Effect of procysteine on aging-associated changes in hepatic GSH and SMase: evidence for transcriptional regulation of smpd3.

In hepatocytes, aging-associated decline in GSH has been linked to activation of neutral SMase (nSMase), accumulation of bioactive ceramide, and inflammation. In this study, we seek to test whether dietary supplementation with the cysteine precursor, L-2-oxothiazolidine-4-carboxylic acid (OTC), would correct the aging-associated differences in hepatic GSH, nSMase, and ceramide. Young and aged mice were placed on a diet that either lacked sulfur-containing amino acids (SAAs) or had 0.5% OTC for 4 weeks. Mice fed standard chow were used as an additional control. SAA-deficient mice exhibited significant aging-associated differences in hepatic GSH, GSH/GSSG, ceramide, and nSMase. C24:1 ceramide, the major ceramide species in liver, was affected the most by aging, followed by the less abundant C16:0 ceramide. OTC supplementation eliminated the aging-associated differences in hepatic GSH and GSH/GSSG ratio. Surprisingly, however, instead of decreasing, the nSMase activity and ceramide increased in the OTC-fed mice irrespective of their age. These effects were due to elevated nSMase-2 mRNA and protein and appeared to be direct. Similar increases were seen in HepG2 cells following treatment with OTC. The OTC-fed aged mice also exhibited hepatic steatosis and triacylglyceride accumulation. These results suggest that OTC is a potent stimulant of nSMase-2 expression and that there may be unanticipated complications of OTC supplementation.


Animals and diets
Young (4 months) and aged (22 months) male C57Bl6 mice were purchased from the National Institute of Aging (Bethesda, MD) and housed in the Association for Assessment and Accreditation of Laboratory Animal Care-approved animal facility at the University of Kentucky Medical Center. The animals were maintained at 12 h light/dark cycle in micro-isolation and had unlimited access to water and standard (Std) chow diet (2918, Teklad-Global, 18% protein rodent diet; Harlan-Teklad, Indianapolis, IN). After a 1 week acclimatization period, mice were randomly placed on three different diets: sulfur-containing amino acid defi cient (SAA def.) control diet (TD.08437, Harlan-Teklad); SAA def. diet supplemented with 5 g/kg OTC (0.5% OTC diet) (TD.99366, Harlan-Teklad); or continued on the Std chow diet for a period of 4 weeks. At the end of the 4 week period, mice were deeply anesthetized, blood was collected by heart puncture, and the serum was obtained in serum separator tubes. Liver and other tissues were dissected, weighed, fl ash-frozen in liquid nitrogen, and kept at Ϫ 80°C for later processing. All experiments with animals were included in our animal protocol, which was approved by the Institutional Animal Care and Use Committee and carried out accordance with the recommendations of American Veterinary Medical Association.

Analysis of ceramide species by HPLC-ESI/tandem mass spectrometry
Total lipid extracts were prepared from 10 mg of liver tissue, using acidifi ed organic solvents ( 21 ) with the addition of C17 ceramide (Avanti Polar Lipids) as an internal standard. The organic phase was collected, evaporated under N 2 , and then the dried lipids were reconstituted in methanol. Ceramide species were quantitated by HPLC-ESI/tandem mass spectrometry using a 4000 Q-Trap hybrid linear ion trap triple-quadrupole mass spectrometer as described earlier ( 21 ). Recovery was determined by reference to the internal standards and quantifi cation was accomplished using calibration curves generated using synthetic standards for each ceramide species (Avanti Polar Lipids). Total ceramide mass was calculated as the sum of all detected ceramide species.

Determination of hepatic glutathione content
Whole liver homogenates (20%) were prepared in a buffer containing 50 mM Tris-HCl (pH, 7.4), 1 mM EDTA, and protease The effects GSH has on cellular functions are mediated through changes in the cellular redox state, as well as through direct protein modifi cations causing allosteric activation or inhibition of function. Studies have suggested that GSH also targets a lipid-signaling enzyme, neutral SMase (nSMase)-2 (24)(25)(26)(27)(28). nSMase-2 is a SM-hydrolyzing enzyme that is activated during cellular stress by cytokines such as TNF-␣ and IL-1 ␤ to generate ceramide ( 29,30 ). The latter is a bioactive metabolite that activates distinct sets of downstream target proteins, including Stress-activated protein kinase/c-Jun N terminal kinase (SAPK/JNK), ERK, Protein phosphatase 2A (PP2A), and Protein phosphatase 1 (PP1) (31)(32)(33)(34), as well as p38 ( 35 ) and Protein Kinase R (PKR) ( 36 ), which in turn propagate infl ammation or induce cell death and growth arrest. Several studies have suggested that in vitro GSH acts as an inhibitor of nS-Mase-2 activity, which helps maintain low basal nSMase activity in a healthy state. Addition of GSH to cell lysates, or to partially purifi ed nSMase-2, inhibits nSMase enzymatic activity, suggesting that it might be a direct inhibitor of the enzyme (26)(27)(28). It has been suggested that depletion of cellular GSH in conditions associated with elevated oxidative stress (including aging) triggers the activation of nSMase-2 and the accumulation of ceramide, and thus causes cellular death/infl ammation ( 28 ).
This model was previously studied in the context of aging using primary hepatocytes ( 28 ). Hepatocytes isolated from aged rats showed very low GSH levels, accounting for only 30% of the GSH content of hepatocytes from young rats. This was paralleled by a 30-40% increase in total cellular nSMase activity. Gene silencing approaches indicated that nSMase-2 was responsible for this agingassociated increase in nSMase activity. Treatment of hepatocytes from young animals with buthionine sulphoximine, an inhibitor of GSH synthase, decreased cellular GSH content and activated nSMase-2 to values similar to those observed in aged hepatocytes. Inversely, the addition of N -acetyl cysteine to hepatocytes from aged animals successfully restored GSH content and nSMase-2 activity. The upregulation of hepatic infl ammatory response observed in the aged animals decreased after replenishment of cellular GSH or inhibition of nSMase-2. These experiments suggested that age-associated depletion of GSH in old animals might be important for the onset of a chronic infl ammatory phenotype via upregulation of nSMase-2 activity. The link between GSH and nSMase-2, however, has never been studied in vivo. Thus, the goal of the present study was to test the relations between aging, hepatic GSH, and nSMase-2 activity in mice. The specifi c hypothesis was that restoration of GSH content in aged livers by dietary supplementation with a cysteine pro-drug, OTC, would inhibit nSMase-2 activity and would have benefi cial effects on hepatic functions in the aged mice. The study, however, unexpectedly showed that OTC has novel GSH-independent effects on nSMase-2 mRNA levels. These data are the fi rst evidence for transcriptional regulation of nSMase-2 mRNA in the liver, which is physiologically relevant and determines the overall hepatic nSMase activity and ceramide content. USA, Inc., Torrance, CA) and frozen for future processing. Frozen liver tissue samples were sectioned (8-10 m), stained with either hematoxylin and eosin (H and E) or Oil Red-O, and observed under a light microscope.

Triacylglycerol determination
The mass of triacylglycerols (TAGs) in the liver was measured in total lipid extracts using an L-type Triglyceride M kit from Wako Chemicals USA, Inc. (Richmond, VA) following the manufacturer's instructions.
Quantitative real-time PCR analyses were performed using TaqMan® gene expression assays (Applied Biosystems) specifi c for mouse (Mm00491359_m1) and human (Hs00920354_m1) nSMase-2 gene. Reactions were done in 96-well plates according to the manufacturer's instructions and analyzed in absolute quantitation experiments using Applied Biosystems's 7500 real-time PCR system equipped with SDSv1.2x software. Expression assays for ␤ -actin (Hs99999903_m1, human; Mm00607939_s1, mouse) were used as endogenous normalization controls. Levels of nSMase-2 mRNA were normalized to ␤ -actin mRNA.

