Joirrnal of Neitrochernisrry Raven Press, Ltd., New York Q 1992 International Society for Neurochernistry

Tetrahydrobiopterin Turnover in Cultured Rat Sympathetic Neurons: Developmental Profile, Pharmacologic Sensitivity, and Relationship to Norepinephrine Synthesis Gregory Kapatos, Kei Hirayama, and “Hiroyuki Hasegawa Cellular and Clinical Neurobiology Program, Department of Psychiatry, Wayne State University School ofhfedicine, Detroit. Michigan, U.S.A.;and *Department of Bioscience, The Nishi Tokyo University, Yamanashi, Japan

Abstaacb: We have examined the turnover of 5,6,7,8-tetizhydrobiopterin (BH4) and the effect of decreasing BH4 levels on in situ tyrosine hydroxylase (TH) activity and norepinephrine (NE) content in a homogeneous population of NE-containing neurons derived from the superior cervical ganglion (SCG) ofthe neonatal rat and maintained in tissue culture. Initial studies indicated that the level of BH4 within SCG cultures increased fourfold between 5 and 37 days in vitro (DIV). This increase in BH4 levels was determined to result from an increase in the rate of BH4 biosynthesis without a change in the rate of degradation. Regardless of culture age, the BH4 content of SCG neurons was observed to turn over with a half-life of -2.5 h. BH4 synthesis by SCG neurons was found to be five times more sensitive to inhibition by 2,4-diamino-6-hydroxypyrimidine (DAHP) and 25 times less sensitive to inhibition by N-acetylserotonin than was previously reported for CNS neurons in culture. Under basal conditions, the rates of in situ T H activity and BH4 biosynthesis were similar. In response to inhibition of BH4 biosynthesis by DAHP and a 90-95% decrease in BH4 levels, in situ T H activity declined by 75%. NE levels declined

by 30%following a 24-h period ofinhibition of BH4 synthesis. After 2 days of BH4 synthesis inhibition, the level of NE was decreased by 47%. On treatment days 3 and 4, the decline in NE content plateaued at 24% of control levels. In contrast, treatment of cultures for 24 h with the direct-acting inhibitor of TH, a-methyl-p-tyrosine, produced an 84% decline in NE content that was maintained over the 4-day treatment period, indicating that the slow decline in NE content following inhibition of BH4 synthesis was not the result of the slow turnover rate of NE. These results demonstrate that despite an almost complete loss of BH4, sympathetic neurons were able to maintain neurotransmitter content, albeit at reduced levels, by retaining a level of T H activity above the value that might have been predicted based on the reduced level of BH4. Key Words: 5,6,7,8Tetrahydrobiopterin-Norepinephrine-Superior cervical ganglia in culture-Tyrosine hydroxylase-Turnover. Kapatos G. et al. Tetrahydrobiopterin turnover in cultured rat sympathetic neurons: Developmental profile, pharrnacologic sensitivity, and relationship to norepinephrine synthesis. J. Neurochem. 59, 2048-2055 ( 1992).

The basal activity of tyrosine hydroxylase (TH; EC tion of this enzyme by phosphorylation (Iuvone et al., 1985) are both dependent, in part, on the intracellular concentration of the pteridine cofactor, tetrahydrobiopterin [6-(R)-(~-erythro-1’,2’-dihydroxypropy1)-2amino-4-hydroxy-5,6,7,8-tetrahydropteridine; BH41. Although the concentration of BH4 within monoamine-containing neurons is unknown, it has been estimated to be 18 pM within cells of the adrenal medulla (Abou-Donia et al., 1986), 30 pLMwithin cells of

the pineal gland (Kapatos et al., 1982), and 1 15 pM within N 1E 1 15 neuroblastoma cells (Kapatos and Kaufman, 1983), all values well within the range of the Michaelis constant for BH4 of the activated form of TH (Zigmond et al., 1989). The cellular mechanisms responsible for maintenance of the concentration of BH4 within a range permissive for the allosteric regulation of TH have not yet been identified. A low biosynthetic rate, perhaps due to inhibition of BH4 biosynthesis by BH4, has been proposed (Kapatos and Kaufman, 1983). Re-

Received February 7, 1992; revised manuscript received April 27, 1992; accepted May 1 I , 1992. Address correspondence and reprint requests to Dr. G. Kapatos at Center for Cell Biology, Sinai Hospital, 6767 West Outer Drive, Detroit, MI 48235,U.S.A.

Abbreviations used: AMPT, a-methyl-p-tyrosine; BH4,5,6,7,8tetrahydrobiopterin; DAHP, 2,4diamino-6-hydroxypyrimidine; DIV, days in vitro; DOPA, 3,4-dihydroxyphenylalanine;NAS, Nacetylserotonin; NE, norepinephrine; SCG, superior cervical ganglion; TH, tyrosine hydroxylase.

1.14.16.2) (Kaufman, 1974) and the allosteric regula-

2048

BH4 TURNOVER IN CULTURED SYMPATHETIC NEURONS cent observations, however, indicate that the rate of synthesis of BH4 within neurons derived from the CNS and maintained in culture is actually quite rapid and that the low intracellular concentration of BH4 is the result of an equally rapid rate of BH4 degradation (Kapatos, 1990). Because of the heterogeneity of neuronal cell types within cultures of the CNS and the question of the cellular localization of BH4, no definitive conclusions could be reached from this earlier work regarding the turnover of BH4 within monoamine-containing neurons. Therefore, we have examined BH4 and norepinephrine (NE) turnover within a homogeneous population of neurons derived from the superior cervical ganglion (SCG) of the neonatal rat and maintained in tissue culture.

