Immunohistochemical localization of beta1 and beta4 integrins in mouse endometrium during implantation and early pregnancy

Implantation presents the remarkable synchronisation between the development of embryo and differentiation of endometrium. Cell-cell adhesion is an important phenomenon taking place during blastocyst implantation in uterine membrane. We think that the investigation of existence and the level of integrins in women can be a guide for treatment of infertility. Our purpose in this study was to show expression beta1 and beta4 integrins on gestational days 4, 6, 12 by immunohistochemical methods and to investigate whether beta4 integrin is a useful marker for receptivity. beta1 and beta4 integrin were exhibited on surface epithelium on gestational day 4. On the other hand, strong beta4 immunoreactivity was detected on surface epithelium and glandular cells on gestational day 12 but no beta1 reactivity was present in the surface epithelium and glandular cells on day 12. In conclusion, both beta1 and beta4 integrins may have a role in implantation process because positive immunoreactivity was seen on apical membrane of surface epithelium on day 4 when implantation occurred. The localization to apical pole of surface epithelium suggest a role for beta1, beta4 integrins in initial embryo and endometrium interaction. It does not seem that beta1 integrin has a role supporting pregnancy since expression of beta4 on surface epithelium and glandular epithelium disappeared on day 12. beta4 integrin expression increasing on day 12 of pregnancy leads us to think a possible functional role supporting pregnancy.     

Isoform-specific effects of the beta2 subunit on voltage-gated sodium channel gating

Voltage-gated sodium channels (Nav) are complex glycoproteins comprised of an alpha subunit and often one to several beta subunits. We have shown that sialic acid residues linked to Nav alpha and beta1 subunits alter channel gating. To determine whether beta2-linked sialic acids similarly impact Nav gating, we co-expressed beta2 with Nav1.5 or Nav1.2 in Pro5 (complete sialylation) and in Lec2 (essentially no sialylation) cells. Beta2 sialic acids caused a significant hyperpolarizing shift in Nav1.5 voltage-dependent gating, thus describing for the first time an effect of beta2 on Nav1.5 gating. In contrast, beta2 caused a sialic acid-independent depolarizing shift in Nav1.2 gating. A deglycosylated mutant, beta(2-DeltaN), had no effect on Nav1.5 gating, indicating further the impact of beta2 N-linked sialic acids on Nav1.5 gating. Conversely, beta(2-DeltaN) modulated Nav1.2 gating virtually identically to beta2, confirming that beta2 N-linked sugars have no impact on Nav1.2 gating. Thus, beta2 modulates Nav gating through multiple mechanisms possibly determined by the associated alpha subunit. Beta1 and beta2 were expressed together with Nav1.5 or Nav1.2 in Pro5 and Lec2 cells. Together beta1 and beta2 produced a significantly larger sialic acid-dependent hyperpolarizing shift in Nav1.5 gating. Under fully sialylating conditions, the Nav1.2.beta1.beta2 complex behaved like Nav1.2 alone. When sialylation was reduced, only the sialic acid-independent depolarizing effects of beta2 on Nav1.2 gating were apparent. Thus, the varied effects of beta1 and beta2 on Nav1.5 and Nav1.2 gating are apparently synergistic and highlight the complex manner, through subunit- and sugar-dependent mechanisms, by which Nav activity is modulated.     

Transforming growth factor beta 1-specific binding proteins on human vascular endothelial cells.

