Silymarin and milk thistle extract may prevent the progression of diabetic nephropathy in streptozotocin-induced diabetic rats

Abstract

Objectives: To investigate the effect of silymarin and milk thistle extract on the progression of diabetic nephropathy (DN) in rats. Methods: Diabetes was induced with a single intraperitoneal (IP) injection of streptozotocin (STZ) (60 mg/kg). Silymarin (100 mg/kg/d) or the extract (1.2 g/kg/d) was gavaged for 4 weeks. Blood glucose (BS), serum urea (S_u), serum creatinine (S_cr), and 24-h urine protein (Up) were measured and glomerular filtration rate (GFR) was calculated. Concentration of thiobarbituric acid reactive species (TBARS) and activities of glutathione peroxidase (GPx), superoxide dismutase (SOD), and catalase (CAT) were evaluated in the renal tissue. Results: Data were expressed as mean ± SEM. Silymarin or the extract had no significant effect on BS, S_cr, and GFR. Both milk thistle extract and silymarin, respectively, decreased S_u (mg/dL) (87.1 ± 7.78, p < 0.001; 84.5 ± 7.15, p < 0.001), Up (mg) (5.22 ± 1.56, p = 0.014; 5.67 ± 0.86, p = 0.034), and tissue TBARS (nmol/mg protein) (0.67 ± 0.04, p < 0.001; 0.63 ± 0.07, p < 0.001) in diabetic rats, compared to diabetic control (DC) (S_u: 131.0 ± 4.55, Up: 8.3 ± 0.84, TBARS: 0.94 ± 0.06). Both the extract and silymarin could increase the activity of CAT (IU/mg protein) (25.5 ± 4.0, p = 0.005; 20 ± 1.8, p = 0.16) and GPx (IU/mg protein) (0.86 ± 0.05, p = 0.005; 0.74 ± 0.04, p = 0.10), respectively, in diabetic rats compared to DC (CAT = 14.4 ± 2.0, GPx = 0.57 ± 0.02). Conclusion: Milk thistle extract, to a lesser extent silymarin, can attenuate DN in rats possibly by increasing kidney CAT and GPx activity and decreasing lipid peroxidation in renal tissue.

Keywords:

• diabetic nephropathy
• silymarin
• milk thistle extract
• antioxidants
• rat

INTRODUCTION

Diabetic nephropathy (DN) is a frequent complication of type 1 and type 2 diabetes, involving about 30% of all diabetic patients.¹ If left untreated, 20–40% of diabetic patients with microalbuminuria will progress to overt nephropathy and 20% of them will develop end-stage renal failure within 20 years.² Albuminuria, microscopic and macroscopic, correlates with risk of renal failure and mortality.³

Oxidative stress is one of the major pathophysiological mechanisms involved in the development of DN according to many in vivo and in vitro studies.^4,5 Hyperglycemia not only stimulates the production of reactive oxygen species (ROS) but also attenuates antioxidative mechanisms through glycosylation of antioxidative enzymes.^6,7 Hyperglycemia can directly increase the production of hydrogen peroxide and also increase lipid peroxidation in the glomerulus.⁸ Furthermore, hyperglycemia can produce advanced glycosylated end products (AGEs) through glycosylation of proteins in blood and tissues.^9–11

Unfortunately, despite glycemic and blood pressure control, many diabetic patients are still prone to developing kidney disease. Consequently, there is an urgent need to identify additional therapeutic strategies to attenuate the progression of nephropathy in this population.

Silymarin, the dominant flavonoid in Silybum marianum (milk thistle) seed extract, is known for its antioxidant, anti-inflammatory, and antihyperglycemic properties.¹²

In this study, we examined the effect of silymarin and milk thistle extract on the renal function and oxidative stress in streptozotocin (STZ)-induced rat models of DN.