MATERIALS AND METHODS Cell culture Rat pups 1 or 2 days old, obtained from timed pregnant Sprague-Dawley dams (Hilltop Farms, Scottdale, PA, U.S.A.), were killed by cervical dislocation. Unless stated otherwise, all tissue culture reagents were obtained from GIBCO/BRL (Grand Island, NY, U.S.A.). SCG tissues were removed bilaterally and collected in Ham’s F12 medium containing 25 mM HEPES, 100 units/ ml of penicillin, and 100 pg/ml of streptomycin. Ganglia from 20 animals were transferred to 2 ml of a solution composed of calcium and magnesium-free Hanks’ balanced salts, 25 rnM HEPES, 25 mM glucose, 2 mM glutamine, and antibiotics (Sol A) to which was added 0.1% trypsin, 0.190 collagenase, and 0.01% DNase (Worthington Biochemical Co., Freehold, NJ, U.S.A.). Ganglia were incubated at 37°C for 15 min and then stripped of their outer membranes, and incubation was continued for 30 min. Following collection of tissue by centrifugation, partially digested ganglia were resuspended in 2 ml of Sol A containing 0.3% trypsin and incubated at 37°C for an additional 20 min. Trypsin was then inactivated by addition of 1.5 ml of Sol A containing 0.6% soybean trypsin inhibitor. Ganglia were centrifuged, resuspended in 1.5 ml of Ham’s F12 medium containing 25 rnMHEPES and antibiotics, and triturated with fire-polished Pasteur pipettes of decreasing bore diameter. Cells were resuspended in growth medium composed of Ham’s F12 medium containing 10% heat-inactivated fetal calf serum, 5.4 pA4 linoleic acid, 2 mM glutamine, antibiotics, and 100 ng/ml of 7s nerve growth factor (Collaborative Research, Bedford, MA, U.S.A.) (Kessler, 1985). Cell number and viability were determined, and cell density was adjusted to 200,000 viable neurons/ml. Two hundred microliters of the cell suspension containing 40,000 neurons was dispensed into individual wells of 96-well tissue culture plates previously coated with bovine type I collagen (Collaborative Research). Cultures were maintained at 37°C in a humidified atmosphere of 5% CO, and received a 50% medium change every other day. To eliminate nonneural cells, cultures were incubated with 10 pA4 cytosine P-Darabinofuranoside (Sigma, St. Louis, MO, U.S.A.) for 4 days. Cultures were typically taken for biochemical analysis after 1227 days in vitro (DIV).

2049

BH4 quantification The biopterin content of individual cultures was determined by a modification of the method of Fukishima and Nixon ( 1980) as described previously (Kapatos, 1990). In brief, culture plates were placed on ice, individual wells were rinsed with ice-cold Dulbecco’s phosphate-buffered saline, and cells were harvested under subdued lighting by sonication in 150 p1 of ice-cold 0.1 M HCl containing a known amount of 2-amino-4-hydroxypterin (pterin) as an internal standard. Preliminary studies showed that endogenous pterin levels were barely detectable; they did not attain levels > 1% of the pterin added. Following sonication, 10 pl of a 1% iodine/290 potassium iodide solution in 0.1 MHCl was added to each well. After 1 h at room temperature in the dark, residual iodine was reduced by addition of 35 pl of a 1% ascorbic acid solution in 10%perchloric acid. The entire tissue culture plate was then centrifuged to precipitate protein. Aliquots of the supernatants were filtered and then directly analyzed for biopterin and pterin content by reverse-phase HPLC with fluorescence detection. In most experiments, the precipitates were dissolved in 1 M NaOH and assayed for protein content by the method of Bradford ( 1976) with bovine serum albumin as the standard. In some experiments, the protein content of each well was not determined but was estimated from the average of previous cultures of similar age. BH4 content was corrected for recovery of the pterin internal standard and expressed as femtomoles of BH4 per well or femtomoles of BH4 per microgram of protein.

NE levels and in situ TH activity Culture plates were placed on ice, and individual wells were rinsed with ice-cold Dulbecco’s phosphate-buffered saline. To each well was added 0.1 ml ofO. 1 Mperchloric acid containing a known amount of 3,4-dihydroxybenzylamine (Sigma) as an internal standard. Following sonication and centrifugation, NE and 3,4-dihydroxybenzylaminein the supernatant were quantified by HPLC with electrochemical detection (Michaud et al., 1981). NE levels were corrected for recovery of 3,4-dihydroxybenzylamine and are presented as femtomoles of NE per microgram of protein. In situ T H activity was monitored as a measure of NE biosynthesis by following the accumulation of 3,4-dihydroxyphenylalanine (DOPA) after inhibition of L-aromatic amino acid decarboxylase activity. Cultures received 75 p1 of serum-free growth medium containing the L-aromatic amino acid decarboxylase inhibitor NSD- 1015 (Sigma) at 150 p M and were returned to the incubator for 3 h. This reaction was terminated on ice, 75 pI of 0.2 M perchloric acid containing a-methyl-DOPA (Sigma) as an internal standard was added directly to the culture medium, and cells and medium were mixed by sonication. Following centrifugation, a-methyl-DOPA and DOPA were extracted by alumina chromatography and quantified by HPLC with electrochemical detection (Michaud et al., 198 I). DOPA accumulation was linear for at least 5 h of incubation. DOPA levels in the absence of NSD- 1015 represented < 1% of that accumulated following incubation with the inhibitor. In situ TH activity was corrected for recovery of amethyl-DOPA and expressed as femtomoles of DOPA per microgram of protein per hour.

