DNA interaction with human serum albumin studied by affinity capillary electrophoresis and FTIR spectroscopy

The question addressed in this study is how does the protein-DNA complexation affect the structure and dynamics of DNA and protein in aqueous solution. We examined the interaction of calf-thymus DNA with human serum albumin (HSA) in aqueous solution at physiological conditions, using constant DNA concentration of 12.5 mM (phosphate) and various HSA contents 0.25 to 2% or 0.04 to 0.3 mM. Affinity capillary electrophoresis and FTIR spectroscopic methods were used to determine the protein binding mode, the association constant, sequence preference, and the biopolymer secondary structural changes in the HSA-DNA complexes. Spectroscopic evidence showed two types of HSA-DNA complexes with strong binding of K(1) = 4.5 x 10(5) M(-1) and weak binding of K(2) = 6.10 x 10(4) M(-1). The two major binding sites were located on the G-C bases and the backbone PO(2) group. The protein-DNA interaction stabilizes the HSA secondary structure. A minor alteration of B-DNA structure was observed, while no major protein conformational changes occurred.     

Interactions of atrazine and 2,4-D with human serum albumin studied by gel and capillary electrophoresis, and FTIR spectroscopy.

The herbicides 6-chloro-N-ethyl-N'-(1-methylethyl)-1,3,5-triazine-2,4-diamine (atrazine) and 2,4-dichlorophenoxyacetic acid (2,4-D) are widely used in agricultural practice to fight dicotyledon weeds mainly in maize, cereals, and lucerne. As a result, these compounds are found not only in the plants, soil, and water, but also in the cultivated ground in the following years as well as in agricultural products such as fruits, milk, butter, and sugar beet. The toxicological effects of herbicides occur in vivo, when transported to the target organ through the bloodstream. It has been suggested that human serum albumin (HSA) serves as a carrier protein to transport 2,4-D to molecular targets. This study was designed to examine the interaction of atrazine and 2,4-D with HSA in aqueous solution at physiological pH with herbicide concentrations of 0.0001-1 mM, and final protein concentration of 1% w/v. Gel and capillary electrophoresis, UV-visible and Fourier transform infrared spectroscopic methods were used to determine the drug binding mode, the drug binding constant, and the protein secondary structure in aqueous solution. Structural analysis showed that different types of herbicide-HSA complexes are formed with stoichiometric ratios (drug/protein) of 3:1 and 11:1 for atrazine and 4.5:1 and 10:1 for 2,4-D complexes. Atrazine showed a weak binding affinity (K=3.50 x 10(4) M(-1)), whereas two bindings (K(1)=2.50 x 10(4) M(-1) and K(2)=8.0 x 10(3) M(-1)) were observed for 2,4-D complexes. The herbicide binding results in major protein secondary structural changes from that of the alpha-helix 55% to 45--39% and beta-sheet 22% to 24--32%, beta-anti 12% to 10--22% and turn 11% to 12--15%, in the drug-HSA complexes. The observed spectral changes indicate a partial unfolding of the protein structure, in the presence of herbicides in aqueous solution.     

Resveratrol binding to human serum albumin

Resveratrol (Res), a polyphenolic compound found largely in the skin of red grape and wine, exhibits a wide range of pharmaceutical properties and plays a role in prevention of human cardiovascular diseases [Pendurthi et al., Arterioscler. Thromb. Vasc. Biol. 19, 419-426 (1999)]. It shows a strong affinity towards protein binding and used as inhibitor for cyclooxygenase and ribonuclease reductase. The aim of this study was to examine the interaction of resveratrol with human serum albumin (HSA) in aqueous solution at physiological conditions, using a constant protein concentration (0.3 mM) and various pigment contents (microM to mM). FTIR, UV-Visible, CD, and fluorescence spectroscopic methods were used to determine the resveratrol binding mode, the binding constant and the effects of pigment complexation on protein secondary structure. Structural analysis showed that resveratrol bind non-specifically (H-bonding) via polypeptide polar groups with overall binding constant of K(Res) = 2.56 x 10(5) M(-1). The protein secondary structure, analysed by CD spectroscopy, showed no major alterations at low resveratrol concentrations (0.125 mM), whereas at high pigment content (1 mM), major increase of alpha-helix from 57% (free HSA) to 62% and a decrease of beta-sheet from 10% (free HSA) to 7% occurred in the resveratrol-HSA complexes. The results indicate a partial stabilization of protein secondary structure at high resveratrol content.     

