Reliable analysis of the interaction between ... - ACS Publications

Subscriber access provided by Kaohsiung Medical University

Article

Reliable analysis of the interaction between specific ligands and immobilized beta2-adrenoceptor by adsorption energy distribution Qian Li, Xiaohui Ning, Yuxin An, Brett J. Stanley, Yuan Liang, Jing Wang, Kaizhu Zeng, Fuhuan Fei, Ting Liu, Huanmei Sun, Jiajun Liu, Xinfeng Zhao, and Xiaohui Zheng Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b00214 • Publication Date (Web): 08 Jun 2018 Downloaded from http://pubs.acs.org on June 9, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 41

Reliable analysis of the interaction between specific ligands and immobilized

beta2-adrenoceptor

by

adsorption

energy

distribution

Qian Li,† Xiaohui Ning,‡ Yuxin An,§|| Brett J Stanley,¶ Yuan Liang,† Jing Wang,† Kaizhu Zeng,† Fuhuan Fei,† Ting Liu,† Huanmei Sun,† Jiajun Liu,† Xinfeng Zhao,†* and Xiaohui Zheng†

†

Key Laboratory of Resource Biology and Biotechnology in Western China, Ministry

of Education, College of Life Sciences, Northwest University, Xi’an 710069, China ‡

Institute of Analytical Science, Northwest University, Xi’an 710069, China

§

National Chromatographic R. & A. Center, Key Laboratory of Separation Science

for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China || ¶

University of Chinese Academy of Sciences, Beijing 100049, China Department of Chemistry & Biochemistry, California State University, San

Bernardino, CA 92407-2397, USA

*Corresponding author: Xinfeng Zhao Fax: +86-29-88302686 Tel.: +86-29-88302686 E-mail: [email protected] (X. Zhao)


Page 3 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract Although a comparatively robust method, immobilized protein-based techniques have displayed limited precision and inconsistent results due to a lack of strategy for the accurate selection of drug adsorption models on the protein surface. We generated the adsorption data of three drugs on immobilized beta2-adrenoceptor (β2-AR) by frontal affinity chromatography-mass spectrometry (FAC-MS) and site-specific competitive FAC-MS. Using adsorption energy distribution (AED) calculations; we achieved the best adsorption models for the binding of salbutamol, terbutaline and pseudoephedrine to immobilized β2-AR. The Langmuir model proved to be desirable for describing the adsorptions of salbutamol and terbutaline on immobilized β2-AR; while the bi-Langmuir model was favorable to characterize the adsorption of pseudoephedrine on the receptor. Relying on the accurate determination of association constants, we presented an efficient approach for β2-AR ligand screening based on the loss of breakthrough time of an indicator drug caused by the inclusion of competitive drugs in the mobile phase. We concluded that the current strategy enables the reliable and accurate analysis of G protein-coupled receptor (GPCR)-drug interaction. The percentage change in the breakthrough time for drugs can provide useful information for estimating their binding affinity to the receptor. This approach builds a powerful platform for high-throughput ligand screening.

Keywords:

Receptor-drug

interaction;

ligand

screening;

adsorption

energy

distribution; beta2-adrenoceptor; frontal affinity chromatography-mass spectrometry



Page 4 of 41

G protein-coupled receptors (GPCRs) have been extensively investigated for their unique roles in drug discovery and the regulation of known physiological functions.1-2 Considerable effort has been devoted to creating methodologies for GPCR-drug interaction analysis to search for new drug leads.3-4 Existing methods are typically classified into the following categories: biochemical, biophysical and theoretical, genetic, and computational.5-9 Each approach displays its own strengths and weaknesses, especially with respect to its precision, sensitivity and specificity. Improvements in these methodologies can greatly facilitate the precise and high-throughput analysis of GPCR-drug interaction. High-performance

affinity

chromatography

(HPAC)

based

on

diverse

immobilized receptors provides a powerful platform for GPCR-drug interaction analysis.10-11 Successful uses of this methodology depend on both experimental determination and theoretical data processing models. In typical cases, zonal elution and frontal analysis are ideal mathematical models for experimental design and GPCR-drug binding analysis.12-13 The mathematical models behind the two methods are based on linearization of the Langmuir equations.14 To address this issue, Guiochon and co-workers have introduced an approach to estimate protein-ligand interactions by the nonlinear regression of rival adsorption models. The best model is judged by statistical evaluations using F-test.14 The limitation of such an approach is the need to select a certain model to describe a certain adsorption mechanism using statistics (i.e., fitting). As an advanced numerical tool, adsorption energy distribution (AED) has been


Page 5 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


proposed as an essential and important model to determine the degree of heterogeneity in interactions.15-23 This approach has been successfully applied to experimental protein-ligand data produced by surface plasmon resonance16 and quartz crystal microbalance technologies.21-23 These applications have demonstrated the advantages of this strategy for acquiring real information about the heterogeneity of protein-ligand interactions, which facilitates the selection of the correct model to describe the adsorption mechanism. Because of these advantages, we hypothesized that the extension of the AED approach to frontal affinity chromatography-mass spectrometry (FAC-MS) is a powerful tool for the precise and high-throughput clarification of GPCR-drug interactions, and even the determination of the binding parameters using a drug mixture. Beta2-adrenoceptor (β2-AR) is a main therapeutic target of drugs to treat diseases of the respiratory system.24 This work applied systematic AED calculations to experimental data generated by direct FAC-MS and site-specific competitive FAC-MS studies with the immobilized receptor as the stationary phase. We found that the accuracy and precision of the method are much greater than those of classical chromatographic methods such as zonal elution and frontal analysis. The percentage change in the breakthrough time of the drugs can provide useful information for estimating their binding affinity to the receptor. We expect this method to improve the speed, accuracy, and precision of GPCR-drug interaction analysis.

