Wu-tou decoction (Q) is composed ofAconiti Radix Cocta, Ephedrae Herba,Paeoniae Radix Alba,Astragali Radix and Glycyrrhiza Radix Preparata. It has been widely applied in the treatment of rheumatoid arthritis for thousands years in China. Published studies show that it could obviously decrease the acute exudative inflammation of mice induced by xylene,the foot swelling of rat induced by albumen and acetic acid-induced writhing number in mice. It could also reduce the peripheral blood CD⁴⁺ percentage and increase the peripheral blood CD⁸⁺ percentage in adjuvant arthritis rat ^{[1, 2]}. Pharmacology studies also show that Q could down-regulate the level of TNF-α,IL-1β,IL-17,TNFα,VEGF,PGE2 and inhibit the expression of COX-2 ^[3]. Studies on metabonomics revealed that Q might prevent the pathological process of adjuvant induced arthritis through regulating the disturbed metabolic pathways ^[4].

In the traditional Chinese medical system,Aconitum species are incompatible with Pinelliae Rhizoma and Fritillariae. But due to the multi-component,multi-target nature and synergistic therapeutic efficacies of traditionalChinesemedicines,the specificmechanismof herb-herb interactions is not clear as yet ^[5]. The possible mechanisms are as follows: chemical components are changed in the process of decoction; the absorption and metabolism are changed in vivo. Recently,most research focused on herb-herb interactions that could produce the poisonous ingredients or derive new compounds in vitro ^[6]. While little attention was paid to the variations aroused by incompatible herbs on the bioavailability of the formulae’s effective constituents in vivo.With regard to Aconitum carmichaeli Debx,the researchers concentrated on the facts that incompatible herb produce toxic compounds,such as the case of diester-diterpene alkaloids that can be increased when boiled together ^{[7, 8]}. In the clinical application,Aconiti Radix Cocta containing much less diester-diterpene alkaloids is more commonly used. So it is significant to study whether the incompatible herbs havenegativeimpact onthe therapeutic effects ofWu-tou decoction.

Caco-2 cell monolayers are a well-accepted model in vitro for the assessment of the intestinal absorption potential of drugs ^[9]. As we know,oral administration is a traditional mode of traditional Chinese medicine,thus oral absorption plays an important role in the pharmacological actions. Recently,the mechanisms of alkaloids in the intestinal absorption have been studied ^{[10, 11]}. But little research focused on the impacts of incompatible herbs on the absorption of alkaloids in formula. In this work,Caco-2 cells were used to investigate whether the incompatible herbs including Fritillariae Cirrhosae Bulbus,Fritillariae Thunbergii Bulbus and Pinelliae Rhizoma affected the permeability of alkaloids acting as the main effective components in the formula through the intestine after oral dosing in gastrointestinal tract,and thus whether the incompatible herbs could attenuate pharmacodynamics actions of Wu-tou decoction.

In this study,ultra performance liquid chromatography- electrospray ionization-mass spectrometry (UPLC-ESI-MS) was applied to study the permeability of alkaloids in Wu-tou decoction in Caco-2 cells with or without the incompatible herbs including Fritillariae Cirrhosae Bulbus,Fritillariae Thunbergii Bulbus and Pinelliae Rhizoma. By comparing the changes of permeability using different ratios in vitro,the effects of different incompatible herbs on the Wu-tou decoction were analyzed to illustrate the possible mechanisms. 2. Experimental

2.1. Materials and methods

Aconiti Radix Cocta,Ephedrae Herba,Paeoniae Radix Alba, Astragali Radix,Glycyrrhiza Radix Preparata,Fritillariae Cirrhosae Bulbus,Fritillariae Thunbergii Bulbus,and Pinelliae Rhizoma were purchased from Beijing Huamiao Chinese Medicine Engineering Development Center (Beijing,China). All herb medicines were identified by Prof. Shumin Wang (Changchun University of Chinese Medicine). Dulbecco’s modified Eagle medium (DMEM) with high glucose medium,penicillin,streptomycin,trypsin,DMSO,HEPES, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma-Aldrich,FBS (Thermo), acetonitrile,methyl alcohol and acetate were obtained from Fisher Scientific (Fair Lawn,NJ,USA). Water was prepared on a Milli-Q water purification system (Millipore). All other chemicals were of analytic grade or better. 2.2. UPLC-ESI-MS analysis