Transforming growth factor beta (TGF beta) regulates the growth of human umbilical vein endothelial cells (HUVEC) differently depending on the isoform of TGF beta and the culture conditions. The cells are resistant to growth inhibition by TGF beta when the cells are cultured on substratum coated with gelatin. However, the proliferation of HUVEC cultured on substratum without a gelatin coating is inhibited by TGF beta, depending on the isoform and concentration of TGF beta. Binding assays with 125I-TGF beta 1 reveal that HUVEC contain a single class of high-affinity (Kd = 4.4 pM) TGF beta 1 binding sites with 8500 sites per cell. Affinity cross-linking studies demonstrate that HUVEC express 180 and 80 kDa TGF beta 1 binding sites that do not bind TGF beta 2. The reduction and the removal of glycosaminoglycans does not affect the electrophoretic mobility of the 180-kDa binding protein cross-linked with 125I-TGF beta 1. Therefore, the 180-kDa TGF beta 1 binding protein is not related to the type III TGF beta receptor, but might be a novel TGF beta 1-specific receptor/binding protein expressed on vascular endothelial cells. The expression of TGF beta 1 binding sites is not affected by the presence or absence of the gelatin coating on the culture substratum. The data suggest that a gelatin coating does not regulate the susceptibility of HUVEC to TGF beta 1 at the level of the receptor/binding proteins, and that growth inhibition of HUVEC by TGF beta 1 is linked to the regulation of extracellular matrices required for the interaction between the cells and the substratum.     

Biosynthesis of blood group I and i antigens in rat tissues. Identification of a novel beta 1-6-N-acetylglucosaminyltransferase.

The beta-galactoside beta 1-6- and beta 1-3-N-acetylglucosaminyltransferases (beta 1-6GnT and beta 1-3GnT) that synthesize blood group I and i antigens were identified in rat tissues, using pyridylaminated lacto-N-neotetraose (Gal beta 1-4GlcNAc beta 1-3Gal beta 1-4Glc-PA) as an acceptor. The products of the transferase reactions were separated on high performance liquid chromatography. The products of the transferase reactions were identified by 1H NMR as (formula; see text) and GlcNAc beta 1-3Gal beta 1-4GlcNAc beta 1-3Gal beta 1-4Glc-PA. The product for beta 1-6GnT was also identified by methylation analysis. Kinetic experiments were carried out using rat serum for beta 1-3GnT and partially purified enzyme from rat intestine for beta 1-6GnT. beta 1-3GnT has a pH optimum of 7.5 and requires Mn2+ for optimal activity. beta 1-6GnT has a pH optimum of 7.0 and does not require Mn2+. Studies on the substrate specificity of each enzyme indicated that the preferred substrate for beta 1-3GnT had the general structure Gal beta 1-4GlcNAc-OR and, for beta 1-6GnT, Gal beta 1-4GlcNAc beta 1-3Gal-OR where R = sugar. This is the first demonstration that the beta 1-6GnT acts on an internal galactose of lacto-N-neotetraose and paragloboside, and the enzyme appears to be a novel enzyme in terms of substrate specificity.     

The human beta 3 alcohol dehydrogenase subunit differs from beta 1 by a Cys for Arg-369 substitution which decreases NAD(H) binding

The beta 3 beta 3 (formerly called beta Indianapolis) and beta 1 beta 1 isoenzymes of human alcohol dehydrogenase differ substantially in their catalytic properties. Specifically, the KM value for NAD+ of beta 3 beta 3 is 70 times greater than that of beta 1 beta 1, and the Ki value for NADH is 35 times greater than that of beta 1 beta 1. To identify the structural basis of these catalytic differences, we sequenced regions of the beta 3 subunit and the beta 3 gene. beta 3 differs from beta 1 by the substitution of Cys for Arg-369. Based on x-ray crystallography of horse ADH, Arg-369 should interact with the nicotinamide phosphate moiety of NAD(H). The Cys for Arg-369 substitution would decrease the enzyme's affinity for coenzyme and, thus, account for the observed kinetic differences between beta 3 beta 3 and beta 1 beta 1.     

Beta 1 and beta 3 integrins have different roles in the adhesion and migration of vascular smooth muscle cells on extracellular matrix.