Serum ALP and ALT activity assays
In vitro quantitative determination of ALP and ALT activities was carried out in mouse serum using DiscretePak™ ALP and DiscretePak™ ALT reagent kits (Catachem, Bridgeport, CT) following the manufacturer's recommendations in 10 l (for ALT) or 6 l (for ALP) of serum. Absorbance was monitored at 340 nm (for ALT) and 405 nm (for ALP) for up to 10 min. Final values were calculated according to the instructions and normalized per gram of liver.

Impact of diets on body and liver weight
Studies in hepatocytes isolated from young and aged animals consistently show a 50-60% decrease in GSH content during aging. The goal of this study was fi rst to test whether similar differences are seen in vivo and to determine if dietary supplementation with a GSH precursor, OTC, could alleviate the aging-associated differences. Assessment of hepatic GSH content in animals, however, is diffi cult. Fluctuations in endogenous GSH content due to the different nutritional status of each animal, as well as differences in metabolism, has led to inconsistency in the results of several major studies ( 37-44 ) with a similar goal. and phosphatase inhibitors. GSH+GSSG and GSSG contents were determined in 50 l aliquots using the Bioxytech® GSH/ GSSG-412™ assay kit according to the manufacturer's instructions. GSH was calculated as GSSG values were subtracted from the total glutathione (GSH+GSSG).

Cell culture
HepG2 cells (ATCC, Manassas, VA) were maintained in MEM (Gibco, Invitrogen) supplemented with 10% FBS (Atlanta-Biologicals, Atlanta, GA). Cells were seeded in 6-well plates (100,000 cells/well) and allowed 18 h to attach to the bottom. Cell culture medium was replaced with fresh medium supplemented with the respective concentrations of OTC in the beginning of the treatment and every third day after that. Cells were harvested 1, 3, or 6 days after treatment, as indicated.

SMase activity assays
Liver homogenates were prepared in a buffer containing 50 mM Tris-HCl (pH, 7.4), 1 mM ␤ -mercaptoethanol ( ␤ -Me), 0.1% Triton X-100, phosphatase inhibitors (1 mM Na 2 VO 4 , 1 mM NaF), and 1:100 (v/v) protease inhibitor cocktail (Sigma-Aldrich). nSMase and acidic SMase (ASMase) activities were measured using C 6 -NBD-SM from Molecular Probes Inc. as a substrate. Liver lysates (80 g protein/assay) or HepG2 cell lysates in PBS (20 g protein/assay) were incubated in a reaction buffer containing 50 mM Tris-HCl (pH 7.4), 7.5 mM MgCl 2 , and 10 M C 6 -NBD-SM, for nSMase activity. Because the inclusion of ␤ -Me in the homogenates/assays might infl uence the OTC effects (in case these were mediated via redox-sensitive mechanisms), preliminary assays were done where the concentration of ␤ -Me in the lysis or assay buffers was varied (from 0 to 10 mM). The results from these studies indicated that albeit ␤ -Me inclusion affected the basal nSMase activity to some extent, it did not alter the elevation seen in the OTC-treated groups. For ASMase activity, liver lysates were incubated in 420 mM acetate buffer (pH 4.5) containing 10 M C 6 -NBD-SM. Reactions were allowed to proceed for 1 h at 37°C and stopped by the addition of 1 ml of methanol. After another incubation at 37°C for 30 min, the samples were centrifuged at 1,000 g for 5 min and the generation of the fl uorescent product, C6-NBD-ceramide ((6-((N-(7-Nitrobenz-2-Oxa-1,3-Diazol-4-yl)amino)hexanoyl)Sphingosine), was monitored by a reverse phase HPLC using methanol:water: phosphoric acid (850:150:0.15, by volume) as a mobile phase ( 30 ).

