Microtubule-associated [corrected] protein 7 increases the membrane expression of transient receptor potential vanilloid 4 (TRPV4)

The molecular mechanism of the transmission of changes in the shape of the cell surface to ion channels remains obscure. Ca2+ influx induced by cell deformity is inhibited by actin-freezing reagents, suggesting that the actin microfilament couples with an ion channel. Transient receptor potential vanilloid 4 (TRPV4) is a candidate in the calcium-permeable, swelling-activated mechanosensitive channel in heterogeneously expressed cells. To investigate the mechanosensitive molecular complex, we found that microtubule-associated protein 7 (MAP7) is the mouse TRPV4 C-terminal binding protein. MAP7 was coimmunoprecipitated with TRPV4. The results of a pull-down assay demonstrated that the alignment of amino acids 798-809 of TRPV4 was important in this interaction. TRPV4 and MAP7 colocalized in the lung and kidney. The coexpression of these two molecules resulted in the redistribution of TRPV4 toward the membrane and increased its functional expression. The alignment of amino acids 798-809 of TRPV4 was also important in the functional expression. The activated current was abolished by actin-freezing but not by microtubule-freezing reagents. We therefore believe that MAP7 may enhance the membrane expression of TRPV4 and possibly link cytoskeletal microfilaments.     

Biological activities of homologous loop regions in the laminin alpha chain G domains

Laminin alpha chains (alpha1-alpha5 chains) have diverse chain-specific biological functions. The LG4 modules of laminin alpha chains consist of a 14-stranded beta-sheet (A-N) sandwich structure. Several biologically active sequences have been identified in the connecting loop regions. Here, we evaluated the biological activities of the loop regions of the E and F strands in the LG4 modules using five homologous peptides from each of the mouse alpha chains (EF-1: DYATLQLQEGRLHFMFDLG, alpha1 chain 2747-2765; EF-2: DFGTVQLRNGFPFFSYDLG, alpha2 chain 2808-2826; EF-3: RDSFVALYLSEGHVIFALG, alpha3 chain 2266-2284; EF-4: DFMTLFLAHGRLVFMFNVG, alpha4 chain 1511-1529; EF-5: SPSLVLFLNHGHFVAQTEGP, alpha5 chain 3304-3323). These homologous peptides showed chain-specific cell attachment and neurite outgrowth activities. Well organized actin stress fibers and focal contacts with vinculin accumulation were observed in fibroblasts attached on EF-1, whereas fibroblasts on EF-2 and EF-4 showed filopodia with ruffling. Fibroblast attachment to EF-2 and EF-4 was mediated by syndecan-2. In contrast, EF-1 promoted alpha2beta1 integrin-mediated fibroblast attachment and inhibited fibroblast attachment to a recombinant laminin alpha1 chain LG4-5. The receptors for EF-3 and EF-5 are unknown. Further, when the active core sequence of EF-1 was cyclized, utilizing two additional cysteine residues at both the N and C termini through a disulfide bridge, the cyclic peptide significantly enhanced integrin-mediated cell attachment. These results indicate that integrin-mediated cell attachment to the EF-1 sequence is conformation-dependent and that the loop structure is important for the activity. The homologous peptides, which promote either integrin- or syndecan-mediated cell attachment, may be useful for understanding the cell type- and chain-specific biological activities of the laminins.     

