The swinging movement of the distal histidine residue and the autoxidation reaction for midge larval hemoglobins Satoshi Kamimura 1 , Ariki Matsuoka 2 , Kiyohiro Imai 3 and Keiji Shikama 1,4 1 Biological Institute, Graduate School of Life Sciences, Tohoku University, Sendai, Japan; 2 Fukushima Medical University, Fukushima, Japan; 3 Laboratory of Nanobiology, Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan; 4 PHP Laboratory for Molecular Biology, Nakayama-Yoshinari, Sendai, Japan Some insects have a globin exclusively in their fast-growing larval stage. This is the case in the 4th-instar larva of Toku- nagayusurika akamusi, a common midge found in Japan. In the polymorphic hemoglobin comprised of 11 separable components, hemoglobin VII (Ta-VII Hb) was of particular interest. When its ferric met-form was exposed to pH 5.0 from 7.2, the distal histidine was found to swing away from the E7 position. As a result, the iron(III) was converted from a hexacoordinate to a pentacoordinate form by a concom- itant loss of the axial water ligand. The corresponding spectral changes in the Soret band were therefore followed by stopped-flow and rapid-scan techniques, and the observed first-order rate constants of k out ¼ 25 s )1 and k in ¼ 128 s )1 were obtained for the outward and inward movements, respectively, of the distal histidine residue in 0.1 M buffer at 25 °C. For O 2 affinity, Ta-VII Hb showed a value of P 50 ¼ 1.7 Torr at pH 7.4, accompanied with a remarkable Bohr effect (dH + ¼ )0.58) almost equal to that of mammalian hemoglobins. We have also investigated the stability property of Ta-VII HbO 2 in terms of the autoxi- dation rate over a wide range of pH from 4 to 11. The resulting pH-dependence curve was compared with those of another component Ta-V HbO 2 and sperm whale MbO 2 , and described based on a nucleophilic displacement mech- anism. In light of the O 2 binding affinity, Bohr effect and considerable stability of the bound O 2 against acidic autoxidation, we conclude that T. akamusi Hb VII can play an important role in O 2 transport and storage as the major component in the larval hemolymph. Keywords: Insect (midge) Hb; distal (E7) histidine; pH jump; swinging movement; heme oxidation. In previous papers, we reported that the hemoglobin from the 4th-instar larva of Tokunagayusurika akamusi,acom- mon midge (Diptera) found in eutrophic lakes in Japan, is comprised of as many as 11 separable components (IA, IB, II, III, IV, V, VIA, VIB, VII, VIII and IX) on a DEAE- cellulose column, and found that these can be classified into two groups on the basis of their presence or absence of the distal (E7) histidine residue. For instance, hemoglobin VII consists of 150 amino-acid residues and contains the usual distal histidine at position 64, whereas component V replaces it by an isoleucine at position 66 [1]. This is certainly one of the unique characters of T. akamusi hemoglobins, as all the Chironomus hemoglobins have a distal histidine residue. Among the T. akamusi hemoglobins, component VII was of particular interest. When its ferric met-form was placed in acidic pH range, its Soret peak was considerably blue-shifted and accompanied by a marked decrease in intensity, indicative of the hemoglobin being converted into a structure quite similar to that of Aplysia (sea hare) myoglobin lacking the usual distal histidine residue. The pH-dependent Soret magnetic circular dichroism (CD) spectra also revealed that hemoglobin VII is in an equilibrium between a hexacoor- dinate and a pentacoordinate structure for its ferric heme iron [2]. We attributed this to a transition of an iron-ligated water molecule that is hydrogen-bonded to the distal histidine, to a water-free iron with the histidine swung away from its E7 position. In the present paper, we describe the swinging movement of the distal histidine residue in T. akamusi hemoglobin VII followed by stopped-flow rapid mixing techniques. To demonstrate any unusual character of midge larval hemo- globins, we also examine the autoxidation rate over a wide range of pH, as well as the oxygen equilibrium parameters, in 0.1 M buffer at 25 °C. These examinations will provide us with new insights about the unique characters of midge larval hemoglobins and also the biochemical properties of heme proteins in general. Materials and methods Chemicals Butyl-Toyopearl (650 M ) was a product of Tosoh (Tokyo). CM-cellulose (CM-32) and DEAE-cellulose (DE-32) were purchased from Whatman. Mes, Mops, Taps, Caps and Tris for buffer systems, and all other chemicals, were of reagent grade from Wako Pure Chemicals (Osaka). Solu- tions were made with deionized and glass-distilled water. Preparation of midge larval hemoglobin components As described previously [1], the hemoglobin from the 4th- instar larva of T. akamusi is comprised of as many as 11 Correspondence to K. Shikama, PHP Laboratory for Molecular Biology, Nakayama-Yoshinari 1-16-8, Sendai 989-3203, Japan. E-mail: shikama@mail.cc.tohoku.ac.jp (Received 13 September 2002, revised 28 January 2003, accepted 4 February 2003) Eur. J. Biochem. 270, 1424–1433 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03498.x separable components on a DEAE-cellulose column. Of these, hemoglobin VII, a major component, was isolated as follows. Frozen larvae (20 g in each experiment) were thawed quickly and homogenized with Polytron (Kinema- tica, Switzerland) in a 3-volume (v/w) of 50 m M Tris/HCl buffer (pH 8.4). After insoluble materials had been removed by centrifugation, the extract was fractionated with ammo- nium sulfate between 55% and 100% saturation. The hemoglobin was centrifuged down at 30 000 g for 30 min, dissolved in a minimum volume of 50 m M Tris/HCl buffer (pH 8.4), and subjected to a Butyl-Toyopearl column (5 · 10 cm) equilibrated with ammonium sulfate of 40% saturation at pH 8.4. The elution was then carried out with a linear concentration gradient of ammonium sulfate from 40% to 20% saturation in 50 m M Tris/HCl buffer (pH 8.4). In this procedure, components IV, VIA, VIB and VII were still not fully separated from each other. To isolate hemoglobin VII, the combined protein solution was there- fore developed on a DEAE-cellulose column (2 · 7cm), which had been equilibrated with 5 m M Tris/HCl buffer (pH 8.4). In the elution with 25 m M Tris/HCl buffer pH 7.9, a small peak containing components IV, VIA and VIB appeared first, and a large major