MATERIALS AND METHODS

Animal experiments

The ethical committee of Shiraz University of Medical Sciences approved all our animal experiments. Male Sprague–Dawley rats (227 ± 16 g) bred and raised at the university animal quarters were housed five in each cage and given a rat chow diet (Pars Dam, Tehran, Iran) and water ad libitum.

Preparation of milk thistle extract

Two hundred grams of ground milk thistle seeds were soaked twice in 200 mL of 85% ethanol for 48 hours. The alcoholic extract obtained was concentrated in a rotary evaporator (Eyela, Tokyo Rikakikai Co., Tokyo, Japan) at 40°C and then freeze-dried in a lyophilizer (Zibrus Technology, Bad Grund, Germany).

Study design

Six groups of rats, each containing 10 animals, were used in this study: Group 1, or the healthy control (HC), in which normoglycemic animals were receiving 1 mL of vehicle (60% ethanol + 5% tween-40) per day; Group 2, or healthy extract (HE), in which normoglycemic animals were receiving 1.2 g/kg milk thistle extract in 1 mL vehicle per day; Group 3, or healthy silymarin (HS), in which normoglycemic rats were receiving 100 mg/kg silymarin in 1 mL vehicle per day; Group 4, or diabetic controls (DCs), in which diabetic rats were receiving 1 mL of vehicle per day; Group 5, or diabetic extract (DE), in which diabetic rats were receiving 1.2 g/kg milk thistle extract in 1 mL vehicle per day; and Group 6, or diabetic silymarin (DS), in which diabetic rats were receiving 100 mg/kg silymarin in 1 mL vehicle per day.

Diabetes was induced by a single intraperitoneal (IP) dose of 60 mg/kg STZ. One week after STZ injection, blood glucose (BS) was determined by a glucometer (Accu-Chek®, Roche Diagnostics, Basel, Switzerland) using a drop of blood from the caudal vein. Rats with a BS level of more than 280 mg/dL were considered diabetic.¹³ Diabetic animals were kept in separate cages and were given access to rat chow and water ad libitum. Four weeks after STZ administration, gavage feeding was started on the six groups of rats as indicated above. Gavage feeding was done at 9:00 AM every day and carried out for a period of 4 weeks.

Estimation of serum glucose and kidney function

BS was estimated by a glucometer 4 weeks after STZ injection and at the end of the gavage feeding. Animals were then housed individually in metabolic cages at the end of gavage feeding and 24-hour urine was collected. Creatinine concentration in pooled urine samples was determined by the Jaffe method using Pars Azemoon kits (Tehran, Iran). Urinary protein was determined by Bradford method.¹⁴ A blood sample was collected for serum urea and creatinine determination using commercially available kits (Pars Azemoon, Tehran, Iran) according to the procedure of Newman and Burtis.¹⁵ Glomerular filtration rate (GFR) was determined in terms of creatinine clearance, using serum and urine creatinine and 24-hour urine volume.

Preparation of kidney homogenates

The rats were killed by decapitation and their kidneys were removed and washed with cold normal saline. Kidney homogenate was prepared according to the procedure of Zal et al.¹⁶

The clear supernatant obtained upon centrifugation was kept at −70°C until used for the estimation of thiobarbituric acid reactive species (TBARS), and for the measurement of the activities of Cu, Zn-superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) enzymes.

Estimation of lipid peroxides and antioxidant enzymes

Lipid peroxidation was assessed by measuring the level of TBARS in kidney supernatants using the colorimetric method described by Hagar et al.¹⁷ Concentration of TBARS was calculated and expressed as nmol/mg protein of the kidney supernatant using 1,1,3,3-tetraethoxypropane (TEP) as standard.

GPx activity of the kidney supernatant was estimated by continuous monitoring of the regeneration of reduced glutathione (GSH) from oxidized glutathione in the presence of glutathione reductase (GR) and NADPH according to a procedure described by Fecondo and Augusteyn.¹⁸ The activity of GPx in the kidney supernatant was expressed as μmol NADPH oxidized/min/mg protein.

