pH-Induced motional and conformational changes of amino acids. A

Y. Van Haverbeke, R. N. Muller, and L. Vander Elst. J. Phys. Chem. , 1984, 88 (21), pp 4978–4980. DOI: 10.1021/j150665a037. Publication Date: Octobe...
0 downloads 0 Views 352KB Size
4978

J . Phys. Chem. 1984, 88, 4978-4980

the Na2 in the molecule larger than the Na2(lZ,+) bond length but smaller than the Na2+(ZZg+).Indeed from the above geometry we obtain -6.3 bohrs which is 6.0 < 6.3 < 7.0, where the two extremes refer to Na2 and Naz+ bond lengths, respectively. Our MCSCF+CI calculation predicts a stabilization of the CNaz in the 3A2state with respect to C- and Na2+of +0.214 hartree. From Coulomb’s law and using the above charge distribution and geometry we obtain 2 X 0.5/4.6 = 0.217 hartree. Conclusions CNa, like CLi, has a 4Z- ground state and is quite ionic. It is bound by at least 21 kcal/mol which at the MCSCF level of theory is 16 kcal/mol less bound than CLi. Our calculations suggest that unlike CLi the 211 state of CNa is at best very slightly bound. The ground state of CNa2 seems to be the highly bent 3Azstate which like CLiz is characterized by electron donation

from the CJ, bonding molecular orbital of the alkali diatomic to the formally empty 2p, orbital of C. As with CLiz there seems to be a one-electron bond between the alkali atoms and an essentially ionic interaction between the resulting alkali dimer cation and C-. This 3A2state is bound relative to the separated atoms by at least 35 kcal/mol and is approximately 7 kcal/mol below the linear 3Z; state. These relative energies are summarized in Figure 6. While the energy separations reported in this work will change with more elaborate calculation we expect the physical picture of the bonding in the 42-state of CNa and the 3A2state of CNaz to survive. Acknowledgment. A.M. thanks the North Atlantic Treaty Organization for financial support in the form of a NATO Science Fellowship. Registry No. CNa, 91585-80-7; Na,C, 91550-38-8.

pH-Induced Motional and Conformational Changes of Amino Acids. A Reexamination by Deuterium Longitudlnal Nuclear Relaxation Y. Van Haverbeke,* R. N. Muller, and L. Vander Elst Department of Organic Chemistry, Faculty of Medicine, State University of Mons, 7000 Mons, Belgium (Received: October 27, 1983; In Final Form: April 18, 1984)

A description of the dynamics of simple amino acids in aqueous solution has been inferred from deuterium relaxation. The deuterium longitudinal relaxation rate, which is dominated by the quadruplar mechanism, is nearly insensitive to paramagnetic impurities and proves therefore to be an excellent mobility probe for [2H5]glycineand 2- [ZHl]alanine. The experimental results are related to the extent of intra- and intermolecular interactions at different pH values.

Introduction Nuclear magnetic resonance studies of conformational and motional behavior of simple amino acids like glycine and alanine have been reported in several papers’-s during the past decade, but experimental data and conclusions are partly discordant. From a I3C N M R investigation performed on solutions free of paramagnetic impurities, Pearson et al.s reported that dipole-dipole relaxation rates of glycine carbons do not change between pH 3 and pH 10 for 1.0 M solutions of glycine; this observation would lead to the conclusion that the molecular motion remains unchanged. In later work, Valentine et ale6observed that I7O line widths of enriched glycine and alanine decrease from pH 1 to pH 7 but increase in the higher pH range. The authors attribute the line-width reduction to a higher mobility of the zwitterion, which is of smaller size because of intramolecular ionic interaction and being less solvated. Recently, Gerothanassis et alasdiscarded this assumption and claimed that the observed line-width decrease is due only to 170 (1) J. P. Behr and J. M. Lehn, J . Chem. SOC.,Perkin Trans. 2, 1488 (1972). (2) R. A. Cooper, R. L. Lichter, and J. D. Roberts, J. Am. Chem. Soc., 95,3724 (1973). (3) I. M. Armitage, H. Huber, H. Pearson, and J. D. Roberts, Proc. Natl. Acad. Sci. U.S.A., 71,2096 (1974). (4) T. K. Leipert and J. H. Noggle, J . Am. Chem. SOC.,97,269 (1975). (5) H. Pearson, D. Gust, I. M. Armitage, H. Huber, J. D. Roberts, R. E. Stark, R. R. Vold, and R. L. Vold, Proc. Natl. Acad. Sei. U.S.A.,72,1599 (1975). (6) B. Valentine, T. St. Amour, R. Walter, and D. Fiat, J. Magn. Reson., 38, 413 (1980). (7) I. P. Gerothanassis, R. Hunston, and J. Lauterwein, Helv. Chim. Acta, 65, 1764 (1982). (8) I. P. Gerothanassis, R. Hunston, and J. Lauterwein, Helu. Chim. Acta, 65, 1774 (1982).

