Microcalorimetry as an Effective Tool for the Determination of Thermodynamic Characteristics of Fulvic–Drug Interactions

Abstract

The presence of pharmaceuticals in the environment can result in potentially dangerous situations. In soils and sediments, pharmaceuticals can be partially immobilized by interactions with humic substances. Interactions, thus, can strongly affect their mobility and bioavailability. An investigation of the thermodynamic aspects of the interactions is largely missing. Thermodynamic parameters are usually calculated on the basis of sorption experiments. Our study is focused on the direct measurements of the heat effect of interactions between fulvic acids and chosen drugs. Well-characterized fulvic sample standards provided by the International Humic Substances Society were used. Ibuprofen, diclofenac, and sulphapyridine were chosen as drugs. Isothermal titration calorimetry provided a complete set of thermodynamic characteristics of underlying processes—interaction enthalpy, entropy, and Gibbs energy. All studied interactions were found to be exothermic with heat liberation between −496 and −9938 J/mol. The lowest enthalpies were obtained for sulphapyridine and the highest ones for ibuprofen (on average). Changes in Gibbs energy were very similar for all studied interactions (20–28 kJ/mol). The highest change in entropy was determined for ibuprofen (73 J/mol·K); values obtained for diclofenac and sulphapyridine were comparable (57 and 56 J/mol·K, respectively).

1. Introduction

Pharmaceuticals are indispensable in modern human and veterinary medicine to secure the health of a growing world population and the consumed livestock. They can appear in nature as a result of manufacturing processes, disposal of unused products, and animal excretion [1,2,3]. Therefore, they can be detected in natural waters, sediments and soils and recognized as emerging environmental contaminants [4,5,6,7,8,9]. They can undergo sorption on natural soil and water components, degradation, and chemical transformation into structurally similar compounds [10,11,12,13]. Drugs are not completely eliminated in the human body; therefore, their residues can leave the body and pass into water, sediments, and soils [14,15]. The potential degradation of drugs depends on their chemical character and physicochemical properties of their environment, e.g., soil properties and quality of organic matter. The migration ability and toxicity of pharmaceuticals in nature can be significantly affected by their interactions with soil components.

An investigation of thermodynamic aspects of the interactions between humic substances and drugs is relatively scarce. Interactions are studied mainly from the point of view of the adsorptions of pharmaceuticals on humic substances [10,11,13,16,17,18], interactions of drugs with dissolved humic substances [19,20], the effect of humic substances on the interactions of drugs with other sorbents [21,22,23,24,25], and the diffusion of drugs in model hydrogels with incorporated humic substances [26]. Isothermal titration calorimetry (ITC) was used for the direct measurement of interactions of humic substances in more studies, e.g., [27,28,29,30,31,32,33], which were focused mainly on their complexation with metal ions [27,28,29,30]. Some studies dealt with interactions with arsenite [31], TiO₂ [32], and lysozyme [33]. Khil’ko and Semenova [34] used ITC for the investigation of interactions of humic acid salts with papaverine, benzohexonium, and B-group vitamins. Determined heats of the complexation of sodium humates with the drug preparations in a ratio of 1:1 were very similar (–ΔH = 3.8–4.5 kJ/mol). The interactions occurred between the negatively charged functional groups of humic acids and the positively charged centers of drug molecules. Other interactions (such as hydrogen bonds and hydrophobic interactions) can also occur. The authors assumed that the number of moles of drug preparations capable of interacting with a natural polymer macromolecule is determined by their molecular structure peculiarities and charges. Xu et al. [35] studied the interaction between sulfamethazine (antibiotic) and fulvic acids by nuclear magnetic resonance combined with surface plasmon resonance and isothermal titration calorimetry. They stated that hydrophobic interactions together with π-π conjunction play significant roles in the binding. Based on calorimetry, they determined the binding enthalpy (ΔH) as −0.55 kJ/mol and calculated the change in Gibbs energy (ΔG) as −23.10 kJ/mol. Similar values of ΔG were obtained by means of other methods: −18.95 kJ/mol (nuclear magnetic resonance) and −20.85 kJ/mol (surface plasmon resonance). From the point of view of drug interactions, the highest attention is paid to serum albumin [36,37,38,39,40], hemoglobin [41], chitosan [1], and activated carbon [42]. Measured changes in enthalpy (ΔH) for serum albumin were between −19.6 and −27.2 kJ/mol for ibuprofen [37,40] and −12.9 and −37.5 kJ/mol for diclofenac [36,38,39]. The value of ΔH obtained for ibuprofen and hemoglobin was −18.8 kJ/mol [41], for diclofenac and chitosan +22.1 kJ/mol [1], and for diclofenac and activated carbon −25 kJ/mol [42]. As can be seen, all interactions except the interactions of diclofenac with chitosan have an exothermic character with ΔH values in tens of kJ/mol in magnitude (absolute values) [36,37,38,39,40,41,42]. In contrast, the ΔH value for interactions between sulfamethazine and fulvic acids was much lower.

