Pharmacokinetics and pharmacodynamics of glycyrrhetinic acid with Paeoniflorin after transdermal administration in dysmenorrhea model mice.
Chinese peony-Liquorice Decoction (PLD), a famous traditional Chinese medicine preparation containing Chinese peony (the roots of Paeonia lactiflora Pall, family Paeoniaceae) and Liquorice (roots of Glycyrrhiza uralensis Fisch, family Leguminosae), has been widely used in Asia for hundreds of years. The main ingredients in PLD are paeoniflorin (PF, Fig. lb) from Chinese peony and glycyrrhetinic acid (GA, the aglycone which is hydrolysed from glycyrrhizic acid by intestinal bacterias after oral ingestion, Fig. la) from Liquorice (Wang et al. 1998).
Recent years, some researches have reported the pharmacological mechanism of PF and GA respectively, that PF exerts its anti-inflammatory action by inhibiting the production of proinflammatory cytokines (Chang et al. 2009; Wu et al. 2012) and GA has been proven to be antispasmodic on tissues in vitro (Kaifuchi et al. 2015; Tsuji et al. 2012). Furthermore, it has been defined that GA was the main component in PLD to inhibit pain accompanying cramps, compared to PF (Lee et al. 2013).
Although PLD or GA-PF has obvious pharmacological activities, its oral administration also results in serious side effects, such as hypertension, gastrointestinal irritation and pseudoaldosteronism disease (Ploeger et al. 2000; Raphael and Kuttan 2003). Moreover, an analgesic therapy often needs continuous effect, while the hydrophilicity of PF indicated it might show poor retention in systemic circulation after single administration. And transdermal therapeutic systems (US) could offer an alternative route that bypasses the gut, leading to a safe therapy, and are commonly considered to be suitable to transport drugs in steady flux. Presently available transdermal patches can be classified into two categories on the basis of their design: reservoir-type and matrix-type patches, and matrix-type patches combine the drug, adhesive and mechanical backbone of the patch into a simpler design (Prausnitz et al. 2004). The transdermal drug delivery of GA and PF has been investigated in our previous study (Hao et al. 2010). The pharmacokinetic comparison of PLD with different combination of the two plant materials in rats has been investigated (Xu et al. 2013). Considering the pharmacologic effect may not directly parallel the plasma drug concentration, a simple PK study focusing on concentration-time cannot provide an exact description of the therapeutic process, and a pharmacokinetics/pharmacodynamics (PK/PD) study appears to be a more reasonable way to evaluate.
In the present study, the GA-PF TTS, the matrix-type transdermal patches were prepared, and optimized through in vitro permeation experiments and in vivo analgesic effect experiments with dysmenorrhea model mice (DM). In addition, the primary assessment of the synergism for the two ingredients was made and the topical distribution and the PK/PD for this formulation were also investigated.
Materials and methods
Glycyrrhetinic acid (GA, HPLC purity>98%, Lot # ZL0806008) was purchased from Zelang Pharmaceutical Technology Co. Ltd. (Nanjing, China); paeoniflorin (PF, HPLC purity>95%, Lot # 081,012) was purchased from Lanbei Pharmaceutical Technology Co. Ltd. (Chengdu, China); meloxicam tablets were purchased from Boehringer Ingelheim Pharma GmbH & Co. KG (Ingelheim, Germany); gardenoside (HPLC purity>95%, Lot # 100,382-200,802) and tanshinone I1A (HPLC purity>98%, Lot # 110,766-200,416) were purchased from National Institutes for Food and Drug Control (Beijing, China); triethanolamine, ethanol, butanol, monopotassium phosphate, sodium tetraborate and glycerin were purchased from Bodi Drug Manufacturing Co. Ltd. (Tianjin, China); sodium hydrate, sucrose, amidulin and acetonitrile, which were of HPLC grade, were purchased from Kemiou Technology Co. Ltd. (Tianjin, China); isopropyl myristate was obtained from China National Medicines Co. Ltd. (Shenyang, China); and orthophosphoric acid was purchased from Zhiao Chemical Reagents Research Institute (Anshan, China). All other chemicals and solvents were of reagent grade.
Chinese Kun Ming (KM) mice ([female]) weighing 20-25 g were supplied by Dalian Medical University were used in all experiments. The mice were housed and cared for according to the guidelines and polices of China, Liaoning Province and Dalian University of Technology. Approval for this project was granted by the Institutional Ethical Committee for Biological and Medical Experimentation of Dalian University of Technology.
