- Email: [email protected]
Oxygen affinities (p50) of myoglobins from four vertebrate species (Canis familiaris, rattus norvegicus, mus musculus and Gallus domesticus) as determined by a kinetic and an equilibrium method
~
Comp. Biochem. Physiol. Vol. I lOB,No. 1, pp. 193-199, 1995 ElsevierScienceLtd Britain 0305-0491/95$9.50+ 0.00
Pergamon
PrintedinGreat
0305-0491(94)00119-7
Oxygen affinities (Ps0) of myoglobins from four vertebrate species (Canis familiaris, Rattus norvegicus, Mus musculus and Gallus domesticus) as determined by a kinetic and an equilibrium method Yasunori Enoki, Kazuhito Matsumura, Yoshimi Ohga, Hisaharu Kohzuki and Masatoshi Hattori Second Department of Physiology, Nara Medical University, Kashihara, Nara 634, Japan Rate constants of the reaction with oxygen of myoglobins from four vertebrate species (Canis familiaris, Rattus norvegicus, Mus musculus and Gallus domesticus) and the isolated ~A and /~A chains of human adult hemoglobin (HbA) were determined by the stopped-flow-spectrophotomeric method. Half-saturation oxygen pressure (Ps0) of the proteins calculated from the rate constants, assuming a simple bimolecular reaction model, agreed very well with those directly determined by oxygen equilibria. The proteins used were freshly prepared, and fully characterized by electrophoretic and ultracentrifugai analyses. Sulphydryl groups in the Hb chains were ascertained to be completely regenerated. Key words: Myoglobins; Hemoglobin chains; Vertebrate; Oxygen affinities (Ps0); Oxygen equilibrium; Stopped-flow spectrophotometry; Oxygen binding rates; Oxygen dissociation rates.
Comp. Biochem. Physiol. IIOB, 193-199, 1995.
Introduction The physiological function of myoglobin, curate determination. Furthermore, a relawhich, in contrast to hemoglobin, exhibits tively large amount of sample is needed with neither homotropic nor heterotropic allos- a rather long measurement time. Although teric characteristics, can be described solely some recently developed instruments, using by its oxygen affinity. The oxygen affinity of a thin layer technique, achieve an oxygen the protein has usually been determined by an equilibrium curve with as little as 5 g 1sample oxygen equilibrium approach, which poses in tens of minutes or less, a favourable choice several difficulties and disadvantages. For is the kinetic approach. The kinetic method, example, considerable metmyoglobin forma- which is applicable to non-cooperative protion during the measurement, due to a high teins like myoglobins and also isolated subtendency to autoxidation, might prevent ac- units of hemoglobins, require much less time with a small amount of sample and allows the determination of oxygen affinity with less metmyoglobin formation. Correspondence to: Y. Enoki, Second Department of So far several kinetic studies on the Physiology, Nara Medical University, Kashihara, reaction of oxygen with myoglobins and Nara 634, Japan. Received 11 January 1994; accepted 10 June 1994. isolated hemoglobin chains have been 193
194
Y. Enoki et al.
reported, in which equilibrium constants calculated from kinetic constants are compared with those from the oxygen equilibrium measurements (Antonini, 1965; Wittenberg et al., 1965; Brunori et al., 1966; Brunori and Schuster, 1969; Noble et al., 1969). Further research, all studying noncooperative invertebrate hemoglobins, is found in recent literature (Amiconi et al., 1972; Dilorio et al., 1986; Smit et al., 1986; Kraus and Wittenberg, 1990). Most of the results, however, show discrepancies between both lines of determination, which have not yet been satisfactorily explained. Hypotheses, albeit not proved yet, were proposed that the departure from the simple reaction model might be due to the formation of short-lived intermediates or conformational changes following the simple bimolecular reaction of the proteins and ligands (Antonini, 1965; Brunori et al., 1966). Another reason for the discrepancy, if only partial, might be contamination and denaturation of the samples, where commercially available material was used after insufficient purification. It should be also stressed that the kinetic measurement has its own merit. The similar oxygen affinities could be achieved by very different balances of the combination and dissociation rates, as actually observed in non-cooperative hemoglobins isolated from the bivalve mollusc, Lucina pectinata (Kraus and Wittenberg, 1990). The kinetic information is extremely valuable in such a case and irreplaceable by the equilibrium measurement. In this paper, the kinetic constants for the reaction with oxygen of freshly prepared canine, rat, murine and chicken myoglobins and also the isolated chains (ctA and fl A) from human adult hemoglobin are reported, and the oxygen affinities (Ps0) are calculated. We also examine whether the Ps0 thus derived agree with those separately determined by the direct equilibrium measurements or not. Agreement between both lines of the determination is excellent.
reported (Enoki et al., 1984). Muscles (usually 100-200 g) excised in small pieces were extracted with three-fold volumes of chilled water by using a juice mixer, and centrifuged to obtain a clear supernate (18,000 rpm x 20 min, 0°C). The extract was heated for 10min at 50°C in a thermostatted bath under continuous shaking. The denatured material thus developed was spun off by refrigerated centrifugation after CO flushing. This heat denaturation step (Bfinning and Hamm, 1969), instead of the commonly used salting-out procedure, saved much labour and time in effectively removing hemoglobin, a troublesome contaminant, in the extract. The extract was concentrated by an ultrafiltration apparatus (Toyo model UHP 90), loaded onto a 3 x 9 5 c m column of Sephadex G-75 (Pharmacia) equilibrated with 0.05 M Tris-HC1-0.05M NaC1-0.001 M EDTA (pH 7.5), eluted with the same buffer and the absorbance of the effluent was monitored at 540 and 280 nm. The resultant myoglobin fractions were pooled, concentrated and dialysed overnight against the equilibrating buffer (pH 8.3) for chromatofocusing. The dialysed material, after refrigerated centrifugation, was applied onto a PBE 94 (Pharmacia) column equilibrated with CO:-free 0.025 M Tris-acetate buffer (pH 8.3) and eluted with PB 96 buffer (Pharmacia), 13-fold diluted and pH adjusted to 6.0 with acetic acid. An Amicon CFI0 x 30 column with a 20 ml gel bed was used for up to 100mg myoglobin. The elution was monitored at 540 and 280 nm and myoglobin fractions, as judged by visible absorption spectrum, were pooled, and dialysed overnight against deionized water. PB 96 buffer in the sample was then removed by hydrophobic interaction chromatography with Sepharose CL-4B (Pharmacia) according to the manufacturer's instruction. After dialysis against deionized water and concentration, myoglobin solution was stored in a fine droplet form at -135°C. Reduction of ferric myoglobin, if necessary, was performed with dithionite in a mixed bed BioMaterials and Methods Rad AG501-X8 (D) column (Bauer and Preparation o f myoglobins Pacyna, 1975). All the chromatographic Myoglobins were purified by a heat procedures were conducted in the cold denaturation-gel filtration procedure com- (4°C) and the buffers were chilled and bined with chromatofocusing as previously saturated with CO.
