- Email: [email protected]
Photo-dissociation quantum yields of mammalian oxyhemoglobin investigated by a nanosecond laser technique
Biochemical and Biophysical Research Communications 353 (2007) 953–959 www.elsevier.com/locate/ybbrc
Photo-dissociation quantum yields of mammalian oxyhemoglobin investigated by a nanosecond laser technique Ning-li Yang a, Shu-yi Zhang a,*, Pao-kuang Kuo a, Min Qu a, Jian-wen Fang b, Jia-huang Li c, Zi-chun Hua c a
Laboratory of Modern Acoustics, Institute of Acoustics, Nanjing University, Nanjing 210093, China b Department of Physics, Zhejiang Normal University, Jinhua 321004, China c Laboratory of Pharmaceutical Biotechnology, Nanjing University, Nanjing 210093, China Received 12 December 2006 Available online 22 December 2006
Abstract The photo-dissociations of oxyhemoglobin of several mammals, such as human, bovine, pig, horse, and rabbit, have been studied. By means of optical pump-probe technique, the quantum yields for photo-dissociation of these oxyhemoglobin have been determined at pH 7 and 20 °C. A nanosecond laser at 532 nm is used as the pumping source, and a xenon lamp through a monochrometer provides a probe light at 432 nm. The experimental results show that the quantum yields of these mammalian oxyhemoglobin are different from each other, especially for that of rabbit. By analyzing the amino acid sequences and tetramer structures as well as the flexibility and hydrophobicity of the different hemoglobin, possible explanations for the differences are proposed. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Oxyhemoglobin; Photo-dissociation; Quantum yield
Hemoglobin (Hb) is an oxygen carrier in blood. The Hb tetramer consists of two a and two b subunits, where the bchain contains eight helices (A–H) while the a-chain contains the same helices except D helix [1]. Each subunit binds a heme, which can bind a dioxygen. The heme is located at the cranny of the E-helix and F-helix and is exposed to solvent. The interactions between different subunits (a to b chain) are strong, while interactions between the same subunits (a to a chain or b to b chain) are weak. The dissociation of ligands from Hb by light has been studied for many years. But the most studied liganded Hbs are the carbon-monoxide hemoglobin (HbCO) and carbonmonoxide myoglobin (MbCO) [2–4], but a few studies are concerning HbO2. The choice of CO as a ligand in these studies probably comes from the fact that the photo-dissociation quantum yield of HbCO is much higher than that
*
Corresponding author. Fax: +86 25 83313690. E-mail address: [email protected] (S. Zhang).
0006-291X/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.12.118
of oxyhemoglobin (HbO2), and consequently it is more difficult to work with the oxygen derivative of Hb [5]. The quantum yield of photo-dissociation is the efficiency of the transfer from the reactants to the products generated by light irradiation, which is defined as the molecular number of photoproduct species divided by the absorbed photon number [6]. Because the quantum yield is a very sensitive parameter, which may easily be affected by the experimental conditions including temperature, pressure, density and pH state (if in the aqueous state), etc., of the solutions, it would be better to determine the quantum yield value under a specific kind of experiments. The quantum yields of the photo-dissociation of oxygen from HbO2, which are quite small for most of mammals, just have been reported rarely although the mechanisms of the association and dissociation of Hb and ligand molecules have been frequently researched [7], but a few studies are concerning HbO2 [8,9]. In this paper, an optical pump-probe method [10] is used to study the photo-dissociation of several kinds of
954
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
mammalian HbO2 solutions, including human, bovine, porcine, equine, and rabbit HbO2 solutions. The different quantum yields of the photo-dissociation in nanosecond range for these kinds of mammals are obtained, especially much difference for that of rabbit compared with others. The differences of the quantum yields among the five mammals are also discussed by analyzing the discrepancies of the amino acid sequences and tetramer structures although the structures of the Hb of the mammals are very similar with each other. Materials and methods Materials. The Hbs of human, pig, horse, rabbit (Sigma, USA), and bovine (Mingzhu, Dongfang Biological Technology, China) are used as experimental samples. They are all in the form of lyophilized powders and have the same water content, iron percentage, rate of purity, and solubility. The HbO2 solutions are prepared by conventional process as 4.0 3.5
Hb HbO2
3.0
OD
2.5 2.0 1.5 1.0 0.5 0.0 -0.5 300
400
500
600
700
Wavelength (nm) Fig. 1. The absorption spectra of Hb and HbO2 in buffer at pH 7.0.
follows: (1) the Hb powder is dialyzed against 50 mM phosphate buffer at pH 7.0; (2) the solution is reduced with sodium dithionate (Na2S2O4); (3) HbO2 is formed by bubbling O2 through the solutions at 1 atm pressure. Before the experiments, in order to check the oxy- and deoxy-forms of Hbs in the solutions, the absorption spectra are measured with UV-3100 Spectrophotometer, and the optical densities of all the solutions under the light illustration at wavelength 532 nm are determined, as shown in Fig. 1. It is found that the absorption spectra of all these five mammalian Hb and oxy-Hb are almost the same. For comparison, HbCO is also formed by bubbling CO through the solution, which is also prepared with the same conditions. Apparatus and method. The apparatus used to measure the quantum yields is shown in Fig. 2. The sample of HbO2 is put in a quartz cuvette with cross-sectional area of 10 mm2. A Q-switched Nd:YAG laser operating at 532 nm, 10 Hz and pulse width 8 ns is used as the excitation source. The pulsed-laser beam is divided into two beams with a 50/50 beam splitter. One of the divided beams is the pump beam. It is focused by a cylindrical lens onto the sample cuvette. The other is directed into a monitoring pulse ergometer. Meanwhile, a continuous light beam from a xenon lamp through a monochrometer is directed through the cuvette perpendicular with the pump beam and is matched in width. Based on the optical spectra of the oxy- and deoxy-Hb samples shown in Fig. 1, the optical wavelength of the probe beam is chosen at 432 nm, where the difference of optical densities between oxy- and deoxy-Hb is the largest and also sufficiently away from the wavelength of the pumping beam, so the sensitivity for measuring the photo-dissociation will be the highest. In order to suppress the interference of the stray reflections of the pumping beam, the probe beam is then made to pass through multiple 432 nm bandpass interference filters and a second monochrometer and then to a photomultiplier. A digital oscilloscope records the counting rates for further processing and analyzing by a computer. For exploring the photo-dissociation reactions of HbO2 in the solution, the variation of optical density at 432 nm during and after the pumping laser irradiation is detected, which means the variation of optical density is induced by the photolysis reaction of the HbO2. In order to eliminate the error produced by the fluctuation of the pumping laser intensity and the noise of the system, the detected signal is averaged by 300 pulses to improve the signal to noise ratio. All the measurements are taken from the same Hb concentrations, i.e., at the optical density 0.17, of HbO2 solutions for human, bovine, pig, horse, and rabbit, and at 20 °C. Quantum yield calculation. The photo-dissociation reactions involved in the HbO2 solutions may be expressed as
Fig. 2. Experimental system.
