Reduction of Perlecan Synthesis and Induction of TGF-1 in Human Peritoneal Mesothelial Cells Due to High Dialysate Glucose Concentration: Implication in Peritoneal Dialysis
Susan Yung,
Xiao-Rui Chen,
Ryan CW Tsang,
Qing Zhang and
Tak Mao Chan
Division of Nephrology, Department of Medicine, Queen Mary Hospital, The University of Hong Kong, Hong Kong SAR, China
Correspondence to Professor Tak-Mao Chan, Department of Medicine, University of Hong Kong, Queen Mary Hospital, Pokfulam Road, Hong Kong SAR, China. Phone: 852-2855-4041; Fax: 852-2872-5828; E-mail: dtmchan{at}hkucc.hku.hk
ABSTRACT. Prolonged exposure of the peritoneal mesothelium tohigh dialysate glucose concentrations reduces anionic sitesthat are critical to its selective permeability, thereby impairingthe peritoneal transport properties in patients on long-termperitoneal dialysis (PD). Perlecan, an anionic heparan sulfateproteoglycan, is pivotal to the selective permeability of basementmembranes, and high glucose concentrations modulate its synthesisin mesangial cells. The effect of glucose on perlecan expressionin the peritoneal mesothelium has not been established. We investigatedperlecan expression in peritoneal biopsies from patients onPD, and the effect of high glucose concentrations on perlecansynthesis in cultured human peritoneal mesothelial cells (HPMC).Peritoneal biopsies from PD patients showed reduced perlecanexpression compared with controls. Exposure of HPMC to highglucose concentrations resulted in a dose-dependent reductionin the synthesis of perlecan polypeptide and its depositioninto the extracellular matrix. These effects were mediated inpart through the induction of TGF-1. Characterization studiesshowed that perlecan synthesized by HPMC contained solely heparansulfate glycosaminoglycan (HS GAG) chains, and [35S]-incorporationstudies demonstrated progressive reduction of their de novosynthesis with increasing glucose concentrations (68142 ±3658, 48147 ± 2517, 31468 ± 5781, and 25575 ±3621 cpm/µg cellular protein for 5 mM, 30 mM, 75 mM, and120 mM D-glucose, respectively; P < 0.001 for 5 mM versus30 mM D-glucose, and P < 0.0001 for 5 mM versus 75 mM or120 mM D-glucose). Both the length and the charge density ofthe HS GAG chains remained unchanged. Reduction of peritonealperlecan expression in long-term PD was attributed to high dialysateglucose concentrations, which induced TGF-1 and reduced perlecansynthesis in HPMC. Since perlecan can sequester growth factors,thereby modulating cell migration and differentiation perturbationof peritoneal perlecan expression contributes to the structuraland functional changes of the peritoneum in long-term PD.
The peritoneal membrane comprises an overlying monolayer ofmesothelial cells, beneath which lies the interstitium thatcontains fibroblasts, collagen fibrils, and a capillary network.Mesothelial cells play important roles in both peritoneal transportfunctions and remodeling, the latter through its controlledsynthesis of cytokines, chemokines, growth factors, prostaglandins,and matrix proteins (1–4). Injury to the mesothelium representsan important initiating step, leading to the progressive histologicchanges of the peritoneum in peritoneal dialysis (PD). The abundantanionic sites normally present in the mesothelium and the underlyingbasement membrane prevent excessive albumin wastage (5–8).Proteoglycans (PG), which are anionic macromolecules presentubiquitously on the cell surface and in basement membrane (9,10),are putatively major contributors to the anionic sites in theperitoneum. Glycosaminoglycans (GAG), the highly negativelycharged carbohydrate moiety of PG, have been identified in naturalperitoneal fluid and serve to reduce friction and prevent adhesionof the abdominal viscera (11,12). Moreover, we have previouslydemonstrated PG synthesis and secretion in human peritonealmesothelial cells (HPMC) in vitro (13).
PG are composed of a protein backbone to which one or more GAGchain(s) is/are attached, and they serve multi-faceted functionspertaining to filtration and permeability characteristics ofbiologic membranes, cell adhesion, differentiation, proliferation,sequestration of growth factors, and collagen fibrillogenesis(14,15). PG synthesis is modulated by inflammation and chemicalor bacterial insult. Long-term PD is associated with chronicinflammation in the peritoneal membrane leading to progressivestructural and functional deterioration (16–18). Prolongedexposure to high dialysate glucose concentrations is among theimportant etiological factors leading to these changes. Thereis accumulating evidence that some of the pathogenetic mechanismsmay be similar to those in diabetes mellitus, for example, cellhypertrophy, increased synthesis of transforming growth factor-1(TGF-1) and matrix proteins, and increased basement membranepermeability induced by high ambient glucose concentrations(19,20). Altered synthesis of perlecan, a large heparan sulfate(HS) PG that plays a critical role in the selective permeabilityof basement membranes, mitogenesis, angiogenesis, and cell adhesion,has been demonstrated in diabetes mellitus (9). The effect ofglucose on peritoneal perlecan synthesis remains to be elucidated.
We therefore examined in vivo peritoneal perlecan expressionin peritoneal biopsies from patients on PD, and we investigatedthe etiological significance of dialysate glucose and TGF-1on perlecan synthesis in cultured HPMC.
All materials were of the highest grade and were purchased fromSigma (Tin Hang Technology Ltd, Hong Kong) unless otherwisestated. Tissue culture flasks were from Falcon (Becton-Dickinson,Gene Company, Hong Kong), and culture media and supplementswere from Invitrogen Life Technologies (Hong Kong). TGF-1 DuoSetELISA Development System and TGF-1 neutralizing antibody werepurchased from R&D Systems (TWC Biosearch International,Hong Kong). Carrier free Na2[35S]SO4 (950 to 1350 Ci/mmol) waspurchased from Amersham Biosciences China Ltd, Hong Kong. Perlecanantibodies included a kind gift (EY10) from Professor John Hassell(Tampa, Florida) and one purchased from Zymed (7B5 clone, OnwonTrading Ltd, Hong Kong).
Peritoneal Biopsy Studies
Parietal peritoneal biopsies were obtained with informed consentfrom five patients at the time of initial insertion of the PDcatheter and from six patients who had been on PD for more than12 mo at the time of catheter re-insertion due to blockade offluid flow by omentum or malpositioning. Catheter re-insertionbiopsies subsequent to previous removal necessitated by severeperitonitis were excluded from this study. Paraffin sections(10 µm) were incubated with antibody against perlecan(1:50) overnight at room temperature (18). Samples were thenwashed in PBS and incubated with horseradish peroxidase–conjugatedsecondary antibody for 1 h at 37°C. Signal detection wasby the peroxidase-anti-peroxidase (PAP) method (Dako, Gene Company,Hong Kong), with hematoxylin counterstaining and examinationunder a Leica DMLB microscope (Schmidt BioMed Tech (HK) Ltd,Hong Kong). Specimens were photographed with Kodak Max 100 filmusing a MP560 camera (Schmidt BioMed Tech (HK) Ltd, Hong Kong).For all biopsies in both sample groups, the staining intensityat six separate points along the length of the mesothelium wasmeasured in a semiquantitative manner using ChemiGenius analysissoftware (Syngene, Gene Company, Hong Kong), and the mean stainingintensity (expressed as arbitrary densitometry unit (DU) calculatedfor each specimen.