CAT activity in the kidney supernatant was assayed by a spectrophotometric method using H₂O₂ as a substrate according to the procedure described by Aebi.¹⁹ The specific activity of the enzyme was expressed as μmol of H₂O₂ consumed/min/mg kidney supernatant protein using a molar extinction coefficient of 43.6 L/mol/cm for H₂O₂.

Cu, Zn-SOD was measured according to the procedure by Misra and Fridovich.²⁰ One unit of SOD is the amount of the enzyme that produces 50% inhibition in the auto-oxidation of adrenaline. The specific activity of the enzyme was expressed as U/mg protein.

Statistical analysis

Data were expressed as mean ± SEM. Using SPSS version 12 software, data were analyzed by one-way analysis of variance (ANOVA) and least significant difference (LSD) test, and p < 0.05 was considered significant.

RESULTS

Table 1 shows the effects of 4 weeks of milk thistle extract and silymarin gavage feeding on BS level and renal function parameters in different groups of rats. The DC had far higher BS than the HC, as expected. Neither extract nor silymarin had any significant effects on BS of normoglycemic animals. Similarly, there was no significant difference among the glucose levels of DC and diabetic rats receiving silymarin or the extract.

TABLE 1.  The effects of milk thistle extract and silymarin gavage feeding on blood glucose level and renal function in diabetic nephropathic rats (n = 10)

Experimental groups	Serum glucose (mg/dL)	Serum urea (mg/dL)	Serum creatinine (mg/dL)	eGFR (mL/min)	24-hour urine protein (mg)
Normoglycemic control (HC)	119.3 ± 1.07^a	57.9 ± 4.36^a	0.76 ± 0.13^a	1.65 ± 0.77^a	2.7 ± 0.43^a
Normoglycemic + extract (HE)	108.4 ± 2.6^a	43.7 ± 3.67^a	1.12 ± 0.08^a	0.62 ± 0.13^a	3.49 ± 0.31^a,c
Normoglycemic + silymarin (HS)	119.5 ± 1.36^a	60.7 ± 2.18^a	0.91 ± 0.16^a	1.43 ± 0.92^a	3.37 ± 0.31^a,c
Diabetic control (DC)	562.9 ± 11.4^b	131 ± 4.55^b	1.24 ± 0.48^a	0.61 ± 0.1^a	8.28 ± 0.84^b
Diabetic + extract (DE)	524.3 ± 20.7^b	87.1 ± 7.78^c	1.31 ± 0.34^a	0.951 ± 0.32^a	5.22 ± 1.56^c
Diabetic + silymarin (DS)	492.7 ± 17.5^b	84.5 ± 7.15^c	0.75 ± 0.08^a	0.63 ± 0.08^a	5.67 ± 0.86^c

Notes: Data are expressed as mean ± SEM. In each column, values with different superscript letters are significantly different at p < 0.05. eGFR, estimated glomerular filtration rate.

Serum urea (S_u) was significantly higher in the DC compared to HC. The DE and DS groups both had a significantly lower serum urea than the DC (p < 0.001).This shows that both silymarin and the milk thistle extract can significantly lower the serum urea level. Serum urea was not significantly different between the DE and the DS groups (p = 0.778). Neither silymarin nor milk thistle extract had an effect on serum urea concentration of the normoglycemic animals.

Both S_cr and GRF did not differ significantly among various groups. Results illustrated that the DC group had a significantly higher 24-hour urine protein than the HC (p < 0.001), and the 24-hour urine protein decreased significantly in both the DE and DS groups compared to DC (p = 0.014 and 0.034, respectively). There was no significant difference among 24-hour urine protein of HC, HE, or HS groups.

