d) Inhibition of AChE by Complexed Fluorides
Using the example of the inhibition of AChE we pursued the question of whether complexed fluorides inhibit an enzyme more strongly than the fluoride amounts contained within them if all the fluoride was in ionized form. If this is the case, and if the existence of such compounds in the organism can be supported or even proven, then vagotonic fluoride effects in a physiologically justifiable concentration range might possibly be understood in this way.
We therefore studied the inhibitory effect of the complexes dealt with earlier using AChE from human erythrocytes, PChE from human serum, and purified AChE from bovine erythrocytes, obtainable commercially. We were initially interested in the dependence of the inhibition on the concentration of the complexes and then, with the help of the remaining processes discussed in section III,A,2, tried to make statements about the inhibition kinetics.
First we investigated the inhibition by hexafluorosilicate of AChE from bovine erythrocytes (Serva), AChE from human erythrocytes (using intact cells), and PChE from human serum (using non-purified serum). The results are presented in figure 22. The inhibition of PChE again depicts a non-monotonic course (see figure 9).
Figure 22 – Cholinesterase Inhibition as a Function of Hexafluorosilicate Concentration
To study whether the kinetics are homogeneous within the concentration range used in the experiment, we plotted (vo/v)-1 against the concentration of inhibitor. The results are presented in Figure 23. In the initial section the lines run linearly. Curve 2 has a critical start value and therefore does not come out of the origin. A certain initial concentration of inhibitor is therefore necessary for inhibition to begin. We could make this observation in all analogous investigations of intact erythrocytes in a Ringer's solution.
Figure 23 - Dependence of (vo/v) -1 on the Concentration of Inhibitor.
(See Figure 22 for a legend for the curves)
In contrast to the analogous curves with NaF as the inhibitor (figure 10), these curves leave the linear domain above a certain concentration, and thereby also leave the domain of validity of equations 16-18 for n=1. In order to determine if a change occurs in the order of the complex building reactions between enzyme and inhibitor within the concentration range of the experiment we plotted the coordinates from figure 23 in double logarithmic form. According to equations 19-21, sections of straight line should arise if the number of inhibitor binding regions in the enzyme is constant within the concentration range of the experiment. The slope of these lines should be a measure of the number of binding regions.
Figure 24 – Log-Log Plot of (vo/v) -1 vs. the Concentration of Inhibitor.
In the case of the purified enzyme (1), two straight lines can be drawn to approximate the course of the curve. The SiF62- concentration at the bending point is 1.43 x 10-3 M. The slopes have the following values:
in the region of lower concentration: n=0.81
in the region of higher concentration: n=1.23
According to these values, the number of binding sites spontaneously increased at the given concentration.
Curves 2 and 3, which are derived from measurements of human erythrocytes and serum, are curved across their entire length. One possibility is that the number of binding sites on the enzyme is constantly changing, which would mean that in these functional groups, which are frequently represented in the large protein molecule, accumulation occurs in a non-specific way. The other possibility is that the relatively small buffering capacity of the Ringer's solution does not hold its ground against the hydrolysis of the complex, so that the pH shifts, which leads to an increase in the inhibitory capacity of the fluoride ions that arose from hydrolysis, since the formation of free HF would increase. A clear kink in the curves, at least in the case of the PChE from serum, is nonetheless visible here as well. Apparently, when a certain concentration of hexafluorosilicate (that is to say its partially hydrolyzed product) is reached, the form of the enzyme binding changes. This change blocks substrate binding. To uncover the nature of this binding we investigated the dependence of inhibition on the concentration of substrate in a double reciprocal plot.
Figure 25 - Lineweaver-Burk Diagram of the AChE-ACh System in a Phosphate-Citrate Buffer at pH 7.7.
AChE-ACh + 0.71 x 10-3 M Na2SiF6
AChE-ACh + 1.42 x 10-3 M Na2SiF6
According to this figure, the inhibition is non-competitive. Using equation 8 we can calculate the inhibitor-constant to be: KI = (1.82 +/- 0.06) x 10-3 M. The next figure shows the same plot with human serum.
