- Email: [email protected]
High glucose reduces albumin uptake in cultured proximal tubular cells (LLC-PK1)
Diabetes Research and Clinical Practice 65 (2004) 217–225
High glucose reduces albumin uptake in cultured proximal tubular cells (LLC-PK1) Fukashi Ishibashi Ishibashi Clinic, 1-9-41-2 Kushido Hatsukaichi Hiroshima 738-0033, Japan Received 10 September 2003; received in revised form 20 January 2004; accepted 2 February 2004
Abstract In this study, we clarify that high glucose inhibits albumin uptake in cultured LLC-PK1 cells. LLC-PK1 cells cultured for 6 days with 5.5–27.8 mM d-glucose were challenged by fluorescein isothiocyanate (FITC)-conjugated human albumin (HA). FITC-HA binding and uptake were inhibited by >5.5 mM glucose (5.5 mM > (P < 0.01) 11.0 mM > (P < 0.05) 16.7 mM ∼ = 27.8 mM). Analysis of FITC-HA binding and uptake at 5.5 and 16.7 mM d-glucose (high glucose, HG) showed decreased affinity (Km for binding: 35.5 mg/l versus 52.6 mg/l, Km for uptake; 41.3 mg/l versus 55.6 mg/l) and maximal velocity (Bmax —0.33 g versus 0.27 g/30 min/mg protein; Umax —4.40 g versus 3.48 g/60 min/mg protein) at HG. A comparison of the time courses of FITC-HA binding and uptake at 5.5 mM glucose and at HG showed that HG suppressed them beyond 15 min (P < 0.05–0.001). Phlorizin (>0.25 mM) completely reversed the HG-induced inhibition of FITC-HA binding and uptake. High glucose decreased mRNA of GLUT-1 and SGLT-1, but did not influence that of SGLT-2. The simultaneous presence of Vitamin E (10−6 M), Vitamin C (10−6 M) and reduced glutathione (0.25 mM) reversed the suppressed FITC-HA binding and uptake by HG, while any one or two of these molecules, and various inhibitors of advanced glycation end products, failed to do so. In conclusion, a high glucose milieu causes inhibition of albumin binding and uptake in proximal tubular cells by increasing metabolic oxidative stress through excessive glucose flux via the sodium glucose transporter. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Proximal tubular cells; Albumin uptake; High glucose; Phlorizin; Oxidative stress
1. Introduction Microalbuminuria is the most reliable predictor of diabetic nephropathy in patients with type 1 [1] as well as type 2 [2] diabetes. In developing microalbuminuria, impaired tubular reabsorption of albumin plays a role besides glomerular hyperfiltration of albumin [3]. However, no direct in vitro evidence exists that a high glucose environment reduces the uptake
E-mail address: [email protected] (F. Ishibashi).
of albumin in cultured proximal tubular cells. Glucose enters into tubular epithelial cells via glucose transporter-1 (GLUT-1) basolaterally independent of insulin, resulting in a direct relationship between a high plasma glucose concentration and the intracellular glucose level [4]. In addition, the excess glucose in the glomerular filtrate in diabetes leads to enhanced proximal tubular glucose influx via sodium glucose transporters (SGLT) from tubular lumen [5]. Although histological study confirmed that renal function correlates better with tubular and interstitial changes than with glomerular changes [6,7],
0168-8227/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.02.003
218
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
most studies concerning the development of diabetic nephropathy have focused on morphological and functional changes in the glomerulus, and especially in mesangial cells [8,9]. Hence, the present study was performed to show that ambient high glucose inhibits the binding and uptake of fluorescein isothiocyanate (FITC)-human albumin (HA) in cultured porcine proximal tubular epithelial cells (LLC-PK1 cells). Furthermore, in order to address the inhibitory mechanisms, the modification of glucose transporters by high glucose, and the influence of SGLT blocker and antioxidants on the high-glucose induced inhibition of albumin binding and uptake were examined.
2. Materials and methods The cell culture medium was Medium 199 (M199, Gibco, NY, USA) containing either physiological (5.5 mM) or high d-glucose concentrations (11.0–27.8 mM). During the culture period, d-glucose levels in the medium were kept constant by adding a 50% d-glucose solution according to the measured d-glucose concentration. The media were supplemented with 100 g/ml streptomycin, 100 u/ml penicillin (Sigma, MO, USA) and 5% fetal calf serum (FCS, Biological Industries, Kibbutz Beit Haemek, Israel). All other reagents were obtained from Sigma (MO, USA).
min C (VitC), reduced glutathione, N-acetyl-cysteine, taurine or aminoguanidine, for 10 min at 37 ◦ C. After removal of the preincubation media, the cells were pulsed with preincubation media supplemented with FITC-HA, and incubated for 60 min at 37 ◦ C for uptake. The uptake represents albumin bound to putative albumin binding protein as well as that trapped in the endocytic pathway to lysosome. It has been well known that at 4 ◦ C endocytic steps beyond albumin binding at apical membrane are completely inhibited in LLC-PK1 [10] and other cultured proximal tubular cells [11,12]. Therefore, for binding cells were incubated with FITC-HA at 4 ◦ C for 30 min. FITC-HA remaining in the incubation media was removed by rinsing three times with ice-cold PBS. The cells were scraped with a rubber policeman into 1 ml of ice-cold PBS and washed twice by centrifugation at 4 ◦ C at 1000 cpm for 5 min. Cell pellets disintegrated by 0.1% Triton X-100 in PBS were used in the fluorescence and protein assays. Non-specific uptake was determined by measuring FITC-HA uptake in the presence of 1000 times cold HA. For the determination of the inulin space in cultured LLC-PK1 cells, FITC-inulin (300 g/ml) was pulsed for 60 min at 37 ◦ C. FITC-inulin uptake was determined as described for FITC-HA uptake. The substances employed in the present study did not alter the inulin space (4.6–5.2 l/mg protein). 4.1. RT-PCR analysis of mRNA of GLUT-1, SGLT-1 and SGLT-2
3. Cell culture LLC-PK1 cells were obtained from Dainipponseiyaku (Osaka, Japan) and used between passages 144 and 156. Culture media were replaced every 48–72 h, and the cells were incubated in a humidified atmosphere of 5% CO2 in air at 37 ◦ C.