Liver histology
Immediately upon dissection, part of the liver tissue was embedded in OCT compound from Tissue-Tek® (Sakura Finetek  Diets were purchased from Harlan-Teklad. SAA def. and 0.5% OTC diets were custom-made amino aciddefi ned diets. Std chow was an 18% protein rodent diet from natural sources.
a Amino acid profi le and other ingredients in the diets are presented as g/kg of dry weight diet.
In an attempt to alleviate these inherent limitations, three different dietary regimens were used with young (4 months) and aged (22 months) C57Bl6 mice. In the fi rst group, mice were fed regular chow diet and served as a Std control. To diminish the effects of the GSH and GSH precursors present in the regular rodent chow, a second group of mice was placed on a diet of defi ned amino acid composition that was devoid of the two GSH precursors, cysteine and methionine (SAA def. diet). A third group, consisting of age-and gender-matched animals, was fed the same SAA def. diet supplemented with OTC (see Table 1 for the exact composition of the diets). Mice were placed on these diets for 4 weeks. Signifi cant diet-and aging-associated differences in the body and liver weight (supplementary Table I Fig. I) remained well below the range typically associated with toxicity (i.e., between 2,000 and 10,000 units/l).

Aging-associated changes in hepatic GSH/GSSG levels and nSMase activity in vivo
In mice fed the Std diet, no statistically signifi cant agingassociated changes in hepatic GSH content or GSH/GSSG ratio were measurable ( Fig. 1A, B ). The aging-associated differences in nSMase activity also did not reach statistical signifi cance ( Fig. 1C ). These data are in contrast to our earlier observations for primary hepatocytes, where a 70% drop in the total GSH content and almost 2-fold elevation in nSMase activity were observed in hepatocytes isolated from aged rats as compared with those from young rats ( 28 ).
Aging-associated differences in GSH levels and in nSMase activity were readily apparent in mice on the SAA def. diet. As expected, there was an overall decline in total hepatic GSH content in all SAA def. diet-fed mice. However, the hepatic GSH content ( Fig. 1D ) and GSH/GSSG ratio ( Fig. 1E ) in aged animals were lower than those in young animals by 35 and 200%, respectively. nSMase activity was 2-fold higher in the aged animals than that measured in the young animals ( Fig. 1F ). Thus, the changes observed in mice on a SAA def. diet refl ect the aging-associated changes in GSH and nSMase activity observed in isolated hepatocytes.

Aging-associated accumulation of C16:0 and C24:1 ceramides
Measurement of hepatic levels of ceramide, the product of nSMase activity, in the liver of mice fed the regular chow diet showed a slight tendency to increase with aging, but the differences did not reach statistical signifi cance. Fig. 1. Aging-associated changes in GSH/GSSG levels, nSMase activity, and ceramide species in liver. Young (5 months) and aged (23 months) C57Bl6 mice were randomly placed on either a Std chow or a SAA def. diet for 4 weeks. Twenty percent homogenates from livers were prepared and used for analyses. Hepatic GSH and GSSG (A, B, D, E) were measured in individual animals (n = 3) using a Bioxytech® GSH/GSSG-412™ assay kit as outlined in the Materials and Methods. nSMase activity (C, F) was measured in whole liver homogenates (40 g of protein/assay) using a fl uorescently labeled analog NBD-SM as a substrate. The mean values for each individual mouse are shown. Quantifi cation of the different ceramide species in the liver of young and aged animals was done by HPLC-ESI/tandem mass spectrometry (G). Mean values ± SD are shown (n = 3), * P < 0.05, ** P < 0.01 according to Student's t -test.
However, statistically signifi cant changes were seen when mass spectrometry was used to individually quantify ceramides with different FA chain length. Notably, the two major ceramide species in the liver, C24:1 and C16:0 ceramide, were elevated with aging. The levels of C20:1 ceramide seemed to decrease with aging, while levels of C18:1, C18:0, C20:0, and C24:0 ceramide were not affected ( Fig. 1G ). The same ceramide species were elevated in aged SAA def. diet-fed mice as compared with young mice on the same diet. These data are consistent with an aging-associated stimulation of the activity of nSMase-2 specifi cally, rather than that of nSMase-1, because C24:1-SM is the preferred substrate of the former ( 45 ).

