Mutual effects of proton and sodium chloride on oxygenation of liganded human hemoglobin Oxygen affinities of the a and b subunits Sergei V. Lepeshkevich and Boris M. Dzhagarov Institute of Molecular and Atomic Physics, National Academy of Sciences of Belarus, Minsk, Belarus Normal adult human hemoglobin (HbA) is the classic textbook example of an allosteric protein. The HbA molecule is a heterotetramer consisting of two a sub- units and two b subunits, a 2 b 2 , which are arranged, around a central water-filled cavity, as a pair of ab dimers [1]. Each subunit carries one heme group to which one oxygen molecule binds reversibly. Oxygen- ation of HbA in solution or inside red blood cells is cooperative, i.e. the oxygen affinity for each subunit rises as the other hemes in the same tetramer became saturated with oxygen [2,3]. This cooperative inter- action has been explained as the result of a shift in the equilibrium between two quaternary structures: from the unliganded structure of the low-affinity (T-state) to the high-affinity structure characteristic of the fully sat- urated molecule (R-state) [1,2,4]. Recently, a new inter- pretation of the molecular mechanism of cooperativity and allostery of HbA has been deduced [5–7]. It was shown that ‘stripped’ HbA is a surprisingly inert, mod- erately cooperative O 2 carrier with limited functional diversity if heterotropic effectors are absent. Further- more, it was shown that HbA exhibits amazing func- tional diversity in terms of O 2 affinity, cooperativity and the Bohr effect only in the presence of heterotropic allosteric effectors including hydrogen, phosphates and chloride ions. Such functional diversity is generated primarily by the tertiary structural constraints caused by interaction of the effectors with HbA, especially with oxy-HbA, rather than the T ⁄ R quaternary struc- tural transition. Allosteric effectors allow HbA to take up and release oxygen in response to changing physio- logical conditions. Because tetrameric hemoglobin consists of two types of subunits, differing in structure, knowledge of ligand affinities for each subunit type in Keywords a and b subunits; affinity; human hemoglobin; molecular oxygen; sodium chloride Correspondence B.M. Dzhagarov, Institute of Molecular and Atomic Physics, National Academy of Sciences of Belarus, 70 Nezavisimosti Ave, Minsk 220072, Belarus Fax: +375 17 284 0030 Tel: +375 17 284 1620 E-mail: bmd@imaph.bas-net.by (Received 1 July 2005, revised 2 October 2005, accepted 6 October 2005) doi:10.1111/j.1742-4658.2005.05008.x The different effects of pH and NaCl on individual O 2 -binding properties of a and b subunits within liganded tetramer and dimer of human hemo- globin (HbA) were examined in a number of laser time-resolved spectro- scopic measurements. A previously proposed approach [Dzhagarov BM & Lepeshkevich SV (2004) Chem Phys Lett 390, 59–64] was used to determine the extent of subunit dissociation rate constant difference and subunit affinity difference from a single flash photolysis experiment. To investigate the effect of NaCl concentration on the association and dissociation rate constants we carried out a series of experiments at four different concentra- tions (0.1, 0.5, 1.0 and 2.0 m NaCl) over the pH range of the alkaline Bohr effect. As the data suggest, the individual properties of the a and b sub- units within the completely liganded tetrameric hemoglobin did not depend on pH under salt-free conditions. However, different effects NaCl on the individual kinetic properties of the a and b subunits were revealed. Regula- tion of the O 2 -binding properties of the a and b subunits within the ligan- ded tetramer is proposed to be attained in two quite different ways. Abbreviations BR, bimolecular recombination; GR, geminate recombination; HbA, human hemoglobin. FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6109 the different conformational forms of HbA is a key factor in the complete description of the sigmoidal be- havior of HbA oxygenation. Taking advantage of the photosensitivity of the heme Fe–ligand bond [8–14], flash photolysis has been used extensively in kinetic studies of oxygenated hemo- globins. Recently, the rates of O 2 association [14–16] to both a and to b subunits within triliganded HbA and the efficiency of O 2 escape [15,16] from these sub- units within the completely oxygenated tetramer were obtained using laser photolysis. In a recent study [17], having the individual parameters of bimolecular recombination (BR) for each subunit type within native HbA, it has been proposed that the relative O 2 affinity for HbA subunits could be determined using a single flash photolysis experiment. The approach is based on determination of the bimolecular association rate constant of O 2 rebinding and the quantum yield of BR, c. The latter value is defined as the ratio of the number of O 2 , which succeed in escaping into the sur- rounding medium after photodissociation, to the num- ber of the absorbed light quanta. It should be pointed out that the ratio of the number of the dissociated O 2 molecules to that of the absorbed light quanta defines the primary quantum yield of photodissociation, c 0 [16]. As soon as we can determine the a and b subunit heterogeneity in oxygenation, it is a great importance to find the effect of different heterotropic effectors on the individual parameters of oxygenation for each sub- unit type within the tetrameric protein. Recently [18], effects on the individual properties of the a and b sub- units within oxygenated HbA have been revealed at the pyridoxal 5¢-phosphate modification. In this study, we first attempt to evaluate the mutual effects