■ THEORETICAL SECTION



Frontal Affinity Chromatography. As a widespread method for determining single-component isotherms, FAC involves step-wise increases in the concentrations of ligands that percolate through a column to form various breakthrough curves. If the immobilized protein has only one type of adsorption site, the interaction is described by the double-reciprocal form of the Langmuir isotherm equation (eq. (1)):25-26 1 1 1 = + (1) ∗ ∙ ∙ []

where q* is the adsorption amount at equilibrium at ligand concentration [C] and can be calculated through supplementary eq. (1). Ka is the association constant and mL is the number of binding sites on the column. Calculation of the Adsorption Energy Distributions. The detailed isotherm models27-30 are described in supporting information (Suppl. eq. (2-10)). Taking the Langmuir isotherm as the local model, we derive eq. (2):28

∗ ([])

= ()

⁄ [] (2) 1 + ⁄ []

The distribution function F(ɛ) is calculated by the expectation maximization (EM) method which is directly applied to the raw data.31-34 This method allows a robust AED estimate, minimizes artifactual information from a numerical standpoint and provides the least biased solution available. F(ɛ) is discretized using N grid points in the energy space between ɛmin and ɛmax. The amount q([Cj]) at concentration [Cj] is iteratively determined by the following equation: ()*

#!" $[% ]& = ' # (( ) ∙ ()+

,⁄ [% ] ∙ ∆ j ∈ [1, 1]; 3 ∈ [1, 4] (3) 1 + , ⁄ [% ]


Page 6 of 41

Page 7 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


where △ɛ and ɛi are defined in supplementary eq. (11) and (12). Then the AED function at the kth iteration can be updated as eq. (4): #7+ (( )

=

# (( )

%)8

' %)+

, ⁄ [% ] 9:; $[% ]& ∙ ∆ # (4) 1 + , ⁄ [% ] !" $[% ]&

Calculation of band profiles. The equilibrium-dispersive (ED) model is usually used to calculate the overloaded band profiles.35 The mass balance equation for a single component in this model is expressed as eq (5):35 =

>D > > > + +A − C! D = 0 (5) >? >? >@ >@

where u, t and z are the mobile phase velocity, the time and the axial position in the column, respectively; C and q are the sample concentrations in the mobile and stationary phases; f represents the phase ratio calculated by f=(1- ɛ)/ɛ , where ɛ is the total column porosity; and Da is the apparent dispersion term expressed as: C! =

=G (6) 24

where L is the column length and N is the number of theoretical plates. ■ EXPERIMENTAL SECTION Materials and Reagents. Standard solutions of salbutamol (batch No. 100328-200703), terbutaline (batch No. 100273-201202) and pseudoephedrine (batch No. 171237-200505) were obtained from the National Institutes for Food and Drug Control (Beijing, China). Amino polystyrene microspheres (8 µm, 300 Å) were purchased from Wuxi Knowledge & Benefit Sphere Tech. Co., Ltd. (Wuxi, China). HPLC-grade methanol was purchased from Fisher Scientific. Other chemical reagents were analytically pure unless stated otherwise. Apparatus. HPAC/MS/MS analyses were performed using an Agilent 1100



Page 8 of 41

LC-MSD system (Agilent Technologies, Waldbronn, Germany), which consists of a binary pump, a column thermostat, a diode-array detector, and an ion trap MS with an electrospray ionization (ESI) MS interface. The make-up organic solvent was handled by an Agilent 1260 Infinity isocratic pump. Data acquisition and processing were carried out using Chemstation 4.2 software. Chromatographic System. The column was prepared by attaching purified β2-AR to

N,

N’-carbonyldiimidazole-activated

polystyrene

amino

microspheres.36

Ammonium acetate (30 mM, pH 7.4) was used as the mobile phase at a flow rate of 200 µL/min. The eluents from the column were subjected to ESI-MSD. To enhance the ionization intensity of the target analyte, we designed a post-column make-up strategy (Figure S1) by the addition of methanol at 400 µL/min. Other MS conditions included: 10 L/min of drying gas (nitrogen) at 350 °C; 50 psi nebulizing gas (nitrogen); +3.5 kV capillary voltage; 100 - 1000 m/z mass scan range. Procedures. Frontal Analysis. Series concentrations of salbutamol, terbutaline and pseudoephedrine prepared by ammonium acetate buffer (30 mM, pH 7.4) were separately flushed through the β2-AR column to generate breakthrough curves. Multiple reaction monitoring (MRM) mode was applied to record the ion signals in positive ion mode. The three drugs produced ion patterns of m/z 240.0[M+H]+ 221.8[M+H-H2O]+, 226.2[M+H]+ 141.5[M+H-NHC(CH3)3]+ and 166.8[M+H]+ 148.8[M+H-H2O]+. The drug mobile phase concentrations were selected based on their therapeutic and monolayer adsorption ranges37: 0.015, 0.03, 0.06, 0.12, 0.24, 0.30, 0.45, 0.60, 0.90, 1.20, 1.50, 3.00, 4.50, 9.00 and 12.00 µmol/L for salbutamol;


