Prostaglandin E2 potentiates heat shock-induced heat shock protein 72 expression in A549 cells

Abstract

The heat shock (HS) response is an important cytoprotective response comprising the expression of heat shock proteins (HSPs) and orchestrated by the heat/stress-induced transcription factor, heat shock factor-1 (HSF-1). Previous studies suggest that the activation threshold and magnitude of the HS response may be modified by treatment with arachidonic acid (AA). We analyzed the effect of exogenous AA and its metabolites, PGE(2), LTD(4), and 15-HETE on HSF-1-dependent gene expression in A549 human respiratory epithelial-like cells. When added at 1microM, PGE(2) much more than LTD(4), but not 15-HETE increased activity of a synthetic HSF-1-dependent reporter after HS exposure (42 degrees C for 2h), but had no effect in the absence of HS. Exposing A549 cells to HS stimulated the release of PGE(2) and treatment with the cyclooxygenase inhibitor, ibuprofen, reduced HS-induced HSF-1-dependent transcription. PGE(2) increased HS-induced HSP72 mRNA and protein expression but EMSA and Western blot analysis failed to show an effect on HSF-1 DNA binding activity or post-translational modification. In summary, we showed that HS stimulates the generation of PGE(2), which augments the generation of HSPs. The clinical consequences of this pathway have yet to be determined.

Figure 1. Effect of PGE₂ and other stimuli on HS-induced activation of a HS-responsive reporter construct

A549 cells stably transfected with a HS-responsive luciferase reporter plasmid were incubated for 6 h with or without a 2 h 42°C HS and coincident treatment with medium alone (none), vehicle (0.4% EtOH), 1 μM of AA or indicated AA metabolite (A), the indicated concentration of PGE₂ (B), the indicated concentration of AA (C), 20 μM AA (D) or 30 min pretreatment with 5 μM ibuprofen (E). The cells were then lysed and luciferase measured (A-C, E) and expressed as fold-change compared with untreated controls without HS or PGE₂ measured by EIA (D). Mean±SE, n = 4 (A–C, E) or 3 (D); * and † denotes p<0.05 vs. untreated, HS-exposed controls and cells receiving the same treatment but without HS, respectively.

Figure 2. Effect of PGE₂ on HS-induced activation of HSP72 expression

A549 cells were incubated for 6 h with or without a 2 h, 42°C HS and with 1 μM PGE₂ or 0.4% ethanol and HSP72 mRNA was quantified by real-time RT-PCR standardized to GAPDH (A) and protein was quantified by Western blotting with β-tubulin as a control (representative of 4 similar gels in B). The bands were quantified by direct imaging of the chemilluminescent bands, expressed as a ratio to β-tubulin, and as fold change vs untreated cells without HS (C). Data are expressed as mean±SE, n = 4; * and † denotes p<0.05 vs. untreated, HS-exposed controls and cells receiving the same treatment but without HS, respectively.

Figure 3. Effect of PGE₂ on HSF-1 activation

Nuclear extracts were collected from A549 cells treated with 1 μM PGE₂ or 0.4% ethanol with or without HS at 42°C and analyzed for HSF-1 DNA binding activity by EMSA (representative of 4 similar autoradiographs in A; HSF-1-shifted complex indicated by arrow). HSF-1 binding was confirmed by supershifting with anti-HSF-1 antibody (ss, lane 9; indicated by arrowhead) and competition with 30X excess of unlabeled hspa1a oligonucleotide (cc, lane 10). Competition with a nonspecific (ns) oligonucleotide is shown as a negative control in lane 11. B. Covalent modification of HSF-1 was analyzed by measuring the elecrophoretic mobility of HSF-1 in nuclear extracts by Western blotting; a representative of 3 similar blots is shown.