Study of captopril pharmacokinetics in rabbit blood with microdialysis based on online generated Au nanoclusters and pepsin–captopril interaction in luminol chemiluminescence

Abstract

A luminol–HAuCl₄–pepsin (Pep) flow injection-chemiluminescence (FI-CL) system was explored to determine captopril (CAP) based on the CL intensity inhibition effect and applied to study CAP pharmacokinetics in rabbits with microdialysis. HAuCl₄ and pepsin (Pep) could significantly enhance the luminol chemiluminescence (CL) intensity. It was found that sub-nanometre Au nanoclusters (AuNCs) were generated in the luminol–HAuCl₄–Pep reaction solution. A possible mechanism for AuNCs generation is given. By means of the FI-CL and molecular docking (MD) methods, the Pep–CAP interaction was systematically studied. The results showed that CAP might enter into Pep active site Asp32 with the binding constant (K) 1.7 × 10⁶ L mol⁻¹, which could effectively inhibit the CL intensity. The CL intensity could be remarkably inhibited by CAP and the decrement of CL intensity was linearly correlated to the logarithm of CAP concentration in the range of 3.0 pmol L⁻¹ to 0.1 μmol L⁻¹, with a detection limit of 1.0 pmol L⁻¹ (3σ). This proposed approach was successfully applied to determine CAP in rabbit’s blood during the 16 h after intragastric administration with an elimination ratio of 45.9% and recovery ratios from 89.0% to 112.0%. The pharmacokinetic results showed that CAP could be rapidly absorbed into blood with a peak concentration (C_max) of 9.63 ± 1.45 μg mL⁻¹ at a maximum peak time (T_max) of 0.75 ± 0.08 h; the elimination half-life of 3.19 ± 0.13 h and the elimination rate constant of 7.27 ± 0.41 L g⁻¹ h⁻¹ in rabbits were derived, respectively.

---

Introduction

Protein–drug interactions have become a hot topic in the fields of medicine, chemistry and biology for drug discovery, screening, design and development.^1–3 Recently, numerous works have been performed on predicting the binding sites of drugs to proteins and analyzing the interaction patterns between them.^4,5 Pepsin (Pep) (MW: 34.5 kD) is a monomeric, two domain, mainly l-protein, with a high percentage of acidic residues (43 out of 327).⁶ The catalytic sites are Asp32 and Asp215 for the Pep to be active.⁷ Pep, as a digestive protease, is most efficient at cleaving peptide bonds between hydrophobic and aromatic amino acids such as phenylalanine (Phe), tryptophan (Trp) and tyrosine (Tyr). The interaction behavior of Pep with bisphenol A,⁸ nobiletin⁹ and fleroxacin¹⁰ was systemically investigated using fluorescence spectroscopy, UV-visible absorption, resonance light scattering, synchronous fluorescence spectroscopy, 3D spectroscopy and molecular docking (MD), while the relative interaction parameters, such as binding constants and thermodynamic parameters were given.

Gold nanoclusters (AuNCs) have attracted substantial research interest in the fields of chemistry,^11,12 materials,^13,14 biology,^15–17 and medicine.¹⁸ Considerable efforts have been devoted to exploring synthesis methods for stability, functionality and solubility of AuNCs.^19–22 Synthesis methods for AuNCs with biological macromolecule-mediators such as DNA,²³ peptides,²⁴ bovine serum albumin (BSA)²⁵ and Pep²⁶ in alkaline solution have been reported. In order to overcome the time-consuming nature of AuNC biosynthesis, technologies such as microwaves²⁷ and photolithography²⁸ have gradually been applied in the biosynthesis of AuNCs. The photochemical induced effect (PCIE), as a photo-induced effect,^29,30 could not only quickly induce AuNC generation in solution under biological macromolecule-mediation, but also endow AuNCs with some special photoelectric properties. In view of its special nature, PCIE will open a new way for the biosynthesis of AuNCs in solution. There are no reports of online generated AuNCs in a flow inject-chemiluminescence (FI-CL) system and the application on protein–drug interaction.

Captopril (CAP, Fig. 1) has a significant antihypertensive effect as an angiotensin converting enzyme inhibitor (ACEI), which could improve cardiac function in patients.³¹ The methods commonly used for determining CAP in vivo are liquid chromatography-mass spectrometry (LC-MS) or liquid chromatography-column derivatization-UV detection (LC-CD-VWD).^32,33 But their time-consuming nature, expensive instrumentation and low sensitivities are bottlenecks of the above methods for CAP determination in vivo. CL methods have gradually become the general and practical methods for determining CAP with high sensitivity and wide dynamic range.³⁴ Recently, Tzanavaras has reviewed a variety of flow related methods for CAP determination in both pharmaceutical and biological samples.³⁵

Fig. 1 The structure of CAP.

It has been reported that HAuCl₄ as a co-reactant can remarkably increase luminol CL intensity.^36,37 To date, no flow inject-chemiluminescence (FI-CL) approach combined with HAuCl₄ and Pep has been designed and developed for drug analysis in vivo and Pep–drug interaction. In this work, we developed a luminol–HAuCl₄–Pep FI-CL approach for CAP determination, and applied the proposed approach to study CAP pharmacokinetics in rabbits with microdialysis. The aims of the present study were to: (1) investigate the mechanism of the complex enhancement effect of CL and the complex quench effect of CL in the luminol–HAuCl₄–Pep/CAP CL system; (2) develop a luminol–HAuCl₄–Pep FI-CL approach to study CAP pharmacokinetics in rabbits with microdialysis.

Experimental section

Chemicals and reagents

All the reagents used were analytical grade. Water was purified using a Milli-Q system (Millipore, Bedford, MA, USA) with a resistivity of 18.2 MΩ cm⁻¹ and used throughout the whole experiment. Luminol (Fluka, Biochemika, Switzerland) and Pep (porcine gastric mucosa, 010M7006V, Sigma-Aldrich, St. Louis, MO, USA) were used without further purification. CAP was purchased from the National Institute of Control of Pharmaceutical and Biological Products, China. Chloroauric acid (HAuCl₄, analytical grade) was purchased from Shanghai Reagent Factory, China; Capoten tablets (Sino-American Shanghai Squib Co., Ltd, China, H20010430) were purchased from a local dispensary.

