- Email: [email protected]
Sodium bicarbonate treatment reduces renal injury, renal production of transforming growth Factor-β, and urinary transforming growth Factor-β excretion in rats with doxorubicin-induced nephropathy
Sodium Bicarbonate Treatment Reduces Renal Injury, Renal Production of Transforming Growth Factor-, and Urinary Transforming Growth Factor- Excretion in Rats With Doxorubicin-Induced Nephropathy Edmara A. Baroni, PhD, Roberto S. Costa, MD, Rildo Volpini, BS, and Terezila M. Coimbra, MD ● The aim of this study is to investigate the effect of sodium bicarbonate on doxorubicin-injected rats. Thirty female Wistar rats were injected with doxorubicin (3.5 mg/kg of body weight, intravenously) and 30 rats with 0.15 mol/L of sodium chloride solution (group C). Fifteen days later, we replaced the drinking water with a 0.15-mol/L sodium bicarbonate solution for 10 of the animals injected with doxorubicin (group AD-B). Three months after the beginning of treatment, urine samples were collected to quantify albumin, creatinine, and transforming growth factor- (TGF-). The rats were killed, and the kidneys were removed for histological, morphometric, immunohistochemical, and RNA studies. All doxorubicin-injected animals showed structural renal changes. However, these alterations were less intense in rats treated with doxorubicin plus sodium bicarbonate (P F 0.05). The percentage of glomerulosclerosis was 0.11% ⴞ 0.08% in group C, 14.7% ⴞ 12.8% in group AD (rats treated with doxorubicin only), and 4.38% ⴞ 1.9% in group AD-B, and the percentage of tubulointerstitial damage was 0.01% ⴞ 0.03% in group C, 54.6% ⴞ 20.3% in group AD, and 16.6% ⴞ 10.3% in group AD-B. The immunostaining for TGF- in the renal cortex and glomeruli was more intense in the animals injected with doxorubicin only. A greater renal cortical TGF- messenger RNA content was observed in the animals injected with only doxorubicin that did not receive sodium bicarbonate (P F 0.05). These animals also presented a greater rate of urinary TGF- excretion reported as picograms of TGF- per milligram of urinary creatinine (P F 0.05), which was 202 ⴞ 11 pg/mg in group C, 1, 103 ⴞ 580 pg/mg in group AD, and 299 ⴞ 128 pg/mg in group AD-B. However, albuminuria was more intense in the sodium bicarbonate–treated animals (P F 0.05). The animals from group AD also showed higher immunostaining scores for vimentin and albumin in tubule cells (P F 0.05). In conclusion, treatment with sodium bicarbonate reduces structural renal damage, albumin reabsorption, and renal TGF- production in rats with doxorubicin-induced nephropathy. 娀 1999 by the National Kidney Foundation, Inc. INDEX WORDS: Doxorubicin nephropathy; urine alkalinization; transforming growth factor (TGF)-; tubulointerstitial damage.
A
DRIAMYCIN (doxorubicin; Farmitalia, Carlo Erba RJ, Brazil) injection induces marked and persistent proteinuria and glomerular structural changes in rats.1 Glomerular epithelial cell lesions that result in loss of the foot processes are observed in the kidneys of these animals. These alterations progress to glomeruloFrom the Departments of Physiology and Pathology, Medical School of Ribeira˜o Preto, USP, Ribeira˜o Preto, SP, Brazil. Received July 31, 1998; accepted in revised form March 5, 1999. Presented in part at the 31st Annual Meeting of American Society of Nephrology, Philadelphia, PA, October 25-28, 1998. Supported in part by Fundaca˜o de Amparo a` Pesquı´sa do Estado de sa˜d Paulo, Brazil. R.S.C. and T.M.C. are recipients of Conselho Nacional de Desenvolvimento Cientı´ficoe Technolo´gico, DF, Brazil, fellowships. Address reprint requests to Terezila M. Coimbra, MD, Departamento de Fisiologia, Faculdade de Medicina de Ribeira˜o Preto, USP, 14049-900-Ribeira˜o Preto-SP, Brazil. E-mail: [email protected]
娀 1999 by the National Kidney Foundation, Inc. 0272-6386/99/3402-0020$3.00/0 328
sclerosis and tubulointerstitial fibrosis.1-4 Transforming growth factor- (TGF-) has been considered the principal cytokine involved in the pathogenesis of renal fibrosis.5-10 Some studies have shown increased renal TGF- production in rats with nephropathy induced by doxorubicin.11,12 Treatment with sodium bicarbonate provokes urine alkalinization and extracellular volume expansion that may reduce protein reabsorption by tubular cells.13,14 Tubular reabsorption of protein depends on the electrostatic interaction between positively charged groups of protein and anionic sites on the luminal membrane or membrane coat.15 The modifications in protein charges caused by the change in urinary pH must interfere with protein binding to the luminal membrane, which is the first step of endocytosis. Changes in protein concentrations at the endocytic sites because of extracellular volume expansion also influence protein reabsorption.16 Some evidence suggests a role of proteinuria and renal handling of proteins in the development of tubu-
American Journal of Kidney Diseases, Vol 34, No 2 (August), 1999: pp 328-337
NaHCO3 TREATMENT AND RENAL INJURY
lointerstitial damage.17,18 It has been suggested that proteins filtered by glomeruli could be toxic for the tubule cells.19,20 Studies with tubule cell cultures have shown that some proteins, such as immunoglobulin G (IgG) and albumin, induce increased production of inflammatory and vasoactive factors.21,22 This observation can explain why greater proteinuria is frequently associated with more severe tubular damage.23 Urine alkalinization can also decrease the precipitation of proteins at the distal tubule level, another factor that may contribute to the nephrotoxicity of the filtered proteins. It was observed in vitro that when pH decreased from 6.0 to 4.5, some proteins, such as albumin, coprecipitate with Tamm-Horsfall mucoprotein.24 The aim of this study is to investigate the effect of sodium bicarbonate on renal injury and renal TGF- production and excretion in rats with nephropathy induced by doxorubicin. We also analyzed the relationship between the rate of urinary TGF- excretion and structural renal changes observed in these animals. MATERIALS AND METHODS
Animals and Experimental Protocols Sixty female Wistar rats weighing 180 to 220 g were used. Thirty rats were injected with a single dose of doxorubicin, 3.5 mg/kg of body weight, through the femoral vein, and 30 with 0.15 mol/L of sodium chloride solution (control group). Fifteen days after the injections, we selected 23 of the animals injected with doxorubicin with albuminuria of 4.0 to 38.0 mg/24 h of albumin, and the experimental and control animals were divided into four groups that underwent different treatments. The control animals (group C) and doxorubicin-injected animals (group AD) did not receive any treatment. For the animals from groups C-B (control) and AD-B (doxorubicin injected), we replaced the drinking water with a 0.15 mol/L of sodium bicarbonate solution. There was no difference in mean urine albumin levels between the different groups at 15 days. All animals had free acess to normoproteic (18%) chow. The average amount of food consumed by the groups was determined on alternate days after the treatment was started using the following formula: IMG ⫽ (Ro ⫺ Ri)/N where: IMG ⫽ average amount of food consumed by the group, Ro ⫽ total amount of chow offered to the group, Ri ⫽ total amount of chow not consumed by the group, and N ⫽ number of animals in the group. This amount was not modified by the different kinds of treatment. After treatment with sodium bicarbonate, the rats showed a much higher urinary pH (8 to 9) than before injection (6.0 to 7.0). Monthly urine samples were collected to determine albumin excretion and urinary volume. Systolic blood pressure was
329
measured by the tail-cuff method 75 days after the injection. The animals were killed with excess anesthesia 105 days after the injection. The organs were perfused with phosphatebuffered saline (PBS) solution (0.15 mol/L of sodium chloride, 0.01 mol/L of sodium phosphate buffer, pH 7.2), and the kidneys were removed for messenger RNA (mRNA), immunohistochemical, histological, and morphometric studies.
Renal Function Studies Urinary and plasma albumin were quantified by electroimmunoassay using a specific antibody against rat albumin.25 Glomerular filtration rate (GFR) was measured by inulin clearance26 in five control animals, six doxorubicin-treated animals, and six animals treated with doxorubicin and sodium bicarbonate at 105 days after the beginning of treatment. The animals were anesthetized intraperitoneally with 50 mg/kg of sodium thionembutal. After tracheostomy, the femoral artery and vein were cannulated to collect blood samples and inject fluids, and both ureters were cannulated to collect urine. The animals received a priming inulin dose of 12 mg/100 g, followed by a maintenance dose of 30 mg/100 g/h. After stabilization for approximately 60 minutes, urine was collected for 60 minutes and blood sampled at 30 and 60 minutes. Plasma and urine inulin was measured by the method of Fu¨hr et al,26 and creatinine was measured by the Jaffe´ reaction.27
Antibodies The primary antibodies used were: (1) a purified IgG fraction of polyclonal rabbit antibody against TGF-1 (Sta Cruz Biotechnology, Sta Cruz, CA) specific for TGF-1 that shows no cross-reactivity with TGF-2 or TGF-3 and is reactive for mouse, rat, and human TGF-1; (2) a purified IgG fraction of polyclonal rabbit antirat fibronectin (Chemicon; International INC, Temecula, CA); (3) a monoclonal mouse antivimentin antibody (Dako Corporation, Glostrup, Denmark), and (4) a polyclonal antirat albumin antibody produced in our laboratory.14
Immunohistochemical Studies Eleven doxorubicin-injected animals and five control animals were killed 105 days after injection. The rats were submitted to left intraventricular perfusion with PBS (0.15 mol/L of sodium chloride, 0.01 mol/L phosphate buffer, pH 7.2) until the kidneys were blanched. The kidneys were then perfused with 4% formalin for 2 hours, postfixed in Bouin’s solution for an additional 4 to 6 hours, rinsed with 70% ethanol to eliminate picric acid, dehydrated through a graded series of ethanol solutions, embedded in Paraplast (improved paraffin embedding media; Sigma, St. Louis, MO), sectioned into 3-µm slices, deparaffinized, and subjected to immunohistochemical staining.28 The sections were incubated overnight at 4°C with a 1/30 anti–TGF-1 polyclonal antibody or 1/800 antirat fibronectin antibody or for 1 hour at room temperature with a 1/1,500 monoclonal mouse antivimentin or 1/5,000 antirat albumin antibody. The reaction product was detected with an avidinbiotin-peroxidase complex (Vector Laboratories, Burlingame, CA). The color reaction was developed with 3,
330
3-diamino-benzidine (Sigma Chemical Company, St Louis, MO), and the material was counterstained with Harris hematoxylin or methyl green, dehydrated, and mounted. Nonspecific protein binding was blocked by incubation with 5% goat serum in PBS for 30 minutes. Controls for TGF- antibody consisted of replacing anti–TGF-1 antibody with normal rabbit IgG or anti–TGF-1 antibody preincubated with cellulose-bound human TGF-1 protein (Boehringer Mannheim Biochemica, Mannheim, Germany). For the other antibodies, the controls consisted of replacing the primary antibody with normal rabbit (polyclonal antibodies) or mouse IgG (monoclonal antibody). For evaluation of the immunoperoxidase stains for TGF-, fibronectin, vimentin, and albumin, each glomerular area or tubulointerstitial grid field was graded semiquantitatively, and the mean score per biopsy was calculated. Each score reflects mainly changes in the extent, rather than intensity, of staining and depended on the percentage of glomerular tuft area or grid field showing positive staining: 0, absent or less than 5%; I, 5% to 25%; II, 25% to 50%; III, 50% to 75%; and IV, ⬎75%.29
BARONI ET AL
fragments were labeled to a specific activity of 1 ⫻ 109 disintegrations/min/µg using the random primer method.35 Autoradiograms from these blots were quantified by optic densitometry. The densitometric ratio between TGF-1 mRNA signals and 28 S was calculated, and data are expressed in comparison to control, designated as 100%.