4. FITC-HA binding and uptake Cell monolayers grown on 35 mm culture dishes for 6 days were washed twice with 1.2 ml of phosphate buffered saline (PBS) supplemented with 5.5 mM d-glucose, and preincubated with PBS containing varying concentrations of d-glucose (5.5–27.8 mM), either alone or with phlorizin, Vitamin E (VitE), Vita-
Total RNA was extracted from LLC-PK1 cells maintained at 5.5 or 16.7 mM glucose with or without 0.25 mM phlorizin for 6 days using QIAamp RNA Mini kit (QIAGEN, Tokyo, Japan). Primers were synthesized chemically on the basis of sequences of GLUT-1(sense: 5 -TGTCGCTGTTCGTGGTGGAA3 , antisense: 5 -GACCAGGAGCACCGTGAAGA-3 , 357bps), SGLT-1(sense: 5 -CCGGTTGGAGCTTCTCTGTT-3 , antisense: 5 -AAGACCCCACCAGCATGATT-3 , 441bps) and SGLT-2 (sense: 5 -ATACTGGTCGTCCTGGCAAT-3 , antisense: 5 -GATGT TCACTATAGTCTTCC-3 , 317bps). RT-PCR was performed using a ReverTra Ace RT-PCR kit (TOYOBO, Osaka, Japan). Amplification was performed according to the following profile: denaturation at 95 ◦ C for 30 s, annealing at 58 ◦ C for 30 s, extension
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
at 72 ◦ C for 1 min, 25 cycles. The PCR products were separated by electrophoresis in 1.6% agarose gels, and amplified bands were detected by ethidium bromide staining. Data are presented as means ± S.E.M. The significance of difference was tested by either paired or unpaired t-tests. Differences were considered significant for P < 0.05.
219
5. Results A d-glucose concentration higher than 11.0 mM suppressed FITC-HA binding (Fig. 1A) and uptake (Fig. 1B) in LLC-PK1 cells, with maximal suppression obtained at 16.7 mM. The same concentration range of d-mannitol did not influence the FITC-HA uptake (Fig. 1B).
Fig. 1. (A) Influence of high glucose on FITC-HA binding in LLC-PK1 cells. After subcultivation in 5.5–27.8 mM d-glucose for 6 days, the cells were pulsed with FITC-HA (30 mg/l) at 4 ◦ C for 30 min. (B) Influence of high glucose and d-mannitol on FITC-HA uptake in LLC-PK1 cells. After subcultivation in varying concentrations of d-glucose or d-mannitol, the cells were pulsed with FITC-HA (30 mg/l) at 37 ◦ C for 60 min in the presence of the same concentrations of d-glucose or d-mannitol. Values are means ± S.E. of five dishes.
Fig. 2. (A) Concentration-dependent binding of FITC-HA in LLC-PK1 cells. After subcultivation in 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with varying concentrations of FITC-HA (10–500 mg/l) at 5.5 or 16.7 mM d-glucose at 4 ◦ C for 30 min. Values are means of four dishes. (B) Scatchard plot of the concentration-dependent binding of FITC-HA to LLC-PK1 cells at 5.5 mM (䊊) or 16.7 mM d-glucose (䊉).
220
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
Fig. 2A shows a comparison of the concentration dependency of FITC-HA binding in LLC-PK1 cells grown in 5.5 and 16.7 mM d-glucose. d-Glucose at 16.7 mM reduced the binding at any substrate concentration. A Scatchard analysis of the binding kinetics (Fig. 2B) showed that high glucose reduced the affinity (Km = 35.5 mg/l at 5.5 mM d-glucose versus Km = 52.6 mg/l at 16.7 mM d-glucose) and the maximum velocity (Bmax = 0.33 g/mg protein/30 min at 5.5 mM d-glucose versus Bmax = 0.27 g/mg protein/30 min at 16.7 mM d-glucose) of albumin binding. A comparison of the concentration dependency of FITC-HA uptake at 5.5 and 16.7 mM d-glucose showed that high glucose reduced FITC-HA uptake at any FITC-HA concentration (Fig. 3A). The kinetic analysis revealed that high glucose reduced the affinity (Km = 41.3 mg/l at 5.5 mM d-glucose versus 55.6 mg/l at 16.7 mM d-glucose) and the maximal velocity Umax = 4.4 g mg protein/60 min at 5.5 mM d-glucose versus Umax = 3.48 g/mg protein/60 min at 16.7 mM d-glucose) of uptake (Fig. 3B). The time course of FITC-HA binding (A) and uptake (B) at 5.5 and 16.7 mM d-glucose is compared in Fig. 4, and shows that high glucose significantly suppressed FITC-HA binding and uptake beyond 15 min.
As shown in Fig. 5, the inhibition of SGLT by phlorizin ameliorated the high glucose-induced inhibition of FITC-HA binding (Fig. 5A) and uptake (Fig. 5B) in a dose dependent manner. A phlorizin concentration greater than 0.1 mM reversed the high glucose-induced inhibition. The mRNA levels of GLUT-1, SGLT-1 and SGLT-2 in LLC-PK1 maintained at 5.5 mM glucose or 16.7 mM glucose with or without 0.25 mM phlorizin were determined by PCR amplification (Fig. 6). High glucose decreased mRNA of GLUT-1 and SGLT-1 by 30%, but did not influence that of SGLT-2. The addition of phlorizin did not influence the mRNA levels of GLUT-1 and SGLTs in cells maintained at 16.7 mM glucose. The influence of various antioxidants alone or in combination on the high glucose-induced inhibition of FITC-HA binding (Fig. 7A) and uptake (Fig. 7B) was examined. VitC, VitE or reduced glutathione alone or couple of any two drugs were unable to restore the suppression of FITC-HA binding and uptake by 16.7 mM d-glucose, over a wide range of antioxidant concentrations (VitC, 10−7 –2.5 × 10−5 M; VitE, 10−7 –2.5 × 10−5 M; reduced glutathione, 0.1–2.5 mM). On the other hand, the simultaneous presence of the three antioxidants completely
Fig. 3. (A) Concentration-dependent uptake of FITC-HA in LLC-PK1 cells. After subcultivation in 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with varying concentrations of FITC-HA (10–500 mg/l) at 5.5 or 16.7 mM d-glucose at 37 ◦ C for 60 min. Values are means of four dishes. (B) Scatchard plot of the concentration-dependent uptake of FITC-HA in LLC-PK1 cells at 5.5 mM (䊊) or 16.7 mM d-glucose (䊉).
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
221
abolished the inhibition by high glucose (Fig. 7A and B). As shown in Fig. 8, N-acetyl-cysteine, taurine and aminoguanidine were unable to ameliorate the inhibition of FITC-HA uptake by high glucose. Neither of them influenced high glucose-induced reduction of FITC-HA binding at all (figure not shown).
6. Discussion
Fig. 4. Comparison of the time courses of FITC-HA binding (A) and uptake (B) in LLC-PK1 cells maintained at 5.5 and 16.7 mM d-glucose. After subcultivations at 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with FITC-HA (30 mg/l) at 5.5 mM (䊊) or 16.7 mM (䊉) d-glucose at 4 ◦ C for binding and at 37 ◦ C for uptake for 3–60 min. Values were means ± S.E. of four dishes. ∗ P < 0.05, ∗∗ P < 0.01, ∗∗∗ P < 0.001 vs. 5.5 mM glucose.
