Studies on the Phenylalanine Hydroxylase System in Viva AN IN VW0 ASSAY BASED ON THE LIBERATION OF DEUTERIUM OR TRITIIJM INTO THE BODY WATER FROM RING-LABELED I/PHENYLALANINE

The rate of release of deuterons into the body water from 2,3,4,5,6-pentadeutero-L-phenylalanine has been shown to be a valid measure of the activity of the phenylalanine hydroxylase system in vivo. At a dose of 0.5 g/kg, the rate of release of deuterons is linear for 60 to 90 min. Male rats, which had previously been shown to have 22 to 25% more phenylalanine hydroxylase activity in liver extracts than female rats, produced deuterons from deuterated phenylalanine at a rate 20 to 30% greater than female rats. p-Chlorophenylalanine, which irreversibly inhibits phenylalanine hydroxylase in uivo, caused a similar degree of inhibition of the rate of deuteron formation as was found when phenylalanine hydroxylase was measured in extracts from the same group of animals. Methotrexate, which inhibits the phenylalanine hydroxylase system by preventing regeneration of the tetrahydropteridine cofactor, caused parallel inhibition of the in uivo assay as well as when the conversion of phenylalanine to tyrosine was measured in liver slices.

From the National Institute of Mental Health, Laboratory of Neurochemistry, Bethesda, Maryland 20014 The rate of release of deuterons into the body water from 2, 3,4,5,6-pentadeutero-L-phenylalanine has been shown to be a valid measure of the activity of the phenylalanine hydroxylase system in vivo. At a dose of 0.5 g/kg, the rate of release of deuterons is linear for 60 to 90 min. Male rats, which had previously been shown to have 22 to 25% more phenylalanine hydroxylase activity in liver extracts than female rats, produced deuterons from deuterated phenylalanine at a rate 20 to 30% greater than female rats. p-Chlorophenylalanine, which irreversibly inhibits phenylalanine hydroxylase in uivo, caused a similar degree of inhibition of the rate of deuteron formation as was found when phenylalanine hydroxylase was measured in extracts from the same group of animals. Methotrexate, which inhibits the phenylalanine hydroxylase system by preventing regeneration of the tetrahydropteridine cofactor, caused parallel inhibition of the in uivo assay as well as when the conversion of phenylalanine to tyrosine was measured in liver slices.
Randomly ring-tritiated phenylalanine can be used interchangeably with ring-deuterated phenylalanine if greater sensitivity is needed in the in uiuo assay for phenylalanine hydroxylase. However, a dose of 20 to 30 wCi/kg is required.
The in uivo deuterium release assay described in this paper should be useful in studying the physiological control of the phenylalanine hydroxylating system. It also may be of value in differentiating between individuals who are heterozygotes for phenylketonuria and those who are homozygotes for hyperphenylalaninemia.
Although it had been thought that there is a total lack of hepatic phenylalanine hydroxylase activity in phenylketonuria patients. it recently has been shown that a patient with the classical form of the disease has about 0.27% of the normal enzyme activity (2). This finding raises the possibility that measures that could enhance this low level of activity might be of therapeutic value in phenylketonuria and has provided an impetus for the study of the in uiuo regulation of phenylalanine hydroxylase.
In order to evaluate the effectiveness in uiuo of procedures that are known to be able to markedly activate the purified hydroxylase, e.g. exposure of the enzyme to phospholipids or to limited proteolysis (3), as well as to aid in the exploration of other possible modes for regulating the activity of this enzyme, it was necessary to develop a method which could be used to measure accurately the enzyme activity without resorting to a liver biopsy. Currently, the best method available for assessing the amount of phenylalanine activity in uivo is the phenylalanine tolerance test (4). This method, however, is relatively insensitive and cannot be used to distinguish between a heterozygote for phenylketonuria and an individual with hyperphenylalaninemia.
A possible method for the in uiuo measurement of phenylalanine hydroxylase would be one based on the formation of tritiated water from p-tritiophenylalanine upon hydroxylation.
However, since nearly all of the isotope in the para position migrates to the meta position in the reaction (5), this assay would require the use of fully ring-labeled phenylalanine. Although this method would be useful in animal studies, it might necessitate the use of undesirably large doses of radioactivity in humans. An alternative approach would be the use of fully ring-deuterated phenylalanine as the substrate. Hydroxylation then would lead to the release of deuterium, a nontoxic, nonradioactive substance, which can be measured by a number of physical methods (6).
This paper describes the preparation of L-phenylalanine (ring-deuterated) and its use in an in uivo assay for phenylalanine hydroxylase.
