1. Introduction
Acridinium aromatic esters (AEs) belong to the systems capable of efficient chemical transformation into electronically excited products, i.e., chemiluminescence (CL) [
1,
2]. Unlike popular chemiluminogens such as derivatives of luminol, the main advantages of AEs are the relatively high quantum yield of emission with a low background signal and not needing the use of a catalyst for initiating CL, as well as relatively quick and easy control of its dynamics [
3,
4]. All these aspects have rendered AEs broadly employed as CL labels or indicators in medical/pharmaceutical [
5,
6], chemical and biochemical [
7,
8], and environmental and food analytics today [
9]. CL labels based on them have been used to determine the concentration of hormones [
10,
11], antibodies and antigens [
12,
13,
14], enzymes [
15], antioxidants [
4,
16] and nucleic acid fragments [
17] at an ultra-sensitive level (limits of detection at the level of 10
−19 mole of analyte or below [
18]).
Our previous computational studies have indicated that AEs readily undergo oxidation in the presence of hydrogen peroxide in alkaline media, initiating the emission of light (
Figure 1) [
19,
20]. In the first step of this process, the OOH
− ions, formed from the oxidant H
2O
2 in an alkaline environment, attack the electrophilic centre of AE, i.e., the carbon atom in position 9 of the acridinium moiety (the C atom with the highest electron deficit [
19,
20]). Next, due to bond rearrangements, a cyclic intermediate product is formed: the acridane dioxethanone derivative. Its decarboxylation occurs, and an electronically excited product, 10 substituted-acridan-9-one, is formed. Returning to the ground state, the latter emits radiation in the visible range (ca. 450 nm in water).
One of the key parameters that significantly influences the efficiency of the emission of light of AEs is the appropriate acid–base equilibrium of its aqueous solution before the initiation of CL (
Figure 1). At a high pH, so-called pseudobases of AE are formed [
20]. The latter reacts slowly with OOH
− ions, but this process does not produce electronically excited CL products [
19,
20]. In an alkaline environment, AEs can undergo hydrolysis to form 10 substituted-9-carboxy acridinium acid salt [
20]. At low pH, the acid–base equilibrium is shifted towards the acridinium cation. The effectiveness of CL is higher when the cationic form of AE is at the maximum level in the system. Therefore, slightly acidic conditions must be maintained before the CL emission is started.
One way to increase the efficiency of the CL process is to use organised media, which can play a significant role by accelerating the transformation’s kinetics and increasing the emission efficiency, resulting in higher sensitivity of luminometric measurements [
21].
Organised media are widely employed in chemical synthesis and analysis to modify the solubility and environment, appropriately orient solute molecules of a bulk solution, enhance reactivity and control the course of chemical or photochemical transformations and biological processes [
22,
23,
24,
25,
26]. The two types of organised media were researched to improve the CL efficiency of chemiluminogenic acridinium salts [
21]. These were micellar media formed by surfactants, which, at specific concentrations exceeding the critical micelle concentration (CMC), combine forming aggregates termed micelles and cyclodextrins consisting of a macro-cyclic structure of glucose subunits joined by α-1,4-glycosidic bonds. Surfactants can be divided into four main types depending on their charge and structure: cationic, anionic, amphoteric and nonionic [
22]. Their specific physicochemical properties, such as their propensity to undergo self-aggregation processes leading to micelle formation, are related to their ability to interact between the hydrophobic and hydrophilic parts of the solute molecule [
23]. The equilibrium between micelles and monomer molecules can be affected by external factors, leading to changes in micelle stability and, thus, changes in the properties of a given system. The main factors influencing the CMC of surfactants include pH, temperature, ionic strength, solvent properties, electrolytes or non-electrolytes, and the technique used to measure the CMC [
27,
28,
29]. Therefore, from a practical point of view, it is essential to control the form (monomers or micelles) in which the surfactant molecules are present in the system under investigation. According to that, the determination of the CMC is a widely researched topic today, and various methods are available for identifying the micellar environment [
30].
In the 1980s, Grayeski and co-workers [
31,
32] reported that the efficiency of the CL reaction of acridine derivatives was increased by adding cyclodextrins as well as surfactants such as cetyltrimethylammonium chloride (CTAC) to the solution. The latter is the most commonly used micellar medium, which reduces the emission time of acridinium CL to less than 5 s and substantially increases its efficiency [
33,
34,
35]. Natrajan’s group, investigating various types of surfactants, suggested that they affect light emission from AEs through two general mechanisms, depending on their charge and the influence of their micelles on mono- or bimolecular types of transformations [
33]. The studies indicated that the maximum surfactant effect is obtained for the relatively hydrophobic acridinium moiety. However, it was also noticed that hydrophilic functional groups (in particular, sulfobetaine zwitterions) substituted in acridinium moiety in the presence of CTAC reduced the emission time while increasing the CL efficiency. The most significant improvement in the CL parameters of acridinium derivatives was achieved when the hydrophobic acridinium ring was combined with a hydrophilic leaving (phenyl) group [
34]. The same group of researchers obtained interesting results by investigating the CL properties of fluorinated acridinium labels (AL) in the presence of cationic (CTAC) and anionic agents (sodium perfluorooctanoate, SPFO) and their mixtures. The studies indicated that the mixed system (CTAC and SPFO) at low mole fractions of SPFO leads to enhanced emission [
36]. The authors suggested that enhancement in AL light yield is sensitive to the polarity of the micellar interface.
