Identification and Characterization of Aldose Reductase in Cultured Rat Mesangial Cells RYUICHI KIKKAWA, KISABURO UMEMURA, MASAKAZU HANEDA, NOBUYUKI KAJIWARA, SHIRO MAEDA, CHIHIRO NISHIMURA, YUKIO SHIGETA
Although the enhanced activity of the polyol pathway has been detected in diabetic glomeruli, the intraglomerular localization of this pathway has not yet been well defined. In this study, we attempted to identify aldose reductase, a key enzyme of the polyol pathway, in cultured rat mesangial cells and to characterize the properties of this enzyme using enzymological and immunological methods. When the aldose reductase (DL-glyceraldehyde-reducing) activity was analyzed in mesangial cell extract, the Lineweaver-Burk plot showed concave downward curvature, and the Michaelis constant was 0.83 mM DL-glyceraldehyde, and this activity was noncompetitively inhibited by an aldose reductase inhibitor, ICI-128,436. The enzyme activity was enhanced by the addition of sulfate ion and partially suppressed by barbital. The enzyme cross-reacted with the antisera against rat lens and testis aldose reductases on Ouchterlony plate, and migrated to the region of molecular weight of about 36,500 Da on Western blotting. The presence of aldose reductase mRNA was also confirmed by Northern analysis using cDNA for rat aldose reductase, 10Q. From these results, it was concluded that the aldose reductase may exist in rat glomerular mesangial cells and may play a role in the development of diabetic glomerulopathy, though the coexistence of aldehyde reductase(s) may not be fully ruled out. Diabetes 41:1165-71, 1992
From the Third Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga, Japan; and the National Eye Institute, National Institutes of Health, Bethesda, Maryland. Address correspondence and reprint requests to Dr. R. Kikkawa, Third Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga 520-21, Japan. Received for publication 8 October 1991 and accepted in revised form 30 March 1992.
DIABETES, VOL. 41, SEPTEMBER 1992
T
he presence of the polyol pathway in various tissues has been suggested to have pathological importance in the development of diabetic complications including diabetic nephropathy (1-3). Beyer-Mears et al. (4) have reported the intraglomerular polyol accumulation in experimental diabetic rats. Recently, we also biochemically demonstrated the presence of the polyol pathway in glomerular mesangial cells (5) and suggested that the impairment of glomerular contractility in diabetes could be attributed to the mesangial cell dysfunction induced by the abnormality in the polyol pathway (6). Because the previous report failed to detect aldose reductase, a key enzyme of the polyol pathway in mesangial cells (7), we attempted in this study to confirm the presence of aldose reductase (aldehyde reductase 2, EC 1. 1. 1. 21) and to further characterize the enzyme in mesangial cells using enzymological and immunological techniques as well as a cDNA probe for rat aldose reductase.
RESEARCH DESIGN AND METHODS Mesangial cell culture. Male Sprague-Dawley rats weighing 50-100 g were decapitated and renal glomeruli were isolated by sequential sieving as previously reported (8). The glomerular preparation with a purity over 95% was used for the experiment. Glomerular mesangial cells were obtained by culturing these glomeruli with RPMI1640 medium supplemented with 20% fetal bovine serum (Gibco, Grand Island, NY), 100 U/ml penicillin G and 100 |xg/ml streptomycin sulfate as previously described by Kreisberg and Kamovsky (9). The cells were identified as mesangial cells by the following criteria: 1) The cells had D-amino acid oxidase, determined by incubating them with a medium containing D-valine substituted for L-valine; 2) they were resistant to puromycin
1165
Aldose Reductase in Mesangial Cells
FIG. 1. Transmission electron microscopic observation of mesangial cells stained with peroxidase-conjugated heavy melomyosin. The longitudinal section (A) or the cross section (B) shows the bundles of actin filaments running parallel to the plasma membrane.