All of the patches used in this work were prepared as follows. Appropriate amounts (listed in Table 1) of DURO-TAK[R] 87-4098 or 87-2677, GA, PF, and triethanolamine were added to butanol and mixed with permeation enhancer isopropyl myristate (Hao et al. 2010), then left to stand at 40[degrees]C in a water bath for 30 min to remove air bubbles. The mixture was spread onto the release liner (Scotchpak 9748, 3 M, MN, USA) at a constant thickness of 0.5 mm, let stand at room temperature for 20 min, and then dried at 60[degrees]C for 30 min. After cooling, the patches were covered by a backing layer (Cotran 9701, 3 M, MN, USA) and placed in an aluminium foil bag. DURO-TAK[R] is a series of commonly used acrylate-vinylacetate pressure sensitive adhesives supplied in organic solvent solutions, 87-2677 contains -COOH as functional group, and 87-4098 contains no functional group. Scotchpak 9748 release liner is fluoropolymer coated polyester film. Cotran 9701 backing layer is flexible polyurethane film that is occlusive and breathable.
In vitro permeation experiments
These experiments were performed using a modified Franz type horizontal glass diffusion cell consisting of two half- cells equipped with star-head magnetic stirrers and water jackets connected to a water bath (KX-5P Permcell horizontal diffusion cells, Dalian Kexiang Instruments Co., Ltd, China). Each cell had a volume of 5.0 ml and an effective diffusion area of 0.627 [cm.sup.2]. Skins and receptor solutions were prepared as described in the literature (Hao et al. 2010). The skins applied in the experiments sourced from female KM mice. The hair of abdominal skin was removed carefully with clippers and shaver, after the animals were sacrificed by dislocating the spinal cord. The skin was excised and the fat and sub-dermal tissue were removed. After measuring the thickness the skins were stored at -20[degrees]C until being used in 1 week. The excised back skin attached to a patch was mounted to the cell such that the dermal side of the skin faced the receiver fluid. After securely clamping the cell assembly together, the receptor compartment was filled with 5.0 ml receptor solution which was phosphate buffer solution (PBS, pH 7.4) containing 40% (v/v) ethanol (40% ethanol PBS) to maintain sink conditions and continuously stirred at approximately 700 rpm. The temperature of the cell was maintained at 37 [+ or -] 0.5[degrees]C with thermostatically controlled water, which was circulated through a water jacket surrounding the cell body throughout the experiments. Care was taken to ensure that no air bubbles remained in the water jacket. At predetermined time intervals, 0.2 ml of receptor solution was collected for analysis and replaced with the same volume of fresh solution. The samples were analysed by HPLC, and the experiments were replicated at least three times.
Optimization of compatibility proportion
To produce dysmenorrhea model mice, mice were administered (2 mg/kg) diethylstilbestrol solution by gavage for 12 days and injected oxytocin solution (20 U/kg) intraperitoneally 60 min after the last administration of diethylstilbestrol to produce a writhing response. Mice that produced more than two writhes (number of writhes, NW > 2) over a 30 min period were selected for the experiments.
The DM (n = 90) were randomly divided into nine groups: (1) positive control group, the mice were administered a meloxicam suspension (5 mg/kg) by gavage each day for three days and received a oxytocin injection 60 min after the last administration: (2) negative control group, the mice were attached a blank patch without GA and PF each day for three days, and received an oxytocin injection 12 h after the last administration; (3) blank control group, the mice received nothing and received an oxytocin injection at the same time as the mice in the treated groups; and (4)~(9) GA-PF patch groups, the mice were attached GA-PF patches with different compatibility proportions each day for three days and received an oxytocin injection 12 h after the last administration.
For all of the experimental mice, the analgesic effect was displayed by the pain inhibition ratio (PIR) and prolongation of the antalgic latency ([DELTA]AL), which were calculated according to formulas (1) and (2):
PIR = ([NW.sub.negative] - [NW.sub.administration])/[NW.sub.negative] x 100% (1)
[DELTA]AL = [AL.sub.administrati0n] - [AL.sub.negative] (2)
where [NW.sub.negative] is the NW exhibited by the mice in the negative control group, [NW.sub.administratjon] is the NW exhibited by the mice in the positive control group or GA-PF patch groups, [DELTA]AL is the prolongation of the AL, [AL.sub.negative] is the AL of the negative control group, and [AL.sub.administration] is the AL of the positive control group or GA-PF patch groups.
The interaction was evaluated with the Bliss Independence (BI) criterion (Afeltra et al. 2004; Cazzola et al. 2014) and calculated according to formula (3):
[E.sub.A+B] = [E.sub.A] + [E.sub.B] - [E.sub.A] x [E.sub.B] (3)
where [E.sub.A+B] is the theoretical combination of drugs A and B which is an expected value, and [E.sub.A] and [E.sub.B] are the experimental percentages of inhibition of each drug acting alone, respectively. If the pharmacological effect exerted by combination of A and B is higher than the expected value, the interaction is synergistic; if it is lower than the expected value, the interaction is antagonistic. The BI criterion assumed the two compounds act independently from one another, it is adequate to our work, which the two ingredients own different mechanism and both showed analgesic effect.
