Glucocorticoid regulation of human pulmonary surfactant protein-B (SP-B) mRNA stability is independent of activated glucocorticoid receptor

Abstract

Adequate expression of surfactant protein-B (SP-B) is critical in the function of pulmonary surfactant to reduce alveolar surface tension. Expression of SP-B mRNA is restricted to specific lung-airway epithelial cells, and human SP-B mRNA stability is increased in the presence of the synthetic glucocorticoid dexamethasone (DEX). Although the mechanism of SP-B mRNA stabilization by DEX is unknown, studies suggest involvement of the glucocorticoid receptor (GR). We developed a dual-cistronic plasmid-based expression assay in which steady-state levels of SP-B mRNA, determined by Northern analysis, reproducibly reflect changes in SP-B mRNA stability. Using this assay, we found that steady-state levels of SP-B mRNA increased greater than twofold in transfected human-airway epithelial cells (A549) incubated with DEX (10(-7) M). DEX-mediated changes in SP-B mRNA levels required the presence of the SP-B mRNA 3'-untranslated region but did not require ongoing protein synthesis. The effect of DEX on SP-B mRNA levels was dose dependent, with maximal effect at 10(-7) M. DEX increased levels of SP-B mRNA in cells lacking GR, and the presence of the GR antagonist RU486 did not interfere with the effect of DEX. Surprisingly, other steroid hormones (progesterone, estradiol, and vitamin D; 10(-7) M) significantly increased SP-B mRNA levels, suggesting a common pathway of steroid hormone action on SP-B mRNA stability. These results indicate that the effect of DEX to increase SP-B mRNA stability is independent of activated GR and suggests that the mechanism is mediated by posttranscriptional or nongenomic effects of glucocorticoids.

Fig. 1.

Analysis of dexamethasone (DEX)-induced changes in steady-state levels of surfactant protein-B (SP-B) or red fluorescent protein (RFP) mRNA in A549 cells transfected with pCMVGFP-hspB:N or pCMV-GFP-RFP, where CMV is cytomegalovirus and GFP is green fluorescent protein. A: schematic of pCMVGFP-hspB:N and pCMV-GFP-RFP plasmid shown in scale. Black boxes represent identical, independent CMV E1 promoters; arrows denote the direction of transcription. Gray arrows represent the coding regions for human SP-B (hSP-B), RFP, and GFP. Hatched boxes represent the SP-B mRNA 3′-untranslated region (hSP-B 3′-UTR), rabbit β-globin 3′-UTR (globin 3′-UTR) and the SV40 large T antigen mRNA 3′-UTR (SV40pA). The positions of the polyadenylation signals are indicated. B: analysis of GFP mRNA stability in A549 cells. Cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of DEX (10⁻⁷ M) and then treated with actinomycin D (10 μg/ml) for 24 h. RNA was isolated at time = 0 and 24 h and subjected to Northern analysis for the presence of GFP mRNA and cyclophilin mRNA (for use as a loading control, only at t = 0 h). Shown is a typical image of the analysis generated by phosphorimaging. Levels of GFP mRNA at t = 0 h were set as 1 and normalized GFP levels (means ± SE) at 24 h relative to levels at t = 0 h are shown (N = 6, 2 independent experiments). C: effect of DEX on steady-state levels of SP-B and RFP mRNA in A549 cells. Cells were transfected with pCMVGFP-RFP or pCMVGFP-hspB:N and incubated in the absence or presence of DEX (10⁻⁷ M). RNA was isolated and subjected to Northern analysis for the presence for SP-B, RFP, or GFP mRNA. Shown is a typical image of the analysis generated by phosphorimaging. The RFP mRNA/GFP mRNA or SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from DEX-treated samples was normalized to this average. Shown are normalized SP-B levels (means ± SE) in DEX-treated samples relative to levels in untreated samples (N ≥ 10, 3 independent experiments; *P < 0.01).

Fig. 2.

Analysis of DEX-induced changes in steady-state levels of SP-B mRNA as a function of time and in the presence of a protein synthesis inhibitor. A: changes in steady-state SP-B mRNA levels in the absence or presence of DEX as a function of time. A549 cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of DEX (10⁻⁷ M). RNA was isolated from the samples at the indicated times and subjected to Northern analysis for the presence of SP-B and GFP mRNA. Shown is a line graph of DEX-induced SP-B mRNA/GFP mRNA levels (means ± SE) relative to untreated samples, normalized as 1 (N ≥ 7, 2 independent experiments; *P < 0.01). B: effect of DEX on steady-state levels of SP-B mRNA in A549 cells transfected with pCMVGFP-hspB:N in the absence or presence of cyclohexamide (CHX). A549 cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of DEX (10⁻⁷ M) and in the absence or presence of cyclohexamide (100 μg/ml) for 24 h. RNA was isolated and subjected to Northern analysis for the presence of SP-B and GFP mRNA, and the SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from DEX-treated samples was normalized to this average. Shown are normalized SP-B mRNA levels in DEX-treated samples relative to levels in untreated samples (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.03 relative to controls without DEX).

