Induction of Galphas contributes to the paradoxical stimulation of cytosolic phospholipase A2alpha expression by cortisol in human amnion fibroblasts

Abstract

Cytosolic phospholipase A (cPLA(2alpha)) catalyzes the formation of arachidonic acid in prostaglandin synthesis. In contrast to the well-described down-regulation of cPLA(2alpha), up-regulation of cPLA(2alpha) by glucocorticoids has been reported in human amnion fibroblasts, which may play a key role in parturition. The mechanisms underlying this paradoxical induction of cPLA(2alpha) by glucocorticoids remain largely unknown. Using cultured human amnion fibroblasts, we found that the induction of cPLA(2alpha) by cortisol required ongoing transcription and synthesis of at least one other protein. The induction of cPLA(2alpha) by cortisol was abolished by mutagenesis of a glucocorticoid response element (GRE) in the promoter. The same GRE was found mediating the classical inhibition of cPLA(2alpha) expression by cortisol in human fetal lung fibroblasts (HFL-1). Cortisol increased Galpha(s) expression in amnion fibroblasts but not in HFL-1 cells. Inhibition of Galpha(s) with NF449 attenuated the phosphorylation of cAMP response element-binding protein-1 (CREB-1) and the induction of cPLA(2alpha) by cortisol in amnion fibroblasts. Both glucocorticoid receptor (GR) and CREB-1 were found bound to the GRE upon cortisol stimulation of amnion fibroblasts. The induction of cPLA(2alpha) by cortisol was blocked by GR antagonist RU486 or protein kinase A inhibitor H89 or dominant-negative CREB-1. In conclusion, cortisol activates the cAMP/protein kinase A/CREB-1 pathway via Galpha(s) induction, and the phosphorylated CREB-1 interacts with GR at the GRE to promote cPLA(2alpha) expression in amnion fibroblasts.

Fig. 1.

A and B, Concentration-dependent stimulation of cPLA_2α mRNA and protein expression by cortisol, which could be blocked by GR antagonist RU486 in human amnion fibroblasts. The top panel of B is a representative immunoblot; bottom panel shows the mean data. C and D, The stimulation of cPLA_2α mRNA expression by cortisol (1 μm) was blocked by mRNA transcription inhibitor DRB (75 μm) and protein synthesis inhibitor CHX (10 μm) in human amnion fibroblasts. n = 3–7. *, P < 0.05; **, P < 0.01 vs. vehicle control; #, P < 0.05 vs. cortisol without RU486.

Fig. 2.

A and B, Inhibition of cPLA_2α mRNA and protein expression by cortisol (F, 1 μm) in HFL-1 cells. The top panel of B is the representative immunoblot; bottom panel shows the mean data. C, Inhibition of cPLA_2α promoter (−505 bp) activity by cortisol (F, 1 μm) could be diminished either by introduction of triple-nucleotide mutations (−505mut) into the putative GRE at −485 to −490 bp or by complete removal of the GRE as −458 bp in HFL-1 cells. D, Cortisol (F, 1 μm) treatment did not affect the expression of Gα_s protein in HFL-1 cells. The top panel of D is the representative immunoblot; bottom panel shows the mean data. n = 3–4. *, P < 0.05 vs. control (Ctr). E, ChIP demonstrated that cortisol (1 μm) caused no obvious binding of CREB to the GRE in the cPLA_2α promoter as revealed by gel electrophoresis of the PCR products amplified from the DNA fragments precipitated by CREB antibody in HFL-1 cells. The top panel is a representative gel of three individual experiments; bottom panel illustrates the positions and sequences of the primers used for PCR.

Fig. 3.

Bioinformatic analysis of the putative GRE in cPLA_2α promoter (−595 bp). Boxed nucleotides indicate the positions of the putative GREs, and the arrows indicate the sites of antisense (pointing to the right) and sense (pointing to the left) primers used for PCR subcloning of 5′-end truncated cPLA_2α promoters.