The effects of poly(ethylene glycol) on the solution structure of human serum albumin

Protein physical and chemical properties can be altered by polymer interaction. The presence of several high affinity binding sites on human serum albumin (HSA) makes it a possible target for many organic and polymer molecules. This study was designed to examine the interaction of HSA with poly(ethylene glycol) (PEG) in aqueous solution at physiological conditions. Fourier transform infrared, ultraviolet-visible, and CD spectroscopic methods were used to determine the polymer binding mode, the binding constant, and the effects of polymer complexation on protein secondary structure.The spectroscopic results showed that PEG is located along the polypeptide chains through H-bonding interactions with an overall affinity constant of K = 4.12 x 10(5) M(-1). The protein secondary structure showed no alterations at low PEG concentration (0.1 mM), whereas at high polymer content (1 mM), a reduction of alpha-helix from 59 (free HSA) to 53% and an increase of beta-turn from 11 (free HSA) to 22% occurred in the PEG-HSA complexes (infrared data). The CDSSTR program (CD data) also showed no major alterations of the protein secondary structure at low PEG concentrations (0.1 and 0.5 mM), while at high polymer content (1 mM), a major reduction of alpha-helix from 69 (free HSA) to 58% and an increase of beta-turn from 7 (free HSA) to 18% was observed.     

Interaction of cisplatin drug with RNase A.

cis-Pt(NH3)2Cl2 (cisplatin) is an antitumor drug with many severe toxic side effects including enzymatic structural changes associated with its mechanism of action. This study is designed to examine the interaction of cisplatin drug with ribonuclease A (RNase A) in aqueous solution at physiological pH, using drug concentration of 0.0001 mM to 0.1 mM with final protein concentration of 2% w/v. Absorption spectra and Fourier transform infrared (FTIR) spectroscopy with its self-deconvolution, second derivative resolution enhancement and curve-fitting procedures were used to characterize the drug binding mode, association constant and the protein secondary structure in the cisplatin-RNase complexes. Spectroscopic results show that at low drug concentration (0.0001 mM), no interaction occurs between cisplatin and RNase, while at higher drug concentrations, cisplatin binds indirectly to the polypeptide C=O, C-N (via H2O or NH3 group) and directly to the S-H donor atom with overall binding constant 5.66 x 10(3)M(-1). At high drug concentration, major protein secondary structural changes occur from that of the alpha-helix 29% (free enzyme) to 20% and beta-sheet 39% (free enzyme) to 45% in the cisplatin-RNase complexes. The observed structural changes indicate a partial protein unfolding in the presence of cisplatin at high drug concentration.     

Retinol and retinoic acid bind human serum albumin: stability and structural features

Vitamin A components, retinol and retinoic acid, are fat-soluble micronutrients and critical for many biological processes, including vision, reproduction, growth, and regulation of cell proliferation and differentiation. The cellular uptake of Vitamin A is through specific interaction of a plasma membrane receptor with serum retinol-binding protein. Human serum albumin (HSA), as a transport protein, is the major target of several micronutrients in vivo. The aim of present study was to examine the interaction of retinol and retinoic acid with human serum albumin in aqueous solution at physiological conditions using constant protein concentration and various retinoid contents. FTIR, UV–vis, CD and fluorescence spectroscopic methods were used to determine retinoid binding mode, the binding constant and the effects of complexation on protein secondary structure.

Structural analysis showed that retinol and retinoic acid bind non-specifically (H-bonding) via protein polar groups with binding constants of K_ret = 1.32 (±0.30) × 10⁵ M⁻¹ and K_retac = 3.33 (±0.35) × 10⁵ M⁻¹. The protein secondary structure showed no alterations at low retinoid concentrations (0.125 mM), whereas at high retinoid content (1 mM), an increase of -helix from 55% (free HSA) to 60% and a decrease of β-sheet from 22% (free HSA) to 18% occurred in the retinoid–HSA complexes. The results point to a partial stabilization of protein secondary structure at high retinoid content.     

