TGF-β binding in human Wharton’s jelly

Our previous study reported that TGF-β may be isolated from human Wharton’s jelly (WJ) in a form of soluble, high molecular complex(es). We decided to study the effect of extracellular matrix degradation and reduction of disulphide bridges reduction on the release of TGF-β from WJ. The WJ prepared from the umbilical cords of newborns delivered at term by healthy mothers was homogenised and treated with hyaluronidase, collagenase, heparinase, chondroitinase and β-mercaptoethanol, the resulting extracts were then submitted to TGF-β immunoassay and SDS/PAGE followed by Western immunoblotting. The effect of metalloproteinase activation on TGF-β was also studied. Pre-treatment of WJ homogenates with hyaluronidase or collagenase markedly increased the extractability of TGF-β, but did not dissociate the complexes. In contrast, the action of β-mercaptoethanol resulted in the release of free TGF-β; but activation of metalloproteinases resulted in the disappearance of this factor. We conclude that TGF-β1 is bound through disulphide bonds to an extracellular matrix component of WJ. The large amount of collagen fibrils and hyaluronate molecules which surround the cells scattered in WJ may prevent the access of extracting solution to TGF-β causing a low extractability of this factor. Although hyaluronate and collagen do not bind TGF-β directly, they may present a barrier that prevents the diffusion of TGF-β in WJ and results in its concentration around the cells thereby facilitating its interaction with membrane receptors and subsequent stimulation of cell division and synthesis of extracellular matrix components.     

The critical interaction of the metallopeptidase PHEX with heparan sulfate proteoglycans

The PHEX gene (phosphate-regulating gene with homologies to endopeptidase on the X chromosome) identified as a mutated gene in patients with X-linked hypophosphatemia (XLH), encodes a protein (PHEX) that shows striking homologies to members of the M13 family of zinc metallopeptidases. In the present work the interaction of glycosaminoglycans with PHEX has been investigated by affinity chromatography, circular dichroism, protein intrinsic fluorescence analysis, hydrolysis of FRET substrates flow cytometry and confocal microscopy. PHEX was eluted from a heparin-Sepharose chromatography column at 0.8 M NaCl showing a strong interaction with heparin. Circular dichroism spectra and intrinsic fluorescence analysis showed that PHEX is protected by glycosaminoglycans against thermal denaturation. Heparin, heparan sulfate and chondroitin sulfate inhibited PHEX catalytic activity, however among them, heparin presented the highest inhibitory activity (K_i = 2.5 ± 0.2 nM). Flow cytometry analysis showed that PHEX conjugated to Alexa Fluor 488 binds to the cell surface of CHO-K1, but did not bind to glycosaminoglycans defective cells CHO-745. Endogenous PHEX was detected at the cell surface of CHO-K1 colocalized with heparan sulfate proteoglycans, but was not found at the cell surface of glycosaminoglycans defective cells CHO-745. In permeabilized cells, PHEX was detected in endoplasmic reticulum of both cells. In addition, we observed that PHEX colocalizes with heparan sulfate at the cell surface of osteoblasts. This is the first report that the metallopeptidase PHEX is a heparin binding protein and that the interaction with GAGs modulates its enzymatic activity, protein stability and cellular trafficking.     

Correlation of C4ST-1 and ChGn-2 expression with chondroitin sulfate chain elongation in atherosclerosis

Subendothelial retention of lipoproteins by proteoglycans (PGs) is the initiating event in atherosclerosis. The elongation of chondroitin sulfate (CS) chains is associated with increased low-density lipoprotein (LDL) binding and progression of atherosclerosis. Recently, it has been shown that 2 Golgi enzymes, chondroitin 4-O-sulfotransferase-1 (C4ST-1) and chondroitin N-acetylgalactosaminyltransferase-2 (ChGn-2), play a critical role in CS chain elongation. However, the roles of C4ST-1 and ChGn-2 during the progression of atherosclerosis are not known. The aim of this study was to analyze the expression of C4ST-1 and ChGn-2 in atherosclerotic lesions in vivo and determine whether their expression correlated with CS chain elongation.Low-density lipoprotein receptor knockout (LDLr KO) mice were fed a western diet for 2, 4, and 8 weeks to stimulate development of atherosclerosis. The binding of LDL and CS PG in this mouse model was confirmed by chondroitinase ABC (ChABC) digestion and apolipoprotein B (apo B) staining. Gel filtration analysis revealed that the CS chains began to elongate as early as 2 weeks after beginning a western diet and continued as the atherosclerosis progressed. Furthermore, quantitative real-time polymerase chain reaction (qRT-PCR) showed that the mRNA levels of C4ST-1 and ChGn-2 increased after 8 weeks of this diet. In contrast, the mRNA levels of their homologs, C4ST-2 and ChGn-1, were unchanged. In addition, immunohistochemical analysis demonstrated that the expression of C4ST-1 and ChGn-2 appeared to have similar site-specific patterns and coincided with biglycan expression at the aortic root.Our results suggested that C4ST-1 and ChGn-2 may be involved in the elongation of CS chains in the arterial wall during the progression of atherosclerosis. Therefore, modulating their expression and activity might be a novel therapeutic strategy for atherosclerosis.     

Caveolae-mediated endocytosis of the glucosaminoglycan-interacting adipokine tartrate resistant acid phosphatase 5a in adipocyte progenitor lineage cells

Adipogenesis depends on growth factors controlling proliferation/differentiation of mesenchymal stem cells (MSCs). Membrane binding and endocytosis of growth factors are often coupled to receptor activation and downstream signaling leading to specific cellular responses. The novel adipokine tartrate-resistant acid phosphatase (TRAP) 5a exhibits a growth factor-like effect on MSCs and pre-adipocytes and induces hyperplastic obesity in vivo. However its molecular interaction with pre-adipocytes remains unknown. Therefore, this study aimed to investigate membrane interaction of TRAP and its endocytosis routes in pre-adipocytes. Confocal and/or electron microscopy were used to detect TRAP in untreated or TRAP 5a/b treated pre-adipocytes under conditions that allow or inhibit endocytosis in combination with co-staining of endocytotic vesicles. TRAP interaction with heparin/heparan sulfate was verified by gel filtration. It could be shown that TRAP 5a, but not 5b, binds to the membrane of pre-adipocytes where it co-localizes with heparin-sulfate proteoglycan glypican-4. Also in vitro, TRAP 5a exhibited affinity for both heparin and heparan sulfate with heparin inhibiting its enzyme activity. Upon caveolae-mediated endocytosis of saturating levels of TRAP 5a, TRAP 5a co-localized intracellularly with glypican-4 and late endosomal marker Rab-7 positive vesicles. The protein was also located in multivesicular bodies (MVBs) but did not co-localize with lysosomal marker LAMP-1. TRAP 5a endocytosis was also detectable in pre-osteoblasts, but not fibroblasts, embryonic MSCs or mature adipocytes. These results indicate that TRAP 5a exhibits binding to cell surface, endocytosis and affinity to glucosaminoglycans (GAGs) in pre-adipocyte and pre-osteoblast lineage cells in a manner similar to other heparin-binding growth factors.     

Chondroitin Sulfate and Chondroitin/Keratan Sulfate Proteoglycans of Nervous Tissue: Developmental Changes of Neurocan and Phosphacan

Abstract: We have studied developmental changes in the structure and concentration of the hyaluronic acid-binding proteoglycan, neurocan, and of phosphacan, another major chondroitin sulfate proteoglycan of nervous tissue that represents the extracellular domain of a receptor-type protein tyrosine phosphatase. A new monoclonal antibody (designated 1F6), which recognizes an epitope in the N-terminal portion of neurocan, has been used for the isolation of proteolytic processing fragments that occur together with link protein in a complex with hyaluronic acid. Both link protein and two of the neurocan fragments were identified by amino acid sequencing. The N-terminal fragments of neurocan are also recognized by monoclonal antibodies (5C4, 8A4, and 3B1) to epitopes in the G1 and G2 domains of aggrecan and/or in the hyaluronic acid-binding domain of link protein. The presence in brain of these N-terminal fragments is consistent with the developmentally regulated appearance of the C-terminal half of neurocan, which we described previously. We have also used a slot-blot radioimmunoassay to determine the concentrations of neurocan and phosphacan in developing brain. The levels of both proteoglycans increased rapidly during early brain development, but whereas neurocan reached a peak at approximately postnatal day 4 and then declined to below embryonic levels in adult brain, the concentration of phosphacan remained essentially unchanged after postnatal day 12. Keratan sulfate on phosphacan-KS (a glycoform that contains both chondroitin sulfate and keratan sulfate chains) was not detectable until just before birth, and its peak concentration (at 3 weeks postnatal) was reached ～1 week later than that of the phosphacan core protein. Immunocytochemical studies using monoclonal antibodies to keratan sulfate (3H1 and 5D4) together with specific glycosidases (endo-β-galactosidase, keratanase, and keratanase II) also showed that with the exception of some very localized areas, keratan sulfate is generally not present in the embryonic rat CNS.     

Perlecan Binds to the β-Amyloid Proteins (Aβ) of Alzheimer's Disease, Accelerates Aβ Fibril Formation, and Maintains Aβ Fibril Stability

