Replacement of Arg in the conserved N-terminal RLFDQxFG motif affects physico-chemical properties and chaperone-like activity of human small heat shock protein HspB8 (Hsp22)

The small heat shock protein (sHsp) called HspB8 (formerly, Hsp22) is one of the least typical sHsp members, whose oligomerization status remains debatable. Here we analyze the effect of mutations in a highly conservative sequence located in the N-terminal domain of human HspB8 on its physico-chemical properties and chaperone-like activity. According to size-exclusion chromatography coupled to multi-angle light scattering, the wild type (WT) HspB8 is present as dominating monomeric species (~24 kDa) and a small fraction of oligomers (~60 kDa). The R29A amino acid substitution leads to the predominant formation of 60-kDa oligomers, leaving only a small fraction of monomers. Deletion of the 28–32 pentapeptide (Δ mutant) results in the formation of minor quantities of dimers (~49 kDa) and large quantities of the 24-kDa monomers. Both the WT protein and its Δ mutant efficiently bind a hydrophobic probe bis-ANS and are relatively rapidly hydrolyzed by chymotrypsin, whereas the R29A mutant weakly binds bis-ANS and resists chymotrypsinolysis. In contrast to HspB8 WT and its Δ mutant, which are well phosphorylated by cAMP-dependent and ERK1 protein kinases, the R29A mutant is poorly phosphorylated. R29A mutation affects the chaperone-like activity of HspB8 measured in vitro. It is concluded that the irreplaceable Arg residue located in the only highly conservative motif in the N-terminal domain of all sHsp proteins affects the oligomeric structure and key properties of HspB8.

Overexpression of miR-126 promotes the differentiation of mesenchymal stem cells toward endothelial cells via activation of PI3K/Akt and MAPK/ERK pathways and release of paracrine factors

The endothelial cell (EC)-specific miRNA, miR-126, is known to promote angiogenesis in response to angiogenic factors by repressing negative regulators of signal transduction pathways; however, whether miR-126 might regulate the differentiation of stem cells toward endothelial lineage remains unknown. To answer this question, in this study mesenchymal stem cells (MSCs) harvested from C57BL/6 mouse bone marrow were transfected with miR-126 (MSCmiR-126) using recombinant lentiviral vectors. Results showed the para-secretion and the expression levels of phosphorylated PI3K p85, Akt, p38, ERK1 protein in the MSCmiR-126 group were dramatically increased when compared with the control group. With half culture medium refreshed every 3 days, a small number of 6-day-cultured MSCmiR-126 differentiated into endothelial-like cells and most of 9-day-cultured MSCmiR-126 formed a cobblestone-like structure. These differentiated cells evidently expressed EC-specific makers and possessed mature ECs function, while inhibition of paracrine factors suppressed the MSC-EC differentiation. Strikingly, the increased secretion of MSCmiR-126 and their endothelial-differentiated potential were deprived by using a PI3K or MEK chemical inhibitor. Our results suggest that overexpression of miR-126 agumenting the endothelial differentiation of MSCs might in part be attributable to the activation of PI3K/Akt and MAPK/ERK pathways and an increased release of paracrine factors.

Regulation of ERK1 gene expression by coactivator proteins

RARs (retinoic acid receptors) mediate the effect of their ligand RA (retinoic acid) on gene expression. We previously showed that RA inhibited cellular proliferation in part by decreasing expression of the mitogen activated protein kinase ERK1 (extracellular signal regulated kinase 1). However, the mechanism by which RA regulates ERK1 expression is largely uncharacterized. The present study characterizes coactivator-mediated regulation of RA target gene expression by analysing ERK1 promoter activation. CBP (CREB-binding protein) and PCAF (p300/CBP associated factor) are transcriptional coactivators that interact with nuclear hormone receptors such as RARs. CBP and PCAF differentially regulated ERK1 expression in stable clones. CBP clones expressed higher ERK1 protein levels, proliferated faster in culture and were resistant to RA-mediated growth inhibition. PCAF clones expressed lower levels of ERK1 protein and cells grew more slowly than controls. CBP and PCAF regulation of the ERK1 promoter was dependent on two Sp1 (specificity protein 1) sites located between −86 and −115 bp. Immunoprecipitation and yeast two-hybrid analysis revealed that PCAF interacted with Sp1 via CBP. A putative p53 binding site at −360 bp functioned as a major repressor of ERK1 promoter activity even in the absence of exogenous p53 expression. CBP and PCAF occupancy of the proximal ERK1 promoter was dramatically decreased by RA treatment. PCAF mediated inhibition of ERK1 expression was due to decreased stability of the kinase mRNA. We conclude that CBP and PCAF coactivators mediate ERK1 gene expression at both the transcriptional and post-transcriptional level.