UPLC-ESI-MS system (Acquity-xevo Waters),chromatographic separations were performed on an ODS C18 column (50 mm×2.1 mm,1.7 μm,Waters Acquity UPLC BEH). The column temperature and the flow rate were 35 ℃ and 0.3 mL/ min,respectively. The mobile phase contained solvent A (5 mmol/L acetic acid/ammonia water (pH 10.5)) and solvent B (methanol/ acetonitrile (1:1,volume ratio)). The gradient elution program was as follows: 0-2 min,30%-50% (B),2-4 min,50%-52% (B),4-8 min, 52%-65% (B),8-9 min,65%-100% (B),9-12 min 100% (B).

The mass spectrometric analysis for the alkaloids in this study was as follows: positive ionization mode,multiple reaction monitoring (MRM),capillary voltage: 2.5 kV,desolvation temperature: 350 ℃,desolvation gas (nitrogen) flow: 700 L/h, cone gas flow: 50 L/h,collision gas (argon) flow: 0.15 mL/min, source temperature: 120 ℃. 2.3. Cell culture

Human colon adenocarcinoma-derived Caco-2 cell line was purchased from the cell bank of Chinese academy of sciences. Cells from passage 35-50 were routinely cultured for the permeability experiments. Caco-2 cells were seeded at a density of 1×10⁵ cells/ cm² on transwell polycarbonate insert filters (1.12 cm² of surface, 0.4 μm of pore size,12 mm of diameter,Corning Costar Corporation) in 12-well plates. After seeding,the Caco-2 cells were fed with 0.5 mL and 1.5 mL of culture medium in the apical and basolateral sides for 21-25 days and the medium was refreshed every other day for the first 2 weeks and every day for the rest time of the experiment. The monolayer cells with transepithelial electrical resistance (TEER),which exceeded 400 Ω cm²,were used for the permeability experiment. The integrity of Caco-2 monolayers was confirmed by the paracellular flux of lucifer yellow,which was <1% per hour. 2.4. Permeability studies and sample preparation

Powders of Aconiti Radix Cocta 2 g,Ephedrae Herba 3 g,Paeoniae Radix Alba 3 g,Astragali Radix 3 g and Glycyrrhizae Radix Preparata 3 g were immersed in deionized water,which was 10 times of the above total weight for 1 h and then heated to reflux for 1.5 h. After filtering,the residue of the decoction was heated to reflux for another 1.5 h in eight times of total weight deionized water. The filtered extraction was then blended and concentrated to 0.5 g crude drug/mL. After Alcohol precipitation,the solutions were centrifuged and the precipitates were removed. The supernatant was concentrated and freeze-dried into powders. Crude powders of Fritillariae Cirrhosae Bulbus (C),Fritillariae Thunbergii Bulbus (Z) and Pinelliae Rhizoma (X) were mixed with Wu-tou decoction powders in the ratio of 2:1,1:1 and 1:2,respectively. Then the mixtures were refluxed and processed with the same method mentioned above. Then the powders were dissolved with HBSS (pH 7.4) to the required concentrations. The monolayers were washed twice with HBSS and then incubated for 30 min with HBSS in a CO2 incubator at 37 ℃. For the absorption study,500 μL of samples was added to the apical side (AP) and 1.5 mL of HBSS was added to the basolateral side (BL). Then,100 mL of solution from the basolateral side were withdrawn at the time of 20,40,60,90,120 min, respectively and individually replaced with the same volume of fresh pre-warmed corresponding buffer. For the efflux study, samples were added to the basolateral side,which acted as the donating side. The apical side filled with 1.5 mL HBSS worked as the acceptor compartment. The procedures were opposite with the above mentioned absorption study. The solution collected in this study was dried to power and again,dissolved with methanol/ acetonitrile (1:1,volume ratio). Reserpine was added to each sample as an internal standard (IS). 2.5. Statistical analysis

