281

Biochimica et Biophysica Acta, 428 (1976) 2 8 1 - - 2 9 0 © Elsevier Scientific Publishing C o m p a n y , A m s t e r d a m -- Printed in The Netherlands

BBA 27854

LOW TEMPERATURE PHOTODISSOCIATION STUDIES OF FERROUS HEMOGLOBIN AND MYOGLOBIN COMPLEXES BY MOSSBAUER SPECTROSCOPY K. S P A R T A L I A N a, G. L A N G a and T. Y O N E T A N I b

a Department of Physics, The Pennsylvania State University, University Park, Pa. 16802 and u Department of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania, Philadelphia, Pa. 19174 (U.S.A.) ( Received N o v e m b e r 3rd, 1975)

Summary STFe-enriched complexes of hemoglobin and myoglobin with CO and O5 were photodissociated at 4.2 ° K, and the resulting spectra were compared with those of the deoxy forms. Differences in both quadrupole splitting and isomer shift were noted for each protein, the photoproducts having smaller isomer shift and larger quadrupole splitting than the deoxy forms. The photoproducts of HbCO and HbO2 had narrow absorption lines, indicating a well-defined iron environment. The corresponding myoglobin species had broader absorption lines, as did both deoxy forms. The weak absorption lines of photodissociated NO complexes appeared to be wide, possibly indicating magnetic interaction with the unpaired electron of the nearby NO.

Introduction The formation of a quickly reacting form of hemoglobin (Hb) on photodissociation of HbCO was first reported by Gibson [1]. In a number of recent papers the study of photodissociated heine complexes has been extended to extremely low temperatures. This has made it possible to slow the recombination and study its kinetics in greater detail [2--8]. It has also made possible a close spectroscopic examination of the dissociated state because the dissociation may be performed at 4.2°K, producing a complex which remains stable as long as the temperature is kept below 10 ° K. In view of the role of allosteric effects in the cooperative oxygen binding by hemoglobin, the possibility of producing and examining a ligand-free iron site in a protein having the ligated conformation Abbreviations: Mb, m y o g l o b i n ; Hb: h e m o g l o b i n ; MbCO: m y o g l o b i n carbon m o n o x i d e ; Mb*(CO), the presumably un]igated species f o r m e d b y p h o t o d i s s o c i a t i o n of MbCO; etc.

282 makes such studies of obvious importance. A fundamental question arises: does photodissociation produce a special state of the iron site? Only a probe which is sensitive to iron alone, regardless of its state, can answer this unambiguously. This is relevant to an understanding of the binding kinetics because, for example, relaxation of the binding site to a state of lower affinity might become an attractive alternative to the postulated multiple barrier well for the ligand molecule in accounting for details of the recombination [8]. Yonetani et al. [5] and Iizuka et al. [6] found that in artificial hemoglobin and myoglobin (Mb), made by substituting Co for Fe, it was n o t possible to distinguish the photodissociated 02 complex from the ordinary deoxy form by spectrophotometric means. However, the photolysis products CoMb* (O2) and CoHb* (O2) were found to have EPR signatures clearly different from those of deoxy CoMb and CoHb. In spectrophotometric studies [3,7] of the EPR-silent iron proteins the same group found that a distinction could be made between d e o x y and photodissociated forms only in the absorption in the far infrared region near 760 nm. In the position of this "conformational band" d e o x y myoglobin and hemoglobin were clearly distinct from each other and from the six photodissociated species which were investigated. The latter, made from the 02, CO, and NO complexes, all had conformational band maxima at nearly the same wavelength. In the present work we introduce MSssbauer spectroscopy as a complementary m e t h o d for investigating the iron site in these materials. This technique has the advantage that it senses only the energy level shifts of the iron nucleus, and hence probes only the binding site. Furthermore, the relative simplicity of the process and the generally good level of understanding which this has allowed make it possible in most cases to determine whether one, two, or many iron environments are present and to assign the spin states unambiguously.