However, it should be considered that the mentioned agents (surfactants) can be toxic, especially to aquatic organisms, when released into the environment. In this context, degradable cationic surfactants containing amide and carbonate groups can effectively substitute classical surfactants. In particular, the carbonate surfactant (dodecyl-3-(trimethylazane)propyl carbonate chloride) is especially promising because it mimics the behaviour of CTAC in affecting the CL of AE. The latter substance has limited stability in the final (post-reaction) mixture but is characterised by excellent long-term stability in acidic solutions. Therefore, it could effectively replace CTAC in systems to enhance the CL of AEs in automated immunoassays, being a more environmentally friendly alternative to CTAC [
37].
Although the noticeable effect of surfactants on improving the CL efficiency of AEs is demonstrated in many experimental studies [
20,
31,
32,
33,
34,
35,
36,
37], there are no reports explaining at the molecular level why surfactants significantly alter the emissive properties of acridinium salts. The answer to the last question became the motive to undertake the proposed research experimentally and computationally. Molecular dynamics (MD)-based approaches can help characterise the potential interactions of surfactant molecules with intermediate products formed during the transformation of AEs in the process of their chemiluminogenic oxidation. To our knowledge, no computational studies have been conducted so far providing potential explanations for the atomistic mechanism of interaction of AEs with surfactants and their influence on the parameters of the CL process, which renders such an attempt very promising. In this work, we present a relatively simple chemiluminogenic compound—acridinium alkyl phenyl ester—as a representative and effective indicator to identify micellar systems and to distinguish the charge of micelle surfaces. The interactions between AE molecules and product(s) of its transformations with ionic surfactants may thus be a diagnostic tool for determining specific features of luminogenic systems, such as pre- or post-micellar concentration ranges, type of surfactant introduced, etc. The latter can be utilised in designing new CL systems with beneficial features potentially useful in various fields, such as pharmaceutical, biotechnology or materials science.
3. Materials and Methods
3.1. Reagents and Working Solutions
All commercial reagents of analytical grade were used without additional purification. Hexadecyltrimethylammonium chloride (CTAC, ≥99%), hexadecyltrimethylammonium bromide (CTAB, ≥99%), sodium dodecyl sulphate (SDS, ≥99%) and anhydrous acetonitrile were purchased from Sigma Aldrich (St. Louis, MI, USA). Analytical weights (0.1 M) of hydrochloric acid, nitric acid(V) and sodium hydroxide were purchased from Chempur (Piekary Śląskie, Poland). Hydrogen peroxide 30% (p.a.) was purchased from Stanlab (Lublin, Poland). Ultrapure water (conductivity below 0.2 μS cm−1) (Beckman, Brea, CA, USA) was used to prepare all solutions. Aqueous solutions of surfactants for CL and FL measurements were prepared to obtain values before and above the CMC point, namely, 0.4 mM/0.8 mM and 4 mM/8 mM for CTAC and CTAB, and 10 mM/20 mM/40 mM/60 mM for SDS. Solutions of surfactants (8 mM CTAC, 8 mM CTAB, and 80 mM SDS) for conductometric measurements were prepared in an aqueous mixture containing 0.25 mM HCl and 0.5 mM HNO3.
Before CL and FL measurements, 0.1% H
2O
2 in 1 mM HNO
3 and 0.2 M NaOH in water were prepared. The details concerning synthesis, chemical analyses and spectroscopic features (MS, NMR) of the chemiluminogenic substrate selected for this work, 10-methyl-9-[(2-methylphenoxy)carbonyl]acridinium trifluoromethanesulphonate (2MeX,
Figure 2), have been described in our former work [
19]. The 5 mM stock solution of 2MeX in anhydrous acetonitrile was kept in a freezer (253.15 K), and immediately before the experiments, the working solutions were prepared by its dilution to
c = 0.1 μM with 1 mM HCl.
