RELATIONSHIP BETWEEN THE RATE OF AXOPLASMIC TRANSPORT AND SUBCELLULAR DISTRIBUTION OF ENZYMES INVOLVED IN THE SYNTHESIS OF NOREPINEPHRINE

The net rate of proximo-distal transport of tyrosine hydroxylase, dopamine β-hydroxylase, DOPA decarboxylase and choline acetyltransferase was determined by measuring the accumulation of these enzymes proximal to a ligature of the rat sciatic nerve. The rate of accumulation was constant for at least 12 h. For the enzymes involved in the biosynthesis of norepinephrine the rate of transport was correlated to their subcellular distribution and a close correlation between these two parameters was found. Dopamine β-hydroxylase, an enzyme mainly localized in the particulate fraction of the sciatic nerve, showed the fastest rate of transport (1·94 mm/h) whereas DOPA decarboxylase, exclusively located in the high-speed supernatant fluid, gave the slowest (0·63 mm/h) rate of transport. Tyrosine hydroxylase, predominantly located in the non-particulate fraction of the sciatic nerve was transported much slower (0·75 mm/h) than dopamine β-hydroxylase but still significantly (P < 0.005) faster than DOPA decarboxylase. The subcellular distribution of dopamine β-hydroxylase in ganglia did not differ significantly (0·45 > P > 0·40) from that in the sciatic nerve, but in nerve endings a greater proportion of dopamine β-hydroxylase was localized in particulate fractions. Tyrosine hydroxylase and DOPA decarboxylase were found exclusively in the non-particulate fractions of ganglia. In the nerve endings of the effector organs a small but consistent portion of tyrosine hydroxylase was found in particulate fractions, whereas DOPA decarboxylase was exclusively localized in the high-speed supernatant fluid.     

Role of Aromatic l-Amino Acid Decarboxylase for Dopamine Replacement by Genetically Modified Fibroblasts in a Rat Model of Parkinson's Disease

Abstract: Investigations of gene therapy for Parkinson's disease have focused primarily on strategies that replace tyrosine hydroxylase. In the present study, the role of aromatic l -amino acid decarboxylase in gene therapy with tyrosine hydroxylase was examined by adding the gene for aromatic l -amino acid decarboxylase to our paradigm using primary fibroblasts transduced with both tyrosine hydroxylase and GTP cyclohydrolase I. We compared catecholamine synthesis in vitro in cultures of cells with tyrosine hydroxylase and aromatic l -amino acid decarboxylase together versus cocultures of cells containing these enzymes separately. l -DOPA and dopamine levels were higher in the cocultures that separated the enzymes. To determine the role of aromatic l -amino acid decarboxylase in vivo, cells containing tyrosine hydroxylase and GTP cyclohydrolase I were grafted alone or in combination with cells containing aromatic l -amino acid decarboxylase into the 6-hydroxydopamine-denervated rat striatum. Grafts containing aromatic l -amino acid decarboxylase produced less l -DOPA and dopamine as monitored by microdialysis. These findings indicate that not only is there sufficient aromatic l -amino acid decarboxylase near striatal grafts producing l -DOPA, but also the close proximity of the enzyme to tyrosine hydroxylase is detrimental for optimal dopamine production. This is most likely due to feedback inhibition of tyrosine hydroxylase by dopamine.     

Role of tyrosine, DOPA and decarboxylase enzymes in the synthesis of monoamines in the brain of the locust

