2.1 Regents and materials

AGP (≥ 99% agarose gel electrophoresis, 41 kDa), KH₂PO₄, hydrochloric acid, sodium nitrate, acetanilide (AC), dipotassium hydrogen phosphate dihydrate, and sodium hydroxide were all supplied by Sigma-Aldrich, Germany. Fluka (Steinheim, Germany) was the supplier for DAC, CHL, and Propranolol (PRO). Aldrich (Steinheim, Germany) was the provider of 1-ethyl-3-methylimidazolium tetrafluoroborate (BMImBF₄), 1-decyl-3-methylimidazolium chloride (DMImCl), 1-ethyl-3-methylimidazolium chloride (EMImCl), 1-ethyl-3-methylimidazolium bromide (EMImBr), 1-butyl-3-methylimidazolium chloride (BMImCl), and 1-hexyl-3-methylimidazolium chloride (HMImCl). AMilli-Rx apparatus (Millipore, Milford, MA, USA) was utilised in order to purify the water used throughout this work.

2.2 Instrumentations

The HPLC system utilised in this work was a PerkinElmer series 200 LC binary solvent delivery (San Diego, Canada) with a series 200 vacuum degasser. The injector used was a Rheodyne injection valve (model 7725) with series 200 UV/V which is a variable wavelength detector. The AGP column (2.1 × 100 mm, particle size 5 μm) was supplied by Sigma-Aldrich (Germany). The data attained was gathered via a Total Chrom^TM Chromatography data handling System. A software was used with the PerkinElmer 600 series Link Interface to governor the HPLC system. The temperature set throughout this work was maintained at 37.00 °C via the use of a column oven (Model 200, PerkinElmer, Canada). For the mobile phase filtration, a pump supplied by the Scientific Technical Supplies (Model D-6072 Dreieich, West Germany) was utilised.

An Agilent 7100 System (Waldbronn, Germany) with PDA as a detector was used for the ACE work. A bare fused-silica capillary (Agilent, Moers, Germany) with total length 64.55 cm, effective length 56.00 cm and an internal diameter (ID) 50.00 µm was used. Chemstation software was used in order to acquire the electropherograms obtained for this work.

In order to weigh the material used in this work, a Mettler PM 400 analytical balance (Switzerland) was utilised. A Jenway 3510 pH meter (UK) was employed in order to adjust the pH values for the solutions used.

2.3 Preparation of solutions

The stock solution of the phosphate buffer at a physiological pH 7.45 was made through mixing 80.25 mL K₂HPO₄ (67.00 mmol L⁻¹) with 19.80 mL KH₂PO₄ (67.00 mmol L⁻¹). The stock solutions of DAC, CHL, and AC were prepared by dissolving 6.25 mg, 6.24 mg, 37.50 mg in 25.00 mL of the phosphate buffer at the physiological (pH 7.45) respectively. Prior to the dilution step, 1.00 mL of AC with a concentration of 300.00 μg mL⁻¹ was added to the solutions used for the analysis. As to the working solutions, they were made through the stock solutions in which an appropriate dilution of the stock solutions was made in 5.00 mL of the buffer (phosphate) at the physiological (pH 7.45). The solution AGP was made through dissolving the proper amount of AGP in a solution containing 67.00 mmol L⁻¹of KH₂PO₄ (pH 7.45) and 100.00 μmol L⁻¹ of EMImCl solutions. The solutions prepared were under the lab conditions, 25 ± 2 °C, and filtered via syringe filters (0.22 μm) prior to the ACE analysis.

2.4 Operating conditions

3.1 Study of AGP binding with CHL and DAC by ILATP/ACE

To check the validity of our established ILATPS/ACE system, the well-known PRO-AGP binding model was used (Abd El-Hady and Albishri, 2018). In this study, seven concentrations of AGP ranging between 5.00 and 100.00 µmol L⁻¹ were dissolved in a BGE containing 100.00 μmol L⁻¹ EMImCl and 67.00 mmol L⁻¹ K₂HPO₄ (pH 7.45). The mobility of PRO was normalised against the measured value for AC (300.00 µg mL⁻¹) which was utilised as an electroosmotic flow marker (EOF) for calculating the mobility ratio, cancelling out its effects due to the viscosity change in the BGE following the addition of AGP. The non-linear regression approach was utilised to calculate the apparent binding constants between PRO and AGP (Abd El-Hady and Albishri, 2018). The ACE apparent binding constant calculated was 2.90 × 10⁵ L mol⁻¹ which does not differ significantly (P > 0.05) from the reference value (3.00 × 10⁵ L mol⁻¹) (Albani and Riva, 1989; Abd El-Hady and Albishri, 2018) indicating the suitability of the system developed.

Several rinsing procedures were utilised so as to avoid any changes in the migration time and peak areas of the analytes investigated which ultimately would have a negative impact on the reproducibility of the values obtained. The procedures included methanol, sodium hydroxide and hydrochloric acid to avoid any likelihood of EMImCl IL and AGP adsorption on the inner surface of capillary during the ACE analysis. The relative standard deviation (RSDs) obtained for the migration times and peak areas of the analytes investigated were ≥3 (n = 5) in the case of methanol and sodium hydroxide indicating the negative impact of these solutions (methanol and sodium hydroxide). Whereas when HCl solution was used, no changes were observed illustrating the suitability of this solution (HCl) for rinsing. Hence, the use of 0.10 mol L⁻¹ HCl for 1 min followed by the deionised water for 2.00 min and then the BGE for 3.00 min was used to enhance the reproducibility of the migration times and peak areas yielding good precision values (RSDs < 0.9%, n = 5). This improvement of analysis was thought to be owing to the ability of HCl to eradicate any adsorbed molecules of EMImCl and/or AGP that might have existed on the interior surface of the capillary.

The developed ILATPS/ACE system was also employed to determine the apparent binding constants of two significant anti-cancerous drugs (CHL and DAC) in relation to the AGP protein. The effect of various types of alkyl chain imidazolium ILs combined with their concentrations ranging between 10.00 and 1000.00 μmol L⁻¹, below their critical micelle concentrations (CMC), (Abd El-Hady et al., 2014) on CHL-AGP and DAC-AGP bindings were investigated as shown in Tables 1 and 2. As shown in the tables, the results related to the performance of studying ILs revealed that the utilisation of EMImCl up to 100.00 μmol L⁻¹ demonstrated no significant deviation for the CHL-AGP and DAC-AGP apparent binding constants with the ILATPS/ACE in comparison with those measured by the HPAC. However, at 500.00 μmol L⁻¹of EMImCl, a change with a significant differentiation (P < 0.05) was observed and this differentiation was increased as the concentration increased reaching 1000.00 μmol L⁻¹, as illustrated in Tables 1 and 2. This variance might due to the water-miscible EMImCl IL acting almost like the chaotropic agents, in which, high concentrations are needed to denature the AGP to a complete unfolded state. Yet, in contradiction of the compounds investigated, high concentrations of EMImCl might lead to the AGP’s oligomerisation (Baker and Heller, 2009). Moreover, it was observed that utilising a longer alkyl chain C_nMImCl (C₂–C₁₀) combined with their higher concentrations up to 1000.00 μmol L⁻¹ greatly changed the apparent binding constants of the CHL-AGP and DAC-AGP (P < 0.05) in comparison with those binding values by HPAC, as reported in Tables 1 and 2. The reason for these changes might have been due to the increase of ILs lipophilicity from BMImCl to DMImCl, which as a result, un-preferably promoted their hydrophobic interactions with AGP. That is to say, IL cations with longer alkyl side chains affected the stability of AGP structure due to the increase in IL hydrophobic surface area contact with AGP molecule, and subsequently, has led to the AGP apparent binding constants deviation (Silva et al., 2014). In order to support this explanation; several techniques including fluorescence, UV–Vis, and FT-IR spectroscopy have been utilised and it has been concluded that the lipophilic force could alter the protein’s structure (Huang et al., 2013; Yan et al., 2012). In addition, the ILs anions effect on the CHL-AGP and DAC-AGP binding analysis were also examined. It was seen that the best results for the apparent binding constants were achieved in the presence of chloride and bromide anions unlike for tetrafluoroborate anion as indicated in Tables 1 and 2. The reason for this result could be attributed to the difference in size and electronegativity of the anions investigated. Therefore, the ionic liquid EMImCl is the optimum choice for studying the apparent binding constants of AGP with the drugs investigated. Fig. 1(A) and (B) illustrate the binding electropherograms between CHL and AGP in the absence and presence of EMImCl respectively. Fig. 2(A) and (B) demonstrate the binding electropherograms between DAC and AGP in the absence and presence of EMImCl respectively. Additionally, the traces “a” displays the CHL and AGP’s migration times when the AGP and EMImCl were absent, as seen in Fig. 1A and 2A. In the same wavelength, the traces “a” displays the CHL and AGP’s migration times when the AGP and EMImCl were present as seen in Fig. 1B and 2B. It was evidently seen that both CHL and DAC had a slower migration time than observed with the AC marker under the normal CZE mode, as shown in traces “a” in Fig. 1A and 2A. This behavior might owe to the dissociation of the COOH group forming a negatively charged CHL possessing a pK_a of 5.80 as well as the phenomenon of tautomerisation of the amide group in DAC possessing a pK_a of 4.45 under the physiological pH 7.45 (Abd El-Hady et al., 2015).

Close

Fig. 1

Illustration of the apparent binding between CHL and AGP absence (A) and presence of EMImCl (100 μmol L⁻¹) (B). Experimental conditions: 100.00 µmol L⁻¹ DAC; AGP concentrations (0) (a), (25.00 µmol L⁻¹) (b), (50.00 µmol L⁻¹) (c) and (100 µmol L⁻¹) (d), phosphate buffer (67 mmol L⁻¹) at pH 7.45, applied voltage 20.00 kV with a positive polarity, capillary’s temperature 25.00 °C, detection wavelength 254.00 nm, injection pressure 25.00 mbar for injection time 10.00 s.

Export to PPT

Close

Fig. 2

Illustration of the binding between DAC and AGP when EMImCl was absent (A) and when EMImCl was present (100.00 μmol L⁻¹) (B). Experimental conditions: 100.00 µmol L⁻¹ DAC; AGP concentrations (0) (a), (25.00 µmol L⁻¹) (b), (50.00 µmol L⁻¹) (c) and (100 µmol L⁻¹) (d), phosphate buffer (67.00 mmol L⁻¹) at pH 7.45, applied voltage 20.00 kV with a positive polarity, capillary’s temperature 25.00 °C, detection wavelength 254.00 nm, injection pressure 25.00 mbar for injection time 10.00 s.

Export to PPT

The mobility ratios for CHL and DAC without the addition of EMImCl were 0.55 and 0.52, respectively. The same mobility ratios of CHL and DAC were attained when EMImCl (100.00 μmol L⁻¹) was present. Thus, the interactions between the investigated drugs or AGP and EMImCl IL are thought to be very minimal. This is because there was no shift observed in the mobility ratios for both drugs when the IL was added. Additionally, an insignificant shift in the AC’s migration time (EOF marker) was observed when the IL or AGP was present in comparison with their absence. This finding might be referred to the change of BGE’s viscosity after the addition of IL or AGP. Yet, the effects of the mobility ratio in the apparent binding calculations was cancelled out. It was also seen that the investigated analytes’ migration times were clearly changed after adding various concentrations of AGP ranging from 25.00 µmol L⁻¹ trace “b” to 100.00 µmol L⁻¹ trace “d” in the presence of 100.00 μmol L⁻¹ EMImCl (B) in comparison with their absences (A). These observations facilitated, consequently, the use of ACE for the calculation the apparent binding constant of CHL-AGP and DAC-AGP, as demonstrated in Tables 1 and 2. In addition, the HPAC technique was utilised as a verification approach in order to validate and increase the reliability and level of the confidence of the calculated apparent binding constants of CHL-AGP and DAC-AGP attained by the proposed ILATPS-ACE system. In HPAC, employing a AGP immobilized stationary phase is commonly practiced without the need for any stabilising agents such as ILs (Hage et al., 2011, 2009; Li et al., 2016). In the HPAC experiments, a model called zonal elution was utilised to gain information about the drug-protein binding equilibria. This was conducted through the injection of a 10.00 µL of 100.00 µmol L⁻¹ CHL or 100.00 µmol L⁻¹ DAC (Hage et al., 2011). The model was based on the interaction of the drugs investigated with the AGP immobilised stationary phase in the presence of 100% (v/v) K₂HPO₄ (67.00 mmol L⁻¹) at pH 7.45 as the mobile phase (Hage et al., 2011). In order to examine the extent of this interaction between the drugs investigated and the AGP immobilised stationary phase, the plots of 1/k on Y-axis versus [drug] on X-axis was constructed (Hage et al., 2011, 2009; Li et al., 2016). As demonstrated in Tables 1 and 2; the calculated apparent binding constants by HPAC were found to be very close to those obtained by the ILATP-ACE when EMImCl (100.00 μmol L⁻¹) was present without any substantial variance (P > 0.05) (Tables 1 and 2). Hence, the HPAC values attained were in line with the ILATP/ACE analysis indicating that EMImCl has indeed improved the stability of AGP in ACE analysis without any negative impact on the binding sites of AGP with CHL and DAC drugs. In addition, this alternative technique (HPAC) has, consequently, indeed increased the confidence and reliability of the results obtained and proposed by the ILATPS-ACE system for calculating the apparent binding values of CHL-AGP and DAC-AGP.