aminonucleoside (10 txg/ml) but susceptible to mitomycin C (10 M-g/ml); 3) they had a large number of intracellular actin bundles stained by peroxidase-conjugated heavy melomyosin (Fig. 1); 4) they had receptors for angiotensin II (All), which were analyzed according to the method of Kitamura et al. (10); 5) they contracted in response to All; and 6) they possessed the cell-surface antigen of Thy-1, but not the la antigen (9,11). Cultured cells from the 5th to 15th passages were used for the following experiments. During these passages, the cells were morphologically identical and the activity of aldose reductase was kept constant. Properties of aldose reductase activity. Cultured cells in dishes were homogenized in pH 7.4 20 mM sodiumphosphate buffer, with a Polytron homogenizer (Kinematica, Lucerne, Switzerland) for 30 s, and the homogenate was centrifuged at 2000 x gr for 15 min. The supernatant was used for the measurement of aldose reductase (AR) activity. AR activity was assayed by the method of Gabbay and Kinoshita using DL-glyceraldehyde as a substrate (13). In brief, the reaction mixture consisted of 135 mM Na/K phosphate buffer (pH 6.2), 200 n-M NADPH, 4 mM DL-glyceraldehyde, and an enzyme sample in a final volume of 1 ml. The decrease in absorbance at 340 nm
1166
was monitored for 20 min at 30°C after the addition of substrate, and the enzyme activity was expressed as micromoles NADPH oxidized per minute. Kinetic studies were performed in the presence or absence of AR inhibitor, ICI-128,436. The enzyme activity was also examined in the presence of lithium sulfate, ammonium sulfate, or barbital to determine whether its properties are similar to those of other tissue ARs. Furthermore, the inhibition of AR activity by antiserum against rat lens AR was studied with the addition of increasing amounts of antisera. Namely, 200 |xl of the enzyme sample was preincubated with antisera or normal goat sera (0.5-25 nl) at 37°C for 15 min. Then, the incubated sample (180 |xl) was assayed for its AR activity. Partial purification of aldose reductase. The supernatant of mesangial cell homogenate (~35 mg protein) was processed for a 30-70% ammonium sulfate precipitation. The precipitate dissolved in a minimum amount of distilled water (-10 ml) was dialyzed against 0.1 M Na/K phosphate buffer (pH 6.2) containing 5 mM mercaptoethanol and 0.02% sodium azide (buffer A). The dialyzed sample was concentrated on ultrafilter with UP-20 membrane to 1 ml (-7 mg protein), which was placed on a Sephadex G-100 column (10 x 300 mm) and eluted with buffer A at a rate of 0.25 ml/min (1 ml/tube). Fractions containing AR activity were pooled and concentrated on a ultrafilter with UP-20 membrane (Toyoroshi, Tokyo, Japan). The final preparation was used as partially purified AR of mesangial cell. The protein determination was performed according to the method of Lowry et al. (12). Ammonium sulfate precipitations (30-70%) of rat lens and testis homogenates were also conducted, and the precipitates were used for comparison with mesangial cell AR. Immunodiffusion and Western blot analysis of AR. The partially purified ARs from mesangial cells, lens, and testis (protein concentration 4 - 5 mg/ml in each sample) were subjected to double immunodiffusion against antirat testis or lens AR antisera. Furthermore, the mesangial AR preparation was applied to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) conducted according to the method of Laemmli (14) using 14% acrylamide gel. Thereafter, the separated proteins in the gel were transferred to a sheet of nitrocellulose filter (0.45 M-m pore size, Schleicher & Schuell, Dassel, Germany) by Western blot technique (15). After transfer, the sheet was reacted with anti-rat testis AR antiserum (x500 dilution), then visualized with alkaline phosphatase-conjugated anti-rabbit IgG goat serum. The molecular weight of mesangial cell AR was estimated from the position of molecular weight markers. Detection of aldose reductase mRNA. Mesangial cells in culture were rinsed twice with ice-cold phosphatebuffered saline (pH 7.4), scraped off and solubilized in the denaturing solution (4 M guanidinium thiocyanate, 25 mM sodium citrate (pH 7.4), 0.5% sodium lauroylsarcosine, 0.1 M 2-mercaptoethanol). Total RNA was then isolated using the method described by Chomczynski and Sacchi (16). Rat testis was also homogenized in the denaturing solution and total RNA was prepared in the
DIABETES, VOL 41, SEPTEMBER 1992
R. Kikkawa and Associates
TABLE 1 The effect of lithium sulfate, ammonium sulfate, or barbital on aidose reductase activity in mesangial cell Enzyme activity (%)
Activator or inhibitor
100.0 121.0 123.1 142.9 60.0
None Li2SO4 (0.1 M) Li2SO4 (0.3 M) (NH4)2SO4 (0.3 M) Barbital (1.0 mM)
±5.9 ±2.6* ± 4.3* ± 10.0* ± 8.6*
Activities were determined in the presence of 4 mM DL-glyceraldehyde and 200 n-M NADPH. Values represent relative activities to that without inhibitor or activator (none) taken as 100%, and were expressed as mean ± SD (n = 3). The absolute value of none was 13.45 x 10~3 jjumol NADPH oxidized/min/mg protein. *P< 0.01 vs. none.