Pharmacokinetics, pharmacodynamics, and topical distribution
Abdominal hairs removed from DM (n = 128) were randomly divided into eight groups. From each group, six DM were selected randomly and divided equally into a negative control group, a positive control group and a blank control group. The ten remaining DM from each group were treated with 10%-10% GA-PF patches. A single patch with an area of about 1 [cm.sup.2] contained 2.0 mg GA and 2.0 mg PF respectively. The patches used to treat the negative control and GA-PF patch-treated animals were removed at 48 h after administration. Oxytocin solution was injected intraperitoneally at 4, 8, 12, 24, 36, 48, 60 and 72 h, and subsequently, the antalgic latency and number of the writhing response over a 30 min period were recorded by blind experimenters. Blood (0.5 ml) was collected from the retro-orbital sinus through heparinized capillary tubes; the mice were then sacrificed and the skins under the patches were sampled. Plasma samples were separated immediately by centrifugation at 1700 g for 10 min then the supernatants and skin specimens were separately stored at -20[degrees]C until analysis.
Fifty microliters of internal standard solution and 50 [micro]l of methanol were added to 50 [micro]l of plasma sample in 1.5ml Eppendorf tubes. The mixture was vortexed for 60s and separated by centrifugation at 1700 g for 10 min. The supernatant was then collected for analysis by LC-MS/MS.
Skin samples (50 mg) were cut into small pieces in a 10-ml centrifuge tube, and 1.0 ml of methanol was added. The sample was homogenized at 18,000 rpm for 1 min, and the supernatant was obtained by centrifugation at 1700 g for 10 min. The supernatant was treated using the protocol used for the plasma samples and analysed by HPLC-UV as reported previously (Sun et al. 2012).
HPLC-UV and LC-MS/MS condition
GA and PF were analysed separately due to the huge difference in their retention behaviours on a reversed-phase column. Gardenoside and tanshinone IIA were selected as internal standards for GA and PF, respectively.
A Shimadzu instrument (LC 2010A, LC solution workstation) equipped with a UV detector and a Diamonsil C18 column (5 [micro]m, 4.6 mm x 150mm) were used. The column was maintained at 30[degrees]C and the flow rate was 1.0 ml/min. The volume of injection was 40 [micro]l. GA was analysed with acetonitrile-methanol-0.1 % orthophosphoric acid solution (50:35:15, v/v/v) at 254 nm, and PF was analysed with acetonitrile-0.1 % orthophosphoric acid solution (16:84, v/v) at 230 nm.
Please note: Some tables or figures were omitted from this article.
Abbreviations: PLD, Chinese peony-Liquorice Decoction; GA, glycyrrhetinic acid; PF, paeoniflorin; TTS, transdermal therapeutic system; PK, pharmacokinetics; PD, pharmacodynamics; PSB, phosphate buffer solution; HPLC, high-performance liquid chromatography; DM, dysmenorrhea model mice; NW, number of writhes exhibited by mice; PIR, pain inhibition ratio; AL, antalgic latency; BI, Bliss independence; UV, ultraviolet; LC-MS/MS, liquid chromatography tandem mass spectrometry; ESI, electrospray ionization; CID, collision-induced dissociation; SRM, selected reaction monitoring; QC, quality control; [C.sub.max]. the peak plasma drug concentration; [t.sub.max], time to peak plasma drug concentration; [AUC.sub.0-t], area under the time-concentration curve from zero to a definite time t; [AUC.sub.0-[infinity]], area under the time-concentration curve from zero to infinite time; [t.sub.1/2], half-life of elimination; MRT, mean residence time; AIC, Akaike information criterion; [E.sub.max], maximum pharmacological effect; [EC.sub.50]. drug concentration that produces 50% of the maximum pharmacological effect; [k.sub.e0], transfer rate constant out of the effect compartment.
LC-MS/MS methods were developed for the determination of GA and PF in plasma. The HPLC system consisted of an Accela 1250 pump (Thermo Scientific Inc., MA, USA) connected to a Hypersil GOLD ODS column (150 mmx2.1 mm i.d., 5 [micro]m, Thermo Scientific, USA) with a C18 guard column (4 mmx3.0mm i.d., Phenomenex, Torrance, CA, USA). Isocratic chromatography was performed using a mobile phase of 5 mmol/1 ammonium acetate with 0.5% acetic acid-methanol, 60:40 (v/v) for PF, and 90:10 (v/v) for GA. The flow rate was 300 [micro]l/min and the column temperature was maintained at 30 [degrees]C. The injection volume was 20 [micro]l.
