Neurochemical Research, Vol. 23, No. 7, 1998, pp. 1011-1020

The Effect of Phenylalanine on DOPA Synthesis in PC12 Cells Frank R. DePietro1 and John D. Fernstrom1,2,3 (Accepted November 24, 1997)

DOPA synthesis from phenylalanine was studied in PC 12 cells incubated with m-hydroxybenzylhydrazine, to inhibit aromatic L-amino acid decarboxylase. DOPA synthesis rose with increasing concentrations of either phenylalanine or tyrosine; maximal rates (~55 pmol/min/mg protein for tyrosine; ~40 pmol/min/mg protein for phenylalanine) occurred at a medium concentration of ~10 MM for either amino acid. The Km for either amino acid was about 1 MM (medium concentration). At tyrosine concentrations above 30 MM, DOPA synthesis declined; inhibition was observed at higher concentrations for phenylalanine (>300 MM). These effects were most notable in the presence of 56 mM potassium. Measurements of intracellular phenylalanine and tyrosine suggested the Km for either amino acid is 20-30 MM; maximal synthesis occurred at 120-140 MM. In the presence of both phenylalanine and tyrosine, DOPA synthesis was inhibited by phenylalanine only at a high medium concentration (1000 MM), regardless of medium tyrosine concentration. The inhibition of DOPA synthesis by high medium tyrosine concentrations was antagonized by high medium phenylalanine concentrations (100, 1000 MM). Together, the findings indicate that for PC 12 cells, phenylalanine can be a significant substrate for tyrosine hydroxylase, is a relatively weak inhibitor of the enzyme, and at high concentrations can antagonize substrate inhibition by tyrosine.

KEY WORDS: Phenylalanine; tyrosine; DOPA; tyrosine hydroxylase; PC12 cells.

INTRODUCTION

Phe was found to be as good as Tyr as a substrate for tyrosine hydroxylase (3). Studies in vitro in brain synaptosomes (4-6) and intact cells (7), and in vivo in whole brain (8) supported this conclusion. However, despite this evidence, recent studies have suggested that Phe could not be hydroxylated by tyrosine hydroxylase (using enzyme purified from PC 12 cells) (9,10) [though this finding is now disputed (11,12)], and that this amino acid is a significant inhibitor of the enzyme (13,14). The issue thus remains unresolved. In the present studies, we have therefore examined further the putative roles of Phe as substrate for and/or inhibitor of tyrosine hydroxylase in intact PC 12 cells. The PC 12 cell line is derived from a rat adrenal medullary pheochromocytoma, and is known to provide a good model for studying catecholamine synthesis in intact cells (15-17). A cell line was chosen for study,

Tyrosine hydroxylase, the initial enzyme in catecholamine biosynthesis, was originally reported to use tyrosine (Tyr) as its only substrate (1). However, phenylalanine (Phe) was subsequently found also to be a substrate, albeit a poor one (2). Following the identification of tetrahydrobiopterin as the natural cofactor for tyrosine hydroxylase, and its use in kinetic studies of the enzyme (instead of the synthetic cofactors used in earlier work), 1

Department of Molecular Biology and Biochemistry. Departments of Psychiatry, Pharmacology and Neuroscience, University of Pittsburgh School of Medicine, Pittsburgh PA. 3 Address reprint requests to: John D. Fernstrom, Ph.D., Western Psychiatric Institute & Clinic, Room 1620, 3811 O'Hara Street, Pittsburgh PA 15213. Phone: (412) 624-2032; FAX: (412) 624-3696; E-Mail: [email protected]. 2

1011 0364-3190/98/0700-1011$15.00/0 © 1998 Plenum Publishing Corporation

1012

rather than conducting experiments in vivo, since Phe is rapidly converted to Tyr in vivo, making it extremely difficult to control the concentrations of each amino acid presented to catecholamine-producing cells (18). We have focused experimentally on tyrosine hydroxylase in the PC 12 cell, using a method developed originally to measure tyrosine hydroxylase activity in vivo (17,19): the linear accumulation of dihydroxyphenylalanine (DOPA) following pharmacologic inhibition of aromatic L-amino acid decarboxylase. And, because Tyr hydroxylation rate and catecholamine synthesis in vivo are known to be sensitive to Tyr concentrations only in neurons that are firing actively (20), PC 12 cells have been studied under both basal and activated conditions. We reasoned that effects of Phe as a substrate might be most evident under conditions in which Tyr's effects have been observed. Finally, the PC 12 cell was selected because of the recent controversy over this cell's ability to synthesize catecholamines from Phe (9-11).

EXPERIMENTAL PROCEDURE Culture Conditions. PC 12 cells (#CRL 1721; American Type Culture Collection, Rockville MD) were maintained in 100 mm collagencoated (21) tissue culture dishes at 37.5°C in an atmosphere of 5% CO2 and 100% humidity. They were fed 3 times weekly with 10 ml of a growth medium consisting of RPMI medium 1640 with 25 mM L-HEPES buffer and L-glutamine (#22400-055, Gibco-BRL, Grand Island NY) supplemented with 10% heat-inactivated equine serum (#A3311-L, HyClone Laboratories, Inc., Logan, UT), 5% fetal bovine serum (#A1115-L, HyClone), and 100 units/ml penicillin G/100 Mg/ml streptomycin sulfate (#15140-015, Gibco/BRL) (15). The cells were passaged weekly (split ratio 1:5): following mild trypsinization (0.025% trypsin in phosphate-buffered saline) for 10 min at 37°C, they were triturated, diluted into growth medium, and then replated. Typical cell densities after one week of growth were 30-40 X 106 cells/dish. New cells were thawed from frozen stocks every 2 months or 10 passages (whichever came first). The cells were studied four days after replating onto 35 mm collagen-coated dishes (21). To achieve almost complete dispersion of cells, replating was preceded by incubation of 100 mm dishes in 0.125% trypsin at 37°C for 7 minutes, followed by trituration and a second 7 minute incubation of the resulting cell suspension, after which the cells were again triturated and added to a 5-fold excess of growth medium. Following centrifugation at 600 X g, the cells were dispersed in fresh growth medium and replated at 2 x 106 cells/35mm dish. Cell viability was >95%, as determined by trypan blue exclusion. The growth medium was replaced three days later. On day three, for DOPA studies, dishes were washed in 1 ml low-potassium (low-K+) Krebs-Ringer HEPES buffer (KRH: 120 mM NaCl, 4.8 mM K.C1, 10 mM D-glucose, 2.5 mM CaCl2, 1.3 mM MgSO4, and 25 mM HEPES buffer, pH 7.4 at 37°C), and preincubated for 30 minutes at 37°C in 0.95 ml of KRH containing appropriate concentrations of amino acids and stimulants (see results). When high K+ concentrations were employed (56 mM KG), the NaCl concentration was adjusted to maintain iso-osmolarity. Thereafter, 0.05 ml of 600 (MM m-hydrox-