The metabolic transformation of tyrosine (TYR) by the decarboxylase and hydroxylase enzymes was investigated in the central nervous system of the locust, Locusta migratoria. It has been demonstrated that the key amino acids, 3,4-dihydroxyphenylalanine (DOPA), 5-hydroxytryptophan (5HTP) and tyrosine are decarboxylated in all part of central nervous system. DOPA and 5HTP decarboxylase activities show parallel changes in the different ganglia, but the rank order of the activity of TYR decarboxylase is different. Enzyme purification has revealed that the molecular weights of TYR decarboxylase and DOPA/5HTP decarboxylase are 370,000 and 112,000, respectively. The decarboxylation of DOPA by DOPA/5HTP decarboxylase is stimulated, whereas the decarboxylation of DOPA by TYR decarboxylase is inhibited in the presence of the cofactor pyridoxal-5'-phosphate. TYR hydroxylase could not be detected and 3H-TYR is found to be metabolised to tyramine (TA), but not to DOPA. The haemolymph contains a significant concentration of DOPA (120 pmol/100 microl haemolymph), and the ganglia incorporates DOPA from the haemolymph by a high affinity uptake process (K(M)=12 microM and V(max)=24 pmol per ganglion/10 min). Our results suggest that no tyrosine hydroxylase is present in the locust CNS and the DOPA uptake into the ganglia by a high affinity uptake process as well as the DOPA decarboxylase enzyme may be responsible for the regulation of the ganglionic dopamine (DA) level. Two types of decarboxylases exist, one of them decarboxylating DOPA and 5HTP (DOPA/5HTP decarboxylase), other decarboxylating TYR (TYR decarboxylase). The DOPA/5HTP decarboxylase enzyme present in the insect brain may correspond to the 5HTP/DOPA decarboxylase in vertebrate brain, whereas TYR decarboxylase is characteristic only for the insect brain.     

Adrenergic enzymes in cultured mouse neuroblastoma: absence of detectable aromatic-L-amino-acid decarboxylase.

The relative activities of tyrosine hydroxylase, aromatic-l -amino-acid decarboxylase and dopamine beta-hydroxylase were established in a number of clones of neuroblastoma cells isolated from the uncloned mouse C-1300 tumor. One clone, NBD-2, was chosen for further analysis on the basis of its relatively high activities of tyrosine hydroxylase and dopamine beta-hydroxylase. The levels of these enzymes, and monoamine oxidase and catechol O-methyltransferase, were at least 20-80 fold lower in the neuroblastoma culture than in mouse superior cervical ganglion. More importantly, aromatic-l -amino-acid decarboxylase activity was not even detectable in any neuroblastoma clone examined. Based on the relative sensitivities of the tyrosine hydroxylase and aromatic-l -amino-acid decarboxylase assays and on the ratio of these two enzymes in the mouse ganglion, decarboxylase activity is more than 10 fold lower in the cultured cells than would be predicted on the basis of tyrosine hydroxylase activity. Dialysis and mixing studies with neuroblastoma extracts and partially purified aromatic-l -amino-acid decarboxylase did not reveal the presence of any endogenous inhibitors that could account for the low level of decarboxylase activity in the cultured cells. During growth of the neuroblastoma cells to confluency, only one enzyme, monoamine oxidase, exhibited an elevated specific activity on the basis of cell number. However, when based on the amount of protein, the specific activity of all measurable enzymes increased in culture-because cell protein decreased 5 fold during growth to confluency. These findings are discussed with respect to individual cell function.     

AXONAL TRANSPORT OF CATECHOLAMINE SYNTHESIZING AND METABOLIZING ENZYMES

The rates of accumulation of the catecholamine synthesizing and metabolizing enzymes proximal to a ligation on the sciatic nerve of the rat were studied. Dopamine-β hydroxylase (EC 1.14.2.1) and tyrosine hydroxylase (EC 1.14.3a) accumulated at a similar rapid rate, and catechol-O-methyl-transferase (EC 2.1.1.6), choline acetyltransferase (EC 2.3.1.6) and monoamine oxidase (EC 1.4.3.4) accumulated at the same slow rate, whereas DOPA decarboxylase (EC 4.1.1.26) accumulated at an intermediate rate. Based on clearance of the rapidly accumulating enzymes, absolute flow rates were estimated to be: 106-167 mm/24 h for tyrosine hydroxylase; 138-185 mm/24 h for dopamine-β-hydroxylase; and 36-86 mm/24 h for DOPA decarboxylase. In contrast, the mean rate of transport of the slowly accumulating enzymes (monomine oxidase, catechol-O-methyltransferase and choline acetyltransferase) was approximately 3 mm/24 h. Colchicine and vinblastine completely blocked the axonal transport of both the rapidly and slowly transported enzymes. Studies of the subcellular distribution of each enzyme failed to confirm the suggestion that particulate enzymes are transported rapidly and soluble enzymes slowly. Our results suggest that the transport and inactivation of dopamine-β-hydroxylase, DOPA decarboxylase, and tyrosine hydroxylase are under different controls than monoamine oxidase and catechol-O-methyltransferase.     

