2.5. Differential Expression of Antioxidant Defense and Oxidative Stress Related Genes
Quantitative Real Time-PCR array analysis of the relative expression of 84 oxidative stress and antioxidant defense genes in rat liver was conducted after 28 days of treatment. No analyses were conducted after 90 days as the 28 days exposure was consider sufficient to effect changes regarding these genes in the liver. Significant (p
< 0.05) changes detected in the expression of seven genes as a result of GRE intake are summarized in Table 5
Fold change a and p-values of the genes affected by dietary consumption of GRE b during the 28 study with male Fischer rats (n = 3 per group).
Fold change a and p-values of the genes affected by dietary consumption of GRE b during the 28 study with male Fischer rats (n = 3 per group).
|Function Grouping and Name of Gene||Symbol||FC c||p-Value d|
|(i) Antioxidant defense related genes|| || |
| Glutathione Peroxidases|
| Glutathione peroxidase 2||Gpx2||+1.80 c||0.04|
| Glutathione peroxidase 3||Gpx3||−1.20 d||0.04|
|(i) Genes involved in reactive oxygen species (ROS) metabolism|
| Oxidative Stress Responsive Genes|
| Aminoadipate-semialdehyde synthase||Aass||+1.32||0.01|
| Apolipoprotein E||Apoe||−1.27||0.05|
| Isocitrate dehydrogenase 1||Idh1||+1.30||0.04|
| NAD(P)H dehydrogenase, quinone 1||Nqo1||+1.68||0.02|
| Other genes involved in superoxide metabolism|
| Neutrophil cytosolic factor 2||Ncf2||−4.78||0.03|
GRE significantly (p < 0.05) affected the expression of two genes encoding for antioxidant defense, including the 1.8 fold up-regulation of glutathione peroxidase 2 (Gpx2) and a 1.2 fold down-regulation of Gpx3.
GRE significantly (p < 0.05) up-regulated some of the oxidative stress responsive genes including aminoadipate-semialdehyde synthase (Aass), isocitrate dehydrogenase 1 (Idh1) and NAD(P)H dehydrogenase, quinone 1 (Nqo1) 1.32, 1.30 and 1.68 fold, respectively, while the expression of apolipoprotein E (Apoe) was down-regulated 1.27 fold (p = 0.05). Neutrophil cytosolic factor 2 (Ncf2), involved in superoxide metabolism, was down-regulated 4.78 fold by GRE.
GRE treatment did not significantly (p ≥ 0.05) affect any of the genes for oxygen transporters.
Studies in cell cultures and experimental animals indicated that antioxidants may cause adverse toxic effects, particularly when administered or consumed at high dose levels. Studies on flavanol-enriched green tea (unfermented Camellia sinensis
)) preparations, containing mainly the potent antioxidant, (−)-epigallocatechin gallate (EGCG), have been reported to exhibit hepato- and nephrotoxicity effects when administered as a single high dose to mice (1500 mg/kg), rats (2000 mg/kg) and dogs (1500 mg/kg) [13
]. In humans, consumption of high doses of tea-based (C. sinensis
) dietary supplements showed elevated ALT and Tbili levels, which were resolved following cessation of the supplement consumption [15
]. It became necessary that investigations on specific health effects of antioxidants should incorporate studies to establish “safety levels” to avoid toxic or adverse health effects [16
No effect on the body weight gain and the relative liver and kidney weights of the rats was observed as a result of the 28 and 90 day GRE treatments. Although the baseline Tbili was higher in the younger rats and Dbili, ALT and creatinine levels increased significantly (p
< 0.05) in the older rats, these parameters were not altered by GRE. A significant (p
= 0.03) increase in ALP was evident in the rats chronically exposed to GRE. Although increased serum ALP is associated with drug-induced cholestasis [17
], the gamma glutamyl-transferase (GGT) activity, considered to be a more reliable marker for cholestasis [18
], was not significantly (p
≥ 0.05) increased by GRE. A specific role of the rooibos flavonoids in the disruption of biliary function is not known at present and different manifestations thereof could prevail depending on the age of the rats and the level and duration of GRE exposure.
