Biochimica et Biophysica Acta, 1203 (1993) 267-275 267 © 1993 Elsevier Science Publishers B.V. All rights reserved 0167-4838/93/$06.00

BBAPRO 34625

Dynamics and thermodynamics of acid-alkaline transitions in metmyoglobins lacking the usual distal histidine residue

Yasuhiko Y a m a m o t o a,,, Tomohiko Suziki h and Hiroshi Hori c

a Department of Biomolecular Engineering, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 227 (Japan), b Department of Biology, Faculty of Science, Kochi University, Kochi (Japan) and c Department of Biophysical Engineering,

Faculty of Engineering Science, Osaka University, Osaka (Japan)

(Received 5 May 1993)

Key words: Nuclear magnetic resonance spectroscopy; Electron paramagnetic resonance spectroscopy; Myoglobin; Acid-alkaline transition; Distal residue

The kinetics of the acid-alkaline transition in the ferric myoglobins from the gastropodic mollusc Dolabella auricularia and the shark Mustelusjaponicus, which possess the distal Val E7 and Gin E7, respectively, has been investigated using the paramagnetic 1H-NMR saturation transfer measurements in order to gain insight into functional properties of these non-His distal residues. Both myoglobins possess the penta-coordinated heme below the pK of the transition (7.8 and 10.0 for Dolabella and Mustelus myoglobins, respectively) and bind OH- above the pK. The pH dependence of the transition rates and the relatively high activation barrier (58 + 9 kJ/mol) for the dissociation of the Fe-bound OH- in Dolabella myoglobin indicate a strong interaction between the bound ligand and the guanidino NH proton of the mrg El0 in Dolabella myoglobin. Such a strong interaction between Fe-bound OH- and the Arg El0 side-chain in Dolabella myoglobin is also manifested in the EPR spectra. For Mustelus myoglobin, the pH and temperature dependence studies on the kinetics strongly suggest that the distal Gin E7 in this myoglobin does not contribute significantly to stabilize the Fe-bound ligand.

Introduction

The acid-alkaline transition in ferric myoglobin (Mb) has been of particular importance in understand- ing the structure-function relationship of Mb because its reflects characteristics of both structural and ligand binding properties of the heme active site [1-4]. Upon the transition from acidic form to alkaline form in the ferric state of mammalian Mb, the Fe-bound ligand changes from H 2 0 to O H - , with concomitant change of the spin state from S = 5 / 2 to S = 1 /2 [5]. The rapid transition in mammalian Mb has been inter- preted in terms of p ro tona t ion /depro tona t ion process of the distal His E7 imidazole bound to the Fe-bound ligand via a hydrogen bond [6]. On the other hand, the Mbs from the sea hare Aplysia limacina [7,8] and the shark Galeorhinus japonicus [9] exhiigit a relatively slow transition, which has been attributed to the absence of bound H 2 0 in the acidic form. Hence the acid-al-

* Corresponding author. Fax: +81 45 9217794. Abbreviations: Mb, myoglobin; metMb, met myogiobin; metMbOH-, hydroxymet myoglobin; ppm, parts per million; NOE, nuclear Over- hauser effect; Me, methyl group.

kaline transition in these Mbs leads to simultaneous changes in both the coordination and the spin states of the heme iron.

The thermodynamics of the acid-alkaline transition in ferric hemoproteins has been analyzed in detail using a variety of physicochemical methods, including 1H-NMR [1,3], optical spectroscopy [2,10,11], EPR [12], and magnetic susceptibility measurements [10]. But the kinetics of the transition has not been fully discussed in conjunction with molecular structure of the active site in Mb at present because of the lack of a methodology to yield quantitative kinetic parameters, although the temperature-jump method [7] and the analyses of I H- NMR parameters [3,8,9,13] have shown to provide some qualitative kinetics. We have recently demonstrated that the paramagnetic ~H-NMR saturation transfer experiments allow quantitative characterization of the kinetics of the acid-alkaline transition in Mb, if the transition rate is comparable to the paramagnetic re- laxation rate of the heme peripheral side-chain proton of ferric Mb [14].

The Mb from the triturative stomach of Dolabella auricularia [15], a gastropodic mollusc belonging to the Aplysiidae, exhibits high sequence homology (about 70%) with the more commonly characterized mollusc

268

Mb of Aplysia limacina [7,8,13,16-22] in addition to the presence of the distal Val E7. The mollusc Mb exhibits relatively high 0 2 affinity, compared with the Mb possessing the distal Val E7 [23]. Recent studies have revealed that the guanidino NH proton of the Arg El0 in the mollusc Mb is hydrogen-bonded to the bound O z and hence the O z affinity of the mollusc Mb is controlled by a molecular mechanism completely different from that of mammalian Mb [20,24,25]. We call the Arg El0 in mollusc Mb the 'pseudo-distal' residue. Kinetic study of the acid-alkaline transition in mollusc metMb would provide further insight into the functional properties of Arg El0 in this Mb under the simultaneous change of the coordination and the spin states of the heme iron.