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959 k
FeO2 0 Fe þ O2 k
where k and k 0 are the reaction rates of the dissociation and rebind processes, respectively. A first order exponential decay expression A exp (t/s) is used to mathematically fit the rebinding process, where A is the peak value of the signal represented photo-dissociation process and s is the ligand rebinding relaxation time. The quantum yield q is defined as the molecular number of photoproducts divided by the absorbed photon number, then Z K 1 t=s A e dt; ð1Þ q¼ E 0 where E is the energy of the beam irradiating the sample, and K is the constant of the experimental system. Therefore it is clear that q¼K
As : E
ð2Þ
As the pumping pulse energy is in the range of milli-Joule range, the photo-dissociation process is in linear with the energy variation. Changing the pumping pulse energy by means of an optical attenuator and recording the energy values by the pulse ergometer, a set of exponential decay curves can be obtained, and then a set of As for different pumping energies can be fitted, by which the q/K can be obtained from the slope of the plot of As against E. In order to eliminate the effect of the system constant K, the quantum yields of photo-dissociation HbO2 of human, bovine, pig, horse, and rabbit are obtained by comparing with the quantum yield for photo-dissociation HbCO of human, which is similar to the experiments of Saffran et al. [5] and Duddell et al. [11]. Keeping all the conditions including temperature, pressure, pH state, and then the constant K invariable during the experiments, the ratio of the quantum yields of HbO2 of human, bovine, pig, horse, and rabbit to that of human HbCO can be obtained. As the quantum yields of HbCO are determined, the HbO2 of the five mammals can also be determined at all.
955
exponential decay expression A exp (t/s) is used to mathematically fit the rebinding process, where A is the peak value of the signal represented photo-dissociation process and s is the ligand rebinding relaxation time. The experimental measured plots of As VS E for human, bovine, pig, horse, and rabbit HbO2 and human HbCO at the optical density 0.17 under the wavelength 532 nm are illustrated in Fig. 4, which shows the amount of photoproducts As of the five mammals HbO2 and human HbCO under an exposure of pulses with different energies E. All the measurements are taken at 20 °C. The experimental points for each HbO2 lie on a straight line whose slope is proportional to the quantum yield. All of the measured results of photolyses of HbO2 solutions for human, bovine, pig, horse, and rabbit, as well as HbCO of human obtained under the same conditions are listed in Table 1, in which qs/qco is the relative quantum yields, qs represents the quantum yields of HbO2 of human, bovine, pig, horse, and rabbit, and qco represents that of human HbCO. The absolute values of the quantum yields (QY) of different mammals can be obtained if the quantum yield of the HbCO of human is determined. Based on the results of Saffran et al., the quantum yields of human HbCO are taken as 0.46 at 20 °C [5], then the quantum yields of human, bovine, pig, horse, and rabbit are also given in Table 1. It can be seen from the data of Table 1 that the photo-dissociation quantum yields of these mammalian oxy-Hbs are different from each other, especially so for that of rabbit.
Results Discussion When the laser pulse is irradiating on the HbO2 solutions, the ligand is dissociated from the ligand-heme, but the ligand then rebinds to the heme in the dark duration. As an example, the measured decay curves of the rabbit HbO2 and human HbCO are shown in Fig. 3. A first order
a
b
3.0
Experimental result Fitted result
2.5
4.4
Experimental result Fitted result
4.0 3.6
2.0
signal (a.u.)
signal (a. u.)
In order to explain the differences of the quantum yields of photolyses among the mammalian HbO2, the amino acid sequences, tetramer structures, and the hydrophobicity and flexibility of the different Hb are considered.
1.5
3.2 2.8
1.0 2.4
0.5 2.0
0.0 -0.2
0.0
0.2
0.4
0.6
0.8
t (ms)
1.0
1.2
1.4
1.6
1.8
-0.2 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
t (ms)
Fig. 3. Time-resolved absorption plot obtained in buffer solution at 432 nm: (a) rabbit HbO2; (b) human HbCO.
1.8
2.0
956
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
600
HbCO HbO2
Aτ (mv.μs)
500
b=1.542
HbO2 HbO2
400
HbO2
b=1.332
HbO2 300
b=0.502 200
b=0.424 b=0.386
100
b=0.463 100
150
200
250
300
350
400
450
500
550
E (μJ) Fig. 4. As VS E at optical density 0.17 (b, slope of the line).
Table 1 The ratio of quantum yield (qs/qco), the relaxation time s, and the quantum yields (QY) of different HbO2 Sample
qs/qco
Optical density
s (ls)
QY
Human HbCO Human HbO2 Bovine HbO2 Pig HbO2 Horse HbO2 Rabbit HbO2
1.0 0.29 ± 5% 0.32 ± 5% 0.35 ± 5% 0.38 ± 5% 1.16 ± 5%
0.17 0.17 0.17 0.17 0.17 0.17
120 ± 5% 120 ± 5% 120 ± 5% 150 ± 5% 180 ± 5% 430 ± 5%
0.46 0.13 0.15 0.16 0.17 0.53
1. The amino acid sequences of the Hbs of human, bovine, pig, horse, and rabbit are taken from swissprot (http://us.expasy.org/sprot/), and all of them are aligned using GeneDOC program (Karl Nicholas, Pittaburgh supercomputing center, USA) as shown in Fig. 5 [12]. Fig. 5a and b show the multiple alignments of a globin chains and b globin chains, respectively. By comparison, it shows that the Hbs exhibit above 80% sequence identity to each other (white letters in black shade) and additional several similarities with each other (white letters in gray shade), which indicate that these mammalian Hbs have high homology. However, analyses of amino acid sequences of the five mammal Hbs have revealed 18 variant residues of amino acid residues in the a-chain and 18 in the b-chains. The differences in the primary structure of these globins are located in these well-defined positions: 4, 8, 15, 20, 35, 53, 57, 60, 63, 64, 68, 71, 82, 111, 113, 115, 116, and 131 in a-chain and 2, 5, 9, 12, 21, 50, 51, 56, 69, 70, 73, 76, 87, 116, 117, 121, 125, and 129 in b-chain, in which the difference for rabbit seems more apparent. Most of these variants are located in the surface of the protein and exposed to solvent, so the differences are easily to be displayed.