As described above, an investigation of thermodynamic aspects of the interactions between humic substances and drugs is scarce, and thermodynamic parameters determined for interactions between pharmaceuticals and humic substances were determined only for humic acid salts interactions with papaverine, benzohexonium, and B-group vitamins [34] and sulfamethazine and fulvic acids [35]. In consideration of the relevance of thermodynamic characteristics of interactions of humic substances as key constituents of natural organic matter with pharmaceuticals as a specific type of pollutant occurring in natural systems (besides manufacturing processes) as a result of disposal of unused drugs and excretion of unchanged pharmaceuticals by the human or animal body. Therefore, the novelty of our paper is to determine the thermodynamic parameters of the interactions based on the direct measurements of heat effects by means of ITC. Our paper can be considered as an initial study of thermodynamic aspects of interactions between humic substances and pharmaceuticals, which can provide valuable information about their fate and behavior in nature and help to understand the possibilities of their immobilization and reduction of their mobility in natural systems. In this study, we use microcalorimetry for the direct measurement of heat effects of interactions between three types of pharmaceuticals and three fulvic acids extracted from different matrices. Fulvic acids were chosen for this purpose for their high content of functional groups and good solubility in water. The aim was to determine experimentally changes in enthalpy (ΔH) and to calculate changes in Gibbs energy (ΔG) and entropy (ΔS) in order to characterize the interaction from the point of view of thermodynamics. There is still a lack of information about the interactions between humic substances (as an important constituent of natural organic matter) and pharmaceuticals, which can occur in nature as a result of manufacturing processes, disposal of unused products, and animal excretion. The interactions can affect the mobility of pharmaceuticals in natural systems and cause a decrease in the content of free drug particles in water and soil solutions. The aim of this study is to better understand the character of studied interactions and an evaluation of this method as a complementary tool to investigate the drug–fulvic interactions in detail. Microcalorimetry can provide a quantitative understanding of the thermodynamics underlying the partitioning phenomenon in terms of the energetics of interactions.

2. Materials and Methods

2.1. Chemicals

Ibuprofen (CAS 15687-27-1, ≥98%), diclofenac (CAS 15307-86-5, ≥98%), and sulphapyridine (144-83-2, ≥99%) were purchased from Sigma-Aldrich (St. Luis, MO, USA). Pharmaceuticals were used in the form of aqueous solutions with concentrations equal to 24.2 mM (ibuprofen), 1.44 mM (sulphapyridine), and 67.5 μM (diclofenac). The concentrations were strongly affected by the solubility of individual pharmaceuticals.