DePietro and Fernstrom ybenzylhydrazine (NSD-1015) was added, mixed with gentle swirling, and the dishes were incubated for an additional 15 minutes. At the end of this period, dishes were placed on ice, and 0.1 ml of 2 N perchloric acid (containing 2.5 mM EDTA and 5 mM sodium metabisulfite) was added to each dish (containing both cells and medium). Cells were scraped from the dish surface using a polypropylene policeman (Bel-Art Products, Pequannock NJ), and 1 ml of the resulting suspension was placed in a 1.5 ml tube and centrifuged at 16,000 X g for 15 minutes. The resulting supernatant was used to measure DOPA levels, and the pellet was retained for protein determinations. The procedure for amino acid studies was almost identical: On day three after replating, the growth medium was removed, and the dishes were washed three times with 1 ml warm (37°C) KRH to remove extracellular amino acids. Dishes were then incubated for 30 minutes at 37°C in 1 ml of KRH containing the appropriate concentrations of amino acids and/or stimulants. (NSD-1015 was not included, since preliminary studies showed that the drug had no significant effect on intracellular Tyr or Phe). Incubations were terminated by transferring dishes onto an ice-water bath, removing the incubation media, and then washing each dish rapidly with three 1-ml volumes of ice-cold KRH (the total time for all three washes was less than 40 seconds, to prevent loss of intracellular amino acids (22-24)). One-mi of ice-cold 10% (w/w) trichloroacetic acid (TCA) was then added to each dish, the cells were scraped off of dishes with a polypropylene policemen, and 900 Ml of the resulting suspensions were pelleted by centrifugation at 16,000 X g for 15 minutes. 850 Ml of each supernatant was mixed with 100 Ml of an internal standard solution (3 MM G-methyl-leucine/3 MM norvaline), and then lyophilized overnight. The pellets were retained for protein determinations (25). Chemical Procedures. The DOPA in perchloric acid supernatants was separated and quantitated by HPLC-electrochemical detection. An Econosphere-C18 5 Jim, 150 X 4.6 mm column (Alltech Associates, Deerfield IL) was employed; the mobile phase was 100 mM sodium phosphate (pH 3.0) containing 200 MM octyl sodium sulfate, 500 MM EDTA and 5% methanol, run at 1.2 ml/minute at room temperature using a Waters 510 HPLC pump (Waters Associates, Milford MA). The electrochemical detector was a BAS LC-4B/17AT with an LC22A preheater module (Bioanalytica! Systems, West Lafayette IN); the detector had an applied voltage of 0.7 V; the preheater was set to 30DC. Under these conditions, DOPA typically eluted 5-6 minutes after sample injection. For quantitation of Phe and Tyr, the lyophilized samples were redissolved in 100 Ml H2O; 60 Ml was then mixed with 60 Ml of Fluoraldehyde Reagent Solution (Pierce, Rockford IL) and allowed to react 1 minute at room temperature. The reaction was halted with 30 Ml IM sodium acetate, and samples were filtered (0.45 Mm Microfilterfuge Tubes, Rainin, Woburn MA). 100 Ml of each filtrate was immediately injected onto the HPLC. The HPLC system employed a reversed-phase Adsorbosphere OPA-HR column (5 Mm, 150 mm X 4.6 mm) equipped with a direct-connect cartridge guard column containing an OPA 5 Mm packing material insert (all from Alltech). Separation of the amino acids was achieved by gradient elution using two buffers (buffer A: 90% 100 mM sodium acetate [pH 6.45], 5% acetonitrile, 5% methanol [v:v:v]; buffer B: 30% 100 mM sodium acetate [pH 6.45], 25% acetonitrile, 45% methanol). After sample injection, a linear gradient was run starting at 95% buffer A: 5% buffer B and ending 60 minute later at 8% buffer A: 92% buffer B. The flow rate was 2.2 ml/minute, run at room temperature, using two Waters 6000A pumps and a Waters 680 Gradient Controller. At the completion of each run, the column was restored to starting conditions with a 5 minute reverse gradient. Fluorescent compounds in the eluant were detected on-line using an Hitachi F-2000 fluorescence spectrophotometer (excitation wave-

Phenylalanine Hydroxylation in PC12 Cells

Fig. 1. Effect of Tyr and Phe concentration on DOPA Synthesis in PC12 Cells. PC12 cells were split (at 2 X 106 cells/dish) and grown in 35 mm dishes in RPMI 1640 medium for 3 days. The growth medium was then removed, and replaced with KRH buffer containing either low (4.8 mM, open symbols) or high (56 mM, closed symbols) concentrations of KC1, and Tyr (panel A: 0, 1, 3, 6, 10, 30, 60, 100 or 300 MM) or Phe (panel B: 0, 1, 3, 10, 30, 60, 100, 300 or 1000 MM). Thereafter, dishes were preincubated at 37°C for 30 minutes, after which NSD-1015 was added to a final concentration of 30 MM. The cells were then incubated for 15 minutes, and then terminated as described in the methods section. Each point represents the mean ± s.e.m. for 6 dishes taken from 3 identical studies (duplicate dishes run in each study).

length, 344 nm; emission wavelength, 440 run) equipped with an 18 Ml micro-flow cell (Hitachi Instruments, Danbury CT). The fluorescent signal was quantitated using a Waters/Dynamic Solutions Baseline 810 chromatography workstation (v. 3.3). Using this procedure, Tyr and Phe were reliably separated from other o-phthalaldehyde-positive compounds, and had retention times of 15 and 28 minutes, respectively (the internal standards, norvaline and G-methyl-leucine, had retention times of 25 and 35 minutes, respectively). This method produced linear fluorescence curves from 0.5 to 200 pmol for both Tyr and Phe (0.5 pmol was the smallest amount tested, since baseline samples typically contained ~5 pmol of either amino acid).