An underlying oxidative stress, however, seems to prevail after 28 days when considering the expression of the oxidative stress and antioxidant response genes. This is despite the fact that none of the oxidative stress markers of lipid peroxidation in the liver, i.e.
, conjugated dienes and malondialdehyde, were significantly (p
≥ 0.05) altered. This could imply a lack of sensitivity of these markers when considering oxidative stress. In this regard the respective up- and down-regulation of Gpx2
involved in hydrogen peroxide degradation is of particular interest. The Gpx2
gene is also a target for NF-E2-related factor 2 (Nrf2), a transcription factor that regulates important antioxidant and phase II detoxifying genes [19
suppresses cyclooxygenase activity by removal of hydroperoxides required for enzyme activation, thereby facilitating an anti-inflammatory function [20
]. The up-regulation of Gpx2
by GRE may indicate an anti-inflammatory response as a result of an underlying oxidative stress due to the high levels of flavonoids consumed. Down-regulation of Gpx3
by GRE could facilitate oxidative stress due to reduced ROS quenching. This gene is down-regulated during neoplastic transformation as compared to healthy tissue, where it presumably plays a role as a tumor suppressor [20
] and therefore its down-regulation may be unfavorable during various stages of cancer development. Of interest is the up-regulation of peroxidase, Aass
, encoding a catalyzing peroxidase protein enzyme that plays an antioxidant protective role in cells and is involved in lysine degradation [21
]. The latter is known to cause lipid peroxidation, reducing the level of GSH and gluthathione peroxidases, thereby impairing antioxidant defenses mechanisms. It would appear that the hyper-expression of the Aass
gene is associated with protection of the liver against the harmful action of ROS involving the GSH redox cycle. Although the activity of the antioxidant enzymes, CAT and SOD, were not altered, GR activity was significantly (p
< 0.05) increased after 28 days by GRE, while it reduced the GSH level in the liver after 90 days, suggesting modulation of the GSH redox cycle. The decrease in GSH is therefore of relevance as polyphenols are known to be an important determining factor when considering their pro-oxidant activity and their interaction with glutathione metabolism [22
]. It is known that monophenol-type flavonoids is prone to cause GSH oxidation, while flavonoids with a catechol group on the B-ring may lead to GSH conjugation. GSH oxidation seems not to be involved, since the GSSG level was not altered. The formation of GSH conjugates via flavonoid quinone formation is therefore more likely. Studies in hepatocytes indicated that flavonoids with the catechol group possessing low redox potentials, such as luteolin and quercetin, depleted hepatocyte GSH (Galati et al.
, 2002). Therefore, despite the increase in expression of Gpx2
, an increase in the GSSG levels was not observed, which could be related to the reduction in the GSH levels via the interaction with the rooibos flavonoids. Of interest was that the GR activity was significantly increased after 28 days, but tended to stabilize after 90 days. GSH depletion has been proposed as a potential strategy to sensitize the cell to phenoxyl radical-induced oxidative stress and mitochondrial membrane potential collapse [26
]. However, it may adversely affect the GSH redox cycle under normal physiological conditions due to the potential pro-oxidant activity [24
]. In this regard, pro-oxidant activity of rooibos aqueous extracts, their crude polymeric fractions and pure aspalathin has been reported while “fermentation” (i.e.
, oxidation) of rooibos decreased the pro-oxidant activity associated with the concomitant decrease in the aspalathin content [28
]. Major GRE flavonoids containing a B-ring catechol arrangement are aspalathin, isoorientin, orientin and the quercetin glycosides, i.e.