Fig. 3.

DEX-induced changes in steady-state levels of SP-B mRNA requires portions of the human SP-B mRNA 3′-UTR. A: schematics of the various expression cassettes harboring deletions of the SP-B 3′-UTR shown in scale. The gray arrow denotes the coding region for human surfactant protein-B, whereas the thin black lines represent the 3′-UTR. The hatched box represents the rabbit β-globin 3′-UTR. The positions of restriction endonuclease cleavage sites used to generate deletions of the SP-B 3′-UTR (shown by dashed lines) are indicated. AATAAA indicates the position of the polyadenylation signal of SP-B mRNA. B: effect of DEX on steady-state levels of SP-B mRNA in A549 cells transfected with plasmids shown above. The SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from DEX-treated samples was normalized to this average. Shown are normalized SP-B levels in DEX-treated samples relative to levels in untreated samples (means ± SE; *P < 0.01 relative to untreated controls, #P < 0.01 relative to SP-B mRNA levels from pCMVGFP-hspB:N treated with DEX). C: effect of deletions of the SP-B mRNA 3′-UTR on intrinsic stability of SP-B mRNA expressed from A549 cells transfected with plasmids shown above. The SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in pCMVGFP-hspB:N samples was set as 1; the ratio from other samples was normalized to this average. Shown are the normalized SP-B levels (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.01 relative to SP-B mRNA levels from pCMVGFP-hspB:N).

Fig. 4.

Effects of DEX concentration and other steroid hormones on steady-state levels of SP-B mRNA. A: changes in steady-state levels of SP-B mRNA in response to various concentrations of DEX. A549 cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of indicated concentrations of DEX. RNA was isolated and subjected to Northern analysis for the presence of SP-B and GFP mRNA, and the SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from DEX-treated samples was normalized to this average. Shown are normalized SP-B levels in DEX-treated samples relative to levels in untreated samples (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.02 relative to untreated controls; #P < 0.02 relative to levels at 10⁻⁷ M DEX). B: changes in steady-state levels of SP-B mRNA in response to other steroid hormones. A549 cells were transfected with pCMVGFP-hspB:N and incubated for 36 h in the absence (Con) or presence of various steroid hormones (10⁻⁷ M), including DEX, betamethasone (BMZ), hydrocortisone (HCT), progesterone (PROG), β-estradiol (E₂), and vitamin D (VitD). RNA was isolated and subjected to Northern analysis. The levels of SP-B and GFP were quantified, and the SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from the hormone-treated samples was normalized to this average. Shown are normalized SP-B levels in treated samples relative to levels in untreated samples (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.03 relative to untreated controls).

Fig. 5.

DEX-induced changes in steady-state levels of SP-B mRNA in various cells that lack or possess expression of the glucocorticoid receptor (GR). A: detection of GR mRNA in the cell lines. RT-PCR analysis was performed on RNA isolated from A549, human epithelial kidney (HEK)293, ECV304, and HeLa cells as described in materials and methods. The size of the expected 825-bp band is indicated by the arrow. RT control indicates reaction performed without reverse transcriptase. B: effect of DEX on steady-state levels of SP-B mRNA in various cells. A549, HEK293, ECV304 and HeLa cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of DEX (10⁻⁷ M). Cells that express GR are indicated by +, whereas cells that lack expression are indicated by -. RNA was isolated, subjected to Northern analysis for the presence of SP-B and GFP mRNA, and the SP-B mRNA/GFP mRNA ratio was determined in each sample. The average SP-B mRNA/GFP mRNA ratio in untreated samples was set as 1; the ratio from DEX-treated samples was normalized to this average. Shown are normalized SP-B levels in DEX-treated samples relative to levels in untreated samples (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.01 relative to untreated controls).

Fig. 6.

DEX-induced changes in steady-state levels of SP-B mRNA in the presence of RU486. A: A549 cells were transfected with pCMVGFP-hspB:N and incubated in the absence or presence of indicated concentrations of DEX and/or the GR antagonist RU486 for 36 h. RNA was isolated subjected to Northern analysis. The levels of SP-B and GFP were quantified, and the SP-B mRNA/GFP mRNA ratio was determined in each sample. The average ratio in untreated samples was set as 1; the ratio from treated samples was normalized to this average. Shown are normalized SP-B levels in DEX-treated samples relative to levels in untreated samples (means ± SE; N ≥ 7, 2 independent experiments; *P < 0.02 relative to untreated controls). B: analysis of the GR-responsive human serum/glucocorticoid regulated kinase 1 (sgk1) gene in A549 cells in the absence or presence of DEX and/or RU486. Northern analysis of RNA from the cells used in A was performed for detection of sgk1 and cyclophilin (cyclo) RNA. Shown is a typical image of the analysis generated by phosphorimaging. Numbers in parenthesis indicate the fold change from control of the ratio of sgk1/cyclophilin signal.