Novel mechanisms of protein synthesis in diabetic nephropathy--role of mRNA translation.

Ambient protein levels are affected by both synthesis and degradation. Synthesis of a protein is regulated by transcription and messenger RNA (mRNA) translation. Translation has emerged as an important site of regulation of protein expression during development and disease. It is under the control of distinct factors that regulate initiation, elongation and termination phases. Regulation of translation occurs via signaling reactions, guanosine diphosphate-guanosine triphosphate binding and by participation of non-coding RNA species such as microRNA. Recent work has revealed an important role for translation in hypertrophy, matrix protein synthesis, elaboration of growth factors in in vivo and in vitro models of diabetic nephropathy. Studies of translation dysregulation in diabetic nephropathy have enabled identification of novel therapeutic targets. Translation of mRNA is a fertile field for exploration in investigation of kidney disease.

Heterogeneous nuclear ribonucleoprotein K contributes to angiotensin II stimulation of vascular endothelial growth factor mRNA translation.

ANG II rapidly increases VEGF synthesis in proximal tubular epithelial cells through mRNA translation. The role of heterogeneous nuclear ribonucleoprotein K (hnRNP K) in ANG II regulation of VEGF mRNA translation initiation was examined. ANG II activated hnRNP K as judged by binding to poly(C)- and poly(U)-agarose. ANG II increased hnRNP K binding to VEGF mRNA at the same time as it stimulated its translation, suggesting that hnRNP K contributes to VEGF mRNA translation. Inhibition of hnRNP K expression by RNA interference significantly reduced ANG II stimulation of VEGF synthesis. ANG II increased hnRNP K phosphorylation on both tyrosine and serine residues with distinct time courses; only Ser302 phosphorylation paralleled binding to VEGF mRNA. Src inhibition using PP2 or RNA interference inhibited PKCdelta activity and prevented hnRNP K phosphorylation on both tyrosine and serine residues and its binding to VEGF mRNA. Under these conditions, ANG II-induced VEGF synthesis was inhibited. ANG II treatment induced redistribution of both VEGF mRNA and hnRNP K protein from light to heavy polysomal fractions, suggesting increased binding of hnRNP K to VEGF mRNA that is targeted for increased translation. This study shows that hnRNP K augments efficiency of VEGF mRNA translation stimulated by ANG II.

Angiotensin II inhibits insulin-stimulated phosphorylation of eukaryotic initiation factor 4E-binding protein-1 in proximal tubular epithelial cells.

Interaction between angiotensin II, which binds a G-protein-coupled receptor, and insulin, a ligand for receptor tyrosine kinase, was examined in renal proximal tubular epithelial cells. Augmented protein translation by insulin involves activation of eukaryotic initiation factor 4E (eIF4E) which follows the release of the factor from a heterodimeric complex by phosphorylation of its binding protein, 4E-BP1. Angiotensin II (1 nM) or insulin (1 nM) individually stimulated 4E-BP1 phosphorylation. However, pre-incubation with angiotensin II abrogated insulin-induced phosphorylation of 4E-BP1, resulting in persistent binding to eIF4E. Although angiotensin II and insulin individually activated phosphoinositide 3-kinase and extracellular signal-regulated kinase (ERK)-1/-2-type mitogen-activated protein (MAP) kinase, pre-incubation with angiotensin II abolished insulin-induced stimulation of these kinases, suggesting more proximal events in insulin signalling may be intercepted. Pretreatment with angiotensin II markedly inhibited insulin-stimulated tyrosine phosphorylation of insulin-receptor beta-chain and insulin-receptor substrate 1. Losartan prevented angiotensin II inhibition of insulin-induced ERK-1/-2-type MAP kinase activation and 4E-BP1 phosphorylation, suggesting mediation of the effect of angiotensin II by its type 1 receptor. Insulin-stimulated de novo protein synthesis was also abolished by pre-incubation with angiotensin II. These data show that angiotensin II inhibits 4E-BP1 phosphorylation and stimulation of protein synthesis induced by insulin by interfering with proximal events in insulin signalling. Our data provide a mechanistic basis for insulin insensitivity induced by angiotensin II.

Activation of renal signaling pathways in db/db mice with type 2 diabetes.

