Simultaneous Characterization of Intravenous and

Original Papers

1121

Authors

Larissa Lachi-Silva 1, Sherwin K. B. Sy 2, 3, Alexander Voelkner 2, João Paulo Barreto de Sousa 4, João Luis C. Lopes 4, Denise B. Silva 4, 5, Norberto P. Lopes 4, Elza Kimura 1, Hartmut Derendorf 2, Andrea Diniz 1

Affiliations

The affiliations are listed at the end of the article

Key words " lychnopholide l " pharmacokinetic l " transit compartment l " model‑based development l " pharmacognosy l " population pharmacokinetic l

Abstract !

The pharmacokinetic properties of a new molecular entity are important aspects in evaluating the viability of the compound as a pharmacological agent. The sesquiterpene lactone lychnopholide exhibits important biological activities. The objective of this study was to characterize the pharmacokinetics of lychnopholide after intravenous administration of 1.65 mg/kg (n = 5) and oral administration of 3.3 mg/kg (n = 3) lychnopholide in rats (0.2 ± 0.02 kg in weight) through nonlinear mixed effects modeling and non-compartmental pharmacokinetic analysis. A highly sensitive analytical method was used to quantify the plasma lychnopholide concentrations in rats. Plasma pro-

Introduction !

received revised accepted

March 11, 2015 May 19, 2015 May 25, 2015

Bibliography DOI http://dx.doi.org/ 10.1055/s-0035-1546214 Published online July 28, 2015 Planta Med 2015; 81: 1121–1127 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0032‑0943 Correspondence Prof. Andrea Diniz Departamento de Farmácia Universidade Estadual de Maringá Av. Colombo, 5790, Bloco K68 Maringá, Paraná, 87020900 Brazil Phone: + 55 44 30 11 49 37 Fax: + 55 44 30 11 48 35 [email protected]

Between 1981 and 2010, over a span of thirty years, the United States Food and Drug Administration granted marketing approval to 175 anticancer drugs; 48.6 % of which are either natural products or derivatives of natural products [1]. This trend reflects an increasing importance of natural products as sources for medicine as well as their utility as a structural chemistry model for synthesis of novel drug candidates [2, 3]. The knowledge of the pharmacokinetic behavior of new compounds, whether they are natural or synthesized products, is an important step in evaluating their feasibility to become novel drugs [4–7]. In pharmaceutical drug development, a model-based approach has been implemented to efficiently bring drug products from the bench to the clinic [8, 9]. This approach also played important roles in key decision-making processes [10]. The application of a model-based approach to * Dedicated to Professor Dr. Dr. h. c. mult. Adolf Nahrstedt on the occasion of his 75th birthday.

tein binding of this compound was over 99 % as determined by a filtration method. A two-compartment body model plus three transit compartments to characterize the absorption process best described the disposition of lychnopholide after both routes of administration. The oral bioavailability was approximately 68%. The clearance was 0.131 l/min and intercompartmental clearance was 0.171 l/min; steady-state volume of distribution was 4.83 l. The mean transit time for the absorption process was 9.15 minutes. No flip-flop phenomenon was observed after oral administration. The pharmacokinetic properties are favorable for further development of lychnopholide as a potential oral pharmacological agent.

the development of natural products, however, has a lot of “catching-up” to do [6]. The modeling approach when applied to natural products can significantly improve our understanding of how to design trials to evaluate their safety and efficacy as well as understanding their pharmacokinetic behavior. Among the many chemical classes of natural compounds, the sesquiterpene lactones include approximately 5000 distinct structures, and the biological activities of many of these compounds are yet to be characterized [11]. The biological activities of the lactone lychnopholide (LYC) was recently characterized, including its antimicrobial [12], antiparasitic [13], anti-inflammatory [14] and antineoplastic effects [15]. Its chemical struc" Fig. 1) consists of alquil-α, β, γ, δ-unsatuture (l rated and α-methylene-γ-lactone, which is thought to be responsible for its cytotoxic activities as well as the inhibition of nuclear factorkappa B (NF-κB) [16]. An internal Michael-type reaction, which is a nucleophilic addition to an unsaturated carbonyl group, can occur in the goyazendolide moiety, creating eremantholide

Lachi-Silva L et al. Simultaneous Characterization of …

Planta Med 2015; 81: 1121–1127

Downloaded by: University of Florida. Copyrighted material.