same fashion. Twenty micrograms RNA were denatured with glyoxal at 50°C for 60 min, separated on a 1% agarose gel, and transferred to nylon filter. The filter was then hybridized with EcoRI fragments of rat aidose reductase cDNA 10Q (17), labeled with 32P by multiprime DNA labeling system (Amersham, UK), by the method described previously (18). RESULTS Enzymological properties of mesangial cell AR. ARlike activity of mesangial cell homogenate was significantly activated by sulfate ions up to 40%, and was
partially (40%) inhibited by barbital (Table 1). The Lineweaver-Burk plot of this enzyme exhibited concave downward curvature, and the Michaelis constant of this enzyme in the higher range of the substrate was 0.83 mM of DL-glyceraldehyde (Fig. 2A). Moreover, a potent AR inhibitor, ICI-128,436 inhibited the enzyme activity noncompetitively (Fig. 28). Furthermore, the AR-like activity was significantly inhibited by the addition of antiserum against rat lens AR, whereas the control goat serum did not inhibit the AR-like activity significantly as shown in Fig. 3. Immunochemical properties of partially purified AR. Partially purified AR in mesangial cells was obtained by the Sephadex G-100 column chromatography of ammonium sulfate precipitate (Fig. 4). A couple of fractions around the peak of AR activity eluted after the major protein peak were used as partially purified mesangial cell AR for following experiments. The AR preparation of mesangial cells produced precipitation bands on the Ouchterlony plate with antiserum against testis AR. Moreover, those precipitation bands were fused with those bands between partially purified AR from lens or testis and antiserum (Fig. 5). A similar observation was obtained with the antiserum against rat lens AR (data not shown). In the SDS-PAGE and western blot procedures using partially purified mesangial cell AR and anti-testis AR antibody, only a single band of immunoreactive AR was observed at the point of molecular weight, 36,500 kD (Fig. 6), which was clearly separated
B
I
Q.
c E •a
a>
4 -
N •p
'x o
xa. a< "5 E
2 -
h 10
1/[ Glyceraldehyde (mM) ]
1/[Glyceraldehyde (mM) ]
FIG. 2. Llneweaver-Burk plot of the mesangial cell aidose reductase with DL-glyceraldehyde and NADPH as substrates. A: the reaction system contains 66,100, or 200 \t,M of NADPH with varying concentrations of DL-glyceraldehyde. B: the reaction performed in the presence or absence of an aidose reductase Inhibitor, ICI-128,436 (1
Although the enhanced activity of the polyol pathway has been detected in diabetic glomeruli, the intraglomerular localization of this pathway has not yet been well defined. In this study, we attempted to identify aldose reductase, a key enzyme of the polyol pathway, in cultured rat mesangial cells and to characterize the properties of this enzyme using enzymological and immunological methods. When the aldose reductase (DL-glyceraldehyde-reducing) activity was analyzed in mesangial cell extract, the Lineweaver-Burk plot showed concave downward curvature, and the Michaelis constant was 0.83 mM DL-glyceraldehyde, and this activity was noncompetitively inhibited by an aldose reductase inhibitor, ICI-128,436. The enzyme activity was enhanced by the addition of sulfate ion and partially suppressed by barbital. The enzyme cross-reacted with the antisera against rat lens and testis aldose reductases on Ouchterlony plate, and migrated to the region of molecular weight of about 36,500 Da on Western blotting. The presence of aldose reductase mRNA was also confirmed by Northern analysis using cDNA for rat aldose reductase, 10Q. From these results, it was concluded that the aldose reductase may exist in rat glomerular mesangial cells and may play a role in the development of diabetic glomerulopathy, though the coexistence of aldehyde reductase(s) may not be fully ruled out. Diabetes 41:1165-71, 1992
From the Third Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga, Japan; and the National Eye Institute, National Institutes of Health, Bethesda, Maryland. Address correspondence and reprint requests to Dr. R. Kikkawa, Third Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga 520-21, Japan. Received for publication 8 October 1991 and accepted in revised form 30 March 1992.
DIABETES, VOL. 41, SEPTEMBER 1992
T
he presence of the polyol pathway in various tissues has been suggested to have pathological importance in the development of diabetic complications including diabetic nephropathy (1-3). Beyer-Mears et al. (4) have reported the intraglomerular polyol accumulation in experimental diabetic rats. Recently, we also biochemically demonstrated the presence of the polyol pathway in glomerular mesangial cells (5) and suggested that the impairment of glomerular contractility in diabetes could be attributed to the mesangial cell dysfunction induced by the abnormality in the polyol pathway (6). Because the previous report failed to detect aldose reductase, a key enzyme of the polyol pathway in mesangial cells (7), we attempted in this study to confirm the presence of aldose reductase (aldehyde reductase 2, EC 1. 1. 1. 21) and to further characterize the enzyme in mesangial cells using enzymological and immunological techniques as well as a cDNA probe for rat aldose reductase.
RESEARCH DESIGN AND METHODS Mesangial cell culture. Male Sprague-Dawley rats weighing 50-100 g were decapitated and renal glomeruli were isolated by sequential sieving as previously reported (8). The glomerular preparation with a purity over 95% was used for the experiment. Glomerular mesangial cells were obtained by culturing these glomeruli with RPMI1640 medium supplemented with 20% fetal bovine serum (Gibco, Grand Island, NY), 100 U/ml penicillin G and 100 |xg/ml streptomycin sulfate as previously described by Kreisberg and Kamovsky (9). The cells were identified as mesangial cells by the following criteria: 1) The cells had D-amino acid oxidase, determined by incubating them with a medium containing D-valine substituted for L-valine; 2) they were resistant to puromycin
1165
Aldose Reductase in Mesangial Cells
FIG. 1. Transmission electron microscopic observation of mesangial cells stained with peroxidase-conjugated heavy melomyosin. The longitudinal section (A) or the cross section (B) shows the bundles of actin filaments running parallel to the plasma membrane.
aminonucleoside (10 txg/ml) but susceptible to mitomycin C (10 M-g/ml); 3) they had a large number of intracellular actin bundles stained by peroxidase-conjugated heavy melomyosin (Fig. 1); 4) they had receptors for angiotensin II (All), which were analyzed according to the method of Kitamura et al. (10); 5) they contracted in response to All; and 6) they possessed the cell-surface antigen of Thy-1, but not the la antigen (9,11). Cultured cells from the 5th to 15th passages were used for the following experiments. During these passages, the cells were morphologically identical and the activity of aldose reductase was kept constant. Properties of aldose reductase activity. Cultured cells in dishes were homogenized in pH 7.4 20 mM sodiumphosphate buffer, with a Polytron homogenizer (Kinematica, Lucerne, Switzerland) for 30 s, and the homogenate was centrifuged at 2000 x gr for 15 min. The supernatant was used for the measurement of aldose reductase (AR) activity. AR activity was assayed by the method of Gabbay and Kinoshita using DL-glyceraldehyde as a substrate (13). In brief, the reaction mixture consisted of 135 mM Na/K phosphate buffer (pH 6.2), 200 n-M NADPH, 4 mM DL-glyceraldehyde, and an enzyme sample in a final volume of 1 ml. The decrease in absorbance at 340 nm
1166
was monitored for 20 min at 30°C after the addition of substrate, and the enzyme activity was expressed as micromoles NADPH oxidized per minute. Kinetic studies were performed in the presence or absence of AR inhibitor, ICI-128,436. The enzyme activity was also examined in the presence of lithium sulfate, ammonium sulfate, or barbital to determine whether its properties are similar to those of other tissue ARs. Furthermore, the inhibition of AR activity by antiserum against rat lens AR was studied with the addition of increasing amounts of antisera. Namely, 200 |xl of the enzyme sample was preincubated with antisera or normal goat sera (0.5-25 nl) at 37°C for 15 min. Then, the incubated sample (180 |xl) was assayed for its AR activity. Partial purification of aldose reductase. The supernatant of mesangial cell homogenate (~35 mg protein) was processed for a 30-70% ammonium sulfate precipitation. The precipitate dissolved in a minimum amount of distilled water (-10 ml) was dialyzed against 0.1 M Na/K phosphate buffer (pH 6.2) containing 5 mM mercaptoethanol and 0.02% sodium azide (buffer A). The dialyzed sample was concentrated on ultrafilter with UP-20 membrane to 1 ml (-7 mg protein), which was placed on a Sephadex G-100 column (10 x 300 mm) and eluted with buffer A at a rate of 0.25 ml/min (1 ml/tube). Fractions containing AR activity were pooled and concentrated on a ultrafilter with UP-20 membrane (Toyoroshi, Tokyo, Japan). The final preparation was used as partially purified AR of mesangial cell. The protein determination was performed according to the method of Lowry et al. (12). Ammonium sulfate precipitations (30-70%) of rat lens and testis homogenates were also conducted, and the precipitates were used for comparison with mesangial cell AR. Immunodiffusion and Western blot analysis of AR. The partially purified ARs from mesangial cells, lens, and testis (protein concentration 4 - 5 mg/ml in each sample) were subjected to double