Presentation of results Three to five individual culture wells were analyzed for each group within an experiment, and, with the exception of J. Neurochem.. Vol. 59. No. 6. 1992

G. KAPATOS ET AL.

2050

the developmental time course and BH4 turnover studies, which were performed once, experiments were conducted at least twice with cultures produced on different plating days. Data from representative experiments are presented as mean k SD values. Kinetic parameters describing BH4 metabolism were determined as previously described (Kapatos, 1990).In analysis ofthe first-order decline ofBH4 content following synthesis inhibition, the fractional rate constant k was calculated by nonlinear regression analysis (ENZFITTER; Elsevier). In all cases, data were fit best by a first-order equation. Knowledge of k and the steady-state level of BH4 enabled calculation of the remaining parameters required to describe BH4 metabolism. Under steady-state conditions, K = k[BH4],, where K is the rate constant for BH4 synthesis (in fmol/pg of protein/h), k is the fractional rate constant for loss (h-I), and [BH4Io is the resting concentration of BH4 (in fmol/pg of protein). Turnover time, the time required for the synthesis ofan amount ofBH4 equal to that stored in the neurons at steady state, was calculated as the reciprocal of k. The half-life was calculated as (ln2)lk. In studies of the rate of recovery of BH4 following termination of synthesis inhibition, the first-order rate constant k was determined based on the equations [BH4], = K/k(l - e-"") and In[BH4]o/([BH4]o - [BH4],) = kt, where K and k are as described above and the slope of the plot of the latter function yields k directly (Sladeczek and Bockaert, 1983).

RESULTS In preliminary studies, phase-contrast microscopy and immunohistochemistry for TH demonstrated that treatment of cultures with the mitotic inhibitor cytosine p-Darabinofuranoside eliminated support cells and resulted in cultures composed exclusively of TH-positive neurons. This is an important consideration because nonneuronal cells, such as fibroblasts, may contain low levels of BH4 (Werner et al., 1990). The kinetic parameters describing BH4 turnover reported here are therefore derived from a homogeneous population of NE-containing neurons. In addition, the amount of protein per culture well was found to remain relatively constant between 12 and 27 DIV (7.1 k 1.2 and 5.2 +- 0.5 pg, respectively). Thus, the observations described here would also appear to be based on a culture system that has stabilized in terms of neuron number and size. Initial studies demonstrated that the BH4 content of SCG cultures increased from 860 -+ 64 to 3,775 k 245 fmol per well, or greater than fourfold, between 5 and 37 DIV (Fig. 1). As determined by the differential oxidation technique (Fukushima and Nixon, 1980),>98% ofthe biopterin synthesizedby these neurons was in the fully reduced tetrahydro form. This increase in BH4 levels during a period where protein content remained relatively constant suggested that some alteration in BH4 metabolism, perhaps the result of an increase in synthesis rate or a decline in the rate of degradation, was responsible for the observed increase in cofactor levels. The following experiments were designed to test this possibility. J. Neurochem., Vol. 59, No. 6. 1992

0 0

10

20

30

3

DAYS IN VlTRO

FIG. 1. BH4 content of sympathetic neurons maintained in culture. Neurons were plated on six separate 96-well plates at a density of 40,000cells per well. Cultures were harvested following 5, 10, 15, 24, 28, and 36 DIV, and BH4 content was determined. Data are mean -c SD (bars) values of five determinations.

2,4-Diamino-6-hydroxypyrimidine (DAHP), an inhibitor of the first and rate-limiting enzyme in BH4 biosynthesis, GTP cyclohydrolase (GAI et al., 1978), and N-acetylserotonin (NAS), an inhibitor of the last enzyme in BH4 biosynthesis, sepiaptenn reductase (Katoh et al., 1982), have been used previously to establish rate constants for BH4 synthesis and degradation within neurons maintained in culture (Kapatos, 1990). To examine the sensitivity of SCG neurons to DAHP and NAS and to determine the concentration of these compounds necessary to produce maximal inhibition of BH4 synthesis, 12-DIV cultures were incubated for 2 h with 0.2-10 mM DAHP or 0.2-2.5 mM NAS. This range of concentrations was chosen based on previous studies of CNS neurons maintained in culture (Kapatos, 1990).The 2-h incubation time was chosen to insure that initial rates of BH4 decline were evaluated. At all concentrations tested, DAHP produced a decrease in BH4 content (Fig. 2). The maximal decline was observed at a minimal DAHP concentration of 1 mM. All further studies used a DAHP concentration of 10 mM. Although less potent than DAHP, NAS was also able to produce a concentration-dependent decline in BH4 levels and at 2.5 mM was equipotent to 1 mM DAHP (Fig. 2, inset). Because this concentration of NAS was 25-fold higher than that required to produce a similar level of inhibition in central neurons (Kapatos, 1990), no other studies with NAS will be reported here. Two experimental paradigms, both based on steady-state kinetic analyses following the inhibition of BH4 biosynthesis and previously demonstrated to yield equivalent results (Kapatos, 1990), were used to detect differences in BH4 turnover between 12- and 27-DIV cultures. In the first study, the decline-fromsteady-state paradigm was used. Cultures after 12 DIV were incubated with 10 mM DAHP for periods ranging from 0.33 to 5.75 h. Control cultures that received a complete medium change and DAHPtreated cultures were harvested at these times. The BH4 content of control cultures remained constant