To our knowledge, this study is the first to demonstrate that ambient high glucose reduces albumin binding and uptake in proximal tubular epithelial cells in vitro, and that the inhibition by phlorizin of excess glucose entry via SGLT, or a combination of antioxidants, completely reverses the reduced albumin binding and uptake. High glucose caused a dose-dependent reduction in FITC-HA binding and uptake (Fig. 1). Since the same concentration of d-mannitol did not influence FITC-HA uptake at all, the reduced FITC-HA binding and uptake caused by high glucose is not due to increased osmosis. When FITC-HA binding and uptake were analyzed by Scatchard plot, it was linear as found in previous many reports using cultured proximal tubular cells [11,13,14]. The Km values for albumin binding and
Fig. 5. Influence of phlorizin on high glucose-induced suppression of FITC-HA binding (A) and uptake (B) in LLC-PK1 cells. After subcultivation with 16.7 mM d-glucose and phlorizin (0.1–1.0 mM) for 6 days, the cells were pulsed with FITC-HA (30 mg/l) in the presence of 16.7 mM d-glucose and phlorizin (0.1–1.0 mM) at 4 ◦ C for 30 min for binding and at 37 ◦ C for 60 min for uptake. Values were means ± S.E. of five dishes.
222
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
Fig. 6. Comparison of mRNA levels of GLUT-1, SGLT-1 and SGLT-2 by PCR amplification in LLC-PK1 cultured at 5.5 mM glucose or 16.7 mM glucose with or without 0.25 mM phlorizin. GLUT-1: lane 1—at glucose 5.5 mM; lane 2—at glucose 16.7 mM; lane 3—at glucose 16.7 mM ± 0.25 mM phlorizin. SGLT-1: lane 4—at glucose 5.5 mM; lane 5—at glucose 16.7 mM; lane 6—at glucose 16.7 mM ± 0.25 mM phlorizin. SGLT-2: lane 7—at glucose 5.5 mM; lane 8—at glucose 16.7 mM; lane 9—at glucose 16.7 mM ± 0.25 mM phlorizin.
uptake at 5.5 mM d-glucose seemed to be similar to the albumin concentration in the tubular fluid that has been determined in animals [15]. On the other hand, the Km values at 16.7 mM d-glucose increased by 48.2 and 34.6%, respectively. Furthermore, high glucose reduced Bmax (binding) and Umax (uptake). The compromised sensitivity and velocity of albumin handling induced by high glucose in proximal tubular cells, which are exposed to excess albumin from the glomerulus, result in microalbuminuria. These findings indicate that in diabetes, where hyperglycemia and glycosuria increase glucose flux into proximal tubular cells via GLUT-1 and SGLTs, the proximal tubules will reabsorb albumin less effectively. The present study revealed that in LLC-PK1, high glucose down-regulates GLUT-1 and SGLT-1 at transcription level, but leaves SGLT-2 unaffected as reported previously [16,17]. Phlorizin, a SGLT inhibitor, did not influence SGLTs in cells maintained by high glucose. In our study, phlorizin (0.25–1.0 mM) reversed the reduction in FITC-HA binding and uptake by high glucose, and T-1095, a derivative of phlorizin, has been shown to prevent microalbuminuria and tubular morphological changes without influencing glomerular architecture in STZ-diabetic rats [18]. This indicates that an excess glucose flux mainly through SGLT-2 in diabetes plays a pivotal role in the development of impaired tubular reabsorption of albumin and microalbuminuria.
Fig. 7. Influences of various antioxidants alone or in combination on suppression of FITC-HA binding (A) and uptake (B) by high glucose in LLC-PK1 cells. After subcultivation for 6 days with 5.5 or 16.7 mM d-glucose and various antioxidants, the cells were pulsed with FITC-HA (30 mg/l) in the presence of the same concentration of d-glucose and antioxidants at 4 ◦ C for 30 min for binding and at 37 ◦ C for 60 min for uptake. Values are means±S.E. of five dishes. Eth, ethanol; VitE, 10−6 M Vitamin E; VitC, 10−6 M Vitamin C; Gluta, 0.25 mM reduced glutathione.
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
223
Fig. 8. Effects of various inhibitors of AGE on high glucose-induced inhibition of FITC-HA uptake in LLC-PK1 cells. After subcultivation in 16.7 mM glucose and varying concentrations of either N-acetyl-cysteine, taurine or aminoguanidine for 6 days, the cells were pulsed with FITC-HA (30 mg/l) in the presence of 16.7 mM d-glucose and the same concentrations of the AGE inhibitors. Values are means ± S.E. of five dishes.
The process of albumin binding and uptake in proximal tubular cells is still not fully understood. Although an albumin-binding protein in proximal tubular cells has yet to be identified unequivocally, megalin [19], cubilin [20] and a 32 kDa albumin-binding protein reported by us [21] are all possible candidates. As a first step, it is likely that albumin binds to the putative albumin-binding protein at the apical membrane. Because inhibition of SGLT by phlorizin reversed the high-glucose-induced suppression of binding, it appears that excessive glucose entry through SGLT may disturb albumin binding to the putative albumin-binding protein. A comparison of the time courses of FITC-HA binding and uptake at 5.5 and 16.7 mM d-glucose showed that high glucose suppressed binding as well as uptake beyond 15 min (Fig. 4). Since endosomal and lysosomal acidification has been considered to be a rate-limiting step of albumin uptake [22,23], the reduction in albumin uptake by high glucose beyond 15 min might occur because the high glucose induces insufficient acidification of endosomes and lysosomes. This may occur because of reduced activity of proton-translocating adenosine triphosphatase, one of the key factors in acidification [24], since it has been shown that high glucose lowers the ATP level in cultured proximal tubular cells [25]. However, it remains unclear whether the insufficient acidifica-
tion of late endosome and lysosome impairs albumin binding. The reduced number of recycled putative albumin binding protein on apical membrane caused by impaired dissociation between albumin molecule and albumin binding protein due to the insufficient acidification in endosome might be one of mechanisms, because endosomal alkalinisation by bafilomycin A1 reduces albumin binding in OK cells [26]. The etiological role of oxidative stress in the development of diabetic tubular lesions has been postulated [27,28]. Hence, the influence of antioxidants on the high glucose-induced inhibition of FITC-HA binding and uptake in LLC-PK1 cells was examined. VitC, VitE or reduced glutathione alone or a combination of any two drugs was unable to ameliorate inhibition of FITC-HA binding and uptake by high glucose, but simultaneous presence of three antioxidants completely reversed the inhibition. Although previous investigations have shown the antioxidant action of VitE alone against high glucose-induced oxidative stress in vivo and in vitro [29], recent studies have failed to verify the counteraction by VitE alone against the oxidative stress [30]. Because VitE acts as a pro-oxidant [31], especially in hyperglycemia, the simultaneous presence of another antioxidant, such as VitC and reduced glutathione, is required to overcome the oxidative stress [32]. However, it was beyond the scope of this study to make clear the mechanisms through which
224
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
the simultaneous presence of three drugs overcomes the oxidative stress due to high glucose. The AGE inhibitors (N-acetyl-cysteine, taurine and aminoguanidine) have been reported to prevent high glucose-induced morphological and functional changes in the kidney [30,33]. In the present study, none of these AGE inhibitors ameliorated the high glucose-induced reduction in FITC-HA binding and uptake in LLC-PK1 cells. This may be because exposure to high glucose for 6 days might be insufficient for AGE induction. In conclusion, ambient high glucose suppressed albumin binding and uptake in cultured proximal tubular epithelial cells. Excess glucose entry via SGLT and the resultant metabolic oxidative stress play a role in high glucose-induced impaired albumin reabsorption in proximal tubular epithelial cells.
