Interactions between Dansyl Amino Acids and Human Serum Albumin

Differential scanning calorimetry was used to show phase transition in the HSA stationary phase at pH = 7 and 7.5. ... Eric Peyrin, Yves Claude Guilla...
0 downloads 0 Views 192KB Size
Download PDF
Anal. Chem. 1997, 69, 4979-4984

Interactions between Dansyl Amino Acids and Human Serum Albumin Using High-Performance Liquid Chromatography: Mobile-Phase pH and Temperature Considerations Eric Peyrin, Yves Claude Guillaume,* and Christiane Guinchard

Laboratoire de Chimie Analytique, Faculte´ de Me´ decine et Pharmacie, Place Saint Jacques, 25030 Besancon Cedex, France

The reversed-phase liquid chromatography (RPLC) retention mechanism of a series of dansyl amino acids was investigated over a wide range of mobile-phase pH and column temperatures using human serum albumin (HSA) as a chiral stationary phase. Thermodynamic constants for the transfer of a solute from the mobile to the HSA stationary phases were determined. Different van’t Hoff plot shapes were observed with different mobile-phase pH values, indicating a change in the retention mechanism. Enthalpy-entropy compensation revealed that the solute retention mechanism was independent of the compound molecular structure, the same at four pH values (5.5, 6, 6.5, and 8), but changed at pH ) 7 and 7.5. Differential scanning calorimetry was used to show phase transition in the HSA stationary phase at pH ) 7 and 7.5. A new theory was presented to explain that the HSA protein structure balance between a disordered and an ordered solid-like state. Variations of column temperature and mobile-phase pH tend to cause this phase transition between these two states, explaining the observed thermodynamic constant variations with pH and temperature. In the last review period, the number of new protein- and biomolecule-based stationary phases has increased. Hermansson and Grahn1 investigated the basicity of a new chiral column based on silica-immobilized cellobio hydrolase I (CBH I) using 32 different drugs and endogenous compounds. In a similar study, Henriksson et al.2 immobilized intact and fragmented cellobiose hydrolase II (CBH II) chiral stationary phases (CSPs). Aubry et al.3 prepared a new CSP based on mixed immobilized human serum albumin (HSA) and an R1-acid glycoprotein (AGP) and were able to show that this new mixed CSP had a wider range of applications than the individual HSA or AGP CSPs. Gasparrini et al.4 prepared a new CSP by covalently attaching a synthetic C3 symmetry, O-allyl-protected tyrosyl macrocycle to γ-mercaptopropyl silica gel and obtained extremely high separation factors for (1) Hermansson, J.; Grahn, A. J. Chromatogr. 1994, 687, 45. (2) Henriksson, H.; Joensson, S.; Isaksson, R.; Petersson, G. Chirality 1995, 7, 415. (3) Aubry, A. F.; Markoglou, N.; Descorps, V.; Wainer, I. W.; Felix, G. J. Chromatogr. 1994, 685, 1. (4) Gasparrini, F.; Misiti, D.; Villani, C.; Borchardt, A.; Burger, M. T.; Still, W. C. J. Org. Chem. 1995, 60, 4314. S0003-2700(97)00432-0 CCC: $14.00