RESULTS Initial studies were conducted to optimize the conditions for obtaining linear DOPA accumulation. Poor linearity and reproducibility were obtained when the incubation was begun immediately after cells were switched from RPMI 1640 to KRH medium. A preincubation period eliminated this problem: when cells were incubated in KRH for 30 minutes prior to initiation of DOPA accumulation (by the addition of NSD-1015, 30 MM final concentration), DOPA accumulated linearly for at least 30 minutes in the presence of low (4.8 mM) or high (56 mM) K+ concentrations (using either Tyr or Phe [10 MM] as substrate). Accordingly, a 30 minute preincubation period and a 15 minute incubation period

1013 (following NSD-1015 addition) were selected for quantitating DOPA synthesis. A study was then conducted to examine the influence of NSD-1015 concentration on DOPA accumulation. Cells were preincubated for 30 minutes in KRH, containing either 4.8 mM or 56 mM K+ and 10 MM Phe, and then incubated for 15 minutes with NSD-1015 (0, 5, 10, 20, 30, 50, 100, 300 MM final concentration). Maximal DOPA synthesis (~20 pmol/min/mg protein at 4.8 mM K+; ~50 pmol/min/mg protein at 56 mM K+) was evident even at 5 pM NSD1015, and remained at these levels to at least the 100 MM concentration of the drug. A study using Tyr produced like results. No DOPA accumulated in the absence of NSD-1015. Based on these results, an NSD-1015 concentration of 30 MM was employed in subsequent studies. This dose produced no change in cellular dopamine content during the 15 minute incubation period. The next series of studies examined the relationship between medium substrate concentration and DOPA synthesis. Cells were preincubated for 30 minutes in one of several concentrations of either Tyr or Phe, and either 4.8 or 56 mM K+. NSD-1015 was then added (to 30 MM), and the incubation proceeded for 15 minutes. This study was replicated on three separate occasions, and gave very similar results. The data were therefore combined (Figure 1). In the presence of low K+ concentrations, DOPA synthesis rate increased as medium Tyr concentrations rose to 6 MM, plateaued to 30 MM, and then declined progressively to 300 MM. The peak rate of synthesis was about 22 pmol/min/mg protein. In the presence of high K+, the same general relationship was obtained, except that synthesis rates at any Tyr concentration were greater at 56 mM than at 4.8 mM K+ (Fig. 1A): the maximal DOPA synthesis rate at 56 mM K+ was about 55 pmol/min/mg protein. At 4.8 mM K+, DOPA synthesis also increased with increasing medium Phe concentration, with the peak rate ( ~ 1 5 pmol/min/mg) protein occurring at 10 MM Phe. At higher Phe concentrations, DOPA synthesis declined. At 56 mM K+, DOPA synthesis rates were much greater at any medium Phe concentration, reaching peak values of about 40 pmol/min/mg protein at 10 MM medium Phe (Fig. 1B). DOPA synthesis declined at elevated Phe concentrations, but the effect was marked only at 1000 MM Phe. We next examined the influence on DOPA synthesis of incubating PC 12 cells in the presence of varying concentrations of both Tyr and Phe. The results of one such study are presented in Fig. 2, and employed high K+ to activate the cells. The left panels plot DOPA synthesis as a function of medium Phe concentration (0, 1,

1014

Fig. 2. Effect of Co-varying Medium Tyr and Phe concentrations on DOPA Synthesis under Conditions of Basal and High Potassium Conditions. Cells were split (at 2 X 106 cells/dish) and grown in 35 mm dishes in RPMI 1640 medium for 3 days. The medium was then removed, and replaced with KRH buffer containing low (open circles) or high (56 mM, closed circles) KC1 concentrations, and Tyr (0, 1, 6, 30, or 300 MM) and Phe (0, 1, 10, 100, or 1000 MM) such that at each Tyr concentration, all concentrations of Phe were tested. This study therefore involved 50 conditions, and 100 dishes, since duplicates were run at each condition. Thereafter, dishes were preincubated at 37°C for 30 minutes, after which NSD-1015 was added to a final concentration of 30 MM. The cells were then incubated for 15 minutes, and then terminated as described in the methods section. Panels A —> E plot DOPA synthesis vs medium Phe concentration (panel A, 0 MM Tyr; panel B, 1 MM Tyr; panel C, 6 MM Tyr; panel D, 30 MM Tyr; panel E, 300 MM Tyr); panels F -> J plot the same data as DOPA synthesis vs medium Tyr concentration (panel F, 0 MM Phe; panel G, 1 MM Phe; panel H, 10 MM Phe; panel I, 100 MM Phe; panel J, 1000 MM Phe). The lines connect the mean values of the duplicate data points for each condition.

