Rac1 modification by an electrophilic 15-deoxy Δ(12,14)-prostaglandin J2 analog.

Vascular endothelial cells (ECs) are important for maintaining vascular homeostasis. Dysfunction of ECs contributes to cardiovascular diseases, including atherosclerosis, and can impair the healing process during vascular injury. An important mediator of EC response to stress is the GTPase Rac1. Rac1 responds to extracellular signals and is involved in cytoskeletal rearrangement, reactive oxygen species generation and cell cycle progression. Rac1 interacts with effector proteins to elicit EC spreading and formation of cell-to-cell junctions. Rac1 activity has recently been shown to be modulated by glutathiolation or S-nitrosation via an active site cysteine residue. However, it is not known whether other redox signaling compounds can modulate Rac1 activity. An important redox signaling mediator is the electrophilic lipid, 15-deoxy-Δ(12,14)-prostaglandin J2 (15d-PGJ2). This compound is a downstream product of cyclooxygenase and forms covalent adducts with specific cysteine residues, and induces cellular signaling in a pleiotropic manner. In this study, we demonstrate that a biotin-tagged analog of 15d-PGJ2 (bt-15d-PGJ2) forms an adduct with Rac1 in vitro at the C157 residue, and an additional adduct was detected on the tryptic peptide associated with C178. Rac1 modification in addition to modulation of Rac1 activity by bt-15d-PGJ2 was observed in cultured ECs. In addition, decreased EC migration and cell spreading were observed in response to the electrophile. These results demonstrate for the first time that Rac1 is a target for 15d-PGJ2 in ECs, and suggest that Rac1 modification by electrophiles such as 15d-PGJ2 may alter redox signaling and EC function.

Cell signalling by oxidized lipids and the role of reactive oxygen species in the endothelium.

The controlled formation of ROS (reactive oxygen species) and RNS (reactive nitrogen species) is now known to be critical in cellular redox signalling. As with the more familiar phosphorylation-dependent signal transduction pathways, control of protein function is mediated by the post-translational modification at specific amino acid residues, notably thiols. Two important classes of oxidant-derived signalling molecules are the lipid oxidation products, including those with electrophilic reactive centres, and decomposition products such as lysoPC (lysophosphatidylcholine). The mechanisms can be direct in the case of electrophiles, as they can modify signalling proteins by post-translational modification of thiols. In the case of lysoPC, it appears that secondary generation of ROS/RNS, dependent on intracellular calcium fluxes, can cause the secondary induction of H2O2 in the cell. In either case, the intracellular source of ROS/RNS has not been defined. In this respect, the mitochondrion is particularly interesting since it is now becoming apparent that the formation of superoxide from the respiratory chain can play an important role in cell signalling, and oxidized lipids can stimulate ROS formation from an undefined source. In this short overview, we describe recent experiments that suggest that the cell signalling mediated by lipid oxidation products involves their interaction with mitochondria. The implications of these results for our understanding of adaptation and the response to stress in cardiovascular disease are discussed.

Mechanisms of signal transduction mediated by oxidized lipids: the role of the electrophile-responsive proteome.

Cellular redox signalling is mediated by the post-translational modification of proteins by reactive oxygen/nitrogen species or the products derived from their reactions. In the case of oxidized lipids, several receptor-dependent and -independent mechanisms are now emerging. At low concentrations, adaptation to oxidative stress in the vasculature appears to be mediated by induction of antioxidant defences, including the synthesis of the intracellular antioxidant glutathione. At high concentrations apoptosis occurs through mechanisms that have yet to be defined in detail. Recent studies have revealed a mechanism through which electrophilic lipids, formed as the reaction products of oxidation, orchestrate these adaptive responses in the vasculature. Using a proteomics approach, we have identified a subset of proteins in cells that we term the electrophile-responsive proteome. Electrophilic modification of thiol groups in these proteins can initiate cell signalling events through the transcriptional activation of genes regulated by consensus sequences for the antioxidant response element found in their promoter regions. The insights gained from our understanding of the biology of these mechanisms will be discussed in the context of cardiovascular disease.

Nitric oxide and cell signaling: modulation of redox tone and protein modification.

Reactive oxygen species (ROS) and reactive nitrogen species (RNS) have an impact on many cellular processes, often serving as signal transducers in both physiological and pathological situations. These small molecules can act as ligands for receptors as is the case for nitric oxide and guanylate cyclase. However, they can also modify proteins, changing their function and establishing a baseline for other signals in a process that we have termed "redox tone." In this review, we discuss the different mechanisms of redox cell signaling, and give specific examples of RNS participation in cell signaling via classical and redox tone pathways.

Design, characterization, and structure of a biologically active single-chain mutant of human IFN-gamma.

