Toxicological risks of Rhizoma paridis saponins in rats involved NF-κB and Nrf2 signaling

Abstract

The aim of the study is to evaluate the safety of long-term use of Rhizoma paridis saponins (RPS). After 90 day administration of RPS in rats, it induced liver and lung injury through the over-expression of reactive oxygen species (ROS) and pro-inflammatory cytokines, and by down-regulating the levels of antioxidant and detoxification enzymes. Meanwhile, RPS treatment also activated the self-protective transcription of Nrf2 and elevation of GSH and HO-1 expression to inhibit worsening tissue conditions in the rats. After 30 days' recovery, the abnormalities in liver and lungs disappeared, accompanied by the return of phase II enzyme, pro-inflammatory cytokine, and nuclear factor levels to normal. In conclusion, 350 mg kg⁻¹ d⁻¹ of RPS induced toxicity and detoxicity reactions involving NF-κB and Nrf2 signaling. Our work provides useful data for the correct administration of RPS and minimizing the danger of toxic herbal product use.

---

Introduction

Paris polyphylla var. yunnanensis (Fr.) Hand.-Mazz. (PPY), a traditional Chinese medicinal herb, has the effects of heat-clearing, detoxification, anti-inflammatory and pain relief in folk medicine.¹ It is reported that Rhizoma paridis saponins (RPS) belonging to the steroidal saponins are the main and active components in Paris polyphylla.^2–4 In traditional clinical usage, excessive ingestion of RPS could cause side effects such as nausea, vomiting, diarrhea, and even heart palpitations and convulsions. As RPS becomes more popular for their activity, there have been increasing concerns about the safety and potential toxicity of RPS.

Our preliminary research revealed that the main components in RPS include diogeninyl and pennogenyl saponins.⁵ Dioscin, one of the diogeninyl compounds contained in RPS, is shown to cause slight gastrointestinal tract distension during the treatment period and hemolytic anemia in hematology assessments.⁶ Meanwhile, RPS possesses sedative–hypnotic activity, gastric stimulation⁷ and hepatotoxicity⁸ side effects. However, scientific information regarding the toxic effects of RPS is limited. Taking into account these drawbacks, it is critical to understand the toxicological property associated with long-term use of RPS.

The production of reactive oxygen species (ROS) has been identified to contribute to drug-induced liver, heart, renal and brain toxicity.⁹ To manage oxidative stress, cells possess antioxidant protection mechanisms, which primarily consist of classical antioxidant enzymes such as superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), and an additional group of enzymes, termed phase II detoxifying enzymes, that include glutathione-S-transferase (GST) and heme oxygenase-1 (HO-1).¹⁰ In this research, we focused on the long-term toxicity of RPS involving detoxifying enzymes and antioxidant enzymes.

Materials and methods

Drugs

The dried rhizoma of Paris polyphylla var. yunnanensis were collected in September 2010 from Lijiang, Yunnan Province, China, and identified by Professor Gao. A voucher specimen (GWCL201009) was deposited at the School of Pharmaceutical Science and Technology at Tianjin University, Tianjin, China. Rhizoma paridis saponins (RPS) were prepared as previously described.⁵

Experimental animals

Eighteen male Sprague-Dawley rats at six weeks of age were purchased from the Laboratory Animal Center of Academy of Military Medical Sciences (Beijing, China quality certification number: SCXK (Jun) 2012-0004). All the experiments involving rats were approved by the local Animal Ethics Committee and performed in strict compliance with the ethical guidelines issued by national legislations of China and local guidelines. After one week of acclimatization, the rats were housed in a room maintained at 24 ± 1 °C, relative humidity of 55 ± 5%, artificial lighting from 8:00 to 19:00 and an air-exchange rate of 18 times per hour. The animals were kept in stainless-steel wire-mesh cages and allowed free access to tap water and diet. Before carrying out the animal experiment, we measured the body weight of each rat. The rats were randomly allocated into three groups. The high dose of RPS (RPSH) was orally administered using 350 mg of RPS in 10 mL of 0.9% sodium chloride per kg of body weight every day for 90 days. The lower dose of RPS (RPSL) was fed orally using 50 mg of RPS in 10 mL of 0.9% sodium chloride per kg of body weight every day for 90 days. Normal control rats were administered appropriate vehicles. During the experiment, the body weight of each rat was measured every week. Food consumption, water consumption, urine volume, urine density and urine pH were monitored every other day during the course of the study.