Kinetically determined Pso of myoglobins
Preparation of adult human hemoglobin chains
195
dithionite had been loaded immediately before applying the sample. During the course of elution, reduction of the ferric forms, if any, as well as the deoxygenation was accomplished. For oxygen dissociation, the oxygenated proteins were rapidly mixed with 0.2 percent dithionite in the N2-saturated oxygen-free buffer and the reaction was followed at 415nm. The protein concentration was around 10#M on a heme basis after mixing. All experiments were carried out at 20°C and pH 7.0 in 0.1 M potassium phosphate buffer.
Native ~A and fl A chains were isolated from freshly prepared Hb A by the procedure of Bucci and Fronticelli (1965) after minor modifications (Enoki et al., 1983). Cation exchange chromatography with CM32 (Whatman) and anion exchange chromatography with DE32 (Whatman) were used for isolation of flA and ~A, respectively. Removal of p-chloromercuribenzoate was carried out by eluting the mercurated chains through a Sephadex G10-column (2.5 × 30cm) to which were preloaded fl-mercaptoethanol (Tyuma Oxygen equilibrium measurements Oxygen equilibrium studies were peret al., 1966). Sulphydryl groups of the formed according to a spectrophotometric isolated chains were verified as being fully procedure (Enoki, 1959) at 20°C. The regenerated by the titration with pprotein concentration was around 100 #M chloromercuribenzoate (Boyer, 1954). on a heme basis. Chemicals were of guaranteed grade and Physicochemical analyses purchased from Nacalai Tesque (Kyoto). Starch gel electrophoresis with TrisN 2 gas was of Zero-A Grade (99.999 EDTA-borate buffer (pH 8.6) system (Smithies, 1955) was used for characteriz- percent or higher purity) and further ation of the isolated myoglobins and hemo- purified by bubbling through 0.1 M globin chains. The gels were stained with 10 vanadous sulphate solution. percent Amido Black 10B. The molecular homogeneity was also ascertained by sedi- Results mentation velocity analysis of the proteins The starch gel electrophoresis of (0.5 percent) in 0.1 M potassium phosphate myoglobins and hemoglobin chains used in (pH 7.0) using a Hitachi analytical ultracenthis study is shown in Fig. I. The proteins trifuge 282. can be concluded to be sufficiently purified, Kinetic measurements of oxygen combi- except for murine myoglobin which shows a little contamination by a slower minor nation and dissociation component. The minor component, which Kinetic measurements were performed persisted after repeated attempts at with a stopped-flow spectrophotometer purification, was later found to have an RA401 (Union Giken, Osaka) as described identical oxygen binding property as the previously (Matsumura et al., 1990). Using main component (Enoki et al., unpublished an observation tube of 2 mm path length data). Further purification, therefore, was and driving pressure of 7 kg/cm 2 the dead time of the apparatus was 0.9msec (Matsumura et al., 1990). For oxygen + combination, deoxymyoglobins or deoxyhemoglobin chains were rapidly mixed with oxygen solution of known concentrations in 0.1 M potassium phosphate buffer (pH 7.0) and the reaction process was followed at Origin 430nm. The deoxygenated proteins were prepared by passing the samples through an Fig. 1. Starch gel electrophoresis of the myoglobins anaerobic Sephadex G-25 (Pharmacia) and isolated hemoglobin chains used in the present column (0.9 x 25 cm) equilibrated and study. From left: canine, rat, murine and chicken myoglobins and ~A and flA chains of Hb A. eluted with the N2-saturated oxygen-free Tris-EDTA-borate buffer system (pH 8.6) and buffer, to which 10-fold molar excess of Amido Black 10B stain. p
196
Y. Enoki et al. (a)
not attempted in this study. Sedimentation velocity analysis revealed that the sedimenting boundary of myoglobins was monodisperse with S20.wvalues around 1.7 S (data not shown). The reaction of myoglobins or isolated hemoglobin chains with oxygen can be described in a simple scheme,
1.0'
0.5 i0.2
k'
Mb +
0 2~ k
MbO2,
K = k '/k.
0.1
(1)
I
For oxygen combination studies, excess of oxygen over the proteins was employed so that the reaction could be regarded as pseudo-first-order. In addition, oxygen concentration is so high that the 'off' rate constant (k) hardly makes any contribution to the observed reaction rate, i.e. the backward reaction can be ignored. Thus the rate equation can be written as d ln[Mb] dt - k '[02].
I
I
0.8 Time ( m s e c )
I
1.6
Co) 1.0
0.5
i0.2
0.1 I
(2)
Oxygen binding is a rapid process which results in a loss of a significant part of the total reaction during the dead time of the present apparatus. However, sufficient absorbance change could be still observed and a fairly good first-order kinetic trace was obtained as shown in Fig. 2a. For chicken myoglobin in this case the oxygen combination rate constant (k ') was determined as 18 x 10 6 M - ' sec- 1. When the oxygenated proteins were rapidly mixed with dithionite solution and free oxygen kept at zero, the reaction should follow the first-order kinetics and the equation must be d l n [ M b O j _ k. dt
I
0
(3)
Actually the semilogarithmic plot for chicken oxymyoglobin showed the firstorder time course (Fig. 2b) and an oxygen dissociation rate constant (k) of 18 sec-' was obtained. The rate constants obtained for canine, rat, murine and chicken myoglobins and ~A and fla chains are summarized in Table 1. Since myoglobins and isolated hemoglobin chains exhibit no cooperativity (hyperbolic oxygen equilibrium curve) without any heterotropic interactions, the reaction with oxygen must be a single step bimolecular reaction as described in
I
0
I
40 80 Time (mse~)
I
120
Fig. 2. First-order plots of oxygen combination (a) to and oxygen dissociation (b) from chicken gizzard myoglobin I (main component). [Mb]: 10pM in 0.I M potassium phosphate (pH 7.0), 20°C. (a) [02]: 7 0 # M (b) ['Na2S204]: 6mM.
equation (1). Then, the oxygen affinity as conventionally expressed by Ps0 (the oxygen pressure for half-oxygenation), which, in turn, corresponds to the reciprocal of the association equilibrium constant (K), should be given by the ratio k / k ': Ps0 = 1/K = k / k ".