Experimental

Sample preparation Myoglobin was prepared from horse heart according to the m e t h o d of Yamazaki et al. [9]. Hemoglobin was prepared from human blood according to the m e t h o d of Drabkin [ 1 0 ] . STFe-enriched p r o t o h e m e was prepared as described previously [11]. STFe-enriched myoglobin and hemoglobin were prepared by the chemical exchange of the prosthetic group as described previously [ 12]. STFe-enriched p r o t o h e m e containing myoglobin and hemoglobin were reduced with dithionite and passed through a column of Sephadex G-25 equilibrated with air-saturated 0.1 M phosphate buffer, pH 7.0, to remove excess dithionite and to obtain the oxy compounds. The d e o x y c o m p o u n d s were prepared from the corresponding o x y c o m p o u n d s by reduction with a minimal a m o u n t of dithionite. The nitric complexes were prepared by anaerobic reduction with dithionite in the presence of 0.1 M NaNO2 as described elsewhere [13]. The final MSssbauer samples (2--4 mM in heme) were prepared in 0.1 M potassium phosphate buffer, pH 7.0, transferred into the MSssbauer sample holders, and immediately frozen in liquid nitrogen.

283 Instru m e n tation

All samples were maintained at 4.2°K by immersion in liquid helium in a cryostat of a design described elsewhere [14]. The samples were in the frozen water solution form, 1/8 inch thick, in polythene sample holders of the kind described in ref. 14. In order to allow for the illumination of the samples, the cryostat was modified in the following manner: A copper radiation shield which could be rotated from the outside was placed around the fixed 77°K shield protecting the tailpiece of the liquid helium chamber. This movable shield had two perpendicular sets of windows, one set open and one covered with iron-free aluminium foil. Illumination was effected in the "open" position with the sample already immersed in the helium. Rotation of the movable shield to the "closed" position after the completion of the illumination cycle blocked the incoming blackbody radiation, thus ensuring low helium comsumption during the running time of the experiment. All experiments were run with the illumination off during the accumulation of the MSssbauer spectra. The illumination of the samples was provided by a slide projector with a 150 W incandescent lamp (Sylvania CEW). All samples were illuminated for 2 min on each side. No attempt was made to quantify the illumination because the high sample concentration and polycrystalline nature of the frozen matrix gave rise to large and unpredictable absorption. For each sample, spectra were taken at 4.2°K before illumination, after illumination and after annealing. Annealing consisted of removing the sample from the liquid helium, immediately immersing it in liquid nitrogen and keeping it therein for 20 min before replacing it in the cryostat. The MSssbauer spectra were taken in horizontal transmission geometry using two independent constant acceleration spectrometers operated in connection with 256 channel analyzers in the time scale mode. The sources used were kept at room temperature and consisted of 50 mCi of STCo diffused in rhodium foil. The spectrometers were calibrated against metallic iron foils and zero velocity was taken as the centroid of their room temperature MSssbauer spectra. In these calibration spectra linewidths of about 0.23 mm/s were normally observed. Results

Figs. 1 and 2 show the zero field MSssbauer spectra at 4.2 ° K of oxyhemoglobin and oxymyoglobin prior to illumination (top), after illumination (middle} and after annealing (bottom}. The illuminated samples of HbO2 and MbO2 clearly show the appearance of two extra peaks in addition to the usual quadrupole split pattern of the oxygenated species. These extra peaks, hereafter referred to as the photodissociated component of the spectrum, have roughly the same energy separation AE as the original component, but distinctly different isomer shifts 8. Annealing the illuminated samples caused the photodissociated components; Hb*(O2) and Mb*(O2}, to vanish. Figs. 3 and 4 show the zero field MSssbauer spectra at 4.2°K of HbCO and MbCO before illumination (top) and after illumination (middle). The bottom spectrum in each case is from the corresponding deoxy material and is included for comparison. It is immediately seen that the photodissociated components

284

/

I (a)

2%

. -

-,.

I IN

(a)

-% ..... '"

• .

I -2

I -t

SOURCE

I IN

..

(c)

I

-'--

(b) •.'