Human prorenin has "gate and handle" regions for its non-proteolytic activation

We investigated the mechanism for non-proteolytic activation of human prorenin using five kinds of antibodies. Each of the antigens, L1PPTDTTTFKRI11P, T7PFKRIFLKRMP17P, I11PFLKRMPSIRESLKER26P, M16PPSIRESLKER26P, and G27PVDMARLGPEWSQPM41P, was designed from the tertiary structure of predicted prorenin. These antibodies were labeled anti-01/06, anti-07/10, anti-11/26, anti-16/26, and anti-27/41, respectively, for their binding specificities. Inactive recombinant human prorenin (0.1 nM) bound to various concentrations of anti-01/06, anti-11/26, and anti-27/41 antibodies at 4 degrees C with equilibrium dissociation constants of 138, 41, and 22 nM, respectively. However, intact prorenin (0.1 nM) did not show significant binding to 200 nM anti-07/10 and anti-16/26 antibodies for 20 h. Ninety percent of prorenin (0.1 nM) was found to be non-proteolytically activated by incubation with anti-11/26 antibodies (200 nM) at 4 degrees C for 20 h. Prorenin was not active even under complex with either anti-01/06 or anti-27/41 antibodies. Prorenin was also reversibly activated at pH 3.3 and 4 degrees C for 25 h. The acid-activated prorenin bound to anti-07/10 and anti-16/26 antibodies as well as to anti-01/06, anti-11/15, and anti-27/41 antibodies at neutral pH and 4 degrees C in 2 h. Their dissociation constants were 13, 40, 8.6, 3.6, and 14 nM, respectively. The acid-activated prorenin was re-inactivated by incubation at pH 7.4 and 4 degrees C in 50 h. Anti-07/10 and anti-11/26 antibodies inhibited such re-inactivation at 25 degrees C by more than 90% and 50%, respectively, whereas other kinds of antibodies did not prevent the re-inactivation at 25 degrees C. These results indicate that prorenin has "gate" (T7PFKR10P) and "handle" (I11PFLKR15P) regions critical for its non-proteolytic activation.     

Functional domains essential for Gs activity in prostaglandin EP2 and EP3 receptors

The interaction of cell surface hormone receptors with heterotrimeric G proteins is crucial for hormonal actions. The domains of the receptor, which interact with and activate G protein, have been extensively studied. However, precise molecular mechanisms underlying regulation of the receptor-induced G protein activation are still poorly understood. Prostaglandin E(2) (PGE(2)) receptors comprise of four subtypes, EP1, EP2, EP3 and EP4. Among them, EP2 and EP4 couple to Gs and EP3 to Gi. To assess the functional domains essential for Gs activation in prostanoid receptors, EP2, EP3beta and each intracellular loop- (IC-) interchanged EP2/EP3 chimeras were tested for agonist binding and functional responses. In EP2 receptor, substitution of IC1 or IC3 resulted in loss of binding activity, while substitution of IC2, N- (IC2N) or C-terminal half region of IC2 (IC2C) had no effects on the binding activity. Wild-type EP2 and IC2C-substituted EP2 showed agonist-induced Gs activity, but IC2- and IC2N-substituted EP2 failed to elicit Gs activity upon agonist stimulation. On the other hand, in EP3 receptor substitution of IC1 resulted in loss of PGE(2) binding, while substitution of IC2, IC3, IC2N or IC2C had no effects on binding activity. Wild-type EP3beta, IC3- or IC2C-substituted EP3 failed to show Gs activity upon agonist stimulation, but IC2- or IC2N-substituted EP3 chimera showed agonist-dependent Gs activity. These results indicated that the second intracellular loop of the EP2 plays an essential role in activation of Gs.     

Heparan sulfate is required for bone morphogenetic protein-7 signaling

Although genetic studies have suggested that heparan sulfate (HS) is involved in bone morphogenetic protein (BMP)-mediated embryonic morphogenesis, it is unclear whether HS is directly involved in BMP-mediated signaling. Here, we investigate the involvement of HS in BMP-7 signaling. We show that HS and heparin chains specifically bind to BMP-7. Digestion of cell-surface HS with heparitinase interferes with BMP-7-mediated Smad phosphorylation in ROS 17/2.8 osteoblastic cells. Inhibiting sulfation of cell-surface HS with chlorate also causes interruption of Smad phosphorylation. Addition of exogenous heparin to ROS 17/2.8 cells prevents BMP-7-mediated Smad phosphorylation rather than enhances the BMP-7 signal, suggesting that HS should be anchored on the plasma membrane for BMP signaling. Moreover, BMP-7 binding to ROS 17/2.8 cells is inhibited by chlorate treatment and exogenous application of heparin. These results demonstrate that BMP-7 specifically binds to cell-surface HS and the BMP-7-HS interaction is required for BMP-7 signaling.     

