ZHAO Furong, GUO Ming,2*, SHAO Dongwei, XIA Qihan

(1. College of Forestry and Bio-technology, Zhejiang A & F University, Hangzhou 311300, China; 2. College of Science, Zhejiang A & F University, Hangzhou 311300, China; 3. School of Pharmacy, Jiamusi University, Jiamusi 154007, China; 4. Department of Mechanical and Power Engineering, University of North Denton, Denton 76201, USA)

Abstract: Matrine (MT) is an alkaloid widely used in the treatment of tumor diseases. It is the main medicinal ingredient in the dried roots of kuh-seng (Sophora flavescens Ait). However, there have been few studies on its transport mechanism. Serum albumin (SA) is the most abundant protein in blood. SA combines easily with many substances, including MT. MT and human serum albumin (HSA) were analyzed by capillary electrophoresis (CE) under in vitro conditions. The capillary tubing was 50 μm. The total length of the capillary was 60 cm, the total effective length was 50 cm. The interaction models of ligand-receptor binding were constructed by the mobility and frontal analysis (FA) methods. The purpose of establishing the interaction model was to study the binding of MT and SA. The phosphate buffer solution (PBS, 0.02 mol/L) was prepared in double distilled water. All solutions were prepared in PBS (0.02 mol/L). All solutions were filtered twice through a 0.45 μm microporous membrane, degassed for 5 min at a time. In the mobility method, different gradient MT solutions were used as running buffers. Their concentrations were 1.0×10-4-1.0×10-3mol/L, with the gradient of 1.0×10-4mol/L. And the HSA solution containing (0.5% (v/v)) acetone was used as test sample. Its concentration was 1.0×10-5mol/L. The nonlinear fitting method was used to obtain the binding parameters of MT and HSA. In the FA method, different gradient MT-HSA solutions were used as test samples. Their concentrations were 1.0×10-4-1.0×10-3mol/L, with the gradient of 1.0×10-4mol/L. And the PBS solution (0.02 mol/L) was used as running buffer. Then three equations were used to obtain the binding parameters of MT and HSA. And the applicability of the models was analyzed using the binding parameters. These three equations were nonlinear regression equation, Scatchard linear equation, and Klotz linear equation. Using the mobility method, the apparent binding constant KB was 8.072×103 mol/L. According to the FA method, three apparent binding constants were obtained for MT and HSA. The apparent binding constant KB of HSA and MT by nonlinear regression equation, Scatchard linear equation and Klotz linear equation were 1.434×103, 1.781×103 and 2.133×103mol/L. The comparison was as follows, KB(nonlinearregressionequation)r(nonlinearregressionequations)>r(Scatchardlinearequations). The results showed that both the methods were all suitable for analyzing the MT-SA system. The FA method could calculate the apparent binding constants and the numbers of binding sites. Therefore, it was more suitable for the analysis of MT and HSA. And the Klotz linear equation was the best fit for the theoretical model among the three equations. The combined parameters indicated that the interaction of MT with HSA had only one binding site. And the binding of MT with HSA was stable. This experimental method could be used to determine the binding status of MT and HSA. It is useful to further explore the binding mechanism of MT and HSA. This work provides valuable information on the interaction mechanism of typical alkaloids with SA. It will be useful in studies of the blood transport mechanisms of alkaloids.

Key words: capillary electrophoresis (CE); theoretical model; combined parameters; matrine (MT); serum albumin (SA)