Loss of active MEK1-ERK1/2 restores epithelial phenotype and morphogenesis in transdifferentiated MDCK cells

Constitutive activation of the MAPK/ERK kinase (MEK)1-ERK2 signaling module in Madin-Darby canine kidney (MDCK)-C7 cells disrupts their ability to form cystlike structures in collagen gels and induces an invasive, myofibroblastlike phenotype. However, the reversibility of these cellular events, as well as the relative role of both MEK isoforms (MEK1 and MEK2) and both ERK isoforms (ERK1 and ERK2) during these processes, has not yet been investigated. We now report that loss of constitutively active MEK1 (caMEK1) and, thus, loss of active ERK1/2 in C7caMEK1 cells is associated with increased MEK2 protein expression, reexpression of ERK1 protein, and epithelial redifferentiation of these cells. The morphological changes toward an epithelial phenotype in these revertant cell lines (C7rev4, C7rev5, C7rev7) are reflected by the upregulation of epithelial marker proteins, such as E-cadherin, β-catenin, and cytokeratin, by the loss of α-smooth muscle actin expression, and by the ability of these epithelial revertants to form well-organized spherical cysts when grown in three-dimensional collagen gels. Further evidence for a role of the MEK1-ERK1/2 module in epithelial-mesenchymal transition was obtained from the analysis of two novel, spontaneously transdifferentiated MDCK-C7 cell clones (C7e1 and C7e2 cells). In these clones, increased MEK1/2-ERK1/2 phosphorylation, reduced MEK2 protein expression, and loss of ERK1 protein expression is associated with phenotypic alterations similar to those observed in transdifferentiated C7caMEK1 cells. C7e1 cells at least partially regained some of their epithelial characteristics at higher passages. In contrast, C7e2 cells maintained a transdifferentiated phenotype at high passage, were unable to generate cystlike epithelial structures, and retained invasive properties when grown on a three-dimensional collagen matrix. We conclude that in renal epithelial MDCK-C7 cells, stable epithelial-to-mesenchymal transition (EMT) is associated with loss of ERK1 protein expression, reduced MEK2 protein expression, and increased basal ERK2 phosphorylation. In contrast, loss of active MEK1-ERK1/2 results in increased MEK2 protein expression and reexpression of ERK1 protein, concomitant with the restoration of epithelial phenotype and the ability to form cystic structures.

Differential expression and activation of MAP kinases in dedifferentiated MDCK-focus cells

Mitogen-activated protein kinases (MAPK) play a key role in the regulation of cellular processes such as cell growth, cell differentiation, and apoptosis. However, the specific function of single isoforms of the MAPK family in renal epithelial cell differentiation and/or proliferation has not been investigated so far. We now report stable reduction of extracellular signal-regulated kinase 1 (ERK1) protein expression and lack of serum-induced ERK1 activation in alkali-dedifferentiated Madin-Darby canine kidney-C7 focus (MDCK-C7F) cells compared with their parental epithelial MDCK-C7 cells. The changes in ERK1 protein expression and activation were accompanied by a small rise in c-jun NH2-terminal kinase 1 (JNK1) protein expression but slightly decreased basal and anisomycin-stimulated JNK1 activity. In contrast, ERK2 phosphorylation, as assessed by using an antibody which detects phosphorylated tyrosine 204 of both ERK1 and ERK2, as well as enzymatic ERK2 activity, was substantially increased in untreated and fetal calf serum-treated MDCK-C7F cells, although ERK2 protein expression remained unchanged. Differential expression and activation of ERK1, ERK2, and JNK1 were accompanied by an inhibition of serum-induced MDCK-C7F cell proliferation. Together, our results demonstrate an association between changes in the activation of certain MAPK and alkali-induced stable MDCK-C7 cell dedifferentiation. Moreover, these data provide evidence for distinct signaling functions of ERK1 and ERK2 in these cells.