Table 2 illustrates the effects of 4 weeks of milk thistle extract and silymarin gavage feeding on TBARS and antioxidant enzymes. The DC had a significantly higher level of TBARS than the HC (p = 0.001). The mean levels of kidney TBARS in the DE and DS groups were significantly lower than that of the DC group (p < 0.001). There was no significant difference between the mean levels of TBARS in the DE and DS groups. Both extract and silymarin did not have a significant effect on the kidney TBRS levels of normoglycemic animals.

TABLE 2.  The effects of milk thistle extract and silymarin gavage feeding on kidney thiobarbituric acid reactive species (TBARS) and antioxidant enzyme activities in diabetic nephropathic rats (n = 10)

Experimental groups	TBARS (nmol/mg protein)	CAT (U/mg protein)	GPx (U/mg protein)	SOD (U/mg protein)
Normoglycemic control (HC)	0.47 ± 0.03^a	40.8 ± 2.8^a	1.76 ± 0.06^a	24.7 ± 0.54^a
Normoglycemic + extract (HE)	0.52 ± 0.03^a	34.8 ± 1.64^a	ND	ND
Normoglycemic + silymarin (HS)	0.56 ± 0.02^a	33.8 ± 3.28^a	ND	ND
Diabetic control (DC)	0.94 ± 0.06^b	14.4 ± 2.0^b	0.57 ± 0.02^b	17.4 ± 1.04^b
Diabetic + extract (DE)	0.67 ± 0.04^c	25.5 ± 4.0^c	0.87 ± 0.05^c	17.61 ± 0.84^b
Diabetic + silymarin (DS)	0.63 ± 0.07^c	20 ± 1.8^b,c	0.74 ± 0.04^b,c	15.9 ± 0.73^b

Notes: Data are expressed as mean ± SEM. In each column, values with different letter superscripts are significantly different at p < 0.05. One international unit (U) of catalase (CAT) activity equals 1 μmol of H₂O₂ consumed per minute. One U of glutathione peroxidase (GPx) equals to 1 μmol of NADPH oxidized per minute. One U of superoxide dismutase (SOD) equals that amount of enzyme producing 50% inhibition of the auto-oxidation of adrenaline. ND, not determined.

The results confirm that the DC group had a significantly lower kidney CAT activity than the healthy rats (p < 0.001). Although both silymarin and extract increased kidney CAT activity in diabetic rats, this increase was only statistically significant for the rats that had consumed the extract (p = 0.006). Again the activity of CAT in normoglycemic animals was not affected by either extract or silymarin. According to our results, the DC had a significantly lower kidney GPx level than the HC (p < 0.001). Although both silymarin and the extract increased the mean kidney GPx level in diabetic rats, this increase was only statistically significant for the diabetic rats which consumed the extract (p = 0.005). There was no significant difference between the mean GPx levels of the DE and DS groups.

The DC had a significantly lower kidney SOD level than the HC (p < 0.001). However, neither milk thistle extract nor silymarin could significantly increase the SOD levels compared to the DC group (p = 0.863 and 0.228, respectively).

DISCUSSION

Some studies indicate that ROS play a key intermediate role in the pathophysiology of DN, one of the important microvascular complications of diabetes mellitus.^4,5 Hyperglycemia, the main determinant of the initiation and progression of DN, generates more reactive oxygen metabolites and attenuates antioxidative mechanisms through nonenzymatic glycosylation of antioxidant enzymes^6,7 contributing to oxidative stress.

Flavonoids are naturally occurring phenolic compounds with strong antioxidant properties because of the presence of aromatic hydroxyl groups and are widely distributed in plants. They are scavengers of ROS and reactive nitrogen species and, therefore, inhibit peroxidation reactions.^21,22 They also have protective effects against oxidative stress. For example, they protect macrophages from oxidative stress by keeping glutathione in its reduced form.²³ Silymarin, a flavonoid extracted from the milk thistle (S. marianum), with its powerful antioxidant properties, is active against oxidative stress and may induce a positive effect on diabetic metabolic abnormalities. In support of this hypothesis, several experimental and clinical studies indicate that substances with antioxidant properties have favorable effects on oxidative metabolic derangement of hyperglycaemia.^24–26