peak containing component VII in the ferric met-form was the second. The oxy-form of component VII was finally obtained by changing the buffer to 50 m M at pH 7.9. All these procedures were carried out at low temperature (0–4 °C) as far as possible. In the same way, hemoglobin V was also prepared as described previously [3]. Identification of each component was made using PAGE, according to the specifications of Riggs [4]. The concentration of T. akamusi hemoglobin was deter- mined, after conversion into cyanomet-form, by using an absorption coefficient of 10.6 m M )1 Æcm )1 at 540 nm for both components V and VII [1]. Stopped-flow and rapid-scan spectroscopy Rapid scan experiments were carried out in an Otsuka stopped-flow spectrophotometer (RA-2000) equipped with a 10-mm light path cell and two sample reservoirs (3 mL). A 350-lL solution of ferric hemoglobin VII (10 l M heme) in 10 m M Tris/HCl buffer, pH 7.2, was mixed with an equal volume of 0.2 M Mes/NaOH buffer pH 5.0. The absorption spectra were recorded over 350–450 nm by means of a rapid scanning photodiode array capable of 512/8192 counts in 1/3 ms. Reaction temperature was controlled by a water bath (Lauda NM-454 L) maintained to within ± 0.1 °C. The pH of the reaction mixture was carefully checked, before and after the run, with a Hitachi-Horiba pH meter (model F-22). Data sets were saved on an attached computer (DELL, OptiPLex GXM 5133) for further analysis. Time courses of the absorbance changes recorded at different wavelengths or different pH values were simultaneously fitted to a series of exponentials so as to treat the reaction as 1st-order processes. Autoxidation rate measurements According to our standard procedure, the rate of autox- idation of oxyhemoglobin was measured in 0.1 M buffer at 25 °C over a wide pH range (4–12) and in the presence of 1m M EDTA. For example, a 1-mL solution containing 0.2 M appropriate buffer and 2 m M EDTA was placed in a test tube and incubated in a water bath maintained at 25 (± 0.1) °C. The reaction was started by adding an equal volume of fresh HbO 2 solution (40 l M ), and the changes in the absorption spectrum over 450–700 nm were recorded on the same chart at measured intervals of time. For the final state of each run, the hemoglobin was completely converted to the ferric met-form by the addition of potassium ferricyanide. The buffers used were acetate, Mes, Mops, Taps, Caps and phosphate (pK 3 ). The pH of the reaction mixture was carefully checked, before and after the run, with a Hitachi-Horiba pH meter (model F-22). Spectral measurements Absorption spectra were recorded in a Hitachi two- wavelength double-beam spectrophotometer (U-3210 or U-3300) or in a Beckman spectrophotometer (DU-650), each being equipped with a thermostatically controlled (within ± 0.1 °C) cell holder. CD spectra were obtained in a Jasco spectropolarimeter (J-720) equipped with a ther- mostatically controlled cell holder. In the Soret region, recordings were made with 10 l M hemoglobin (as heme) in a 1-mm cell and at the scale setting of 0.002 degrees per cm on the chart. Oxygen equilibrium measurements Oxygen equilibrium curves were obtained in 50 m M Tris/ HClbufferplus0.10 M KCl at 25 °C by using the automatic oxygenation apparatus developed by Imai [5]. The optical absorption and oxygen partial pressure data were acquired in real time on a microcomputer, and the results were stored on disks for further analysis. Results Rapid-scan spectroscopy for the swinging movement of the distal histidine residue in hemoglobin VII Eleven separable components of T. akamusi hemoglobin can be divided into two groups on the basis of the presence or absence of the distal (E7) histidine residue, which plays an important role in the stability of the bound dioxygen. Among those components, hemoglobins V and VII are the major ones in each different group, and make up 15% and 30%, respectively, of the total hemoglobin concentration in the 4th-instar larval hemolymph. Figure 1 represents the complete amino acid sequences of both hemoglobins. As is clear, T. akamusi hemoglobin VII (Ta-VII) consists of 150 amino-acid residues and contains the usual distal histidine at position 64, whereas hemo- globin V (Ta-V) replaces it by an isoleucine at position 66 in the 152 sequence. In addition, the B10 residues appear to be Phe32 in Ta-V and Leu30 in Ta-VII, based on the structure of Chironomus Hb. When the aquomet-form of T. akamusi hemoglobin VII was placed in an acidic pH range, the Soret peak was considerably blue-shifted and accompanied by a marked decrease in intensity, indicative of the protein being Ó FEBS 2003 Swinging of the distal histidine in midge Hb (Eur. J. Biochem. 270) 1425 converted from a hexacoordinate to a pentacoordinate structure. Using stopped-flow rapid mixing techniques, we have therefore studied this intramolecular isomerization reaction in 0.1 M buffer at 25 °C. Figure 2 shows such time- resolved Soret absorption spectra of ferric Ta-VII Hb, when its 10 l M solution in 10 m M Tris/HCl buffer, pH 7.2, was mixed with an equal volume of 0.2 M Mes/NaOH buffer, pH 5.0, in the stopped-flow apparatus at 25 °C. After the pH was decreased from 7.2 to 5.0, the spectra were recorded on a computer every 3 ms over a 500-s period and a range of 350–450 nm. In the three-dimensional display, the spectrum scanned 0.05 s after mixing still retained the usual Soret absorption with maximum centered at 407 nm, characteristic of the sixcoordinate ferric species. After 0.80 s, however, the peak was dramatically shifted to 397 nm with a set of isosbestic points at 396 and 426 nm, and accompanied by a marked decrease in intensity. All these features indicate that the protein was converted completely from a sixcoordinate form to a fivecoordinate one without formation of any intermediate species [6–8]. For more detailed study of this pH-dependent conversion, it would be much more informative to inspect some repre- sentative spectra at selected pH values, just corresponding to the two-dimensional spectra at selected times during the course of the reaction. In a previous paper [2], we have already reported such a series of spectral changes that led to the spectroscopic titration curve with a midpoint pH of 6.3. However, these spectral changes of Ta-VII were not due to acid denaturation of the protein, but were totally reversible with pH. Indeed, its CD magnitude at 222 nm