The functional roles of the distal Gln E7 in Mb have received a considerable interest, because, considering similarity in both the molecular size and ability to form the hydrogen bond between the side-chains of Gin and His residues, the distal Gin E7 is expected to play similar roles to His E7 in mammalian Mb [9,26-29]. The Mb from the shark Mustelus japonicus possesses Gin E7 (Suzuki, T., unpublished results) as Galeorhi- nus japonicus [30] and Asian elephant Mbs [31]. Only the D- and E-helices of Mustelus japonicus Mb have been sequenced at present and the comparison of the amino-acid sequence between Mustelus japonicus and Galeorhinus japonicus Mbs revealed a single replace- ment at El3, Asn for the former and Arg for the latter (Suzuki, T., unpublished results). Since El3 residue is at least 1.1 nm away from the sixth coordination site of the heme iron, it is unlikely that the replacement of this residue alters the functional properties of Mb significantly. Furthermore, the sequence homology be- tween the Mbs from the sharks Mustelus antarcticus and Galeorhinus japonicus is about 90% [32] and there- fore Mustelus japonicus Mb is expected to possess similarly high homology with the latter Mb.

In the present paper, we report the results of 1H- NMR and EPR studies of the acid-alkaline transition in Dolabella and Mustelus Mbs, which not only re- vealed the ligation state of the heine iron in both acidic and alkaline forms of the Mbs, but also provided the kinetics and thermodynamics of the transition. The interaction of Fe-bound ligand with non-His distal residues in the present Mbs has been inferred from the obtained data.

Materials and Methods

Sample preparation. The oxy-form of Dolabella Mb was extracted from its triturative stomach and purified according to the method previously described [15]. Mustelus Mb was isolated from red muscle and purified

as previously reported [33]. The Mb samples were oxidized by the addition of 5-fold molar excess of potassium ferricyanide (Sigma Chemical). metMb was separated from the residual reagents with a Sephadex G-50 (Sigma Chemical) column equilibrated with 10 mM Bis-Tris buffer (Sigma Chemical) (pH 6.8). Equine Mb was purchased from Sigma Chemical and used without further purification. Mb solution was concen- trated to ~ 1 mM and then solvent was exchanged to 2H20 in an Amicon ultrafiltration cell. pZH of the sample was adjusted using 0.2 M NaOZH or 2HCI and the pZH was measured using a Toko model TP-10 pH-meter with a Toko type CE103C electrode. The isotope effect was not considered to correct the pZH value.

NMR spectroscopy. ~H-NMR spectra were recorded using a JEOL GSX-500 FT-NMR spectrometer operat- ing at a ~H frequency of 500 MHz. A typical spectrum consisted of 3000 transients with 8 Kbyte data points over 60 kHz band width, and a 6.3 ~s 90 ° pulse. The residual water resonance was suppressed with 50 ms presaturation decoupler pulse. Intrinsic spin-lattice re- laxation time (T~ intr) was measured using the satura- tion-recovery method with a selective saturation pulse. Saturation-transfer experiments were carried out by selectively saturating a desired peak for a variety of time and the steady-state value of the saturation trans- fer factor (I/Io; I and I 0 are the signal intensities of a peak A without and with the saturation of a peak B which is connected to peak A by dynamic process, respectively) was achieved for the saturation time > 50 ms. The spectra that resulted from the saturation transfer experiments were presented in the form of difference spectra. The saturation- transfer factor was calculated by integrating the peak area. The signal-to- noise of the spectra was improved by apodization which introduced a 50 Hz line broadening. The chemical shifts are given relative to sodium 2,2-dimethyl-2-sila- pentane-5-sulphonate with the residual HzHO as in- ternal reference.

EPR spectroscopy. EPR spectra were recorded at 15 K on a home-built EPR spectrometer using a Varian X-band cavity operating at 9.23 GHz. The temperature was controlled using an Oxford flow cryostat (ESR-900). Experiments were carried out at an incident microwave power of 5 mW with a field modulation of 1 mT at 100 kHz. Microwave frequency was calibrated with a mi- crowave frequency counter (Takada Riken, Model TR5212). The magnetic field strength was determined by a nuclear magnetic resonance of protons in water. Accuracy of the g values was within + 0.01. No buffer was used to prepare EPR samples in order to avoid the 'buffer effect' in the spectra, recently reported by Ikeda-Saito et al. [34] and the pH was adjusted by 0.2 N NaOH or 0.2 N HC1.

Determination of the kinetic parameters for acid-al- kaline transition via saturation transfer factor. The acid-alkaline transition in Mbs used in the present study is expressed as:

k+ metMb + O H - ~ > m e t M b O H -

k_

where k÷and k_ are the association and dissociation rates, respectively. In the saturation transfer experi- ments, the pair of spectra recorded without and with the saturation of a heme methyl proton resonance of metMbOH- yields I 0 and I for the signal intensity of the corresponding heme methyl proton resonance of metMb, respectively. The quantity I/Io, called the saturation factor, is related to k_ and Z~ ntr of the heme methyl proton resonance of metMb by the for- mula [35]:

k_ = (T~m~) - l ( 1 - I l I o ) l ( I l I o ) . (1)

The k÷is related to the equilibrium constant of the reaction, K, by the following equation:

K = k + / k _ = [me tMbOH- ] / [ m e t M b ] [ O H - ], (2)

where [metMbOH-], [metMb] and [OH-] are the con- centrations of metMbOH-, metMb and OH-, respec- tively. Therefore, k+can be calculated from k_, [OH-] and [metMbOH-]/[metMb], which is determined from the intensity of the heme methyl proton resonances arising from metMbOH- and metMb.