Molecular cloning, expression, and characterization of the human mitogen-activated protein kinase p44erk1.

p44erk1 is a member of a family of tyrosyl-phosphorylated and mitogen-activated protein (MAP) kinases that participate in cell cycle control. A full-length erk1 cDNA was isolated from a human hepatoma cell line (Hep G2) library. The erk1 cDNA clone shared approximately 96% predicted amino acid identity with partial sequences of rodent erk1 cognates, and the erk1 gene was assigned to human chromosome 16 by hybrid panel analysis. Human erk1 expressed in Escherichia coli as a glutathione S-transferase fusion (GST-Erk1) protein was substantially phosphorylated on tyrosine in vivo. It underwent further autophosphorylation in vitro (up to 0.01 mol of P per mol) at the regulatory Tyr-204 site and at additional tyrosine and serine residues. Threonine autophosphorylation, presumably at the regulatory Thr-202 site, was also detected weakly when the recombinant kinase was incubated in the presence of manganese, but not in the presence of magnesium. Before and after cleavage of the GST-Erk1 protein with thrombin, it exhibited a relatively high level of myelin basic protein phosphotransferase activity, which could be reduced eightfold by treatment of the kinase with the protein-tyrosine phosphatase CD45, but not by treatment with the protein-serine/threonine phosphatase 2A. The protein-tyrosine kinase p56lck catalyzed phosphorylation of GST-Erk1 at two autophosphorylations sites, including Tyr-204, and at a novel site. A further fivefold stimulation of the myelin basic protein phosphotransferase activity of the GST-Erk1 was achieved in the presence of a partially purified MAP kinase kinase from sheep platelets. Under these circumstances, there was primarily an enhancement of the tyrosine phosphorylation of GST-Erk1. This MAP kinase kinase also similarly phosphorylated a catalytically compromised version of GST-Erk1 in which Lys-71 was converted to Ala by site-directed mutagenesis.

Molecular cloning, expression, and characterization of the human mitogen-activated protein kinase p44erk1

p44erk1 is a member of a family of tyrosyl-phosphorylated and mitogen-activated protein (MAP) kinases that participate in cell cycle control. A full-length erk1 cDNA was isolated from a human hepatoma cell line (Hep G2) library. The erk1 cDNA clone shared approximately 96% predicted amino acid identity with partial sequences of rodent erk1 cognates, and the erk1 gene was assigned to human chromosome 16 by hybrid panel analysis. Human erk1 expressed in Escherichia coli as a glutathione S-transferase fusion (GST-Erk1) protein was substantially phosphorylated on tyrosine in vivo. It underwent further autophosphorylation in vitro (up to 0.01 mol of P per mol) at the regulatory Tyr-204 site and at additional tyrosine and serine residues. Threonine autophosphorylation, presumably at the regulatory Thr-202 site, was also detected weakly when the recombinant kinase was incubated in the presence of manganese, but not in the presence of magnesium. Before and after cleavage of the GST-Erk1 protein with thrombin, it exhibited a relatively high level of myelin basic protein phosphotransferase activity, which could be reduced eightfold by treatment of the kinase with the protein-tyrosine phosphatase CD45, but not by treatment with the protein-serine/threonine phosphatase 2A. The protein-tyrosine kinase p56lck catalyzed phosphorylation of GST-Erk1 at two autophosphorylations sites, including Tyr-204, and at a novel site. A further fivefold stimulation of the myelin basic protein phosphotransferase activity of the GST-Erk1 was achieved in the presence of a partially purified MAP kinase kinase from sheep platelets. Under these circumstances, there was primarily an enhancement of the tyrosine phosphorylation of GST-Erk1. This MAP kinase kinase also similarly phosphorylated a catalytically compromised version of GST-Erk1 in which Lys-71 was converted to Ala by site-directed mutagenesis.