A TSQ Quantum Ultra EMR triple-quadrupole mass spectrometer (Thermo Scientific, San Jose, CA, USA) equipped with an electrospray ionization (ESI) source was used in the positive ion mode with selected reaction monitoring (SRM) for the quantitative analysis. The spray voltage was set to 3.6 kV, and the capillary temperature was maintained at 275 [degrees]C. Nitrogen was used as the sheath gas (30 Arb) and auxiliary gas (10 Arb) for nebulization. For collision-induced dissociation (CID), argon was used as the collision gas at a pressure of 1.2 mTorr. The collision energy was 20 eV for all of the compounds. Quantification was performed using SRM, monitoring the transitions m/z 471.26 [right arrow] 189.18 for glycyrrhetinic acid, m/z 295.28 [right arrow] 234.09 for tanshinone IIA, m/z 498.29 [right arrow] 179.04 for paeoniflorin, and m/z 406.12 [right arrow] 149.13 for gardenoside (Wen et al. 2012).
Calibration curves for GA and PF were prepared through the analysis of calibration samples prepared by spiking blank plasma or skin homogenate (50 p.1) with standard solutions (GA or PF in methanol, 50 [micro]l), and 50[micro]l of the internal standard solution (1.9[micro]g/ml tanshinone IIA or 24.89 [micro]g/ml gardenoside in methanol for the HPLC-UV method and 190.0 [micro]g/ml tanshinone IIA or 248.9 [micro]g/ml gardenoside in methanol for the LC-MS/MS method). Quality control (QC) samples were similarly prepared by spiking QC solutions from a different weighing of the reference substance. Each analytical run included a set of calibration samples, a duplicate set of QC samples, and unknown samples.
Pharmacokinetic parameters, such as the peak plasma drug concentration ([C.sub.max]) and time to peak plasma drug concentration ([t.sub.max]), were read visually from the concentration-time profiles. Other pharmacokinetic parameters, such as the area under the time-concentration curve from zero to a definite time t ([AUC.sub.0-t]), the area under the time-concentration curve from zero to infinite time ([AUC.sub.0-[infinity]]), the half-life of elimination ([t.sub.1/2]), and the mean residence time (MRT), were obtained through a non-compartmental analysis with the Kinetica program (Kinetica 5.0 fully functional trial version, Thermo Fisher Scientific Inc., MA, USA). The [AUC.sub.0-t] and [AUC.sub.0-[infinity]] were calculated by the linear trapezoidal rule.
PK/PD parameters estimation
Kinetica 5.0 was used to calculate the compartmental pharmacokinetic parameters and to build the PK/PD models. Both one- and two-compartment models were characterized, and the most appropriate pharmacokinetic models were determined using the Akaike information criterion (AIC). The PK/PD model with a separated effect compartment was adopted for the data analysis of the data. The effect compartment drug concentration ([C.sub.e]) was calculated using formula (4) (Shargel et al. 2006):
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)
where [D.sub.0] is the dose, [V.sub.e] is the volume of the effect compartment, [k.sub.1e] is the transfer rate constant for drug movement from the central compartment to the effect compartment, [k.sub.e0] is the transfer rate constant out of the effect compartment, and k is the elimination rate constant from the central compartment.
The S-[E.sub.max] model was determined as detailed in formula (5):
E = [E.sub.max] x [C.sub.e.sup.n] /([C.sub.e.sup.n] + [EC.sub.50]) (5)
where E is the pharmacological effect, [E.sub.max] is the calculated maximum effect, [EC.sub.50] is the drug concentration in the effect compartment yielding half of the maximal effect, and n is the sigmoidicity factor, an exponent describing the number of drug molecules that combine with each receptor molecule.
Results and discussion
Screening of formulation
GA-PF transdermal patches prepared in this study were matrix-type, and skin permeability usually governs the rate of drug delivery (Prausnitz et al. 2004). The in vitro permeation experiments with animal skins and diffusion cells were commonly employed to evaluate the patches preliminarily. Duro-Tak[R] 87-4098 or 87-2677 was selected and evaluated for the in vitro mouse skin permeation studies. The cumulative permeation profiles of GA and PF from the two different adhesives are shown in Fig. 2. The steady and continuous permeation profiles of GA and PF from all of the patches were observed, and the results indicated that the components were dissolved in all of the adhesives with a uniform distribution.
A previous study (Hao et al. 2010) proved that PF is able to penetrate through skin easily without any chemical enhancer, whereas GA is blocked by the skin. Therefore, an organic base and an enhancer were added to improve the permeation profile of GA. The cumulative permeation amount of GA from patch whose matrix was Duro-Tak[R] 87-4098 was 1.33-fold greater than that from patch whose matrix was Duro-Tak[R] 87-2677, and it was 2.2-fold greater the cumulative permeation amount of PF. According to the results of the in vitro permeation experiments, the different pressure-sensitive adhesives had only a slight effect on the cumulative permeation amount of GA, whereas the permeation of PF was inhibited by the formation of hydrogen bonds between the--COOH in Duro-Tak[R] 87-2677 and the -OH of the saccharide group in paeoniflorin (Kim et al. 2000) . Therefore, Duro-Tak[R] 87-4098 without any functional group was an optimal choice for GA-PF transdermal patches.