Extracellular matrix receptors on ductus arteriosus smooth muscle cells (SMC) must enable the cells to migrate through both interstitial and basement membrane matrices to form intimal mounds during postnatal ductus closure. We examined the role of beta 1 and beta 3 integrin receptors on SMC adhesion and migration. Using a new assay to measure cell migration, we found that lamb ductus arteriosus SMC attach to and migrate over surfaces coated with fibronectin (FN), laminin (LN), vitronectin (VN), and collagens I (I) and IV (IV). Blocking antibodies, specific to different integrin complexes, showed that SMC adhesion to FN, LN, I, and IV depended exclusively on functioning beta 1 integrins with little, if any, contribution by the alpha V beta 3 integrin; on the other hand, cell migration over these substrates depended to a large extent on the alpha V beta 3 receptor. Immunofluorescent staining demonstrated that during the early phase of SMC migration, the beta 1 integrins organized rapidly into focal plaques that, with time, gradually covered the cell's basal surface; on the other hand, the beta 3 receptor remained concentrated at all times at the cell's margins. Ligand affinity chromatography and immunoprecipitation techniques identified a unique series of beta 1 integrins binding to each matrix component: FN (alpha 5 beta 1, alpha 3 beta 1, alpha V beta 1), LN (alpha 1 beta 1, alpha 7 beta 1), VN (alpha V beta 1), I (alpha 1 beta 1, alpha 2 beta 1), and IV (alpha 1 beta 1). In contrast, the beta 3 integrin, alpha V beta 3, bound to all the substrates tested: FN, LN, VN, I, and IV. The results indicate that beta 1 and beta 3 integrins may play different roles in attachment and migration as SMC move through the vascular extracellular matrix to produce obliteration of the ductus arteriosus lumen.     

The alpha v beta 1 integrin functions as a fibronectin receptor but does not support fibronectin matrix assembly and cell migration on fibronectin

The fibronectin receptor, alpha 5 beta 1, has been shown to be required for fibronectin matrix assembly and plays an important role in cell migration on fibronectin. However, it is not clear whether other fibronectin binding integrins can take the place of alpha 5 beta 1 during matrix assembly and cell migration. To test this, we expressed the human alpha v subunit in the CHO cell line CHO-B2 that lacks the alpha 5 subunit. We found that the human alpha v combined with CHO cell beta 1 to form the integrin alpha v beta 1. Cells that expressed alpha v beta 1 attached to and spread well on fibronectin-coated dishes, but did so less well on vitronectin-coated dishes. This, along with other data, indicated that alpha v beta 1 functions as a fibronectin receptor in CHO-B2 cells. The alpha v beta 1-expressing cells failed to produce a fibronectin matrix or to migrate on fibronectin, although the same cells transfected with alpha 5 do produce a matrix and migrate on fibronectin. The affinity of the alpha v beta 1-expressing cells for fibronectin was fourfold lower than that of the alpha 5 beta 1- expressing cells. In addition, alpha v beta 1 was distributed diffusely throughout the cell surface, whereas alpha 5 beta 1 was localized to focal adhesions when cells were seeded onto fibronectin-coated surfaces. Thus, of the two fibronectin receptors, alpha v beta 1 and alpha 5 beta 1, only alpha 5 beta 1 supports fibronectin matrix assembly and promotes cell migration on fibronectin in the CHO-B2 cells. Possible reasons for this difference in the activities of alpha v beta 1 and alpha 5 beta 1 include the lower affinity of alpha v beta 1 for fibronectin and the failure of this integrin to localize in adhesion plaques on a fibronectin substrate. These results show that two integrins with similar ligand specificities and cell attachment functions may be quite different in their ability to support fibronectin matrix assembly and cell motility on fibronectin.     

Properties of the interspecies hybrids between haptoglobin alpha and beta subunits

1. Subunits alpha isolated from human haptoglobin were recombined with beta subunits of equine haptoglobin, and vice versa. Both hybrid proteins were separated on electrophoresis in polyacrylamide gel into four bands with mobilities corresponding to tetramers 2alpha.2beta, trimers 2alpha.beta, and dimers alpha.beta, in addition to free subunits beta. 2. The binding ability of haemoglobin and the antigenic specificity of tetramers depended on the origin of beta subunit. 3. Reduction of native and hybrid proteins with 2-mercaptoethanol led to gradual formation of alpha.beta, alpha, and beta; the components 2alpha.beta and 2alpha appeared in trace amounts.     