2. The structures of the five mammalian Hbs are compared using X-ray crystallographic analyses and Insight II (Accelrys, San Diego, USA) shown in Fig. 6. The X-ray structures of human HbO2, bovine HbCO, pig HbO2, and horse HbCO are extracted from PDB (ID: 1gzx [9], 1fsx [13], 1qpw [14], and 1y8i [15]). Because there are no X-ray structure data of bovine and horse HbO2, and the structure of HbCO is similar to that of HbO2 [12], then bovine and horse HbCO are used instead. On the other hand, using the tertiary structure of rabbit Hb modeled by the Swiss-model server [16–18], the tetramer structure of rabbit HbO2 is built also using Insight II by authors. Fig. 6a shows the general tetramer structure of Hb (left) and the superposition of tetramer structures of the five Hbs (right) including human HbO2 (grey), bovine HbCO (red), pig HbO2 (green), horse HbCO (orange), and rabbit HbO2 (cyan), by which the root-mean-square derivations of the tetramers of Hbs of bovine, pig, horse, and rabbit superimposed on that of human Hb for all backbone atoms ˚ , respectively. Fig. 6b shows are 2.6, 2.5, 3.0, and 0.5 A the superposition of a-subunits (left) and b-subunits (right) of the five Hbs, and Fig. 6c shows the active sites of the five Hbs. a-Subunit heme group and environment is shown in the left, and the right shows b-subunit heme group and environment. From the figures, it is clear that the structures of the Hb of the five mammals are quite different although they are similar with each other roughly. Especially, in some parts, such as in Fig. 6c, the differences of the active structures in the b-chain for rabbit are also more apparent. 3. Based on the analyses of the amino acid sequences and structures, there is a lot of differences among the five mammalian Hb. It is well known that the changed residues 4, 8, and 15 in a-chain and 2, 5, 9, and 12 in b-chain
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
957
Fig. 5. The alignments include the sequences of Hb from human, bovine, horse, pig, and rabbit. The number on the right indicates the amino acid position in each sequence. The amino acid identities in the alignment are shown in black shade and the relevant amino acid similarities are indicated by gray shade. (a) Alignment of a chains; (b) alignment of b chains. * Indicates the residue’s number to be multiple of 10.
are located at the A-helix of Hb subunits (see Fig. 6b). It was reported that the mammalian Hb with hydrophilic residues located at the second position of the N-terminal of A-helix of the b-subunit have high intrinsic oxygen affinity [13,19–21]. The variants of hydrophobic residues in this position would lead to the differences of hydrophobic interactions between the N-terminal residues and the hydrophobic core of the b-subunits, which results in the shifting of the N-terminal residues toward the center of the molecular dyad. It will affect the affinity of oxygen. 4. The variant residues a 53, 57, 60, 64, 68, 71 and residues b 56, 69, 70, 73, 76 are located in the outward E-helix of a- and b-chains, respectively. Meanwhile the distal His (a His 58 and b His 63) coordinated heme is also lain in this helix, and the residues in this helix may affect the stability of heme binding site (see Fig. 6c), which make an impact on oxygen association or dissociation [9,13–15].
5. Besides, there are only variant residues b 69 and 70 involved in the surroundings of the heme. The residues in this position are GA (human), DS (bovine), QS (pig), HS (horse), and AA (rabbit). According to the flexibility and hydrophobicity of each peptide predicted using ExPASy server (http://us.expasy.org/cgi-bin/protscale.pl) [22,23], the rabbit Hb has higher hydrophobic characteristics and the lower flexibility than other Hbs in this region, which may affects the surroundings of heme. 6. Finally, the variant fragments of residues a 111–120, i.e., xHxPxxF (N/T) P (A/S) (x: any residues), in the G- and H-helix corners at a1 and b1 (a2–b2) interface, makes contact with variant regions of residues 115–118, i.e., (A/S) xx (F/L), in the b subunit. The hydrogen bond was observed between a Pro114 (C@O) and b Arg 116 (NH1) in bovine Hb, pig Hb, and horse Hb, while no hydrogen bond was observed in this region of human Hb. In rabbit Hb three hydrogen bonds were formed:
958
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
Fig. 6. The superposition of structures of five Hbs: (a) the general tetramer structures of five Hbs (left) and the superposition of tetramer structure of five Hbs (right): human HbO2 (grey), bovine HbCO (red), pig HbO2 (green), horse HbCO (orange), and rabbit HbO2 (cyan); (b) the superposition of a- and b- subunits of the corresponding proteins; (c) the active sites of the Hb structures of five Hbs. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
side-chain of a asn111 (ND2) to b Gly119 (C@O), a asn111 (OD1) to b Lys (NH of backbone), and a Val107(C@O) to b Ser115 (OH). The diversity of regions mentioned above causing different interactions between subunits may impact the Hbs’s allosteric effectors, and lead to difference of oxygen affinity. Thus, the most strong oxygen affinity of human Hb will lead to the lowest quantum yield. While the weakest oxygen affinity of rabbit Hb will lead the highest quantum yield. 7. In addition, several research groups reported the differences of structures and oxygen affinities of Hb among different kinds of species. With a developed thin-layer microcalorimeter method, Johnson et al. studied the
stepwise enthalpies associated with the binding of four O2, which show that the enthalpy change of photo-dissociation of Hb(O2)4 for human and for bovine HbO2 are not the same [24]. We also have successfully got the enthalpy and conformational volume changes related to the tertiary relaxation in photo-dissociation of five mammalian HbO2 investigated by photoacoustic calorimetry [25]. On the other hand, Liang et al. studied the structure of Hb of bar-headed goose, and found that the quaternary structure of deoxy-Hb of the goose shows obvious differences from that of human, and compared to its closely related lowland species, greylag goose, the stripped Hb from the bar-headed goose
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
shows a slightly higher oxygen affinity [26]. The varieties of the results prove the existence of the differences among different HbO2.