Optimization of GA-PF proportion
A wide variety of animal models have been developed to evaluate the analgesic effect of substances and formulations before clinical test. The writhing test of rodents was considered to have advantages than paw drawal test and tail flick test (Barrett 2015; Gregory et al. 2013). Thus the writhing behaviour evoked by spasm of uterine tissue in mice dysmenorrhea model was employed in the present research, since the PLD has showed effect on inhibiting smooth muscle contractions of human pregnant uterine tissue in vitro (Tsuji et al. 2012).
Patches with different proportions of GA and PF were prepared, and the corresponding proportions are shown in Table 1. The pharmacological effects in DM were evaluated in terms of the inhibition ratio of the writhing response and the prolongation of the antalgic latency to obtain the optimal proportion compatibility. Significant differences between the negative control group and the 15*5%, 10*30%, 5*15% and 0*15% (GA-PF) patches-treated groups were observed (Table 2). The patches with a proportion compatibility of 10*10% and 5%-15% proportion compatibility were selected and shared similar pharmacological effects (PIR = 75.0% and 73.7%, [DELTA]AL = 16.4 [+ or -]13.2 min and 16.5 [+ or -]11.3 min, respectively), which were higher than those of the meloxicam suspension, the positive control (PIR = 61.8%, [DELTA]AL = 9.5 [+ or -] 11.6 min). Although the 5*15% GA-PF patch also exerted a satisfactory analgesic effect, more GA made the mixture system unstable due to the poor solubility of GA. Therefore, the 10*10% GA-PF patch was selected as the optimal formulation. Both the positive control group and the GA-PF patch group exhibited prolonged antalgic latency, but no statistically significant difference was observed (p > 0.05), which might have resulted from the great individual differences. Increasing the number of experimental animals could help obtain a clearer result. Further investigation and subsequent discussion should be based on the PIR results reported in this manuscript.
The analgesic mechanisms of the two ingredients were different. GA has been proven to exert antispasmodic effects on tissues in vitro (Kaifuchi et al. 2015; Lee et al. 2013; Tsuji et al. 2012), and PF exerts its anti-inflammatory action by inhibiting the production of pro-inflammatory cytokines (Chang et al. 2009; Wu et al. 2012). Dysmenorrhea pain was caused by uterine muscle contractions, and in the present study, GA exhibited the main analgesic effect, whereas the combination of GA and PF exerted a synergism, since the effect of 10*10% GA-PF administration (PIR = 75%) and 15*5% GA-PF administration (PIR = 73.7%) were higher than their theoretical combination [E.sub.A+B] based on formula 3 (the expected PIR<61.1%, calculated on 15%0% GA-PF administration and 0*10% GA-PF administration).
Pharmacokinetics and topical distribution
An analysis of the pharmacokinetics and the distribution in skin and tissue under skin in mice was performed. The pharmacokinetics of GA and PF were both fitted to one-compartment models with lag time. The corresponding pharmacokinetic parameters are listed in Table 3, and the mean concentration-time curves of skin, tissue and plasma are shown in Fig. 3.
As shown in Fig. 3(a), for GA a lag time was observed and the plasma concentration achieved a [C.sub.max] of 336.1 ng/ml at 8 h, then maintained for almost 16 h. After the patches were removed, the plasma concentration level remained steady. Fig. 3(d) shows that the pharmacokinetic behaviour of PF was similar to that of GA in absorption phase ([C.sub.max] = 197.2 ng/ml, [t.sub.max] = 8 h, with a lag time), and the plasma concentration was maintained for approximately 4h before the elimination phase. However, after removing the patches, the plasma concentration of PF declined rapidly, an effect the might be attributable to the elimination of PF in systemic circulation because PF is relatively water-soluble (solubility in water> 3.0x[10.sup.6] [micro]g/ml). To the best of our knowledge there are no reports on the pharmacokinetics of a combination of GA and PF in mice, whereas the researches on rats have been reported, in which the [t.sub.1/2] of GA is 14.2 h (Sun et al. 2012), and the [t.sub.1/2] of PF is 3.3 h (Liu et al. 2013). Relative to the kinetics after oral administration, the time of action in rodents after transdermal administration was prolonged, whether GA ([t.sub.1/2] = 22.7 h after transdermal administration) or PF ([t.sub.1/2] = 15.4 h after transdermal administration). The GA-PF transdermal patches showed advantages over oral formulations.
Fig. 3 also showed that the level of GA in skin and plasma was higher than PF, while it was lower in muscle, which displayed the different distribution process of the two ingredients. After the patches were removed, GA skin concentration declined although it was still detected in the following 24 h, whereas the plasma and muscle concentration levels maintained steady. This finding indicates that GA was released from the skin to the plasma, suggesting a reservoir effect. In the contrary, PF concentration declined in muscle and plasma and maintained in skin after the patches removed. That was considered as a result of skin binding.