Figure 26 - Lineweaver-Burk Diagram of PChE-ACh system in Veronal/HCL Buffer, pH 7.4
The course of the inhibition is mixed-competitive. In this case the inhibitor constant derived from equation 11 is: KI = (1.53 +/- 0.07) x 10-3M. The kinks in curves 1 and 2 after 1/[S] = 3.25 or [S] = 0.31 x 10-3 M are particularly conspicuous. This situation probably again arises from the individual enzymes' differing affinities for the substrate. The linear course of curve 3 might be due to the components that caused a kink in the two lower curves already being completely inhibited at this concentration of inhibitor. Only inhibitor concentrations of [ACh] > 0.31 x 10-3 M were used to calculate the inhibitor constants. These "constants" are, however, not real dissociation constants, but rather a cumulative value. One can only use them to describe an inhibitory effect of a MgSiF6 solution on PchE, based on equation 15.
The mixed competitive characteristic stems from the fact that the F ions that were freed during the partial hydrolysis of SiF62- cause a competitive inhibition of HF, with which they are in equilibrium. Meanwhile, the residual complex causes a non-competitive inhibition. That this observation did not appear in the measurement represented in Figure 25 is in and of itself astonishing. It might be because the inhibitory effect of free fluoride on the PChE is of greater importance in relation to the Si-complex than in the case of AChE from bovine erythrocytes, where the inhibitory effect of the residual complex covers that of the free F-.
We made a very interesting observation when we simultaneously added the complex from aqueous solution (in which the hydrolysis runs distinctly more slowly due to the low pH level that develops) and the substrate to the buffered enzyme solution. At this moment the hydrolysis must approach a constant end-value in a manner analogous to that depicted in figure 16. One can, however, assume that the solution will become saturated more quickly because the complex is already in solution, while in the other case it would first have to dissolve. We otherwise ran the measurement as described for figure 26.
Figure 27 - Lineweaver-Burk Diagram of the AChE-ACh System with the Addition of MgSiF6
The inhibition is competitive and unusually strong. The inhibitor constant has a value of KI = 2.9 x 10-5 M and is thereby 52 times smaller than for the measurement given in figure 25, in which the Na2SiF6 solution was added a half hour before the substrate was added. The difference surely can not be solely explained by the use of different cations (Na+ and Mg2+ respectively) or different buffers (citrate-phosphate buffer - pH 7.7 and Veronal/HCl buffer - pH 7.4 respectively). Apparently there were ions present shortly after initiation of hydrolysis that are highly active with regard to the AChE and can compete with the substrate at the active site of the enzyme. Whereas after some time passes, during which an aging of the hydrolysis product begins, a form develops, perhaps through chain elongation, that binds to the enzyme outside of the active site. The inhibition capacity simultaneously diminishes significantly.
We must now also take into account that the enzymatic inhibitions, as they are represented in figure 22, consist of two factors, one of which is triggered by free F- (which acts by way of HF), and the other of which arises from the residual complex. Since we have measured the degree of hydrolysis [dissociation] of SiF62- at pH 7.4 (see figures 16 and 17) and also know the inhibition by fluoride of active ACh enzymes in human blood at pH 7.4 (see figure 9) we can, through subtraction, determine the portion produced by the residual complexes. This is shown in figures 28 and 29. The lower curves correspond to the portion of the total inhibition represented by the residual complexes. We used a middle level of hydrolysis of a = 0.6 as a basis.