Suwannee River Fulvic Acids (2S101F), Pahokee Peat Fulvic Acids (1S103F), and Nordic Lake Fulvic Acids (1R105F) were purchased from the International Humic Substances Society (St. Paul, MN, USA) which provides their basic characteristics as the elemental composition, contents of functional groups, and other characteristics (for details see ref. [43]). These fulvic samples were also studied from the point of view of their content and strength of acidic functional groups, interactions, and conformational changes, e.g., refs. [44,45,46,47,48]. Fulvic acids were used in the form of aqueous solutions with a concentration of 1 g dm⁻³. Molar concentrations of fulvic acids were determined as the total molar content of acidic functional groups per liter. Details about the contents of functional groups are listed in

Table 1. Contents of acidic functional groups in samples of fulvic acids.

Fulvic Acids	COOH (mmol/g)	OH (mmol/g)	Total Acidity (mmol/g)
Suwanee River	5.77	1.47	7.24
Pahokee Peat	6.25	1.09	7.34
Nordic Lake	5.67	1.61	7.28

.

2.2. Isothermal Titration Calorimetry

A MicroCal PEAQ-ITC (Malvern Panalytical Ltd., Malvern, UK) was used for calorimetric measurements. According to our previous experiences [29], we proposed the following measuring procedure. An aqueous solution of fulvic acid (0.2 cm³) was titrated by drug solutions (at 0.002 cm³). The total addition was 20 doses. The titrant was injected at 150 s intervals. The whole experiment took 50 min (procedure A).

The proposed procedure was applied to the measurement with all three pharmaceuticals. However, it was found that it is not suitable for sulphapyridine and diclofenac, probably because of the low concentrations of their solutions. The procedures were modified and optimized to the following one. An aqueous solution of pharmaceutical (0.2 cm³) was titrated by fulvic solutions (at 0.002 cm³). The total addition was 20 doses. The titrant was injected at 300 s intervals. The whole experiment took 100 min (procedure B). The concentration of fulvic acids was 0.1 g dm⁻³ in the case of titration of diclofenac solution.

The temperature was set at 25.0 ± 0.1 °C in all experiments. For homogeneous mixing in the cell, the stirrer speed was kept constant at 750 rpm. The data were processed with MicroCal PEAQ-ITC Analysis software (Version 1.21) from Malvern Instruments. Measurements were triplicated. Data in figures are presented as examples of given measurements because of very small standard deviation bars. Results in tables are presented as average values with error of determination.

3. Results and Discussion

As mentioned above, procedure A was first applied to all three pharmaceuticals and fulvic acids. The experiment course, as well as the obtained experimental data, met the requirements only in the case of ibuprofen (see the example of experimental data in Figure 1). As can be seen, interactions between ibuprofen and fulvic acids are exothermic and change in binding enthalpy is a negative value. The minimum in ΔH value was indicated around the molar ratio equal to 0.2. In order to calculate changes in Gibbs energy and entropy, we adopted the simple mathematical model presented for interactions of pharmaceuticals with chitosan [1] and hemoglobin [41], as well as humic substances with metal ions [29]. According to the model, the change in Gibbs energy ΔG can be expressed as

ΔG = –RTlnK,

(1)

where R is the (universal) molar gas constant, T is the absolute temperature, and K is the equilibrium binding constant called the association constant (reciprocal value of the dissociation constant of the formed metal–fulvic complex). The heat released or consumed during the binding is related to the number of binding sites available, the binding constant, and the thermodynamic parameters of the binding equilibrium. The K_a constant can be determined from the slope of the dependence of integrated raw calorimetric data. Peak areas per mole of added substance can be plotted versus the molar ratio of pharmaceuticals and fulvic acids [1,41]. Applying change in Gibbs energy (ΔG) and enthalpy (ΔH) to the principal equation of thermodynamics, the entropy (ΔS) can be determined [1,41] according to Equation (2):

ΔG = ΔH − TΔS,

(2)

Equations (1) and (2) thus can provide the major thermodynamic parameters for the physicochemical characterization of the interactions between studied pharmaceuticals and fulvic acids as representatives of natural organic matter.