BH4 TURNOVER IN CULTURED SYMPATHETIC NEURONS

205 1

+ ++- o z

J

0 0

1

2

3

4

5

6

7

2.4-DIAMINO-6-HYDROXYF'YRlMlOlNE

8

9

10

0

1

over this incubation period (1,788 -t 173 fmol per well, 35 1 fmol/pg of protein), indicating that the repeated transfer of the culture plate to and from the incubator, which is required for harvesting individual culture wells at different time points, did not alter BH4 levels (Fig. 3). BH4 levels declined by 80% following 5.75 h of incubation in the presence of DAHP. Nonlinear regression analysis of these data was fit best by an equation describing a first-order decline from steady state, from which could be derived a fractional rate constant of loss (k)equal to 0.27 1 f0.009/h. This value for k indicated that -27% of the intracellular pool of BH4 within 12-DIV cultures turns over per hour. The first-order decline in BH4 content following inhibition of synthesis also suggests that a single pool of BH4 exists within these neurons. An alternative method to investigate BH4 metabolism was used with cultures generated from the same cell plating as above but maintained in culture for an additional 15 days. This method required monitoring the recovery of BH4 following a period ofBH4 synthesis inhibition. Cultures after 27 DIV were incubated with 10 mMDAHP for 18 h. Untreated cultures contained 3,559 & 84 fmol of BH4 per well (698 fmol of BH4/pg of protein), a level twice that observed at 12 DIV. An 18-h incubation with DAHP decreased the BH4 content to t10% of control values. After an extensive washout of inhibitor, cultures were placed in fresh medium, returned to the incubator, and harvested 1-7 h later, and the BH4 content was determined. BH4 levels recovered without a pause during this period, indicating that the level of any residual intracellular DAHP had been reduced to a concentration below that required to produce persistent inhibition of BH4 biosynthesis (Fig. 4). Levels of BH4 in control cultures were again relatively unaffected by

3

4

5

6

INCUBATION TIME (HOURS)

(mM)

FIG. 2. Effect of incubation with 0.2-1 0 mM DAHP on levels of BH4. Cultures after 12 DIV were incubated for 2 h with DAHP in growth medium and harvested, and BH4 content was determined. Results are presented as percentages of control levels (1,050 f 216 fmol of BH4/pg of protein). Data are mean -t SD (bars) values of four determinations. Inset: Effect of incubation with 0.2-2.5 mM NAS on levels of BH4. Cultures after 12 DIV were incubated for 2 h with NAS in growth medium and harvested, and BH4 content was determined. Results are presented as percentagesof control levels (2,011 -t 127 fmol of BH4/pg of protein). Data are mean f: SD (bars) values of three determinations.

2

FIG. 3. Time course of the decline in BH4 levels following incubation with DAHP. Culturesafter 12 DIV were incubatedwith 10 mM DAHP in growth medium and harvested 0.33,0.75, 1.25,2.0,3.0, 4.25, and 5.75 h later, and BH4 content was determined (0). Control cultures received a medium change and were harvested at identical times (0).Results are presented as percentages of control levels (1,788 f 173 fmol of BH4 per well, 351 f 41 fmol of BH4/pg of protein). Data are mean f SD (bars) values of four determinations.

repeated handling of the culture plate. Nonlinear regression analysis of the recovery of BH4 allowed calculation of a first-order fractional rate constant of 0.289 t 0.060/h, a value virtually identical to that obtained by the decline-from-steady-state paradigm. The effect of a 90-95% decrease in BH4 content on the capacity of sympathetic neurons to synthesize and maintain levels of NE was also investigated. Initial experiments investigatedthe in situ activity of TH as a measure of NE biosynthesis by monitoring the accumulation of DOPA after inhibition of L-aromatic amino acid decarboxylase. Cultures were incubated with 10 mM DAHP for 24 h, at which time fresh medium containing DAHP and L-aromatic amino acid decarboxylase inhibitor was added. Cultures were then returned to the incubator for 3 h, after which cells and medium were acidified and harvested,

0'

'

'

'

0

1

2

.

3

'

'

'

'

4

5

6

7

J

INCUBATION TIME (HOURS) FIG. 4. Time course of the recovery of BH4 levels following termination of incubation with DAHP. Cultures after 27 DIV were incubated with 10 mM DAHP for 18 h, rinsed extensively in growth medium to washout residual DAHP, incubated in fresh growth medium, and harvested 1, 2, 3, 4, 5,6, and 7 h later, and BH4 content was determined (0).Control cultures received identical medium changes but were not incubated with DAHP (0).Results are presented as percentages of control levels (3,559 f 84 fmol of BH4 per well, 698 f 17 fmol of BH4/pg of protein). Data are mean k SD (bars) values of four determinations.

J. Neurochem.. Vol. 59. No. 6, I992

G. KAPATOS ET AL.

2052

TABLE 1. Summary of rate constants for BH4 and NE turnover in cultures of rat SCG Steady-state level (frnol/pg of protein)

Fractional rate constant

k (h-')

Synthesis rate K (fmollpg of protein/h)

Turnover time (h)

Half-life (h)

BH4 12 DIV 27 DIV

351 + 4 1 698 f 17

0.271 0.01 0.289 f 0.06

95 20 1

3.69 3.46

2.56 2.40

N E (range)

2,621-4,639

0.05-0.07

227-307

15-20

10-14

*

BH4 steady-state levels and fractional rate constants are given as mean t SE values. NE values are given as ranges.

and DOPA production was quantified. Under basal conditions, the in situ rate of TH ranged from 227 k 21 to 307 -t 25 fmol of DOPA/pg of protein/h in 12- and 2 1-DIV cultures, respectively. The rate of NE biosynthesis was therefore roughly equivalent to the rate of BH4 biosynthesis (95-201 fmol/pg of protein/ h) determined in cultures of similar age (Table 1). Depletion cf BH4 levels by 90-95% was found to decrease in situ tyrosine hydroxylation by 75% (Fig. 5). The long-term effect of BH4 depletion and the accompanying decline in NE synthesis on NE levels were then studied. Cultures after 12 DIV were treated with 10 mM DAHP for 1-4 days and harvested for NE analysis. Similar cultures were incubated with a direct-acting inhibitor of TH, 200 pM a-methyl-p-tyrosine (AMPT). Long-term treatment with DAHP or AMPT appeared not to alter cell viability over the 4-day incubation period. Under control conditions, the content of NE ranged from 2,621 456 to 4,639 +- 33 1 fmol/pg of protein. Although not determined directly, based on average steady-state levels of NE and rates of NE synthesis, the average fractional rate constant for NE was estimated to be 0.06 h-' (Table 1). From this estimate of k,the average turnover time and half-life of NE were estimated to be 18 and 12 h, respectively (Table 1). Incubation for 24 h with DAHP, however, produced only a 30%decline in NE

*

levels rather than the 75% decline predicted from the estimated half-life (Fig. 6). Following 2 days of DAHP treatment, the level of NE declined by 47%. On treatment days 3 and 4, NE content reached a plateau at 24% of control levels. Treatment of cultures with AMPT produced an 84% decline in NE content after 24 h of incubation; this reduction was maintained over the treatment period. Thus, the slow decline in NE content following inhibition of BH4 biosynthesis was not the result of the slow turnover rate of NE.

DISCUSSION Under steady-state conditions the rate of BH4 biosynthesis and loss are equivalent. Knowledge of the steady-statelevel of BH4 and the fractional rate of loss permitted calculation of the rate of BH4 biosynthesis. Steady-state levels and kinetic parameters describing BH4 turnover in SCG neurons maintained in culture are presented in Table 1. The decline-from-steadystate and recovery-of-steady-state paradigms yielded fractional rate constant, k, values that were identical for 12- and 27-DIV cultures, indicating that at either DIV -28% of the BH4 pool turns over every hour.

100

I z

100

0

kjzi

0-0

DAMP AMPT

0-0

TO \

a

1

2

80

30 32 00

2;

60 40

Q-

8 n

20

01 0

3

4

TREATMENT PERIOD (DAYS) 0

FIG. 5. Effect of inhibition of BH4 biosynthesison in situ TH activity. Cultures after 21 DIV were incubated with 10 mM DAHP for 18 h in growth medium. Medium was then replaced with serumfree growth medium containing 10 mM DAHP and 150 pM NSD1015, and cultures were returned to the incubator for 3 h. Cultures were then harvested by sonication, and DOPA accumulation was determined as a measure of in situ TH activity. Results are presented as percentages of control levels (307 f 25 fmol of DOPA/pg of protein/h). Data are mean t SD (bars) values of four determinations.

J. Neurochem.. Vol. 59, No. 6,1992

FIG. 6. Time course of the decline in NE levels in response to inhibition of BH4 biosynthesis by DAHP or AMPT. Cultures after 12 DIV were incubated with 10 mM DAHP in growth medium for 1-4 days, at which time medium was removed and cells were harvested for NE analysis (0).Alternatively, in a separate experiment, cultures were incubated with 200 pM AMPT in growth medium for 1-4 days, at which time medium was removed and cells were harvested for NE analysis (0).Results are presented as percentages of control levels (4,639 2 331 fmol of NE/pg of protein for DAHP and 4,485 ? 515 fmol of NE/pg of protein for AMPT). Data are mean -C SD (bars) values of four determinations.