of pH (over the range of the alkaline Bohr effect) and NaCl on the individual oxygen binding properties of the a and b subunits within liganded tetrameric HbA. In the current literature, there are controversial results con- cerning the pH dependence of the fourth Adair con- stant [19–23]. Change in pH, over the range of the alkaline Bohr effect, does not seem to have any signifi- cant effect on the dissociation [19] and association [20] rate constants, suggesting pH-independent properties for the liganded hemoglobin. This suggestion is con- trary to kinetic results [21,22], showing that the rate of O 2 binding to the triliganded hemoglobin is pH dependent. The pH dependence of the fourth Adair constant was also shown by Imai and Yonetani [23] by determination of the hemoglobin affinity for the fourth ligand molecule. The purpose of this study was twofold. First, the mutual effects of pH and NaCl on the bimolecular association rate constant of O 2 rebinding and the quantum yield of BR for the a and b subunits within the liganded dimer and tetramer of hemoglobin were determined. Second, pH and NaCl effects on the total protein affinity to oxygen and on the subunit affinity were studied [17]. The results indicate that the allosteric effectors modulate the O 2 rebinding to the a and b subunits in two quite different ways. Results Photo-induced HbA reoxygenation was studied when a small amount (0.3–0.5%) of O 2 was released from fully saturated HbA. Bearing in mind the contribution from geminate recombination (GR), we assumed that the primary photodissociation level did not exceed 5% [16]. Such a photoexcitation level was used to ensure the experimental conditions when, statistically, each photo-deoxygenated hemoglobin molecule loses only one molecule of oxygen after photo-irradiation and the tetrameric protein remains in its original state [24,25]. In fact, after photodissociation in the hemoglobin solu- tion, two reactions are initiated simultaneously. One occurs with the participation of the a subunit within HbA and the other with participation of the b subunit: ðaO 2 ; bO 2 ÞðaO 2 ; bO 2 ÞÀ! hv ða; bO 2 ÞðaO 2 ; bO 2 ÞþO 2 À! k 0 a ðaO 2 ; bO 2 ÞðaO 2 ; bO 2 Þ ðaO 2 ; bO 2 ÞðaO 2 ; bO 2 ÞÀ! hv ðaO 2 ; bÞðaO 2 ; bO 2 ÞþO 2 À! k 0 b ðaO 2 ; bO 2 ÞðaO 2 ; bO 2 Þ ð1Þ where (aO 2 , bO 2 )(aO 2 , bO 2 ) denotes the oxyhemoglob- in molecule. In Scheme 1, the oxygenated subunits are shown together with O 2 . The central terms in Scheme 1 represent the case of free O 2 motion in the solution. Here k¢ a and k¢ b are, respectively, the rate constants of BR for the a and b subunits within tri- liganded HbA. Time courses for O 2 rebinding are shown in Figs 1 and 2. The transient absorption decays were analyzed using a standard least-squares technique using home- made software for PC. After kinetic normalization, analysis showed that the time courses for the HbA re- oxygenation over the microsecond (0–4000 ls) time range are fitted with a biexponential function: DA norm ¼ a a Á expðÀk 0 a Á½O 2 ÁtÞþa b ÁexpðÀk 0 b Á½O 2 ÁtÞ ð2Þ where DA norm is a normalized change in optical density of the sample and a a , a b , k¢ a and k¢ b are the ampli- tudes and rate constants of BR. The quantity [O 2 ]is the concentration of molecular oxygen dissolved in the pH and NaCl effects on HbA subunits oxygenation S. V. Lepeshkevich and B. M. Dzhagarov 6110 FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS buffer. Based on considerations described previously [13–16], these two exponential processes are assigned to BR of the a and b subunits within HbA (Model 1). The quantum yield of these processes are defined as c a(b) ¼ 2Æa a(b) Æc. Here c, the quantum yield of BR for tetrameric HbA, is determined using a relative method discussed previously [15]. HbA in 10 mm Tris ⁄ HCl, pH 7.4, buffer is used as a reference standard, for which c ¼ 0.023 ± 0.003 was obtained [16]. Also, the efficiency of O 2 escape from the protein matrix after photodissociation, d, is calculated as the ratio of the quantum yield of BR, c, to that of the primary quan- tum yield of photodissociation, c 0 . Having the individual parameters of bimolecular oxygenation for each subunit type within HbA, the extent of subunit dissociation rate constant difference (k 2 /k 1 ) and the magnitude of subunit affinity difference (K 2 /K 1 ) [17] was calculated using the formulas: k 2 k 1 ¼ k th2 c 02  Á k th1 c 01  À1 Á c 2 c 1 ð3Þ and K 2 K 1 ¼ k th2 c 02  Á k th1 c 01  À1 Á k 0 2 k 0 1 Á c 1 c 2 ð4Þ respectively. Here the subscripts 1 and 2 correspond to two compared subunits within the tetramers as well as within the dimers in similar or different conformations. The kinetic rate, k th , represents the thermal bond- breaking rate. As concluded previously [17], for each oxygenated subunit type in the different conformational forms of the protein, the thermal bond-breaking rate, Fig. 1. Effect of chloride on hemoglobin oxygenation at (A) pH 8.5, (B) pH 7.4 and (C) pH 6.8. Time courses for the recombination of hemoglobin with oxygen in the absence of NaCl (a) and at a NaCl concentration of 0.5 M (b), and 2.0 M (c). Insets shows residuals (a) (b), and (c) from the double exponential fits of the curve (a) (b), and (c), respectively. Excitation wavelength, k exc ¼ 532 nm; detection wavelength, k det ¼ 430 nm. Conditions: 10 mM Tris ⁄ HCl buffer, at 21 °C. Protein concentration, 100 lm in heme. Fig. 2. Normalized time courses for the oxygenation of hemoglobin at pH 6.8 (a, b) and pH 8.5 (c) in the presence of 2.0 M NaCl. Heme concentration: 100 l M (a, c), and 20 lM (b). Excitation wavelength, k exc ¼ 532 nm; detection wavelength, k det ¼ 430 nm. Conditions: 10 m M Tris ⁄ HCl buffer, at 21 °C. S. V. Lepeshkevich and B. M. Dzhagarov pH and NaCl effects on HbA subunits oxygenation FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6111 k th , can be considered constant with an accuracy of 9%. Dzhagarov et al. [26] determined the value, c 0 , for the a and b subunits within oxygenated HbA to be equal to that for the isolated chains, c 0 ¼ 0.23 ± 0.03. Therefore, the ratio of k th /c 0 in Eqns (3) and (4) can be considered constant for each oxygenated subunit type in different conformational forms of tetrameric and dimeric HbA. Knowledge of the association rate constants, k¢ 2(1) , and the quantum yields of BR, c 2(1) , is required only to find simultaneously the extent of subunit dissociation rate constant difference and the magnitude of subunit affinity difference from a single flash photolysis experiment. As soon as we are able to determine the magnitude of the a and b subunit affinity difference (Eqn 4), it seems very important to introduce the total tetramer (dimer) affinity, K t , for the last ligand binding step: K t ¼ K 1 Á K 2 K 1 þ K 2 ð5Þ Here, K 1 and K 2 correspond to the affinity of O 2 bind- ing to the a and b subunits within the triliganded (monoliganded) tetramer (dimer), respectively. Hence it is straightforward to show that the extent of protein total affinity difference can be determined as: K t ðp 1 Þ K t ðp 2 Þ ¼ 1 þ K 1 ðp 2 Þ K 2 ðp 2 Þ K 1 ðp 2 Þ K 1 ðp 1 Þ þ K 1 ðp 2 Þ K 2 ðp 1 Þ ð6Þ Here, p 1 and p 2 correspond to two compared proteins. Subscripts 1 and 2 correspond to two different types of subunits within considered proteins. Rates of O 2 binding to the a and b subunits within liganded hemoglobin measured over the pH range of the alkaline Bohr effect The bimolecular oxygenation parameters measured at different proton concentrations are given in Table 1. In this set of experiments, the HbA concentration is 100 lm in heme. At such a concentration no more than 10%, by weight, of the hemoglobin is in the dimer form [27–30]. However, dimer formation does not appear to affect the measured values of hemo- globin oxygenation because under these pH conditions the kinetic parameters for the last step in ligand bind- ing to tetrameric HbA and those for binding to dimer- ic HbA are almost identical [13–15]. As seen from Table 1, the individual properties of the a and b sub- units do not depend on pH. Small nonprincipal scat- tering of the kinetic parameters, observed at a number of pH values, can be considered ‘error bars’ for the results. Therefore, for later use, the averaged bimole- cular oxygenation parameters in the salt-free buffers (Table 1, Average) can be considered as follows. The BR rate constant for the a subunits within triliganded HbA and the BR quantum yield for the a subunits within completely oxygenated HbA fall in the range 30±3lm )1 Æs )1 and 0.012 ± 0.003, respectively. The association rate constant and the BR quantum yield for the b subunits are found to lie in the range 66±3lm )1 Æs )1 and 0.036 ± 0.006, respectively. The data show an essential ligand-rebinding difference between the a and b subunits. On average, one in every 10 photodissociated O 2 molecules succeeds in escaping from the protein matrix of the triliganded HbA (Table 1, <d>), but only one in every 20 ligands leaves the a subunits (Table 1, <d a >), and in every six ligands leaves the b subunits (Table 1, <d b >). Using Eqns (3) and (4), the dissociation rate con- stant, k, and the O 2 affinity, K , can be derived for both the a and b subunits from the averaged parameters of HbA oxygenation (Table 1, Average). The association and dissociation rate constants for the b subunits are found to exceed 2.2 ± 0.3- and 3.1 ± 0.9-fold, respectively, the corresponding values obtained for the a subunits within HbA. We also found that the O 2 affinity for the a subunits is 1.4 ± 0.3 times higher than that for the b subunits. The data are in a good agreement with previous data [13,14]. Mutual effects of pH and NaCl on the total protein affinity To investigate the effect of NaCl on O 2 binding to the a and b subunits within liganded HbA we carried out a series of experiments at four NaCl concentrations of 0.1, 0.5, 1.0 and 2.0 m. The rate constant of BR and the quantum yield of BR, in the presence of NaCl, gave a direct evidence of significant functional hetero- geneity for the a and b subunits in the last ligand- binding step (Table 2). The change in the total protein affinity to oxygen is derived from the rate constant and quantum yield of BR using Eqns (4) and (6). The NaCl effect is seen at a concentration of 0.1 m (results not shown). At both pH 6.8 and 7.4, the total protein affinity to oxygen decreases as the NaCl concentration is increased up to 0.5 m with respect to the absence of NaCl (Fig. 3A1 and A2), the largest change being at pH 6.8. However, as the salt concentration continues to be increased up to 2.0 m at these pH values (Fig. 3, B1 and C1; B2 and C2), the affinity does not decrease further but increases. At pH 8.5 (Fig. 3, A3, B3 and C3), there is a constant increase in the affinity as a function of increasing NaCl concentration. The tendency for an pH and NaCl effects on HbA subunits oxygenation S. V. Lepeshkevich and B. M. Dzhagarov 6112 FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS increase by a factor of 1.35 ± 0.37 in the oxygen affi- nity is found at 2.0 m NaCl. Effect of NaCl concentration on the rates of O 2 binding to the a and b subunits within liganded dimer Sodium chloride is known to promote the dissociation of liganded hemoglobin [2]. In addition to tetramer– dimer dissociation, the dimer oxygenation is assumed to be altered with increasing