Culture of HPMC
HPMC were obtained by enzymatic digestion of omental specimensfrom non-uremic patients who underwent abdominal surgery. HPMCwere cultured in Medium-199 supplemented with 10% FCS as describedpreviously (13,18). All experiments were performed on cellsof the second passage that had been growth-arrested for 72 h.In experiments to investigate the effect of physiologic andelevated concentrations of glucose on cell morphology and perlecanexpression, HPMC were preconditioned in 5 mM, 30 mM, 75 mM,or 120 mM D-glucose for 2 wk; 30 mM represented the intra-peritonealglucose concentration after 1 h of equilibration after the instillationof conventional peritoneal dialysis solution, and 75 mM and120 mM represented the original glucose concentrations in hypertonicdialysates. Mannitol and L-glucose, at concentrations identicalto those of D-glucose, were used as the hexose control. To investigatethe role of high glucose-induced TGF-1 on perlecan synthesis,parallel experiments were performed with the prior additionof TGF-1 neutralizing antibody (1 µg/ml) or control IgG(1 µg/ml) to HPMC for 2 h. Cell morphology was monitoredthroughout the in vitro culture. HPMC were cultured in 75-cm2flasks to measure TGF-1 secretion, de novo synthesis of [35S]-labeledperlecan, or perlecan core protein synthesis by Western blotting,or on glass coverslips for immunohistochemical staining.
TGF-1 Secretion by HPMC
Aliquots of supernatant (1 ml) from HPMC (106 cells) culturedunder control or experiment conditions for time periods up to48 h were lyophilized and reconstituted in 100 µl of H20,and TGF-1 was measured with the DuoSet ELISA, which was specificto TGF-1 and had a detection range of 15 to 2000 pg/ml.
Assessment of Cell Proliferation
HPMC were cultured in triplicate in 96-well plates under controland experiment conditions until 80% confluent, and growth-arrestedwas for 72 h. Cell proliferation was monitored for time periodsup to 48 h, by the addition of MTT (0.5 mg/ml, final concentration)during the last 4 h (21). The resultant formazan was dissolvedovernight in 10% (wt/vol) SDS containing 0.01 M HCl, and theoptical density was measured at 595 nm, with a reference wavelengthof 690 nm.
Cell Protein Quantitation
HPMC cultured in triplicate in 96-well plates under controlor experiment conditions were lysed in 4 M urea buffer containing20 mM sodium acetate (pH 6.0) and 1% (vol/vol) Triton X-100(50 µl), and the protein concentration was determinedusing a modified Lowry assay (BioRad, Hong Kong).
Isolation and Western Blot Analysis of Perlecan Core Protein
HPMC were cultured under control or experiment conditions in10% FCS that had previously been cleared of PG and GAG by passageover a DEAE-Sephacel column (22). Cell surface and intracellular[35S]-labeled macromolecules (lysate fraction) were extractedwith 0.2% (vol/vol) Triton X-100 in 25 mM Tris-HCl, pH 7.5,while [35S]-labeled cytoskeletal/matrix molecules (matrix fraction)were extracted with 4 M urea, 25 mM Tris-HCl, pH 7.5, containing1% Triton X-100 (22,23). Proteinase inhibitors were includedin all samples (13). Ten micrograms of total protein contentof each sample was precipitated with ethanolic potassium acetateand then digested with buffer alone, protease-free chondroitinABC lyase, or a combination of heparinase I, II, and III (22).Digested samples were denatured in sample buffer at 95°Cfor 5 min, separated on a 3 to 12% SDS-PAGE acrylamide gel,and transferred onto nitrocellulose membrane using a mini-geltransfer system (Bio-Rad, Hong Kong) at 100 V for 1 h at 4°C.Equal loading of proteins was confirmed by staining of the membraneswith 1% (wt/vol) Ponceau S solution. Membranes were immunoblottedwith perlecan antibody (diluted 1:1000), and the respectivehorseradish peroxidase–conjugated secondary antibody (diluted1:10,000). The bands were visualized by enhanced chemiluminescence(Amersham Biosciences China Ltd, Hong Kong), semiquantitatedby densitometry using ChemiGenius analysis software, and expressedas arbitrary densitometric unit unless otherwise stated.
Immunohistochemical Staining
HPMC exposed to control or experiment conditions were fixedwith cold acetone/methanol (1:1) for 5 min before washing withPBS. Cells were blocked with 3% BSA in PBS for 1 h at room temperature,followed by washing with PBS for 30 min. Cells were incubatedwith perlecan antibody (diluted 1:100) for 1 h at 37°C ina humidified chamber (24). HPMC were washed in PBS and thenincubated with FITC-conjugated secondary antibody for 1 h at37°C in a darkened humidified chamber. After washing withPBS, HPMC were mounted with fluorescence mountant (DAKO, GeneCompany, Hong Kong), and epifluorescence was viewed using anAxiovert Zeiss 135 inverted microscope (Zeiss, Gold PacificLtd, Hong Kong).
Metabolic Labeling and Differential Extraction of [35S]-Labeled Macromolecules
Confluent growth-arrested HPMC were labeled with 50 µCi/mlNa2[35S]-sulfate for 24 h in low-sulfate medium as describedpreviously (13), and the conditioned medium was decanted. Cellsurface and intracellular [35S]-labeled macromolecules (lysatefraction) and cytoskeletal/matrix molecules (matrix fraction)were extracted as described previously (22,23). Proteinase inhibitorswere included in all samples. [35S]-labeled PG in the lysateand matrix fractions were buffer exchanged into 8 M urea, 25mM Bis-Tris, pH 6.0, containing 0.1% (wt/vol) CHAPS and appliedto DEAE-Sephacel columns (1.0 x 2.5 cm) equilibrated in thesame buffer (13). Unbound material was washed with 8 M ureabuffer containing 0.3 M NaCl (25 ml) to remove [35S]-labeledglycoproteins and contaminating proteins, and [35S]-labeledPG were eluted with 4 M guanidine-HCl buffer. [35S]-labeledPG were dialyzed into 8 M urea buffer containing 0.15 mM NaCland further purified on an Uno Q column, equilibrated in thesame buffer, and interfaced with a Biologic Workstation (BioRad,Hong Kong). Samples were washed with 8 M urea buffer containing(a) 0.1 M NaCl (10 ml) and then (b) 0.4 M NaCl (10 ml) beforeelution of [35S]-labeled PG with urea buffer containing 1.5M NaCl (15 ml). Fractions (0.5 ml) were collected and monitoredfor radioactivity by scintillation beta counting. Total eluted[35S]-labeled PG were concentrated by ethanolic potassium acetateprecipitation in the presence of carrier chondroitin sulfate(50 µg/ml) and heparin (50 µg/ml), washed with 95%(vol/vol) ethanol, dried under nitrogen, and stored at –20°Cuntil hydrodynamic size analysis by Sepharose CL-4B or 6B gelfiltration chromatography under dissociative conditions.