Individual embryonic fibroblasts express multiple beta chains in association with the alpha v integrin subunit. Loss of beta 3 expression with cell confluence.

The alpha chain of the vitronectin receptor, alpha v, has been found in association with the integrin subunits beta 1, beta 3, or beta 5 on different cell types. We show here that cultured embryonic fibroblasts simultaneously display alpha v beta 3, alpha v beta 1, and alpha v in association with two other beta subunits, one of which is probably beta 5. Polymerase chain reaction analysis of single cells isolated by micromanipulation identified mRNA for alpha v, beta 1, beta 3, and beta 5 in six of eight clones. Immunoprecipitation of iodinated cell surface proteins with a monoclonal antibody to alpha v indicated that the relative proportions of the different beta chains in association with alpha v varied, particularly between two different cell lines. The cytokines platelet-derived growth factor, transforming growth factor beta 1, and tumor necrosis factor alpha did not appear to alter this ratio although tumor necrosis factor alpha increased the surface expression of the alpha v-associated integrins; but overnight culture in basic fibroblast growth factor caused a lower expression of alpha v beta 1 and alpha v beta 5 with no reduction in alpha v beta 3 expression. When the cell cultures were grown to complete confluence, surface expression of beta 3 was abolished, and the expression of an unknown beta chain (beta u) became more prominent. This effect was not overcome by culturing confluent cells with basic fibroblast growth factor. Affinity column chromatography showed that alpha v beta 5 bound to vitronectin but alpha v beta 1 did not, whereas alpha v beta 1 but not alpha v beta 5 bound to fibronectin. These results suggest that, on individual cells, the beta subunits found in association with alpha v may vary according to the proliferative capacity of the cell and that the promiscuous beta 3 subunit is progressively replaced by beta subunits of individual ligand specificity.     

Altered localization and cytoplasmic domain-binding properties of tyrosine-phosphorylated beta 1 integrin

We describe a novel approach to study tyrosine-phosphorylated (PY) integrins in cells transformed by virally encoded tyrosine kinases. We have synthesized a peptide (PY beta 1 peptide) that represents a portion of the cytoplasmic domain of the beta 1 integrin subunit and is phosphorylated on the tyrosine residue known to be the target of oncogenic tyrosine kinases. Antibodies prepared against the PY beta 1 peptide, after removal of cross-reacting antibodies by absorption and affinity purification, recognized the PY beta 1 peptide and the tyrosine-phosphorylated form of the intact beta 1 subunit, but did not bind the nonphosphorylated beta 1 peptide, the nonphosphorylated beta 1 subunit or other unrelated tyrosine-phosphorylated proteins. The anti- PY beta 1 antibodies labeled the podosomes of Rous sarcoma virus- transformed fibroblasts, but did not detectably stain nontransformed fibroblasts. The localization of the tyrosine phosphorylated beta 1 subunits appeared distinct from that of the beta 1 subunit. Adhesion plaques were stained by the anti-beta 1 subunit antibodies in Rous sarcoma virus-transformed fibroblasts plated on fibronectin, whereas neither podosomes nor adhesion plaques were labeled on vitronectin or on uncoated plates. Anti-phosphotyrosine antibodies labeled podosomes, adhesion plaques and cell-cell boundaries regardless of the substratum. One of the SH2 domains of the p85 subunit of phosphatidylinositol-3- kinase bound to the PY beta 1 peptide, but not to the non- phosphorylated beta 1 cytoplasmic peptide. Other SH2 domains did not bind to the PY beta 1 peptide. These results show that the phosphorylated form of the beta 1 integrin subunit is detected in a different subcellular localization than the nonphosphorylated form and suggest that the phosphorylation on tyrosine of the beta 1 subunit cytoplasmic domain may affect cellular signaling pathways.     

