Systemic Exposure to and Disposition of Catechols Derived...

1521-009X/43/5/679–690$25.00 http://dx.doi.org/10.1124/dmd.114.061473DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 43:679–690, May 2015Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics

Systemic Exposure to and Disposition of Catechols Derived fromSalvia miltiorrhiza Roots (Danshen) after Intravenous Dosing

DanHong Injection in Human Subjects, Rats, and Dogs

Meijuan Li, Fengqing Wang, Yühong Huang, Feifei Du, Chenchun Zhong, Olajide E. Olaleye,Weiwei Jia, Yanfen Li, Fang Xu, Jiajia Dong, Jian Li, Justin B. R. Lim, Buchang Zhao, Lifu Jia, Li Li,

and Chuan Li

Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China (M.L., F.W., F.D., C.Z., O.E.O., W.J., F.X.,J.D., J.L., J.B.R.L., L.L., C.L.); University of Chinese Academy of Sciences, Shanghai, China (M.L., L.L., C.L.); Second Affiliated

Hospital, Tianjin University of Traditional Chinese Medicines, Tianjin, China (Y.H., Y.L.); Buchang Pharmacy Group, Xi’an,Shaanxi Province, China (B.Z., L.J.); and Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences,

Beijing, China (C.L.)

Received October 1, 2014; accepted February 10, 2015

ABSTRACT

DanHong injection is a Danshen (Salvia miltiorrhiza roots)-basedinjectable solution for treatment of coronary artery disease andischemic stroke. Danshen catechols are believed to be responsible forthe injection’s therapeutic effects. This study aimed to characterizesystemic exposure to and elimination of Danshen catechols in humansubjects, rats, and dogs receiving intravenous DanHong injection. Atotal of 28 catechols were detected, with content levels of 0.002–7.066mM in the injection, and the major compounds included tanshinol,protocatechuic aldehyde, salvianolic acid B, rosmarinic acid, salvia-nolic acids A and D, and lithospermic acid with their daily doses ‡10mmol/subject. After dosing, tanshinol, salvianolic acid D, and lith-ospermic acid exhibited considerable exposure in human subjects andrats. However, only tanshinol had considerable exposure in dogs. The

considerable exposure to tanshinol was due to its having the highestdose, whereas that to salvianolic acid D and lithospermic acid was dueto their relatively long elimination half-lives in the human subjects andrats. Protocatechuic aldehyde and rosmarinic acid circulated in thebloodstream predominantly as metabolites; salvianolic acids A and Bexhibited low plasma levels with their human plasma metabolites littleor not detected. Tanshinol and salvianolic acid D were eliminatedmainly via renal excretion. Elimination of other catechols involvedhepatobiliary and/or renal excretion of their metabolites. Methylationwas found to be the primary metabolism for most Danshen catecholsand showed intercompound and interspecies differences in rate anddegree in vitro. The information gained here is relevant to pharmaco-logical and toxicological research on DanHong injection.

Introduction

Traditionally, herbal medicines are orally administered. However,there are 134 types of herbal injections that are approved by the ChinaFood and Drug Administration (China FDA) for clinical use. Theherbal injections are prepared from a single herb or from a mixture of

several herbs and contain multiple bioactive compounds. The Chinesepharmaceutical industry is required to manufacture herbal injectionproducts in compliance with good manufacturing practice. Unlikeorally administered herbal medicines, herbal injections are injecteddirectly into the bloodstream and potentially have immediate effects.Many approved herbal injections are extensively used in Chinesehospitals for treating cardiovascular or infectious diseases and foralleviating related symptoms; these herbal products appear to exhibitcomparable incidence of side effects with synthetic drug injections.However, active compounds responsible for the therapeutic effectshave remained poorly defined for most herbal injections. This hampersinvestigating the mechanisms of action, optimizing the herbaltherapies, and assessing the safety and associated potential herb-drug interactions. Bioactive herbal compounds that exhibit consider-able levels of body exposure after administration are most likely toaccount for the pharmacological effects of an herbal injection andform the basis of therapeutic efficacy. Accordingly, pharmacokinetic(PK) studies should be designed for herbal injections and implemented

This work was supported by the National Science Fund of China forDistinguished Young Scholars [Grant 30925044]; the National Science andTechnology Major Project of China “Key New Drug Creation and ManufacturingProgram” [Grants 2009ZX09304-002 and 2013ZX09201020001003]; and theNational Basic Research Program of China [Grant 2012CB518403].

This work was previously presented as a poster presentation at the followingworkshop: Li et al. (2014) In vivo chemical profiles and metabolic profiles of phenolicacids after intravenous administration of DanHong injection in different species. 5th

Asian Pacific Regional Meeting of ISSX; 2014 May 9–12; Tianjin, China.Justin B. R. Lim was a visiting student from Ernest Mario School of Pharmacy,

Rutgers, the State University of New Jersey.dx.doi.org/10.1124/dmd.114.061473.

ABBREVIATIONS: ALDH, aldehyde dehydrogenase; AUC0-‘, area under plasma concentration-time curve to infinity; COMT, catechol-O-methyltransferase; Cum.Ae-B,0–8h, cumulative amount excreted into bile collected 0–8 hours after dosing started; Cum.Ae-U,0–24h, cumulative amountexcreted into urine collected 0–24 hours after dosing started; fe-B, fraction of dose excreted into bile; fe-U, fraction of dose excreted into urine; PK,pharmacokinetic; SAM, S-adenosylmethionine; SULT, sulfotransferase; t1/2, elimination half-life; TOF-MS, time-of-flight mass spectrometer; UGT,uridine 59-diphospho-glucuronosyltransferase.

679

http://dmd.aspetjournals.org/content/suppl/2015/02/10/dmd.114.061473.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on June 14, 2018

dmd.aspetjournals.org

Dow

nloaded from

http://dx.doi.org/10.1124/dmd.114.061473

http://dx.doi.org/10.1124/dmd.114.061473

http://dmd.aspetjournals.org/content/suppl/2015/02/10/dmd.114.061473.DC1

http://dmd.aspetjournals.org/

in human subjects and laboratory animals to reveal body exposure tothe herbal compounds after dosing. The body exposure to herbalcompounds depends on the doses of the compounds from theadministered injection and their pharmacokinetics and disposition afterdosing. In the first step of PK studies of an herbal injection, threequestions should be answered: 1) what chemical compounds from theherbal injection are injected into the bloodstream and how much arethey dosed? 2) Which herbal compounds and herb-related metabolitesexhibit considerable systemic exposure after dosing and how long aretheir elimination half-lives? 3) How are the herbal compounds that areconsiderably dosed from the injection principally eliminated? The firststep of PK studies provides information on which compounds derivedfrom the herbal injection are worth further investigation with respectto pharmacokinetics, pharmacology, and toxicology. The follow-upPK studies should focus on these important herbal substances andshould be carried out to define their detailed pharmacokinetics andassociated interindividual variations, to characterize their principalroutes of elimination, to quantify the contribution of enzymes andtransporters to their pharmacokinetics and disposition, and to assessassociated potential herb-drug and herb-herb interactions. Here, theaforementioned “first-step” PK studies were designed for DanHonginjection and implemented in human subjects, rats, and dogs.DanHong injection is prepared from a 3:1 mixture of Salvia

miltiorrhiza roots (Danshen in Chinese) and Carthamus tinctoriusflowers (Honghua). It is an injectable solution available as a sterile,nonpyrogenic parenteral dosage form for intravenous injection. Thisherbal injection is approved by the China FDA for the treatment ofatherosclerotic coronary artery disease and acute ischemic stroke. Severalreports on meta-analyses of randomized trials show that DanHonginjection is effective in treating angina pectoris, congestive heart failure,and acute ischemic stroke (Peng et al., 2010, 2011; Yang and Zeng,2012). DanHong injection contains multiple water-soluble polyphenolsof Danshen-origin (Liu et al., 2013; Li et al., 2013; Xie et al., 2014). TheDanshen polyphenols contain one or more ortho-dihydroxy benzenerings, known as catechols, and have one or more carboxylic acidfunctional groups, except for protocatechuic aldehyde. They are caffeicacid derivatives, occurring as monomers (such as tanshinol and caffeicacid), dimers (such as rosmarinic acid and salvianolic acid D), trimers(such as salvianolic acid A and lithospermic acid), and tetramers (such assalvianolic acid B) (Jiang et al., 2005; Li et al., 2009). Cell- and isolated-tissue-based studies have shown that Danshen catechols exhibitedvasodilating, endothelial protective, cardioprotective, antithrombotic,antioxidant, and anti-inflammatory properties; tanshinol and otherDanshen catechols also decreased blood pressure in rats and attenuatedvenular thrombosis, cardiac hypertrophy, and methionine-inducedhyperhomocysteinemia in rats (Jiang et al., 2005; Han et al., 2008;Cao et al., 2009; Ho and Hong, 2011; Wang et al., 2013a). Despiteincreasing knowledge of the chemistry and pharmacology of DanHonginjection, the systemic exposure to and disposition of its bioactivecompounds after dosing remain to be elucidated. This report presents theDanshen-derived chemical profile of DanHong injection and thesystemic exposure to and elimination of Danshen catechols. It alsopresents comparative in vitro metabolic stability of Danshen catecholstoward various metabolic reactions and their in vitro methylationprofiles. The information gained here will be relevant to the optimizationof DanHong injection-based therapy.

Materials and Methods

DanHong Injection. DanHong injection was manufactured by HezeBuchang Pharmaceutical Co., Ltd. (Heze, Shandong Province, China) witha China FDA ratification no. of GuoYaoZhunZi-Z20026866. Each milliliter of

DanHong injection was prepared from 0.75 g of Salvia miltiorrhiza roots Danshenand 0.25 g of Carthamus tinctorius flowers (Honghua). To prepare the herbalinjection, Danshen and Honghua were mixed and extracted with water. Afterfiltration and concentration under reduced pressure, the concentrated extract wasprecipitated with ethanol and centrifuged. The resulting supernatant was treatedwith active carbon, evaporated under reduced pressure to remove the ethanol,adjusted to pH 6.5–7.5, ultrafiltered, and sterilized. The finished product ofDanHong injection was standardized to contain not less than 0.35 mg/mlconcentration of tanshinol and protocatechuic aldehyde combined. Samples of sixDanHong injection lots (lot numbers: 13011037, 13072016, 13082026, 13092035,13102030, and 13122024) were obtained for analysis, and all the products weremanufactured in 2013. DanHong injection from lot number 13011037 was used inthe current human and animal studies.