Simultaneous Characterization of Intravenous and Oral Pharmacokinetics of Lychnopholide in Rats by Transit Compartment Model*

Original Papers

Fig. 1 Chemical structure of lychnopholide.

subtypes. Eremantholydes has no δ-unsaturated and α-methylene-γ-lactone functional groups, reducing the number of possible interactions with proteins and/or enzymes. Several authors suggest a correlation between the presence of δ-unsaturated and α-methylene-γ-lactone with the biological activity of sesquieterpene lactones by the mechanism of alkylation through a Michael-type reaction with the proteinʼs cystein residues [17, 18]. Even with a higher binding property of goyazendolides to macromolecules, such as proteins, than eremantholides, which could block absorption, it was recently demonstrated that one of these sesquieterpene lactones from the goyazendolide moiety, 15-deoxygoyazensolide, was able to cross biological barriers after intraperitoneal administration. The authors confirmed that this alquil-α, β, γ, δ-unsaturated and α-methylene-γ-lactone can be detected in the systemic circulation in animal models; this sesquiterpene lactone was found in rat plasma, 45 min after intraperinoteal administration [19]. Despite the characterization of LYC pharmacological activities and proof that alquil-α, β, γ, δ-unsaturated and α-methylene-γlactone can pass through biological membranes, the pharmacokinetic information of LYC is still unavailable. The purpose of this study was to evaluate the preclinical pharmacokinetics of LYC after intravenous (IV) and oral administrations in rats using nonlinear mixed effect modeling.

Results !

For the protein binding studies, no binding of LYC to the membrane was observed. Lower concentrations (50, 100, and 200 ng/ ml) were evaluated. However, due to a high protein binding rate, the free concentrations at these concentrations were not quantifiable. The protein binding rates of LYC in rat plasma at three LYC concentration levels (500, 1500, and 3000 ng/ml) are shown in " Table 1. For all of the concentrations evaluated, the protein l binding was greater than 99 % and no significant difference between these rates was found (p > 0.05). The free plasma concentration of LYC was estimated to be around 0.48 %. After both IV bolus and oral administrations in rats, the plasma concentrations of LYC were determined by a validated HPLC‑MS/MS method (no shown data). The time course of LYC concentrations after both IV and oral administrations were simultaneously characterized by a two-compartment body model. The model provided reasonable fits to the observed concentration-time profiles, wherein the concentration-time course after the IV route was better characterized by a biexponential decay, whereas the concentrationtime profiles after the oral route had a monoexponential decline

Lachi-Silva L et al. Simultaneous Characterization of …

Table 1 Plasma protein binding of lychnopholide at concentrations of 500, 1500, and 3000 ng/ml (n = 3 per concentration). Concentrations (ng/mL)

Protein binding (%) (mean ± SD)

500 1500 3000

99.99 ± 0.03 99.44 ± 0.09 99.14 ± 0.10

" Fig. 2). The conditional weighted rein its elimination phase (l siduals against time and population fitted data that can be seen " Fig. 2 showed that most of the data lie within at the bottom of l 2 units of the zero ordinate. The points that were less than − 2 were the data from the oral route. Even though the conditional weighted residuals may not be evenly distributed along both time and population predicted concentrations, the fit was considered reasonable given that the drug distribution pattern after the two routes of administration were significantly different. " Fig. 3 shows the individual-predicted and the population-prel dicted concentration-time profiles overlaid with the observed data. The individual-predicted profiles tightly conform to the observed data from both routes of administration. The populationpredicted values and the observed data were very close in the intravenous route, whereas the differences between the two were larger after oral administration. Visual inspection of the prediction- and variability-corrected visual predictive check (pvc-VPC) plot showed a good correlation between the 95 % prediction interval obtained by simulation from the final model and the ob" Fig. 4). In addition, the shrinkage absolute values served data (l were between 4.9 and 6.8 %. The single set of population pharmacokinetic parameter estimates to characterize the concentration profiles for both admin" Table 2. The median and 95 % confidence istrations are listed in l intervals of the parameter estimates were generated from 2000 bootstrap resampling procedures. All the population estimates were encompassed within the 95 % confidence interval obtained from bootstrap. The systemic clearance was 0.131 l/min and the volumes of the central and peripheral compartments were 2.06 and 2.77 l, respectively. The mean steady-state volume of distribution, which is the sum of the two volumes, was 4.83 l. The model estimated oral bioavailability was 68 %, whereas 64% was the bioavailability estimated by the non-compartmental ap" Table 3 shows the non-compartmental parameters. proach. l The systemic clearance and steady-state volume of distribution values estimated after IV and oral administrations by non-compartmental analysis were in good agreement with parameters " Table 2). Comobtained from the compartmental approach (l parisons of other parameters resulting from the two approaches indicate good consistency.