exhibited a constant value of )16 500 (± 500) degÆcm 2 Æ dmol )1 over the pH range of 7.5–4.6 [2]. We have therefore concluded that the observed Soret absorbance changes can be attributed to a transition of an iron-ligated water molecule that is hydrogen-bonded to the distal histidine residue at position 64, to a water-free iron with the histidine swung away from the E7 position. Figure 3 represents such a swinging movement of the distal histidine residue in a very schematic way. Unfortunately, we cannot indicate where the distal histidine moves to in T. akamusi Hb VII, as the X-ray crystal structure is not yet available. However, it is interesting to note that this transformation reaction is accompanied by a complete reversal of the sign of the Soret CD signal. At pH 7.0, ferric Hb VII gave the CD spectrum containing a weak but distinct negative Soret signal, as shown in Fig. 4. At pH 5.0, on the other hand, the protein exhibited a well-developed, positive CD lobe with maximum centered at 406 nm. These findings strongly suggest that the swinging movement of the distal histidine would exert some Fig. 1. Amino acid sequences of T. akamusi Hb components V and VII. ThemarkersareusedtoindicatetheproximalF8-His(#),thedistalE7 residue (*), and the identical residues between both components (:). The B10 residues appear to be Phe32 in Ta-V and Leu30 in Ta-VII. Fig. 2. Time-resolved Soret absorption spectra of T. akamusi Hb VII after the pH-drop from 7.2 to 5.0 at 25 °C. The ferric met-species in 10 m M Tris buffer pH 7.2 was mixed with an equal volume of 0.2 M Mes buffer pH 5.0 in the stopped-flow apparatus. The first spectrum was for the Soret band scanned 0.05 s after mixing, while the last one is for 0.8 s later. The final heme concentration was 4.1 l M at pH 5.0. 1426 S. Kamimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 effects on the amino-acid chromophores in the very vicinity of the heme moiety, so as to change the optical rotatory dispersion of the Soret band. No such pH-dependent reversal of the CD signal (or the Cotton effect) was observed in another component Ta-V, as well as in sperm whale myoglobin. In this conversion reaction, we have also found the involvement of a single dissociable group (AH) having a pK a ¼ 6.3 at 25 °C. At a glance, this pK a value was likely that of the distal histidine residue, whose protonation would be associated with the rupture of hydrogen bonding to the coordinated water molecule. In sperm whale aquomet- myoglobin, however, no spectral change was observed in the Soret peak until the heme moiety began to separate abruptly due to the globin denaturation at pH values below 4.5. From the effect of temperatures on the pK a value, the AH group involved was found to have thermodynamic param- eters characteristic for the ionization of a carboxyl group, although its pK a value does not lie in the normal range [2]. Anyway, our next step was to clarify how rapidly the distal histidine swings away from the E7 position in T. akamusi Hb VII. Stopped-flow measurements for the distal histidine swinging in hemoglobin VII As demonstrated previously [2], the Soret absorption spectra of ferric Hb VII changed from a hexacoordinate form (k max ¼ 407 nm) to a pentacoordinate one (k max ¼ 397 nm) with a set of isosbestic points at 396, 426, 576 and 640 nm, indicative of no production of any intermediate species between the two forms. To observe the time courses of the transformation reaction of ferric Hb VII more directly, for this paper we have studied the absorbance changes by selected wavelengths such as 410, 407 and 380 nm. Figure 5 gives such an example for the spectral track followed at 407 nm (Soret peak of the ferric high-spin species) up to a full development of the reaction. The pH-drop experiment was carried out at 25 °C by mixing a 10-l M ferric Hb solution in 0.01 M Tris/HCl buffer (pH 7.2) with an equal volume of 0.2 M Mes/NaOH buffer (pH 5.0). The dead time of the apparatus was 3 ms, and the noise level was indicated in the first place as the absorbance fluctuations for the residual solution of the preceding run. Thus, the conver- sion process could be displayed as a single exponential decay over the whole course of the reaction. At the same time, it was of importance to know whether the moved residue could come back again to the E7 position with pH. We have therefore carried out similar experiments at 25 °C. A 10-l M ferric Hb VII solution in 10 m M Mes/ NaOH buffer (pH 5.0) was mixed with an equal volume of 0.2 M Tris/HCl buffer (pH 7.2). As soon as the pH was jumped from 5.0 to 7.2, the absorbance at 407 nm increased exponentially with increasing appearance of the hexacoordinate species. In formulating the pH-induced swinging movement of the distal histidine residue, at least six kinetic micro constants will be required, as follows: Fig. 3. A schematic representation for the pH-dependent swinging-out of the distal histidine residue. By a concomitant loss of the axial water molecule, T. akamusi Hb VII is transformed from a hexacoordinate to a pentacoordinate (or vacant-type) species. Fig. 4. CD spectra of T. akamusi ferric Hb VII at pH 7.0 and 5.0. Reversal of the sign of the Soret signal occurred between pH 7.0 (continuous line) and 5.0 (broken line). Heme concentration was 10 l M each in 50 m M phosphate buffer. ð1Þ Ó FEBS 2003 Swinging of the distal histidine in midge Hb (Eur. J. Biochem. 270) 1427 where AH represents the dissociable group of the regulatory amino acid residue (probably a carboxyl group of the heme propionate in this case), and the asterisk is for the unstable intermediate species for each of the two forms. The equilibrium between the hexacoordinate[-AH]* species and the pentacoordinate[-AH] form is not guaranteed, because the swung-out histidine is found in a completely different Soret CD environment. The a is the molar fraction of the hexacoordinate form present at a given pH value. In this reaction scheme, k out represents the apparent first-order rate constant for the outward movement, while k in is for the inward or swing-back movement of the distal histidine residue, as the protonation and deprotonation processes for the AH group involved would be too fast to become rate- limiting in the swinging movement. Moreover, the Soret absorption used here is completely silent in such protona- tion and deprotonation processes of a carboxyl group of the heme propionate, so