0022-3654/84/2088-4978$01.50/0

quadrupole coupling constant variation. In addition, these authors reported that if EDTA” is added to the solutions, no line-width increase is observed in the higher pH range extending from pH 7 to pH 13. This observation is consistent with other reports of the marked tendency of amino acids toward association with certain paramagnetic Among the possible nuclear probes for molecular dynamic investigation, deuterium is one of the most owing to its dominant quadrupole nuclear relaxation mechanism. In the present work, we report a study of deuterium longitudinal relaxation rates of glycine and alanine observed in a pH range extending from 0 to 13. Experimental Section [ZHS]Glycineand ~ , ~ - 2 - [ ~ H ~ ] a l a were n i n e purchased from Aldrich and Merck Sharpe & Dohme, respectively. The pH of (9) N. Ishida, A. Okubo, H. Kawai, S . Yamazaki, and S . Toda, Agric. Biol. Chem., 44, 263 (1980). (10) B. Henry, J. C. Boubel, and J. J. Delpuech, Polyhedron, 1, 113 (1982). (11) H. H. Mantsch, H. Saito, and I. C. P. Smith, Prog. Nucl. Magn. Reson. Spectrosc., 11, 211 (1977). (12) C. Brevard and J. P. Kintzinger “NMR and the Periodic Table”, R. K. Harris and B. E. Mann, Eds., Academic Press, New York, 1978, p 107. (1 3) Y. Van Haverbeke, A. Maquestiau, R. N. Muller, and L. Vander Elst, Spectrochim. Acta, Part A , 36A,627 (1980). (14) Y.Van Haverbeke, A. Maquestiau, R. N. Muller, and L. Vander Elst, Can. J . Chem., 59, 1701 (1981). (15) I. C. P. Smith and H. H. Mantsch in “NMR Spectroscopy: New Methods and Applications”, G. C. Levy, Ed., American Chemical Society, Washington, DC, 1982, ACS Symp. Ser. No. 191, p 97. (16) H. C. Jarrell, I. C. P. Smith in “The Multinuclear Approach to NMR Spectroscopy”,J. B. Lambert and F. G. Riddell, Eds., Reidel, Dordrecht. The Netherlands, 1983, NATO AS1 Ser. C. 103, p 133.

0 1984 American Chemical Societv

The Journal of Physical Chemistry, Vol. 88, No. 21, 1984 4979

Motional and Conformational Changes of Amino Acids TABLE I ionic ______

PH

form

R,, s-I

1012r,:

s

lo~zrc/%, s

lr

~o~~v,,,,~m 3

[ZH5]Glycine(0.34 M) 0 7 13

G+

2.70 f 0.14

G* G-

1.97 f 0.10 1.75 f 0.09

0 7 13

A+

4.95 f 0.25 3.72 f 0.19 3.67 f 0.18

6.3 f 0.6 (7.0)c 4.6 f 0.4 (4.0, 5.5)c9d 4.1 f 0.4 (3.9)"

1.22 f 0.02 1.09 f 0.02 1.25 f 0.03

5.2 f 0.6 4.2 f 0.5 3.3 f 0.4

10.2 f 1.5 (11.2)e 8.6 f 1.3 (7.2)' 7.1 f 1.1 (7.1)e

9.3 f 1.0 7.7 f 0.9 6.9 f 0.8

15.9 f 2.4 13.8 f 2.1 (10.1)s 12.7 f 1.9

2-[2H,]Alanine(0.34 M) A* A-

11.6 f 1.0 8.7 f 0.8 8.6 f 0.8

1.25 f 0.03 1.13 f 0.02 1.24 f 0.03

O r c obtained by using a value of 170 kHz for the quadrupolar coupling constant. 1. CReference24. /References 21-23. gReferences 22, 23, and 25.

V, deduced by using rs = 1.65 A.26 cReference4. dReference

1

1.5

5

10

P H-

Figure 1. Plot of deuterium relaxation rate of the methylene deuteriums of [ZH5]glycine(0.34 M) vs. pH.

aqueous solution is adjusted by adding HCl, NaOH, or KOH at known amounts of stock solutions of amino acids. The samples (1 mL) were contained in 10-mm tubes. The relaxation rates Rl were measured by the inversion recovery Fourier transform technique (IRFT)I7 or the fast inversion recovery Fourier transform technique (FIRFT)I8 on a Bruker WP-60 spectrometer (frequency 9.2 MHz, temperature 30 OC, 32-100 transients recorded). Viscosity measurements were performed at 30.5 O C on a viscosimeter KPG of Ubbelhode.