(4)
Table 1. Rate constants for reactions of the vertebrate myoglobins and the isolated hemoglobin chains with oxygen
Myoglobins Canine Rat Murine Chicken gizzard I* III?
k ' x 10 -6
k
(M-- 1 sec - 1)
(sec- 1)
20 + 1.7 (8) 22 + 0.8 (5) 17 _ 0.5 (5)
16 + 0.5 (5) 22 + 0.6 (6) 22 __+0.5 (7)
18-1-1.3(6) 19 + 0.7 (5)
18+__0.4(6) 18 + 0.5 (7)
38 _+ 6 (6) 42 _ 4 (6)
19 _+ 1.6 (6) 12 + 0.4 (6)
Hb chains aA fl A
Conditions: 0.1 M potassium phosphate, pH 7.0, 20°C. *Major, tminor components. Concentration of myoglobins: 10#M. Mean +_ SEM, figures in parentheses are number of determinations.
Kinetically determined Ps0 of myoglobins 1.0
curves based on the kinetic data are compared with the actual equilibrium data in Figs 3 and 4. Agreement of both lines of the data is satisfactory in both myoglobins and hemoglobin chains. The values of Ps0 calculated from the rate constants (Table 1) for the present four myoglobins and two hemoglobin chains are listed in Table 2 along with those from the direct equilibrium measurements. It can be seen that the values from both lines of determination are in excellent agreement with each other.
~. 0.5
I I 1.0 2.0 PO2 (ramHg)
0
Fig. 3. Oxygen equilibrium curve for chicken gizzard myoglobin I calculated from the kinetically determined Ps0 as compared with the actually obtained equilibrium data. The curve was drawn assuming the Ps0 value of 0.54mmHg. Equilibrium experiments (Enoki et al., 1984): (©) 0.1 M Tris-HCi, pH 7.5, ( 0 ) 0.1 M Tris-HC1-0.5 M NaC1, pH 7.5 and (A) 0.1 M potassium phosphate, pH 7.6. [Mb]: 1.7 x 10 -4 M, 20°C.
The ratio k / k ', having the dimension of M, can be converted to pressure in mmHg, where PO2 of 1 mmHg was assumed to correspond to 1.82 # M dissolved oxygen at 20°C from the solubility coefficient (Hodgman, 1952). According to equation 4, Ps0 for chicken myoglobin was determined to be 0.54mmHg. Oxygen equilibrium 1.0
t~ 0.5
I
0
I
0.2 0.4 PO2 (nun Hg)
197
I
0.6
Fig. 4. Oxygen equilibrium curve for the isolated fl A chain of Hb A calculated from the kinetically determined Ps0 as compared with the actually determined equilibrium data. The curve was drawn assuming the Ps0 value of 0.16mmHg. Equilibrium experiments: (©) 0.05 M bis-Tris--HC1, pH 7.3, ( 0 ) 0.05 M bis-Tris-HCI-0.4 M NaCI, pH 7.3 and (/x) 0.05M bis-Tris-HCl--0.7M NaCI, pH 7.3. [flA]: 1.7 x 10 -4 M, 20°C.
Discussion Relatively few studies have been reported for the comparison of the directly (equilibrium) and kinetically determined K or Ps0 in vertebrate myoglobins and hemoglobin chains (Wittenberg et al., 1965; Antonini, 1965; Brunori et al., 1966; Brunori and Schuster, 1969; Noble et al., 1969). A survey of these results seems to show a good agreement of both lines of K in 0cA chains but not in myoglobins and fl Achains with their SH regenerated. A fairly good agreement of the Ks in myoglobins from a gastropod mollusc, Aplysia, was argued to be fortuitous in view of small number of determinations (Wittenberg et al., 1965). The observed discrepancies, which should not be present in the simple bimolecular reaction mode as expected from the results of temperature-jump relaxation experiments (Brunori and Schuster, 1969), have not been satisfactorily explained. One reason for the discrepancy might be that the samples used in the experiments were not pure and not functionally native. In these previous reports, however, we could not find sufficient characterization data of the sample materials to judge these concerns. Insufficient -SH regeneration might be another reason in the case of fl A chains. Indeed, the extent of regeneration was reported to be as low as 50-80 percent, and there were quite large differences of K (Ps0) between fl SHand fl PM8(Brunori et al., 1966; Brunori and Schuster, 1969). In our experience, the regeneration with thioglycolate as used in these studies is mostly incomplete and possibly forms mixed disulphide. More
198
Y. Enoki et aL Table 2. Comparison of Ps0 values from the kinetic and equilibrium determinations for the vertebrate myoglobins and the isolated hemoglobin chains Ps0 (mmHg) determined from Myoglobins Canine Rat Murine Chicken gizzard I III Skeletal muscle
Kinetics*
Equilibriumt
0.48 + 0.06 (10) 0.59+0.14(10) 0.72 + 0.10 (10)
0.48 ___0.03 (8) 0.58+0.02(8) 0.69 + 0.05 (7)
0.54 + 0.04 (10) 0.51 + 0.03 (10) --
0.50 + 0.01 (9)~ -0.50 +__0.01 (6):~
Hb chains 0~A
0.28 + 0.04 (10)
0.33 + 0.01 (7)
j~A
0.16-t-0.01(10)
0.16"1-0.01(11)
*Conditions are the same as in Table 1. tDetermined in Tris-HC1 buffer at various pHs and ionic strengths and 20°C. Concentration of myoglobins and hemoglobin chains for the equilibrium experiments: 1.7 x 10-4M (on heme). :~Enoki et al. (1984). Mean + SEM, figures in parentheses are number of determinations.