Quantification of TGF- in Urine Samples The ureters from 12 control rats and 16 doxorubicininjected rats were cannulated, and urine samples were collected over a period of 30 minutes for TGF- and creatinine determination. The samples were immediatly treated with 1 mmol/L of phenylmethylsulphonyl fluoride (Sigma Chemical Company, St Louis, MO) and acidified to measure total TGF- (latent plus active). Quantification of TGF- in these samples was performed by enzyme-linked immunosorbent assay using kits from Promega (Promega Corporation, Madison, MO),36 and creatinine was determined by Jaffe´’s reaction.27 Urinary TGF- is reported as picograms of TGF- per milligram of urinary creatinine to correct the variation in urine concentration.
Light Microscopy The kidneys from 14 control animals and 15 doxorubicintreated rats killed 105 days after injection were fixed in alcoholic Bouin for 12 hours and processed for paraffin embedding. Three-micrometer histological sections were stained with hematoxylin-eosin and Masson’s trichrome and examined under the light microscope. The percentage of segmental glomerulosclerosis or tubulointerstitial damage was determined by morphometry with a light camera connected to an image analyzer (KS-300; Kontron Electronik, Munich, Germany). Forty subcapsular and 20 juxtaglomerular glomeruli or 20 fields from the cortex were measured in a section of each kidney.30 Encircled damaged areas were traced on a videoscreen and determined by computerized morphometry. Each glomerular or tubulointerstitial field was graded quantitatively, and the mean value was calculated for each biopsy. The glomerular and tubulointerstitial immunostain for TGF- in the renal cortex was also evaluated by morphometry.
Statistical Analysis Urine and plasma albumin levels, blood pressure, glomerular filtration rate, urinary volume, plasma creatinine level, renal cortical TGF- mRNA, percentage of glomerulosclerosis or tubulointerstitial damage, and the immunohistochemical scores obtained under control and experimental conditions were submitted to analysis of variance with multiple comparisons by the Tukey test. Data concerning urinary albumin excretion, percentage of glomerulosclerosis, and immunohistochemical scores for TGF- in the glomeruli and fibronectin in the cortical tubulointerstitium were log10 transformed before analysis, as required by the Bartlett test. Pearson’s correlation coefficient was calculated to determine r. P less than 0.05 is considered statistically significant.
RESULTS
RNA Analysis
Immunohistochemical and Light Microscopy Studies
Cortical renal tissue was obtained from 6 control animals and 16 doxorubicin-injected animals killed 105 days after injection. The samples were prepared by the guanidine/ phenol/chloroform technique described by Chomczynsky and Sacchi.31 The RNA obtained was subjected to electrophoresis in formaldehyde-agarose gels and transferred to nitrocellulose filters (Schleicher & Shuell Inc, Keene, NH). RNA was also analyzed by dot blot hybridization after loading onto a blotting apparatus under denaturing conditions.32 Thirty micrograms of total RNA was loaded for Northern analysis and 10, 5, and 2.5 µg were loaded for dot blot analysis. The plasmid with complementary DNA (cDNA) encoding a portion of human TGF-1 RNA was used to detect transcripts for TGF-1.33 A plasmid containing portions of rat 28 S ribosomal ribonucleic acid (RNA)34 was used to normalize the TGF-1 signal and to adjust for inequalities in RNA loading and/or transfer. The purified
Light microscopy studies showed glomeruli with global or segmental sclerosis in the kidneys from rats with doxorubicin-induced nephropathy (P ⬍ 0.05). The tubulointerstitial compartment showed tubular atrophy and/or dilatation with intraluminal casts, interstitial mononuclear cell infiltration, and fibrosis (P ⬍ 0.05). These lesions were less intense in the rats treated with sodium bicarbonate (P ⬍ 0.05; Table 1). The results of the immunohistochemical studies showed the presence of moderate immunostaining for TGF- in the cytoplasm of some tubular cells in the renal cortex from control rats. We did not observe any reaction in the glomeruli
NaHCO3 TREATMENT AND RENAL INJURY
331
Table 1. Blood Pressure, Plasma Albumin, and Renal Functional, Histological, and Immunohistochemical Data From Control, Doxorubicin Treated, and Doxurubicin Plus Sodium Bicarbonate Treated Rats Groups
C (n ⫽ 5)
AD (n ⫽ 6)
AD-B (n ⫽ 6)
Blood pressure (mm Hg) Urinary volume (mL/24 h) GFR (mL/min/100 g) Plasma creatinine (mg/dL) Plasma albumin (mg/mL) GS (%) TID (%) TGF--C (%) TGF--G Fib-CTI Fib-G Albumin Vimentin
106.4 ⫾ 7.6 5.54 ⫾ 1.61 0.76 ⫾ 0.07 0.59 ⫾ 0.03 29.49 ⫾ 6.26 0.11 ⫾ 0.08 0.01 ⫾ 0.03 19.60 ⫾ 3.65 1.00 ⫾ 1.00 0.99 ⫾ 0.06 1.68 ⫾ 0.18 0.02 ⫾ 0.03 0.00 ⫾ 0.00
102.9 ⫾ 9.9 17.80 ⫾ 11.32* 0.53 ⫾ 0.19* 0.76 ⫾ 0.28 15.78 ⫾ 5.23* 14.70 ⫾ 12.80* 54.60 ⫾ 20.30* 44.17 ⫾ 4.90* 13.00 ⫾ 4.8* 1.49 ⫾ 0.24* 2.95 ⫾ 0.38* 1.30 ⫾ 0.31* 1.28 ⫾ 0.32*
101.5 ⫾ 4.6 44.55 ⫾ 17.95*† 0.62 ⫾ 0.09 0.67 ⫾ 0.17 14.76 ⫾ 3.19* 4.38 ⫾ 1.90*† 16.70 ⫾ 10.30† 44.60 ⫾ 9.13* 6.60 ⫾ 2.70* 1.14 ⫾ 0.02*† 2.28 ⫾ 0.32† 0.30 ⫾ 0.10† 0.54 ⫾ 0.34*†
Abbreviations: GFR, glomerular filtration rate; GS, glomeruli with glomerulosclerosis; TID, tubulointerstitial damage; TGF--C, immunostaining for TGF- in the renal cortex; TGF--G, immunostaining scores for TGF- in the glomeruli; Fib-CTI, immunostaining scores for fibronectin the cortical tubulointerstitium; Fib-G, immunostaining scores for fibronectin in the glomeruli; C, control rats; AD, rats treated with doxorubicin; AD-B, rats treated with doxorubicin plus sodium bicarbonate. *P ⬍ 0.05 v control. †P ⬍ 0.05 v AD.
of these animals (Fig 1A). The staining intensity was greater in the kidneys of the rats treated only with doxorubicin (Fig 1B). In these animals, we observed a marked immunohistochemical reaction for TGF- in the tubular cells and glomeruli localized primarily in the podocytes, whereas in the interstitial cells, the reaction was less intense (Fig 1B). The same pattern of reaction was verified in the glomeruli of rats treated with
Fig 1. Immunolocalization of TGF-1 in the renal cortex of (A) a control rat and (B) a rat treated with doxorubicin or (D) doxorubicin plus sodium bicarbonate 105 days after injection using a TGF-1 polyclonal antibody. Note the intense staining present (B) in podocytes and (A, B, D) in the cytoplasm of some tubule cells. (C) Immunolocalization of TGF-1 in the renal cortex of a control rat using a TGF-1 antibody preincubated with TGF- protein. Staining was abolished by this treatment. (Original magnification ⴛ2.)
sodium bicarbonate, although at lower intensity (Fig 1D). The intensity of tubulointerstitial reaction for TGF- was also less in the animals treated with sodium bicarbonate than in the rats treated only with doxorubicin. However, we did not find any difference in the immunostaining score for TGF- between these groups because these scores reflect mainly changes in immunostaining extent, rather than intensity (Table 1).
332
No reaction was observed in the cortex from these animals when the TGF-1 antibody was replaced with normal rabbit IgG or with a TGF- antibody previously absorbed with a TGF-1 protein (Fig 1C), thus showing the specificity of the reaction. The immunohistochemical analysis also showed that the animals treated only with doxorubicin had greater fibronectin expression in the glo-
BARONI ET AL
meruli and cortical tubulointerstitium and greater immunostaining scores for vimentin and albumin in the tubular cells from the renal cortex than the animals treated with doxorubicin plus sodium bicarbonate (Fig 2). The increase of vimentin expression in the cytoplasm of tubular cells was associated with an albumin overload in these cells. Because the expression of vimentin is an indicator of tubular cell regeneration suggestive of
Fig 2. Immunolocalization of (A, B) fibronectin, (C, D) vimentin, and (E, F) albumin in the renal cortex of a rat treated with (B, D, F) doxorubicin or (A, C, E) doxorubicin plus sodium bicarbonate 105 days after injection. Note the immunostaining for fibronectin in glomeruli and cortical tubulointerstitium is more intense in B and the immunostaining for albumin and vimentin in tubular cells is more intense in D and F. (Original magnification; [A, B, C, D] ⴛ120; [E, F] ⴛ280.)