Results

1H-NMR spectra of metMbs The hyperfine-shifted portions of the 500 MHz 1H-

NMR spectra of Dolabella and Mustelus metMbs at 30°C and pH 7.0 are illustrated in the traces A and B of Fig. 1, respectively, and are compared with that of equine metMb (trace C). Four heme methyl proton resonances, peaks Mi-M4, and six single-proton reso- nances, peaks A1-A6, are clearly resolved below 30 ppm in the traces A an B. The ~H-NMR spectrum of Dolabella metMb is similar to that of Aplysia metMb [14]. The downfield hyperfine shifted resonances in trace A have been assigned from the observation of nuclear Overhauser effect (NOE) connectivities (re- suits not shown) and from the signal assignments of Aplysia metMb [8]. Furthermore, the spectrum of Mustelus metMb resembles that of the previously re- ported shark Galeorhinus metMb [9] and the resolved heme peripheral proton resonances in trace B were also assigned on the basis of NOE connectivities (re- suits not shown) and from the signal assignments of Galeorhinus metMb [9]. From its line width and shift,

269

3-Me 1-Me~ I - - 1 8-Me 5-Me | 14.ct '

6 -o~' , I 7-c~ meso 7 -(t

C . . -

,-, !IA, II A, IIA A,A, 1/ ':> StXty

..A__ ..... ' A, A~ I

12o too 80 6o 40 20 o -2o -4o

Fig. 1. Hyperfine-shifted portions of the 500 MHz tH-NMR spectra of Dolabella (A), Mustelus (B), and equine (C) metMbs in 2H20 at 30°C and pH 7. Peaks labeled by (* ) arise from the heme disorder [36] and those labeled by (e) are due to impurities. Heme methyl proton signals are labeled by M i and peaks Ai are single-proton

resonances.

upfield-shifted broad peaks at about -25 ppm in the traces A and B are assignable to meso-proton and peak A 7 may be attributed to the fl-CH 2 proton of the proximal His F8. The shifts of the assigned heme peripheral side-chain proton resonances are summa- rized in Table I. Peaks with less than single-proton intensity, marked by an asterisk in Fig. 1, are attributed to the reversely oriented heme arising from the heme orientational disorder [36].

There are three characteristic spectral differences between Dolabella or Mustelus and equine metMbs, i.e., (1) the line width of the resonances of the former

TABLE I

Assignment and chemical shifts in DolabeUa, Mustelus, and equine metMbs in 21420 at 30°C (pH 7)

Signal Dolabella Mustelus Equine b

8-Me 96.6 ( - 1.2) a 97.9 ( - 2.8) 90.6 5-Me 88.7 (3.1) 87.4 ( - 7 . 7 ) 84.0 3-Me 76.0 ( - 1.6) 74.0 ( - 7.9) 71.2 1-Me 55.4 (11.3) 54.6 (6.3) 51.5 2-a 39.9 (11.0) 34.5 (17.9) 31.4 4-a 51.5 (0.1) 45.2 (12.8) 45.2 6-a 53.8 (11.4) 59.9 ( - 4 . 2 ) 57.9 6-a' 48.0 (6.1) 39.3 (4.3) 45.2 7-a 61.9 (33.1) 81.3 (23.3) 74.0 7-a ' 45.1 ( - 16.8) 40.5 ( - 17.4) 31.4

a The numbers in parentheses are the shifts to T - I ._, 0. b Obtained from [9].

270

metMbs is narrower than that of the latter, (2) the average heme methyl proton shifts of Dolabella metMb (79.2 ppm) and Mustelus metMb (78.5 ppm) are more than 4 ppm larger than that of equine metMb (74.3 ppm) at 30°C, and (3) the meso-proton signals of the former proteins are observed in the upfield-shifted hyperfine region, while that of the latter resonates in the downfield region. These spectral characteristics are consistent with those for the penta-coordinated heme in metMb [37].

Furthermore, in the optical absorption spectra, a slight hypsochromic and a large hypsochromic effect on the Soret band, i.e., 402 nm and 97.5 m M - ] c m -1 for Dolabella metMb [29] and 396 nm and 98.5 m M - l c m - 1 for Mustelus metMb (Suzuki, T., unpublished results), compared with those of equine metMb, 408 nm and 188 m M - l c m - 1 [11], also support the absence of bound H 2 0 in the ferric high-spin form of these Mbs [16].

ply dependence of 1H-NMR spectra The pH dependence of the 500 MHz 1H-NMR

spectra of DolabeUa and Mustelus metMbs at 30°C is shown in Fig. 2. In both metMbs, the resonances arising from the alkaline form appear in the chemical shift region 20-40 ppm at this temperature and the spectral pattern is highly similar to that of the hydrox- ymet form ( m e t M b O H - ) of mammalian Mb [3,38]. With increasing pH, the signal intensity for the reso- nances of alkaline form increases at the expense of the signals of the acidic form. The fact that the two sets of the signals from both the acidic and alkaline forms are separately observed indicates that the acid-alkaline transition in these metMbs is slower compared with the NMR time scale. These results are in contrast to the rapid acid-alkaline transition in mammalian Mb [3,38].

8.86

' ~ o . . . . ~ . . . . ~ . . . . - ~ ' ' ' ~6o . . . . ~ . . . . ~ . . . . -5o' '

A B Fig. 2. pH dependence of the hyperfine-shifted portions of the 500 MHz 1H-NMR spectra of DolabeUa (A) and Mustelus (B) metMbs in 2H20 at 30°C. With increasing pH, the signals, at 20-40 ppm, arising from the alkaline form gain intensity at the expense of the

signals of the acidic form.