Fig. 4 illustrates the effect-time curve of PIR and the plasma concentration-time curves of GA. The maximum effect appeared at 24 h showing a lag compared with the peak concentrations ([t.sub.max] = 8 h). Because the antispasmodic effect of PLD on uterine tissue has been described as a direct action in vitro (Tsuji et al. 2012), an effect compartment model was used to explain the asynchrony between the plasma concentration and the pharmacological effect. The AIC value was 61.95 and the MSC value was 58.35. The concentrations of GA in the effect compartments calculated based on the model were synchronous with PIR (Fig. 4). The construction of a PK/PD model could contribute to the estimation and prediction of relevant parameters associated with the onset, magnitude and time courses of the dose-concentration-effect of a formulation (Perez-Urizar et al. 2000). In this case, the maximum effect would be observed at 24 h after dosing, and this effect would then be maintained for the following 48 h, which differs from the inferences obtained from the pharmacokinetic analysis. In the experiment, no obvious fluctuations were observed after removing the patches. The characteristics of the formulation suggest a continuous and stable analgesic effect.
Previous studies have proven that certain analgesics induce tolerance (Chu et al. 2012; Rezende et al. 2010; Tsiklauri et al. 2010), and patients who receive a continuous infusion rather than intermittent intramuscular bolus injections tend to develop more pronounced tolerance to the analgesic effects (Marshall et al. 1985). In a PK/PD study, a clockwise ECT hysteresis loop reflected drug tolerance which induced by receptor desensitization or the production of counter-regulatory substances (Shang et al. 2006). The pharmacological responses were plotted as a function of the plasma concentrations in Fig. 5, and the curves showed counterclockwise hysteresis loops, suggesting that no tolerance was induced by the GA-PF patches. Moreover, the counterclockwise hysteresis loops might contribute to a slow initial diffusion of the drug to the target organ (Table 4, the low Ke0), which may explain why the pharmacological effect selected in the analysis lagged.
In the present study, GA-PF transdermal patches for analgesic therapy were prepared and optimized, and its in vitro permeation, the synergism, the topical distribution and PK/PD characteristics were assessed. The results suggested that a synergistic effect of GA and PF in 10%GA-10%PF patch and 15%GA5%PF patch. Moreover, the classical effect compartment model could be applied to describe the PK/PD of the two ingredients, and the 10%GM0%PF patch could provide a stable and continuous analgesic effect lasts for at least 48 h in dysmenorrhea model mice with a single dose. Thus, the formulation might be suitable for topical analgesic and spasmolysis therapy.
Received 24 December 2015
Revised 15 April 2016
Accepted 16 May 2016
Conflicts of interest
The authors declare no conflict of interest.
The study has been financed by Liaoning province natural science funds (2014020015) and project of outstanding talent support program in universities of Liaoning province (LR2014002).
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.phymed.2016.05.005.
Afeltra, J., Vitale, R.G., Mouton, J.W., Verweij, P.E., 2004. Potent synergistic in vitro interaction between nonantimicrobial membrane-active compounds and itraconazole against clinical isolates of aspergillus fumigatus resistant to itraconazole. Antimicrob. Agents Ch. 48, 1335-1343.
Barrett, J.E., 2015. The pain of pain: challenges of animal behavior models. Eur. J. Pharmacol. 753, 183-190.
Cazzola, M., Calzetta, L., Page, C.P., Rogliani, P., Facciolo, F., Gavalda, A., Matera, M.G., 2014. Pharmacological characterization of the interaction between aclidinium bromide and formoterol fumarate on human isolated bronchi. Eur. J. Pharmacol. 745, 135-143.
Chang, Y., Wei, W., Zhang, L., Xu, H.M., 2009. Effects and mechanisms of total glucosides of paeony on synoviocytes activities in rat collagen-induced arthritis. J. Ethnopharmacol. 121, 43-48.
Chu, L.F., D'Arcy, N., Brady, C., Zamora, A.K., Young, C.A., Kim, J.E., Clark, J.D., 2012. Analgesic tolerance without demonstrable opioid-induced hyperalgesia: a double-blinded, randomized, placebo-controlled trial of sustained-release morphine for treatment of chronic nonradicular low-back pain. Pain 153, 1583-1592.
Gregory, N.S., Harris, A.L., Robinson, C.R., Dougherty, P.M., Fuchs, P.N., Sluka, K.A., 2013. An overview of animal models of pain: disease models and outcome measures. J. Pain 14, 1255-1269.
Hao, J., Sun, Y., Wang, Q, Tong, X., Zhang, H., Zhang, Q., 2010. Effect and mechanism of penetration enhancement of organic base and alcohol on glycyrrhetinic acid in vitro. Int. J. Pharm. 399, 102-108.