NaHCO3 TREATMENT AND RENAL INJURY
333
recent injury,37 this damage may be related to an increase in tubular albumin reabsorption. mRNA Analysis Dot blot analysis showed a marked increase in TGF- mRNA content in the renal cortex from rats injected with doxorubicin (Fig 3) compared with control animals (P ⬍ 0.05). The renal cortex levels of TGF- mRNA in rats treated with doxorubicin plus sodium bicarbonate were similar to those observed in control animals. There was no difference in renal cortex levels of TGF- mRNA between the two groups of control animals (Fig 3). Northern blot analysis was also performed to check the specificity of the TGF- probe, and the results essentially confirmed those obtained by dot blot analysis (Fig 4).
Fig 4. Autoradiographs of the Northern blot analysis of renal cortical RNA from a control rat (C) and from rats killed 105 days after doxorubicin injection (AD). RNA was hybridized with a TGF-1 cDNA probe (first line). The line at the bottom of the figure corresponds to the same blot hybridized with the 28 S cDNA probe.
Rate of Urinary TGF- Excretion The rate of urinary TGF- excretion was greater in animals treated with doxorubicin than control rats (P ⬍ 0.05; Fig 5). However, treatment with sodium bicarbonate prevented this increase (P ⬍ 0.05). We observed a significant positive correlation between the rate of urinary TGF- excretion and incidence of glomerular sclerosis (r ⫽ 0.74; P ⫽ 0.014) and tubulointerstitial damage (r ⫽ 0.76; P ⫽ 0.029) in rats injected with doxorubicin.
Fig 3. Quantification of renal cortical TGF-1 RNA in doxorubicin-injected rats (AD) and control rats (C) by dot blot hybridization with a cDNA TGF-1 probe. A densitometric ratio between the TGF- mRNA area and the 28 S area was calculated, and data are expressed in comparison to the control (C), with the mean ⴞ SD control value designated as 100%. Autoradiograms of the individual results shown at the top of the graphs represent the TGF- mRNA data from a control rat (C), a control rat treated with sodium bicarbonate (C-B), a rat treated with doxorubicin (AD), and a rat treated with doxorubicin plus sodium bicarbonate (AD-B). The line below corresponds to the same blots in the same assay, hybridized with the 28 S cDNA probe.
Fig 5. Rate of urinary TGF- excretion for control rats (C) and for rats treated with doxorubicin (AD) or doxorubicin plus sodium bicarbonate (AD-B) at 105 days after doxorubicin injection.
334
BARONI ET AL
Fig 6. Urinary albumin excretion (UAE) in rats injected with sodium chloride solution and treated with sodium bicarbonate (C-B) or untreated (C) and in rats injected with doxorubicin and treated with sodium bicarbonate (AD-B) or untreated (AD).
Urinary and Plasma Albumin and Blood Pressure Urinary albumin excretion showed a significant and progressive increase in both groups of animals injected with doxorubicin compared with the respective controls (P ⬍ 0.05; Fig 6). However, these increases were more intense in the animals treated with sodium bicarbonate (P ⬍ 0.05). Plasma albumin levels were less in all doxorubicin-treated animals compared with controls (P ⬍ 0.01). However, there was no difference between the two groups of rats treated with doxorubicin (Table 1). Blood pressure was not modified by the different kinds of treatment (Table 1). Renal Function We did not observe any difference in plasma creatinine levels between the animals from different groups (Table 1). However, GFR values (mean ⫾ SD) were less in the animals treated only with doxorubicin than in controls (Table 1), probably because inulin clearance is a more precise method for evaluation of renal function. Treatment with sodium bicarbonate prevented the decrease in GFR observed in animals treated with doxorubicin. DISCUSSION
Our data show that all animals treated with doxorubicin (groups AD and AD-B) presented
glomerulosclerosis and tubulointerstititial damage. However, these disturbances were less intense in the animals treated with sodium bicarbonate. We observed increased immunohistochemical TGF-1 staining in the cytoplasm of some tubular cells of the kidney from doxorubicin-injected rats, whereas the reaction was less intense in animals treated with sodium bicarbonate. We also found immunoreactivity for TGF-1 in the glomeruli (podocytes) of these animals. The increased immunoreaction intensity for TGF- protein in the renal cortex from these animals was associated with the enhancement of renal cortex TGF- mRNA content, suggesting an increased production of this polypeptide by the kidney. This enhancement was prevented by treatment with sodium bicarbonate. The immunostaining score for fibronectin in the glomeruli and cortical tubulointerstitium was also greater in the animals treated only with doxorubicin. An increase in TGF- mRNA levels and TGF- protein in doxorubicin-treated rats has been previously observed in recent studies11,12 and was related to the structural renal changes observed in these animals. TGF- has been considered one of the major cytokines involved in the regulation of extracellular matrix synthesis and degradation. It was also verified that the introduction of a plasmid containing cDNA for TGF- through the renal artery induced glomerulosclerosis.7 The present immunohistochemical studies showed the presence of an immune reaction for TGF-1 in the glomeruli from doxorubicintreated rats primarily localized in the podocytes. It has been shown that podocytes can produce various cytokines, including TGF-.38,39 Although the role of TGF- in podocyte injury has not been determined, extracellular matrix exposure by glomerular basement membrane denudation provoked by podocyte injury could result in adhesions between glomerular tuft and Bowman’s capsule and segmental sclerotic lesions. There is evidence that renal handling of proteins present in animals with nephropathy induced by doxorubicin may contribute to increased renal TGF- production and development of tubulointerstitial damage.17,18 A low-protein diet attenuates the proteinuria and progression of tubulointerstitial lesions and reduces renal TGF- production in rats injected with doxorubicin.40-44
NaHCO3 TREATMENT AND RENAL INJURY
An increase in the renal expression of genes that promote interstitial inflammation and renal TGF- production was found in rats with proteinoverload proteinuria.19 Therefore, the reduction of protein reabsorption by tubular cells induced by sodium bicarbonate treatment may prevent the increase in renal TGF- production, as well as the progression of renal disease. We found increased vimentin expression in tubular cells from the renal cortex associated with albumin overload. Because vimentin expression is an indicator of tubular cell regeneration suggestive of recent injury, the tubulointerstitial damage present in these animals may be related to the increase in reabsorption of this protein by tubular cells. Greater immunostaining scores for albumin were observed in the cytoplasm of the tubular cells from animals treated only with doxorubicin than in animals treated with doxorubicin plus sodium bicarbonate. Treatment with sodium bicarbonate induced urine alkalinization and extracellular volume expansion that can reduce the handling of proteins by tubular cells.14,16 The greater amount of urinary albumin excretion and greater immunostaining scores for albumin in the tubular cells observed in the animals treated only with doxorubicin compared with those treated with doxorubicin plus sodium bicarbonate, despite identical plasma albumin levels in the two groups, confirm this hypothesis. Urine alkalinization can decrease the positively charged groups of the proteins, interfering with their accessibility to the endocytic sites, binding to the luminal tubular membrane, and consequently with the efficiency of their tubular reabsorption. Although most charged groups of albumin are negative under normal conditions, this macromolecule also has several positive groups that may be reduced with the increase of intratubular fluid pH. The renal handling of other cationic or slighty anionic plasma proteins, such as IgG or IgG fragments, can also be reduced by this treatment. The extracellular volume expansion present in these animals can also contribute to the reduction of renal albumin handling because of the decrease in protein concentrations at the endocytic sites provoked by the reduction in the rate of sodium reabsorption. The increase of urinary volume observed in bicarbonate-treated rats confirms this hypothesis. Urine alkalinization can also decrease the precipitation of pro-
335
teins at the distal tubule level, another factor that may contribute to the nephrotoxicity of the filtered proteins.24 High sodium ingestion by the animals treated with sodium bicarbonate can also provoke a reduction in renal angiotensin level. It was observed that angiotensin can increase TGF- production by mesangial cells.45 Some studies suggested angiotensin contributes to the renal changes observed in animals with doxorubicin nephropathy.46,47 However, this effect was not verified in other studies.43,48,49 There is also some evidence that complement has a role in proteinuria-associated tubulointerstitial injury.50 Deposition of C3 and C5b-9 was detected in the kidney of rats with nephrosis induced by aminonucleosides. Therefore, changes in urinary pH may decrease complement activation, reducing the tubulointerstitial damage. These effects of sodium bicarbonate on the renal changes induced by doxorubicin were not observed by Nagashima et al,51 who did not observe any protective effect of sodium bicarbonate treatment on renal function and structure in doxorubicin-treated rats. However, these investigators used a different strain of rats (Dawley rats), greater doses of doxorubicin (2.5 mg/kg of body weight, administered twice to the rats after a 20-day interval), and started treatment with sodium bicarbonate at the begining of the experiment. The rate of urinary TGF- excretion was increased in the rats treated only with doxorubicin, but treatment with sodium bicarbonate prevented this increase. We observed a significant positive correlation between the rate of urinary TGF- excretion and incidence of glomerulosclerosis and tubulointerstitial damage score in rats with nephropathy induced by doxorubicin. Some observations suggesting urinary TGF- derives from renal biosynthesis and not from plasma have been reported.10,52-54 The amount of total TGF- (active plus latent) is negligible in plasma, and the amount of active TGF- is undetectable.52 Noh et al53 found urinary TGF- activity is related to the extent of scarring in rabbits with crescentic nephritis. We also detected a correlation between the amount of TGF- in urine and the incidence of glomerular sclerosis and tubulointerstitial lesions in rats with subtotal renal ablation and patients with glomerulonephritis.10,54 The rate of urinary TGF- excretion was
336
BARONI ET AL
greater in the group treated only with doxorubicin, which showed more alterations in renal structure (glomerular sclerosis and tubulointerstitial lesions). There was no correlation between albuminuria and urinary TGF- excretion. The animals treated with sodium bicarbonate showed greater albuminuria, although they did not present an increase in the rate of urinary TGF- excretion. Therefore, the determination of this polypeptide in urine could be a useful noninvasive procedure for the evaluation of renal TGF- production and disease progression. In conclusion, these data show treatment with sodium bicarbonate reduces renal TGF- production and the rate of urinary TGF- excretion in rats with doxorubicin-induced nephropathy, an effect related to the protective action of this treatment on renal structure, and suggest the determination of this polypeptide in urine could be a useful procedure for the evaluation of disease progression. ACKNOWLEDGMENT The authors thank Dr Euclides Braga Malheiros for advice and helpful discussion and Cleonice G.A. da Silva and Erika Dellaiogno for expert technical assistance.