BH4 TURNOVER IN CULTURED SYMPATHETIC NEURONS The fractional rate constant for loss may reflect processes as diverse as enzymatic degradation of BH4 (Rembold, 1982) and diffusion out of the cell. The first-order decline of BH4 content can also be taken as evidence that a single pool of BH4 exists within these cultures. Because the cultures contain a homogeneous population of neuronal cell bodies and nerve terminals, this suggests that the turnover rate of BH4 is similar in these cellular compartments. The turnover-time (3.6 h) and half-life (2.5 h) of BH4, both derived from k, were also equivalent at 12 and 27 DIV. The synthesis rate of BH4 ( K ) , calculated as the product of k and the steady-state level of BH4, was twice as rapid in 27- (210 fmol of BH4/pg of protein/ h) as in 12- (95 fmol of BH4/pg of protein/h) DIV cultures. The twofold increase in the steady-state level of BH4 between 12 and 27 DIV appeared therefore to result from a doubling in the rate of BH4 synthesis without a similar acceleration in the rate of BH4 degradation. Although the rates of BH4 and NE biosynthesis (227-307 fmol/pg of protein) were similar, the eightfold difference in steady-state levels of NE and BH4 could be attributed to the fourfold slower rate of NE turnover (Table 1). Inasmuch as developmental increases in BH4 content have been observed previously in rat brain and pineal gland in vivo and have been correlated with increases in the activity of GTPcyclohydrolase (Kapatos et al., 1983), the developmental control of BH4 biosynthesis may be common in many cell types. There are several notable differences between CNS and SCG neurons regarding BH4 metabolism and the response to inhibitors of BH4 biosynthesis. Owing to the heterogeneity of neuronal cell types present in cultures of the CNS, no comparison between central and SCG neurons can be made regarding the rate of BH4 biosynthesis. Comparisons can be made, however, between fractional rate constants, because these values are independent of whether BH4 content is expressed on a per well, protein, or cell basis. BH4 metabolism within CNS neurons maintained in culture exhibited an average k of 0.16 h-' with a turnover time and half-life of 6.6 and 4.5 h, respectively (Kapatos, 1990). The turnover rate observed for BH4 within SCG neurons was thus almost twice as rapid as that determined for central neurons. Although the basis for this difference has not been established, it illustrates that BH4 metabolism can differ between populations of BH4containing cells. Indeed, diversity in BH4 metabolism across cell types may be responsible, in part, for the differential sensitivities of phenylalanine, catecholamine, and indoleamine metabolism to genetic abnormalities of BH4 biosynthesis in humans (Kaufman et al., 1983; Mclnnes et al., 1984). Cellular heterogeneity of BH4 metabolism is also suggested by differences in the ability of the compounds NAS and DAHP to inhibit BH4 biosynthesis.

2053

Concentrations of 100 pM NAS and 5 mM DAHP have previously been shown to produce maximal levels of inhibition in central neurons (Kapatos, 1990). In contrast, as reported here, maximal levels of inhibition of BH4 biosynthesis in peripheral neurons were obtained at concentrations of 2.5 mM NAS and 1 mM DAHP. BH4 synthesis by SCG neurons was therefore 25 times less sensitive to NAS and five times more sensitive to DAHP than in central neurons. NAS is also relatively ineffective at inhibiting BH4 biosynthesis in certain cell lines (Smith et al., 1985) and adrenal chromaffin cells (see Abou-Donia et al., 1986). Although these differences could reflect an artifact of tissue culture or the metabolism or mechanism by which NAS enters the cell, they are also consistent with a difference in the BH4 biosynthetic pathway within central and peripheral neurons. This premise is supported by clinical reports of a novel form of hyperphenylalaninemia, characterized by a defect in hepatic BH4 biosynthesis, that does not involve the CNS (Hreidarsson et al., 1982; Blau et al., 1990). The differential sensitivity to NAS exhibited by central and peripheral neurons may find its biochemical basis in the seemingly redundant utilization of the enzymes sepiapterin reductase and 6-pyruvoyl reductase in the reduction of BH4 intermediates during the last two steps of the BH4 biosynthetic pathway (Milstien and Kaufman, 1989; Levine et al., 1990; but see Smith and Nichol, 1986). Based on this hypothesis, cells from the periphery would express equal levels of both of these enzymes and, as a result, BH4 synthesis would be less sensitive to inhibition of sepiaptenn reductase by NAS, whereas cells from the CNS would express a preponderance of sepiapterin reductase and BH4 synthesis would thus be more sensitive to NAS. This possibility is supported by the observation that a concentration of NAS (200 p M ) that was unable to inhibit endogenous BH4 synthesis by SCG was able to inhibit the conversion of exogenous sepiapterin to BH4 (Kapatos et al., 1990). Nerve impulse flow has been shown to regulate NE biosynthesis within sympathetic neurons, in part by producing a stable activation of TH that can be detected in cell-free homogenates as an increase in the affinity of the enzyme for the pteridine cofactor (Zigmond et al., 1989).For such a mechanism to be operative in the intact neuron, the concentration of BH4 must also be subsaturating (Iuvone et al., 1985). The results of studies that have increased BH4 levels suggest that the concentration of cofactor may be limiting for monoamine biosynthesis (Kettler et al., 1974; Pollock et al., 1981; Miwa et al., 1985; Sawada et al., 1986). Based on these observations and the rapid rate of BH4 metabolism, it was proposed that alterations in the intracellular concentration of BH4 brought about, for example, by the hypothetical coupling of BH4 metabolism to nerve impulse flow could rapidly