Resveratrol Binding to Human Serum Albumin

Abstract

Resveratrol (Res), a polyphenolic compound found largely in the skin of red grape and wine, exhibits a wide range of pharmaceutical properties and plays a role in prevention of human cardiovascular diseases [Pendurthi et al., Arterioscler. Thromb. Vasc. Biol. 19, 419–426 (1999)]. It shows a strong affinity towards protein binding and used as inhibitor for cyclo- oxygenase and ribonuclease reductase. The aim of this study was to examine the interaction of resveratrol with human serum albumin (HSA) in aqueous solution at physiological conditions, using a constant protein concentration (0.3 mM) and various pigment contents μM to mM). FTIR, UV-Visible, CD, and fluorescence spectroscopic methods were used to determine the resveratrol binding mode, the binding constant and the effects of pigment complexation on protein secondary structure.

Structural analysis showed that resveratrol bind non-specifically (H-bonding) via polypeptide polar groups with overall binding constant of K_Res = 2.56× 10⁵ M^?1. The protein secondary structure, analysed by CD spectroscopy, showed no major alterations at low resveratrol concentrations (0.125 mM), whereas at high pigment content (1 mM), major increase of α-helix from 57% (free HSA) to 62% and a decrease of β-sheet from 10% (free HSA) to 7% occurred in the resveratrol-HSA complexes. The results indicate a partial stabilization of protein secondary structure at high resveratrol content.     

Polyamine analogues bind human serum albumin

Polyamine analogues show antitumor activity in experimental models, and their ability to alter activity of cytotoxic chemotherapeutic agents in breast cancer is well documented. Association of polyamines with nucleic acids and protein is included in their mechanism of action. The aim of this study was to examine the interaction of human serum albumin (HSA) with several polyamine analogues, such as 1,11-diamino-4,8-diazaundecane (333), 3,7,11,15-tetrazaheptadecane.4HCl (BE-333), and 3,7,11,15,19-pentazahenicosane.5HCl (BE-3333), in aqueous solution at physiological conditions using a constant protein concentration and various polyamine contents (microM to mM). FTIR, UV-visible, and CD spectroscopic methods were used to determine the polyamine binding mode and the effects of polyamine complexation on protein stability and secondary structure. Structural analysis showed that polyamines bind nonspecifically (H-bonding) via polypeptide polar groups with binding constants of K333 = 9.30 x 10(3) M(-1), KBE-333 = 5.63 x 10(2) M(-1), and KBE-3333 = 3.66 x 10(2) M(-1). The protein secondary structure showed major alterations with a reduction of alpha-helix from 55% (free protein) to 43-50% and an increase of beta-sheet from 17% (free protein) to 29-36% in the 333, BE-333, and BE-3333 complexes, indicating partial protein unfolding upon polyamine interaction. HSA structure was less perturbed by polyamine analogues compared to those of the biogenic polyamines.     

3'-azido-3'-deoxythymidine binding to ribonuclease A: model for drug-protein interaction

Ribonuclease A (RNase A) with several high affinity binding sites is a possible target for many organic and inorganic molecules. 3'-Azido-3'-deoxythymidine (AZT) is the first clinically effective drug for the treatment of human immunodeficiency virus (HIV) infection. The drug interactions with protein and nucleic acids are associated with its mechanism of action in vivo. This study was designed to examine the interaction of AZT with RNase A under physiological conditions. Reaction mixtures of constant protein concentration (2%) and different drug contents (0.0001-0.1 mM) are studied by UV-visible, FTIR, and circular dichroism spectroscopic methods in order to determine the drug binding mode, the drug binding constant, and the effects of drug complexation on the protein and AZT conformations in aqueous solution. The spectroscopic results showed one major binding for the AZT-RNase complexes with an overall binding constant of 5.29 x 10(5) M(-1). An increase in the protein alpha helicity was observed upon AZT interaction, whereas drug sugar pucker remained in the C2'-endo/anti conformation in the AZT-RNase complexes.     

