Dietary Analysis Essay

ABSTRACT
Background: Botanical dietary supplements are used by 80% of American women aged 40 to 60 years old to reduce menopausal symptoms. Previous researches have not proven concrete associations between the consumption of botanical extracts and liver failure, even though many people are concerned about liver toxicity caused by an intake of botanical products.
Objective: The objective of this paper is to identify the molecular structure of chemical constituents of commonly used botanicals and to investigate the relationship of botanical supplements to liver toxicity.
Design: Botanical chemical constituents were identified from the Natural Products Alert Database for the periods coverings at least 1975 through 2003. This database includes

natural clinical studies, metabolism data, and ethno-medical information on more than 20,000 species of plants collected through the UIC Health Science Library, public journals, as well as foreign journals published in non-English languages. Using a specialized computational software program, (Q) SAR models and chemo-informatics analyses were conducted using silico screening of the chemical constituents in the botanicals for various hepatotoxicity endpoints. Five models were employed, including 1) BioEpisteme(version 4.1); 2) Derek for Windows; 3) Lead-scope Model Applier(LMA); 4) MetaDrugg(version 6); and 5) MC4PC(version 2.1.0.11)
Results: From four popular botanical products, including black cohosh, red clover, chastberry, and hops, a total 577 chemicals were identified. Among the 577 chemicals, 37 chemicals are duplicated, and 540 are unique chemicals. Four chemical constituents, including protocatechuic acid(PA), benzofuran, compound 4-vinylphenol, xanthohumol I(XI) and glycosides combined with a flavone (or isoflavone) have been proven to have hepatic toxicity.
Conclusion: This paper is worthwhile for identifying data relating to molecular structure of four selected botanical products. Endpoints of liver toxicity that were predicted from available computational hepatoxicity models, and verified in animal tests have revealed the following information on the proper dosages to decide hepatic toxicity. Benzofuran’s lethality dosage is 250mg/kg/day. 4-Vynylphenol’s lethality dosage is 200mg/kg/day. Flavone or isoflaveone coupled with glycosides are important structure for predicting positive effects. After reviewing this minimum level of toxicity, future researches can also design functional decision experiments on potential toxicity through mixing of other molecule level of chemical components.

INTRODUCTION
Black cohosh, red clover, chastberry, and hops are commonly used botanical products for American women aged 40 and 60 years old. Although the endpoints of toxicity for each botanical product are well known, molecular level of materials that may impart changes in toxicity is poorly characterized. This paper is designed to allow consumers to decide how much of these botanical products they can ingest without risking toxicity and to allow future researchers to design study guideline, in which other molecule level of chemical components can be mixed to predict and prove the outcome of toxicity.

According to the North American Menopause Society, American women aged 40 and 60 years old, who experience menopausal symptoms, prefer to take botanical products as an alternative to Menopausal hormone therapy (MHT). This is because of the reported side effects of MHT therapy, which include hot flashes, night sweats, rapid heartbeat, migraines, vaginal atrophy, and joint pain. Hormone treatment such as estrogen therapy can relieve menopausal symptoms but can also cause side effects and health risks.

Most frequently used botanicals for treating the symptoms of menopause are the root and rhizome of black cohosh, the fruit of chasteberry, the dried female flowering part of hops, and the flowering tops of red clover. Even though there is not concrete evidence of associations between consumption of botanical products and liver failure, many people are generally fearful of the higher level of liver toxicity associated with a long-term consumption of botanical products. However, molecular levels of materials that may impart changes in toxicity are poorly characterized, even though the end points of toxicity for each botanical product are well known.

However, safety risks are not clearly described on many dietary product labels. Recently, the United States Food and Drug Administration (FDA) recognized the seriousness of the risk that was caused by the intakes of chemical components and has been trying to ensure transparency through predictable regulatory guidelines and procedures, in support of development of nano-technology analysis .

Therefore, we need to identify associations between commonly used botanical products and liver toxicity. This paper is designed to help consumers decide what dosages botanical products they can properly take to avid the risk of toxicity and to assist researchers in designing future researchers in designing the guideline, in which other molecule level of chemical components can be mixed to predict and prove the outcome of toxicity.