ionic strength of the sol- vent [31]. Therefore, to investigate the effect of NaCl on O 2 binding to tetrameric HbA, the contribution of O 2 rebinding with dimer to the total protein oxygen- ation must be taken into account and the effect of NaCl concentration on the rates of O 2 binding to the a and b subunits within the dimer must be found. It is well known [31] that an increase in the dimer fraction with protein dilution at a fixed high salt level can imply a change in the time course for total protein oxygenation. We took this as a starting point for our investigation. Thus, to estimate the contribution of O 2 rebinding with dimer to the total protein oxygenation we performed the following experiment. At pH 6.8 and 8.5 with 2.0 m NaCl (Fig. 2), the HbA concentration is reduced from 100 to 20 lm in heme. Under these con- ditions, the fraction of monomers in solution can be neglected [32]. However, there is an appreciable amount of dimer. Such dilution, at pH 6.8, must lead to an increase in the dimer fraction from 40 ± 75 to 75 ± 90% [27,28,33]. As it can be seen from Table 2, this expected increase in dimer fraction at pH 6.8 leads to a notice- able increase in the association rate constant for a sub- units, k¢ a . It should be emphasized that the protein solution after dilution at pH 6.8 (Figs 3D1, 4C), exhib- its the individual properties of the a and b subunits intermediate between those of the protein solution before the dilution at pH 6.8 (Figs 3C1, 4B) and pH 8.5 (Fig. 3C3, 4D). However, the absence of any detectable changes in oxygenation at the dilution at pH 8.5 (Table 2; Fig. 3C3, D3) suggests that, under these conditions, oxyhemoglobin is completely in the form of dimer. Thus, the dimer oxygenation properties can be determined at pH 8.5 (2.0 m NaCl) at a hemo- globin concentration of 100 or 20 lm in heme. Taking into account the almost identical ligand-binding prop- erties of the subunits within liganded tetrameric and dimeric HbA under salt-free conditions [13,14], the tendency for the increase by a factor of 1.35 ± 0.37 in the total dimer affinity to oxygen, K t , can be found with increasing the salt concentration at pH 8.5 (Fig. 3A3, B3, C3). The value agrees reasonably well with that ($ 1.4) obtained for the liganded dimer in a variety of salt solutions at pH 7.4 and quoted previ- ously [31]. The observed tendency for an increase in the total dimer affinity is caused mainly by the increase by a factor of 2.2 ± 0.9 in the O 2 affinity of the a sub- units within oxygenated dimer (Fig. 4D3). In turn, this a subunit affinity increase results from the remarkable decrease by 1.8 ± 0.6 times in the dissociation rate constant at an insignificant change in the association rate constant (Fig. 4, D2 and D1, respectively). Also, at pH 8.5, the rebinding study reveals an increase in the association and dissociation rate constant for the b subunit within dimer by a factor of 1.62 ± 0.09 and 1.5 ± 0.4, respectively (Fig. 4, D4 and D5). At such rate constant variation, b subunit affinity does not change noticeably (Fig. 4, D6). As a result, at the highest salt level the b subunit within the liganded dimer exhibited a threefold lower affinity than that for the a subunit. Table 1. Kinetic parameters for oxygen rebinding to the oxygenated forms of human hemoglobin after laser photolysis. Protein concentra- tions are 100 l M on a per heme basis. Conditions: 10 mM Tris ⁄ HCl buffer, at 21 °C. pH k¢ a lM )1 Æs )1 k¢ b lM )1 Æs )1 a a % a b % c a , d a a ·10 )2 , ·10 )2 c b , d b a ·10 )2 , ·10 )2 c, d a ·10 )2 , ·10 )2 8.5 31±2 66±3 24±3 76±3 1.1±0.2 [4.8 ± 1.0] 3.4 ± 0.4 [15 ± 3] 2.2 ± 0.3 [9.7 ± 1.6] 7.4 27.9 ± 1.3 63.2 ± 0.9 22 ± 2 78 ± 2 1.01 ± 0.16 [4.4 ± 0.9] 3.6 ± 0.4 [16 ± 3] 2.3 ± 0.3 [10.0 ± 1.7] 6.8 29.0 ± 1.5 65.0 ± 1.8 26 ± 3 74 ± 3 1.3 ± 0.2 [5.5 ± 1.1] 3.7 ± 0.5 [16 ± 3] 2.5 ± 0.3 [10.7 ± 1.8] Average b <30 ± 3> <66 ± 3> <24.5 ± 4.5> <75.5 ± 4.5> <1.2 ± 0.3> <[5.1 ± 1.6]> <3.6 ± 0.6> [16 ± 4] <2.4 ± 0.5> [10 ± 2] a The efficiency of O 2 escape from the protein matrix, d , is presented in square brackets. For the kinetic parameters the uncertainties are presented as 95% confidence intervals. b The average bimolecular oxygenation parameters are given in the row ‘Average’ in the angled brackets. S. V. Lepeshkevich and B. M. Dzhagarov pH and NaCl effects on HbA subunits oxygenation FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6113 Effect of NaCl concentration on the rates of O 2 binding to the a and b subunits within liganded tetramer As evident from the experiment at pH 8.5 and the pre- vious one at pH 7.4 [31], the effect of NaCl concentra- tion on dimer oxygenation is manifested as a slight increase in the total protein affinity to oxygen but not as an affinity decrease. Therefore, the reduction in pro- tein affinity at 0.5 m NaCl at pH 6.8 and 7.4 cannot be attributable solely to dimerization at the increased salt concentration or to the moderate change in the dimer oxygenation. From this reasoning, the ligand-binding properties of the completely liganded tetrameric HbA should be considered as sensitive to proton and NaCl concentra- tions. Therefore, our study indicates that the hemo- globin solution is comprised of dimers and tetramers, whose ligand-binding properties are dependent on the buffer conditions. As the chloride concentration increases at pH 8.5, complete dissociation of tetrameric hemoglobin to dimer without a detectable change in the O 2 -binding properties of the tetramer may be inferred to take place. By contrast, at pH values of 6.8 and 7.4, addition of NaCl to a concentration of 0.5 m results not only in an increase in the dimer fraction [33], but also