The apparent permeability coefficient (P_app) was calculated according to Eq. (1):

The mean of at least three experiments and their standard deviation (SD) were used to express the values of P_app,i.e.,the data were expressed as mean±SD. Statistical analysis was performed with one-way analysis of variance (ANOVA) followed by Dunnet’s multiple comparison tests as a post hoc analysis. The level of significance test was p < 0.05. 2.6. MTT cytotoxicity assay

MTT colorimetric assay is widely used to determine cell growth and cell cytotoxicity ^[12]. The cells were harvested and seeded at a density of 1×10⁴ cells per well in 96-well plates and incubated for 48 h at 37 ℃. The drugs with different concentrations dissolved by DMEM were added into 96-well plates after culture medium removed,in which,five wells for one concentration. After 2.5 h, before adding 100 μL of MTT (1 mg/mL) into each well,the solution was removed. Four hours later,100 μL of DMSO was added to dissolve the formazan crystals and the absorbance was measured at 570 nm by a microplate reader (TECAN GENios). Cell viability (%) was calculated from Eq. (3)

In this research,the compatibility of Wu-tou decoction individually with Fritillariae Cirrhosae Bulbus (named as QC), Fritillariae Thunbergii Bulbus (named as QZ) and Pinelliae Rhizoma (named as QX) was observed. In the case of Fig. 1,the ratios of concentrations for QC,QZ and QX were all equal to 1:2. This means that in the three cases above the concentrations ofWu-toudecoction were the same. According to the results in Fig. 1,when the concentrations of the three componentswere less than 25mg/mL,a high cell viability,typically >98%,was observed and indicated that the drug was non-toxic to the cells. As shown in Fig. 1,the cell viability gradually declined as the concentration increased.

Fig. 1

	Download: JPG larger image
Fig 1. Cytotoxicity results of drugs at different concentrations in Caco-2 cells using the MTT assay. QC 12 (Wu-tou decoction: Fritillariae Cirrhosae Bulbus = 1:2),QZ 12 (Wu-tou decoction: Fritillariae Thunbergii Bulbus = 1:2),QX 12 (Wu-tou decoction: Pinelliae Rhizoma = 1:2).

According to the reported paper ^[13],the alkaloids,benzoylaconitine (BA),benzoylmesaconitine (BM),benzoylhypaconitine (BH) could decrease the rate of primary or secondary foot swelling and reduce the blood sedimentationaswell as the number ofwhite blood cells in the rheumatoid arthritis rats. In addition,they could decrease the amount of MDA and raise the activity of SOD. Monoester alkaloids bind more strongly to cyclooxygenase-2 (COX-2) than diester-diterpene alkaloids revealed by centrifugal ultrafiltration mass spectrometry. These results suggest monoester alkaloids had better anti-inflammatory activity. In the diester-diterpene alkaloids, hypaconitine (HA) had the strongest inhibitory activity of COX-2 ^[14]. Recent reports indicated that appropriate concentration of HA could apparently decrease the apoptosis of myocardial cells caused byH₂O₂ throughreducing the expressionof apoptosisunilateral ^[15]. Fuziline (FU) and niaolin (NE) were found to have the potential to repair the myocardial injury and resist the hemorrhagic shock ^{[16, 17]}. FU could dilate blood vessels,lower the blood pressure and exert the analgesic effect ^[18]. Also,NE had the role of analgesia and local anesthesia ^[19] and the analgesia effect of deoxyaconitine (DA) was observed ^[20]. Thus these compounds were considered the medicinal effective ingredients. In this research,compounds mentioned above were taken as the main alkaloid constituents in Wu-tou decoction for quantitative study by multiple reaction monitoring (MRM),and at the same time,reserpine was added in the formula as an internal standard. The UPLC-MS/MS spectrum of the Wu-tou decoction is shown in Fig. 2. MS/MS detection was performed in the multiple-reaction monitoring (MRM) mode.