Chemicals and Reagents. Analytical reference standards of caffeic acid;lithospermic acid; protocatechuic acid; protocatechuic aldehyde; rosmarinicacid; salvianolic acids A, B, C, and D; tanshinol; chrysin, (2)-epicatechin;7-hydroxyflavone; isovanillin; isovanillin acid; and vanillic acid were purchasedfrom Tauto Biotech (Shanghai, China). The purity of the preceding compoundswas $98%. Pentobarbital was obtained from Shanghai Westang Biotechnology(Shanghai, China).

S-adenosylmethionine (SAM), uridine 59-diphospho-glucuronic acid, 39-phosphoadenosine-59-phosphosulfate, tris-hydroxymethyl-aminomethane, nicotin-amide adenine dinucleotide (NAD+), taurocholic acid, formic acid, and methanolwere obtained from Sigma-Aldrich (St. Louis, MO). Rat liver cytosol andmicrosomes and dog liver cytosol and microsomes were prepared in-house bydifferential centrifugation. Human liver cytosol and microsomes were obtainedfrom Corning Gentest (Woburn, MA). (2)-Epicatechin was used as the positivecontrol to assess the catechol-O-methyltransferase (COMT) activities of thehuman, rat, and dog liver cytosols (Zhu et al., 2000; Li et al., 2012), and the invitro t1/2 values were 0.46, 0.08, and 0.21 hour, respectively. 7-Hydroxyflavonewas used as the positive control to assess the sulfotransferase (SULT) activities ofthe human, rat, and dog liver cytosols (Li et al., 2012; Hu et al., 2013), and the invitro t1/2 values were 0.07, 0.29, and 0.10 hour, respectively. Chrysin was used asthe positive control to assess the UDP-glycosyltransferase (UGT) activities of thehuman, rat, and dog liver microsomes (Li et al., 2012; Hu et al., 2013), and the invitro t1/2 values were 0.08, 0.05, 0.05 hour, respectively. Isovanillin was used asthe positive control to assess the aldehyde dehydrogenase (ALDH) activities ofthe human, rat, and dog liver cytosols (Panoutsopoulos et al., 2004), and the invitro t1/2 values were 0.04, 0.05, and 0.05 hour, respectively.

Human Study. The protocol for human study was approved by the ethicscommittee of clinical investigation at Second Affiliated Hospital of TianjinUniversity of Traditional Chinese Medicine (Tianjin, China). The study wasregistered in Chinese Clinical Trials Registry (www.chictr.org) with a registrationnumber of ChiCTR-ONRC-13003571. Healthy volunteers gave written informedconsent to participate in the study. The human subjects were considered to be ingood health on the basis of medical history, physical examination, vital signsmeasurement, electrocardiogram, and clinical laboratory tests. They were requiredto be aged between 18 and 35 years, have a body mass index between 19 and 24kg/m2, be a nonsmoker, and not be allergic to products prepared from Danshenand/or Honghua. Synthetic drugs and herbal products were prohibited startingfrom 2 weeks before the study began until the end of the study period. In addition,alcoholic beverages were prohibited starting from 2 days before the study beganuntil the end of the study period.

A single-period, open-label human study was performed at a national clinicalresearch center of the Second Affiliated Hospital of Tianjin University ofTraditional Chinese Medicine (Tianjin, China). Six male subjects were asked to fastovernight before receiving a 120-minute intravenous infusion dose of DanHonginjection (40 ml/person, a label daily dose) the next morning. Before dosing, 40 mlof DanHong injection was diluted in 250 ml of 5% glucose injection(GuoYaoZhunZi-H12020021; China Otsuka Pharmaceutical Co., Ltd., Tianjin,China). Serial blood samples (;1 ml, collected in heparinized tubes) were takenfrom an antecubital vein catheter at 0, 0.25, 0.5, 1, 2, 2.17, 2.5, 3.5, 5, 11, and 24hours after the intravenous infusion was started. The blood samples wereheparinized and centrifuged to obtain plasma fractions. 0.1 ml aliquots of theresulting plasma samples of the same time point were pooled and stored at 270�Cuntil analysis. Urine samples were collected predose and at 0–2, 2–6, 6–10, and10–24 hours postdose and weighed. One milliliter aliquots of the urine sampleswithin the same time interval were pooled before storage at 270�C without use of

680 Li et al.

at ASPE



Dow

nloaded from

http://www.chictr.org


any preservative. Serum alanine aminotransferase, aspartate aminotransferase, totalprotein, albumin/globulin ratio, total bilirubin, and direct bilirubin were monitoredas liver function markers for the human subjects before and 24 hour after dosingDanHong injection; serum creatinine and blood urea nitrogen were also assessed tomonitor human renal function.

Laboratory Animals. All animal studies were conducted in compliance withthe Guidance for Ethical Treatment of Laboratory Animals (Ministry of Scienceand Technology of China, 2006). The experimental protocols were approved bythe Institutional Animal Care and Use Committee at Shanghai Institute of MateriaMedica (Shanghai, China). Male Sprague-Dawley rats (230–270 g) were obtainedfrom Sino-British SIPPR/BK Laboratory Animal (Shanghai, China). Rats weremaintained in a unidirectional airflow room at 20–24�C and relative humiditybetween 30% and 70% with a 12 hour light/dark cycle. Rats were given filteredtap water and commercial rat chow ad libitum and allowed to acclimate to thefacilities and environment for 3 days before use. Rats received in-house femoral-artery-cannulation for blood sampling or bile duct cannulation for bile sampling(Chen et al., 2013; Cheng et al., 2013). After surgery, rats were housedindividually and allowed to regain their preoperative body weights before furtheruse. During the bile collection period, a sodium taurocholate solution (pH 7.4)was infused into the duodena of bile-duct-cannulated rats at 1.5 ml/h. After use,rats were euthanized with CO2.

Male beagle dogs (7.8–9.0 kg) were obtained from the Teaching and ResearchFarm of Agricultural College, Shanghai Jiao Tong University (Shanghai, China)and were individually housed in stainless steel cages (1.7 � 1.0 � 0.9 m), whichcould be used to collect urine samples. The dogs were maintained at a controlledtemperature of 20–24�C, relative humidity between 40% and 70%, and a 12-hourcycle of light and dark. They were given commercial diets, except for anovernight fasting period before dosing, and filtered tap water ad libitum. The dogswere acclimated to the facilities for 2 weeks before use.

Rat Studies. Conscious and freely moving rats received a single intravenousbolus dose of DanHong injection at 4 ml/kg. The dose of the herbal injection forthe rats was derived from the clinical daily dose (40 ml/person) according to dosenormalization by body surface area (Reagan-Shaw et al., 2008). Serial bloodsamples (;150 ml; immediately before and 5, 15, and 30 minutes and 1, 2, 4, 6,8, 11, and 24 hours after dosing) were collected from three femoral arterycannulated rats into heparinized tubes. The blood samples were centrifuged toobtain the plasma fractions and the resulting plasma samples (0.05 ml) of thesame time point were pooled. Bile samples were collected from three bile ductcannulated rats before and 0–1, 1–2, 2–4, 4–6, and 6–8 hours after dosing. Threerats that did not undergo any surgical treatment were housed individually inmetabolic cages. Urine samples were collected before and 0–4, 4–8, and 8–24hours after dosing; samples (0.5 ml) of the same collection period from differentrats were pooled and stored at 270�C until analysis.

Dog Study. Dogs received a single intravenous bolus dose of DanHonginjection at 1.2 ml/kg. The dog dose was derived from the clinical daily dose ofthe herbal injection (40 ml/person) according to dose normalization by bodysurface area (Reagan-Shaw et al., 2008). Serial blood samples (;500 ml;immediately before and 5, 15, 30 minutes and 1, 2, 4, 6, 8, 11, and 24 hours afterdosing) were collected from three dogs into heparinized tubes. The blood sampleswere centrifuged to obtain the plasma fractions and the resulting plasma samples(0.1 ml) of the same time point were pooled. Urine samples were also collectedfrom the dogs before and 0–4, 4–8, and 8–24 hours after dosing. The urinesamples (0.5 ml) from the same collection period were pooled. The samples fromdogs were stored at 270�C until analysis.

In Vitro Metabolism Studies. Comparative metabolic stability of severalDanshen catechols toward COMT, SULT, UGT, and ALDH (ALDH forprotocatechuic aldehyde only) was assessed with human, rat, and dog livercytosols and microsomes. The test Danshen catechol compounds were tanshinol,protocatechuic aldehyde, protocatechuic acid, salvianolic acid B, rosmarinic acid,salvianolic acids A and D, and lithospermic acid. Before use, the COMT, SULT,and ALDH activities of the liver cytosols were assessed with the referencecompounds (2)-epicatechin, 7-hydroxyflavone, and isovanillin, respectively; theUGT activities of liver microsomes was assessed with chrysin. The liver cytosolwas fortified with an appropriate cofactor (SAM for methylation, 39-phosphoadenosine-59-phosphosulfate for sulfation, and NAD+ for oxidation),and the liver microsomes were fortified with uridine 59-diphospho-glucuronicacid for glucuronidation. The reaction was initiated by adding the test compound(10 mM) at 37�C and terminated at 0, 15 (for rat enzyme-mediated reactions

only), 30 (for dog enzyme-mediated reactions only), and 60 minutes (for humanenzyme-mediated reactions only) by adding a double volume of ice-coldmethanol. The time of incubation for tanshinol was double that for the otherDanshen catechol compounds. Negative control reactions were carried out byincubating mixtures that excluded the cytosol or the microsomes. The quenchedincubation was centrifuged at 1369 g for 10 minutes, and 5 ml of the supernatantwas analyzed by liquid chromatography/mass spectrometry. Metabolic stabilitywas calculated as percentage of unchanged compound remaining after incubation.When determining the metabolic stability of protocatechuic aldehyde towardCOMT and SULT present in human liver cytosol, isovanillin (100 mM) was usedto inhibit the catalytic activity of the coexisting ALDH.