Page 9 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.05, 0.10, 0.20, 0.40, 0.80, 1.00, 2.00, 3.00, 4.00, 5.00, 10.00, 25.00 and 50.00 µmol/L for terbutaline; and 0.01, 0.02, 0.40, 0.80, 1.60, 4.00, 10.00, 60.00, 160.00, 800.00 and 1600.00 µmol/L for pseudoephedrine. Site-specific competitive studies were performed by including fixed concentrations of terbutaline (0.05 µmol/L), pseudoephedrine (0.05 µmol/L) and their mixture (0.05 µmol/L for each drug) as competitors in the mobile phase. We recorded the breakthrough curves of the analyte (salbutamol) when the three independent competitors were applied to the mobile phase. The other analysis conditions remained identical to those of the direct FAC-MS study. To accurately obtain the inflection point, we treated the raw data of each extracted ion breakthrough curve using a third-degree polynomial equation (y = ax3 + bx2 + cx + d). The breakthrough time was determined by the second derivative. Procedures. Expectation Maximization. AED calculations were performed by EM using MATLAB (R2012a). To digitize the energy distribution of salbutamol, we assigned 200 grid points in the energy space between bmin = 8.33 mM-1 and bmax = 66667 mM-1. We changed the parameters bmin = 0.1/Cmax and bmax = 1/Cmin to 2-20000 and 0.625-100000 mM-1 when terbutaline and pseudoephedrine were analyzed. Here, Cmax and Cmin represented the highest and lowest drug concentrations. We ran the EM algorithm with 105 to 108 iterations. To obtain a qcal value closer to qexp, we continued the iteration by re-calculating eq. (3) with the new assumption (expectation) and correcting the distribution by eq. (4) (maximization). Procedures. Modeling of band profiles. To record the overloaded profiles of



salbutamol, terbutaline and pseudoephedrine, 100 µL of each drug (50 mM) was injected with duration of 50 s. The absorbance data from the detector was transformed into a concentration unit using a calibration curve based on the concentration plateaus of the frontal analysis. The column hold-up volume V0 was determined to be 0.36 mL. The phase ratio was calculated to be 1.307 through the relationship of f=(Vtotal- V0)/V0, where Vtotal is the total volume of the empty column. The initial concentration of the solute and adsorbate in the column is zero. The boundary conditions used are the classical Danckwerts type one at the inlet and outlet of the column.14 The calculations were realized by MATLAB (R2012a) using the best-fit isotherm parameters of each drug. Procedures. Application of FAC-MS in receptor ligand screening. Paeoniflorin (0.05 µmol/L), liquiritin (0.05 µmol/L) and a mixture of paeoniflorin, liquiritin, terbutaline, pseudoephedrine, propranolol and carazolol (0.05 µmol/L for each drug) were used as competitors. The percentage change in the breakthrough time of the indicator (salbutamol, 2.0 µmol/L) was determined to assess the affinity of those competitors to immobilized β2-AR.

■ RESULTS AND DISCUSSION Characterization of the Affinity Column. The β2-AR column (Figure S2) was characterized by the retention behaviors of sodium nitrite, terazosin (a specific ligand of α1-AR) and specific ligands of β2-AR (salbutamol and terbutaline) (Figure S3). The diverse retention times confirmed the recognition specificity of the immobilized


Page 10 of 41

Page 11 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


receptor. In comparison with the reported immobilized β2-AR,38 the utilization of amino polystyrene microspheres demonstrated longer retention times, reduced peak width, and decreased broadening for the same ligands. We interpreted that the stationary phase consisting of polymer spheres has higher biocompatibility than the silica gel. Adsorption isotherm model selection. Frontal analysis is canonical for collecting adsorption isotherm data.39 It requires various concentrations of drugs to saturate the column. An insufficient concentration range can cause biased values of binding parameters. The other concern about frontal analysis is the assumption of a single-site binding model, which differs greatly from real multi-site interaction. Such issues are often addressed by fitting the raw data to local models of adsorption such as the Langmuir,27 bi-Langmuir29 and Tóth30 models, prior to the calculation of the binding parameters. To simplify the above robust fitting procedures, previous reports have recommended the combination of AED calculations and Scatchard analysis.18-19 The pioneering work of Guiochon and Fornstedt has stated that AED calculations should be applied in biological systems when equilibrium is completely reached. For success in real biological system in non-steady state conditions, they should be replaced by the Adaptive Interaction Distribution Algorithm (AIDA) with a kinetic kernel function.40-41 This work applied frontal analysis to obtain the adsorption data. The very similar plateau levels provided proof of the steady state condition, thus confirming the rationality of the AED application with a four step strategy,18-19 as follows: (I) the adsorption isotherms are categorized as to that by visual inspection of