BACKGROUND: Altered regulation of signaling pathways may contribute to the pathogenesis of renal disease. We examined renal cortical signaling pathways in type 2 diabetes. METHODS: The status of renal cortical signaling pathways was examined in control and db/db mice with type 2 diabetes in the early phase of diabetic nephropathy associated with renal matrix expansion and albuminuria. RESULTS: Tyrosine phosphorylation of renal cortical proteins was increased in diabetic mice. Renal cortical activities of phosphatidylinositol 3-kinase (PI 3-kinase) in antiphosphotyrosine immunoprecipitates, Akt (PKB), and ERK1/2-type mitogen-activated protein (MAP) kinase activities were significantly augmented sixfold (P < 0.01), twofold (P < 0.0003), and sevenfold (P < 0.001), respectively, in diabetic mice compared with controls. A part of the increased renal cortical PI 3-kinase activity was due to insulin receptor activation, as PI 3-kinase activity associated with beta chain of the insulin receptor was increased nearly fourfold (P < 0.0235). Additionally, the kinase activity of the immunoprecipitated insulin receptor beta chain was augmented in the diabetic renal cortex, and tyrosine phosphorylation of the insulin receptor was increased. In the liver, activities of PI 3-kinase in the antiphosphotyrosine immunoprecipitates and Akt also were increased threefold (P < 0.05) and twofold (P < 0.0002), respectively. However, there was no change in the hepatic insulin receptor-associated PI 3-kinase activity. Additionally, the hepatic ERK1/2-type MAP kinase activity was inhibited by nearly 50% (P < 0.01). CONCLUSIONS: These studies demonstrate that a variety of receptor signaling pathways are activated in the renal cortex of mice with type 2 diabetes, and suggest a role for augmented insulin receptor activity in nephropathy of type 2 diabetes.

Insulin regulation of protein translation repressor 4E-BP1, an eIF4E-binding protein, in renal epithelial cells.

BACKGROUND: Augmented protein translation by insulin involves activation of eukaryotic initiation factor 4E (eIF4E) that follows release of eIF4E from a heterodimeric complex by phosphorylation of its inhibitory binding protein, 4E-BP1. We examined insulin regulation of 4E-BP1 phosphorylation in murine proximal tubular epithelial cells. METHODS AND RESULTS: Insulin (1 nmol/L) increased de novo protein synthesis by 58 +/- 11% (P < 0.001). Insulin also augmented 4E-BP1 phosphorylation and phosphatidylinositol 3-kinase (PI 3-kinase) activity in antiphosphotyrosine immunoprecipitates. This could be prevented by PI 3-kinase inhibitors, Wortmannin, and LY294002. Insulin also activated Akt that lies downstream of PI 3-kinase. Rapamycin abrogated 4E-BP1 phosphorylation in response to insulin, suggesting involvement of mammalian target of rapamycin (mTOR), a kinase downstream of Akt. Insulin-stimulated phosphorylation of 4E-BP1 was also inhibited by PD098059, implying involvement of Erk-1/-2 mitogen-activated protein (MAP) kinase. An increase in Erk-1/-2 type MAP kinase activity by insulin was directly confirmed in an immunokinase assay and was found to be PI 3-kinase dependent. CONCLUSIONS: In proximal tubular epithelial cells, insulin augments 4E-BP1 phosphorylation, which is PI 3-kinase and mTOR dependent. The requirement for Erk-1/-2 MAP kinase activation for 4E-BP1 phosphorylation by insulin suggests a cross-talk between PI 3-kinase and Erk-1/-2-type MAP kinase pathways.

Regulation of renal laminin in mice with type II diabetes.