© 1997 American Chemical Society

boc-protected amino acid derivatives. Massolini et al.5 immobilized hen egg yolk riboflavin protein on silica gel for the separation of chiral drug enantiomers in the reversed-phase mode, and Sinibaldi et al.6 bonded an aminopropyl derivative of the ergot alkaloid (+)-terguride to silica and demonstrated the resolution of dicarboxylic acid, 2-arylcarboxylic acid, and amino acid derivatives. For the multiple interaction protein phases, the mechanisms of interaction are barely understood. Hermansson and Grahn7 studied the chromatographic properties of 29 basic drugs on an AGP column while varying the pH and the concentration of inorganic ions in the mobile phase. Hermansson and Hermansson8 employed eight antiinflammatory activity molecules as model compounds for a thorough study of organic and inorganic modifier-induced effects on enantioselectivity and retention on a chiral AGP column. Karlsson et al.9 investigated the influence of charged and uncharged modifiers and mobile-phase pH on the separation of (R)- and (S)-felodipine on an AGP column. Haginaka et al.10,11 isolated fractions of chicken ovomucoid10 and bovine serum albumin (BSA),11 and Pinkerton et al.12 isolated fractions of turkey ovomucoid in order to determine which fraction contained chiral recognition ability. Allenmark13 attempted to gain further insight into the mechanism by which BSA discriminates between organic acids by using radioisotopically labeled samples, and Loun and Hage14 characterized the thermodynamic processes involved in the binding and separation of (R)- and (S)-warfarin on a HSA column by using frontal analysis. In this paper, the retention mechanism of a series of dansyl amino acids using HSA as stationary phase was investigated over a wide range of mobilephase pH and column temperatures. The shapes of van’t Hoff plots and differential scanning calorimetry (DSC) were used to assess changes in the retention process in relation to temperature and mobile-phase pH. The thermodynamic constants of transfer of these compounds from the mobile (bulk solvent) to the stationary phases were determined. To understand the depen(5) Massolini, G.; De Lorenzi, E.; Ponci, M. C.; Gandini, C.; Caccialanza, G.; Monaco, H. L. J. Chromatogr. 1995, 704, 55. (6) Sinibaldi, M.; Flieger, M.; Cvak, L.; Messina, A.; Pichini, A. J. Chromatogr. 1994, 666, 471. (7) Hermansson, J.; Grahn, A. J. Chromatogr. 1995, 694, 57. (8) Hermansson, J.; Hermansson, I. J. Chromatogr. 1994, 666, 181. (9) Karlsson, A.; Pettersson, K.; Hernqvist, K. Chirality 1995, 7, 147. (10) Haginaka, J.; Seyama, C.; Kanasugi, N. Anal. Chem. 1995, 67, 2539. (11) Haginaka, J.; Kanasugi, N. J. Chromatogr. 1995, 694, 71. (12) Pinkerton, T. C.; Howe, W. J.; Ulrich, E. L.; Comiskey, J. P. Anal. Chem. 1995, 67, 2354. (13) Allenmark, S. Chirality 1993, 5, 295. (14) Loun, B.; Hage, D. S. Anal. Chem. 1994, 66, 3814.