METHODS
Unique botanical chemical components from four botanical products including black cohosh, red clover, chastberry, and hops, were reviewed based on the publicly available National Products Alert Database. This database includes comprehensive literature dating from at least 1975 through 2003 discussing natural products clinical studies, metabolism data, and ethno-medical information on more than 20,000 species of plants collected through the UIC Health Science Library, public journals, as well as foreign journals published in non-English languages.

In order to analyze the unique chemical constituents of these botanical products at molecular level, five computational software programs were utilized because in silico predictive approaches, (Q) SAR models and chemoinformatics can be complemented.. The five applied models include 1) BioEpisteme(version 4.1); 2) Derek for Windows; 3) Lead-scope Model Applier(LMA); 4) MetaDrug(version6); and 5) MC4PC(version 2.1.0.11).

In prediction of positive toxicity, prediction of positive toxicity followed the criteria set forth in each software program. The Leadscope Model Applier predicted the probability as “positive” when chemical constituents with above 70 % of confidence are positive. In MetaDrug, at least 60% of the identified botanical constituents must be covered by the model training set with test data showing at least 70% confidence. The MC4PC considered the prediction level is significant when at least one component is positive among the components of less than two unknown fragments. In BioEpisteme, chemicals are predicted as positive when at least half of the available sub-medical components are positive.

After reviewing these computational software models, researchers were able to determine toxic positive constituents. Selected molecular levels of constituents were verified in animal tests that are currently available, and minimum level of dosages to affect liver toxicity were determined.

RESULTS
Association between consumption of botanical supplements and liver toxicity was reviewed by three levels’ analysis in order. First, in the review of the NAPALERT database and a FDA Bioscience Library literature search, chemical components were screened and the unique components were extracted. Second, in five available computational software programs that can examine at a 2D molecular structure level to be able to transform into an electronic format, positive toxicology is predicted at molecular level. Third, the predictions in software programs were verified in available animal tests or human tests. These results are depicted in Table 1.

Table 1
Molecular level of chemical components and minimum level of dosage and its verification of lethality
Botanicals 1st Level Screening of Chemical Constituents Predicted Positive Predicted Positive at Molecular Level in Computational Software Programs Verified Lethality in Animal Tests or Human Tests
Red Clover 17 out of 94 chemicals Isoflavone Not concrete dosage is verified
Chasteberry 22 out of 153 chemicals Benzofuran Necrosis in male/female rats and mice at 13 weeks 125,250, 500 oral doses.Lethality at 250mg. Negative liver histopathology at 62.5mg(NTP, 1989)
4-Vynyl phenol Necrosis and GSH depletion in mice, 50-400 acute intraperitoneal injection lethality at 200 (turner et al., 2005; Vogie et al., 2004)
Black Cohosh 37 out of 103 chemicals Protocatechuic Acid(PA) 3.5-fold increase in liver enzymes and 30% decrease in GSH at 50,500 i.p. 6h in mice. Mild increase in liver enzymes after 60 day. 0.1% in drinking water(Nakamura et al., 2001)
Hops 14 out of 227 chemicals Xanthohumol I(XI) No effects on liver enzymes after 4 weeks in mice, 50 µM in drinking water (Vanhoecke et atl., 2005)

1. Chemical Constituents Screened from NAPALERT and FDA database
The first step in screening botanical chemicals is to identify unique chemicals contained in the four major botanicals studied.

Computational software of (Q)SAR is used for screening chemical components that were identified from the review of literature in the NAPALERT database and a FDA Bioscience Library literature search. From this screening of chemical components, positive predictive lists were completed, which showed that there were 37 positive components out of 103 unknown chemical constituents in black cohosh, 17 positive components out of 94 chemical constituents in red clover, 22 positive components out of 153 chemicals constituents in chasteberry, and 14 positive components out of 227 chemical constituents in

hops.