The Golgi localization of phosphatidylinositol transfer protein beta requires the protein kinase C-dependent phosphorylation of serine 262 and is essential for maintaining plasma membrane sphingomyelin levels

Recombinant mouse phosphatidylinositol transfer protein (PI-TP)beta is a substrate for protein kinase C (PKC)-dependent phosphorylation in vitro. Based on site-directed mutagenesis and two-dimensional tryptic peptide mapping, Ser(262) was identified as the major site of phosphorylation and Ser(165) as a minor phosphorylation site. The phospholipid transfer activities of wild-type PI-TP beta and PI-TP beta(S262A) were identical, whereas PI-TP beta(S165A) was completely inactive. PKC-dependent phosphorylation of Ser(262) also had no effect on the transfer activity of PI-TP beta. To investigate the role of Ser(262) in the functioning of PI-TP beta, wtPI-TP beta and PI-TP beta(S262A) were overexpressed in NIH3T3 fibroblast cells. Two-dimensional PAGE analysis of cell lysates was used to separate PI-TP beta from its phosphorylated form. After Western blotting, wtPI-TP beta was found to be 85% phosphorylated, whereas PI-TP beta(S262A) was not phosphorylated. In the presence of the PKC inhibitor GF 109203X, the phosphorylated form of wtPI-TP beta was strongly reduced. Immunolocalization showed that wtPI-TP beta was predominantly associated with the Golgi membranes. In the presence of the PKC inhibitor, wtPI-TP beta was distributed throughout the cell similar to what was observed for PI-TP beta(S262A). In contrast to wtPI-TP beta overexpressors, cells overexpressing PI-TP beta(S262A) were unable to rapidly replenish sphingomyelin in the plasma membrane upon degradation by sphingomyelinase. This implies that PKC-dependent association with the Golgi complex is a prerequisite for PI-TP beta to express its effect on sphingomyelin metabolism.     

Identification of subunits mediating clustering of GABA(A) receptors by rapsyn.

Human embryonic kidney 293 cells transfected with alpha1beta1gamma2, alpha1beta2gamma2, alpha1beta3gamma2, alpha1beta1, alpha1beta2, alpha1beta3, beta3gamma2, or beta3 subunits formed gamma-aminobutyric acidA receptors on the cell surface that could be clustered by rapsyn. In contrast, alpha1, beta1, beta2, or gamma2 subunits, or alpha1gamma2 subunit combinations could not be detected on the surface of transfected cells and could not be clustered by rapsyn. Experiments investigating the ability of rapsyn to cluster chimeras consisting of the N-terminus of the beta3 subunit and the remaining part of the alpha1, beta2 or gamma2 subunits indicated that the intracellular domains of beta1, beta2, beta3 or gamma2 subunits, but not those of alpha1 subunits are able to form sites mediating clustering by rapsyn. These results demonstrate that rapsyn has the potential to cluster the majority of GABA(A) receptor subtypes via beta or gamma2 subunits. Further experiments will have to clarify the physiological importance of this observation.     

Sodium channel beta1 subunits promote neurite outgrowth in cerebellar granule neurons

Many immunoglobulin superfamily members are integral in development through regulation of processes such as growth cone guidance, cell migration, and neurite outgrowth. We demonstrate that homophilic interactions between voltage-gated sodium channel beta1 subunits promote neurite extension in cerebellar granule neurons. Neurons isolated from wild-type or beta1(-/-) mice were plated on top of parental, mock-, or beta1-transfected fibroblasts. Wild-type neurons consistently showed increased neurite length when grown on beta1-transfected monolayers, whereas beta1(-/-) neurons showed no increase compared with control conditions. beta1-mediated neurite extension was mimicked using a soluble beta1 extracellular domain and was blocked by antibodies directed against the beta1 extracellular domain. Immunohistochemical analysis suggests that the beta1 and beta4 subunits, but not beta2 and beta3, are expressed in cerebellar Bergmann glia as well as granule neurons. These results suggest a novel role for beta1 during neuronal development and are the first demonstration of a functional role for sodium channel beta subunit-mediated cell adhesive interactions.     

