Reactivity of ferrous myoglobin at low pH.

The rates of reaction of myoglobin with carbon monoxide at low pH are reported. The pH versus rate profile of these kinetics resembles that found for heme model compounds, revealing an increase in combination rate at low pH. These facts suggest that CO binding by myoglobin changes from a mechanism of "direct ligant association" at pH 5 to a mechanism, similar to that proposed for heme model compounds, which assumes a tetracoordinated intermediate as a result of the protonation of the proximal imidazole.


SUMMARY
The rates of reaction of myoglobin with carbon monoxide at low pH are reported. The pH uersus rate profile of these kinetics resembles that found for heme mode1 compounds, revealing an increase in combination rate at low pH. These facts suggest that CO binding by myoglobin changes from a mechanism of "direct ligand association" at pH 5 to a mechanism, similar to that proposed for heme model compounds, which assumes a tetracoordinated intermediate as a result of the protonation of the proximal imidazole.
Ligand binding properties of a great number of simple monomeric hemoproteins with different ligands have been studied in the past (1). Apart from the large differences in the equilibrium and kinetic constants for the various ligands, such as 0, and CO, it is well known that the reactivity of different hemoproteins for the same ligand displays great variability.
Thus, if we limit ourselves to CO, it may be recalled that the combination rate constant for the binding of this ligand to the ferrous derivative is as low as 5 x lo3 M-I s-l for horse radish peroxidase (2) and as high as 3 x lo7 M-' s-l for Chironomus erythrocruorin (3). A structural interpretation of these large differences is not available although on a number of findings, they cannot be attributed, solely or even largely, to steric effects encountered on the distal side.
In this paper, new experiments to elucidate the structural basis of intrinsic ligand reactivity in simple hemoproteins are reported. The idea behind these experiments originated from a series of investigations on mode1 compounds which have indicated that tetracoordinated ferrous heme, as can be obtained by protonation of the iron-binding imidazole, has a very high "on" constant for CO (4, 5). The central theme of this study is, therefore, to obtain direct information on the rate of CO combination of myoglobin at very acid pH. This is an attempt to gain evidence for the existence and properties of tetracoordinated ferrous iron which could originate from protonation of the proximal imidazole. at pH between 2.5 and 6; (b) measuring the spectrum of the acid form of myoglobin before denaturation has time to occur; (c) measuring the rate of CO combination to this form of the protein, again before denaturation, by mixing myoglobin with an acid buffer containing CO.

AND DISCUSSION
Following the acid denaturation of myoglobin, both as the reduced and as the carbonylated derivatives, at different final pH values, we found that the time course of denaturation is complex as previously reported for Hb (6, 71, that ferrous Mb denaturates more rapidly (kden = 3 s' at pH 3) than COMb, and that the rate of the latter process decreased as the CO concentration is increased. Fig. 1 reports the spectrum of ferrous Mb at time zero after mixing with an acid buffer (final pH 2.4), and for comparison, the spectra of native ferrous Mb and of fully denaturated ferrous Mb. The spectrum at time zero was built from the spectrum of the fully denaturated form plus the optical density changes measured in the flow experiment (corrected for the dead time of the instrument).
It can be seen that the spectrum at time zero is different from that of native Mb at neutral pH, the absorption maximum being shifted to the blue by approximately 10 nm, the maximum extinction being increased by approximately 50%. If we assume that protonation is very fast, i.e. it is completed in the mixing time, the spectrum at time zero represents the optical properties of a modified "acid" myoglobin.
Based upon spectra of four-coordinate mesoheme in benzene (8), we expect the four-coordinate protoheme to have a Soret absorption at the same position as that of carboxymyoglobin. The spectrum of the acid form of deoxymyoglobin as shown in Fig. 1 is consistent with this assignment.
Under identical conditions, we have followed combination with CO of the acid form of ferrous Mb, by rapid mixing experiments, in which a known amount of CO was present in the strongly acid buffer so that the protein finds itself confronted with the ligand immediately after the pH jump. The results are as follows: (a) the time course of the absorbance change and the CO concentration dependence show that the observed process reflects a bimolecular reaction over the entire pH range explored; (b) below pH 5, the rate of combination with CO increases as the pH decreases. No correction was made for the simultaneous denaturation whose time course is slow with respect to the CO combination. The dashed line is drawn for (log k -1.40) ucrs'sus pH from data on aqueous cetyltrimethylammonium bromide suspensions of the 3-Dimidazolyllpropyl amide of mesoheme monomethyl ester from Ref. 4.
where K = (Mb)(H+)/(MbH+) is the dissociation constant for the proton, I,' and I,' are the second order rate constants for the CO combination according to Scheme 1. Equation 2 can be written in exponential form as: k = l,'lO-"K + 1,' lo-"" lO-PK + 10-P" The data were fitted with Equation 3 using for I,' the value of 6 X loj M-' SK' The fit, shown in Fig. 2 as a continuous line, provides (a) the intrinsic pK for the promnation step which is 3.45, as compared to the apparent pK -4 and (b) the value of 1,' = 1.4 x 10' M-I s-l. A plot of (log k -