recent literature demonstrates better agreement between results of equilibrium and kinetic measurements (Amiconi et al., 1972; DiIorio et al., 1986; Smit et al., 1986; Kraus and Wittenberg, 1990). The objects of these studies, however, were all invertebrate (arthropod, flatworm and mollusc) hemoglobins. In our present study, all the samples were freshly prepared and used immediately. Characterization by gel electrophoresis (Fig. 1) and ultracentrifugal analysis showed sufficient purity of the samples. We did not attempt further elimination of a minor component in murine myoglobin (Fig. 1), since this was shown to have an identical oxygenation behaviour as the major component (Enoki et al., unpublished results). Further structural and functional characterization of the minor component will be published elsewhere. For SH regeneration of the hemoglobin chains, we used a column method in which the mercurated chains were treated to remove p-chloromercuribenzoate with #-mercaptoethanol passing each other during elution through a Sephadex G-10 column (Tyuma et al., 1966). This procedure functions with sufficiently short contact of the chains with the thiol agent and, therefore, avoids the danger of mixed disulphide formation. SH
titres, after regeneration, were 1.01/chain for 0cA (100 percent regeneration) and 1.94+0.04/chain ( N = 5 ) for #A (97% regeneration). Various buffers (Tris, bis-Tris and phosphate) of various concentrations and pH were used in the present kinetic and equilibrium studies. Furthermore, there was 10-fold or more difference in protein concentration between both lines of the experiments. These differences, however, should not affect the oxygenation behaviour, since it has been well established that neither myoglobin nor isolated hemoglobin chain show any solvent and concentration effect in their oxygen binding (Rossi-Fanelli and Antonini, 1958; Antonini, 1965). The excellent agreement (Table 2) between the Ps0 values derived from the rate constants ( k / k ') and those from the direct equilibrium measurements is reasonably consistent with the expected simple bimolecular reaction model. It should be noted that the kinetic and equilibrium determinations were carried out independently by two different individuals, and furthermore, for chicken myoglobin, at different times. Apart from the present scope of investigation, the kinetic analysis is an important means for full understanding of the physio-
Kinetically determined Ps0 of myoglobins logical f u n c t i o n . As r e p o r t e d b y K r a u s a n d W i t t e n b e r g (1990), the three h e m o g l o b i n s isolated f r o m the b a c t e r i a l s y m b i o n t - h a r b o u r i n g gill o f the bivalve mollusc, Lucina pectinata, h a v e similar o x y g e n affinities a n d , t h e r e f o r e , similar k / k ' ratios, b u t v e r y different rate constants. Information r e g a r d i n g kinetic c o n s t a n t s is e x t r e m e l y i m p o r t a n t in such a case. T o c o n c l u d e , the kinetic a p p r o a c h to oxygenation reaction of myoglobins and h e m o g l o b i n chains is n o t o n l y essential for f u n d a m e n t a l u n d e r s t a n d i n g o f the physiological f u n c t i o n , b u t also offers a r a p i d a n d a c c u r a t e m e a n s to d e t e r m i n e their o x y g e n affinity (Ps0). T h e kinetic a p p r o a c h to the d e t e r m i n a t i o n o f Ps0 is n o w successful in a variety o f avian m y o g l o b i n s .
References Amiconi G., Antonini E., Brunori M., Formaneck H. and Huber R. (1972) Functional properties of native and reconstituted hemoglobins from Chironomus thummi thummi. Eur. J. Biochem. 31, 52-58. Antonini E. (1965) Interrelationship between structure and function in hemoglobin and myoglobin. Physiol. Rev. 45, 123-170. Bauer C. and Pacyna B. (1975) The conversion of trivalent to divalent iron in hemoglobin of various species. Analyt. Biochem. 65, 445-448. Boyer P. D. (1954) Spectrophotometric study of the reaction of protein -SH with organic mercurials. J. Am. Chem. Soc. 76, 4331-4337. Brunori M., Noble R. W., Antonini E. and Wyman J. (1966) The reactions of the isolated ~t and fl chains of human hemoglobin with oxygen and carbon monoxide. J. biol. Chem. 241, 5238-5243. Brunori M. and Schuster T. M. (1969) Kinetic studies of ligand binding to hemoglobin and its isolated subunits by the temperature jump relaxation method. J. biol. Chem. 244, 4046-4053. Bucci E. and Fronticelli C. (1965) A new method for the preparation of ~ and fl subunits of human hemoglobin. J. biol. Chem. 240, PC551-PC558. Biinning K. and Hamm R. (1969) Priiparation von
199
reinen Myoglobin mittels der Gelchromatographie. J. Chromatogr. 43, 450-456. Dilorio E. E., Meier U. T., Smit J. D. J. and Winterhalter K. H. (1986) Kinetics of oxygen and carbon monoxide binding to liver fluke (Dicrocoelium dendriticum) hemoglobin. An extreme case? J. biol. Chem. 260, 2160-2164. Enoki Y. (1959) Salt effects on hemoglobin-oxygen equilibrium. J. Nara Med. Assoc. 10, 345-355. Enoki Y., Ochiai T., Ohga Y., Kohzuki H. and Sakata S. (1983) Preparation of native ~t and fl subunits from canine hemoglobin. Biochim. biophys. Acta 744, 71-75. Enoki Y., Ohga Y., Kawase M. and Nakatani A. (1984) Identical myoglobin is present in both skeletal and smooth muscles of chicken. Biochim. biophys. Acta 789, 334-341. Hodgrnan C. D. (1952) Handbook of Chemistry and Physics, 34th Edn. p. 1532. Chemical Rubber Co., Cleveland, OH. Kraus D. W. and Wittenberg J. B. (1990) Hemoglobins of the Lucina pectinata bacteria symbiosis. I. Molecular properties, kinetics and equilibria of reactions with ligands. J. biol. Chem. 265, 1643-1653. Matsumura K., Enoki Y., Kohzuki H. and Sakata S. (1990) A simple procedure for determination of the dead time of a stopped-flow instrument. Jpn. J. Physiol. 40, 567-571. Noble R. W., Gibson Q. H., Brunori M., Antonini E. and Wyman J. (1969) The rates of combination of the isolated chains of human hemoglobin with oxygen. J. biol. Chem. 244, 3905-3908. Rossi-Fanelli A. and Antonini E. (1958) Studies on the oxygen and carbon monoxide equilibria of human myoglobin. Archs Biochem. Biophys. 77, 478-492. Smit J. D. G., Sick H., Peterhans A. and Gersonde K. (1986) Acid Bohr effect ofa monomeric haemoglobin from Dicrocoelium dendriticum. Mechanism of the allosteric conformation transition. Eur. J. Biochem. 155, 231-237. Smithies O. (1955) Zone electrophoresis in starch gels. Biochem. J. 61, 629-641. Tyuma I., Benesch R. E. and Benesch R. (1966) The preparation and properties of the isolated ct and fl subunits of hemoglobin A. Biochemistry 5, 2957-2962. Wittenberg B. E., Brunori M., Antonini E., Wittenberg J. B. and Wyman J. (1965) Kinetics of the reactions of Aplysia myoglobin with oxygen and carbon monoxide. Archs Biochem. Biophys. 111, 576-579.