Regional development of catecholamine biosynthesis in rat brain

Abstract— The ontogenetic development of norepinephrine and dopamine and their associated biosynthetic and degradative enzymes was investigated in five anatomical regions of the rat brain. Clear regional differences were found in the development of both norepinephrine and tyrosine hydroxylase (EC 1.14.3.1). In the case of both norepinephrine and tyrosine hydroxylase, brainstem structures achieved adult levels well before forebrain structures. The development of DOPA decarboxylase (EC 4.1.1.26), monoamine oxidase (EC 1.4.3.4) and catechol-0-methyl transferase (EC 2.1.1.6) did not appear to differmarkedly from area to area. Further analysis of the data revealed that in forebrain structures both the amines and the biosynthetic enzymes developed concurrently. By contrast, in the brainstem structures, there was a dissociation of amine and enzyme development with development of tyrosine hydroxylase, in particular, markedly preceding that of norepinephrine and of DOPA decarboxylase. The bases for both the lower amine levels in the infant brain and the regional developmental differences are discussed in relation to the anatomical organization of the central catecholamine-containing neurons.     

THE EFFECT OF HYPOXIA ON MONOAMINE SYNTHESIS, LEVELS AND METABOLISM IN RAT BRAIN

Abstract— Rats were exposed to 5.6% oxygen environments for up to 2 h. The accumulation of brain DOPA and 5-hydroxytryptophan at 30 min after decarboxylase inhibition was used to estimate cerebral tryosine and tryptophan hydroxylase activity, respectively, in vivo. There was a continuing decrease in tryosine hydroxylase activity during the 2 h in whole brain as well as five brain regions. Tryptophan hydroxylase activity declined during the 1st h, but then increased towards control levels during the 2nd h. There was an increase in brain tryptophan during the 2nd h as well. In whole brain and the five brain regions, there was no significant change in the levels of noradrenaline, dopamine or 5-hydroxytrypamine. During a 1 h exposure to 5.6% oxygen, there was decreased accumulation of noradrenaline, dopamine and 5-hydroxytryptamine after MAO inhibition and decreased accumulation of homovanillic acid and 5-hydroxyindoleacetic acid after probenecid administration. The dercreased synthesis and metabolism of the monoamines is most likely attributable to insufficient brain tissue oxygen as a substrate for the two hydroxylase enzymes.     

Sex steroids do not alter sex differences in tyrosine hydroxylase activity of dopaminergic neurons in vitro

Summary In order to distinguish the effects of genetic sex from those of sex hormones on the sexual differentiation of dopaminergic neurons, catecholamine synthesis was studied in gender-specific cultures of embryonic day-14 rat diencephalon. In addition to embryos from normal dams, embryos were used whose mothers had been treated with the estrogen antagonist tamoxifen or the testosterone antagonist cyproterone acetate on days 12 and 13 of gestation. Cultures from embryos of untreated dams were fed daily with a medium containing 17-estradiol or testosterone. After 10 days in vitro, cultures were immunostained for tyrosine hydroxylase and the accumulation of dihydroxyphenylalanine (DOPA) was measured in the presence of the DOPA decarboxylase inhibitor NSD 1015. Rates of DOPA synthesis, unlike the numbers of tyrosine hydroxylase-immunoreactive neurons, were markedly higher in female cultures under all experimental conditions. Treatment of dams with antisteroids prior to removal of the embryos had no influence on these results. Treatment of cultures with both steroids decreased DOPA formation in a dose-dependent manner without altering the sex difference. These results suggest that cultured diencephalic dopaminergic neurons develop sex differences in the activity of tyrosine hydroxylase. This sexual dimorphism is initiated independently of the action of gonadal steroid hormones. Sex hormones exert an additional modulatory influence on the activity of the enzyme but do not abolish or reverse sex differences. Therefore, the concept of a purely epigenetic mode of sexual differentiation of the mammalian brain needs to be broadened to incorporate other mechanisms, such as the cell-autonomous fulfillment of a sex-specific genetic program.     