Discussion !

Lychnopholide is a promising natural product with important in vivo and in vitro pharmacological properties as discussed previously [20]. Formulated as a nanostructured parenteral dosage form, lychnopholide was shown to be efficacious against Chagas disease in trypomastigote-infected mouse model [21]. Our study, for the first time, characterizes both the oral and parenteral pharmacokinetics of lychnopholide using a transit compartment model. Considering the scarcity of pharmacokinetic information of sesquiterpene lactones, we utilized both nonlinear mixed ef-

Planta Med 2015; 81: 1121–1127

Downloaded by: University of Florida. Copyrighted material.

1122

Original Papers

1123

Fig. 3 Individual-predicted (solid line), population-predicted (dashed line), and observed (closed circle) concentration-time profiles of lychnopholide in rats modeled by a two-compartment body model with three transit compartments for its absorption process. Animal IDs 1 through 5 were administered 1.65 mg/kg by intravenous bolus route via the left tail vein and animal IDs 6 to 8 were administered 3.30 mg/kg orally by gavage.

fect modeling and non-compartmental pharmacokinetic to investigate the pharmacokinetic behavior of this compound. We evaluated whether the oral pharmacokinetics of lychnopholide could potentially be in a flip-flop situation given that the elimination profile of the oral route follows a one-compartment body model while the parenteral route behaves more closely to a two-compartment model. A flip-flop phenomenon is a case

wherein the absorption process is slower than elimination and could potentially present a bias for the assumptions made during modeling. The terminal slopes of both the oral and parenteral concentration-time profiles were compared. The terminal elimination phase for both routes of administration had a similar halflife, which allowed us to rule out that flip-flop was occurring in the oral administration. The drug distribution following the par-

Lachi-Silva L et al. Simultaneous Characterization of …

Planta Med 2015; 81: 1121–1127

Downloaded by: University of Florida. Copyrighted material.

Fig. 2 Model diagnostics showing observed versus individual and population fitted lychnopholide concentrations and conditional weighted residuals versus time and population fitted values. The solid lines are the expected lines of unity (top) and zero lines (bottom).

Original Papers

Fig. 4 Prediction- and variability-corrected visual predictive check with 2.5th and 97.5th percentiles (dashed lines) and median (solid line) of the simulated profiles, and 90 % confidence interval (shaded area) of the median and prediction intervals as well as the observed lychnopholide concentrations (closed circles).

enteral route was a rapid process that occurred prior to the time of maximum concentration in the oral route and thus the twocompartment characteristics after IV bolus administration were not observed after the oral administration. In order to utilize a single set of parameters to characterize the pharmacokinetics of lychnopholide after both oral and parenteral routes, the concentration-time profiles from both routes of administration were modeled simultaneously assuming a twocompartment body model plus three transit compartments to characterize its absorption process [22–25]. The rapid clearance of lychnopholide suggests that more than one process could be involved, including renal excretion, hepatic metabolism, tissue distribution, and possibly active tubular secretion. The lychnopholide metabolic pathway is not well characterized yet, given that this compound was just recently isolated. The distribution of lychnopholide to the peripheral compartment characterized

Parameter Structural model parameters Number of transit compartments Mean transit time (MTT, min) Clearance (CL, l/min) Volume of central compartment (Vc, l) Intercompartmental clearance (Q, l/min) Volume of peripheral compartment (Vp, l) Bioavailability (F, %) Interindividual variability %CV of CL (ωCL) %CV of Vc (ωVc) Residual variability Additive error parameter Proportional error parameter