A novel giant gene CSMD3 encoding a protein with CUB and sushi multiple domains: a candidate gene for benign adult familial myoclonic epilepsy on human chromosome 8q23.3-q24.1

We identified a novel giant gene encoding a transmembrane protein with CUB and sushi multiple domains on the human chromosome 8q23.3-q24.1 in which benign adult familial myoclonic epilepsy type 1 (BAFME1/FAME, OMIM:601068) has been mapped. This giant gene consists of 73 exons and spans over 1.2Mb on the genomic DNA region. It showed significant homology to two genes, CSMD1 gene on 8p23 and CSMD2 gene on 1p34, at reduced amino acid sequence level and hence we designated as CSMD3. The CSMD3 gene was expressed mainly in adult and fetal brains. We performed mutation analysis on the CSMD3 gene for seven patients with BAFME1/FAME, but no mutation was found in the coding sequence of the CSMD3 gene. Comparative genomic analysis revealed a conserved family of CSMD genes in the mouse and fugu genomes. Possible functions of the CSMD gene family are discussed.     

Akt is an endogenous inhibitor toward tumor necrosis factor-related apoptosis inducing ligand-mediated apoptosis in rheumatoid synovial cells

Akt is known to be activated in the rheumatoid synovial tissues. We examined here functional role of Akt during tumor necrosis factor-related apoptosis inducing ligand (TRAIL)-mediated apoptosis in rheumatoid synovial cells. Rheumatoid synovial cells in vitro were rapidly committed to apoptosis in response to TRAIL in mitochondria-dependent manner whereas Akt and extracellular signal-regulated kinase (ERK) were also phosphorylated. TRAIL-mediated apoptosis in synovial cells was significantly increased through inactivation of Akt by LY294002, however, that process was not so changed by adding ERK inhibitor, PD98059. Platelet-derived growth factor (PDGF) clearly phosphorylated both Akt and ERK in synovial cells, and PDGF pretreatment markedly suppressed TRAIL-mediated synovial cell apoptosis. The use of not PD98059 but LY294002 abrogated PDGF-mediated inhibitory effect toward TRAIL-induced apoptosis in synovial cells. The above protective effect of Akt was confirmed by the use of short interfering RNA (siRNA)-directed inhibition of Akt. Our data suggest that Akt is an endogenous inhibitor during TRAIL-mediated synovial cell apoptotic pathway, which may explain that synovial cells in situ of the rheumatoid synovial tissues are resistant toward apoptotic cell death in spite of death receptor expression.     

The structure and function of human dipeptidyl peptidase IV,possessing a unique eight-bladed beta-propeller fold

Dipeptidyl peptidase IV (DPPIV) is a serine protease, a member of the prolyl oligopeptidase (POP) family, and has been implicated in several diseases. Therefore, the development of DPPIV selective inhibitors, which are able to control the biological function of DPPIV, is important. We determined the crystal structure of human DPPIV at 2.6A resolution. The molecule consists of a unique eight-bladed beta-propeller domain in the N-terminal region and a serine protease domain in the C-terminal region. Also, the large "cave" structure, which is thought to control the access of the substrate, is found on the side of the beta-propeller fold. Comparison of the overall amino acid sequence between human DPPIV and POP shows low homology (12.9%). In this paper, we report the structure of human DPPIV, especially focusing on a unique eight-bladed beta-propeller domain. We also discuss the way for the access of the substrate to this domain.     

Caspase-mediated cleavage of JNK during stress-induced apoptosis

The c-Jun N-terminal kinases (JNKs) are a subfamily of the mitogen-activated protein kinases (MAPKs). The JNKs are encoded by three separate genes (jnk1, jnk2, and jnk3), which are spliced alternatively to create 10 JNK isoforms that are either p46 or p54 in size. In this study, we found that the p52 form of JNK emerged in human leukemia MOLT-4 or U937 cells following X-irradiation or heat treatment. The accumulation of p52 coincided with the reduction of p54 JNK. On the other hand, the amounts of p46 JNK did not change by X-irradiation. Induction of the p52 form of JNK also paralleled the appearance of the active form of caspase-3 and was suppressed by a caspase-specific inhibitor, Ac-DEVD-CHO, but not by Ac-YVAD-CHO. In vitro cleavage assays indicated that recombinant human JNK1beta2 and JNK2beta2 were cleaved by caspase-3, and that the mutation of aspartic acid at position 413 of JNK1beta2 or 410 of JNK2beta2 to alanine abolished the cleavage. Altogether, our results demonstrated that p54 JNKs, at least JNK1beta2 and JNK2beta2, were new selective targets of caspases in JNK splicing variants, and suggested that the p52 form could serve as a marker of apoptosis.