in a change in tetramer oxygenation. Subsequent increases in salt concentration at these pH values leads, for the most part, to the further tetramer dissociation. Referring to Fig. 3, the largest decrease in the total protein affinity to oxygen and, consequently, the lar- gest change in the O 2 -binding properties of the ligan- ded tetramer are observed at a NaCl concentration of 0.5 m at pH 6.8. Here, the ligand-binding properties of the tetramer can be found if two initial conditions are imposed: (a) $ 30% of the hemoglobin is in the dimer form under these buffer conditions [33]; and (b) total dimer affinity does not vary practically above 0.5 m NaCl [31], so the ligand-binding properties of the Table 2. Mutual effects of proton and NaCl on hemoglobin oxygenation. Conditions: 10 mM Tris ⁄ HCl buffer, at 21 °C. pH Protein conc. l M (in heme) NaCl conc. M k¢ a lM )1 Æs )1 k¢ b lM )1 Æs )1 a a % a b % c a , d a a ·10 )2 , ·10 )2 c b , d b a ·10 )2 , ·10 )2 c, d a ·10 )2 , ·10 )2 8.5 100 None 31±2 66±3 24±3 76±3 1.1±0.2 [4.8 ± 1.0] 3.4 ± 0.4 [15 ± 3] 2.2 ± 0.3 [9.7 ± 1.6] 0.5 31 ± 4 76 ± 3 18 ± 4 82 ± 4 1.0 ± 0.3 [4.3 ± 1.3] 4.4 ± 0.6 [19 ± 3] 2.7 ± 0.3 [12 ± 2] 1.0 32 ± 3 84 ± 2 15 ± 2 85 ± 2 0.84 ± 0.15 [3.7 ± 0.8] 4.7 ± 0.6 [20 ± 4] 2.8 ± 0.3 [12 ± 2] 20 2.0 2.0 35 ± 3 38 ± 5 106.4 ± 1.3 100 ± 5 10.8 ± 1.6 11 ± 2 89.2 ± 1.6 89 ± 2 0.6 ± 0.2 [2.8 ± 1.0] 0.66 ± 0.14 [2.9 ± 0.7] 5.4 ± 0.7 [23 ± 4] 5.3 ± 0.7 [23 ± 4] 3.0 ± 0.4 [13 ± 2], 3.0 ± 0.4 [13 ± 2] 7.4 100 None 27.9 ± 1.3 63.2 ± 0.9 22 ± 2 78 ± 2 1.01 ± 0.16 [4.4 ± 0.9] 3.6 ± 0.4 [16 ± 3] 2.3 ± 0.3 [10.0 ± 1.7] 0.5 16.6 ± 1.1 64.6 ± 1.5 13.9 ± 1.2 86.1 ± 1.2 0.85 ± 0.13 [3.7 ± 0.7] 5.3 ± 0.6 [23 ± 4] 3.1 ± 0.4 [13 ± 2] 1.0 11.9 ± 1.1 70 ± 2 11.3 ± 0.4 88.7 ± 0.4 0.69 ± 0.09 [3.0 ± 0.5] 5.4 ± 0.7 [24 ± 4] 3.1 ± 0.4 [13 ± 2] 2.0 16.1 ± 1.1 94 ± 2 10.0 ± 0.3 90.0 ± 0.3 0.66 ± 0.08 [2.9 ± 0.5] 5.9 ± 0.7 [26 ± 4] 3.3 ± 0.4 [14 ± 2] 6.8 100 None 29.0 ± 1.5 65.0 ± 1.8 26 ± 3 74 ± 3 1.3 ± 0.2 [5.5 ± 1.1] 3.7 ± 0.5 [16 ± 3] 2.5 ± 0.3 [10.7 ± 1.8] 0.5 8.5 ± 0.2 59.1 ± 0.5 14.5 ± 0.3 85.5 ± 0.3 0.92 ± 0.11 [4.0 ± 0.7] 5.4 ± 0.7 [24 ± 4] 3.2 ± 0.4 [14 ± 2] 1.0 10.2 ± 0.7 72.3 ± 1.4 11.8 ± 0.2 88.2 ± 0.2 0.77 ± 0.09 [3.3 ± 0.6] 5.7 ± 0.7 [25 ± 4] 3.3 ± 0.4 [14 ± 2] 2.0 12.9 ± 0.4 91.2 ± 1.0 11.8 ± 0.3 88.2 ± 0.3 0.78 ± 0.10 [3.4 ± 0.6] 5.8 ± 0.7 [25 ± 4] 3.3 ± 0.4 [14 ± 2] 20 2.0 19 ± 4 90 ± 2 9 ± 2 91 ± 2 0.54 ± 0.14 [2.4 ± 0.7] 5.5 ± 0.7 [24 ± 4] 3.0 ± 0.4 [13 ± 2] a The efficiency of O 2 escape from the protein matrix, d, is presented in square brackets. For the kinetic parameters the uncertainties are presented as 95% confidence intervals. pH and NaCl effects on HbA subunits oxygenation S. V. Lepeshkevich and B. M. Dzhagarov 6114 FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS dimer in the presence of 0.5 m NaCl can be considered to be equal to those found at the highest salt level at pH 8.5. Thus, the quantum yield of BR for the a and b subunits within completely oxygenated tetrameric HbA (c a and c b ) are found to lie in the range of 0.011 ± 0.002 and 0.061 ± 0.013, respectively. The rate constant of BR for the a and b subunits within triliganded HbA are k¢ a ¼ 6.9 ± 0.3 lm )1 Æs )1 and k¢ b ¼ 47 ± 3 lm )1 Æs )1 , respectively. Using Eqns (4) and (6), total tetramer affinity is found to be reduced by a factor of three in the pres- ence of NaCl. This decrease is seen to be due to the a and b subunit affinity reduction of 4.0 ± 1.6 and 2.4 ± 0.9 times, respectively. The association rate con- stant for the a subunits is decreased in 4.4 ± 0.5 times in the presence of NaCl, whereas the dissociation rate constant does not vary virtually. In contrast, the b subunits exhibit a larger 1.7 ± 0.6 times dissociation rate constant and a lower 1.40 ± 0.11 times associ- ation rate constant in the presence of NaCl with respect to the absence of NaCl. Discussion It has long been known that the binding of various heterotropic effectors including chloride ions modu- lates the O 2 affinity and cooperative function of HbA [34–36]. Previous measurements [35] have suggested at least two classes of chloride-binding sites. Over the range 0.1–2.5 m NaCl, oxygenated hemoglobin binds chloride ions at high-affinity sites with an intrinsic binding constant of $ 10 m )1 . The data [35,37] have Fig. 3. Total oxygen affinity (K t ) of liganded hemoglobin as a func- tion of NaCl concentration. Bars 1, 2 and 3 are the relative changes in K t at pH 6.8, 7.4 and 8.5, respectively. As a reference, the total oxygen affinity for the liganded hemoglobin under salt-free condi- tions was taken. The uncertainties are presented as 95% confid- ence intervals. Conditions: 10 m M Tris ⁄ HCl buffer, at 21 °C. (A) Protein (100 l M in heme) at 0.5 M NaCl. (B) Protein (100 lM in heme) at 1.0 M NaCl. (C) Protein (100 lM in heme) at 2.0 M NaCl. (D) Protein (20 l M in heme) at 2.0 M NaCl. Fig. 4. The parameters of O 2 binding to the a and b subunits within liganded hemoglobin as a function of NaCl concentration. Bars 1, 2 and 3 are the relative changes in the association (k¢), dissociation (k) rate constants, oxygen affinity (K) for the a subunits, respectively. Bars 4, 5 and 6 are the changes in k¢, k and K for the b subunits, respectively. As a reference, the averaged parameters of O 2 rebinding (Table 1, Aver- age) under salt-free conditions were taken. Uncertainties are presented as 95% confidence intervals. Conditions: 10 m M Tris ⁄ HCl buffer, at 21 °C. (A) Protein (100 l M in heme) at pH 6.8 at 0.5 M NaCl. (B) Protein (100 lM in heme) at pH 6.8 at 2.0 M NaCl. (C) Protein (20 lM in heme) at pH 6.8 at 2.0 M NaCl. (D) Protein (100 lM in heme) at pH 8.5 at 2.0 M NaCl. S. V. Lepeshkevich and