Kinetic Properties and Electron Paramagnetic Resonance Spectra

Proc. Nat. Acad. Sci. USA Vol. 71, No. 6, pp. 2470-2472, June 1974

Kinetic Properties and Electron Paramagnetic Resonance Spectra of the Nitric Oxide Derivative of Hemoglobin Components of Trout (Salmo irideus) (nitric oxide binding/Root effect/pH effects) MAURIZIO BRUNORI, GIANCARLO FALCIONI, AND GIUSEPPE ROTILIO Laboratory of Molecular Biology and Institute of Biochemistry, University of Camerino and C.N.R. Centre of Molecular Biology, Institutes of Biochemistry and Chemistry, University of Rome, Italy Communicated by Jeifries Wyman, January 16, 1974

ABSTRACT The binding of nitric oxide to hemoglobin ponents from trout (Salmo irideu8), i.e., Hb trout I and components of trout (Salmo irideus), i.e., Hb trout I and Hb previously (9-11), Hb trout I is Hb trout lV, has been studied by optical and electron trout IV. As reported paramagnetic resonance spectroscopy. Kinetic studies characterized by cooperative ligand binding (n is about 2.6 show that the Root effect in Hb trout IV is operative also for 02), and by an independence of the shape and position for NO, since a large increase in the dissociation velocity of the 02 binding curve of solvent composition, and in particu- constant (U4) is observed as the pH is decreased below 7. lar of pH. On the other hand, Hb trout IV is characterized Moreover, the time course of the displacement of NO by by a dramatic change of both the shape and position of the CO is heterogeneous, suggesting that a and f3 chains may have different j4 values. Low-temperature X-band electron ligand binding curve with pH. Thus, as the pH is decreased paramagnetic resonance spectra have been recorded with from about 8 to about 6, the value of n in the Hill equation Hb trout I and IV saturated with NO at different pH values. (which is an empirical measure of cooperativity) drops from The spectra of Hb trout IV are strongly pH-dependent. about 2.5 to one or less. The high-pH form (pH 8.1) shows axial symmetry and no Among the various hemoglobin com- resolved hyperfine splitting, while the low-pH form is ponents from trout blood, this is the only one which dis- rhombic with a hyperfine splitting of 6.5 G in the g, region. plays the Root effect, a functional property characteristic of The latter form reflects a more distorted site with a more fish hemoglobins (12). significant delocalization of the unpaired electron on the The study of NO binding by these two components, and of proximal histidine; both features indicate a destabilization of the ligand binding at low pH. On the other hand, their EPR spectra at different saturations and pH values, spectra of Hb trout I are axial at both pH values, with has been used in the present work to gain insight on the molec- hyperfine splitting of 16.5 G, indicating that the site is not ular mechanism of the Root effect. distorted and interacts with the ligand very strongly at either pH. MATERIALS AND METHODS It is well established that the reaction of nitric oxide (NO) Preparation of hemoglobin from trout and separation of the with mammalian hemoglobins yields a derivative (HbNO) components was achieved as previously reported (9). which resembles, in its general properties, oxy and carbonmon- The totally or partially liganded NO derivatives were ob- oxide hemoglobin (HbO2 and HbCO) (1). Although the af- tained by mixing anaerobically a known amount of hemo- finity for NO is almost 106 times higher than that for 02, globin with a known volume of water equilibrated with pure binding of this ligand is characterized by the presence of both NO gas (concentration in solution, 2 mM at 20°). All the homotropic and heterotropic interaction effects, similar to the solutions were carefully degassed, and to ensure complete binding of other ligands by ferrous hemoglobins (2, 3). removal of 02, a small amount of dithionite (about 0.1%) was The NO derivative of hemoglobin, as well as of other heme- added to the buffered hemoglobin solution. proteins, displays an electron paramagnetic resonance (EPR) Addition of different aliquots of NO was made at two pH spectrum due to the presence of an unpaired electron asso- values, i.e., 7.2 and 9.0. After 1-3 hr, the various solutions, ciated with the NO nitrogen atom (4-8). Recently, the corresponding to each fractional saturation, were mixed with structural properties of the NO derivatives of hemoglobin a known volume of concentrated buffer to yield final values and the isolated hemoglobin chains have been elucidated of 6.3, 7.0, and 8.1. The solutions were frozen with liquid through the work of several investigators (4-7). Among other nitrogen (or with cold isopentane at -130°) within 10-15 things, it has been shown that the EPR spectrum of NO sec after mixing, directly inside the EPR tubes, with the hemoglobin is the sum of the spectra of the two component appropriate buffer. chains, a and ,B, which have different spectral features. Thus, Absorption spectra were obtained with a Cary 14 or a there are good indications that the paramagnetic signal arising Beckman DK1 spectrophotometer. pH values were all mea- from the bound NO can be used to monitor the structural sured at room temperature with a Radiometer pH 4 meter. properties of the complex and to study the interactions of the X-band EPR spectra were taken at -160° in a Varian V- NO nitrogen atom with the heme iron and with the N atom of 4502-14 instrument equipped with the 100-kHz field modula- the proximal histidine. tion unit and the variable temperature accessory. The mi- This note reports experiments on the NO-binding properties crowave frequency was 9.15 GHz and the microwave power and the EPR spectra of two of the isolated hemoglobin com- approximately 5 mW. The field modulation amplitude was Abbreviation: EPR, electron paramagnetic resonance. approximately 4 G. 2470 Downloaded by guest on September 29, 2021 Proc. Nat. Acad. Sci. USA 71 (1974) Trout Hemoglobin NO Spectra 2471