that unstable intermediates, even if present, could not be detected by our spectrophotometric techniques. The swinging process, irrespective of its direction, was therefore followed by a plot of absorbance data at 407 nm as –ln{(A t ) A 1 )/(A 0 ) A 1 )} vs. time t after mixing. Figure 6 represents such first-order plots for the pH- induced conversion reaction of ferric Hb VII, from a hexacoordinate to a pentacoordinate form and vice versa, in 0.1 M buffer at 25 °C. In both semilog plots, data points were taken from several different runs of the same experiment and could easily be extended on the straight line for at least three half-lives (although not shown in Fig. 6 but clear from Fig. 5). From the slope of each straight line, we obtained the observed first-order rate constants of k out ¼ 25 s )1 for the swing-out movement and k in ¼ 128 s )1 for the swing-in process, respectively, of the distal histidine residue. Consequently, the swing-away movement takes place with a half-life period of t 1/2 ¼ 27 ms, this being less rapid than the swing-back process with t 1/2 ¼ 5ms. At present, no mechanistic explanation can be given for this rate difference between the two constants k out and k in ,aswe do not know yet exactly where the distal histidine moved to. Nevertheless, it is true that its new position is in favor of making the residue swing back again to the original E7 position more easily by the pH-jump. In the pH-shift experiments to obtain the first-order rate constant of the outward or inward movement, it was essential to check carefully both the initial and final pH values of the solution. As the reaction is started whether from the pH-drop or jump, the protein adjusts itself to the rapid change in pH to reach an equilibrium. Consequently, the rate constant (k out or k in ) observed at any intermediate pH value for the conversion involves the reacting popula- tion of the hexacoordinate or pentacoordinate form. Both molar fractions can be deduced from the spectroscopic titration curve as a function of pH, as described previously [2]. When the pH-drop experiments were carried out, for instance, from pH 7.2 to 6.5, or 6.0, or 5.5, each observed rate constant was associated with each reacted fraction of hexacoordinate species in such a manner as to lead to the required rate constant, that is to say k out for the full conversion reaction. The situation was the same with the pH-jump experiments. Therefore, the practical way to reach the intrinsic value of k out or k in was to measure the swinging reaction by jumping the pH from one extreme to the other so as to complete the isomerization reaction, irrespective of the direction. We have found such extreme limits to be Fig. 6. First-order plots for the pH-dependent transformation reaction of ferric Hb VII in 0.1 M buffer at 25 °C. In these plots, k out represents the outward movement of the distal histidine from its E7 position, while k in is for the inward or backward movement to the original position. Fig. 5. Time courses for the pH-induced transformation reaction of ferric Hb VII followed by the absorbance changes at 407 nm. The experimental conditions for mixing were the same as is in Fig. 2, and the pH was jumped down from 7.2 to 5.0 at 25 °C. The final heme concentration was 5 l M . 1428 S. Kamimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 sufficient at pH 7.2 and 5.0, respectively [2]. Certainly, four pH units should be required to complete the pK-dependent conversion of one form to the other up to a 99% level. Unfortunately, our case is not so ideal. The midge ferric hemoglobin VII is subjected to acid denaturation at pH values < 4.5, and to the alkaline transition at pH values > 7.5. In these situations, a pH shift of two units (between pH 5 and 7) was a possible way for this material. In the swinging reaction of the distal histidine residue, we also assumed that the ligand water dissociation or association step would be too fast to become rate-limiting. Oxygen binding and stability properties of T. akamusi hemoglobins: comparison of components V and VII Table 1 summarizes the oxygen equilibrium parameters of T. akamusi hemoglobins V and VII at three different pH values and 25 °C. Overall oxygen affinity was expressed in terms of the oxygen pressure to half saturate the protein, P 50 (Torr). Bohr coefficient was also calculated as dH + by the difference in log(P 50 ) between pH 6.4 and 8.4. At neutral pH, both components have P 50 values quite similar to those of mammalian myoglobins. At acidic pH, however, their O 2 affinities became lower, particularly in Ta-VII. As a result, hemoglobin VII exhibited a remarkable Bohr effect almost equal to that of human hemoglobin (dH + ¼ )0.48). In this sense, Hb VII can play a central role in oxygen supply as the major component of the larval hemolymph. It is in the ferrous form that hemoglobin or myoglobin can bind molecular oxygen reversibly and carry out its physiological function. Under air-saturated conditions, however, the oxygenated form of hemoglobin or myoglobin is easily oxidized to its ferric met-form with generation of the superoxide anion as follows: HbðIIÞðO 2 ÞÀ! k obs meHbðIIIÞþO À 2 ð2Þ where k obs represents the observed first-order rate constant at a given pH value [9]. Therefore, the rate of the autoxidation reaction is written by: Àd½HbO 2  dt ¼ k obs ½HbO 2 ð3Þ This process was followed by a plot of experimental data as )ln([HbO 2 ] t /[HbO 2 ] 0 )vs.timet, where the ratio of HbO 2 concentration after time t to that at time t ¼ 0canbe monitored by the absorbance changes at a-peak of the oxygenated species (578 nm in the case of T. akamusi HbO 2 ). Figure 7 shows such an example for the spectral changes with time in the autoxidation reaction of T. aka- musi Hb VII in 0.1 M Mops/NaOH buffer at pH 7.2 and 25 °C. The spectra evolved with a set of isosbestic points (at 525 and 593 nm) to the final state of the run, which was identified as a typical acidic met-form. Inserted is the first- order plot monitored at 578 nm to obtain the rate constant of k obs ¼ 0.28 · 10 )1 h )1 from the slope of the straight line. This autoxidation rate was several times higher than those of sperm whale MbO 2 (k obs ¼ 0.50 · 10 )2 h )1 ) and human psoas MbO 2 (k obs ¼ 0.83 · 10 )2 h )1 ), but several times lower than that of Aplysia (sea mollusca) MbO 2 (k obs ¼ 0.11 h )1 ) under the same conditions [10]. As the ferric met-species