by intercompartmental clearance was rapid, which explains the rapid initial decline after the parenteral route and also suggests that lychnopholide could potentially be distributed to tissue compartments where the site of action for its pharmacological effects resides. Assuming that the plasma volume of rats is around 0.03 l/kg [26], the difference in its steady-state volume of distribution is 700-fold (~ 24 l/kg). For other sesquiterpene lactones, the volumes of distribution were 353.5, 15.36, and 11.69 l/kg for artemisinin, alantolactone, and isoalantolactone, respectively [27, 28]. The latter two lactones have a comparable volume of distribution to lychnopholide. The estimated octanol : water partition coefficients (logKo : w) were 2.84, 2.85, 3.27, and 3.35 for lychnopholide, artenisinin, alantolactone, and isoalantolactone, respectively [29]. The large magnitude in logKo : w suggests high permeability, corroborating with our pharmacokinetic findings. Despite high permeability, lychnopholide is poorly soluble. Its solubility in water was estimated to be 20.33 mg/l [29]. Lychnopholide has an oral bioavailability of approximately 68 %, achieving a maximum concentration of 50.7 ± 1.6 ng/ml after the administration of 3.3 mg/kg by body weight. The mean transit time for the absorption through the three transit compartments was approximately 9 min. The use of a transit compartment resolved some issues with model convergence using alternative models, which we speculate that these alternate models were compensating for the one-compartmental behavior in the oral data against a two-compartment model for the intravenous data. Several alternative absorption models were evaluated, including first-order with lag time, parallel first- and zero-order and sequential first- and zero-order absorption models. These models, however, did not result in satisfactory model convergence. Lychnopholide, as a parthenolide, contains functional groups that are known for high protein affinity, acting as an alkylating agent of the free thiol groups on the cysteine residues of protein molecules. The alkyl-α, β, γ, δ-unsaturated and α-methylene-γ-lactones are responsible for protein binding as well as for their biological activities [30]. This study determined that the plasma protein binding of lychnopholide was > 99 %, which is compatible with another lactone parthenolide whose protein binding was over 98% [31]. Lychnopholide exhibited good oral bioavailability, with rapid oral absorption, distribution, and clearance. These pharmacokinetic properties are favorable for the development of lychnopholide as an oral pharmacological agent.

Mean value (%RSE)

Median [95% CI]‡

3 9.15 (1.4%) 0.131 (4.9%) 2.06 (37%) 0.171 (4.3%) 2.77 (2.2%) 0.68 (5.1%)

3 9.10 [8.90, 9.32] 0.132 [0.124, 0.147] 2.07 [1.96, 2.17] 0.169 [0.159, 0.176] 2.76 [2.65, 2.85] 0.68 [0.64, 0.74]

10 (77 %) 88 (14 %)

10 [4, 14] 78 [85, 94]

0.99 (55%) 0.0611 (18 %)

1.02 [0.32, 1.65] 0.0605 [0.0457, 0.0712]

RSE = relative standard error; CI = confidence interval; ‡Median [95% CI] obtained from 2000 bootstrap resampling procedures

Lachi-Silva L et al. Simultaneous Characterization of …

Planta Med 2015; 81: 1121–1127

Table 2 Population pharmacokinetic model parameters of the simultaneous fit of IV bolus and oral lychnopholide concentration-time profiles.

Downloaded by: University of Florida. Copyrighted material.

1124

Original Papers

Parameters (unit) Intravenous bolus administration Dose (µg) AUC0-∞ (µg min l−1) (min) t1/2 CL (l · min−1) Vss (l) Oral administration Dose (µg) AUC0-∞ (µg min l−1) t1/2 (min) Cmax (µg · l−1) (min) Tmax CL (l · min−1) F (%)

Mean ± SD

Median [range]

353 ± 5.8 2447 ± 303 27.4 ± 2.7 0.15 ± 0.02 4.30 ± 0.15

350 [350, 363] 2483 [1935, 2701] 27.4 [23.2, 30.2] 0.14 [0.13, 0.18] 4.34 [4.13, 4.5]

655 ± 42 2940 ± 158 37.9 ± 9.4 50.7 ± 1.6 10 ± 0 0.13 ± 0.01 64

677 [606, 681] 2955 [2775, 3090] 34.2 [30.1, 48.6] 50.2 [49.3, 52.4] 10 [10, 10] 0.13 [0.12, 0.15]

1125

Table 3 Pharmacokinetic parameters of lychnopholide after IV bolus (n = 5) and oral administrations (n = 3) from non-compartmental analysis.

AUC = area under the curve; t1/2 = half-life; CL = clearance; Vss = steady-state volume of distribution; Cmax = maximum concentration;

Materials and Methods

Pharmacokinetic studies

!