REFERENCES 1. Bertani T, Poggi A, Pozzoni R, Delaini F, Sacchi G, Thoua Y, Mecca G, Remuzzi G, Donati MB: Adriamycininduced nephrotic syndrome in rats—Sequence of pathologic events. Lab Invest 46:16-23, 1982 2. Bertani T, Cutillo F, Zoja C, Broggine M, Remuzzi G: Tubulo-interstitial lesions mediate renal damage in Adriamycin glomerulopathy. Kidney Int 30:488-496, 1986 3. Bertani T, Rocchi G, Sacchi G, Mecca G, Remuzzi G: Adriamycin-induced glomerulosclerosis in the rat. Am J Kidney Dis 7:12-19, 1986 4. Okuda S, Oh Y, Tsuruda H, Onoyama E, Fujimi S, Fujishima M: Adriamycin-induced nephropathy as a model of chronic progressive glomerular disease. Kidney Int 29:502510, 1986 5. Border WA, Okuda S, Languino LR, Sporn MB, Ruoslahti E: Suppression of experimental glomerulonephritis by antiserum against transforming growth factor-1. Nature 346:371-374, 1990 6. Coimbra TM, Wiggins RC, Noh JW, Merritt S, Phan SH: Transforming growth factor- production in antiglomerular basement disease in rabbit. Am J Pathol 138:223– 234, 1991 7. Isaka Y, Fujiwara Y, Ueda N, Kaneda Y, Kamada T, Imai E: Glomerulosclerosis induced by in vivo transfection of transforming growth factor- or platelet-derived growth factor gene into the rat kidney. J Clin Invest 92:2597-2601, 1993 8. Sharma K, Ziyadeh FN: The emerging role of trans-
forming growth factor  in kidney diseases. Am J Physiol 266:F829-F842, 1994 9. Bertoluci MC, Schmid H, Lachat JJ, Coimbra TM: Transforming growth factor  in the development of rat diabetic nephropathy. A 10-month study with insulin-treated rats. Nephron 74:189-196, 1996 10. Monteiro de Freitas AS, Coimbra TM, Costa RS, Baroni EA: Urinary transforming growth factor- excretion and renal production of TGF- in rats with subtotal renal ablation: Effect of enalapril and nifedipine. Nephron 78:302309, 1998 11. Tamaki K, Okuda S, Miyazono K, Nakayama M, Fujishima M: Matrix-associated latent TGF- with latent TGF- binding protein in the progressive process in Adriamycin-induced nephropathy. Lab Invest 73:81-89, 1995 12. Tamaki K, Okuda S, Ando T, Iwamoto T, Nakayama M, Fujishima M: TGF-1 in glomerulosclerosis and interstitial fibrosis of Adriamycin nephropathy. Kidney Int 45:525536, 1994 13. Chiu PJS, Miller GJ, Long JF, Waity JA: Renal uptake and nephrotoxicity of gentamicin during urinary alkalinization in rats. Clin Exp Pharmacol Physiol 6:317326, 1979 14. Coimbra TM, DeGiacobbi G, Gouveia MA, Lachat JJ, Carvalho IF: Effect of urinary alkalinization on renal changes produced by cationic albumin. Nephron 38:261266, 1984 15. Christensen EI, Carone FA, Rennke HG: Effect of molecular charge on endocytic uptake of ferritin in renal proximal tubule cells. Lab Invest 44:351-358, 1981 16. Maack T, Hyang Park C, Camargo MJ: Renal filtration, transport, and metabolism of proteins in Seldin DW, Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 3005-3038 17. Eddy AA, McCulloch L, Liu E, Adams J: A relationship between proteinuria and acute tubulointerstitial disease in rats with experimental nephrotic syndrome. Am J Pathol 138:1111-1123, 1991 18. Eddy AA, Michael AF: Acute interstitial nephritis associated with aminonucleoside nephrosis. Kidney Int 33: 14-23, 1988 19. Eddy AA, Giachelli CM: Renal expression of genes that promote interstitial inflammation and fibrosis in rats with protein-overload proteinuria. Kidney Int 47:1546-1557, 1995 20. Eddy AA, McCulloch L, Adams J, Liu E: Interstitial nephritis induced by protein-overload proteinuria. Am J Pathol 135:719-733, 1989 21. Zoja C, Donadelli R, Colleoni S, Figliuzzi M, Bonazzola S, Morigi M, Remuzzi G: Protein overload stimulates RANTES production by proximal tubular cells depending on NF-B activation. Kidney Int 53:1608-1615, 1998 22. Zoja C, Morigi M, Figliuzzi M, Bruzzi I, Oldroyd S, Benigni A, Ronco PM, Remuzzi G: Proximal tubular cell synthesis and secretion of endothelin-1 on challenge with albumin and other proteins. Am J Kidney Dis 26:934-941, 1995 23. Remuzzi G, Ruggenenti P, Benigni A: Understanding the nature of renal disease progression. Kidney Int 51:2-15, 1997 24. Kant KS, Pesce AJ, Clyne DH, Pollack VE: Urinary
NaHCO3 TREATMENT AND RENAL INJURY
cast formation—pH-dependence and interaction of TammHorsfall mucoprotein with myoglobin, hemoglobin, Bence Jones protein, and albumin. Kidney Int 10:559A, 1976 (abstr) 25. Laurell CB: Electroimmunoassay. Scand J Clin Lab Invest 124:S21-S23, 1972 (suppl 1) 26. Fu¨hr Y, Kaczmarczk Y, Kruttgen GD: Eine einfache colorimetrische Methode zur Inulin-Bestimmung fu¨r NierenClearance-Untersuchungen bei Stoffwechselgesunden und Diabetikern. Klin Wochenschr 33:729-730, 1955 27. Haygen HN: The determination of ‘‘endogenous creatinine’’ in plasma and urine. Scand J Clin Lab Invest 5:48-57, 1953 28. Flanders KC, Thompson NL, Cissel DS, Van Obberghen-Schilling E, Baker CC, Kass ME, Ellingsworth LR, Roberts AB, Sporn MB: Transforming growth factor 1: Histochemical localization with antibodies to different epitopes. J Cell Biol 108:653-660, 1989 29. Kliem V, Johnson RJ, Alpers CE, Yoshimura A, Couser WG, Koch KM, Floege J: Mechanism involved in the pathogenesis of tubulointerstitial fibrosis in 5/6-nephrectomized rats.Kidney Int 49:666-678, 1996 30. Russ JC: The Image Processing Handbook (ed 2). Ann Arbor, MI, CRC, 1995 31. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Anal Biochem 162:156-159, 1987 32. Maniatis T, Fritsh EF, Sambrook J: Molecular Cloning: A Laboratory Manual (ed 2). Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press, 1989 33. Derynck RJ, Jarrett JA, Chen EY, Eaton DH, Bell JR, Assoian RK, Roberts AB, Sporn MB, Goeddel DV: Human transforming growth factor- complementary DNA sequence and expression in normal and transformed cells. Nature 316:701-705, 1985 34. Chan YL, Olivera J, Wood IG: The structure of rat 28 S ribosomal acid inferred from the sequence of nucleotides in a gene. Nucleic Acids Res 11:7819-7831, 1983 35. Feinberg AP, Vogelstein B: A technique for labelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13, 1983 36. Danielpour D, Kim KY, Dart LL, Watanabe S, Roberts AB, Sporn MB: Sandwich enzyme-linked immunosorbent assays quantitate and distinguish two forms of transforming growth factor-beta in complex biological fluids. Growth Factors 2:61-65, 1989 37. Gro¨ne H-J, Weber K, Gro¨ne E, Helmcken U, Osborn M: Coexpression of keratin and vimentin in damaged and regenerating tubular epithelia of the kidney. Am J Pathol 129:1-8, 1987 38. Kriz W, Elger M, Nagata M, Kretzler M, Uiker S, Koeppen Hageman I, Tenschert S, Lemley KV: The role of podocytes in the development of glomerular sclerosis. Kidney Int 45:S64-S72, 1994 (suppl 45) 39. Floege J, Johnson RJ, Alpers CE, Fatemi-Nainie S, Richardson CA, Gordon K, Couser WG: Visceral glomerular epithelial cells can proliferate in vivo and synthesize platelet-derived growth factor- chain. Am J Pathol 142:637650, 1993 40. Remuzzi G, Zoja C, Remuzzi A, Rossini M, Battaglia
337
C, Broggini M, Bertani T: Low-protein diet prevents glomerular damage in Adriamycin-treated rats. Kidney Int 28:21-27, 1985 41. Motomura K, Okuda S, Sanai T, Ando T, Onoyama K, Fujishima M: Importance of early initiation of dietary protein restriction for the prevention of experimental progressive renal disease. Nephron 49:144-149, 1988 42. Okuda S, Nakamura T, Yamamoto T, Ruoslahti E, Border WA: Dietary protein restriction rapidly reduces TGF-1 expression in experimental glomerulonephritis. Proc Natl Acad Sci U S A 88:9765-9769, 1991 43. Barretti P, Viero RM, Soares VA: Effects of dietary protein, angiotensin I converting enzyme inhibition and mesangial overload on the progression of Adriamycininduced nephropathy. Braz J Med Biol Res 28:39-50, 1995 44. Nakayama M, Okuda S, Tamaki K, Fujishima M: Short- or long-term effects of a low-protein diet on fibronectin and transforming growth factor-beta synthesis in Adriamycin-induced nephropathy. J Lab Clin Med 127:29-39, 1996 45. Kagami S, Border WA, Miller DE, Noble NA: Angiotensin II stimulates extracellular matrix protein synthesis through induction of transforming growth factor- expression in rat glomerular mesangial cells. J Clin Invest 93:24312437, 1994 46. Irwin KC, Brouhard BH, Satoh S, Stowe NT: Effects of enalapril on Adriamycin-induced nephrosis. Pediatr Nephrol 6:448-450, 1992 47. Jovanovic D, Dimitrijevic J, Varagic J, Jovovic D, Starcevic A, Djukanovic L: Effects of captopril on morphologic changes in kidney spontaneously hypertensive rats with Adriamycin nephropathy. Ren Fail 20:451-458, 1998 48. Hall RL, Wilke WL, Fettman MJ: The progression of Adriamycin-induced nephrotic syndrome in rats and the effect of captopril. Toxicol Appl Pharmacol 82:164-174, 1986 49. Beukers JJ, Hoedemaker PJ, Weening JJ: A comparison of the effects of converting-enzyme inhibition and protein restriction in experimental nephrosis. Lab Invest 59:631-640, 1988 50. Nomura A, Morita Y, Maruyama S, Hotta N, Nadai M, Wang L, Hasegawa T, Matsuo S: Role of complement in acute tubulointerstitial injury of rats with aminonucleoside nephrosis. Am J Pathol 151:539-547, 1997 51. Nagashima A, Okuda S, Tamaki K, Fujishima M: Influence of high-salt diet on glomerular injury and the preventive effects of amiloride in Adriamycin nephropathy. Nephron 71:87-94, 1995 52. Grainger DJ, Mosedale DE, Metcalfe JC, Weissberg PL, Kemp PR: Active and acid-activatable TGF- in sera, platelets and plasma. Clin Chim Acta 235:11-31, 1995 53. Noh JW, Wiggins RC, Phan SH: Urine transforming growth factor- activity is related to the degree of scarring in crescentic nephritis in rabbit. Nephron 63:73-78, 1993 54. Dominguez GCS, Costa RS, Dantas M, Kimachi T, Piucci CR, Coimbra TM: Transforming growth factor  activity in urine from patients with glomerulonephritis is related to their functional and structural changes. Nephrology 4:31-36, 1998