thus produced cannot bind molecular oxygen, the observed rate constant (k obs ) of autoxidation provides us with a useful measure of the stability of the bound dioxygen. In this way, if the values of k obs are plotted against the pH of the solution, we can obtain a stability profile of HbO 2 or MbO 2 in terms of the autoxidation rate as a function of pH. Figure 8 represents two such profiles for T. akamusi hemoglobins V and VII in 0.1 M buffer at 25 °C, compared with sperm whale MbO 2 as a reference. In sperm whale MbO 2 , it is clear that the rate of autoxidation increases rapidly with increasing hydrogen ion concentration, that a minimum rate appears at pH 9.2, and that a further increase occurs at higher pH values. On the other hand, Ta-V HbO 2 was quite susceptible to autoxidation over the whole range of pH values studied. At pH 9.0, for instance, its rate was Table 1. Oxygen equilibrium parameters of T. akamusi hemoglobins in 50 m M Tris buffer plus 0.1 M KCl at 25 °C. Heme concentration: 60 l M . Hb component pH P 50 (Torr) dH + Ta-V 8.4 0.57 7.4 0.79 )0.20 6.4 1.43 Ta-VII 8.4 0.63 7.4 1.7 )0.58 6.4 9.1 Fig. 7. Spectral changes with time for the autoxidation of T. akamusi Hb VII in 0.1 M Mops buffer at pH 7.2 and 25 °C. Scans were made at 3-h intervals after the fresh HbO 2 was placed in air-saturated buffer and in the presence of 1 m M EDTA. Inserted is the first-order plot monitored at 578 nm. HbO 2 concentration: 20 l M heme. Ó FEBS 2003 Swinging of the distal histidine in midge Hb (Eur. J. Biochem. 270) 1429 25 times higher than that of sperm whale MbO 2 .Further- more, its pH-dependence was unusual. The rate also increased with increasing hydrogen ion concentration but much less so than in sperm whale MbO 2 .Rather,Ta-V HbO 2 exhibited a distinct rate-saturation below pH 6. This strongly suggests that the mode of action of the proton is quite different between the two proteins. In sperm whale MbO 2 , the rate increases so rapidly at acidic pH that a value close to n ¼ )1 is always found for the slope of log(k obs )vs. pH. This is a definite indication of the involvement of a very strong acid-catalysis performed by the distal histidine residue [3,7,10–14]. In marked contrast to this, Ta-V HbO 2 stands with a slope of n ¼ )0.4 for the acidic autoxidation. In fact, this protein has an isoleucine at position 66 in place of the usual distal histidine residue (Fig. 1). In the autoxidation reaction, pH can affect the rate in many different ways. Recent kinetic and thermodynamic studies of the stability of mammalian oxymyoglobins have shown that the autoxidation reaction is not a simple, dissociative loss of O 2 – from MbO 2 but is due to a nucleophilic displacement of O 2 – from MbO 2 byawater molecule or a hydroxyl ion that can enter the heme pocket from the surrounding solvent. The iron is thus converted to the ferric met-form, and the water molecule or the hydroxyl ion remains bound to the Fe(III) at the sixth coordinate position so as to form aqua- or hydroxide-metMb. Even the complicated pH-profile for the autoxidation rate can thereby be explained primarily in terms of the following three types of displacement processes [7,10–14]: MbðIIÞðO 2 ÞþH 2 O À! k 0 MbðIIIÞðOH 2 ÞþO À 2 ð4Þ MbðIIÞðO 2 ÞþH 2 O þ H þ À! k H MbðIIIÞðOH 2 ÞþHO 2 ð5Þ MbðIIÞðO 2 ÞþOH À À! k OH MbðIIIÞðOH À ÞþO À 2 ð6Þ In these equations, k 0 is the rate constant for the basal displacement by H 2 O, k H is the rate constant for the proton- catalyzed displacement by H 2 O, and k OH is the rate constant for the displacement by OH – . The extent of the contribution of these elementary processes to the observed or overall autoxidation rate, k obs in Eqn (3), can vary with the concentrations of H + and OH – ions. Consequently, the stability of MbO 2 exhibits a very strong pH-dependence having a parabolic part, as typically seen in sperm whale myoglobin. To determine definitely the kinetic and thermo- dynamic parameters contributing to each k obs vs. pH profile therefore, we have proposed some mechanistic models for each case. The rate equations derived from these were tested for their fit to the experimental data with the aid of a computer, according to our previous specifications [11,13]. Based on such a nucleophilic displacement mechanism, the pH profile of sperm whale MbO 2 has already been analyzed completely in terms of an Ôacid-catalyzed two-state modelÕ [3,7,13]. In this model, it is assumed that a single, dissociable group, AH with pK 1 , is involved in the reaction. Consequently, there are two states of the MbO 2 , represented by A and B, at molar fractions of a and b (¼ 1–a) respectively, which are in equilibrium with each other but which differ in dissociation state for the group AH. These forms can be oxidized to metMb by displacement of O 2 – from MbO 2 by an entering water molecule or hydroxyl ion. Using the rate constants defined above, the observed rate constant, k obs in Eqn (3), could be reduced to: k obs ¼ k A 0 ½H 2 Oþk A H ½H 2 O½H þ  ÈÉ ðaÞ þ È k B 0 ½H 2 Oþk B H ½H 2 O½H þ þk B OH ½OH À  É ðbÞð7Þ where a ¼ ½H þ  ½H þ þK 1 and b ¼ð1 À aÞ¼ K 1 ½H þ þK 1 ð8Þ By iterative least-squares procedures inserting various values for K 1 , the adjustable parameter in Eqn (8), the best fit to the experimental values of k obs was obtained as a function of pH (Fig. 8). In this way, the rate constants and the acid dissociation constant involved in the autoxidation reaction of sperm whale MbO 2 were established in 0.1 M buffer at 25 °C, as already reported previously [3,7,10]. From those results, it has become evident that the proton- catalyzed processes with the rate constants k A H and k B H Fig. 8. pH-profiles for the stability of T. akamusi oxyhemoglobins and sperm whale MbO 2 in 0.1 M buffer at 25 °C. The logarithmic values of the observed first-order rate constant, k obs in h )1 , for the autoxidation reaction are plotted against the pH of the solution. The pH-profile of sperm whale MbO 2 is taken from our previous paper [7,13]. Heme concentration: 20 l M for midge; 50 l M for sperm whale. 