-5

•

2%

(b)

2%

• . (c)

I

I

I

I

I

0

I

2

5

-3

VELOCITY (MM/SEC)

.,

I -2

I -I

SOURCE

Fig. 1. Zero-field MSssbauer spectra at 4.2°K luminated HbO 2 after annealing at 77°K.

of (a) oxyhemoglobin,

F i g . 2 . Z e r o - f i e l d M 6 s s b a u e r s p e c t r a at 4 . 2 ° K • o luminated MbO 2 after annealing at 77 K.

of (a) oxymyoglobin,

~ I%

I

I

L

0

2

3

VELOCITY (MM/SEC)

HbO 2 ; (b) illuminated

H b O 2 ; ( e ) il-

MbO 2 , (b) illuminated

M b O 2 ~ ( e ) il-

are much stronger than in the previous t w o figures. The spectra of annealed material are not shown, but each was identical with the corresponding top ~pectrum, indicating as expected the disappearance of the photodissociated component. Comparison of the photodissociated c o m p o n e n t with the spectrum of d e o x y hemoglobin directly below it in Fig. 3 indicates that the left line of the Hb*(CO) spectrum is aligned with the left edge of the much wider corresponding hemoglobin line, while the right hand lines appear to coincide. Thus Hb* (CO) has a larger quadrupole splitting AE and a smaller isomer shift ~ than hemoglobin. The central broad peak of the hemoglobin spectrum is attributed to hemochromogen impurities c o m m o n l y present in deoxyhemoglobin samples. Unlike hemoglobin, myoglobin has a MSssbauer spectrum which appears symmetric about its center. The individual absorption lines, however, are not symmetric, being steeper on their outer edges. This same line asymmetry is present in Mb*(CO), at least in the right hand line, which is clearly defined. It is obvious from the figure that the left line of Mb*(CO) lies at a more negative velocity than its counterpart in myoglobin. Thus Mb*(CO) has the larger AE and smaller 5. Fig. 5 shows the spectra at 4.2°K of HbNO and MbNO before and after illumination. An external magnetic field of a few hundred Oersted was applied perpendicular to the ~f-ray beam. The NO complexes are Kramers salts, in contrast

285

I "

I I%

(a)

(0)

I I% .o

•, (b)

.

~

(b)

.

•

..."

•

I%

"' '. ,

.4. o.

..

"

•

•

I%

o

o.

f I%

It)

I

(c)

I%

°•

L"

,

,

,

i

i

,

i

-3

-2

-~

0

~

z

3

SOURCE VELOCITY (MM/SEC)

I -3

J -2

1 -I

SOURCE

J 0

I

L

I

2

I 3

VELOCITY (MM/SEC)

Fig. 3. Zero-field M6ssbauer spectra at 4.2°K of (a) c a r b o x y h e m o g l o b i n , HbCO; (b) illuminated HbCO; (c) d e o x y h e m o g l o b i n . Fig. 4. Zero-field M6ssbauer spectra at 4.2°K of (a) carboxymyoglobin, MbCO; (b) illuminated MbCO: (c) deoxymyoglobin.

with the other materials examined here. Their complex spectra result from magnetic interaction between the iron nuclear m o m e n t and the unpaired halfintegral electron spin. The applied field serves to decouple these, in the sense that its interaction with the electron spin is large compared with that of the iron nucleus or any other nearby nuclei. It therefore simplifies the hyperfine interaction Hamiltonian which determines the M6ssbauer spectrum. Differences can be seen between the "before" and "after" spectra for each protein, but the photodissociated c o m p o n e n t is not easily resolved as in the case of the 02 and CO complexes. However, the absorption is seen to increase at energies corresponding to the line positions of the respective d e o x y spectra (shown by bars in the figures). Interpretation and Discussion When the absorption lines of M6ssbauer spectra overlap, attempts to establish their positions accurately by least-squares fitting are appreciably influenced by the assumed lineshape. Many of the lines we have observed have widths greater than the natural linewidth attributable to the finite lifetime of the excited nuclear level. A small part of this is instrumental; a least-squares Lorentzian fit to calibration spectra of thin iron foils yields typically a line-

286

" ~ P ~ " "~":~~,~

~ i

(a]

~

:

.;.~.~.~.

0.5%

'-. ~,

(b)

~,.

~'...

z

'...

~. &: ".,,,,:

'-

x/"

~-"',':,"v-.,

"% (d)

/

.:,-

~'

~,,.. -. (c)

.~'-':.

•

.

% .

""

"%

:",':-"

o,5%

-.

/¢.

":

","

""

"-,'

I 0.5%

?":"

,-

.~..-..,~...~.

"

:-:.

.'

I

"" " "

~

: