1607289
Enzyme 1991;45:188-193
©1991 S. Karger AG, Basel 0013-9432/91/0454-0188S2.75/0
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers in Cataract M.J.C. Crabbed S.T. Hoeb “Department of Microbiology, School of Animal and Microbial Sciences, University of Reading, and b Nuffield Laboratory of Ophthalmology, University of Oxford, UK
Key Words. Bendazac • Steroids • Cyanate • Aspirin • Aldehyde dehydrogenase • Aldose reductase • Cataract Abstract. Human lens was found to contain aldehyde dehydrogenase at a level of activity similar to that of bovine lens, namely 1.76 ± 0.51 IU/g. The enzyme, which appears to be a tetramer of 229 kD, was less susceptible to inhibition by cataractogenic agents than the bovine enzyme. The lipid peroxidation product malondialdehyde was a good substrate of the human lens enzyme. The in vitro aldose reductase reaction, which we have shown is caused by glyceraldehyde-stimulated free-radical NADPH oxidation, is inhibited by the potential anti-cataract agents, bendazac acid and bendazac lysine; these compounds also inhibit ferricytochrome c reduction in the presence of DL-glyceraldchyde and scavenge superoxide rad icals. These results are consistent with the hypotheses that aldehyde dehydrogenase is a protective enzyme in the human lens, and that the peroxy radical scavenging effects of bendazac acid and bendazac lysine contribute to their anti-cataract activity.
The enzyme aldose reductase has been implicated in the formation of cataract, par ticularly in diabetics, while a protective role in cataract has been assigned to the enzyme aldehyde dehydrogenase [1]. We have shown previously that levels of ‘aldose reductase’ found in the lens are due to artifacts of monosaccharide autoxidation, and that compounds known as ‘aldose reductase in hibitors’ probably act in a number of ways, including peroxy radical scavenging, that do
not relate to any specific enzyme inhibition [2] , Oxygen-derived free radicals have been implicated in many disease processes affect ing the eye, including cataract, while free radical scavengers have been shown to de toxify superoxide and hydroxyl radicals in ocular tissues [1]. Bendazac and its ana logues have been reported to prevent ultraviolet-induced dénaturation of lens proteins [3] , to prevent X-ray-induced damage to the rabbit lens [4], to prevent cyanate binding to lens proteins and cyanate-induced phaseseparation opacities [5], to scavenge hy Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Introduction
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers
droxyl and superoxide radicals [6], and to exhibit anti-cataract activity [7], An NAD+-dependent aldehyde dehydro genase that has been purified to homogene ity from bovine lens shows many features common to cytoplasmic liver aldehyde dehy drogenases, including a high Km for acetalde hyde, high sensitivity to disulfiram inactiva tion, deviation from Michaelis-Menten ki netics and a pi value of 5.2 [8], However, there are significant differences in native molecular weight (dimers rather than te tramere), amino acid composition, and the kinetic effects of metal ion-chelating re agents [9], While the baboon lens contains a typical cytoplasmic aldehyde dehydrogenase of unknown molecular weight [10], little or no information in available on the human lens enzyme. In this paper we attempt to study the human lens enzyme, its susceptibility to in hibition, its action on malondialdehyde, and the use of bendazac and its analogues as peroxy radical scavengers in an in vitro al dose reductase reaction.
189
was applied to the column, the homogenizing buffer described above being used as eluent. The fraction volume was 1 ml. Aldehyde Dehydrogenase Assay This was as described previously [8, 11], using sodium pyrophosphate buffer (50 mmol/1, pH 8.4), NAD+ (0.4 mmol/1) and substrate (acetaldehyde, 10 mmol/1 unless otherwise stated) at 37 °C. No alco hol dehydrogenase was detected under these condi tions when ethanol (10 or 35 mmol/1) was substituted for acetaldehyde. 1 IU of enzyme catalyses the pro duction of 1 pmol of NADH/min at pH 8.4 and 37 °C. Aldose Reductase Assay This was performed as described previously [12].