3.2. Conductometric Measurements
Conductometric measurements were performed using a microtitration unit (Cerko Lab System, Gdynia, Poland). The unit included a 5 mL syringe (Hamilton, Gdynia, Poland) and a CD-201 conductometric cell (Hydromet, Gliwice, Poland). The syringe was calibrated using a weight calibration method. The conductometric electrode was standardised using standards, specifically aqueous KCl solutions with conductivities of 84 and 200 μS cm−1 (Hamilton, Poland). The measurements were conducted at 298.15 ± 0.10 K and controlled using the Lauda E100 circulation thermostat. The reagents (CTAC, CTAB, and SDS) were dissolved directly in the solution containing 0.25 mM HCl and 0.5 mM HNO3. The experiment consisted of injecting, at 20 s intervals, 0.01 mL of the titrant solution (8 mM CTAC/8 mM CTAB/80 mM SDS) into the reaction cell, which initially contained a 5 mL mixture of 0.25 mM HCl and 0.5 mM HNO3. Two measurements were taken for each system, and the resulting data were analysed using the computer program Excel (Microsoft 365, Version 2405).
3.3. Measurements of Chemiluminescence
Measurements of CL were performed using a Fluoroskan Ascent FL microplate reader (Labsystems, Vantaa, Finland) with the detector tailored on maximal sensitivity (PMT voltage = 1000 mV). A 25 μL solution of 2MeX (c = 0.1 μM in 1 mM HCl) and 25 μL of surfactant (CTAC, CTAB, and SDS) solution in water were distributed onto a 96-well white polystyrene plate. After placing the plate in the apparatus, from the first dispenser, 50 μL of 0.1% H2O2 in 1 mM HNO3 was added to each well, and the plate was then shaken (10 s) at 600 rpm and incubated at 298.15 K for 2 min. The CL was induced and measured by adding 50 μL of 0.2 M NaOH to each well (n = 5 replicates for each concentration). The resolution of CL measurements was set at 20 ms for systems containing CTAC/CTAB and 300 ms for systems with SDS. The number of points was tailored to obtain a complete reaction profile in each case (100–1000 points).
Each measurement of CL was repeated at least three times (n = 3–5) before the results were presented. The relative standard deviations (RSD) were calculated for each data group and are given in the captions denoting figures or tables. In all the CL data gathered for this work, the RSD factor was 0.3–6.7%. The results were processed using Excel software (Microsoft 365).
3.4. Measurements of Fluorescence
The FL emission spectra of post-CL reaction mixtures were recorded at room temperature using a Cary Eclipse spectrofluorimeter (Varian, Palo Alto, CA, USA) with the detector’s sensitivity (PMT voltage) set at 600 mV and excitation/emission slits of 5 nm. The spectra were recorded using a standard 1 cm quartz cuvette in the 375–650 nm range, with the excitation wavelength set at 365 nm. The tested mixtures contained 1 mL of 2MeX (c = 50 μM) in 1 mM HCl, 2 mL of 0.1% H2O2 in 1 mM HNO3, 2 mL of 0.2 M NaOH and 1 mL of 8 mM CTAC/8 mM CTAB/60 mM SDS or UP water. The FL spectra were recorded just after the completion of the CL emission (n = 5 replicates for each system) and processed with the aim of the Excel software (Microsoft 365).
UV–Vis absorption spectra of 2MeX acridinium ester in 1 mM HCl with participation of surfactants were registered using a Lambda 40 Perkin-Elmer spectrophotometer and standard 1 cm quartz cuvettes.
3.5. Molecular Dynamics-Based Analysis
The structures of 2MeX, 2Me-OOH, [2Me-OO]
−, Me-Aon and both surfactant molecules CTA and SDS were first built in Avogadro [
40,
41] and further optimised in Gaussian16 [
42] at DFT(B3LYP)/6-31G(d) [
43,
44,
45,
46] level of theory. Then, the RESP procedure [
47] was applied to calculate the atomic partial point charges of these molecules compatible with the gaff force field [
48] implemented in the AMBER20 package [
49]. The missing bonded parameters for the equilibrium bond and angle values for [2Me-OO]
− were directly obtained from the Gaussian geometry optimisation, and the force constants were chosen to be characteristic for these types of bonds and angles to keep the geometry in the molecular dynamics (MD) simulation in agreement with the results from quantum chemical optimisation (30 kcal mol
−1 Å
−2 and 10 kcal mol
−1 rad
−2, respectively).