This study examines the regulation of renal laminin in the db/db mouse, a model of type II diabetes characterized by extensive remodeling of extracellular matrix. Immunohistochemistry demonstrated an increase in the contents of laminin chains including beta1 chain in the mesangium and tubular basement membranes at 1, 2, 3, and 4 mo of diabetes. Immunofluorescence with an antibody against the recently discovered laminin alpha5 chain showed that in the normal mouse, the protein had a restricted distribution to the glomerular and tubular basement membranes with scant expression in the mesangium of older mice. In the diabetic mouse, the laminin alpha5 chain content of the glomerular and tubular basement membranes was increased, with marked expression in the mesangium. Northern analysis revealed a significant decrease in the renal cortical contents of alpha5, beta1, and gamma1 chain mRNA in the diabetic mice compared to control, at each of the time points. In situ hybridization showed decreased abundance of alpha5 transcripts in the glomeruli of diabetic mice compared to nondiabetic controls. Analysis of mRNA changes by Northern and in situ hybridization studies demonstrated that the reduction in laminin transcripts involved both glomerular and tubular elements. These observations demonstrate that laminin accumulation in the db/db mice with type II diabetes is due to nontranscriptional mechanisms. Because previous investigations in rodents with type I diabetes have shown that the increase in renal laminin content was associated with a corresponding increment in laminin chain transcript levels, it appears that the mechanisms underlying augmentation in renal matrix laminin content may be distinct in the two types of diabetes.

Successful treatment of valproic acid overdose with hemodialysis.

A 43-year-old woman took a large amount of depakote (divalproex, a slow-release form of valproate), became comatose, and developed severe hypotension refractory to fluid resuscitation and high-dose vasopressors. The serum valproic acid (VPA) concentration on admission was 1,380 microgram/mL (therapeutic range, 50 to 100 microgram/mL). She also had metabolic acidosis, thrombocytopenia, and normal renal and liver functions. Hemodialysis was initiated 4 hours after presentation. After 6 hours of hemodialysis with a high-flux dialyzer, her serum VPA concentration decreased from 940 microgram/mL to 164 microgram/mL, coincident with improvement in clinical status. The half-life of VPA was reduced to 2.4 hours with hemodialysis, whereas it was 7.2 hours before the procedure. Hemodialysis could be a valuable therapeutic intervention in VPA toxicity.

Cyclic AMP regulates basement membrane heparan sulfate proteoglycan, perlecan, metabolism in rat glomerular epithelial cells.

Perlecan, the basement membrane heparan sulfate proteoglycan (HSPG), has been fully cloned from mouse and human tissues. When a cRNA probe of murine perlecan cDNA was employed in RNase protection assay to test whether rat glomerular epithelial cells (GEC) constitutively express perlecan, several bands of hybridization were seen, suggesting that sequences between rat and murine perlecan may not be identical. Using primers based on published cDNA sequences of murine and human perlecan and poly A+ RNA of rat GEC, we synthesized a 497 bp product (RPD-I) by RT-PCR. The deduced aminoacid sequence showed an 85% and 88% homology with domain I of murine and human perlecan, respectively. The three putative sites containing the consensus sequence SGD for attachment of heparan sulfate chains were fully conserved in the rat perlecan as was a site (NFT) for attachment of N-linked oligosaccharide. RPD-I detected a > 9.5 kb transcript of perlecan in RNA of GEC, similar in size to that present in rat glomeruli. Employing a riboprobe synthesized from RPD-I in RNase protection assay we examined whether dbcAMP regulated perlecan expression in the GEC. At 1, 6, 24 and 48 h of incubation, 1 mM dbcAMP caused 43%, 32%, 47% and 40% reduction in mRNA abundance of perlecan, respectively. Immunoprecipitation showed a corresponding reduction of 61%, 70% and 65% in the synthesis of 35SO4 labeled basement membrane HSPG by the GEC following 12, 24 and 48 h of incubation with dbcAMP. Following incubation for 1 and 24 h prostaglandins, PGE1 and PGE2 (1 uM), known activators of glomerular adenylate cyclase, reduced perlecan mRNA abundance to a similar extent as dbcAMP on northern analysis. Our results show that glomerular basement membrane HSPG synthesized by the GEC belongs to the perlecan family. Decrease of GEC perlecan gene expression and synthesis by cAMP and prostaglandins may be of relevance to proteinuric states characterized by activation of these mediators.

Sifting the causes of microscopic hematuria.

Among the goals of the primary care workup are detection of serious disease and determination of whether referral--such as to a urologist or nephrologist--is indicated. Diagnosis is facilitated when hematuria is accompanied by other findings; isolated hematuria presents a more complex challenge. Measurement of serum complement levels may be helpful in narrowing the differential.

Regulation of basement membrane heparan sulfate proteoglycan, perlecan, gene expression in glomerular epithelial cells by high glucose medium.