Ferncytochrome c Reduction Ferricytochrome c (5 pmol/1) was incubated with DL-glyceraldehyde (50 mmol/1) in potassium phos phate buffer (100 mmol/1, pH 6.2). The absorbance change at 550 nm was monitored using a Perkin Elmer 550 double-beam spectrophotometer as de scribed previously [13]. Blanks contained all reactants except glyceraldehyde. The inclusion of hypoxanthine (1 mmol/1), xanthine oxidase (10 pi) as described pre viously [13], together with EDTA (2 mmol/1) to lower hydroxyl radical formation through the Haber-Weiss reaction, allowed for the monitoring of peroxy (super oxide) radical scavenging.
Materials and Methods
Gel Chromatography A TSK HW55 column of void volume 72 ml was used at a flow rate of 15 ml/h at 4 ° C. 1 ml of human lens supernatant, protein concentration 39.3 mg/ml,
Results and Discussion The mean activity of aldehyde dehydro genase was 1.76 ± 0.51 IU/g of tissue. This compared closely with a corresponding value of 1.34 ± 0.11 IU/g of tissue with bovine lens. The human lens enzyme showed simi lar pH behaviour to the bovine lens enzyme, with the ratio of activity at pH 8.4 to that at pH 7.5 being 1.73 in both cases. Figure 1 shows the molecular weight determination of native human lens aldehyde dehydroge nase. The elution profile on the calibrated Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Human Lenses Clear human lenses from post-mortem eyes (4 male and 1 female; ages 40-65 years) were kept frozen in closed containers to prevent lyophilization. After thawing, lenses were homogenized (1 ml/lens) in icecold nitrogen-saturated phosphate buffer (20 mmol/1, pH 7.5) containing sodium azide (0.02%), EDTA (2 mmol/1) and ß-mercaptoethanol (10 mmol/1). The homogenate was centrifuged at 25,000 g for 40 min at 4°C.
190
Crabbe/Hoe
very high molecular weight material; a = a-crystallins; ß = ß-crystallins, with little resolution between heavy and light components; y = y-crystallins. Aldehyde dehydrogenase activity is expressed as the increase in absorption per minute at 340 nm due to the reduction of NAD+ to NADH. For each assay 100 pi of the eluant fraction was used in a final volume of 1 ml. Other details are in the text, o = Elution profile of human lens supernatant; ▼ = elution profile of alde hyde dehydrogenase activity.
column indicated a molecular weight of 229 kD, which, with a subunit molecular weight of 57 kD [8] suggests that the human lens enzyme, in common with most other aldehyde dehydrogenases, is a tetramer. However, it differs from the bovine lens en zyme, which is a dimer [8]. Table 1 shows the effects of cataractogenic agents as inhibi tors of the human lens aldehyde dehydroge nase activity. Generally, these agents are much poorer inhibitors of the human en zyme than of the bovine enzyme [11], Pro gesterone (10 (J.mol/1) gave only 30% inhibi
tion of the human enzyme, compared with 75% inhibition of the bovine enzyme [IllPrednisolone, a steroid which has been shown to cause cataract in experimental ani mals [1], was a very weak inhibitor, while cyanate, also known to cause cataract [14] and the protective compound aspirin [15] had only limited effects at concentrations greater than 1 mmol/1. The differences in the inhibitor specificity between the human and bovine enzymes may be related to the differ ences in the native molecular weights, and so potential differences in numbers of subDownloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Fig. 1. Elution of human lens aldehyde dehydroge nase from a calibrated column of TSK HW55 of void volume 72 ml. The arrows give the elution volumes of the molecular weight markers, which were as follows: ferritin (440 kD), 93 ml; catalase (232 kD), 101 ml; yglobulin (150kD), 103 ml; bovine serum albumin (67 kD). 106 ml; ovalbumin (43 kD), 111ml; car bonic anhydrase (31 kD), 119 ml; lysozyme (17kD, not shown), 149 ml. The following regions were iden tified with the aid of polyacrylamide SDS gels; H =
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers
191
Table 1. The effects of compounds upon the activity of the aldehyde dehydrogenase in human lens super natant Compound
Acetaldehyde 10 mmol/1
Control Progesterone (10 pmol/1)1 Prednisolone (100 pmol/1) Cyanate (10 mmol/1) Aspirin (4 mmol/1)
activity IU/g tissue
% activity
activity IU/g tissue
% activity
1.76
100 69 97 80 91
0.80 0.56 0.55 0.62 0.79
100 70 69 78 99
1.21
1.70 1.40 1.6
The assay with progesterone included 10% ethanol, which did not affect the activity of the enzyme.
