Elevated plasma levels of homocysteine are a risk factor for cardiovascular diseases, neural tube defects, and Alzheimer's disease. The transsulfuration pathway converts homocysteine to cysteine, and ∼50% of the cysteine in glutathione is derived from homocysteine in human liver cells, which suggests the hypothesis that defects in the transsulfuration pathway perturb redox homeostasis. To test this hypothesis, we examined a murine model for hyperhomocysteinemia in which the gene encoding the first enzyme in the transsulfuration pathway, cystathionine β-synthase (CBS), has been disrupted. Limited metabolite profiling and CBS expression studies in liver, kidney, and brain reveal tissue-specific differences in the response to Cbs disruption. Homozygous disruption of Cbs lowered cysteine concentration in all three organs. Glutathione concentration was diminished in liver and brain, thus affecting the redox buffering capacity in these organs, whereas the approximately twofold higher glutathione synthesis capacity in kidney helped preserve the glutathione pool size despite loss of the transsulfuration pathway in this organ. In contrast, disruption of a single Cbs allele elicited only minor redox perturbations. Furthermore, the Cbs+/− genotype did not confer a significant disadvantage compared with the Cbs+/+ genotype in hepatocytes challenged by oxidative stress from exposure to tertiary butylhydroperoxide. These studies provide evidence that homozygous disruption of Cbs perturbs redox homeostasis and reduces cysteine levels, raising the possibility that these changes may be important in the etiology of the clinical manifestations of CBS deficiency.
- cystathionine β-synthase
hyperhomocysteinema, a condition in which plasma levels of total homocysteine (tHcy) are elevated, is a risk factor for atherosclerotic diseases, including peripheral arterial occlusive, coronary artery, and cerebrovascular diseases (15, 17, 28). Plasma tHcy is a graded risk factor and it is estimated that for every 5 μmol/l increase in tHcy the odds ratio for coronary artery disease increases 1.6-fold (4). Severe hyperhomocysteinemia can be inherited as an autosomal recessive disorder that results from impairments in enzymes involved in homocysteine metabolism, namely, cystathionine β-synthase (CBS), methionine synthase, or methylenetetrahydrofolate reductase. Homocysteine is an intermediary metabolite in methionine metabolism that has two major fates: transmethylation, which salvages it back to the methionine cycle, and transsulfuration, which commits it to the synthesis of cysteine and also provides an avenue for catabolism (Fig. 1). Homocysteine metabolism is uniquely dependent on multiple B vitamins, including B6, B12, and folic acid. Insufficiency of these nutrients is correlated with hyperhomocysteinemia (11, 29, 34). Aberrations in CBS, which are the single most common genetic cause of severe hyperhomocysteinemia, are associated with pleiotropic clinical manifestations, including mental retardation, ectopia lentis, skeletal abnormalities, osteoporosis, and fatty liver (25). The most frequent cause of morbidity in these patients is premature atherosclerosis with thromboembolic complications.
The transsulfuration pathway converts homocysteine to cysteine, the limiting reagent in the synthesis of GSH, a major intracellular antioxidant in mammals. The intracellular concentration of GSH is high, varying from ∼1 to 10 mM depending on the cell type (3, 33). The transsulfuration pathway has limited tissue distribution and is found mainly in tissues that have the largest stores of GSH and consequently the highest demand on the intracellular cysteine pool. In human liver cells, it is estimated that ∼50% of the cysteine in GSH is derived from homocysteine (24). The quantitatively significant contribution of the transsulfuration pathway to the major intracellular redox buffer suggests the hypothesis that CBS deficiency perturbs redox homeostasis.
A mouse model of Cbs gene disruption has been developed (35). Homozygous Cbs-deficient (Cbs−/−) mice are born at the expected frequency from heterozygote matings but suffer from growth retardation, have severely elevated plasma tHcy, and exhibit markedly reduced viability. In contrast, the heterozygotes (Cbs+/−) have moderately elevated tHcy but are otherwise indistinguishable from wild-type mice. Interestingly, when a combination of dietary (high methionine or low folate) and genetic (Cbs+/+ vs. Cbs+/− mice) factors was used to modulate tHcy levels, endothelial dysfunction was observed and the severity was associated with plasma tHcy concentration (8, 18). In an independent study, heterozygous Cbs deficiency in mice was associated with impaired endothelium-dependent vasodilation and it was suggested that increased oxidative stress may play a role in reducing nitric oxide bioactivity under these conditions (10). Endothelial dysfunction associated with mild hyperhomocysteinemia has also been reported in primates and humans (2, 19, 38).