, rutin, isoquercitrin and quercetin-3-O
Further evidence towards the possible induction of oxidative stress by GRE is the up- and down-regulation of genes involved in ROS metabolism. These include the down-regulation of Ncf2/p67phox
genes and the up-regulation of Ncf2
genes. The role of Ncf2
in the liver is unclear, however, it encodes for a multi-enzyme complex known as NADPH oxidase, which plays an essential role in regulating the activity of neutrophils [29
]. NADPH oxidase produces superoxide anion and other ROS from molecular oxygen, using NADPH as electron donor and influences a multitude of biological functions including host defense and redox signaling [30
-dependent isocitrate dehydrogenase (Idh1) also provides NADPH needed for the regeneration of GSH, as well as for fat and cholesterol synthesis [32
]. In addition, down-regulation of apolipoprotein E (Apoe
) gene expression may also impact on cholesterol homeostasis as Apoe plays a key role in metabolism of cholesterol and triglycerides by binding to receptors in the liver, contributing to the clearance of chylomicrons, very low density lipoprotein (VLDL) and high density lipoprotein (HDL) from plasma [33
]. In contrast, the up-regulation of Idh1
and the corresponding increase in the associated protein, Idh1, are therefore of relevance [34
]. In normal cellular metabolism, Idh1 plays an important role in lipid metabolism, maintaining cellular cholesterol and fatty acid homeostasis through synthesis and degradation. The enzyme is therefore suggested to be a target enzyme for lipid-lowering pharmacological strategies [35
]. An association with cellular response to oxidative insults and an increased activity of the enzyme have also being suggested [34
]. Dysregulation of Idh1
is a common phenomenon in cancer cells as it functions at a crossroad of cellular metabolism in lipid synthesis and cellular defense against oxidative stress and cancer cells may gain from the glucose sensing role of Idh1 [34
]. The outcome of the up-regulation of Idh1
expression by GRE will therefore depend on the specific conditions in terms of oxidative stress and disease. Changes in the expression of these genes could therefore be related to the lowering of serum cholesterol, as well as the reduction in the GSH level in the liver. Subsequent studies should focus on the effect of GRE on the translation protein levels of the associated enzymes. This is of specific interest, as a study in humans who consumed six cups of fermented rooibos daily for six weeks had decreased LDL-cholesterol and increased HDL-cholesterol and GSH levels in the blood [36
gene is a member of the NAD(P)H: quinone oxidoreductase family that prevents the one electron reduction of quinones resulting in the production of radical species. The up-regulation of the gene by GRE suggested it to be part of an oxidative stress response and it is reported to be overexpressed in certain types of malignant tissues in the colon, breast, lung and liver [37
]. Therefore the increased expression of this gene in the liver following GRE consumption is of concern and warrants further investigation. The expression of the Nqo1
gene, as was mentioned for the Gpx2
gene, is highly regulated by Nrf2
directly via an antioxidant response element (ARE) and plays an integral role in cellular responses to oxidative stress [38
]. In mice electrophilic chemicals up-regulated the expression of Nqo1
in an Nrf2-dependent fashion [39
]. The rooibos flavone, isoorientin, up-regulates the expression of Nqo1
in HepG2 cells, which was associated with an increased level of the antioxidant enzyme proteins [40
]. The so-called “protective” effect resulting from up-regulated expression of Nqo1
under the current study conditions could be related to a protective response towards GRE-induced oxidative stress in the liver.
The modulation of the oxidative status in the liver by rooibos flavonoids therefore seems to depend on the dose and the duration of GRE exposure. The current GRE preparation contained 18.4 g aspalathin/100 g extract, approximately 1.3–3 fold the concentration of aqueous unfermented (“green”) rooibos extracts, while the levels of nothofagin and the flavone glucosides, isoorientin, vitexin and isovitexin, were also increased [7
]. The TP content of GRE (39.22 g GAE/100 g extract) was higher than the averages of 35.08 and 35.12 g GAE/100 g extract reported for aqueous extracts of unfermented rooibos by Joubert et al.
] and de Beer et al.