Human serum albumin complexes with chlorophyll and chlorophyllin

Porphyrins and their metal derivatives are strong protein binders. Some of these compounds have been used for radiation sensitization therapy of cancer and are targeted to interact with cellular DNA and protein. The presence of several high-affinity binding sites on human serum albumin (HSA) makes it possible target for many organic and inorganic molecules. Chlorophyll a and chlorophyllin (a food-grade derivative of chlorophyll), the ubiquitous green plant pigment widely consumed by humans, are potent inhibitors of experimental carcinogenesis and interact with protein and DNA in many ways. This study was designed to examine the interaction of HSA with chlorophyll (Chl) and chlorophyllin (Chln) in aqueous solution at physiological conditions. Fourier transform infrared, UV-visible, and CD spectroscopic methods were used to determine the pigment binding mode, the binding constant, and the effects of porphyrin complexation on protein secondary structure. Spectroscopic results showed that chlorophyll and chlorophyllin are located along the polypeptide chains with no specific interaction. Stronger protein association was observed for Chl than for Chln, with overall binding constants of K(Chl) = 2.9 x 10(4)M(-1) and K(Chln) = 7.0 x 10(3)M(-1). The protein conformation was altered (infrared data) with reduction of alpha-helix from 55% (free HSA) to 41-40% and increase of beta-structure from 22% (free HSA) to 29-35% in the pigment-protein complexes. Using the CDSSTR program (CD data) also showed major reduction of alpha-helix from 66% (free HSA) to 58 and 55% upon complexation with Chl and Chln, respectively.     

Interaction of RNase A with VO3- and VO2+ ions. Metal ion binding mode and protein secondary structure

Some of vanadyl complexes have shown potential to inhibit RNase activity by acting as transition state analogue, while at the same time not inhibiting DNase. To gain an insight into the interaction of protein with vanadate (VO3-) and vanadyl (VO2+) ions, the present study was designed to examine the binding of ribonuclase A (RNase A) with NaVO3 and VOSO4 in aqueous solution at physiological pH with metal ion concentrations of 0.001 mM to 1 mM, and protein concentration of 2% w/v. Absorption spectra and Fourier transform infrared (FTIR) spectroscopy with self-deconvolution and second derivative resolution enhancement were used to determine the cation binding mode, association constant and the protein secondary structure in the presence of vanadate and vanadyl ions in aqueous solution. Spectroscopic results show that an indirect metal ion interaction occurs with the polypeptide C = O, C-N (via H2O) with overall binding constants of K(VO3-) = 3.93x10(2) M(-1) and K(VO2+) = 4.20x10(3) M(-1). At high metal ion concentrations, major protein secondary structural changes occur from that of the alpha-helix 29% (free enzyme) to 23-24%; beta-sheet (pleated and anti) 50% (free enzyme) to 64-66% and turn 21% (free enzyme) to 10-12% in the metal-RNase complexes. The observed structural changes indicate a partial protein unfolding in the presence of high metal ion concentration.     

The effects of drug complexation on the stability and conformation of human serum albumin

We report different analytical methods used to study the effects of 3\'-azido-3\'-deoxythymidine, aspirin, taxol, cisplatin, atrazine, 2,4-dichlorophenoxyacetic, biogenic polyamines, chlorophyll, chlorophyllin, poly(ethylene glycol), vanadyl cation, vanadate anion, cobalt-hexamine cation, and As2O3, on the stability and secondary structure of human serum albumin (HSA) in aqueous solution, using capillary electrophoresis, Fourier transform infrared, ultraviolet visible, and circular dichroism (CD) spectroscopic methods. The concentrations of HSA used were 4% to 2% or 0.6 to 0.3 mM, while different ligand concentrations were 1 microM to 1 mM. Structural data showed drugs are mostly located along the polypeptide chains with both specific and nonspecific interactions. The stability of drug-protein complexes were in the order K(VO(2+)) 1.2 x 10(8) M(-1) > K(AZT) 1.9 x 10(6) M(-)1 > K(PEG) 4.1 x 10(5) M(-1) > K(atrazine) 3.5 x 10(4) M(-1) > K(chlorophyll) 2.9 x 10(4) M(-1) > K2,4-D 2.5 x 10(4) M-1 > K(spermine) 1.7 x 10(4) M(-1) > K(taxol) 1.43 x 10(4) M(-1) > K(Co(3+)) > 1.1 x 10(4) M(-1) > K(aspirin) 1.04 x 10(4)i(-1) > K(chlorophyllin) 7.0 x 10(3) M(-1) > K(VO(3)(-)) 6.0 x 103 M(-1) > K(spermidine) 5.4 x 10(3) M(-1) > K(putrescine) 3.9 x 10(3) M(-1) > K(As(2)O(3)) 2.2 x 10(3) M(-1)> K(cisplatin) 1.2 x 10(2) M(-1). The protein conformation was altered (infrared and CD results) with major reduction of alpha-helix from 60 to 55% (free HSA) to 49 to 40% and increase of beta-structure from 22 to 15% (free HSA) to 33 to 23% in the drug-protein complexes. The alterations of protein secondary structure are attributed to a partial unfolding of HSA on drug complexation.     