Summary In conclusion, the photo-dissociations of HbO2 of five mammals (human, bovine, pig, horse, and rabbit) have been studied successfully by using the nanosecond laser pump-probe technique, by which the different quantum yields of photo-dissociation of five mammalian HbO2 have been obtained, especially a big difference in the photo-dissociation of rabbit HbO2 has been observed. Based on the analyses of the amino acid sequences and the structures of the Hb, it is shown that, although these Hbs all belong to the same protein family and have high positives and identities, the differences among them are noticeable, especially for that of rabbit. The differences of the amino acid sequences and the structures probably are the reasons of the disparities of quantum yield related to the photo-dissociation of O2 from HbO2 for the five kinds of mammals. However, up to date, it cannot be identified that the quantum yields are mainly determined by what substructures of the Hb, so it is necessary to study the photo-dissociation reaction of liganded Hb in more details. Acknowledgment This work is supported by National Natural Science Foundation of China, No. 10574073. References [1] S.M. Hanash, G.J. Brew, Advance in Hemoglobin Analysis, Alan, R. Liss, Inc., NY, 1981, pp. 10–54. [2] R.W. Noble, M. Brunori, J. Wyman, E. Antonini, Studies on the quantumn yields of the photodissociation of carbon monoxide from hemoglobin and myoblobin, Bichemistry 6 (1967) 1216– 1222. [3] C.L. Norris, K.S. Peters, A photoacoustic calorimetry study of horse carboxymyoglobin on the 10-nanosecond time scale, Biophys. J. 65 (1993) 1660–1665. [4] R.M. Esquerra, R.A. Goldbeck, S.H. Reaney, A.M. Batchelder, Multiple geminate ligand recombination in human hemoglobin, Biophys. J. 78 (2000) 3227–3239. [5] W.A. Saffran, Q.H. Gibson, Photodiossociation of Ligands from Heme and Heme Proteins, J. Biol. Chem. 252 (1977) 7955–7958. [6] IUPAC Compendium of Chemical Terminology 2nd Edition. (1997).
959
[7] E. Ghelichkhani, R.A. Goldbevk, J.W. Lewis, D.S. Kliger, Nanosecond time-resolved absorption studies of human oxyhemoglobin photolysis intermediates, Biophys. J. 71 (1996) 1596–1604. [8] D.A. Chernoff, R.M. Hochsterasser, A.W. Steele, Germinate recombination of O2 and hemoglobin, Proc. Natl. Acad. Sci. USA 77 (1980) 5606–5610. [9] M. Paoli, R. Liddington, J. Tame, A. Wilkinson, G. Dodson, Crystal structure of T state haemoglobin with oxygen bound at all four haems, J. Mol. Biol. 256 (1996) 775–792. [10] Q.H. Gibson, J.S. Olson, R.E. McKinnie, R.J. Rohlfs, A kinetic description of ligand binding to sperm whale myoglobin, J. Biol. Chem. 261 (1986) 10228–10236. [11] D.A. Duddell, R.J. Morris, J.T. Richards, Nanosecond laser photolysis of aqueous carbon monoxy- and oxyhaemoglobin, Biochim. Biochim. Acta 621 (1980) 1–8. [12] K.B. Nicholas, H.B. Nicholas Jr., D.W. Deerfield, GeneDoc: analysis and visualization of genetic variation, embnew, News 4 (1997) 14. [13] M.K. Safo, D.J. Abraham, The X-ray structure determination of ˚ resolution and its bovine carbonmonoxy hemoglobin at 2.1 A relationship to the quaternary structures of other hemoglobin crystal forms, Protein Sci. 10 (2001) 1091–1099. [14] T.H. Lu, K. Panneerselvam, Y.C. Liaw, P. Kan, C.J. Lee, Structure determination of pig haemoglobin, Acta Crystallogr. D 56 (2000) 403. [15] N. Shibayama, S. Miura, J.R.H. Tame, T. Yonetani, S.Y. Park, Crystal structure of horse carbonmonoxyhemoglobin-bezafibrate complex at ˚ resolution, J. Biol. Chem. 277 (41) (2002) 38791–38796. 1.55-A [16] T. Schwede, J. Kopp, N. Guex, M.C. Peitsch, SWISS-MODEL: an automated protein homology-modeling server, Nucleic Acids Res. 31 (2003) 3381–3385. [17] N. Guex, M.C. Peitsch, SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling, Electrophoresis 18 (1997) 2714–2723. [18] M.C. Peitsch, Protein modeling by E-mail, Biotechnology 13 (1995) 658–660. [19] M.F. Perutz, G. Fermi, A novel allosteric mechanism in Hb, J. Mol. Biol. 233 (1993) 536–545. [20] M.F. Perutz, K. Imai, Regulation of oxygen affinity of mammalian Hbs, J. Mol. Biol. 136 (1980) 183–191. [21] M.F. Perutz, G. Fermi, B. Luisi, B. Shaanan, R.C. Liddington, Stereochemistry of cooperative mechanisms in Hb, Acc. Chem. Res. 20 (1987) 309–321. [22] R.M. Sweet, D. Eisenberg, Correlation of sequence hydrophobicities measures similarity in three-dimensional protein structure, J. Mol. Biol. 171 (1983) 479–488. [23] R. Bhaskaran, P.K. Ponnuswamy, Amino acid scale: average flexibility index, Int. J. Pept. Protein Res. 32 (1988) 242–255. [24] C.R. Johnson, D.W. Ownby, S.J. Gill, K.S. Peters, Oxygen binding constants and stepwise enthalpies for human and bovine hemoglobin at pH 7.6, Biochemistry 31 (1992) 10074–10082. [25] N.L. Yang, S.Y. Zhang, L. Sun, M. Qu, X.J. Shui, H.L. Chen, Enthalpy and conformational volume changes of mammalian oxyhemoglobins investigated by pulsed photoacoustic calorimetry, Presented at World Congress on Ultrasonics–Ultrasonics International 2005 (Beijing, China). [26] Y. Liang, Z. Hua, X. Liang, Q. Xu, G. Lu, The crystal structure of bar-headed goose hemoglobin in deoxy form: the allosteric mechanism of a hemoglobin species with high oxygen affinity, J. Mol. Biol. 313 (2001) 123–137.