References [1] H.H. Parving, B. Oxenball, P.A. Svendsen, J.S. Christensen, A.R. Anderson, Early detection of patients at risk of developing diabetic nephropathy: a longitudinal study of urinary albumin excretion, Acta Endoclinol. (Copenh.) 100 (1982) 550–555. [2] A. Schmitz, Microalbuminuria and mortality in non-insulindependent diabetes, in: C.E. Mogensen (Ed.), The Kidney and Hypertension, Martinus Nijhoff Publishing, Boston, 1988, pp. 65–70. [3] F. Ishibashi, Microalbuminuria in NIDDM is caused by increased excretion of unmodified albumin, Diabetes 45 (1996) 731–736. [4] S.C. Jones, H.J. Saunders, C.A. Pollock, High glucose increases growth and collagen synthesis in cultured tubulointerstitial cells, Diabet. Med. 16 (1999) 932–938. [5] R.C. Harris, B.M. Brenner, J.L. Seifter, Sodium-hydrogen exchange and glucose transport in renal microvillus vesicles from rats with diabetes mellitus, J. Clin. Invest. 77 (1986) 724–733. [6] J.L. Taft, C.J. Nolan, S.P. Yeung, T.D. Hewitson, F.I. Martin, Clinical and histological correlations of decline in renal function in diabetic patients with proteinuria, Diabetes 43 (1994) 1046–1051. [7] K.A. Nath, Tubulointerstitial changes as a major determinant in the progression of renal damage, Am. J. Kidney Dis. 20 (1992) 1–17. [8] T. Kislinger, C. Fu, B. Huber, et al., N-(Carboxymethyl)lysine adducts of proteins are ligands for receptor for advanced glycation end products that activate cell signalling pathways and modulate gene expression, J. Biol. Chem. 274 (1999) 31740–31749.
[9] V. Scivittaro, M.B. Ganz, M.F. Weiss, AGE induce oxidative stress and activate protein kinase C-beta (II) in neonatal mesangial cells, Am. J. Physiol. 278 (2000) F676–F683. [10] Y. Takakura, T. Morita, M. Fujikawa, et al., Characterization of LLC-PK1 kidney epithelial cells as an in vitro model for studying renal tubular reabsorption of protein drugs, Pharm. Res. 12 (1995) 1968–1972. [11] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Functional characterization of albumin binding to the apical membrane of OK cells, Am. J. Physiol. 271 (1996) F286– F291. [12] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Long-term protein exposure reduces albumin binding and uptake in proximal tubule-derived opossum kidney cells, J. Am. Soc. Nephrol. 9 (1998) 960–968. [13] V. Batuman, S. Guan, Receptor-mediated endocytosis of immunoglobulin light chains by renal proximal tubule cells, Am. J. Physiol. 272 (1997) F521–F530. [14] Y. Sasaki, J. Nagai, Y. Kitahara, N. Takai, T. Murakami, M. Takano, Expression of chloride channel, ClC-5, and its role in receptor-mediated endocytosis of albumin in OK cells, Biochem. Biophy. Res. Comm. 282 (2001) 212–218. [15] P.D. Leber, D.J. Marsh, Micropuncture study of concentration and fate of albumin in rat nephron, Am. J. Physiol. 219 (1970) 358–363. [16] C.J. Clancey, J.E. Lever, Differential regulation of three glucose transporter genes in a renal epithelial cell line, J. Cell. Physiol. 185 (2000) 244–252. [17] T. Ohta, K.J. Isselbacher, D.B. Rhoads, Regulation of glucose transporters in LLC-PK1 cells: Effects of d-glucose and monosaccharides, Molec. Cell. Biol. 10 (1990) 6491– 6499. [18] T. Adachi, K. Yasuda, Y. Okamoto, et al., T-1095, a renal Na+ - glucose transporter inhibitor, improves hyperglycemia in streptozotocin-induced diabetic rats, Metabolism 49 (2000) 990–995. [19] E.I. Christensen, H. Birn, P. Verroust, S.K. Moestrup, Megalin- mediated endocytosis in renal proximal tubule, Renal Failure 20 (1998) 191–199. [20] H. Birn, J.C. Fyfe, C. Jacobsen, et al., Cubilin is an albumin binding protein important for renal tubular albumin reabsorption, J. Clin. Invest. 105 (2000) 1353–1361. [21] F. Ishibashi, Characterization of novel human proximal tubular cell albumin binding protein (hABP), Diabetes 51 (suppl. 2) (2002) A183. [22] I. Mellman, R. Fuchs, A. Helenius, Acidification of the endocytic and exocytic pathways, Ann. Rev. Biochem. 55 (1986) 663–700. [23] G. Rudnick, ATP-driven H+ pumping into intracellular organelles, Ann. Rev. Physiol. 48 (1986) 403–413. [24] N. Nelson, W.R. Harvey, Vacuolar and plasma membrane proton-adenosine triphosphatase, Physiol. Rev. 79 (1999) 361–385. [25] M. Symonian, M. Smogorzewski, W. Marcinkowski, E. Krol, S.G. Massry, Mechanisms through which high glucose concentration raises [Ca2+ ]i in renal proximal tubular cells, Kidney Int. 54 (1998) 1206–1213.
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225 [26] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Endosomal alkalinization reduces Jmax and Km of albumin receptor-mediated endocytosis, Am. J. Physiol. 268 (1995) F899–F906. [27] M. Morcos, A.A.R. Sayed, A. Bierhaus, et al., Activation of tubular epithelial cells in diabetic nephropathy, Diabetes 51 (2002) 3532–3544. [28] H. Ha, K.H. Kim, Pathogenesis of diabetic nephropathy: the role of oxidative stress and protein kinase C, Diabet. Res. Clin. Pract. 45 (1999) 147–151. [29] H. Trachtman, Vitamin E prevents glucose-induced lipid peroxidation and increased collagen production in cultured rat mesangial cells, Microvasc. Res. 47 (1994) 232– 239.
225
[30] H. Trachtman, S. Futterweit, J. Maesaka, Taurine ameliorates chronic streptozotocin-induced diabetic nephropathy in rats, Am. J. Physiol. 269 (1995) F429–F438. [31] V.W. Bowry, K.U. Ingold, R. Stocker, Vitamin E in human low-density lipoprotein: when and how this antioxidant becomes a pro-oxidant, Biochem. J. 288 (1992) 341–344. [32] A.J. Sinchair, A.J. Girling, L. Gray, C. LeGuen, J. Lunec, A.H. Barnett, Disturbed handling of ascorbic acid in diabetic patients with and without microangiopathy during high dose ascorbate supplementation, Diabetologia 34 (1991) 171–175. [33] P. Odetti, C. Pesce, N. Traverso, et al., Comparative trial of N-acetyl-cysteine, taurine, and oxerutin on skin and kidney damage in long-term experimental diabetes, Diabetes 52 (2003) 499–505.
High glucose reduces albumin uptake in cultured proximal tubular cells (LLC-PK1) Fukashi Ishibashi Ishibashi Clinic, 1-9-41-2 Kushido Hatsukaichi Hiroshima 738-0033, Japan Received 10 September 2003; received in revised form 20 January 2004; accepted 2 February 2004
Abstract In this study, we clarify that high glucose inhibits albumin uptake in cultured LLC-PK1 cells. LLC-PK1 cells cultured for 6 days with 5.5–27.8 mM d-glucose were challenged by fluorescein isothiocyanate (FITC)-conjugated human albumin (HA). FITC-HA binding and uptake were inhibited by >5.5 mM glucose (5.5 mM > (P < 0.01) 11.0 mM > (P < 0.05) 16.7 mM ∼ = 27.8 mM). Analysis of FITC-HA binding and uptake at 5.5 and 16.7 mM d-glucose (high glucose, HG) showed decreased affinity (Km for binding: 35.5 mg/l versus 52.6 mg/l, Km for uptake; 41.3 mg/l versus 55.6 mg/l) and maximal velocity (Bmax —0.33 g versus 0.27 g/30 min/mg protein; Umax —4.40 g versus 3.48 g/60 min/mg protein) at HG. A comparison of the time courses of FITC-HA binding and uptake at 5.5 mM glucose and at HG showed that HG suppressed them beyond 15 min (P < 0.05–0.001). Phlorizin (>0.25 mM) completely reversed the HG-induced inhibition of FITC-HA binding and uptake. High glucose decreased mRNA of GLUT-1 and SGLT-1, but did not influence that of SGLT-2. The simultaneous presence of Vitamin E (10−6 M), Vitamin C (10−6 M) and reduced glutathione (0.25 mM) reversed the suppressed FITC-HA binding and uptake by HG, while any one or two of these molecules, and various inhibitors of advanced glycation end products, failed to do so. In conclusion, a high glucose milieu causes inhibition of albumin binding and uptake in proximal tubular cells by increasing metabolic oxidative stress through excessive glucose flux via the sodium glucose transporter. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Proximal tubular cells; Albumin uptake; High glucose; Phlorizin; Oxidative stress
1. Introduction Microalbuminuria is the most reliable predictor of diabetic nephropathy in patients with type 1 [1] as well as type 2 [2] diabetes. In developing microalbuminuria, impaired tubular reabsorption of albumin plays a role besides glomerular hyperfiltration of albumin [3]. However, no direct in vitro evidence exists that a high glucose environment reduces the uptake
E-mail address: [email protected] (F. Ishibashi).
of albumin in cultured proximal tubular cells. Glucose enters into tubular epithelial cells via glucose transporter-1 (GLUT-1) basolaterally independent of insulin, resulting in a direct relationship between a high plasma glucose concentration and the intracellular glucose level [4]. In addition, the excess glucose in the glomerular filtrate in diabetes leads to enhanced proximal tubular glucose influx via sodium glucose transporters (SGLT) from tubular lumen [5]. Although histological study confirmed that renal function correlates better with tubular and interstitial changes than with glomerular changes [6,7],
0168-8227/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.02.003
218
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
most studies concerning the development of diabetic nephropathy have focused on morphological and functional changes in the glomerulus, and especially in mesangial cells [8,9]. Hence, the present study was performed to show that ambient high glucose inhibits the binding and uptake of fluorescein isothiocyanate (FITC)-human albumin (HA) in cultured porcine proximal tubular epithelial cells (LLC-PK1 cells). Furthermore, in order to address the inhibitory mechanisms, the modification of glucose transporters by high glucose, and the influence of SGLT blocker and antioxidants on the high-glucose induced inhibition of albumin binding and uptake were examined.
2. Materials and methods The cell culture medium was Medium 199 (M199, Gibco, NY, USA) containing either physiological (5.5 mM) or high d-glucose concentrations (11.0–27.8 mM). During the culture period, d-glucose levels in the medium were kept constant by adding a 50% d-glucose solution according to the measured d-glucose concentration. The media were supplemented with 100 g/ml streptomycin, 100 u/ml penicillin (Sigma, MO, USA) and 5% fetal calf serum (FCS, Biological Industries, Kibbutz Beit Haemek, Israel). All other reagents were obtained from Sigma (MO, USA).
min C (VitC), reduced glutathione, N-acetyl-cysteine, taurine or aminoguanidine, for 10 min at 37 ◦ C. After removal of the preincubation media, the cells were pulsed with preincubation media supplemented with FITC-HA, and incubated for 60 min at 37 ◦ C for uptake. The uptake represents albumin bound to putative albumin binding protein as well as that trapped in the endocytic pathway to lysosome. It has been well known that at 4 ◦ C endocytic steps beyond albumin binding at apical membrane are completely inhibited in LLC-PK1 [10] and other cultured proximal tubular cells [11,12]. Therefore, for binding cells were incubated with FITC-HA at 4 ◦ C for 30 min. FITC-HA remaining in the incubation media was removed by rinsing three times with ice-cold PBS. The cells were scraped with a rubber policeman into 1 ml of ice-cold PBS and washed twice by centrifugation at 4 ◦ C at 1000 cpm for 5 min. Cell pellets disintegrated by 0.1% Triton X-100 in PBS were used in the fluorescence and protein assays. Non-specific uptake was determined by measuring FITC-HA uptake in the presence of 1000 times cold HA. For the determination of the inulin space in cultured LLC-PK1 cells, FITC-inulin (300 g/ml) was pulsed for 60 min at 37 ◦ C. FITC-inulin uptake was determined as described for FITC-HA uptake. The substances employed in the present study did not alter the inulin space (4.6–5.2 l/mg protein). 4.1. RT-PCR analysis of mRNA of GLUT-1, SGLT-1 and SGLT-2
3. Cell culture LLC-PK1 cells were obtained from Dainipponseiyaku (Osaka, Japan) and used between passages 144 and 156. Culture media were replaced every 48–72 h, and the cells were incubated in a humidified atmosphere of 5% CO2 in air at 37 ◦ C.
4. FITC-HA binding and uptake Cell monolayers grown on 35 mm culture dishes for 6 days were washed twice with 1.2 ml of phosphate buffered saline (PBS) supplemented with 5.5 mM d-glucose, and preincubated with PBS containing varying concentrations of d-glucose (5.5–27.8 mM), either alone or with phlorizin, Vitamin E (VitE), Vita-
Total RNA was extracted from LLC-PK1 cells maintained at 5.5 or 16.7 mM glucose with or without 0.25 mM phlorizin for 6 days using QIAamp RNA Mini kit (QIAGEN, Tokyo, Japan). Primers were synthesized chemically on the basis of sequences of GLUT-1(sense: 5 -TGTCGCTGTTCGTGGTGGAA3 , antisense: 5 -GACCAGGAGCACCGTGAAGA-3 , 357bps), SGLT-1(sense: 5 -CCGGTTGGAGCTTCTCTGTT-3 , antisense: 5 -AAGACCCCACCAGCATGATT-3 , 441bps) and SGLT-2 (sense: 5 -ATACTGGTCGTCCTGGCAAT-3 , antisense: 5 -GATGT TCACTATAGTCTTCC-3 , 317bps). RT-PCR was performed using a ReverTra Ace RT-PCR kit (TOYOBO, Osaka, Japan). Amplification was performed according to the following profile: denaturation at 95 ◦ C for 30 s, annealing at 58 ◦ C for 30 s, extension
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
at 72 ◦ C for 1 min, 25 cycles. The PCR products were separated by electrophoresis in 1.6% agarose gels, and amplified bands were detected by ethidium bromide staining. Data are presented as means ± S.E.M. The significance of difference was tested by either paired or unpaired t-tests. Differences were considered significant for P < 0.05.