Alleviation of aging-associated differences in hepatic GSH by dietary OTC supplementation
Substantial elevation in hepatic GSH content was achieved in mice fed with the OTC-supplemented diet. The total GSH content increased up to 3-fold ( Fig. 2A ) in animals of all ages, while the GSH/GSSG ratio was elevated between 2.5-and 3-fold ( Fig. 2B ). These changes were well within the physiological range of GSH, as they closely resembled the values typically observed in mice on a chow diet. More importantly, mice on an OTC diet exhibited no aging-associated differences in hepatic GSH levels and GSH/GSSG ratio. Together these results show that OTC provided in the diet can effectively modulate hepatic GSH content in an age-specifi c manner.

Effects of the OTC diet on hepatic ceramide levels and nSMase activity
Our earlier studies in isolated hepatocytes showed that restoration of normal GSH levels in hepatocytes of aged rats also restored normal nSMase activity and ceramide content ( 28 ), consistent with the proposed role of GSH as an inhibitor of nSMase; therefore, we anticipated that restoration of GSH content in aged animals should inhibit nSMase activity and respectively decrease ceramide content. Surprisingly, however, in spite the signifi cant elevation in hepatic GSH content, all mice on the OTC diet exhibited 5-to 6-fold higher hepatic nSMase activity as compared with the activity observed in mice fed either the Std chow or the SAA def. control diet ( Fig. 3A ). Accordingly, the levels of the SMase product, ceramide, also increased ( Fig. 3B ), with C24:1 ceramide being affected the most ( Fig. 3C ). The assessment of nSMase-2 protein content in the liver revealed that the increases in activity were likely caused by substantial elevation of nSMase-2 protein in young and aged mice fed the OTC diet ( Fig. 4A, B ). These effects of OTC appear to be specifi c for nSMase-2, as neither nSMase-1 nor ASMase were affected, based on the lack of change in nSMase-1 protein content ( Fig. 4C, D ) or ASMase activity ( Fig. 4E ). Furthermore, these effects were most likely GSH-independent, because a comparison of GSH levels and nSMase-2 expression in all groups of animals shows no correlation between GSH content, ceramide levels, or nSMase-2 expression.
Interestingly, the fold increase in nSMase-2 protein levels observed in mice on the OTC diet was up to 5-fold, while the corresponding changes in activity were only 2.5-fold, indicating that change in activity did not follow exactly the change in protein content. Normalization of the nSMase activity measured in vitro to the amount of nSMase-2 protein, instead of the total protein content, indicated that in the OTC-fed mice, hepatic nSMase-2 had lower specifi c activity than the enzyme of SAA def.-fed mice ( Fig. 4F ). One possible reason for this lower specifi c activity might be that GSH, in which content increased substantially in the OTC-fed mice, effectively inhibited the intrinsic activity of nSMase-2, similar to the effects seen in vitro when GSH content was increased via N -acetyl cysteine supplementation. Whatever the explanation, these results provide compelling evidence that OTC regulates the expression of nSMase-2 mRNA in the liver. More importantly, this upregulation was physiologically important, as it resulted in increased nSMase-2 activity and accumulation of ceramide.
Quantitative RT-PCR ( Fig. 5A, B ), as well as real-time PCR ( Fig. 5C ), showed that hepatic nSMase-2 mRNA levels were substantially higher in the OTC-fed young or old mice, indicating that the effects of the diet on nSMase-2 protein were likely due to increased transcription/stability of nSMase-2 mRNA. The observed amplitude of change ranged between 2-fold (based on RT-PCR) and 15-fold (based on real-time PCR), which is not surprising having in mind the signifi cantly better sensitivity of the latter method.

nSMse-2 protein expression in HepG2 cells supplemented with OTC
To begin investigating the mechanism of these effects of OTC on nSMase-2, HepG2 cells were used. Initially, OTC was added to HepG2 cells for up to 6 days at various concentrations ranging from 1 to 10 mM. No toxicity was observed (data not shown), however, concentrations at the higher end (7.5 and 10 nM) tended to acidify the medium. Therefore, OTC was added to HepG2 cells at a concentration of 2.5 mM and nSMase-2 mRNA expression was determined by real-time PCR after 1, 3, and 6 days of treatment ( Fig. 6A ). In agreement with the results seen in vivo, the OTC treatments led to a substantial increase in nSMase-2 mRNA after 3 and 6 days of treatment. The dose dependency of this effect was confi rmed by RT-PCR of nSMase-2 mRNA ( Fig. 6B, D ) or by Western blotting to assess nSMase-2 protein ( Fig. 6C ), whose levels were similarly elevated in the OTC-treated cells. Finally, the OTC-treated cells also exhibited 2-fold higher nSMase activity ( Fig. 6E ). Together these data suggest that OTC is a novel inducer of nSMase-2 expression in the liver.