A
DRIAMYCIN (doxorubicin; Farmitalia, Carlo Erba RJ, Brazil) injection induces marked and persistent proteinuria and glomerular structural changes in rats.1 Glomerular epithelial cell lesions that result in loss of the foot processes are observed in the kidneys of these animals. These alterations progress to glomeruloFrom the Departments of Physiology and Pathology, Medical School of Ribeira˜o Preto, USP, Ribeira˜o Preto, SP, Brazil. Received July 31, 1998; accepted in revised form March 5, 1999. Presented in part at the 31st Annual Meeting of American Society of Nephrology, Philadelphia, PA, October 25-28, 1998. Supported in part by Fundaca˜o de Amparo a` Pesquı´sa do Estado de sa˜d Paulo, Brazil. R.S.C. and T.M.C. are recipients of Conselho Nacional de Desenvolvimento Cientı´ficoe Technolo´gico, DF, Brazil, fellowships. Address reprint requests to Terezila M. Coimbra, MD, Departamento de Fisiologia, Faculdade de Medicina de Ribeira˜o Preto, USP, 14049-900-Ribeira˜o Preto-SP, Brazil. E-mail: [email protected]
娀 1999 by the National Kidney Foundation, Inc. 0272-6386/99/3402-0020$3.00/0 328
sclerosis and tubulointerstitial fibrosis.1-4 Transforming growth factor- (TGF-) has been considered the principal cytokine involved in the pathogenesis of renal fibrosis.5-10 Some studies have shown increased renal TGF- production in rats with nephropathy induced by doxorubicin.11,12 Treatment with sodium bicarbonate provokes urine alkalinization and extracellular volume expansion that may reduce protein reabsorption by tubular cells.13,14 Tubular reabsorption of protein depends on the electrostatic interaction between positively charged groups of protein and anionic sites on the luminal membrane or membrane coat.15 The modifications in protein charges caused by the change in urinary pH must interfere with protein binding to the luminal membrane, which is the first step of endocytosis. Changes in protein concentrations at the endocytic sites because of extracellular volume expansion also influence protein reabsorption.16 Some evidence suggests a role of proteinuria and renal handling of proteins in the development of tubu-
American Journal of Kidney Diseases, Vol 34, No 2 (August), 1999: pp 328-337
NaHCO3 TREATMENT AND RENAL INJURY
lointerstitial damage.17,18 It has been suggested that proteins filtered by glomeruli could be toxic for the tubule cells.19,20 Studies with tubule cell cultures have shown that some proteins, such as immunoglobulin G (IgG) and albumin, induce increased production of inflammatory and vasoactive factors.21,22 This observation can explain why greater proteinuria is frequently associated with more severe tubular damage.23 Urine alkalinization can also decrease the precipitation of proteins at the distal tubule level, another factor that may contribute to the nephrotoxicity of the filtered proteins. It was observed in vitro that when pH decreased from 6.0 to 4.5, some proteins, such as albumin, coprecipitate with Tamm-Horsfall mucoprotein.24 The aim of this study is to investigate the effect of sodium bicarbonate on renal injury and renal TGF- production and excretion in rats with nephropathy induced by doxorubicin. We also analyzed the relationship between the rate of urinary TGF- excretion and structural renal changes observed in these animals. MATERIALS AND METHODS
Animals and Experimental Protocols Sixty female Wistar rats weighing 180 to 220 g were used. Thirty rats were injected with a single dose of doxorubicin, 3.5 mg/kg of body weight, through the femoral vein, and 30 with 0.15 mol/L of sodium chloride solution (control group). Fifteen days after the injections, we selected 23 of the animals injected with doxorubicin with albuminuria of 4.0 to 38.0 mg/24 h of albumin, and the experimental and control animals were divided into four groups that underwent different treatments. The control animals (group C) and doxorubicin-injected animals (group AD) did not receive any treatment. For the animals from groups C-B (control) and AD-B (doxorubicin injected), we replaced the drinking water with a 0.15 mol/L of sodium bicarbonate solution. There was no difference in mean urine albumin levels between the different groups at 15 days. All animals had free acess to normoproteic (18%) chow. The average amount of food consumed by the groups was determined on alternate days after the treatment was started using the following formula: IMG ⫽ (Ro ⫺ Ri)/N where: IMG ⫽ average amount of food consumed by the group, Ro ⫽ total amount of chow offered to the group, Ri ⫽ total amount of chow not consumed by the group, and N ⫽ number of animals in the group. This amount was not modified by the different kinds of treatment. After treatment with sodium bicarbonate, the rats showed a much higher urinary pH (8 to 9) than before injection (6.0 to 7.0). Monthly urine samples were collected to determine albumin excretion and urinary volume. Systolic blood pressure was
329
measured by the tail-cuff method 75 days after the injection. The animals were killed with excess anesthesia 105 days after the injection. The organs were perfused with phosphatebuffered saline (PBS) solution (0.15 mol/L of sodium chloride, 0.01 mol/L of sodium phosphate buffer, pH 7.2), and the kidneys were removed for messenger RNA (mRNA), immunohistochemical, histological, and morphometric studies.
Renal Function Studies Urinary and plasma albumin were quantified by electroimmunoassay using a specific antibody against rat albumin.25 Glomerular filtration rate (GFR) was measured by inulin clearance26 in five control animals, six doxorubicin-treated animals, and six animals treated with doxorubicin and sodium bicarbonate at 105 days after the beginning of treatment. The animals were anesthetized intraperitoneally with 50 mg/kg of sodium thionembutal. After tracheostomy, the femoral artery and vein were cannulated to collect blood samples and inject fluids, and both ureters were cannulated to collect urine. The animals received a priming inulin dose of 12 mg/100 g, followed by a maintenance dose of 30 mg/100 g/h. After stabilization for approximately 60 minutes, urine was collected for 60 minutes and blood sampled at 30 and 60 minutes. Plasma and urine inulin was measured by the method of Fu¨hr et al,26 and creatinine was measured by the Jaffe´ reaction.27
Antibodies The primary antibodies used were: (1) a purified IgG fraction of polyclonal rabbit antibody against TGF-1 (Sta Cruz Biotechnology, Sta Cruz, CA) specific for TGF-1 that shows no cross-reactivity with TGF-2 or TGF-3 and is reactive for mouse, rat, and human TGF-1; (2) a purified IgG fraction of polyclonal rabbit antirat fibronectin (Chemicon; International INC, Temecula, CA); (3) a monoclonal mouse antivimentin antibody (Dako Corporation, Glostrup, Denmark), and (4) a polyclonal antirat albumin antibody produced in our laboratory.14
Immunohistochemical Studies Eleven doxorubicin-injected animals and five control animals were killed 105 days after injection. The rats were submitted to left intraventricular perfusion with PBS (0.15 mol/L of sodium chloride, 0.01 mol/L phosphate buffer, pH 7.2) until the kidneys were blanched. The kidneys were then perfused with 4% formalin for 2 hours, postfixed in Bouin’s solution for an additional 4 to 6 hours, rinsed with 70% ethanol to eliminate picric acid, dehydrated through a graded series of ethanol solutions, embedded in Paraplast (improved paraffin embedding media; Sigma, St. Louis, MO), sectioned into 3-µm slices, deparaffinized, and subjected to immunohistochemical staining.28 The sections were incubated overnight at 4°C with a 1/30 anti–TGF-1 polyclonal antibody or 1/800 antirat fibronectin antibody or for 1 hour at room temperature with a 1/1,500 monoclonal mouse antivimentin or 1/5,000 antirat albumin antibody. The reaction product was detected with an avidinbiotin-peroxidase complex (Vector Laboratories, Burlingame, CA). The color reaction was developed with 3,
330
3-diamino-benzidine (Sigma Chemical Company, St Louis, MO), and the material was counterstained with Harris hematoxylin or methyl green, dehydrated, and mounted. Nonspecific protein binding was blocked by incubation with 5% goat serum in PBS for 30 minutes. Controls for TGF- antibody consisted of replacing anti–TGF-1 antibody with normal rabbit IgG or anti–TGF-1 antibody preincubated with cellulose-bound human TGF-1 protein (Boehringer Mannheim Biochemica, Mannheim, Germany). For the other antibodies, the controls consisted of replacing the primary antibody with normal rabbit (polyclonal antibodies) or mouse IgG (monoclonal antibody). For evaluation of the immunoperoxidase stains for TGF-, fibronectin, vimentin, and albumin, each glomerular area or tubulointerstitial grid field was graded semiquantitatively, and the mean score per biopsy was calculated. Each score reflects mainly changes in the extent, rather than intensity, of staining and depended on the percentage of glomerular tuft area or grid field showing positive staining: 0, absent or less than 5%; I, 5% to 25%; II, 25% to 50%; III, 50% to 75%; and IV, ⬎75%.29
BARONI ET AL
fragments were labeled to a specific activity of 1 ⫻ 109 disintegrations/min/µg using the random primer method.35 Autoradiograms from these blots were quantified by optic densitometry. The densitometric ratio between TGF-1 mRNA signals and 28 S was calculated, and data are expressed in comparison to control, designated as 100%.
