pH DEPENDANT XANES OF CARBONYLHEMOGLOBIN BY FAST DISPERSIVE SPECTROSCOPY

HAL Id: jpa-00226143

https://hal.archives-ouvertes.fr/jpa-00226143

Submitted on 1 Jan 1986

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

pH DEPENDANT XANES OF

CARBONYLHEMOGLOBIN BY FAST DISPERSIVE SPECTROSCOPY

I. Ascone, A. Fontaine, M. Barteri, A. Bianconi, A. Congiu-Castellano

To cite this version:

I. Ascone, A. Fontaine, M. Barteri, A. Bianconi, A. Congiu-Castellano. pH DEPENDANT XANES OF CARBONYLHEMOGLOBIN BY FAST DISPERSIVE SPECTROSCOPY. Journal de Physique Colloques, 1986, 47 (C8), pp.C8-1201-C8-1204. �10.1051/jphyscol:19868236�. �jpa-00226143�

JOURNAL DE PHYSIQUE

Colloque C8, supplgment au n o 12, Tome 47, decembre 1986

pH DEPENDANT XANES OF CARBONYLHEMOGLOBIN BY FAST DISPERSIVE SPECTROSCOPY

I. ASCONE, A, FONTAINE, M. BARTERI*, A. BIANCONI' and A. CONGIU-CASTELLANO*

L.U.R.E., B d t i m e n t 209 D , F-91405 O r s a y C e d e x , France

" ~ i p a r t i m e n t o d i F i s i c a , U n i v e r s i t d " L a S a p i e n z a " , I - 0 0 1 8 5 Rome, I t a l y

Resume : La spectroscopie d'absorption X rapide permise par la gbm6trie dispersive a permis de distinguer trois transitions dans la cinCtique de denaturation de l'hkmoglobine saturee de CO en solution ( S M ) . Le dktachement du CO de lheme ^?i faible pH (2,6) est pric6d6 de l'inclinaison du CO du la protonation, (pH-5,6) puis du dkpliement de la protkine -(pH

-

Abstract : Fast x-Ray dispersive spectroscopy allowed by the dispersive geometric shows three transitions in the kinetics of 5mM carbonylhemoglobin denaturation. The CO detachment at low pH(2.6) is preceeded by CO bending due to the protonation (pH-5.6), and later by protein unfolding (pH-3.3)

Only a few detection systems have been achieved to compete the creditable performance of synchrotron radiation source. In the X-Ray absorption field, the combination of a well-known X-Ray energy dispersive optics and a position, sensitive detecror able to work under high flux conditions enlarges the scope of this spectroscopy.

This paper, devoted to a pH dependant XANES study of the pH-denaturation of carbonylhemoglobin offers the opportunity to discuss the advantages brought by the dispersive scheme applied to biophysics.

The ion XANES spectra shown in fig. 1 has been recorded using a Si 31 1 to give an account of the energy resolution which is as good as it is with a step by step scan using a double detuned crystal.

In addition, because of the lack of mechanical movements, once the optics is tuned for a given absorption edge, the energy scale is found absolutely stable.

The in-situ observation permits a systematic investigation of the time-dependant system and an a-posteriori decision about stages of interest. Anticipating the kinetics discussed below we would like to indicate that we bin the spectra by set of 10 or 5 according to the importance of the change observed in the relevant PH-domain PH dependant XANES of 5 mM carbonylhemoglobin solution.

Knowledge of Fe-C-0 and Fe-02 bond angles in solution is fundamental to the understanding of the reversible molecular binding of diatomic molecules in haem proteins. XAPdES has been used sucessfully to such determination for protein crystals as well as solutions.(2) Different approaches including infrared CO stretching, 13c nuclear magnetic resonance, optical absorption try to correlate the CO bending or tilting with respect ot the porphyrin plane to a steric constraint either from the distal side of the heme pocket or from the proximal side.