Comp. Biochem. Physiol. Vol. I lOB,No. 1, pp. 193-199, 1995 ElsevierScienceLtd Britain 0305-0491/95$9.50+ 0.00
Pergamon
PrintedinGreat
0305-0491(94)00119-7
Oxygen affinities (Ps0) of myoglobins from four vertebrate species (Canis familiaris, Rattus norvegicus, Mus musculus and Gallus domesticus) as determined by a kinetic and an equilibrium method Yasunori Enoki, Kazuhito Matsumura, Yoshimi Ohga, Hisaharu Kohzuki and Masatoshi Hattori Second Department of Physiology, Nara Medical University, Kashihara, Nara 634, Japan Rate constants of the reaction with oxygen of myoglobins from four vertebrate species (Canis familiaris, Rattus norvegicus, Mus musculus and Gallus domesticus) and the isolated ~A and /~A chains of human adult hemoglobin (HbA) were determined by the stopped-flow-spectrophotomeric method. Half-saturation oxygen pressure (Ps0) of the proteins calculated from the rate constants, assuming a simple bimolecular reaction model, agreed very well with those directly determined by oxygen equilibria. The proteins used were freshly prepared, and fully characterized by electrophoretic and ultracentrifugai analyses. Sulphydryl groups in the Hb chains were ascertained to be completely regenerated. Key words: Myoglobins; Hemoglobin chains; Vertebrate; Oxygen affinities (Ps0); Oxygen equilibrium; Stopped-flow spectrophotometry; Oxygen binding rates; Oxygen dissociation rates.
Comp. Biochem. Physiol. IIOB, 193-199, 1995.
Introduction The physiological function of myoglobin, curate determination. Furthermore, a relawhich, in contrast to hemoglobin, exhibits tively large amount of sample is needed with neither homotropic nor heterotropic allos- a rather long measurement time. Although teric characteristics, can be described solely some recently developed instruments, using by its oxygen affinity. The oxygen affinity of a thin layer technique, achieve an oxygen the protein has usually been determined by an equilibrium curve with as little as 5 g 1sample oxygen equilibrium approach, which poses in tens of minutes or less, a favourable choice several difficulties and disadvantages. For is the kinetic approach. The kinetic method, example, considerable metmyoglobin forma- which is applicable to non-cooperative protion during the measurement, due to a high teins like myoglobins and also isolated subtendency to autoxidation, might prevent ac- units of hemoglobins, require much less time with a small amount of sample and allows the determination of oxygen affinity with less metmyoglobin formation. Correspondence to: Y. Enoki, Second Department of So far several kinetic studies on the Physiology, Nara Medical University, Kashihara, reaction of oxygen with myoglobins and Nara 634, Japan. Received 11 January 1994; accepted 10 June 1994. isolated hemoglobin chains have been 193
194
Y. Enoki et al.
reported, in which equilibrium constants calculated from kinetic constants are compared with those from the oxygen equilibrium measurements (Antonini, 1965; Wittenberg et al., 1965; Brunori et al., 1966; Brunori and Schuster, 1969; Noble et al., 1969). Further research, all studying noncooperative invertebrate hemoglobins, is found in recent literature (Amiconi et al., 1972; Dilorio et al., 1986; Smit et al., 1986; Kraus and Wittenberg, 1990). Most of the results, however, show discrepancies between both lines of determination, which have not yet been satisfactorily explained. Hypotheses, albeit not proved yet, were proposed that the departure from the simple reaction model might be due to the formation of short-lived intermediates or conformational changes following the simple bimolecular reaction of the proteins and ligands (Antonini, 1965; Brunori et al., 1966). Another reason for the discrepancy, if only partial, might be contamination and denaturation of the samples, where commercially available material was used after insufficient purification. It should be also stressed that the kinetic measurement has its own merit. The similar oxygen affinities could be achieved by very different balances of the combination and dissociation rates, as actually observed in non-cooperative hemoglobins isolated from the bivalve mollusc, Lucina pectinata (Kraus and Wittenberg, 1990). The kinetic information is extremely valuable in such a case and irreplaceable by the equilibrium measurement. In this paper, the kinetic constants for the reaction with oxygen of freshly prepared canine, rat, murine and chicken myoglobins and also the isolated chains (ctA and fl A) from human adult hemoglobin are reported, and the oxygen affinities (Ps0) are calculated. We also examine whether the Ps0 thus derived agree with those separately determined by the direct equilibrium measurements or not. Agreement between both lines of the determination is excellent.
reported (Enoki et al., 1984). Muscles (usually 100-200 g) excised in small pieces were extracted with three-fold volumes of chilled water by using a juice mixer, and centrifuged to obtain a clear supernate (18,000 rpm x 20 min, 0°C). The extract was heated for 10min at 50°C in a thermostatted bath under continuous shaking. The denatured material thus developed was spun off by refrigerated centrifugation after CO flushing. This heat denaturation step (Bfinning and Hamm, 1969), instead of the commonly used salting-out procedure, saved much labour and time in effectively removing hemoglobin, a troublesome contaminant, in the extract. The extract was concentrated by an ultrafiltration apparatus (Toyo model UHP 90), loaded onto a 3 x 9 5 c m column of Sephadex G-75 (Pharmacia) equilibrated with 0.05 M Tris-HC1-0.05M NaC1-0.001 M EDTA (pH 7.5), eluted with the same buffer and the absorbance of the effluent was monitored at 540 and 280 nm. The resultant myoglobin fractions were pooled, concentrated and dialysed overnight against the equilibrating buffer (pH 8.3) for chromatofocusing. The dialysed material, after refrigerated centrifugation, was applied onto a PBE 94 (Pharmacia) column equilibrated with CO:-free 0.025 M Tris-acetate buffer (pH 8.3) and eluted with PB 96 buffer (Pharmacia), 13-fold diluted and pH adjusted to 6.0 with acetic acid. An Amicon CFI0 x 30 column with a 20 ml gel bed was used for up to 100mg myoglobin. The elution was monitored at 540 and 280 nm and myoglobin fractions, as judged by visible absorption spectrum, were pooled, and dialysed overnight against deionized water. PB 96 buffer in the sample was then removed by hydrophobic interaction chromatography with Sepharose CL-4B (Pharmacia) according to the manufacturer's instruction. After dialysis against deionized water and concentration, myoglobin solution was stored in a fine droplet form at -135°C. Reduction of ferric myoglobin, if necessary, was performed with dithionite in a mixed bed BioMaterials and Methods Rad AG501-X8 (D) column (Bauer and Preparation o f myoglobins Pacyna, 1975). All the chromatographic Myoglobins were purified by a heat procedures were conducted in the cold denaturation-gel filtration procedure com- (4°C) and the buffers were chilled and bined with chromatofocusing as previously saturated with CO.