L,S,

xl/15 ]

I 1 H.S. 10.6

~ 8.9 _ _

7.8 ~

x112

g=6 26o

A x1/15 ] l 6.0

g=6 g~=2 0 200 4()0 (mT)

A

L,S,

10.0

8.6

g=2 4()0 (mT)

B Fig. 3. EPR spectra of Dolabella metMb (A) and Mustelus metMb (B) at the indicated pH. In the spectra of Dolabella metMb, low-spin species I(g~ = 2.79, gz = 2.12, and g3 = 1.75) and II (gt = 2.68, g2 = 2.09, and g3 = 1.79) are observed. A rhombic high-spin signal (gl = 6.87 and g2 = 5.11, H.S. in the spectra) and a signal of g = 3.37 are also detected at higher pH. For Mustelus metMb, the low-spin signal (gl = 2.54, g2 = 2.16, and g3 = 1.86) exhibits a typical g values for

metMbOH-.

The p K values of 7.8 and 10.0 for the transition in Dolabella and Mustelus metMbs, respectively, were ob- tained from the pH dependence of the signal intensity for the resonances arising from the two forms and are also consistent with the values determined by optical spectroscopy (Suzuki, T., unpublished results). These values are almost equivalent to those of the corre- sponding analogous metMbs, i.e., Aplysia metMb (pK = 7.5) [7] and Galeorhinus metMb (pK = 10.0) [9].

EPR spectra The EPR spectra of Dolabella metMb at various pH

values are shown in Fig. 3(A). At pH 6.0, this Mb exhibits an axial ferric high-spin EPR spectrum with g ± = 6.0 and gtt = 2.0 and an additional signal at g = 5.6, which disappears above pH 6.8. The g = 5.6 signal may arise from an admixture of S = 5 / 2 and 3 / 2 states as Ikeda-Saito et al. [34] reported. With increasing pH, ferric low-spin species appear at the expense of the high-spin species. The pH dependence of the g j_ = 6.0 signal intensity essentially parallels to that of the high- spin signal intensity observed in the NMR spectra. There exist two low-spin species: species I (gl = 2.79, g2=2.12, and g3 = 1.75) and species II (gl =2.68, g2 = 2.09, and g3 = 1.79), with a fractional ratio of 1 : 1. Since the ligation of O H - to Aplysia metMb has shown to exhibit the ferric low-spin EPR spectrum with gl = 2.67, g2 = 2.15, and g3 = 1.77 [39], both species can be identified as me tMbOH- , although their g values are similar to those for the heme-iron coordi- nated to an imino nitrogen. At pH > 10, a new rhombic high-spin species with low signal intensity (gl = 6.87

and g2 = 5.11) and other component with the signal of g = 3.37 are observed in the spectra. The origin of these signals is not clear at the moment .

The EPR spectra of Mustelus metMb are illustrated in Fig. 3(B). The acidic form exhibits a typical axial high-spin spectrum and the alkaline form shows a low-spin spectrum with gl = 2.54, g2 = 2.16, and g3 =

1.86. These g values are similar to those of sperm whale m e t M b O H - ( g l = 2.53, g2 = 2.17, and g3 = 1.86) [39]. Therefore, the alkaline form of Mustelus metMb is identified as m e t M b O H - .

Heme methyl proton signal assignment o f Mustelus m e t M b O H - via saturation transfer

The saturation-transfer difference spectra resulted from the saturation of individual heme methyl proton signal of Mustelus m e t M b O H - (pH 10.2), at 40°C are

A

100 80 60 40 20

Fig. 4. saturation-transfer difference spectra of Mustelus metMb in 2H20 at 40~C and pH 10.2. (A) Reference spectrum. (B) Saturation of the 8-Me signal of the acidic form exhibits the saturation transfer to peak a of the alkaline form. (C) Saturation of the 5-Me signal of the acidic form exhibits the saturation transfer to peak b of the alkaline form. (D) Saturation of the 3-Me signal of the acidic form exhibits the saturation transfer to peak c of the alkaline form. (E) Saturation of the 1-Me signal of the acidic form exhibits the satura- tion transfer to peak d of the alkaline form. The peak indicated by

an arrow is saturated.

271

TABLE II

Assignments and chemical shifts of heme methyl proton resonances in Dolabella (pH 9.3), Mustelus, pH 11.1, and equine, MbOH -s at 30°C (pn 12.4)

Signal Dolabella Mustelus Equine

8-Me 36.9 (20.1) a 36.9 (18.8) 36.9 5-Me 37.8 (14.6) 36.9 (18.8) 36.9 3-Me 32.4 (22.1) 34.1 (23.2) 32.8 1-M 26.5 (13.6) 26.4 (17.6) 25.8

a The numbers in parentheses are the shifts to T- 1 ~ 0 .

shown in Fig. 4. The saturation of the 8-Me signal leads to partial saturation of peak a (trace B). The traces C - E exhibit the saturation-transfer connectivi- ties between the 5-Me signal and peak b (trace C), between the 3-Me signal and peak c (trace D), and between the 1-Me signal and peak d (trace E). Al- though the 2-a proton signal (peak A 6 in the trace 1B) resonates under peak c, no N O E is expected between 2-a and 3-Me. Therefore, peaks a ~ d are assigned to the heme 8-, 5-, 3-, and 1-Me proton resonances of Mustelus m e t M b O H - , respectively. The shifts of the heme methyl proton resonances of Dolabella and Mustelus m e t M b O H - s , together with that of equine m e t M b O H - , are listed in Table II. In contrast with the case for the acidic form, the average heme methyl proton shift (about 33 ppm) for the alkaline form is almost independent of the proteins.

