Introduction
Herbs are widely used to prevent and treat diseases without a medical prescription based on the belief that these treatments are considered natural and safe [
]. However, hepatotoxicity associated with herbal or botanical use is increasingly being recognized as the use of these medicines has become widespread in many countries [
,
]. The popular Chinese herb Heshouwu (dried root of
Polygonum multiflorum Thunb., family Polygonaceae) has been used for thousands of years in China to prevent aging and hair graying. This herb has also been commonly used as herbals and dietary supplements (HDS) in Europe and United States in recent years. Nevertheless, reports of Heshouwu-induced liver injury are increasing [
]. The China Food and Drug Administration recently warned of the risks of Heshouwu-containing drugs and strengthened the regulations of these products. In fact, the hepatotoxicity of Heshouwu was reported in 1996 in Hong Kong, China [
] and in 2006 in the UK by the Medicines and Healthcare Products Regulatory Agency. Despite the significant increase in Heshouwu-induced liver injuries, our previous investigations in suspected clinical patients have revealed that such injuries occur only in a minority of patients and are related to idiosyncratic hepatotoxicity [
,
]. The hepatotoxicity of Heshouwu and its related preparations (Shou Wu Pian, Shen Min, etc.) have also been summarized and recorded in the LIVERTOX
® database, a comprehensive resource for idiosyncratic drug-induced liver injury (IDILI), produced by the National Institute of Diabetes and Digestive and Kidney Diseases and National Library of Medicine [
].
Although IDILI often occurs in a minority of patients (generally<1%) [
], it is one of the leading causes of drug development failure and withdrawal of drugs from the market. This problem remains challenging for human health and for pharmaceutical companies and regulatory agencies; it is unpredictable and the mechanisms underlying IDILI are generally unknown. IDILIs are often found after marketing or in the final phase of a clinical study since IDILIs cannot be evaluated in pre-clinical drug safety assessment using healthy animals [
]. Thus, the development of animal models is needed to assess those drugs that cause IDILI. The inflammatory stress hypothesis has provided some of the first animal models of idiosyncratic hepatotoxicity in which nontoxic doses of IDILI-causing drugs are rendered hepatotoxic upon coexposure to a nontoxic but modestly inflammatory dose of bacterial endotoxin (lipopolysaccharide (LPS)) [
,
]. The LPS model has been successfully used to evaluate several drugs known to cause IDILI in humans, such as trovafloxacin, ranitidine, sulindac, chlorpromazine, halothane, monocrotaline, amiodarone, and diclofenac in toxicological assessment [
]. In our previous work, experimental results also demonstrate that combined treatment with nontoxic dose of LPS and therapeutic dose of Heshouwu resulted in acute idiosyncratic liver injury in rats, whereas the solo use of each could not induce any observable hepatotoxicity [
].
However, the hepatotoxic chemicals attributed to Heshouwu-induced idiosyncratic hepatotoxicity remain in dispute [
]. This herb has been shown to contain abundantly the natural stilbene 2,3,5,4'-tetrahydroxy
trans-stilbene-2-O-β-glucoside (
trans-SG) as its major bioactive constituent. Although a recent report named
trans-SG as the major hepatotoxic component in Heshouwu [
], limited evidence exists to confirm this conclusion. The
trans-isomer of stilbenes, such as
trans-SG, can be transformed by ultraviolet light or sunlight into its
cis-isomer (Fig. 1A). Interestingly, the
cis-form of some stilbenes reportedly exhibits stronger cytotoxicity than its
trans-isomer [
,
]. Conversely, the possible association of the ultraviolet-induced
cis-
trans isomerization of
trans-SG with the idiosyncratic hepatotoxicity of Heshouwu is unknown. Opportunities for exposure to ultraviolet radiation in the course of the sunlight drying, preparation, and preservation of the herb are known to be abundant.
This paper provided a representative example of structural isomers having divergent toxicity potentials. With direct experimental evidences, we demonstrated that ultraviolet radiation transformed natural trans-SG into a hepatotoxic cis-isomer. Thus, we provided a rationale for controlling photoisomerization in drug discovery and clinical use of Heshouwu or the other stilbene-related medications.