Proteinuria in diabetic nephropathy has been correlated with reduction in heparan sulfate proteoglycan (HSPG) content of the glomerular basement membrane. We have previously shown that the underlying mechanism probably involves reduction in the synthesis by glomerular epithelial cells. In this study we explored whether high glucose medium regulates basement membrane HSPG gene expression. Northern analysis demonstrated that rat glomerular epithelial cells in vitro constitutively express mRNA for basement membrane HSPG, similar to that observed in rat kidney glomerulus. RNase protection assay showed that incubation of glomerular epithelial cells with 30 mM glucose for 24 h and 7 days resulted in reduction in HSPG mRNA abundance. The decrease in mRNA abundance correlated with reduction in the synthesis of 35SO4-labeled basement membrane HSPG as measured by immunoprecipitation. Reduction in synthesis of HSPG could not be entirely accounted for by decrease in mRNA abundance, suggesting both transcriptional and posttranscriptional mechanisms may be involved in reduction of glomerular basement membrane HSPG synthesis by glomerular epithelial cells in diabetic nephropathy.

Effect of insulin on high-glucose medium-induced changes in rat glomerular epithelial cell metabolism of glycoconjugates.

We have previously reported that incubation of rat glomerular epithelial cells in vitro for 8 days with 30 mM glucose without insulin results in reduction in the synthesis of a cell layer heparan sulfate proteoglycan (HSPG) species that has hydrodynamic size and antigenic characteristics of glomerular basement membrane HSPG (1994, Arch. Biochem. Biophys 309, 149-159). In these studies, reduction in HSPG synthesis could be attributed either to high-glucose medium or to insulin deficiency. In this study we investigated the effects of insulin replacement on changes in glomerular epithelial cell metabolism of glycoconjugates induced by high-glucose medium. Addition of pharmacologic concentrations of insulin prevented the following changes induced by 30 mM glucose: (a) increment in 35SO4 incorporation into macromolecules in cell layer and medium, (b) increment in the synthesis of low anionic macromolecules, probably glycoproteins, in both cell layer and medium, (c) increment in synthesis of small-sized glycosaminoglycans (Kav 0.75 on Sephrose CL-4B) associated with the cell layer. Insulin was unable to correct the 30 mM glucose-induced reduction in the synthesis of cell layer HSPG that resembles glomerular basement membrane HSPG. Physiologic concentrations of insulin did not affect any of the changes in glycopeptide metabolism induced by 30 mM glucose. These findings suggest that (a) inhibition of glomerular epithelial cell synthesis of 35SO4-labeled low anionic macromolecules, probably glycoproteins, may be involved in insulin-induced reversal of glomerular hypertrophy seen in early diabetes, and (b) mechanisms other than insulin lack are involved in reduction in the synthesis of glomerular basement membrane HSPG in diabetic nephropathy.

Regulation of rat glomerular epithelial cell proteoglycans by high-glucose medium.

In diabetic nephropathy the heparan sulfate proteoglycan (HSPG) content of the glomerular basement membrane (GBM) is reduced but the cellular mechanisms involved have not been studied. Glomerular epithelial cells (GEC) are thought to be the source of HSPG present in the GBM. In this study we examined if proteoglycan metabolism of the rat GEC in culture is dysregulated in a metabolic environment simulating diabetes. Following incubation for 8 days with a serum-supplemented medium containing 30 mM glucose and no added insulin, a significant increase in the overall synthesis of 35SO4-labeled molecules by the GEC was seen compared to control monolayers incubated with medium containing 5 mM glucose and insulin. Ion exchange chromatography revealed that 30 mM glucose did not alter the anionic charge density of proteoglycans, but significantly increased the amount of 35S-labeled low-anionic macromolecules in the medium; mannitol induced similar changes. Sepharose CL-4B chromatography, glycosaminoglycan analysis and immunoprecipitation of control cell layer proteoglycans demonstrated the presence of HSPG of hydrodynamic size, Kav 0.4, resembling rat GBM HSPG in size and antigenic nature. Incubation of GEC with 30 mM glucose resulted in a significant reduction (58%) in this HSPG species, an effect not seen with equimolar mannitol. Additionally, 30 mM glucose induced a significant increment in synthesis of a small HS species (Kav 0.71 on Sepharose CL-4B column) present in the cell layer. Our findings suggest that both osmotic and nonosmotic mechanisms are operative in dysregulation of glycopeptide metabolism by high-glucose medium and that reduced synthesis by the GEC may contribute to decreased content of GBM HSPG in diabetic nephropathy.