In this study, we employed the Cbs-deficient mouse model for limited metabolite profiling, which reveals tissue-specific differences in manifestations of CBS deficiency. Interestingly, changes in redox status as reflected by the [GSH]-to-[GSSG] ratio and in the GSH pool size are observed in the heterozygous and homozygous Cbs-deficient mice, respectively. Hepatocytes from Cbs+/− mice exhibit lower viability compared with Cbs+/+ mice when exposed to tertiary butyl hydroperoxide (t-BuOOH). In addition, cysteine levels are decreased in the homozygous knockouts. These results suggest that some of the clinical manifestations of CBS deficiency may be mediated by cysteine deficiency and perturbations in redox homeostasis.
MATERIALS AND METHODS
This study was approved by the Institutional Animal Care and Use Committees at the University of Nebraska, the University of Iowa, and Baylor College of Medicine. Breeding colonies of CBS-deficient mice (in a C57BL/6J background) were maintained at the University of Iowa and age-matched Cbs+/+ and +/− mice were shipped to Nebraska where they were maintained on the same diets purchased from Harlan Teklad: LM-485 chow (control diet), which contains 6.7 mg/kg folic acid and 4.0 g/kg methionine, and TD00205 chow (high-methionine/low-folate diet), which contains 0.2 mg/kg folic acid and 8.2 g/kg methionine. Genotyping for the targeted CBS allele was performed by the polymerase chain reaction (35). All the mice were littermates. Males and females were separated at the time of weaning (3 wk of age), housed in cages containing 2–4 animals, and provided with unlimited amounts of food and water. Male and female mice were studied at the same time, and metabolites from the two genders were assayed at the same time.
Animals were anesthetized with an intraperitoneal injection of Beuthanasia-D Special containing 9.5 mg pentobarbital sodium and 1.25 mg phenytoin sodium (Schering-Plough Animal Health) per mouse, and blood was collected from the retroorbital vein and immediately transferred to weighed sample tubes containing 50 μl EDTA solution (50 mM in PBS). Blood samples were centrifuged and plasma samples were collected and stored at −80°C until further use for measurement of tHcy, methionine, and cysteine. The mice were killed by exsanguination, and the brain, kidneys, and liver were removed rapidly for ex vivo studies, freeze-clamped in liquid nitrogen, and stored at −80°C until further use. Male and female mice were killed on the same day and in no particular order. Organs from 2-wk-old Cbs−/− and +/+ mice were removed as described above and shipped frozen from Baylor College of Medicine to the University of Nebraska for analysis.
Analysis of CBS expression and metabolite levels in the liver, kidney, and brain of Cbs+/+ and +/− mice was performed on 7- to 20-wk-old mice. Because of the low viability of Cbs−/− mice, they were killed at 2 wk of age and compared with Cbs+/+ mice of the same age.
The frozen organs were pulverized in liquid nitrogen using a porcelain mortar and pestle, and the powder was collected in weighed sample tubes containing a known volume of the desired sample preparation solution as described below. Tubes were vortexed gently and weighed, and the samples were used for analysis or stored at −80°C until further use.
Measurement of CBS activity.
CBS activity was measured in powdered organs suspended in a 1.15% KCl solution using the ninhydrin assay as described in Ref. 16.
Measurement of tissue metabolite levels.
To measure tissue concentrations of reduced and oxidized GSH and cysteine, protein in powdered organs was precipitated with metaphosphoric acid solution (containing 16.8 mg/ml HPO3, 2 mg/ml EDTA, and 9 mg/ml NaCl) and removed by centrifugation. Thiol metabolites in protein-free extracts were derivatized with monoiodoacetic acid followed by 2,4-dinitrofluorobenzene and analyzed by HPLC using μ-Bondapak-NH2 300 × 3.9 mm column (Waters) as described previously (24, 27).