As can be seen in Figure 1, the interactions between Nordic Lake Fulvic Acids and ibuprofen are exothermic, with the minimum in ΔH corresponding with the molar ratio between 0.20 and 0.25. Similar data were obtained for the other two fulvic samples. Pahokee Peat Fulvic Acids had a minimum in the same molar ratio interval (0.20–0.25), and Suwanee River Fulvic Acids in a slightly lower interval between 0.18 and 0.20. Shapes of measured curves were similar for all three used fulvic samples, as well.

Thermodynamic parameters calculated for the interactions between ibuprofen and fulvic acids are listed in

Table 2. Thermodynamic parameters determined for the interactions of fulvic acids with ibuprofen.

Fulvic Acids	ΔH (J/mol)	ΔG (kJ/mol)	ΔS (J/mol·K)
Suwanee River	−5032 ± 277	−27.7 ± 1.5	76.2 ± 4.2
Pahokee Peat	−496 ± 43	−20.6 ± 0.1	67.7 ± 0.4
Nordic Lake	−1693 ± 149	−23.7 ± 0.2	74.0 ± 0.2

. Although interactions of ibuprofen with all three used fulvic acids are exothermic, the values of ΔH are different. Higher changes in enthalpy (in absolute values) were obtained for fulvic acids isolated from aqueous sources (Suwanee River and Nordic Lake) which are comparable with ΔH values determined for interactions of humic acids with papaverine, benzohexonium, and B-group vitamins (–ΔH = 3.8–4.5 kJ/mol) [34] and lower in absolute values comparing with the interactions of ibuprofen with bovine serum albumin (–ΔH = 19.6–27.2 kJ/mol) [37,40] and hemoglobin (–ΔH = 18.8 kJ/mol) [41]. Both serum albumin and hemoglobin have many more binding sites in their mass units, which probably resulted in the higher heat liberation caused by their interactions. The ΔH value determined for Pahokee Peat Fulvic Acids is lower in magnitude (while ΔG and ΔS values are comparable). It is not easy to explain the differences in ΔH between individual fulvic acids. The total acidities (total contents of acidic functional groups) are very similar for all fulvic samples. Nevertheless, Pahokee Peat Fulvic Acids have slightly higher acidity compared to the other two fulvic samples and contain more carboxylic groups than the others (see Table 1). While the ratio between COOH and OH contents is 5.73, the ratios for Suwanee River and Nordic Lake Fulvic acids are 3.92 and 3.52, respectively. Moreover, the pK_a values of Pahokee Peat Fulvic Acids are the highest for COOH (3.99) and the lowest for OH (9.57) functional groups (in comparison with other fulvic samples: 3.76 and 3.79; 9.84, and 9.67) [43]. It should be noted that all presented values should be considered as apparent mean values, including functional groups with their different surrounding structures and different strengths of the functional groups (pK_a).

In contrast, the measurements with sulphapyridine and diclofenac were not successful. Although the obtained experimental data showed an exothermic character of the interactions, they were chaotic without an obvious trend (see Figure 2). Similar chaotic data were obtained for sulphapyridine. As can be seen, the molar ratio is very low, probably due to the low solubility of diclofenac (and similarly for sulphapyridine). Therefore, an equilibrium saturation of binding sites in fulvic acids is not achievable.

In consideration of these findings, we decided to change the measurement procedure and titrate the solutions of pharmaceuticals with fulvic acids (procedure B). The bound amounts of pharmaceuticals, thus, can be determined on the basis of added fulvic acids (to be more accurate, on the basis of functional groups contained in added fulvic acids). The example of data obtained for interactions between Suwanee River Fulvic Acids with sulphapyridine by procedure B is shown in Figure 3.

As can be seen in Figure 3, procedure B can provide experimental data in a larger range of molar ratios, which is required for the determination of thermodynamic parameters. Interactions have an exothermic character, similarly as in the case of ibuprofen (procedure A) and diclofenac (procedure B). Results obtained by means of procedure B are listed in

Table 3. Thermodynamic parameters determined for the interactions of fulvic acids with diclofenac.