Male Wistar rats, weighting 0.200 ± 0.020 kg, were supplied by the animal facilities of Universidade Estadual de Maringá, Maringá, Paraná, Brazil. The protocol was approved by the local animal ethics committee (Comissão de Ética no Uso de Animais da Universidade Estadual de Maringá; CEUA/UEM – 098/2013). The animals were housed under controlled temperature (22 ± 2 °C) and light (12-h dark-light cycle), and were provided with water and food freely. The animals were separated into two groups. Group 1 (n = 5) received 1.65 mg/kg in 0.5 ml solution through the left tail vein via intravenous bolus administration. Group 2 (n = 3) received 3.30 mg/kg in 1 ml of LYC solution via oral administration by gavage. Blood samples (200 µl) were collected in heparinized microtubes from the right tail vein at 0, 5, 10, 15, 30, 45, 60, 90, and 120 min after administration. Samples were immediately centrifuged at 2140×g for 15 min at 4 °C and plasma was transferred to another microtube and kept at − 80 °C until HPLC‑MS/ MS analysis. The solution containing the highest soluble concentration of LYC (0.75 mg · ml−1) was prepared in water : Tween80:dimethyl sulfoxide (80 : 13.3 : 6.7, v/v/v) and the doses (1.65 mg/kg and 3.30 mg/kg) that were administered to the animals were determined by the administered volume of the solution (0.5 ml and 1 ml for intravenous bolus and oral administration, respectively).

Materials Lychnopholide was provided by Research Nucleo for Natural Products and Synthesis (NPPNS) of the Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo. This lactone was isolated and characterized with a purity of 96 %. Lychnopholide solution (0.75 mg/ml) was prepared with DMSO : Tween80 : water (6.7 : 13.3 : 80).

Analytical method Plasma lychnopholide concentrations were quantified by Waters Alliance e2965 linked to a mass spectrometer, QuattroPremier XE, with electrospray ionization. The data acquisition and postrun analyses were performed by MassLynx v. 4.1 (Waters Corp). The analytical column was ACE3 C18–300 (50 mm×2.1 mm, 3 µm) and the guard column was ACE C18–300. The mobile phase was prepared with methanol : water added with 0.1 % formic acid (80 : 20, v/v). The elution system was isocratic and the flow rate was 0.4 ml/min. The injection volume was 10 µl. Electrospray ionization in positive ion mode with multiple reaction monitoring (MRM) was used for quantification of ion transitioning m/z 359 → 259 for LYC and m/z 285 → 154 for internal standard (IS), which is diazepam. Method validation was done in accordance with regulatory guidelines for including parameters selectivity, lower limit of quantification (LLOQ), matrix effect, linearity, accuracy, precision, recovery and stability (data not shown).

Plasma protein binding Lychnopholide protein binding was evaluated at concentrations of 50, 100, 200, 500, 1500, and 3000 ng/ml by a filtration method, as previously described [32]. For each concentration of plasma sample, there was a corresponding sample prepared in aqueous solution. Samples of blank plasma and water were analyzed as well. All these solutions were allowed to reach equilibrium for 20 min at 36.5 °C in a water bath. The samples (500 µl) were then transferred to Microcon-10kDa with an Ultracel YM-10 membrane and centrifuged at 10 000 rpm for 40 min at 23 °C. The free drug concentration in ultrafiltrate (100 µl) was quantified by the HPLC‑MS/MS system described above.

Nonlinear mixed effects model and non-compartmental analysis The population pharmacokinetic model for lychnopholide was simultaneously fitted to the plasma concentration-time profiles in rats after both oral and parenteral administrations using NONMEM® 7.2.0 and the first-order conditional estimation (FOCE) method with η-interaction. A two-compartment model with three transit compartments linked to the central compartment " Fig. 5). The by first-order transit rate constants was explored (l structural model had a total of five compartments with the following pharmacokinetic parameters: apparent clearance (CL/F), intercompartmental clearance (Q/F), volume of the central compartment (VC/F), volume of the peripheral compartment (VP/F), mean transit time (MTT), and a bioavailability term (F). The MTT was defined as the number of transit compartments plus 1 then divided by the transit rate constant (ktr). F was a logit transformed parameter so that the boundary was between 0 and 1:

Lachi-Silva L et al. Simultaneous Characterization of …

Planta Med 2015; 81: 1121–1127

Downloaded by: University of Florida. Copyrighted material.