-2.0-

o -2.5-

0 .- (D 0 -3.0- -,3

6 6.2 6.4 6.6 6.8 7.0 7.2 7.4 pH

FIG. 1. Dependence on pH on the dissociation velocity constant for NO (jO) from fully saturated Hb trout IV. Conditions: 200 and 0.1 M potassium phosphate buffer; CO at 1 atmosphere (101 kPa); protein concentration about 2.5 puM in heme; obser- vation wavelength = 420 nm. 0 and @ refer to different experiments. MAGNETIC FIELD (GAUSS) 2. EPR of the nitric oxide derivative of Hb RESULTS AND DISCUSSION FIG. spectra trout IV at three pH values: a = 8.1; b = 7.0; c = 6.3, in 0.25 The nitric oxide derivative of both Hb trout I and IV has an M potassium phosphate buffer. Protein concentration 630 /M optical absorption spectrum in the visible and Soret regions in heme; fractional saturation = 100%. which is very similar to that characteristic of the same derivative of mammalian hemoglobin. The following maxima were basis of the broad similarities of the binding properties among observed: a band at X = 570 nm; , band at X = 540 nm; y different ligands of ferrous hemoglobin (1, 16). (Soret) band at X = 418 nm. The time course of the displacement of NO by CO in com- At pH > 7.2 the binding of NO to trout hemoglobin is pletely liganded Hb trout IV is clearly heterogeneous, as is characterized by high affinity. At the protein concentrations reported for the replacement of 02 by CO in human hemo- used (from 0.5 to 1 mM in heme), the fractional saturation globin (17). It would seem, therefore, that in Hb trout IV the with the ligand, measured spectrophotometrically, corre- two types of chains in the tetramer are characterized by sponds within a few percent to that calculated on the basis slightly different dissociation velocity constants for NO and of the stoichiometry of binding (one NO per heme iron) and that j4 (a) and j4 (a) differ by a factor < 5. The data reported of the known concentrations of the protein and the ligand. in Fig. 1 correspond, at each pH, to the value of the initial Hb trout IV is characterized by a large drop in affinity for dissociation velocity constant. both 02 and CO as the pH is decreased below about 7.5 (9, Fig. 2 reports the low-temperature (-160°) EPR spectra of 10). To gain evidence that in Hb trout IV the Root effect is Hb trout IV saturated with NO at three pH values, i.e., 8.1, operative also in the case of NO, a series of kinetic experi- 7.0, and 6.3. It will be seen that the shape of the spectrum is ments was undertaken to determine the pH dependence of the strongly pH-dependent. The spectra of the solutions at higher dissociation velocity constant for NO, using CO as a replacing and lower pH values (8.1 and 6.3) are characterized by the ligand. For all the experiments reported below, the molar following properties. The high-pH form shows axial symmetry ratio of the two ligands (CO/NO) was sufficiently high to re- and no resolved hyperfine structure; it resembles the spectrum sult in a complete replacement, according to: of the NO derivative of the y chains of human Hb (6). The low-pH form has rhombic character and a more resolved hy- HbNO + CO -HbCO + NO. perfine structure, with a splitting of 6.5 G in the gz region; The method provides a measure of the dissociation velocity this form resembles the a-NO chains of human Hb (6). These constant from a molecule which is completely liganded at are the dominant features of the spectra, although it cannot all times, and, therefore, yields a rate constant which cor- be excluded that each of the species contains a certain (minor) responds to the last step in the Adair scheme (i.e., jo in the proportion of the other component. case of the reaction of NO with tetrameric hemoglobin) (1, On the basis of previous work on the NO derivatives of 2, 13). various hemeproteins (4-8), an interpretation of the signifi- Fig. 1 depicts the dependence of logj4 on pH and shows that cance of the spectra may be given. In particular, the form as the pH is decreased from 7.4 to 6.3 the value of the dissocia- observed at pH 6.3, in view of its rhombic symmetry, reflects tion velocity constant increases about 50-fold. Thus the half- a more distorted site than that at pH 8.1. Moreover, the time for dissociation, which is about 30 min at pH 7.2, be- presence of a hyperfine structure, which can be attributed to comes less than 1 min at pH 6.3. This dramatic increase in the interaction with the N of the proximal histidino (5), and the off constant with decrease in pH resembles that reported which is evident in the low-pH form, indicates a more signifi- for 02 and CO in the case of Hb trout IV (10, 14) and Hb from cant delocalization of unpaired electron density on the proxi- carp (15), both of which are characterized by the Root ef- mal histidine, and therefore, a decreased interaction with the fect. Therefore, the experiments reported in Fig. 1 provide nitrogen of the ligand. very strong evidence that in Hb trout IV the Root effect is It should be noted that at each pH the same spectral fea- operative also for NO, as may have been anticipated on the tures were observed at various saturations from about 20% Downloaded by guest on September 29, 2021 2472 Biochemistry: Brunori et al. Proc. Nat. Acad. Sci. USA 71 (1974)