Quantification of TGF- in Urine Samples The ureters from 12 control rats and 16 doxorubicininjected rats were cannulated, and urine samples were collected over a period of 30 minutes for TGF- and creatinine determination. The samples were immediatly treated with 1 mmol/L of phenylmethylsulphonyl fluoride (Sigma Chemical Company, St Louis, MO) and acidified to measure total TGF- (latent plus active). Quantification of TGF- in these samples was performed by enzyme-linked immunosorbent assay using kits from Promega (Promega Corporation, Madison, MO),36 and creatinine was determined by Jaffe´’s reaction.27 Urinary TGF- is reported as picograms of TGF- per milligram of urinary creatinine to correct the variation in urine concentration.
Light Microscopy The kidneys from 14 control animals and 15 doxorubicintreated rats killed 105 days after injection were fixed in alcoholic Bouin for 12 hours and processed for paraffin embedding. Three-micrometer histological sections were stained with hematoxylin-eosin and Masson’s trichrome and examined under the light microscope. The percentage of segmental glomerulosclerosis or tubulointerstitial damage was determined by morphometry with a light camera connected to an image analyzer (KS-300; Kontron Electronik, Munich, Germany). Forty subcapsular and 20 juxtaglomerular glomeruli or 20 fields from the cortex were measured in a section of each kidney.30 Encircled damaged areas were traced on a videoscreen and determined by computerized morphometry. Each glomerular or tubulointerstitial field was graded quantitatively, and the mean value was calculated for each biopsy. The glomerular and tubulointerstitial immunostain for TGF- in the renal cortex was also evaluated by morphometry.
Statistical Analysis Urine and plasma albumin levels, blood pressure, glomerular filtration rate, urinary volume, plasma creatinine level, renal cortical TGF- mRNA, percentage of glomerulosclerosis or tubulointerstitial damage, and the immunohistochemical scores obtained under control and experimental conditions were submitted to analysis of variance with multiple comparisons by the Tukey test. Data concerning urinary albumin excretion, percentage of glomerulosclerosis, and immunohistochemical scores for TGF- in the glomeruli and fibronectin in the cortical tubulointerstitium were log10 transformed before analysis, as required by the Bartlett test. Pearson’s correlation coefficient was calculated to determine r. P less than 0.05 is considered statistically significant.
RESULTS
RNA Analysis
Immunohistochemical and Light Microscopy Studies
Cortical renal tissue was obtained from 6 control animals and 16 doxorubicin-injected animals killed 105 days after injection. The samples were prepared by the guanidine/ phenol/chloroform technique described by Chomczynsky and Sacchi.31 The RNA obtained was subjected to electrophoresis in formaldehyde-agarose gels and transferred to nitrocellulose filters (Schleicher & Shuell Inc, Keene, NH). RNA was also analyzed by dot blot hybridization after loading onto a blotting apparatus under denaturing conditions.32 Thirty micrograms of total RNA was loaded for Northern analysis and 10, 5, and 2.5 µg were loaded for dot blot analysis. The plasmid with complementary DNA (cDNA) encoding a portion of human TGF-1 RNA was used to detect transcripts for TGF-1.33 A plasmid containing portions of rat 28 S ribosomal ribonucleic acid (RNA)34 was used to normalize the TGF-1 signal and to adjust for inequalities in RNA loading and/or transfer. The purified
Light microscopy studies showed glomeruli with global or segmental sclerosis in the kidneys from rats with doxorubicin-induced nephropathy (P ⬍ 0.05). The tubulointerstitial compartment showed tubular atrophy and/or dilatation with intraluminal casts, interstitial mononuclear cell infiltration, and fibrosis (P ⬍ 0.05). These lesions were less intense in the rats treated with sodium bicarbonate (P ⬍ 0.05; Table 1). The results of the immunohistochemical studies showed the presence of moderate immunostaining for TGF- in the cytoplasm of some tubular cells in the renal cortex from control rats. We did not observe any reaction in the glomeruli
NaHCO3 TREATMENT AND RENAL INJURY
331
Table 1. Blood Pressure, Plasma Albumin, and Renal Functional, Histological, and Immunohistochemical Data From Control, Doxorubicin Treated, and Doxurubicin Plus Sodium Bicarbonate Treated Rats Groups
C (n ⫽ 5)
AD (n ⫽ 6)
AD-B (n ⫽ 6)
Blood pressure (mm Hg) Urinary volume (mL/24 h) GFR (mL/min/100 g) Plasma creatinine (mg/dL) Plasma albumin (mg/mL) GS (%) TID (%) TGF--C (%) TGF--G Fib-CTI Fib-G Albumin Vimentin
106.4 ⫾ 7.6 5.54 ⫾ 1.61 0.76 ⫾ 0.07 0.59 ⫾ 0.03 29.49 ⫾ 6.26 0.11 ⫾ 0.08 0.01 ⫾ 0.03 19.60 ⫾ 3.65 1.00 ⫾ 1.00 0.99 ⫾ 0.06 1.68 ⫾ 0.18 0.02 ⫾ 0.03 0.00 ⫾ 0.00
102.9 ⫾ 9.9 17.80 ⫾ 11.32* 0.53 ⫾ 0.19* 0.76 ⫾ 0.28 15.78 ⫾ 5.23* 14.70 ⫾ 12.80* 54.60 ⫾ 20.30* 44.17 ⫾ 4.90* 13.00 ⫾ 4.8* 1.49 ⫾ 0.24* 2.95 ⫾ 0.38* 1.30 ⫾ 0.31* 1.28 ⫾ 0.32*
101.5 ⫾ 4.6 44.55 ⫾ 17.95*† 0.62 ⫾ 0.09 0.67 ⫾ 0.17 14.76 ⫾ 3.19* 4.38 ⫾ 1.90*† 16.70 ⫾ 10.30† 44.60 ⫾ 9.13* 6.60 ⫾ 2.70* 1.14 ⫾ 0.02*† 2.28 ⫾ 0.32† 0.30 ⫾ 0.10† 0.54 ⫾ 0.34*†
Abbreviations: GFR, glomerular filtration rate; GS, glomeruli with glomerulosclerosis; TID, tubulointerstitial damage; TGF--C, immunostaining for TGF- in the renal cortex; TGF--G, immunostaining scores for TGF- in the glomeruli; Fib-CTI, immunostaining scores for fibronectin the cortical tubulointerstitium; Fib-G, immunostaining scores for fibronectin in the glomeruli; C, control rats; AD, rats treated with doxorubicin; AD-B, rats treated with doxorubicin plus sodium bicarbonate. *P ⬍ 0.05 v control. †P ⬍ 0.05 v AD.
of these animals (Fig 1A). The staining intensity was greater in the kidneys of the rats treated only with doxorubicin (Fig 1B). In these animals, we observed a marked immunohistochemical reaction for TGF- in the tubular cells and glomeruli localized primarily in the podocytes, whereas in the interstitial cells, the reaction was less intense (Fig 1B). The same pattern of reaction was verified in the glomeruli of rats treated with
Fig 1. Immunolocalization of TGF-1 in the renal cortex of (A) a control rat and (B) a rat treated with doxorubicin or (D) doxorubicin plus sodium bicarbonate 105 days after injection using a TGF-1 polyclonal antibody. Note the intense staining present (B) in podocytes and (A, B, D) in the cytoplasm of some tubule cells. (C) Immunolocalization of TGF-1 in the renal cortex of a control rat using a TGF-1 antibody preincubated with TGF- protein. Staining was abolished by this treatment. (Original magnification ⴛ2.)
sodium bicarbonate, although at lower intensity (Fig 1D). The intensity of tubulointerstitial reaction for TGF- was also less in the animals treated with sodium bicarbonate than in the rats treated only with doxorubicin. However, we did not find any difference in the immunostaining score for TGF- between these groups because these scores reflect mainly changes in immunostaining extent, rather than intensity (Table 1).
332
No reaction was observed in the cortex from these animals when the TGF-1 antibody was replaced with normal rabbit IgG or with a TGF- antibody previously absorbed with a TGF-1 protein (Fig 1C), thus showing the specificity of the reaction. The immunohistochemical analysis also showed that the animals treated only with doxorubicin had greater fibronectin expression in the glo-
BARONI ET AL
meruli and cortical tubulointerstitium and greater immunostaining scores for vimentin and albumin in the tubular cells from the renal cortex than the animals treated with doxorubicin plus sodium bicarbonate (Fig 2). The increase of vimentin expression in the cytoplasm of tubular cells was associated with an albumin overload in these cells. Because the expression of vimentin is an indicator of tubular cell regeneration suggestive of
Fig 2. Immunolocalization of (A, B) fibronectin, (C, D) vimentin, and (E, F) albumin in the renal cortex of a rat treated with (B, D, F) doxorubicin or (A, C, E) doxorubicin plus sodium bicarbonate 105 days after injection. Note the immunostaining for fibronectin in glomeruli and cortical tubulointerstitium is more intense in B and the immunostaining for albumin and vimentin in tubular cells is more intense in D and F. (Original magnification; [A, B, C, D] ⴛ120; [E, F] ⴛ280.)
NaHCO3 TREATMENT AND RENAL INJURY
333
recent injury,37 this damage may be related to an increase in tubular albumin reabsorption. mRNA Analysis Dot blot analysis showed a marked increase in TGF- mRNA content in the renal cortex from rats injected with doxorubicin (Fig 3) compared with control animals (P ⬍ 0.05). The renal cortex levels of TGF- mRNA in rats treated with doxorubicin plus sodium bicarbonate were similar to those observed in control animals. There was no difference in renal cortex levels of TGF- mRNA between the two groups of control animals (Fig 3). Northern blot analysis was also performed to check the specificity of the TGF- probe, and the results essentially confirmed those obtained by dot blot analysis (Fig 4).
Fig 4. Autoradiographs of the Northern blot analysis of renal cortical RNA from a control rat (C) and from rats killed 105 days after doxorubicin injection (AD). RNA was hybridized with a TGF-1 cDNA probe (first line). The line at the bottom of the figure corresponds to the same blot hybridized with the 28 S cDNA probe.
Rate of Urinary TGF- Excretion The rate of urinary TGF- excretion was greater in animals treated with doxorubicin than control rats (P ⬍ 0.05; Fig 5). However, treatment with sodium bicarbonate prevented this increase (P ⬍ 0.05). We observed a significant positive correlation between the rate of urinary TGF- excretion and incidence of glomerular sclerosis (r ⫽ 0.74; P ⫽ 0.014) and tubulointerstitial damage (r ⫽ 0.76; P ⫽ 0.029) in rats injected with doxorubicin.