Materials and methods
Plants and reagents
The dry roots of Polygonum multiflorum Thunb. (PM) were purchased from Beijing Lüye Pharmaceutical Company, and voucher specimen is deposited in China Military Institute of Chinese Medicine. The PM was verified to have met the standards specified byChinese Pharmacopoeia. The trans-SG and cis-SG reference substances were provided by the National Institutes for Food and Drug Control and the Chengdu Chroma-Biotechnology Co., Ltd., respectively. LPS and sodium pentobarbital (Cat#P3761) were purchased from Sigma Chemical Company. LPS was derived from Escherichia coli, 055: B5 (Lot#113M4068V). Assays kits for the detection of plasma alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were purchased from Jiancheng Biological Technology, Co., Ltd. (Nanjing, China). All other reagents and solvents were of the highest grade commercially available.
Study design
This study aimed to screen and identify the culprit responsible for Heshouwu-induced hepatotoxicity. The chemical fingerprints were comparably investigated between regular Heshouwu samples and the Heshouwu herbal liquor belonging to a DILI patient to preliminarily find suspect constituent with association to liver injury. Then, the constituent knock-out/knock-in strategy, which compared a target constituent with a whole-herb extract in a holistic manner, was used to track the major hepatotoxic compound in Heshouwu in a stepwise manner in the LPS-treated idiosyncratic rat model. The hepatotoxicity of cis-SG was further confirmed by evaluating the idiosyncratic hepatotoxicity of isolated cis-SG and trans-SG. The difference in hepatotoxicity between cis-SG and trans-SG was also studied by a metabolomics approach to elucidate the potential mechanism by which cis-SG induces liver injury.
Sample collection and preparation
The PM extracts were prepared through extraction eight times using 50% ethanol–water solution via the cold soaking method, which was repeated twice for 48 h each time. The extracted liquid was mixed, filtered, concentrated under reduced pressure, and dried under vacuum [
].
Trans-SG was weighed precisely, dissolved in methanol to a concentration of 5mg/ml, placed in a plane dish, and irradiated with UV light at 365 nm for 60 min (Fig. 1F–1I). After volatilizing methanol in the fume cupboard, the samples were sealed and preserved in light-proof bags [
].
Constituent “knock-out” and “knock-in”
For constituent “knock-out” (Fig. 2A), the crude extract was diluted with deionized water to a suitable concentration. The diluted extract was then extracted with an equal volume of chloroform five times and subjected to vacuum drying to obtain the chloroform extract (CH). The ethyl acetate extract (EA) was obtained by further extracting the remaining aqueous fraction with an equal volume of ethyl acetate nine times. The remaining ethyl acetate-extracted aqueous portion was evaporated to obtain the Heshouwu residue (RE).
For constituent “knock-in” (Fig. 2D), the negative extract was prepared by mixing the CH extract with the RE extract according to the ratio of the contents of the crude drugs. Then, the EA extract was added into the negative extract at a ratio of 0-, 0.25-, 0.5-, or 1-fold of the content of crude drugs in the EA extract (Fig. 2E).
Assessment of idiosyncratic hepatotoxicity in rats
Male Sprague–Dawley (SD) rats weighing 200±10 g were obtained from the Laboratory Animal Center of the Academy of Military Medical Sciences (License No. SYXK 2007-004) (Beijing, China). They were housed in the Laboratory Animal Center of the 302 Military Hospital (animal ethics’ number 2014095D) and maintained under a 12-h light/dark cycle in a controlled temperature (25±2 °C) and humidity (50%–60%) environment for a period of 1 week before use. The animals were allowed access to food and water ad libitum. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964Helsinki Declarationand its later amendments or comparable ethical standards. All applicable institutional and/or national guidelines for the care and use of animals were followed.
The assessment of idiosyncratic hepatotoxicity was based on our previously reported rat model, which was modified from the literature. In a typical procedure, the animals were intragastrically administered with different extracts of Heshouwu or an equivalent volume of normal saline, followed by a tail vein injection of LPS (2.8 mg/kg, Sigma) or normal saline 3 h later. At 7 h after the injection of LPS, blood and liver samples were collected. The rat liver was fixed in 10% neutral buffered formalin for at least 72 h before being processed for histologic analysis. Paraffin-embedded sections were cut to 4 mm-thick pieces and stained with hematoxylin and eosin for microscopic examination. The abovementioned method also was used to evaluate cis-SG and trans-SG of idiosyncratic liver injury in rats.