Stock solutions of CAP (1.0 mmol L⁻¹) and Pep (100.0 μmol L⁻¹) were prepared in purified water and stored at 4 °C. Working standard solutions of CAP and Pep were prepared daily by diluting the stock solution appropriately with purified water. A stock solution of luminol (2.5 × 10⁻² mol L⁻¹) was prepared by dissolving 0.44 g luminol in 100 mL NaOH (1.0 × 10⁻¹ mol L⁻¹) solution in a brown calibrated flask. A stock solution of HAuCl₄ (2.5 × 10⁻² mol L⁻¹) was prepared by dissolving 1.0 g HAuCl₄ in 100 mL purified water and stored at 4 °C.

Apparatus

The apparatus (Model IFFM-E, Xi’an Remax Electronic Science-Tech. Co. Ltd) of the FI-CL system consisted of a sampling system, photomultiplier tube (PMT), and a PC with an IFFM-E client system (Remax, Xi’an, China). A polytetrafluoroethylene (PTFE) tube (1.0 mm i.d.) was used to carry the solutions. The microdialysis system was composed of a CMA/100 microinjection pump, a CMA/140 micro-fraction injector (CMA, Stockholm, Sweden) and microdialysis probes (CMA/20, Beijing Ying Bo Li Da Technology Development Co., Ltd, China). The UV-Vis absorption spectra (225–800 nm) were collected using a U-3010 spectrophotometer system (Hitachi, Japan). The TEM images were obtained using a Tecnai G² F20 S-TWIJEM-2010 transmission electron microscope (FEI, USA) operated at 200 kV.

The profile for different systems with static injection CL

The static injection CL method was used to evaluate the CL kinetics progress for different CL systems. Using the permutations method, four different FI-CL systems were designed to study the different CL mechanisms. For the luminol–dissolved oxygen/CAP CL system, 100 μL luminol solution was directly injected into the dissolved oxygen solution in the absence or presence of CAP. For the luminol–HAuCl₄/CAP and luminol–Pep/CAP CL systems, 100 μL luminol solution was injected into the HAuCl₄ solution and the Pep solution in the absence or presence of CAP, respectively. For the luminol–HAuCl₄–Pep/CAP CL system, the Pep was first mixed with CAP, then with HAuCl₄ to form the HAuCl₄–Pep/CAP solution in the presence of CAP; finally 100 μL luminol solution was injected into the above HAuCl₄–Pep/CAP solution. The CL intensity was measured by the PMT (negative voltage was set as 400 V) for 40 s.

The procedure of luminol–HAuCl₄–Pep CL combined with microdialysis

In the luminol–HAuCl₄–Pep CL system, five flow lines were inserted into the solutions of luminol, carrier (purified water), HAuCl₄, Pep and CAP, respectively, and the solutions were propelled by peristaltic pumps. Luminol (100 μL) was quantitatively injected into the mixed solution of HAuCl₄, Pep and samples by six-way valve, then the mixture was delivered into the flow cell producing CL emission which was detected by PMT (negative voltage was set as 700 V). The concentration of CAP was quantified by the decrease of CL intensity (ΔI = I₀ − I_s), where I_s and I₀ are CL signals in the presence and in the absence of CAP samples, respectively.

A retrograde calibration technique was used for the assessment of the in vivo recovery rate of the luminol–HAuCl₄–Pep CL system (Fig. 2). Two hours post probe implantation, which served as a stabilization period, the perfuse (C_perf) and dialysate (C_dial) concentrations of CAP were determined using the luminol–HAuCl₄–Pep FI-CL system. The relative loss of CAP during retro-dialysis (L_retro) or relative recovery (R_dial) by dialysis was then calculated as follows: L_retro = R_dial = (C_perf − C_dial)/C_perf.

Fig. 2 Scheme of CAP determination in rabbit blood of the luminol–HAuCl₄–Pep FI-CL system with microdialysis.

Molecular docking

The MD of Pep–CAP was performed with the open-free soft of Autodock 4.2 using a semi-flexible docking mode. The crystal structure of Pep (PDB entry 1YX9) was obtained from the Protein Data Bank. The 3D structure of CAP was generated using ChemDraw 10.0 and Chem3D 10.0 soft (Cambridge Soft, USA); and the energy-minimized conformation was obtained by the Gasteiger–Huckel charges with a gradient of 0.005 kcal mol⁻¹.³⁸ With the aid of AutoDock tools, the ligand root of CAP was detected and rotatable bonds were free-defined. The grid box with 60 Å × 60 Å × 60 Å along x, y, z axes of 0.375 Å spacing was set in the whole process of MD. The population size and the maximum number of energy evaluation were set as 1.5 × 10² and 2.5 × 10⁶, respectively. The Lamarckian genetic algorithm was applied for docking simulations. The conformation with the lowest binding energy was analyzed using PyMOL 1.6.0.0.

Method validation for CAP determination

The proposed method was validated regarding its selectivity, linearity, the limit of detection (LOD), accuracy, precision, recovery and stability. The linearity of methods were constructed between the relative CL intensity and the different concentrations of CAP. The LOD was considered as the final concentration that produced a signal-to-noise (S/N) ratio of 3. The precision and accuracy of the method were assessed by performing replicate analyses of CAP with anti-coagulant citrate dextrose (ACD) solution consisting of citric acid 3.5 × 10⁻³ mol L⁻¹, sodium citrate 7.5 × 10⁻³ mol L⁻¹, and dextrose 13.6 × 10⁻³ mol L⁻¹. The precision was determined from inter-day and intra-day using six determinations of low, medium and high concentrations and expressed as relative standard deviation (RSD%). The extraction recovery rate was determined by calculating the ratio between the amounts of the drug-free samples and those spiked with known amounts of CAP into drug-free samples. The stability of the sample was assessed by measuring the analysis data of CAP standard samples with high, medium and low concentrations under ambient conditions. To evaluate its selectivity, different foreign species were added into a standard solution of CAP, and the impact of foreign substances on the standard CAP solution assessed.

Pharmacokinetic study of CAP in rabbits

A Capoten tablet was stripped of the outer sugar coating and ground to a powder. The Capoten tablet powder (1.00 g) was accurately weighed and placed in a beaker, 50 mL deionized water was added with ultrasound for 30 min, and a constant volume was kept with purified water added to a 100 mL brown volumetric flask, chilled in the dark.

Male rabbits (1.8–2.2 kg, n = 5) were purchased from the Laboratory Animal Center of Xi’an Jiaotong University (Xi’an, P.R. China) and housed in a cage with free access to food and water available ad libitum. The animals were acclimatized for at least one week with a 12 h light/dark cycle. All experimental rabbit surgery procedures were approved by the institutional animal experimentation committee of Xi’an Jiaotong University.