219
5. Results A d-glucose concentration higher than 11.0 mM suppressed FITC-HA binding (Fig. 1A) and uptake (Fig. 1B) in LLC-PK1 cells, with maximal suppression obtained at 16.7 mM. The same concentration range of d-mannitol did not influence the FITC-HA uptake (Fig. 1B).
Fig. 1. (A) Influence of high glucose on FITC-HA binding in LLC-PK1 cells. After subcultivation in 5.5–27.8 mM d-glucose for 6 days, the cells were pulsed with FITC-HA (30 mg/l) at 4 ◦ C for 30 min. (B) Influence of high glucose and d-mannitol on FITC-HA uptake in LLC-PK1 cells. After subcultivation in varying concentrations of d-glucose or d-mannitol, the cells were pulsed with FITC-HA (30 mg/l) at 37 ◦ C for 60 min in the presence of the same concentrations of d-glucose or d-mannitol. Values are means ± S.E. of five dishes.
Fig. 2. (A) Concentration-dependent binding of FITC-HA in LLC-PK1 cells. After subcultivation in 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with varying concentrations of FITC-HA (10–500 mg/l) at 5.5 or 16.7 mM d-glucose at 4 ◦ C for 30 min. Values are means of four dishes. (B) Scatchard plot of the concentration-dependent binding of FITC-HA to LLC-PK1 cells at 5.5 mM (䊊) or 16.7 mM d-glucose (䊉).
220
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
Fig. 2A shows a comparison of the concentration dependency of FITC-HA binding in LLC-PK1 cells grown in 5.5 and 16.7 mM d-glucose. d-Glucose at 16.7 mM reduced the binding at any substrate concentration. A Scatchard analysis of the binding kinetics (Fig. 2B) showed that high glucose reduced the affinity (Km = 35.5 mg/l at 5.5 mM d-glucose versus Km = 52.6 mg/l at 16.7 mM d-glucose) and the maximum velocity (Bmax = 0.33 g/mg protein/30 min at 5.5 mM d-glucose versus Bmax = 0.27 g/mg protein/30 min at 16.7 mM d-glucose) of albumin binding. A comparison of the concentration dependency of FITC-HA uptake at 5.5 and 16.7 mM d-glucose showed that high glucose reduced FITC-HA uptake at any FITC-HA concentration (Fig. 3A). The kinetic analysis revealed that high glucose reduced the affinity (Km = 41.3 mg/l at 5.5 mM d-glucose versus 55.6 mg/l at 16.7 mM d-glucose) and the maximal velocity Umax = 4.4 g mg protein/60 min at 5.5 mM d-glucose versus Umax = 3.48 g/mg protein/60 min at 16.7 mM d-glucose) of uptake (Fig. 3B). The time course of FITC-HA binding (A) and uptake (B) at 5.5 and 16.7 mM d-glucose is compared in Fig. 4, and shows that high glucose significantly suppressed FITC-HA binding and uptake beyond 15 min.
As shown in Fig. 5, the inhibition of SGLT by phlorizin ameliorated the high glucose-induced inhibition of FITC-HA binding (Fig. 5A) and uptake (Fig. 5B) in a dose dependent manner. A phlorizin concentration greater than 0.1 mM reversed the high glucose-induced inhibition. The mRNA levels of GLUT-1, SGLT-1 and SGLT-2 in LLC-PK1 maintained at 5.5 mM glucose or 16.7 mM glucose with or without 0.25 mM phlorizin were determined by PCR amplification (Fig. 6). High glucose decreased mRNA of GLUT-1 and SGLT-1 by 30%, but did not influence that of SGLT-2. The addition of phlorizin did not influence the mRNA levels of GLUT-1 and SGLTs in cells maintained at 16.7 mM glucose. The influence of various antioxidants alone or in combination on the high glucose-induced inhibition of FITC-HA binding (Fig. 7A) and uptake (Fig. 7B) was examined. VitC, VitE or reduced glutathione alone or couple of any two drugs were unable to restore the suppression of FITC-HA binding and uptake by 16.7 mM d-glucose, over a wide range of antioxidant concentrations (VitC, 10−7 –2.5 × 10−5 M; VitE, 10−7 –2.5 × 10−5 M; reduced glutathione, 0.1–2.5 mM). On the other hand, the simultaneous presence of the three antioxidants completely
Fig. 3. (A) Concentration-dependent uptake of FITC-HA in LLC-PK1 cells. After subcultivation in 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with varying concentrations of FITC-HA (10–500 mg/l) at 5.5 or 16.7 mM d-glucose at 37 ◦ C for 60 min. Values are means of four dishes. (B) Scatchard plot of the concentration-dependent uptake of FITC-HA in LLC-PK1 cells at 5.5 mM (䊊) or 16.7 mM d-glucose (䊉).
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
221
abolished the inhibition by high glucose (Fig. 7A and B). As shown in Fig. 8, N-acetyl-cysteine, taurine and aminoguanidine were unable to ameliorate the inhibition of FITC-HA uptake by high glucose. Neither of them influenced high glucose-induced reduction of FITC-HA binding at all (figure not shown).
6. Discussion
Fig. 4. Comparison of the time courses of FITC-HA binding (A) and uptake (B) in LLC-PK1 cells maintained at 5.5 and 16.7 mM d-glucose. After subcultivations at 5.5 or 16.7 mM d-glucose for 6 days, the cells were pulsed with FITC-HA (30 mg/l) at 5.5 mM (䊊) or 16.7 mM (䊉) d-glucose at 4 ◦ C for binding and at 37 ◦ C for uptake for 3–60 min. Values were means ± S.E. of four dishes. ∗ P < 0.05, ∗∗ P < 0.01, ∗∗∗ P < 0.001 vs. 5.5 mM glucose.
To our knowledge, this study is the first to demonstrate that ambient high glucose reduces albumin binding and uptake in proximal tubular epithelial cells in vitro, and that the inhibition by phlorizin of excess glucose entry via SGLT, or a combination of antioxidants, completely reverses the reduced albumin binding and uptake. High glucose caused a dose-dependent reduction in FITC-HA binding and uptake (Fig. 1). Since the same concentration of d-mannitol did not influence FITC-HA uptake at all, the reduced FITC-HA binding and uptake caused by high glucose is not due to increased osmosis. When FITC-HA binding and uptake were analyzed by Scatchard plot, it was linear as found in previous many reports using cultured proximal tubular cells [11,13,14]. The Km values for albumin binding and
Fig. 5. Influence of phlorizin on high glucose-induced suppression of FITC-HA binding (A) and uptake (B) in LLC-PK1 cells. After subcultivation with 16.7 mM d-glucose and phlorizin (0.1–1.0 mM) for 6 days, the cells were pulsed with FITC-HA (30 mg/l) in the presence of 16.7 mM d-glucose and phlorizin (0.1–1.0 mM) at 4 ◦ C for 30 min for binding and at 37 ◦ C for 60 min for uptake. Values were means ± S.E. of five dishes.
