Toxicology Evaluation of Realgar-Containing Niu-Huang-Jie-Du Pian as Compared to Arsenicals in Cell Cultures and in Mice

Niu-Huang-Jie-Du Pian (NHJD) is a widely used traditional Chinese medicine containing realgar (As4S4). Realgar has been included in many traditional medicines, but is often taken as arsenite for risk assessment. To evaluate true risk of realgar and realgar-containing NHJD, their toxicity was compared with common arsenicals. In cultured cells, the LC50 for NHJD (1200 μM) and realgar (2000 μM) was much higher than arsenite(35 μM), arsenic trioxide (280 μM), and arsenate (400 μM). Acute toxicity in mice showed more severe liver and kidney injury after arsenite or arsenate, but was mild after realgar and NHJD, corresponding to cellular and tissue arsenic accumulation. The expressions of arsenic-sensitive stress gene metallothionein-1 were increased 3–7-folds after arsenite or arsenate, but were unaltered after NHJD and realgar. Thus, realgar and NHJD are much less toxic than arsenite and arsenate. The use of total arsenic to evaluate the safety of realgar and realgar-containing NHJD is inappropriate.


Introduction
Realgar (90% as As 4 S 4 ) has been used in traditional Chinese medicines [1,2] and in Indian Ayurvedic medicines [3] for thousands of years and claimed to have therapeutic effects in these remedies. However, arsenic (As) is a highly toxic substance and its risk in traditional remedies is of concern [4][5][6]. Hundreds of traditional medicines are forbidden in the USA or European market because of the contents of As are higher than the allowable limits for food and drugs [2], and over 20% of online-sold Ayurvedic medicines were demanded for rigorous regulation for heavy metal contents [6]. The Chinese Pharmacopeia Committee has reduced the allowable As contents in traditional medicine recipes by as much as 65%, but As contents are still thousands-fold over the health food standards. More studies are therefore recommended to evaluate the true risk of metal-containing traditional medicines [7].
We have recently shown that chemical form of realgar (As 4 S 4 ) is a major determinant for its disposition and toxicity. For example, realgar-containing An-Gong-Niu-Huang Wan ( ) was much less toxic in cultured cells [8], in acute animal studies [9], and in subchronic toxicology studies [10,11]. The present study was undertaken to evaluate the true risk of a realgar-containing NHJD, a most popular patent Chinese medicine, in cultured cells and in intact animals, as compared to common arsenicals. The results 2 ISRN Toxicology fortified our prior conclusions that chemical forms of metals are very important in determining the disposition and toxicity of metal-containing traditional medicines.

Cell Cultures and Treatments.
Human pharyngeal carcinoma FaDu cells were obtained from Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China) and cultured in DMEM media supplemented with 10% fetal bovine serum (FBS) with penicillin and streptomycin. Cells were cultured at 37 • C in a 5% CO 2 -humidified atmosphere. All chemicals were dissolved in dimethyl sulfoxide (DMSO) in 10-30 mM and serially diluted in culture media before addition to the cultures at 5-6 different concentrations as indicated. For cytotoxic assay, chemicals were added to 96wells when initiating the culture, and 48 h later, the cytotoxicity was measured by the MTS assay as described previously [8]. For As uptake and accumulation studies, cells were exposed to the same concentration (100 μM) for 30-120 min, and cellular As was determined after thoroughly washing.

Animals and Treatments.
Adult male and female Kunming mice (22 ± 2 g) were purchased from the Animal Center of the Third Military Medical University (Chongqing, China). Mice were kept in a regulated environment (22 ± 1 • C, 50 ± 2% humidity) with a 12 h : 12 h light:dark cycle. All animal procedures follow the WHO Guidance of Humane Care and Use of Laboratory Animals.
Mice were orally administered with NHJD (600 mg/kg), realgar (600 mg/kg), and the equal amount of As doses as sodium arsenite (36 mg/kg) and sodium arsenate (88 mg/kg). The NHJD 600 mg/kg is approximately 2 times of clinical dose (2000 mg/day/60 kg person, taken mouse-human extrapolation factor of 10). Animals were monitored closely for clinical symptoms after gavage, and tissues were harvested for analysis 8 hrs later. The doses of arsenicals selection were based on our recent publications [9,12].

Blood
Biochemistry. Serum was separated from whole blood by standing for 1 hr, and blood biochemistry was determined with an autoanalyzer (GLAMOUR1600). The serum activities of alanine aminotransferase (ALT) and concentrations of blood urea nitrogen (BUN) were quantified to evaluate the hepatotoxicity and nephrotoxicity of animals treated with NHJD and various arsenicals.