In a study performed by Anjaneyulu et al., the effect of 4 week administration of quercetin, a potent bioflavonoid widely distributed throughout vegetables and fruits, was evaluated on STZ-induced diabetic nephropathic rats. Quercetin markedly increased the amount of antioxidative enzymes (SOD, GPx, and CAT) and decreased the level of malondialdehyde (MDA) which is an index for lipid peroxidation.²⁷

Till now, no study has been performed to evaluate the effect of milk thistle extract or silymarin on DN. However, because silymarin is a flavonoid and a free radical scavenger similar to quercetin, we expected the same results on diabetic nephropathic rats.

In this study, STZ-injected rats had typical characteristics of DM, such as hyperglycemia, and increased urination. A significant increase in urinary protein excretion indicated the presence of nephropathy.

Although flavonoids have been shown to have antihyperglycemic effect, there are many controversies in this regard. Most probably, the potency, dose, duration of use, and the measurement of BS under fed or starved states can contribute to these controversial results regarding BS.

Chakravarthy et al. have shown that a flavonoid, (–)-epicatechin, normalizes BS levels and promotes β-cell regeneration in pancreatic islet cells of alloxan-treated rats.²⁸ However, the study performed by Sheehan et al. showed no effect on BS by administering (–)-epicatechin in alloxan-induced diabetic rats.²⁹ Similarly, in a study by Bone et al., the beneficial effects of (–)-epicatechin in STZ-diabetic animals could not be demonstrated.³⁰ Quercetin, an antioxidant bioflavonoid was capable of augmenting liver glycogen content and significantly reducing serum glucose concentration in a study performed by Nuraliev et al. on alloxan-induced diabetic rats.³¹ In a similar study, Vessal et al. have shown that quercetin decreased the plasma glucose level of STZ-induced diabetic rats, although it had no effect on plasma glucose level of normal animals.³² Conversely, Anjaneyulu et al. reported that quercetin did not significantly reduce BS level in STZ-induced diabetic rats.²⁷

Maghrani et al. showed that 20 mg/kg of the aqueous extract of S. marianum exhibited potent hypoglycemic effects on STZ-induced diabetic rats.³³ Similarly, Vengerovskii et al. reported a decrease in BS in STZ-induced diabetic rats by administration of 70 mg/kg silymarin.³⁴ Similar results have been shown by Soto et al. on alloxan-induced diabetic rats.³⁵

In this study, no significant difference was observed in BS of STZ-induced diabetic rats after the administration of silymarin and milk thistle extract. This was similar to the study performed by Anjaneyulu et al. regarding the effect of quercetin²⁷ and the recent study on the effect of mangiferin, a polyphenol, on STZ-induced diabetic rats.³⁶

Although neither S_cr nor GFR were significantly different among various groups, urine protein excretion, however, was significantly increased in diabetic rats compared to HC. Neither the extract nor silymarin had any significant effect on 24-hour urine content of the normoglycemic animals. The DE and DS groups both had a significantly lower 24-hour urine protein excretion than the DC (Table 1). This decrease in protein excretion was more pronounced with the milk thistle extract. This effect shows, for the first time, that silymarin and to a greater extent, milk thistle extract, may attenuate the progression of DN in diabetic rats. According to the fact that microalbuminuria is the first sign of DN, it was not surprising that GFR was not significantly decreased in diabetic rats. We can conclude that a 4-week period is not sufficient to affect GFR and serum creatinine. In a study performed by Singh et al., on STZ-induced diabetic rats, the decrease in GFR was significant only after a minimum of 8 weeks, whereas urine protein excretion increased significantly after 4 weeks of STZ injection.³⁷ Should we have extended the duration of the study, we might have noticed a decrease in GFR in the diabetic rats and could determine the effect of silymarin and the extract on this measure of kidney function. The increase in urea was, however, significant in the diabetic group. This can be due to the fact that glucosuria can induce an osmotic diuresis and dehydration, predisposing the rat to a prerenal acute renal failure (ARF) state. This can explain the increase in serum urea in DC rats relative to serum creatinine (Table 1).