Analytical Chemistry, Vol. 69, No. 24, December 15, 1997 4979

dence of these constants on the pH, a model which takes into account the structure (organization) of the apolar side chains inside the HSA cavity core with which the compound interacts was developed. EXPERIMENTAL SECTION Apparatus. The HPLC system consisted of a Merck Hitachi pump (L7100, Nogent-sur-Marne, France), an Interchim Rheodyne injection valve (Model 7125, Montlucon, France) fitted with a 20 µL sample loop, and a Merck L4500 diode array detector. A HSA protein chiral Shandon column (150 mm × 4.6 mm) was used with a controlled temperature in an Interchim Crococil oven (No. 701). After each use, the column was stored at 4 °C until further use. Mobile-phase flow rate was kept at 1 mL/min. DSC studies were performed using a T.A.2000 Instruments DuPont system (Paris, France). The apparatus was calibrated for temperature by melting high-purity indium. The instrument was flushed with nitrogen. Experiments were run over the temperature range -5 to 45 °C, with a heating rate of 10 °C/min. Solvents and Samples. HPLC grade acetonitrile (Merck) was used without further purification. Sodium hydrogen phosphate and sodium dihydrogen phosphate were supplied by Prolabo (Paris, France). Water was obtained from an Elgastat option water purification system (Odil, Talant, France) fitted with a reverse osmosis cartridge. (1) DL-Dansylnorvaline, (2) DL-dansylvaline, (3) DL-dansylphenylalanine, and (4) DL-dansyltryptophan were obtained from Sigma Aldrich (Saint Quentin, France) and were made fresh daily at a concentration of 20 mg/L in acetonitrile. Sodium nitrate was used as a dead time marker (Merck). The mobile phase consisted of 0.05 M sodium phosphate bufferacetonitrile (85:15 v/v). The mobile-phase pH was adjusted to the values of 5.5, 6, 6.5, 7, 7.5, and 8. Twenty microliters of each solute and a mixture of these were injected, and the retention times were measured. For DSC measurements, HSA stationary phase was obtained from Shandon (Eragny-sur-Oise, France), and native silica was supplied by Merck. Then, 20 mg of each was suspended and mixed in the mobile phase at all pH values. These suspensions were stored at 4 °C for 24 h. After filtration through a Pyrex filter (No. 3), 5-10 mg samples were quickly transferred into aluminum crucibles, which were then sealed and weighed. Temperatures Studies. Compound retention factors were determined over the temperature range -3 to 27 °C. The chromatographic system was allowed to equilibrate at each temperature for at least 1 h prior to each experiment. To study this equilibration, the compound retention time of the D-dansylvaline was measured every hour for 7 h and again after 22, 23, and 24 h. The maximum relative difference of the retention time of this compound was always 0.8%, making the chromatographic system sufficiently equilibrated for use after 1 h. All the solutes were injected three times at each temperature and pH. Once the measurements were completed at the maximum temperature, the column was immediately cooled to ambient conditions to minimize the possibility of any denaturation of the immobilized HSA. Van’t Hoff Plot Calculations. Solute retention is usually expressed in terms of the retention factor k′ by the well-known equation15 (15) Guillaume, Y. C.; Guinchard, C. Anal. Chem. 1996, 68, 2869.

4980

Analytical Chemistry, Vol. 69, No. 24, December 15, 1997

ln k′ ) -∆H°/RT + ∆S°*

(1)

∆S°* ) ∆S°/R + ln φ

(2)

where ∆H° (respectively ∆S°) is the enthalpy (respectively entropy) of transfer of the solute from the mobile to the stationary phase, T the temperature, R the gas constant, and φ the phase ratio (volume of the mobile phase divided by the volume of the stationary phase). ln k′ vs 1/T is called a van’t Hoff plot. For a linear plot, the slope and intercept were respectively -∆H°/R and ∆S°*. For a nonlinear van’t Hoff plot, these thermodynamic data can be calculated using the following method. If an equation can be obtained for the best fit of a curved van’t Hoff plot, then the partial derivative of ln k′ with respect to 1/T will yield a second equation, which represents the negative enthalpy divided by R in relation to temperature. Using eq 1, ∆S°* can, therefore, be determined at a particular temperature. RESULTS AND DISCUSSION Van’t Hoff Plots. (a) For pH ) 5.5, 6, 6.5, and 8. The van’t Hoff plots were all linear for D- and L-dansyl amino acids, showing that there was no change in the retention mechanism. The correlation coefficients (r) for the fits were over 0.989. The typical standard deviations of the slope and intercept obtained were respectively 0.005 and 0.03. Figure 1 shows the van’t Hoff plots for the solute L-dansylphenylalanine at the four pH values. Table 1 contains a complete list of ∆H° and ∆S°* values for all solutes at pH ) 5.5, 6, 6.5, and 8. Both ∆H° and ∆S°* were always negative. (b) For pH ) 7 and 7.5. The van’t Hoff plots for all DLdansyl amino acids showed distinct changes in slope, which is indicative of a modification of the solute retention mechanism. All the curved plots were a good fit using a second-order polynomial. The correlation coefficient of these fits were in excess of 0.986. Figure 2 shows the van’t Hoff plots for the solute L-dansylphenylalanine for these two pH values. The change appeared at a low temperature, T*, between 1 and 7 °C. Table 2 contains ∆H° and ∆S°* values at different temperatures for all solutes with these two pH values. When T was less than or equal to the critical value T*, ∆H° and ∆S°* were positive, and when T g T*, ∆H° and ∆S°* were negative. Enthalpy-Entropy Compensation. The compensation enthalpy-entropy can be expressed by the formula15