Blood samples (0.5 mL) were collected into heparinized tubes from each rat by puncture of the retro-orbital sinus on the 30^th, 60^th, 90^th and 120^th days. Blood was immediately processed for plasma by centrifugation at 3500g for 15 min. Plasma was frozen and maintained at −20 °C until analysis. After the different days of collection, all the rats were sacrificed in batch. Autopsies were harvested. Portions of each tissue were fixed in 10% formalin (pH 7.4) for histology and snap-frozen in liquid nitrogen for the oxidative stress test.

Histopathological examination

For the histopathological examination, portions of the liver, kidney, heart, spleen, brain and lung tissues were fixed in 10% formalin, and after proper dehydration, the tissues were embedded in paraffin wax. Five μm-thick sections were prepared and stained with hematoxylin and eosin. Each organ was randomly cut into 6 histological sections. Histopathology examination was completed using Nikon eclipse TE2000-U microscope and performed by a pathologist who was unaware of whether tissues were treated.

Biochemical analyses

Serum levels of alkaline phosphatase (ALP), aspartate transaminase (AST), alanine transaminase (ALT), gamma glutamyl transpeptidase (γ-GT), blood urea nitrogen (BUN) and creatinine (Cr) were measured according to the manufacturer's instructions using detection kits obtained from Nanjing Jiancheng Institute of Biotechnology (Nanjing, China).

Oxidative stress in the liver tissues

Livers fixed by snap-freezing in liquid nitrogen were rapidly washed and homogenized in ten volumes (v/w) of ice-cold saline solution. The homogenate was centrifuged at 3000 rpm for 10 min. Its supernatant was used as a total liver homogenized sample (10% homogenate). The levels of malondialdehyde (MDA), superoxide dismutase (SOD), reduced glutathione (GSH), oxidized glutathione (GSSG) and glutathione-S-transferase (GST) enzyme activities were measured by spectrophotometry according to the manufacturer's instructions using commercially available kits (Nanjing Jiancheng Bioengineering Institute).

Reactive oxygen species (ROS) release assay

The tissue cell samples were prepared following the ROS assay kit manufacturer's instructions (Applygen Technologies Inc., Beijing, China). Intracellular ROS concentration was measured using the oxidant-sensitive fluorescent probe, 2,7-dichlorofluorescin (DCFH) diacetate (DA). Cells were mixed with 400 nM DCFHDA for 30 min and then washed with 1× PBS. The fluorescence of the oxidation product dichlorofluorescein (DCF) was imaged using C6 Software.

Assay of nuclear 8-hydroxy-2-deoxyguanosine (8-OHdG) level

8-OHdG is produced by the oxidative damage of DNA and serves as an established marker of oxidative stress. Each urine sample was centrifuged at 3000 rpm for 20 min, and the supernatant was used for measuring 8-OHdG concentration. The level of 8-OHdG was determined by an ELISA kit following the manufacturer's instructions (Huiying Co., Shanghai, China).

Measurements of COX-2, TNF-α and IL-6 levels by ELISA assays

Plasma samples were analyzed for COX-2 (Huole Co., Shanghai, China), TNF-α and IL-6 (Suer Co., Shanghai, China) with rat ELISA kits following the manufactures' instructions. Each sample was assayed in duplicate, and the values were within the linear portion of the standard curve.

RT-PCR analysis

Total RNA was isolated from rat liver and lung using TRIzol (Life Technologies Inc.) according to the manufacturer's instructions. The quality of RNA was assessed by the absorbance of the samples at 260 and 280 nm. cDNA synthesis was performed using RevertAid™ M-MuLV RT (Fermentas, Hanover, MD, USA) according to the supplier's protocol. Resulting reverse transcription products were stored at −80 °C until analysis. A series of cDNA was amplified in PCR reactions consisting of 1× Taq polymerase buffer with 1.5 mM of MgCl₂ (Promega, Madison, WI), 200 mM of each dNTP, 10 pmol of each primer pair, and 1 U of Taq DNA polymerase (Promega). Polymerase chain reaction products were electrophoresed on 3.0% agarose gel and visualized after ethidium bromide staining.