Gel Filtration Chromatography
To determine the hydrodynamic size of [35S]-labeled PG, aliquotsof the lysate and matrix fractions from HPMC cultured undercontrol or experiment conditions were treated with either bufferalone, chondroitin ABC lyase, or heparinase I, II, and III (13)and chromatographed on Sepharose CL-4B columns (0.05 x 1.2m)equilibrated in 4 M guanidine-HCl buffer (13). To confirm thePG nature of the [35S]-labeled molecules, samples were digestedwith papain (15 U, 100 µl) at 60°C overnight, andthe enzyme inactivated by the addition of 4 M guanidine buffer(400 µl) before gel filtration chromatography. Samples(0.5 ml) were collected at a rate of 2.4 ml/h, and radioactivitywas monitored by scintillation beta counting. The void (Vo)and total (Vt) volume of each column was calibrated using dextranblue and unincorporated [35S]-sulfate, respectively. The hydrodynamicsize of individual PG/GAG species was calculated using the equation[(Ve – Vo)/(Vt – Vo)], where Ve represented thefraction number of the eluted sample/peak.
Analysis of Perlecan HS GAG Chain Modification
The length and anionic charge of perlecan HS GAG chains synthesizedunder control and experiment conditions were assessed by gelfiltration chromatography and anionic exchange chromatography,respectively.
Altered hydrodynamic size of HS GAG chains was detected by papaindigestion to release the HS chains, followed by Sepharose CL-6Bchromatography. Vo and Vt were calibrated as with SepharoseCL-4B.
Changes in electronegativity were detected by applying the samplesto a Uno Q column interfaced with a BioRad Workstation, followedby elution with a linear NaCl gradient of 0.15 to 1.5 M in 8M urea buffer. Fractions (0.5 ml) were collected and monitoredfor radioactivity.
Statistical Analyses
Results are expressed as mean ± SD. Unless otherwisestated, all experiments were performed three times using HPMCfrom four separate donors. Statistical analyses were performedusing Prism version 3 GraphPad software (San Diego, CA). Between-groupdata were compared by the t test. P < 0.05 was consideredstatistically significant.
Perlecan Expression in Peritoneal Biopsies
Perlecan expression was prominent in the mesothelium of peritonealbiopsies obtained at commencement of PD (Figure 1, A and C;90.70 ± 16.05 DU; mean submesothelial thickness, 73.20± 22.39 µm), but it was notably reduced in biopsiesfrom patients on long-term PD (median PD duration, 25.33 ±8.28 mo) (Figure 1, B and C; mean perlecan staining, 25.16 ±11.83 DU; mean submesothelial thickness, 368.33 ± 42.14µm; P < 0.001). Long-term PD patients also showed increasedperlecan staining in the submesothelial fibrous tissue, comparedwith the former group (Figure 1B).
Figure 1. Representative perlecan expression in peritoneal biopsy specimens (A) from patients before commencement of peritoneal dialysis (PD) or (B) from patients on long-term PD. Paraffin sections were subjected to cytochemical staining using the peroxidase-anti-peroxidase (PAP) method. Immunoreactive products were visualized using diaminobenzidine as chromagen and counterstained with hematoxylin. Constitutive expression of perlecan was observed in peritoneal specimens obtained from new PD patients (A), identified predominantly within the peritoneal mesothelium (arrow). Peritoneal specimens from patients on long-term PD (B) demonstrated significant reduction in perlecan expression within the mesothelium, but increased expression within the submesothelium (*). Scale bar, 200 µm. The mean intensity staining of perlecan in the mesothelium of each biopsy specimen obtained from patients at the time of initial catheter insertion () or catheter re-insertion (•) was analyzed in a semiquantitative manner using ChemiGenius analysis software and its association with submesothelial thickness determined (C). Data represents the mean intensity staining of six separate measurements for each specimen.
In in vitro experiments, HPMC were pre-conditioned in eitherphysiologic (5 mM) or elevated (30 mM, 75 mM, or 120 mM) D-glucoseconcentration for 2 wk. Prior studies had demonstrated thatshorter durations of glucose exposure did not alter perlecansynthesis.
Effect of High Glucose Concentrations on HPMC Morphology, Proliferation, and Total Protein Content
HPMC were preconditioned in 5 mM, 30 mM, 75 mM, or 120 mM D-glucosefor 2 wk. Under these conditions, the cells became confluentafter 7 to 9 d of culture. HPMC cultured in 5 mM D-glucose exhibiteda homogenous, epithelial-like, cobblestone appearance (Figure 2A).In contrast, HPMC became progressively hypertrophic, multi-nucleated,and multi-vacuolated under increasing glucose concentrations(Figure 2, B–D), while proliferation was reduced (Figure 3).Total protein synthesis by HPMC remained unaltered underthe different glucose concentrations (Table 1). Mannitol atconcentrations identical to D-glucose did not affect HPMC morphology(Figure 2, E–H), proliferation (Figure 3B), or proteinsynthesis (Table 1). Similarly, no change was observed in controlcells exposed to L-glucose (data not shown).
Figure 2. Morphology of human peritoneal mesothelial cells (HPMC) cultured under physiologic and elevated glucose concentrations. HPMC were preconditioned in 5 mM (A and E), 30 mM (B and F), 75 mM (C and G), or 120 mM (D and H) D-glucose (A–D) or mannitol (E–H) for 2 wk. Cells cultured under 5 mM D-glucose adopted a homogenous population of cobblestone epithelial-like cells (A). With increasing glucose concentrations, cells became enlarged, multi-nucleated, and multi-vacuolated (*). Moreover, areas of the mesothelial monolayer were denuded in the presence of elevated glucose (depicted by arrow). No significant difference was observed in HPMC cultured with different concentrations of mannitol (E–H) compared with 5 mM D-glucose. Representative images of four separate experiments. Original magnification, x100.
Figure 3. The effect of D-glucose and mannitol on HPMC proliferation. Cell proliferation was analyzed in 80% confluent HPMC pre-conditioned with 5 mM (), 30 mM (), 75 mM (), or 120 mM () D-glucose (A) or mannitol (B) for selective time periods up to 48 h. While cells proliferated in a time-dependent manner with 5 mM D-glucose, HPMC proliferation was reduced progressively with increasing glucose concentrations (*P < 0.001, 5 mM D-glucose compared with 30 mM, 75 mM, or 120 mM at the same time point). No significant difference was observed in cells cultured with different concentrations of mannitol compared with 5 mM D-glucose. Data represent mean ± SD of four separate experiments.