1430 S. Kamimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 promote most of the autoxidation reaction of sperm whale MbO 2 above the basal processes in water with the rate constants k A 0 and k B 0 . The reductive displacement of the bound dioxygen as O 2 – by H 2 O can proceed without any protonation, but it is clear that the rate is enormously accelerated with the proton assistance, by a factor of 4.7 · 10 6 mol )1 for state A and by 1.1 · 10 8 mol )1 for state B. In this proton-catalysis formulated by Eqn (5), the distal E7 histidine (the dissociable group AH with pK 1 ¼ 6.2), which forms a hydrogen bond to the bound dioxygen [15], appears to facilitate the effective movement of a catalytic proton from the solvent to the bound, polarized dioxygen via the imidazole ring by a proton-relay mechanism [10,14]. On the other hand, Ta-V HbO 2 was characterized by the distinct rate-saturation behavior below pH 6 (Fig. 8). We have therefore measured the rate at more than 70 different pH values, from 4 to 12, and finally established the best fit to the experimental values of k obs by a simple Ôtwo-state modelÕ [3]. In this mechanism, we assumed that a single, dissociable group (AH with pK 1 ) is also involved in the reaction but in a different way. As a result, the pH profile for the autoxida- tion rate of this insect protein could be described by the following equation: k obs ¼ k A 0 ½H 2 O ÈÉ ðaÞþ k B 0 ½H 2 Oþk B OH ½OH À  ÈÉ ðbÞð9Þ Using the same fitting procedures, we have thus obtained the rate constants and the acid dissociation constant involved in the autoxidation reaction of Ta-V HbO 2 in 0.1 M buffer at 25 °C, as reported in the previous paper [3]. In this kinetic formulation, one of the most remarkable features was that Ta-V HbO 2 does not show any proton- catalyzed process having the term of k H [H 2 O][H + ], such as the one that can play a dominant role in the autoxidation reaction of most mammalian myoglobins or hemoglobins having the usual distal histidine residue. Instead, Ta-V HbO 2 contained a dissociable group (AH with pK 1 ¼ 6.2) that is responsible for a small rate increase at the acidic pH side. To characterize thermodynamically this dissociable group AH, we have investigated the effect of temperatures on the K 1 value, by analyzing respective pH-profiles obtained at three different temperatures. As a result, the enthalpy change of practically zero was deduced from the slope of the van’t Hoff plot. Thus, the resulting thermodynamic parameters were: DG° ¼ 33.1 kJÆmol )1 , DH° ¼ 0kJÆmol )1 and DS° ¼ )111 JÆmol )1 ÆK )1 in 0.1 M buffer at 25 °C [3]. Although its pK a value does not lie in the normal range, these parameters are those expected for the ionization of a carboxyl group, and we suggest that the most probable candidate is a carboxyl group of the heme propionates, just as in the previous case of Aplysia MbO 2 . In fact, when the protoheme was esterified with methanol to block its propionic acid side-chains, Aplysia myoglobin completely lost such a rate increase, with pK 1 ¼ 6.1 [7]. Along with this line of evidence, our great interest was in the stability property of Ta-VII HbO 2 , as in its ferric met- form the distal histidine is found to swing away from the E7 position on the acidic pH side. As demonstrated in Fig. 8, the pH profile of Ta-VII HbO 2 showed an intermediate character between sperm whale MbO 2 and Ta-V HbO 2 . With increasing hydrogen ion concentration, the rate increased rapidly as in sperm whale MbO 2 , but began to deviate from the theoretical line having a slope of n ¼ )1, and finally reached a saturation level below pH 5. Among these unusual kinetic properties, of most interest was that the rate deviation occurred with a midpoint pH of approximately 6, this being very close to the pH value (¼ 6.4) at which half of the conversion reaction from a hexacoordinate to a pentacoordinate structure is completed in the ferric met-form. At the present time, we unfortunately failed to formulate such extremely complexed pH-profile of Ta-VII HbO 2 by a simple equation, but we suggest strongly that the swinging-away movement of the distal histidine residue proceeds in the oxygenated form of Ta-VII, also, and this would certainly contribute to the protection of Ta-VII HbO 2 from an accelerated proton-catalysis in the acidic autoxidation. For detecting the swinging movement of the distal histidine residue, the Soret absorption spectro- scopy is silent in the oxygenated form of Mb or Hb. Therefore, another approach will be needed for more detailed kinetics of the autoxidation reaction and histidine swinging in Ta-VII HbO 2 , and this remains open to our future study. Discussion Physiological properties of T. akamusi hemoglobins Some insects have a globin in their fast-growing larval stage, but lose it after metamorphosis in favor of the diffusion of gaseous oxygen through hollow tracheal tubes. This is the case in the midge (Diptera, Chironomidae), and the Chironomidae is one of the largest insect families. Among the midge hemoglobins, extensive work has been carried out with several species of Chironominae, such as Chironomus thummi thummi [16–19], and Chironomus thummi piger [20,21].Indeed,theC. thummi thummi (CTT) Hb-III was the first invertebrate Hb whose X-ray structure was determined at high resolution. In its crystal structure, displaying the common globin fold, the heme group is rotated by 180° and the heme cavity in the deoxy form has an unusual open gate conformation at pH 7.0, with the distal His able to swing out of the cavity [16,17]. This Hb has therefore been the subject of structural, spectral and functional studies [18,19]. On the other hand, T. akamusi, a common species found in Japan, belongs to a different subfamily (Orthocladiinae) from Chironominae, and its larva is quite unique in morphology and ecological behavior. In the Chironomid group, the young hatch from the colorless, transparent egg as wormlike larvae. The larva grows through four instars (stages separated by a molt) without change of shape. As for T. akamusi, the Japanese word ÔakamusiÕ means blood- worm, which comes from the fact that a large amount of hemoglobin is synthesized into the hemolymph of the 4th- instar larva. This small bloodworm (15–18 mm in length and 1.5 mm in diameter) begins to burrow into polluted and extremely hypoxic mud flats of lakes to have a long period (more than half a year) of diapause. The burrow can reach up to 80 cm in depth, and e 0 ¢ ¼ 0 V in the oxidation- reduction potential. After diapause, the matured and sex- differentiated larva crawls up above the ground again, and undergoes a pupal molt in which the shape alters com- pletely. The brown pupa is encased in a cuticle, and the Ó FEBS 2003 Swinging of the distal histidine in midge Hb (Eur. J. Biochem. 