Non‐genomic glucocorticoid signaling via G αs contributes to one‐third of their canonical genomic effects

Background and Objective We recently found that glucocorticoids produce rapid nongenomic signals within minutes by increasing cAMP production. G αs knockdown virtually eliminates these glucocorticoid‐stimulated cAMP responses. The objective of this study was to define the contribution of this nongenomic signaling to the glucocorticoid transcriptional activity in human airway smooth muscle cells (HASMs). Methods Control or G αs knockdown HASM cells from 6 different donors were treated with either vehicle or 1 μM budesonide for 24 hr. Total RNA was extracted and sequenced with an Illumina HiSeq 4000 sequencer. 100 bp single‐end reads were then aligned and quantified with Rsubread package. Differential gene lists were generated with DESeq2 package for the following treatment conditions: (1) vehicle‐treated control HASM cells, (2) budesonide‐treated control HASM cells, (3) vehicle‐treated G αs knockdown HASM cells, and (4) budesonide‐treated G αs knockdown HASM cells. ASSIGN , a pathway profiling toolkit, was used to generate signatures and predict pathway activities. Predicted budesonide activity was correlated using Pearson’s correlation. Overlap in differentially expressed genes was assessed using Venn diagrams, and a Gene Set Enrichment Analysis (GSEA) was applied using fgsea package. Finally, to find similar patterns of nongenomic transcriptional activity in existing public dataset, we probed the ConnectivityMap (CMAP) database with our nongenomic gene list. Results After internal and external in silico validations, we used the differentially expressed gene lists to generate two signatures: a 140‐gene budesonide (genomic + nongenomic) signature and a 121‐gene (genomic only) budesonide signature. Both signatures accurately estimated high budesonide activity in budesonide‐treated HASMs derived from both asthma and non‐asthma donors in an independent dataset, GSE94335 (Pearson’s correlation 0.9995; p‐value < 0.0001). There was no significant enrichment of any KEGG or GO gene‐sets for the signature genes after adjusting the p‐value for multiple comparisons. Ninety‐two genes were shared between budesonide and G αs dependent budesonide signatures. Forty‐eight genes were unique to the budesonide signature, indicating that transcriptional activity of these genes (26 upregulated and 22 downregulated) is dependent on signaling via G αs . We evaluated these 48 genes in their ability to identify glucocorticoid treatment in the CMAP database. Connectivity score (CS), a quantitative score between a query gene‐list and a perturbagen, was the highest (98.97) for “glucocorticoid receptor (GR) agonist” among the 171 pharmacologic classes available in the database. High CSs (>90) were seen for GR agonists in many different cell lines, including prostate, melanoma, lung, and hepatocellular cancer cell lines. Conclusion Of the 140‐gene budesonide signature, transcriptional changes in one‐third (48) of the genes depend upon non‐genomic signaling via G αs . This 48‐gene non‐genomic signature is present in many cell types treated with various glucocorticoids, implying that rapid, non‐genomic signaling by glucocorticoids is a common phenomenon. Support or Funding Information This work was supported by NIH grant GM107094