One of the most common forms of functional interaction among transcription factors is the more than additive effect at promoters harboring multiple copies of a response element. The mechanisms that enable or control synergy at such compound response elements are poorly understood. We recently defined a common motif within the negative regulatory regions of multiple factors that operates by regulating their transcriptional synergy. We have identified such a synergy control (SC) motif embedded within the "attenuator domain" of CCAAT/enhancer-binding protein ␣ (C/EBP␣), a key regulator of energy homeostasis and cellular differentiation. A Lys 159 3 Arg substitution within the SC motif does not alter C/EBP␣ activity from a single site but leads to enhanced transactivation from synthetic or natural compound response elements. The sequence of SC motifs overlaps with the recently defined consensus SUMO modification site, and we find that the SC motif is the major site of both SUMO-1 and SUMO-3 modification in C/EBP␣. Furthermore, the disruption of SC motif function is accompanied by loss of SUMO but not ubiquitin modification. C/EBP␣ interacts directly with the E2 SUMO-conjugating enzyme Ubc9 and can be SUMOylated in vitro using purified recombinant components. Notably, we find that PIASy has E3-like activity and enhances both SUMO-1 and SUMO-3 modification of C/EBP␣ in vivo and in vitro. Our results indicate that SUMO modification of SC motifs provides a means to rapidly control higher order interactions among transcription factors and suggests that SUMOylation may be a general mechanism to limit transcriptional synergy.

Diverse functions for three soybean (Glycine max L. Merr.) cysteine proteinase inhibitors (CysPls) are inferred from unique characteristics of differential regulation of gene expression and inhibitory activities against specific Cys proteinases. Based on northern blot analyses, we found that the expression in leaves of one soybean CysPl gene (LI) was constitutive and the other two (N2 and R I ) were induced by wounding or methyl jasmonate treatment. Induction of N2 and R I transcript levels in leaves occurred coincidentally with increased papain inhibitory activity. Analyses of kinetic data from bacterial recombinant CysPl proteins indicated that soybean CysPls are noncompetitive inhibitors of papain. The inhibition constants against papain of the CysPls encoded by the wound and methyl jasmonate-inducible genes (57 and 21 nM for N2 and R1, respectively) were 500 to 1000 times lower than the inhibition constant of L1 (19,000 nM). N2 and R1 had substantially greater inhibitory activities than L l against gut cysteine proteinases of the third-instar larvae of western corn rootworm and Colorado potato beetle. Cysteine proteinases were the predominant digestive proteolytic enzymes i n the guts of these insects at this developmental stage. N2 and R1 were more inhibitory than the epoxide trans-epoxysuccinyl-~-leucylamide-(4-guanidino)butane (E-64) against western corn rootworm gut proteinases (50% inhibition concentration = 50, 200, and 7000 nM for N2, R1, and E-64, respectively). However, N2 and R1 were less effective than E-64 against the gut proteinases of Colorado potato beetle. These results indicate that the wound-inducible soybean CysPls, N2 and R1, function in host plant defense against insect predation, and that substantial variation in CysPl activity against insect digestive proteinases exists among plant CysPl proteins.Proteinaceous CysPIs, which specifically inhibit sulfl~y-dryl proteinase activities, are distributed ubiquitously among animal, plant, and microorganism species. The an-