In vitro methylation was further examined individually for the precedingDanshen catechol compounds by incubating with human, rat, and dog livercytosols fortified with SAM. The incubation broth contained 0.5 mg protein/mlliver cytosol, 50 mM Tris-HCl buffer (pH 7.4), 1 mM dithiothreitol, 2 mMMgCl2,and 1 mM SAM and was preincubated for 3 minutes at 37�C before initiation ofreaction by adding the test compound (10 mM). After incubation for 0, 5, 15, 30,60, 90, and 120 minutes at 37�C, the reactions were terminated by adding a doublevolume of ice-cold methanol. After centrifugation, the supernatant (5 ml) wasanalyzed to profile unchanged substrate and methylated metabolites. In thedetermination of in vitro t1/2, the concentrations of the test Danshen catechol in theincubation broth over time were calculated as percentage of compound remaining,using the initial concentration (the incubation time, 0 minutes) as 100%. The slopeof the linear regression from log percentage remaining versus incubation timerelationship, i.e., 2k, was used in the calculation of in vitro t1/2, according to thefollowing equation:

In vitro t1=2 ¼ 0:693=k ð1Þ

Analysis of Danshen Catechol Compounds and the Metabolites. AWaters Synapt G2 high definition time-of-flight mass spectrometer (TOF-MS;Manchester, UK) was interfaced via a Zspray/LockSpray ESI source witha Waters Acquity UPLC separation module (Milford, MA). The massspectrometer was operated in resolution mode giving a resolving power ofaround 20,000. The ESI source worked in the negative ion mode under thefollowing conditions: capillary voltage at 22.5 kV, source temperature of 120�C,desolvation gas at 800 l/hour and 400�C, sampling cone at 25 V, and extractioncone at 4.0 V. The mass spectrometer was externally calibrated over a range ofm/z50–1000 using a 5 mM sodium formate solution at 20 ml/min and mass shiftsduring acquisition were corrected using leucine enkephalin (m/z 554.2615) asa lockmass. MSE data acquisition (in centroid mode, m/z 50–1000) was achievedusing a trap collision energy of 4 V, a trap collision energy ramp of 15–30 Vsimultaneously, and a scan time of 0.3 seconds. Chromatographic separation ofDanshen catechol compounds in DanHong injection was achieved on a WatersAcquity UPLC HSS T3 1.8-mm column (100� 2.1 mm i.d.; Dublin, Ireland; keptat 45�C) using a mobile phase that consisted of solvent A (methanol/water, 1:99,v/v, containing 25 mM formic acid) and solvent B (methanol/water, 99:1, v/v,containing 25 mM formic acid). The mobile phase was delivered at 0.3 ml/min. Agradient program was used as follows: 0–2 minutes, at 2% solvent B; 2–32minutes, from 2 to 98% solvent B; 32–37 minutes, at 98% solvent B; and 37–42minutes, at 2% solvent B. For quantitative analysis of Danshen compounds,a modified gradient elution program with a mobile phase flow rate of 0.3 ml/minwas used as follows: 0–10 minutes, from 2 to 70% solvent B; 10–12 minutes,from 70 to 98% solvent B; and 12–15 minutes, at 2% solvent B. Samplepreparation for human, rat, and dog samples was performed using methanol asprecipitant with a volumetric precipitant-to-sample ratio of 3:1. After precipitationand centrifugation, the supernatants were reduced to dryness by centrifugalevaporation under reduced pressure. The residues were reconstituted in 50%methanol for LC/TOF-MSE-based analysis.

Profiling of Danshen compounds in the samples of DanHong injection wasconducted. The detected compounds were identified by comparing with thereference standards, with respect to accurate molecular mass, fragmentation, andchromatographic retention time. When the reference standard was not available,the compound identification was achieved by comparing the compound’s accuratemolecular mass, fragmentation, and elution order with the reported data ofDanshen compounds. To support the metabolite profiling, an Accelrys metabolitedatabase (version 2012.1; Accelrys, Inc., San Diego, CA) was used to get priorknowledge of likely metabolic pathways of Danshen compounds. Metabolite

Systemic Exposure to Catechols from Intravenous DanHong Injection 681

at ASPE



Dow

nloaded from


profiling was conducted with the human, rat, and dog samples according to themolecular mass gains or losses predicted for the possible metabolites comparedwith those of the parent compounds and according to the generation of fragmentions from the precursor deprotonated molecules by collision-induced dissociation.The MS fragmentation data were also used to assign the metabolite structures.Methylated metabolites were characterized by the comparative fragmentationprofiles of metabolite and parent compound. Glucuronides and sulfates werecharacterized by neutral losses of 176 and 80 Da, respectively. Quantification ofDanshen catechols and the metabolites in the biologic samples was performedusing the available reference standards for calibration or using the calibrationcurve of an analog (also called “brother compound”) that bore close structuralsimilarity to the analyte (Li et al., 2012). By using the reference standards, matrix-matched calibration curves (13.7–10,000 nM) were constructed using weightedlinear regression of the peak area against the corresponding nominal analyteconcentration (nM), which showed good linearity (r2 . 0.99). Assay validationwas carried out according to the United States Food and Drug Administrationguidance on bioanalytical validation (2013) to prove that the bioanalytical assayswere reliable for the intended applications.

PK Data Analysis. The area under plasma concentration-time curve up to thelast measured time point (AUC0-t) was calculated by the trapezoidal rule. TheAUC0-‘ was generated by extrapolating the AUC0-t to infinity using theelimination rate constant (k) and the last measured concentration. The eliminationhalf-life (t1/2) was calculated using the relationship 0.693/k. The cumulativeamounts excreted into urine (Cum.Ae-U) and bile (Cum.Ae-B) were calculated bynumeric integration of the amount excreted per collection interval. The fractionsof dose excreted into urine (fe-U) and bile (fe-B) were established by Cum.Ae-U/doseand Cum.Ae-B/dose, respectively.

Results

Danshen Catechols Present in DanHong Injection. Threeliterature references by Liu et al. (2007), Wang et al. (2007), and Liet al. (2009) give the most comprehensive information about chemicalconstituents present in Danshen. Among the Danshen constituents,hydrophilic catechols and lipophilic diterpene quinones are pharma-cologically relevant. However, only the Danshen catechols, but not thediterpene quinones, were detected in DanHong injection. This isprobably because of water extraction of the herb involved inmanufacturing the injection. As shown in Table 1 and Fig. 1, a totalof 28 Danshen catechol compounds were detected in six lots ofDanHong injection with tanshinol (6) present in the highest contentlevel (6.139–7.066 mM). Other major Danshen compounds presentwere protocatechuic aldehyde (1; 1.474–1.840 mM), salvianolic acidB (26; 1.014–1.245 mM), rosmarinic acid (10; 0.684–0.795 mM),salvianolic acids A (16; 0.563–0.886 mM) and D (13; 0.467–0.704mM), lithospermic acid (22; 0.235–0.322 mM), salvianolic acid Bisomer (25; 0.159–0.227 mM), and salvianolic acid E (27; 0.136–0.154 mM). These were major Danshen catechols, which have LogPvalues of20.28–2.69 (calculated using Pallas software, Pallas 3.7.1.1;CompuDrug International, Sedona, AZ). The remaining Danshencompounds, which were also detected in the herbal injection samples,were salvianolic acid B isomer (24); lithospermic acid isomers (18, 19,and 20); caffeic acid (3); protocatechuic acid (2); salviaflaside (17);salvianolic acids H/I (21), F (7), J (23), G (8), and C (15); salvianicacid C (11) and salvianic acid C isomer (12); isosalvianolic acid C(14); prolithospermic acid (9); isoferulic acid (5); ferulic acid (4); and4-methoxyl salvianolic acid B (28). These minor Danshen compoundsexhibited content levels of 0.002–0.076 mM with LogP values of0.17–3.02. Lot-to-lot variability of DanHong injection was alsoinvestigated with the major constituents ($0.136 mM). The herbalinjection showed consistency in the content levels of 6, 1, 26, 10, and27 from lot to lot. This was indicated by relative standard deviations of4.3–9.4%. Such relative standard deviations values for 16, 13, 22, and25 were greater, i.e., 12.0–16.3%.

DanHong injection (lot number, 13011037) was used in the currentstudy; the content levels of tanshinol (6), protocatechuic aldehyde (1),salvianolic acid B (26), rosmarinic acid (10), salvianolic acid A (16),salvianolic acid D (13), and lithospermic acid (22) were 7.066, 1.833,1.013, 0.769, 0.886, 0.704, and 0.247 mM, respectively. Clinical dailydose levels of the Danshen catechols from the herbal injection could bedivided into three classes: .100, 10–100, and ,10 mmol/subject (Fig.1C). The dose level of 6 was 282.8 mmol/subject, which was 52.7% ofthe total dose of Danshen catechols detected in the injection (Fig. 1D); 6was a Danshen catechol of class I. The dose levels of 1, 26, 10, 16, 13,and 22 were 10.0–73.2 mmol/subject, the sum of which accounted for40.7% of the total dose of Danshen catechols detected in the injection;these compounds were of class II. The remaining Danshen catecholswere of class III with dose levels of 0.1–7.2 mmol/subject, the sum ofwhich accounted for only 6.6% of the total dose of Danshen catecholsdetected in the injection.Unchanged Danshen Catechols Detected in Plasma and Urine

Samples of Human Subjects Receiving an Intravenous InfusionDose of DanHong Injection. In the current study, pooled humanplasma samples and pooled human urine samples were used for thedetection of unchanged Danshen catechols and related metabolites. Thiswas because the detection of Danshen compounds depended on factors,such as assay sensitivity and interindividual variations in the eliminationmediated by drug metabolizing enzymes and/or drug transporters. Thesamples were pooled to minimize the possibility of failed detection ofimportant herbal compounds and their metabolites due to the potentialinfluence of interindividual variation in their elimination. A total of 11Danshen catechols were detected in human plasma samples during andafter intravenous infusion of DanHong injection; they were not detectedin the control plasma samples before dosing. Tanshinol (6) was detectedin considerable amount in the plasma samples after dosing started.Although salvianolic acid D (13) and lithospermic acid (22) wereDanshen catechols of class II with content levels being only 10.0 and3.5% of that of 6, respectively, their levels of systemic exposure werecomparable with that of 6 (Fig. 2A). The other Danshen catecholsdetected in plasma, including salvianolic acids B (26) and A (16),rosmarinic acid (10), and protocatechuic acid (2), were at low AUC0-‘

levels (Fig. 2A). The observed high levels of systemic exposure to 13and 22 were probably associated with their longer t1/2 values (2.5 and5.7 hours, respectively) compared with those of 6 (0.6 hour) and theother Danshen catechols (around 0.2 hour). Only trace amounts ofprotocatechuic aldehyde (1), lithospermic acid isomers (19 and 20), andsalvianolic acid C (15) were detected in plasma. Plasma concentrations of6, 13, and 22 over time during and after an intravenous infusion dose ofDanHong injection in human subjects are shown in Supplemental Fig. 1.A considerable amount of tanshinol (6) from intravenously dosed