Delocalization of the electron density on the N of the proximal histidine may be indicative of a situation in which the complex of the NO molecule with the protein site is desta- bilized. It may be tempting to speculate that in the case of Hb trout IV the EPR signal observed at low pH with the NO derivative reflects a molecule which, albeit still liganded, is in a conformational state characteristic of a "low-affinity" form. In any case, it can be concluded that the ligand-bound form of Hb trout IV undergoes a pH-dependent structural change, clearly indicated by the EPR spectra, and that this structural change occurs in a time scale of less than a few seconds, possibly of fractions of seconds. Our sincere appreciation to Mr. M. Rossi, of the Troticoltura "Rossi," Sefro, Macerata (Italy) for providing trout of an inbred strain. It is a pleasure to thank Dr. W. Blumberg for stimulating MAGNETIC FIELD (GAUSS) discussions. FIa. 3. EPR spectra of the nitric oxide derivative of Hb 1. Antonini, E. & Brunori, M. (1971) Hemoglobin and Myo- trout I at two pH values: a = 8.1; b = 6.3, in 0.25 M potassium globin in Their Reactions with Ligands (North Holland, phosphate buffer. Protein concentration 800 ,uM in heme; Amsterdam). 2. Gibson, 0. H. & Roughton, F. J. W. (1957) J. Physiol. 136, fractional saturation = 50%. 507-518. 3. Chien, J. C. W. (1973) Biochem. Biophys. Res. Commun. 52, up to 100%. In addition, the fractional saturation calculated 1338-1340. 4. Kon, H. (1968) J. Biol. Chem. 243, 4350-4357. from the amplitude of the EPR signal agrees very closely with 5. Kon, H. &. Kataoka, N. (1969) Biochemistry 8, 4757-4762. that obtained on the same solutions by optical spectra, show- 6. Henry, Y. &. Banerjee, R. (1973) J. Mol. Biol. 73, 469-482. ing that no significant dissociation of NO occurred during 7. Shiga, T., Hwang, K. & Tyuma, I. (1969) Biochemistry 8, mixing and freezing. 378-383. 8. Yonetani, T., Yamamoto, H., Erman, J. E., Leigh, J. S. & Fig. 3 shows the low-temperature EPR spectra of Hb trout Reed, G. H. (1972) J. Biol. Chem. 247, 2447-2455. I-NO at two pH values (6.3 and 8.1). At both pH values, the 9. Binotti, I., Giovenco, S., Giardina, B., Antonini, E., species is characterized by axial symmetry, and shows the Brunori, M. & Wyman, J. (1971) Arch. Biochem. Biophys. three-line pattern of the NO hyperfine splitting (16.5 G). 142, 274-280. These features indicate that the site is not very distorted at 10. Brunori, M., Bonaventura, J., Bonaventura, C., Giardina, B., Bossa, F. & Antonini, E. (1973) Mol. Cell. Biochem. 1, either pH value, and that the unpaired electron always in- 189-196. teracts very strongly with the N of nitric oxide. 11. Wyman, J. (1972) Curr. Top. Cell Regul. 6, 209-226. These findings are in substantial agreement with expecta- 12. Riggs, A. (1970) "Properties of fish hemoglobins," Fish tions based on the evidence that the Root effect is operative Physiol. 4, 209-252. 13. Gibson, Q. H. & Roughton, F. J. W. (1955) Proc. Roy. Soc. in Hb trout IV also for NO (see above). Thus the spectra of Ser. B 143, 310-322. the NO derivative of Hb trout I are essentially pH-indepen- 14. Giardina, B., Brunori, M., Binotti, I., Giovenco, S. & dent, and show that, if anything, the stability of the NO Antonini, E. (1973) Eur. J. Biochem. 39, 571-579. complex may be increased going from high to low pH. On the 15. Noble, R. W., Parkhurst, L. J. & Gibson, Q. H. (1970) J. other at low the NO derivative of Hb trout Biol. Chem. 245, 6628-6633. hand, pH displays 16. Wyman, J. (1964) Advan. Protein Chem. 19, 223-286. a very different EPR spectrum, which indicates that the 17. Olson, J. S., Andersen, M. E. & Gibson, Q. H. (1971) J. "site" is in some way strained. Biol. Chem. 246, 5919-5923. Downloaded by guest on September 29, 2021