To measure ascorbate concentrations, protein was precipitated with 5% trichloroacetic acid solution in powdered organ samples and removed by centrifugation. Ascorbate in protein-free extracts was measured by reduction of ferric ion by ascorbic acid and monitoring formation of ferrous ion as a red-orange α,α′-dipyridyl complex (26, 39).
Measurement of plasma metabolite levels.
tHcy, methionine, and cysteine were assayed by stable isotope dilution capillary gas chromatography/mass spectrometry as previously described (31).
Measurement of GSH synthesis capacity.
GSH synthesis capacity was measured using the fluorescent method described in Refs. 12 and 13 with minor modifications. Tissue powder (100–150 mg) was suspended in 1 ml sample buffer containing (in mM) 100 Tris·HCl (pH 7.3), 150 KCl, 20 MgCl2, and 2 EDTA, and samples were vortexed and centrifuged. The supernatant was diluted 1:1 with the sample buffer and incubated at 37°C for 10 min. Then 50–200 μl of the sample was added to a thermostated spectrofluorimeter cuvette (maintained at 37°C) containing 0.78 ml of sample buffer supplemented with (in mM) 12.8 glutamate, 12.8 glycine, 3.8 ATP, and 0.13 monochlorobimane (Molecular Probes). In the presence of glutathione-S-transferase, which is present in cell extracts, monochlorobimane reacts with reduced GSH, forming a fluorescent product with maximal excitation at 390 nm and maximal emission at 478 nm. GSH synthesis in the cuvette was initiated by addition of 20 μl of a cysteine (15 mM) and DTT (15 mM) mixture, and the increase in fluorescence was monitored over 10 min.
The background fluorescence was measured in the same sample diluted 1:1 with the sample buffer containing 10 mM buthionine sulfoximine, a γ-glutamylcysteine ligase inhibitor, and incubated at 37°C for 10 min. Then the change in fluorescence was measured as described above, and the background rate was subtracted from the experimental rate.
A calibration factor used for conversion of fluorescence intensity to GSH concentration was determined for each sample. To this end, the sample diluted 1:1 with the sample buffer and preincubated at 37°C for 10 min was mixed with 0.78 ml of the sample buffer containing 0.13 mM monochlorobimane, and the initial fluorescence intensity was measured after 5 min incubation at 37°C. Then 10 μl of 1 mM GSH solution was added to the cuvette, and the final fluorescence intensity was measured.
Western blot analysis of CBS.
For Western blot analysis of CBS protein levels, tissue powder was suspended in lysing buffer [0.1 M sodium phosphate, pH 7.4, containing 0.1% Triton X-100, 10 μl/ml protease inhibitor cocktail for use with mammalian cells and tissue extracts (Sigma), 25 μg/ml tosyllysine chloromethylketone, 25 μg/ml phenylmethylsulfonyl fluoride, 27 μg/ml aprotinin, and 10 μg/ml leupeptin]. The mixture was incubated on ice for 30 min and then centrifuged at 15,000 g for 10 min at 4°C, and equal amounts of protein were loaded per lane. Polyclonal antibodies against CBS, available in our laboratory, were employed for detection of this enzyme. The secondary antibody, anti-rabbit horseradish peroxidase (Amersham Biosciences), was diluted 1:25,000 and detected using the Chemiluminescent Peroxidase Substrate kit (Sigma). Protein bands (run at least in triplicate) were quantitated using the Quantity One software from Bio-Rad and normalized to the level of actin in each sample, which was employed as an equal loading control.
Isolation and incubation of hepatocytes.
Hepatocytes were isolated from Cbs+/+ and Cbs+/− mice as follows. Livers were perfused for 5 min through the vena porta with KHR solution (in mM: 115 NaCl, 5 KCl, 1 KH2PO4, and 25 HEPES, pH 7.4), saturated with 95% O2-5% CO2 at 39°C, followed by 7–8 min perfusion with the same solution containing 2 mM CaCl2 and 56 μg/ml collagenase-protease mixture (Liberase Blendzyme 3, Roche). The final activities of collagenase and protease in the perfusing solution were 0.203 and 108 U/ml, respectively. The rate of perfusion was 7 ml/min. Then the liver was dispersed in KHR solution containing 2 mM CaCl2 and 1% BSA at 37°C. Hepatocytes were filtered through a 100-μm nylon mesh, sedimented by low-speed centrifugation, washed with 45 ml of 30% Percoll in KHR with 2 × 30 ml KHR-CaCl2-BSA, and kept on ice. Cell concentration and viability were determined using a hemocytometer and Trypan blue staining, respectively.