DanHong injection was recovered in human urine, with a fe-U of 63.3%.Although the fe-U value of salvianolic acid D (13; 62.7%) was almost thesame as that of 6, its Cum.Ae-U,0–24h was only 8.6% of that of 6 (Fig.2B). The difference in urinary Cum.Ae-U,0–24h between the two catecholswas associated mainly with the difference in their content levels inDanHong injection; the content level of 13 was 10.0% of that of 6 (Fig.1B). Only 3.8% of lithospermic acid (22) from intravenously dosedDanHong injection was excreted unchanged into the urine, and its Cum.Ae-U,0–24h was quite low (Fig. 2B). Other plasma Danshen catechols[including salvianolic acid B (26) and rosmarinic acid (10)] were alsodetected in urine, with fe-U values of 0.2 and 18.2%, respectively;salvianolic acid A (16) was not detected in urine.Metabolites of Danshen Catechols Detected in Plasma and Urine

Samples of Human Subjects Receiving the Intravenous InfusionDose of DanHong Injection. Metabolites detected in the humansamples after dosing were those of the major Danshen catechols. As

682 Li et al.

at ASPE



Dow

nloaded from

http://dmd.aspetjournals.org/lookup/suppl/doi:10.1124/dmd.114.061473/-/DC1


shown in Fig. 3A, the major Danshen catechols could be divided intothree categories according to their circulating patterns: 1) tanshinol (6),salvianolic acid D (13), and lithospermic acid (22): circulating mainly asparent compound with minor or major metabolites; 2) protocatechuicaldehyde (1) and rosmarinic acid (10): circulating in the bloodstreampredominantly as metabolites with a minor parent compound; and 3)

salvianolic acids B (26) and A (16): only minor parent compoundcirculating in the bloodstream with metabolites very little or notdetected. The major metabolites of Danshen catechols were themethylated metabolites and their further sulfated and glucuronizedproducts. Sulfates and glucuronides of Danshen catechols were alsodetected in plasma after dosing. Table 2 shows LC/TOF-MSE-based

TABLE 1

Detection of Danshen catechol compounds in DanHong injection

ID CompoundLC/TOF-MSE Data

Molecular Mass Molecular FormulatR [M2H]2 (m/z) Fragmentation profilea (m/z)

min Da

1 Protocatechuic aldehyde 8.38 137.0241 109.0311, 108.0233a 138.0317 C7H6O3

2 Protocatechuic acid 7.11 153.0191 109.0295,a 108.0221 154.0266 C7H6O4

3 Caffeic acid 11.30 179.0342 135.0451,a 134.0370 180.0423 C9H8O4

4 Ferulic acid 14.12 193.0505 178.0258, 134.0370,a 133.0305 194.0579 C10H10O4

5 Isoferulic acid 14.75 193.0506 178.0257, 134.0373,a 133.0292 194.0579 C10H10O4

6 Tanshinol 6.63 197.0455 179.0346,a 135.0454, 123.0453 198.0528 C9H10O5

7 Salvianolic acid F 11.01 313.0718 295.0621, 269.0818,a 203.0376,159.0446 314.0790 C17H14O6

8 Salvianolic acid G 15.92 339.0502 311.0565, 295.0614,a 267.0659 340.0583 C18H12O7

9 Prolithospermic acid 11.67 357.0606 313.0706, 203.0349,a 159.0458 358.0689 C18H14O8

10 Rosmarinic acid 16.85 359.0765 197.0453, 179.0345, 161.0242a 360.0845 C18H16O8

11 Salvianic acid C 12.07 377.0869 359.0774, 197.0459,a 179.0352, 161.0237 378.0951 C18H18O9

12 Salvianic acid C isomer 12.52 377.0862 359.0772, 197.0448,a 179.0355, 161.0244 378.0951 C18H18O9

13 Salvianolic acid D 16.32 417.0827 373.0917, 197.0447, 179.0345, 175.0393a 418.0900 C20H18O10

14 Isosalvianolic acid C 19.50 491.0975 311.0563,a 293.0455, 249.0556 492.1056 C26H20O10

15 Salvianolic acid C 21.27 491.0978 311.0559, 293.0455,a 267.0663 492.1056 C26H20O10

16 Salvianolic acid A 18.51 493.1131 313.0716, 295.0611,a 185.0246 494.1213 C26H22O10

17 Salviaflaside 15.79 521.1292 359.0773,a 197.0453, 161.0241 522.1373 C24H26O13

18 Lithospermic acid isomer 13.48 537.1038 493.1134, 313.0721, 295.0607a 538.1111 C27H22O12

19 Lithospermic acid isomer 13.80 537.1030 493.1133, 313.0726, 295.0603,a 185.0249 538.1111 C27H22O12

20 Lithospermic acid isomer 14.32 537.1039 493.1134, 313.0717, 295.0604,a 185.0244 538.1111 C27H22O12

21 Salvianolic acid H/I 15.38 537.1031 339.0502,a 295.0617, 185.0243 538.1111 C27H22O12

22 Lithospermic acid 16.78 537.1036 493.1141, 295.0607,a 185.0241 538.1111 C27H22O12

23 Salvianolic acid J 17.97 537.1037 493.1142, 295.0609,a 185.0244 538.1111 C27H22O12

24 Salvianolic acid B isomer 16.03 717.1461 519.0933,a 339.0511, 321.0408 718.1534 C36H30O16

25 Salvianolic acid B isomer 16.36 717.1448 519.0932,a 339.0502, 321.0407 718.1534 C36H30O16

26 Salvianolic acid B 17.41 717.1455 519.0935,a 339.0504, 321.0399 718.1534 C36H30O16

27 Salvianolic acid E 18.18 717.1454 519.0938,a 339.0503, 321.0403 718.1534 C36H30O16

28 4-Methoxyl salvianolic acid B 18.38 731.1599 533.1101,a 353.0676, 335.0564 732.1690 C37H32O16

tR, chromatographic retention time.aProduct ion of base peak.

Fig. 1. LC/TOF-MSE-based detection and measurement ofDanshen catechols present in DanHong injection. (A) Chro-matogram of DanHong injection; (B) average content levels ofDanshen catechols in different lots of DanHong injection inmillimoles; (C) clinical daily doses of Danshen catechols fromintravenous DanHong injection (lot number, 13011037) inmmol/subject; (D) percentages of individual catechols in totalcontent level of Danshen catechols in DanHong injection.Danshen catechols (shown as compound ID) in (B) and (C) areranked according to their average content levels in different lotsof the herbal injection; their names and detection informationare shown in Table 1.


at ASPE



Dow

nloaded from


detection of the individual metabolites of Danshen catechols in humanplasma and/or urine during and after the intravenous infusion dose ofDanHong injection, as well as the metabolite ID notation system.Figures 3, A and B, exhibit plasma AUC0-‘ and Cum.Ae-U,0–24h ofmetabolites in comparison with the respective data of their parentDanshen catechols. Figure 4 depicts major metabolic pathwaysproposed for Danshen catechols from intravenously dosed DanHonginjection in humans. The major metabolite of 6 was M6M; thismetabolite was eliminated mainly via renal excretion. M13M wasa detected metabolite of 13 and its Cum.Ae-U,0–24h was 6.6% of theamount of 13 dosed from DanHong injection. The major metabolite of22 was M222M, which was poorly eliminated via renal excretion.Protocatechuic aldehyde has been reported to convert into protocatechuicacid by ALDH (Panoutsopoulos and Beedham, 2005; Xu et al., 2007).Consistent with this, the total Cum.Ae-U,0–24h of protocatechuic acid (2)and its metabolites (including M2M, M2M-G-1, M2M-G-2, M2S-1, andM2S-2) after dosing exceeded the amount of 2 dosed from DanHonginjection by 28-fold. This suggested that 1 was eliminated mainly byoxidation into 2, which was further methylated, glucuronized, andsulfated before renal excretion of 2 and the metabolites. The major

metabolites of 10 included M10M-2, M102M, M102M-S-1, M102M-S-2,M102M-G-1, and M102M-G-2; these metabolites were eliminated con-siderably via renal excretion with a total Cum.Ae-U,0–24h being 44.3% ofthe amount of 10 dosed from DanHong injection. Like in the plasmasamples, no metabolite of 26 was detected in the associated urinesamples. Unchanged 16 from DanHong injection was not detected inurine; only four minor urinary metabolites (M16M-G-1, M16M-G-2,M16M-G-3, M16G-2) were detected for 16. The total Cum.Ae-U,0–24h ofthese metabolites was only 1.0% of the amount of 16 dosed fromDanHong injection. These data suggested that 26, 16, and theirmetabolites in humans were not eliminated mainly via renal excretion.Danshen Catechols and Metabolites Detected in Plasma, Urine,

and Bile Samples of Rats and Dogs after Dosing DanHongInjection. For rats and dogs, samples of the same kind and from thesame collection time points/intervals were pooled to minimize thepossibility of failed detection of important herbal compounds and theirmetabolites; this sample pooling treatment was similar to that forhuman subjects. Like for human subjects, tanshinol (6), salvianolicacid D (13), and lithospermic acid (22) were the most abundantDanshen compounds detected in rat plasma samples after dosingDanHong injection (Fig. 2C). In rats, the t1/2 values of 13 and 22 (1.8and 1.9 hours, respectively) were longer than that of 6 (0.3 hour).Protocatechuic aldehyde (1), rosmarinic acid (10), salvianolic acids B(26) and A (16) were also detected in the rat plasma samples but at lowAUC0-‘ levels (Fig. 2C). Their t1/2 values ranged from 0.2 to 0.8 hour.In rats, both tanshinol (6) and salvianolic acid D (13) were