DePietro and Fernstrom 10, 100, 1000 MM) at each of five medium Tyr concentrations (0, 1, 6, 30, 300 MM, panels A -> E, respectively). The right panels plot the same data set, as a function of medium Tyr concentration at each of the five medium Phe concentrations (panels F —> J). Panel A shows the substrate product curve for Phe in the absence of Tyr. Adding Tyr to the medium, up to 30 MM (Panels B -> D), has the primary effect of increasing DOPA synthesis at each medium Phe concentration, with DOPA synthesis rates rising to the same maximal rates achieved in the presence of Tyr alone (about 60 pmol/min/mg protein). At the highest Tyr concentration (300 MM, panel E), however, synthesis rates drop; only at 1000 MM Phe is no reduction in synthesis noted in the presence of this high concentration of Tyr (compare the 1000 MM Phe points in panels A -> E of Fig. 2). The right panels in Fig. 2 (panels F —> J) show that as the Phe concentration rises, DOPA synthesis at low Tyr concentrations increases progressively, falling only at the highest Phe concentration (panel J). Peak synthesis rates achieved at 6-30 MM Tyr do not rise as the Phe concentration is increased, but they also do not decline, except at the highest Phe concentration (1000 MM; panel J). Finally, at the highest concentration of Tyr (300 MM), increasing medium Phe causes DOPA synthesis to increase, even at the highest Phe concentration (compare 300 MM points proceeding from panels F through J). Essentially identical results were obtained (Fig. 3) when this study was repeated using 2-chloroadenosine, a purinergic agonist (26) and a known activator of tyrosine hydroxylase (27). Further studies were conducted to quantitate intracellular Tyr and Phe concentrations under the conditions employed in these experiments. First, intracellular Phe and Tyr concentrations were monitored during the preincubation period, to ascertain when steady-state concentrations were achieved. When cells were switched from RPMI to KRH containing 10 MM Phe, the intracellular Phe concentration rose and declined rapidly to attain plateau levels within 5 min, which were maintained for at least 60 minutes (Fig. 4A). Tyr concentrations declined immediately, also reaching a plateau within 5 minutes (Fig. 4B). When incubated in the presence of 10 MM Tyr, a similar pattern emerged, though intracellular Tyr concentrations reached plateau values more slowly (Fig. 4D). Phe concentrations dropped rapidly, reaching a plateau within 5 minutes (Fig. 4C). Intracellular Tyr and Phe levels were then quantitated over the range of medium Tyr and Phe concentrations employed in the doseresponse studies of DOPA synthesis. Duplicate dishes of cells were incubated for 30 minutes in KRH medium containing either Phe (0, 1, 10, 100, 1000 MM) or Tyr

Phenylalanine Hydroxylation in PC 12 Cells

Fig. 3. Effect of Co-varying Medium Tyr and Phe concentrations on DOPA Synthesis under Basal Conditions and Incubation in the Presence of 2-Chloroadenosine. Cells were split (at 2 X 106 cells/dish) and grown in 35 mm dishes in RPMI 1640 medium for 3 days. The medium was then removed, and replaced with KRH buffer containing low potassium (open circles) or 10 MM 2-chloroadenosine (closed circles), and Tyr (0, 1, 6, 30, or 300 MM) and Phe (0, 1, 10, 100, or 1000 MM) such that at each Tyr concentration, all concentrations of Phe were tested. Duplicates were run for each condition (50 conditions, 100 dishes). Thereafter, dishes were preincubated at 37°C for 30 minutes, after which NSD-1015 was added to a final concentration of 30 MM. The cells were then incubated for 15 minutes, and then terminated as described in the methods section. Panels A —> E plot DOPA synthesis vs medium Phe concentration (panel A, 0 MM Tyr; panel B, 1 MM Tyr; panel C, 6 MM Tyr; panel D, 30 MM Tyr; panel E, 300 MM Tyr); panels F -> J plot the same data as DOPA synthesis vs medium Tyr concentration (panel F, 0 MM Phe; panel G, 1 MM Phe; panel H, 10 MM Phe; panel 1, 100 MM Phe; panel J, 1000 MM Phe). The lines connect the mean values of the duplicate data points for each condition.

1015

Fig. 4. Time-Course of the Changes in Intracellular Tyr and Phe concentrations in PC 12 Cells incubated in the presence of 10 MM Tyr or Phe. Dishes of cells were split (at 2 X 106 cells/dish) and grown in 35 mm dishes in RPMI 1640 medium for 3 days. The medium was then removed, and replaced with KRH buffer containing either low (4.8 mM, open circles) or high (56 mM, closed circles) concentrations of KC1, and 10 MM Tyr or Phe. In one study (10 MM Phe, panels A and B), dishes were incubated for 1, 3, 5, 15, 30, 45 or 60 minutes at 37°C, and terminated as described in the methods section. In a separate study, (10 MM Tyr, panels C and D), dishes were incubated for 5, 15, 30, 45 or 60 minutes at 37°C, and then terminated. In both studies, cells were harvested from duplicate dishes at 0 minutes, to obtain basal concentrations of each amino acid. Panels A and C present intracellular Phe concentrations, and panels B and D intracellular Tyr concentrations. Each point represents the value obtained for a single dish; duplicate dishes were run at each amino acid concentration for each condition (basal, high K+). The lines connect the mean values of the duplicate data points for each condition.

(0, 1, 6, 30, 100, 300 MM), and were then harvested for amino acid determinations. Incubation in the presence of no added Tyr or Phe produced intracellular levels for each of 175-180 pmol/mg protein, or about 16 MM (using an intracellular water value for PC 12 cells of 11 Ml/mg protein (17)). At 10 MM medium Phe (the concentration at which DOPA synthesis reached its maximum), intracellular Phe rose to about 1300 pmol/mg protein, or about 120 MM (intracellular Tyr was about 375 pmol/mg protein, or about 35 MM) (Fig. 5, C & D). At 1000 MM medium Phe (the concentration at which notable inhibition of DOPA synthesis occurred), the intracellular Phe reached about 15,000 pmol/mg protein, or about 1400 MM (intracellular Tyr was 500-800