Temperature dependence o f the spectra The shift to T-1 ~ 0 for the Curie plot, shift vs. the

reciprocal absolute temperature, is given in Table I. These shifts are not close to the values for the corre- sponding resonances of the diamagnetic compound [40]. Large deviation of the shift to T-1 ~ 0 from the dia- magnetic shift for the heme vinyl and propionate a- proton resonances reflects the tempera ture depen- dence of their conformation with respect to the heme plane, which influences the contact-shift contribution [411.

It has been shown that, at ambient temperature, m e t M b O H - exhibits a thermal equilibrium between the high-spin, S = 5 /2 , and low-spin, S = 1/2, states, due to the intermediate field strength of the O H - ligand [42-44]. The shift to T-1 ._. 0, determined from the Curie plot, in Table II exhibits a large deviation from the diamagnetic shift of 3.1 ppm for the heme methyl proton resonance [40], clearly indicating the contribution of other thermally accessible spin states to the electronic structure of the heme iron.

Determination of acid-alkaline transition rate via satu- ration transfer measurements

The rate of the acid-alkal ine transition of these Mbs have been determined using the paramagnetic

272

o 0 . 9 , \

0 . . . . 50 . . . . 100 . . . . 150' ' ' T,. . / m s

B

A Fig. 5. (A) The effect of the saturation time (TIR R) on the saturation transfer factor (I/I0). The 1-Me resonance at 26.1 ppm of the alkaline form in Mustelus Mb at 40°C and pH 10.2 was saturated for different TmRs to observe the saturation transfer to the 1-Me resonance at 53.8 ppm of the acidic form. (B) Plot of saturation transfer factor ( I / I 0) against the saturation time (TmR). The

steady-state I / I 0 is observed with TmR > 50 ms.

1H-NMR saturation transfer method [14]. The T~ ntr

values of the 3-Me resonance of Dolabella metMb at pH 6.3 are 3.0, 3.3, and 3.5 ms at 21, 30, and 35°C, respectively, and the values for the 1-Me resonance of Mustelus metMb at pH 8.0 are 6.7, 6.8, and 7.2 ms at 30, 35 and 40°C, respectively, with experimental errors of about __+ 10%. The absence of the alkaline form in the Mb samples at the pH value used in the measure- ments ensures that the determined T~ntrvalue is free from the effects of the acid-alkaline transition. The T~ ntr values of these metMbs are larger than non-selec- tive T~ value of about 2 ms for the hexa-coordinated heme in metMb [45].

The saturation-transfer difference spectra recorded with the saturation of the 1-Me signal of Mustelus metMbOH- for a variety of time are illustrated in Fig. 5(A) and the saturation factor (I/Io) is plotted against the saturation time (TmR) in Fig. 5(B). The plot shows that the steady-state I / I o value is observed with TIR R > 50 ms.

According to Eqns. 1 and 2, the kinetic parameters, k+and k , for the two metMbs at three different pH values and temperatures were determined and the results are summarized in Table III. The k_ of Mustelus metMb exhibits some pH dependence (0.27 pH at 30°C), while that of Dolabella metMbs is almost i n d e p e n d e n t of p H .

T h e ac t i va t i on e n e r g y ( A E ) fo r t h e d i s soc i a t i on o f

t h e b o u n d O H - l igand in Dolabella a n d Mustelus m e t M b s is o b t a i n e d f r o m t h e A r r h e n i u s p lo t s s h o w n in

Fig. 6. T h e A E v a l u e s o f 58 + 9 a n d 27 +_ 3 k J / m o l

w e r e o b t a i n e d for Dolabella a n d Mustelus m e t M b s ,

r e spec t ive ly , and a re e s sen t i a l ly i n d e p e n d e n t o f p H

n e a r t h e p K va lue .

TABLE III

Kinetic parameters for the acid-alkaline transition and the low-spin fraction in Dolabella and Mustelas ferric myoglobins

Myoglobin pH

Dolabella

Muste/us

Temper- k+ k_ Ro H a

ature (M -1 s -1 ) (s -s) (°C)

7.4 21 (1.9+0.4)×108 86+ 9 0.36_+0.03 7.4 30 (3.0+0.7)×10 s 170+20 0.315:0.03 7.4 35 (4.8-+1.2)x10 s 280+30 0.30-+0.03 7.8 21 (1.3+0.2)X108 895 :8 0.485:0.03 7.8 30 (2.6_+0.5)X10 s 190+20 0.475:0.03 7.8 35 (4.2-+0.9)×108 310+30 0.465:0.03 8.3 21 (1.7+0.5)x108 1005:15 0.775:0.03 8.3 30 (3.45:1.0)×108 1805:30 0.795:0.03 8.3 35 (5.35:2.0)x10 s 2405:50 0.815:0.03

9.2 30 (2.95:0.8)x105 13+ 3 0.265:0.01 9.2 35 (3.4+0.8)x105 165:3 0.255:0.01 9.2 40 (3.65:_0.8)x105 195:3 0.235:0.01 9.6 30 (2.3+0.5)×105 17+ 3 0.355:0.01 9.6 35 (1.9+0.4)×105 18+ 3 0.305:0.01 9.6 40 (2.4+0.4)×105 23+ 3 0.295:0.01

10.2 30 (1.8+0.3)×105 24-+ 3 0.545:0.01 10.2 35 (2.05:0.2)×105 31+ 3 0.515:0.01 10.2 40 (2.15:0.3)x105 345:3 0.495:0.01

a Roll; the fractional contents of metMbOH- ( = [metMbOH- ]/([metMbOH- ] + [metMb])), determined from NMR spectra.