Fulvic Acids	ΔH (J/mol)	ΔG (kJ/mol)	ΔS (J/mol·K)
Suwanee River	−5356 ± 295	−24.6 ± 1.3	64.7 ± 3.6
Pahokee Peat	−3302 ± 117	−20.4 ± 0.1	57.4 ± 0.2
Nordic Lake	−8975 ± 95	−23.5 ± 0.1	48.9 ± 0.5

and

Table 4. Thermodynamic parameters determined for the interactions of fulvic acids with sulphapyridine.

Fulvic Acids	ΔH (J/mol)	ΔG (kJ/mol)	ΔS (J/mol·K)
Suwanee River	−3709 ± 204	−23.9 ± 1.3	67.8 ± 3.7
Pahokee Peat	−9938 ± 645	−23.7 ± 0.2	46.2 ± 1.5
Nordic Lake	−7770 ± 249	−23.8 ± 4.3	53.7 ± 1.8

.

While changes in Gibbs energy (ΔG) and entropy (ΔS) are comparable for all pharmaceuticals and fulvic samples, liberated heats (ΔH) are bigger in absolute values for diclofenac and sulphapyridine. Comparing our results with values published for interactions of diclofenac with other bio-colloids, no evident accordance with results obtained for chitosan [1], activated carbon [42], bovine serum albumin [36,38,39], and human serum albumin [39] was found. In contrast to our results and other published data [36,38,39], interactions of diclofenac with chitosan were endothermic, with ΔH equal to +22.1 kJ/mol. Calculated values of ΔG and ΔS were equal to −25.2 kJ/mol and +16 J/mol K, respectively. The change in Gibbs energy, thus, was similar to our results; the entropy was lower but in the same magnitude as the value determined for other bio-colloids [36,38,39]. The variance between heat liberation obtained for the interactions of diclofenac with bovine serum albumin is unusually wide. While Sharma et al. [36] determined the change in enthalpy equal to −103 kJ/mol, which is a very low value in comparison with other results, Talele et al. [38] reported that their heat liberation was ten times higher (−10.2 kJ/mol). Bou-Abdallah et al. [39] stated that their measurements provided ΔH equal to −32 kJ/mol. Our values of ΔH are lower in magnitude; only enthalpy determined for interactions of diclofenac with Nordic Lake Fulvic Acids approximated the results published by Talele et al. [38]. Changes in entropy (ΔS) were similarly high (252 J/mol K) in the case of results published by Sharma et al. [36] compared to other authors: 56 J/mol K [38] and 19.6 J/mol K [39]. Our results are comparable with values obtained by Talele et al. [38] and higher in comparison with the results of Bou-Abdallah et al. [39]. While Sharma et al. [36] did not publish their value of ΔG, other authors published results very similar to our findings (−26.9 kJ/mol [38] and −26.1 kJ/mol [39]). Investigations by Bou-Abdallah et al. [39] also provided results for interactions of diclofenac with human serum albumin: −18.6 kJ/mol (ΔH), −25.7 kJ/mol (ΔG), and 23.9 J/mol K (ΔS). The results, thus, were comparable with the values published for bovine serum albumin in [38,39]. Masson et al. [42] studied the adsorption of several micropollutants (including diclofenac) on activated carbon. Their values of ΔH (−25 kJ/mol) and (ΔG −15.4 kJ/mol) are comparable with results published for bovine and human serum albumin in [38,39]. In contrast, the study with activated carbon [42] provided a positive value of ΔS (+23.9 J/mol K), which is characteristic of the spontaneous process.