On the day of experiment, each rabbit was initially anesthetized with chloral hydrate solution (1.0 mg kg⁻¹, subcutaneous) and catheters were positioned within the jugular vein toward the right atrium and then perfused with ACD solution. The flow rate of ACD was set at 3.0 μL min⁻¹ by a microinjection pump for blood microdialysis. The rabbit’s body temperature was maintained at 37 °C with a heating blanket. Following a 2 h stabilization period after surgery, 1.16 mg kg⁻¹ CAP was administered via intragastric (i.g.) administration. The dialysates were collected every 25 min for 16 h and preserved at −4 °C in a refrigerator. The concentration of CAP in the dialysate was determined by the luminol–HAuCl₄–Pep FI-CL system. CAP microdialysis concentration (C_m) was converted to unbound concentration (C_u) as follows: C_u = C_m/L_retro.

Results and discussion

Relative CL intensity–time profile

The relative CL intensity–time profiles of different photochemical reaction systems are shown in Fig. 3. It can be seen from the CL intensity–time profiles that the maximum time (T_max) for reaching maximum CL intensity (I_max) of the luminol–dissolved oxygen and luminol–HAuCl₄ CL systems (curves 2 and 6) was 3.0 s with the I_max values of 45 and 550, respectively; compared with the luminol–dissolved oxygen and luminol–HAuCl₄ CL system, the T_max of the luminol–Pep and luminol–HAuCl₄–Pep CL systems (curves 4 and 8) were shortened from 3.0 s to 2.8 s, and the corresponding CL intensities for curves 4 and 8 were 104 and 1143, respectively. From these results, we could speculate that Pep could accelerate the electron transfer rate due to the proton process of luminol or luminol–HAuCl₄ and Pep in alkaline solution, which could lead to the shortage of T_max for curves 4 and 8. Compared with the CL system of curves 2 and 4, HAuCl₄ as the co-reactant could remarkably increase the luminol CL intensity. The reason might be attributed to the Au nuclei generated in the alkaline solution, which could cause the quantum confinement effect mediated by the PCIE of luminol. In the presence of CAP (C_CAP = 10.0 pmol L⁻¹), it could sharply quench from 1143 to 1017 for the luminol–HAuCl₄–Pep CL system with a quenching ratio of 11.0%. The CL systems of luminol–dissolved oxygen, luminol–HAuCl₄ and luminol–Pep had almost no or slightly inhibitory effect. This result could explain why the luminol–HAuCl₄–Pep CL system had a higher sensitivity for minor changes in the confirmation of Pep, which was mediated by the interaction between Pep and CAP. Interesting, the CL intensity for curves 1–6 extinguished during 40 s, while curves 7 and 8 had more stable CL intensities than other CL intensities and lasted for 80 s before extinguishing.

Fig. 3 Kinetic curves for different CL reactions. Curve 1: luminol–dissolved oxygen-CAP (C_CAP = 5.0 nM); Curve 2: luminol–dissolved oxygen; Curve 3: luminol–Pep–CAP (C_CAP = 7.0 pM); Curve 4: luminol–Pep; Curve5: luminol–HAuCl₄–CAP (C_CAP = 30.0 pM); Curve 6: luminol–HAuCl₄. Curve 7: luminol–HAuCl₄–Pep–CAP (C_CAP = 10.0 pM); Curve 8: luminol–HAuCl₄–Pep. The corresponding concentrations of luminol, HAuCl₄ and Pep were 2.5 × 10⁻⁴, 2.5 × 10⁻⁵ and 1.0 × 10⁻⁶ mol L⁻¹, respectively.

The CL intensity could be enhanced and inhibited when alkaline luminol was mixed with different HAuCl₄–Pep solution and HAuCl₄–Pep/CAP solution. Different CL response intensities could be obtained (Fig. 3). But of all the mentioned CL systems, the luminol–HAuCl₄–Pep and luminol–HAuCl₄–Pep/CAP CL intensities could be significantly enhanced and inhibited compared with other CL systems. The mechanisms of luminol–dissolved oxygen, luminol–Pep and luminol–HAuCl₄ have been explained in previous reports.^39,40 Here we mainly focused on the luminol–HAuCl₄–Pep and luminol–HAuCl₄–Pep/CAP CL systems to explain the possible mechanism of AuNCs generated in alkaline solution and the interaction of Pep/CAP in the luminol–HAuCl₄–Pep CL system.

CL mechanism for luminol–HAuCl₄–Pep/CAP system

Complex enhancement effect of CL. To further confirm the above possible mechanism, the different reaction solutions were investigated using TEM and UV-Vis absorption under fixed reactant concentration (Fig. 4 and 5). HR-TEM results showed sub-nanometre AuNCs could be generated in the luminol–HAuCl₄–Pep CL reaction solution with average diameter distributed in the range 1–2 nm. The UV-Vis results showed the absorption wavelength of Pep was 256 nm, which is smothered by the peak absorption of luminol; the characteristic absorption of luminol had decreasing tendency at 285 nm and 325 nm; meanwhile, a new absorption peak at 545 nm was produced, which is a Au atom character absorption peak.⁴¹ Based on this characterization, we could make the conclusion that HAuCl₄ firstly formed Pep–Au(III) complex in the CL system, then reduced to Au nuclei and eventually formed the sub-nanometre AuNCs by the PCIE of luminol,^42,43 when injected into alkaline luminol solution with a flow rate of 2.0 mL mol⁻¹.

Fig. 4 HR-TEM images of sub-nanometre AuNCs generated in the luminol–HAuCl₄–pepsin CL reaction solution.

Fig. 5 UV-Vis graphic of the luminol–HAuCl₄–Pep system with a pH of 10.5. The corresponding concentrations of luminol, HAuCl₄ and Pep were 2.5 × 10⁻⁴, 2.5 × 10⁻⁵ and 1.0 × 10⁻⁶ mol L⁻¹, respectively.

Combined with the results of TEM and UV-Vis, in the profile of the luminol–HAuCl₄–Pep CL system, it was found that Pep–Au(III) complex was firstly formed on mixing Pep with HAuCl₄; then this was reduced on addition into alkaline luminol solution (Fig. 6). The detailed mechanism could refer to our previous report about photochemically induced formation of Au nanomaterial with size and shape controlled by the luminol–Pep CL reaction.⁴⁴ The AuNCs generated in the luminol–HAuCl₄–Pep CL system were conducted in the fixed alkaline luminol solution with a pH value of 10.5. From the perspective of the experimental performance, the alkaline luminol had the optimal luminous efficiency at a pH of 10.5, which could improve the detection sensitivity. Due to their properties of sub-nanometre size and good hydrophilicity, AuNCs could effectively prevent pipeline blockage, which is caused by the effect of deposition and aggregation of the AuNCs, to disturb the sensitivity and reproducibility for CAP determination.

Fig. 6 Schematic illustration of the CL enhancement and AuNC generation mechanism in luminol–HAuCl₄–Pep CL reaction.

Complex quench effect of CL. Using the established model of protein–drug interaction,⁴⁵ the corresponding binding parameters for CAP to Pep, K and n, were 1.7 × 10⁶ L mol⁻¹ and 0.87, respectively. The results showed that a 1 : 1 Pep–CAP complex was formed. The thermodynamic parameters of CAP to Pep were calculated using the van’t Hoff equation.⁴⁶ The results indicated that ΔH^o > 0, ΔS^o > 0 and ΔG^o < 0 at different temperatures (Table 1). It could be deduced that the binding force was mainly on the hydrophobic interaction.⁴⁷ The MD studies could give some insight into the protein–drug interactions.^48,49 In the presence of CAP, the docked conformation of Pep/CAP is shown in Fig. 7. The docked pose shows that CAP might enter into the active site cavity of Pep and form a hydrogen bond to Asp32 with the bond distance of 3.2 Å. For the Pep/CAP complex, the inhibition constants, free energy of binding and accessible surface area (ASA) were 8.7 × 10⁵ L mol⁻¹, −30.44 kJ mol⁻¹ and 119.38 Ų, respectively (Table 2).

Table 1 Thermodynamic parameters of Pep–CAP

Temperature/K	K/L mol^−1	n	ΔH/kJ mol^−1	ΔS/J mol^−1 K^−1	ΔG/kJ mol^−1
283	8.1 × 10^5	0.67			−32.03
288	9.8 × 10^5	0.75			−33.02
293	1.3 × 10^6	0.84	24.48	199.68	−34.03
298	1.7 × 10^6	0.87			−35.02
313	2.2 × 10^6	0.89			−38.02

---

Fig. 7 The structure of CAP binding to Pep.

Table 2 Binding results of CAP to Pep by FI-CL model and molecular docking

FI-CL		Molecular docking
n	0.87	Binding site (n)	Asp32
KE (L mol^−1)	1.7 × 10^6	Bind distance (Å)	3.2
ΔH^o (kJ mol^−1)	24.48	KM (L mol^−1)	8.7 × 10^5
ΔS^o (kJ mol^−1 K^−1)	199.68	ASA (Å^2)	119.38
ΔG^o (298.13 K) (kJ mol^−1)	−35.03	ΔG^o (298.13 K) (kJ mol^−1)	−30.44

---

In the luminol–HAuCl₄–Pep/CAP CL system, CAP solution was first mixed with Pep solution under neutral conditions, and then mixed successively with HAuCl₄ solution, and alkaline luminol solution (pH = 10.5). In the profile of the luminol–HAuCl₄–Pep/CAP CL system, it was found that the CL intensity of luminol–HAuCl₄–Pep/CAP could effectively be inhibited compared with the CL intensity of luminol–HAuCl₄–Pep. The reason might be attributed to the Pep–CAP interaction. In the Pep conformation, the Asp32 is a negatively charged polar residue located on the interface cleft of Pep.⁵⁰ For the structure of CAP, the thiol group (–SH) and carboxyl group (–COOH) make it easy to form a hydrophilic microenvironment due to the strong electron-withdrawing effect. In the Pep/CAP complex (Fig. 7), the O atom of the carbonyl group in CAP could bind to the Asp32 of Pep with a hydrogen bond and enter into the hydrophilic center of Pep, which is formed by the two activity sites, Asp32 and Asp215 on each cleft of Pep. The interaction of Asp32 and the carbonyl group was accessible to protonation between Pep and CAP. The thiol group of CAP as the special group plays a key function for drug-efficacy in vivo. The thiol group in the Pep/CAP complex had no binding with other groups of Pep. These results indicate that the bioactivity of CAP is not influenced by the activity center of Pep, when in the process of protein–drug interaction between Pep and CAP. The flexible loop formed with Asp32 and Asp215, which is commonly known as the “flap”, could induce the conformational change of Pep.⁵¹ On Pep–CAP interaction, the amino acid residues on the Pep surface such as Asp11, Asp159, Glu4, Glu13 and Asp118 would be embedded into the interior of Pep. The abnormally high pK_a values of Asp11 and Asp159 gradually tended to normal value due to the protonation process between Pep and CAP. This change would directly promote reduction potential reduction, which could inhibit Au atom aggregation on the surface of Pep with other amino acid residues, produce the CL quenching effect.

In short, the possible mechanism for producing sub-nanometre AuNCs, enhancing the CL intensity and inhibiting the CL intensity on adding CAP might be as follows:

(1) Under alkaline conditions, Au³⁺ and negatively charged amino acid residues could form a Pep–Au³⁺ complex on the surface of Pep. Due to the unstable conformation of Pep in alkaline solution, the negatively charged polar residues with abnormally high pK_a values could promote Au³⁺ flow into the vicinity of negatively charged amino acid residues on the principle of charge density matching and form sub-nanometre AuNCs.^52,53 Meanwhile, microscopic changes of Pep confirmation could accelerate the electron transfer rate of excited 3-aminophthalate, giving the enhancement of the CL intensity of luminol and producing the complex enhancement effect of CL (CEC).

(2) CAP could bind to Asp32 with a hydrogen bond, and enter into the hydrophilic center of Pep, leading to Pep’s conformational change, reducing the reduction potential on Pep’s surface. Based on the cascading effect of Pep/CAP interaction, the electron transfer rate of excited 3-aminophthalate was inhibited, and produced the complex quench effect of CL (CQC).

CL experimental conditions. In order to obtain the optimum performance for developing the luminol–HAuCl₄–Pep FI-CL approach, the concentration of luminol, NaOH, HAuCl₄, Pep, the flow rate and mixing tube length were systematically optimized for the whole experiment (Table 3). The effects of luminol concentration from 5.0 × 10⁻⁷ to 2.5 × 10⁻⁴ mol L⁻¹ were tested. It was found that on increasing the luminol concentration, the CL signal increased steadily until a luminol concentration of 2.5 × 10⁻⁴ mol L⁻¹, and tended to be stable, thus 2.5 × 10⁻⁴ mol L⁻¹ was chosen as the optimum luminol concentration. Due to the alkaline medium-dependent nature of the luminol CL reaction,⁵⁴ NaOH solution with concentrations ranging from 5.0 × 10⁻³ to 2.0 × 10⁻¹ mol L⁻¹ were tested. It was found that NaOH had the ability to increase the sensitivity of the CL system; 2.5 × 10⁻² mol L⁻¹ NaOH was finally chosen as the optimum concentration. Pep had a linear relationship for enhancing the luminol CL intensity from 1.0 × 10⁻¹⁰ mol L⁻¹ to 1.0 × 10⁻⁶ mol L⁻¹. Based on the fundament of the luminol–Pep CL response, HAuCl₄ could sharply increase the luminol–Pep CL signal in the range of 1.0 × 10⁻⁶ to 1.5 × 10⁻⁴ mol L⁻¹ in the luminol–HAuCl₄–Pep CL system. The concentration of Pep and HAuCl₄ were finally set as 1.0 × 10⁻⁶ mol L⁻¹ and 2.5 × 10⁻⁵ mol L⁻¹, respectively, for the consideration of the whole system of sensitivity, background noise and possible pipe clogging effect of the AuNCs. Meanwhile, the flow rate of the system was the key factor for obtaining a good sensitivity, signal-to-noise, and preventing the CL spectrum broadening; the flow rate was set at 2.0 mL min⁻¹. The mixing tube was set at 10.0 cm for good sensitivity and reproducibility.

Table 3 Optimum performance of luminol–HAuCl₄–Pep FI-CL system

Optimum factors	Optimum range	Final setting
Luminol	5.0 × 10^−7 to 2.5 × 10^−4 mol L^−1	2.5 × 10^−4 mol L^−1
NaOH	5.0 × 10^−3 to 2.0 × 10^−1 mol L^−1	2.5 × 10^−2 mol L^−1
HAuCl4	1.0 × 10^−6 to 1.5 × 10^−4 mol L^−1	2.5 × 10^−5 mol L^−1
Pep	1.0 × 10^−10 to 1.0 × 10^−6 mol L^−1	1.0 × 10^−6 mol L^−1
Flow rate	1.0–5.0 mL min^−1	2.0 mL min^−1
Mixing tube	5–20 cm	10 cm

---

In vivo recovery of CAP from microdialysis probe. Microdialysis as a micro-flow system has been successfully used for continuous in vivo sampling in biomedical, pharmacological and neuroscience studies.^55,56 In the process of sample collection, the flow rate is the key factor for the dialysis recovery ratio. Here, in order to obtain a better microdialysis recovery ratio, different flow rates of dialysate were studied under fixed CAP concentration (210.0 pg mL⁻¹). The results are shown in Fig. 8; the recovery ratio largely decreased from 53% to 38.9%, while the flow rate increased from 0.05 to 3.0 μL min⁻¹. After 3.0 μL min⁻¹, the recovery ratio reached a near-steady-state period. The reason for this result would be that, at the low flow rate, concentration was the key factor; while for the high flow rate, it was mainly controlled by the size of the porous membrane.⁵⁷ With comprehensive consideration of various factors, such as recovery ratio, dialysis pressure, cut-off ratio and membrane permeation, the flow rate was eventually set as 3.0 μL min⁻¹ for the whole microdialysis experiment.

Fig. 8 Relationship curve between flow rate and recovery ratio for a fixed concentration of CAP (210.0 pg mL⁻¹) in the microdialysis system.

It is well known that the efficiency of microdialysis is mainly controlled by some distinct factors, which include surface area of the dialysis membrane, molecule cut-off rate and CAP concentration. The length of the blood probe membrane was 2.5 cm and the molecular weight of CAP was 217.29, which was much less than the maximum cut-off limitation (3.5 kD). To evaluate the in vivo recovery, the microdialysis probe was placed in three unknown concentrations of CAP blood (CAP standard solution + blank serum) (2.1, 21.0, 210.0 and 2100.0 pg mL⁻¹) with 3.0 μL min⁻¹ of ACD solution for microdialysis to determine the factual CAP concentration according to the linear relationship of the luminol–HAuCl₄–Pep CL system. The results of average recoveries of three different CAP concentrations were 36.8 ± 0.02, 38.9 ± 0.05, 41.2 ± 0.08 and 42.8 ± 0.13% in blood, respectively. The data indicated that the recovery for microdialysis probes in blood shared no significant differences among the concentration ranges of 2.1, 21.0, 210.0 and 2100.0 pg mL⁻¹. These results suggest that the recovery for microdialysis probes is concentration independent, which could be applied to the study of CAP pharmacokinetics in rabbits.

Method validation

Linearity and LOD. A series of CAP standard solutions were injected into the manifold depicted in the above mentioned Fig. 2 with the different CL systems of luminol–dissolved oxygen, luminol–HAuCl₄, luminol–Pep and luminol–HAuCl₄–Pep. The decrement of the CL was proportional to the concentration of CAP, and the correspondent linear equation and limit of detection (LOD) are listed in Table 4. Compared with the luminol–dissolved oxygen and luminol–HAuCl₄ systems, the luminol–Pep CL system had a good sensitivity of 2.3 pmol L⁻¹ for CAP. Apparently, the luminol–HAuCl₄–Pep CL system has a wide linear range and good sensitivity for CAP determined from 3.0 pmol L⁻¹ to 0.1 μmol L⁻¹, with a LOD of 1.0 pmol L⁻¹.

Table 4 Linear and LOD for CAP in different luminol FI-CL systems

FI-CL	Linear equation	Linear range	LOD	R^2
Dissolved oxygen	ΔI = 11.205 ln C + 60.96	10.0 nM to 30 μM	3.0 nM	0.998
Pep	ΔI = 35.40 ln C + 948.5	7.0 pM to 3.0 nM	2.3 pM	0.983
HAuCl4	ΔI = 53.08 ln C + 1445.2	30.0 pM to 100.0 nM	10.0 pM	0.988
HAuCl4–Pep	ΔI = 125.63 ln C + 3310.2	3.0 pM to 0.1 μM	1.0 pM	0.995

---

Stability, precision and accuracy. The operational stability were tested for 210.0, 21.0 and 2.1 pg mL⁻¹ in the correspondent FI-CL system, and the relative CL intensities (ΔI = I_s − I₀) were recorded. The experiments were performed for 5 days with the FI-CL system regularly used over 8 h per day and the results are listed in Table 5. It was found that ΔI remained stable under the fluctuation of I₀ and the RSD were less than 5.0% with satisfying stability of the correspondent FI-CL system. At a flow rate of 2.0 mL min⁻¹, a complete determination of CAP, including sampling and washing, could be accomplished in 0.5 min, given a throughout of 120 h⁻¹ with a RSD of less than 5.0%. Inter-day and intra-day precision and accuracy data are shown in Table 6. The RSD of inter-day and intra-day precision for different CAP concentrations were less than 4.78, and the recovery rate for different concentrations of CAP in ACD solution and blank dialysis were more than 95.42%, indicating the overall reproducibility, precision and accuracy of the proposed method.