Results and Discussion The curves obtained for longitudinal relaxation rates (R,) vs. pH of both amino acids (Figures 1 and 2) exhibit the same profile as those of I7O line-width variation measured on solutions of similar range of concentrations containing nevertheless, it should be pointed out that the variation of deuterium relaxation rate is of a lower extent (35% and 25%). (It is striking that the graphs observed do not exhibit the expected shape for titration curves whereas this particular shape is obtained for simple aliphatic carboxylic acids and amineseZ8) The influence of paramagnetic impurities on observed deuterium relaxation rates is negligible: from the data reported by Espersen et al.,I9 one can calculate that, for a maximum concentration of paramagnetic Mn(I1) ions estimated to M, the a-deuterium s-I. relaxation rates enhancement would be less than 4 X The relaxation rate of deuterium is given by

R1 = l / T I = 3/27r2(ezqQ/h)z~, (1) where rC is the rotational correlation time and e2qQ/h is the quadrupolar coupling constant expressed in hertz." Its value for a deuterium bonded to an sp3 carbon atom is ca. 170 kHz and can be assumed to be insentitive to pH and conformational changes."J5 In the approximation of an isotropic motion, the (17) R. Freeman and H. D. W. Hill, J . Chem. Phys., 53, 4103 (1971). (18) D. Canet, G. C. Levy, and I. R. Peat, J. Mugn. Reson., 18, 199 (1975). (19) W.G. Espersen and R. B. Martin, J . Phys. Chem., 80,161 (1976).

5

10

PH

Figure 2. Plot of deuterium relaxation rate of ~,~-2-[*H~]alanine (0.34 M) vs. pH.

rotational correlation time can be described as a function of molecular volume (V,), solution viscosity ( q ) , and microviscosity coefficient (fJZo

where qs is the water viscosity, qr is equal to q / q s , and V, = 4 / 3 ~ r 2 . Gierer et aLzosuggested a relation between this microviscosity coefficient and the effective solute (ro) and solvent (rJ radii

(3) Thus, from determinations of relaxation rates, measurements of solution viscosity, and knowledge of solvent radius, it is possible to deduce the effective molecular volume of the solute by solving eq 4 as successive approximations. Values calculated by this way 6-7 s r0

+

(+ 1

:;)-3

-Y3rr,3-rckT qros ro

=0

(4)

are in good agreement with those obtained by other methods like viscosity and density measurement^^^-^^ or virial coefficientsZ5 (Table I). Whereas the approximation of an isotropic motion (20) A. Gierer and K. Wirtz, Z . Naturforsch., A , SA, 532 (1953). (21) W. Devine and B. M. Lowe, J. Chem. SOC.A , 2113 (1971). (22) F. J. Millero, A. Lo Surdo, and C. Shin, J . Phys. Chem., 82,784 (1978). (23) J. Kirchnerova, P. G. Farell, and J. T. Edward, J . Phys. Chem., 80, 1974 (1976). (24) F. Shahidi, J . Solution Chem., 12,295 (1983). (25) J. J. Kozak, W. S. Knight, and W. Kauzmann, J . Chem. Phys., 48, 675 (1968). (26) T.R.Camp, "Water and Its Impurities", Reinhold, New York, 1963.

4980

J . Phys. Chem. 1984,88, 4980-4987

is probably questionable in the case of these simple amino acids, it can reasonably be admitted that nuclei centrally located are less sensitive to segmental motions. On transition from the cationic form to the zwitterion, a reduction of 7Jq1 is observed which expressed in percentage is nearly equal for both amino acids (Table I). This higher mobility can be related to a decrease of intermolecular associations (solutesolute as well as solute-solvent), owing to intramolecular ionic interaction between the COO- and NH3+ groups. The pH dependence of carboxylic group autoassociation is w e l l - k n o ~ nand ,~~ its influence on 2H relaxation rates has been observed in the case of [2H3]aceticacid for which the relaxation is 1.25 times more efficient a t pH 3 than at p H 7.28 The fact that the percentage decreases of 2H relaxation rates strongly suggests that not are less than those reported for 1706*8 only variation of the 170quadrupolar coupling constant on deprotonation but also the structural and dynamic effects cited above have to be taken into account to explain the whole I7Oline-width variation. (27) (a) G. R. Nash and C. B. Monk, J. Chem. Soc., 4274 (1957); (b) D.

L.Martin and F. J. C. Rossotti, Proc. Chem. SOC.,London, 60 (1959); (c)

P. Waldstein and L. A. Blatz, J. Phys. Chem. 71, 2271 (1967). (28) Y. Van Haverbeke, R. N. Muller, and L. Vander Elst, unpublished results.