Fig 3. Quantification of renal cortical TGF-1 RNA in doxorubicin-injected rats (AD) and control rats (C) by dot blot hybridization with a cDNA TGF-1 probe. A densitometric ratio between the TGF- mRNA area and the 28 S area was calculated, and data are expressed in comparison to the control (C), with the mean ⴞ SD control value designated as 100%. Autoradiograms of the individual results shown at the top of the graphs represent the TGF- mRNA data from a control rat (C), a control rat treated with sodium bicarbonate (C-B), a rat treated with doxorubicin (AD), and a rat treated with doxorubicin plus sodium bicarbonate (AD-B). The line below corresponds to the same blots in the same assay, hybridized with the 28 S cDNA probe.
Fig 5. Rate of urinary TGF- excretion for control rats (C) and for rats treated with doxorubicin (AD) or doxorubicin plus sodium bicarbonate (AD-B) at 105 days after doxorubicin injection.
334
BARONI ET AL
Fig 6. Urinary albumin excretion (UAE) in rats injected with sodium chloride solution and treated with sodium bicarbonate (C-B) or untreated (C) and in rats injected with doxorubicin and treated with sodium bicarbonate (AD-B) or untreated (AD).
Urinary and Plasma Albumin and Blood Pressure Urinary albumin excretion showed a significant and progressive increase in both groups of animals injected with doxorubicin compared with the respective controls (P ⬍ 0.05; Fig 6). However, these increases were more intense in the animals treated with sodium bicarbonate (P ⬍ 0.05). Plasma albumin levels were less in all doxorubicin-treated animals compared with controls (P ⬍ 0.01). However, there was no difference between the two groups of rats treated with doxorubicin (Table 1). Blood pressure was not modified by the different kinds of treatment (Table 1). Renal Function We did not observe any difference in plasma creatinine levels between the animals from different groups (Table 1). However, GFR values (mean ⫾ SD) were less in the animals treated only with doxorubicin than in controls (Table 1), probably because inulin clearance is a more precise method for evaluation of renal function. Treatment with sodium bicarbonate prevented the decrease in GFR observed in animals treated with doxorubicin. DISCUSSION
Our data show that all animals treated with doxorubicin (groups AD and AD-B) presented
glomerulosclerosis and tubulointerstititial damage. However, these disturbances were less intense in the animals treated with sodium bicarbonate. We observed increased immunohistochemical TGF-1 staining in the cytoplasm of some tubular cells of the kidney from doxorubicin-injected rats, whereas the reaction was less intense in animals treated with sodium bicarbonate. We also found immunoreactivity for TGF-1 in the glomeruli (podocytes) of these animals. The increased immunoreaction intensity for TGF- protein in the renal cortex from these animals was associated with the enhancement of renal cortex TGF- mRNA content, suggesting an increased production of this polypeptide by the kidney. This enhancement was prevented by treatment with sodium bicarbonate. The immunostaining score for fibronectin in the glomeruli and cortical tubulointerstitium was also greater in the animals treated only with doxorubicin. An increase in TGF- mRNA levels and TGF- protein in doxorubicin-treated rats has been previously observed in recent studies11,12 and was related to the structural renal changes observed in these animals. TGF- has been considered one of the major cytokines involved in the regulation of extracellular matrix synthesis and degradation. It was also verified that the introduction of a plasmid containing cDNA for TGF- through the renal artery induced glomerulosclerosis.7 The present immunohistochemical studies showed the presence of an immune reaction for TGF-1 in the glomeruli from doxorubicintreated rats primarily localized in the podocytes. It has been shown that podocytes can produce various cytokines, including TGF-.38,39 Although the role of TGF- in podocyte injury has not been determined, extracellular matrix exposure by glomerular basement membrane denudation provoked by podocyte injury could result in adhesions between glomerular tuft and Bowman’s capsule and segmental sclerotic lesions. There is evidence that renal handling of proteins present in animals with nephropathy induced by doxorubicin may contribute to increased renal TGF- production and development of tubulointerstitial damage.17,18 A low-protein diet attenuates the proteinuria and progression of tubulointerstitial lesions and reduces renal TGF- production in rats injected with doxorubicin.40-44
NaHCO3 TREATMENT AND RENAL INJURY
An increase in the renal expression of genes that promote interstitial inflammation and renal TGF- production was found in rats with proteinoverload proteinuria.19 Therefore, the reduction of protein reabsorption by tubular cells induced by sodium bicarbonate treatment may prevent the increase in renal TGF- production, as well as the progression of renal disease. We found increased vimentin expression in tubular cells from the renal cortex associated with albumin overload. Because vimentin expression is an indicator of tubular cell regeneration suggestive of recent injury, the tubulointerstitial damage present in these animals may be related to the increase in reabsorption of this protein by tubular cells. Greater immunostaining scores for albumin were observed in the cytoplasm of the tubular cells from animals treated only with doxorubicin than in animals treated with doxorubicin plus sodium bicarbonate. Treatment with sodium bicarbonate induced urine alkalinization and extracellular volume expansion that can reduce the handling of proteins by tubular cells.14,16 The greater amount of urinary albumin excretion and greater immunostaining scores for albumin in the tubular cells observed in the animals treated only with doxorubicin compared with those treated with doxorubicin plus sodium bicarbonate, despite identical plasma albumin levels in the two groups, confirm this hypothesis. Urine alkalinization can decrease the positively charged groups of the proteins, interfering with their accessibility to the endocytic sites, binding to the luminal tubular membrane, and consequently with the efficiency of their tubular reabsorption. Although most charged groups of albumin are negative under normal conditions, this macromolecule also has several positive groups that may be reduced with the increase of intratubular fluid pH. The renal handling of other cationic or slighty anionic plasma proteins, such as IgG or IgG fragments, can also be reduced by this treatment. The extracellular volume expansion present in these animals can also contribute to the reduction of renal albumin handling because of the decrease in protein concentrations at the endocytic sites provoked by the reduction in the rate of sodium reabsorption. The increase of urinary volume observed in bicarbonate-treated rats confirms this hypothesis. Urine alkalinization can also decrease the precipitation of pro-
335
teins at the distal tubule level, another factor that may contribute to the nephrotoxicity of the filtered proteins.24 High sodium ingestion by the animals treated with sodium bicarbonate can also provoke a reduction in renal angiotensin level. It was observed that angiotensin can increase TGF- production by mesangial cells.45 Some studies suggested angiotensin contributes to the renal changes observed in animals with doxorubicin nephropathy.46,47 However, this effect was not verified in other studies.43,48,49 There is also some evidence that complement has a role in proteinuria-associated tubulointerstitial injury.50 Deposition of C3 and C5b-9 was detected in the kidney of rats with nephrosis induced by aminonucleosides. Therefore, changes in urinary pH may decrease complement activation, reducing the tubulointerstitial damage. These effects of sodium bicarbonate on the renal changes induced by doxorubicin were not observed by Nagashima et al,51 who did not observe any protective effect of sodium bicarbonate treatment on renal function and structure in doxorubicin-treated rats. However, these investigators used a different strain of rats (Dawley rats), greater doses of doxorubicin (2.5 mg/kg of body weight, administered twice to the rats after a 20-day interval), and started treatment with sodium bicarbonate at the begining of the experiment. The rate of urinary TGF- excretion was increased in the rats treated only with doxorubicin, but treatment with sodium bicarbonate prevented this increase. We observed a significant positive correlation between the rate of urinary TGF- excretion and incidence of glomerulosclerosis and tubulointerstitial damage score in rats with nephropathy induced by doxorubicin. Some observations suggesting urinary TGF- derives from renal biosynthesis and not from plasma have been reported.10,52-54 The amount of total TGF- (active plus latent) is negligible in plasma, and the amount of active TGF- is undetectable.52 Noh et al53 found urinary TGF- activity is related to the extent of scarring in rabbits with crescentic nephritis. We also detected a correlation between the amount of TGF- in urine and the incidence of glomerular sclerosis and tubulointerstitial lesions in rats with subtotal renal ablation and patients with glomerulonephritis.10,54 The rate of urinary TGF- excretion was
336
BARONI ET AL
greater in the group treated only with doxorubicin, which showed more alterations in renal structure (glomerular sclerosis and tubulointerstitial lesions). There was no correlation between albuminuria and urinary TGF- excretion. The animals treated with sodium bicarbonate showed greater albuminuria, although they did not present an increase in the rate of urinary TGF- excretion. Therefore, the determination of this polypeptide in urine could be a useful noninvasive procedure for the evaluation of renal TGF- production and disease progression. In conclusion, these data show treatment with sodium bicarbonate reduces renal TGF- production and the rate of urinary TGF- excretion in rats with doxorubicin-induced nephropathy, an effect related to the protective action of this treatment on renal structure, and suggest the determination of this polypeptide in urine could be a useful procedure for the evaluation of disease progression. ACKNOWLEDGMENT The authors thank Dr Euclides Braga Malheiros for advice and helpful discussion and Cleonice G.A. da Silva and Erika Dellaiogno for expert technical assistance.