Table 1. Protein synthesis in HPMC after exposure to D-glucose or mannitol at different concentrationsa
Effect of High Glucose Concentrations on TGF-1 Secretion
Under physiologic or high glucose concentrations, over 95% ofTGF-1 secreted by HPMC was in the latent form and necessitatedacid-activation for detection. Glucose increased total TGF-1secretion in a time- and dose-dependent manner (Figure 4A).Exposure to 30 mM, 75 mM, or 120 mM D-glucose resulted in asignificant increase in TGF-1 secretion at 24h (2.01 ±0.23, 3.86 ± 0.13, 3.94 ± 0.45, and 4.14 ±0.59 pg/µg cellular protein for 5 mM, 30 mM, 75 mM, and120 mM D-glucose, respectively; P < 0.05 comparing 5 mM andhigher glucose levels) and 48 h (2.85 ± 0.17, 4.18 ±0.27, 4.57 ± 0.64, and 5.28 ± 0.67 pg/µgcellular protein for 5 mM, 30 mM, 75 mM, and 120 mM D-glucose,respectively; P < 0.05 comparing 5 mM and higher glucoselevels). Exposure of HPMC to mannitol did not alter TGF-1 secretionsignificantly, compared with 5 mM D-glucose (Figure 4B).
Figure 4. TGF-1 secretion by HPMC under physiologic or experiment conditions. HPMC were cultured in 5 mM (), 30 mM (), 75 mM (), or 120 mM () D-glucose (A) or mannitol (B) for time periods up to 48 h. TGF-1 secretion increased with increasing glucose concentrations in a time dependent manner (A). Mannitol at identical concentrations did not modulate TGF-1 secretion compared with physiologic glucose concentration (B). Data represent mean ± SD of four separate experiments. *P < 0.001 and **P < 0.005, 5 mM D-glucose versus 30 mM, 75 mM, or 120 mM D-glucose concentration.
Perlecan Core Protein Synthesis
Expression of perlecan core protein was investigated in thelysate and matrix fractions of HPMC exposed to control or experimentconditions. Western blot analysis showed no perlecan polypeptidein the lysate fractions (data not shown), but only in the matrixfractions (Figure 5). Perlecan core protein was not synthesizedas free core protein, but solely attached to HS GAG chains (Figure 5,compare lanes 1–4 with lanes 9–12). The expressionof perlecan core protein was progressively reduced with increasingglucose concentrations (Figure 5, lanes 9–12). No changewas observed with mannitol (Figure 5, lanes 13–16).
Figure 5. Characterization of perlecan core protein using Western blot analysis. Non-labeled matrix proteoglycans (PG) were purified from HPMC cultured under control or experiment conditions. D-glucose (lanes 1–12) stimulated samples were electrophoresed on 3 to 12% SDS-PAGE acrylamide gels, before (lanes 1–4) or after treatment with chondroitin ABC lyase (Ch’ ABC lyase, lanes 5–8) or heparinase I, II, and III (Hep’ase, lanes 9–12), and immunoblotted with a specific monoclonal antibody to perlecan (A). Mannitol samples treated with heparinase are also shown (panel A, lanes 13–16). The intensity of bands representing heparinase-released perlecan core proteins synthesized by HPMC under increasing concentrations of D-glucose () or mannitol () was assessed by densitometric scan and expressed as arbitrary units (panel B). Representative blots of three separate experiments. *P < 0.01, 5 mM D-glucose versus 30 mM D-glucose; **P < 0.001, 5 mM D-glucose versus 75 mM or 120 mM D-glucose, P < 0.01 and P < 0.001, D-glucose versus mannitol at the same hexose concentration.
Localization of Perlecan Expression
Immunohistochemical staining experiments were performed to localizeperlecan expression in HPMC cultured under control or experimentconditions for periods up to 2 wk (Figure 6). Perlecan exhibiteda perinuclear distribution in HPMC exposed to 5 mM D-glucosefor 24 h. In contrast, its deposition within the extracellularmilieu was observed after 3 d of culture (Figure 6B). After2 wk, an extensive perlecan matrix was noted (Figure 6C). Underhigh glucose concentrations (30 mM), perlecan expression wassignificantly reduced in a time-dependent manner (Figure 6, D–F).Mannitol at corresponding concentrations did notalter perlecan expression (Figure 6 G–I).
Figure 6. Immunohistochemical analysis of HPMC for perlecan expression under physiologic or experiment conditions. HPMC were cultured in 5 mM D-glucose (A–C), 30 mM D-glucose (D–F), or 30 mM mannitol (G–I) onto glass coverslips for 1 (panels A, D, and G), 3 (B, E, and H), or 14 (C, F, and I) days, before fixation with acetone/methanol. Cells were subsequently stained for perlecan. With increasing time in culture, HPMC cultured in 5 mM D-glucose deposited an intense matrix containing perlecan (A-C). In comparison, elevated D-glucose reduced mesothelial perlecan expression in a time-dependent manner (D–F). Mannitol did not affect perlecan synthesis compared with 5 mM D-glucose. Representative images from three individual experiments. Original magnification, x400.
Other investigators have reported the presence of agrin, anotherHSPG, in basement membranes (25). Using an antibody directedagainst the core protein of agrin (a kind gift of Dr. Jacobvan der Born, Department of Molecular Cell Biology, VU UniversityMedical Centre, Amsterdam, the Netherlands), we did not detectagrin in the lysate or matrix fractions of HPMC (data not shown).
[35S]-Labeling of PG Synthesized de novo by HPMC
[35S]-sulfate labeling experiments were performed for qualitativeand quantitative characterization of perlecan. Confluent HPMCmonolayers preconditioned under 5 mM, 30 mM, 75 mM, or 120 mMD-glucose were labeled with [35S]-sulfate for 24 h. [35S]-labeledPG from the lysate and matrix fractions were purified from glycoproteins,proteins, and non-incorporated [35S]-sulfate by dialysis andDEAE-Sephacel anionic exchange chromatography. Conditions forion-exchange chromatography were optimized to separate the highlycharged PG from the weakly charged glycoproteins.
Under 5 mM, 30 mM, 75 mM, or 120 mM D-glucose, 95.5 ±5.2%, 95.6 ± 4.8%, 94.2 ± 5.9%, and 95.1 ±4.9%, respectively, of [35S]-labeled macromolecules from thelysate and 92.1 ± 4.2%, 91.8 ± 4.8%, 94.2 ±3.5%, and 91.8 ± 7.1%, respectively, of [35S]-labeledmacromolecules from the matrix fraction bound to DEAE-Sephacel.Since total cell protein content was unaltered under controlor elevated D-glucose concentrations, [35S]-labeled macromoleculeswere expressed as cpm/µg total protein unless otherwisestated. HPMC incorporated 4.11 ± 0.76 x 105 cpm/µgtotal cell protein when cultured in 5 mM D-glucose, of which3.42 ± 0.15 x 105 and 0.68 ± 0.03 x 105 cpm/µgtotal protein was isolated from the lysate and matrix fractions,respectively. De novo synthesis of [35S]-labeled cell associatedmacromolecules decreased for 30 mM, 75 mM, and 120 mM D-glucose(lysate fraction: 3.08 ± 0.24 x 105, 2.77 ± 0.21x 105, and 2.76 ± 0.15 x 105 cpm/µg cell protein,respectively; matrix fraction: 0.48 ± 0.02 x 105, 0.31± 0.05 x 105, and 0.25 ± 0.03 x 105 cpm/µgcell protein, respectively). These macromolecules were characterizedas PG by a shift in their elution profiles on Sepharose CL-4Bunder dissociative conditions after papain digestion (Figure 7).Using selective enzymatic digestion, a large HSPG that wasresistant to chondroitin ABC lyase but sensitive to heparinaseI, II, and III treatment was detected and identified exclusivelyin the matrix fraction, thereby corroborating our Western blotdata. The intact perlecan molecule possessed a hydrodynamicsize (Kav) of 0.10 to 0.25 on Sepharose CL-4B. The length ofperlecan HS GAG chains was unaltered in the presence of elevatedglucose concentrations, as demonstrated by papain treatmentfollowed by gel filtration chromatography on Sepharose CL-6B(Table 2). The charge density of these macromolecules was alsounchanged (data not shown).