Metabolomics analysis
After sample preparation and sample handling, all plasma samples were subjected to LC-MS analysis. Chromatography analysis was performed using an Agilent 1290 series UHPLC system. For mass spectrometry, the Agilent 6550 Q-TOF/MS with an electrospray ionization source (ESI) in both positive and negative mode was used. All data were preprocessed with Profinder (version B.06.00, Agilent Technologies, USA). Then, SIMCA-P 11.0 version (Umetrics AB, Umea, Sweden) were used for multivariate statistical analyses. In the PLS-DA model, only VIP (variable influence on the projection) values>1 were selected as potential biomarker. Finally, identification of these peaks was based on METLIN database, and metaboAnalyst 3.0 was used to identify the enriched pathway disturbed by cis-SG. Full methods and any associated operational details are available in the supplementary materials.
Statistical analysis
The data were analyzed with the SPSS software program (version 17.0, Chicago, IL, USA). Unless otherwise indicated, data are expressed as mean±SD. One-way analysis of variance (ANOVA) was used for statistical analysis of the results. The differences were considered to be statistically significant when P<0.05 and highly significant when P<0.01.
Results
Transformation of cis-SG in the Heshouwu herbal liquor used by a DILI patient
Compared with a regular Heshouwu sample, a higher content of
cis-SG was found in the Heshouwu herbal liquor belonging to a DILI patient (Fig. 1C), which was self-made by soaking Heshouwu in a commercial liquor (alcohol degree= 52% vol), containing about 0.1 g raw Heshouwu per ml liquor. After nearly one month, the patient began to consume it. The patient refused to take any prescription medications, but drank the herbal liquor (40–80 ml once per day) for about two months before he experienced fatigue, poor appetite, and jaundice. The patient was diagnosed with herb-induced liver injury associated to Heshouwu, according to the American College of Gastroenterology clinical guidelines for DILI diagnosis and management [
] and the China Association of Chinese Medicine clinical guidelines for diagnosis and treatment of herb-induced liver injury [
]. Moreover, his detailed information was recorded in the supplementary materials (Supplementary Table 1 and Supplementary Fig. 1). In addition, sunlight transformed
trans-SG into
cis-SG when an extracted solution of the regular Heshouwu sample was exposed to sunlight (Fig. 1E). We further demonstrated that the pure
trans-SG compound in solution was transformed into
cis-SG only in quartz, transparent glass or transparent polyethylene bottles, whereas either light shielding or brown glass bottles prevented isomerization (Fig. 1F–1J). Unsurprisingly, the aforementioned Heshouwu DILI patient had stored his Heshouwu liquor in transparent glass on a windowsill exposed to sunlight throughout his course of treatment with the liquor, and he started consuming it after the liquor was exposed to sunlight for about one month. Thus, the evident transformation of the patient’s Heshouwu liquor into
cis-SG can be explained. Furthermore, a higher content of
cis-SG was also detected in the herb or its relevant Heshouwu-containing preparations, belonging to the actual liver intoxication patients associated with Heshouwu, compared with the general collected Heshouwu samples (unpublished data). These results provided a clinical clue of the relationship between a high level of transformation into
cis-SG and liver injury associated with Heshouwu.
A stilbene-containing ethyl acetate (EA) extract displayed comparable idiosyncratic hepatotoxicity to Heshouwu whole extract
Given the probable association between
cis-SG and idiosyncratic liver injury, we first determined whether an isolated stilbene extract fraction displayed similar idiosyncratic hepatotoxicity to the whole extract of Heshouwu in rats. In these experiments, we used a recently described idiosyncratic hepatotoxicity rat model [
] in which a nontoxic dose of lipopolysaccharide (LPS) was used as a sensitizer to induce idiosyncratic hepatotoxicity similar to that reported in literature [
]. A constituent knock-out strategy was applied (Fig. 2A).