2. Computational Software Analysis

a. LMA Test and Predictive Scaffolds
A LMA test is used for identifying a molecule level of constituents and predicting negative and positive components through in silico prediction, which can then be interpreted by visual review and quantitative analysis. LMA uses eight criteria for calculation; parent molecular weight, AlogP, polar suface area, hydrogen bond donors, number of rotational bonds and Lipinski score. In order to result in a positive, probability must be at least 70 % positive. However, black cohosh is not analyzed with LMA because their constituents are not predicted positive in LMA test. LMA can analyze atom weight and illustrate the features of the molecule that are not represented in the QSAR model. According to LMA analysis, positive structures are predicted when the sugar moiety of glycosides is coupled with a flavone(or isoflavone) backbone, and the 4-hydroxyacetophenone in flavonoid backbones.

Distinguishing features between positive and negative prediction are dependent on 4 common structural features; the sugar moiety of glycosides, and compounds containing flavone and isoflavone backbones.

b. Percent Concentrations and Exposure Level
From the positive predictions of (Q)SAAR software model and LMA testing, research identified average exposure level of botanical constituents to be determined as having positive liver effect. Average of dosage for calculation is based on a standard body weight of 60 kilogram.

c. Ledscope Structural Analysis
Each of the end points of toxicity are confirmed with toxicity data. Leadscope analysis focuses on functionality. From the analysis of functionality, it was found that the presence of glycosides only significantly contributes to the positive prediction when coupled with a flavone(or isoflavone) backbone. Leadscope enterprise generates the scaffold, which can distinguish components from negative constituents. The outcome assures that isoflavone is only important constituent when liked with the glucoside, in which the negative constituents are eliminated.

d. Computational Prediction and Structural Analysis
Researchers found that isoflavone is the main positive constituent that affects liver toxicity through identifying botanical data and using LMA and Leadscope Enterprise. Researchers confirmed the same outcome by reviewing secondary resources of other research papers. Although other researchers do not control dosage at 60mm for a daily intake, other research papers are based on an experiment of a long-take of selected botanical products, which outcome is compared with non-taken group. In animal test, liver toxicity is determined by the weight increase of the liver.
However, this test has limits when mixing multiple components because it is unable to examine a mixture of chemicals in toxic data and the (Q)SAR approach.

3. Proof of Validation of Each Chemical Constituent: Prediction is compared with animal testing data
a. Protocatechuic acid(PA)
Protocatechuic acid(PA), contained in black cohosh, was identified as a positive chemical components through reviewing the QSAR computational modeled endpoints. Literature study of animal studies proved that the toxicity outcome was the same as predicted in computational models. In several in-vivo mouse toxicological studies, it demonstrates that PA induces depletion of hepatic glutathinoe(GSH) by at most 30% after consumption of a dose of 500mg/kg. Plasma ALT and AST activities are significantly increased after administrating of both the low (50mg/kg bw) and high dose (500mg/kg bw) groups. Animal study concluded that liver injury was possible through the biotransformation of PA to form a reactive quinone intermediate that may covalently bind to the nucleophilic residues of thiol groups or GSH leading to liver toxicity. Computational prediction of liver toxicity was supported by experimental test data in animal.

b. Benzofuran
Benzofuran was identified as a positive toxic chemical components in chasteberry. Two software programs using Dfw and MetaDrug were used. The National Toxicology Program(NTP) concluded that benzofuran is a clear carcinogen in male and female mice in liver and other tissues. In NTP studies with benzofuran, different doses of 62.5mg/kg, 125mg/kg, 250mg/kg, and 500mg/kg were administered in liquid form to male and female rats. 250mg/kg doses were determined to be the minimum level sufficient to induce lethality. For example, subjects given 250mg/kg and 500mg/kg doses group were observed as exhibitng hepatic necrosis in both male and female rats, whereas subjects in the 62.6mg/kg low dose group did not show any liver lesions. Subjects in the 125mg/kg doses group were observed with hepatic necrosis in only male rats. A two year bio-study by the NTP supports this conclusion by providing evidences of tubular cell adenocarcinomas of the kidney in female rats.