Small ubiquitin-like modifier (SUMO) modification of sequence-specific transcription factors has profound regulatory consequences. By providing an intrinsic inhibitory function, SUMO isoforms can suppress transcriptional activation, particularly at promoters harboring multiple response elements. Through a comprehensive structure-function analysis, we have identified a single critical sector along the second beta sheet and the following alpha helix of SUMO2. This distinct surface is defined by four basic residues (K33, K35, K42, R50) that surround a shallow pocket lined by aliphatic (V30, I34) and polar (T38) residues. Substitutions within this area specifically and dramatically affected the ability of both SUMO2 and SUMO1 to inhibit transcription and revealed that the positively charged nature of the key basic residues is the main feature responsible for their functional role. This highly conserved surface accounts for the inhibitory properties of SUMO on multiple transcription factors and promoter contexts and likely defines the interaction surface for the corepressors that mediate the inhibitory properties of SUMO.Eukaryotic transcriptional control relies strongly on the power of combinatorial regulation (53). In a given cell, genes can draw unique sets of regulatory factors from a relatively small pool based on their specific arrangement of cis-acting elements. The often synergistic functional interactions between these factors give rise to the complex patterns of gene expression required for the appropriate physiological performance of cells (6, 9). The regulatory repertoire of each factor can be further expanded by structural alterations induced by binding of ligands or posttranslational modifications, as in the case of the glucocorticoid receptor (GR) and other steroid hormone receptors (53). Subtle mutations that disturb ligand binding or covalent attachment can lead to profound alterations in the outcome of functional interactions among factors. For example, we have identified a short regulatory motif embedded in GR and many other sequence-specific regulators that is both necessary and sufficient to limit their transcriptional synergy (23). Disruption of these conserved synergy control (SC) motifs selectively enhances synergistic activation at compound response elements without altering the activity driven from a single site. In order to exert their effects, SC motifs need to be recruited to multiple independent sites on the DNA, and this feature accounts for their selective effects at compound response elements (20). The function of SC motifs is not circumscribed to inhibit synergistic activation driven by the factor in which they are embedded (in-cis effects) but are also capable of affecting separate activators bound nearby (intrans effects). Based on the convergence between SC motifs and the consensus sequence for posttranslational modification by members of the small ubiquitin-like (Ubl) modifier (SUMO) family, we (20, 47) and others (51) have shown that SC motifs function as SUMO acceptor sites and...

SummaryPlant cysteine proteinase inhibitors (phytocystatins) have been implicated as defensive molecules against Coleopteran and Hemipteran insect pests. Two soybean cystatins, soyacystatin N (scN) and soyacystatin L (scL), have 70% sequence identity but scN is a much more potent inhibitor of papain, vicilin peptidohydrolase and insect gut proteinases. When these cystatins were displayed on phage particles, papain-binding affinity and CPI activity of scN were substantially greater than those of scL, in direct correlation with their relative CPI activity as soluble recombinant proteins. Furthermore, scN substantially delayed cowpea weevil (Callosobruchus maculatus (F.)) growth and development in insect feeding bioassays, whereas scL was essentially inactive as an insecticide. Papain biopanning selection of phage-displayed soyacystatins resulted in a 200-1000-fold greater enrichment for scN relative to scL. These results establish that binding affinity of cystatins can be used in phage display biopanning procedures to select variants with greater insecticidal activity, illustrating the potential of phage display and biopanning selection for directed molecular evolution of biological activity of these plant defensive proteins.

No abstract

Abstract-GATA sequences are required for the optimal expression of endothelial cell-specific genes, including endothelin-1 (ET-1). We have identified PIASy in a search for new GATA-2 interacting proteins that can regulate GATA-2-mediated endothelial gene expression. Notably, among the cell populations comprising vascular walls, PIASy mRNA is selectively expressed in endothelial cells, and its expression can be regulated by angiogenic growth factors. We show that GATA-2 is covalently modified by small ubiquitin-like modifier (SUMO)-1 and -2 and that PIASy, through its E3 SUMO ligase activity, preferentially enhances the conjugation of SUMO-2 to GATA-2. Through a functional analysis, we demonstrate that PIASy potently suppresses the activity of the GATA-2-dependent human ET-1 promoter in endothelial cells. The suppressive effect of PIASy requires the GATA-binding site in the ET-1 promoter and depends on its interaction with GATA-2, which requires both N-terminal (amino acids 1-183) and C-terminal (amino acids 414-510) sequences in PIASy. We conclude that PIASy enhances the conjugation of SUMO-2 to GATA-2 and that the interaction of PIASy with GATA-2 can modulate GATA-mediated ET-1 transcription activity in endothelial cells through a RING-like domain-independent mechanism. (Circ Res. 2003;92:1201-1208.)

No abstract

No abstract