overloaded peaks; (II) the raw adsorption isotherm data are plotted as Scatchard plots; (III) the corresponding AEDs are calculated; (IV) the fitting procedure of models to the raw adsorption isotherm data. AED has been conveniently applied to immobilized proteins including lipoprotein23 and Cel7A.42 The paramount importance of GPCRs makes GPCR-drug interaction analysis of long-standing interest in the fields of analytical science, life sciences and the pharmaceutical industry.43 To our knowledge, no extension of AED to such analysis has been reported. We introduced the method into GPCRs with immobilized β2-AR as a probe. The categorization of the adsorption isotherms of salbutamol, terbutaline and pseudoephedrine was made by visual inspection of the raw adsorption isotherm data (Figure 1a-c) and overloaded peaks (Figure 3). All three ligands appeared to have a convex type I isotherm. To achieve reliable model discriminations, we applied Scatchard analysis to the raw data (Figure 1d-e)18-19, 23 Salbutamol and terbutaline displayed linear plots, indicating that the Langmuir model is favorable for their adsorption on immobilized β2-AR. Pseudoephedrine exhibited a concave curve, confirming a heterogeneous adsorption model. The heterogeneity of the bi-Langmuir means different types of adsorption sites with diverse adsorption energies, while the Tóth model describes adsorption when only one type of interaction occurs, even though the interaction is heterogeneous.44 We performed AED calculations to straightforwardly distinguish the unimodal and multimodal AEDs of those adsorptions. Figure 2 displayed successive stages of the calculation of AEDs from the raw data by


Page 12 of 41

Page 13 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


direct FAC-MS. These stages corresponded to increasing numbers of iterations applied in the algorithm. The increasing number of iterations changed the non-convergent wide peaks to sharp spikes. The unimodal energy distributions were observable for salbutamol and terbutaline. Pseudoephedrine presented a bimodal AED with one site at low energy and the other at high energy. Based upon these results, we declared that the Langmuir model is reasonable for the adsorption of salbutamol and terbutaline, and adsorption of pseudoephedrine favored the bi-Langmuir model over the Tóth model on β2-AR. We further verified the reasonability of the Langmuir model for salbutamol adsorption by site-directed competitive FAC-MS coupled with AED calculations (Figure S4). The spikes of unimodal energy distributions remained sharp when the number of iterations reached 107. The raw adsorption data of the three drugs were fitted to rival models to obtain the statistical evaluations (Table S1). Since a higher Fisher parameter means a better correlation of experimental data, we calculated the Fisher ratio to compare the nested and non-nested models using supplementary equations (14) and (15). As summarized in Table S2, the results were well in line with the AED calculations in terms of best model for the three drugs on β2-AR column. To further inspect whether the estimated adsorption isotherm parameters are physically reasonable, we utilized the desired model of each drug to predict their band profiles (Figure 3). The experimental profiles of salbutamol and terbutaline agreed well with the simulation using Langmuir models; while the Bi-Langmuir model was acceptable for describing the adsorption behavior of pseudoephedrine on the β2-AR



Page 14 of 41

column. The overlaps, defined in supplementary equation (16), between the experimental and calculated profiles were 98.8% for salbutamol, 99.2% for terbutaline and 99.4% for pseudoephedrine. This convincingly verified the precision of AED calculations for model selection in GPCRs-drug interaction analysis. Direct FAC-MS assays. We plotted the curves of the double-reciprocal of the adsorbed amounts versus drug concentrations (Figure 4) to measure the binding parameters (Table S3) by eq. (1). The linear regression coefficients of salbutamol and terbutaline (R2 = 0.9999 for both drugs) demonstrated high consistency with Scatchard plots and AED calculations. The association constants of salbutamol and terbutaline are two orders of magnitude larger than the data obtained by traditional frontal analysis44 and also more consistent with the previous values documented by radio-ligand binding assays.45 The Scatchard plot of pseudoephedrine (Figure 1f) presented a fluctuation point at 1.0 × 10-5 M, while the other points showed two linear relationships with correlation coefficients of 0.96 and 0.98 by eq. (1). The association constants for the high- and low-affinity binding sites were (1.64±0.08) × 105 M-1 and (0.9±0.1) × 103 M-1. Vansal and Feller46 prove that 1R, 2R-pseudoephedrine and 1S, 2S-pseudoephedrine have diverse affinities of 7 and 10 µM to human β2-AR. The corresponding association constants were 1.4 × 105 M-1 and 1.0 × 105 M-1.46 Such values are similar to the high binding affinity by FAC-MS in this work. The slight discrepancy between them is attributed to the diverse origination of the

receptors. Concentrating on

chromatographic methods, we declared that those variations in the association


Page 15 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


constants were also caused by the stationary phase, the methodological sensitivity and the mathematical models. The β2-AR stationary phase based on amino polystyrene microspheres exhibited better biocompatibility than the silica gel-conjugated receptors.36 The coupling of FAC to MS is applicable for investigations at a lower concentration range. This technique enables the examination of a broad affinity range. For instance, the lowest salbutamol concentrations analyzed in this work were ten times lower than those by FAC alone. Such low concentrations are crucial to accurately reveal affinity interactions between the immobilized receptor and the drugs. Analogous situations also occurred for terbutaline and pseudoephedrine. Site-Specific Competitive FAC-MS studies. The use of an indicator to monitor real-time competition in receptor-drug interactions and to screen drug candidates has been well established owing to its capacity to elucidate the exact binding domain of an unknown drug on an immobilized protein.47 Figure 5 illustrates the breakthrough curves of the indicator (salbutamol) with and without the application of a competitor (terbutaline and pseudoephedrine). We observed a loss of 3.0 and 4.0 min in breakthrough time when the competitors were included in the mobile phase. This result demonstrated that the two drugs competitively displace the same binding site with salbutamol. The percentage change in the breakthrough time was utilized to assess the binding affinity of the competitor using eq. (7): I (%) =

@+ − @ × 100% (7) @+ − @

where t1 and t represent the breakthrough times of the indicator in the absence or presence of the competitor; t0 is the void time of the chromatographic system.



The percentage change in the breakthrough time of salbutamol in the presence of pseudoephedrine and terbutaline was 16.5% and 12.4%, respectively, indicating that the binding affinity of terbutaline to β2-AR is much stronger than pseudoephedrine. This result implies that the percentage change in the breakthrough time of an indicator is possibly may preliminarily forecast the affinity of the competitors. When the mixture of two drugs was used as the competitor, the breakthrough time of salbutamol shifted from 26 min to 20 min, a percentage change of 24.8%. In theory, this value is equal to the sum of percentage changes when a single drug was applied. The slight deviation was reasonable because all the drugs in the current work belong to quaternary ammonium compounds. Inclusion of these compounds enhanced the ionic strength of the mobile phase, and thus contributed to a slight reduction of the indicator breakthrough time. Table S4 summarizes the binding parameters of salbutamol to β2-AR achieved by site-specific competitive FAC-MS. The high consistency of the association constants of salbutamol and β2-AR obtained by direct and site-specific competitive FAC-MS confirmed the feasibility and accuracy of the competitive method. This makes site-specific competitive FAC-MS superior for simultaneously identifying the binding site and forecasting the binding strengths of a drug to the immobilized receptor by simple experiments. β2-AR ligand screening. Paeoniflorin and liquiritin were confirmed to be the agonists of β2-AR in Shaoyao-Gancao decoction.48 To further explore the ability of site-direct competitive FAC-MS to screen bioactive compounds in a complex system, we


Page 16 of 41

Page 17 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


analyzed a mixture of six ligands containing paeoniflorin and liquiritin. As shown in supplementary Figure 5, this mixture shifted the breakthrough curve of salbutamol to the left, and thus revealed that at least one compound in the mixture competed same binding site with salbutamol on immobilized β2-AR. Application of propranolol and carazolol (antagonists of β2-AR) to the mobile phase demonstrated the identical breakthrough time of salbutamol. When paeoniflorin or liquiritin was applied, salbutamol shifted its breakthrough time to the left from 26 min to 25 or 24 min with a percentage change of 4.1% and 8.2%, respectively. This result proved that the two ligands are agonists of β2-AR with the same binding site to salbutamol. By these percentage change values, the association constants of paeoniflorin and liquiritin with β2-AR were estimated to be 1.52× 104 M-1 and 3.03× 104 M-1. These values were in good agreement with the data by frontal analysis ((2.16 ± 0.10) × 104 M-1 for paeoniflorin and (2.95 ± 0.15) × 104 M-1 for liquiritin) in previous work.48 The results indicated that FAC-MS is convenient for screening site-specific bioactive compounds in a complex matrix. Radio-ligand binding assays are canonical for probing the ligand of GPCRs. Despite the exact determination of binding affinities, this assay is powerless to distinguish an agonist or an antagonist. Efforts are growing

to develop signaling-dependent,

cell-based functional assays to pursue more precious information for screening ligands of GPCRs.49 These techniques include the cAMP assay, the IP3/IP1 and Ca2+ assays, and the reporter assay. The main drawback of these methods is the inevitable use of radio-ligands. Our works provides a promising and environment friendly



method for simultaneously screening and distinguishing agonists and antagonists of immobilized GPCRs from a complex system.

■ CONCLUSIONS We introduced adsorption energy distribution (AED) calculations into the development of a systematic strategy for selecting an appropriate adsorption model of ligands on immobilized β2-AR. Unlike other approaches applied to similar biological systems, AED calculation requires that equilibrium be reached, and must be tested with FA chromatograms. If not reached, AIDA should be used instead of AED. Our method is promising for the pursuit of real information about drug adsorption on the receptor, and so enables highly-efficient elucidation of GPCR-drug interaction. Due to the accurate selection of adsorption models, the loss in breakthrough time of an indicator drug proves to be dependent on the inclusion of site-specific competitors in the mobile phase. The percentage change in the breakthrough time of an indicator enables the screening of potential ligands by estimating their affinities to the receptor when a mixture sample is applied. Altogether, we concluded that the current strategy is advantageous to distinguish agonists and antagonists of immobilized GPCRs from a complex sample. This most noteworthy feature makes immobilized-receptor-based methodologies a powerful platform for high-content and high-throughput screening of ligands.