Arsenic Determination.
Total arsenic contents in cells and tissues were analyzed by atomic fluorescence spectrometry (AFS) as described previously [12]. Briefly, tissues were completely digested in nitric acid at 170 • C for 2.5 hrs and brought to 25 mL with distilled water, and 5 mL of the sample was mixed with 1 mL 5% thiourea-ascorbic acid solution. Following 30 min incubation, aliquots were used for quantification of As contents with atomic fluorescence spectrometry (Kechuang Haiguan Instrument Co. Ltd, Beijing, China). These assays were performed at the Guizhou Chemical Analysis Center of Chinese Academia of Sciences [12].
2.6. RNA Isolation and Real-Time RT-PCR Analysis. RNA isolation and real-time RT-PCR analysis approximate 50-100 mg tissue was homogenized in 1 mL TRIzol agent (Invitrogen, Carlsbad, Calif, USA), and total RNA was extracted according to manufacturer's instructions, followed by purification with RNeasy columns (Qiagen, Valencia, Calif, USA). The quality of RNA was determined by the 260/280 ratios, and by gel electrophoresis to visualize the integrity of 18S and 28S bands. Total RNA was reverse transcribed with MMLV reverse transcriptase and oligo-dT primers. The PCR primers were designed with Primer Express software (Applied Biosystems, Foster City, Calif, USA) as MT-1 (BC027262), forward: AATGTGCCCAGGGCTGTGT; reverse: GCTGGGTTGGTC-CGATACTATT. The Power SYBR Green Mater Mix (Applied Biosystems, Foster City, CA, USA) was used for real-time RT-PCR analysis. The cycle threshold (Ct) values of the interested genes were first normalized with β-actin of the same sample and expressed as percentage of controls.

Statistical Analysis.
For cytotoxicity analysis, means and standard error were calculated from 5 separate cultures, and the LC 50 values were estimated via graphics. For animal studies, means and standard error of 6 mice were calculated. Data were analyzed using a one-way analysis of variance (ANOVA), followed by Duncan's multiple range test. The significant level was set at P < 0.05 in all cases.

Accumulation of Arsenic in Human Pharyngeal FaDu
Cells. Confluent FaDu cells were exposed to NHJD and arsenicals, all at the 100 μM concentrations for 30 min, and cultures were thoroughly washed and cellular uptake of As were determined as described in Methods. The results ( Figure 2) show dramatic difference in cellular arsenic accumulation, NHJD, and realgar treatments resulted in 100-200 ng As/mg cellular protein, while sodium arsenite and sodium arsenate resulted in 1360 and 470 ng As/mg  protein, respectively. Similar results were also obtained after 60 min and 120 min exposure (data not shown).

Blood Biochemistry of NHJDP and Arsenical Treatments in Mice.
Mice were orally administered with NHJD (600 mg/ kg), realgar (600 mg/kg, equivalent to reaglar in NHJD), and the equal amount of As doses of sodium arsenite Data are mean ± SE, n = 6. * P < 0.05.
(36 mg/kg) and sodium arsenate (88 mg/kg). Animals were killed 8 hr later and blood biochemistry was performed. The results ( Table 1) show that the elevations of serum ALT and BUN after arsenite and arsenate, but these parameters were unaltered after NHJD, and realgar. Histopathology was consistent with blood biochemistry, showing more severe liver and kidney damage after arsenite and arsenate, but mild or absent after NHJD and realgar [13].

Accumulation of Arsenic in Liver and Kidneys after NHJD and Arsenical Treatments in Mice.
Mice were orally administered with NHJD (600 mg/kg), realgar (600 mg/kg, equivalent to realgar in NHJD), and the equal amount of As doses of sodium arsenite (36 mg/kg) and sodium arsenate (88 mg/kg). Animals were killed 8 hr later and tissue As accumulation was determined. The results (Figure 3) show that the dramatic difference in tissue arsenic accumulation, NHJD and reaglar treatments resulted in approximately 200 ng As/g liver, while arsenite (6200 ng/g liver) and arsenate (3320 ng/g liver) produced significant As accumulation. Renal As contents were under detection limits for NHJD and realgar, but reached 3350 ng/g kidney after arsenite and 1500 ng/g kidney after arsenate. Figure 4 shows the expression of metallothionein-1 (MT-1) in liver and kidney. MT is a small, cysteine-rich, metalbinding protein playing an important role in metal detoxication [14]. MT overexpression is a sensitive biomarker for arsenic-induced stress [15]. As shown in Figure 4, hepatic MT-1 was increased 5-7-folds after arsenite and arsenate, but was not altered after NHJD and realgar. As shown in Figure 4, bottom, renal MT-1 transcript levels were also increased after arsenite (5-fold) and arsenate (2-fold), respectively. In comparison, NHJD and realgar did not produce significant elevation in renal MT-1 transcripts.

Discussion
The present study clearly demonstrated that realgar and realgar-containing NHJD were much less toxic than sodium arsenite and arsenate in cultured cells and in mice, as evidenced by LC 50 values, the elevated serum ALT and BUN concentrations in mice. Toxicokinetically, much less As was accumulated in the cells or in tissues after realgar and NHJD. Furthermore, the expression of stress-related genes, namely, MT-1, was increased 2-7-folds after sodium arsenite and arsenate in the liver and kidneys, but was basically unaltered after realgar and NHJD, consistent with tissue damage. The present study fortified our recent observations in cultured cells [8] and in intact animals [9][10][11][12][13], indicating that realgar and realgar-containing NHJD are different from sodium arsenite and arsenate and clearly demonstrate that the chemical forms of arsenicals underlies their disposition and toxicity potentials. Arsenic has been used as a remedy and a poison since ancient times. In addition to the use of arsenic compounds in cancer chemotherapy [16,17], realgar has been included in 22 oral patent traditional remedies according to Pharmacopeia of China [1], and was claimed to have many beneficial effects for various diseases [1-3, 16, 17]. However, the major concern is their toxicity potential, which limits many realgar-containing remedies. NHJD is a most popular realgar-containing traditional Chinese medicine (http://www.wickpedia.org/) available in Chinese grocery stores in the USA and in the Europe and is used for many inflammatory diseases [1]. To critically evaluate its toxicity potential is important for safely use of realgar and realgarcontaining traditional medicines in the treatment of various inflammatory diseases.
It is generally assumed that the severity of poisoning is related to the total amount of poison ingested, and assessment of health risk associated with arsenic exposure from human ingestion of traditional medicines has typically taken this tactic [4][5][6][7]. However, the present study clearly showed that realgar was poorly accumulated into the cells and tissues and was unable to reach a critical concentration to cause tissue damage as compared to the same amount of arsenic given as arsenite and arsenate, with tremendous differences in their tissue contents. The disposition of these arsenicals in the body depends on various key factors including solubility, absorption, distribution, and excretion [2]. For example, the average total arsenic in NHJD is about 7 ± 1% (i.e., 70,000 ppm), corresponding to 28 mg arsenic per pill, of which only 1 mg arsenic finds its way into the blood stream, that is, only 4% of intake realgar is bioavailable [19]. Compared to realgar, 80% of orally given sodium arsenite and arsenate can be absorbed from the gastrointestinal tract, with much higher plasma As levels and tissue distribution [2,18]. The bioavailability is a critical determinant of efficacy and toxicity of arsenical compounds. Thus, it is the amount of toxicants to the target organ, rather than the amount ingested or inhaled that makes a poison [20].
Various stresses have been proposed as an important mechanism involved in arsenic toxicity and carcinogenesis.
MT is a small, cysteine-rich, metal-binding protein playing an important role in metal toxicity [14]. Induction of MT by arsenicals is dependent on arsenic forms, with sodium arsenite is the most potent and efficient inducer followed by arsenate and organic arsenicals, such as MMA and DMA [21], and MT-null mice were susceptible to chronic arsenic-induced hepatotoxicity and nephrotoxicity [22]. MT induction is considered an adaptive mechanism to toxic metal-induced oxidative stress [14,15] and can be used as an indicator for arsenic-induced stress. In the present study, hepatic and renal MT was induced by sodium arsenite and arsenate, but was basically unaltered by realgar and NHJD, fortifying the observations from blood biochemistry that realgar and NHJD are much less acutely toxic than sodium arsenite and arsenate.
Traditional medicines are based on empirical experience and have their own theory and are generally safe at clinical doses. The regulation of metal-containing traditional medicines has been a topic of debate [2-7, 13, 23-25]. We have been invited to write two reviews on this topic for arsenic (J Pharmcol Exp Ther [2]) and for Hg (Exp Biol Med [26] and have challenged this bias by studying the relative safety of Liu-Shen-Wan ( , As) [12], An-Gong-Niu-Huang-Wan ( , As+Hg) [9][10][11], and Zhu-Sha-An-Shen-Wan ( , Hg) [27]. In the present study, NHJD is unique in that it is the most popular traditional medicine available not only in China [1], but also in Chinese grocery stories worldwide. To study its relative safety is of clinical significance.
Realgar is less toxic than arsenic oxide yet effective in cancer chemotherapy based on the recent literature [23]. However, this does not imply that realgar and NHJD are nontoxic. High dose of realgar for the long-term use did produce toxicity to the liver and kidneys [24], similarly, longterm use NHJD was reported to produce hepatotoxicity and nephrotoxicity [25]. In the evaluation of realgar and realgarcontaining NHJD in the treatment of various diseases, the balance of the benefit and risk is important. "Dose makes a poison." Although realgar and NHJD are less toxic than sodium arsenite and arsenate, high-dose and longterm administration should be avoided to reduce undesired adverse effects and toxicity [2,18].
In summary, the present study clearly demonstrated that realgar and realgar-containing NHJD are much less toxic to the cultured cells and to the intact animals, as compared to sodium arsenite and arsenate. The chemical forms of arsenicals determine their tissue accumulation and toxicity potentials, and thus the use of the total content of As to evaluate realgar-containing traditional medicines is inappropriate.