MD simulations were performed in AMBER20 [
49] for eight systems, each of which consisted of one molecule of either 2MeX or 2Me-OOH or [2Me-OO]
− or Me-Aon (chemical species appearing on the reaction path [
19]) with 10 molecules of surfactant (CTAC or SDS) placed randomly. A truncated octahedron TIP3P periodic box of 15 Å water layer from the box’s border to solute was used to solvate the molecular systems. Cl
− or Na
+ counter ions were used to neutralise the system’s charge. Two energy minimisation steps were performed: first, 500 steepest descent cycles and 103 conjugate gradient cycles with 100 kcal mol
−1 Å
−2 harmonic force restraints on the solute, and second, 3 × 10
3 steepest descent cycles and 3 × 10
3 conjugate gradient cycles without any restraints. Then, the system was heated from 0 to 300 K for 10 ps with harmonic force restraints of 100 kcal mol
−1 Å
−2 solute. Finally, the system was equilibrated at a constant temperature of 300 K and constant pressure of 10
5 Pa for 100 ps. The productive MD run was performed in the same isothermal isobaric ensemble for 1 μs. The particle mesh Ewald method for treating electrostatics and the SHAKE algorithm for all the covalent bonds containing hydrogen atoms were applied. The obtained trajectories were analysed with the ccptraj module of AMBER20 [
49]. Linear interaction energy (LIE) binding free energy calculations used the dielectric constant of 80. The visualisation and statistical analysis were performed in VMD [
50] and R package [
51], respectively.
3.6. The DFT Calculations
Optimised geometry of the studied compounds was performed utilising DFT [
43] at the B3LYP [
44,
45] level of theory in conjunction with the 6-31G(d,p) [
52] basis sets. The harmonic vibrational frequencies, characterising the stationary points, were evaluated to ensure that the obtained structures correspond to true minima on the potential energy surface. The calculations were achieved in the selected solvent (aqueous phase) by applying the polarisable continuum model (PCM) [
53,
54] with the Gaussian16 program [
42], and the output files were visualised employing the ChemCraft program package [
55].
The kinetics of CL decays were determined using a classical graphical method, and the thermodynamic parameters (enthalpy and Gibbs free energies) of the studied processes were calculated from the classical Hess equation [
39].
4. Conclusions
Plate measurements of CL emission indicate interesting and different features of the acridinium ester-based aqueous system in the presence of ionic surfactants (CTAC, CTAB and SDS). This study generally aimed to explain, at the molecular level, the observed variabilities of the emission parameters of the chemiluminogenic substrate in the presence of the above surfactants. The selected substrate, 9-(2-methylphenoxy)carbonyl-10-methylacridinium triflate (2MeX), is characterised by good emission efficiency and stability in aqueous solutions, being employed in luminescence analysis as a CL indicator and fragment of CL labels.
The CL parameters varied with the type of surfactant (cationic/anionic) and the pre-and post-micellar environment. Cationic surfactants exert a pronounced influence on the emission dynamics of CL, with a moderate impact on its efficiency. In contrast, the anionic surfactant exhibited an opposite effect, significantly slowing the emission dynamics while enhancing its efficiency. The CL kinetic constants and integral efficiencies in the presence of ionic surfactants were assessed before and after the CMC points and determined conductmetrically at comparable conditions. It has been observed that near and above the point of CMC, there is a significant (ca. 18-fold) increase in the emission rate constant for CTAC and a ca. 15-fold decrease in this parameter for SDS.
Molecular dynamics analysis disclosed potential interactions of surfactant molecules with the intermediate products appearing during 2MeX chemiluminogenic transformations. The distances between the acridinium cation and the surfactant molecules and their number in the interacting aggregates were comparable for all the systems. At the same time, differences have been observed in the binding strength between 2MeX with CTAC and SDS, revealing that SDS molecules interact more strongly with the acridinium cation than CTAC ones.
Quantum chemical calculations at the DFT levels of theory have proven that the reaction of oxidant (e.g., H2O2 and its anionic form, OOH−) with an anionic surfactant (SDS) is not thermodynamically preferred, both in the aqueous and gaseous phases. Additionally, fluorescence emission measurements performed on the post-CL reaction mixtures excluded the significant influence of investigated surfactants on the emissive features of the final emitter, which is 10-methylacridan-9-one.
In summary, research carried out employing experimental methods (CL and FL measurements) and computational ones (MD and DFT) indicate that interactions leading to significant changes in the observed emissive properties of the acridinium chemiluminogenic system occur at the reaction’s mechanism stage in the H2O2/NaOH aqueous environment, i.e., as a result of different interactions of the CL substrate (acridinium cation) and products of its oxidation, with cationic or anionic surfactant molecules. At the atomistic level, differences between CTAC and SDS interactions with the acridine-based molecular systems were detected due to the charge differences of the used surfactants.
The study has provided significant insights into the role of surfactants in the emissive features of acridinium chemiluminescence. The results can contribute to the rational optimisation of new chemiluminogenic systems, underscoring their practical implications. They also indicate that the acridinium ester may be an original and specific indicator of the surfactant type (cationic vs. anionic) and the system type (pre- or post-micellar).