Fig. 2

	Download: JPG larger image
Fig 2. MRM total ion chromatogram of the alkaloids in Wu-tou decoction.

In the Caco-2 cell monolayer,apparent permeability coefficient (P_app) was related to the absorption in the intestine. P_app(AP→BL) and P_app(BL→AP) values of drugs reflect the absorption and excretion in the intestine,respectively. As shown in Table S1 in Supporting information,in the cases of Fritillariae Cirrhosae Bulbus, only the absorptions of BA,HA,DA and NE decreased in QC 21 compared to Q. And P_app(AP→BL) values of all the alkaloids increased in QC 11 and QC 12 compared to Q,which indicated that the absorption of the alkaloids in Wu-tou decoction increased in the presence of Fritillariae Cirrhosae Bulbus. But the augment of monoester alkaloids in QC 11 and QC 12 were quite close. QC 11 induced the greatest increase in diester-diterpene alkaloids and alkylolamine diterpenoid alkaloids among the cases of QC. In the BL-AP direction,the reductions of the P_app(BL→AP) values of HA, DA,NE in QC 21 and QC 12 led to less efflux. While in the QC 11,the P_app(BL→AP) values of all the seven alkaloids increased comparing to Q.

It is necessary to observe the efflux ratio variations of a drug for its absorption and excretion occurring simultaneously. Low efflux ratios would result in high bioavailability. The results in Table S1 reveal that the bioavailability of diester-diterpene alkaloids increased along with quantity of Fritillariae Cirrhosae Bulbus. At the same time,Fritillariae Cirrhosae Bulbus decreased the efflux ratio of FU in all cases and thus enhanced the bioavailability of FU in the formula. In the cases,QC 21 and QC 11,the efflux ratios suggest that Fritillariae Cirrhosae Bulbus would decrease the bioavailability of monoester alkaloids and NE in Wu-tou decoction. While in QC 12 containing most Fritillariae Cirrhosae Bulbus,the efflux ratio was decreased leading to the increase of bioavailability of the seven alkaloids compared to Q. It means that Fritillariae Cirrhosae Bulbus might further improve the therapeutic effects. 3.4. Transport of the alkaloids in Wu-tou decoction with different ratios of Fritillariae Thunbergii Bulbus

As shown in Table S2 in Supporting information,in the case ofQZ 21,the P_app(AP→BL) values of HA,DA and NE individually reduced, therefore their absorptions were lower than Q. Moreover,the absorption of the monoester alkaloids was higher than Q because of the improvement of the P_app(AP→BL) values. The alterations of alkaloids in QZ 11 were quite different. The P_app(AP→BL) values of BA,HA and NE in QZ 11 were almost the same as Q. The absorptions of other alkaloids in QZ 11 increased. That is to say,Fritillariae Thunbergii Bulbus could raise the absorption of BH,BM,FU and DA. In the case of QZ 12,the P_app(AP→BL) values of alkaloid BA were not changed and the other five alkaloids increased except NE which decreased instead. Fritillariae Thunbergii Bulbus increased the absorptions of most of the alkaloids in the formula. In the opposite direction,the P_app(BL→AP) values ofHAandDAboth wentdownin QZ 21,QZ 11 and QZ 12,respectively,hence the excretion of the diester-diterpene alkaloids in the Wu-tou decoction obviously decreased because of the addition of Fritillariae Thunbergii Bulbus. Although the P_app(BL→AP) value of BH increased in QZ,the values gradually decreased as the dose of Fritillariae Thunbergii Bulbus increased. It is considered that the excretion of BH decreased with the increase of Fritillariae Thunbergii Bulbus.