I

5

4

3

2

Mb

~ Mustelus Mb

pH = 10.2

I I I I I

3.1 3.2 3.3 3.4 3.5

T "1 ×10"3/K "1 Fig. 6. Arrhenius plot of k_ for DolabeUa and Mustelus Mbs. The values of 58 5:9 and 27 + 3 kJ/mol are obtained for the activation

energy (AE) in Dolabella and Mustelus Mbs, respectively.

Discussion

Ligation state of Mb It has been shown that the hyperfine shift of the

heme methyl proton resonance is largely influenced by the pseudo-contact contribution which is modulated through the zero-field splitting characteristic of the 6A state of ferric hemoproteins [46]. Since large zero-field splitting results in large downfield-shift bias for the heme methyl proton resonances [37], the larger average heme methyl proton shifts for Dolabella and Mustelus metMbs than that for equine metMb indicate that the zero-field splitting increases in the order, equine < Mustelus <Dolabella metMb. This conclusion is also supported by the analysis results of the NMR relax- ation time. The relaxation time of the heme methyl proton resonance is primarily modulated by the elec- tron spin relaxation which in turn is determined by the zero-field level of the complex. The fact that both the narrower line width and larger T1 tntr values for the resonances of Dolabella and Mustelus metMbs than those of equine metMb indicates the lower symmetry of the coordination geometry around the heme iron in the former metMbs. In addition, the results of the optical absorption spectra and relatively slower acid- alkaline transition also support the absence of Fe- bound H20 in Dolabella [30] and Mustelus (Suzuki, T., unpublished results) metMbs.

It has been shown that nature of the distal E7 residue is the controlling factor in stabilizing the bound H20 [37]. In the absence of the distal His in DolabeUa Mb, the guanidyl side-chain of Arg El0 interacts with Fe-bound anionic ligand [24]. But obviously this side- chain cannot be a proton acceptor for the bound H20.

Although only a single set of the resonances arising from Dolabella metMbOH- is observed in the NMR spectra at ambient temperature, two sets of spectra, i.e., species I ( g l = 2.79, gz = 2.12, and g3 = 1.75) and species II (gl = 2.68, g2 = 2.09, and g3 = 1.79), are detected in the EPR spectra. From the g values, both species may be recognized as the 'H-type' [12,47] or the 'N,N-type' [48] which has been interpreted as the in- dicative of the ligation of a nitrogen at the sixth coordi- nation site of the heme iron. The side-chain of Arg El0 is a possible candidate for the ligand at the sixth site in Dolabella Mb. But the direct ligation of Arg El0 to the heme iron cannot account for the pK value (7.8) of the acid-alkaline transition. The alkaline form of Aplysia Mb, which has shown to possess Fe-bound OH- by X-ray study [16], exhibits the EPR signal with gl = 2.67, g2---2.17, and g3 = 1.86 [39]. The greater spread of the g values has been attributed to a greater electronegativity of the ligand environment. We, there- fore, concluded that both species I and II of Dolabella Mb originates from metMbOH- interacted with Arg

273

El0. The appearance of the two EPR components may be attributed to a freezing effect in the EPR sample.

The absence of Fe-bound H20 in Mustelus metMb dictates that its distal Gin E7 does not contribute to stabilize the bound H20. Met form of human Mb mutant with Gin E7 possesses the bound H20 [34] and elephant metMb has shown to be partially penta-coor- dinated at neutral pH [37]. These results clearly indi- cate that functional properties of the distal residue cannot be understood through the site-directed muta- tion on various Mbs. It could be concluded that the side-chain carbonyl oxygen of Gin E7 in Mustelus metMb is not close enough to the bound H20 to act as a proton donor. The ligation of OH- in the alkaline form of Mustelus metMb is clearly manifested in the EPR g values (gl = 2.54, g2 = 2.16, and g3 = 1.86).

pK value of acid-alkaline transition Similar pK values for Dolabella and Aplysia metMbs

indicate that the affinity of OH- to the pentacoordi- nated heme iron in the mollusc Mbs is primarily con- trolled by the interaction of bound OH- with the Arg El0. This conclusion is also supported by the study of Cutruzzola et al. [25], which demonstrated that the replacement of His E7 by Val in sperm whale Mb, His(E7) ~ Val, alters the pK value from 9.0 to 10.2 and that the double mutant, His(E7)~ Val and Thr(E10) ~ Arg, exhibits the value of 8.8.

The fact that the pK value of the shark metMb is similar to that of the His(E7)~ Val mutant indicates .that the distal Gln E7 in the shark Mb does not interact strongly with bound OH-. This is in contrast to the case of elephant Mb which exhibits the pK value of 8.5 [27]. Since both shark and elephant Mbs possess Thr residue at the El0 position [31], the differ- ence in pK between these two metMbs could be solely attributed to the difference in the interaction between bound OH- and the Gln E7. The relatively low pK value for elephant metMb dictates that the side-chain NH proton of the Gin E7 in elephant Mb is strongly hydrogen-bonded to bound OH-. Hence the Gin E7 in Asian elephant metMb is likely to be slightly closer to the bound OH- than that in shark metMb. But quanti- tative comparison of the spatial relationship between the heme and the Gln E7 in these Mbs must await more detailed structural determination of their distal heine pocket.

