SUMMARY Protein-DNA interactions (PDIs) mediate a broad range of functions essential for cellular differentiation, function, and survival. However, it is still a daunting task to comprehensively identify and profile sequence-specific PDIs in complex genomes. Here, we have used a combined bioinformatics and protein microarray-based strategy to systematically characterize the human protein-DNA interactome. We identified 17,718 PDIs between 460 DNA motifs predicted to regulate transcription and 4,191 human proteins of various functional classes. Among them, we recovered many known PDIs for transcription factors (TFs). We also identified a large number of new PDIs for known TFs, as well as for previously uncharacterized TFs. Remarkably, we found that over three hundred proteins not previously annotated as TFs also showed sequence-specific PDIs, including RNA binding proteins, mitochondrial proteins, and protein kinases. One of such unconventional DNA-binding proteins, MAPK1, acts as a transcriptional repressor for interferon gamma-induced genes.

The transcription factor RUNX2 (Cbfa1/AML3/Pebp-2␣A) is a critical regulator of osteoblast differentiation. We investigated the effect of the inflammatory cytokine tumor necrosis factor ␣ (TNF) on the expression of RUNX2 because TNF is known to inhibit differentiation of osteoblasts from pluripotent progenitor cells. TNF treatment of fetal calvaria precursor cells or MC3T3-E1 clonal pre-osteoblastic cells caused a dose-dependent suppression of RUNX2 steady state mRNA as measured by reverse transcription-PCR. The IC 50 for TNF inhibition was 0.6 ng/ml. TNF suppression of RUNX2 mRNA was confirmed using Northern analysis. The effect of TNF was studied using isoform-specific primers that flanked unique regions of two major RUNX2 isoforms. TNF suppressed expression of the mRNA coding for the shorter MRIPV isoform by >90% while inhibiting expression of the mRNA for the longer MASNS isoform by 50%. RUNX2 nuclear content was evaluated by electrophoretic mobility shift assay using a rat osteocalcin promoter binding sequence as probe and by Western analysis. TNF reduced nuclear RUNX2 protein. Inhibition of new protein synthesis with cycloheximide failed to prevent TNF inhibition of RUNX2 mRNA, suggesting that a newly translated protein did not mediate the TNF effect. RUNX2 mRNA half-life was 1.8 h and reduced to 0.9 h by TNF. The effect of TNF on RUNX2 gene transcription was evaluated using a 0.6-kb RUNX2 promoter-luciferase reporter in MC3T3-E1 cells. TNF caused a dose-dependent inhibition of transcription to 50% of control values. The inhibitory effect of TNF was preserved with deletions to nucleotide ؊108 upstream of the translational start site; however, localization downstream of nucleotide ؊108 was obscured by loss of basal activity. Our results indicate that TNF regulates RUNX2 expression at multiple levels including destabilization of mRNA and suppression of transcription. The disproportionate inhibition of RUNX2 nuclear protein suggests that additional post-transcriptional mechanisms may be occurring. Suppression of RUNX2 by TNF may decrease osteoblast differentiation and inhibit bone formation in TNF excess states.The inflammatory cytokine tumor necrosis factor-␣ (TNF) 1 has been shown to contribute to bone loss through a variety of mechanisms that increase bone resorption and decrease bone formation. TNF has a major role as an inflammatory mediator in rheumatoid arthritis where increased bone resorption causes periarticular bone loss, and in postmenopausal osteoporosis in which there is generalized bone loss (1-5). In addition to the effects of TNF on bone resorption, TNF also inhibits the bone-forming function of osteoblasts. In mature osteoblasts TNF inhibits the expression of the skeletal matrix proteins type I collagen and osteocalcin, causes resistance to the genomic action of 1,25-dihydroxyvitamin D 3 , and increases the production of matrix metalloproteinases and pathologic paracrine factors (6 -11). We have shown previously that TNF inhibits the differentiation of new osteoblasts from precursor cells (1...

Geosmin is responsible for the characteristic odor of moist soil, as well as off-flavors in drinking water and foodstuffs. Geosmin is generated from farnesyl diphosphate (FPP, 2) by an enzyme that is encoded by the SCO6073 gene in the soil organism Streptomyces coelicolor A32 (ref. 3). We have now shown that the recombinant N-terminal half of this protein catalyzes the Mg2+-dependent cyclization of FPP to germacradienol and germacrene D, while the highly homologous C-terminal domain, previously thought to be catalytically silent, catalyzes the Mg2+-dependent conversion of germacradienol to geosmin. Site-directed mutagenesis confirmed that the N- and C-terminal domains each harbor a distinct, independently functioning active site. A mutation in the N-terminal domain of germacradienol-geosmin synthase of a catalytically essential serine to alanine results in the conversion of FPP to a mixture of sesquiterpenes that includes an aberrant product identified as isolepidozene, which was previously suggested to be an enzyme-bound intermediate in the cyclization of FPP to germacradienol.