1016

Fig. 5. Dose-Response Relationship between Medium Tyr or Phe Concentration and Intracellular Tyr and Phe Concentrations in PC 12 Cells. Dishes of cells were split (at 2 X 106 cells/dish) and grown in 35 mm dishes in RPMI 1640 medium for 3 days. The medium was then removed, and replaced with KRH buffer containing either low (4.8 mM, open circles) or high (56 mM, closed circles) concentrations of KG, and different concentrations of Tyr (0, 1, 6, 30, 100 and 300 MM) [panels A and B] or Phe (0, 1, 10, 100 and 1000 MM) [panels C and D]. Thereafter, dishes were incubated for 30 minutes at 37°C, and terminated as described in the methods section. Panels A and C present intracellular Tyr cone Mirations, and panels B and D intracellular Phe concentrations. Each point represents the value obtained for a single dish; duplicate dishes were run at each amino acid concentration for each condition (basal, high K+). The lines connect the mean values of the duplicate data points for each condition.

pmol/mg protein, or 45-75 MM). At 6 MM medium Tyr (the concentration at which DOPA synthesis reached its maximum), intracellular Tyr rose to about 1200 pmol/mg protein, or about 110 MM (intracellular Phe was about 150 pmol/mg protein, or about 14 MM). At 100 MM medium Tyr (a concentration at which inhibition of DOPA synthesis was clearly evident), intracellular Tyr reached about 10,000 pmol/mg protein, or about 900 H.M (intracellular Phe was still around 150 pmol/mg protein, or 14 MM) (Fig. 5).

DISCUSSION These studies show that PC 12 cells synthesize DOPA when the only amino acid provided in the incubation medium is Phe. As medium Phe concentrations

DePietro and Fernstrom are increased, DOPA synthesis rate increases to attain a maximum value of about 75% that observed when cells are incubated in Tyr (Fig. 1). Intracellular Tyr concentrations also rise, further suggesting that these cells are capable of hydroxylating Phe (Fig. 5). Significant "substrate inhibition" of DOPA synthesis (28) occurs only when the medium Phe concentration is very high (1000 MM).At all concentrations of Phe examined, DOPA synthesis is stimulated substantially in the presence of high potassium concentrations or 2-chloroadenosine. And, when PC 12 cells are incubated in the presence of varying concentrations of both Phe and Tyr, under either basal or high potassium conditions, or in the presence of 2-chloroadenosine, a purinergic agonist (26), Phe acts primarily to promote DOPA synthesis, causing inhibition only at pharmacologically high concentrations (1000 MM) (Fig. 2 and 3). Together, these results support the hypothesis that Phe, like Tyr, behaves as a substrate for tyrosine hydroxylase in PC 12 cells, acting as an inhibitor only at high concentrations (via "substrate inhibition"). Only one other study appears to have examined the relationship between intracellular substrate concentration and tyrosine hydroxylation in PC 12 cells. This study focused exclusively on Tyr (17), showing that the DOPA synthesis rate plateaued at 20 uM (the lowest medium substrate concentration tested), which corresponded to an intracellular Tyr concentration of 50-75 MM. DOPA synthesis then fell gradually but progressively as the medium Tyr concentration was increased to 200 MM (intracellular Tyr = 470 uM). The present study expands on these observations: first, it examined much lower Tyr concentrations, finding that the plateau in DOPA synthesis occurs at 6 MM Tyr (medium concentration), and allowing an estimation of the apparent Tyr Km for DOPA synthesis: using the activated curve in Figure 1A, the Vmax is about 50 pmol/min/mg protein, giving a Km value a little lower than 1 MM medium tyrosine, which corresponds to an intracellular Tyr level of ~220 pmol/mg protein, or ~20 MM (using an intracellular water value of 11 Ml/mg protein for PC12 cells (17)). A Km that is a little below 20 MM agrees surprisingly well with recent estimates obtained using purified PC 12 cell tyrosine hydroxylase (11). The K™ appears to be similar under nonstimulating conditions (Figure 1A), a finding consistent with observations that activation does not modify the Tyr K m o f tyrosine hydroxylase (28). Second, it carried out the medium Tyr concentration to 300 MM, allowing the observation of frank substrate inhibition, which was barely evident in the earlier study at 200 MM Tyr (17). Nevertheless, both PC 12 cell studies concur that intracellular Tyr concentrations must reach 400-500 MM before substrate inhibition emerges, a level higher than that

Phenylalanine Hydroxylation in PC12 Cells described for purified PC12 cell Tyr hydroxylase (—150 (uM (11)). This in vivo-in vitro difference may [a] be real, [b] reflect a degree of imprecision in the quantitation of intracellular amino acid concentrations, or [c] indicate that the concentrations of reactant amino acids in the proximity of the enzyme differ somewhat from those represented by gross intracellular concentrations, issues that cannot be resolved at present. Third, it showed that DOPA synthesis was increased by activating agents (high K+, 2-chloroadenosine) at all Tyr concentrations tested, with no apparent effect on the substrate Km. And fourth, it examined intracellular Tyr levels under low and high K+ conditions, and at a variety of medium Tyr concentrations: intracellular Tyr concentrations were found to be lower under high K+ than low K+ conditions, an effect observed at all Tyr concentrations tested (and seen previously by Vaccaro et al. (17)). Our study then compared the effects of Tyr on DOPA synthesis with those obtained for Phe. Qualitatively similar effects on DOPA synthesis were noted in the presence of Phe: DOPA synthesis rose with increasing medium Phe concentration, plateaued, and then declined. And, high K+ and 2-chloroadenosine stimulated DOPA production at all medium Phe concentrations examined. In addition, several other features were apparent: first, maximal DOPA synthesis in the presence of Phe was about 75% that in the presence of Tyr. Maximal DOPA synthesis rates were achieved at medium (and intracellular) Phe concentrations only a little higher than those for Tyr. Using the data in Fig. 1 and 5, the Km for Phe can be approximated: with a Vmax in activated cells of about 40 pmol/min/mg protein, the Km is a little below 1 uM medium Phe (Fig. 1B), which corresponds to an intracellular Phe level of about 350 pmol/mg protein (Fig. 5D), or 32 uM. The Km for Phe would thus be about 30 uM, a value smaller than that reported for the purified enzyme (100 uM (11)). However, the Km value derived by Ribeiro et al. (11) was for the combined hydroxylation of Phe to Tyr and DOPA; calculated for DOPA production alone, as was the case in the present study, the Km for Phe would have been much lower, approximating that for Tyr (11)). From this perspective, the in vitro and in vivo findings appear compatible. A Km calculation for total hydroxylation of Phe (Tyr + DOPA) in the present study was not possible, since medium Tyr concentrations were not determined. Intracellular Tyr concentrations, however, provide an indication that the Phe Km would be larger if total hydroxylation could have been calculated, since they continued to increase as extracellular Phe concentrations were raised to 1000 uM Phe (Fig. 5C). Hence, the in vivo findings again appear to support results obtained using the puri-