Figure 7. Representative elution profiles of PG/GAG synthesized by HPMC under physiologic glucose concentration. Aliquots of lysate (A) and matrix (B) fractions were chromatographed on Sepharose CL-4B under dissociative conditions before (•) and after papain digestion (). Sensitivity of these macromolecules to papain digestion confirmed their PG nature. Treatment of samples with chondroitin ABC lyase () revealed difference species of HSPG/GAG synthesized by HPMC. Quantitative but not qualitative changes in PG synthesis by HPMC were observed with increasing concentrations of D-glucose (Table 2).
Table 2. [35S]-labeled perlecan synthesis by HPMC under physiological or high glucose concentrationsa
The Role of Glucose-Induced TGF-1 on Perlecan Synthesis
Experiments were performed in which HPMC were exposed to physiologicor high glucose concentrations in the presence or absence ofTGF-1 neutralizing antibody or control IgG (1 µg/ml),followed by Western blot analysis and immunohistochemistry.Since 30 mM D-glucose represented the intraperitoneal glucoseconcentration after equilibration of dialysate, subsequent seriesof experiments concentrated on this glucose concentration.
TGF-1 neutralizing antibody or control IgG per se did not alterperlecan core protein expression in HPMC cultured under 5 mMD-glucose (Figure 8, compare lanes 7–9). While controlIgG did not alter perlecan expression in HPMC exposed to 30mM D-glucose, TGF-1 neutralizing antibody partially restoredperlecan core protein synthesis in these cells (Figure 8, lanes10–12). These results were corroborated by immunohistochemicalstudies, which showed that TGF-1 neutralizing antibody increasedextracellular perlecan expression under 30 mM D-glucose (Figure 9).
Figure 8. Western blot analysis of perlecan expression showing the effect of glucose and TGF-1 neutralization antibody. Non-labeled matrix PG were purified from HPMC cultured under 5 mM (lanes 1–3 and 7–9) or 30 mM (4–6 and 10–12) D-glucose in the presence or absence of TGF-1 neutralization antibody (1 µg/ml) or control IgG (1 µg/ml) as described in the Materials and Methods. Aliquots of samples were electrophoresed on a 3 to 12% SDS-PAGE acrylamide gels before (panel A, lanes 1–6) and after heparinase I, II, and III digestion (panel A, lanes 7–12). The intensity of perlecan core protein expression under control and experiment conditions was assessed by densitometric scan and expressed as arbitrary units (panel B). Representative blots of three individual experiments.
Figure 9. Immunohistochemical analysis of perlecan expression under control or experiment conditions. Ninety percent confluent growth-arrested HPMC were incubated with either TGF-1 neutralization antibody (B, E, and H) or control IgG (C, F, and I) for 2 h before incubation with 5 mM D-glucose (A–C), 30 mM D-glucose (D–F), or 30 mM mannitol (G–I) for 3 d and fixed with methanol/acetone. Synthesis and localization of perlecan was assessed using immunohistochemistry. While TGF-1 neutralization antibody or control IgG had no effect on HPMC cultured in 5 mM D-glucose (A–C) or 30 mM mannitol (G–I), addition of TGF-1 neutralization antibody to cells exposed to 30 mM D-glucose ameliorated the reduction of perlecan synthesis. Deposition of perlecan was noted in the ECM (E). Control IgG had no effect on perlecan expression. Representative images of three separate experiments. Original magnification x400.
The mesothelium is the most extensive serous membrane in thebody. It contains abundant anionic sites and intercellular junctionsthat enable it to act as a biologic barrier to the passage ofplasma molecules (26,27). The introduction of PD over 20 yrago has allowed us to manipulate the peritoneal membrane toact as a dialyzing organ in which ultrafiltration and diffusionoccurs. The constant exposure of the mesothelium to unphysiologicalPD solutions, however, results in the progressive deteriorationof the peritoneal membrane, with subsequent loss of peritonealfunction and selective regulation of molecules. Possible mechanismsof peritoneal dysfunction include the altered expression ofintercellular junctions (tight junctions, gap junctions, adherensjunctions, and desmosomes) within the mesothelium. This hasbeen highlighted in a recent study by Ito et al. (27), who demonstratedan inverse association between elevated glucose induction ofTGF-1, and synthesis of proteins that constitute intercellularjunctions, namely ZO-1, E-cadherin, and -catenin. Increasedtransperitoneal passage of proteins as a result of reduced anionicsites within the mesothelium and the underlying basal laminaby chemical or bacterial injury has also been documented (7,8).The constituents contributing to these anionic sites, or theirchange during PD, have not been defined. Perlecan has a strongnet negative charge and is a ubiquitous macromolecule in basementmembranes. We therefore investigated its expression in the normalperitoneal membrane and that under the influence of peritonealdialysate.
Our original observations showed intense perlecan expressionin peritoneal biopsies from patients at the commencement ofPD, before exposure to peritoneal dialysates. In contrast, mesothelialperlecan expression was significantly reduced in chronic PDpatients and was associated with thickening of the submesothelium.That perlecan expression within the mesothelium is decreasedduring clinical PD is in accordance with the observation fromShostak et al. (28), who demonstrated reduced charge densityin rat mesothelium during experimental PD, accompanied by increasedpermeability to plasma protein. Perlecan is a large PG thatpossesses either HS or CS GAG chains, and it has a wide tissuedistribution. Apart from conferring selective permeability throughits negative charge, there are recent data to substantiate animportant role of perlecan in maintaining the mechanical andfunctional integrity of basement membranes (29). Perlecan expressionis often changed in diseases that affect the basement membrane.In diabetic mellitus, the altered perlecan expression in thekidney is related to hyperglycemia (9,30). It is therefore pertinentto examine the effect of high dialysate glucose levels on peritonealperlecan expression.
HPMC cultured in vitro provide a means to study the effectsof peritoneal dialysate on cell proliferation and synthesisof macromolecules. Previous studies have demonstrated that invivo morphologic and immunohistochemical characteristics werepreserved in cultured HPMC (31). Our results showed constitutivesynthesis of perlecan by HPMC under physiologic glucose concentration.The perlecan thus synthesized had a hydrodynamic size of 0.15to 0.25 (on Sepharose CL-4B), comprising a core protein of Mrapproximately 173 kD and exclusively HS GAG chains of approximately30 kD. Differential cellular extraction, Western blot analysis,and immunohistochemistry demonstrated perlecan deposition inthe extracellular matrix at confluence of HPMC. Exposure ofHPMC to increasing glucose concentrations progressively reducedde novo perlecan synthesis, while there was no qualitative changein the core protein or GAG chains. This quantitative reductionin perlecan synthesis was selective, since total cell proteinsynthesis remained unaltered. In addition, these changes werespecific to high concentrations of D-glucose and not relatedto osmolality, as indicated by the results with mannitol. Inview of the unaltered anionic charge density and the lengthof GAG chains, a reduction in GAG synthesis was the likely mechanismleading to the decreased sulfate incorporation under the influenceof high glucose concentrations.
Other investigators have demonstrated altered perlecan synthesisin glomerular cells and perlecan deposition in glomerular basementmembrane (GBM) upon exposure to supra-physiologic glucose concentrations(32,33). An overall reduction in HSPG with preservation of thecharge density and the length of GAG chains has been noted withinthe diabetic GBM or glomerular epithelial cells, similar toour findings in HPMC (33–36). However, van Det et al.(37) have reported reduced sulfation of the HS GAG chains butunaltered HSPG core protein synthesis in mesangial and epithelialcells upon exposure to elevated glucose concentrations. Whilethe biologic implications of reduced GAG chain synthesis orunder-sulfation of GAG chains remain to be fully elucidated,these changes could have important implications on cell proliferation,differentiation, or activation, and angiogenesis, since theGAG moiety of perlecan engages in the coupling of growth factorsand affects their stability, conformation, proteolysis, andsequestration (38). These indirect wide-ranging effects of perlecanare further underscored by the recent data that showed thatits core protein exhibited similar growth factor sequestrationactions (39).
Fibroblasts and endothelial cells synthesize perlecan core proteinwith apparent Mr of 400 to 470 kD (40,41), while the core proteinsynthesized in mesangial cells is of Mr 250 to 300 kD (22).We observed that the perlecan core protein synthesized in HPMCwas significantly smaller (Mr, 173 kD). The relatively lowerMr was not attributed to protease digestion during its isolationbecause proteinase inhibitors were incorporated in all procedures,nor was it the result of internal degradation since de novosynthesis of [35S]-methionine labeled perlecan core proteingave identical results (Yung and Davies, unpublished data).The core protein of perlecan comprises five distinct domains:HS attachment domain, LDL-like domain, two different laminindomains, and N-CAM-like domain that contains potential alternativesplicing sites, thereby generating variations of the perlecanmolecule (42). Size variants of perlecan have been observedin the Engelbreth-Holm-Swarm tumor matrix (43,44), and alternativesplicing has been demonstrated in the nematode homologue ofmammalian perlecan UNC-52 and mouse skeletal muscle cells (45).It is thus possible that alternative splicing or posttranscriptionalmodification in HPMC may result in a truncated form of perlecanwith smaller molecular weight. In this context, we have previouslydemonstrated that HPMC constitutively synthesized stromelysin(46), an enzyme that degraded the core protein of proteoglycans,thereby releasing GAG-peptide fragments.
The role of TGF-1 in mediating the glucose-induced changes inperlecan synthesis is exemplified by the results obtained withTGF-1 neutralizing antibody and the induction of TGF-1 secretionby HPMC in the presence of high glucose concentrations. Thelatter has also been shown in other cell types (47,48). Iozzoet al. (49) have demonstrated TGF-1–responsive sequencesbetween –461 and –285 bp in the promotor regionof human perlecan, which could explain the effect of exogenousTGF-1 on perlecan expression in human colon carcinoma cells,uterine epithelial cells, aortic endothelial cells, and cornealfibroblasts (47,48). We and others (50–52) have demonstratedthat TGF-1 was secreted predominantly in the inactive or latentform in HPMC, and activation could be achieved by an increasein temperature, extreme pH, altered glycosylation, integrinbinding, plasmin, reactive oxygen species, or thrombospondin-1.The exact mechanism involved in TGF-1 activation in HPMC isundefined. Thromobspondin-1 plays a pivotal role in high-glucose–inducedTGF-1 activation in mesangial cells (52). Vischer et al. (53)have shown that perlecan binds thrombospondin-1 through itsHS GAG chains, thus sequestering this glycoprotein at the apicalsurface of endothelial cells, before its internalization throughreceptor-mediated endocytosis. It is therefore appropriate toinvestigate whether perlecan may modulate the activation ofTGF-1 by presenting a high local concentration of thrombospondin-1,thereby implicating a two-way relationship between perlecanand TGF-1.
In summary, we have presented novel evidence that the normalperitoneal membrane expresses perlecan, which is synthesizedby HPMC, and that perlecan synthesis decreases after exposureto high glucose concentrations, leading to the reduced perlecanexpression in patients on PD. The interrelationship betweenperlecan, TGF-1, and high dialysate glucose concentrations,together with the indirect effects of perlecan on cellular migrationand differentiation, represent pathogenetic mechanisms leadingto the progressive structural and functional alterations ofthe peritoneal membrane during long-term PD.
Acknowledgments
This work was supported by the Hong Kong Research Grants CouncilEarmarked Research Grant [7240/98M] and the Wai Hung CharityFoundation. The authors thank Dr. Kent-Man Chu, Dr. Kin-WahChu, Dr. Judy Ho, Dr. Sai Man Chu, Dr. Po Chor Tam, and theirsurgical teams for the collection of omentum specimens. We aregrateful to Mr. Jack Leung for his assistance in the isolationand culture of peritoneal mesothelial cells. Part of this workwas presented in abstract form at the 33rd American Societyof Nephrology and IX Congress of the International Society forPeritoneal Dialysis (J Am Soc Nephrol 11: 540A, 2000 and PeritDial Int 21: S129, 2001).
Footnotes
Drs. Yung and Chen contributed equally to this work.