The efflux ratio in Table S2 showed that the bioavailability of BH and BA in QZ 21 increased but the bioavailability of BM and NE reduced compared to Q. the bioavailability of all the alkaloids except BA in QZ 11 was improved. In case of QZ 12,the bioavailability of alkaloids except NE was improved. But the increased bioavailability in above two cases was not equal to the case in QC 12. It has been shown that most alkaliods of Aconitum carmichaeli Debx were the substrates of P-glycoprotein (P-gp) ^{[10, 11]}. Thus,we presumed that peimine and peiminine in fritillary as the inhibitors of P-gp ^{[21, 22]} might account for the higher bioavailability of alkaloids above. 3.5. Transport of the alkaloids in Wu-tou decoction with different ratios of Pinelliae Rhizoma

In QX 21,the P_app(AP→BL) values of FU and DA in Wu-tou decoction increased,leading to the absorption increases of FU and DA. But the P_app(AP→BL) values of monoester alkaloids and DA slightly increased in QX 11. And for absorption,in QX 12,the values of BH,BM and FU slightly increased. After adding Pinelliae Rhizoma in the Wu-tou decoction,the P_app(AP→BL) values of HA and NE in three ratios (see Table S3 in Supporting information) were lower than Q. That is to say,the absorption of HA and NE decreased in these cases. On the whole,the variations of the alkaloids in QX were not apparent in the AP to BL direction as QC or QZ. In the BL→AP direction,the P_app(BL→AP) values of BA and FU in QX 21 decreased compared to Q. In the above seven alkaloids,the efflux of HA in QX 11 and BA in QX 12 did not alter compared to Wu-tou decoction,but others increased. These results in the direction of BL to AP demonstrated that Pinelliae Rhizoma caused the excretion increases of the most alkaloids.

The efflux ratio in Table S3 shows that only the bioavailability of BA,FU and DA in QX 21 and BM in QX 12 increased. In short,the addition of Pinelliae Rhizoma decreased the bioavailability of the alkaloids in Wu-tou decoction,which illustrated that the efficacious components in Wu-tou decoction were less available. The possible reason for the phenomenon could be that Pinelliae Rhizoma plays a role on the transport proteins such as ATPbinding cassette transporters in the intestine. Further research is needed to explain the mechanism. 4. Conclusion

The present study provides the basis for the study on the mechanism of the eighteen incompatible medications in Chinese herbs thorough Caco-2 human intestinal cell monolayers using UPLC-ESI-MS. The aim of this work was to clarify how the incompatible herbs impact the bioavailability of active ingredients in the intestine and then affected the therapeutic effects of Wu-tou decoction. The results demonstrate that when Pinelliae Rhizoma was added into Wu-tou decotion,the main influence was to decrease the absorption of the most alkaloids and increase the excretion of all the alkaloids in Wu-tou decoction. In the codecoction of Fritillariae Thunbergii Bulbus and Wu-tou decoction,as exiting Fritillariae Thunbergii Bulbus,the bioavailability of the three monoester alkaloids being the main effective components in Wutou decoction increased. Adding Fritillariae Cirrhosae Bulbus to Wutou decoction could enhance the bioavailability of HA,DA and NE. When the ratio was 1:2,the bioavailability of all seven medicinal ingredients was improved in the presence of Fritillariae Cirrhosae Bulbus. The comparison of QC and QZ shows that although both could improve the active ingredients in Wu-tou decoction,QC improved the bioavailability of alkaloids more efficiently than QZ.

In summary,Pinelliae Rhizoma is an incompatible herb that reduces the therapeutic effects of Wu-tou decoction. All three ratios of Wu-tou decoction and Fritillariae Cirrhosae Bulbus could increase the absorption of all the alkaloids in Wu-tou decoction and especially the QC 12,which could increase the pharmacodynamic action of Wu-tou decoction. Fritillariae Thunbergii Bulbus impacted some alkaloids but its effects on the bioavailability of alkaloids were inferior to that of Fritillariae Cirrhosae Bulbus.

This work was supported by the National Natural Sciences Foundation of China (No. 81274046) and National Basic Research Program of China (‘‘973 Program’’) (Nos. 2011CB505300, 2011CB505305).

Supplementary data associated with this article can be found,in the online version,at http://dx.doi.org/10.1016/j.cclet.2014.12.008.