Thyroxine Degradation during Lipoxidase-catalyzed Peroxidation of Linoleic Acid*

SUMMARY The possibility has been examined that thyroxine may be degraded by an intermediate free radical in the oxidation of linoleate catalyzed by lipoxidase. Such a reaction has observed and the lipid radical with which thyroxine reacts suggested enzyme attached A similar reaction has been inferred to occur during thyroxine degradation peroxidizing microsomal lecithin. autoxidation lecithin, it a similar reaction with enzyme-bound RO; oxidation The results out the degradation

Such a reaction has been observed and the lipid radical with which thyroxine reacts is suggested to be enzyme attached RO; proposed by Tappel (in W. 0. LUIVDBERG (Editor), A&oxidation and antioxidants, Vol. I, Interscience Publishers, New York, 1961, p. 325). A similar reaction has been inferred to occur during thyroxine degradation by peroxidizing microsomal lecithin. Although thyroxine serves as an antioxidant during autoxidation of lecithin, it was predicted that a similar reaction with enzyme-bound RO; might not influence oxidation rate or extent.
The results bear out the prediction. The purpose of this and prior studies of thyroxine degradation in systems in vitro has been to investigate the mechanism of reaction of thyroxine with biological materials.
In light of recent studies of the mechanism of deiodination of iodoaryl compounds, it is suggested that the lipid radical RO; might either initiate degradation by electron abstraction or react with a terminal product of thyroxine after deiodination. The enabling property of thyroxine making these reactions probable is its predicted capacity to take up, transiently hold, and yield electrons.
Thyroxine is degraded during the peroxidation of liver microsomes (I, 2), the oxidation-reduction of flavins (3), and the ironinitiated autoxidation of lecithin (2). It serves as an antioxidant during the latter reaction and appears to react with the chainpropagating species of autoxidation. In order to study this latter possibility further, a source of similar lipid radical was sought which would not involve the presence of a metal or require indirect inferences of chain-propagated oxidation. The oxidation of linoleate catalyzed by soybean lipoxidase was chosen as such a source of oxidized lipid radical.
The propagating lipid radical of autoxidation may be designated as RO;. Since a similar radical is proposed as an intermediate of lipoxidase-catalyzed oxidation of linoleate (4) This is inferred because enzyme-attached R0.j has no propagating capacity and the proposed thyroxine reaction is viewed as occurring following the lipid reaction with oxygen.
The results of these studies bear out these predictions.
Thyroxine is degraded during lipoxidase-catalyzed oxidation of linoleate. This degradation is the consequence of reaction of thyroxine with an intermediate product, presumably enzyme-bound, and thyroxine has no measurable influence on oxidation rate of the reaction.

Methods
Basic Lipoxidase-Linoleate Reaction-Linoleic acid, 1.44 x low6 mole, in 0.3 ml of alcohol was added to 11.3 ml of 0.1 M sodium phosphate buffer at pH 7.4. Stock thyroxine solution, 1 x 1OP M in ethanol, was added in amounts from 0 to 0.40 ml. Ethanol was added to bring the final volume of ethanol to 0.7 ml. The reaction was begun by the addition of 0.5 ml of phosphate buffer containing varying amounts of lipoxidase. The thyroxine concentration varied between 8 x 10M7 and 3.2 X lo-' MI.
Oxygen Uptake-Oxygen uptake was measured polarographically in a 12.5-ml constant temperature chamber at 23" with a Teflon-covered platinum electrode.
Measurement of Iodide Released from @-Phenyl Ring of Thyroxine-The basic methodology used has been previously described (1).
Ultraviolet Light Absorption-Ultraviolet light absorption as a measure of conjugated diene production was measured as follows. Reaction temperature was 23'. The reaction to whidh thyroxine was added is indicated by the dashed line. The chamber contained 3.3 X 10-s mole of oxygen at loo'% saturation. Disappearance of 4Ooj, of contained oxygen represents an uptake by linoleate of 1.32 X lO+ mole of oxygen or 92'% of the predicted value. There is no effect of thyroxine on the rate or extent of the oxidation. 2. Rate of oxygen uptake with different concentrations of linoleate with and without thyroxine.
Reactions included 20 mg per liter of lipoxidase and 8 X 1OaB M thyroxine if added. Maximal reaction rates were estimated during the first 90 set of the reaction. Temperature, 23". Reactions to which thyroxine was added are shown by the dashed line.