Matrine (MT, C15H24N2O), an alkaloid [1-3], is the main medicinal ingredient in the dried roots of kuh-seng (SophoraflavescensAit) [4-6], the structural is shown in Fig. 1. MT has many pharmacological activities [7,8], including anti-viral [9], anti-tumor [10], anticancer [11], and anti-inflammatory activity [12], and has good therapeutic effects on the human digestive, central nervous, and cardiovascular systems[13-15]. Therefore, there has been much interest in the development of MT as a pharmaceutical in recent years [16,17]. Generally, MT reaches the target cell through blood transport. In the transport process, MT binds to the most abundant carrier protein in the plasma to achieve dynamic balance, which means that MT can be released to act at a particular location [18,19]. The biological characteristics of the interaction between MT and serum albumin (SA) affect the pharmacological actions of MT to a certain extent [20], and a study of the interaction between MT and SA is of great importance to elucidate the pharmacology, pharmacokinetics, and toxicity of MT [21,22]. Human serum albumin (HSA) is not only the main transport protein in the blood [23,24], but its tertiary structure, protein molecular mass, and solubility are well known, so it can be used as a model protein to study the interaction between drugs and proteins [25,26]. Therefore, a study of the interaction between MT and HSA to simulate the absorption of human drugs will be useful to understand the binding of drugs to HSA at the molecular level [27]. This work can provide valuable theoretical data for pharmaceutical research.

Fig. 1 Structural formula of MT

The aim of this study is to use effective analytical methods to determine the parameters of the interaction between MT and HSA. To date, a variety of technologies have been used to research the interactions between drugs and biomolecules at the molecular level [28-30], including spectral and chromatographic technologies. Spectral technology has the advantages of low cost and high sensitivity [31,32]. Chromatographic techniques can be used to study interactions, and also to separate systems to reduce interference from coexisting compounds [33]. Capillary electrophoresis (CE) not only has the advantages of both spectrophotometry and chromatography, but also has other advantages, such as high automation, high separation efficiency, simple sample preparation, and has the ability to maintain the activity of biological media in a simulated physiological environment [34-37]. Therefore, CE is often used to determine ligand-receptor binding constants. For example, Mozafari et al. [38] studied the interaction between SA and heparin at 296 and 310 K by affinity CE, indicating that CE could maintain the activity of biological media in a simulated physiological environment. Zhang et al. [39] used three different CE methods to study the interaction between loureirin B and HSA, indicating that diverse CE methods could be applied to the analysis of the interaction between small molecules and biomacromolecules. To our knowledge, there have been few reports on the interaction between MT and HSA using CE methods, and there have been rarely reported that use several theoretical equations to describe the interaction between MT and SA. It is of interest to study the interaction between MT and HSA to understand the binding and the transport mechanisms of MT.

Therefore, in this study, the mobility and frontal analysis (FA) methods were used to study the interaction between MT and HSA. Then, the corresponding binding constants were obtained, the results were analyzed in order to obtain the most suitable method, and the interaction between MT and HSA was investigated. The results could provide further mechanisms for studying the effects of MT and other drugs on the human body and cells.

1 Materials and methods

1.1 Apparatus and chemicals

All the CE experiments were carried out by using a P/ACETMMDQ capillary electrophoresis apparatus (Beckman-Coulter Company, USA) equipped with a UV detector. The capillary tubing was of uncoated fused silica with 50 μm inside and outside diameter and total effective length of 50 cm. The glass instruments were cleaned by KQ-250DB type numerical control ultrasonic cleaner from Kun Shan Ultrasonic Instruments Co., Ltd. (Kunshan, China). Distilled water came from 1810D automatic double pure water distiller from Shanghai Shensheng Technology Co., Ltd. (Shanghai, China). The samples were weighed by analytical balance from Sedolis Scientific Instruments (Beijing) Co., Ltd. (Beijing, China).

MT (≥98.0%), NaOH (≥96.0%), Na2HPO4·12H2O (≥99.0%), and NaH2PO4·2H2O (≥99.0%) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The HSA (purity 96%～99%) was purchased from Sigma (USA). Acetone was purchased from Hangzhou Shuanglin Chemical Reagent Factory. The experimental water was two times distilled. All chemicals were of analytical grade.