Interaction of cisplatin drug with Na,K-ATPase: drug binding mode and protein secondary structure

cis-Pt(NH(3))(2)Cl(2) (cisplatin) is an antitumor drug with many severe toxic side effects including enzymatic changes associated with its mechanism of action. This study was designed to examine the interaction of cisplatin drug with the Na(+), K(+)-dependent adenosine triphosphatase (Na,K-ATPase) in H(2)O and D(2)O solutions at physiological pH, using drug concentrations of 0.1 microM to 1 mM. UV absorption spectra and Fourier transform infrared difference spectroscopy with its self-deconvolution, second derivative resolution enhancement and curve-fitting procedures were applied to characterize the drug binding mode, the drug binding constant and the protein secondary structure in the cisplatin-ATPase complexes. Spectroscopic evidence showed that at low drug concentration (0.1 microM), cisplatin binds mainly to the lipid portion of the enzyme, whereas at higher drug contents, the Pt cation interaction is through the polypeptide C==O and C-N groups with overall binding constant of K=1.93 x 10(4) M(-1). At high cisplatin concentration (1 mM), drug binding results in protein secondary structural changes from that of the alpha-helix 19.8%; beta-pleated 25.6%; turn 9.1%; beta-antiparallel 7.5% and random 38%, in the free Na,K-ATPase to that of the alpha-helix 22.2%; beta-pleated 23.2%; turn 9.4%; beta-antiparallel 2.2% and random 43%, in the cis-Pt-ATPase complexes.     

Study on the binding of Arsenazo-TB to human serum albumin by Rayleigh light scattering technique and FT-IR

The interaction between Arsenazo-TB and human serum albumin (HSA) was studied by Rayleigh light scattering (RLS) technique and Fourier transformed IR (FT-IR). The binding parameters of Arsenazo-TB with HSA were studied at different temperature of 288, 298, 308, 318 K under the optimum conditions. It is indicated by the Scatchard plots that the binding constant K decreased from 5.03 x 10(7) to 7.13 x 10(6) and the maximum binding number N reduced from 53 to 36 with the increasing of the temperature. The binding process was exothermic, enthalpy driven and spontaneous, as indicated by the thermodynamic analyses, and the major part of the binding energy is hydrophobic interaction. The free energy change deltaG0, the enthalpy change deltaH0 and the entropy change deltaS0 of 288 K were calculated to be -42.46 kJ/mol, -49.17 kJ/mol and 318.15 J/mol K, respectively. The alterations of protein secondary structure in the presence of Arsenazo-TB in aqueous solution were quantitatively calculated from FT-IR spectroscopy with reductions of alpha-helix from 57% to 40% and with increases of beta-sheet from 36% to 39%, beta-turn from 7% to 21%.     

Probing the interaction of anticancer drug temsirolimus with human serum albumin: molecular docking and spectroscopic insight

The binding interaction between temsirolimus, an important antirenal cancer drug, and HSA, an important carrier protein was scrutinized making use of UV and fluorescence spectroscopy. Hyper chromaticity observed in UV spectroscopy in the presence of temsirolimus as compared to free HSA suggests the formation of complex between HSA and temsirolimus. Fluorescence quenching experiments clearly showed quenching in the fluorescence of HSA in the presence of temsirolimus confirming the complex formation and also confirmed that static mode of interaction is operative for this binding process. Binding constant values obtained through UV and fluorescence spectroscopy reveal strong interaction; temsirolimus binds to HSA at 298 K with a binding constant of 2.9 × 10⁴ M^?1implying the strength of interaction. The negative Gibbs free energy obtained through Isothermal titration calorimetry as well as quenching experiments suggests that binding process is spontaneous. Molecular docking further provides an insight of various residues that are involved in this binding process; showing the binding energy to be -12.9 kcal/mol. CD spectroscopy was retorted to analyze changes in secondary structure of HSA; increased intensity in presence of temsirolimus showing changes in secondary structure of HSA induced by temsirolimus. This study is of importance as it provides an insight into the binding mechanism of an important antirenal cancer drug with an important carrier protein. Once temsirolimus binds to HSA, it changes conformation of HSA which in turn can alter the functionality of this important carrier protein and this altered functionality of HSA can be highlighted in variety of diseases.     