Cells were suspended in Williams E Media (Sigma, W-4125) supplemented with 5% FBS, NaHCO3 (2.2 g/l), HEPES (15 mM), penicillin (100 000 U/l), streptomycin (100 mg/l), and gentamicin (5 mg/ml) at a concentration of ∼0.3 × 106 viable cells per milliliter. The suspension was dispensed aseptically into a 24-well cell culture plate (0.5 ml/well) coated with collagen I. Plates were incubated for cell adhesion for 3 h at 37°C and in a 5% CO2 atmosphere. The medium was then removed and cells were washed with Williams E media lacking FBS and maintained in this media (0.5 ml/well) for the following 12 h at 37°C and 5% CO2 atmosphere. After the medium was changed, t-BuOOH solution in ethanol was added to cells (2.5 μl/well). To control cells, only ethanol (2.5 μl/well) was added.
After 12 h of incubation, the medium was removed and cells were washed with PBS and 0.3 ml/well of staining solution was added. The staining solution was prepared using the LIVE/DEAD Kit (Molecular Probes) and contained 2 μM calcein AM and 3 μM ethidium homodimer-1 in PBS. Cells were incubated with the staining solution for 30 min at 37°C. Then 250 μl of staining solution was removed from each well, and digitized images from the central area of each well were collected using a dual-band filter (FITC/TRITC) and a ×4 objective lens (the final magnification was ×40) in a Nikon Eclipse TE300 inverted microscope with a MagnaFire color digital camera (Optronics Inc). All images were recorded using the same microscope and camera settings. Images were analyzed quantitatively using the SIS AnalySIS-Opti imaging analysis program (Soft Imaging System, Lakewood, CO) using the same preset thresholds for the areas labeled by green fluorescence in each frame. For each experiment, the average area covered by green fluorescence in two to four samples was determined and presented as a percentage of that obtained from the control samples run in parallel. The total number of experiments performed with Cbs+/+ and Cbs+/− mice were 10 and 8, respectively.
Statistical analysis was performed using the one-way ANOVA test included in Origin 7.0 software. A P value of <0.05 was used to define statistical significance.
Plasma concentrations of metabolites.
Plasma tHcy concentration in Cbs+/− mice was approximately twofold higher than in Cbs+/+ mice (Table 1) as reported previously (35). In contrast, methionine and cysteine levels were not significantly affected by the Cbs genotype. Both Cbs+/+ and Cbs+/− mice maintained on a high-methionine/low-folate diet displayed tHcy levels that were significantly higher than in animals maintained on a control diet (P < 10−7). Plasma levels of methionine and cysteine were not influenced by the high-methionine/low-folate diet (Table 1).
Tissue distribution of CBS.
As expected, CBS activity in different tissues varied over a significant range, being highest in the liver, approximately threefold lower in kidney, and ∼30-fold lower in brain (Table 2). The average CBS activity in the liver of Cbs+/− mice was ∼40% of the value in Cbs+/+ mice (P < 10−13) as seen previously (35) and was associated with a corresponding decrease in CBS protein levels as determined by Western blot analysis (Fig. 2A). Interestingly, although CBS activity (Table 2) tended to be decreased in the brain of Cbs+/− mice relative to Cbs+/+ mice, the difference did not reach statistical significance (P > 0.1). In contrast, CBS expression in brain as judged by Western analysis was decreased more than twofold in Cbs+/− animals (Fig. 2C).
Conspicuous differences in CBS activity and in CBS protein levels were discernable in the kidneys of males vs. female mice (Table 2, Fig. 2B). In Cbs+/+ mice, CBS activity in male kidneys was 2.7-fold higher than in female kidneys (P < 10−6). In heterozygous males, kidney CBS activity and protein levels were decreased to ∼50% of control values (P < 10−6). CBS activity in the kidneys of female Cbs+/+ and +/− mice was not significantly different (P > 0.2), but the expression level as judged by Western analysis was more than twofold lower in Cbs+/− than wild-type females (Fig. 2B).