Kinetically determined Pso of myoglobins
Preparation of adult human hemoglobin chains
195
dithionite had been loaded immediately before applying the sample. During the course of elution, reduction of the ferric forms, if any, as well as the deoxygenation was accomplished. For oxygen dissociation, the oxygenated proteins were rapidly mixed with 0.2 percent dithionite in the N2-saturated oxygen-free buffer and the reaction was followed at 415nm. The protein concentration was around 10#M on a heme basis after mixing. All experiments were carried out at 20°C and pH 7.0 in 0.1 M potassium phosphate buffer.
Native ~A and fl A chains were isolated from freshly prepared Hb A by the procedure of Bucci and Fronticelli (1965) after minor modifications (Enoki et al., 1983). Cation exchange chromatography with CM32 (Whatman) and anion exchange chromatography with DE32 (Whatman) were used for isolation of flA and ~A, respectively. Removal of p-chloromercuribenzoate was carried out by eluting the mercurated chains through a Sephadex G10-column (2.5 × 30cm) to which were preloaded fl-mercaptoethanol (Tyuma Oxygen equilibrium measurements Oxygen equilibrium studies were peret al., 1966). Sulphydryl groups of the formed according to a spectrophotometric isolated chains were verified as being fully procedure (Enoki, 1959) at 20°C. The regenerated by the titration with pprotein concentration was around 100 #M chloromercuribenzoate (Boyer, 1954). on a heme basis. Chemicals were of guaranteed grade and Physicochemical analyses purchased from Nacalai Tesque (Kyoto). Starch gel electrophoresis with TrisN 2 gas was of Zero-A Grade (99.999 EDTA-borate buffer (pH 8.6) system (Smithies, 1955) was used for characteriz- percent or higher purity) and further ation of the isolated myoglobins and hemo- purified by bubbling through 0.1 M globin chains. The gels were stained with 10 vanadous sulphate solution. percent Amido Black 10B. The molecular homogeneity was also ascertained by sedi- Results mentation velocity analysis of the proteins The starch gel electrophoresis of (0.5 percent) in 0.1 M potassium phosphate myoglobins and hemoglobin chains used in (pH 7.0) using a Hitachi analytical ultracenthis study is shown in Fig. I. The proteins trifuge 282. can be concluded to be sufficiently purified, Kinetic measurements of oxygen combi- except for murine myoglobin which shows a little contamination by a slower minor nation and dissociation component. The minor component, which Kinetic measurements were performed persisted after repeated attempts at with a stopped-flow spectrophotometer purification, was later found to have an RA401 (Union Giken, Osaka) as described identical oxygen binding property as the previously (Matsumura et al., 1990). Using main component (Enoki et al., unpublished an observation tube of 2 mm path length data). Further purification, therefore, was and driving pressure of 7 kg/cm 2 the dead time of the apparatus was 0.9msec (Matsumura et al., 1990). For oxygen + combination, deoxymyoglobins or deoxyhemoglobin chains were rapidly mixed with oxygen solution of known concentrations in 0.1 M potassium phosphate buffer (pH 7.0) and the reaction process was followed at Origin 430nm. The deoxygenated proteins were prepared by passing the samples through an Fig. 1. Starch gel electrophoresis of the myoglobins anaerobic Sephadex G-25 (Pharmacia) and isolated hemoglobin chains used in the present column (0.9 x 25 cm) equilibrated and study. From left: canine, rat, murine and chicken myoglobins and ~A and flA chains of Hb A. eluted with the N2-saturated oxygen-free Tris-EDTA-borate buffer system (pH 8.6) and buffer, to which 10-fold molar excess of Amido Black 10B stain. p
196
Y. Enoki et al. (a)
not attempted in this study. Sedimentation velocity analysis revealed that the sedimenting boundary of myoglobins was monodisperse with S20.wvalues around 1.7 S (data not shown). The reaction of myoglobins or isolated hemoglobin chains with oxygen can be described in a simple scheme,
1.0'
0.5 i0.2
k'
Mb +
0 2~ k
MbO2,
K = k '/k.
0.1
(1)
I
For oxygen combination studies, excess of oxygen over the proteins was employed so that the reaction could be regarded as pseudo-first-order. In addition, oxygen concentration is so high that the 'off' rate constant (k) hardly makes any contribution to the observed reaction rate, i.e. the backward reaction can be ignored. Thus the rate equation can be written as d ln[Mb] dt - k '[02].
I
I
0.8 Time ( m s e c )
I
1.6
Co) 1.0
0.5
i0.2
0.1 I
(2)
Oxygen binding is a rapid process which results in a loss of a significant part of the total reaction during the dead time of the present apparatus. However, sufficient absorbance change could be still observed and a fairly good first-order kinetic trace was obtained as shown in Fig. 2a. For chicken myoglobin in this case the oxygen combination rate constant (k ') was determined as 18 x 10 6 M - ' sec- 1. When the oxygenated proteins were rapidly mixed with dithionite solution and free oxygen kept at zero, the reaction should follow the first-order kinetics and the equation must be d l n [ M b O j _ k. dt
I
0
(3)
Actually the semilogarithmic plot for chicken oxymyoglobin showed the firstorder time course (Fig. 2b) and an oxygen dissociation rate constant (k) of 18 sec-' was obtained. The rate constants obtained for canine, rat, murine and chicken myoglobins and ~A and fla chains are summarized in Table 1. Since myoglobins and isolated hemoglobin chains exhibit no cooperativity (hyperbolic oxygen equilibrium curve) without any heterotropic interactions, the reaction with oxygen must be a single step bimolecular reaction as described in
I
0
I
40 80 Time (mse~)
I
120
Fig. 2. First-order plots of oxygen combination (a) to and oxygen dissociation (b) from chicken gizzard myoglobin I (main component). [Mb]: 10pM in 0.I M potassium phosphate (pH 7.0), 20°C. (a) [02]: 7 0 # M (b) ['Na2S204]: 6mM.
equation (1). Then, the oxygen affinity as conventionally expressed by Ps0 (the oxygen pressure for half-oxygenation), which, in turn, corresponds to the reciprocal of the association equilibrium constant (K), should be given by the ratio k / k ': Ps0 = 1/K = k / k ".