Potential role of OTC-induced nSMase-2 protein expression in hepatic steatosis
Close histological examination of the livers indicates that the OTC-fed mice begin to develop fatty liver or steatosis. H and E staining shows intact hepatic morphology ( Fig. 7A ); however, an increased proportion of white coloration, indicative for fatty deposit, was evident in the OTC-fed mice, especially at old age. The nature of these deposits was confi rmed by Oil Red-O staining, indicating an increase in the accumulation of neutral fat (TAG and esterifi ed cholesterol) in the liver of aged OTC-fed mice ( Fig. 7B ). These results suggested that at least in aged mice, OTC diet may cause increased TAG synthesis, and respectively, the development of fatty liver. The direct measurements of hepatic TAG mass confi rmed that there Fig. 2. Effect of dietary OTC supplementation on hepatic GSH content. Young and aged C57Bl6 mice were placed on either a 0.5% OTC or a SAA def. control diet for 4 weeks. Hepatic GSH (A) and GSH/GSSG (B) contents were measured in whole liver homogenates prepared from individual animals (n = 3-4 per group) using a Bioxytech® GSH/GSSG-412™ assay kit. Data are shown as mean values ± SD. Statistical analyses were done by two-way ANOVA. The main effects of the diet and age are indicated with an asterisk (* P < 0.05). No interaction between both factors was present. Fig. 3. Effect of dietary OTC supplementation on nSMase activity and ceramide levels in mouse livers. Young and aged C57Bl6 mice were placed on either a 0.5% OTC or a SAA def. control diet for 4 weeks. nSMase activity (A) was assessed in whole liver homogenates (40 g of protein/assay) using a fl uorescently labeled NBD-SM as a substrate. Total ceramide (B) and ceramide species with different FA length (C) were measured in total lipid extracts by HPLC-ESI/ tandem mass spectrometry. Mean values ± SD are shown (n = 3-4 mice in each group). Statistical analysis was done by two-way ANOVA followed by Bonferroni posttest analyses. The statistical signifi cance of the main effects (aging and diet) are shown with asterisks (* P < 0.05, ** P < 0.01, *** P < 0.001). The statistical signifi cance of the interaction effect is shown with number signs (## P < 0.001).
is a statistically signifi cant elevation of this lipid class in aged animals on the OTC diet ( Fig. 7C ). Based on earlier observations by ourselves and others, these effects could be linked to the observed upregulation of nSMase-2 protein. An elevation in hepatic ceramide has been observed and causatively linked to the development of the disease in various models of hepatic steatosis, including diet-and alcohol-induced TAG accumulation (46)(47)(48).

DISCUSSION
This study investigates the link between hepatic GSH, aging, and SMase activity. Excessive utilization of GSH, either for toxin conjugation or ROS scavenging, is characteristic of many pathophysiological conditions and may lead to chronic depletion of GSH and accumulation of GSSG in the tissue. Aging, in particular, is associated with a substantial increase in the GSSG/GSH ratio in the liver, the main site of GSH synthesis; however, whether this refl ects depletion of GSH or only elevation in GSSG has not been clearly shown. Direct measurements of GSH in isolated hepatocytes indicated that hepatocytes from aged animals exhibit 2-to 3-fold lower levels of GSH as compared with young animals. Some in vivo studies confi rm these results, reporting a net decrease in GSH content in the livers of 24-month-old rats as compared with 4-monthold rats (37)(38)(39)(40)(41)(42). Other studies, however, found no change in hepatic GSH with age ( 43,44 ). The results from our study indicate that the dietary supply of cysteine and cysteine precursors may be an important factor in determining whether or not hepatic GSH content will decrease with aging. Apparently the availability of dietary GSH precursors helps to maintain GSH content at any age, while feeding for 4 weeks with a diet depleted of cysteine precursors reveals signifi cant aging-associated decline. It is likely that while on a SAA def. diet, the aged mice exhaust available GSH much faster than young mice due to increased demand of ROS scavenging and/or defects in their ability to synthesize GSH. Because only male mice were used in our experiments, it is unclear whether these observations refl ect a trend common for both genders. One should not exclude the possibility that in females, the changes might be blunted, as male mice have been found to experience more dramatic age-associated change in GSH content than female mice ( 49 ). Fig. 4. Effect of dietary OTC supplementation on nSMase-2 protein in mouse livers. Young and aged C57Bl6 mice were placed on either a 0.5% OTC or a SAA def. control diet for 4 weeks. nSMase-2 (A, B) and nSMase-1 (C, D) protein levels in whole liver lysates were determined by Western blot using specifi c rabbit polyclonal antibodies against each protein.
Cyclophilin-A was used to control for equal loading and for normalization. Mean values ± SD (n = 3-4 animals per group) of the ratio between the respective nSMase protein and the loading control are shown. ASMase activity (E) was determined in whole liver homogenates as described in the Materials and Methods. nSMase-2 specifi c activity (F) was calculated as the total nSMase activity measured for each individual mouse divided by the nSMase-2 protein amount quantifi ed by Western blotting. Data are shown as mean values ± SD (n = 3-4 animals per group). Statistical analyses were done according to two-way ANOVA and Bonferroni posttest analyses. The statistical signifi cance of the main effect of diet and aging is shown by asterisks (* P < 0.05, ** P < 0.01, and *** P < 0.001). No statistically signifi cant interaction effects were present.
An aging-associated elevation in ceramide content accompanies GSH depletion and is observed in liver, brain, plasma, and other tissues. While accumulation of ceramide can lead to increased infl ammation and cell death, current evidence suggest that the FA make-up of ceramide is a key determinant of ceramide biological activity. It seems that ceramides with various chain lengths can be generated in a specifi c manner and have distinct bioactivities ( 50 ). The effects of aging on the FA composition of ceramide have been investigated in several models including mouse CD4+ T-cells ( 51 ), cardiomyocytes ( 52 ), and skeletal muscle ( 53 ). Our mass spectrometry-based analysis of ceramide species in the livers of aged and young mice reveals that aging leads to the accumulation of C16:0 and C24:1 ceramides and has no effect on the less abundant C18:0, C18:1, and C24:0 ceramides, while the level of C20:0 ceramide is modestly suppressed.
Studies in cell culture have also found that total nSMase activity is elevated with aging. The siRNAi approach indicated that nSMase-2, but not nSMase-1, accounted for these increases ( 54 ). Our analyses in mouse livers show similar differences that, however, reached statistical signifi cance only in mice on the SAA def. diet but not in mice on Std chow. One possible explanation for this failure to observe strong aging-associated changes in nSMase activity in vivo could be the heterogeneity of the cell population making up the hepatic extracts. Furthermore, there is no specifi c assay to monitor the enzymatic activity of nSMase-2. In fact the assay refl ects the sum of all hepatic nSMases, including the main housekeeping enzyme, nSMase-1, which is not affected by aging but is more abundant than nSMase-2. Whatever the case, the expected elevation in nSMase activity with aging was clearly seen when the mice were placed briefl y on the SAA def. diet and the aging-associated differences in hepatic GSH levels were detected.