Photo-dissociation quantum yields of mammalian oxyhemoglobin investigated by a nanosecond laser technique Ning-li Yang a, Shu-yi Zhang a,*, Pao-kuang Kuo a, Min Qu a, Jian-wen Fang b, Jia-huang Li c, Zi-chun Hua c a
Laboratory of Modern Acoustics, Institute of Acoustics, Nanjing University, Nanjing 210093, China b Department of Physics, Zhejiang Normal University, Jinhua 321004, China c Laboratory of Pharmaceutical Biotechnology, Nanjing University, Nanjing 210093, China Received 12 December 2006 Available online 22 December 2006
Abstract The photo-dissociations of oxyhemoglobin of several mammals, such as human, bovine, pig, horse, and rabbit, have been studied. By means of optical pump-probe technique, the quantum yields for photo-dissociation of these oxyhemoglobin have been determined at pH 7 and 20 °C. A nanosecond laser at 532 nm is used as the pumping source, and a xenon lamp through a monochrometer provides a probe light at 432 nm. The experimental results show that the quantum yields of these mammalian oxyhemoglobin are different from each other, especially for that of rabbit. By analyzing the amino acid sequences and tetramer structures as well as the flexibility and hydrophobicity of the different hemoglobin, possible explanations for the differences are proposed. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Oxyhemoglobin; Photo-dissociation; Quantum yield
Hemoglobin (Hb) is an oxygen carrier in blood. The Hb tetramer consists of two a and two b subunits, where the bchain contains eight helices (A–H) while the a-chain contains the same helices except D helix [1]. Each subunit binds a heme, which can bind a dioxygen. The heme is located at the cranny of the E-helix and F-helix and is exposed to solvent. The interactions between different subunits (a to b chain) are strong, while interactions between the same subunits (a to a chain or b to b chain) are weak. The dissociation of ligands from Hb by light has been studied for many years. But the most studied liganded Hbs are the carbon-monoxide hemoglobin (HbCO) and carbonmonoxide myoglobin (MbCO) [2–4], but a few studies are concerning HbO2. The choice of CO as a ligand in these studies probably comes from the fact that the photo-dissociation quantum yield of HbCO is much higher than that
*
Corresponding author. Fax: +86 25 83313690. E-mail address: [email protected] (S. Zhang).
0006-291X/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.12.118
of oxyhemoglobin (HbO2), and consequently it is more difficult to work with the oxygen derivative of Hb [5]. The quantum yield of photo-dissociation is the efficiency of the transfer from the reactants to the products generated by light irradiation, which is defined as the molecular number of photoproduct species divided by the absorbed photon number [6]. Because the quantum yield is a very sensitive parameter, which may easily be affected by the experimental conditions including temperature, pressure, density and pH state (if in the aqueous state), etc., of the solutions, it would be better to determine the quantum yield value under a specific kind of experiments. The quantum yields of the photo-dissociation of oxygen from HbO2, which are quite small for most of mammals, just have been reported rarely although the mechanisms of the association and dissociation of Hb and ligand molecules have been frequently researched [7], but a few studies are concerning HbO2 [8,9]. In this paper, an optical pump-probe method [10] is used to study the photo-dissociation of several kinds of
954
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
mammalian HbO2 solutions, including human, bovine, porcine, equine, and rabbit HbO2 solutions. The different quantum yields of the photo-dissociation in nanosecond range for these kinds of mammals are obtained, especially much difference for that of rabbit compared with others. The differences of the quantum yields among the five mammals are also discussed by analyzing the discrepancies of the amino acid sequences and tetramer structures although the structures of the Hb of the mammals are very similar with each other. Materials and methods Materials. The Hbs of human, pig, horse, rabbit (Sigma, USA), and bovine (Mingzhu, Dongfang Biological Technology, China) are used as experimental samples. They are all in the form of lyophilized powders and have the same water content, iron percentage, rate of purity, and solubility. The HbO2 solutions are prepared by conventional process as 4.0 3.5
Hb HbO2
3.0
OD
2.5 2.0 1.5 1.0 0.5 0.0 -0.5 300
400
500
600
700
Wavelength (nm) Fig. 1. The absorption spectra of Hb and HbO2 in buffer at pH 7.0.
follows: (1) the Hb powder is dialyzed against 50 mM phosphate buffer at pH 7.0; (2) the solution is reduced with sodium dithionate (Na2S2O4); (3) HbO2 is formed by bubbling O2 through the solutions at 1 atm pressure. Before the experiments, in order to check the oxy- and deoxy-forms of Hbs in the solutions, the absorption spectra are measured with UV-3100 Spectrophotometer, and the optical densities of all the solutions under the light illustration at wavelength 532 nm are determined, as shown in Fig. 1. It is found that the absorption spectra of all these five mammalian Hb and oxy-Hb are almost the same. For comparison, HbCO is also formed by bubbling CO through the solution, which is also prepared with the same conditions. Apparatus and method. The apparatus used to measure the quantum yields is shown in Fig. 2. The sample of HbO2 is put in a quartz cuvette with cross-sectional area of 10 mm2. A Q-switched Nd:YAG laser operating at 532 nm, 10 Hz and pulse width 8 ns is used as the excitation source. The pulsed-laser beam is divided into two beams with a 50/50 beam splitter. One of the divided beams is the pump beam. It is focused by a cylindrical lens onto the sample cuvette. The other is directed into a monitoring pulse ergometer. Meanwhile, a continuous light beam from a xenon lamp through a monochrometer is directed through the cuvette perpendicular with the pump beam and is matched in width. Based on the optical spectra of the oxy- and deoxy-Hb samples shown in Fig. 1, the optical wavelength of the probe beam is chosen at 432 nm, where the difference of optical densities between oxy- and deoxy-Hb is the largest and also sufficiently away from the wavelength of the pumping beam, so the sensitivity for measuring the photo-dissociation will be the highest. In order to suppress the interference of the stray reflections of the pumping beam, the probe beam is then made to pass through multiple 432 nm bandpass interference filters and a second monochrometer and then to a photomultiplier. A digital oscilloscope records the counting rates for further processing and analyzing by a computer. For exploring the photo-dissociation reactions of HbO2 in the solution, the variation of optical density at 432 nm during and after the pumping laser irradiation is detected, which means the variation of optical density is induced by the photolysis reaction of the HbO2. In order to eliminate the error produced by the fluctuation of the pumping laser intensity and the noise of the system, the detected signal is averaged by 300 pulses to improve the signal to noise ratio. All the measurements are taken from the same Hb concentrations, i.e., at the optical density 0.17, of HbO2 solutions for human, bovine, pig, horse, and rabbit, and at 20 °C. Quantum yield calculation. The photo-dissociation reactions involved in the HbO2 solutions may be expressed as
Fig. 2. Experimental system.
