IGF-I stimulation of proteoglycan synthesis by chondrocytes requires activation of the PI 3-kinase pathway but not ERK MAPK

Abstract

The IGF-I (insulin-like growth factor-I) signalling pathway responsible for regulation of proteoglycan synthesis in chondrocytes has not been defined and is the focus of the present study. Chondrocytes isolated from normal human articular cartilage were stimulated with IGF-I in monolayer culture or in suspension in alginate. IGF-I activated members of both the PI3K (phosphoinositide 3-kinase) pathway and the ERK (extracellular-signal-regulated kinase)/MAPK (mitogen-activated protein kinase) pathway. The PI3K inhibitors LY294002 and wortmannin blocked IGF-I-stimulated Akt phosphorylation without blocking ERK phosphorylation and this was associated with complete inhibition of proteoglycan synthesis. A decrease in IGF-I-stimulated proteoglycan synthesis was also observed upon inhibition of mTOR (mammalian target of rapamycin) and p70S6 kinase, both of which are downstream of Akt. The MEK (MAPK/ERK kinase) inhibitors PD98059 and U0126 blocked IGF-I-stimulated ERK phosphorylation but did not block the phosphorylation of Akt and did not decrease proteoglycan synthesis. Instead, in alginate- cultured chondrocytes, the MEK inhibitors increased IGF-I-stimulated proteoglycan synthesis when compared with cells treated with IGF-I alone. This is the first study to demonstrate that IGF-I stimulation of the PI3K signalling pathway is responsible for the ability of IGF-I to increase proteoglycan synthesis. Although IGF-I also activates the ERK/MAPK pathway, ERK activity is not required for proteoglycan synthesis and may serve as a negative regulator.

Figure 1. IGF-I stimulation activates PI3K and ERK/MAPK signalling pathways in chondrocytes

Confluent chondrocytes in primary monolayer culture were switched to serum-free media, cultured overnight and then treated with IGF-I or vehicle as a control. Cell lysates were prepared and used for immunoblot analysis. (A) Dose–response for 10 min with 0, 0.2, 0.78, 3.125, 12.5 and 50 ng/ml IGF-I respectively. Cell lysates were immunoblotted with an antibody directed to phosphorylated Akt (Ser⁴⁷³). (B) Time course with 50 ng/ml IGF-I or 5 ng/ml IL-1β and immunoblotting with antibodies directed to phosphorylated ERK (pERK) or Akt [pAKT (473)] and, as control, non-phosphospecific (total) antibodies. (C) Time course with 50 ng/ml IGF-I and immunoblotting with phosphospecific antibodies and non-phosphospecific antibodies as labelled. Antibodies directed to two different phosphorylation sites on p70S6 kinase were tested (Thr³⁸⁹ and Thr⁴²¹/Ser⁴²⁴). Antibodies directed to a second phosphorylation site on Akt (Thr³⁰⁸) were also tested.

Figure 2. Effect of PI3K and MAPK inhibitors on IGF-I stimulation of Akt and ERK phosphorylation in chondrocytes

Chondrocytes were pretreated for 1 h with the PI3K inhibitor LY294002 at 25 μM or the MEK inhibitor PD98059 at 25 μM or both, followed by stimulation with 50 ng/ml IGF-I for 15 min. Cell lysates were prepared and used for immunoblotting. (A) Stimulation of Akt phosphorylation (pAKT Ser⁴⁷³) in cells cultured as monolayers. Samples were blotted for total Akt as a control. (B) Stimulation of ERK phosphorylation (pERK) in monolayer culture and total ERK as a control. (C) Stimulation of Akt phosphorylation (pAKT Ser⁴⁷³) in chondrocytes cultured in alginate. (D) Stimulation of ERK phosphorylation (pERK) in chondrocytes cultured in alginate.

Figure 3. Effect of mTOR inhibition on IGF-I stimulation of p70S6 kinase, Akt and ERK phosphorylation

Chondrocytes cultured in monolayer were pretreated for 1 h with the indicated doses of rapamycin followed by stimulation with 50 ng/ml IGF-I for 15 min. Cell lysates were prepared and used for immunoblotting. (A) Stimulation of P70S6 kinase phosphorylation (pP70S6kinase). The phospho-P70S6 kinase band is indicated by an arrow. The upper band is most probably P85S6 kinase, which is also detected by the antibody. Samples were blotted for total P70S6 kinase as a control. (B) Stimulation of Akt phosphorylation (pAKT Ser⁴⁷³) and total Akt as a control. (C) Stimulation of ERK phosphorylation (pERK) and total ERK as a control.

Figure 4. Inhibition of PI3K but not ERK blocks IGF-I-stimulated proteoglycan synthesis in chondrocytes cultured as monolayers or in alginate

Cultures were preincubated for 1 h with 25 μM LY294002, 25 μM PD98059 or both (LYPD) and then treated overnight with 50 ng/ml IGF-I. Proteoglycan synthesis was measured using [³⁵S]sulphate incorporation (c.p.m.) and DNA was analysed by a dye binding assay. Results are expressed as percentage of control treated with vehicle in the place of IGF-I. Chondrocytes were cultured either as monolayers (A–C) or in alginate (D) (n=6; aged 56, 57, 62, 69, 70 and 76 years); *P<0.001, IGF-I versus IGF-I+LY294002; **P=0.03, IGF-I versus IGF-I+LY294002; ***P=0.009, IGF-I versus IGF-I+PD98059. (E) Alginate cultures from four additional donors (aged 75, 81, 69 and 82 years) were preincubated for 30 min with 50 nM wortmannin or 10 μM U0126 and then treated overnight with or without 50 ng/ml IGF-I and analysed for proteoglycan synthesis as above. (*P=0.04, IGF-I versus IGF-I+wortmannin). (F) Results for proteoglycan synthesis in the cell-associated matrix for the experiment shown in (E). **P=0.002, IGF-I versus IGF-I+U0126.

Figure 5. Inhibition of mTOR or p70S6 kinase decreases proteoglycan synthesis in IGF-I-treated chondrocytes

Chondrocytes in monolayer culture were preincubated with the indicated concentrations of rapamycin (A) or Tos-Phe-CH₂Cl (B), and sulphate incorporation in response to IGF-I was measured as detailed in Figure 4. Data obtained in these experiments were collected from human donors; n=6, aged 62, 63, 66, 70, 72 and 78 years. *P<0.002, IGF-I versus IGF-I+rapamycin or Tos-Phe-CH₂Cl.

Figure 6. Inhibition of p70S6 kinase results in chondrocyte death

Chondrocytes were treated with (A) Tos-Phe-CH₂Cl or (B) rapamycin as detailed in Figure 5 and cell survival was measured using the LIVE/DEAD cell-survival assay.