Tmax = time of maximum concentration; F = bioavailability

Original Papers

intervals of the pharmacokinetic parameters, the fixed effect, and the random effect parameters. The bias of each parameter was evaluated by comparing the median value derived from the bootstrap and the final pharmacokinetic parameter mean values. The non-compartmental analysis was performed using Phoenix WinNonlin (version 6.3; Pharsight). The linear trapezoidal rule was used for integration. Descriptive statistics was used to summarize the pharmacokinetic parameters of the analysis.

Acknowledgements ! Fig. 5 Schematic representation of lychnopholide pharmacokinetic model.

 y ¼ log

F 1F

The authors thank FAPESP (Process number: 2012/18031–7, 2009/54098–6 and 2014/50265–3) – Biota, CNPq and CAPES for funding the study.

Conflict of Interest



!

There is no conflict of interest.

expðyÞ and F ¼ y1 ¼ 1þexpðyÞ

Affiliations F is only determined if the profile comes from the oral route, otherwise, F is assumed to be 1. Exponential interindividual variability terms were included in CL and VC only, using: Pi = P × exp(ηi), as this approach was optimal for convergence, where: P represents the population mean and ηi describes the interindividual variability, which was assumed to be independently and normally distributed with mean zero and variance: ω2p. The subscript i represents the individual rats. Since the study was not a crossover study, F was designated as a population value. Model selection was based on the minimum of the objective function value, which is approximately − 2 log-likelihood, precision of the parameter estimates expressed as relative standard error [RSE(%)], and goodness-of-fit plots as well as plots of individual predicted, population-predicted, and observed concentration versus time for each individual animal. The residual variability was described by a combined additive and proportional error model. The subroutine was Advan6 with a tolerance of 9. The drug concentrations were designated for the central compartment. For orally administered lychnopholide, the dose was introduced into the first transit compartment, whereas the dose was injected into the central compartment for the parenteral route of administration. This strategy allowed for the simultaneous fit of the profiles from both oral and IV bolus administrations. The predictability of the final model was evaluated using pvc-VPC with PerlspeaksNONMEM (PsN) version 3.5.5 running ActivePerl 5.12. The plasma concentration-time profiles were simulated in 1000 replicates using the final population pharmacokinetic model. The median, 95% prediction interval, and 90 % confidence interval of the prediction interval and median were computed and overlaid with the observed data to evaluate the predictive performance of the final model. The majority of the individual observations should be enclosed within the 2.5th and 97.5th percentiles of the simulated data if the final model adequately describes the original data. The pvc-VPC plots were generated using the Xpose4 package running R 3.1.2. The model was evaluated for stability by a bootstrap resampling technique, which involved resampling from the original data. A total of 2000 replicates of the data were generated by bootstrap in PsN for NONMEM analysis, using the final model, to obtain the median and 95 % confidence

Lachi-Silva L et al. Simultaneous Characterization of …

1

2

3

4

5

Preclinical Pharmacokinetic Laboratory, Department of Pharmacy, Maringa State University, Maringa, PR, Brazil Department of Pharmaceutics, College of Pharmacy, University of Florida, Gainesville, FL, USA Biostatistics Master Program, Department of Statistics, Maringa State University, Maringa, PR, Brazil NPPNS (Núcleo de Pesquisa em Produtos Naturais e Sintéticos), Departamento de Física e Química, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brazil LAPNEM (Laboratório de Produtos Naturais e Espectrometria de Massas), Centro de Ciências Biológicas e da Saúde (CCBS), Universidade Federal de Mato Grosso do Sul (UFMS), Campo Grande, MS, Brazil

