and incubated for 2 h at 37°C in the presence or in the absence of 100 ng/ml IGF-II. Cells were washed, fixed, and stained with β-gal. Attachment of cells transfected with PI 3-kinase cDNA was expressed as percentage (average and SD) of the number of attached cells that were transfected with vector alone, set at 100. Data are expressed as mean ± SEM. (B) β_1A- or β_1C-CHO clones were stimulated with IGF-I or IGF-II for 10 min, lysed and immunoblotted with Ab to Akt or Ab to phospho-Akt. Experiments were repeated at least twice with similar results. Results using representative clones are shown.

Figure 9.

IGF binding to IGF-IR mediates adhesion of β_1C expressing cells to LN-1. (A) β_1A- and β_1C-CHO clones were transiently transfected with 486/STOP or vector alone. Cells were cultured in the absence of tet for 48 h. Cells were plated on either LN-1 or FN and incubated for 2 h at 37°C in the presence or in the absence of IGF-II. After incubation, cells were fixed and stained with β-gal. The number of attached cells transfected with 486/STOP cDNA was expressed as percentage of the number of attached cells transfected with vector alone, set at 100. (B and C) β_1C- and β_1A-PC3 clones were cultured in the absence of tet for 48 h. Cells were plated on BSA or LN-1 (B) or FN (C) at 37°C for 2 h in the presence or in the absence of IGF-I or α-IR3 or ni-mIgG. The experiments were repeated at least three times with similar results. Data are expressed as mean ± SEM.

Figure 10.

A model for the biological effects of the cross-talk between IGF-IR and β₁ integrins. This model shows that the β_1C integrin forms a complex with and activates Gab1/Shp2; this results in recruitment of Shp2 to IGF-IR and consequently, IGF-IR dephosphorylation. The signaling events stimulated by β_1C expression results in increased cell adhesion to laminin, reduced cell proliferation and inhibition of tumor growth. When down-regulation of β_1C occurs, IGF-IR remains tyrosine phosphorylated and associated with β_1A integrin, and this results in increased tumor growth and cell proliferation, but in reduced cell adhesion to LN.