Measurements of the rate of release of deuterons into the body water, after administration of this compound to rats, have shown that this assay is a valid one for Co. All other chemicals were reagent grade. Preparation of L-Phenylalanine-D,-Phenylalanine was deuterated by a modification of the acid-catalyzed exchange procedure described by Moss and Schoenheimer (7). To 10 g of L-phenylalanine were added 10 ml of D,O. The mixture was cooled in an ice bath while 100 g of sulfuric acid-D, were added slowly. The mixture then was flushed with nitrogen, sealed, and stirred for 7 days at room temperature. The yellow solution then was slowly added to 500 g of ice water, keeping the temperature below 20". A solution of 255 g of barium acetate in 500 ml of water was added, and the precipitate of BaSO, was removed by centrifugation.
The supernatant fraction was passed through a Dowex 50 (H+, 100 to 200 mesh) column (5 x 20 cm) to remove any remaining sulfate and sulfonated phenylalanine.
The column was washed with 5 column volumes of water and the deuterated phenylalanine was eluted with 1 column volume of 2 N NH,OH. The NH,OH eluate was concentrated on a flash evaporator to a small volume at 40". Ethanol was added to precipitate the product, which then recrystallized from a water-ethanol mixture.
The yield was generally between 4 to 5 g of greater than 90% ring-deuterated phenylalanine. The ultraviolet spectrum of the deuterated phenylalanine was identical with that of authentic phenylalanine. Upon direct introduction into the probe of a chemical ionization mass spectrometer, a parent ion of m/e = 170 (170/165 > 9) was found and a peak at m/e = 125 (125/120 > 9), corresponding to loss of CO,. This second ion also indicated that there was no racemization in the deuteration process, since there was no detectable deuterium incorporated in the cy position. This conclusion was further confirmed by a determination of an NMR spectrum in 1 N DCl, which showed a multiplet at 3.3 d (methylene, 2H), a multiplet at 4.38 6 (a-CH, 1H) and a multiplet 7.4 d (Ar-H, <0.2H).
The deuterium content was determined by integration of the NMR spectrum.
The deuterated phenylalanine was found to have the same V,., and K, values as untreated L-phenylalanine for purified rat liver phenylalanine hydroxylase when measured as described (8). There was also 1 mol of deuterium produced for each mole of tyrosine formed in the in uitro assay.   We have shown that methotrexate inhibits the phenylalanine hydroxylase system by inhibiting dihydropteridine reductase (16), thereby preventing the regeneration of the tetrahydropterin.
This inhibition is usually not evident in assays done on extracts, where excess regenerating system is added routinely. It has been shown, however, that methotrexate inhibits the conversion of phenylalanine to tyrosine in liver slices (12). To see if the in viuo assay can detect the lowered hydroxylase activity caused by inhibition of the system by methotrexate, two groups of rats were given 2.5 mg/kg of methotrexate intraperitoneally.
After 1 hour, the animals were killed and the conversion of phenylalanine to tyrosine in liver slices was determined.
There was 50% inhibition compared to salineinjected controls. The second group of rats was given a load of L-phenylalanine-D, 1 hour after the methotrexate injection and the rate of formation of deuterium was measured. There was 55% inhibition compared to controls. Here again, the results obtained with the in viva assay are in excellent agreement with those of the in vitro assay.
In Fig. 2 are shown the results of giving rats a load of tritiated phenylalanine (3 &i/rat, 0.5 g of L-phenylalanine/kg). The appearance of 3H in the body water is linear with time and is inhibited by methotrexate (2.8 mg/kg) to a similar extent as in the deuterium release assay as shown in Table I. To test whether phenylalanine hydroxylation or a subsequent step in the pathway for conversion of tyrosine to CO, and H,O is rate-limiting, the rate of release of deuterons from 2, 3,4,5,6-pentadeutero-L-phenylalanine and from 2,3,5,6-tetradeutero-nL-tyrosine was compared. Deuterium was released from tyrosine-D, in the rat at a rate at least 2 times greater than from phenylalanine-D,.
The first enzyme in the metabolic pathway for the conversion of tyrosine to CO, and H,O is tyrosine transaminase.
As a further test of the hypothesis that phenylalanine hydroxylase is the rate-limiting enzyme in catabolism of phenylalanine, a group of rats was treated with triamcinolone (100 mg/kg, intraperitoneally in 1 ml of 0.9% NaCl) to induce tyrosine transaminase.
After fasting for 14 hours, two animals were killed and their liver tyrosine transaminase levels were measured (17). The remainder were given a load of tritiated phenylalanine as described above and the rate of appearance of tritiated water in the body water was measured. Liver tyrosine transaminase levels were increased B-fold over the untreated controls and the rate of appearance of 3H in the body water was identical to that shown in Fig. 2. These results strongly indicate that phenylalanine hydroxylation is the rate-limiting enzyme in the pathway for the catabolism of phenylalanine.