c. Compound 4-vinylphenol
The compound 4-vinylpheonol, contained in the chasteberry botanical, was identified as positive in MC4PC and MetaDrug. In animal study, doses of 50, 100, 200, and 400 mg of 4-vynylphenol/kg bw were administered. In all the dosages level, significant increases of serum sorbitol dehydrogenase(SDH) were observed. Effective biochemical factors of liver injury were correlated with hepatic histopathological evidence of centrilobular necrosis and hepatocyte swelling, and lethality was observed at doses of 200mg/kg bw and above.

d. Xanthohumol I(XI)
Xanthohumol I(XI), contained in the hops botanical, was predicted positive in the gall bladder and liver enzyme computational models of MC4PC and in the liver necrosis model of the MetaDrug software. In a 4-week safety study, female rats that were administered 500mg and 1000mg of XI/kg - bw/day demonstrated decreased liver weight (30-40%). The researcher( Hussong et.at., 2005) concluded that XI is weakly hepatotoxic.

e. Isoflavone backbone
There is a phenolic substructure to be found except for benzofuran. Computational screening for liver toxicity and analysis by LMA and leadscope enterprise showed that isoflavonoids are commonly present in phtoestrogens and herbals. Computational screening also identified that glycosides coupled with a flavone(or isoflavone), and the 4-hydroxyacetophenone in flavonoid backbones are important structural features in predicting positive botanical constituents. This outcome was verified by epidemiology and experimental studies that show isoflavonoids are anti-oxidative and have anti-carcinogenic properties. Although the casual relationship between isoflavonoid consumption and liver toxicity is unknown, several study results have suggested that possible liver effects are induced by long-term isoflavonoid intake. Although the casual relationship between isoflavonoid consumption and liver toxicity is unknown, several study results have suggested possible liver effects are induced by long-term isoflavonoid intake.

DISCUSSION
Computational software analysis and verification by animal tests proved that molecular level of chemical components in each four botanical products including red clover; chasteberry; black cohosh; and hops can affect liver toxicity.

The minimum level of intake to affect liver toxicity was also verified in animal tests. A chestberry has two positive components; Benzofuran and 4-Vinyl phenol. Benzofuran proved lethality at 250mg of daily dosage and 4-Vinyl phenol proved lethality at 200mg of daily dosage. Protocatechuic acid that is contained in Black cohosh affected liver enzymes at 0.1% intake by drinking water for 60days. Xanthohumol I that is contained in hops proved no effects on liver enzyme levels after 4 weeks in mice. However, 50uM in drinking water predicted positive. Isoflavone in Redclover proved that glycosides coupled with a flavone affect liver toxicity after a long

intake.

Theses results proved that four botanicals have association with liver toxicity. They also proves which molecular levels of chemical components affect liver toxicity and its results are verified in animal tests. However, the initial reason of selecting four botanicals is to prove liver toxicity affecting to Women aging 40 to 60 years. Test does not compare botanical intake group’s liver toxicity with the non-intake group who chose menopausal hormone therapy under same condition. Moreover, metabolic factors of female that are different from rats in experiments need to be reviewed.

CONCLUSION
In this paper, predictive liver toxicity data are confirmed by existing human exposure in case of having human study data and animal studies. The decision factor of toxicity does not include a safety factor for species extrapolation from animal to human.

This paper is worthwhile to use identified data of molecular structure of four selected botanical products. Moreover, even though botanical products themselves are different from the products that are used in this paper, unique chemical components can be used to experiment with mixture of other chemical components. From analyzing endpoints toxicity, future researchers can design a research with mixing other molecule level of components. Also, it can prove other combination levels of toxicity to be affected to different parts of the body.

RECOMMENDATIONS
This paper will help future researchers identify all the chemical components for each botanical product.
For the future development, metabolites and potential synergistic effects need to be considered. However, this research has already screened molecular level of chemical components. Future research can experiment with the mixture level for testing in animal tests or human tests. Metabolites can be also screened based on this paper. For identifying metabolites and chemometrics (chemoinformatics) of botanicals using GC/HPLC-MS-MS, other technologies are suggested to obtain more comprehensive chemical profiles.