The Stimulation of Rat Liver Phenylalanine Hydroxylase by Phospholipids

Abstract The maximum velocity of rat liver phenylalanine hydroxylase is stimulated by the following phospholipids, in order of decreasing potency: lysolecithin, lysophosphatidylserine, phosphatidylserine, and sphingomyelin. Lysolecithin stimulates the maximum velocity 50-fold when the naturally occurring pterin, tetrahydrobiopterin, is used as cofactor, and only 1.15-fold when the synthetic cofactor, 6,7-dimethyltetrahydropterin, is employed. In the presence of tetrahydrobiopterin, lysolecithin also converts the initial velocity-phenylalanine concentration relationship from a sigmoidal to a hyperbolic curve.

Glucose was found to be the major carbohydrate component, whereas xylosc and hexosamine occurred to a lesser degree. The ratios of the three sugars varied from 3 to 5 moles of glucose and 1 to 2 moles each of xylose and hexosamine per mole of the agglutinin.
The mannose content decreased gradually with the degree of purity of the wheat germ agglutinin.
The crystalline wheat germ agglutinin probably has no mannose or at best 1 mole per mole of the agglutinin.
More detailed analytical data on the amino acid and carbohydrate composition will be reported elsewhere.
Wheat germ agglutinin preparations obtained from a Sephades G-75 (2) or a DEAE-cellulose chromatography gave two bands on polyacrylamide gel electrophoresis (Fig. 3C). The two bands were always of about the same relative intensity.
However, a single band, corresponding to the slower moving band of the two (Fig. 3, A and C), was observed upon electrophoresis of the active material eluted from a carboxymethylcellulose column. The faster moving band (Fig. 3B), which was also obtained in pure form after chromatography on a carboxymethylcellulose column, was found to have no agglutinating activity for either L1210 leukemia cells or human red blood group Al or B cells. Amino acid analysis showed a clear difference between the two proteins, e.g. wheat germ agglutinin contained 40 residues of half-cystine and 3 residues of valine, whereas the faster moving band contained 14 residues of half-cystine and 22 residues of valine per 23,500 g. Therefore, the main impurity associated wit,h wheat germ agglutinin up to the DEAE-cellulose step is probably not a suburlit of wheat germ agglutinin or a similar kind of phytohemngglutinin.
A comparison with other agglutinins, concanavalin ;Z and soybean agglutinin, both known to agglutinate also primarily but not exclusively transformed cells, indicates that these two agglutinins do not contain any cysteine, whereas carbohydrates have been observed in soybean agglutinin (mannose and glucosamine) but not in concanavalin A (11)(12)(13).
On the other hand, only the nonagglutinating, inactive subunit of concanavalin -% (active molecule, 55,000 and 100,000; inactive subunit, 27,000) (14,15) has a molecular weight in the range of wheat germ agglutinin.
An s-ray cryst,allographie study on wheat germ agglutinin is now in progress in Dr. R. Langridge's laboratory at Princeton University, and preliminary results will be reported (16).

SUMMARY
The maximum velocity of rat liver phenylalanine hydroxylase is stimulated by the following phospholipids, in order of decreasing potency: lysolecithin, lysophosphatidylserine, phosphatidylserine, and sphingomyelin. Lysolecithin stimulates the maximum velocity SO-fold when the naturally occurring pterin, tetrahydrobiopterin, is used as cofactor, and only 1.15-fold when the synthetic cofactor, 6,7-dimethyltetrahydropterin, is employed.
In the presence of tetrahydrobiopterin, lysolecithin also converts the initial velocityphenylalanine concentration relationship from a sigmoidal to a hyperbolic curve.
Rat liver phenylalanine hydroxylase requires a reduced pterin cofactor as a source of electrons for the conversion of phenylalsnine to tyrosine (1). When a synthetic pterin, 6,7dimethyltetrahydropterin, is used as cofactor, the initial velocity-phenylalanine concentration curve is of the classical Xlichaelis-blenten hyperbolic type (2). When the naturally occurring pterin, tetrahydrobiopterin (3), is employed as cofactor, phenylalanine gives a sigmoidal saturation curve and a maximum velocity much lower than that obtained with the synthetic cofactor (2).