Besides its conversion to tyrosine, incorporation into proteins is one of the important reactions in the metabolism of phenylalanine.
It was of interest to know, therefore, how sensitive the rate of phenylalanine hydroxylation might be to competition from this alternate metabolic pathway. If variations in the rate of protein synthesis could alter the rate of phenylalanine hydroxylation (by altering the concentration of phenylalanine available to the hydroxylase), this sensitivity to competition from other metabolic pathways might limit the usefulness of the deuterium release assay. To evaluate this possibility, protein synthesis in rats was inhibited by administration of cycloheximide (3.5 mg/kg). The rate of release of deuterium from L-phenylalanine-D, was unaffected by this treatment, an indication that agents that can lead to changes in the endogeneous pool size of amino acids have no effect on the in oiuo hydroxylase assay. These results also indicate that in rats the rate of protein synthesis is low compared to the rate of conversion of phenylalanine to tyrosine.

DISCUSSION
The phenylalanine hydroxylase system in mammalian liver catalyzes the following reactions (1) If the rate-limiting reaction in the pathway leading to the complete oxidation of phenylalanine is the conversion of phenylalanine to tyrosine, the rate of appearance of deuterium in the body water after the administration of 2, 3,4,5,6-pentadeutero-r,-phenylalanine would be a quantitative measure of the in uiuo activity of the phenylalanine hydroxylase system. If a reaction subsequent to the phenylalanine hydroxylation step were rate-limiting, however, the rate of appearance of only one out of five of the deuteriums (i.e. the one released during the hydroxylation reaction per se) would be a measure of the hydroxylation step, whereas the rate of appearance of the major part of the deuterium would be a measure of other reactions in the metabolic sequence. This latter circumstance might complicate the interpretation of deuterium release data. Because of these considerations, it was necessary not only to delineate the general characteristics of the in uiuo deuteriumrelease assay, but also to try to determine whether or not phenylalanine hydroxylation is the rate-limiting step in the catabolic pathway.
The finding that deuterium is released from ring-deuterated tyrosine faster than it is released from ring-deuterated phenylalanine indicates that the conversion of phenylalanine to tyrosine is the rate-limiting step in the catabolic pathway.
The results obtained with the two inhibitors of the phenylalanine hydroxylating system, p-chlorophenylalanine and methotrexate, provide independent evidence that the hydroxylasecatalyzed step is the rate-limiting one in the sequence that leads to deuterium release. From the data in Fig. 1, based on the deuterium release assay, the rate of tyrosine formation in the whole animal can be calculated and is equal to 0.15 pmol of tyrosine/min/animal. From the data in Table I for liver slices, and from a separate assay on liver extracts carried out in the presence of tetrahydrobiopterin (the rate shown in Table I for extracts was determined in the presence of 6,7-dimethyltetrahydropterin), it can be calculated that the rate of conversion of phenylalanine to tyrosine in liver slices and extracts is 0.13 and 0.11 wmol of tyrosine/min/animal, respectively. It should be noted that these three values, which are in reasonable agreement, were not corrected for the temperature difference between the in uiuo assay, carried out at the body temperature of the rat, and the two in vitro assays, which were carried out at 25".
In summary, a simple, stable-isotope, in uiuo assay for phenylalanine hydroxylase activity has been developed. That the rate of release of deuterium after a single dose of Lphenylalanine-D, is a valid measure of the phenylalanine hydroxylase system in viva is shown by the parallel results obtained with this assay and with the more direct assays for 4785 hydroxylase activity in liver slices and extracts. Since the assay, as it would be applied to humans, should be as innocuous as the commonly employed phenylalanine tolerance test, requiring only a few small samples of blood (1 ml/time point), and equipment found in most laboratories, it may find widespread use. It should be especially useful in differentiating between individuals who are heterozygotes for phenylketonuria, with about 15% of normal phenylalanine hydroxylase levels, and those who are homozygotes for hyperphenylalaninemia, with about 5% of normal phenylalanine hydroxylase activity (18). The standard phenylalanine tolerance test cannot with certainty discriminate between these two conditions. The formation of tritiated water from ring-tritiated phenylalanine can also be used for an in uiuo assay for phenylalanine hydroxylase. The greater sensitivity of this assay could outweigh the potential hazards of the radioactive label when attempting to differentiate between a homozygote for phenylketonuria with a small amount of residual phenylalanine hydroxylase activity and the hyperphenylalaninemic. Further experiments are also in progress on the use of this assay in evaluating alternate methods of therapy for phenylketonuria which are based on increasing the level of activity of the phenylalanine hydroxylase system in viuo.