Metabolic fate of monovalent and multivalent antibodies of Heymann nephritis following formation of surface immune complexes on glomerular epithelial cells.

Heymann nephritis of the rat has several similarities to human idiopathic membranous glomerulo-nephropathy, and therefore serves as an animal model for study of the human disease. The disease process in the rat is initiated by autoantibodies forming in situ immune deposits on the surface of the glomerular epithelial cell in the lamina rara externa region of the glomerular basement membrane. It is now understood that multispecific antibodies (like anti-gp600) form the nephritogenic deposits which cause proteinuria, while MoAb (like anti-gp330, anti-90 kD and anti-gp70) form only transient deposits which do not give rise to proteinuria. The reasons for differences between pathologic effects of mono- and multispecific antibodies are not known. Following characterization of putative antigens of Heymann nephritis by immunofluorescence, immunogold and immunoblot techniques, we have investigated the metabolic handling of a multispecific (anti-600), monospecific antibody (anti-gp70) and a MoAb (anti-gp330) by the cultured glomerular epithelial cell to gain an insight into the mechanism of nephritogenicity of multispecific antibody. Anti-gp600 reacted to multiple antigens in the 330-kD and 70-kD regions; anti-gp70 reacted to only the 70-kD region. Ultrastructurally, all three types of antigens were present on the plasma membrane, concentrated in the microvillar region. The cells were incubated with the antibodies, and clearance of the antibodies from the cells was evaluated. Following binding, all three antibodies were internalized by the cells. However, it was found that monoclonal anti-gp330 MoAb was cleared most rapidly from the cell (t1/2 5 min), followed by anti-gp70 (t1/2 30 min). Anti-gp600 was cleared at four times slower rate than anti-gp70 (t1/2 2h) by the cell. While anti-gp330 MoAb and monospecific anti-gp70 antibody were expelled from the cell in 25-32% digested form, the multispecific anti-gp600 was digested to the extent of only 5-8%. No immune complexes were detected in the medium with any of the three antibodies. The shed label was in intact IgG form. It is concluded that multispecific antibodies of Heymann nephritis are nephritogenic because of their slower clearance and digestion and therefore, higher accretion rate on the surface of the glomerular epithelial cell. The differential handling of the antibodies by the glomerular epithelial cell offers an explanation for the in vivo differences in the nephritogenicity of these antibodies observed earlier.

Glomerular endothelial cell proteoglycans--regulation by TGF-beta 1.

Glomerular endothelial cells (G-Endo) may participate in the synthesis of glomerular basement membrane. We studied the metabolism of proteoglycans and its regulation by TGF-beta 1 in bovine G-Endo. The synthesis of cell layer and medium-associated [35S]SO4-labeled macromolecules was significantly increased by TGF-beta 1. On ion exchange chromatography, 80-90% of the 35S-labeled macromolecules in the control medium and cell layer were found to be proteoglycans; TGF-beta 1 increased their synthesis by 1.8-fold but did not alter the anionic charge density. On dissociative Sepharose CL-4B chromatography, the medium proteoglycans were distributed into two peaks of Kav 0.22 (M-I) and 0.44 (M-II). Digestion procedures showed that 39% and 57% of peaks M-I and M-II consisted of HSPG, the remainder being CS/DSPG. The HSPG in peak M-II has the same hydrodynamic size as HSPG present in bovine glomerular basement membrane (N. Parthasarathy and R. G. Spiro, 1987, J. Biol. Chem. 259:12749). Cell layer proteoglycans also resolved into two peaks of Kav 0.33 (C-I) and 0.63 (C-II), respectively, on Sepharose CL-4B chromatography; 39 and 45% of these peaks were made of HSPG. TGF-beta 1 increased the synthesis of CS/DSPG in the media by three-fold without affecting the synthesis of HSPG. Our data suggest that G-Endo may participate in synthesis of glomerular matrix HSPG under physiologic conditions; regulation of G-Endo proteoglycans by TGF-beta 1 may be of relevance in glomerular diseases characterized by matrix expansion.