According to the differences in the chemical polarity of anthraquinone and stilbene glycoside, chloroform (CH) and ethyl acetate (EA) were selected as the solvents for extracting these two kinds of constituents from whole extract of Heshouwu (PM). The CH and EA extracts were obtained sequentially, as was the residue extract (RE). In HPLC fingerprints, we observed that the CH and EA extracts and the RE contained chemicals with different deduced polarities (Fig. 2B). The CH extract contained only anthraquinones, the EA extract contained mostly stilbenes and trace anthraquinone glycoside, and the RE extract contained neither anthraquinones nor stilbenes (Supplementary Table 2). The dosage of PM for evaluating idiosyncratic hepatotoxicity was set according to a previous work [
], and those of the other extract fractions (EA, CH, and RE) were set according to their extraction yields to achieve a dosage comparable to that of the PM extract.
As reported previously [
], low-dosage LPS treatment did not induce a significant (
P>0.05) increase in plasma aspartate aminotransferase (AST) and alanine aminotransferase (ALT), nor were histologic changes induced in the liver, demonstrating that nontoxic LPS alone was not associated with liver injury (Fig. 2C, Supplementary Fig. 2A and Supplementary Fig. 3). Administration of either PM or the other isolated extract fractions (EA, CH and RE) to normal rats failed to induce a significant (
P>0.05) increase in plasma ALT or liver histologic changes. In addition, administration of the CH or RE extract to LPS-treated rats failed to significantly (
P>0.05) increase plasma ALT or AST. By contrast, administration of either the PM or EA extract to LPS-treated rats resulted in liver histologic changes and markedly increased plasma ALT and AST compared with both the normal and LPS control groups (Fig. 2C, Supplementary Fig. 2A and Supplementary Fig. 3). These data suggested that the isolated stilbene fraction (EA extract), and not the anthraquinone fraction (CH extract) or other fraction (RE extract), probably played the dominant role in the pathogenesis of the idiosyncratic hepatotoxicity of Heshouwu.
To further determine the role of stilbenes in the pathogenesis of the idiosyncratic hepatotoxicity of Heshouwu, we administered stepwise amounts of the EA extract added to a mixture of the other fractions (the CH and RE extracts) to LPS-treated rats (Fig. 2D and 2E). Administration of the mixture of CH and RE with 0% (without EA), 25% or 50% of the full amount of EA did not result in any significant increase in ALT in either the normal or LPS-treated rats. By contrast, administration of the mixture of CH, EA, and the full amount of EA (equal to the whole extract of Heshouwu) to LPS-treated rats resulted in significantly increased ALT and AST as well as liver histologic changes, compared with all of the control groups (Fig. 2F, Supplementary Fig. 2B and Supplementary Fig. 3). These data, along with the aforementioned results, demonstrated that the stilbene-containing EA extract played a key role in the pathogenesis of the idiosyncratic hepatotoxicity of Heshouwu.
Isolated cis-SG displays comparable idiosyncratic hepatotoxicity to the EA extract
To determine which of the two isomeric stilbenes was associated with idiosyncratic hepatotoxicity, we administered isolated cis- and trans-SG to LPS-treated rats to compare the hepatotoxic intensity of the two compounds. Results showed that the administration of either cis-SG (at a dosage comparable with its content in the EA extract) or the EA extract resulted in significantly increased plasma ALT and AST and liver histologic changes (Fig. 3A and 3B). By contrast, the administration of trans-SG (at a dosage comparable with its content in the EA extract) was not associated with liver injury. In addition, the administration of emodin glycoside (EG), a trace component in the EA extract, did not display liver injury. These results demonstrated that cis-SG was the major hepatotoxic component in the EA extract. We also determined plasma IFN-g, TNF-a, and IL-6 contents, and results showed a significant increase in these inflammatory cytokines in the LPS/cis-SG group compared with the LPS/trans-SG group (Fig. 3C).
Cis-SG alters distinct enriched pathways from trans-SG
We performed metabolomics analysis to reveal the enriched pathways involved in the divergent types of idiosyncratic liver injury induced by cis- and trans-SG. Initially, unsupervised principal component analysis (PCA) showed that the trend in the LPS/cis-SG group deviated from the normal control group to a much greater extent than the trend in the LPS/trans-SG group (Fig. 4A and 4B) and was almost completely distinct from the trend in the LPS/trans-SG group (Fig. 4A). To explore and compare the metabolic alteration of LPS-treated rats treated with cis-SG and trans-SG, multivariate statistical analysis including PCA (Fig. 4C) and partial least squares-discriminant analysis (PLS-DA) were applied to LPS/cis-SG and LPS/trans-SG groups(Fig. 4D). To identify differentially expressed metabolites, the loading plot of supervised PLS-DA was used to investigate differences between the LPS/cis-SG and LPS/trans-SG groups. Fig. 4D and 4E display the result of PLS-DA model derived from the data of ESI+ analysis. The R2Y and Q2Y (predictive ability) were 0.964 and 0.510, respectively. Permutation tests with 100 iterations were further performed. As shown in Fig. 4F, the model is not over-fitted. Similarly, the PLS-DA model derived from the data of ESI− analysis was constructed (Supplementary Fig. 4). The LPS/cis-SG group can be separated from LPS/trans-SG group clearly both in the two models. All parameters of PCA and PLS-DA model are listed in Supplementary Table 3.
The metabolite ions with a VIP value>1 were kept for further study. Potential biomarkers were identified and are displayed in the loading plots (Fig. 4E and Supplementary Fig. 4C), variables with a VIP value>1 were marked with a red square. Subsequently, metabolites that differed significantly between LPS/trans-SG and LPS/cis-SG groups were selected as candidate potential biomarkers. The criteria were further restricted to features with an average intensity difference of 1.5-fold between LPS/trans-SG and LPS/cis-SG groups. Finally, 11 metabolites in the positive mode and 10 metabolites in the negative mode in LC-MS (a total of 19 metabolites, with two metabolites both found in the ESI+ and ESI− mode), were tentatively identified by using Biofluid Metabolites Database (http://metlin.scripps.edu). They are summarized in Table 1, and the relative mean peak areas of different metabolites are graphed in Supplementary Fig. 5. The relative concentration of 19 endogenous metabolites was significantly affected by LPS/cis-SG treatment, so these metabolites can be considered as potential markers for biological pathway analysis.
The changed concentrations of differentially expressed metabolites (Table 1) suggested that the disturbed enriched pathways in idiosyncratic liver injury rats, such as arginine and ornithine metabolism, vitamin B6 metabolism, pyrimidine metabolism, pantothenate, and CoA biosynthesis, were affected by the administration of cis-SG. As is shown in Supplementary Table 4 and Supplementary Figs. 5 and 6, among the eight LPS/cis-SG-disturbed enriched pathways, the most affected were those of arginine and ornithine metabolism, vitamin B6 metabolism, and pyrimidine metabolism.
Discussion
Despite the potent beneficial effects of natural stilbenes in herbs [
], concerns regarding the nephro- and hepato-toxicity of such compounds have risen in recent years [
]. Stilbenes were initially identified as phytoalexins [
], which are produced in response to stresses such as wounding, pathogen attack, or ultraviolet radiation, as well as increased phytotoxic activity, in plants [
]. Stilbenes exist predominantly in the
trans-form with a relatively low energy level [
,
]. Differences in the pharmacological and cytotoxic effects of (
Z)- and (
E)-3,4,3′,5′-tetramethoxystilbene isolated from
Eugenia rigida have been reported [
], and
cis-form of some stilbenes have been shown to possess more potent cytotoxicity than the
trans-isomers
in vitro [
,
]. In the present study, using a constituent knock-out/knock-in strategy [
,
], we tracked the major idiosyncratic hepatotoxic compound in Heshouwu in a stepwise manner, from the total extract to the stilbene-containing EA extract and then to
cis-SG (Figs. 2 and 3). Results indicated that
cis-SG was an ultraviolet-transformed isomer much more idiosyncratic hepatotoxic than
trans-SG, which is the predominant form of natural stilbene in Heshouwu. Furthermore, transformation to
cis-SG was evident in the Heshouwu herbal liquor of a DILI patient (Fig. 1C), which provided clinical evidence of an association between a high level of
cis-SG transformation and idiosyncratic liver injury.
In a previous retrospective investigation of patients with Heshouwu-induced liver injury, we found that a considerable proportion of the patients took wine-soaked Heshouwu [
,
]. Considering that the process of wine-soaking of herbal medicines usually takes several months or even years, these medicines are likely to be exposed to sunlight or ultraviolet light. In China, most people usually preserve wine-soaked Heshouwu in a transparent glass container and place it on a windowsill where sunlight could shine through the window during sunny days. As indicated in our experiment (Fig. 1F–1H), sunlight or ultraviolet exposure transformed
trans-SG into
cis-SG and consequently increased the risk of liver injury. Notably, the patient described herein had, in fact, exposed his Heshouwu liquor to sunlight about one month, began to digest it, had a high level of
cis-SG transformation; thus, his hepatotoxicity symptoms were explainable. After further examination, we also found the evident transformation of
cis-SG in other clinically obtained Heshouwu samples consumed by actual patients with Heshouwu liver intoxication (unpublished data).
Notably, liver injury induced by Heshouwu is a multifactorial disease. The administration of Heshouwu or SG (either
trans- or
cis-SG) alone at clinical dosages cannot sufficiently induce liver injury in normal subjects (Fig. 3A); only a combination of hepatic immune activation and exposure to
cis-SG, which was transformed from
trans-SG, can induce idiosyncratic liver injury. Hepatic immune activation can occur under certain circumstances, such as in the presence of (1) intestinal injury-induced endotoxin overabsorption into the liver, which activates Kupffer cells and immune responses [
]; and (2) pre-existing liver inflammation and immune stress. Thus, in persons with hepatic immune activation or intestinal injury, the intake of herbs or drugs containing hepatotoxic
cis-stilbenes can be deduced to result in exposure to liver injury risk. Thus, elevated concentrations of
cis-stilbenes in isolation may not lead to liver injury in normal subjects; nonetheless, in the setting of hepatic immune activation, experimental evidence indicates that
cis-stilbenes are key factors in the pathogenesis of the idiosyncratic hepatotoxicity of Heshouwu.
Based on the idiosyncratic hepatotoxicity phenotype associated with
cis-SG versus
trans-SG, relevant differentially expressed metabolites biomarkers and enriched pathways were investigated. As shown in Fig. 5 and Supplementary Fig. 6, arginine and ornithine metabolism, vitamin B6 metabolism, and pyrimidine metabolism were the most affected pathways. Previous studies have shown that metabolic disorder of vitamin B6 occurs parallel to major hepatocyte destruction, and patients with liver disease often went hand-in-hand with metabolic defect in vitamin B6 [
,
]. Other research has revealed that pyrimidine metabolism is closely related to pathological condition of liver [
,
]. Taking together, we speculated that the perturbation of vitamin B6 metabolism and pyrimidine metabolism were the result of liver damage. Notably, the downregulation of arginine and omithine was more evident in the
cis-SG group compared with the
trans-SG group, indicating arginine and ornithine metabolism disorders. Previous studies have shown that an early response to inflammation is the conversion of arginine to the cytostatic molecule nitric oxide (NO), and the breakdown of arginine to urea and ornithine by arginase also contributed to inflammatory response [
,
], potentially explaining the markedly higher inflammation and liver injury observed in the
cis-SG group than in the
trans-SG group. Moreover, arginase, an important enzyme that can convert arginine into ornithine and urea, has been shown to be inducible in macrophages by LPS [
]. Arginase is crucially involved in various aspects of inflammation and has been shown either to be responsible for or to participate in, for example, inflammation triggered immune dysfunction, immunosuppression, and immunopathology of infectious diseases [
]. The disorders of arginine and ornithine metabolism, together with the administration of LPS, were likely to amplify immuno-inflammatory responses and further aggravate liver injury. Associated with the aforementioned immune activation involved in the pathogenesis of liver injury, interactions between the immune and metabolic systems may play a key role in the biological mechanism of
cis-SG-induced idiosyncratic hepatotoxicity.
This study has some limitations. Despite experimental evidence demonstrating that idiosyncratic liver injury can be induced by either Heshouwu or cis-SG, systematic clinical elucidation and evidence are still needed. Additionally, the exact biological mechanism underlying Heshouwu (or cis-SG)-induced idiosyncratic hepatotoxicity still requires clarification.
In summary, the present study provided experimental evidences that sunlight or ultraviolet radiation transformed natural trans-stilbene into an idiosyncratic hepatotoxic cis-isomer. Interestingly, trans- and cis-stilbene show the opposite effects similar to Yin (meaning dark) and Yang (meaning sunlight), which suggested the metaphor of sunlight-derived isomerization. Therefore, the present study provided a rationale for controlling photoisomerization in drug discovery and clinical use of stilbenes and related phytomedicines, as well as a representative example of structural isomers having divergent toxicity potentials.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(690KB)