1017 fied enzyme. Overall, the findings are clearly consistent with the idea that Phe is acting as a substrate for tyrosine hydroxylase. Like the findings of Ribeiro et al. (11) and Andersson et al. (12), therefore, our results do not agree with reports that PC 12 cell tyrosine hydroxylase cannot hydroxylate Phe (9,10). Second, Phe appeared less efficacious than Tyr as an inhibitor of tyrosine hydroxylase, when extracellular substrate concentrations were the basis for comparison: at 60 uM medium Tyr, inhibition of DOPA synthesis was just evident, and appeared substantial at 300 (AM. In contrast, inhibition by Phe was marginal at 300 uM, and appeared to be significant only at 1000 uM. However, when intracellular substrate concentrations were the basis for comparison (using the data in Fig. 5), this difference in inhibitory efficacy narrowed considerably: at 60 (AM medium Phe, intracellular Tyr concentration was between 350-700 JAM, while at 300 JAM extracellular Phe, intracellular Phe was around 500 uM (extrapolating in Fig. 5D). Thus, the threshold intracellular concentrations for inhibition appeared quite similar. At an extracellular Tyr concentration of 300 (AM, intracellular Tyr was about 2000 uM; at an extracellular Phe concentration of 1000 (AM, intracellular concentrations were about 1700 (AM. Hence, intracellular concentrations for clear inhibition also appeared similar. If the intracellular substrate concentrations obtained in the present study are like those in the local vicinity of tyrosine hydroxylase molecules within the cell, then Phe would not appear to be remarkably different from Tyr in efficacy with regard to substrate inhibition. If our findings are correct, the inhibitory concentration of Tyr would appear to be somewhat larger than that noted by Ribeiro et al. (11) for PC 12 cell tyrosine hydroxylase in vitro; a comparison for Phe is not possible, since they did not examine substrate concentrations above 200 (AM. It should be noted that the inhibition of DOPA synthesis by high extracellular Phe concentrations is unlikely the result of high intracellular concentrations of Tyr. Even at 1000 (AM extracellular Phe, intracellular Tyr concentrations reached values below 800 pmol/mg protein (Fig. 5C). This concentration corresponds to an extracellular Tyr concentration well below 6 uM (Fig. 5A), which is not in the range of Tyr concentrations associated with substrate inhibition (Fig. 1 A). Phe is thus not inhibiting DOPA synthesis by raising intracellular Tyr concentrations to high levels; clearly, it is also not inhibiting the enzyme by reducing intracellular Tyr concentrations. And third, as noted above, the intracellular Tyr concentration rose as medium Phe concentration was increased (Fig. 5C). This Tyr most likely derived from the

1018 hydroxylation of Phe, since no Tyr was present in the medium, and thus serves as further evidence that tyrosine hydroxylase can hydroxylate Phe. In vivo, a similar phenomenon (Tyr accumulation following intraperitoneal Phe administration) has recently been described in microdialysis fluid obtained from rat brain (29) (though this study may be flawed, since hepatic Phe hydroxylase was not inhibited, and the observed rise in dialysate Tyr could thus have derived from Tyr produced in the liver from Phe (18)). Conceivably, this finding may further suggest that at least some of the Tyr derived from Phe may disassociate from the enzyme, and mix with endogenous Tyr pools. If so, this observation would suggest that tyrosine hydroxylase does not uniformly convert Phe to DOPA without the release of the intermediate Tyr, as has been suggested based on studies in synaptosomes (5). It would fit with the proposal that the order of binding of substrates and cofactors to tyrosine hydroxylase requires that tetrahydropterin bind before the amino acid substrate (30,31) (presumably requiring for Phe conversion to Tyr that the product disassociate before it could be further hydroxylated). Of course, the possibility also exists that this Tyr simply represents amino acid bound to tyrosine hydroxylase, which was released on homogenization of the cells. The present studies do not rule out such a possibility. The dual-substrate experiments provide insight into how Tyr and Phe interact to influence overall DOPA synthesis rate. The studies were conducted using two different means of activating PC 12 cell tyrosine hydroxylase (high potassium, Fig. 2; 2-chloroadenosine, Figure 3), and produced very similar results, suggesting the general validity of the findings. The only inhibition of DOPA synthesis was observed at concentrations of each substrate at which "substrate inhibition" would be expected. For example, Fig. 1A suggests that Tyr should inhibit DOPA synthesis at or above 60-100 MM medium Tyr concentrations. In Fig. 2 and 3, in the curves relating DOPA synthesis to medium Phe concentration (left panels), the maximal rates of DOPA synthesis are not diminished until the Tyr concentration reaches 300 MM (panel E). At lower Tyr concentrations, DOPA synthesis appears enhanced when Tyr is coincubated with Phe; for example, compare panel A, where no Tyr is present, with panel D, where 30 MM Tyr is present. Not only is DOPA synthesis higher in panel D at lower concentrations of medium Phe, it is higher at the concentration of Phe associated with maximal DOPA production. These results thus support the notion that when inhibition is occurring, it is due to substrate inhibition, rather than some other mechanism. The situation appears the same for Phe. In the right panels of Fig. 2 and 3, each curve