Figure 28. Human Erythrocyte AChE Inhibition Due to MgSiF6 in Ringer’s Solution
(Total and Fraction Assigned to Residual Complex)
Figure 29. Human Serum PChE Inhibition Due to MgSiF6 in Ringer’s Solution
(Total and Fraction Assigned to Residual Complex)
One can recognize that the inhibitory effect of a [SiF6]2- solution is stronger than the corresponding amount of free F- that it releases. In the case of AChE the total inhibition by the complex is nearly twice as large as that of the free fluoride ions. In the case of PChE the F ions represent the larger portion; however, here too the residual complex accounts for a significant portion of the total inhibition.
Inhibition of AChE by Additional Fluoride Complexes
In order to discover if other fluoride complexes could inhibit AChE and if there is a relationship to the size as well as the charge of the complexes, we studied the effects of the following complexes: BF4-, AlF63-, GeF62-, and PF6-. We described the hydrolytic behavior of these complexes in the previous chapter.
The dissociation level of these compounds in veronal/HCl buffer at pH 7.4 was a = 0.83, which is exactly equivalent to five fluoride ions splitting from the complex. When we performed the inhibition tests with AChE from human erythrocytes in Ringer's solution we observed something that was unique to this complex. Since the buffering capacity of the Ringer's solution was not sufficient to counteract the H+ ions set free by the hydrolysis, the pH value shifted into the acidic region. This shift should, according to figure 12, result in a decrease of the self-saponification rate of the ACh, since this ester is saponified from OH- through a catalytic effect. We, however, observed the opposite. Despite a decrease in the OH- concentration, the saponification rate increased with rising GeF62- in the absence of enzyme.
Figure 30 - Self Saponification Rate of ACh as a Function of GeF62- Concentration and the pH Value in Ringer's Solution.
Reaction time = 0.5 hours
The increase in the self saponification rate of the ACh could be brought about by the catalytic effect of a germanium complex. GeF62- that is incompletely hydrolyzed in a more acidic medium. Separated F- ions are not replaced by OH-, so that the end product of the hydrolysis is GeF4 in this case.
Because of the two unoccupied d-orbitals, this compound has the characteristics of a strong "Lewis acid", which can catalyze saponification reactions that run according to a SN2 mechanism through its polarizing effect. In this case it exerts an "electron pull" on the carbonyl oxygen of the ACh and thereby strengthens the positive partial charge on the C atom. The following schematic gives an overview of the probable course of the reaction:
The positive carbon reacts out of this transition state upon addition of an H2O molecule! An intra-molecular rearrangement of the bonds then leads to the products of reaction, choline and acetic acid, whereby the catalyst is reformed.
Because of the pH shift that arose when K2GeF6 dissolved in Ringer's solution, we could not study the effect of this substance on the AChE of intact erythrocytes. We therefore used PChE from human serum in a phosphate-citrate buffer. The next figure shows the course of the enzymatic inhibition as well as the difference curve, which we derived by subtracting the F- portion.
As one can see, the inhibition is predominantly caused by the fluoride, which means that in this case the residual complex is hardly in a position to cause an enzymatic inhibition.
All the other complexes, BF4-, PF6-, and AlF63-, only inhibit the enzymes as much as the proportion of fluoride ions released by their hydrolysis. This means that only the bivalent representatives of the fluoride complexes can cause an inhibition while the inhibition strongly decreases upon transition from SiF62- to GeF62-.
The effect of fluoride on the cholinesterases can therefore be increased when fluoride is used in a Si complex (e.g. as MgSiF6) instead of in ionized form (e.g. as NaF). The [SiF6] 2- ion is particularly effective when it is not hydrolyzed until it reaches the place where it acts, since apparently reactive intermediate products form that can cause a competitive inhibition of the AChE. This can be the case when the substance was previously absorbed in the stomach, where hydrolysis does not take place because of the acidic medium that is predominant there. But the fact that the remaining fluoride complexes do not display such an effect does not mean that they are of no biological importance. Especially in the case of AlF63-, which is widespread in nature, comprehensive studies of its effects on many possible biological processes should be carried out. This would probably contribute to an understanding that the role of fluoride in the nature of organic life is not limited to just the effectiveness of free fluoride ions.