Table 5 Stability of the FI-CL system under different concentrations of CAP

Time day	Io	RSD%	Is (2.1)/pg mL^−1	RSD%	Is (21.0)/pg mL^−1	RSD%	Is (210.0)/pg mL^−1	RSD%
1^st	1132	1.5	1004	1.6	715	2.6	426	3.8
2^nd	1125	1.8	997	2.3	708	3.1	419	3.5
3^rd	1122	1.9	994	2.6	705	3.2	416	4.1
4^th	1135	2.1	1007	2.3	718	2.3	429	4.3
5^th	1130	2.3	1002	2.0	713	2.5	424	4.6

---

Table 6 Precision and accuracy of the proposed method

QC samples/pg mL^−1	Precision (n = 5)		ACD solution (n = 5)		Blank dialysis (n = 5)
	Inter-day	Intra-day	Found	Recovery	Found	Recovery
2.10	2.08 ± 0.12	2.06 ± 0.22	2.08 ± 0.12	99.04%	2.02 ± 0.19	96.19%
21.0	20.32 ± 1.69	20.09 ± 2.89	20.32 ± 1.69	96.76%	20.04 ± 1.95	95.42%
210.0	202.87 ± 3.89	196.87 ± 4.78	202.87 ± 3.89	96.60%	200.96 ± 4.61	95.69%

---

Interference studies. The interference of foreign species was tested by analyzing a standard solution of CAP into which increasing amounts of potential interfering substances were added. The tolerable concentrations of foreign species with respect to 50 nmol L⁻¹ CAP for interference at the 5.0% level were less than 100 μmol L⁻¹ for methanol and ethanol; 5.0 μmol L⁻¹ for I⁻, SO₄²⁻, PO₄³⁻, BrO₃⁻, glucose and citric acid; 3.0 μmol L⁻¹ for Mg²⁺, Ca²⁺, Zn²⁺ and Ba²⁺; and Fe³⁺/Fe²⁺; 160 μmol L⁻¹ for chloral hydrate, respectively.

In order to eliminate disturbance of the dialysate and improve the sensitivity of CAP determination in rabbit’s blood dialysate, the interference of blank blood dialysate was tested by diluting serial blank blood dialysate with purified water to gain similar intensity to that of the luminol–HAuCl₄–Pep CL. In the presence of blank blood dialysate, the relative CL intensity in the luminol–HAuCl₄–Pep system was approximately equal to the CL intensity in the absence of blank blood while diluted 5 × 10³ times with purified water.

Pharmacokinetics of CAP in blood. Following the above method described, an aliquot (20 μL) of rabbit blood dialysate was determined after appropriate dilution (dilution factor = 5 × 10³); the results are listed in Table 7. It could be seen that the recovery ratios vary from 89.0% to 112% with RSDs less than 5.0%. The CAP concentration–time fitting curves for 5 rabbits after i.g. administration are shown in Fig. 9.

Table 7 Determination of CAP in rabbit blood for 16 h^a

Interval/h	Added/found pg mL^−1	RSD%	Recovery%	Content/ng mL^−1	Eliminate%
a 2.32 mg CAP was administered i.g. to rabbit (2.0 kg).
0.00	0/—	—	99.3	—	—
	2.10/2.02	0.5
0.08	0/3.15	0.9	98.28	18.53 ± 0.83	0.01
	21.00/24.09	0.6
0.16	0/28.51	2.8	106.43	40.07 ± 0.84	0.19
	21.0/51.34	1.2
0.25	0/78.42	2.3	118.43	59.51 ± 0.79	0.38
	210.0/302.87	1.2
0.50	0/172.55	1.8	96.29	171.22 ± 0.72	0.89
	210.0/376.15	1.8
0.75	0/151.84	2.7	107.38	411.55 ± 0.53	1.87
	210.0/373.05	1.9
1.00	0/129.7	1.9	99.44	1003.59 ± 0.47	8.02
	210.0/339.00	3.2
1.50	0/103.12	1.6	105.17	1595.53 ± 0.09	12.93
	210.0/318.45	2.5
2.00	0/83.39	1.9	104.44	1277.57 ± 0.14	9.14
	210.0/297.09	3.3
3.00	0/57.83	1.6	100.10	942.46 ± 0.29	4.38
	210.0/267.89	4.1
4.00	0/36.26	2.8	109.09	661.09 ± 0.35	3.12
	21.0/60.56	3.7
5.00	0/23.58	2.6	109.57	460.35 ± 0.39	2.26
	21.0/46.84	3.4
6.00	0/7.20	2.9	115.82	320.30 ± 0.44	1.42
	2.1/10.44	4.6
8.00	0/3.97	4.1	107.17	155.87 ± 0.58	0.73
	2.1/6.35	4.2
10.00	0/1.16	3.5	113.97	75.18 ± 0.63	0.45
	2.1/3.42	3.2
12.00	0/0.63	2.8	95.14	36.45 ± 0.87	0.26
	2.1/2.70	4.0
14.00	0/0.44	3.5	98.53	18.54 ± 0.89	0.14
	2.1/2.53	4.3
16.00	0/0.31	3.6	100.38	8.15 ± 1.02	0.05
	2.1/2.41	2.8
Total elimination rate					45.9

---

Fig. 9 CAP concentration–time fitting curves for 5 rabbits after intragastric (i.g.) administration.