■ ASSOCIATED CONTENT


Page 18 of 41

Page 19 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Supporting Information The Supporting Information is available free of charge on the ACS Publications website. The fundamental theories of frontal affinity chromatography, isotherm models, calculation of the adsorption energy distributions, fisher ratio calculation and overlap calculation. Optimization of drug concentrations in frontal analysis; schematic of the FAC-MS system for monitoring molecular interactions; diagram of entrapment of β2-AR on amino polystyrene microspheres, chromatograms of four dugs on β2-AR column; AED calculations of salbutamol on

β2-AR

column

by

site-specific

competitive

FAC-MS;

FAC-MS

chromatograms showing the effects of paeoniflorin, liquiritin and a mixture of six compounds on the indicator salbutamol; raw adsorption isotherms and Scatchard plots of three drugs at higher drug concentrations; adsorption isotherm parameters of three drugs on β2-AR column; fisher ratio parameters of diverse adsorption models for three drugs on β2-AR column; association parameters for the binding of three drugs to β2-AR obtained by direct FAC-MS; binding parameters of salbutamol to β2-AR obtained by site-specific competitive FAC-MS; adsorption isotherm parameters of three drugs on β2-AR column at higher drug concentrations up to 50 mM.

■ AUTHOR INFORMATION Corresponding Author



Page 20 of 41

*Tel.: +86-29-88302686. E-mail: [email protected]. ORCID Qian Li: 0000-0002-1837-0208 Xiaohui Ning: 0000-0002-9405-8822 Yuxin An: 0000-0003-4478-4479 Brett J Stanley: 0000-0002-2785-5012 Yuan Liang: 0000-0002-7829-1114 Jing Wang: 0000-0003-1109-341X Kaizhu Zeng: 0000-0002-8566-8901 Fuhuan Fei: 0000-0002-4514-5735 Ting Liu: 0000-0002-1849-7655 Huanmei Sun: 0000-0001-5394-9808 Jiajun Liu: 0000-0002-8363-226X Xinfeng Zhao: 0000-0001-8824-738X Xiaohui Zheng: 0000-0003-0738-4943 Notes The authors declare no competing financial interest.

■ ACKNOWLEDGMENTS We thank the National Natural Science Foundation of China (21705126, 21775119 and 81702832), the Postdoctoral Science Foundation of China (2017M620467), the Key

Research

and

Development

Program


of

Shaanxi

Province

Page 21 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(2017ZDCXL-SF-01-02-01) and the Ministry of Education of the People's Republic of China (IRT_15R55) for their support of this work.

■ REFERENCES (1) Ritter, S.L.; Hall, R.A. Nat. Rev. Mol. Cell Bio. 2009, 10, 819-830. (2) Christopoulos, A. Nat. Rev. Drug Discov. 2002, 1, 198-210. (3) Daulat, A.M.; Maurice, P.; Jockers, R. Trends Pharmacol. Sci. 2009, 30, 72-78. (4) Vass, M.; Kooistra, A.J.; Ritschel, T.; Leurs, R.; de Esch, I.J.; De, G.C. Curr. Opin. Pharmacol. 2016, 30, 59-68. (5) Hillger, J.M.; Schoop, J.; Boomsma, D.I.; Slagboom, P.E.; Ijzerman, A.P.; Heitman, L.H. Biosens. Bioelectron. 2015, 74, 233-242. (6) Zhang, R.; Xie, X. Acta Pharmacol. Sin. 2012, 33, 372-384. (7) Jacobson, K.A. Biochem. Pharmacol. 2015, 98, 541-555. (8) Hauser, A.S.; Attwood, M.M.; Raskandersen, M.; Schiöth, H.B.; Gloriam, D.E. Nat. Rev. Drug Discov. 2017, 16, 829-842. (9) Graaf, C. de; Kooistra, A.J.; Vischer, H.F.; Katritch, V.; Kuijer M.; Shiroishi. M.; Iwata, S.; Shimamura, Tatsuro.; Stevens, R.C.; Esch, I.J.P. de; Leurs, R. J. Med. Chem. 2011, 54, 8195-8206. (10) Calleri, E.; Ceruti, S.; Cristalli, G.; Martini, C.; Temporini, C.; Parravicini, C.; Volpini, R.; Daniele, S.; Caccialanza, G.; Lecca, D.; Lambertucci, C.; Trincavelli, M.L.; Marucci, G.; Wainer, I.W.; Ranghino, G.; Fantucci, P.; Abbracchio, M.P.; Massolini, G. J. Med. Chem. 2010, 53, 3489-3501.