In vivo release of dopa and dopamine from genetically engineered cells grafted to the denervated rat striatum

Fibroblastic 3T3 and endocrine RIN cells were genetically modified by infection with a recombinant retrovirus encoding the form I of human tyrosine hydroxylase (TH) and selection in tyrosine-free medium. These cells were grafted to rats unilaterally lesioned with 6-hydroxy-dopamine. Both cell types survived implantation into the striatum, expressed TH immunoreactivity, and as assessed by microdialysis 8-9 days after implantation, secreted high amounts of DOPA and/or dopamine into the surrounding host striatum. The modified 3T3 cells secreted large amounts of DOPA that was efficiently decarboxylated to dopamine by the host striatal tissue; the newly synthesized dopamine was stored only to a limited extent in the denervated striatum. The modified RIN cells synthesized dopamine that was stored intracellularly and released in a regulated fashion. The grafted DOPA-secreting cells produced 4-5 times higher extracellular dopamine levels than the dopamine-secreting cells, and they were more efficient in reducing apomorphine-induced rotation. No effect was observed with either cell type on amphetamine-induced turning behavior.     

Concomitant elevation of tyrosine hydroxylase and dopamine beta-hydroxylase by cyclic AMP in cultured mouse neuroblastoma cells.

The effects of cyclic AMP analogues and of phosphodiesterase inhibitors were investigated in neuroblastoma cells (NBD-2) cloned from the C-1300 tumor. 8Br-cAMP and phosphodiesterase inhibitors that elevated cAMP induced large (greater than 15 fold) and specific increases in tyrosine hydroxylase and dopamine beta-hydroxylase activity. In contrast, catechol O-methyltransferase, monoamine oxidase and aromatic-l -amino-acid decarboxylase were unaffected by the cAMP altering drugs. Similarly, AChE was unaffected and only a small increase in choline acetyltransferase (3 fold) was observed. The increases in tyrosine hydroxylase and dopamine beta-hydroxylase were similar with respect to dose response relationships and with respect to time course of onset. Only those phosphodiesterase inhibitors that elevated cAMP (papaverine and Ro20-1724 as opposed to theophylline) were effective in elevating tyrosine hydroxylase and dopamine beta-hydroxylase. Further, the doses optimal for elevating cAMP coincided with the optimal doses for elevating the two enzymes. Theophylline had no influence either upon NBD-2 cell cAMP levels or upon tyrosine hydroxylase and dopamine beta-hydroxylase activity. The changes in protein synthesis rates produced by the cAMP altering drugs were temporally distinct from the changes in either tyrosine hydroxylase or dopamine beta-hydroxylase. These results suggest that the intracellular messenger compound cAMP is involved in the specific regulation of both tyrosine hydroxylase and dopamine beta-hydroxylase in adrenergic cells.     

3,4-Dihydroxyphenylalanine (DOPA) metabolism and retinoic acid induced differentiation in human neuroblastoma

In mature cells of the sympathetic nervous system and the adrenal gland, the activity of dihydroxyphenylalanine decarboxylase (DDC) is higher than that of tyrosine hydroxylase and 3,4-dihydroxyphenylalanine (DOPA) does not accumulate in the cells. On the other hand, it is known that in some neuroblastoma cells there is a relative deficiency of DDC, resulting in accumulation and secretion of DOPA. Such a relative deficiency of DDC is a characteristic of neural cells at an early stage of neural crest development, suggesting the neuroblastoma are cells arrested in early neural crest development. If this were the case, it is possible that agents such as retinoic acid (RA) could induce neuroblastoma to differentiate into mature cells with respect to their metabolism of catecholamines. We have measured the effect of RA on the metabolism of DOPA and expression of tyrosine hydroxylase and DDC in human neuroblastoma cell lines, CHP-126, CHP-134, IMR-32, NB-59, and LA-N-5. When the cell cultures were treated with RA, they showed wide variations in response as measured by morphological change, growth inhibition, enzyme activities and DDC, but does not increase DDC relative to tyrosine hydroxylase. It is concluded that RA does not induce biochemical differentiation of the neuroblastoma into mature cells even when there are extensive morphological changes and suppression of growth rate.     

REGULATION OF TYROSINE HYDROXYLASE ACTIVITY IN MOUSE NEUROBLASTOMA CLONE N1E-115

Abstract— Mouse neuroblastoma (clone N1E-115) cells in the logarithmic growth phase were incubated for 12 days. From early log phase to late stationary phase, the specific activity of tyrosine hydroxylase (EC 1.14.3a) increased greater than 30-fold. The increase in tyrosine hydroxylase per cell and per dish was 12- and 2700-fold, respectively. When cell division was stopped by removing serum or by adding 0.1 m m -5-fluorodeoxyuridine and 0.1 m m -uridine, the enzyme activity was also found to increase. These results show that tyrosine hydroxylase is regulated in neuroblastoma clone N1E-115.     