1.2 Standards and sample preparation

The phosphate buffer solution (PBS, 0.02 mol/L, pH 7.4) was prepared by appropriately diluting the NaH2PO4(0.2 mol/L, solution A) and Na2HPO4(0.2 mol/L, solution B), and the pH was adjusted with 0.1 mol/L NaOH. The MT stock solution (5×10-3mol/L) was obtained by dissolving 0.031 0 g MT powder in PBS in a 25 mL volumetric flask. A series of different concentrations of MT solution (1.0×10-4-1.0×10-3mol/L, increasing concentration gradient with 1.0×10-4mol/L) were prepared by the PBS. The 1.0×10-4mol/L and 1.0×10-5mol/L sample solution of HSA were prepared by the PBS, while 0.5% (v/v) acetone was added to 1.0×10-5mol/L sample solution of HSA as a neutral marker. The 1.0×10-4mol/L sample solution of HSA was mixed with a series of different concentrations of MT solution (1.0×10-4-1.0×10-3mol/L, increasing concentration gradient with 1.0×10-4mol/L), to get MT-HSA solutions with different concentrations.

All solutions were prepared in double distilled water, filtered with a 0.45 μm cellulose membrane filter and exhale twice for 5 min at a time.

1.3 Experimental conditions and methods

Before each experiment, at the beginning of the day, the equipment was sequentially rinsed for 5 min each with 0.1 mol/L NaOH, double distilled water and blank buffer, then 5 min with running buffer. The capillaries were conditioned for 5 min each between 0.1 mol/L NaOH, double distilled water and running buffer. The samples at each concentration were tested three times to obtain the average value.

In the mobility method, a sample of HSA (1.0×10-5mol/L) containing 0.5% acetone in buffer solution was injected into the capillary at a pressure of 3 447.4 Pa for 3 s. The electrophoresis was carried out in running buffer (a series of different concentrations of MT solution) for 13 min. The conditions used for the CE were as follows: voltage, 15 kV; temperature, 25 ℃; detection wavelength, 214 nm.

In the FA method, a series of different concentrations of MT-HSA solutions in buffer solution were injected into the capillary at a pressure of 3 447.4 Pa for 10 s. The electrophoresis was carried out in running buffer (PBS) for 12 min. The conditions used for the CE were as follows: voltage, 15 kV; temperature, 25 ℃; detection wavelength, 214 nm.

All experimental data were obtained through the nonlinear regression, Scatchard linear equation, and Klotz linear equation. All pictures in the article were drawn by origin software.

2 Results and discussion

The interaction between MT and HSA was elucidated using mobility and FA methods at the molecular level. Theoretical models were developed and the binding constants were calculated. In this experiment, MT and HSA were in a reversible equilibrium reaction as shown in equation (1), where P represents HSA, D represents MT, and DnP represents MT-HSA. The experimental process of CE included a binding reaction and dissociation equilibrium reaction of HSA and MT, and the separation of the components. The mass load ratio of free HSA, free MT, and MT-HSA, contributed to the different electrophoresis migration rates. Therefore, the mobility and FA methods could be used to separate the free HSA, free MT, and MT-HSA in the reversible equilibrium system. During the study of the interaction between MT and HSA in CE, a blank control contrast test was performed.

For the binding reaction system between MT and HSA, a reaction model describing the relationship between the electrophoresis mobility of each substance can be developed, which can be expressed as follows:

(1)

For the complex DnP, the apparent constant is

(2)

The average electrophoretic mobility (μ, cm2/(V·s)) values of the receptor P are between the electrophoretic mobility of the free receptor P (μP, cm2/(V·s), in the absence of ligand D and the electrophoretic mobility of the conjugate DnP (μDnP, cm2/(V·s)), so it can be expressed as:

(3)

Through a series of derivations, equation (4) can be obtained whenn=1, and the specific derivation process is as described previously [40].

(4)

Equation (4) is a Scatchard linear equation expressed by electrophoretic mobility, linearly fitting 1/[D] with 1/(μ-μP), and the ratio of the intercept and slope of the fitting curve is the apparent binding constantKB. However, there is no linear relationship betweenμand D in the actual fitting situation, thus the nonlinear fitting method is used to analyze the measured data, which can reduce the error in the calculated binding constant especially. It can be seen from equation (4) that different [D] values have differentμvalues. The migration time is obtained from the electropherogram. The equation for calculating the effective mobility can be referred to the literature [41].

The electropherograms of different concentrations of MT measured by this mobility method are shown in Fig. 2.