(11) Zhao, X.F.; Li, Q.; Chen, J.J.; Xiao C.N.; Bian, L.J.; Zheng J.B.; Zheng X.H.; Li, Z.J.; Zhang, Y.Y. J. Chromatogr. A 2014, 1339, 137-144. (12) Li, Q.; Wang, J.; Zheng, Y.Y.; Yang, L.J.; Zhang, Y.J.; Bian, L.J.; Zheng, J.B.; Li, Z.J.; Zhao, X.F.; Zhang, Y.Y. J. Chromatogr. A. 2015, 1401, 75-83. (13) Temporini, C.; Ceruti, S.; Calleri, E.; Ferrario, Silvia.; Moaddel, R.; Abbracchio, M.P.; Massolini, G. Anal. Biochem. 2009, 384, 123-129. (14) Guiochon, G.; Shirazi, D.G.; Felinger, A.; Katti, A.M. Fundamentals of Preparative and Nonlinear Chromatography, 2nd ed.; Academic Press: Boston, 2006. (15) Götmar, G.; Zhou, D.; Stanley, B.J.; Guiochon G. Anal. Chem. 2004, 76, 197-202. (16) Sandblad, P.; Arnell, R.; Samuelsson, J.; Fornstedt, T. Anal. Chem. 2009, 81, 3551-3559. (17) Stanley, B.J.; Krance, J.; Roy, A. J. Chromatogr. A 1999, 865, 97-109. (18) Fornstedt, T. J. Chromatogr. A 2010, 1217, 792-812. (19) Samuelsson, J.; Arnell, R.; Fornstedt, T. J. Sep. Sci. 2009, 32, 1491-1506. (20) Samuelsson, J.; Undin, T.; Törncrona, A.; Fornstedt, T. J. Chromatogr. A 2010, 1217, 7215-7221. (21) Lipponen, K.; Stege, P.W.; Cilpa, G.; Samuelsson, J.; Fornstedt, T.; Riekkola, M.L. Anal. Chem. 2011, 83, 6040-6046. (22) Hernández, V.A.; Samuelsson, J.; Forssén, P.; Fornstedt, T. J. Chromatogr. A 2013, 1317, 22-31. (23) Cilpa-Karhu, G.; Lipponen, K.; Samuelsson, J.; Öörni, K.; Fornstedt, T. Riekkola.


Page 22 of 41

Page 23 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


M.L. Analy. Biochem. 2013, 443, 139-147. (24) Rosenbaum, D.M.; Cherezov, V.; Hanson, M.A.; Rasmussen, S.G.F.; Thian, F.S.; Kobilka, T.S.; Choi, H.J.; Yao, X.J. Weis, W.I.; Stevens, R.C.; Kobilka, B.K. Science 2007, 318, 1266-1273. (25) Zhong, G.; Sajonz, P.; Guiochon, G. Ind. Eng. Chem. Res. 1997, 36, 506-509. (26) Loun, B.; Hage, D.S. Anal. Chem. 1994, 66, 3814-3822. (27) Cremer, E.; Huber, G.H. Angew. Chem. 1961, 73, 461-465. (28) Jaroniec, M. Adv. Colloid Interfac. 1983, 18, 149-225. (29) Graham, D. J. Phys. Chem. 1953, 57, 665-669. (30) Tóth, J. Acta Chem. Hung. 1971, 69, 311-317. (31) Stanley, B.J.; Guiochon, Georges. Langmuir 1994, 10, 4278-4285. (32) Quiñones, I.; Stanley, B.J.; Guiochon G. J. Chromatogr. A 1999, 849, 45-60. (33) Shepp, L.A.; Vardi, Y. IEEE T. Med. Imaging. 1982, 1, 113-122. (34) Stanley, B.J.; Krance J. J. Chromatogr. A 2003, 1011, 11-22. (35) Ruthven, D.M. Principles of Adsorption and Adsorption Processes, Wiley, New York, 1984. (36) Liang, Y.; Wang, J.; Fei, F.F.; Sun, H.M.; Ting, Liu.; Li, Q.; Zhao, X.F.; Zheng, X.H. J. Chromatogr. A 2018, 1538, 17-24. (37) Regenthal, R.; Krueger, M.; Koeppel, C.; Preiss, R. J. Clin. Monit. 1999, 15, 529-544. (38) Zhao, X.F.; Li, Q.; Xiao, C.N.; Zhang, Y.J.; Bian, L.J.; Zheng, J.B.; Zheng, X.H.; Li, Z.J.; Zhang, Y.Y; Fan, T.P. Anal. Bioanaly. Chem. 2014, 406, 2975-2985.



(39) Gritti, F.; Gotmar, G.; Stanley, B.J.; Guiochon, G. J. Chromatogr. A 2003, 988, 185-203. (40) Zhang, Y.; Forssén, P.; Fornstedt, T.; Gulliksson, M.; Dai, X. Inverse Probl. Sci. En. 2017, 8, 1-26. (41) Forssén, P.; Multia, E.; Samuelsson, J.; Andersson, M.; Aastrup, T.; Altun, S.; Wallinder, D.; Wallbing, L.; Liangsupree, T.; Riekkola, M. L.; Fornstedt, T. Anal. Chem. 2018, 90, 5366-5374. (42) Götmar, G.; Stanley, B.J.; Fornstedt, T.; Guiochon, G. Langmuir 2003, 19, 6950-6956. (43) Hauser, A.S.; Attwood, M.M.; Rask-Andersen, M.; Schiöth, H.B.; Gloriam, D.E. Nat. Rev. Drug Discov. 2017, 16, 829-842. (44) Zhao, X.F.; Huang, J.J.; Li, Q.; Wei, L.S.; Zheng, J.B.; Zheng, X.H.; Li, Z.J.; Zhang, Y.Y. J. Mol. Recognit. 2013, 26, 252-257.(45) Baker, J.G. Brit. J. Pharmacol. 2010, 160, 1048-1061. (46) Vansal, S.S.; Feller, D.R. Biochem. Pharmacol. 1999, 58, 807-810. (47) Slon-Usakiewicz, J.J.; Dai, J.R.; Ng, W.; Foster, J.E.; Deretey, E.; Toledo-Sherman, L.; Redden, P.R.; Pasternak, A.; Reid, N. Anal. Chem. 2005, 77, 1268-1274. (48) Li, Z.H.; Gao, H.Y.; Li, J.Y.; Zhang, Y.Y. J. Sep. Sci. 2017, 40, 2558-2564. (49) Zhang, R.; Xie, X. Acta Pharmacol. Sin. 2012, 33, 372-384.