The pharmacokinetic parameters for CAP in dialysate were obtained with the aid of DAS soft. Various parameters such as area under curve (AUC), peak plasma concentration (C_max), time to reach the peak (T_max), and elimination rate constant (K_el), absorption and elimination half-life (T_1/2), the total mean residence time (MRT) and absorption efficiency were calculated for each rabbit. AUC_0–∞ was calculated as AUC_0–t + C_last/K_el, where C_last is the last measurable concentration. The volume of distribution was obtained as dose/AUC_0–∞. The mean pharmacokinetic parameters for 5 rabbits are listed in Table 8. The results showed that the maximum peak concentration of CAP was 9.63 ± 1.45 μg mL⁻¹ at 0.75 ± 0.08 h and a detectable quantity (6.23 ± 0.4 pg mL⁻¹) was found at 16 h after its i.g. administration, which was higher than the limit of quantity (LOQ) of the analytical approach. Meanwhile, the value of the higher AUC_0–t (613.16 ± 5.09 mg L⁻¹ h⁻¹), V₁/F (2.02 ± 0.17 L g⁻¹) and lower MRT_0–t time (12.47 ± 0.39 h) showed that CAP has a strong ability to penetrate biological membranes and is mainly distributed in the rich blood of organs and tissues in vivo, excreted with the prototype with CL/F (7.27 ± 0.41 L g⁻¹ h⁻¹). After a single i.g. of CAP, it could rapidly be absorbed from the gastro-intestinal tract with peak blood level of 0.8 μg mL⁻¹ in about an hour period with the minimal absorption of 75%. Up to 50% of CAP was metabolized through the liver, oxidized to its respective disulfides, and the remaining was excreted in the urine.^58,59 The results showed that CAP could be in line with the two-compartment open model in vivo.

Table 8 Pharmacokinetic parameters of CAP in rabbits after intragastric (i.g.) administration (n = 6)^a

Parameters	Sample no.
	1^#	2^#	3^#	4^#	5^#	M ± SD
a AUC0–t, under the curve up to the last time (t); Tmax, the time to reach peak concentration; Cmax, the maximum plasma concentration; V1/F, the apparent volume of distribution; K, the apparent rate constant; t1/2α and t1/2β, the apparent absorption and elimination half-life; MRT0–t, mean residence time.
t1/2α (h)	0.187	0.203	0.219	0.196	0.256	0.212 ± 0.07
t1/2β (h)	3.56	3.17	3.24	2.88	3.11	3.192 ± 0.13
V1/F (L g^−1)	3.24	2.84	3.02	3.15	2.87	2.02 ± 0.17
CL/F (L g^−1 h^−1)	8.12	7.01	6.35	7.24	7.63	7.27 ± 0.41
AUC0–t (mg L^−1 h^−1)	569.42	486.13	721.36	688.29	600.58	613.16 ± 5.09
Tmax (h)	0.79	0.72	0.76	0.75	0.71	0.75 ± 0.08
Cmax (μg mL^−1)	7.34	8.18	9.24	11.36	11.02	9.63 ± 1.45
MRT0–t (h)	12.36	11.45	13.27	12.22	13.06	12.47 ± 0.39

---

Statistical analysis of the main pharmacokinetic parameters, such as AUC_0–t, V₁/F, MRT_0–t, CL/F and C_max was performed using one-way analysis of variance (ANOVA) followed by Spss19.0 soft. The statistical results showed that the pharmacokinetic parameters had no significant specific difference (P > 0.05).

Conclusions

Based on the luminol CL intensity enhanced by Pep and HAuCl₄, a luminol–HAuCl₄–Pep FI-CL system was constructed for the first time. In the proposed FI-CL system, sub-nanometre AuNCs were generated in the luminol–HAuCl₄–Pep CL reaction solution, which could potentially enhance the CL intensity; a possible mechanism for enhancing the CL intensity with HAuCl₄ and Pep was given. This proposed FI-CL approach was successfully applied to study the CAP pharmacokinetics in rabbits with microdialysis. The results showed that CAP has a C_max (9.63 ± 1.45 μg mL⁻¹) at T_max (0.75 ± 0.08 h), corresponding to t₁/_2β (3.19 ± 0.13 h) and CL/F (7.27 ± 0.41 L g⁻¹ h⁻¹) in rabbits in vivo. The pharmacokinetic results showed that CAP fits a two-compartment open model in rabbits.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21275118) and the Open Fund from the Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education.