Interactions of pure beta gamma-subunits of G-proteins with purified beta 1-adrenoceptor

The role of the beta gamma-subunits in the interaction of G-proteins was examined with beta 1-adrenoceptors purified from turkey erythrocytes and pure beta gamma-subunits prepared from turkey erythrocytes and bovine brain. On a non-denaturing polyacrylamide gel, the mobility of beta gamma-subunits was increased when incubated with beta 1-adrenoceptor and the beta 1-adrenergic agonist 1-(-)-isoproterenol, whereas on incubation with the antagonist 1-alprenolol the mobility was unchanged. Furthermore, the beta 1-adrenoceptor was retarded on a Sephadex G-50 column equilibrated with beta gamma-subunits and agonist. No retardation occurred in the presence of antagonist. These data suggest a direct interaction of activated beta 1-adrenoceptors with isolated beta gamma-subunits of G-proteins.     

Analysis of gene structure and antigen determinants of DR2 antigens using DR gene transfer into mouse L cells

Three HLA class II DR beta genes and one DR alpha gene from the DR2 haplotype were cloned in cosmid vectors. The DR beta II gene might be a pseudogene lacking the first exon that encodes the leader peptide. The DR beta I and DR beta III genes were expressed, together with the DR alpha-chain, after transfection into mouse L cells. Restriction enzyme mapping of the DR beta genomic clones and reactivity of their products expressed on the L cell transfectant against mAb showed that the DR beta I and DR beta III genes encoded the nonpolymorphic and polymorphic DR beta chain, respectively. This arrangement is the reverse of that observed in other haplotypes, such as DR3, 4 and 6. The alignment of the HLA class II genes including the DR beta genes on the chromosome 6, however, was consistent with other haplotypes, e.g., centromere-DX beta-DX alpha-DV beta-DQ beta-DQ alpha-DR beta I-DR beta II- DR beta III-DR alpha-telomere. These results suggest that the susceptibility to mutations or gene conversions responsible for genetic polymorphisms depends on the gene itself and not on its location. Furthermore, absorption experiments of anti-DR2 allosera by the DR alpha/DR beta transfectants revealed that the so-called DR2 specificities were determined by multiple epitopes although both the DR beta I and DR beta III genes behaved similarly with DR2-specific antibodies.     

Characterization of novel complexes on the cell surface between integrins and proteins with 4 transmembrane domains (TM4 proteins).

Here we identified several new integrin/TM4 protein complexes on the cell surface. By immunoprecipitation using nonstringent conditions, and by reciprocal immunoprecipitation, we found that alpha 3 beta 1 and alpha 6 beta 1 integrins but not alpha 2 beta 1, alpha 5 beta 1, or alpha 6 beta 4 integrins associated with CD9 and CD81 in alpha 3 beta 1/CD81, alpha 3 beta 1/CD9, alpha 6 beta 1/CD81, and alpha 6 beta 1/CD9 complexes. Also, cross-linking experiments established that alpha 3 beta 1/CD81, alpha 3 beta 1/CD9, and alpha 3 beta 1/CD63 associations occur on the surface of intact cells and suggested that a critical interaction site is located within extracellular domains. Cross-linking in conjunction with reimmunoprecipitation indicated that larger multi-component alpha 3 beta 1/TM4/TM4 complexes (alpha 3 beta 1/CD9/CD63, alpha 3 beta 1/CD81/CD63, and alpha 3 beta 1/CD9/CD81) also could be detected on the cell surface. Immunofluorescent staining showed redistribution of alpha 3 beta 1/TM4 complexes toward the periphery of cells plated on various extracellular matrix substrates and also showed that these complexes were localized in cell footprints. Staining of human tissues yielded additional results consistent with co-localization of alpha 3 beta 1 and CD9, CD63, and CD81 proteins. In conclusion we suggest that the prevalence of integrin/TM4 complexes in diverse cellular environments is indicative of their general physiological importance.     

Identification of testis specific calcineurin beta subunit isoform by a monoclonal antibody and detection of a specific six amino acid sequence.