Kinetics of acid-alkaline transition The k of Dolabella Mb is much less than the

dissociation rates of O 2 and N 3 from the mutant His(E7) ~ Val [23,25], reflecting a strong interaction between the bound OH- and Arg El0 in this Mb. The pH insensitivity of its k also supports the stability of the Fe-bound OH- in Dolabella metMb. On the other hand, the k of Mustelus metMb exhibits some pH

274

dependence (0.27 pH-1), although the slope is much less than the unity. Therefore, there appears to be a chemical environment in the distal pocket of Mustelus Mb, which contributes to stabilize the bound OH- ligand, even in the absence of a particular amino-acid side-chain to interact with.

Thermodynamics of acid-alkaline transition The AE values for the dissociation process of the

bound OH- are 58_ 9 and 27 _+ 3 kJ /mol for Dola- bella and Mustelus metMbs, respectively. The AE dif- ference of about 30 kJ /mol between the two metMbs accounts for much more than the value (4-8 kJ/mol) found for the stabilization energy of the hydrogen bonding between Fe-bound ligand and the distal His E7 in mammalian Mb [49] and therefore, in addition to the interaction between the bound OH- and Gln E7, protein distal and proximal effects have to be consid- ered to interpret these AE values.

In conclusion, the present data clearly indicate that a guanidino NH proton of Arg El0 in DolabeUa Mb serves as a strong proton donor for the bound OH- and such interaction stabilizes the Fe-ligand bond. These results are consistent with the previous conclu- sion drawn for the functional properties of the Arg El0 in the mollusc Mb [20,24,50]. Our EPR data, together with those of Aplysia metMbOH- [39], demonstrate not only that the magnetic properties of the heme iron is largely perturbed through the strong interaction be- tween the bound OH- and Arg El0 in the mollusc Mb, but also that the g value cannot be a definitive probe for the ligation of OH- to the heme iron in metMb, where the bound ligand is strongly hydrogen bonded to a distal amino-acid side-chain.

On the other hand, the results for the shark Mb suggests that its distal Gln E7 is not interacting strongly with Fe-bound ligand.

Acknowledgements

One of the authors (Y.Y.) thanks Professors R. Chfij6 and Y. Inoue for their encouragement. This work is supported in part by a grant in aid for Scientific Research 04225221 (priority area) and 04680270 for the Japanese Ministry of Education, Science and Cul- ture (H.H.).

References

1 Iizuka, T. and Morishima, I. (1975) Biochim. Biophys. Acta 400, 143-163.

2 Sono, M. and Asakura, T. (1976) J. Biol. Chem. 251, 2664-2670. 3 McGrath, T.M. and La Mar, G.N. (1978) Biochim. Biophys. Acta

534, 99-111. 4 Gadsby, P.M.A., Peterson, J., Foote, N., Greenwood, E.C. and

Thomson, A.J. (1987) Biochem. J. 246, 43-54. 5 Antonini, E. and Brunori, M. (1971) in Hemoglobin and Myo-

globin in Their Reactions with Ligands, pp. 40-54, North-Hol- land, Amsterdam.

6 Caughey, W.S. (1966) in Hemes and Hemoproteins (Chance, B., Estabrook, R.W. and Yonetani, T., eds.) pp. 276-277, Academic Press, New York.

7 Giacometti, G.M., Da Ros, A., Antonini, E. and Brunori, M. (1975) Biochemistry 14, 1584-1588.

8 Pande, U., La Mar, G.N., Lecomte, J.T.L., Ascoli, F., Brunori, M., Smith, K.M., Pandey, R.K., Parish, D.W. and Thanabal, V. (1986) Biochemistry 25, 5638-5646.

9 Yamamoto, Y., Osawa, A., Inoue, Y., Chili6, R. and Suzuki, T. (1990) Eur. J. Biochem. 192, 225-229.

10 George, P., Beetlestone, J. and Griffith, J.S. (1964) Rev. Mod. Phys. 36, 441-458.

11 Hanania, G.I.H., Yeghiayan, A. and Cameron, B.F. (1966) Biochem. J. 98, 189-192.

12 Peisach, J., Blumberg, W.E., Wittenberg, B.A., Wittenberg, J.B. and Kampa, L. (1969) Proc. Natl. Acad. Sci. USA, 934-939.

13 Giacometti, G.M., Ascenzi, P., Brunori, M. and Bolognesi, M. (1981) J. Mol. Biol. 146, 363-374.

14 Yamamoto, Y., Chfij6, R., Inoue, Y. and Suzuki, T. (1992) FEBS Lett. 310, 71-74.

15 Suzuki, T. (1986) J. Biol. Chem. 261, 3692-3699. 16 Giacometti, G.M., Ascenzi, P., Brunori, M., Rigatti, G., Gia-

cometti, G. and Bolognesi, M. (1981) J. Mol Biol. 151, 315-319. 17 Bolognesi, M., Cannillo, E., Ascenzi, P., Giacometti, G.M., Merli,

A. and Brunori, M. (1982) J. Mol. Biol. 158, 305-315. 18 Bolognesi, M., Coda, A., Gatti, G., Ascenzi, P. and Brunori, M.

(1985) J. Mol. Biol. 183, 113-115. 19 Bolognesi, M., Onesti, S., Gatti, G., Coda, A., Ascenzi, P. and

Brunori, M. (1989) J. Mol. Biol. 205, 529-544. 20 Bolognesi, M., Coda, A., Frigerio, F., Gatti, G., Ascenzi, P. and

Brunori, M. (1990) J. Mol. Biol. 213, 621-625. 21 Rousseau, D.L, Ching, Y.-C, Brunori, M. and Giacometti, G.M.