Topley N, Brown Z, Jorres A, Westwick J, Davies M, Coles GA, Williams JD: Human peritoneal mesothelial cells synthesize interleukin-8: synergistic induction by interleukin-1 and tumor necrosis factor-. Am J Pathol 142: 1876–1886, 1993[Abstract]
Offner FA, Feichtinger H, Stadlmann S, Obrist P, Marth C, Klingler P, Grage B, Schmahl M, Knabbe C: Transforming growth factor-beta synthesis by human peritoneal mesothelial cells. Induction by interleukin-1. Am J Pathol 148: 1679–1688, 1996[Abstract]
Robson RL, McLoughlin RM, Witowski J, Wilkinson TS, Jones SA, Topley N: Differential regulation of chemokine production in human peritoneal mesothelial cells: IFN-gamma controls neutrophil migration across the mesothelium in vitro and in vivo. J Immunol 167: 1028–1038, 2001[Abstract/Free Full Text]
Coene M-C, van Hove C, Claeys M, Herman AG: Arachidonic acid metabolism by cultured mesothelial cells. Biochem Biophys Acta 710: 437–445, 1982[Medline]
Gotloib L, Bar Sella P, Jaichenko J, Shostack A: Ruthenium-red-stained polyanionic fixed charges in peritoneal microvessels. Nephron 47: 22–28, 1987[Medline]
Gotloib L, Shostack A, Jaichenko J: Ruthenium-red-stained anionic charges of rat and mice mesothelial cells and basal lamina: The peritoneum is a negatively charged dialyzing membrane. Nephron 48: 65–70, 1988[Medline]
Gotloib L, Shustak A, Jaichenko J: Loss of mesothelial electronegative fixed charges during murine septic peritonitis. Nephron 51: 77–83, 1989[Medline]
Conde-Knape K: Heparan sulfate proteoglycans in experimental models of diabetes: A role for perlecan in diabetes complications. Diabetes Metab Res Rev 17: 412–421, 2001[CrossRef][Medline]
Park PW, Reizes O, Bernfield M: Cell surface heparan sulfate proteoglycans: selective regulators of ligand receptor encounters. J Biol Chem 275: 29923–29926, 2000[Free Full Text]
Staprans I, Piel F, Felts JM: Analysis of selected plasma constituents in continuous ambulatory peritoneal dialysis effluent. Am J Kidney Dis 7: 490–494, 1986[Medline]
Davies M, Stylianou E, Yung S, Thomas GJ, Coles GA, Williams JD: Proteoglycans of CAPD-dialysate fluid and mesothelium. In: CAPD: Host Defence, Nutrition and Ultrafiltration, edited by Coles GA, Davies M, Williams JD, Basel, Karger, 1990, pp 134–141
Yung S, Thomas GJ, Stylianou E, Williams JD, Coles GA, Davies M: Source of peritoneal proteoglycans. Human peritoneal mesothelial cells synthesize and secrete mainly small dermatan sulfate proteoglycans. Am J Pathol 146: 520–529, 1995[Abstract]
Miao HQ, Ishai-Michaeli R, Atzmon R, Peretz T, Vlodavsky I: Sulfate moieties in the subendothelial extracellular matrix are involved in basic fibroblast growth factor sequestration, dimerization, and stimulation of cell proliferation. J Biol Chem 271: 4879–4886, 1996[Abstract/Free Full Text]
Fuki IV, Iozzo RV, Williams KJ: Perlecan heparan sulfate proteoglycan: A novel receptor that mediates a distinct pathway for ligand catabolism. J Biol Chem 275: 25742–25750, 2000[Abstract/Free Full Text]
Honda K, Nitta K, Horita S, Yumura W, Nihei H: Morphological changes in the peritoneal vasculature of patients on CAPD with ultrafiltration failure. Nephron 72: 171–176, 1996[Medline]
Williams JD, Craig KJ, Topley N, von Ruhland C, Fallon M, Newman GR, Mackenzie RK, Williams GT: Morphologic changes in the peritoneal membrane of patients with renal disease. J Am Soc Nephrol 13: 470–479, 2002[Abstract/Free Full Text]
Chan TM, Leung JKH, Tsang RCW, Liu ZH, Li LS, Yung S: Emodin meliorates glucose-induced matrix synthesis in human peritoneal mesothelial cells. Kidney Int 64: 519–533, 2003[CrossRef][Medline]
Chen S, Cohen MP, Lautenslager GT, Shearman CW, Ziyaeh FN: Glycated albumin stimulates TGF-beta 1 production and protein kinase C activity in glomerular endothelial cells. Kidney Int 59: 673–681, 2001[CrossRef][Medline]
Coimbra TM, Janssen U, Grone HJ, Ostendorf T, Kenter U, Schmidt H, Brabant G, Floge J: Early events leading to renal injury in obese Zucker (fatty) rats with type II diabetes. Kidney Int 57: 167–182, 2000[CrossRef][Medline]
Yung S, Liu ZH, Lai KN, Li LS, Chan TM: Emodin ameliorates glucose-induced morphologic abnormalities and synthesis of transforming growth factor beta1 and fibronectin by human peritoneal mesothelial cells. Perit Dial Int 21: S41–S47, 2001[Abstract/Free Full Text]
Yung S, Woods A, Chan TM, Davies M, Williams JD, Couchman JR: Syndecan-4 up-regulation in proliferative renal disease is related to microfilament organization. FASEB J published online 10.1096/fj. 00–0794fje, 2001
Woods A, Couchman JR, Höök M: Heparan sulfate proteoglycans of the rat embryo fibroblasts. A hydrophobic form may link cytoskeleton and matrix components. J Biol Chem 260: 10872–10879, 1985[Abstract/Free Full Text]
Yung S, Davies M: Response of the human peritoneal mesothelial cell to injury: An in vitro model of peritoneal wound healing. Kidney Int 54: 2160–2169, 1998[CrossRef][Medline]
Raats CJ, van den Born J, Berden JH: Glomerular heparan sulfate alterations: Mechanisms and relevance for proteinuria. Kidney Int 57: 385–400, 2000[Medline]
Wasteson A: A method for the determination of the molecular weight and molecular-weight distribution of chondroitin sulphate. J Chromat 59: 87–97, 1971[CrossRef][Medline]
Ito T, Yorioka N, Yamamoto M, Kataoka K, Yamakido M: Effect of glucose on intercellular junctions of cultured human peritoneal mesothelial cells. J Am Soc Nephrol 11: 1969–1979, 2000[Abstract/Free Full Text]
Shostak A, Wajsbrot V, Gotloib L: Protective effect of aminoquanidine upon capillary and submesothelial anionic sites. Microvasc Res 61: 166–178, 2001[CrossRef][Medline]
Costell M, Gustafsson E, Aszodi A, Morgelin M, Bloch W, Hunziker E, Addicks K, Timpl R, Fässler R: Perlecan maintains the integrity of cartilage and some basement membranes. J Cell Biol 147: 1109–1122, 1999[Abstract/Free Full Text]
Hansen PM, Chowdhury T, Deckert T, Hellgren A, Bain SC, Pociot F: Genetic variation of the heparan sulfate proteoglycan gene (perlecan gene). Association with urinary albumin excretion in IDDM patients. Diabetes 46: 1658–1659, 1997[Medline]
Hjelle JT, Golinska BT, Waters DC, Steidley KR, McCarroll DR, Dobbie J: Isolation and propagation in vitro of peritoneal mesothelial cells. Perit Dial Int 9: 341–347, 1989
Kasinth BS, Grellier P, Ghosh Chouduury G, Abboud SL: Regulation of basement membrane heparan sulfate proteoglycan, perlecan, gene expression in glomerular epithelial cells by high glucose medium. J Cell Physiol 167: 131–136, 1996[CrossRef][Medline]
Brown DM, Klein DJ, Michael AF, Oegema TR: 35S-glycosaminoglycan and 35S-glycopeptide metabolism by diabetic glomeruli and aorta. Diabetes 31: 418–425, 1982[Abstract]
Templeton DM, Fan MY: Posttranscriptional effects of glucose on proteoglycan expression in mesangial cells. Metabolism 45: 1136–1146, 1996[CrossRef][Medline]