It was recently reported that when DMPH41 is used as the cofactor and tryptophan as the substrate for crude rat liver phenylalanine hydroxylase, the velocity-tryptophan concentration curve is sigmoidal (4). It was also reported that high concentrations (i.e. 0.5 W) of propanol convert the kinetics from the sigmoidal to the Michaelis-l\lent.en hyperbolic form. We were interested in determining whether a similar effect of propanol could be observed with phenylalanine hydroxylase in the presence of its normal substrate, phenylalanine, and the naturally occurring cofactor, BH4. We found that not only does propanol change the kinetics from sigmoidal to hyperbolic under these conditions, but it also increases the maximum velocity several-fold.
In a search for a naturally occurring compound that might be a more potent activator of the hydroxylase, we foulid that lysolecithin and lysophosphatidylserine at low concentrations are capable of greatly stimulating the For most of the assays the enzymatic rate was followed spectrophotometrically by mea.surement of the phenylalaninedependent oxidation of TPNH (5). In these assays, the complete reaction mixture consisted of the following components in a final volume of 1 ml: potassium phosphate, pH 6.9, 0.1 M; TPNH, 0.2 InM; catalase (Boehringer-Mannheim Corp.), 100 pg; dihydropteridine reductase (purrfied through the calcium phosphat,e gel step, (l)), 200 pg; tetrahydropt.erin (concentration indicated in legends) ; phenylalanine (concentration indicated in legends) and phenylalanine hydroxylase (purified through Sephades G-200 step and about 90% pure (6)), 18 pg. Although there was the same concentration of hydroxylase in each experiment, the basal and lysolecithin-stimulated rates varied, apparently due to a loss in activity from freezing and thawing the hydroxylase.
The degree of Iysolecithin stimulation, however, was nearly the same in all experiments.
When turbid components such as lipid suspensions were present in the assay, the reaction rate was determined by a fluorometric modification of the nitrosonaphthol assay for tyrosine (7). In this assay, in addition to the components present for the spectrophotometric assay, glucose B-phosphate, 2.0 mM, and gIucose 6-phosphate dehydrogenase (Boehringer-Mannheim Corp.), 10 pg, were included to maintain the TPNH in its reduced state. All assays were carried out at 25". Lysolecithin and lysophosphatidylserine are soluble in water in the concentrations used in the assay. The other lipids, however, are insoluble in water and were susbtentled by sonicat,ion (with a Sonifier Cell Disruptor, Branson Sonic Power Co., Corm.; 20 s at setting 3). Since the lysolecithin used was prepared commercially by treatment of egg white lecithin with phospholipase iz, it was important to establish its purity.
The lysolecithin migrated as a single spot in two thin layer chromatography systems. The adsorbant ITas Silica Gel II. The solvent for one system was chloroform- were similar to those reported for lysolecithin in the two systems (8). The concentration of the lipids was determined from the phosphorous content of ashed samples (9).
In the presence of BH4, high concentrations of I-propanol (1 M) stimulat.ed phenylalanine hydroxylase several-fold. Since it was found that butanol was three times more effective than propanol, we studied compounds containing longer carbon chains. Fatty acids containing 14 carbons or less did not stimulate.
Fatty acids containing 16 carbons or more were more effective than butanol in stimulating phenylalanine hydroxylase.
The greatest stimulation was observed with phospholipids.
Lysolecithin and 1ysophosphatidyIserine were the most active compounds tested. Table I (Experiment 2) shows that low levels of lysolecithin and lysophosphatidylserine stimulated phenylalanine hydroxylase about 20.fold when the enzyme was saturated with its substrates (phenylalanine, BH4, and oxygen). Lysolecithin stimu-Iation was studied in greater detail because this compound is more readily available than lysophosphatidylserine.