], respectively. Increased polyphenol content of GRE resulted in increased antioxidant activity in the FRAP, DPPH and ORAC assays when compared to aqueous unfermented rooibos extracts [42
]. The increased activity is mainly attributed to the high aspalathin content, known to be a potent rooibos antioxidant when compared to the radical scavenging activity of quercetin and EGCG [43
A 10 week study with fermented and unfermented rooibos as sole source of drinking fluid significantly (p
< 0.05) reduced GSSG levels in the liver of rats, while GSH was markedly increased, resulting in a significant (p
< 0.05) increase in the GSH/GSSG ratio [44
]. Differences in the TP intake need to be considered as the dietary intake of 6.2 mg GAE/100 g bw/day for the current study was 2.5 times lower than that of the 10 week study when unfermented rooibos was consumed (16.2 mg GAE/100 g bw/day). The low TP exposure for 10 weeks increased the GSH level while in the current study, the lower dose over a period of 90 days reduced the GSH level. A three-week study in rats indicated that fermented rooibos as sole drinking fluid prevented lipid oxidation and oxidative stress effected by the hepatotoxic carcinogen, fumonisin B1
in rats at a TP exposure level of 6.4 mg GAE/100 g bw/day [44
]. However, unfermented rooibos (TP of 16.1 mg GAE/100 g bw/day) synergistically increased the hepatotoxic effect of fumonisin B1
], which was attributed to pro-oxidant effects as demonstrated for rooibos extracts and aspalathin [28
]. Both the fermented and unfermented rooibos significantly (p
< 0.05) reduced the FB1
-induced increase in the liver GSH level. In the current study an underlying oxidative stress appears to exist in rat liver following the consumption of GRE mainly via disruption of the GSH redox cycle.
Of interest was the significant (p
< 0.001) decrease in the total serum iron levels by GRE after the 90 day dietary treatment, while no effect was evident after 28 days. No adverse effects on the serum Fe level was recorded at a three-fold lower TP level in the 10 week study using aqueous extract of unfermented rooibos as sole drinking fluid [44
]. Disruption of iron absorption from the gut at a chronic high dose of exposure could have important implications in humans regarding physiological conditions associated with anemia. The high levels of TP consumed and the formation of non-transportable polyphenol-iron complexes [46
] appears to be responsible for reducing the serum iron levels in the current study. However, studies in humans fail to provide evidence of a reduction in serum iron following consumption of fermented rooibos infusions thus far [36
]. Daily consumption of six cups of fermented rooibos for six weeks did not affect the total serum iron levels of the subjects [36
]. Based on the TP content of one cup (200 mL) of rooibos infusion and an average bw of 70 kg, the study subjects were exposed to an average of 5 mg TP/kg bw/day [36
]. This is approximately 12.5 fold lower when compared to the current rat study (62.7 mg/kg bw/day). The difference in the exposure to aspalathin, the major rooibos antioxidant, was even greater (295 fold), considering that the rats consumed 29.5 mg aspalathin/kg bw/day compared to 0.1 mg aspalathin/kg bw/day for humans. The aspalathin content of the rooibos infusions in the human study was not quantified, but analysis of a large number of rooibos production batches, collected over three seasons, provided the expected average aspalathin content of a cup of rooibos infusion (1.2 mg aspalathin/200 mL) [49
]. The dose of aspalathin received by the rats in the current study (90 days) translates to an human equivalent dose of approximately 4.8 mg/kg bw/day based on the body surface area normalization method [50
]. Therefore, a threshold of exposure to rooibos polyphenols seems to exist when considering possible adverse and/or positive health outcomes. The current study provided evidence that consumption of high levels of an aspalathin-enriched rooibos extract could lead to some adverse effects in the liver, but more research is needed to determine the total polyphenol and/or aspalathin thresholds for these outcomes. Subsequent studies should focus on the protein levels of the specific genes that were upregulated to provide better insight into subtle changes related to the redox status in the liver and the kidneys effected by the polyphenol-enriched rooibos flavonoids.