strate/inhibitor-binding sites. Free radicalmediated lipid peroxidation, another poten tial cause of cataract, produces malondialdehyde. This aldehyde was a good substrate of the human lens enzyme, at an activity of 1.04 IU/g of tissue (1 mmol/1 concentration). Substrate inhibition was observed with malondialdehyde and, as was observed with the bovine enzyme [11], there was no irrevers ible inactivation by malondialdehyde. With malondialdehyde levels in the lens at about 2 nmol/g wet weight [16], the human lens therefore appears to contain levels of alde hyde dehydrogenase which would be ade quate to prevent cataract due to aldehydes binding to lens proteins [ 17], although it may not protect against intermediates of any free radical mechanisms which have been impli cated in cataract. Such mechanisms include monosaccha ride-induced autoxidation [13], which we have shown causes an ‘aldose reductase’ reaction by oxidizing NADPH to NADP+ in the presence of glyceraldehyde [12], Aldose
reductase had been thought to be a key en zyme in the aetiology of cataract in diabetics [18], although this has been severely criti cised [19], and aldose reductase inhibitors have recently been suggested to act by bind ing ‘decompartmentalized’ transition metals [20]. Both bendazac acid and bendazac ly sine at 5-mmol/l concentrations inhibited glyceraldehyde-stimulated NADPH oxida tion by 68%. This inhibition occurred in both the presence and absence of the NADPH-binding protein glutathione reduc tase, which enhances the free radical-me diated oxidation of the nucleotide. Figure 2a shows the effect of bendazac and its ana logues in modulating the rate of ferricytochrome c reduction, under conditions which allow DL-glyceraldehyde to autoxidise to produce hydroxyl, superoxide and carboncentred free radicals [13]. Reduced gluta thione and Trolox, a soluble vitamin E ana logue, scavenged the radicals produced by the autoxidizing glyceraldehyde, thus lower ing the rate of reduction. Bendazac acid and Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
l
0.25 mmol/1
192
Crabbe/Hoe
100
Enzyme 1991;45:188-193
©1991 S. Karger AG, Basel 0013-9432/91/0454-0188S2.75/0
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers in Cataract M.J.C. Crabbed S.T. Hoeb “Department of Microbiology, School of Animal and Microbial Sciences, University of Reading, and b Nuffield Laboratory of Ophthalmology, University of Oxford, UK
Key Words. Bendazac • Steroids • Cyanate • Aspirin • Aldehyde dehydrogenase • Aldose reductase • Cataract Abstract. Human lens was found to contain aldehyde dehydrogenase at a level of activity similar to that of bovine lens, namely 1.76 ± 0.51 IU/g. The enzyme, which appears to be a tetramer of 229 kD, was less susceptible to inhibition by cataractogenic agents than the bovine enzyme. The lipid peroxidation product malondialdehyde was a good substrate of the human lens enzyme. The in vitro aldose reductase reaction, which we have shown is caused by glyceraldehyde-stimulated free-radical NADPH oxidation, is inhibited by the potential anti-cataract agents, bendazac acid and bendazac lysine; these compounds also inhibit ferricytochrome c reduction in the presence of DL-glyceraldchyde and scavenge superoxide rad icals. These results are consistent with the hypotheses that aldehyde dehydrogenase is a protective enzyme in the human lens, and that the peroxy radical scavenging effects of bendazac acid and bendazac lysine contribute to their anti-cataract activity.