Use of Microdialysis for Monitoring Tyrosine Hydroxylase Activity in the Brain of Conscious Rats

An on-line microdialysis system was developed which monitored the 3,4-dihydroxyphenylalanine (DOPA) formation in the striatum during infusion of a submicromolar concentration of an L-aromatic amino-acid decarboxylase inhibitor (NSD 1015). The absence of DOPA in dialysates of 6-hydroxydopamine-pretreated rats and the disappearance of DOPA after administration of alpha-methyl-p-tyrosine indicated that the dialyzed DOPA was derived from dopaminergic nerve terminals. Next we investigated whether the steady-state DOPA concentration in striatal dialysates could be considered as an index of tyrosine hydroxylase activity. The increase in DOPA output after intraperitoneal administration of haloperidol or gamma-butyrolactone and the decrease in DOPA output after intraperitoneal administration of apomorphine are in excellent agreement with results of postmortem studies, in which a decarboxylase inhibitor was used to measure the activity of tyrosine hydroxylase. The effect of haloperidol on DOPA formation was not visible when a U-shaped cannula (0.80 mm o.d.) was used. Some methodological problems related to microdialysis of the haloperidol-induced increase in DOPA formation are discussed. We concluded that the proposed model is a powerful and reliable in vivo method to monitor tyrosine hydroxylase activity in the brain. The method is of special interest for investigating the effect of compounds which are not able to pass the blood-brain barrier. As an application of the method in the latter situation, we report the effect of infusion the neurotoxin 1-methyl-4-phenylpyridinium ion (10 mmol/L infused over 20 min) on the activity of striatal tyrosine hydroxylase.     

Changes in enzyme patterns produced by high potassium concentration and dibutyryl cyclic AMP in organ cultures of sympathetic ganglia

—Preliminary experiments had shown that acetylcholine, the putative mediator of trans-synaptic induction of tyrosine hydroxylase (TH) and dopamine β-hydroxylase (DBH) in vivo, did not lead to an increase in these enzyme activities in mouse superior cervical ganglia kept in organ culture. It was the aim of the present study to evaluate whether increases in tyrosine hydroxylase and dopamine β-hydroxylase evoked by other stimuli such as potassium or dibutyryl cyclic AMP in such an in vitro system are representative for in vivo trans-synaptic induction where changes in the levels of enzymes involved in norepinephrine synthesis or degradation are strictly confined to TH and DBH. In the presence of elevated concentrations of potassium or 5 mm dibutyryl cyclic AMP under organ culture conditions TH and DBH as well as DOPA decarboxylase and monoamine oxidase were significantly (P < 0.025) increased. The increase in total activities of TH and DBH were completely, those of DOPA decarboxylase and monoamine oxidase partially, inhibited by cycloheximide. In the presence of high concentrations of potassium, the total protein content of the ganglia was 28 per cent higher than in culture controls while dibutyryl cyclic AMP had no significant effect. Cycloheximide alone caused the protein content to fall to 70 per cent of that in control cultures. The loss of protein in the presence of cycloheximide was not accompanied by a simultaneous loss of TH, DOPA decarboxylase or monoamine oxidase, but DBH was decreased. Potassium was shown to increase the incorporation of [³H]leucine into TCA-insoluble protein during an early culture period but dibutyryl cyclic AMP showed no such effect. An increase in the rate of incorporation of [³H]leucine into protein was seen in both the control and elevated potassium cultures after 48 h. This increase did not occur in the presence of dbcAMP. The difference in enzyme patterns under conditions of elevated potassium and dibutyryl cyclic AMP and the fact that no changes in the levels of endogenous cyclic AMP were observed during exposure to 54 mm -potassium for a time period sufficient to initiate changes ultimately leading to elevated TH levels argues against the mediation of the potassium-induced enzyme increases by cAMP. Since changes in enzyme patterns caused by potassium and dbcAMP were not similar to patterns seen in vivo under conditions of trans-synaptic induction we conclude that use of this system as an in vitro model for in vivo trans-synaptic induction necessitates great caution.     