(4)
Table 1. Rate constants for reactions of the vertebrate myoglobins and the isolated hemoglobin chains with oxygen
Myoglobins Canine Rat Murine Chicken gizzard I* III?
k ' x 10 -6
k
(M-- 1 sec - 1)
(sec- 1)
20 + 1.7 (8) 22 + 0.8 (5) 17 _ 0.5 (5)
16 + 0.5 (5) 22 + 0.6 (6) 22 __+0.5 (7)
18-1-1.3(6) 19 + 0.7 (5)
18+__0.4(6) 18 + 0.5 (7)
38 _+ 6 (6) 42 _ 4 (6)
19 _+ 1.6 (6) 12 + 0.4 (6)
Hb chains aA fl A
Conditions: 0.1 M potassium phosphate, pH 7.0, 20°C. *Major, tminor components. Concentration of myoglobins: 10#M. Mean +_ SEM, figures in parentheses are number of determinations.
Kinetically determined Ps0 of myoglobins 1.0
curves based on the kinetic data are compared with the actual equilibrium data in Figs 3 and 4. Agreement of both lines of the data is satisfactory in both myoglobins and hemoglobin chains. The values of Ps0 calculated from the rate constants (Table 1) for the present four myoglobins and two hemoglobin chains are listed in Table 2 along with those from the direct equilibrium measurements. It can be seen that the values from both lines of determination are in excellent agreement with each other.
~. 0.5
I I 1.0 2.0 PO2 (ramHg)
0
Fig. 3. Oxygen equilibrium curve for chicken gizzard myoglobin I calculated from the kinetically determined Ps0 as compared with the actually obtained equilibrium data. The curve was drawn assuming the Ps0 value of 0.54mmHg. Equilibrium experiments (Enoki et al., 1984): (©) 0.1 M Tris-HCi, pH 7.5, ( 0 ) 0.1 M Tris-HC1-0.5 M NaC1, pH 7.5 and (A) 0.1 M potassium phosphate, pH 7.6. [Mb]: 1.7 x 10 -4 M, 20°C.
The ratio k / k ', having the dimension of M, can be converted to pressure in mmHg, where PO2 of 1 mmHg was assumed to correspond to 1.82 # M dissolved oxygen at 20°C from the solubility coefficient (Hodgman, 1952). According to equation 4, Ps0 for chicken myoglobin was determined to be 0.54mmHg. Oxygen equilibrium 1.0
t~ 0.5
I
0
I
0.2 0.4 PO2 (nun Hg)
197
I
0.6
Fig. 4. Oxygen equilibrium curve for the isolated fl A chain of Hb A calculated from the kinetically determined Ps0 as compared with the actually determined equilibrium data. The curve was drawn assuming the Ps0 value of 0.16mmHg. Equilibrium experiments: (©) 0.05 M bis-Tris--HC1, pH 7.3, ( 0 ) 0.05 M bis-Tris-HCI-0.4 M NaCI, pH 7.3 and (/x) 0.05M bis-Tris-HCl--0.7M NaCI, pH 7.3. [flA]: 1.7 x 10 -4 M, 20°C.
Discussion Relatively few studies have been reported for the comparison of the directly (equilibrium) and kinetically determined K or Ps0 in vertebrate myoglobins and hemoglobin chains (Wittenberg et al., 1965; Antonini, 1965; Brunori et al., 1966; Brunori and Schuster, 1969; Noble et al., 1969). A survey of these results seems to show a good agreement of both lines of K in 0cA chains but not in myoglobins and fl Achains with their SH regenerated. A fairly good agreement of the Ks in myoglobins from a gastropod mollusc, Aplysia, was argued to be fortuitous in view of small number of determinations (Wittenberg et al., 1965). The observed discrepancies, which should not be present in the simple bimolecular reaction mode as expected from the results of temperature-jump relaxation experiments (Brunori and Schuster, 1969), have not been satisfactorily explained. One reason for the discrepancy might be that the samples used in the experiments were not pure and not functionally native. In these previous reports, however, we could not find sufficient characterization data of the sample materials to judge these concerns. Insufficient -SH regeneration might be another reason in the case of fl A chains. Indeed, the extent of regeneration was reported to be as low as 50-80 percent, and there were quite large differences of K (Ps0) between fl SHand fl PM8(Brunori et al., 1966; Brunori and Schuster, 1969). In our experience, the regeneration with thioglycolate as used in these studies is mostly incomplete and possibly forms mixed disulphide. More
198
Y. Enoki et aL Table 2. Comparison of Ps0 values from the kinetic and equilibrium determinations for the vertebrate myoglobins and the isolated hemoglobin chains Ps0 (mmHg) determined from Myoglobins Canine Rat Murine Chicken gizzard I III Skeletal muscle
Kinetics*
Equilibriumt
0.48 + 0.06 (10) 0.59+0.14(10) 0.72 + 0.10 (10)
0.48 ___0.03 (8) 0.58+0.02(8) 0.69 + 0.05 (7)
0.54 + 0.04 (10) 0.51 + 0.03 (10) --
0.50 + 0.01 (9)~ -0.50 +__0.01 (6):~
Hb chains 0~A
0.28 + 0.04 (10)
0.33 + 0.01 (7)
j~A
0.16-t-0.01(10)
0.16"1-0.01(11)
*Conditions are the same as in Table 1. tDetermined in Tris-HC1 buffer at various pHs and ionic strengths and 20°C. Concentration of myoglobins and hemoglobin chains for the equilibrium experiments: 1.7 x 10-4M (on heme). :~Enoki et al. (1984). Mean + SEM, figures in parentheses are number of determinations.