Kaifuchi, N., Omiya, Y., Kushida, H., Fukutake, M., Nishimura, H., Kase, Y., 2015. Effects of shakuyakukanzoto and its absorbed components on twitch contractions induced by physiological Ca2+ release in rat skeletal muscle. J. Nat. Med. 69, 287-295.
Kim, J., Cho, Y., Choi, H., 2000. Effect of vehicles and pressure sensitive adhesives on the permeation of tacrine across hairless mouse skin. Int. J. Pharm. 196, 105-113.
Lee, K.K., Omiya, Y., Yuzurihara, M., Kase, Y., Kobayashi, H., 2013. Antispasmodic effect of shakuyakukanzoto extract on experimental muscle cramps in vivo: role of the active constituents of Glycyrrhizae radix. J. Ethnopharmacol 145, 286-293.
Liu, H., Shan, J., Kang, A., Yun, F., Chen, L., Zhang, W., Di, L, 2013. Effects of glycyrrhizic add and glycyrrhetinic acid on in vivo pharmacokinetic parameters of paeoniflorin in rats. Chin. Tradit. Herbal Drugs 44, 1610-1614.
Marshall, H., Porteous, C., McMillan, I., MacPherson, S.G., Nimmo, W.S., 1985. Relief of pain by infusion of morphine after operation: does tolerance develop? Brit Med. J. 291, 19-21.
Perez-Urizar, J., Granados-Soto, V., Flores-Murrieta, F.J., Castaneda-Hernandez, G., 2000. Pharmacokinetic-Pharmacodynamic Modeling: Why? Arch. Med. Res. 31, 539-545.
Ploeger, B.A., Meulenbelt, J., Dejongh, J., 2000. Physiologically based pharmacokinetic modeling of glycyrrhizic acid, a compound subject to presystemic metabolism and enterohepatic cycling. Toxicol Appl. Pharmacol. 162, 177-188.
Prausnitz, M.R., Mitragotri, S., Langer, R., 2004. Current status and future potential of transdermal drug delivery. Nat. Rev. Drug Discov. 3, 115-124.
Raphael, T.J., Kuttan, G., 2003. Effect of naturally occurring triterpenoids glycyrrhizic acid, ursolic acid, oleanolic acid and nomilin on the immune system. Phytomedicine 10, 483-489.
Rezende, R.M., Paiva-Lima, P., Dos Reis, W.G., Camelo, V.M., Bakhle, Y.S., de Francischi, J.N., 2010. Celecoxib induces tolerance in a model of peripheral inflammatory pain in rats. Neuropharmacology 59, 551-557.
Shang, G.W., Liu, D.N., Yan, L.H., Cui, X.Y., Zhang, K.P., Qi, C., Chen, J., 2006. Nociceptive stimulus modality-related difference in pharmacokinetic-pharmacodynamic modeling of morphine in the rat. Pharmacol. Biochem. Behav. 85, 464-473.
Shargel, L., Sussana, Wu-Pong, Yu, A.B.C., 2006. Relationship between pharmacokinetics and pharmacodynamics. In: Applied biopharmaceutics and pharmacokinetics (Simplified Chinese translation edition). McGraw-Hill Education (Asia) Co. and Chemical Industry Press, Beijing, pp. 406-409.
Sun, Y., Yang, Y., Li, D., Pu, T., Xie, X., Wang, Q., 2012. effect of paeoniflorin on the pharmacokinetics of glycyrrhizic acid and glycyrrhetinic acid in rats. Chinese Pharm. J. 47, 599-603.
Tsiklauri, N., Viatchenko-Karpinski, V., Voitenko, N., Tsagareli, M.G., 2010. Non-opioid tolerance in juvenile and adult rats. Eur. J. Pharmacol. 629, 68-72.
Tsuji, S., Yasuda, K., Sumi, G., Cho, H., Tsuzuki, T., Okada, H., Kanzaki, H., 2012. Shakuyaku-kanzo-to inhibits smooth muscle contractions of human pregnant uterine tissue in vitro. J. Obstet. Gynaecol. Res. 38, 1004-1010.
Wang, Z., Nishioka, M., Kurosaki, Y., Nakayama, T., Kimura, T., 1998. Gastrointestinal absorption characteristics of glycyrrhizin from glycyrrhiza extract. Biol. Pharm. Bull. 18, 1238-1241.
Wen, J., Qiao, Y., Yang, J., Liu, X., Song, Y., Liu, Z., Li, F., 2012. UPLC-MS/MS determination of paeoniflorin, naringin, naringenin and glycyrrhetinic acid in rat plasma and its application to a pharmacokinetic study after oral administration of Si-Ni-San decoction. J. Pharm. Biomed. Anal. 66, 271-277.