DePietro and Fernstrom plots DOPA synthesis against medium Tyr concentration. No inhibition of DOPA synthesis is evident until panel J, which represents an incubation in the presence of 1000 MM Phe. At lower Phe concentrations, DOPA synthesis appears enhanced by Phe at low Tyr concentrations, and unaffected at higher Tyr concentrations. For example, panel I represents coincubation in 100 MM Phe. The highest rate of DOPA synthesis in panel I is comparable to that in Panel F, in which the incubation was conducted in the absence of Phe. But note that at 0 MM and 1 MM Tyr, the presence of 100 MM Phe (panel I) is associated with a much higher rate of DOPA production than when no Phe is present (panel F). As for Tyr, it appears that Phe does not inhibit DOPA production until it reaches the concentration range associated with "substrate inhibition" for this amino acid. Hence, it appears that Phe exerts no significant degree of "competitive inhibition" of Tyr under these conditions. A form of competition does appear to occur, however, which is evident in panels A —> F: As medium Tyr concentrations rise from 0 to 30 MM (panels A —> D), maximal DOPA synthesis rates rise to values achieved in the presence of no added Phe (panel F). This effect suggests that Tyr may be displacing Phe as substrate, consistent with the notion that Tyr has a greater affinity for tyrosine hydroxylase than does Phe. If Phe had a greater affinity, one might expect to see peak DOPA synthesis rates in panels F —> I decline, since Phe would bind preferentially to the enzyme, but hydroxylate to DOPA at a slower rate than would Tyr (Fig. 1). Overall, the findings indicate that Phe and Tyr, even when incubated together, produce inhibition of the hydroxylase in PC 12 cells only at medium concentrations at which each would produce substrate inhibition (300 MM for Tyr, 1000 MM for Phe). Based on these latter considerations, it is of interest to note that at physiologic concentrations of Tyr and Phe (such as those occurring in plasma: 50-150 MM (32)), inhibition from Phe would not be expected, though some inhibition from Tyr might occur. At very high plasma Phe concentrations, such as occur in homozygous phenylketonuria (e.g., ~1400 MM (33)), inhibition of DOPA synthesis would be expected, though inhibition would probably be modest in hyperphenylalanemia (plasma Phe ~ 400 MM (33)). Of course, it is important to note in these dual substrate studies that we could not determine the actual contribution of each amino acid to overall DOPA synthesis. Studies employing radiolabeled amino acids are required for this refinement; these are currently under way. Though Phe was originally posited over thirty years ago to be an inhibitor of Tyr hydroxylase (1), it was soon shown to act as a substrate (34), being hydroxyl-

Phenylalanine Hydroxylation in PC12 Cells ated to both Tyr and DOPA by purified enzyme preparations in the presence of the natural pterin cofactor (3,5). Subsequent studies of cellular elements (synaptosomes) (4-6), cells (7), and whole animals (8,29) affirmed this fact. But, recent work in vivo (13,14) and in enzyme preparations (9,10) has argued that the amino acid should be viewed as an inhibitor of the enzyme, and not a substrate. While other recent studies of purified Tyr hydroxylase reaffirm that Phe is indeed a substrate for the enzyme (11,12), the issue remains incompletely resolved in the "natural setting" (e.g., inside the cell). The present studies have examined Phe's effects on DOPA synthesis inside the whole cell, showing the relationship to be qualitatively and quantitatively similar to that described for purified enzyme preparations (11). The results suggest that like Tyr, Phe behaves as a substrate for DOPA synthesis, inhibiting DOPA synthesis only at high concentrations, in a manner consistent with "substrate inhibition". Our findings thus support the general notion that Phe functions as a substrate for Tyr hydroxylase, rather than simply as an inhibitor.

ACKNOWLEDGMENTS These studies were supported in part by grants from the National Institutes of Health (HD24730) and the NutraSweet Company. F.R.D. was supported by in part by an NIMH Training Grant (MH18273).

REFERENCES 1. Nagatsu, T., Levitt, M., and Udenfriend, S. 1964. Tyrosine hydroxylase: the initial step in norepinephrine biosynthesis. J. Biol. Chem. 239:2910-2917. 2. Ikeda, M., Levitt, M., and Udenfriend, S. 1965. Hydroxylation of phenylalanine by purified preparations of adrenal and brain tyrosine hydroxylase. Biochem. Biophys. Res. Commun. 18:482-488. 3. Shiman, R., Akino, M., and Kaufman, S. 1971. Solubilization and partial purification of tyrosine hydroxylase from bovine adrenal medulla. J. Biol. Chem. 246:1330-1340. 4. Karobath, M., and Baldessarini, R. J. 1972. Formation of catechol compounds from phenylalanine and tyrosine with isolated nerve endings. Nature New Biol. 236:206-208. 5. Katz, I., Lloyd, T., and Kaufman, S. 1976. Studies on phenylalanine and tyrosine hydroxyiation by rat brain tyrosine hydroxylase. Biochim. Biophys. Acta 445:567-578. 6. Kapatos, G., and Zigmond, M. J. 1977. Dopamine biosynthesis from L-tyrosine and L-phenylalanine in rat brain synaptosomes: preferential use of newly accumulated precursors. J. Neurochem. 28:1109-1119. 7. Fukami, M. H., Haavik, J., and Flatmark, T. 1990. Phenylalanine as substrate for tyrosine hydroxylase in bovine adrenal chromaffin cells. Biochem. J. 268:525-528. 8. Bagchi, S. P., and Zarycki, E. P. 1970. In vivo formation of tyrosine from phenylalanine in brain. Life Sci. 9(I): 111—119. 9. Dix, T. A., Kuhn, D. M., and Benkovic, S. J. 1987. Mechanism of oxygen activation by tyrosine hydroxylase. Biochemistry. 26: 3354-3361.