CBS protein and activity could not be detected in tissues of Cbs−/− animals (not shown). The average CBS activity in the liver and brain of 2-wk-old Cbs+/+ mice was 84.9 ± 23.7 and 3.0 ± 0.4 mmol·h−1·kg tissue−1, respectively, which is similar to the values seen in the 7- to 20-wk-old mice (Table 2).
Effect of heterozygous CBS gene disruption on GSH-dependent redox homeostasis.
To test the hypothesis that disruption of the transsulfuration pathway perturbs GSH-linked redox homeostasis, we monitored the GSH, GSSG, and cysteine levels in heterozygous and homozygous Cbs-deficient mice relative to wild-type controls. The [GSH]-to-[GSSG] ratio is an indicator of intracellular redox status and varies in a tissue-specific manner. A significant difference in GSH concentration was not observed in organs of Cbs+/+ vs. +/− animals (Table 3) or between males and females (not shown). However, the [GSH]-to-[GSSG] ratio was decreased in kidney (P < 0.02) and tended to be decreased in liver (P < 0.06) but not in brain (P > 0.2) of Cbs+/− mice compared with Cbs+/+ mice.
The capacity for GSH synthesis as indicated by the combined activities of γ-glutamylcysteine ligase and GSH synthetase in liver (12.6 ± 5.5 in Cbs+/+ vs. 11.5 ± 4.1 mmol· h−1·kg tissue−1 in Cbs+/-) and kidney (22.7 ± 5.0 in +/+ vs. 23.6 ± 2.8 mmol·h−1·kg tissue−1 in Cbs+/−) was indistinguishable between Cbs+/+ and +/− mice. Interestingly, the observed variation in brain GSH concentration between mice of a given genotype was found to be very small and suggests that the GSH level is tightly regulated in murine brain. The concentration of the precursor amino acid cysteine did not differ significantly between Cbs+/+ and +/− mice in the liver, kidney, or brain. However, a sex-specific difference in cysteine levels that parallels the difference in CBS activity in kidney (Table 2) was observed with the concentration being higher in male vs. female kidney (P < 0.02) (Table 3).
Effect of homozygous Cbs gene disruption on GSH-dependent redox homeostasis.
In contrast to the heterozygotes, the homozygous Cbs-deficient mice displayed diminished GSH levels in the liver (1.6-fold lower, P < 0.01) and brain (1.4-fold lower, P < 10−5) but not kidney (P > 0.4) (Table 4). A diminution in cysteine levels was also observed in liver and brain of Cbs−/− mice. In kidney, the comparison between Cbs+/+ and Cbs−/− mice was slightly more complex because of the underlying gender effect. Male Cbs+/+ mice tended to have a higher concentration of cysteine than female Cbs+/+ mice (P < 0.08) and this effect was lost in Cbs−/− mice (Table 4). It should be noted that due to the limited viability of Cbs−/− mice, 2-wk-old mice were employed and the comparison involved only three mice in each category, i.e., male and female. Significantly, the GSH concentration but not the [GSH]-to-[GSSG] ratio was altered in liver or brain of Cbs−/− mice (Table 4).
Ascorbate concentration in wild-type and Cbs knockouts.
Metabolism of GSH and ascorbate appear to be related and mice, unlike humans, are able to synthesize this antioxidant vitamin (14, 21, 22). We therefore decided to examine the ascorbate concentrations in the three groups of animals but did not observe a significant genotype-dependent difference in any organ (Tables 3 and 4).
Viability of hepatocytes from Cbs+/+ and Cbs+/− mice under oxidative stress conditions.
Under steady-state conditions, disruption of a single Cbs allele has a modest effect on the redox state altering the [GSH]-to-[GSSG] ratio but not the GSH pool size (Table 3). When the viability of hepatocytes challenged with 100 μM t-BuOOH was compared, a difference was observed between the Cbs+/+ and Cbs+/− genotypes (Fig. 3). However, this difference (P < 0.07) did not reach statistical significance. Furthermore, a difference was not observed at lower (50 μM, Fig. 3B) or higher (200 μM, not shown) concentrations of the organic peroxide. It is likely that because the flux through CBS is low under steady-state conditions (24), an increase in flux can accommodate the oxidative challenge at low t-BuOOH concentrations under our experimental conditions. At higher peroxide (200 μM) concentrations, viability is very low (<20% after 12 h) and the cells are apparently overwhelmed with oxidative stress obscuring genotypic differences at the Cbs locus, at least under the conditions used in this study.