The enzyme aldose reductase has been implicated in the formation of cataract, par ticularly in diabetics, while a protective role in cataract has been assigned to the enzyme aldehyde dehydrogenase [1]. We have shown previously that levels of ‘aldose reductase’ found in the lens are due to artifacts of monosaccharide autoxidation, and that compounds known as ‘aldose reductase in hibitors’ probably act in a number of ways, including peroxy radical scavenging, that do
not relate to any specific enzyme inhibition [2] , Oxygen-derived free radicals have been implicated in many disease processes affect ing the eye, including cataract, while free radical scavengers have been shown to de toxify superoxide and hydroxyl radicals in ocular tissues [1]. Bendazac and its ana logues have been reported to prevent ultraviolet-induced dénaturation of lens proteins [3] , to prevent X-ray-induced damage to the rabbit lens [4], to prevent cyanate binding to lens proteins and cyanate-induced phaseseparation opacities [5], to scavenge hy Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Introduction
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers
droxyl and superoxide radicals [6], and to exhibit anti-cataract activity [7], An NAD+-dependent aldehyde dehydro genase that has been purified to homogene ity from bovine lens shows many features common to cytoplasmic liver aldehyde dehy drogenases, including a high Km for acetalde hyde, high sensitivity to disulfiram inactiva tion, deviation from Michaelis-Menten ki netics and a pi value of 5.2 [8], However, there are significant differences in native molecular weight (dimers rather than te tramere), amino acid composition, and the kinetic effects of metal ion-chelating re agents [9], While the baboon lens contains a typical cytoplasmic aldehyde dehydrogenase of unknown molecular weight [10], little or no information in available on the human lens enzyme. In this paper we attempt to study the human lens enzyme, its susceptibility to in hibition, its action on malondialdehyde, and the use of bendazac and its analogues as peroxy radical scavengers in an in vitro al dose reductase reaction.
189
was applied to the column, the homogenizing buffer described above being used as eluent. The fraction volume was 1 ml. Aldehyde Dehydrogenase Assay This was as described previously [8, 11], using sodium pyrophosphate buffer (50 mmol/1, pH 8.4), NAD+ (0.4 mmol/1) and substrate (acetaldehyde, 10 mmol/1 unless otherwise stated) at 37 °C. No alco hol dehydrogenase was detected under these condi tions when ethanol (10 or 35 mmol/1) was substituted for acetaldehyde. 1 IU of enzyme catalyses the pro duction of 1 pmol of NADH/min at pH 8.4 and 37 °C. Aldose Reductase Assay This was performed as described previously [12].
Ferncytochrome c Reduction Ferricytochrome c (5 pmol/1) was incubated with DL-glyceraldehyde (50 mmol/1) in potassium phos phate buffer (100 mmol/1, pH 6.2). The absorbance change at 550 nm was monitored using a Perkin Elmer 550 double-beam spectrophotometer as de scribed previously [13]. Blanks contained all reactants except glyceraldehyde. The inclusion of hypoxanthine (1 mmol/1), xanthine oxidase (10 pi) as described pre viously [13], together with EDTA (2 mmol/1) to lower hydroxyl radical formation through the Haber-Weiss reaction, allowed for the monitoring of peroxy (super oxide) radical scavenging.
Materials and Methods
Gel Chromatography A TSK HW55 column of void volume 72 ml was used at a flow rate of 15 ml/h at 4 ° C. 1 ml of human lens supernatant, protein concentration 39.3 mg/ml,
Results and Discussion The mean activity of aldehyde dehydro genase was 1.76 ± 0.51 IU/g of tissue. This compared closely with a corresponding value of 1.34 ± 0.11 IU/g of tissue with bovine lens. The human lens enzyme showed simi lar pH behaviour to the bovine lens enzyme, with the ratio of activity at pH 8.4 to that at pH 7.5 being 1.73 in both cases. Figure 1 shows the molecular weight determination of native human lens aldehyde dehydroge nase. The elution profile on the calibrated Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Human Lenses Clear human lenses from post-mortem eyes (4 male and 1 female; ages 40-65 years) were kept frozen in closed containers to prevent lyophilization. After thawing, lenses were homogenized (1 ml/lens) in icecold nitrogen-saturated phosphate buffer (20 mmol/1, pH 7.5) containing sodium azide (0.02%), EDTA (2 mmol/1) and ß-mercaptoethanol (10 mmol/1). The homogenate was centrifuged at 25,000 g for 40 min at 4°C.