Kanwar YS, Rosenzweig LJ, Linker A, Jakubowski ML: Decreased de novo synthesis of glomerular proteoglycans in diabetes: biochemical and atoradiographic evidence. Proc Natl Acad Sci USA 80: 2272–2277, 1983[Abstract/Free Full Text]
Kolm V, Sauer U, Olgemooller B, Schleicher ED: High glucose-induced TGF-beta 1 regulates mesangial production of heparan sulfate proteoglycan. Am J Physiol 270: F812–F821, 1996
van Det NF, van den Born J, Tamsma JT, Verhagen NA, Berden JH, Bruijn JA, Daha MR, van der Woude FJ: Effects of high glucose on the production of heparan sulfate proteoglycan by mesangial and epithelial cells. Kidney Int 49: 1079–1089, 1996[Medline]
Aviezer D, Hecht D, Safran M, Eisinger M, David G, Rayon A: Perelcan, basal lamina proteoglycan, promotes basic fibroblast growth factor receptor binding, mitogenesis and angiogenesis. Cell 79: 1005–1013, 1994[CrossRef][Medline]
Mongiat M, Otto J, Oldershaw R, Ferrer F, Sato JD, Iozzo RV: Fibroblast growth factor-binding protein is a novel partner for the perlecan protein core. J Biol Chem 276: 10263–10271, 2001[Abstract/Free Full Text]
Kaji T, Yamada A, Sawako M, Yamamoto C, Fujiwara Y, Wight TN, Kinsella MG: Cell density-dependent regulation of proteoglycan synthesis by transforming growth factor-1 in cultured bovine aortic endothelial cells. J Biol Chem 275: 1463–1470, 2000[Abstract/Free Full Text]
Brown CT, Nugent MA, Lai FW, Trinkaus-Randall V: Characterization of proteoglycans synthesized by cultured corneal fibroblasts in response to transforming growth factor and fetal calf serum. J Biol Chem 274: 7111–7119, 1999[Abstract/Free Full Text]
Noonan DM, Hassell JR: Perlecan, the large low-density proteoglycan of basement membranes: Structure and variant forms. Kidney Int 43: 53–60, 1993[Medline]
Hassell JR, Leyshon WC, Ledbetter SR, Tyree B, Suzuki S, Kato M, Kimata K, Kleinman HK: Isolation of two forms of basement membrane proteoglycans. J Biol Chem 260: 8098–8105, 1985[Abstract/Free Full Text]
Larrain J, Alvarez J, Hassell JR, Brandan E: Expression of perlecan, a proteoglycan that binds myogenic inhibitory basic fibroblast growth factor, is down regulated during skeletal muscle differentiation. Exp Cell Res 234: 405–412, 1997[CrossRef][Medline]
Mullen GP, Rogalski TM, Bush JA, Gorji PR, Moerman DG: Complex patterns of alternative splicing mediate the spatial and temporal distribution of perlecan/UNC-52 in Caenorhabditis elegans. Mol Biol Cell 10: 3205–3221, 1999[Abstract/Free Full Text]
Martin J, Yung S, Robson RL, Steadman R, Davies M: Production and regulation of matrix metalloproteinases and their inhibitors by human peritoneal mesothelial cells. Perit Dial Int 20: 524–533, 2000[Abstract/Free Full Text]
Dodge GR, Kovalszky I, Hassell JR, Iozzo RV: Transforming growth factor beta alters the expression of heparan sulfate proteoglycan in human colon carcinoma cells. J Biol Chem 265: 18023–18029, 1990[Abstract/Free Full Text]
Nugent MA, Edelman ER: Transforming growth factor beta 1 stimulates the production of basic fibroblast growth factor binding proteoglycans in Balb/c3T3 cells. J Biol Chem 267: 21256–21264, 1992[Abstract/Free Full Text]
Iozzo RV, Pillarisetti J, Sharma B, AD, Danielson KG, Mauviel A: Structural and functional characterization of the human perlecan gene promoter, transcriptional activation by transforming growth factor-beta via a nuclear factor-1 binding element. J Biol Chem 272: 5219–5228, 1997[Abstract/Free Full Text]
Lyons RM, Gentry LE, Purchio AF, Moses HL: Mechanism of activation of latent recombinant transforming growth factor beta 1 by plamsin. J Cell Biol 110: 1361–1367, 1990[Abstract/Free Full Text]
Munger JS, Huang X, Kawakatsu H, Griffiths MJ, Dalton SL, Wu J, Pittet JF, Kaminski N, Garat C, Matthay MA, Rifkin DB, Sheppard D: The integrin alpha v beta binds and activates latent TGF beta 1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell 96: 319–328, 1999[CrossRef][Medline]
Poczatek MH, Hugo C, Darley-Usmar V, Murphy-Ullrich JE: Glucose stimulation of transforming growth factor- bioactivity in mesangial cells is mediated by thrombospondin-1. Am J Pathol 157: 1353–1363, 2000[Abstract/Free Full Text]
Vischer P, Feitsma K, Schon P, Volker W: Perelcan is responsible for thrombospondin 1 binding on the cell surface of cultured porcine endothelial cells. Eur J Cell Biol 73: 332–343, 1997[Medline]
Received for publication November 25, 2003.
Accepted for publication January 27, 2004.
1 in Human Peritoneal Mesothelial Cells Due to High Dialysate Glucose Concentration: Implication in Peritoneal Dialysis
Top
Abstract
Introduction
1. Characterization studies showed that perlecan synthesized by HPMC contained solely heparan sulfate glycosaminoglycan (HS GAG) chains, and [35S]-incorporation studies demonstrated progressive reduction of their de novo synthesis with increasing glucose concentrations (68142 ± 3658, 48147 ± 2517, 31468 ± 5781, and 25575 ± 3621 cpm/µg cellular protein for 5 mM, 30 mM, 75 mM, and 120 mM D-glucose, respectively; P < 0.001 for 5 mM versus 30 mM D-glucose, and P < 0.0001 for 5 mM versus 75 mM or 120 mM D-glucose). Both the length and the charge density of the HS GAG chains remained unchanged. Reduction of peritoneal perlecan expression in long-term PD was attributed to high dialysate glucose concentrations, which induced TGF-
. Am J Pathol 142: 1876–1886, 1993[Abstract]
) or catheter re-insertion (•) was analyzed in a semiquantitative manner using ChemiGenius analysis software and its association with submesothelial thickness determined (C). Data represents the mean intensity staining of six separate measurements for each specimen.
), 30 mM (
), 75 mM (
), or 120 mM (
) D-glucose (A) or mannitol (B) for selective time periods up to 48 h. While cells proliferated in a time-dependent manner with 5 mM D-glucose, HPMC proliferation was reduced progressively with increasing glucose concentrations (*P < 0.001, 5 mM D-glucose compared with 30 mM, 75 mM, or 120 mM at the same time point). No significant difference was observed in cells cultured with different concentrations of mannitol compared with 5 mM D-glucose. Data represent mean ± SD of four separate experiments.
), 30 mM (
) D-glucose (A) or mannitol (B) for time periods up to 48 h. TGF-
) was assessed by densitometric scan and expressed as arbitrary units (panel B). Representative blots of three separate experiments. *P < 0.01, 5 mM D-glucose versus 30 mM D-glucose; **P < 0.001, 5 mM D-glucose versus 75 mM or 120 mM D-glucose, 
P < 0.01 and 
P < 0.001, D-glucose versus mannitol at the same hexose concentration.
). Sensitivity of these macromolecules to papain digestion confirmed their PG nature. Treatment of samples with chondroitin ABC lyase (