(1989) J. Biol. Chem. 264, 7878-7881. 22 Qin, J., La Mar, G.N., Ascoli, F., Bolognesi, M. and Brunori, M.

(1992) J. Mol. Biol. 224, 891-897. 23 Rohlfs, R.J., Mathews, A.J., Carver, T.E., Olson, J.S., Springer,

B.A., Egeberg, K.D. and Sligar, S.G. (1990) J. Biol Chem. 265, 3168-3176.

24 Yamamoto, Y., Iwafune, K., Chili6, R., Inoue, Y., Imai, K. and Suzuki, T. (1992) J. Mol. Biol. 228, 343-346.

25 Cutruzzolfi, F., Travaglini, C., Ascenzi, P., Bolognesi, M., Sligar, S.G. and Brunori, M. (1991) FEBS Lett. 282, 281-284.

26 Nagai, K., Luisi, B., Shih, D., Miyazaki, G., Imai, K., Poyart, C., De Young, A., Kwiatkowsky, L., Noble, R.W., Lin, S.-H. and Yu, N.-T. (1987) Nature 329, 858-860.

27 Bartnicki, D.E., Mizukami, H. and Romero-Herrera, A.E. (1983) J. Biol Chem. 258, 1599-1602.

28 Krishnamoorthi, R., La Mar, G.N., Mizukami, H. and Romero, A. (1984) J. Biol. Chem. 259, 265-270.

29 Krishnamoorthi, R., La Mar, G.N., Mizukami, H. and Romero, A. (1984) J. Biol. Chem. 259, 8826-8831.

30 Suzuki, T. (1987) Biochim. Biophys. Acta 914, 170-176. 31 Romero-Herrera, A.E., Goodman, M., Dene, H., Bartnicki, D.E.

and Mizukami, H. (1981) J. Mol. Evol. 17, 140-147. 32 Suzuki, T., Suzuki, T. and Yata, T. (1985) Aust. J. Biol. Sci. 38,

347-354. 33 Suzuki, T. and Kisamori, T. (1984) Comp. Biochem. Physiol. 78 B,

163-166. 34 Ikeda-Saito, M., Hori, H., Anderson, L.A., Prince, R.C., Picker-

ing, l.J., GeorgE~3S.N., Sanders II, C.R., Lutz, R.S., McKelvey, E.J. and Mattera, R. (1992) J. Biol. Chem., 267, 22843-22852.

35 Krishna, N.R., Huang, D.H., Glickson, J.D., Rowan III, R. and Walter, R. (1979) Biophys. J., 26, 345-366.

36 La Mar, G.N., Budd, D.L., Viscio, D.B., Smith, K.M. and Langry, K.C. (1978) Proc. Natl. Acad. Sci. USA 75, 5755-5759.

37 Rajarathnam, K., La Mar, G.N., Chiu, M.L., Sligar, S.G., Singh, J.P. and Smith, K.M. (1991) J. Am. Chem. Soc. 113, 7886-7892.

38 Iizuka, T. and Morishima, I. (1974) Biochim. Biophys. Acta 371, 1-13.

39 Rotilio, G., Calabrese, L., Giacometti, G.M. and Brunori, M. (1971) Biochim. Biophys. Acta 236, 234-237.

40 WiJthrich, K. and Baumann, R. (1973) Helv. Chim. Acta 57, 336-350.

41 La Mar, G.N. (1979) in Biological Applications of Magnetic Resonance (Shulman, R.G., ed.) pp. 305-343, Academic Press, New York.

42 Beetlestone, J. and George, P. (1964) Biochemistry 3, 707-714. 43 Iizuka, T. and Kotani, M. (1969) Biochim. Biophys. Acta 181,

275 -286. 44 Iizuka, T. and Kotani, M. (1969) Biochim. Biophys. Acta 194,

351-363.

275

45 Unger, S.W., Lecomte, J.T.J. and La Mar, G.N. (1985) J. Magn. Reson. 64, 521-526.

46 La Mar, G.N. and Walker, A.F. (1973) J. Am. Chem. Soc. 95, 6950-6956.

47 Blumberg, W.E. and Peisach, J. (1971) in Probes of Structure and Function of Macromolecules and Membranes (Chance, B., Yone- tani, T. and Mildvan, A.S., eds.), Vol. 2, pp. 215-228, Academic Press, New York.

48 Yoshimura, T. and Ozaki, T. (1984) Arch. Biochem. Biophys. 230, 466-482.

49 Mims, M.P., Porras, A.G., Olson, J.S., Noble, R.W. and Peterson, J.A. (1983) J. Biol. Chem. 258, 14219-14232.

50 Mattevi, A., Gatti, G., Coda, A., Rizzi, M., Ascenzi, P., Brunori, M. and Bolognesi, M. (1991) J. Mol. Recogn. 4, 1-6.