In Vitro Investigations of Acetohexamide Binding to Glycated Serum Albumin in the Presence of Fatty Acid

The interaction of drugs with human serum albumin (HSA) is an important element of therapy. Albumin affects the distribution of the drug substance in the body, as well as its pharmacokinetic and pharmacodynamic properties. On the one hand, inflammation and protein glycation, directly associated with many pathological conditions and old age, can cause structural and functional modification of HSA, causing binding disorders. On the other hand, the widespread availability of various dietary supplements that affect the content of fatty acids in the body means that knowledge of the binding activity of transporting proteins, especially in people with chronic diseases, e.g., diabetes, will achieve satisfactory results of the selected therapy. Therefore, the aim of the present study was to evaluate the effect of a mixture of fatty acids (FA) with different saturated and unsaturated acids on the affinity of acetohexamide (AH), a drug with hypoglycaemic activity for glycated albumin, simulating the state of diabetes in the body. Based on fluorescence studies, we can conclude that the presence of both saturated and unsaturated FA disturbs the binding of AH to glycated albumin. Acetohexamide binds more strongly to defatted albumin than to albumin in the presence of fatty acids. The competitive binding of AH and FA to albumin may influence the concentration of free drug fraction and thus its therapeutic effect.


Introduction
The pace of life, improper eating habits (energy rich, unbalanced diet), and reduction of physical activity to a total minimum has caused an explosion in the number of diabetes all over the world. Diabetes, a civilization related chronic disease, considered a 21st century epidemic, remains a challenge for modern medicine. It is associated with disorders of the mechanisms regulating blood glucose level. Reaserchers argue that the complication and progression of the disease are closely related to the toxic effects of glucose and its metabolites, as one of the pathogenic factors. In patients with the increased blood glucose level, the glycation process intensifies, especially glycation of human serum albumin (HSA) [1]. During the peristent state of hyperglycemia, advanced glycation-end products (AGEs) arise, which are responsible for numerous tissue and organ pathologies [2]. The decrease in glycemia level is crucial to the prevention and inhibition of chronic diabetes (macro-and microvascular) disease progression and complications. This goal is achieved by the change of lifestyle and using anti-diabetic drugs. Acetohexamide (1-((p-acetylphenyl)sulfonyl)-3-cyclohexylurea, AH, Scheme 1) is a first-generation sulfonylurea derivative used to treat type 2 diabetes, especially in people whose Scheme 1) is a first-generation sulfonylurea derivative used to treat type 2 diabetes, especially in people whose blood glucose cannot be controlled by diet. AH is very well absorbed from the gastrointestinal tract and binds to albumin in Sudlow sites I and II [3]. The choice of acetohexamide from many different anti-diabetic drugs was dictated by its physicochemical properties. Its absorbance at wavelengths used to excite the fluorescence of albumin was small enough that allowed high minimize inner filter effect. This is the basic factor limiting the use of the fluorescent technique in drug-protein interactions studies.   HSA belongs to the group of soluble proteins most abundant in blood, where its concentration reaches 40-50 g/L. HSA constitutes 55-65% of all proteins present in plasma [4]. Due to the low molecular weight (66.5-66.9 kDa) compared to other plasma globulins, HSA is involved in many physiological processes that include the regulation of osmotic pressure, pH control, and transport to target tissues of many biologically active compounds (e.g., hormones, acids fatty acids, vitamins, and drugs) [5,6]. The transport of ligands, both exogenous and endogenous in a complex with albumin, is manifested by an increase in the solubility of these compounds in serum, a decrease in their toxicity and protection against their oxidation. In the HSA molecule there are two main specific binding sites for exo-and endogenous substances, referred to as site I (located in subdomain IIA) and II (in subdomain IIIA) according to the Sudlow nomenclature [5,7]. Another small drug binding site in HSA is the digitoxin site [8]. Carter et al. described six binding sites for structurally different ligands: two for binding small, heterocyclic or aromatic carboxylic acids (located in IIA and IIIA subdomains), two for binding of long chain fatty acids and small anionic compounds (located in domain I and III in subdomains B) and two sites for the binding of metal ions, subdomain IIIA. The albumin subdomain IA does not have ligand binding properties, despite homology to IIA and IIIA subdomains that bind most drugs [9]. HSA as a carrier protein is the main transporter of saturated  Scheme 1) is a first-generation sulfonylurea derivative used to treat type 2 diabetes, especially in people whose blood glucose cannot be controlled by diet. AH is very well absorbed from the gastrointestinal tract and binds to albumin in Sudlow sites I and II [3]. The choice of acetohexamide from many different anti-diabetic drugs was dictated by its physicochemical properties. Its absorbance at wavelengths used to excite the fluorescence of albumin was small enough that allowed high minimize inner filter effect. This is the basic factor limiting the use of the fluorescent technique in drug-protein interactions studies.   