Sequence of oxygen binding by hemoglobin ( hemoglobin subunits / spin - labeling / electron paramagnetic resonance / heme - heme interactions ) TOSHio ASAKURA AND PUI

A nitroxide spin-label probe was attached directly to a propionic acid group of heme in either the a or the P chain of hemoglobin. The electron paramagnetic resonance (EPR) spectrum of the spin label is altered by the spin-state change of the heme iron to which the spin label is attached. These hybrid hemoglobins showed normal optical and functional properties, indicating that the attachment of the spin label did not perturb the function of hemoglobin. Upon deoxygenation of a-heme-spinlabeled hemoglobin, EPR signals changed proportionally with oxygen saturation (determined by measuring absorption spectra) This result indicates that there is no binding preference between the a and P chains of hemoglobin. However, the cross plot for the fraction of the EPR changes vs. the fraction of oxygen saturation deviated significantly from the diagonal straight line in response to the addition of 2,3diphosphoglycerate and inositol hexaphosphate. The deviation indicated that the EPR change precedes the optical change at low oxygen tension. This result implies that, in the presence of organic phosphate, oxygen binds preferentially to the a subunit of deoxyhemoglobin. This conclusion was supported by the result obtained with P-heme-spin-labeled hemoglobin: the direction of the deviation for 0-heme-spin-labeled hemoglobin in the presence of diphosphogiycerate and inositol hexaphosphate was opposite to that obtained for a-heme-spin-labeled hemoglobin. However, the curve deviated even in the absence of organic phosphate. This deviation for (B-heme-spin-labeled hemoglobin can be explained by the intersubunit interaction of hemoglobin. From these results, it was concluded that, in the absence of organic phosphate, oxygen combines with the a and I chains with equal probability whereas, in the presence of organic phosphate, oxygen binds preferentially to the a chains of hemoglobin. Although the structural, spectral, and functional nonequivalence of isolated a and 13 chains of hemoglobin is well known, the affinity of these individual chains for oxygen in native tetrameric hemoglobin is not well understood. Published data concerning the sequence of oxygen binding by the four hemoglobin subunits are conflicting. Perutz (1) speculated that oxygen will first combine with the a subunit because there is ample room in the heme pocket of a chain in both the oxy and deoxy forms. Gray and Gibson (2) reported that the 13 chain binds carbon monoxide at a faster rate than does the a chain. We have concluded from the analysis of oxygen equilibrium curves of hemoglobin that the oxygen affinities of unlike chains are relatively similar in stripped hemoglobin but differ greatly in the presence of 2,3-diphosphoglycerate (2,3-DPG) and inositol hexaphosphate (InsP6) (3). However, we could not determine which of the two chains has a higher affinity for oxygen. On the basis of NMR studies of normal hemoglobin (Hb A), Johnson and Ho (4) reported that the a subunit binds oxygen more strongly than does the 13 subunit in the presence of organic phosphates. In contrast, by similar The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. 5462 NMR studies, Huang and Redfield (5) did not detect a clear binding preference for oxygen by the a subunit relative to the 13 subunit. These discrepancies may be attributed to the lack of a technique by which the changes specific to each chain can be measured accurately without affecting the oxygen binding properties of hemoglobin. The heme-spin-label method may be an advantageous method by which to study the sequence of oxygen binding by hemoglobin. The direct attachment of a spin label to the propionic group of heme at either the a or the 13 chain does not affect the oxygen binding properties of hemoglobin (6), although the spin label can sense changes in paramagnetism during the oxy-deoxy transition of the specific heme to which it is attached. This paper describes studies on the sequence of oxygen binding to hemoglobin by using aor 13-heme-spin-labeled hemoglobin (a2SL02 or a212SL). EXPERIMENTAL PROCEDURE Heme-Spin-Labeled Hybrid Hemoglobin. The oxy form of heme-spin-labeled hemoglobin in which all four hemoglobin subunits contain mono-spin-labeled protoheme was prepared by a method described elsewhere (6, 7). The a and 13 chains containing spin-labeled heme were isolated from the hemespin-labeled hemoglobin by the method of Bucci and Fronticelli (8). p-Chloromercuribenzoate was removed from the subunits according to DeRenzo et al. (9). The aand 13-heme-spin-labeled hemoglobin subunits were then recombined with nonspin-labeled partner subunits prepared by the same technique from native hemoglobin. This hemoglobin was purified by chromatography on CM-cellulose to remove excess hemoglobin subunits and small amounts of denatured product (6). Measurements. Electron paramagnetic resonance (EPR) spectra of spin-labeled hemoglobins were measured in a quartz flat cell (Varian) by using a Varian E-9 EPR spectrometer at ambient temperature. The EPR spectra were analyzed with a Nicolet Instrument computer, model 1074. The oxygen saturation of the sample in the quartz flat cell was measured optically with a Perkin-Elmer Coleman 126 spectrophotometer. The oxygen equilibrium curves of hemoglobins were determined with an Imai-type automatic apparatus (6, 10). Reagents. Protohemin and other reagents were purchased from Sigma. The spin-label, 2,2,5,5-tetramethyl-3-aminopyrrolidine-l-oxyl, was purchased from Eastman. Abbreviations: 2,3-DPG, 2,3-diphosphoglycerate; InsP6, inositol hexaphosphate; EPR, electron paramagnetic resonance; a2SL,#2, a-heme-spin-labeled hemoglobin; a2,2SL, fl-heme-spin-labeled hemoglobin. Address reprint requests to: Room 8085, Division of Hematology, Children's Hospital of Philadelphia, Philadelphia, PA 19104. Proc. Natl. Acad. Sci. USA 75 (1978) 5463