A mutant form of human interferon-gamma (IFN-gamma SC1) that binds one IFN-gamma receptor alpha chain (IFN-gamma R alpha) has been designed and characterized. IFN-gamma SC1 was derived by linking the two peptide chains of the IFN-gamma dimer by a seven-residue linker and changing His111 in the first chain to an aspartic acid residue. Isothermal titration calorimetry shows that IFN-gamma SC1 forms a 1:1 complex with its high-affinity receptor (IFN-gamma R alpha) with an affinity of 27(+/- 9) nM. The crystal structure of IFN-gamma SC1 has been determined at 2.9 A resolution from crystals grown in 1.4 M citrate solutions at pH 7.6. Comparison of the wild-type receptor-binding domain and the Asp111-containing domain of IFN-gamma SC1 show that they are structurally equivalent but have very different electrostatic surface potentials. As a result, surface charge rather than structural changes is likely responsible for the inability of the His111-->Asp domain of to bind IFN-gamma R alpha. The AB loops of IFN-gamma SC1 adopt conformations similar to the ordered loops of IFN-gamma observed in the crystal structure of the IFN-gamma/IFN-gamma R alpha complex. Thus, IFN-gamma R alpha binding does not result in a large conformational change in the AB loop as previously suggested. The structure also reveals the final six C-terminal amino acid residues of IFN-gamma SC1 (residues 253-258) that have not been observed in any other reported IFN-gamma structures. Despite binding to only one IFN-gamma R alpha, IFN-gamma SC1 is biologically active in cell proliferation, MHC class I induction, and anti-viral assays. This suggests that one domain of IFN-gamma is sufficient to recruit IFN-gamma R alpha and IFN-gamma R beta into a complex competent for eliciting biological activity. The current data are consistent with the main role of the IFN-gamma dimer being to decrease the dissociation constant of IFN-gamma for its cellular receptors.

S100A1 utilizes different mechanisms for interacting with calcium-dependent and calcium-independent target proteins.

While previous studies have identified target proteins that interact with S100A1 in a calcium-dependent manner as well as target proteins that interact in a calcium-independent manner, the molecular mechanisms of S100A1-target protein interaction have not been elucidated. In this study, point and deletion mutants of S100A1 were used to investigate the contribution of carboxyl terminal amino acids to S100A1 interaction with calcium-dependent and calcium-independent target proteins. First, a recombinant rat S100A1 protein (recS100A1) expressed in bacteria exhibited physical and chemical properties indistinguishable from native S100A1. Next, proteins lacking the carboxyl-terminal nine residues of recS100A1 (Delta85-93), or containing alanine substitutions at Phe 88 (F88A), Phe 89 (F89A), or Trp 90 (W90A), both Phe 88 and Phe 89 (F88/89A), or all three aromatic residues (F88/89A-W90A) were recombinantly expressed. Like recS100A1, F88A, F89A, and W90A proteins interacted with phenyl-Sepharose in a calcium-dependent manner. However, the Delta85-93 protein did not interact with phenyl-Sepharose, indicating that a phenyl-Sepharose-binding region (PSBR) of recS100A1 had been disrupted. The F88/89A and F88/89A-W90A proteins exhibited reduced calcium-dependent interaction with phenyl-Sepharose when compared with recS100A1, demonstrating that the carboxyl-terminal aromatic residues Phe 88, Phe 89, and Trp 90 comprise the PSBR of S100A1. Fluorescence studies showed that the Delta85-93 protein exhibited reduced calcium-dependent interaction with the dodecyl CapZ peptide, TRTK, while W90A bound TRTK with a Kd of 5.55 microM. These results demonstrate that the calcium-dependent target protein-binding site and the PSBR are indistinguishable. In contrast to the calcium-dependent target TRTK, activation of the calcium-independent target protein aldolase A by the point and deletion mutant S100A1s was indistinguishable from native S100A1. These results demonstrate that carboxyl-terminal residues are not required for S100A1 modulation of calcium-independent target protein aldolase A. Alltogether, these results indicate that S100A1 utilizes distinct mechanisms for interaction with calcium-independent and calcium-dependent target proteins.

The role of cysteine residues in S100B dimerization and regulation of target protein activity.