Fig. 2 Electropherograms of interaction between MT and HSA by the mobility method

In the blank sample test, only two positive peaks, one for acetone and the other for HSA, appeared under the electric field. In the sample injection test, the carrier electrolytes were the buffers with different concentrations of MT in the capillaries, the injected sample was the HSA buffer containing acetone, and the reversible reaction between MT and HSA occurred continuously in the CE. Therefore, a negative peak and two positive peaks were formed, the negative peak corresponded to the concentration of free MT in the sample that was lower than the concentration of MT in the carrier electrolyte buffer solution. As can be seen from Fig. 2, with increasing MT concentrations, the migration time of the acetone does not increase significantly, indicating that there is no obvious interaction between acetone and MT. The migration time of HSA does not change significantly, which indicates that the increase in MT concentration does not appreciably affect the migration behavior of HSA. It should be noted that the UV absorption peaks of HSA and MT-HSA show an increasing trend with the increasing of MT concentration, and the results can be calculated based on the peak height or peak area. In Fig. 2, the peak heights of HSA and MT-HSA does not changed significantly with the increasing of MT concentration. Because the ultraviolet absorption height of HSA was much larger than MT, and the amount of HSA-MT generated after the binding reaction was smaller, the absorption peak generated after the combination of HSA and MT was masked by the HSA. Therefore, the peak height observed at the experimental wavelength seemed to have little change, which the relative value of the increased peak height was smaller and the increased peak height was masked by a larger absolute peak height. It was found through CE experiments that the intensity of the UV absorption peaks for MT and HSA/MT-HSA showed an increasing trend with an increase in MT concentration, which indicated that MT and HSA had combined to form a complex.

The results are shown in Table 1. The migration times of HSA and acetone at different concentrations were obtained by changing the concentration of MT in the operating buffer, and the average effective mobility of HSA could be calculated.

Table 1 Effective mobilities of HSA in operating buffers with different concentrations of MT

According to the calculation method proposed by Gu et al. [41], the effective mobility of HSA is opposite to the direction of the electroosmotic flow. As can be seen from Table 1,μeffchanges from -1.904×10-4±0.012 50 cm2/(V·s) to -1.999×10-4±0.029 30 cm2/(V·s) with an increase in MT concentration, which indicate that there is an increasing interaction between MT and HSA with an increase in MT concentration. The horizontal axis is the concentration of MT, the vertical axis isμeff, the nonlinear fitting is carried out according to equation (4), and a nonlinear fitting graph is obtained (Fig. 3). As can be seen from Fig. 3, with the exception of the large deviation of some experimental data, most experimental data points are close to the fitting curve. By fitting, the apparent binding constantKBof MT-HSA was 8.072×103L/mol. The advantage of the mobility method was the low relative sample consumption, however, the number of binding sites could not be determined. Therefore, we should use other methods in CE to determine the number of binding sites between them.

Table 2 Binding parameters of MT and HSA by different equations

Fig. 3 Nonlinear fitting curve of the MT-HSA

In the FA method, the carrier electrolyte was the PBS buffer in the capillaries and the injected samples were the MT-HSA buffers. At the beginning of the injection, there were three compounds in the zone, free HSA, free MT, and MT-HSA, as the complex was the product of equilibrium. In the process of electrophoresis, the free MT and MT-HSA were always in the process of a continuous dynamic combination and dissolution equilibrium. Due to the different migration rates of different substances during electrophoresis, the UV absorption peaks of HSA showed a downward trend. The migration rates of free HSA and MT-HSA were greater than MT, so the area zone of the sample components changed from the initial superposition state to the separation state. The mass-to-charge ratios of HSA and MT-HSA were similar, so the peaks of the HSA and MT-HSA were not completely separated, but the free MT was completely separate, thus two independent positive peaks were detected. The results are shown in Fig. 4.