Western blot assay

Total proteins from livers and lungs were isolated using the tissue protein extraction kit (Bio-Rad, USA), and the obtained protein was quantified using the Bradford Assay Kit (BioRad, Hercules, CA). The protein samples (20–100 μg) were separated on 12% SDS-PAGE. Proteins were transferred to PVDF membranes (Millipore Corp., USA) and probed with GST-α, GAPDH, HO-1 (Bioworld Technology, USA), GST-μ, GST-π (Boster, China), and Nrf2, NF-κB (Santa Cruz Biotechnology, USA) antibodies followed by appropriate secondary antibodies. Equal protein loading was checked by quantifying GAPDH. The relative optical densities of the bands were quantified using Kodak Imaging Program and Image-Pro Plus software.

Immunohistochemistry

Nrf2 in livers were detected by direct immunostaining. All the sections were examined and photographed under Nikon eclipse TE2000-U microscope. Li J. and Man SL examined the immunohistochemistry slides independently. Li J. utilized an immunohistochemical score (IHS) to estimate each group's immunoreactivity.

Statistical analysis

Statistical evaluation was conducted by using the SPSS 17.0 for Windows package software. Data are expressed as the means ± standard error mean (SEM). One-way variance analysis and Duncan multiple range test were used to determine significantly different groups. p values less than 0.05 were considered as significant differences for all statistical calculations.

Results

General observations

None of the animals died during the study period. However, compared with the control group, the rats in RPS group exhibited general variations, such as body trembling, smaller body weight gain (Fig. 1), less physical activity, diarrhea and some altered feces. Meanwhile, the fur of the rats in the RPS treated group was not as smooth and glossy as the ones in the control group. However, no statistical differences were observed for the urinalysis data (Table 1).

Fig. 1 Body weight and food/water consumption changes of SD rats during the 90 day toxicological assessment. (A) Body weight changes; (B) food consumption changes; (C) water consumption changes during the 90 day toxicological assessment. *p < 0.05, compared with normal group.

Table 1 Urinalysis values of SD rats

Groups	pH	Urine volume (mL/24 h)	Specific gravity (g mL^−1)	Urine protein (mg mL^−1)
Normal	9.14 ± 0.15	4.28 ± 2.05	0.99 ± 0.02	16.00 ± 2.56
RPS H	8.91 ± 0.56	5.38 ± 1.54	0.99 ± 0.03	17.66 ± 3.10
RPS L	9.27 ± 0.18	4.57 ± 1.93	1.00 ± 0.03	15.24 ± 2.70

---

Hematology and clinical biochemistry

Selected hematology data are shown in Table 2. No significant changes were observed for any of the parameters examined. Clinical biochemistry data are shown in Table 3. After ninety-day oral administration of 350 mg kg⁻¹ d⁻¹ of RPS, a significant increase in concentration of transaminases and alkaline phosphatase indicated liver function failure. The levels of ALP, ALT and AST were decreased after withdrawal of RPS for thirty days of recovery. BUN and Cr levels were significantly increased on the 30^th day of RPS administration compared with that in the normal group, and returned to normal on the 90^th day of treatment.

Table 2 Hematology values of male SD rats

	WBC (×10^9/L)	LY (×10^9/L)	RBC (×10^12/L)	HCT (%)	MCV (fL)	HGB (g dL^−1)	MCH (pg)	PDW (%)	MPV (fL)	PCT (%)	P-LCR (%)
Normal	10.2 ± 0.5	9.5 ± 0.3	9.3 ± 0.1	60.7 ± 0.4	65.0 ± 0.4	167.5 ± 0.9	18.0 ± 0.3	7.2 ± 0.0	15.7 ± 0.4	0.9 ± 0.1	65.0 ± 2.2
RPS H	12.1 ± 2.0	10.8 ± 1.8	9.5 ± 0.2	60.9 ± 0.8	64.2 ± 0.4	170.2 ± 2.4	18.0 ± 0.2	7.3 ± 0.1	16.1 ± 0.8	0.9 ± 0.1	67.5 ± 4.8
RPS L	10.8 ± 0.4	9.3 ± 0.0	9.3 ± 0.1	60.2 ± 0.5	64.7 ± 0.4	163.0 ± 2.0	17.5 ± 0.2	7.2 ± 0.0	17.8 ± 0.1	0.6 ± 0.0	76.8 ± 0.6