Lysolecithin produced detectable activation at 0.01 mM and full activation at 0.15 n?M (Fig. 1). Lysolecithin obtained from t,wo different commercial sources gave nearly the same degree of stimulation for a given concentration of the lipid. The data in Table I (Experiment 1) also demonstrate that at high concenkations, phosphatidylserine and sphingomyelin also stimulated the hydroxylase.
Phosphatidylethanolamine and lysophosphatidylethanolamine had no effect. Egg white lecithin was highly inhibitory and this inhibition was overcome by lysolecithin.
In the presence of lecithin, however, a higher concentration of lysolecithin was required to give full activation.
It should be mentioned that lysolecithin cannot substitute for the phenylalanine hydroxylase-stimulating protein that was recently described (10). Indeed, a synergistic effect is observed when phenylalanine hydroxylase-stimulating protein and lysolecithin are both added to the hydroxylase.2 2'262 Phenylalanine Hydroxylase Xt&ulat.ion by Phospholipids Vol. 247, FIG . 2 (left and center). a, phenylalanine hydroxylase activity as a function of phenylalanine concentration with and without 1 rn~ lysolecithin.
The cofactor was BHI at 0.03 mm. b, the reciprocal of the velocities obtained with lysolecithin depicted in a were plotted DBTSUS the reciprocal of the phenylalanine concentrations. Fig. 20, shows that the saturation curve for phenylnlanine was sigmoidsl when T%Hk was used as cofact.or (as previously reported (2)). Lysolecithin converted the kinetics for phenylalanine saturation to the hyperbolic type with substrate inhibition above 0.1 nlM phenylalanine (Fig. 2b). Lysolecithin stimulated the maximum velocity 50.fold and reduced the K, for phenylalanine from 0.3 to 0.2 mM.
At rat serum levels of phenylalanine, 0.078 f 0.0075 mm (n = 30),3 lysolecithin stimulated the rate of phenylalanine hydroxylation over loo-fold when the natural cofactor was employed.
When the synthetic pterin, DMI'H4, was used as the cofactor, lysolecithin increased the maximum velocity only 1.15.fold (Fig. 3). 'G'nder these conditions, lysolecithin decreased the Km for phenylalanine from 1.3 to 0.8 mN. A comparison of the results in Fig. 2 with those of Fig. 3 shows that without lysolecithin the rate of hydroxylation in the presence of DXIPH4 was 12 times the rate obtained with l3H4, but with lysolecithin the rate of hydroxylation was four times faster with EHI than with DMPH+ We also studied the effect of lecithin in the presence of DMPH,.
With this synthetic cofactor even 5 rnht lecithin did not inhibit the hydroxylase.
Therefore, both the activation by lysolecithin and inhibition by lecithin occur only with the natural cofactor RH4.
It is known that a large number of enzymes can be activated by phospholipids (11,12). The mechanism of activation, however, is obscure.
Since we have recently found that surfactants, such as sodium dodecyl sulfate, can also activate phenylalanine hydrosylaseP the activation by phospholipids of t'his enzyme is probably related to their detergent properties.
It, should be noted t,hat, in contrast to the effect of phospholipids, the activation by SDS is seen only wit,hin a narrow concentration range; the maximum stimulation requires three t.imes more SDS t.han phospholipid, and concentrations of SDS greater than 0.5 mM inactivate the enzyme.
Our results demonstrate that two naturally occurring phospholipids, lysolecithin and lecithin, are remarkably effective activators and inhibitors, respectively, of rat liver phenylalanine hydroxylase and suggest the possibility that these lipids might regulate the in vivo activity of the hydroxylase. Since 2';: of liver phospholipid is lysolecithin (13), its concentration in this tissue (assuming uniform distribution within the liver a C. B. Storm and S. Kaufman, unpublished observations. 4 D. B. Fisher and 8. Kaufman, unpublished observations.