Differential effects of ERK and p38 signaling in BMP-2 stimulated hypertrophy of cultured chick sternal chondrocytes

Abstract

BACKGROUND: During endochondral bone formation, the hypertrophy of chondrocytes is accompanied by selective expression of several genes including type X collagen and alkaline phosphatase. This expression is stimulated by inducers including BMPs and ascorbate. A 316 base pair region of the type X collagen (Col X) promoter has been previously characterized as the site required for BMP regulation. The intent of this study was to examine the role of Mitogen Activated Protein (MAP) and related kinase pathways in the regulation of Col X transcription and alkaline phosphatase activity in pre-hypertrophic chick chondrocytes. RESULTS: Using a luciferase reporter regulated by the BMP-responsive region of the type X collagen promoter, we show that promoter activity is increased by inhibition of extra-cellular signal regulated kinases 1 or 2 (ERK1/2). In contrast the ability of BMP-2 to induce alkaline phosphatase activity is little affected by ERK1/2 inhibition. The previously demonstrated stimulatory affect of p38 on Col X was shown to act specifically at the BMP responsive region of the promoter. The inhibitory effect of the ERK1/2 pathway and stimulatory effect of the p38 pathway on the Col X promoter were confirmed by the use of mutant kinases. Inhibition of upstream kinases: protein kinase C (PKC) and phosphatidylinositol 3-(PI3) kinase pathways increased basal Col X activity but had no effect on the BMP-2 induced increase. In contrast, ascorbate had no effect on the BMP-2 responsive region of the Col X promoter nor did it alter the increase in promoter activity induced by ERK1/2 inhibition. The previously shown increase in alkaline phosphatase activity induced by ascorbate was not affected by any kinase inhibitors examined. However some reduction in the alkaline phosphatase activity induced by the combination of BMP-2 and ascorbate was observed with ERK1/2 inhibition. CONCLUSION: Our results demonstrate that ERK1/2 plays a negative role while p38 plays a positive role in the BMP-2 activated transcription of type X collagen. This regulation occurs specifically at the BMP-2 responsive promoter region of Col X. Ascorbate does not modulate Col X at this region indicating that BMP-2 and ascorbate exert their action on chondrocyte hypertrophy via different transcriptional pathways. MAP kinases seem to have only a modest effect on alkaline phosphatase when activity is induced by the combination of both BMP-2 and ascorbate.

Figure 1

Effects of BMP-2 and the ERK inhibitor, UO126, on the Col X promoter and alkaline phosphatase activity in chick cephalic sternal chondrocytes. A: Activity of the b2/640 type X collagen promoter; 24 hrs after seeding cells were transfected with PGLb2/640 and pRLnull luciferase vectors, 5 hrs after transfection 4 μM U0126 or vehicle (DMSO) was added. BMP-2 was added to selected wells after a further hour. Values are mean ± S.D of the mean ratio of promoter to empty vector fluorescence units, for 6 experiments assayed in triplicate. B: Alkaline phosphatase activity; 24 hrs after seeding, medium was changed and 4 μM U0126 or vehicle was added. BMP-2 was added to selected wells after a further hour. Cell extracts were prepared 72 hrs later. Data was obtained using 5 different isolates of chondrocytes assayed in triplicate. Values are mean ± SEM of 12–15 samples normalized to within experiment controls treated with BMP-2 but no inducers, *:p < 0.01 group differs from non BMP-2 treated group within inhibitor treatment, +: p < 0.05 that group differs from group with no UO126.

Figure 2

Effects of MAP kinase manipulation on b2/640 type X collagen promoter activity, in chick cephalic sternum chondrocytes. A and B: Mutant kinases; 24 hrs after seeding, cells were transfected with luciferase vectors and 0.5–1 μg mutant kinase DNA. After 5 hrs medium was changed (A) or medium changed and BMP-2 added (B). C and D: Inhibitor treatments; 5 hours after transfection with luciferase vectors, medium was changed and kinase inhibitor or vehicle (DMSO) was added as indicated. In D 30 ng/ml BMP-2 was added one hr after medium change. Data obtained using at least 2 independent isolates of chondrocytes, assayed in triplicate. Values are mean ± SEM of luciferase ratios of type X collagen promoter activity to control vector, normalized to BMP-2 treated controls. *:p < 0.05 that luciferase ratio differs from non BMP-2 treated group, +: p < 0.05 that luciferase ratio differs from group with no MAP kinase manipulation.

Figure 3

Effects of ascorbate and MAP kinase inhibitors on the Col X promoter and alkaline phosphatase activity in chick cephalic sternal chondrocytes. A: Activity of the b2/640 type X collagen promoter; 24 hrs after seeding cells were transfected with luciferase vectors, 5 hrs after transfection 4 μM U0126 or vehicle were added. Ascorbate was added to selected wells after a further hour. Values are mean ± S.D of the mean ratio of promoter to empty vector fluorescence units, for 3 experiments assayed in triplicate. B: Alkaline phosphatase activity; 24 hrs after seeding, medium was changed and 4 μM U0126 or vehicle was added. Ascorbate or ascorbate with BMP-2 was added to selected wells after a further hour. Cell extracts were prepared 72 hrs later. Data was obtained using 5 different isolates of chondrocytes assayed in triplicate. Values are mean ± SEM of 12–15 samples normalized to within experiment controls treated with BMP-2 but no inducers. *: p < 0.05 that group differs from non-supplemented group within inhibitor treatment. The increase in ALP caused by BMP-2 addition is shown, this increase is significantly smaller in UO126 treated cells, +:p < 0.05.

Figure 4

Possible mechanism for kinase involvement in BMP stimulated type X collagen expression. ERK1/2 is proposed to suppress activated Smad translocation to the nucleus as described by Kretschmer et al. [30] and thereby inhibit Col X transcription. ERK-specific Runx2 phosphorylation is required for regulation of typical osteoblastic genes but may not be required for Col X expression. However p38 facilitates Col X transcription, possibly via Runx2 phosphorylation and although was not shown to be involved in short term increases in ALP in our model, has been shown to be involved in long term increases in ALP over 15 days of chondrocyte differentiation in culture (question mark) [29].