Fig. 4 Electropherograms of interaction between MT and HSA by FA method

It can be seen from Fig. 4 that two distinct UV absorption peaks are obtained in the FA method. These ultraviolet absorption peaks represent MT and HSA/MT-HSA. It should be noted that the UV absorption peaks of HSA show a decreasing trend with the increasing of MT concentration, and the results can be calculated based on the peak height or peak area. In Fig. 4, the peak heights of HSA do not change significantly with the increasing of MT concentration. Because the ultraviolet absorption level of HSA was much higher than MT, and the amount of HSA-MT generated after the binding reaction was lower, the absorption peak generated after HSA and MT combine was masked by the HSA. Therefore, the peak height observed at the experimental wavelength seemed to have little change. The reason was that the relative value of the decreased peak height was smaller and the decreased peak height was masked by a larger absolute peak height. Using the recorded experimental data on the peak area, we could see that with an increase in MT concentration, the intensity of the UV absorption peaks of MT increased gradually, while the intensities of the UV absorption peaks of HSA decreased, which indicated that MT and HSA had combined to form a complex. The peak time of UV absorption of MT was before that of HSA which was because of the reversible binding process between MT and HSA in the reaction system. Under the conditions of this experiment, because HSA and MT-HSA were negatively charged, the direction of their electrophoresis was opposite to that of the electroosmosis. Meanwhile the charge-mass ratio of MT was smaller than that of HSA/MT-HSA, so the migration rate of MT was greater than that of HSA/MT-HSA, and the peak time of UV absorption of MT before that of HSA/MT-HSA. Because the mass-to-charge ratios of HSA and MT-HSA complexes were similar, HSA and MT-HSA complexes were not completely separated, so only one peak appeared later.

The electropherograms of different concentrations of MT (1×10-4-1×10-3mol/L) were obtained under the same experimental conditions. The peak areas (A) were used to analyze the concentration (C, 10-5mol/L) by linear regression, and a good linear relationship curve was obtained. The linear regression equation wasA=11 873C-1 039.9, and the correlation coefficient (r) was 0.999 4.

It was assumed that the combination of the protein and MT was in an ideal state and the reactions were independent of each other. The equilibrium relationship can be expressed by the following multi-level equilibrium equation:

(5)

WhereRis the binding ratio,mis the total number of binding sites,niis the number ofi-type sites,Ki(L/mol) is the binding constant of thei-type site,Cb(mol/L) is the concentration of bound small molecules,Cf(mol/L) is the concentration of free small molecules, andCP(mol/L) is the total protein concentration.Cfhas a nonlinear relationship withR. If the small molecule has the same affinity for all binding sites (m=1), equation (5) can be transformed into a nonlinear equation ofCfandR:

(6)

Equation (6) can be reduced to a linear equation ofRandR/Cf:

(7)

In this Scatchard linear equation [42], the apparent binding constantKBand the binding sitencan be directly obtained from the slope and intercept of the fitting curve, respectively. Klotz and Hunston proposed the Klotz linear equation of interaction [43]:

(8)

It can be seen from Table 2 that the MT-HSA interaction model constructed using three different equations (nonlinear regression, Scatchard linear equation, and Klotz linear equation) can effectively reflect the effect of the system in the FA method. The apparent binding constants obtained by the three equations were different,KB(6)r(6)>r(7), it was shown that the Klotz linear regression equation was the optimal equation for the MT-HSA system. The number of calculated binding sites was similar, all of which were approximately 1.0, indicating that there was only a single binding site for MT with HSA.

To our knowledge, there have been few reports using CE to analyze the binding of MT to HSA, and even fewer reports using multiple CE methods to calculate the apparent binding constant and the numbers of binding sites. It is found that the two methods can be used for the analysis of the interaction between MT and HSA, the FA method is the more suitable for investigating the MT-HSA system.

3 Conclusion

In our study, the interaction between MT and HSA was analyzed by CE, which had the advantage of low sample consumption, while providing comprehensive information and accurate and rapid analysis of results. This experimental method can be used to determine the binding status of MT and HSA, which is useful to further explore the binding mechanism of MT and HSA. In addition, this method provides a useful basis for elucidating the molecular mechanisms of other similar alkaloid-protein binding systems.