eliminated mainly via renal excretion (fe-U, 54.3 and 53.6%,respectively), rather than hepatobiliary excretion (fe-B, 0.5 and 1.0%,respectively), after dosing DanHong injection (Fig. 2, D and E). Theirmajor metabolites M6M, M6S, and M13M exhibited lower levels ofsystemic exposure than their respective parent compounds (Fig. 3C).These metabolites were also eliminated mainly via renal excretion(Fig. 3, D and E); the total Cum.Ae-U,0–24h of M6M and M6S and Cum.Ae-U,0–24h of M13M were 29.1 and 20.0% of the amounts of 6 and 13,respectively, dosed from DanHong injection. Lithospermic acid (22)was not eliminated mainly via renal excretion (fe-U, 2.7%) orhepatobiliary excretion (fe-B, 1.2%) (Fig. 2, D and E). Its metabolitesM222M and M22M were substantially detected in rat plasma and bileafter dosing (Fig. 3, C and E). The hepatobiliary excretion of thesemetabolites were significantly greater than their renal excretion (Fig.3, D and E); their total Cum.Ae-B,0–8h and total Cum.Ae-U,0–24h were76.9 and 4.7%, respectively, of the amount of 22 dosed fromDanHong injection. Protocatechuic aldehyde (1) was eliminatedmainly via oxidation into protocatechuic acid (2) followed by furthermethylation and sulfation; sulfation also considerably occurred in ratsfor 1 (Fig. 3, D and E). The urinary recovery of 2 and the follow-upmetabolites (M2M-S, M2S-1, and M2S-2) in rats exceeded the amountof 2 dosed from DanHong injection by 109-fold, which was markedlygreater than that observed in human subjects (28-fold). This was, atleast in part, because the rat diet contained a considerable amountof p-hydroxybenzoic acid ethylester. It has been reported thatp-hydroxybenzoic acid esters can be hydrolyzed and hydroxylated intoprotocatechuic acid (Wang et al., 2013b). Rosmarinic acid (10) waseliminated more extensively via renal excretion (fe-U, 25.2%) thanhepatobiliary excretion (fe-B, 0.12%) (Fig. 2, D and E). Unlike in thehuman subjects, 10 exhibited a higher level of systemic exposure thanits metabolites (M102M, M102M-S-1, M102M-S-2, M102M-G-1, andM102M-G-2) in rats (Fig. 3C). These metabolites were eliminated viaboth renal and hepatobiliary excretion (Fig. 3, D and E); their totalCum.Ae-U,0–24h and total Cum.Ae-B,0–8h were 29.0 and 49.4%,respectively, of the amount of 10 dosed from DanHong injection.Salvianolic acid B (26) was eliminated more extensively via

Fig. 2. Systemic exposure to and excretion of Danshen catechols in human subjects(A and B), rats (C–E), and dogs (F and G) receiving intravenous DanHong injection(lot number, 13011037). Danshen catechols (shown as compound ID) in (A–G) areranked according to their average content levels (Fig. 1B); their names are shown inTable 1. Red, black, and green bars are used to show the plasma, urinary, and biliarylevels of the Danshen catechols after dosing, respectively. The symbol “�” (in red)denotes the Danshen compound that was not detected. The human subjects receiveda 120-minute intravenous infusion dose of the herbal injection at 40 ml/subject. Therats and the dogs received an intravenous bolus dose of the herbal injection at 4.0and 1.2 ml/kg, respectively. Because of the sample pooling, no standard deviation isshown.

684 Li et al.

at ASPE



Dow

nloaded from


hepatobiliary excretion (fe-B, 8.1%) than renal excretion (fe-U, 1.0%)(Fig. 2, D and E). Its major metabolites included M26M, M262M-1,M262M-2, M263M-1, and M263M-2; neither sulfate nor glucuronide of

26 was detected. These metabolites were eliminated mainly viahepatobiliary excretion rather than renal excretion (Fig. 3, D and E);their total Cum.Ae-B,0–8h and total Cum.Ae-U,0–24h was 90.1 and 0.6%,

Fig. 3. Systemic exposure to and excretion of metabolites of major Danshen catechols [tanshinol (6), protocatechuic aldehyde (1), rosmarinic acid (10), salvianolic acid A(16), salvianolic acid D (13), lithospermic acid (22), and salvianolic acid B (26)] in human subjects (A and B), rats (C–E), and dogs (F and G) receiving intravenousDanHong injection (lot number, 13011037). The parent Danshen catechols are ranked according to their average content levels (Fig. 1B), and their respective metabolites areshown on the right (in a left-to-right ranking order of methylated metabolites, sulfates and glucuronides of methylated metabolites, sulfates and glucuronides of parentcompounds, and oxidized metabolites). Only metabolites (shown as metabolite ID) exhibiting plasma AUC0-‘ values and/or excretory amounts, which were 10% of those oftheir respective parent compounds at least in one species, are shown in this figure. The metabolite ID is used to indicate the compound as a metabolite, its parent compound,the type of metabolism, and metabolite isomer. For instance, “M10” in M102M-S-1 denotes “metabolite of Danshen catechol 10 (rosmarinic acid)”; the subscript number andletter “2M” denotes “dimethylation”; the subscript letter “S” denotes “sulfation”; and the last subscript number “1” denotes “the first eluted metabolite isomer.” The subscriptletter “G” for other metabolite IDs denotes “glucuronidation.” Red/pink, black/gray, and green/light green bars represent the plasma, urinary, and biliary levels of the parentDanshen catechols/their metabolites, respectively. The symbol “�” (in red) denotes the Danshen compounds that was not detected. The names of metabolites and associateddetection information are shown in Table 2 (for human metabolites) and Supplemental Tables 1 (for rat metabolites) and 2 (for dog metabolites). The human subjects receiveda 120-minute intravenous infusion dose of the herbal injection at 40 ml/subject. The rats and the dogs received an intravenous bolus dose of the herbal injection at 4 and 1.2ml/kg, respectively. Because of the sample pooling, no standard deviation is shown.


at ASPE



Dow

nloaded from




respectively, of the amount of 26 dosed from DanHong injection.Salvianolic acid A (16) was eliminated more extensively viahepatobiliary excretion (fe-B, 9.3%) than renal excretion (fe-U, 0.5%)(Fig. 2, D and E). Unlike in the human subjects, the metabolitesM16M-G-1, M162M-G-1, M162M-G-3, and M16G-1 appeared to havehigher levels of systemic exposure than unchanged 16 in rats (Fig.3C). In addition to these plasma metabolites, M16M-G-2, M16M-G-3,and M162M-G-2 were considerably detected in bile. Like the parentcompound, the detected metabolites were eliminated mainly viahepatobiliary excretion (Fig. 3, D and E); their total Cum.Ae-B,0–8h was71.8% of the amount of 16 dosed from DanHong injection.Supplemental Table 1 shows LC/TOF-MSE-based detection of theindividual metabolites of Danshen catechols in rat plasma, urine, and/or bile after dosing DanHong injection.In dogs, only plasma tanshinol (6) was considerably detected after

dosing DanHong injection (Fig. 2F). Unlike in the human subjects andin the rats, plasma salvianolic acid D (13) and lithospermic acid (22)were detected at low levels in dogs after dosing. Their t1/2 values were0.2 and 0.5 hour, respectively, which were shorter than that of 6 (1.0hour). Salvianolic acids B (26) and A (16) and rosmarinic acid (10)were also detected in plasma but at low levels of systemic exposure,with t1/2 of 0.2–0.3 hour (Fig. 2F). Only trace amounts ofprotocatechuic aldehyde (1) was detected in canine plasma. The fe-Uvalues of 6, 13, and 22 were 34.5, 43.6, and 5.8%, respectively (Fig. 2G).

M6M and M6S appeared to be substantially formed in dogs (Fig. 3G);their Cum.Ae-U,0–24h were 14.1 and 48.4%, respectively, of the amount of6 dosed from DanHong injection. However, their plasma AUC0-‘ valueswere lower than that of the parent compound 6 (Fig. 3F). The eliminationof 1 mainly involved oxidation into protocatechuic acid (2) followed byfurther methylation and sulfation (Fig. 3, F and G). The urinary recoveryof 2 and the follow-up metabolites (M2M and M2M-S) in dogs exceededthe amount of 2 dosed from DanHong injection by 12 fold. Multiplemetabolites of 10 were detected in plasma but at lower levels than theparent compound (Fig. 3F); they included M10M-1, M10M-2, M102M,M102M-S-1, andM102M-S-2. The Cum.Ae-U,0–24h of unchanged 10 and thetotal Cum.Ae-U,0–24h of its metabolites were 5.1 and 21.9%, respectively,of the amount of 10 dosed from DanHong injection. The metabolites of13, 22, 16, and 26 were detected at low levels in canine plasma and urineafter dosing (Fig. 3, F and G). Supplemental Table 2 shows LC/TOF-MSE-based detection of the individual metabolites of Danshen catecholsin canine plasma and/or urine after dosing DanHong injection.Comparative Metabolism of Danshen Catechols Mediated by

Hepatic COMT, SULT, UGT, and ALDH. In vitro methylation,sulfation, and glucuronidation of the Danshen catechols (tanshinol,protocatechuic aldehyde, protocatechuic acid, salvianolic acid B,rosmarinic acid, salvianolic acid A, salvianolic acid D, and lithospermicacid) were compared, with respect to percentage of compound remainingafter incubation with human, rat, and dog hepatic enzymes for a certain

TABLE 2

Detection of major metabolites of tanshinol (6), protocatechuic aldehyde (1), rosmarinic acid (10), salvianolic acid A (16), salvianolic acid D (13), and lithospermic acid (22)in human plasma and/or urine samples collected during and after an intravenous infusion dose of DanHong injection

IDa Parent compound/metaboliteLC/TOF-MSE Data

Molecular Mass Molecular FormulatR [M2H]2 (m/z) Fragmentation profileb (m/z)

min Da

Metabolites of tanshinol (6)6 Tanshinol 6.61 197.0455 179.0346,b 135.0454, 123.0453 198.0528 C9H10O5

M6M Methyltanshinol 9.90 211.0611 193.0507, 150.0323,b 134.0371 212.0684 C10H12O5

Metabolites of protocatechuic aldehyde (1)1 Protocatechuic aldehyde 8.36 137.0240 109.0311, 108.0233b 138.0317 C7H6O3

2 Protocatechuic acid 7.07 153.0194 109.0295b 154.0266 C7H6O4

M2M Methylprotocatechuic acid 10.8 167.0348 123.0451b 168.0422 C8H8O4

M2M-G-1 Methylprotocatechuic acid glucuronide 8.04 343.0660 167.0345,b 123.0443 344.0743 C14H16O10

M2M-G-2 Methylprotocatechuic acid glucuronide 8.96 343.0671 167.0351,b 123.0441 344.0743 C14H16O10

M2S-1 Protocatechuic acid sulfate 10.3 232.9767 153.0193,b 109.0291 233.9838 C7H6SO7

M2S-2 Protocatechuic acid sulfate 10.7 232.9759 153.0195,b 109.0295 233.9838 C7H6SO7