∆G°β ) ∆H° - β∆S°

(3)

where ∆G°β is the Gibbs free energy of a physicochemical interaction at a compensation temperature β. Combining eqs 1-3, the following equations are obtained:

ln k′T ) ln k′β - ∆H°/R(1/T - 1/β)

(4)

ln k′β ) -∆G°β/(Rβ) + ln φ

(5)

A plot of ln k′T (for T ) 293 K) versus -∆H° was tested for each of the DL-dansyl amino acid compounds when the pH had values of 5.5, 6, 6.5, and 8. All the correlation coefficients for the linear fits were at least equal to 0.972. For example, for L-dansylphenylalanine, r was 0.978. This degree of correlation confirmed that the retention mechanism for the solute molecule was independent

Table 1. Thermodynamic Parameters ∆H° (kJ/mol) and ∆S°* with Standard Deviations (in Parentheses) at Different Bulk Solvent pH for the DL-Dansyl Amino Acids Transfer from the Bulk Solvent to the HSA Protein pH

L,d.nora L,d.val L,d.try L,d.phe D,d.nor D,d.val D,d.try D,d.phe

L,d.nor L,d.val L,d.try L,d.phe D,d.nor D,d.val D,d.try D,d.phe a

5.5

6.0

6.5

8

-6.4(0.1) -5.3(0.1) -9.7(0.4) -9.1(0.1) -7.7(0.1) -5.5(0.1) -12.2(0.3) -10.7(0.1)

∆H° -6.9(0.2) -5.7(0.2) -10.4(0.3) -9.6(0.1) -8.2(0.2) -5.9(0.1) -12.9(0.4) -11.1(0.3)

-8.5(0.3) -7.6(0.3) -10.8(0.4) -10.3(0.2) -10.5(0.5) -7.6(0.3) -13.7(0.5) -12.1(0.1)

-8.9(0.1) -8.7(0.1) -11.4(0.3) -10.8(0.2) -12.1(0.4) -11.9(0.4) -14.4(0.4) -12.9(0.1)

-2.3(0.1) -2.1(0.1) -2.8(0.1) -2.3(0.2) -2.9(0.2) -2.5(0.2) -3.6(0.2) -2.9(0.2)

-3.2(0.1) -3.4(0.2) -4.0(0.3) -3.5(0.3) -4.3(0.5) -3.5(0.4) -4.8(0.1) -3.8(0.2)

-0.7(0.05) -0.6(0.04) -1.5(0.1) -1.3(0.1) -1.3(0.09) -1.5(0.1) -2.2(0.2) -1.8(0.1)

∆S°* -0.9(0.04) -0.9(0.1) -2.1(0.1) -1.9(0.1) -1.7(0.1) -1.8(0.1) -2.7(0.1) -2.0(0.1)

For example, D,d.nor ) D-dansylnorvaline.

(b) for pH ) 7 and 7.5, native silica scans showed no peaks. However, DSC curves of the HSA stationary phase showed thermal features, with a change in baseline ending approximatively at +4 °C (Figures 4 and 5).

Figure 1. Van’t Hoff plots for L-dansylphenylalanine at pH ) 5.5 ([), 6 (9), 6.5 (2), and 8 (×) for its transfer from the bulk solvent to the HSA stationary phase.