3.2 Stability of AGP during storage of BGE

In order to examine the stability of the AGP solution (25.00 µmol L⁻¹) at the ambient conditions for a period of 30 days, 100.00 and 200.00 µmol L⁻¹ of EMImCl were prepared. Table 3 demonstrates the mobility ratios of CHL and DAC during the investigated storage period as well as compares the results obtained in the absence of EMImCl. The comparison was made between the preserved protein solution containing the IL versus freshly prepared protein solution in the absence of IL. It was observed that the mobility ratios were quite constant in the presence of 100.00 µmol L⁻¹ EMImCl compared with those in the absence of IL up to 30 days storage time under the ambient temperature. Moreover, the use of 100.00 µmol L⁻¹ IL might decrease the cost of AGP’s consumption material as well as the time required for the ACE analysis without the need for a daily preparation of AGP solution as commonly practiced. However, a high concentration of EMImCl (200.00 µmol L⁻¹) has failed to reserve the constant values of mobility ratios for the period studied. These observations demonstrated the vital impact of the concentration of EMImCl on the reservation of the AGP’s structure in the solutions. Furthermore, storing the AGP solution prepared in EMImCl/phosphate solution at 4.00 °C refrigeration temperature has indeed increased the AGP stability up to 90 days (data not shown). At 4.00 °C, a two-phases of IL-rich phase were created as well as a negatively charged AGP bound strongly to the positively charged EMIm in the IL-rich phase. This has led to, in turn, the π–π bonding with the imidazolium ring resulting in maintaining the AGP’s activity (Abd El-Hady and Albishri, 2018). Furthermore, the physiological pH 7.45 has also prevented the conformational changes of AGP and helped to maintain it. This behavior of EMImCl-ATPS involving its phase diagrams in K₂HPO₄ at different temperatures was previously investigated (Zafarani-Moattar and Hamzehzadeh, 2007). In general, ILATPS is based on a phase-splitting promoter (usually an inorganic salt) in order to facilitate the formation of two immiscible aqueous phases from a homogeneous aqueous solution of a hydrophilic ionic liquid. Therefore, an ionic liquid-rich and an inorganic salt-rich phase could be obtained, both of which often have water as the main component. Furthermore, the same occurred in the EMImCl-ATPS system, where the use of K₂HPO₄ salt-induced EMIm cation salted out and formed the upper IL-rich phase and K₂HPO₄-rich lower phase. Meanwhile, the water structure induces the chloride anions to increase the dielectric constant of the aqueous phase. This leads to more promoting of the low dielectric EMIm cations focusing in the upper phase concurs with more focusing of K₂HPO₄ salt in the lower phase. This helps to preserve the charge balance. The K⁺ ions are concentrated in the lower phase to balance Cl⁻ anions (Zafarani-Moattar and Hamzehzadeh, 2007). With the addition of AGP, it promptly distributed into the upper IL-rich phase possessing the strongly bounded cationic EMIm via electrostatic attractions and π–π bonding. In addition, the impact of temperature on the separation of the bi-phasic system was also observed. On one hand, when the temperature is less than 4.00 °C, the formation of bi-phasic system is promoted; on the other hand, higher temperatures accelerate the movement of the IL and salt through the biphasic boundary resulting in one-phase’s formation. In the literature, the effect of temperature on the separation of the bi-phasic system was attributed to the ions’ hydration strength. The cause of this unusual phenomenon might be seen in a minor reduction of the Gibbs free energy of hydration for the IL ions with the reduction of the temperature (Du et al., 2007). Thus, the ions solubility of the IL goes down and therefore, the separation of the bi-phasic system occurs. After that, the formed bi-phasic system could be swiftly switched to a one-phase via shaking for a period of 3.00 min at the ambient temperature prior to the CE’s analysis (Zafarani-Moattar and Hamzehzadeh, 2007).