REGULATION OF 92 KDA TYPE-IV COLLAGENASE EXPRESSION BY THE JUN AMINOTERMINAL KINASE-REGULATED AND THE EXTRACELLULAR SIGNAL-REGULATED KINASE-DEPENDENT SIGNALING CASCADES

The 92 kDa type IV collagenase (MMP-9), which degrades type IV collage n, has been implicated in tissue remodeling. The purpose of the curren t study was to determine the role of Jun amino-terminal kinase (JNK)- and extracellular signal-regulated kinase- (ERK)-dependent signaling c ascades in the regulation of MMP-9 expression. Towards this end, we fi rst determined the transcriptional requirements for MMP-9 promoter act ivity in a cell line (UM-SCC-1) which is an avid secretor of this coll agenase. Transfection of these cells with a CAT reporter driven by pro gressive 5' deleted fragments of the MMP-9 promoter indicated the requ irement of a region spanning -144 to -73 for optimal promoter activity . DNase I footprinting revealed a protected region of the promoter spa nning nucleotides -91 to -68 and containing a consensus AP-1 motif at -79. Mutation of this AP-1 motif practically abolished the activity of the MMP-9 promoter-driven CAT reporter. Mobility shift assays indicat ed c-Fos and Jun-D bound to this motif and transfection of the cells w ith a mutated c-Jun, which quenches the function of endogenous Jun and Fos proteins, decreased MMP-9 promoter activity by 80%. UM-SCC-1 cell s contained a constitutively activated JNK and the expression of a kin ase-deficient JNK1 reduced the activity of a CAT reporter driven eithe r by the MMP-9 promoter or by three tandem AP-1 repeats upstream of a thymidine kinase minimal promoter. Conditioned medium collected from U M-SCC-1 cells transfected with the dominant negative JNK1 expression v ector diminished 92 kDa gelatinolysis. Similarly, interfering with MEK K, which lies upstream of JNK1, using a dominant negative expression v ector reduced MMP-9 promoter activity over the same concentration rang e which repressed the AP-l-thymidine kinase CAT reporter construct. UM -SCC-1 cells also contained a constitutively activated ERK1. MMP-9 exp ression, as determined by CAT assays and by zymography, was reduced by the co-expression of a kinase-deficient ERK1. Interfering with MEK1, which is an upstream activator of ERK1, either with PD 098059, which p revents the activation of MEK1, or with a dominant negative expression construct, reduced 92 kDa gelatinolysis and MMP-9 promoter activity r espectively. c-Raf-l is an upstream activator of MEK1 and a kinase-def icient c-Raf-l expression construct decreased the activity of a promot er driven by either the MMP-9 promoter or three tandem AP-1 repeats. C onversely, treatment of UM-SCC-1 cells with PMA, which activates c-Raf -l, increased 92 kDa gelatinolysis. These data suggest that MMP-9 expr ession in UM-SCC-1 cells, is regulated by JNK- and ERK-dependent signa ling pathways.