of the pH value chosen from among these four previous values but changed for pH ) 7 and 7.5. Enthalpy-entropy compensation was also used to test the variation of the retention mechanism of the solute molecule with its molecular structure. A plot of ln k′T (for T ) 293 K) calculated for each of the DLdansyl amino acid against -∆H° determined at pH ) 5.5, 6, 6.5, and 8 was drawn. The correlation coefficients for the linear fits were in excess of 0.978; for example, at pH ) 6, r was 0.980. This can be considered to be adequate to verify enthalpy-entropy compensation. Thus, the retention mechanism was independent of both the size and the structure of the molecule for these four pH values. DSC Measurements. DSC was performed on the HSA stationary phase at pH ) 5.5, 6, 6.5, 7, 7.5, and 8, as well as on the native silica from which it was made, to show the presence of phase transitions in the HSA stationary phases. Each experiment was repeated five times showing a perfect reproductibility of the results obtained. Upon the heating of each sample over the temperature range -5 to 45 °C: (a) for pH ) 5.5, 6, 6.5, and 8, both native silica and the protein stationary phase scans showed no peaks or changes in baseline over the temperature range examined. For example, DSC scans of native silica and the HSA stationary phase at pH ) 8 are set out in Figure 3.

THEORY A novel theory was developed to explain the pH retention mechanism dependence due to the existence of this phase transition in the HSA stationary phase. This model was based on the structure of human serum albumin, where the ligands of region binding to human serum albumin are located in hydrophobic cavities in subdomains IIA and IIIA, which exhibit similar chemistry.16 The nonpolar residues are sequestered into the hydrophobic cavities inside the protein and the polar residues are on the surface. In the cavity core, the apolar side chains are packed very tightly,17 corresponding to a solid- rather than a liquid-like state. This fact was exemplified by large differences in enthalpy and entropy of thermal folding of proteins in comparison with the water-oil process of small nonpolar solutes. For example, at 25 °C, the transfer of small nonpolar solutes from water into oil has approximatively zero enthalpy change and a large positive excess entropy change. In contrast, at 25 °C, protein folding has a negative enthalpy change and a small or negative excess entropy.18 There is both an additional negative entropy and enthalpy of folding in excess of that predicted from nonpolar solutes experiments. The residual enthalpy is due to the dominant van der Waals (VDW) interactions among the side chains inside the protein.19 The residual entropy represents an increase in the ordering on folding relative to that expected for (16) He, X. M.; Carter, D. C. Nature 1992, 358, 209. (17) Hazrpaz, Y.; Gerstein, M.; Chotia, C. Structure 1994, 2, 641. (18) Baldwin, R. L. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 8069. (19) Otzen, D. E.; Rheinncker, M.; Fersht, A. R. Biochemistry 1995, 34, 13051.

Analytical Chemistry, Vol. 69, No. 24, December 15, 1997

4981

Table 2. Thermodynamic Parameters ∆H° (kJ/mol) and ∆S°* with Standard Deviations (in Parentheses) at Different Temperatures for L-Dansylphenylalanine Transfer from the Bulk Solvent to the HSA Protein at pH ) 7 and 7.5 pH T (°C)

7

7.5

-3 0 3 6 9 12 15 18 21 24 27

∆H° 12.3(0.4) 8.9(0.2) 7.2(0.4) 5.5(0.2) -2.6(0.1) -4.5(0.2) -6.4(0.3) -7.6(0.2) -9.3(0.4) -11.2(0.6) -15.1(0.5)

7.3(0.3) 5.6(0.3) 3.2(0.2) 0.4(0.1) -8.5(0.2) -12.7(0.5) -17.3(0.6) -19.4(0.5) -21.1(0.6) -23.4(0.7) -27.2(0.4)

-3 0 3 6 9 12 15 18 21 24 27

∆S°* 65.3(0.9) 51.4(0.8) 41.6(0.6) 21.3(0.5) -2.9(0.1) -4.8(0.2) -10.2(0.3) -17.4(0.4) -23.5(0.4) -29.4(0.3) -32.8(0.5)

35.3(0.6) 28.5(0.8) 24.3(0.4) 17.4(0.5) -2.4(0.2) -21.3(0.5) -49.8(0.6) -55.3(0.7) -66.2(0.4) -75.3(0.7) -90.3(0.6)