HSA belongs to the group of soluble proteins most abundant in blood, where its concentration reaches 40-50 g/L. HSA constitutes 55-65% of all proteins present in plasma [4]. Due to the low molecular weight (66.5-66.9 kDa) compared to other plasma globulins, HSA is involved in many physiological processes that include the regulation of osmotic pressure, pH control, and transport to target tissues of many biologically active compounds (e.g., hormones, acids fatty acids, vitamins, and drugs) [5,6]. The transport of ligands, both exogenous and endogenous in a complex with albumin, is manifested by an increase in the solubility of these compounds in serum, a decrease in their toxicity and protection against their oxidation. In the HSA molecule there are two main specific binding sites for exo-and endogenous substances, referred to as site I (located in subdomain IIA) and II (in subdomain IIIA) according to the Sudlow nomenclature [5,7]. Another small drug binding site in HSA is the digitoxin site [8]. Carter et al. described six binding sites for structurally different ligands: two for binding small, heterocyclic or aromatic carboxylic acids (located in IIA and IIIA subdomains), two for binding of long chain fatty acids and small anionic compounds (located in domain I and III in subdomains B) and two sites for the binding of metal ions, subdomain IIIA. The albumin subdomain IA does not have ligand binding properties, despite homology to IIA and IIIA subdomains that bind most drugs [9]. HSA as a carrier protein is the main transporter of saturated HSA belongs to the group of soluble proteins most abundant in blood, where its concentration reaches 40-50 g/L. HSA constitutes 55-65% of all proteins present in plasma [4]. Due to the low molecular weight (66.5-66.9 kDa) compared to other plasma globulins, HSA is involved in many physiological processes that include the regulation of osmotic pressure, pH control, and transport to target tissues of many biologically active compounds (e.g., hormones, acids fatty acids, vitamins, and drugs) [5,6]. The transport of ligands, both exogenous and endogenous in a complex with albumin, is manifested by an increase in the solubility of these compounds in serum, a decrease in their toxicity and protection against their oxidation. In the HSA molecule there are two main specific binding sites for exo-and endogenous substances, referred to as site I (located in subdomain IIA) and II (in subdomain IIIA) according to the Sudlow nomenclature [5,7]. Another small drug binding site in HSA is the digitoxin site [8]. Carter et al. described six binding sites for structurally different ligands: two for binding small, heterocyclic or aromatic carboxylic acids (located in IIA and IIIA subdomains), two for binding of long chain fatty acids and small anionic compounds (located in domain I and III in subdomains B) and two sites for the binding of metal ions, subdomain IIIA. The albumin subdomain IA does not have ligand binding properties, despite homology to IIA and IIIA subdomains that bind most drugs [9]. HSA as a carrier protein is the main transporter of saturated and unsaturated fatty acids (FA). In its structure, it has at least seven asymmetric and heterogeneous fatty acid binding sites, which are also located in the vicinity of most drug binding sites, i.e., in IIA (FA-bindig site 7) and IIIA (FA-binding sites 3 and 4) subdomain of macromolecule [7,10,11]. The FA-binding site with the highest affinity is probably in IIIB albumin subdomain (FA-bindig site 5) [12]. Under physiological conditions, the albumin molecule binds from 0.1 to 2 moles of FA. During fasting or extreme exercises and in some diseases, i.e., diabetes or obesity, the [FA]/[HSA] molar ratio may increase to 6:1 or higher [5,13]. Bojko et al., based on fluorescence studies, concluded that albumin forms a complex with two fatty acid molecules at [FA]/[albumin] molar ratio between 1:1 and 8:1. [14]. Many studies show that fatty acids compete for the binding site in the albumin molecule or interact with drugs [15]. The binding affinity between FA and HSA molecules mainly depends on the chemical properties of the particular acid, i.e., the chain length, both the presence and position of double bonds [16]. In polytherapy, ligands can compete with each other for a binding site and displace each other from albumin binding. The drug that forms a complex with the protein is a pharmacologically inactive spare pool, it is not subject to biotransformation and elimination. Only free drug fractions are responsible for therapeutic effect. Binding of a ligand to albumin can change its spatial conformation, which prevents other ligands from binding, or strengthens existing binding. In addition, the binding of the drug to one binding site can change the structure of other binding sites or their number in the albumin molecule [17,18]. The interaction of drugs with HSA is an important element of therapy, because albumin affects the distribution of the drug substance in the body, as well as its pharmacokinetic and pharmacodynamic properties. On the one hand, inflammation, disease processes, or old age can cause structural and functional modification of HSA, causing binding disorders. On the other hand, the widespread availability of various dietary supplements that affect the content of fatty acids in the body, means that knowledge of the binding activity of transporting proteins, especially in people with chronic diseases, e.g., diabetes, will achieve satisfactory results of the selected therapy. This aspect is important during both the determination of drug doses and with respect to the method of their administration, which is particularly important in multi-drug treatments.
Therefore, the purpose of this work is to evaluate the effect of a mixture of fatty acids (FA) with different saturated and unsaturated acids on the affinity of acetohexamide (AH) for glycated albumin, simulating the state of diabetes in the body. Individual fatty acid mixtures corresponded to the FA content in physiological state and in various clinical states, proceeding with an increased concentration of saturated and unsaturated fatty acids.

Results and Discussion
In the present work, the quenching of albumin fluorophores fluorescence method has been used. This method is a sensitive tool for studying conformational changes in macromolecules caused by low molecular compounds or/and various processes, including glycation. The main goal of the study was to analyze the interaction of acetohexamide (AH), a drug with hypoglycaemic activity and a sulfonylurea derivative of the first generation, with glycated human serum albumin without (af)gHSA and in the presence of fatty acids ((af)gHSA phys , (af)gHSA 4S, (af)gHSA 8S, (af)gHSA 4US , (af)gHSA 8US ). According to Stryer theory, the basic condition for the energy transfer to the drug molecule without radiations is not more than 10 nm distance between human albumin fluorophores (tryptophanyl residue (Trp-214) and tyrosyl residues (Tyrs)) and the drug chromophore, which results in the quenching of macromolecule fluorescence [19]. A comparison of the course of fluorescence quenching curves at excitation wavelength λ ex = 275 nm and λ ex = 295 nm was aimed in order to demonstrate the participation of tryptophanyl residue (Trp-214) or/and tyrosyl residues (Tyrs) in drug-albumin interaction.  [20,21]. The mechanism of drug interaction with albumin can be determined on the basis of Stern-Volmer curves. The linear F 0 /F = f (C drug ) relationship indicates a dynamic or static mechanism of macromolecule fluorescence quenching in the environment of subdomains containing amino acid residues that are involved in the formation of the drug-albumin complex. The positive/negative deviation from the straight line of the Stern-Volmer curve indicates the occurrence of both dynamic and static quenching [22]. Estimation of the distance between the drug molecule and excited albumin fluorophore is possible from the Lehrer modified Stern-Volmer curve (F 0 /∆F = f (1/[C drug ])). Based on this dependence, the Stern-Volmer constant (K SV ) and the fraction of the initial fluorescence of albumin maximally available for the drug (f a ) are determined. An increase in the distance between the albumin fluorophore and the drug molecule is manifested by a decrease in the K SV constant. The nature of drug binding to albumin, i.e., determination of binding sites specificity in individual classes of macromolecule binding sites, was determined on the basis of binding isotherms ((r = f ([L f ])), whose linear course indicates the non-specific binding of the drug to the hydrophobic surface of albumin. The exponential saturation curve, achieving a "plateau", characterizes a specific binding sites with high affinity and low binding capacity. The mixed nature of the binding, i.e., specific and non-specific, is illustrated by an isotherm with a course between linear and reaching a "plateau" [23]. There are many methods for the calculation of association constant (K a ) that characterizes the stability of drug-albumin complex, a determination the number of drug molecules associated with one albumin (n) molecule at equilibrium, or for prediction the existence of one or more independent classes of binding sites. In the present work, the Klotz (1/r = f (1/[L f ])) and Hill (log[r/(1-r)] = f (log[L f ])) equation were used to determine the value of (K a and n) parameters. In addition, in order to determine whether the binding of the drug molecule to albumin influences the increase of its affinity for other macromolecule binding sites, the Hill coefficients (n H ) were determined from the Hill linear relationship.  (Figure 2b) by acetohexamide, which was found in no more than 10 nm proximity [19].
Molecules 2020, 22, x FOR PEER REVIEW 4 of 5 a dynamic or static mechanism of macromolecule fluorescence quenching in the environment of subdomains containing amino acid residues that are involved in the formation of the drug-albumin complex. The positive/negative deviation from the straight line of the Stern-Volmer curve indicates the occurrence of both dynamic and static quenching [22]. Estimation of the distance between the drug molecule and excited albumin fluorophore is possible from the Lehrer modified Stern-Volmer curve (F0/∆F = f(1/[Cdrug])). Based on this dependence, the Stern-Volmer constant (KSV) and the fraction of the initial fluorescence of albumin maximally available for the drug (fa) are determined. An increase in the distance between the albumin fluorophore and the drug molecule is manifested by a decrease in the KSV constant. The nature of drug binding to albumin, i.e., determination of binding sites specificity in individual classes of macromolecule binding sites, was determined on the basis of binding isotherms ((r = f([Lf])), whose linear course indicates the non-specific binding of the drug to the hydrophobic surface of albumin. The exponential saturation curve, achieving a "plateau", characterizes a specific binding sites with high affinity and low binding capacity. The mixed nature of the binding, i.e., specific and non-specific, is illustrated by an isotherm with a course between linear and reaching a "plateau" [23]. There are many methods for the calculation of association constant (Ka) that characterizes the stability of drug-albumin complex, a determination the number of drug molecules associated with one albumin (n) molecule at equilibrium, or for prediction the existence of one or more independent classes of binding sites. In the present work, the Klotz (1/r = f(1/[Lf])) and ) equation were used to determine the value of (Ka and n) parameters. In addition, in order to determine whether the binding of the drug molecule to albumin influences the increase of its affinity for other macromolecule binding sites, the Hill coefficients (nH) were determined from the Hill linear relationship.

The Interaction of Acetohexamide with Human Serum Albumin in the Absence of Fatty Acids (AH-(af)HSA) and the Influence of Physiological Composition of Fatty Acids on AH-(af)HSA Complex (AH-(af)HSAphys)
As shown in Figure 2, an increase in acetohexamide (AH) concentration in the AH-(af)gHSA and AH-(af)gHSAphys systems causes a gradual decrease in macromolecule fluorescence intensity. The observed effect may be associated with the quenching fluorescence of excited fluorophores (tryptophanyl residue (Trp-214) and tyrosyl residues (Tyrs)) of albumin (af)gHSA ( Figure 2a) and (af)gHSAphys ( Figure 2b) by acetohexamide, which was found in no more than 10 nm proximity [19].  In addition, in the AH-(af)gHSA system, after excitation at λex = 275 nm, the shift in the albumin fluorescence emission band towards shorter waves (blue-shift) by 4 nm relative to the spectrum of the drug-free albumin (Δλmax = 327-323 nm) (Figure 2a, in the main view) has been observed. The hypsochromic shift indicates a decrease in the polar nature, and thus the formation of a hydrophobic environment around the tryptophanyl residue (Trp-214) and tyrosyl residues (Tyrs) of (af)gHSA due to the interaction of AH with albumin. As in our previous work, the presence of gliclazide (GLZ), a second generation sulfonylurea derivative, also caused a blue-shift of glycated (gHSAFRC) spectra in GLZ-gHSAFRC [24]. The blue-shift of maximum albumin fluorescence (Δλmax) caused by the presence of AH is associated with a decrease in polarity (increase in hydrophobicity) of albumin fluorophores after binding to the ligand. This indicates the possibility of hydrophobic interactions between the aromatic rings of the acetohexamide molecule and aromatic amino acid rings of the hydrophobic (af)gHSA cavity within IIA (Trp-214, Tyrs) or/and IB, IIB, IIIA (Tyrs) subdomains. For AH-(af)gHSAphys system at excitation wavelength λex = 295 nm, a red-shift in fluorescence emission band versus fluorescence of free-drug albumin (af)gHSAphys by 3 nm (Δλmax = 332-335 nm) has been recorded ( Figure 2b, in the insert). The shift in the emission fluorescence maximum (af)gHSAphys in the presence of AH towards the long-term direction indicates an increase in polarity and the formation of the hydrophilic environment of albumin fluorophore (Trp-214). Similar observations were noticed by Feng et al. [25] during the fluorescence analysis of theasinesin binding to human serum albumin (HSA). The red-shift of HSA fluorescence spectrum with the increase of ligand concentration the authors explained that tryptophan residue was brought to a more hydrophilic environment in the theasinesin-HSA system, and the structure of the hydrophobic subdomain where a tryptophan was placed was not compact and the segment of polypeptide changed its conformation to a more extended state after the addition of theasinesin. The emission of indole Trp-214 may be blue-shifted if the group is buried within a native protein, and its emission may shift to longer wavelengths (red-shift) when protein is unfolded [20]. In addition, in the AH-(af)gHSA system, after excitation at λ ex = 275 nm, the shift in the albumin fluorescence emission band towards shorter waves (blue-shift) by 4 nm relative to the spectrum of the drug-free albumin (∆λ max = 327-323 nm) (Figure 2a, in the main view) has been observed. The hypsochromic shift indicates a decrease in the polar nature, and thus the formation of a hydrophobic environment around the tryptophanyl residue (Trp-214) and tyrosyl residues (Tyrs) of (af)gHSA due to the interaction of AH with albumin. As in our previous work, the presence of gliclazide (GLZ), a second generation sulfonylurea derivative, also caused a blue-shift of glycated (gHSA FRC ) spectra in GLZ-gHSA FRC [24]. The blue-shift of maximum albumin fluorescence (∆λ max ) caused by the presence of AH is associated with a decrease in polarity (increase in hydrophobicity) of albumin fluorophores after binding to the ligand. This indicates the possibility of hydrophobic interactions between the aromatic rings of the acetohexamide molecule and aromatic amino acid rings of the hydrophobic (af)gHSA cavity within IIA (Trp-214, Tyrs) or/and IB, IIB, IIIA (Tyrs) subdomains. For AH-(af)gHSA phys system at excitation wavelength λ ex = 295 nm, a red-shift in fluorescence emission band versus fluorescence of free-drug albumin (af)gHSA phys by 3 nm (∆λ max = 332-335 nm) has been recorded (Figure 2b, in the insert). The shift in the emission fluorescence maximum (af)gHSA phys in the presence of AH towards the long-term direction indicates an increase in polarity and the formation of the hydrophilic environment of albumin fluorophore (Trp-214). Similar observations were noticed by Feng et al. [25] during the fluorescence analysis of theasinesin binding to human serum albumin (HSA). The red-shift of HSA fluorescence spectrum with the increase of ligand concentration the authors explained that tryptophan residue was brought to a more hydrophilic environment in the theasinesin-HSA system, and the structure of the hydrophobic subdomain where a tryptophan was placed was not compact and the segment of polypeptide changed its conformation to a more extended state after the addition of theasinesin. The emission of indole Trp-214 may be blue-shifted if the group is buried within a native protein, and its emission may shift to longer wavelengths (red-shift) when protein is unfolded [20].
Molecules 2020, 22, x FOR PEER REVIEW 6 of 5 Based on the obtained data from the emission fluorescence spectra, the quenching curves of (af)gHSA and (af)gHSAphys (5 × 10 −6 mol•L −1 ) fluorescence in the presence of AH (5 × 10 −6 mol•L −1 -5 × 10 −5 mol•L −1 ), λex = 275 nm and λex = 295 nm, have been drawn (Figure 3).  The course of albumin fluorescence quenching curves illustrates the reduction in fluorescence intensity of human serum albumin without (af)gHSA and with (af)gHSA phys fatty acids with the increase of acetohexamide concentration in AH-albumin system (Figure 3a-d). The presence of fatty acids at physiological concentration affects the ability of acetohexsamide to quench albumin fluorescence. For (af)gHSA a stronger fluorescence quenching than for (af)gHSA phys , at both excitation wavelengths λ ex = 275 nm (Figure 3a) and λ ex = 295 nm (Figure 3b), has been observed. This phenomenon indicates a greater ability of acetohexamide to absorb energy from defatted albumin than in the presence of fatty acids, that change macromolecule. For ligand: albumin molar ratio 10:1, the quenching of the internal (af)gHSA and (af)gHSA phys fluorescence by AH equals to 30.20% and 10.50%, respectively (λ ex = 275 nm), while λ ex = 295 nm equals to 31.70% and 4.60%, respectively. The comparison of fluorescence quenching curves of (af)gHSA and (af)gHSA phys , in the presence of acetohexamide at the excitation wavelength λ ex = 275 nm and λ ex = 295 nm, allowed to indicate the fluorophores (Trp-214 and/or Tyrs) involved in the interaction with AH. Nearly identical course of albumin fluorescence quenching curves in AH-(af)gHSA (Figure 3c) system, at λ ex = 275 nm and λ ex = 295 nm (4.60% difference in quenching of the intrinsic albumin fluorescence), indicates the contribution of a Trp-214 residue (located in IIA subdomain of albumin) or its environment and a negligible contribution of Tyrs residues in the interaction of AH with (af)gHSA in the environment of binding site. A significant difference (56.60%) in quenching of (af)gHSA phys fluorescence registered for λ ex = 275 nm and 295 nm (Figure 3d) indicates the participation in AH interaction with (af)gHSA phys not only Trp-214, but also Tyrs residues. A stronger fluorescence quenching in AH-(af)gHSA phys system at excitation wavelength λ ex = 275 nm than 295 nm may indicate a much easier access of AH to 17 tyrosyl residues (Tyr-30, -84,  -138, -140, -148, -150, -161, -263, -319, -332, -334, -341, -353, -370, -401, -411, -497) located in IB, IIB, IIA, and IIIA [26] subdomains than to the tryptophanyl residue. It is noteworthy that the fluorescence quenching technique is not sufficient to indicate which Tyrs moieties are involved in AH binding. Literature data indicate that in human albumin structure there are two main specific binding sites for endo-and exogenous substances such as acetohexamide.    (Figure 4b), indicates dynamic (collision) and static (formation of a stable ligand-albumin complex in the ground state) quenching fluorescence of glycated, defatted albumin by acetohexamide. Literature data indicate that the negative deviation from the rectilinear course of the Stern-Volmer curve is explained by occuring initially more easily available binding sites for the drug, and after saturating them, those more difficult to access in the macromolecule [27,28].
The rectilinear course of Stern-Volmer plots (F0/F = f[CAH] relationship) obtained for the AH-(af)gHSAphys system may indicate dynamic or static fluorescence quenching mechanism. When the dynamic mechanism takes place, there is a collision of the acetohexamide molecule with albumin (af)gHSAphys fluorophores in excitation state. The static mechanism leads to a decrease in the intensity  Stern-Volmer curves plotted for glycated, defatted serum albumin in the presence of acetohexamide (AH-(af)gHSA system) show a different course from the curves plotted for AH-(af)gHSA phys system both at excitation λ ex = 275 nm (Figure 4a), and λ ex = 295 nm (Figure 4b). A stronger fluorescence quenching (F 0 /F) at whole range of acetohexamide concentrations have been occurred for albumin (af)gHSA compared to the F 0 /F value obtained for glycated, fatted albumin (af)gHSA phys . Negative deviation from the rectilinear relationship F 0 /F = f [C AH ] for AH-(af)gHSA system at λ ex = 275 nm ( Figure 4a) and λ ex = 295 nm (Figure 4b), indicates dynamic (collision) and static (formation of a stable ligand-albumin complex in the ground state) quenching fluorescence of glycated, defatted albumin by acetohexamide. Literature data indicate that the negative deviation from the rectilinear course of the Stern-Volmer curve is explained by occuring initially more easily available binding sites for the drug, and after saturating them, those more difficult to access in the macromolecule [27,28].
The rectilinear course of Stern-Volmer plots (F 0 /F = f [C AH ] relationship) obtained for the AH-(af)gHSA phys system may indicate dynamic or static fluorescence quenching mechanism. When the dynamic mechanism takes place, there is a collision of the acetohexamide molecule with albumin (af)gHSA phys fluorophores in excitation state. The static mechanism leads to a decrease in the intensity of emitted fluorescence when acetohexamide binds to the fluorophore molecule in the non-excited state, reducing the population of excited fluorophores [27].
From the F 0 /F = f [C AH ] relationship for a system with a linear course of the Stern-Volmer curve (AH-(af)gHSA phys system), the Stern-Volmer constants K SV , the biomolecular quenching rate constants k q (k q = K SV /τ 0 ) and maximum available fluorescence fraction of all albumin f a fluorophores were determined. Quenching parameters (K SV , k q = K SV /τ 0 ) and f a for a system with non-linear Stern-Volmer relationship (AH-(af)gHSA system) were determined from F 0 /∆F = f (1/[C AH ]) relationship represented by Stern-Volmer equation modified by Lehrer [29]. The obtained results are given in Table 1. Table 1. Stern-Volmer constants K SV (mol −1 ·L), biomolecular quenching rate constants k q (mol −1 ·L·s −1 ) and maximum available fluorescence fraction of all albumin fluorophores f a calculated for the AH-(af)gHSA and AH-(af)gHSA phys systems; λ ex = 275 nm and λ ex = 295 nm. Higher values of K SV and k q constant obtained for the AH-(af)gHSA system compared to K SV and k q obtained for AH-(af)gHSA phys system (Table 1) indicate the location of acetohexamide molecules closer to the fluorophores of glycated, defatted albumin (af)gHSA than fluorophores of fatted by fatty acid physiological mixture albumin ((af)gHSA phys ). Exciting albumin (af)gHSA and (af)gHSA phys at wavelength λ ex = 275 nm, fluorescence emission was observed not only from the tryptophanyl residue, but also from tyrosyls residues and higher values of K SV constants were obtained than at λ ex = 295 nm excitation wavelength. This demonstrates the contribution of (af)gHSA and (af)gHSA phys tyrosyl residues in AH-albumin interaction. The IIA human serum albumin subdomain contains one tryptophanyl residue (Trp-214) and one tyrosyl residue (Tyr-263), which would indicate that in acetohexamide binding to albumin (af)gHSA and (af)gHSA phys other subdomains, where tyrosyl residues are located (IB, IIB and IIIA subdomains), are also involved. Our observations are consistent with the results obtained by Joseph et al. [3] who, using the high-performance affinity chromatography (HPAC) technique, indicated the IIA and IIIA subdomains, i.e., the I and II binding sites, respectively, as the main binding site of acetohexamide in the glycated human structure serum albumin. The order of fluorescence quenching rate constants k q (10 12 or 10 11 ) determined for AH-(af)gHSA and AH-(af)gHSA phys system clearly indicates a static fluorescence quenching mechanism in the acetohexamide-albumin system (Table 1). According to Lakowicz, when the maximum value of k q constant in the aqueous solution is 1 × 10 10 [M −1 s −1 ], the dynamic fluorescence quenching mechanism occurs [20]. Since the values of K SV constant and f a fraction are inversely proportional, the decrease in K SV is accompanied by an increase in f a . The presence of fatty acids reduces the strength of AH-(af)gHSA phys fluorophores interaction with the increase the availability (f a ) of AH to tryptophanyl and tyrosyl (af)gHSA phys residues.
To determine the nature of the interaction of acetohexamide with (af)gHSA and (af)gHSA phys , saturation curves (binding isotherms) were plotted in the AH-(af)gHSA and AH-(af)gHSA phys systems, λ ex = 275 nm (Figure 5a) and λ ex = 295 nm (Figure 5b). The shape of the binding isotherms (r = f([Lf]) obtained for (af)gHSA albumin indicates the mixed (specific and non-specific) nature of the interaction, which means that acetohexamide binds not only to its specific binding sites in defatted albumin molecule, but also non-specifically interacts with the hydrophobic fragments of its surface [23]. However, the shape of the binding isotherms for albumin saturated with fatty acids indicates the occurrence of only the non-specific nature of acetohexamide binding to (af)gHSAphys (Figure 5a,b). Specific binding is characterized by high affinity and low binding capacity, while non-specific binding with low affinity and unlimited drug binding capacity [23].
Since acetohexamide saturates the binding sites of defatted ((af)gHSA) and fatted ((af)gHSAphys) albumin, using the Klotz (the independent variable is the inverse of the concentration of free ligand fraction, 1/r = f(1/[Lf])) and Hill (the independent variable is the logarithm of ligand free fraction concentration, log[r/(1-r)] = f(log[Lf])) equations, the association constants Ka, which determine the stability of formed drug-albumin complex, were determined. The number of acetohexamide molecules forming a complex with one molecule of both (af)gHSA and (af)gHSAphys at equilibrium state (n) for a specific class of binding sites was also obtained. Moreover, Hill interaction factors (nH) were determined to investigate the possible cooperation of acetohexamide binding to the macromolecule. The data are presented in Table 2. The shape of the binding isotherms (r = f ([L f ]) obtained for (af)gHSA albumin indicates the mixed (specific and non-specific) nature of the interaction, which means that acetohexamide binds not only to its specific binding sites in defatted albumin molecule, but also non-specifically interacts with the hydrophobic fragments of its surface [23]. However, the shape of the binding isotherms for albumin saturated with fatty acids indicates the occurrence of only the non-specific nature of acetohexamide binding to (af)gHSA phys (Figure 5a,b). Specific binding is characterized by high affinity and low binding capacity, while non-specific binding with low affinity and unlimited drug binding capacity [23].
Since acetohexamide saturates the binding sites of defatted ((af)gHSA) and fatted ((af)gHSA phys ) albumin, using the Klotz (the independent variable is the inverse of the concentration of free ligand fraction, 1/r = f (1/[L f ])) and Hill (the independent variable is the logarithm of ligand free fraction concentration, log[r/(1-r)] = f (log[L f ])) equations, the association constants K a , which determine the stability of formed drug-albumin complex, were determined. The number of acetohexamide molecules forming a complex with one molecule of both (af)gHSA and (af)gHSA phys at equilibrium state (n) for a specific class of binding sites was also obtained. Moreover, Hill interaction factors (n H ) were determined to investigate the possible cooperation of acetohexamide binding to the macromolecule. The data are presented in Table 2. Table 2. Association constants K a (mol −1 ·L), mean number of AH moles bound with one mole of (af)gHSA and (af)gHSA phys (n), the Hill coefficient (n H ) in AH-(af)gHSA and AH-(af)gHSA phys system; λ ex = 275 nm, λ ex = 295 nm.
From the straight line of Klotz eguation for AH-(af)gHSA, AH-(af)gHSA phys systems and Hill equation for AH-(af)gHSA, AH-(af)gHSA phys systems, the binding of AH to macromolecules within one class of binding sites has been obtained. For AH-(af)gHSA complex, the association constants K a are more than 10-fold (for λ ex = 275 nm) and more than 5-fold (for λ ex = 295 nm) higher than the constants K a values obtained for AH-(af)gHSA phys complex (Table 2), which indicates that AH has higher affinity for (af)gHSA than for (af)gHSA phys albumin binding sites. This phenomenon is probably caused by the reduction of AH-albumin complex stability in the presence of fatty acids, when both tryptophanyl residue (Trp-214) and tyrosyl residues (Tyrs) have been excited (λ ex = 275 nm). The same results have been received in our other study (data not published). We have concluded that tolbutamide (TB), a hypoglycaemic drug, binds more easily to defatted albumin than in the presence of fatty acid. In the treatment of TB high fat diet increased the therapeutic effect of the drug. The number of binding sites n close to unity indicates the existence of one specific AH binding site in the (af)gHSA molecule. Based on n H value (n H ≈ 1) lack of cooperativity in binding of AH to (af)gHSA has been observed (λ ex = 275 nm and λ ex = 295 nm) ( Table 2). For AH-(af)gHSA phys complex at λ ex = 295 nm, a Hill coefficient value higher than 1 (n H > 1) points to a positive co-operability phenomenon, which probably means that binding of AH in one place increases the affinity of the drug/drugs for other macromolecule binding sites. The occurrence of cooperative binding results from the fact that under physiological conditions, fatty acid molecules (FA4 and FA5) bind to IIIA and IIIB subdomains [31], thus affecting the albumin conformation, i.e., changes the affinity of a macromolecule for acetohexamide. These observations complement the results discussed earlier that the presence of fatty acids induces changes in the structure of the macromolecule. Under physiological conditions, human serum albumin binds a maximum of two fatty acid moles (FA). The formation of an albumin complex with two FA moles is possible when their molar FA:albumin ratio is within 1:1 and 1:8 [31]. Therefore, in order to assess the binding of AH to glycated human serum albumin in various clinical states, proceeding with increased concentration of saturated (FA S ) and unsaturated (FA US ) fatty acids, in the next part of the study albumin with four and eight times more amount of saturated ((af)gHSA 4S , (af)gHSA 8S ) and unsaturated ((af)gHSA 4US , (af)gHSA 8US ) fatty acids compared to physiological values have been analyzed. In the next part of the study, emission fluorescence spectra of glycated albumin at 5 × 10 −6 mol·L −1 concentration containing four ((af)gHSA 4S ) (4.1 × 10 −5 mol·L −1 ) and eight times ((af)gHSA 8S ) (6.9 × 10 −5 mol·L −1 ) more saturated fatty acids (FA S ) and four ((af)gHSA 4US ) (5.9 × 10 −5 mol·L −1 ) and eight times ((af)gHSA 8US ) (1.11 × 10 −4 mol·L −1 ) more unsaturated (FA US ) fatty acids in relation to the physiological value (2 × 10 −5 mol·L −1 ) have been obtained. Fluorescence of macromolecules excited at λ ex = 275 nm and λ ex = 295 nm decreases with the increase of acetohexamide (AH) concentration (data not shown). Lack of shift in the fluorescence emission band of albumin relative to the albumin spectrum in the presence of AH (∆λ max ) may indicate the invariability of the hydrophobic/hydrophilic properties of AH binding site (near the Trp-214 residue in subdomain IIA and Tyrs residues in subdomain IB, IIB, IIA, and IIIA of albumins).
As it was previously mentioned, after excitation of albumin at excitation wavelength λ ex = 295 nm, the observed emission of fluorescence comes almost exclusively from the tryptophan residue, while for λ ex = 275 nm from both the tryptophan residue and tyrosyl residues. Differences in quenching of albumin fluorescence at both excitation wavelengths λ ex = 275 nm and λ ex = 295 nm, in AH-(af)gHSA 4S (Figure 6c, in the main view), AH-(af)gHSA 8S (Figure 6c, in the insert) and AH-(af)gHSA 8US (Figure 6d, in the insert) systems, indicate the simultaneously participation of Trp-214 and Tyrs residues in the interaction of AH with albumin at appropriate binding site. The lack of differences in albumin fluorescence quenching curves in AH-(af)gHSA 4US system (Figure 6d, in the main view) after excitation at both wavelengths indicates the participation mainly Trp-214 residue.
In addition, for each concentration of AH in the drug-albumin system, the degree of albumin fluorescence quenching is greater after excitation at λex = 275 nm than at λex = 295 nm, which would indicate a significant parfticipation of tyrosyl residues in the interaction of AH with (af)gHSA4S, (af)gHSA8S, (af)gHSA4US and (af)gHSA8US. Tyrosyl residues in position 401 (Tyr-401) and 411 (Tyr-411) located in IIIA subdomain of macromolecule play a major role in drugs binding [20,32]. The observed significant contribution of Tyrs residues in the interaction of AH with albumin indicates the IIIA subdomain as the likely acetohexamide binding site. The analysis of Stern-Volmer plots (F0/F = f[CAH]) showed a straight line for AH-(af), AH-(af)gHSA4US, and AH-(af)gHSA8US systems, indicating a collision or static mechanism of albumin quenching fluorescence by AH. However, for AH-(af)gHSA4S system, F0/F = f[CAH] relationship showed a non-linear course (slight deviation in the direction of OY), which would indicate the simultaneous occurrence of dynamic and static fluorescence quenching mechanism, a similar effect that we observed in our previous work [33] concerning the theophylline (Th)-human albumin system study. Quenching parameters (KSV, kq = KSV/τ0) and the fraction of maximum available fluorescence  For albumin with two times more saturated ((af)gHSA 8S ) and unsaturated ((af)gHSA 8US ) fatty acids, a stronger fluorescence quenching with increasing AH concentration has been recorded than for albumin (af)gHSA 4S and (af)gHSA 4US both after excitation λ ex = 275 nm (Figure 6a,b, in the main view) and λ ex = 295 nm (Figure 6a,b, in the insert). This demonstrates the greater ability of AH to absorb energy from excited fluorophores, as a result of conformational changes caused by a higher content of fatty acids in the structure of albumin.
In addition, for each concentration of AH in the drug-albumin system, the degree of albumin fluorescence quenching is greater after excitation at λ ex = 275 nm than at λ ex = 295 nm, which would indicate a significant parfticipation of tyrosyl residues in the interaction of AH with (af)gHSA 4S , (af)gHSA 8S , (af)gHSA 4US and (af)gHSA 8US . Tyrosyl residues in position 401 (Tyr-401) and 411 (Tyr-411) located in IIIA subdomain of macromolecule play a major role in drugs binding [20,32]. The observed significant contribution of Tyrs residues in the interaction of AH with albumin indicates the IIIA subdomain as the likely acetohexamide binding site.

Procedure of Preparation Fatty Acids Solutions for Fluorescence and Absorbance Studies
Defatted human serum albumin ((af)HSA) in the presence of glucose (GLC) at 5 × 10 −6 mol·L −1 and 0.05 mol·L −1 concentrations, respectively, were prepared in phosphate buffer solution (pH 7.4, 0.05 mol·L −1 ) in double-distilled water and in the presence of sodium azide (NaN 3 ) (0.015 mol·L −1 ). Solution of protein was passed through a sterile Millex-GP syringe filters with 0.2 µm pores, and then it was incubated in sterile closed tubes for a period of 21 days at constant temperature of 37 • C. After the incubation period, to remove the excess of unbound glucose, the solutions of glycated human serum albumin (af)gHSA was dialyzed extensively against 0.05 mol·L −1 phosphate buffer at pH 7.4 for 24 h.

Instruments and Measurements Conditions
The fluorescence and absorbance measurements of the samples were recorded at 37 • C using spectrofluorimeter JASCO FP-6500 (Hachioji, Tokyo, Japan) equipped with Peltier thermostat (∆t ± 0.2 • C) (error apparatus ± 1.5 nm) (Peltier thermostat is a part of JASCO FP-6500) and spectrophotometer JASCO V-760 (correcting error of apparatus for wavelength and photometric is equal to ± 0.3 nm and ± 0.002 Abs. at 0.5 Abs) (Hachioji, Tokyo, Japan), respectively. For the measurement standards quartz cuvettes 1 cm × 1 cm × 4 cm were used. The fluorescence spectra presented in the paper were corrected for the solvent dispersion (phosphate buffer) using the Spectra Manager program, and then analyzed using OriginPro version 8.5 SR1 software (Northampton, MA, USA). Finally, light scattering caused by buffer was subtracted from fluorescence of samples in each spectrum using software supplied by JASCO (Spectra Manager). The results of the study were expressed as a mean ± relative standard deviation (RSD) from three independent experiments. Linear regression by fitting experimental data to the appropriate equations were analyzed using OriginPro version 8.5 SR1 software (Northampton, MA, USA).

The Analyzis of Acetohexamide-Albumins Interaction
Based on the calculated fluorescence emission intensities without and in the presence of fatty acids glycated human serum albumin, the quenching curves ( F F 0 vs. ligand:albumin molar ratio, where: F and F 0 is the fluorescence intensity at the maximum wavelength of albumin in the presence and absence of a quencher, respectively) of (af)gHSA, (af)gHSA phys , (af)gHSA 4S, (af)gHSA 8S, (af)gHSA 4US and (af)gHSA 8US in the presence of acetohexamide (AH) have been plotted.
where: f a is the fractional maximum protein fluorescence accessible for the quencher. [(a f )gHSA] is the number of ligands moles bound per mole of protein molecule; [L b ] = ∆F ∆F max ·(a f )gHSA total , ∆F is the difference between F 0 and F, ∆F max (maximal fluorescence change with complete saturation) is evaluated from the linear part of the 1 ∆F vs. 1 [L] ; [(a f )gHSA] is glycated serum albumin concentration [mol·L −1 ] [23].
From the Klotz curves (Equation (4)) [35], in addition to the values of K a association constants, the number of drug molecules bound to one protein molecule (n) was determined.
where: n is the number of binding sites for the independent class of drug binding sites in the glycated albumin molecule. From Hill dependence (Equation (5)) [36] association constants K a and interaction coefficients n H were calculated. log where: n H is the Hill's coefficient.

Binding Sites Visualisation. Molecular Docking Simulation
To visualize the potential binding site of acetohexamide (AH) to glucose complexed human serum albumin (GLC-HSA), molecular docking simulation has been made using the CLC Drug Discovery Workbench ver. 1.0.2. (CLC Bio, a QIAGEN Company: Aarhus, Denmark) [License number: CLC-LICENSE-51JT8-DXYBY-2A3EW-ED80P-DGW80]. The crystallographic structure of glucose complexed human serum albumin (GLC-HSA) has been obtained from the PDB protein structure database (Protein Data Bank) using the PDB ID: 4IW2 [37]. Theoretical structure of atetohexamide (AH) was minimized and optimized by the Austin Model 1 (AM1)-semi-empirical method using the Atomic and Molecular Electronic Structure System (GAMESS) [38]. The AH docking procedure for the GLC-HSA structure was conducted simulating physiological conditions, in accordance with the default program configuration.

Conclusions
Fluorescence spectroscopy is a technique that allows to monitor the intermolecular interactions.
To determine the distribution capacity and binding properties of glycated, defatted albumin ((af)gHSA) and albumin containing a physiological mixture of fatty acids ((af)gHSA phys ), albumin containing 4 times ((af)gHSA 4S ) and 8 times ((af)gHSA 8S ) higher amount of saturated fatty acids (FA S ) and 4 times ((af)gHSA 4US ) and 8 times ((af)gHSA 8US ) higher amount of unsaturated (FA US ) fatty acids in relation to physiological concentration, a number of fluorescent quenching measurements of albumin fluorescence by acetohexamide (AH) were made. Based on the quantitative and qualitative analysis we can conclude that the presence of both saturated and unsaturated fatty acids disturbs the binding of acetohexamide to human serum albumin. Acetohexamide binds more strongly to defatted albumin (af)gHSA than to albumin in the presence of fatty acids. Competitive binding of AH and fatty acids to albumin may influence the concentration of free drug fraction and thus its therapeutic effect. It should be taken into account that under conditions of abnormal fat content in the body, the pharmacokinetics of the drug may be affected.

Conflicts of Interest:
The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.