immunodiffusion against antirat testis or lens AR antisera. Furthermore, the mesangial AR preparation was applied to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) conducted according to the method of Laemmli (14) using 14% acrylamide gel. Thereafter, the separated proteins in the gel were transferred to a sheet of nitrocellulose filter (0.45 M-m pore size, Schleicher & Schuell, Dassel, Germany) by Western blot technique (15). After transfer, the sheet was reacted with anti-rat testis AR antiserum (x500 dilution), then visualized with alkaline phosphatase-conjugated anti-rabbit IgG goat serum. The molecular weight of mesangial cell AR was estimated from the position of molecular weight markers. Detection of aldose reductase mRNA. Mesangial cells in culture were rinsed twice with ice-cold phosphatebuffered saline (pH 7.4), scraped off and solubilized in the denaturing solution (4 M guanidinium thiocyanate, 25 mM sodium citrate (pH 7.4), 0.5% sodium lauroylsarcosine, 0.1 M 2-mercaptoethanol). Total RNA was then isolated using the method described by Chomczynski and Sacchi (16). Rat testis was also homogenized in the denaturing solution and total RNA was prepared in the
DIABETES, VOL 41, SEPTEMBER 1992
R. Kikkawa and Associates
TABLE 1 The effect of lithium sulfate, ammonium sulfate, or barbital on aidose reductase activity in mesangial cell Enzyme activity (%)
Activator or inhibitor
100.0 121.0 123.1 142.9 60.0
None Li2SO4 (0.1 M) Li2SO4 (0.3 M) (NH4)2SO4 (0.3 M) Barbital (1.0 mM)
±5.9 ±2.6* ± 4.3* ± 10.0* ± 8.6*
Activities were determined in the presence of 4 mM DL-glyceraldehyde and 200 n-M NADPH. Values represent relative activities to that without inhibitor or activator (none) taken as 100%, and were expressed as mean ± SD (n = 3). The absolute value of none was 13.45 x 10~3 jjumol NADPH oxidized/min/mg protein. *P< 0.01 vs. none.
same fashion. Twenty micrograms RNA were denatured with glyoxal at 50°C for 60 min, separated on a 1% agarose gel, and transferred to nylon filter. The filter was then hybridized with EcoRI fragments of rat aidose reductase cDNA 10Q (17), labeled with 32P by multiprime DNA labeling system (Amersham, UK), by the method described previously (18). RESULTS Enzymological properties of mesangial cell AR. ARlike activity of mesangial cell homogenate was significantly activated by sulfate ions up to 40%, and was
partially (40%) inhibited by barbital (Table 1). The Lineweaver-Burk plot of this enzyme exhibited concave downward curvature, and the Michaelis constant of this enzyme in the higher range of the substrate was 0.83 mM of DL-glyceraldehyde (Fig. 2A). Moreover, a potent AR inhibitor, ICI-128,436 inhibited the enzyme activity noncompetitively (Fig. 28). Furthermore, the AR-like activity was significantly inhibited by the addition of antiserum against rat lens AR, whereas the control goat serum did not inhibit the AR-like activity significantly as shown in Fig. 3. Immunochemical properties of partially purified AR. Partially purified AR in mesangial cells was obtained by the Sephadex G-100 column chromatography of ammonium sulfate precipitate (Fig. 4). A couple of fractions around the peak of AR activity eluted after the major protein peak were used as partially purified mesangial cell AR for following experiments. The AR preparation of mesangial cells produced precipitation bands on the Ouchterlony plate with antiserum against testis AR. Moreover, those precipitation bands were fused with those bands between partially purified AR from lens or testis and antiserum (Fig. 5). A similar observation was obtained with the antiserum against rat lens AR (data not shown). In the SDS-PAGE and western blot procedures using partially purified mesangial cell AR and anti-testis AR antibody, only a single band of immunoreactive AR was observed at the point of molecular weight, 36,500 kD (Fig. 6), which was clearly separated
B
I
Q.
c E •a
a>
4 -
N •p
'x o
xa. a< "5 E
2 -
h 10
1/[ Glyceraldehyde (mM) ]
1/[Glyceraldehyde (mM) ]
FIG. 2. Llneweaver-Burk plot of the mesangial cell aidose reductase with DL-glyceraldehyde and NADPH as substrates. A: the reaction system contains 66,100, or 200 \t,M of NADPH with varying concentrations of DL-glyceraldehyde. B: the reaction performed in the presence or absence of an aidose reductase Inhibitor, ICI-128,436 (1