Two isoforms of calcineurin beta subunit(beta 1 and beta 2) were identified in rat testis by a monoclonal antibody Va1. Both beta 1 and beta 2 were recovered in calmodulin binding protein fraction and showed calcium shift on SDS-polyacrylamide gel electrophoresis which is the specific character for EF-hand calcium binding protein. beta 2 showed same apparent molecular weight on SDS-PAGE as that of brain calcineurin beta and was found in wide variety of tissues. beta 1 was shown to have six amino acid polypepeptide sequence and it showed higher molecular weight than brain beta and was specific for testis.     

Differential effects of proteases involved in intracellular degradation of amyloid beta-protein between detergent-soluble and -insoluble pools in CHO-695 cells.

The deposition of amyloid beta-protein (A beta or beta A4) is a key feature of Alzheimer's disease. Most studies have focused on the generation of A beta, but little is known about the degradation of A beta. Recent reports suggest that insulin-degrading enzyme (IDE) and neutral endopeptidase (NEP) are involved in the extracellular degradation of A beta. To date, however, far less is known about the degradation of intracellular A beta. To elucidate the protease(s) responsible for the degradation of intracellular A beta, we investigated the effect of various protease inhibitors on A beta in two distinct intracellular pools (i.e., nonionic detergent-soluble and detergent-insoluble pools) in Chinese hamster ovary cells. Treatment with thiol and metal inhibitors resulted in the accumulation of intracellular A beta and oligomers in detergent-soluble and -insoluble fractions. The overexpression of thiol-metalloprotease IDE resulted in a marked reduction in levels of detergent-soluble intracellular A beta as well as extracellular A beta 40 and A beta 42. Moreover, intracellular A beta in the detergent-insoluble fraction extracted with 70% formic acid or 6 M guanidine hydrochloride decreased markedly in the cells overexpressing IDE. In contrast, expression of NEP degraded the A beta in the detergent-insoluble fraction markedly and partially degraded extracellular A beta 40 and A beta 42, but not intracellular soluble A beta. Thiorphan, an inhibitor of NEP, accumulated, albeit to a lesser extent, in insoluble A beta but not in soluble A beta. Thus, IDE appears to degrade intracellular A beta more effectively than does NEP in both the detergent-soluble and -insoluble fractions.     

The integrin alpha5beta1 regulates chondrocyte hypertrophic differentiation induced by GTP-bound transglutaminase 2.

Soluble GTP-bound transglutaminase 2 (TG2) induces hypertrophic differentiation in chondrocyte cultures in a beta1 integrin-dependent fashion. beta1 integrin subfamily consists of 12 heterodimers with 12 different alpha subunits and a beta1 subunit. To identify the specific integrin heterodimer(s) responsible for this process, we specifically blocked individual beta1 integrins on the CH-8 immortalized human chondrocytes during hypertrophic differentiation. Blockade of alpha5beta1 inhibited matrix metalloproteinase 13 (MMP-13), type X collagen expression, alkaline phosphatase activity and matrix calcification by 30-50% associated with weak effects of anti-alpha3beta1 and -alpha4beta1. Anti-alpha1beta1, -alpha2beta1 and -alpha6beta1 had no effect. To examine whether the dominant effect of integrin alpha5beta1 was due to a direct interaction with TG2, we incubated the chondrocytic cells on plates coated with GTP-bound TG2. The immobilized GTP-bound TG2 induced hypertrophic differentiation to the same extent as the soluble GTP-bound TG2, which was also inhibited by anti-alpha5beta1. CH-8 cells grown on plates coated with GTP-bound TG2 demonstrated adherence associated with focal adhesion kinase phosphorylation. These properties were inhibited by anti-alpha5beta1. Furthermore, engagement of alpha5beta1 on CH-8 cells via anti-alpha5beta1 antibody did, in fact, induce differentiation. Although CH-8 cells adhered to GTP-free TG2 via integrin alpha5beta1, the cells failed to undergo hypertrophic differentiation. Thus, integrin alpha5beta1 is critical for the chondrocyte hypertrophic differentiation induced by GTP-bound TG2, and this induction is ligand dependent.