As can be seen, thermodynamic parameters published for interactions of pharmaceuticals with bio-colloids are relatively scarce and diverse. Thermodynamic parameters for interactions of ibuprofen, diclofenac, and sulphapyridine are not available in published works; therefore, the main aim of our study is to determine them and compare them with results that were published for similar materials. Direct measurements of heat effects of interactions of humic substances with pharmaceuticals were carried out by Khil’ko and Semenova [34]. They used ITC for the investigation of interactions between humic acids and papaverine, benzohexonium, and B-group vitamins and determined very similar values of –ΔH for all studied pharmaceuticals (between −3.8 and −4.5 kJ/mol), which is very similar to our results for fulvic acids. Our values are, in most cases, slightly more negative, probably due to the higher content of binding sites in their structure. The content of binding sites can also be considered the reason for higher heat liberations determined for other bio-colloids [36,37,38,39,40,41]. Simultaneously, some published results differ mutually in magnitude (e.g., ΔH for diclofenac and bovine serum albumin [36,38]). Xu et al. [35] used ITC for an investigation of interactions between fulvic acids and sulfamethazine, which belong to sulphonamide antibiotics similar to sulphapyridine studied in this work. Their measurements provided a value of ΔH equal to −550 J/mol, which is higher in magnitude in comparison with our results. Change in values of Gibbs energy (−23.10 kJ/mol) was, in this case, comparable with our results. In contrast, the change in entropy was determined as positive (+75.6 J/mol/K), which indicated that the interactions are spontaneous. Similarly, positive values of ΔS were obtained for interactions of pharmaceuticals with other bio-colloids [35,36,38,39,42]).

The aim of our study is to determine the thermodynamic parameters of interactions between fulvic acids and commonly used pharmaceuticals. Our research in the literature showed that the use of ITC for the investigation of the interactions is scarce, and the values for our studied systems are not available. Therefore, the thermodynamic parameters determined in our study can be compared only with other pharmaceuticals and bio-colloids.

Interactions of fulvic acids with ibuprofen, diclofenac, and sulphapyridine were assessed as exothermic with negative values of changes in enthalpy and Gibbs energy and positive changes in entropy (characteristic for spontaneous processes). The heat liberations averaged thousands of joules per mol (excepting interactions of ibuprofen with Pahokee Peat fulvic Acids), which were lower absolute values in magnitude (compared to interactions of pharmaceuticals with, e.g., bovine and human serum albumin [36,38,39]). The most negative values of ΔH were measured for interactions of sulphapyridine with fulvic acids (on average). The lowest heat liberation was observed for interactions of ibuprofen with fulvic acids (on average). Calculated changes in Gibbs energy were very similar for all realized measurements; they ranged between −20.6 and −27.7 kJ/mol. Values of ΔS were comparable mainly for the interactions of diclofenac and sulphapyridine with fulvic acids; slightly higher values were determined for ibuprofen. This indicated that the lowest heat liberation was connected with the highest changes in entropy. Calculated values of ΔG and ΔS were in agreement with our assumptions and in agreement with results published for similar systems [35,36,38,39,42]).

4. Conclusions

In this study, the heat effect of the interactions between three commonly used pharmaceuticals (ibuprofen, diclofenac, and sulphapyridine) and fulvic acids extracted from different sources (Suwanee River, Pahokee Peat, and Nordic Lake) was measured directly by ITC. Simultaneously, thermodynamic parameters, such as changes in Gibbs energy and entropy, were calculated. Thermodynamic characteristics of the interactions are scarce, and details for systems studied in this work are not available in the literature. The determination of ΔH, ΔG, and ΔS values for the studied interactions, thus, can be considered as the main contribution of this work. Their comparison with results obtained for similar systems (other pharmaceuticals and other bio-colloids) helped to assess obtained values and, in general, the suitability of ITC for the characterization of drug–fulvic interactions from a thermodynamic point of view. It was found that the investigation of the studied systems by ITC has to be modified and optimized for less soluble pharmaceuticals. Optimized procedures were proven to be suitable for these purposes and can be considered as an invitation to apply them in systems containing humic substances and other pharmaceuticals to create a database of thermodynamic characteristics of the interactions of these types of pollutants with natural organic matter.