There is no effect of thyroxine on the initial rates of the reaction. Lipoxidase concentration was 8 mg per liter. Temperature, 23". There are no significant differences among the oxygen uptake rates plotted. The lower dashed curve is that reaction to which thyroxine was added at the outset. The arrow indicating addition of thyroxine indicates an amount added to replace thyroxine degraded during the reaction.
This amounted to 30yo of the starting material.
Since the blank for this reaction contained the same initial amount of thyroxine as the experimental reaction at the outset, optical density could not attain the value of the nonthyroxine control until degraded thyroxine in the experimental flask was replaced. The early lag in optical density of the thyroxine-supplemented reaction is due to rapid degradation of thyroxine during this period.
When replaced later as shown, extinctions were equal in thyroxine-and non-thyroxine-treated reactions.
The basic reaction mixture was modified in that the concentration of Iinoleate was only half that indicated in the above description.
The final concentration of phosphate buffer remained 0.1 M at pH 7.4. Light absorption was measured in a l-cm cuvette at 236 nm. The signal was recorded continuously.
Continuous absorption spectra were assessed after allowing the basic reaction to run to compIetion.
The reaction mixture was then diluted 1:5 with water, phosphate buffer, or phosphate buffer containing 5.6% alcohol.
The diluent made no measurable difference in the absorption spectra. Continuous absorption spectra were determined with the Cary model 14 spectrophotometer.
Measurement of Radioactive Isotopes-Solutions containing Ia11 were assayed in an autogamma spectrometer at the 264 kev peak. Chromatograms on which lalI-and '*C-labeled substances had been separated were assessed in a Geiger-Mueller continuous gas flow strip counter.  Table I, thyroxine neither influences the initial rate nor the extent of oxygen uptake during linoleate oxidation catalyzed by lipoxidase.
Inffuence of Thyroxine on Hydroperoxide Production-The development of absorbance at 236 nm is interpreted as an index of the quantity of conjugated diene monohydroperoxide produced during linoleate oxidation. This is based on observations described by Bergstrom (6) and Privett (7). As is shown in Figs. 4 and 5, thyroxine alters neither the rate of development of extinction at 236 nm nor the quality of the extinction spectrum between 230 and 360 nm. On this basis, it is presumed that systems with and without thyroxine produce the same conjugated diene structure in equal amounts.
Thyroxk Degradation during Linoleate Oxidation-Although there is no effect of thyroxine on the rate of oxygen uptake by linoleate or on the rate of hydroperoxide formation, thyroxine is deiodinated during the oxidation.
In Fig. 6 is shown the direct relationship between thyroxine degraded and oxygen uptake. In Fig. 7 is shown a plot of l/v for thyroxine degraded against l/s for thyroxine added. The apparent K, for thyroxine under these specific conditions is 0.4 X 10m6 M. The apparent V,., for thyroxine with a 1.15 x 10m4 molar concentration of linoleate and 8 mg per liter of enzyme is 0.665 x 10e6 mole per liter per min. Comparing the measured oxygen uptake rate in Fig. 3 with the estimated thyroxine V,,, under the same conditions, there is approximately 1 mole of thyroxine degraded for each 11 moles of linoleate oxidized.
Chromatographic and Electrophoretic Studies-Chromatograms of the lalI-labeled products formed during linoleate oxidation in the presence of thyroxine (3', 5'-1311) reveal that inorganic iodide and thyroxine are the only labeled substances present following are the same as those described under Fig. 6. Since oxygen uptake here is the same as that shown in Fig. 6, the most rapid rate of thyroxine degradation at infinitely high concentrations of thyroxine under these conditions allows 1 mole of thyroxine to be degraded for each 11.4 moles of oxygen taken up or linoleate oxidized. Oleate (which is not oxidized by lipoxidase) incubated with lipoxidase and thyroxine results in neither oxygen uptake nor thyroxine degradation.

Although
these studies indicate that thyroxine is degraded during lipoxidase-catalyzed oxidation of linoleate, no physiological implications are inferred since thyroxine and lipoxidase do not naturally occur together.
The purpose of this and prior studies of thyroxine degradation has been to define the reactivity of thyroxine with biological materials.
For this purpose the plant enzyme has certain advantages.
Although the autoxidation studies suggested that thyroxine reacts with the lipid radical RO;, the radical has not been isolated and the thesis is difficult to test. Because the lipoxidase-linoleate reaction generates the same radical by a different mechanism, predictions can be made that, if supported, would circumstantially strengthen the thesis that thyroxine reacts with RO;.
Evidence that thyroxine participates during the enzyme-catalyzed reaction is shown by its degradation.