Page 24 of 41

Page 25 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions Figure 1. a, b and c are the raw adsorption isotherms of salbutamol (□), terbutaline (○) and pseudoephedrine (△), respectively. d, e and f are Scatchard plots for salbutamol (□), terbutaline (○) and pseudoephedrine (△), respectively. The solid lines are fitted curves, and the hollow symbols are experimental data. Figure 2. AED of salbutamol (a), terbutaline (b) and pseudoephedrine (c) calculated from experimental frontal analysis data using 200 grid points and different iterations. Number of iterations: 1×105 (dash dot line); 1×106 (dot line); 5×106 (dash line); 1×107 (solid line); 1×108 (dash dot dot line). Figure 3. Comparison of calculated (solid line) with experimental (star symbols) band profiles of salbutamol (a), terbutaline (b) and pseudoephedrine (c) on β2-AR column. Injection of each drug was 50 mM during 50 s at a flow rate of 0.2 ml/min. The best fit Langmuir isotherm parameters were used for the simulation of salbutamol and terbutaline. The best fit Bi-Langmuir parameters were used for numerical calculation of pseudoephedrine. Figure 4. Plots of the double-reciprocal of the adsorbed amounts versus concentrations of salbutamol (a, □), terbutaline (b, ○) and pseudoephedrine (c, △). Figure 5. FAC-MS chromatograms of salbutamol without (a) or with a competitor. Competitors: terbutaline (b), pseudoephedrine (c), terbutaline and pseudoephedrine (d).



Table of Contents graphic


Page 26 of 41

Page 27 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1a is the raw adsorption isotherms of salbutamol (□). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)



Figure 1b is the raw adsorption isotherms of terbutaline (○). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)


Page 28 of 41

Page 29 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1c is the raw adsorption isotherms of pseudoephedrine (△). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)



Figure 1d is Scatchard plot for salbutamol (□). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)


Page 30 of 41

Page 31 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1e is Scatchard plot for terbutaline (○). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)



Figure 1f is Scatchard plot for pseudoephedrine (△). The solid lines are fitted curves, and the hollow symbols are experimental data. 82x58mm (300 x 300 DPI)


Page 32 of 41

Page 33 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2a AED of salbutamol (a) calculated from experimental frontal analysis data using 200 grid points and different iterations. Number of iterations: 1×105 (dash dot line); 1×106 (dot line); 5×106 (dash line); 1×107 (solid line); 1×108 (dash dot dot line). 82x58mm (300 x 300 DPI)



Figure 2b AED of terbutaline (b) calculated from experimental frontal analysis data using 200 grid points and different iterations. Number of iterations: 1×105 (dash dot line); 1×106 (dot line); 5×106 (dash line); 1×107 (solid line); 1×108 (dash dot dot line). 82x58mm (300 x 300 DPI)


Page 34 of 41

Page 35 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2c AED of pseudoephedrine (c) calculated from experimental frontal analysis data using 200 grid points and different iterations. Number of iterations: 1×105 (dash dot line); 1×106 (dot line); 5×106 (dash line); 1×107 (solid line); 1×108 (dash dot dot line). 82x58mm (300 x 300 DPI)



Figure 3a Comparison of calculated (solid line) with experimental (star symbols) band profiles of salbutamol (a) on β2-AR column. 82x58mm (300 x 300 DPI)


Page 36 of 41

Page 37 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3b Comparison of calculated (solid line) with experimental (star symbols) band profiles of terbutaline (b) on β2-AR column. 82x58mm (300 x 300 DPI)



Figure 3c Comparison of calculated (solid line) with experimental (star symbols) band profiles of pseudoephedrine (c) on β2-AR column. 82x58mm (300 x 300 DPI)


Page 38 of 41

Page 39 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4a. Plots of the double-reciprocal of the adsorbed amounts versus concentrations of salbutamol (a, □) 82x58mm (300 x 300 DPI)



Figure 4b. Plots of the double-reciprocal of the adsorbed amounts versus concentrations of terbutaline (b, ○) 82x58mm (300 x 300 DPI)


Page 40 of 41

Page 41 of 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4c. Plots of the double-reciprocal of the adsorbed amounts versus concentrations of pseudoephedrine (c, △) 82x58mm (300 x 300 DPI)



Figure 5. FAC-MS chromatograms of salbutamol without (a) or with a competitor. Competitors: terbutaline (b), pseudoephedrine (c), terbutaline and pseudoephedrine (d) 157x68mm (300 x 300 DPI)


Page 42 of 41

Reliable analysis of the interaction between ... - ACS Publications

Recommend Documents