J. Neurochem., Vol. 59, No. 6, 1992

G. KAPATOS ET AL.

2054

modify the ability of the monoaminergic neuron to respond to changes in neurotransmitter release (Kapatos, 1990). The results of the present study suggest otherwise. First, there did not appear to be a clear relationship between the rate of tyrosine hydroxylation and the level of BH4. This observation is in agreement with results obtained previously in PC12 cells (Brautigam et al., 1984) or the rat brain (Suzuki et al., 1988). Second, the level of neurotransmitter did not respond rapidly to alterations in BH4 content. The slow rate of decline in NE following inhibition of BH4 synthesis was not the result of the slow turnover of NE. Rather, continued synthesis of NE, combined with the slow rate of NE turnover, could account for the gradual decline in NE content following BH4 synthesis inhibition. Alternatively, it remains a possibility that DAHP is able to decrease the rate of NE turnover by a mechanism unrelated to its effect on BH4 biosynthesis, perhaps by inhibiting NE release Oi metabolism. This possibility appears to be unlikely, however, because others have investigated the effect of DAHP administration on levels of monoamine metabolites in rat brain and found them to be unaltered (Suzuki et al., 1988). What mechanism might be operating to maintain TH activity at 25% of control levels despite a 90-95% decline in BH4 content? This apparent disparity may be explained by the nonlinear relationship between substrate concentration and enzyme activity that is determined by the Michaelis constant. For example, if the intracellular concentration of BH4 is 100 p M and the K,,, of TH for BH4 is in the range of 2 mM (Miller and Lovenberg, 1985), then a 90% decline in BH4 content would decrease TH activity by 90%. If, on the other hand, the K,,, ofTH for BH4 is reduced to 10 pM, then a 90% decline in BH4 content would decrease the rate of enzyme activity by only 45%. The ability of the neuron to maintain synthesis of NE despite a precipitous decline in BH4 concentration could therefore be explained if the low-K, form ofTH exists within these neurons. Future experiments will examine this possibility. Acknowledgment: The authors would like to thank Dr. Robert Davis of Parke-Davis (Ann Arbor, MI, U.S.A.) for teaching us the SCG dissection and Drs. M. J. Bannon, A. S. Freeman, and R. L. Shoemaker for critical reading of the manuscript. This work was supported by U.S. Public Health Service grant NS 2608 1.

REFERENCES Abou-Donia M. M., Wilson S. P., Zimmerman T. P., Nichol C. A., and Viveros 0. H. (1986) Regulation of guanosine triphosphate cyclohydrolase and tetrahydrobiopterin levels and the role of the cofactor in tyrosine hydroxylation in primary cultures of adrenomedullary chromaffin cells. J. Neurochem. 46, 1190-1 199.

J. Neurmhem.. Vol. 59. No. 6, 1992

Blau N., Endres W., Guardamagna O., Ferrero G. B., Ferraris S., and Ponzome R. (1990) Unusual case of atypical PKU: peripheral or central form of PPH4S deficiency?, in Chemistry and Biology of Pteridines (Curtius H.-Ch., Ghisla S., and Blau N., eds), pp. 4 18-42 I , Walter de Gruyter, Berlin. Bradford M. M. (1 976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72,248-254. BrLutigam M., Dreesen R., and Herken H. (1984) Tetrahydrobiopterin and total biopterin content of neuroblastoma (NIE-I 15, N2A) and pheochromocytoma (PC-12) clones and the dependence of catecholamine synthesis on tetrahydrobiopterin concentration in PC-12 cells. J. Neurochem. 42, 390-396. Fukushima T. and Nixon J. C. (1980) Analysis of reduced forms of biopterin in biological tissues and fluids. Anal. Biochem. 102, 176- 188.

Gil E. M., Nelson J. M., and Sherman A. D. (1978) Biopterin. 111. Purification and characterization of enzymes involved in the cerebral synthesis of 7,s-dihydrobiopterin. Neurochem. Res. 3, 69-88.

Hreidarsson S., Valle D., Holtzman N.,Coyle J., Singer H., Kapatos G., and Kaufman S. (1982) A peripheral defect in biopterin biosynthesis: a new mutant? Pediatr. Res. 16, 192. b o n e P. M., Reinhard J. F. Jr., Abou-Donia M. M., Viveros 0. H., and Nichol C. A. (1985) Stimulation of retinal dopamine biosynthesis in vivo by exogenous tetrahydrobiopterin: relationship to tyrosine hydroxylase activation. Brain Res. 359,392-396.