Previous studies have demonstrated that the two cysteine residues in the calcium-binding protein S100B are required for its extracellular functions. In the present study, a recombinant S100B protein and mutant S100Bs containing one or no cysteine residue(s) have been used to determine the contribution of cysteine residues to S100B dimerization and interaction with the intracellular target proteins aldolase, phosphoglucomutase, and the microtubule associated tau protein. Mutation of C68 to a valine or C84 to a serine, C68 to valine and C84 to serine, or C68 to valine and C84 to alanine did not significantly alter S100B activation of aldolase. However, mutation of C84 to serine resulted in calcium-independent S100B activation of phosphoglucomutase and a loss of S100B inhibition of tau phosphorylation by Ca2+/calmodulin-dependent protein kinase II. The altered functionality of the C84S mutant with phosphoglucomutase and tau was not due to altered physical properties or dimerization state. All of the mutants exhibited heat stability and calcium dependent conformational changes which were identical to recombinant S100B. In addition, S100B proteins containing two, one or no cysteine residues behaved as dimers in size exclusion chromatography experiments in the presence or absence of calcium as well as in the presence or absence of reducing agent. Dynamic light scattering and analytical ultracentrifugation experiments confirmed that dimerization was not affected by calcium or reducing agent. Altogether these results demonstrate that S100B dimerization is not calcium- or sulfhydryl-dependent. In summary, cysteine residues are not necessary for the noncovalent dimerization of S100B, but are important in certain S100B target protein-interactions.

Identification of an S100A1/S100B target protein: phosphoglucomutase.

Phosphoglucomutase was identified as a potential intracellular S100 target protein because it interacted with two members of the S100 family of calcium-modulated proteins, S100A1 and S100B, in gel overlay experiments. These results were confirmed by affinity chromatography experiments demonstrating that S100A1 and S100B bound to phosphoglucomutase-Sepharose in a calcium-dependent manner. In the reverse experiment, phosphoglucomutase bound to S100A1 and S100B-Sepharose in a calcium-dependent manner. S100A1 inhibited phosphoglucomutase activity in a calcium-dependent manner. In contrast, S100B stimulated phosphoglucomutase activity in a calcium-dependent manner. Other calcium-binding proteins (calmodulin, troponin C, parvalbumin, and alpha-lactalbumin) had no effect on phosphoglucomutase. These results suggest that the effects of S100A1 and S100B are not nonspecific effects of low molecular weight, acidic proteins. This is the first report of an S100 target protein whose activity is antagonistically regulated by S100A1 and S100B, suggesting that cellular diversity in intracellular calcium signaling pathways may be due, at least in part, to the complement of S100 proteins expressed in different cell types.

Analysis of S100A1 expression during skeletal muscle and neuronal cell differentiation.

To better understand the mechanisms that regulate the function of the calcium-binding proteins S100A1 and S100B in developing systems, we have examined the level of, subcellular distribution of, and target proteins for these proteins in skeletal muscle (L6S4) and neuronal (PC12) cell lines. Both undifferentiated and differentiated L6 and PC12 cells express S100A1 and not S100B. Whereas S100A1 protein levels were higher in differentiated cells than in undifferentiated cells, steady-state mRNA levels did not change in differentiated L6 cells and decreased in differentiated PC12 cells when compared with undifferentiated cells. These results suggest that posttranscriptional rather than transcriptional mechanisms are responsible for increased S100A1 protein expression in myotubes and neurons. The colocalization of S100A1 staining with wheat germ agglutinin staining suggests that S100A1 is associated with the Golgi apparatus and secretory vesicles in PC12 and L6 cells. Using a gel overlay technique, S100A1-binding proteins were detected in undifferentiated and differentiated PC12 and L6 cells and the patterns observed were similar to those observed in brain and skeletal muscle, respectively. Although changes in the intensity of some binding proteins were detected, the overall pattern did not change when differentiated and undifferentiated cells were compared. These results suggest that the complement of S100A1-binding proteins does not change during differentiation, only the levels of some binding proteins. Altogether, our data demonstrate that the L6 and PC12 cell lines are excellent in vitro model systems for studying S100A1 expression and mechanisms that regulate S100A1 expression, subcellular distribution, and interaction with target proteins.

The S100 protein family: history, function, and expression.

The S100 family of calcium binding proteins contains approximately 16 members each of which exhibits a unique pattern of tissue/cell type specific expression. Although the distribution of these proteins is not restricted to the nervous system, the implication of several members of this family in nervous system development, function, and disease has sparked new interest in these proteins. We now know that the original two members of this family, S100A1 and S100B, can regulate a diverse group of cellular functions including cell-cell communication, cell growth, cell structure, energy metabolism, contraction and intracellular signal transduction. Although some members of the family may function extracellularly, most appear to function as intracellular calcium-modulated proteins and couple extracellular stimuli to cellular responses via interaction with other cellular proteins called target proteins. Interaction of these proteins with target proteins appear to involve cysteine residues (one in S100A1 and two in S100B), as well as a stretch of 13 amino acids, in the middle of the molecule called the linker region, which connects the two EF-hand calcium binding domains. In addition to the amino acid sequence and secondary structures of these proteins, the structures of the genes encoding these proteins are highly conserved. Studies on the expression of these proteins have demonstrated that a complex mixture of transcriptional and postranscriptional mechanisms regulate S100 expression. Further analysis of the function and expression of these proteins in both nervous and nonnervous tissues will provide important information regarding the role of altered S100 expression in nervous system development, function and disease.