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959 k
FeO2 0 Fe þ O2 k
where k and k 0 are the reaction rates of the dissociation and rebind processes, respectively. A first order exponential decay expression A exp (t/s) is used to mathematically fit the rebinding process, where A is the peak value of the signal represented photo-dissociation process and s is the ligand rebinding relaxation time. The quantum yield q is defined as the molecular number of photoproducts divided by the absorbed photon number, then Z K 1 t=s A e dt; ð1Þ q¼ E 0 where E is the energy of the beam irradiating the sample, and K is the constant of the experimental system. Therefore it is clear that q¼K
As : E
ð2Þ
As the pumping pulse energy is in the range of milli-Joule range, the photo-dissociation process is in linear with the energy variation. Changing the pumping pulse energy by means of an optical attenuator and recording the energy values by the pulse ergometer, a set of exponential decay curves can be obtained, and then a set of As for different pumping energies can be fitted, by which the q/K can be obtained from the slope of the plot of As against E. In order to eliminate the effect of the system constant K, the quantum yields of photo-dissociation HbO2 of human, bovine, pig, horse, and rabbit are obtained by comparing with the quantum yield for photo-dissociation HbCO of human, which is similar to the experiments of Saffran et al. [5] and Duddell et al. [11]. Keeping all the conditions including temperature, pressure, pH state, and then the constant K invariable during the experiments, the ratio of the quantum yields of HbO2 of human, bovine, pig, horse, and rabbit to that of human HbCO can be obtained. As the quantum yields of HbCO are determined, the HbO2 of the five mammals can also be determined at all.
955
exponential decay expression A exp (t/s) is used to mathematically fit the rebinding process, where A is the peak value of the signal represented photo-dissociation process and s is the ligand rebinding relaxation time. The experimental measured plots of As VS E for human, bovine, pig, horse, and rabbit HbO2 and human HbCO at the optical density 0.17 under the wavelength 532 nm are illustrated in Fig. 4, which shows the amount of photoproducts As of the five mammals HbO2 and human HbCO under an exposure of pulses with different energies E. All the measurements are taken at 20 °C. The experimental points for each HbO2 lie on a straight line whose slope is proportional to the quantum yield. All of the measured results of photolyses of HbO2 solutions for human, bovine, pig, horse, and rabbit, as well as HbCO of human obtained under the same conditions are listed in Table 1, in which qs/qco is the relative quantum yields, qs represents the quantum yields of HbO2 of human, bovine, pig, horse, and rabbit, and qco represents that of human HbCO. The absolute values of the quantum yields (QY) of different mammals can be obtained if the quantum yield of the HbCO of human is determined. Based on the results of Saffran et al., the quantum yields of human HbCO are taken as 0.46 at 20 °C [5], then the quantum yields of human, bovine, pig, horse, and rabbit are also given in Table 1. It can be seen from the data of Table 1 that the photo-dissociation quantum yields of these mammalian oxy-Hbs are different from each other, especially so for that of rabbit.
Results Discussion When the laser pulse is irradiating on the HbO2 solutions, the ligand is dissociated from the ligand-heme, but the ligand then rebinds to the heme in the dark duration. As an example, the measured decay curves of the rabbit HbO2 and human HbCO are shown in Fig. 3. A first order
a
b
3.0
Experimental result Fitted result
2.5
4.4
Experimental result Fitted result
4.0 3.6
2.0
signal (a.u.)
signal (a. u.)
In order to explain the differences of the quantum yields of photolyses among the mammalian HbO2, the amino acid sequences, tetramer structures, and the hydrophobicity and flexibility of the different Hb are considered.
1.5
3.2 2.8
1.0 2.4
0.5 2.0
0.0 -0.2
0.0
0.2
0.4
0.6
0.8
t (ms)
1.0
1.2
1.4
1.6
1.8
-0.2 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
t (ms)
Fig. 3. Time-resolved absorption plot obtained in buffer solution at 432 nm: (a) rabbit HbO2; (b) human HbCO.
1.8
2.0
956
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
600
HbCO HbO2
Aτ (mv.μs)
500
b=1.542
HbO2 HbO2
400
HbO2
b=1.332
HbO2 300
b=0.502 200
b=0.424 b=0.386
100
b=0.463 100
150
200
250
300
350
400
450
500
550
E (μJ) Fig. 4. As VS E at optical density 0.17 (b, slope of the line).
Table 1 The ratio of quantum yield (qs/qco), the relaxation time s, and the quantum yields (QY) of different HbO2 Sample
qs/qco
Optical density
s (ls)
QY
Human HbCO Human HbO2 Bovine HbO2 Pig HbO2 Horse HbO2 Rabbit HbO2
1.0 0.29 ± 5% 0.32 ± 5% 0.35 ± 5% 0.38 ± 5% 1.16 ± 5%
0.17 0.17 0.17 0.17 0.17 0.17
120 ± 5% 120 ± 5% 120 ± 5% 150 ± 5% 180 ± 5% 430 ± 5%
0.46 0.13 0.15 0.16 0.17 0.53
1. The amino acid sequences of the Hbs of human, bovine, pig, horse, and rabbit are taken from swissprot (http://us.expasy.org/sprot/), and all of them are aligned using GeneDOC program (Karl Nicholas, Pittaburgh supercomputing center, USA) as shown in Fig. 5 [12]. Fig. 5a and b show the multiple alignments of a globin chains and b globin chains, respectively. By comparison, it shows that the Hbs exhibit above 80% sequence identity to each other (white letters in black shade) and additional several similarities with each other (white letters in gray shade), which indicate that these mammalian Hbs have high homology. However, analyses of amino acid sequences of the five mammal Hbs have revealed 18 variant residues of amino acid residues in the a-chain and 18 in the b-chains. The differences in the primary structure of these globins are located in these well-defined positions: 4, 8, 15, 20, 35, 53, 57, 60, 63, 64, 68, 71, 82, 111, 113, 115, 116, and 131 in a-chain and 2, 5, 9, 12, 21, 50, 51, 56, 69, 70, 73, 76, 87, 116, 117, 121, 125, and 129 in b-chain, in which the difference for rabbit seems more apparent. Most of these variants are located in the surface of the protein and exposed to solvent, so the differences are easily to be displayed.