Metabolites of rosmarinic acid (10)10 Rosmarinic acid 16.70 359.0775 197.0453, 179.0345, 161.0242b 360.0845 C18H16O8

M10M-1 Methylrosmarinic acid 16.26 373.0918 197.0446, 175.0392,b 179.0346 374.1001 C19H18O8

M10M-2 Methylrosmarinic acid 18.18 373.0919 197.0452,175.0389, 135.0447b 374.1001 C19H18O8

M102M Dimethylrosmarinic acid 19.47 387.1078 211.0600,b 193.0492, 175.0388 388.1158 C20H20O8

M102M-S-1 Dimethylrosmarinic acid sulfate 21.33 467.0649 387.1076, 211.0605,b 175.0392 468.0726 C20H20SO11

M102M-S-2 Dimethylrosmarinic acid sulfate 21.44 467.0650 387.1085, 211.0601,b 175.0388 468.0726 C20H20SO11

M102M-G-1 Dimethylrosmarinic acid glucuronide 17.17 563.1411 387.1085, 211.0606,b 193.0484 564.1479 C26H28O14

M102M-G-2 Dimethylrosmarinic acid glucuronide 17.59 563.1405 387.1093, 211.0607, 193.0501b 564.1479 C26H28O14

Metabolites of salvianolic acid A (16)16 Salvianolic acid A 18.41 493.1127 313.0716, 295.0611,b 185.0246 494.1208 C26H22O10

M16M-G-1 Methylsalvianolic acid A glucuronide 18.62 683.1606 507.1295,335.0576,295.0599b 684.1690 C33H32O16



M162M-G-2 Dimethylsalvianolic acid A glucuronide 19.76 697.1771 521.1448, 327.0865, 309.0767b 698.1847 C34H34O16

Metabolites of salvianolic acid D (13)13 Salvianolic acid D 16.30 417.0833 373.0917, 197.0447, 175.0393b 418.0900 C20H18O10

M13M Methylsalvianolic acid D 17.79 431.0981 387.1073, 211.0608, 175.0392b 432.1056 C21H20O10

Metabolites of lithospermic acid (22)22 Lithospermic acid 16.75 537.1037 493.1141, 313.0725, 295.0625b 538.1111 C27H22O12

M22M Methyllithospermic acid 18.00 551.1198 507.1316, 327.0869, 309.0774b 552.1268 C28H24O12

M222M Dimethyllithospermic acid 19.22 565.1342 521.1436, 327.0879, 309.0753b 566.1424 C29H26O12

tR, chromatographic retention time.aThe metabolite ID contains a capital letter M and a number corresponding to compound ID, for instance, M6 in M6M-S-1 denotes “metabolite of compound 6 (tanshinol)”; in the subscript, a letter

denotes the reaction undergone by the compound (M denotes methylation; S, sulfation; G, glucuronidation) while a number such as “1” denotes “the first eluted isomer of the metabolite”.bProduct ion of base peak.

686 Li et al.

at ASPE



Dow

nloaded from




length of time. Among the human hepatic enzyme-mediated metabolicreactions, the rates of methylation of the test compounds were greaterthan those of associated sulfation and glucuronidation, except forprotocatechuic aldehyde and salvianolic acid A (Fig. 5A). Sulfation andglucuronidation of salvianolic acid D, lithospermic acid, and salvianolicacid B appeared to be negligible, because the sulfates and glucuronideswere not detected after incubation. Glucuronidation and oxidation ofprotocatechuic aldehyde were significantly faster than its methylation.Glucuronidation of salvianolic acid A was significantly faster than itsmethylation. Similar scenarios were observed for the rat and dog hepaticenzyme-mediated metabolic reactions, except for rat COMT-mediatedmethylation of protocatechuic aldehyde comparably as fast as theALDH-mediated oxidation (Fig. 5, B and C). The in vitro catalyticactivities of the rat and dog enzymes appeared to be higher than those ofthe human enzymes. Accordingly, the incubation times used for the ratand dog enzymes to mediate metabolism reactions were shorter thanthose of the human enzymes, i.e., 15 minutes (30 minutes for tanshinolonly), 30 minutes (60 minutes for tanshinol only), 60 minutes (120minutes for tanshinol only), respectively.The rates of human hepatic COMT-mediated methylation of the

different test Danshen catechol compounds were considerablydifferent (Fig. 6A). This is indicated by in vitro t1/2 values, whichwere ranked as follows: salvianolic acid D (0.31 hour) � lithospermicacid (0.41 hour) � protocatechuic acid (0.44 hour) , salvianolic acidB (0.73 hour) , rosmarinic acid (1.23 hours) , salvianolic acid A(3.25 hours) � protocatechuic aldehyde (3.71 hours) � tanshinol (3.73hours). The results of in vitro metabolite profiling of the incubationsindicated that the methylation degree was catechol moiety numberdependent and incubation time dependent. Tanshinol, protocatechuicaldehyde, and protocatechuic acid contain only one catechol moiety;only monomethylated metabolites (M6M, M1M, and M2M) were

detected for these Danshen catechol compounds. Rosmarinic acid andlithospermic acid contain two unsubstituted catechol moieties. Bothmonomethylated (M10M-1, M10M-2, and M22M) and dimethylatedmetabolites (M102M and M222M) of rosmarinic acid and lithospermicacid were found to be predominant, and the formation of dimethylatedmetabolites was significantly delayed compared with that of mono-methylated metabolites. Salvianolic acid D contains one unsubstitutedcatechol moiety and one substituted catechol moiety; it was convertedpredominantly into a monomethylated metabolite (M13M). Its dimethylatedmetabolite was detected at very low levels. Salvianolic acid B containsthree unsubstituted catechol moieties and salvianolic acid A containstwo unsubstituted catechol moieties and one substituted catecholmoiety. Human COMT-mediated methylation of these catecholcompounds resulted in the formation of monomethylated (M16M-1,M16M-2, andM26M) and dimethylated metabolites (M162M-1,M162M-2,M262M-1, and M262M-2), but only trace amounts of the trimethylatedmetabolites were detected. The rates of methylation mediated by the ratand the dog COMTs were significantly faster than those by the humanenzyme (Fig. 6, B and C). Because of this faster methylation by the ratCOMT, trimethylated metabolites (M163M-1, M163M-2, M263M-1, andM263M-2) were significantly detected for salvianolic acids A and B (Fig.6B). For the dog COMT, these trimethylated metabolites were alsoconsiderably detected only after 120-minute incubation (Fig. 6C). Forthe rat hepatic COMT-mediated methylation, the in vitro t1/2 valueswere ranked as follows: lithospermic acid (0.04 hour) = salvianolic acidB (0.04 hour) = rosmarinic acid (0.04 hour) � salvianolic acid D (0.05hour) , salvianolic acid A (0.09 hour) � protocatechuic acid (0.12hour) , protocatechuic aldehyde (0.24 hour) , tanshinol (0.94 hour).For the dog hepatic COMT-mediated methylation, the in vitro t1/2 valueswere ranked as follows: lithospermic acid (0.10 hour) � salvianolic acidD (0.13 hour), protocatechuic acid (0.25 hour)� rosmarinic acid (0.28

Fig. 4. Proposed major metabolic pathways of tanshinol (6), salvianolic acid D (13), lithospermic acid (22), protocatechuic aldehyde (1), and rosmarinic acid (10) in humansubjects receiving intravenous DanHong injection. The metabolites of salvianolic acids B (26) and A (16) exhibited little or no detection in human plasma and urine afterdosing. The names and detection information of metabolites are shown in Table 2; their systemic exposure levels and excretory amounts are depicted in Fig. 3, A and B.Sulfation of vanillic acid (M2M) and tanshinol (6) mainly took place in rats and dogs after dosing DanHong injection. The chemical structures ofM6M andM6S were inferredto be 3-O-methyltanshinol and tanshinol-3-O-sulfate, respectively, according to the data by Tian et al. (2015). The chemical structures of M22M and M222M were inferred tobe 39-O-methyllithospermic acid and 39,399-O-dimethyllithospermic acid, respectively, according to the data by Wang et al. (2008).


at ASPE



Dow

nloaded from


hour) � salvianolic acid B (0.30 hour) � salvianolic acid A (0.39 hour), protocatechuic aldehyde (1.36 hours) , tanshinol (1.94 hours).

Discussion

Danshen catechols are pharmacologically important for DanHonginjection, because they have been reported to have multiple antianginal

properties. Because systemic exposure to compounds after dosing anherbal injection depended on their dose levels and elimination kinetics,defining the correlations between exposure and influencing factors areimportant in PK research on the injection. The analysis of DanHonginjection revealed a total of 28 Danshen catechols present. Because oftheir different content levels, these Danshen catechols could be dividedinto three classes according to their daily dose levels from the

Fig. 5. Comparative in vitro metabolic stability of Danshen catechols (including tanshinol, protocatechuic aldehyde, protocatechuic acid, salvianolic acid B, rosmarinic acid,salvianolic acid A, salvianolic acid D, and lithospermic acid) toward metabolism reactions mediated by COMT, SULT, UGT, and ALDH (ALDH, for protocatechuicaldehyde only). Human, rat, and dog liver cytosols containing COMT, SULT, and ALDH were used to induce methylation, sulfation, and oxidation of Danshen catechols,whereas human, rat, and dog liver microsomes containing UGT were used to induce glucuronidation of the compounds. Green, brown, black, and purplish-red semi-openbars denote the data of Danshen catechols undergoing methylation, sulfation, glucuronidation, and aldehyde dehydrogenation, respectively. Green, brown, black, andpurplish-red open bars denote the data of positive controls, i.e., (2)-epicatechin for COMT-mediated methylation, 7-hydroxyflavone for SULT-mediated sulfation, chrysinfor UGT-mediated glucuronidation, and isovanillin for ALDH-mediated aldehyde dehydrogenation, respectively. Values represent the means 6 standard deviations (n = 3).

Fig. 6. Methylation of Danshen catechols (including tanshinol, protocatechuic aldehyde, protocatechuic acid, salvianolic acid B, rosmarinic acid, salvianolic acid A,salvianolic acid D, and lithospermic acid) mediated by COMT in SAM-fortified human (A), rat (B), and dog liver cytosols (C) over time. Blue lines/blue open circles denotethe substrate compounds. Green lines/green open circles, green lines/green semi-open circles, and green lines/green solid circles denote monomethylated, dimethylated, andtrimethylated metabolites, respectively. For the methylation of protocatechuic aldehyde, isovanillin was used to inhibit the catalytic activity of ALDH in human liver cytosol.Values represent the means 6 standard deviations (n = 3).