Abstract: Perlecan is a specific heparan sulfate proteoglycan that accumulates in the fibrillar β-amyloid (Aβ) deposits of Alzheimer's disease. Perlecan purified from the Engelbreth-Holm-Swarm tumor was used to define perlecan's interactions with Aβ and its effects on Aβ fibril formation. Using a solid-phase binding immunoassay, freshly solubilized full-length Aβ peptides bound immobilized perlecan at two sites, representing both high-affinity [K_D = ～5.8 × 10^?11M for Aβ (1–40); K_D = ～6.5 × 10^?12M for Aβ (1–42)] and lower-affinity [K_D = 3.5 × 10^?8M for Aβ (1–40); K_D = 4.3 × 10^?8M for Aβ (1–42)] interactions. An increase in the binding capacity of Aβ (1–40) to perlecan correlated with an increase in Aβ amyloid fibril formation during a 1-week incubation period. The high-capacity binding of Aβ (1–40) to perlecan was similarly observed using perlecan heparan sulfate glycosaminoglycans and was completely abolished by heparin, but not by chondroitin-4-sulfate. Using a thioflavin T fluorometry assay, perlecan accelerated the rate of Aβ (1–40) amyloid fibril formation, causing a significant increase in Aβ fibril assembly over a 2-week incubation period at 1 h (2.8-fold increase), 1 day (3.6-fold increase), and 3 days (2.8-fold increase) in comparison with Aβ (1–40) alone. Perlecan also initially accelerated the formation of Aβ (1–42) fibrils within 1 h and maintained significantly higher levels of Aβ (1–42) thioflavin T fluorescence throughout a 2-week experimental period in comparison with Aβ (1–42) alone, suggesting perlecan's ability to maintain amyloid fibril stability. Perlecan's effects on Aβ (1–40) fibril formation and maintenance of Aβ (1–42) fibril stability occurred in a dose-dependent manner and was also mediated primarily by perlecan's glycosaminoglycan chains. Perlecan was the most effective enhancer and accelerator of Aβ fibril formation when compared directly with other amyloid plaque components, including apolipoprotein E, α₁-antichymotrypsin, P component, C1q, and C3. This study, therefore, demonstrates that perlecan not only binds to the predominant isoforms of Aβ, but also accelerates Aβ fibril formation and stabilizes amyloid fibrils once formed, confirming pivotal roles for perlecan in the pathogenesis of Aβ amyloidosis in Alzheimer's disease.     

Release of Chromaffin Granule Glycoproteins and Proteoglycans from Potassium-Stimulated PC12 Pheochromocytoma Cells

Abstract: Cultured PC12 pheochromocytoma cells were labeled with [³H]gIucosamine, and the glycoproteins and proteoglycans released following potassium-induced depolarization were fractionated and characterized. Exposure of PC12 cells for 20 min to a high concentration of potassium (51.5 mM in Krebs-Ringers-HEPES buffer) results in an approximately sixfold increase in the release of labeled glycoproteins and proteoglycans, compared to incubation in physiological levels of potassium (6 mM ). The released complex carbohydrates include chromogranins, dopamine β-hydroxylase, and two chondroitin sulfate/heparan sulfate proteoglycan fractions, which together account for 7.4% of the soluble cell radioactivity. The chromogranins contained galactosyl(β l ± 3 )N-ace tylgalactosamine, as well as several mono- and disialyl O -glycosidically-linked oligosaccharides, and the tetra-saccharide AcNeu(α2 ± 3)Gal(β l ± 3)[AcNeu(α2 ± (6)] GalNAcol, obtained by alkaline borohydride treatment of the chromogranin glycopeptides, accounted for almost half of the total chromogranin labeling. The proteoglycan fractions varied in their relative proportions of chondroitin sulfate (23–68%), heparan sulfate (16–23%), and glycoprotein oligosaccharides (16–54%), which are of the triand tetraantennary and O -glycosidic types. As previously found in the case of proteoglycans from bovine chromaffin granules, the more acidic species has a considerably higher proportion of carbohydrate in the form of sulfated glycosaminoglycans.     

Structural Properties of the Heparan Sulfate Proteoglycans of Brain

The heparan sulfate proteoglycans present in a deoxycholate extract of rat brain were purified by ion exchange chromatography, affinity chromatography on lipoprotein lipase agarose, and gel filtration. Heparitinase treatment of the heparan sulfate proteoglycan fraction (containing 86% heparan sulfate and 10% chondroitin sulfate) that was eluted from the lipoprotein lipase affinity column with 1 M NaCl led to the appearance of a major protein core with a molecular size of 55,000 daltons, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Comparison of the effects of heparinase and heparitinase treatment revealed that the heparan sulfate proteoglycans of brain contain a significant proportion of relatively short N-sulfoglucosaminyl 6-O-sulfate [or N-sulfoglucosaminyl](alpha 1-4)iduronosyl 2-O-sulfate(alpha 1-4) repeating units and that the portions of the heparan sulfate chains in the vicinity of the carbohydrate-protein linkage region are characterized by the presence of D-glucuronic acid rather than L-iduronic acid. After chondroitinase treatment of a proteoglycan fraction that contained 62% chondroitin sulfate and 21% heparan sulfate (eluted from lipoprotein lipase with 0.4 M NaCl), the charge and density of a portion of the heparan sulfate-containing proteoglycans decreased significantly. These results indicate that a population of "hybrid" brain proteoglycans exists that contain both chondroitin sulfate and heparan sulfate chains covalently linked to a common protein core.