Our studies present evidence for transcriptional regulation of nSMase-2 by dietary OTC. The effects are specifi c and are paralleled by a production of functional protein and an elevation in ceramide. NSMase-2 is a signaling enzyme that typically is linked to the onset of apoptosis, cell growth arrest, and infl ammation ( 55 ). Oxidative stress and pro-inflammatory cytokines, including IL-1 ␤ and TNF-␣ , are among the major inducers of nSMase activity; the effects are rapid (typically within 30 min of stimulation)   PCR ( B, D ) . The levels of nSMase-2 protein were analyzed by Western blotting using nSMase-2 specifi c rabbit polyclonal antibody (C). ␤ -Actin was used for normalization and as a control for equal loading. nSMase activity was measured in whole cell homogenates (D). Data are shown as a percent of the activity measured in vehicletreated cells. Data are shown as mean values ± SD (n = 3). One-way ANOVA with Dunnett's multiple comparison test are shown (* P < 0.05, ** P < 0.01, *** P < 0.001). Results are representative of at least three independent experiments. and transient, involving posttranslational mechanisms such as change in phosphorylation status and translocation ( 56,57 ). Our present study shows that the expression of nSMase-2 mRNA and protein is also regulated and could be affected by dietary means. The effects were clearly seen in vivo (in OTC-fed mice) as well as in vitro (in OTC-treated HepG2 cells), and were seemingly GSH-independent. Notably, nSMase-1 was not affected, thus confi rming the specifi city of the changes. More importantly, these changes in expression were functionally important, as they led to elevated hepatic ceramide content. Two very recent studies showed similar stimulation of nSMase-2 mRNA in response to all-trans retinoic acid ( 58 ) and cyclopamine ( 59 ) in cell culture. Together, these studies delineate a novel pathway for the regulation of nSMase activity and function, namely at mRNA level.
The exact mechanisms behind the OTC-induced upregulation of nSMase-2 mRNA are unclear. OTC-mediated activation of Nuclear factor erythroid 2-related transcription factor (NRF-2) has been observed in liver cells ( 23 ) and analysis of the 5 ′ proximal region of the nSMase-2 gene reveals the existence of an NRF-2 consensus binding site. Alternatively, an OTC metabolite could be involved. As discussed earlier, following its metabolic conversion by conditions associated with decreased GSH content ( 19,22,(63)(64)(65)(66)(67)(68). These could be specifi cally important for the aged population, where decreased ability to detoxify common drugs like acetaminophen has been linked to decline in GSH content ( 69 ). Therefore, our fi ndings that OTC upregulates nSMase-2 expression and increases ceramide content (changes that are typically linked to propagation of infl ammation and cell death) were unexpected. It is of note, however, that we also observed an elevation of hepatic TAG levels after the OTC diet, particularly in the aged mice. These changes in TAG synthesis might be causatively linked to the nSMase-2 upregulation, because in other systems, nSMase-2 and ceramide have been found to promote TAG synthesis and lipid droplet formation ( 70 ). Fig. 7. Liver histology and TAG levels. Young and aged C57Bl6 mice were placed on either a 0.5% OTC or a SAA def. control diet for 4 weeks. Freshly dissected liver portions were fl ash-frozen, sliced, and stained with H and E (A) or Oil Red-O (B). Slides are representative of all animals in each group. The scale bar corresponds to 50 m. Hepatic TAG mass was measured in total lipid extracts using an L-Type Triglyceride M kit and normalized to the total phospholipid content (C). Data from an individual mouse are shown. Statistical analysis was done by two-way ANOVA followed by Bonferroni posttest analyses. The statistical signifi cance of the main effects (aging and diet) is shown by asterisks (* P < 0.05, ** P < 0.01). The statistical signifi cance of the interaction effect is shown by a number sign (# P < 0.01). a cytosolic enzyme, 5-oxoprolinase, OTC serves as a precursor of cysteine and glutathione. Our studies, however, indicated that GSH is not the likely mediator because the changes in the GSH levels observed in vivo did not correlate with the changes in nSMase-2 mRNA or protein. Furthermore, in HepG2 cells, the OTC supplementation had very little effect on cellular glutathione levels (data not shown). Other OTC metabolites, however, could play a role. Hepatic cells effi ciently convert OTC to taurine ( 60,61 ), which is an LXR ligand and has been shown to affect hepatic lipid metabolism ( 62 ). OTC could also generate sulfate ( 60,61 ) that, in turn, may affect hepatic functions such as drug detoxifi cation and protein posttranslational modifi cation.
In general, OTC is considered a benefi cial supplement that may help in improving overall health in number of