2. The structures of the five mammalian Hbs are compared using X-ray crystallographic analyses and Insight II (Accelrys, San Diego, USA) shown in Fig. 6. The X-ray structures of human HbO2, bovine HbCO, pig HbO2, and horse HbCO are extracted from PDB (ID: 1gzx [9], 1fsx [13], 1qpw [14], and 1y8i [15]). Because there are no X-ray structure data of bovine and horse HbO2, and the structure of HbCO is similar to that of HbO2 [12], then bovine and horse HbCO are used instead. On the other hand, using the tertiary structure of rabbit Hb modeled by the Swiss-model server [16–18], the tetramer structure of rabbit HbO2 is built also using Insight II by authors. Fig. 6a shows the general tetramer structure of Hb (left) and the superposition of tetramer structures of the five Hbs (right) including human HbO2 (grey), bovine HbCO (red), pig HbO2 (green), horse HbCO (orange), and rabbit HbO2 (cyan), by which the root-mean-square derivations of the tetramers of Hbs of bovine, pig, horse, and rabbit superimposed on that of human Hb for all backbone atoms ˚ , respectively. Fig. 6b shows are 2.6, 2.5, 3.0, and 0.5 A the superposition of a-subunits (left) and b-subunits (right) of the five Hbs, and Fig. 6c shows the active sites of the five Hbs. a-Subunit heme group and environment is shown in the left, and the right shows b-subunit heme group and environment. From the figures, it is clear that the structures of the Hb of the five mammals are quite different although they are similar with each other roughly. Especially, in some parts, such as in Fig. 6c, the differences of the active structures in the b-chain for rabbit are also more apparent. 3. Based on the analyses of the amino acid sequences and structures, there is a lot of differences among the five mammalian Hb. It is well known that the changed residues 4, 8, and 15 in a-chain and 2, 5, 9, and 12 in b-chain
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
957
Fig. 5. The alignments include the sequences of Hb from human, bovine, horse, pig, and rabbit. The number on the right indicates the amino acid position in each sequence. The amino acid identities in the alignment are shown in black shade and the relevant amino acid similarities are indicated by gray shade. (a) Alignment of a chains; (b) alignment of b chains. * Indicates the residue’s number to be multiple of 10.
are located at the A-helix of Hb subunits (see Fig. 6b). It was reported that the mammalian Hb with hydrophilic residues located at the second position of the N-terminal of A-helix of the b-subunit have high intrinsic oxygen affinity [13,19–21]. The variants of hydrophobic residues in this position would lead to the differences of hydrophobic interactions between the N-terminal residues and the hydrophobic core of the b-subunits, which results in the shifting of the N-terminal residues toward the center of the molecular dyad. It will affect the affinity of oxygen. 4. The variant residues a 53, 57, 60, 64, 68, 71 and residues b 56, 69, 70, 73, 76 are located in the outward E-helix of a- and b-chains, respectively. Meanwhile the distal His (a His 58 and b His 63) coordinated heme is also lain in this helix, and the residues in this helix may affect the stability of heme binding site (see Fig. 6c), which make an impact on oxygen association or dissociation [9,13–15].
5. Besides, there are only variant residues b 69 and 70 involved in the surroundings of the heme. The residues in this position are GA (human), DS (bovine), QS (pig), HS (horse), and AA (rabbit). According to the flexibility and hydrophobicity of each peptide predicted using ExPASy server (http://us.expasy.org/cgi-bin/protscale.pl) [22,23], the rabbit Hb has higher hydrophobic characteristics and the lower flexibility than other Hbs in this region, which may affects the surroundings of heme. 6. Finally, the variant fragments of residues a 111–120, i.e., xHxPxxF (N/T) P (A/S) (x: any residues), in the G- and H-helix corners at a1 and b1 (a2–b2) interface, makes contact with variant regions of residues 115–118, i.e., (A/S) xx (F/L), in the b subunit. The hydrogen bond was observed between a Pro114 (C@O) and b Arg 116 (NH1) in bovine Hb, pig Hb, and horse Hb, while no hydrogen bond was observed in this region of human Hb. In rabbit Hb three hydrogen bonds were formed:
958
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
Fig. 6. The superposition of structures of five Hbs: (a) the general tetramer structures of five Hbs (left) and the superposition of tetramer structure of five Hbs (right): human HbO2 (grey), bovine HbCO (red), pig HbO2 (green), horse HbCO (orange), and rabbit HbO2 (cyan); (b) the superposition of a- and b- subunits of the corresponding proteins; (c) the active sites of the Hb structures of five Hbs. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
side-chain of a asn111 (ND2) to b Gly119 (C@O), a asn111 (OD1) to b Lys (NH of backbone), and a Val107(C@O) to b Ser115 (OH). The diversity of regions mentioned above causing different interactions between subunits may impact the Hbs’s allosteric effectors, and lead to difference of oxygen affinity. Thus, the most strong oxygen affinity of human Hb will lead to the lowest quantum yield. While the weakest oxygen affinity of rabbit Hb will lead the highest quantum yield. 7. In addition, several research groups reported the differences of structures and oxygen affinities of Hb among different kinds of species. With a developed thin-layer microcalorimeter method, Johnson et al. studied the
stepwise enthalpies associated with the binding of four O2, which show that the enthalpy change of photo-dissociation of Hb(O2)4 for human and for bovine HbO2 are not the same [24]. We also have successfully got the enthalpy and conformational volume changes related to the tertiary relaxation in photo-dissociation of five mammalian HbO2 investigated by photoacoustic calorimetry [25]. On the other hand, Liang et al. studied the structure of Hb of bar-headed goose, and found that the quaternary structure of deoxy-Hb of the goose shows obvious differences from that of human, and compared to its closely related lowland species, greylag goose, the stripped Hb from the bar-headed goose
N. Yang et al. / Biochemical and Biophysical Research Communications 353 (2007) 953–959
shows a slightly higher oxygen affinity [26]. The varieties of the results prove the existence of the differences among different HbO2.