References 1 Newman DJ, Cragg GM. Natural products as sources of new drugs over the 30 years from 1981 to 2010. J Nat Prod 2012; 75: 311–335 2 Kinghorn AD, Chin YW, Swanson SM. Discovery of natural product anticancer agents from biodiverse organisms. Curr Opin Drug Discov Devel 2009; 12: 189–196 3 Krause J, Tobin G. Discovery, development and regulation of natural products. In: Marianna K, editor. Using old solutions to new problems – natural drug discovery in the 21st century. Rijeka, Croatia: InTech; 2013: 3–36 4 Szychowski J, Truchon JF, Bennani YL. Natural products in medicine: transformational outcome of synthetic chemistry. J Med Chem 2014; 57: 9292–9308 5 Galvanin F, Ballan CC, Barolo M, Bezzo F. A general model-based design of experiments approach to achieve practical identifiability of pharmacokinetic and pharmacodynamic models. J Pharmacokinet Pharmacodyn 2013; 40: 451–467 6 Brantley SJ, Gufford BT, Dua R, Fediuk DJ, Graf TN, Scarlett YV, Frederick KS, Fisher MB, Oberlies NH, Paine MF. Physiologically based pharmacokinetic modeling framework for quantitative prediction of an herbdrug interaction. CPT Pharmacometrics Syst Pharmacol 2014; 3: e107 7 Sy SK, Wang X, Derendorf H. Introduction to pharmacometrics and quantitative pharmacology with an emphasis on physiologically based pharmacokinetics. In: Schmidt S, Derendorf H, editors. Applied pharmacometrics. New York: Springer; 2014: 1–64 8 Lalonde RL, Kowalski KG, Hutmacher MM, Ewy W, Nichols DJ, Milligan PA, Corrigan BW, Lockwood PA, Marshall SA, Benincosa LJ, Tensfeldt TG, Parivar K, Amantea M, Glue P, Koide H, Miller R. Model-based drug development. Clin Pharmacol Ther 2007; 82: 21–32 9 Milligan PA, Brown MJ, Marchant B, Martin SW, van der Graaf PH, Benson N, Nucci G, Nichols DJ, Boyd RA, Mandema JW, Krishnaswami S, Zwillich S, Gruben D, Anziano RJ, Stock TC, Lalonde RL. Model-based drug development: a rational approach to efficiently accelerate drug development. Clin Pharmacol Ther 2013; 93: 502–514 10 Miller R, Ewy W, Corrigan BW, Ouellet D, Hermann D, Kowalski KG, Lockwood P, Koup JR, Donevan S, El-Kattan A, Li CS, Werth JL, Feltner DE, La-

Planta Med 2015; 81: 1121–1127

Downloaded by: University of Florida. Copyrighted material.

1126

11

12

13 14

15

16

17 18

19

20

21

22

londe RL. How modeling and simulation have enhanced decision making in new drug development. J Pharmacokinet Pharmacodyn 2005; 32: 185–197 Chadwick M, Trewin H, Gawthrop F, Wagstaff C. Sesquiterpenoids lactones: benefits to plants and people. Int J Mol Sci 2013; 14: 12 780– 12 805 Barrero A, Oltra J, Justícia J, Raslan D, Silva E. Atividade antibacteriana de furanoeliangólidos/Antibacterial activity of furanoeliangolides. Rev Bras Farmacogn 2002; 12: 7–10 Saúde-Guimarães DA, Faria AR. Substâncias da natureza com atividade anti-Trypanosoma cruzi. Rev Bras Farmacogn 2007; 17: 455–465 Ferrari FC, Ferreira LC, Souza MR, Grabe-Guimaraes A, Paula CA, Rezende SA, Saude-Guimaraes DA. Anti-inflammatory sesquiterpene lactones from Lychnophora trichocarpha Spreng. (Brazilian Arnica). Phytother Res 2013; 27: 384–389 Raslan D, Oliveira A. In vitro antitumor activity of sesquiterpene lactones from Lychnophora trichocarpha. Revista Bras Plant Med 2014; 16: 275–282 Ren Y, Acuna UM, Jimenez F, Garcia R, Mejia M, Chai H, Gallucci JC, Farnsworth NR, Soejarto DD, Carcache de Blanco EJ, Kinghorn AD. Cytotoxic and NF-kappaB inhibitory sesquiterpene lactones from Piptocoma rufescens. Tetrahedron 2012; 68: 2671–2678 Schmidt TJ. Toxic activities of sesquiterpene lactones: structural and biochemical aspects. Curr Org Chem 1999; 3: 4 Schmidt TJ, Khalid SA, Romanha AJ, Alves TM, Biavatti MW, Brun R, Da Costa FB, de Castro SL, Ferreira VF, de Lacerda MV, Lago JH, Leon LL, Lopes NP, das Neves Amorim RC, Niehues M, Ogungbe IV, Pohlit AM, Scotti MT, Setzer WN, de N C Soeiro M, Steindel M, Tempone AG. The potential of secondary metabolites from plants as drugs or leads against protozoan neglected diseases – part II. Curr Med Chem 2012; 19: 2176–2228 Jabor VA, dos Santos MD, Bonato PS, Gouvea DR, Lopes NP. A new highperformance liquid chromatography assay for the determination of sesquiterpene lactone 15-deoxygoyazensolide in rat plasma. Anal Chim Acta 2007; 601: 212–217 Canalle R, Burim RV, Callegari Lopes JL, Takahashi CS. Assessment of the cytotoxic and clastogenic activities of the sesquiterpene lactone lynchnopholide in mammalian cells in vitro and in vivo. Cancer Detect Prev 2001; 25: 93–101 Branquinho RT, Mosqueira VC, de Oliveira-Silva JC, Simoes-Silva MR, Saude-Guimaraes DA, de Lana M. Sesquiterpene lactone in nanostructured parenteral dosage form is efficacious in experimental Chagas disease. Antimicrob Agents Chemother 2014; 58: 2067–2075 Sy SK, de Kock L, Diacon AH, Werely CJ, Xia H, Rosenkranz B, van der Merwe L, Donald PR. N-acetyltransferase genotypes and the pharmaco-