Wu, H.X., Chen, J.Y., Wang, Q.T., Sun, W.Y., Liu, L.H., Zhang, L.L., Wei, W., 2012. Expression and function of beta-arrestin 2 stimulated by IL-lbeta in human fibroblast-like synoviocytes and the effect of paeoniflorin. Int. Immunopharmacol 12, 701-706.
Xu, C.H., Wang, P., Wang, Y., Yang, Y., Li, D.H., Li, H.F., Wu, X.Z., 2013. Pharmacokinetic comparisons of two different combinations of Shaoyao-Gancao Decoction in rats: competing mechanisms between paeoniflorin and glycyrrhetinic acid. J Ethnopharmacol 149, 443-452.
Xue Ding (a), Yuming Sun (b), Qing Wang (a,c) *, Tingting Pu (a), Xiaohui Li (d), Yaqing Pan (a), Yang Yang (a)
(a) School of Pharmaceutical Science and Technology, Dalian University of Technology, Dalian, China
(b) Chemical Analysis and Research Center, Dalian University of Technology, Dalian, China
(c) State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian, China
(d) School of Life Science and Biotechnology, Dalian University of Technology, Dalian, China
* Corresponding author at: School of Pharmaceutical Science and Technology. Dalian University of Technology, No.2 Linggong Road, Dalian, Liaoning 116023, China. Tel.: +86 (0)411-84986176; fax: +86 (0)411-84986175.
E-mail address: email@example.com (Q. Wang).
Table 1 Formulation screening design for the glycyrrhetinic acid (GA)-paeoniflorin (PF) patches. Reagents were added on mass percentage (w/w. %). GA PF 87-4098 87-2677 5 5 68.2 -- 5 5 -- 68.2 -- 10 66.4 -- 3 17 58.9 -- 5 15 58.2 -- 10 10 56.4 -- 15 5 54.6 -- 15 -- 59.6 -- -- -- 76.4 -- GA Butanol Triethanolamine Isopropyl myristate 5 10 1.8 10 5 10 1.8 10 -- 10 3.6 10 3 10 1.1 10 5 10 1.8 10 10 10 3.6 10 15 10 5.4 10 15 10 5.4 10 -- 10 3.6 10 Note: means not included; pressure sensitive adhesives 87-4098 and 87-2677 were weighed based on solid content. Table 2 Analgesic effects on dysmenorrhea model mice (n = 10) after administration meloxicam, blank matrix and glycyrrhetinic acid (GA) - paeoniflorin (PF) patches respectively, (mean [+ or -] S.D.). Writhing Group Prolongation of Numbers of Pain antalgic latency writhing inhibition (min) (in 30 min) ratio (%) Blank control 4.3 [+ or -] 1.7 22.0 [+ or -] 11.2 -- Negative 5.5 [+ or -] 2.0 22.8 [+ or -] 10.0 -- control Positive 9.5 [+ or -] 11.6 8.4 [+ or -] 6.1 * 61.8 control 0%GA-10%PF 9.9 [+ or -] 11.5 19.6 [+ or -] 17.9 13.5 3%GA-17% PF 11.9 [+ or -] 11.6 15.8 [+ or -] 13.9 30.1 5%GA-15% PF 5.6 [+ or -] 2.3 11.0 [+ or -] 4.6 * 50.0 10%GA-10%PF 16.4 [+ or -] 13.2 5.7 [+ or -] 7.8 ** 75.0 15%GA-5%PF 16.5 [+ or -] 11.3 6.0 [+ or -] 7.7 ** 73.7 15%GA-0%PF 7.8 [+ or -] 10.1 10.3 [+ or -] 8.7 * 55.0 Note: "*" means p < 0.05 vs. the negative control group, "**" means p < 0.01 vs. the negative control group. Table 3 Pharmacokinetic parameters of gly-cyrrhetinic acid (GA) and paeoniflorin (PF) in dysmenorrhea model mice (n = 10) after a single application of 4 [cm.sup.2] 10%GA10%PF patch. Parameters Values GA PF [C.sub.max] (ng/ml) 336.1 197.2 [t.sub.max] (h) 8.0 8.0 [t.sub.1/2] (h) 22.7 15.4 MRT (h) 39.6 29.1 [AUC.sub.0-t] (ng/ml*h) 10,601.2 3966.8 [AUC.sub.0-[infinity]] (ng/ml*h) 12,599.1 4254.5 Table 4 Pharmacokinetic/pharmacodynamic parameters of glycyrrhetinic acid from 10% glycyrrhetinic acid -10% paeoniflorin patch after single transdermal administration to dysmenorrhea model mice (n = 10). Parameters (unit) Values [E.sub.max] (%) 46,41 [EC.sub.50 ]([micro]g*[ml.sup.-1]) 0.0052 [K.sub.e0] ([h.sup.-1]) 1.75 x [l0.sup.-6] n 1.11
Please note: Some tables or figures were omitted from this article.