270) 1431 pupal stage terminates with a final or imaginal molt in which the adult, winged midge emerges from the pupal case. The life of the adult midge is restricted only to a one-month period or so. A stage-specific expression of T. akamusi hemoglobin appears to be adaptive for the bloodworm to extend its inhabitable environment. By burrowing deeply into lake mud flats, the bloodworm can protect itself from fish, and thus have a fairly long period of diapause in safety. In light of such unique behavior of the 4th-instar larva of T. aka- musi, it was of particular interest to investigate the molecular properties of the hemoglobin components isolated from the larval hemolymph. The polymorphic forms of T. akamusi hemoglobin would certainly be advantageous to the larval life in O 2 transport and storage under the particular adverse conditions. Molecular properties of T. akamusi hemoglobins The hemoglobin from Tokunagayusurika akamusi consists of at least 11 components, which fall into two approxi- mately equal groups; one (VIA, VIB, VII, VIII and IX) having a distal histidine and the other (IA, IB, II, III, IV and V) lacking in it [1]. We have therefore carried out a matrix analysis to test the sequence homology of the major components V and VII from T. akamusi, with other midge hemoglobins including 10 components from C. thummi thummi [16,18] and three components from C. thummi piger [20,21]. As a result, component VII has a higher percentage identity (40–48%) with the Chironomus Hb sequences rather than does component V (26–27%). Because all the Chiro- nomus hemoglobins have a distal histidine residue, the appearance of component V-type globins may be very specific to the genus Tokunagayusurika. For the distal histidine swinging, Johnson et al.[22] observed it in the structure of myoglobin-ethylisocyanide. They described that when such a bulky ligand bound to ferric sperm whale myoglobin, the distal (E7) histidine swung up and away from the iron atom, just like a swinging door, toward the protein surface. In sperm whale myoglobin, Tian et al.[23]measuredtheonand off rate-constants for O 2 -binding as a function of pH, and reported a dramatic increase in the O 2 -dissociation rate at low pH, where the imidazole side chain of the E7- His becomes protonated, loses a hydrogen bond to the bound O 2 , and moves outward on a microsecond timescale. In T. akamusi Hb VII, a similar movement of the distal histidine could occur in the ferric met-form but on a millisecond timescale. Consequently, this pH- dependent swinging is quite different from the distal His movement controlling the on and off rate processes of O 2 -binding in myoglobins and hemoglobins. In oxygen equilibrium measurements, the most remark- able result is that O 2 affinities of the two components are almost the same at pH 8.4 and the E7-Ile hemoglobin V has rather a higher affinity at lower pH. In all the recombinant myoglobins reported so far, a His to Ile or Leu mutation at the E7 position always causes a dramatic decrease in O 2 affinity, resulting in a very large increase (> 10 Torr) in P 50 value [24]. In this respect, it is possible that the B10-Phe at position 32 is stabilizing the bound O 2 in hemoglobin V (Fig. 1). Among the distal heme pocket residues, the B10 is known to be very relevant for the O 2 -binding property, in addition to the E7 residue [25]. In the autoxidation reaction, Brantley et al.[26]werethe first to use site-directed mutagenesis of sperm whale myoglobin to make clear the possible role(s) of the distal histidine residue. They showed that mutations of the distal His at position 64, such as those of His64fiGly, His64fiVal, His64fiLeu and His64fiGln, caused dramatic increases in the autoxidation rate, but the relative effects of pH were the same with that for the wild-type (His64) myoglobin if the absolute rates were normalized to pH 7.0. However, our examinations for a dozen naturally occurring myoglobins have shown that only the proteins having the usual distal histidine can manifest a very strong proton- catalysis in the autoxidation reaction [3,7]. Using some typical His64 mutants of sperm whale myoglobin, we have therefore carried out detailed pH-dependence studies of the autoxidation rate over the wide range of pH 5–12 in 0.1 M buffer at 25 °C. The resulting pH profiles were then compared with those of the corresponding myoglobins occurring in nature [3]. As a result, sperm whale MbO 2 (wild-type) is approximately 400 times more resistant to autoxidation if compared with the His64fiVal and His64fiGly mutant proteins at pH 7.0. Such a comparison certainly leads us to conclude that the distal (E7) histidine inhibits heme oxidation by obstructing easy access of a water molecule to the FeO 2 center. This is true in a neutral pH range. Nevertheless, it is also true that the rate of autoxidation of sperm whale MbO 2 increases markedly, not only with increasing hydrogen ion concentration but also with increasing hydroxyl ion concentration, as shown in Fig. 8. Consequently, we have proposed that the distal histidine can play a dual role in the nucleophilic displacement of O 2 – from MbO 2 or HbO 2 . One is in a proton-relay mechanism via its imidazole ring. Insofar as we have examined for more than a dozen myoglobins, such a proton-catalyzed process could never be observed in the autoxidation reaction of myoglobins lacking the usual distal histidine residue, no matter what the protein is, the naturally occurring or the distal His mutant [3]. As a matter of fact, even if the distal residue is a histidine, the protein cannot manifest any proton catalysis if the residue is tilted away from the precise E7 position. We have found this to be the case here, for the autoxidation of T. akamusi hemoglobin VII below pH 6.0, as well as for the b-chain in the autoxidation of the human HbO 2 tetramer [11]. The other role of the distal histidine is in the maximum protection of the FeO 2 center against a water molecule or a hydroxyl ion that can enter the heme pocket from the surrounding solvent [26]. In this way, the distal histidine provides the delicate balance of catalytic and steric factors necessary for controlling the reversible oxygen binding to myoglobin and hemoglobin in protic, aqueous solvent. In light of the oxygen equilibrium parameters and considerable resistance to acidic autoxidation, we conclude that T. akamusi