REFERENCES 1. Bertani T, Poggi A, Pozzoni R, Delaini F, Sacchi G, Thoua Y, Mecca G, Remuzzi G, Donati MB: Adriamycininduced nephrotic syndrome in rats—Sequence of pathologic events. Lab Invest 46:16-23, 1982 2. Bertani T, Cutillo F, Zoja C, Broggine M, Remuzzi G: Tubulo-interstitial lesions mediate renal damage in Adriamycin glomerulopathy. Kidney Int 30:488-496, 1986 3. Bertani T, Rocchi G, Sacchi G, Mecca G, Remuzzi G: Adriamycin-induced glomerulosclerosis in the rat. Am J Kidney Dis 7:12-19, 1986 4. Okuda S, Oh Y, Tsuruda H, Onoyama E, Fujimi S, Fujishima M: Adriamycin-induced nephropathy as a model of chronic progressive glomerular disease. Kidney Int 29:502510, 1986 5. Border WA, Okuda S, Languino LR, Sporn MB, Ruoslahti E: Suppression of experimental glomerulonephritis by antiserum against transforming growth factor-1. Nature 346:371-374, 1990 6. Coimbra TM, Wiggins RC, Noh JW, Merritt S, Phan SH: Transforming growth factor- production in antiglomerular basement disease in rabbit. Am J Pathol 138:223– 234, 1991 7. Isaka Y, Fujiwara Y, Ueda N, Kaneda Y, Kamada T, Imai E: Glomerulosclerosis induced by in vivo transfection of transforming growth factor- or platelet-derived growth factor gene into the rat kidney. J Clin Invest 92:2597-2601, 1993 8. Sharma K, Ziyadeh FN: The emerging role of trans-
forming growth factor  in kidney diseases. Am J Physiol 266:F829-F842, 1994 9. Bertoluci MC, Schmid H, Lachat JJ, Coimbra TM: Transforming growth factor  in the development of rat diabetic nephropathy. A 10-month study with insulin-treated rats. Nephron 74:189-196, 1996 10. Monteiro de Freitas AS, Coimbra TM, Costa RS, Baroni EA: Urinary transforming growth factor- excretion and renal production of TGF- in rats with subtotal renal ablation: Effect of enalapril and nifedipine. Nephron 78:302309, 1998 11. Tamaki K, Okuda S, Miyazono K, Nakayama M, Fujishima M: Matrix-associated latent TGF- with latent TGF- binding protein in the progressive process in Adriamycin-induced nephropathy. Lab Invest 73:81-89, 1995 12. Tamaki K, Okuda S, Ando T, Iwamoto T, Nakayama M, Fujishima M: TGF-1 in glomerulosclerosis and interstitial fibrosis of Adriamycin nephropathy. Kidney Int 45:525536, 1994 13. Chiu PJS, Miller GJ, Long JF, Waity JA: Renal uptake and nephrotoxicity of gentamicin during urinary alkalinization in rats. Clin Exp Pharmacol Physiol 6:317326, 1979 14. Coimbra TM, DeGiacobbi G, Gouveia MA, Lachat JJ, Carvalho IF: Effect of urinary alkalinization on renal changes produced by cationic albumin. Nephron 38:261266, 1984 15. Christensen EI, Carone FA, Rennke HG: Effect of molecular charge on endocytic uptake of ferritin in renal proximal tubule cells. Lab Invest 44:351-358, 1981 16. Maack T, Hyang Park C, Camargo MJ: Renal filtration, transport, and metabolism of proteins in Seldin DW, Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 3005-3038 17. Eddy AA, McCulloch L, Liu E, Adams J: A relationship between proteinuria and acute tubulointerstitial disease in rats with experimental nephrotic syndrome. Am J Pathol 138:1111-1123, 1991 18. Eddy AA, Michael AF: Acute interstitial nephritis associated with aminonucleoside nephrosis. Kidney Int 33: 14-23, 1988 19. Eddy AA, Giachelli CM: Renal expression of genes that promote interstitial inflammation and fibrosis in rats with protein-overload proteinuria. Kidney Int 47:1546-1557, 1995 20. Eddy AA, McCulloch L, Adams J, Liu E: Interstitial nephritis induced by protein-overload proteinuria. Am J Pathol 135:719-733, 1989 21. Zoja C, Donadelli R, Colleoni S, Figliuzzi M, Bonazzola S, Morigi M, Remuzzi G: Protein overload stimulates RANTES production by proximal tubular cells depending on NF-B activation. Kidney Int 53:1608-1615, 1998 22. Zoja C, Morigi M, Figliuzzi M, Bruzzi I, Oldroyd S, Benigni A, Ronco PM, Remuzzi G: Proximal tubular cell synthesis and secretion of endothelin-1 on challenge with albumin and other proteins. Am J Kidney Dis 26:934-941, 1995 23. Remuzzi G, Ruggenenti P, Benigni A: Understanding the nature of renal disease progression. Kidney Int 51:2-15, 1997 24. Kant KS, Pesce AJ, Clyne DH, Pollack VE: Urinary
NaHCO3 TREATMENT AND RENAL INJURY
cast formation—pH-dependence and interaction of TammHorsfall mucoprotein with myoglobin, hemoglobin, Bence Jones protein, and albumin. Kidney Int 10:559A, 1976 (abstr) 25. Laurell CB: Electroimmunoassay. Scand J Clin Lab Invest 124:S21-S23, 1972 (suppl 1) 26. Fu¨hr Y, Kaczmarczk Y, Kruttgen GD: Eine einfache colorimetrische Methode zur Inulin-Bestimmung fu¨r NierenClearance-Untersuchungen bei Stoffwechselgesunden und Diabetikern. Klin Wochenschr 33:729-730, 1955 27. Haygen HN: The determination of ‘‘endogenous creatinine’’ in plasma and urine. Scand J Clin Lab Invest 5:48-57, 1953 28. Flanders KC, Thompson NL, Cissel DS, Van Obberghen-Schilling E, Baker CC, Kass ME, Ellingsworth LR, Roberts AB, Sporn MB: Transforming growth factor 1: Histochemical localization with antibodies to different epitopes. J Cell Biol 108:653-660, 1989 29. Kliem V, Johnson RJ, Alpers CE, Yoshimura A, Couser WG, Koch KM, Floege J: Mechanism involved in the pathogenesis of tubulointerstitial fibrosis in 5/6-nephrectomized rats.Kidney Int 49:666-678, 1996 30. Russ JC: The Image Processing Handbook (ed 2). Ann Arbor, MI, CRC, 1995 31. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Anal Biochem 162:156-159, 1987 32. Maniatis T, Fritsh EF, Sambrook J: Molecular Cloning: A Laboratory Manual (ed 2). Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press, 1989 33. Derynck RJ, Jarrett JA, Chen EY, Eaton DH, Bell JR, Assoian RK, Roberts AB, Sporn MB, Goeddel DV: Human transforming growth factor- complementary DNA sequence and expression in normal and transformed cells. Nature 316:701-705, 1985 34. Chan YL, Olivera J, Wood IG: The structure of rat 28 S ribosomal acid inferred from the sequence of nucleotides in a gene. Nucleic Acids Res 11:7819-7831, 1983 35. Feinberg AP, Vogelstein B: A technique for labelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13, 1983 36. Danielpour D, Kim KY, Dart LL, Watanabe S, Roberts AB, Sporn MB: Sandwich enzyme-linked immunosorbent assays quantitate and distinguish two forms of transforming growth factor-beta in complex biological fluids. Growth Factors 2:61-65, 1989 37. Gro¨ne H-J, Weber K, Gro¨ne E, Helmcken U, Osborn M: Coexpression of keratin and vimentin in damaged and regenerating tubular epithelia of the kidney. Am J Pathol 129:1-8, 1987 38. Kriz W, Elger M, Nagata M, Kretzler M, Uiker S, Koeppen Hageman I, Tenschert S, Lemley KV: The role of podocytes in the development of glomerular sclerosis. Kidney Int 45:S64-S72, 1994 (suppl 45) 39. Floege J, Johnson RJ, Alpers CE, Fatemi-Nainie S, Richardson CA, Gordon K, Couser WG: Visceral glomerular epithelial cells can proliferate in vivo and synthesize platelet-derived growth factor- chain. Am J Pathol 142:637650, 1993 40. Remuzzi G, Zoja C, Remuzzi A, Rossini M, Battaglia
337
C, Broggini M, Bertani T: Low-protein diet prevents glomerular damage in Adriamycin-treated rats. Kidney Int 28:21-27, 1985 41. Motomura K, Okuda S, Sanai T, Ando T, Onoyama K, Fujishima M: Importance of early initiation of dietary protein restriction for the prevention of experimental progressive renal disease. Nephron 49:144-149, 1988 42. Okuda S, Nakamura T, Yamamoto T, Ruoslahti E, Border WA: Dietary protein restriction rapidly reduces TGF-1 expression in experimental glomerulonephritis. Proc Natl Acad Sci U S A 88:9765-9769, 1991 43. Barretti P, Viero RM, Soares VA: Effects of dietary protein, angiotensin I converting enzyme inhibition and mesangial overload on the progression of Adriamycininduced nephropathy. Braz J Med Biol Res 28:39-50, 1995 44. Nakayama M, Okuda S, Tamaki K, Fujishima M: Short- or long-term effects of a low-protein diet on fibronectin and transforming growth factor-beta synthesis in Adriamycin-induced nephropathy. J Lab Clin Med 127:29-39, 1996 45. Kagami S, Border WA, Miller DE, Noble NA: Angiotensin II stimulates extracellular matrix protein synthesis through induction of transforming growth factor- expression in rat glomerular mesangial cells. J Clin Invest 93:24312437, 1994 46. Irwin KC, Brouhard BH, Satoh S, Stowe NT: Effects of enalapril on Adriamycin-induced nephrosis. Pediatr Nephrol 6:448-450, 1992 47. Jovanovic D, Dimitrijevic J, Varagic J, Jovovic D, Starcevic A, Djukanovic L: Effects of captopril on morphologic changes in kidney spontaneously hypertensive rats with Adriamycin nephropathy. Ren Fail 20:451-458, 1998 48. Hall RL, Wilke WL, Fettman MJ: The progression of Adriamycin-induced nephrotic syndrome in rats and the effect of captopril. Toxicol Appl Pharmacol 82:164-174, 1986 49. Beukers JJ, Hoedemaker PJ, Weening JJ: A comparison of the effects of converting-enzyme inhibition and protein restriction in experimental nephrosis. Lab Invest 59:631-640, 1988 50. Nomura A, Morita Y, Maruyama S, Hotta N, Nadai M, Wang L, Hasegawa T, Matsuo S: Role of complement in acute tubulointerstitial injury of rats with aminonucleoside nephrosis. Am J Pathol 151:539-547, 1997 51. Nagashima A, Okuda S, Tamaki K, Fujishima M: Influence of high-salt diet on glomerular injury and the preventive effects of amiloride in Adriamycin nephropathy. Nephron 71:87-94, 1995 52. Grainger DJ, Mosedale DE, Metcalfe JC, Weissberg PL, Kemp PR: Active and acid-activatable TGF- in sera, platelets and plasma. Clin Chim Acta 235:11-31, 1995 53. Noh JW, Wiggins RC, Phan SH: Urine transforming growth factor- activity is related to the degree of scarring in crescentic nephritis in rabbit. Nephron 63:73-78, 1993 54. Dominguez GCS, Costa RS, Dantas M, Kimachi T, Piucci CR, Coimbra TM: Transforming growth factor  activity in urine from patients with glomerulonephritis is related to their functional and structural changes. Nephrology 4:31-36, 1998