recent literature demonstrates better agreement between results of equilibrium and kinetic measurements (Amiconi et al., 1972; DiIorio et al., 1986; Smit et al., 1986; Kraus and Wittenberg, 1990). The objects of these studies, however, were all invertebrate (arthropod, flatworm and mollusc) hemoglobins. In our present study, all the samples were freshly prepared and used immediately. Characterization by gel electrophoresis (Fig. 1) and ultracentrifugal analysis showed sufficient purity of the samples. We did not attempt further elimination of a minor component in murine myoglobin (Fig. 1), since this was shown to have an identical oxygenation behaviour as the major component (Enoki et al., unpublished results). Further structural and functional characterization of the minor component will be published elsewhere. For SH regeneration of the hemoglobin chains, we used a column method in which the mercurated chains were treated to remove p-chloromercuribenzoate with #-mercaptoethanol passing each other during elution through a Sephadex G-10 column (Tyuma et al., 1966). This procedure functions with sufficiently short contact of the chains with the thiol agent and, therefore, avoids the danger of mixed disulphide formation. SH
titres, after regeneration, were 1.01/chain for 0cA (100 percent regeneration) and 1.94+0.04/chain ( N = 5 ) for #A (97% regeneration). Various buffers (Tris, bis-Tris and phosphate) of various concentrations and pH were used in the present kinetic and equilibrium studies. Furthermore, there was 10-fold or more difference in protein concentration between both lines of the experiments. These differences, however, should not affect the oxygenation behaviour, since it has been well established that neither myoglobin nor isolated hemoglobin chain show any solvent and concentration effect in their oxygen binding (Rossi-Fanelli and Antonini, 1958; Antonini, 1965). The excellent agreement (Table 2) between the Ps0 values derived from the rate constants ( k / k ') and those from the direct equilibrium measurements is reasonably consistent with the expected simple bimolecular reaction model. It should be noted that the kinetic and equilibrium determinations were carried out independently by two different individuals, and furthermore, for chicken myoglobin, at different times. Apart from the present scope of investigation, the kinetic analysis is an important means for full understanding of the physio-
Kinetically determined Ps0 of myoglobins logical f u n c t i o n . As r e p o r t e d b y K r a u s a n d W i t t e n b e r g (1990), the three h e m o g l o b i n s isolated f r o m the b a c t e r i a l s y m b i o n t - h a r b o u r i n g gill o f the bivalve mollusc, Lucina pectinata, h a v e similar o x y g e n affinities a n d , t h e r e f o r e , similar k / k ' ratios, b u t v e r y different rate constants. Information r e g a r d i n g kinetic c o n s t a n t s is e x t r e m e l y i m p o r t a n t in such a case. T o c o n c l u d e , the kinetic a p p r o a c h to oxygenation reaction of myoglobins and h e m o g l o b i n chains is n o t o n l y essential for f u n d a m e n t a l u n d e r s t a n d i n g o f the physiological f u n c t i o n , b u t also offers a r a p i d a n d a c c u r a t e m e a n s to d e t e r m i n e their o x y g e n affinity (Ps0). T h e kinetic a p p r o a c h to the d e t e r m i n a t i o n o f Ps0 is n o w successful in a variety o f avian m y o g l o b i n s .
References Amiconi G., Antonini E., Brunori M., Formaneck H. and Huber R. (1972) Functional properties of native and reconstituted hemoglobins from Chironomus thummi thummi. Eur. J. Biochem. 31, 52-58. Antonini E. (1965) Interrelationship between structure and function in hemoglobin and myoglobin. Physiol. Rev. 45, 123-170. Bauer C. and Pacyna B. (1975) The conversion of trivalent to divalent iron in hemoglobin of various species. Analyt. Biochem. 65, 445-448. Boyer P. D. (1954) Spectrophotometric study of the reaction of protein -SH with organic mercurials. J. Am. Chem. Soc. 76, 4331-4337. Brunori M., Noble R. W., Antonini E. and Wyman J. (1966) The reactions of the isolated ~t and fl chains of human hemoglobin with oxygen and carbon monoxide. J. biol. Chem. 241, 5238-5243. Brunori M. and Schuster T. M. (1969) Kinetic studies of ligand binding to hemoglobin and its isolated subunits by the temperature jump relaxation method. J. biol. Chem. 244, 4046-4053. Bucci E. and Fronticelli C. (1965) A new method for the preparation of ~ and fl subunits of human hemoglobin. J. biol. Chem. 240, PC551-PC558. Biinning K. and Hamm R. (1969) Priiparation von
199
reinen Myoglobin mittels der Gelchromatographie. J. Chromatogr. 43, 450-456. Dilorio E. E., Meier U. T., Smit J. D. J. and Winterhalter K. H. (1986) Kinetics of oxygen and carbon monoxide binding to liver fluke (Dicrocoelium dendriticum) hemoglobin. An extreme case? J. biol. Chem. 260, 2160-2164. Enoki Y. (1959) Salt effects on hemoglobin-oxygen equilibrium. J. Nara Med. Assoc. 10, 345-355. Enoki Y., Ochiai T., Ohga Y., Kohzuki H. and Sakata S. (1983) Preparation of native ~t and fl subunits from canine hemoglobin. Biochim. biophys. Acta 744, 71-75. Enoki Y., Ohga Y., Kawase M. and Nakatani A. (1984) Identical myoglobin is present in both skeletal and smooth muscles of chicken. Biochim. biophys. Acta 789, 334-341. Hodgrnan C. D. (1952) Handbook of Chemistry and Physics, 34th Edn. p. 1532. Chemical Rubber Co., Cleveland, OH. Kraus D. W. and Wittenberg J. B. (1990) Hemoglobins of the Lucina pectinata bacteria symbiosis. I. Molecular properties, kinetics and equilibria of reactions with ligands. J. biol. Chem. 265, 1643-1653. Matsumura K., Enoki Y., Kohzuki H. and Sakata S. (1990) A simple procedure for determination of the dead time of a stopped-flow instrument. Jpn. J. Physiol. 40, 567-571. Noble R. W., Gibson Q. H., Brunori M., Antonini E. and Wyman J. (1969) The rates of combination of the isolated chains of human hemoglobin with oxygen. J. biol. Chem. 244, 3905-3908. Rossi-Fanelli A. and Antonini E. (1958) Studies on the oxygen and carbon monoxide equilibria of human myoglobin. Archs Biochem. Biophys. 77, 478-492. Smit J. D. G., Sick H., Peterhans A. and Gersonde K. (1986) Acid Bohr effect ofa monomeric haemoglobin from Dicrocoelium dendriticum. Mechanism of the allosteric conformation transition. Eur. J. Biochem. 155, 231-237. Smithies O. (1955) Zone electrophoresis in starch gels. Biochem. J. 61, 629-641. Tyuma I., Benesch R. E. and Benesch R. (1966) The preparation and properties of the isolated ct and fl subunits of hemoglobin A. Biochemistry 5, 2957-2962. Wittenberg B. E., Brunori M., Antonini E., Wittenberg J. B. and Wyman J. (1965) Kinetics of the reactions of Aplysia myoglobin with oxygen and carbon monoxide. Archs Biochem. Biophys. 111, 576-579.