Fig. 4.

A, Stimulation of cPLA_2α promoter (−595 bp) activity by cortisol could be blocked by GR antagonist RU486 in human amnion fibroblasts. B, Deletion of the first 90 nucleotides (as −505 bp) from the 5′ end of −595-bp cPLA_2α promoter did not affect the stimulation of cPLA_2α promoter activity by cortisol (F, 1 μm). Further deletion of the putative GRE at −485 to −490 bp (as −458 bp) diminished the stimulation of cPLA_2α promoter activity by cortisol (F, 1 μm). C, The stimulation of of cPLA_2α promoter activity by cortisol (F, 1 μm) could be abolished by introducing triple-nucleotide mutations (underlined bold letters) into the putative GRE at −485 to −490 bp (boxed letters). n = 3. **, P < 0.01 vs. control (Ctr); #, P < 0.05 vs. cortisol (F) without RU486.

Fig. 5.

ChIP demonstrated that cortisol (1 μm) stimulated the binding of GR to the GRE in the cPLA_2α promoter as revealed by gel electrophoresis of the PCR products amplified from the DNA fragments precipitated by GR antibody in human amnion fibroblasts. The top panel is a representative gel of three individual experiments; bottom panel illustrates the positions and sequences of the primers used for PCR.

Fig. 6.

A, Forskolin (100 μm), an adenylyl cyclase stimulator, increased the level of cPLA_2α and COX-2 mRNA in human amnion fibroblasts. B and C, Inhibition of PKA with H89 (20 μm) attenuated the stimulation of cPLA_2α mRNA and protein expression by cortisol (F, 1 μm) in human amnion fibroblasts. The top panel of C is the representative immunoblot; bottom panel shows the mean data (n = 3). **, P < 0.01 vs. control (Ctr); #, P < 0.05; ##, P < 0.01 vs. cortisol (F) without H89.

Fig. 7.

Stimulation of cPLA_2α mRNA (A, n = 3) and protein (B) expression by cortisol (1 μm) was attenuated by transfection of a vector expressing dn-CREB-1 protein in human amnion fibroblasts. The top panel of B is the representative immunoblot; bottom panel shows the mean data of three individual experiments. *, P < 0.05 vs. the fold induction in cells transfected with negative control vector (NC).

Fig. 8.

A, Representative immunoblot showing GR in the protein complex immunoprecipitated (IP) by GR antibody (Ab) or CREB-1 antibody in human amnion fibroblasts stimulated by cortisol (1 μm). IgG(H), IgG heavy chain; IgG(L), IgG light chain. B, ChIP demonstrated that cortisol (1 μm) stimulated the binding of CREB-1 to the GRE in the cPLA_2α promoter as revealed by gel electrophoresis of the PCR products amplified from the DNA fragments precipitated by CREB-1 antibody in human amnion fibroblasts. The top panel is the representative gel image of three individual experiments; bottom panel illustrates the positions and sequences of the primers used for PCR.

Fig. 9.

A, Induction of Gα_s protein expression by cortisol (1 μm) was blocked by GR antagonist RU486 (RU, 1 μm) in human amnion fibroblasts. B, Stimulation of CREB-1 phosphorylation by cortisol (1 μm) was attenuated by Gα_s protein inhibitor NF449 (20 μm) in human amnion fibroblasts. C, Stimulation of cPLA_2α and COX-2 protein expression by cortisol (1 μm) was attenuated by Gα_s protein inhibitor NF449 (20 μm). The top panels of A–C are the representative immunoblots; bottom panel shows the mean data. n = 3–4. *, P < 0.05 vs. control (Ctr); #, P < 0.05 vs. cortisol without RU486 or NF449.

Fig. 10.

The hypothesized mechanism underlying the induction of cPLA_2α and COX-2 expression by cortisol in human amnion fibroblasts.