Human TRP-1 Has Tyrosine Hydroxylase but no DOPA Oxidase Activity

Human TRP-1 has been immunopurified from normal human melanocytes cultured from black neonatal subjects and used to investigate the catalytic function of TRP-1 for the two substrates, L-tyrosine and L-DOPA. Immunopurified TRP-1 did not demonstrate DOPA staining on SDS/PAGE nor DOPA oxidase (DO) activity with either routine or modified assays. The purified TRP-1 also demonstrated no tyrosine hydroxylase (TH) activity using the routine Pomerantz assay. However, there was apparent TH activity exhibited by immunopurified TRP-1 under conditions with low tyrosine concentration (≤0.8 μCi/ml of ³H-tyrosine), prolonged incubation time (i.e., overnight) and in the absence of the cofactor L-DOPA. Using these latter specific conditions, TH activity was also detected in cell lysates from a tyrosinase-negative albino melanocyte line which exhibited no TH activity with the routine Pomerantz assay. In addition, TH activity under low substrate assay conditions was not exhibited in a melanocyte line derived from a TRP-1 deficient, Brown albino individual. However, the absence of TH in this Brown albino cell line could be compensated for by the addition of L-DOPA to the assay. These results suggested that TRP-1 has some tyrosine hydroxylase but no DOPA oxidase activity. We propose that one function of TRP-1 is to modulate tyrosinase activity by making DOPA available as a cofactor to perpetuate the initial steps in melanogenesis.     

Mast cells express tyrosine hydroxylase and store dopamine in a serglycin-dependent manner

Here we show that mast cells contain dopamine and that mast cell activation causes dopamine depletion, indicating its presence within secretory granules. Dopamine storage increased during mast cell maturation from bone marrow precursors, and was dependent on the presence of serglycin. Moreover, the expression of tyrosine hydroxylase, the key enzyme in dopamine biosynthesis, was induced during mast cell maturation; histidine decarboxylase and tryptophan hydroxylase 1 were also induced. Mast cell activation caused a robust induction of histidine decarboxylase, but no stimulation of tyrosine hydroxylase or tryptophan hydroxylase 1 expression. The present study points toward a possible role of dopamine in mast cell function.     

Pancreatic acinar cells utilize tyrosine to synthesize L-dihydroxyphenylalanine

The pancreatic β cells can synthesize dopamine by taking L-dihydroxyphenylalanine, but whether pancreatic acinar cells synthesize dopamine has not been confirmed. By means of immunofluorescence, the tyrosine hydroxylase -immunoreactivity and aromatic amino acid decarboxylase (AADC)- immunoreactivity were respectively observed in pancreatic acinar cells and islet β cells. Treatment with L-dihydroxyphenylalanine, not tyrosine, caused the production of dopamine in the incubation of INS-1 cells (rat islet β cell line) and primary isolated islets, which was blocked by AADC inhibitor NSD-1015. However, only L-dihydroxyphenylalanine, but not dopamine, was detected when AR42J cells (rat pancreatic acinar cell line) were treated with tyrosine, which was blocked by tyrosine hydroxylase inhibitor AMPT. Dopamine was detected in the coculture of INS-1 cells with AR42J cells after treatment with tyrosine. In an in vivo study, pancreatic juice contained high levels of L-dihydroxyphenylalanine and dopamine. Both L-dihydroxyphenylalanine and dopamine accompanied with pancreatic enzymes and insulin in the pancreatic juice were all significantly increased after intraperitoneal injection of bethanechol chloride and their increases were all blocked by atropine. Inhibiting TH with AMPT blocked bethanechol chloride-induced increases in L-dihydroxyphenylalanine and dopamine, while inhibiting AADC with NSD-1015 only blocked the dopamine increase. Bilateral subdiaphragmatic vagotomy of rats leads to significant decreases of L-dihydroxyphenylalanine and dopamine in pancreatic juice. These results suggested that pancreatic acinar cells could utilize tyrosine to synthesize L-dihydroxyphenylalanine, not dopamine. Islet β cells only used L-dihydroxyphenylalanine, not tyrosine, to synthesize dopamine. Both L-dihydroxyphenylalanine and dopamine were respectively released into the pancreatic duct, which was regulated by the vagal cholinergic pathway. The present study provides important evidences for the source of L-dihydroxyphenylalanine and dopamine in the pancreas.