1019 10. Kuhn, D. M., and Billingsley, M. L. 1987. Tyrosine hydroxylase: Purification from PC-12 cells, characterization and production of antibodies. Neurochem. Int. 11:463—475. 11. Ribeiro, P., Pigeon, D., and Kaufman, S. 1991. The hydroxyiation of phenylalanine and tyrosine by tyrosine hydroxylase from cultured pheochromocytoma cells. J. Biol. Chem. 266:16207-16211. 12. Andersson, K. K., Vassort, C., Brennan, B. A., Que, L. J., Haavik, J., Flatmark, T., Gros, F., and Thibault, J. 1992. Purification and characterization of the blue-green rat phaeochromocytoma (PC 12) tyrosine hydroxylase with a dopamine-Fe(III) complex. Reversal of the endogenous feedback inhibition by phosphorylation of serine-40. Biochem. J. 284:687-695. 13. Milner, J. D., Irie, K., and Wurtman, R. J. 1986. Effects of phenylalanine on the release of endogenous dopamine from rat striatal slices. J. Neurochem. 47:1444-1448. 14. Wurtman, R. J., and Maher, T. J. 1987. Effects of oral aspartame on plasma phenylalanine in humans and experimental rodents. J Neural Transm 70:169-173. 15. Greene, L. A., and Tischler, A. S. 1976. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc. Natl. Acad. Sci. U.S.A. 73:2424-2428. 16. Greene, L. A. and Tischler, A. S. 1982. PC12 pheochromocytoma cultures in neurobiological research. Adv. Cell. Neurobiol. 3:373414. 17. Vaccaro, K. K., Liang, B. T., Perelle, B. A., and Perlman, R. L. 1980 Tyrosine-3-monooxygenase regulates catecholamine synthesis in pheochromocytoma cells. J. Biol. Chem. 255:6539-6541. 18. Fernstrom, M. H., Baker, R. L., and Fernstrom, J. D. 1989. In vivo tyrosine hydroxyiation rate in retina: effects of phenylalanine and tyrosine administration in rats pretreated with p-chlorophenylalanine. Brain Res. 499:291-298. 19. Carlsson, A., Davis, J. N., Kehr, W., Lindqvist, M., and Atack, C. V. 1972. Simultaneous measurement of tyrosine and tryptophan hydroxylase activities in brain in vivo using an inhibitor of the aromatic amino acid decarboxylase. Naunyn Schmied. Arch. Pharmacol. 275:153-168. 20. Fernstrom, J. D., and Fernstrom, M. H. 1988. Tyrosine availability and dopamine synthesis in the retina, pages 59-70, in Bodis-Wollner, I. and Piccolino, M. (eds.), Dopaminergic Mechanisms in Vision, Alan R. Liss, New York. 21. Greene, L. A., Aletta, J. M., Rukenstein, A., and Green, S. H. 1987. PC12 pheochromocytoma cells: culture, nerve growth factor treatment, and experimental exploitation. Meth. Enzymol. 147: 207-216. 22. Eagle, H., Piez, K. A., and Levy, M. 1961. The intracellular amino acid concentrations required for protein synthesis in cultured human cells. J. Biol. Chem. 236:2039-2042. 23. Drummond, R. J., and Phillips, A. T. 1977. Intracellular amino acid content of neuronal, glial, and nonneural cell cultures: the relationship to glutamic acid compartmentation. J. Neurochem. 29: 101-108. 24. Jara, J. R., Martinez-Liarte, J. H., Solano, F., and Penafiel, R. 1990. Transport of L-tyrosine by B16/F10 melanoma cells: the effect of the intracellular content of other amino acids. J. Cell Sci. 97:479-485. 25. Lowry, O. H., Rosebrough, M., Fair, A., and Randall, R. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 26. Fredholm, B. B., Abbracchio, M. P., Burnstock, G., Daly, J. W., Harden, T. K., Jacobson, K. A., Leff, P., and Williams, M. 1994. Nomenclature and classification of purinoceptors. Pharmacol. Rev. 46:143-156. 27. Erny, R. E., Berezo, M. W., and Perlman, R. L. 1981. Activation of tyrosine 3-monooxygenase in pheochromocytoma cells by adenosine. J. Biol. Chem. 256:1335-1339. 28. Kaufman, S., and Kaufman, E. E. 1985. Tyrosine Hydroxylase, pages 251-352, in Blakley, R. L., and Benkovic, S. J. (eds.), Fo-

1020 lates and Pterins, Volume 2: Chemistry and Biochemistry of the Pterins, John Wiley and Sons, New York. 29. Westerink, B. H. C , and De Vries, J. B. 1991. Effect of precursor loading on the synthesis rate and release of dopamine and serotonin in the striatum: A microdialysis study in conscious rats. J. Neurochem. 56:228-233. 30. Fitzpatrick, P. F. 1991. Steady-state kinetic mechanism of rat tyrosine hydroxylase. Biochemistry 30:3658-3662. 31. Meyer, M. M., and Fitzpatrick, P. F. 1992. The amino acid substrate of bovine tyrosine hydroxylase. Neurochem. Int. 21:191196.

DePietro and Fernstrom 32. Fernstrom, J. D., Wurtman, R. J., Hammarstrom Wiklund, B., Rand, W. M., Munro, H. N., and Davidson, C. S. 1979. Diurnal variations in plasma concentrations of tryptophan, tryosine, and other neutral amino acids: effect of dietary protein intake. Am. J. Clin. Nutr. 32:1912-1922. 33. Caballero, B., Mahon, B. E., Rohr, F. J., Levy, H. L., and Wurtman, R. J. 1986. Plasma amino acid levels after single-dose aspartame consumption in phenylketonuria, mild hyperphenylalaninemia, and heterozygous state for phenylketonuria. J. Pediat. 109:668-671. 34. Ikeda, M., Levitt, M., and Udenfriend, S. 1967. Phenylalanine as substrate and inhibitor of tyrosine hydroxylase. Arch. Biochem. Biophys. 120:420-427.