Glomerular epithelial cells synthesize endothelin peptides.

Glomerular mesangial and endothelial cells have been reported to synthesize and secrete endothelin-1 (ET-1). Whether glomerular epithelial cells (GEC) have the ability to synthesize ET-peptides is not known. Employing immunocytochemistry we report that the GEC in vitro constitutively express ET-1 and ET-3. ET-1 synthesis by the GEC was further confirmed by detection of a specific 2.3kb mRNA that hybridizes with rat prepro ET-1 genomic DNA on Northern blot analysis. ET-1 is secreted into the medium in a time-dependent manner as measured by radioimmunoassay and radiobinding assay. Synthesis of endothelin peptides by the GEC may have important implications in the pathogenesis of glomerular diseases where GEC injury figures prominently.

Interaction of polycations with cell-surface negative charges of epithelial cells.

The interaction between various polycations and cultured glomerular epithelial cells was studied by cell electrophoresis. It was shown that the glomerular epithelial cell presents a negatively charged surface which imparts a zeta potential of -29.0 +/- 1.5 mV at the peripheral layer of the plasma membrane. The pH at which the GEC charge became 50% reduced (pKa) was determined to be 3.0. A variety of polycations of various sizes and fixed and flexible geometries were tested for their capacity to neutralize the cell charge. All the polycations except cytochrome c and lysozyme were capable of completely neutralizing the cell. Cytochrome c could maximally neutralize only 50% of charge and lysozyme only 72% of charge. However, reduced and 'relaxed' molecules of cytochrome c and lysozyme efficiently neutralized the cell surface, as did larger sized 'flexible' polylysines. On the basis of these findings, we conclude that all polycations are not equal in their capacity to neutralize the cell surface. Flexible molecules in contrast to molecules with rigid structures were more effective in neutralizing the cell. This may likely be due to the exposure and availability of more cationic groups in a flexible molecule which results in stabilization of interaction with cells.

Progression and remission of renal disease in the Lupus Nephritis Collaborative Study. Results of treatment with prednisone and short-term oral cyclophosphamide.

OBJECTIVE: To describe the clinical course of severe lupus nephritis and to identify the risk factors for progression to renal failure among patients treated with prednisone and short-term courses of low-dose oral cyclophosphamide. DESIGN: Ancillary analyses of data from the Lupus Nephritis Collaborative Study (LNCS). SETTING: University hospital medical centers (14). PATIENTS: The 86 patients who participated in the LNCS (mean follow-up, 136 weeks [2.6 years]) and a subgroup of 63 patients with follow-up of more than 48 weeks (mean follow-up, 160 weeks [3.1 years]). MEASUREMENTS: Initial clinical and pathologic features, response to therapy within 48 weeks, and subsequent clinical events, including development of renal failure. MAIN RESULTS: Renal failure developed in 18 patients (21%). An observed elevation in serum creatinine concentration was the only initial feature predictive of subsequent renal failure. Mean (+/- SD) initial serum creatinine levels were higher in patients who subsequently developed renal failure (244 +/- 134 mumol/L [2.76 +/- 1.52 mg/dL] compared with 163 +/- 103 mumol/L [1.85 +/- 1.17 mg/dL]; P = 0.007). The risk for renal failure was higher among patients with initial serum creatinine levels greater than 106 mumol/L (1.2 mg/dL) (29% compared with 6.5%; P = 0.014). Response to therapy (defined as resolution of initial serum creatinine elevations within 48 weeks) refined the prognosis based on initial serum creatinine determinations. The risk for subsequent renal failure was higher among patients who failed to respond to therapy within 48 weeks (30% compared with 0%; P = 0.015). By comparison, 9% of patients with normal initial serum creatinine levels progressed to renal failure after 48 weeks. CONCLUSIONS: Initial serum creatinine levels and responses to initial therapy with prednisone and short-term cyclophosphamide, as used in the LNCS, can guide further therapy. Patients with normal initial serum creatinine levels or resolution of initial serum creatinine elevations within 48 weeks have a low risk for renal failure and may not require long-term treatment with cyclophosphamide.