---

Table 3 Clinical biochemistry values of SD rats during 90 day treatment and after the recovery period^a

Parameters	Period	Normal	RPS H	RPS L
a *p < 0.05, **p < 0.01, compared with normal group.
ALP (U gprot^−1)	30^th day	16.40 ± 0.85	25.32 ± 1.91*	22.53 ± 1.00*
	90^th day	8.51 ± 0.28	11.04 ± 0.69*	8.51 ± 0.78
	Recovery	2.38 ± 0.39	2.92 ± 0.55	2.64 ± 0.15
ALT (U gprot^−1)	30^th day	10.45 ± 1.20	21.33 ± 2.37*	21.52 ± 3.94*
	90^th day	12.24 ± 4.75	41.29 ± 8.23*	18.56 ± 1.54
	Recovery	27.26 ± 6.02	28.64 ± 3.97	31.41 ± 3.56
AST (U gprot^−1)	30^th day	141.7 ± 5.8	536.7 ± 20.2*	475.9 ± 5.8*
	90^th day	141.7 ± 15.5	644.7 ± 3.4**	256.5 ± 13.5*
	Recovery	163.0 ± 0.6	174.8 ± 1.8*	171.3 ± 0.2*
γ-GT (U L^−1)	30^th day	15.40 ± 0.62	28.53 ± 4.37*	25.27 ± 4.21
	90^th day	12.32 ± 4.14	35.66 ± 7.27*	16.44 ± 3.27
	Recovery	23.41 ± 1.50	47.91 ± 4.70*	31.78 ± 2.12
BUN (mmol L^−1)	30^th day	467.0 ± 143.1	615.8 ± 147.0*	403.0 ± 64.4
	90^th day	638.4 ± 97.1	691.2 ± 75.1	563.1 ± 51.2
	Recovery	492.0 ± 6.9	447.8 ± 8.1	477.4 ± 7.6
Cr (μmol L^−1)	30^th day	53.79 ± 0.57	64.57 ± 2.91*	54.16 ± 6.00
	90^th day	62.59 ± 0.22	56.75 ± 3.71	59.76 ± 7.98
	Recovery	33.83 ± 1.66	41.12 ± 1.11	39.14 ± 3.32

---

Organ weight

Organ weight data are shown in Table 4. Absolute organ weight of the lung was significantly elevated in the RPS treated rats compared with the normal ones (p < 0.05). No statistically significant variations were detected in any other organs. Meanwhile, the lung weight of the recovered RPSH-treated rats was returned to normal.

Table 4 Absolute organ weights of the male SD rats (g)^a

Parameters	Normal	RPS H	RPS L
a *p < 0.05, compared with normal group.
Brain	1.99 ± 0.01	1.90 ± 0.02	1.94 ± 0.05
Heart	1.23 ± 0.01	1.26 ± 0.02	1.24 ± 0.01
Lung	1.36 ± 0.05	1.77 ± 0.10*	1.58 ± 0.11
Liver	11.10 ± 0.42	10.18 ± 0.13	10.86 ± 0.48
Thymus	0.30 ± 0.02	0.27 ± 0.02	0.32 ± 0.03
Spleen	0.52 ± 0.03	0.55 ± 0.06	0.51 ± 0.03
Stomach	1.90 ± 0.03	1.98 ± 0.05	1.97 ± 0.08
Kidney	2.48 ± 0.09	2.26 ± 0.05	2.50 ± 0.09
Adrenals	0.08 ± 0.00	0.07 ± 0.00	0.07 ± 0.01
Prostate	0.89 ± 0.09	0.77 ± 0.02	0.97 ± 0.12
Spermatophore	1.15 ± 0.04	1.06 ± 0.26	1.20 ± 0.06
Testes	3.12 ± 0.08	3.32 ± 0.09	3.23 ± 0.14
Epididymides	0.56 ± 0.01	0.52 ± 0.02	0.57 ± 0.01