23

24

25

26 27

28

29

30

31

32

kinetics and tolerability of para-aminosalicylic acid in drug-resistant pulmonary tuberculosis patients. Antimicrob Agents Chemother, advance online publication 2015 May 11; DOI: 10.1128/AAC.04049-14 de Kock L, Sy SK, Rosenkranz B, Diacon AH, Prescott K, Hernandez KR, Yu M, Derendorf H, Donald PR. Pharmacokinetics of para-aminosalicylic acid in HIV-uninfected and HIV-coinfected tuberculosis patients receiving antiretroviral therapy, managed for multidrug-resistant and extensively drug-resistant tuberculosis. Antimicrob Agents Chemother 2014; 58: 6242–6250 Savic RM, Jonker DM, Kerbusch T, Karlsson MO. Implementation of a transit compartment model for describing drug absorption in pharmacokinetic studies. J Pharmacokinet Pharmacodyn 2007; 34: 711–726 Shilbayeh SA, Sy SK, Melhem M, Zmeili R, Derendorf H. Quantitation of the impact of CYP3A5 A6986G polymorphism on quetiapine pharmacokinetics by simulation of target attainment. Clin Pharmacol Drug Devel, advance online publication 12 January 2015; DOI: DOI: 10.1002/ cpdd.172 Davies B, Morris T. Physiological parameters in laboratory animals and humans. Pharm Res 1993; 10: 1093–1095 Xing J, Yan H, Zhang S, Ren G, Gao Y. A high-performance liquid chromatography/tandem mass spectrometry method for the determination of artemisinin in rat plasma. Rapid Commun Mass Spectrom 2006; 20: 1463–1468 Guo C, Zhang S, Teng S, Niu K. Simultaneous determination of sesquiterpene lactones isoalantolactone and alantolactone isomers in rat plasma by liquid chromatography with tandem mass spectrometry: application to a pharmacokinetic study. J Sep Sci 2014; 37: 950–956 Exposure Assessment Tools and Models. Estimation program interface suite (EPI Suite) version 4.11. Washington, DC: US Environmental Protection Agency; 2012. Software available at http://www.epa.gov/ opptintr/exposure/pubs/episuitedl.htm Branquinho RT, Mosqueira VC, Kano EK, de Souza J, Dorim DD, SaudeGuimaraes DA, de Lana M. HPLC‑DAD and UV-spectrophotometry for the determination of lychnopholide in nanocapsule dosage form: validation and application to release kinetic study. J Chromatogr Sci 2014; 52: 19–26 Wagner S, Kratz F, Merfort I. In vitro behaviour of sesquiterpene lactones and sesquiterpene lactone-containing plant preparations in human blood, plasma and human serum albumin solutions. Planta Med 2004; 70: 227–233 Diniz A, Escuder-Gilabert L, Lopes NP, Villanueva-Camanas RM, Sagrado S, Medina-Hernandez MJ. Characterization of interactions between polyphenolic compounds and human serum proteins by capillary electrophoresis. Anal Bioanal Chem 2008; 391: 625–632

Lachi-Silva L et al. Simultaneous Characterization of …

Planta Med 2015; 81: 1121–1127

1127

Downloaded by: University of Florida. Copyrighted material.

Original Papers