B. M. Dzhagarov pH and NaCl effects on HbA subunits oxygenation FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6115 shown that Cl – interacts strongly with HbA but pro- vide no evidence for binding of Na + up to concentra- tions of 0.5 m. Furthermore, the chloride effect is considered to arise indirectly from alterations in water activity [38,39]. Dimer–dimer interactions (e.g. hydro- gen bonds, salt bridges) within the interface might be, to a certain degree, osmotic-pressure dependent. The considered indirect effect arises when the high chloride concentration alters the water activity and conse- quently the hydration of hemoglobin. Recent X-ray investigations have rekindled interest in the links between oxygenation, salt binding and dimer–dimer interactions. It has been shown that fully liganded human HbA can be crystallized under low- salt conditions with a ‘third quaternary structure’, des- ignated R2 [40], whereas at high salt levels the protein is found in the classical R quaternary structure [41]. Based on extensive structural analysis, it has been pro- posed that the R2 state represents a crystallographi- cally trapped intermediate in the transition between the T- and R-states. Later modeling studies have argued that the R2-state was actually the endpoint of the transition from the T-state. The crystallization, under different conditions, of liganded hemoglobin in R, R2, and intermediate forms suggests that a family of conformers (the R e ensemble) coexist in solution [42]. Moreover, a recent NMR experiment [43] at near- physiological conditions of pH, ionic strength and tem- perature showed that the solution structure of HbCO is a dynamic intermediate between two previously solved R and R2 crystal structures. Most likely, this intermediate structure is similar to the RR2 structure reported previously [44]. On the basis of recent X-ray studies [40–42,44], it has been concluded that the ligan- ded HbA may undergo structural and functional chan- ges in response to subtle changes in the ionic strength, the concentration of allosteric effectors. Summaries of our new and previous results [18] for the a and b subunits within liganded tetrameric HbA modified by the interactions with sodium chloride and pyridoxal 5¢-phosphate are shown in Fig. 5. As evi- dent, the rate constant (k¢) and the quantum yield of BR (c) for the a and b subunits are modulated by the interactions of the allosteric effectors with HbA in quite different ways. The decrease in the association rate constant of BR for the a subunits is seen at a practically unchanged quantum yield of BR. By con- trast, the decrease in the association rate constant for the b subunits occurs with the increase in the quantum yield of BR. The results for the b subunits show that there is an inverse correspondence between k¢ and c. The decreased association rate at increased quantum yield may result from a low probability of binding to the heme once the ligand has entered the protein [11]. This could arise from a decreased rate of bond forma- tion between the ligand localized to the region of the heme pocket and the heme iron. This suggestion is consistent with the NMR data [45,46]. By investigating the ring-current shifted proton resonances in the NMR spectra, it has been shown [45,46] that anions (both phosphate and chloride) can affect the tertiary struc- ture around the ligand-binding site of liganded hemo- globin. The conformation of Val(E11) in the a and b subunits relative to the heme plane is quite dependent on the nature of the anions and the pD of the solution as well as on the nature of the ligand. It has been observed [45], that in the liganded hemoglobin under different buffering conditions, Val(E11)b moves closer to the iron atom in the presence of certain anions. It leads to lowering the access of the dissociated ligand to the heme. Consequently, it leads to increasing the inner-most barrier controlling bond formation between the ligand and the heme-iron. These appears to be a direct relationship between the ability of the anions to shift Val(E11)b closer to the iron atom and its ability to lower the ligand affinity. In this study, different NaCl effects on the associ- ation rate constant and the quantum yield of BR (the efficiency of the ligand escape) for the a and b sub- units within the oxygenated tetramer and dimer of human hemoglobin were revealed. As a consequence, the regulation of the affinity for the a and b subunits within the completely liganded tetrameric hemoglobin is proposed to be achieved in two distinctly different ways. The mechanism of the regulation can be unam- biguously determined by the additional study of the GR, i.e. the ligand rebinding from within the protein. Fig. 5. Correlations between the values of the association rate con- stant of BR (k¢) and the quantum yield of BR (c). The data for the a and b subunits within liganded tetrameric HbA are shown in (A) and (B), respectively. Conditions: (1)10 m M Tris ⁄ HCl buffer, pH 6.8– 8.5, at 21 °C. (2) HbA modified with pyridoxal 5¢-phosphate (PLP- HbA), 1.6 mol PLP per tetrameric HbA, 50 m M K 2 HPO 4 buffer, pH 7.4, at 20 °C. (3) PLP-HbA, 6.0 mol PLP per tetrameric HbA, 50 m M K 2 HPO 4 buffer, pH 7.4, at 20 °C. (4) 10 mM Tris ⁄ HCl buffer, pH 6.8, 0.5 M NaCl, at 21 °C. pH and NaCl effects on HbA subunits oxygenation S. V. Lepeshkevich and B. M. Dzhagarov 6116 FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS The time course and the yield of the geminate phase are both sensitive to the immediate environment of the heme, and to the dynamics of structural changes in the protein. Hence, the analysis of the GR parameters can give a deep insight into