Serum stimulation of fibronectin gene expression appears to result from rapid serum-induced binding of nuclear proteins to a cAMP response element

D. C. Dean, J. J. McQuillan, S. Weintraub

Research output: Contribution to journalArticlepeer-review

62 Scopus citations

Abstract

Fibronectin (FN) mRNA levels increased when quiescent cells (serum starved) were stimulated to undergo the G0/G1 transition by the addition of 20% given fetal calf serum to the media. The 5'-flanking region of the FN gene (position +69 to -510 base pairs (bp)) was fused to the coding region of the chloramphenicol acetyltransferase (CAT), and the fusion gene was used in transfection assays. Expression of FNCAT increased on serum treatment indicating that the region of the FN gene between positions +69 and -510 bp mediated serum responsiveness. Deletion of FN gene 5'-flanking sequences from position -510 to -122 bp eliminated serum responsiveness suggesting that an element between these positions was mediating the effect. Sequences between positions -122 and -510 bp of the FN gene were able to confer serum responsiveness on a herpes virus thymidine kinase promoter-CAT fusion gene (TKCAT) when the FN gene sequences were cloned upstream of TKCAT. The ability to confer serum responsiveness on TKCAT was retained with a smaller 100-bp sequence (position -122 to -222 bp). Both a cAMP response element (position -170 bp) and a nuclear factor-1 binding site (position -155 bp) have been identified within this sequence (Dean, D.C., Blakeley, M.S., Newby, R.F., Ghazal, P., Hennighausen, L., and Bourgeois, S. (1989) Mol. Cell. Biol. 9, 1498-1506). The cAMP response element was serum-responsive when cloned upstream of TKCAT or a minimal FN promoter (deleted to position -56 bp) with the nuclear factor-1 binding site was unresponsive. Therefore, the cAMP regulatory element (CRE) is the serum-responsive element between position -122 and -222 bp. Serum-induced binding of proteins to the CRE was detected in gel retardation assays with extracts from cell lines where FN expression was serum-responsive. However, no serum-induced binding was detected with extracts from the JEG-3 cell line where FN expression was not serum-responsive. Serum-induced binding occurred rapidly, within 15 min, and did not require protein synthesis. The decay of serum-induced binding was relatively slow as increased binding was still detectable 24 h after removal of serum. The CRE also mediates transcriptional stimulation by cAMP, but unlike serum stimulation increased CRE binding activity was not detectable in extracts from cAMP-treated cells (Dean, D.C., Blakeley, M.S., Newby, R.F., Ghazal, P., Hennighausen, L., and Bourgeois, S. (1989) Mol. Cell. Biol. 9, 1498-1506). In addition, the effects of cAMP and serum on transcription from the FN promoter were additive suggesting that the inducers act through independent pathways even though a common DNA element mediates their action.

Original languageEnglish
Pages (from-to)3522-3527
Number of pages6
JournalJournal of Biological Chemistry
Volume265
Issue number6
StatePublished - 1990

Fingerprint

Dive into the research topics of 'Serum stimulation of fibronectin gene expression appears to result from rapid serum-induced binding of nuclear proteins to a cAMP response element'. Together they form a unique fingerprint.

Cite this