688 Li et al.

at ASPE



Dow

nloaded from


intravenous injection. Tanshinol (6; 282.8 mmol/subject) was the onlyDanshen catechol of class I (.100 mmol/subject). Protocatechuicaldehyde (1), salvianolic acids B (26) and A (16), rosmarinic acid (10),salvianolic acid D (13), and lithospermic acid (22) had dose levels of10.0–73.2 mmol/subject; they were Danshen catechols of class II (10–100 mmol/subject). The other 21 minor Danshen catechols (0.1–7.2mmol/subject) were of class III (, 10 mmol/subject). In the humansubjects receiving DanHong injection, Danshen catechols of classes Iand II were all detected in plasma. However, only few Danshencatechols of class III, including lithospermic acid isomers (19 and 20;5.2 and 3.9 mmol/subject, respectively) and protocatechuic acid (2; 1.9mmol/subject), were detected in human plasma after dosing. Similarcompound dose-dependent detection of plasma Danshen catechols tookplace in rats and dogs.Among the detected Danshen catechols, tanshinol (6), salvianolic acid

D (13), and lithospermic acid (22) exhibited the most significantsystemic exposure in the human subjects receiving DanHong injection,suggesting that these herbal compounds are worth special considerationin pharmacological research on DanHong injection. Unlike 6 (the onlyDanshen catechol of class I), 13 and 22 were Danshen catechols of classII and exhibited content levels in the dosed injection lower than the othercatechols of the same class. The considerable systemic exposure to 13and 22 appeared to be correlated with their elimination half-lives (t1/2,2.5 and 5.7 hours, respectively) significantly longer than those of 6 andthe other Danshen catechols of class II (0.2–0.6 hour). Rats and dogs arelaboratory animals extensively used in cardiovascular pharmacologicalresearch. Rats receiving DanHong injection exhibited an exposureprofile of unchanged Danshen catechols similar to the human subjects;6, 13, and 22 exhibited levels of systemic exposure higher than the otherDanshen catechols. In the rats, the t1/2 values of 13 and 22 (around 2hours) were longer than those of 6 and the other Danshen catechols ofclass II (,1 hour). Unlike the human subjects and the rats, dogsexhibited considerable systemic exposure to 6 only after dosingDanHong injection. In the dogs, 13 and 22 exhibited short t1/2 (,0.5hour). The observed PK interspecies difference should be considered inpharmacological research of DanHong injection. In PK studies of otherherbal medicines, significant differences in elimination half-life havealso been found to cause considerably different systemic exposure toherbal compounds; long elimination half-life may considerablycounterbalance influence of poor oral bioavailability on systemicexposure to an herbal compound (Liu et al., 2009; Chen et al., 2013;Jiang et al., 2015). With rapidly increasing application of analyticaltechniques, such as liquid chromatography/mass spectrometry, inresearch on herbal medicines, content levels of compounds in herbalformulations and their elimination half-lives have become more andmore available in literature; such data of herbal compounds could beused together as potential PK marker to predict their systemic exposureafter dosing herbal injections.In addition to the systemic exposure to unchanged Danshen

catechols, their elimination was also investigated in the current study.After dosing DanHong injection, tanshinol (6), salvianolic acid D (13),and lithospermic acid (22) circulated mainly as parent compounds withminor or major metabolites. Elimination of 6 and 13 mainly involvedrenal excretion of the parent compounds, whereas 22 was eliminatedmainly via hepatic uptake and methylation followed by hepatobiliaryexcretion of the metabolites. Unlike these three catechols, protocatechuicaldehyde (1) circulated in the blood stream predominantly asmetabolites, rather than the parent compound, and was first transformedinto protocatechuic acid (2), which was then eliminated by conversioninto sulfates and methylated glucuronides followed by renal excretion.Rosmarinic acid (10) also circulated in the blood stream predominantlyas metabolites, rather than the parent compound. This catechol was

eliminated mainly via metabolism, and only about 20% of dosed 10 waseliminated via renal excretion of the parent compound. The metabolismof 10 involved formation of dimethylated products and their sulfates andglucuronides; these metabolites were eliminated comparably via renalexcretion and hepatobiliary excretion. Salvianolic acids A (16) and B(26) were detected at low plasma levels, and their plasma metabolitesexhibited little or no detection in the human subjects. These twoDanshen catechols appeared to be eliminated mainly via hepatobiliaryexcretion of glucuronized and methylated metabolites of 16 andmethylated metabolites of 26. Zhang et al. (2004) reported severalhepatobiliary metabolites of salvianolic acid B in rats, including 3-monomethyl-, 3,399-dimethyl-, 3,3999-dimethyl-, and 3,399,3999-trimethyl-salvianolic acid B.The Danshen catechols have poor membrane permeability, except for

protocatechuic aldehyde (Lu et al., 2008). During elimination, thesecompounds tend to necessitate transporter-mediated basolateral uptakefrom blood as the first step in hepatobiliary and/or renal tubularelimination. Hepatic OATP1B3 and renal OAT1 were found to mediatecellular uptake of Danshen catechols (the details to be publishedelsewhere). Because of their catechol moieties, most Danshen catecholsunderwent COMT-mediated methylation as the primary metabolicpathway in humans, rats, and dogs. Significant intercompound andinterspecies differences in rate and degree of methylation were observedfor Danshen catechols. Hepatic methylation of tanshinol by humanCOMT was significantly slower than those of salvianolic acid D andlithospermic acid, demonstrating their in vitro t1/2 3.73, 0.31, and 0.41hour, respectively. However, the elimination t1/2 of tanshinol (6; 0.6hour) after dosing DanHong injection in human subjects wassignificantly shorter than those of salvianolic acid D (13; 2.5 hours)and lithospermic acid (22; 5.7 hours). This is probably because thetransporter-mediated hepatic and/or renal tubular uptake from blood,rather than the methylation, was the rate-limiting step in the eliminationof these Danshen catechols. Although methylation of drugs oftenhampers their pharmacological activities (Goldstein et al., 2003), theantianginal properties of the major methylated metabolites of Danshencatechols are unknown.Methylation of xenobiotics with catechol moieties involves COMT-

catalyzed transfer of methyl groups from SAM to the catechol hydroxylgroups in the presence of Mg2+ (Zhu et al., 2000). This producesS-adenosylhomocysteine, which can be further converted to homocys-teine by S-adenosylhomocysteine hydrolase in the liver and kidneys (toless extent). An abnormally high blood concentration of homocysteine isrecognized as a risk factor for vascular diseases and neurodegenerativeconditions (McCully, 2007; Vogel et al., 2009; Smith et al., 2010). Thishas raised concerns regarding untoward effects of hepatic methylation ofxenobiotics on the blood concentration of homocysteine. A typicalexample is hyperhomocysteinemia, which is caused by chronic levodopatreatment of patients with Parkinson’s disease (Kuhn, et al., 1998; Mülleret al., 1999; Miller et al., 2003). The Danshen catechols from intravenousDanHong injection undergo COMT-mediated methylation in the body.Further investigation of the relationship between methylation of Danshencatechols and hyperhomocysteinemia risk may facilitate hazard identi-fication for this herbal injection. This may also improve understanding ofthe endothelial protection properties of Danshen catechols (Chan et al.,2004; Cao et al., 2009; Yang et al., 2010; Zhang et al., 2013).In summary, Danshen catechols present in DanHong injection could

be divided, according to their daily dose levels, into three classes (.100,10–100, and ,10 mmol/subject). Tanshinol (6) was the only Danshencatechol of class I; protocatechuic aldehyde (1), salvianolic acids B (26)and A (16), rosmarinic acid (10), salvianolic acid D (13), andlithospermic acid (22) were of class II; the other 21 minor Danshencatechols were of class III. Unchanged Danshen catechols of classes I and


at ASPE



Dow

nloaded from


II were all detected in plasma after intravenous dosing DanHonginjection; only three catechols of class III were detected in plasma but atvery low levels. Only 6, 13, and 22 exhibited considerable systemicexposure in human subjects receiving the injection. These three catecholsexhibited considerable systemic exposure also in rats after dosing, butonly 6 had considerable exposure in dogs. The considerable exposure to6 was due to its highest daily dose, whereas that to 13 and 22 was due totheir relatively long elimination half-lives. The catechols 1 and 10circulated in the blood stream predominantly as metabolites rather thanthe parent compounds; 16 and 26 were detected at low plasma levels withtheir plasma metabolites little or not detected in the human subjects. Both6 and 13 were eliminated mainly via renal excretion of the parentcompounds; 22 was eliminated mainly via hepatobiliary excretion of itsmethylated metabolites. The other Danshen catechols were eliminatedmainly via hepatobiliary and/or renal excretion of their metabolites.COMT-mediated methylation was found to be the primary metabolismfor most Danshen catechols and showed intercompound and interspeciesdifferences in rate and degree. Because the only known mechanism ofhomocysteine production involves SAM-dependent methylation, furtherstudies are warranted to fully characterize the methylation of Danshencatechols and to address safety concern associated with possiblealteration of blood concentrations of homocysteine related to usingDanHong injection.

Authorship ContributionsParticipated in research design: C. Li, L. Li, Huang.Conducted experiments: L. Li, M. Li, Wang, Huang, Y. Li, Zhong, Du, W. Jia,

Dong, J. Li, Xu.Contributed new reagents: Zhao, L. Jia.Performed data analysis: C. Li, L. Li, M. Li.Wrote or contributed to the writing of the manuscript: C. Li, L. Li, M. Li,

Olaleye, Lim.

References

Cao Y, Chai J-G, Chen Y-C, Zhao J, Zhou J, Shao J-P, Ma C, Liu X-D, and Liu X-Q (2009)Beneficial effects of danshensu, an active component of Salvia miltiorrhiza, on homocysteinemetabolism via the trans-sulphuration pathway in rats. Br J Pharmacol 157:482–490.

Chan K, Chui S-H, Wong D-Y, Ha W-Y, Chan C-L, and Wong R-N (2004) Protective effects ofDanshensu from the aqueous extract of Salvia miltiorrhiza (Danshen) against homocysteine-induced endothelial dysfunction. Life Sci 75:3157–3171.

Chen F, Li L, Xu F, Sun Y, Du F-F, Ma X-T, Zhong C-C, Li X-X, Wang F-Q, and Zhang N-T,et al. (2013) Systemic and cerebral exposure to and pharmacokinetics of flavonols and terpenelactones after dosing standardized Ginkgo biloba leaf extracts to rats via different routes ofadministration. Br J Pharmacol 170:440–457.