modulation of the ligand bind- ing properties of the hemoglobin subunits. The GR study is in progress now. Experimental procedures Materials Oxyhemoglobin was isolated from fresh donor blood using the method described previously [47]. For experiments on stripped HbA, it is necessary to use buffers that do not affect the ligand affinity, for example, the phosphate buf- fers. Therefore, the kinetic experiments were carried out in 10 mm Tris ⁄ HCl buffer, at 21 °C. Three pH conditions were used as follows: 8.5, 7.4, and 6.8. The NaCl effects were carried out at concentrations of 0.1, 0.5, 1.0, and 2.0 m. The solubility of O 2 in water depends strongly on NaCl concentration. Conversions from O 2 partial pressures to molarities of dissolved O 2 were made with the following solubility coefficients [48,49]: 1.80 lmÆmmHg )1 (salt-free buffers), 1.74 lmÆmmHg )1 (0.1 m NaCl), 1.50 l mÆmmHg )1 (0.5 m NaCl), 1.26 lmÆmmHg )1 (1.0 m NaCl), and 0.90 lmÆmmHg )1 (2.0 m NaCl). The HbA concentration was 20 and 100 lm in heme. Time-resolved spectroscopy The bimolecular oxygenation parameters were measured using a kinetic laser spectrometer described previously [15,16]. The second harmonic (532 nm) of an Nd:YAG laser was applied as an exciting light pulse. Transient absorption measurements were performed in the spectral region 430–435 nm. The sensitivity of the detection system allowed us to measure photo-induced absorption changes up to 1 · 10 )5 absorbance units per 2500 shots. Acknowledgements The authors are greatly indebted to Dr Vladimir S. Starovoitov for fruitful discussion. The authors thank Anna V. Chistyakova and Dr Nona V. Konovalova for preparing protein solutions. This work was supported by the Belarusian Republican Foundation for Funda- mental Research (Grant B00-176) and the Belarus State Program of Basic Research (Project ‘Spectr-06’). References 1 Perutz MF, Wilkinson AJ, Paoli M & Dodson GG (1998) The stereochemical mechanism of the cooperative effects in hemoglobin revisited. Annu Rev Biophys Biomol Struct 27, 1–34. 2 Antonini E & Brunori M (1971) Hemoglobin and Myo- globin in Their Reaction with Ligands. North-Holland, Amsterdam. 3 Ackers GK, Doyle ML, Myers D & Daugherty MA (1992) Molecular code for cooperativity in hemoglobin. Science 255, 54–63. 4 Parkhust LJ (1979) Hemoglobin and myoglobin ligand kinetics. Annu Rev Phys Chem 30, 503–546. 5 Henry ER, Bettati S, Hofrichter J & Eaton WA (2002) A tertiary two-state allosteric model for hemoglobin. Biophys Chem 98, 149–164. 6 Yonetani T, Park S, Tsuneshige A, Imai K & Kanaori K (2002) Global allostery model of hemoglobin. Modu- lation of O 2 affinity, cooperativity, and Bohr effect by heterotropic allosteric effectors. J Biol Chem 277, 34508–34520. 7 Viappiani C, Bettati S, Bruno S, Ronda L, Abbruzzetti S, Mozzarelli A & Eaton WA (2004) New insights into allosteric mechanisms from trapping unstable protein conformations in silica gels. Proc Natl Acad Sci USA 101, 14414–14419. 8 Gibson QH (1959) The photochemical formation of a quickly reacting form of haemoglobin. Biochem J 71, 293–303. 9 Austin RH, Beeson KW, Eisenstein L, Frauenfelder H & Gunsalus IC (1975) Dynamics of ligand binding to myoglobin. Biochemistry 14, 5355–5373. 10 Mathews AJ, Rohlfs RJ, Olson JS, Tame J, Renaud J-P & Nagai K (1989) The effects of E7 and E11 mutation on the kinetics of ligand binding to R state human hemoglobin. J Biol Chem 264, 16573–16583. 11 Murray LP, Hofrichter J, Henry ER & Eaton WA (1988) Time-resolved optical spectroscopy and structural dynamics following photodissociation of carbon mono- xyhemoglobin. Biophys Chem 29, 63–76. 12 Peterson ES, Shinder R, Khan I, Juczszak L, Wang J, Manjula B, Acharya SA, Bonaventura C & Friedman JM (2004) Domain-specific effector interactions within the central cavity of human adult hemoglobin in solution and in porous sol–gel matrices: evidence for long-range com- munication pathways. Biochemistry 43, 4832–4843. 13 Unzai S, Eich R, Shibayama N, Olson JS & Morimoto H (1998) Rate constant for O 2 and CO binding to the a and b subunits within the R and T states of human hemoglobin. J Biol Chem 273, 23150–23159. 14 Philo JS & Lary JW (1990) Kinetic investigations of the quaternary enhancement effect and a ⁄ b differences in binding the last oxygen to hemoglobin tetramers and dimers. J Biol Chem 265, 139–143. 15 Lepeshkevich SV, Konovalova NV & Dzhagarov BM (2003) Laser kinetic studies of bimolecular oxygenation reaction of a and b subunits within the R state of human hemoglobin. Biokhimia (Moscow) 68, 676–685. S. V. Lepeshkevich and B. M. Dzhagarov pH and NaCl effects on HbA subunits oxygenation FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6117 16 Lepeshkevich SV, Karpiuk J, Sazanovich IV & Dzha- garov BM (2004) A kinetic description of dioxygen motion within alpha- and beta-subunits of human hemoglobin in the R-state: geminate and bimolecular stages of the oxygenation reaction. Biochemistry 43, 1675–1684. 17 Dzhagarov BM & Lepeshkevich SV (2004) Kinetic stu- dies of differences between a- and b-chains of human hemoglobin: an approach for determination of the chain affinity to oxygen. Chem Phys Lett 390 , 59–64. 18 Lepeshkevich SV, Konovalova NV, Stepuro II & Dzha- garov BM (2005) Laser kinetic studies of bimolecular oxygenation reaction of alpha- and beta-subunits within pyridoxal 5¢-phosphate derivatives of human hemoglobin. J Mol Struct 735–736C, 307–313. 19 Gibson QH & Roughton FJW (1955) The kinetics of dissociation of the first oxygen molecule from fully satu- rated oxyhaemoglobin in sheep blood solutions. Proc R Soc, Series B 143, 310–334. 20 MacQuarrie R & Gibson QH (1972) Ligand binding and release of an analogue of 2,3-diphosphoglycerate from human hemoglobin. J Biol Chem 247, 5686–5694. 