| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Research and Development, Amersham Health, Nycoveien 2, N-0401 Oslo, Norway. Received February 14, 2002; revision requested March 13; revision received April 29; accepted May 24. Address correspondence to the author (e-mail: tore.skotland@amersham.com).
Index terms: Contrast media, effects • Contrast media, experimental studies • Editorials
Radiographic contrast agents show very little activity in most biologic test systems. To investigate if these agents have a biologic effect or to compare the effect of various agents on a given parameter, in vitro test systems with up to 150 mg of iodine per milliliter of the contrast agent are often used. It should be noted that such concentrations correspond to approximately 0.4 mol/L solution of a monomeric agent. This is approximately half the concentration of the injected solutions, which, of course, are rapidly diluted after injection. Not only are the concentrations used in the in vitro test systems very high during the whole incubation period, but also it is important to note that interpretation of data from experiments with such high concentrations has some pitfalls that are neglected in almost all such experiments. The intention of this commentary is to highlight some of these pitfalls.
Effect of Very High Concentrations of Radiographic Contrast Agents on Enzyme Reactions
The density of molecules in solutions containing 75 or 150 mg of iodine per milliliter of a monomeric contrast agent is illustrated in the Figure. As reported earlier and shown in this Figure, there is some molecular clustering of these agents when they are present at high concentrations, although no large molecular aggregates are formed (1). Most enzymes have molecular masses of 15–150 kd. Thus, an enzyme of "medium" size is similar in size to albumin (68 kd), which would occupy a volume corresponding to 16–18 of the small squares shown in the Figure. By looking at this Figure, it is easy to understand that by adding enzyme molecules to such highly concentrated solutions, there may be physical restrictions in getting the substrate in contact with the active site of the enzyme. Consequently, a reduced rate of product formation may be observed. Thus, it looks like an inhibitor has been added, although the effect may be due entirely to the extreme density of molecules in the solution.
|
Another effect to be aware of is the possible presence of activators or inhibitors in the test mixture. For instance, if an activator is present at concentrations producing an activating effect, an increase in the amount of an inert contrast agent may result in reduced physical contact between activator and enzyme, resulting in a reduced rate of product formation. In such cases, there is a danger of interpreting the finding as evidence for the added contrast agent being an enzyme inhibitor, which is clearly incorrect. Conversely, if an enzyme reaction is run in the presence of an inhibitor, an increase in the amount of added contrast agent may result in reduced physical contact between inhibitor and enzyme. The result is an increased rate of product formation, without any activating effect of the contrast agent on the enzyme. It should be noted that such effects due to activators or inhibitors in the test mixture might be larger with high-molecular-weight activators or inhibitors, compared with those with low-molecular-weight activators or inhibitors (similar to the steric hindrance effect described previously for large versus small substrates).
In addition to the effects obtained as a result of the number or size of substrates and of the presence of activators or inhibitors, the rates of association and dissociation for the interactions of these molecules with the enzyme will also be of importance. One would expect the effect of high concentrations of contrast agents to be larger when the interactions between enzyme and substrate or between enzyme and activator or inhibitor are weak, compared with that when they are strong (weak interactions mean that the substrate, activator, or inhibitor spends a longer time not bound to the enzyme, ie, in a state where the high concentration of contrast agent produces a steric hindrance effect). To compare the effect of contrast agents on different enzymes or analytic systems, it is therefore necessary to consider the actual Km value(s) for the substrate(s) and for the concentration(s) used for the substrate(s) (eg, the enzyme will be saturated with substrate at substrate concentrations far higher than the Km value). Accordingly, even though a contrast agent may affect one enzyme but not another at a given contrast agent concentration, the observed effect may not be specific. It could be due solely to a difference in binding constants or size of substrates, activators, or inhibitors of the two enzymes in combination with the physical restriction effect obtained as a result of the extremely high concentration of contrast agent. Moreover, one should not neglect the possibility that enzyme incubation for many minutes in the presence of very high concentrations of such contrast agents may result in solvation effects on the enzyme structure. This may change the catalytic properties of the enzymes, without relevance for the in vivo enzyme activity.
In addition to the interpretation of in vitro data obtained in the presence of high concentrations of contrast agents, I would like to add some comments concerning the use of enzyme kinetic data in general. To make enzyme mechanistic interpretations based on such data, it is essential that the initial rate of product formation (or depletion of substrate) is measured or that the measurements are performed within a period when the enzyme reaction is linear with time. This is essential to exclude complicating factors, such as the effects of possible reverse enzyme reaction, inhibition of the enzyme by the product(s), and progressive inactivation of the enzyme during incubation. Moreover, it is important to stress that steady-state kinetic analysis of a reaction cannot be used to unambiguously establish the reaction mechanism or inhibitory mechanism. If, however, kinetic data are incompatible with a given mechanism, then this mechanism must be rejected. In conjunction with the high molecular density effects described previously, it should be stressed that increasing the amount of contrast agent in the solution may result in a kinetic pattern similar to what might be observed for a competitive inhibitor (increased apparent Km and unchanged kcat). However, whereas a competitive inhibitor acts by binding to the active site and thus reduces the concentration of free enzyme available for substrate binding, the effect of physical restriction will be to reduce the access of the substrate to the active site of the enzyme.