Summary In conclusion, the photo-dissociations of HbO2 of five mammals (human, bovine, pig, horse, and rabbit) have been studied successfully by using the nanosecond laser pump-probe technique, by which the different quantum yields of photo-dissociation of five mammalian HbO2 have been obtained, especially a big difference in the photo-dissociation of rabbit HbO2 has been observed. Based on the analyses of the amino acid sequences and the structures of the Hb, it is shown that, although these Hbs all belong to the same protein family and have high positives and identities, the differences among them are noticeable, especially for that of rabbit. The differences of the amino acid sequences and the structures probably are the reasons of the disparities of quantum yield related to the photo-dissociation of O2 from HbO2 for the five kinds of mammals. However, up to date, it cannot be identified that the quantum yields are mainly determined by what substructures of the Hb, so it is necessary to study the photo-dissociation reaction of liganded Hb in more details. Acknowledgment This work is supported by National Natural Science Foundation of China, No. 10574073. References [1] S.M. Hanash, G.J. Brew, Advance in Hemoglobin Analysis, Alan, R. Liss, Inc., NY, 1981, pp. 10–54. [2] R.W. Noble, M. Brunori, J. Wyman, E. Antonini, Studies on the quantumn yields of the photodissociation of carbon monoxide from hemoglobin and myoblobin, Bichemistry 6 (1967) 1216– 1222. [3] C.L. Norris, K.S. Peters, A photoacoustic calorimetry study of horse carboxymyoglobin on the 10-nanosecond time scale, Biophys. J. 65 (1993) 1660–1665. [4] R.M. Esquerra, R.A. Goldbeck, S.H. Reaney, A.M. Batchelder, Multiple geminate ligand recombination in human hemoglobin, Biophys. J. 78 (2000) 3227–3239. [5] W.A. Saffran, Q.H. Gibson, Photodiossociation of Ligands from Heme and Heme Proteins, J. Biol. Chem. 252 (1977) 7955–7958. [6] IUPAC Compendium of Chemical Terminology 2nd Edition. (1997).
959
[7] E. Ghelichkhani, R.A. Goldbevk, J.W. Lewis, D.S. Kliger, Nanosecond time-resolved absorption studies of human oxyhemoglobin photolysis intermediates, Biophys. J. 71 (1996) 1596–1604. [8] D.A. Chernoff, R.M. Hochsterasser, A.W. Steele, Germinate recombination of O2 and hemoglobin, Proc. Natl. Acad. Sci. USA 77 (1980) 5606–5610. [9] M. Paoli, R. Liddington, J. Tame, A. Wilkinson, G. Dodson, Crystal structure of T state haemoglobin with oxygen bound at all four haems, J. Mol. Biol. 256 (1996) 775–792. [10] Q.H. Gibson, J.S. Olson, R.E. McKinnie, R.J. Rohlfs, A kinetic description of ligand binding to sperm whale myoglobin, J. Biol. Chem. 261 (1986) 10228–10236. [11] D.A. Duddell, R.J. Morris, J.T. Richards, Nanosecond laser photolysis of aqueous carbon monoxy- and oxyhaemoglobin, Biochim. Biochim. Acta 621 (1980) 1–8. [12] K.B. Nicholas, H.B. Nicholas Jr., D.W. Deerfield, GeneDoc: analysis and visualization of genetic variation, embnew, News 4 (1997) 14. [13] M.K. Safo, D.J. Abraham, The X-ray structure determination of ˚ resolution and its bovine carbonmonoxy hemoglobin at 2.1 A relationship to the quaternary structures of other hemoglobin crystal forms, Protein Sci. 10 (2001) 1091–1099. [14] T.H. Lu, K. Panneerselvam, Y.C. Liaw, P. Kan, C.J. Lee, Structure determination of pig haemoglobin, Acta Crystallogr. D 56 (2000) 403. [15] N. Shibayama, S. Miura, J.R.H. Tame, T. Yonetani, S.Y. Park, Crystal structure of horse carbonmonoxyhemoglobin-bezafibrate complex at ˚ resolution, J. Biol. Chem. 277 (41) (2002) 38791–38796. 1.55-A [16] T. Schwede, J. Kopp, N. Guex, M.C. Peitsch, SWISS-MODEL: an automated protein homology-modeling server, Nucleic Acids Res. 31 (2003) 3381–3385. [17] N. Guex, M.C. Peitsch, SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling, Electrophoresis 18 (1997) 2714–2723. [18] M.C. Peitsch, Protein modeling by E-mail, Biotechnology 13 (1995) 658–660. [19] M.F. Perutz, G. Fermi, A novel allosteric mechanism in Hb, J. Mol. Biol. 233 (1993) 536–545. [20] M.F. Perutz, K. Imai, Regulation of oxygen affinity of mammalian Hbs, J. Mol. Biol. 136 (1980) 183–191. [21] M.F. Perutz, G. Fermi, B. Luisi, B. Shaanan, R.C. Liddington, Stereochemistry of cooperative mechanisms in Hb, Acc. Chem. Res. 20 (1987) 309–321. [22] R.M. Sweet, D. Eisenberg, Correlation of sequence hydrophobicities measures similarity in three-dimensional protein structure, J. Mol. Biol. 171 (1983) 479–488. [23] R. Bhaskaran, P.K. Ponnuswamy, Amino acid scale: average flexibility index, Int. J. Pept. Protein Res. 32 (1988) 242–255. [24] C.R. Johnson, D.W. Ownby, S.J. Gill, K.S. Peters, Oxygen binding constants and stepwise enthalpies for human and bovine hemoglobin at pH 7.6, Biochemistry 31 (1992) 10074–10082. [25] N.L. Yang, S.Y. Zhang, L. Sun, M. Qu, X.J. Shui, H.L. Chen, Enthalpy and conformational volume changes of mammalian oxyhemoglobins investigated by pulsed photoacoustic calorimetry, Presented at World Congress on Ultrasonics–Ultrasonics International 2005 (Beijing, China). [26] Y. Liang, Z. Hua, X. Liang, Q. Xu, G. Lu, The crystal structure of bar-headed goose hemoglobin in deoxy form: the allosteric mechanism of a hemoglobin species with high oxygen affinity, J. Mol. Biol. 313 (2001) 123–137.