21 Kwiatkowski LD & Noble RW (1982) The contribution of histidine (HC3) (146b) to the R state Bohr effect of human hemoglobin. J Biol Chem 257, 8891–8895. 22 Dzhagarov BM & Kruk NN (1996) Bohr alkaline effect: regulation of the binding of O 2 to triliganded haemoglo- bin Hb(O 2 ) 3 . Biophys (Russian) 41, 607–612. 23 Imai K & Yonetani T (1975) pH dependence of the Adair constants of human hemoglobin. Nonuniform contribution of successive oxygen binding to the alka- line Bohr effect. J Biol Chem 250, 2227–2231. 24 Sawicki CA & Gibson QH (1976) Quaternary changes in human hemoglobin studied by laser photolysis of car- boxyhemoglobin. J Biol Chem 251, 1533–1542. 25 Gibson QH (1999) Kinetics of oxygen binding to hemo- globin A. Biochemistry 38, 5191–5199. 26 Dzhagarov BM, Galievsky VA, Kruk NN & Yakutovich MD (1999) Photodissociation of oxygenated forms of the native hemoglobin HbA and its isolated a- and b-sub- units and kinetics of molecular oxygen rebinding. Dokl Biophys (translation of Dokl Akad Nauk) 366, 38–41. 27 Guidotti G (1967) Studies on the chemistry of hemo- globin. II. The effect of salts on the dissociation of hemoglobin into subunits. J Biol Chem 242, 3685–3693. 28 Kirshner AG & Tanford C (1964) The dissociation of hemoglobin by inorganic salts. Biochemistry 3, 291–296. 29 Atha DH & Riggs A (1976) Tetramer–dimer dissoci- ation in hemoglobin and the Bohr effect. J Biol Chem 251, 5537–5543. 30 Ip SHC & Ackers GK (1977) Thermodynamic studies on subunit assembly in human hemoglobin. Tempera- ture dependence of the dimer–tetramer association con- stants for oxygenated and unliganded hemoglobin. J Biol Chem 252, 82–87. 31 Doyle ML, Holt JM & Ackers GK (1997) Effect of NaCl on the linkages between O 2 binding and subunit assembly in human hemoglobin: titration of the qua- ternary enhancement effect. Biophys Chem 64, 271–287. 32 Kellett GL & Schachman HK (1971) Dissociation of hemoglobin into subunits. Monomer formation and the influence of ligands. J Mol Biol 59, 387–399. 33 Kellett GL (1971) Dissociation of hemoglobin into sub- units. Ligand-linked dissociation at neutral pH. J Mol Biol 59, 401–424. 34 Eaton WA (1980) The relationship between coding sequences and function in haemoglobin. Nature 284, 183–185. 35 Chiancone E, Norne JE, Forse ´ n S, Antonini E & Wyman J (1972) Nuclear magnetic resonance quadru- pole relaxation studies of chloride binding to human oxy- and deoxyhaemoglobin. J Mol Biol 70, 675– 688. 36 O’Donnell S, Mandaro R, Schuster TM & Arnone A (1979) X-Ray diffraction and solution studies of speci- fically carbamylated human hemoglobin A. Evidence for the location of a proton- and oxygen-linked chlor- ide binding site at valine 1a. J Biol Chem 254, 12204– 12208. 37 Bull TE, Andrasko J, Chiancone E & Forse ´ n S (1973) Pulsed nuclear magnetic resonance studies on 23 Na, 7 Li and 35 Cl binding to human oxy- and carbon monoxy- haemoglobin. J Mol Biol 73, 251–259. 38 Colombo MF, Rau DC & Parsegian VA (1992) Protein solvation in allosteric regulation: a water effect on hemoglobin. Science 256, 655–659. 39 Salvay AG, Grigera JR & Colombo MF (2003) The role of hydration on the mechanism of allosteric regulation: in situ measurements of the oxygen-linked kinetics of water binding to hemoglobin. Biophys J 84, 564–570. 40 Silva MM, Rogers PH & Arnone A (1992) A third quaternary structure of human hemoglobin A at 1.7-A ˚ resolution. J Biol Chem 267, 17248–17256. 41 Shaanan B (1983) Structure of human oxyhaemoglobin at 2.1 A ˚ resolution. J Mol Biol 171, 31–59. 42 Mueser TC, Rogers PH & Arnone A (2000) Interface sliding as illustrated by the multiple quaternary struc- tures of liganded hemoglobin. Biochemistry 39, 15353– 15364. 43 Lukin JA, Kontaxis G, Simplaceanu V, Yuan Y, Bax A & Ho C (2003) Quaternary structure of hemoglobin in solution. Proc Nat Acad Sci USA 100, 517–520. 44 Safo MK & Abraham DJ (2005) The enigma of the liganded hemoglobin end state: a novel quaternary structure of human carbon monoxyhemoglobin. Biochemistry 44, 8347–8359. 45 Lindstrom TR & Ho C (1973) Effects of anions and ligands on the tertiary structure around ligand binding site in human adult hemoglobin. Biochemistry 12, 134– 139. pH and NaCl effects on HbA subunits oxygenation S. V. Lepeshkevich and B. M. Dzhagarov 6118 FEBS Journal 272 (2005) 6109–6119 ª 2005 The Authors Journal Compilation ª 2005 FEBS [...]... Lepeshkevich and < /b> B M Dzhagarov 46 Ho C (1992) Proton < /b> nuclear magnetic resonance studies on < /b> hemoglobin:< /b> cooperative interactions and < /b> partially liganted intermediates Adv Protein Chem 43, 153–311 47 Bucci E & Fronticelli C (1965) A < /b> new method for the < /b> preparation of < /b> a < /b> and < /b> b subunits of < /b> human < /b> hemoglobin < /b> J Biol Chem 240, 551–552 pH and < /b> NaCl effects < /b> on < /b> HbA subunits oxygenation < /b> 48 Haire RN & Hedlund BE (1977) Thermodynamic... oxygenation < /b> 48 Haire RN & Hedlund BE (1977) Thermodynamic aspects of < /b> the < /b> linkage between binding of < /b> chloride < /b> and < /b> oxygen < /b> to human < /b> hemoglobin < /b> Proc Natl Acad Sci USA 74, 4135–4138 49 Griva ZI, Kotz VA & Tomarchenko SL, eds (1967) Handbook of < /b> Chemistry IV Khimia, Leningrad FEBS Journal 272 (2005) 6109–6119 ª 2005 The < /b> Authors Journal Compilation ª 2005 FEBS 6119 . Mutual effects of proton and sodium chloride on oxygenation of liganded human hemoglobin Oxygen affinities of the a and b subunits Sergei. rate con- stant, k, and the O 2 affinity, K , can be derived for both the a and b subunits from the averaged parameters of HbA oxygenation (Table 1, Average).