The oxidation rate is not measurably altered. Thus, these results are consistent with a mechanistic expectation of what might happen were thyroxine to react with enzymebound RO;.
Several questions may be raised in regard to alternative explanations.
Does thyroxine react with any of the initial or terminal products?
May it react with a transient non-enzymebound intermediate?
May it react with an intermediate other than RO; and, fina.lly, are the reactions during the enzymecatalyzed oxidation similar to those that occur during chainpropagated autoxidation?
Because the initial and terminal enzyme products will not effect thyroxine degradation and because degradation takes place only during the course of oxidation, the view is supported that intermediate lipid products are responsible for the reactions. The apparent conformity of the degradation reaction to enzyme kinetics implies a reaction of thyroxine with an enzyme-lipid complex.
This latter speculation depends upon the assumption that thyroxine does not react with a small amount of RO; or related radical that may have been first released from the enzyme. Lipoxidase ordinarily catalyzes the production of the cis, trans-13-monohydroperoxide-conjugated diene from linoleate (6-g).
If any significant quantity of RO; were released from the enzyme, nonenzymatic autoxidation at 23" would proceed rapidly forming the cis, trans-and trans, trans.9-and la-monohydroperoxide-conjugated dienes (9). The influence of thyroxine would be to terminate the chain reaction and slow the rate of oxidation if it were to react in a fashion similar to that in which it reacts with autoxidizing lecithin. Small amounts of thyroxine would cause large decreases in such nonenzymatic oxygen uptake. The fact that oxygen uptake is not influenced at all during this enzymatic reaction supports the view that thyroxine reacts with an enzyme-bound radical to cause its degradation and that there is essentially no parallel, nonenzymatic autoxidation of lipid. If the enzyme-bound intermediates of the lipoxidase-linoleate reaction may be accurateIy depicted as Tapper has inferred (4), it is unlikely that thyroxine reacts with any radical other than RO;.
The proposed enzyme-bound radical intermediates are R', H', and RO;.
If thyroxine were to react with R', oxygen uptake would be inhibited and reaction rate would be slowed.
If it were to react with H', the reaction rate might be slowed or the end product altered.
Neither situation occurs. Unless the lipoxidase-catalyzed reaction involves more complex intermediate products than Tappel proposed, it seems most likely that thyroxine reacts with RO;.
No matter with which of the radical intermediates thyroxine may react initially, it seems probable that the mechanism of degradation is similar in the microsomal, autoxidizing lecithin, and lipoxidase-linoleate systems. The identity of the products of the fi-phenyl ring iodine and cy-phenyl ring-alanine side chain following thyroxine degradation by all three systems supports this view (2,5). Although the reaction mechanism is unknown, several insights have been developed.
Degradation of thyroxine depends in part upon its predicted free radical stability (10) and Borg has demonstrated the occurrence of a relatively stable thyroxine-free radical signal (11). Complete removal of /3phenyl ring iodine atoms without occurrence of partially iodinated intermediates and net reduction of the deiodinated positions has been shown (2,5,12). Heretofore, there has been no model reaction from which a rational scheme could be devised to account for these observations. Two recent reports have been helpful in this regard.
Anbar, studying the effect of hydrated electrons on iodobenzene, showed that this compound reacts with such electrons by ejecting inorganic iodide (13). Bunnett studied the deiodination of iodobenzene in alkaline alcoholic solution containing a free radical capable of electron abstraction (14). He demonstrated that iodine is ejected from the aryl structure as iodide aud that the position vacated is reduced, He devised a stoichiometrically satisfactory theoretical reaction mechanism assuming that the initial reaction involved acquisition by iodobenzene of a solvated electron or an electron transferred from a negatively charged free radical in solution.
The reaction was only observed with iodine-substituted benzene.
Chlorine and bromine substitution could not be similarly removed.
Both of these studies suggest that t,he degradation reactions involving thyroxine may be initiated by either acquisition of an electron or by molecular rearrangement following one electron abstraction.
In either case, a negatively charged free radical would be formed followed by ejection of inorganic iodide, further rearrangement, formation of a second negatively charged radical, and ejection of the second iodide ion. The predicted reaction with lipid radical RO; may then be an initial reaction causing electron abstraction or a terminal reaction involving the deiodinated, electron-depleted thyroxine remnant. In either instance, the reactivity of biological importance may relate to an unusual capacity of thyroxine to take up, transiently hold, and then yield unpaired electrons rather than a specific affinity for reaction with particular biological free radicals.