Recent improvements in XANES calculations performed at Rome, including superior multiple scattering treatment confirms the validity of the assignment of the 16 eV resonance to CO bending tilting

It is this resonance which will be used as a probe for the present study which is concerned with the pH denaturation of the heam protein. The time dependance is used to ramp down the pH value.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19868236

C8-1202 JOURNAL DE PHYSIQUE

ENERGY ( e V )

f i g 1

Fig. 1 a) Iron foil b) Fe

*

, 8 2 7 3 - . 8 2 1 -

. 8 2 2 3 - . 8 2 - .El>,.

,815..

Fig. 2 5 mM carbony hemoglobin solution at pH=1.8 as treated by lactate.

The signal/noise ratio is essentially photon-limited

. ,...

...

. ,. - .,

*.. .

, .. ,. ..

, ..,,.,..,,.-..;. . .. . , i::. . % _ . , .... ^.,?.

,' . . ..:., ..: . :. 7 4 s

~

; .i'-i1.

. ' f

.,

-

According to the initial amount of lactate injected in the 5mM solution we measured two sets of spectra which overlap on the pH axis . Just after lactate injection the solution was spliaed in two volume, one for pH measurements with a standart p-electrode, the other for a 3 mm thick cell with two capton windows. This cell is left without any dead volume and sealed in order to avoid oxygenation.

Measurement took about 70s for each spectrum made of 32 frames each of them taking 2.2 s.

The capability to collect all data of a full spectrum at once in a short time allows to investigate such systems in-siru evolving under one or several constraints.

Our goal is to know about the different steps which may exist starting from the protein solution at pH=7 to a final state at pH=1.8 where the protein is denatured whith a complete CO detachment. This evolution includes protein unfolding which may affect the CO bendingltilting via changes in the environment of the porphyrin cycle.

Fig. 1 and Fig. 2 introduce the boundary conditions of such a study. The superior resolution achieved by the energy dispersive set-up allows to a clear identification in the CO resonance intensity. Fig.1 , b, c show three ion Kedge according to the degree of oxydation of the cation (metallic ion, Feqt. ~e~~ solutions).

The pH=1.8 spectra of figure 2 contain two major informations. i/ The absolute value of the edge jump is 0.015 which is almost 200 times lower than the value optimized for getting superior signallnoise ratio as it is shown for the ion foil. ii/ The overal number of photons (which means the number of frames) used to build one spectrum determines the sensitivity to change in XANES spectra. As deduced from comparison of pH=7 and pHzl.8 spectra of fig. 3 the total change in intensity of the 16 eV-CO resonance is 30% of the edge jump i.e. 45 and the error bar is estimated to be about 10% of this last value (4.5

It is clear from fig. 2 that the sensitivity is photon limited. Our detection system can work at the ms scale. Hence spectra might be collected even with a 1000 times more intense source.

In fig. 2 the glitch at SO eV has been suppressed because of its importance. Several other ones remain in the set displayed on figure 4. It should be possible to get rid of most of these

"glitches" by using a mirror.

In addition of the spectra of CO-hemoglobin solution at pH=7 and the denatured state at pH=1.8, the fig. 3 shows a synthetic ion 11 basket-handled complex synthetized by M.

Momenteau. This complex has been crystallized and X-Ray diffraction data showed a Fe-C-0 configuration. The ion I1 derivative of these complex bearing a ten-membered polymethylene handle shows a great CO aff~nity in the presence of 1-methyl imidazole in toluene. For the fmt time the linear Fe-C-0 configuration has been measured in solution also using XANES.

The intensity of the 16 eV-CO resonance is 20% larger than the total variation observed between pH=7 and pH=1.8. This is consistent with the polarization dependant XANES study performed on a single crystal(2).

Fig. 4 displays the progressive change of the ion K edge versus the pH values. A guide line has been drawn within data of each spectrum to help the reader. The pH-evolution has been assigned to the total change found between pH=7 and pH=1.8. Whether this study is

.a123

-,

. E l - :: .:

pn 1.8

, , , , , , , ,

2 8 30 a8 58 6 0 70 88 90

EC.Vl

0 min

- & 7 b - -

f

Fig. 3 The arrow points out the 16eV CO resonance for a linear FeCO configuration in a synthetic basket-handled complex in solution and the HbCO solution at pH=7 and 1.8

Fig.4 pH-dependant spectra of a 5rnM Hb CO solution

58.4

time

f i g 4

pH-dependant or he-dependant would be a relevant question if the typical time-constant of such evolution under constant pH was not small.

Allis and Steinhardt (4) studied both human and horse carbonylhemoglobin denaturation.

The denaturing agent was HC1 and the Soret Cotton Effect (400-440 nm), and the optical rotatory dispersion (220-240 nm), were the probes. The time constant involved was found within one second or less and then out of the scope of the present study. These two authors distnguished different pH ranges.

i) pH > 3-4 protonation of native COHb

ii) 3 < pH < 3-4 Protein unfolding occurs and the heme group retains its

close association with its nonnal imidazole binding site on the protein, although the environment of the heme has been modified. At pH > 3 data from both time-dependant absorbances give first order kinetics with pH-dependant rate constant

iii) pH < 3 the second-order reaction is superimposed. This is intrepreted as a heme dimerization. This second rate is pHdependant down to pH=2.1. Below this value, the heme is detached from its normal site and the pH dependance disappears.

J O U R N A L D E PHYSIQUE

"I/---.. f i g 5

Fig. 5 pH-dependance of the 16eV CO- resonance showing three structural rcan- s i t ions

Our results are consistent with a three transition scheme (fig. 5), but our technique investigates system from a different side.

The lack of data in the 5-16 pH range aloes not allow to give any characteristic pH value for the transition. Nevertheless a qualitative change is shown which may correspond to a progressive CO bending as a consequence of the COHb protonation.

A second step is found close to pH=3.7 which accelerates the decrease of the CO resonance intensity.

If we follow Allis and Steinhardt who did not see the breaking of the ion-imidazole bond before pH=3, the immediate interpretation is to assign this second step to the protein unfolding.

From both optical measurements this second step was found to be a first order transformation.

A last and fast intensity decrease occurs at pH=2.6 which leads to the disappearance of the CO resonance. This may coincide with the suggested dimerization although the CO detachment did not derive from the reported optical measurements. As well no immediate insight of the imidazole bond breaking comes from XANES.

As every measurement the decrease of CO resonance intensity is not unambiguous. It may come from the CO bending as well as the CO detachment. Since the biochemical reaction leads to pH-dependant equilibrium the CO detachment may be progressive. This kind of ambiguity is often clarified when other set of data are available. Allis and Stenhardt found that a second order rate constant predonimates when the absorbance change is measured at 269 nm.

This is a real clue that dimerization occurs. They speculate about the ion-imidazole bond breading accompanied by both protonation of the imidazole and displacempent of the heme group to another binding site. If such things do exist our results imply than CO detachment takes place.

Then the first two transitions have to be interpreted as mostly due to CO bending, first under protonation, second because of protein unfolding.

Conclusion The major interest of such study is to do in-situ measurements allowing one parameter to change versus time. A 5rnM protein solution is still in the scope of the X-Ray absorption spectroscopy performed in the transmission mode. The dispersive scheme allows data to be collected at once. This a major advantage over the step by step scheme. Nevertheless this kind of study involves small changes of the total cross section of the sample. Then it is necessary to increase the signunoise ratio by accumulating frames. The projected European Source shouldbring a flux 1000 times more intense and the present detector based on photodiode array will still work.

Finally the full set of spectra were collected in less than 1 hour which is short enough for oxygen-sensitive samples, if careful conditionning has been achieved. Stop-flow experiments will be an alternative way to take benefit of the focussing optics but will require a larger quantity of solution.

This work is supported by the Scientific Department of European Economical Community.

1- E. DARTYGE, C. DEPAUTEX, J.M DUBUISSON, A. FONTAINE, A.JUCHA, P.

LEBOUCHER, and G. TOURILLON NIM. A 246 (1986) p 452-60

2- A. BIANCONI, A. CONGIU-CASTELLANO, P.J. DURHAM, S.S HASNAIN, S. PHILIPS Nature (1985) 318, p. 685-7

3- L. RICARD, R. WEISS, M. MOMENTEAU, J Chem. Soc. Corn. 1986 p.819 4- J.W. ALLIS, J. STEINHARDT, Biochemistry 9 , no 11, 1970, p 2286