190
Crabbe/Hoe
very high molecular weight material; a = a-crystallins; ß = ß-crystallins, with little resolution between heavy and light components; y = y-crystallins. Aldehyde dehydrogenase activity is expressed as the increase in absorption per minute at 340 nm due to the reduction of NAD+ to NADH. For each assay 100 pi of the eluant fraction was used in a final volume of 1 ml. Other details are in the text, o = Elution profile of human lens supernatant; ▼ = elution profile of alde hyde dehydrogenase activity.
column indicated a molecular weight of 229 kD, which, with a subunit molecular weight of 57 kD [8] suggests that the human lens enzyme, in common with most other aldehyde dehydrogenases, is a tetramer. However, it differs from the bovine lens en zyme, which is a dimer [8]. Table 1 shows the effects of cataractogenic agents as inhibi tors of the human lens aldehyde dehydroge nase activity. Generally, these agents are much poorer inhibitors of the human en zyme than of the bovine enzyme [11], Pro gesterone (10 (J.mol/1) gave only 30% inhibi
tion of the human enzyme, compared with 75% inhibition of the bovine enzyme [IllPrednisolone, a steroid which has been shown to cause cataract in experimental ani mals [1], was a very weak inhibitor, while cyanate, also known to cause cataract [14] and the protective compound aspirin [15] had only limited effects at concentrations greater than 1 mmol/1. The differences in the inhibitor specificity between the human and bovine enzymes may be related to the differ ences in the native molecular weights, and so potential differences in numbers of subDownloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
Fig. 1. Elution of human lens aldehyde dehydroge nase from a calibrated column of TSK HW55 of void volume 72 ml. The arrows give the elution volumes of the molecular weight markers, which were as follows: ferritin (440 kD), 93 ml; catalase (232 kD), 101 ml; yglobulin (150kD), 103 ml; bovine serum albumin (67 kD). 106 ml; ovalbumin (43 kD), 111ml; car bonic anhydrase (31 kD), 119 ml; lysozyme (17kD, not shown), 149 ml. The following regions were iden tified with the aid of polyacrylamide SDS gels; H =
Aldehyde Dehydrogenase, Aldose Reductase, and Free Radical Scavengers
191
Table 1. The effects of compounds upon the activity of the aldehyde dehydrogenase in human lens super natant Compound
Acetaldehyde 10 mmol/1
Control Progesterone (10 pmol/1)1 Prednisolone (100 pmol/1) Cyanate (10 mmol/1) Aspirin (4 mmol/1)
activity IU/g tissue
% activity
activity IU/g tissue
% activity
1.76
100 69 97 80 91
0.80 0.56 0.55 0.62 0.79
100 70 69 78 99
1.21
1.70 1.40 1.6
The assay with progesterone included 10% ethanol, which did not affect the activity of the enzyme.
strate/inhibitor-binding sites. Free radicalmediated lipid peroxidation, another poten tial cause of cataract, produces malondialdehyde. This aldehyde was a good substrate of the human lens enzyme, at an activity of 1.04 IU/g of tissue (1 mmol/1 concentration). Substrate inhibition was observed with malondialdehyde and, as was observed with the bovine enzyme [11], there was no irrevers ible inactivation by malondialdehyde. With malondialdehyde levels in the lens at about 2 nmol/g wet weight [16], the human lens therefore appears to contain levels of alde hyde dehydrogenase which would be ade quate to prevent cataract due to aldehydes binding to lens proteins [ 17], although it may not protect against intermediates of any free radical mechanisms which have been impli cated in cataract. Such mechanisms include monosaccha ride-induced autoxidation [13], which we have shown causes an ‘aldose reductase’ reaction by oxidizing NADPH to NADP+ in the presence of glyceraldehyde [12], Aldose
reductase had been thought to be a key en zyme in the aetiology of cataract in diabetics [18], although this has been severely criti cised [19], and aldose reductase inhibitors have recently been suggested to act by bind ing ‘decompartmentalized’ transition metals [20]. Both bendazac acid and bendazac ly sine at 5-mmol/l concentrations inhibited glyceraldehyde-stimulated NADPH oxida tion by 68%. This inhibition occurred in both the presence and absence of the NADPH-binding protein glutathione reduc tase, which enhances the free radical-me diated oxidation of the nucleotide. Figure 2a shows the effect of bendazac and its ana logues in modulating the rate of ferricytochrome c reduction, under conditions which allow DL-glyceraldehyde to autoxidise to produce hydroxyl, superoxide and carboncentred free radicals [13]. Reduced gluta thione and Trolox, a soluble vitamin E ana logue, scavenged the radicals produced by the autoxidizing glyceraldehyde, thus lower ing the rate of reduction. Bendazac acid and Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 4:19:02 PM
l
0.25 mmol/1
192
Crabbe/Hoe
100