Human serum albumin interaction with formononetin studied using fluorescence anisotropy, FT-IR spectroscopy, and molecular modeling methods

Interaction of formononetin with a model transport protein, human serum albumin (HSA), has been studied using fluorescence anisotropy, FT-IR spectroscopy, and molecular modeling methods. Upon binding with HSA, the fluorescence spectrum of formononetin exhibits appreciable hypsochromic shift along with an enhancement in the fluorescence intensity. Gradual addition of HSA led to a marked increase in fluorescence anisotropy (r). From the value of fluorescence anisotropy, it is argued that the drug is located in a restricted environment of protein. The binding constant (K approximately 1.6 x 10(5) M(-1)) and the standard free energy change (DeltaG(0) approximately -29.9 kJ/mol) of formononetin-HSA interaction have been calculated according to the relevant fluorescence data. Fourier transform infrared measurements have shown that the secondary structures of the protein have been changed by the interaction of formononetin with HSA. Computational mapping of the possible binding sites of formononetin revealed the molecule to be bound in the large hydrophobic cavity of subdomain IIA.     

Interaction of rhein with human serum albumin investigation by optical spectroscopic technique and modeling studies

The binding of rhein with human serum albumin (HSA) has been studied in detail by spectroscopic method including circular dichroism (CD), Fourier transformation infrared spectra (FT-IR), fluorescence spectra. The binding parameters for the reaction have been calculated according to Scatchard equation at different temperatures. The plots indicated that the binding of HSA to rhein at 303, 310 and 318 K is characterized by one binding site with the affinity constant K at (4.93+/-0.16)x10(5), (4.02+/-0.16)x10(5) and (2.69+/-0.16)x10(5) M-1, respectively. The secondary structure compositions of free HSA and its rhein complexes were estimated by the FT-IR spectra. FT-IR and curve-fitted results of amide I band are in good agreement with the analyses of CD spectra. Molecular Modeling method was used to calculate the interaction modes between the drug and HSA.     

In vitro plasma protein binding and aqueous aggregation behavior of astaxanthin dilysinate tetrahydrochloride

The tetrahydrochloride salt of astaxanthin di-L-lysinate (lys(2)AST) is a highly water-dispersible astaxanthin-amino acid conjugate, with an aqueous dispersibility of > or = 181.6 mg/mL. The statistical mixture of stereoisomers has been well characterized as an aqueous-phase superoxide anion scavenger, effective at micromolar (microM) concentrations. In the current study, the aqueous aggregation behavior and in vitro plasma protein binding [with fatty-acid-free human serum albumin (HSA) and alpha(1)-acid glycoprotein (AGP)] were investigated with a suite of techniques, including circular dichroism (CD) and UV-vis spectroscopy, ultrafiltration, competitive ligand displacement, and fluorescence quenching. Induced CD bands obtained in Ringer buffer solution of HSA demonstrated high affinity monomeric binding of the compound at low ligand per protein (L/P) ratios (in aqueous solution alone the carotenoid molecules formed card-pack aggregates). The binding constant ( approximately 10(6)M(-1)) and the binding stoichiometry (approximately 0.2 per albumin molecule) were calculated from CD titration data. CD displacement and ultrafiltration experiments performed with marker ligands of HSA indicated that the ligand binding occurred at a site distinct from the main drug binding sites of HSA (i.e., Sites I and II). At intermediate L/P ratios, both monomeric and aggregated ("chirally complexed") binding occurred simultaneously at distinct sites of the protein. At high L/P ratios, chiral complexation predominantly occurred on the asymmetric protein template. The tentative location of the chirally-complexed aggregation on the HSA template was identified as the large interdomain cleft of HSA, where carotenoid derivatives have been found to bind previously. Only weak binding to AGP was observed. These results suggest that parenteral use of this highly potent, water-dispersible astaxanthin-amino acid conjugate will result in plasma protein association, and plasma protein binding at sites unlikely to displace fatty acids and drugs bound at well-characterized binding sites on the albumin molecule.