Further pH increase induces the zwitterion to anion transition. Owing to the decrease of solvation on going from -NH3+ to -NH2, a higher mobility and hence a lower "apparent" molecular volume are expected. When the increase of solution viscosity is corrected for in this pH range, the relaxation rate variation confirms this hypothesis which is furthermore corroborated by the behavior of simple aliphatic amines: on deprotonation of the -NH,+ group of l,l-[2H2]butylamine,the 7c/qr ratio variation is ca. -37%'c.28 The difference observed for amino acids is related to the fact that their -NH3+ group is less solvated, owing to the intramolecular interaction with the carboxylate group, leading thus to a less important relaxation rate change on deprotonation. The percentage variation of rC/?l1 observed in the higer pH range is smaller for alanine than for glycine; this is to be attributed to lower hydration of the alanine zwitterion because of steric effects of the methyl group. This effect has already been reported in ultrasonic investigations of various amino acids.29 It has to be noted that, in the pH range located between pH 3 and pH 10, our results confirm those of Pearson et aLs since they showed that the 13C relaxation rates variation is lower than 15%. Registry No. [ZHS]Glycine,4896-77-9; ~ ~ - 2 - [ ~ H ~ ] a l a n3i1024ne, 9 1-6. (29) A. P. Sarvazyan, D. P. Kharakoz, and P. Hemmes, J. Phys. Chem., 83, 1796 (1979).

Aggregation of Uroporphyrin I and Its Metal Derivatives in Aqueous Solution: Raman Difference Spectroscopy and Absorption Spectroscopy J. A. Shelnutt,* M. M. Dobry, Solid State Materials Division, Sandia National Laboratories, Albuquerque, New Mexico 871 85

and J. D. Satterlee University of New Mexico, Albuquerque, New Mexico 871 31 (Received: January 9, 1984) Absorption and resonance Raman difference spectra of uroporphyrin I and several of its metal derivatives are given for monomeric and aggregated forms in aqueous solution. Aggregation, whether induced by addition of molar quantities of salt, acidification, or high porphyrin concentrations (>0.01 M), result in similar changes in both the UV-visible absorption spectrum and the Raman spectrum. In particular, specific large shifts, especially in the Soret band, are found to be characteristic of r--a dimerization and these shifts are only weakly affected by the metal incorporated into the uroporphyrin core. In spite of these large changes in the absorption spectrum, only small -2-3-cm-I shifts are observed for the Raman lines that are well-known indicators of porphyrin electronic and molecular structure. The following paper examines these spectral changes in detail; here, an equilibrium model of the salt-induced dimerization process is described and used to fit the experimental degreeof-aggregation vs. salt-concentrationdata by nonlinear least-squares procedures. The dimerization model is able to accurately fit the experimental data and reasonable values of the parameters of the model result. Variation in one of the parameters quantifies the metal-dependentdifferences in the observed aggregation curves. Acid-induced aggregation is also quantitatively understood in terms of the dimerization model.

Introduction Uroporphyrin and the metallouroporphyrins provide a valuable system for studies of complexation and aggregation phenomena in biological and biomimetic chemistries. This stems from their being monomolecularly dispersed at concentrations used in most spectroscopic studies.lq2 All other porphyrins and metalloporphyrins with carboxylate side chains, including chlorophylls, are aggregated in aqueous solution at these concentrations. Complex formation,'-' for example, can be studied without the (1) Blumberg, W. E.; Peisach, J. J . Biol. Chem. 1965, 240, 870-876. (2) Mauzerall, D. Biochemistry 1965, 4, 1801-1810. (3) Shelnutt, J. A. J. Am. Chem. SOC.1981, 103, 4275-4277. (4) Shelnutt, J. A. J. Am. Chem. SOC.1983, 105, 774-778. (5) Shelnutt, J. A. J . Phys. Chem. 1983, 87, 605-616. (6) Shelnutt, J. A. Inorg. Chem. 1983, 22, 2535-2544. (7) Shelnutt, J. A.; Dobry, M. M. J . Phys. Chem. 1983,87, 3012-3014.

0022-3654/84/2088-4980$01.50/0

complications of either aggregation or the addition of dispersing agents. In this paper the results of a study of aggregation of uroporphyrin free base (H2UroP) and several metallouroporphyrins (MUroP) using Raman difference spectroscopy (RDS) and ultraviolet-visible (UV-visible) absorption spectroscopy are compared to determine the similarities of the aggregation process for distinct metals. On the other hand, differences in aggregation of these metalloporphyrins may shed light on the effect of the metal ion and also the detailed mechanism of aggregation. In a previous paper8 we investigated the aggregation of urohemin (iron(II1) uroporphyrin I) in aqueous solution by proton nuclear magnetic resonance (NMR) spectroscopy. Urohemin is also included in the present study so that the results of the optical (8) Satterlee, J. D.; Shelnutt, J. A. J . Phys. Chem., in press.

0 1984 American Chemical Society