Figure 2. Van’t Hoff plots for L-dansylphenylalanine at pH ) 7 ([) and 7.5 (9) for its transfer from the bulk solvent to the HSA stationary phase.

the water-oil transfer of nonpolar solutes. In effect, the residual entropy of protein denaturation is 18 J mol-1 K-1 at 112 °C,20 which is similar to that of the dissociation of solid diketopiperazine, 16 J mol-1 K-1 at 112 °C,20 and considerably different from that of liquid hydrocarbon dissolution, which is -0.5 J mol-1 K-1 at 112 °C.20 It was assumed that the hydrophobic cavity surface has R1 and R2 as radius curvatures. Following the Laplace law, the pressure difference ∆p above and below the surface was given by

Figure 3. DSC scans at pH ) 8 for (A) native silica and (B) HSA stationary phase.

surface, eq 5 was rewritten as

∆p ) (2γ/r)/(1 ( r/R) ∆p ) γ(1/R1 + 1/R2)

where γ is the surface tension. For a spherical surface (R1 ) R2 ) R) the following was obtained:

The negative sign was for a concave surface. For a planar surface, R ) ∞; therefore,

γ(R)/γ(∞) ) 1/(1 ( r/R) ∆p ) (2γ)/R

(8)

(6)

(9)

(7)

When a water molecule with a radius r came tangentially to the

It was striking that our model led to an equation similar to that of Tolman:21

(20) Murphy, K. P.; Privalov, P. L.; Gill, S. J. Science 1990, 247, 559.

(21) Tolman, R. C. J. Chem. Phys. 1949, 17, 333.

4982

Analytical Chemistry, Vol. 69, No. 24, December 15, 1997

Figure 4. DSC scans at pH ) 7 for (A) native silica and (B) HSA stationary phase.

In the disordered solid-like state, there was a gain of freedom of the apolar chains in the cavity core due to minimization of the VDW interactions between these chains. This state corresponded, therefore, to a low hydrophobic state. These steric constraint restrictions imply a maximal tension and a minimal sinuosity of the protein surface. In the ordered solid-like state, there was a hydrophobic stabilization by maximization of the VDW interactions among the residues of the cavity core. This state corresponded, therefore, to a high hydrophobic state. In this case, the surface tension and sinuosity were respectively minimal and maximal. Theoretical Application to the Thermodynamic Constant Variations with pH and Temperature. (a) For pH ) 7 and 7.5. At Temperature T g T*. In this temperature region, following our theory, the cavity was in a high sinuosity, low surface tension, and high hydrophobic state (ordered solid-like state). Thus, the gain in hydrophobic bonds among the nonpolar residues of the cavity facilitated the gain of a full hydrogen bond among the water molecules surrounding the cavity (hydratation shell water). These water molecules were more constrained than those in the bulk solvent. If S′ and S are the entropy densities in the cavity hydratation shell and in the bulk solvent, respectively, and nT the number of ordered water molecules in the shell at temperature T, it can be assumed that S - S′ was proportional to nT; therefore,

S - S′ ∼ nT

Figure 5. DSC scans at pH ) 7.5 for (A) native silica and (B) HSA stationary phase.

γ(R)/γ(∞) ) 1/(1 + 2δ/R)

(10)

In eq 10, which is based on a complex thermodynamic theory, γ(R) is the surface tension of a convex surface of radius R, and δ is an unknown constant. Choi et al.22 used eq 10 to correlate the size and solubility of gases in nonpolar liquids. Tolman21 postulates that δ has only atomic dimensions. In our simple model (eq 9), 2δ was equal to the radius of a water molecule. The dimensionless parameter s, called “sinuosity”, was also introduced:

s ) L/l

(11)

where L was the length of the cavity along its course between two points and l was the shortest length between the same points. s reflected the state of the protein surface. When the cavity surface was planar, sinuosity had a minimum value of 1, and the surface tension was maximum. In principle, no maximum value existed for s. Sinuosity was related to the information content and symmetry of the cavity surface. Thus, both surface tension and sinuosity characterized the organization of the apolar side chains inside the hydophobic cavity. The existence of a thermal effect observed at pH ) 7 and 7.5 in DSC was attributed to a phase transition between a disordered and an ordered state of the protein structure. (22) Choi, D. S.; Jhon, M. S.; Eyring, H. J. Chem. Phys. 1970, 53, 2608.