References

1. J. P. Renaud and M. A. Delsuc, Curr. Opin. Pharmacol., 2009, 9, 622–628.
2. W. F. J. de Azevedo, R. A. Caceres, I. Pauli, L. F. Timmers, G. B. Barcellos, K. B. Rocha and M. B. Soares, Curr. Drug Targets, 2009, 10, 271–278.
3. S. M. Huang, H. Zhao, J. I. Lee, K. Reynolds, L. Zhang, R. Temple and L. J. Lesko, Clin. Pharmacol. Ther., 2010, 87, 497–503.
4. R. Minai, Y. Matsuo, H. Onuki and H. Hirota, Proteins: Struct., Funct., Bioinf., 2008, 72, 367–381.
5. J. Colinge, U. Rix, K. L. Bennett and G. Superti-Furga, Proteomics Clin. Appl., 2012, 6, 102–116.
6. B. M. Dunn, Chem. Rev., 2002, 102, 4431–4458.
7. D. R. Dee and R. Y. Yada, Biochemistry, 2010, 49, 365–371.
8. H. M. Zhang, J. Cao, Z. H. Fei and Y. Q. Wang, J. Mol. Struct., 2012, 1021, 34–39.
9. H. J. Zeng, T. T. Qi, R. Yang, J. You and L. B. Qu, J. Fluoresc., 2014, 24, 1031–1040.
10. S. Q. Lian, G. R. Wang, L. P. Zhou and D. Z. Yang, Luminescence, 2013, 28, 967–972.
11. H. J. Zhang, T. Watanabe, M. Okumura, M. Haruta and N. Toshima, Nat. Mater., 2011, 11, 49–52.
12. H. Häkkinen, S. Abbet, A. Sanchez, U. Heiz and U. Landman, Angew. Chem., Int. Ed., 2003, 42, 1297–1300.
13. W. Chen and S. W. Chen, Angew. Chem., Int. Ed., 2009, 48, 4386–4389.
14. Z. W. Wang and R. E. Palmer, Nano Lett., 2012, 12, 91–95.
15. C. A. J. Lin, T. Y. Yang, C. H. Lee, S. H. Huang, R. A. Sperling, M. Zanella, J. K. Li, J. L. Shen, H. H. Wang and H. I. Yeh, ACS Nano, 2009, 3, 395–401.
16. S. Palmal and N. R. Jana, Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol., 2014, 6, 102–110.
17. H. H. Wang, C. A. J. Lin, C. H. Lee, Y. C. Lin, Y. M. Tseng, C. L. Hsieh, C. H. Chen, C. H. Tsai, C. T. Hsieh and J. L. Shen, ACS Nano, 2011, 5, 4337–4344.
18. W. Y. Chen, J. Y. Lin, W. J. Chen, L. Y. Luo, D. E. Wei-Guang and Y. C. Chen, Nanomedicine, 2010, 5, 755–764.
19. Z. K. Wu, M. A. MacDonald, J. Chen, P. Zhang and R. C. Jin, J. Am. Chem. Soc., 2011, 133, 9670–9673.
20. C. L. Liu, H. T. Wu, Y. H. Hsiao, C. W. Lai, C. W. Shih, Y. K. Peng, K. C. Tang, H. W. Chang, Y. C. Chien and J. K. Hsiao, Angew. Chem., Int. Ed., 2011, 50, 7056–7060.
21. R. C. Jin, H. F. Qian, Z. K. Wu, Y. Zhu, M. Z. Zhu, A. Mohanty and N. Garg, J. Phys. Chem. Lett., 2010, 1, 2903–2910.
22. Z. K. Wu and R. C. Jin, Nano Lett., 2010, 10, 2568–2573.
23. G. Y. Liu, Y. Shao, K. Ma, Q. H. Cui, F. Wu and S. J. Xu, Gold Bull., 2012, 45, 69–74.
24. J. George and K. G. Thomas, J. Am. Chem. Soc., 2010, 132, 2502–2503.
25. J. P. Xie, Y. G. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888–889.
26. H. Kawasaki, K. Hamaguchi, I. Osaka and R. Arakawa, Adv. Funct. Mater., 2011, 21, 3508–3515.
27. L. Yan, Y. Q. Cai, B. Z. Zheng, H. Y. Yuan, Y. Guo, D. Xiao and M. M. F. Choi, J. Mater. Chem., 2012, 22, 1000–1005.
28. H. Yabu, Chem. Commun., 2011, 47, 1196–1197.
29. S. H. Lee, A. G. Larsen, K. Ohkubo, Z. L. Cai, J. R. Reimers, S. Fukuzumi and M. J. Crossley, Chem. Sci., 2012, 3, 257–269.
30. H. Zeng, X. W. Du, S. C. Singh, S. A. Kulinich, S. Yang, J. He and W. Cai, Adv. Funct. Mater., 2012, 22, 1333–1353.
31. R. F. Popovici, E. M. Seftel, G. D. Mihai, E. Popovici and V. A. Voicu, J. Pharm. Sci., 2011, 100, 704–714.
32. W. M. M. Mahmoud and K. Kümmerer, Chemosphere, 2012, 88, 1170–1177.
33. N. Rastkari, M. Khoobi, A. Shafiee, M. R. Khoshay and R. Ahmadkhaniha, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2013, 932, 144–151.
34. P. Khan, D. Idress, M. A. Moxley, J. A. Corbett, F. Ahmad, G. V. Figura, W. S. Sly, A. Waheed and M. I. Hassan, Appl. Biochem. Biotechnol., 2014, 173, 333–355.
35. P. D. Tzanavaras, Anal. Lett., 2011, 44, 560–576.
36. M. Iranifam, TrAC, Trends Anal. Chem., 2013, 51, 51–70.
37. Q. Q. Li, L. J. Zhang, J. G. Li and C. Lu, TrAC, Trends Anal. Chem., 2011, 30, 401–413.
38. J. Wang, S. H. Zhang, J. Zhang, P. Ren, Y. C. Chen, J. H. Li, W. Wang, Y. Ma, R. F. Shi and C. H. Wang, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2011, 879, 1605–1609.
39. X. J. Tan, Z. M. Wang, D. H. Chen, K. Luo, X. Y. Xiong and Z. H. Song, Chemosphere, 2014, 108, 26–32.
40. Z. F. Zhang, H. Cui, C. Z. Lai and L. J. Liu, Anal. Chem., 2005, 77, 3324–3329.
41. Z. Wu, M. A. MacDonald, J. Chen, P. Zhang and R. Jin, J. Am. Chem. Soc., 2011, 133, 9670–9673.
42. Y. S. Chen and P. V. Kamat, J. Am. Chem. Soc., 2014, 136, 6075–6082.
43. T. P. Yoon, M. A. Ischay and J. Du, Nat. Chem., 2010, 2, 527–532.
44. K. Luo, X. H. Zheng and Z. H. Song, RSC Adv., 2014, 4, 45230–45233.
45. Z. M. Wang and Z. H. Song, Analyst, 2010, 135, 2546–2553.
46. O. Cañadas and C. Casals, Methods Mol. Biol., 2013, 974, 55–71.
47. X. Y. Gao, Y. C. Tang, W. Q. Rong, X. P. Zhang, W. J. Zhao and Y. Q. Zi, Am. J. Anal. Chem., 2011, 2, 250–257.
48. Y. Chen and B. K. Shoichet, Nat. Chem. Biol., 2009, 5, 358–364.
49. A. S. El-Azab, M. A. Al-Omar, A. A. M. Abdel-Aziz, N. I. Abdel-Aziz, M. A. A. El-Sayed, A. M. Aleisa, M. M. Sayed-Ahmed and S. G. Abdel-Hamide, Eur. J. Med. Chem., 2010, 45, 4188–4198.
50. D. Dee, J. Pencer, M. P. Nieh, S. Krueger, J. Katsaras and R. Y. Yada, Biochemistry, 2006, 45, 13982–13992.
51. J. R. Engen, Anal. Chem., 2009, 81, 7870–7875.
52. J. W. Liu, TrAC, Trends Anal. Chem., 2014, 58, 99–111.
53. J. M. Slocik, A. O. Govorov and R. R. Naik, Nano Lett., 2011, 11, 701–705.
54. H. Albrecht, Z. Phys. Chem., 1928, 136, 321–330.
55. P. Nandi and S. M. Lunte, Anal. Chim. Acta, 2009, 651, 1–14.
56. H. Ronald Zielke, C. L. Zielke and P. J. Baab, J. Neurochem., 2009, 109, 24–29.
57. D. Feuerstein, A. Manning, P. Hashemi, R. Bhatia, M. Fabricius, C. Tolias, C. Pahl, M. Ervine, A. J. Strong and M. G. Boutelle, J. Cereb. Blood Flow Metab., 2010, 30, 1343–1355.
58. A. Luczak, R. Zakrzewski and M. Dobrogowski, Bioanalysis, 2012, 4, 1481–1489.
59. G. Donáth-Nagy, S. Vancea and S. Imre, Croat. Chem. Acta, 2011, 84, 423–427.

---