Despite the clear epidemiological evidence implicating hyperhomocysteinemia as a risk factor for occlusive disease in the coronary, cerebral, and peripheral arteries (4–6), as well as neural tube defects (23, 32) and Alzheimer's disease (7), our understanding of homocysteine regulation and of the etiology of homocysteine-related diseases remains poor. Homocysteine is a sulfur-containing amino acid that is an important intermediate in the metabolism of the essential amino acid methionine. Conversion of homocysteine to cysteine via the transsulfuration pathway plays a quantitatively significant role in human liver cells in the biosynthesis of GSH, a major intracellular antioxidant (24). In this study, we found that CBS deficiency perturbs the intracellular cysteine pool and GSH-dependent redox homeostasis and elucidated organ-specific differences in CBS expression and response to Cbs gene disruption. These effects may be relevant to the etiology of some of the clinical manifestations of CBS deficiency that affects four major organ systems: ocular, cardiovascular, central nervous system, and skeletal.
Homocysteine levels in Cbs+/+ and Cbs+/− mice were modulated by a combination of high methionine and low folate in the diet as described previously (8, 18). Although plasma tHcy levels increased three- to sixfold in mice on the high-methionine/low-folate diet vs. the control diet (Table 1), we did not observe a diet-dependent difference in the GSH level in brain, liver, or kidney (not shown) and therefore used mice on normal diets for the remainder of the study.
Disruption of a single CBS allele led to an expected approximately twofold decrease in CBS enzyme activity and protein levels in liver (Fig. 2 and Table 2) as reported previously (35). However, tissue- and gender-specific differences in the manifestations of Cbs gene disruption were observed that have not been described previously. CBS activity in the brain of heterozygotes was 79% of that seen in wild-type mice (Table 2), although CBS protein levels were more than twofold lower (Fig. 2). In kidney, an unexpected gender effect on CBS expression was observed in Cbs+/+ mice, with males exhibiting approximately twofold higher enzyme levels and activity compared with females (Table 2 and Fig. 2B). A similar pattern was observed in two other mouse lines, BALB/c and Black Swiss (data not shown), demonstrating that the sexual dimorphism in kidney CBS expression was not specific to the C57BL/6J strain. Heterozygous disruption of Cbs led to the expected approximately twofold decrease in enzyme activity in kidneys of males but not females, which showed ∼75% of wild-type activity despite a more than twofold decrease in CBS expression (Fig. 2B). Thus in brain and in female kidney, where the steady-state level of CBS is low, compensatory changes in regulation apparently occur on loss of a single allele to help maintain CBS activity. The magnitude of the change in enzyme activity did not match the change in CBS protein levels and suggests a posttranslational mechanism of regulation. Homozygous disruption of the Cbs gene led to undetectable CBS protein levels and activity in all three organs, as expected (not shown).
The physiological significance of increased CBS activity in male mouse kidney is not known. One possibility is that sulfur-containing catabolites formed downstream of the transsulfuration pathway may be used for territorial markings by males. Interestingly, sexual dimorphism, albeit in the opposite direction, has been noted for CBS in a gene expression study of Drosophila development (1).
Disruption of a single Cbs allele was associated with a decrease in the [GSH]-to-[GSSG] ratio in kidney and a borderline significant decrease in the [GSH]-to-[GSSG] ratio in liver (Table 3). The latter is consistent with only a borderline significant advantage conferred by the Cbs+/+ vs. the Cbs+/− genotype on viability of hepatocytes exposed to oxidative stress (Fig. 3). Increased oxidative stress and impaired GSH peroxidase activity has been reported in the liver of Cbs+/− mice (36). In this latter study, administration of the cysteine donor, l-2-oxothiazolidine-4-carboxylic acid, to Cbs+/− mice increased cellular thiol and reduced GSH pools and restored normal vasodilator function (36). These findings are also consistent with previous observations that overexpression of GSH peroxidase rescues hyperhomocysteinemic mice from endothelial dysfunction (37) and that deficiency of GSH peroxidase sensitizes hyperhomocysteinemic mice to endothelial dysfunction (9).