Cheng C, Liu X-W, Du F-F, Li M-J, Xu F, Wang F-Q, Liu Y, Li C, and Sun Y (2013) Sensitiveassay for measurement of volatile borneol, isoborneol, and the metabolite camphor in ratpharmacokinetic study of Borneolum (Bingpian) and Borneolum syntheticum (syntheticBingpian). Acta Pharmacol Sin 34:1337–1348.

Goldstein DS, Eisenhofer G, and Kopin IJ (2003) Sources and significance of plasma levels ofcatechols and their metabolites in humans. J Pharmacol Exp Ther 305:800–811.

Han J-Y, Fan J-Y, Horie Y, Miura S, Cui D-H, Ishii H, Hibi T, Tsuneki H, and Kimura I (2008)Ameliorating effects of compounds derived from Salvia miltiorrhiza root extract on micro-circulatory disturbance and target organ injury by ischemia and reperfusion. Pharmacol Ther117:280–295.

Ho J-H and Hong C-Y (2011) Salvianolic acids: small compounds with multiple mechanisms forcardiovascular protection. J Biomed Sci 18:30–34.

Hu Z-Y, Yang J-L, Cheng C, Huang Y-H, Du F-F, Wang F-Q, Niu W, Xu F, Jiang R-R, and GaoX-M, et al. (2013) Combinatorial metabolism notably affects human systemic exposure toginsenosides from orally administered extract of Panax notoginseng roots (Sanqi). Drug MetabDispos 41:1457–1469.

Jiang R-R, Dong J-J, Li X-X, Du F-F, Jia W-W, Xu F, Wang F-Q, Yang J-L, Niu W, and Li C(2015) Molecular mechanisms governing different pharmacokinetics of ginsenosides and po-tential for ginsenoside-perpetrated herb-drug interactions on OATP1B3. Br J Pharmacol 172:1059–1073.

Jiang R-W, Lau K-M, Hon P-M, Mak T-C, Woo K-S, and Fung K-P (2005) Chemistry andbiological activities of caffeic acid derivatives from Salvia miltiorrhiza. Curr Med Chem 12:237–246.

Kuhn W, Roebroek R, Blom H, van Oppenraaij D, and Müller T (1998) Hyperhomocysteinaemiain Parkinson’s disease. J Neurol 245:811–812.

Li L, Zhao Y-S, Du F-F, Yang J-L, Xu F, Niu W, Ren Y-H, and Li C (2012) Intestinal absorptionand presystemic elimination of various chemical constituents present in GBE50 extract,a standardized extract of Ginkgo biloba leaves. Curr Drug Metab 13:494–509.

Li M-H, Li Q-Q, Zhang C-H, Zhang N, Cui Z-H, Huang L-Q, and Xiao P-G (2013) Anethnopharmacological investigation of medicinal Salvia plants (Lamiaceae) in China. ActaPharm Sin B 3:273–280.

Li Y-G, Song L, Liu M, Hu Z-B, and Wang Z-T (2009) Advancement in analysis of Salviaemiltiorrhizae Radix et Rhizoma (Danshen). J Chromatogr A 1216:1941–1953.

Liu A-H, Guo H, Ye M, Lin Y-H, Sun J-H, Xu M, and Guo D-A (2007) Detection, character-ization and identification of phenolic acids in Danshen using high-performance liquid chro-matography with diode array detection and electrospray ionization mass spectrometry. JChromatogr A 1161:170–182.

Liu H-F, Yang J-L, Du F-F, Gao X-M, Ma X-T, Huang Y-H, Xu F, Niu W, Wang F-Q, and MaoY, et al. (2009) Absorption and disposition of ginsenosides after oral administration of Panaxnotoginseng extract to rats. Drug Metab Dispos 37:2290–2298.

Liu H-T, Wang Y-F, Olaleye O, Zhu Y, Gao X-M, Kang L-Y, and Zhao T (2013) Character-ization of in vivo antioxidant constituents and dual-standard quality assessment of Danhonginjection. Biomed Chromatogr 27:655–663.

Lu T, Yang J-L, Gao X-M, Chen P, Du F-F, Sun Y, Wang F-Q, Xu F, Shang H-C, and HuangY-H, et al. (2008) Plasma and urinary tanshinol from Salvia miltiorrhiza (Danshen) can be usedas pharmacokinetic markers for cardiotonic pills, a cardiovascular herbal medicine. DrugMetab Dispos 36:1578–1586.

McCully KS (2007) Homocysteine, vitamins, and vascular disease prevention. Am J Clin Nutr 86:1563S–1568S.

Miller JW, Selhub J, Nadeau MR, Thomas CA, Feldman RG, and Wolf PA (2003) Effect of L-dopa on plasma homocysteine in PD patients: relationship to B-vitamin status. Neurology 60:1125–1129.

Müller T, Werne B, Fowler B, and Kuhn W (1999) Nigral endothelial dysfunction, homocysteine,and Parkinson’s disease. Lancet 354:126–127.

Panoutsopoulos GI and Beedham C (2005) Enzymatic oxidation of vanillin, isovanillin andprotocatechuic aldehyde with freshly prepared Guinea pig liver slices. Cell Physiol Biochem15:89–98.

Panoutsopoulos GI, Kouretas D, and Beedham C (2004) Contribution of aldehyde oxidase,xanthine oxidase, and aldehyde dehydrogenase on the oxidation of aromatic aldehydes. ChemRes Toxicol 17:1368–1376.

Peng L-H, Sheng C-L, and Yu Z (2010) DanHong injection for ischemic stroke: a systemicreview. Chin Licenced Pharmacist 7:3–11.

Peng L-H, Yu Z, and Sheng C-L (2011) DanHong injection for angina pectoris: a systemicreview. Chin J Evid-Based Med 11:57–63.

Reagan-Shaw S, Nihal M, and Ahmad N (2008) Dose translation from animal to human studiesrevisited. FASEB J 22:659–661.

Smith AD, Smith SM, de Jager CA, Whitbread P, Johnston C, Agacinski G, Oulhaj A, BradleyKM, Jacoby R, and Refsum H (2010) Homocysteine-lowering by B vitamins slows the rate ofaccelerated brain atrophy in mild cognitive impairment: a randomized controlled trial. PLoSONE 5:e12244.

Tian D-D, Jia W-W, Liu X-W, Wang D-D, Liu J-H, Dong J-J, Li L, Du F-F, Xu F, and WangF-Q, et al. (2015) Methlation and its role in disposition of tanshinol, a cardiovascular car-boxylic catechol from Salvia miltiorrhiza roots (Danshen). Acta Pharmacol Sin. doi: 10.1038/aps.2015.20

Vogel T, Dali-Youcef N, Kaltenbach G, and Andrès E (2009) Homocysteine, vitamin B12, folateand cognitive functions: a systematic and critical review of the literature. Int J Clin Pract 63:1061–1067.

Wang J, Xiong X-J, and Feng B (2013a) Cardiovascular effects of salvianolic Acid B. Evid BasedComplement Alternat Med 2013:247948.

Wang L, Wu Y-H, Zhang W, and Kannan K (2013b) Characteristic profiles of urinaryp-hydroxybenzoic acid and its esters (parabens) in children and adults from the United Statesand China. Environ Sci Technol 47:2069–2076.

Wang L, Zhang Q, Li X-C, Lu Y-L, Xue Z-M, Xuan L-J, and Wang Y-P (2008) Pharmacokineticsand metabolism of lithospermic acid by LC/MS/MS in rats. Int J Pharm 350:240–246.

Wang X-H, Morris-Natschke SL, and Lee KH (2007) New developments in the chemistry andbiology of the bioactive constituents of Tanshen. Med Res Rev 27:133–148.

Xie Y-Y, Xiao X, Luo J-M, Fu C, Wang Q-W, Wang Y-M, Liang Q-L, and Luo G-A (2014)Integrating qualitative and quantitative characterization of traditional Chinese medicine in-jection by high-performance liquid chromatography with diode array detection and tandemmass spectrometry. J Sep Sci 37:1438–1447.

Xu M, Zhang Z-C, Fu G, Sun S-F, Sun J-H, Yang M, Liu A-H, Han J, and Guo D-A (2007)Liquid chromatography-tandem mass spectrometry analysis of protocatechuic aldehyde and itsphase I and II metabolites in rat. J Chromatogr B Analyt Technol Biomed Life Sci 856:100–107.

Yang R-X, Huang S-Y, Yan F-F, Lu X-T, Xing Y-F, Liu Y, Liu Y-F, and Zhao Y-X (2010)Danshensu protects vascular endothelia in a rat model of hyperhomocysteinemia. Acta Phar-macol Sin 31:1395–1400.

Yang Y and Zeng L-X (2012) Meta-analysis on the efficacy and safety of DanHong injection inthe treatment of cerebral infarction. Chin J Gen Pract. 11:372–375.

Zhang W-T, He H, Wang H-D, Wang S-J, Li X, Liu Y, Jiang H-Y, Jiang H, Yan Y-D, and WangY-X, et al. (2013) Activation of transsulfuration pathway by salvianolic acid a treatment:a homocysteine-lowering approach with beneficial effects on redox homeostasis in high-fatdiet-induced hyperlipidemic rats. Nutr Metab (Lond) 10:68.

Zhang Y, Akao T, Nakamura N, Hattori M, Yang X-W, Duan C-L, and Liu J-X (2004) Mag-nesium lithospermate B is excreted rapidly into rat bile mostly as methylated metabolites,which are potent antioxidants. Drug Metab Dispos 32:752–757.

Zhu B-T, Patel UK, Cai M-X, and Conney AH (2000) O-Methylation of tea polyphenols cata-lyzed by human placental cytosolic catechol-O-methyltransferase. Drug Metab Dispos 28:1024–1030.

Address correspondence to: Drs. Chuan Li and Li Li, Laboratory for DMPKResearch of Herbal Medicines, Shanghai Institute of Materia Medica, ChineseAcademy of Sciences, 501 Haike Road, Zhangjiang Hi-Tech Park, Shanghai201203, China. E-mail: [email protected] (C. Li) and [email protected] (L. Li)

690 Li et al.

at ASPE



Dow

nloaded from

mailto:[email protected]

mailto:[email protected]


Systemic Exposure to and Disposition of Catechols Derived...

Documents

Transcript of Systemic Exposure to and Disposition of Catechols Derived...