Our study showed that the levels of kidney antioxidant enzymes (CAT, GPx, SOD) were significantly reduced in diabetic rats compared to the HCs (Table 2) as expected. Similarly, as predicted, lipid peroxidation was significantly increased in the kidney of diabetic rats compared to HCs.

Although kidney CAT and GPx levels were increased by both silymarin and milk thistle extract, only the extract could increase these enzymes to a level of statistical significance (Table 2). On the contrary, both silymarin and milk thistle extract significantly reduced the kidney levels of TBARS compared to DC.

The fact that milk thistle extract had a greater effect on reducing proteinuria and increasing CAT and GPx levels is not surprising due to the fact that this extract contains several other flavonoid-type structures (i.e., apigenin, chrisoeriol, eriodictyol, naringenin, quercetin, taxifolin). It is obvious that these flavonoid-type structures may have additive effects on the above parameters. The beneficial effect of quercetin on DN and kidney antioxidant enzymes has been proven in a previous study.²⁷ In a study performed on nephrotoxic rats, both vitamin E and quercetin improved renal function to some extent, but it was the combination of the two that lowered the blood urea nitrogen (BUN) and creatinine levels to those seen in HCs.¹⁶

The effect of milk thistle extract and silymarin on kidney SOD level was somehow unexpected in this study. None of the treatments could make a significant increase in SOD level compared to the DC group (Table 2). In a study performed by Mostafavi-pour et al., no significant change was observed in SOD activity of the liver in hepatotoxic rats, after vitamin E and/or Quercetin administration.³⁸ However, in their study, there was no significant reduction in SOD levels of hepatotoxic rats compared to HCs. In our study, although there was a significant reduction in kidney SOD of diabetic rats compared to HCs, no statistical increase was observed upon administration of silymarin or milk thistle extract. In a study performed by Soto et al. on pancreatic antioxidant enzyme levels of alloxan-induced diabetic rats, the increase in SOD level was observed after a lag-time of 30 days, indicating that possibly this enzyme needs more time to be affected by silymarin.³⁹ On the contrary, an article published on ROS scavenging effects of silibinin dihemisuccinate (SDH), a water-soluble form of the main structural isomer constituent of silymarin, SDH could inhibit superoxide production, but with a very high IC₅₀ value (concentration required for 50% inhibition of superoxide) of 2 mg/mL, compared with SOD that inhibited the reaction with an IC₅₀ of 59 ng/mL.⁴⁰ These results show that most probably a higher dose of silymarin together with a longer duration of treatment could have affected the activity of SOD enzyme. The study that evaluated the levels of antioxidant enzymes in the pancreas of alloxan-induced rats, a silymarin dose of 200 mg/kg/d for 9 weeks was administered, and there was a significant reduction in pancreatic SOD.³⁹

CONCLUSION

Milk thistle extract can attenuate diabetic renal damage (by decreasing urinary protein excretion), possibly due to its antioxidant action. This effect was also shown with silymarin, the most dominant flavonolignan in milk thistle extract, but to a lesser extent.

It is strongly suggested to design another study and evaluate the effects of higher doses of silymarin, and if possible, longer duration of treatment on renal function and oxidative stress. In addition, it is interesting to study the effect of pretreatment with silymarin or extract by starting drugs on day 0 rather than waiting 4 weeks for DN to develop. Determining the effect of silymarin and milk thistle extracts on levels of tumor necrosis factor (TNF)-α (marker of inflammation) and tumor growth factor (TGF)-β (marker of fibrosis) in the kidney of diabetic nephropathic rats would also be of value.

Acknowledgment

This study was financially supported by the Vice Chancellor for Research, Shiraz University of Medical Sciences.