222
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
Fig. 6. Comparison of mRNA levels of GLUT-1, SGLT-1 and SGLT-2 by PCR amplification in LLC-PK1 cultured at 5.5 mM glucose or 16.7 mM glucose with or without 0.25 mM phlorizin. GLUT-1: lane 1—at glucose 5.5 mM; lane 2—at glucose 16.7 mM; lane 3—at glucose 16.7 mM ± 0.25 mM phlorizin. SGLT-1: lane 4—at glucose 5.5 mM; lane 5—at glucose 16.7 mM; lane 6—at glucose 16.7 mM ± 0.25 mM phlorizin. SGLT-2: lane 7—at glucose 5.5 mM; lane 8—at glucose 16.7 mM; lane 9—at glucose 16.7 mM ± 0.25 mM phlorizin.
uptake at 5.5 mM d-glucose seemed to be similar to the albumin concentration in the tubular fluid that has been determined in animals [15]. On the other hand, the Km values at 16.7 mM d-glucose increased by 48.2 and 34.6%, respectively. Furthermore, high glucose reduced Bmax (binding) and Umax (uptake). The compromised sensitivity and velocity of albumin handling induced by high glucose in proximal tubular cells, which are exposed to excess albumin from the glomerulus, result in microalbuminuria. These findings indicate that in diabetes, where hyperglycemia and glycosuria increase glucose flux into proximal tubular cells via GLUT-1 and SGLTs, the proximal tubules will reabsorb albumin less effectively. The present study revealed that in LLC-PK1, high glucose down-regulates GLUT-1 and SGLT-1 at transcription level, but leaves SGLT-2 unaffected as reported previously [16,17]. Phlorizin, a SGLT inhibitor, did not influence SGLTs in cells maintained by high glucose. In our study, phlorizin (0.25–1.0 mM) reversed the reduction in FITC-HA binding and uptake by high glucose, and T-1095, a derivative of phlorizin, has been shown to prevent microalbuminuria and tubular morphological changes without influencing glomerular architecture in STZ-diabetic rats [18]. This indicates that an excess glucose flux mainly through SGLT-2 in diabetes plays a pivotal role in the development of impaired tubular reabsorption of albumin and microalbuminuria.
Fig. 7. Influences of various antioxidants alone or in combination on suppression of FITC-HA binding (A) and uptake (B) by high glucose in LLC-PK1 cells. After subcultivation for 6 days with 5.5 or 16.7 mM d-glucose and various antioxidants, the cells were pulsed with FITC-HA (30 mg/l) in the presence of the same concentration of d-glucose and antioxidants at 4 ◦ C for 30 min for binding and at 37 ◦ C for 60 min for uptake. Values are means±S.E. of five dishes. Eth, ethanol; VitE, 10−6 M Vitamin E; VitC, 10−6 M Vitamin C; Gluta, 0.25 mM reduced glutathione.
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
223
Fig. 8. Effects of various inhibitors of AGE on high glucose-induced inhibition of FITC-HA uptake in LLC-PK1 cells. After subcultivation in 16.7 mM glucose and varying concentrations of either N-acetyl-cysteine, taurine or aminoguanidine for 6 days, the cells were pulsed with FITC-HA (30 mg/l) in the presence of 16.7 mM d-glucose and the same concentrations of the AGE inhibitors. Values are means ± S.E. of five dishes.
The process of albumin binding and uptake in proximal tubular cells is still not fully understood. Although an albumin-binding protein in proximal tubular cells has yet to be identified unequivocally, megalin [19], cubilin [20] and a 32 kDa albumin-binding protein reported by us [21] are all possible candidates. As a first step, it is likely that albumin binds to the putative albumin-binding protein at the apical membrane. Because inhibition of SGLT by phlorizin reversed the high-glucose-induced suppression of binding, it appears that excessive glucose entry through SGLT may disturb albumin binding to the putative albumin-binding protein. A comparison of the time courses of FITC-HA binding and uptake at 5.5 and 16.7 mM d-glucose showed that high glucose suppressed binding as well as uptake beyond 15 min (Fig. 4). Since endosomal and lysosomal acidification has been considered to be a rate-limiting step of albumin uptake [22,23], the reduction in albumin uptake by high glucose beyond 15 min might occur because the high glucose induces insufficient acidification of endosomes and lysosomes. This may occur because of reduced activity of proton-translocating adenosine triphosphatase, one of the key factors in acidification [24], since it has been shown that high glucose lowers the ATP level in cultured proximal tubular cells [25]. However, it remains unclear whether the insufficient acidifica-
tion of late endosome and lysosome impairs albumin binding. The reduced number of recycled putative albumin binding protein on apical membrane caused by impaired dissociation between albumin molecule and albumin binding protein due to the insufficient acidification in endosome might be one of mechanisms, because endosomal alkalinisation by bafilomycin A1 reduces albumin binding in OK cells [26]. The etiological role of oxidative stress in the development of diabetic tubular lesions has been postulated [27,28]. Hence, the influence of antioxidants on the high glucose-induced inhibition of FITC-HA binding and uptake in LLC-PK1 cells was examined. VitC, VitE or reduced glutathione alone or a combination of any two drugs was unable to ameliorate inhibition of FITC-HA binding and uptake by high glucose, but simultaneous presence of three antioxidants completely reversed the inhibition. Although previous investigations have shown the antioxidant action of VitE alone against high glucose-induced oxidative stress in vivo and in vitro [29], recent studies have failed to verify the counteraction by VitE alone against the oxidative stress [30]. Because VitE acts as a pro-oxidant [31], especially in hyperglycemia, the simultaneous presence of another antioxidant, such as VitC and reduced glutathione, is required to overcome the oxidative stress [32]. However, it was beyond the scope of this study to make clear the mechanisms through which
224
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225
the simultaneous presence of three drugs overcomes the oxidative stress due to high glucose. The AGE inhibitors (N-acetyl-cysteine, taurine and aminoguanidine) have been reported to prevent high glucose-induced morphological and functional changes in the kidney [30,33]. In the present study, none of these AGE inhibitors ameliorated the high glucose-induced reduction in FITC-HA binding and uptake in LLC-PK1 cells. This may be because exposure to high glucose for 6 days might be insufficient for AGE induction. In conclusion, ambient high glucose suppressed albumin binding and uptake in cultured proximal tubular epithelial cells. Excess glucose entry via SGLT and the resultant metabolic oxidative stress play a role in high glucose-induced impaired albumin reabsorption in proximal tubular epithelial cells.