Examples from Published Data
Despite the importance of stating that enzymes are measured in a linear phase of an enzyme reaction, seldom is such a statement made for in vitro studies with radiographic contrast agents. An exception to this fact is a recent study by Graf et al (2), who measured initial velocities and described how data deviating from the average measurements, or outliers, were treated. This precise description of the experimental conditions makes it possible to properly evaluate the results. The authors reported an inhibitory effect of iopromide on thrombin activity, and this inhibition was proportional to the contrast agent concentration up to 184 mmol/L, producing maximal inhibition of 44%. However, since enzyme kinetics follow a hyperbolic curve, it is not theoretically possible to have a reversible inhibitor that produces a linear response of up to 44% of the complete inhibition, because of the curvature of hyperbolic plots. The only way to achieve linear inhibition of up to 44% and for this effect to be due to an inhibitor is to add an extremely reactive irreversible inhibitor (it must be extremely reactive because thrombin activity was shown not to decline over time). In this case, such an irreversible inhibitor must be present in very low amounts because of the high amounts of contrast agents added, as compared with the amount of thrombin molecules in the mixture. It is, however, very unlikely that such an active irreversible inhibitor would be present in the contrast agent solution; such an inhibitor would also be expected to produce in vivo effects. The most plausible interpretation of these data is that the very high concentration of radiographic contrast agent in the incubation mixtures made it physically difficult for the substrate to reach the active site of the enzyme. Such an effect should be linear with the added amount of contrast agent, that is, as was reported in the study.
The effect of physical restriction due to high concentrations of contrast agents is a probable explanation of large differences reported by different groups for several enzyme activity measurements, for example, the different effects on thrombin activity reported for some nonionic monomers. For example, it was reported that 80 mmol/L iopromide produced only approximately 20% inhibition of thrombin activity in two test systems in which low-molecular-weight chromogenic substrates were used (2). On the other hand, when thrombin activity was tested by measuring the time for clot formation in the presence of added plasma, clotting time at a similar contrast agent concentration (30 mg of iodine per milliliter) increased approximately 10 times (3). Clearly, the clot formation system involving reactions between several coagulation factors may be influenced much more by physical restrictions in the solution than by enzyme reactions measured with small chromogenic substrates. Furthermore, one may also speculate that the more pronounced effect observed for thrombin activity when testing serum thrombin, as compared with that when using the purified enzyme (2), may be due to the increased physical density of the test solutions that results from added serum proteins. It should be noted that there was no disagreement between thrombin activity measurements in these studies. As discussed herein, such differences are likely to occur when in vitro analyses are performed in the presence of high concentrations of radiographic contrast agents. It is not my intention to state that one of these methods is better than the other.
In another detailed and interesting publication (4), the effect of high concentrations (up to 240 mg of iodine per milliliter) of 11 contrast agents was tested on various enzymes or test systems. With regard to the present discussion, it is interesting to look in detail at the data reported for the three plasminogen activators, that is, streptokinase, urokinase, and tissue plasminogen activator. Similar test systems with chromogenic substrates were used to analyze for these plasminogen activators. In general, much lower concentrations of contrast agents were necessary to inhibit streptokinase, compared with those inhibiting urokinase or tissue plasminogen activator (mean IC50 [concentrations at which 50% inhibition was obtained] values of 53 mg of iodine per milliliter ± 14 [SD] [n = 11] for streptokinase, 174 mg of iodine per milliliter ± 94 [n = 11] for urokinase, and 233 mg of iodine per milliliter ± 150 [n = 6] for tissue plasminogen activator). With regard to this example, it should be noted that streptokinase is not an enzyme, despite its name, but acts through activating the large protein plasminogen (86 kd), thereby changing plasminogen to an active enzyme that is able to hydrolyze other plasminogen molecules or chromogenic substrates. This may explain why streptokinase in these highly concentrated solutions is inhibited at much lower concentrations of contrast agents than is urokinase or tissue plasminogen activator, as the two latter enzymes need to react with only one low-molecular-weight chromogenic substrate. In fact, this difference may have been further accentuated, since plasma was used as the source of the plasminogen used to make the complex with streptokinase, thus also increasing the density of the molecules in the test solution because of the added plasma proteins. Once again, it is not my intention to state that one of these methods is better than another but to point out that such effects are likely to occur when in vitro analyses are performed in the presence of high concentrations of radiographic contrast agents.
The intention of this commentary is to point out some of the pitfalls and risks of erroneously interpreting data obtained by using in vitro test systems with high concentrations of radiographic contrast agents. The main focus has been on a simple enzyme reaction system and how factors such as the number of substrates and the size and binding constants of substrates, inhibitors, and activators may have a large effect on the enzyme rate in such highly concentrated mixtures. Thus, the effect of the added contrast agent need not be a direct effect of the agent on the enzyme but rather an indirect effect due to the extremely concentrated incubation mixture used, that is, because of physical constraints that restrict contact between the enzyme and substrate, activator, or inhibitor in the mixture. It should be noted that if cells are involved in the test systems, interpretation might be even more complicated than it might be for simple enzyme reaction systems. Finally, it should be mentioned that the effects observed in such highly concentrated solutions should be easily reversed at dilution. Such dilution will, of course, also take place immediately after injection of these agents into patients, since the injected dose is rapidly diluted in the blood volume. The period that the blood enzymes will be in contact with high concentrations of these agents after bolus injection will thus be for only a few seconds, much less than the long incubation times used to observe the reported effects in in vitro experiments.
ACKNOWLEDGMENTS
I am grateful to Per Christian Sontum, PhD, for helpful discussions and for producing the Figure.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  L. L. Graf, D. A. Young, D. C. Kressin, R. A. Marlar, G. B. Jacob, P. H. Hinderling, and T. Skotland Importance of Using Controls in in Vivo Experiments * Dr Skotland responds: Radiology, October 1, 2004; 233(1): 297 - 299. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