---

Histopathological examination

Microscopic examination of organs was performed on animals from different groups. Histopathological examinations of livers and lungs showed some abnormalities in RPSH groups. Liver section from control rats showed normal liver histologic architecture with the central vein and surrounding normal hepatocytes. However, occasional binucleated cells, massive hepatocellular necrosis, diffuse necrotic cells, and infiltration of inflammatory cell were observed in the RPSH-treated livers, probably due to drug toxicity (Fig. 2B1). In addition, histopathology revealed mild chronic lymphocytic interstitial infiltrates and fibrinous exudate occupying the RPSH-treated alveolar spaces. The alveoli were lined by plump vacuolated pneumocytes (Fig. 2B2). There were no abnormalities in other organs.

Fig. 2 Representative sections (HE staining) of the rat livers and lungs (×200). (A) Normal group; (B) RPSH group on the 90^th day; (C) RPSL group; (D) RPSH after recovery period. (A1–D1) Liver tissues; (A2–D2) lung tissues. Pathological finding in RPSH-treated liver is microfoci of necrosis (N) in (B1). Arrows indicate different effects on hepatocytes, such as an increased number of binucleate hepatocytes in (B1), hypertrophy and hyperplasia of Kupffer cells in (C1), and single normal hepatocytes in (D1). Arrow indicates intra-alveolar fibrin and mild chronic interstitial infiltrate in (B2).

Oxidative stress on DNA and lipid

8-OHdG, a marker of oxidative DNA damage, was significantly elevated in RPSH-treated rats (p < 0.05) and returned to normal at recovery. MDA, a product of lipid peroxidation, was also remarkably increased in RPSH-treated groups (p < 0.05) and is reduced at recovery (Fig. 3).

Fig. 3 Effect of RPS on the levels of 8-OHdG and MDA in SD rats. Each value is expressed as mean ± S.E.M.

Effects of RPS on the oxidative stress of liver tissues

ROS generation was monitored through increases in fluorescence intensity of dichlorofluorescein. As shown in Fig. 4, over-expressed levels of ROS were detected in liver cells of the RPS-treated group. Therefore, increasing evidence indicated that ROS was accumulated by RPS-treated mice. The related antioxidant enzymes, including SOD, CAT and GSH, were detected and are shown in Table 5. Levels of SOD and CAT were significantly decreased in RPS-treated groups (p < 0.05). In contrast, the level of GSH was remarkably increased (p < 0.05), which may be related to the self-protective mechanisms due to the liver oxidative stress induced by RPS.

Fig. 4 Effect of RPS on the production of cellular ROS in liver tissues. Liver cells were labeled with DCF-DA and examined for ROS production by flow cytometer. (A) 90 days, (B) recovery.

Table 5 Antioxidant index values in liver of male SD rats^a

Parameters	90 days			Recovery
	Normal	RPS-L	RPS-H	Normal	RPS-L	RPS-H
a Different superscript letters represent significant differences between two groups (p < 0.05).
CAT (U mgprot^−1)	15.64 ± 0.16^a	13.83 ± 0.02^b	9.24 ± 0.05^c	14.31 ± 0.28^a	12.93 ± 0.15^ab	9.71 ± 1.84^b
SOD (U mgprot^−1)	1.18 ± 0.03^a	0.64 ± 0.06^b	0.51 ± 0.03^b	1.00 ± 0.00^a	0.73 ± 0.02^b	0.63 ± 0.01^c
GSH (mg gprot^−1)	27.95 ± 0.78^a	46.64 ± 1.21^b	55.21 ± 0.98^c	27.43 ± 1.11^a	37.29 ± 1.05^b	39.90 ± 0.22^b
GSH/GSSG	11.53 ± 1.09^a	11.27 ± 1.10^a	13.67 ± 0.22^b	11.60 ± 1.19^a	12.20 ± 1.56^a	12.83 ± 1.16^a

---

Effects of RPS on the expression of pro-inflammatory cytokines

Exposure to RPS had a significant dose–response effect on markers of acute inflammation after 90 days of administration. These markers include IL-6, TNF-α, COX-2 and NF-κB (Fig. 5). However, they returned to normal after the 30 day recovery period.

Fig. 5 Effect of RPS treatment on pro-inflammatory factor expression in rats. (A) and (B) Relative expression of pro-inflammatory factors by RT-PCR assay. (C) Western blot evaluation of these pro-inflammatory factors for liver and lung tissues treated with different doses of RPS. The protein expression levels of NF-κB were normalized against GAPDH. (D) Measurements of levels of COX-2, TNF-α and IL-6 by ELISA assay. Different letters represent significant differences between two groups (p < 0.05).

Effects of RPS on the levels of hepatic phase II detoxification enzymes

Glutathione-S-transferases (GSTs) and heme oxygenase-1 (HO-1) are important phase II enzymes that display anti-oxidant, detoxification, anti-inflammatory and cytoprotective activities.¹¹ As Fig. 6 shows, rats receiving 50 mg kg⁻¹ of RPS and 350 mg kg⁻¹ of RPS showed significant inhibition of GST mRNA expression compared with those treated with the control diet. Meanwhile, RPSH significantly decreased GSTα and GSTπ protein expression. In contrast, the mRNA and protein expressions of HO-1, which is another important anti-inflammatory and cytoprotective enzyme, were up-regulated by RPSH administration. Nrf2 plays a vital role during the activation of genes encoding antioxidant proteins, and phase II detoxifying enzymes were also up-regulated. All the levels returned to normal after 30 day recovery. Immunostaining results showed a clear enrichment of Nrf2 protein both in the cytoplasmic and nuclear fraction in hepatocytes upon RPS treatment. Meanwhile, Nrf2 protein exhibited a higher expression in the nuclear fraction of hepatocytes after 30 day recovery compared with normal groups (Fig. 6C).

Fig. 6 Effect of RPS on the mRNA and protein expression of detoxification enzymes in liver tissues. (A) and (D) Relative mRNA expression of detoxification enzymes by RT-PCR assay. (B) and (E) Western blot evaluation of detoxification enzymes. The protein expression levels of GSTs, HO-1 and Nrf2 were normalized against GAPDH. (C) Immunohistochemical analyses of Nrf2 in liver tissues (×400). Positive staining was observed as a dark brown color. (C1) Normal, (C2) RPSH, (C3) RPSH recovery groups. Arrows in C2 indicate nucleus and cytolymph-positive cells. Arrow in C3 indicates nucleus-positive cells. Different letters mean significant differences among three groups (p < 0.05).

Discussion

As stated in our previous report, RPS, as toxic natural products, are often used as anticancer drugs. To date, at least 120 species of poisonous natural products have been identified, of which more than half have been found to possess remarkable anticancer properties.¹² Due to the medicinal value and widespread use of RPS,¹ it is critical to evaluate the toxicity of RPS with the long-term use.

The results from the 90 day long-term toxicity study showed changing trends on the rat's individual body weight (Fig. 1) due to lower food and water consumption. This change may be related to the properties of saponins, including the gastric stimulatory side effect.⁷ Meanwhile, there were no significant changes in the hematology data and urine analysis (Tables 1 and 2).

Clinical biochemistry data show that the concentration of transaminases and alkaline phosphatase, which are suggestive of hepatic damage, were significantly increased after ninety-day administration of RPS (Table 3). Although there were no changes in liver weight, several abnormalities appeared in RPSH liver tissues (Fig. 2B1). These evidences are consistent with the clinical biochemistry data. In addition, the absolute weight of lung tissues was significantly increased (Table 4). Using the histopathological evaluation, several mild chronic lymphocytic interstitial infiltrates and fibrinous exudates were observed in the RPSH alveolar spaces (Fig. 2B2). All these indicated that high doses of RPS induced liver and lung injury.

As we know, oxidative stress has been regarded as a direct and mechanistic indicator of hepatotoxic potential.¹³ In our study, RPS induced the overproduction of ROS in liver tissues (Fig. 4), which may result in oxidant–antioxidant imbalance, thus damaging cellular components such as DNA and lipids.¹⁴ Therefore, 8-OHdG, a marker of oxidative DNA damage, and MDA, a product of lipid peroxidation, were both significantly elevated in RPSH-treated rats (p < 0.05) (Fig. 3).

Oxidative stress also plays a critical role in the induction mechanisms of pro-inflammatory cytokines. Oxidative stress and inflammation may interact with each other to promote both DNA damage and activation of NF-κB, inducing many acute or chronic liver and lung injuries.¹⁵ In the experiment, RPS induced tissue injury accompanied with ROS-activated NF-κB and its oxidant products. NF-κB subsequently activated the expression of COX-2, TNF-α and IL-6 (Fig. 5). Meanwhile, these pro-inflammatory factors enhanced oxidative stress. ROS generated by these pro-inflammatory cytokines further induced NF-κB activation and overproduced pro-inflammatory cytokines. Consequently, this interaction between oxidative stress and inflammation induced the liver and lung injury in the RPSH-treated group.

In the normal condition, ROS overproduction would activate Nrf2/ARE signaling to protect the body against oxidants.¹⁶ Hence, Nrf2-regulated genes are responsible for increased cellular anti-oxidative or detoxification systems. However, some toxicants suppress the activation of Nrf2 and promote the transcription of NF-κB to induce tissue injury. In this research, the inherent defensive enzymes such as CAT and SOD, as important anti-oxidant enzymes of liver, and GSTs, as anti-oxidant and detoxification enzymes, were decreased by high-dose RPS treatment (p < 0.05). Fortunately, Nrf2 was activated by RPS administration accompanied by elevated levels of HO-1 (Fig. 6C–E), which is an inducible enzyme activated by Nrf2 that plays a central role against inflammation and oxidative stress.¹⁷ The increased Nrf2 protein in both cytoplasm and nucleus suggests that not only the total Nrf2 protein levels, but also the activated Nrf2 are significantly induced by RPS, since translocation of Nrf2 protein into the nuclei is considered as the starting point of Nrf2 pathway activation.¹⁸ In addition, GSH, an antioxidant that detoxifies small molecular substances in livers¹⁹ that is indirectly regulated by Nrf2, was significantly increased in RPSH groups (Table 4). The increased levels of GSH, HO-1 and Nrf2, at least in part, support the self-protection of rats against RPSH treatment-induced oxidative stress and inflammatory conditions. Furthermore, after 30 days' recovery, abnormalities in liver and lung tissues disappeared. The levels of ROS, phase II enzymes, pro-inflammatory cytokines, and nuclear factors such as NF-κB and Nrf2 were all returned to normal.

Our findings suggest the toxicity and detoxifying mechanisms induced by toxic doses of RPS shown schematically in Fig. 7. In summary, RPS, a hypotoxic drug, decreased the body weight of rats and induced liver and lung injury through the over-expression of ROS and pro-inflammatory cytokines as well as down-regulation of antioxidant and detoxifying enzymes. Meanwhile, RPSH activated the self-protective mechanism in rats via Nrf2 transcription and the elevated expression of GSH and HO-1. All these reduced the degree of damage in liver and lung tissues. After 30 days' recovery, abnormalities in liver and lung tissues disappeared. The levels of ROS, phase II enzymes, pro-inflammatory cytokines, and nuclear factors were all returned to normal. The results of the study demonstrated for the first time that the toxicity and detoxicity mechanisms of RPS, an anticancer drug isolated from traditional toxic Chinese medicines, involve NF-κB and Nrf2 signaling. Our work provides useful data for the correct administration of RPS and minimizes the danger of toxic herbal product use.

Fig. 7 Proposed mechanism of toxic RPS dose-induced oxidant stress and inflammation disorders in rats.

Abbreviations

ALT	Alanine amino transferase
AST	Aspartate amino transferase
BUN	Blood urea nitrogen
CAT	Catalase
COX-2	Cyclooxygenase-2
Cr	Creatinine
ELISA	Enzyme-linked immunosorbent assay
GSH	Reduced glutathione
GSSG	Oxidized glutathione
GST	Glutathione-S-transferase
γ-GT	γ-Glutamyl transferase
HO-1	Heme oxygenase-1
IL6	Interleukin 6
MDA	Malonaldehyde
Nrf-2	Nuclear factor-2 erythroid related factor-2
8-OHdG	8-Hydroxy-2′-deoxyguanosine
ROS	Reactive oxygen species
RPS	Rhizoma paridis saponins
RT-PCR	Reverse transcription polymerase chain reaction
SOD	Superoxide dismutase
TNF-α	Tumor necrosis factor α

Acknowledgements

This work was supported by grants 81373904 and 81202952 from the National Natural Science Foundation of China, Drug Creation Projects 2013ZX09103002-010 and 2014ZX09301307-018 from Science and Technology in China and 13JCQNJC13400 from Tianjin Natural Science Foundation in China.

References

1. S. Man, Y. Wang, Y. Li, W. Gao, X. Huang and C. Ma, Chin. Herb. Med., 2013, 5, 33–46.
2. L. P. Kang, K. Yu, Y. Zhao, Y. X. Liu, H. S. Yu, X. Pang, C. Q. Xiong, D. W. Tan, Y. Gao, C. Liu and B. P. Ma, J. Pharm. Biomed. Anal., 2012, 62, 235–249.
3. S. Man, W. Fan, Z. Liu, W. Gao, Y. Li, L. Zhang and C. Liu, Steroids, 2014, 84, 17–21.
4. S. Man, W. Fan, W. Gao, Y. Li, Y. Wang, Z. Liu and H. Li, J. Ethnopharmacol., 2014, 151, 407–412.
5. S. Man, W. Gao, Y. Zhang, X. Jin, C. Ma, X. Huang and Q. Li, Anal. Bioanal. Chem., 2009, 395, 495–505.
6. T. Xu, S. Zhang, L. Zheng, L. Yin, L. Xu and J. Peng, Food Chem. Toxicol., 2012, 50, 1279–1287.
7. Z. Liu, W. Gao, S. Man, J. Wang, N. Li, S. Yin, S. Wu and C. Liu, J. Ethnopharmacol., 2012, 144, 67–72.
8. S. Man, P. Qiu, J. Li, L. Zhang and W. Gao, Environ. Toxicol., 2015 DOI:10.1002/tox.22215.
9. C. V. Pereira, S. Nadanaciva, P. J. Oliveira and Y. Will, Expert Opin. Drug Metab. Toxicol., 2012, 8, 219–237.
10. N. Izigov, N. Farzam and N. Savion, Free Radical Biol. Med., 2011, 50, 1131–1139.
11. H. Raza, FEBS J., 2011, 278, 4243–4251.
12. S. Man, W. Gao, C. Wei and C. Liu, Phytother. Res., 2012, 26, 1449–1465.
13. A. J. Shuhendler, K. Pu, L. Cui, J. P. Uetrecht and J. Rao, Nat. Biotechnol., 2014, 32, 373–380.
14. R. E. Pacifici and K. J. Davies, Gerontology, 1991, 37, 166–180.
15. X. H. Li, X. Gong, L. Zhang, R. Jiang, H. Z. Li, M. J. Wu and J. Y. Wan, Mediators Inflammation, 2013, 2013, 354087.
16. Q. Ma, Annu. Rev. Pharmacol. Toxicol., 2013, 53, 401–426.
17. A. Minelli, S. Grottelli, A. Mierla, F. Pinnen, I. Cacciatore and I. Bellezza, Int. J. Biochem. Cell Biol., 2012, 44, 525–535.
18. S. Gao, X. Duan, X. Wang, D. Dong, D. Liu, X. Li, G. Sun and B. Li, Food Chem. Toxicol., 2013, 59, 739–747.
19. S. M. Kumar, K. Swaminathan, D. L. Clemens and A. Dey, Eur. J. Nutr., 2015, 54, 223–234.

---