Kapatos G. (1990) Tetrahydrobiopterin synthesis rate and turnover time in neuronal cultures from embryonic rat mesencephalon and hypothalamus. J. Neurochem. 55, 129- 136. Kapatos G. and Kaufman S. (1983) Inhibition of pterin biosynthesis in the adrenergic neuroblastoma N I E 1 15 by tetrahydrobiopterin and folate, in Chemistry and Biology of Pteridines (Blair J. A., ed), pp. 171-175. Walter de Gruyter, Berlin. Kapatos G., Kaufman S., Weller J. L., and Klein D. C. (1982) Development and regulation of rat pineal biopterin content, in Developmental Neurobiology of the Melatonin Rhythm Cenerating Sysrem (Klein D. C., ed), pp. 165-272. S. Karger, Basel. Kapatos G., Kaufman S., Weller J. L., and Klein D. C. (1983) The development of tetrahydrobiopterin and guanosine-5'-triphosphate cyclohydrolase: differential patterns in rat brain and pineal gland. Brain Res. 258, 351-355. Kapatos G., Hasegawa H., Hirayama K., and Kemski V. ( I 990) Differential tetrahydrobiopterin metabolism in central and peripheral monoamine neurons maintained in culture, in Chemistry and Biology ofpteridines (Curtius H.-Ch., Ghisla S., and Blau N., eds), pp. 376-379. Walter de Gruyter, Berlin. Katoh S., Sueoka T., and Yamada S. (1982) Direct inhibition of brain sepiapterin reductase by a catecholamine and an indoleamine. Biochem. Biophys. Res. Commun. 105, 75-8 1. Kaufman S. (1974) Properties of the pterin-dependent aromatic amino acid hydroxylases, in Ciba Foundation Symposium, Vol. 22: Aromatic Amino Acids in the Brain, pp. 85- I 15. Elsevier, Amsterdam. Kaufman S., Kapatos G., Rizzo W. B., Schulman J. D., Tamarkin L., and Van Loon G. L. (1983) Tetrahydropterin therapy for hyperphenylalaninemia caused by defective synthesis of tetrahydrobiopterin. Ann. Neurol. 14, 308-3 15. Kessler J. A. (1985) Differential regulation ofpeptide and catecholamine characters in cultured sympathetic neurons. Neuroscience 15, 827-839. Kettler R., Bartholini G., and Pletscher A. (1974) In vivo enhancement of tyrosine hydroxylation in rat striatum by tetrahydrobiopterin. Nature 249, 476-478. Levine R. A., Kapatos G., Kaufman S., and Milstien S. (1990) Immunological evidence for the requirement of sepiapterin reductase for tetrahydrobiopterin biosynthesis in brain. J. Neurochem. 54, 1218-1224. McInnes R. R., Kaufman S., Warsh J. J., Van Loon G. R., Milstien S., Kapatos G., Soldin S., Walsh P., MacGregor D., and Han-

BH4 TURNOVER IN CULTURED SYMPATHETIC NEURONS ley W. B. (1984)Biopterin synthesis defect: treatment with L-Dopa and 5-hydroxytryptophan compared with therapy with a tetrahydropterin. J. Clin. Invest. 73,458-469. Michaud R. L., Bannon M. J., and Roth R. H. (1981)A sensitive and specific method using C, (octyl) columns for the analysis of catecholamines by ion-pair reversed-phase liquid chromatography with amperometric detection. J. Chromafogr. 225, 335-345. Miller L.P.and Lovenberg W. (1985)The use ofthe natural cofacin the analysis of nontor 6(R)-~-erythro-tetrahydrobiopterin phosphorylated and phosphorylated rat striatal tyrosine hydroxylase at pH 7.0.Neurochem. Int. 7, 689-697. Milstien S.and Kaufman S.(1989)The biosynthesis of tetrahydrobiopterin in rat brain: purification and characterization of 6pyruvoyl tetrahydropterin (2'-0x0) reductase. J. Biol. Chem. 264,8066-8073. Miwa S., Watanabe Y.,and Hayaishi 0. (1985) 6R-~-erythre 5,6,7,8-Tetrahydrobiopterinas a regulator of dopamine and serotonin biosynthesis in the rat brain. Arch. Biochem. Biophys. 239,234-24 1. Pollock R. J., Kapatos G., and Kaufman S.(1 98 1) Effect of cyclic AMP-dependent protein phosphorylating conditions on the pH-dependent activity of tyrosine hydroxylase from beef and rat striata. J. Neurochem. 37, 855-860. Rembold H. (I 982) Balancing of mammalian biopterin pools, in Biochemical and Clinical Aspects of Pferidines, Vol. I (Wachter H., Curtius H.-Ch., and Pfleiderer W., eds), pp. 5163.Walter de Gruyter, Berlin. Sawada M., Sugimoto T., Matsuura S., and Nagatsu T. (1 986)(6R)-

2055

Tetrahydrobiopterin increases the activity of tryptophan hydroxylase in rat raphe slices. J. Neurochem. 47, 1544-1 547. Sladeczek F. and Bockaert J. (1983)Turnover in vivo of alphaladrenergic receptors in rat submaxillary glands. Mol. PharmaCOI. 23,282-288. Smith G. K. and Nichol C. A. (1986)Synthesis, utilization, and structure of the tetrahydropterin intermediates in the bovine adrenal medullary de novo biosynthesis of tetrahydrobiopterin. J. Biol. Chem. 261, 2725-2737. Smith G.K., Duch D. S., and Nichol C. A. (1985)Inhibition of the tetrahydropterin pathway for biopterin cofactor biosynthesis by N-acetylserotonin in cell cultures and in enzyme preparations and studies on the mechanism of tetrahydropterin formation, in Biochemical and Clinical Aspects of Pteridines, Vol. 4 (Wachter H. and Curtius H., eds), pp. 175-193. Walter de Gruyter, Berlin. Suzuki S., Watanabe Y . , Tsubokura S., Kagamiyama H., and Hayaishi 0. (1988)Decrease in tetrahydrobiopterin content and neurotransmitter amine biosynthesis in rat brain by an inhibitor ofguanosine triphosphate cyclohydrolase. Brain Res. 446, 1-10. Werner G., Werner E. R., Fuchs D., Hausen A., Reibnegger G., and Wachter H. (1990)Tetrahydrobiopterindependent formation of nitrite and nitrate in murine fibroblasts. J. Exp. Med. 172, 1599- 1607. Zigmond R. E.,Schwarzschild M. A., and Rittenhouse A. R. (1989) Acute regulation of tyrosine hydroxylase by nerve activity and by neurotransmitters via phosphorylation. Annu. Rev. Neurosci. 12, 415-461.

J. Neurochem., Vol. 59. No. 6,1992