Homozygous disruption of the CBS gene resulted in 1.6- and 1.4-fold lower concentrations of GSH in the liver and brain, respectively, indicating lowered redox buffering capacity due to loss of the transsulfuration pathway (Table 4). However, the [GSH]-to-[GSSG] ratio was unchanged, indicating that the reduced GSH pool size did not lead to perturbations in the redox status per se, at least under steady-state conditions. In contrast, the GSH concentration in kidney was not affected by loss of the transsulfuration pathway. In this organ, intracellular cysteine concentrations are higher in adult mice (Table 4) and the GSH synthesis capacity as measured by the activities of γ-glutamyl cysteine ligase and GSH synthetase is twofold higher than in liver. Together, these factors are likely to shield the kidney from loss of the transsulfuration pathway-derived pool of intracellular cysteine. The absence of an effect of complete disruption of the transsulfuration pathway on GSH concentration begs the question as to what the role of this pathway is in the kidney. Its importance could lie in catabolism of the sulfur-containing amino acids, methionine and homocysteine, leading to sulfate formation or to taurine biosynthesis.
Although the change in GSH level in the brain of Cbs−/− mice is smaller than in liver, the ascorbate concentration is correspondingly higher (Table 4). A similar situation is seen in a heterozygous knockout for γ-glutamyl-cysteine ligase in mouse, where the GSH level is diminished by 20% and is accompanied by a compensatory increase in the ascorbate pool by 30% (30). Ascorbate levels did not vary in the liver of Cbs−/− mice, although the GSH concentration was lower.
The transsulfuration pathway provides an avenue for conversion of the essential amino acid, methionine, to the nonessential one, cysteine. Decreased cysteine levels were observed in Cbs−/− mice (Table 4). These observations have important implications for some of the clinical manifestations unique to homozygous Cbs deficiency in humans, including skeletal and ocular connective tissue abnormalities. These clinical features are not observed in hyperhomocysteinemic patients with severe hyperhomocysteinemia due to impairments in enzymes in the transmethylation branch of the pathway, which implicates an etiological role for cysteine deficiency rather than elevated homocysteine. In support of this hypothesis, in vitro studies on cultured arterial smooth muscle cells demonstrated that low cysteine but not high homocysteine inhibits deposition of fibrillin-1 (20). Our studies reveal the magnitude of cysteine deficiency despite the availability of cysteine from the diet in a mouse model for homozygous CBS deficiency.
In conclusion, our studies reveal organ-specific adjustments to allelic disruption of the Cbs gene in the heterozygous and homozygous states. Importantly, a lower redox buffering capacity is observed in liver and brain of homozygotes as judged by a lower GSH concentration, but reduced GSH concentration in the brain may be offset by increased ascorbate in the same organ. Furthermore, cysteine concentration is decreased in all three organs in Cbs−/− mice that were examined in this study. In contrast, the redox perturbations are minor in Cbs+/− mice and the presence of somewhat more oxidizing conditions in the liver and kidney of heterozygotes is suggested by a lower [GSH]-to-[GSSG] ratio. Understanding the mechanistic basis for the tissue (and gender)-specific responses to CBS deficiency will be important for elucidating their connection to the etiology of the consequent spectrum of clinical manifestations.
This work was supported by grants from the National Institutes of Health (DK-64959 to R. Banerjee, AG-09834 to S. Stabler, HL-67033 to H. Wang, and HL-63943 to S. R. Lentz); an Established Investigator Award from the American Heart Association to R. Banerjee; a Junior Faculty Scholar Award from American Society of Hematology to H. Wang; support from the Office of Research and Development, Department of Veterans Affairs to S. R. Lentz; and a postdoctoral fellowship from the American Heart Association to S. Dayal.
We gratefully acknowledge Dr. C. Casey, University of Nebraska Medical Center, for assistance with hepatocyte isolation.
- Copyright © 2004 the American Physiological Society