References [1] H.H. Parving, B. Oxenball, P.A. Svendsen, J.S. Christensen, A.R. Anderson, Early detection of patients at risk of developing diabetic nephropathy: a longitudinal study of urinary albumin excretion, Acta Endoclinol. (Copenh.) 100 (1982) 550–555. [2] A. Schmitz, Microalbuminuria and mortality in non-insulindependent diabetes, in: C.E. Mogensen (Ed.), The Kidney and Hypertension, Martinus Nijhoff Publishing, Boston, 1988, pp. 65–70. [3] F. Ishibashi, Microalbuminuria in NIDDM is caused by increased excretion of unmodified albumin, Diabetes 45 (1996) 731–736. [4] S.C. Jones, H.J. Saunders, C.A. Pollock, High glucose increases growth and collagen synthesis in cultured tubulointerstitial cells, Diabet. Med. 16 (1999) 932–938. [5] R.C. Harris, B.M. Brenner, J.L. Seifter, Sodium-hydrogen exchange and glucose transport in renal microvillus vesicles from rats with diabetes mellitus, J. Clin. Invest. 77 (1986) 724–733. [6] J.L. Taft, C.J. Nolan, S.P. Yeung, T.D. Hewitson, F.I. Martin, Clinical and histological correlations of decline in renal function in diabetic patients with proteinuria, Diabetes 43 (1994) 1046–1051. [7] K.A. Nath, Tubulointerstitial changes as a major determinant in the progression of renal damage, Am. J. Kidney Dis. 20 (1992) 1–17. [8] T. Kislinger, C. Fu, B. Huber, et al., N-(Carboxymethyl)lysine adducts of proteins are ligands for receptor for advanced glycation end products that activate cell signalling pathways and modulate gene expression, J. Biol. Chem. 274 (1999) 31740–31749.
[9] V. Scivittaro, M.B. Ganz, M.F. Weiss, AGE induce oxidative stress and activate protein kinase C-beta (II) in neonatal mesangial cells, Am. J. Physiol. 278 (2000) F676–F683. [10] Y. Takakura, T. Morita, M. Fujikawa, et al., Characterization of LLC-PK1 kidney epithelial cells as an in vitro model for studying renal tubular reabsorption of protein drugs, Pharm. Res. 12 (1995) 1968–1972. [11] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Functional characterization of albumin binding to the apical membrane of OK cells, Am. J. Physiol. 271 (1996) F286– F291. [12] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Long-term protein exposure reduces albumin binding and uptake in proximal tubule-derived opossum kidney cells, J. Am. Soc. Nephrol. 9 (1998) 960–968. [13] V. Batuman, S. Guan, Receptor-mediated endocytosis of immunoglobulin light chains by renal proximal tubule cells, Am. J. Physiol. 272 (1997) F521–F530. [14] Y. Sasaki, J. Nagai, Y. Kitahara, N. Takai, T. Murakami, M. Takano, Expression of chloride channel, ClC-5, and its role in receptor-mediated endocytosis of albumin in OK cells, Biochem. Biophy. Res. Comm. 282 (2001) 212–218. [15] P.D. Leber, D.J. Marsh, Micropuncture study of concentration and fate of albumin in rat nephron, Am. J. Physiol. 219 (1970) 358–363. [16] C.J. Clancey, J.E. Lever, Differential regulation of three glucose transporter genes in a renal epithelial cell line, J. Cell. Physiol. 185 (2000) 244–252. [17] T. Ohta, K.J. Isselbacher, D.B. Rhoads, Regulation of glucose transporters in LLC-PK1 cells: Effects of d-glucose and monosaccharides, Molec. Cell. Biol. 10 (1990) 6491– 6499. [18] T. Adachi, K. Yasuda, Y. Okamoto, et al., T-1095, a renal Na+ - glucose transporter inhibitor, improves hyperglycemia in streptozotocin-induced diabetic rats, Metabolism 49 (2000) 990–995. [19] E.I. Christensen, H. Birn, P. Verroust, S.K. Moestrup, Megalin- mediated endocytosis in renal proximal tubule, Renal Failure 20 (1998) 191–199. [20] H. Birn, J.C. Fyfe, C. Jacobsen, et al., Cubilin is an albumin binding protein important for renal tubular albumin reabsorption, J. Clin. Invest. 105 (2000) 1353–1361. [21] F. Ishibashi, Characterization of novel human proximal tubular cell albumin binding protein (hABP), Diabetes 51 (suppl. 2) (2002) A183. [22] I. Mellman, R. Fuchs, A. Helenius, Acidification of the endocytic and exocytic pathways, Ann. Rev. Biochem. 55 (1986) 663–700. [23] G. Rudnick, ATP-driven H+ pumping into intracellular organelles, Ann. Rev. Physiol. 48 (1986) 403–413. [24] N. Nelson, W.R. Harvey, Vacuolar and plasma membrane proton-adenosine triphosphatase, Physiol. Rev. 79 (1999) 361–385. [25] M. Symonian, M. Smogorzewski, W. Marcinkowski, E. Krol, S.G. Massry, Mechanisms through which high glucose concentration raises [Ca2+ ]i in renal proximal tubular cells, Kidney Int. 54 (1998) 1206–1213.
F. Ishibashi / Diabetes Research and Clinical Practice 65 (2004) 217–225 [26] M. Gekle, S. Mildenberger, R. Freudinger, S. Silbernagl, Endosomal alkalinization reduces Jmax and Km of albumin receptor-mediated endocytosis, Am. J. Physiol. 268 (1995) F899–F906. [27] M. Morcos, A.A.R. Sayed, A. Bierhaus, et al., Activation of tubular epithelial cells in diabetic nephropathy, Diabetes 51 (2002) 3532–3544. [28] H. Ha, K.H. Kim, Pathogenesis of diabetic nephropathy: the role of oxidative stress and protein kinase C, Diabet. Res. Clin. Pract. 45 (1999) 147–151. [29] H. Trachtman, Vitamin E prevents glucose-induced lipid peroxidation and increased collagen production in cultured rat mesangial cells, Microvasc. Res. 47 (1994) 232– 239.
225
[30] H. Trachtman, S. Futterweit, J. Maesaka, Taurine ameliorates chronic streptozotocin-induced diabetic nephropathy in rats, Am. J. Physiol. 269 (1995) F429–F438. [31] V.W. Bowry, K.U. Ingold, R. Stocker, Vitamin E in human low-density lipoprotein: when and how this antioxidant becomes a pro-oxidant, Biochem. J. 288 (1992) 341–344. [32] A.J. Sinchair, A.J. Girling, L. Gray, C. LeGuen, J. Lunec, A.H. Barnett, Disturbed handling of ascorbic acid in diabetic patients with and without microangiopathy during high dose ascorbate supplementation, Diabetologia 34 (1991) 171–175. [33] P. Odetti, C. Pesce, N. Traverso, et al., Comparative trial of N-acetyl-cysteine, taurine, and oxerutin on skin and kidney damage in long-term experimental diabetes, Diabetes 52 (2003) 499–505.