with S′ e S

(12)

Thus, the weak polar solute DL-dansyl amino acid came into more and more contact with the hydrophobic cavity and unavoidably less contact with the water molecules in the bulk solvent which were less ordered. The decrease in solute-bulk solvent hydrophobic interactions and the increase in the solute-cavity van der Waals interactions would, thus, explain the decreasing negative values of ∆H° and ∆S°* when the temperature increased (Table 2). At Temperature T e T*. In this temperature region, according to our theory, the cavity was in a low sinuosity, high surface tension, and low hydrophobic state (disordered solid-like state). Therefore, it was better for the weak polar solute to be in the bulk solvent where the water molecules surrounding it were well ordered and had good hydrogen bonds (S′ g S). The increase in solute-bulk solvent hydrophobic interactions and the decrease in the solute-cavity VDW interactions would, thus, explain the increasing positive values of ∆H° and ∆S°* when the temperature decreased (Table 2). (b) For pH ) 5.5, 6, 6.5, and 8. In this case, as seen previously, all the van’t Hoff plots were all linear, and no phase transitions were observed in the DSC measurements. Therefore, according to our theory, the HSA protein stationary phase was always in an ordered solid-like state. Thus, as for pH ) 7 and 7.5 with T g T*, ∆H° and ∆S°* were negative values (Table 1). (c) For the Entire pH and Temperature Range (pH ) 5.5, 6, 6.5, 7, 7.5, 8, and -3 e T e 27 °C). The acid group of the DL-dansyl amino acid was not protonated. When the pH increased, the negative net charge of the protein increased. Thus, the intensity of the Coulombic interactions between the solute molecule and the cavity decreased and resulted in lower ∆H° and ∆S°* values (Tables 1 and 2). The interactions involving delocalized electrons of aromatic ring systems made a significant Analytical Chemistry, Vol. 69, No. 24, December 15, 1997

4983

contribution to the negative sign of both ∆H° and ∆S°*. In the low dielectric interior of the protein,23 such interactions are more likely to be further enhanced, resulting, in most cases, in even more negative ∆H° and ∆S°* values for DL-dansyltryptophan and DL-dansylphenylalanine than those obtained for the other compounds (Tables 1 and 2). The negative ∆H° and ∆S°* values increased in magnitude with increasing polarizability of an π electron charge cloud, as observed for DL-dansyltryptophan g DLdansylphenylalanine (Tables 1 and 2). In summary, both chromatographic temperature studies and DSC have been used to obtain information to elucidate the role of the HSA stationary phase on the retention mechanism of dansyl amino acids in RPLC. Enthalpy-entropy compensation revealed that the solute retention mechanism of dansyl amino acids is independent of the compound molecular structure, identical at (23) Ross, P. D.; Subramanian. S. Biochemistry 1981, 20, 3096.

4984

Analytical Chemistry, Vol. 69, No. 24, December 15, 1997

pH ) 5.5, 6, 6.5, and 8, but changes at pH ) 7 and 7.5. It was demonstrated by DSC that this was due to a phase transition in the protein stationary phase. A new theory, which takes into account both the geometry of the binding cavity and its hydrophobic character, explains the solute retention mechanism observed at pH ) 7 and 7.5. ACKNOWLEDGMENT We thank Mireille Thomassin for her technical assistance.

Received for review May 6, 1997. Accepted September 25, 1997.X AC9704321 X

Abstract published in Advance ACS Abstracts, November 15, 1997.