hemoglobin VII, the major component, can play an important role in O 2 transport and storage against the extremely acidic and hypoxic adversity. For the pH of the larval hemolymph, there is no report of its direct measurement, but there is a strong possibility that it drops to around pH 4. Under anaerobic conditions, the end-products 1432 S. Kamimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 of carbohydrate metabolism by the Chironomus larva are known to involve large amounts of lactic and succinic acids [27]. Multiplicity of other components in the O 2 -binding affinity, Bohr effect, and FeO 2 stability would also be advantageous for the insect larva to overcome various stringent circumstances. Acknowledgements This work was partly supported by grants-in-aid for Scientific Research (07640896, 10440248 and 12640659) from the Ministry of Education, Culture and Science of Japan. References 1. Fukuda, M., Takagi, T. & Shikama, K. (1993) Polymorphic hemoglobin from a midge larva (Tokunagayusurika akamusi)can be divided into two different types. Biochim. Biophys. Acta 1157, 185–191. 2. Akiyama, K., Fukuda, M., Kobayashi, N., Matsuoka, A. & Shikama, K. (1994) The pH-dependent swinging-out of the distal histidine residue in ferric hemoglobin of a midge larva (Tokuna- gayusurika akamusi). Biochim. Biophys. Acta 1208, 306–309. 3. Suzuki,T.,Watanabe,Y H.,Nagasawa,M.,Matsuoka,A.& Shikama, K. (2000) Dual nature of the distal histidine residue in the autoxidation reaction of myoglobin and hemoglobin: Com- parison of the H64 mutants. Eur. J. Biochem. 267, 6166–6174. 4. Riggs, A. (1981) Preparation of vertebrate blood hemoglobins. Methods Enzymol. 76, 14–18. 5. Imai, K. (1981) Measurement of accurate oxygen equilibrium curves by an automatic oxygenation apparatus. Methods Enzymol. 76, 438–449. 6. Shikama, K. & Matsuoka, A. (1989) Spectral properties unique to the myoglobins lacking the usual distal histidine residue. J. Mol. Biol. 209, 489–491. 7. Shikama, K. & Matsuoka, A. (1994) Aplysia myoglobin with unusual properties: Another prototype in myoglobin and hae- moglobin biochemistry. Biol.Rev.(Cambridge)69, 233–251. 8. Matsuoka, A., Kobayashi, N. & Shikama, K. (1992) The Soret magnetic circular dichroism of ferric high-spin myoglobins. A probe for the distal histidine residue. Eur. J. Biochem. 210,337– 341. 9. Gotoh, T. & Shikama, K. (1976) Generation of the superoxide radical during autoxidation of oxymyoglobin. J. Biochem. (Tokyo) 80, 397–399. 10. Shikama, K. (1998) The molecular mechanism of autoxidation for myoglobin and hemoglobin: a venerable puzzle. Chem. Rev. 98, 1357–1373. 11. Tsuruga, M., Matsuoka, A., Hachimori, A., Sugawara, Y. & Shikama, K. (1998) The molecular mechanism of autoxidation for human oxyhemoglobin: Tilting of the distal histidine causes nonequivalent oxidation in the b chain. J. Biol. Chem. 273, 8607– 8615. 12. Yasuda, J P., Ichikawa, T., Tsuruga, M., Matsuoka, A., Sugawara, Y. & Shikama, K. (2002) The a1b1 contact of human hemoglobin plays a key role in stabilizing the bound dioxygen: Further evidence from the iron valency hybrids. Eur. J. Biochem. 269, 202–211. 13. Shikama, K. & Matsuoka, A. (1986) Aplysia oxymyoglobin with an unusual stability property: Kinetic analysis of the pH depen- dence. Biochemistry 25, 3898–3903. 14. Shikama, K. (1988) Stability properties of dioxygen-iron(II) porphyrins: An overview from simple complexes to myoglobin. Coord. Chem. Rev. 83, 73–91. 15. Phillips, S.E.V. & Schoenborn, B.P. (1981) Neutron diffraction reveals oxygen-histidine hydrogen bond in oxymyoglobin. Nature (London) 292, 81–82. 16. Huber, R., Epp, O., Steigemann, W. & Formanek, H. (1971) The atomic structure of erythrocruorin in the light of the chemical sequence and its comparison with myoglobin. Eur. J. Biochem. 19, 42–50. 17. Steigemann, W. & Weber, E. (1979) Structure of erythrocruorin in different ligand states refined at 1.4 A ˚ resolution. J. Mol. Biol. 127, 309–338. 18. Weber, R.E., Braunitzer, G. & Kleinschmidt, T. (1985) Functional multiplicity and structural correlations in the hemoglobin system of larvae of Chironomus thummi thummi (Insect, Diptera): Hb components CTT I, CTT II, CTT III, CTT IV, CTT VI, CTT VIIB, CTT IX and CTT X. Comp. Biochem. Physiol. 80B, 747– 753. 19. Di Iori, E.E., Tavernelli, I.W. & Yu, W. (1997) Dynamic prop- erties of monomeric insect erythrocruorin III from Chironomus thummi thummi: Relationships between structural flexibility and functional complexity. Biophys. J. 73, 2742–2751. 20. Hankeln, T., Rozynek, P. & Schmidt, E.R. (1988) The nucleotide sequence and in situ localization of a gene for a dimeric hae- moglobin from the midge Chironomus thummi piger. Gene 64,297– 304. 21. Rozynek, P., Hankeln, T. & Schmidt, E.R. (1988) Structure of a hemoglobin gene cluster and nucleotide sequence of three hemoglobin genes from the midge Chironomus thummi piger (Diptera, Insecta). Biol. Chem. Hoppe-Seyler 370, 533–542. 22. Johnson, K.A., Olson, J.S. & Phillips, G.N. Jr (1989) Structure of myoglobin-ethylisocyanide: histidine as a swinging door for ligand entry. J. Mol. Biol. 207, 459–463. 23. Tian, W.D., Sage, J.T. & Champion, P.M. (1993) Investigations of ligand association and dissociation rates in the ÔopenÕ and ÔclosedÕ states of myoglobin. J. Mol. Biol. 233, 155–166. 24. Springer, B.A., Sligar, S.G., Olson, J.S. & Phillips, G.N. Jr (1994) Mechanisms of ligand recognition in myoglobin. Chem. Rev. 94, 699–714. 25. Zhao, X., Vyas, K., Nguyen, B.D., Rajarathnam, K., La Mar, G.N., Li, T., Phillips, G.N. Jr, Eich, R.F., Olson, J.S., Ling, J. & Bocian, D.F. (1995) A double mutant of sperm whale myoglobin mimics the structure and function of elephant myoglobin. J. Biol. Chem. 270, 20763–20774. 26. Brantley, R.E. Jr, Smerdon, S.J., Wilkinson, A.J., Singleton, E.W. & Olson, J.S. (1993) The mechanism of autooxidation of myo- globin. J. Biol. Chem. 268, 6995–7010. 27. Wilps, H. & Zebe, E. (1976) The end-products of anaerobic car- bohydrate metabolism in the larvae of Chironomus thummi thummi. J. Comp. Physiol. 112, 263–272. Ó FEBS 2003 Swinging of the distal histidine in midge Hb (Eur. J. Biochem. 270) 1433 . The swinging movement of the distal histidine residue and the autoxidation reaction for midge larval hemoglobins Satoshi Kamimura 1 ,. in the oxygenated form of Mb or Hb. Therefore, another approach will be needed for more detailed kinetics of the autoxidation reaction and histidine swinging